Chemosphere 224 (2019) 816e826

Contents lists available at ScienceDirect

Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere

In-situ reductive degradation of chlorinated DNAPLs in contaminated

groundwater using polyethyleneimine-modified zero-valent iron
nanoparticles
Ndumiso Vukile Mdlovu, Kuen-Song Lin*, Chung-Yu Chen, Fikile Agath Mavuso,
Sikhumbuzo Charles Kunene, Maria Janina Carrera Espinoza
Department of Chemical Engineering and Materials Science/Environmental Technology Research Center, Yuan Ze University, ChungeLi District, Taoyuan
City 32003, Taiwan

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

On-site degradation of 1,2eDCE, TCE,

and PCE by PEIeZVIN was conducted.

1,2eDCE, TCE, and PCE were suc-
cessfully remediated to nontoxic
species.

2-D images of successful decontami-
nation of DNAPLs was confirmed by
RIP.

PEIeZVIN particles showed larger
surface areas and less aggregations
than that of ZVIN.

PEIeZVIN showed higher DNAPLs
removal efficiency than ZVIN.

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

Article history: Zero-valent iron nanoparticles (ZVIN) have found applications in many strategies for on-site soil and
Received 23 November 2018 groundwater decontamination. A number of studies have reported the prospective utilization of ZVIN in
Received in revised form the reduction of chlorinated organic compounds such as dense non-aqueous phase liquids (DNAPLs) in
19 February 2019
groundwater. Due to their bioaccumulation and carcinogenesis, DNAPLs in groundwater are a human
Accepted 22 February 2019
health hazard and pose environmental risks. Therefore, decontamination of these contaminants is
Available online 27 February 2019
Handling Editor: Daniel CW Tsang necessary. This study presents the in-situ remediation of trichloroethylene (TCE), perchloroethene (PCE),
Handling Editor: and 1,2-dichloroethene (1,2eDCE) DNAPLs through the direct injection of polyethylenimine (PEI)-coated
ZVIN (PEIeZVIN composite materials) to facilitate the reduction of contaminants in low-permeability
Keywords: media. A field test was conducted at the premises of a petrochemical company, situated in the Miaoli
Zero-valent iron nanoparticles County of Northern Taiwan that discharged significant amounts of DNAPLs. After in-situ injection and
Polyethylenimine one-day of reaction with groundwater contaminants, ZVIN was further characterized to examine its
DNAPLs efficacy in the reduction of pollutants. After the direct injection of PEIeZVIN, a notable reduction in the
Injection method concentration of DNAPLs was recorded with conversion from toxic to non-toxic substances. Use of re-
Decontamination sistivity image profiling (RIP) technique suggested similar conductivity data for the PEI-coated ZVIN
In-situ groundwater decontamination
suspension and groundwater samples. X-ray absorption near edge structure (XANES) and X-ray

* Corresponding author.
E-mail addresses: s1078804@mail.yzu.edu.tw (N.V. Mdlovu), kslin@saturn.yzu.
edu.tw (K.-S. Lin), s945242@mail.yzu.edu.tw (C.-Y. Chen), s1065231@mail.yzu.edu.
tw (F.A. Mavuso), s1078805@mail.yzu.edu.tw (S.C. Kunene), s1078802@mail.yzu.
edu.tw (M.J. Carrera Espinoza).

https://doi.org/10.1016/j.chemosphere.2019.02.160
0045-6535/© 2019 Elsevier Ltd. All rights reserved.
 N.V. Mdlovu et al. / Chemosphere 224 (2019) 816e826
absorption fine structure (EXAFS) studies depicted that the oxidation of ZVIN and PEIeZVIN was
occurring after the reductive reaction with contaminated groundwater. The reacted samples had bond
distance values of 1.98, 2.00, 1.96, and 1.94 Å. Combining floating surface-coated ZVIN and RIP technique
seems promising and environmentally attractive.
1. Introduction
Groundwater forms about 33% of all fresh water around the
world and owing to its outstanding quality than that of surface
water; groundwater is extensively used for human consumption,
agriculture, and industrial purposes (Mdlovu et al., 2018; Mo et al.,
2017). However, increasing industrialization and urbanization have
led to escalated levels of contaminants released into groundwater.
The discharge of harmful compounds and chemicals into ground-
water resulting from industrial water pollution has posed a huge
environmental risk affecting human health worldwide (Fu et al.,
2014; Mo et al., 2017). This problem has denied millions of people
the right to clean water around the globe. Resolution 64/292 of the
United Nations General Assembly dictates that clean drinking wa-
ter is critical to the fulfillment of all human rights (Heller, 2015).
In addition, chlorinated volatile organic compounds (CVOCs)
form a major part of contaminants from industrial waste. These
contaminants, which include perchloroethene (PCE), trichloroeth-
ylene (TCE), and 1,2edichloroethene (1,2eDCE), are denser than
water and contain low aqueous solubility; they are referred
collectively as dense non-aqueous phase liquids (DNAPLs) (Crane
and Scott, 2012; Lin et al., 2018; Lucattini et al., 2018). DNAPL
contaminants are dominant at contaminated sites and are very
difficult to degrade (Lin et al., 2018). Furthermore, the discharge of
DNAPLs from large numbers of public and private industries pose
substantial difficulties for the treatment and long-term manage-
ment of sites (Kavanaugh et al., 2003; Lin et al., 2018). These
DNAPLs have a film chemical structure which is very difficult to
degrade, and they are hazardous to the environment and human
health. Their high density and low solubility lead to their rapid
deposition below the groundwater as an impermeable layer. Thus,
the deposited DNAPLs become an endless source of groundwater
pollution, thereby posing serious risks to the environment and
human health (Bahrami et al., 2018; Lin et al., 2018; Lucattini et al.,
2018). Disruption of DNAPLs leads to contaminated plumes of low
permeability lenses or layers, groundwater, and pooling (US EPA,
1996; Mdlovu et al., 2018).
Over the last few decades, nanotechnology has been proven to
be a promising pathway for the treatment of contaminated
groundwater. Such technology can help develop technically and
economically feasible alternatives that overcome the limitations of
conventional methods. This approach entails the use of nano-scale
materials to develop nanoparticles that can degrade contaminants
(Qu et al., 2013; Chaturvedi and Dave, 2019).
Nano-scale zero-valent iron (ZVIN) has appeared as a promising
remediation method for aquifers contaminated by DNAPLs (Kim
et al., 2017; Sheu et al., 2018). The application of ZVIN for the
decontamination of harmful pollutants from groundwater and
contaminated sites has gained significant attention as a conse-
quence their large surface area and good reactivity, leading to the
easy reduction of contaminants (Lin and Chen, 2016; Gil-Díaz et al.,
2017; Kim et al., 2017). The injection of ZVIN reportedly leads to the
substantial degradation of total chlorinated composites such as TCE
and PCE. The high surface area of ZVIN escalates its reactivity with
pollutants. In addition, the oxidation of ZVIN will lead to the
817
© 2019 Elsevier Ltd. All rights reserved.
transfer of electrons to the contaminants to reduce them to non-
toxic elements. (Busch et al., 2015; Soukupova et al., 2015; Kim
et al., 2017; Litter et al., 2018). Regrettably, ZVIN is very prone to
agglomeration which is a phenomenon attributed to the quick
oxidation of metallic iron and strong magnetic properties. There-
fore, these newly developed aggregates are characterized as having
smaller surface active areas and lower reaction levels, limiting the
movement and stability of ZVIN in groundwater (Busch et al., 2015;
Ezzatahmadi et al., 2017; Huo et al., 2017; Teng et al., 2017; Litter
et al., 2018). An in-situ study revealed that the injected ZVIN
transported only a short distance from the injection area. This
problem is due to the aggregated ZVIN, which cannot flow
smoothly with moving groundwater to reach the contaminated
sites (Schrick et al., 2004; Lee et al., 2014; Gil-Díaz et al., 2017).
To enhance stability and mobility and avoid the aggregation and
oxidation of ZVIN in field applications, several studies have been
carried out to fabricate and modify the surface of ZVIN with poly-
meric materials (San Roma
n et al., 2016; Hu et al., 2018; Tasharrofi
et al., 2018). The coating of ZVIN with polymeric material protects it
from dissolved hydrogen and oxygen, thereby facilitating the ability
of ZVIN to decontaminate a wide range of contaminants (Golshirazi
et al., 2017; Lin et al., 2018; San Roma
n et al., 2016). Furthermore,
the limitations of ZVIN can be addressed by polyethylenimine (PEI)
coating on the surface of ZVIN particles to create PEIeZVIN nano-
composite material. Numerous studies have reported that polymers
can play a paramount role in modifying the surface of ZVIN
(Phenrat et al., 2008; Ravikumar et al., 2018). Interestingly, PEI,
which possesses an extensive range of functional groups, is highly
soluble in water, chemically stable, and strongly binds with metals.
The high macromolecular size and polyelectrolyte nature of PEI
render them appropriate for sequestering metal ions from dilute
solutions (Gil-Díaz et al., 2017; Andreev et al., 2018).
In Taiwan, treatment of contaminated soil and groundwater,
particularly those contaminated by chlorinated organic com-
pounds, is of prime concern with regard to environmental protec-
tion (Liu et al., 2016; Mdlovu et al., 2018). In the Miaoli County
located in Northern Taiwan, groundwater and soil are polluted with
DNAPLs at very high concentrations discharged from a local
petrochemical company (Yang et al., 2002; Mdlovu et al., 2018). The
Environmental Protection Agency (EPA) of Taiwan ordered the
company to carry out a decontamination works about two years
ago. Therefore, the excavation works were conducted for the on-
site decontamination soil and groundwater.
Hence, the objectives of this study were to assess the efficiency
of a ZVIN-mediated treatment process for 1,2eDCE, TCE, and PCE
contaminated aquifer. Moreover, the in-situ pumping of PEIeZVIN
by a direct injection technique was piloted and proceeded by
monitoring of the reactive period and stability of the injected
PEIeZVIN suspension in the plume. Groundwater monitoring was
conducted to observe the dissolution of iron species, DNAPL con-
taminants (1,2eDCE, TCE, and PCE), reduction mechanisms, and
other geochemical factors. By taking the contaminated water
samples from the treatment area to the laboratory, the reactivity
test of ZVIN suspension gathered from the aquifer was performed.
Moreover, XANES and EXAFS spectroscopy determined the
818 N.V. Mdlovu et al. / Chemosphere 224 (2019) 816e826
mechanisms and corrosion products of DNAPL degradation on the
uncoated ZVIN and PEIeZVIN surfaces.
2. Materials and methods
2.1. Preparation of ZVIN and PEIeZVIN nanoparticles
The ZVIN particles were synthesized in the laboratory through a
chemical reduction process described by Lin et al. (2018) and
Mdlovu et al. (2018), whereby sodium borohydride was used as a
reducing agent following the reaction below:
FeSO4 þR(strong reducing agent)/Fe(0)(nano-metal)þRSO4 (1)
The two-step reaction of the Iron(II) sulfate aqueous solution
with the reducing is shown below:

FeðH2 OÞ2þ
7 þ 2e /Feð0ÞY þ 7H2 O (2)
2BH 
4 þ 6H2 O/2BðOHÞ3 þ 2e þ 7H2 (3)
The two reactions above combined to form the full reaction as
follows:

FeðH2 OÞ2þ
7 þ 2BH4 /Feð0ÞY2BðOHÞ3 þ H2 O þ 7H2 (4)
To avoid the danger of H2 emission during the synthesis of the
nanoparticles, the synthesis of ZVIN was done under a high per-
formance fume cupboard of gas extraction at 60 feet per minute (ft/
s).
Briefly, we dissolved FeSO4$7H2O (15 g) in 100 mL of 70%
deionized (DI) water/30% ethanol. 3.8 N Sodium hydroxide (NaO-
H(aq)) was used to adjust the pH value to 6.8. Sodium borohydride
(NaBH4) powder (2.7 g) was added into the mixture and continuous
stirring was done for 1 h, and then filtered through a 0.22 mm filter.
The obtained mixture was kept for 24 h under 200  C in an airtight
stainless-steel autoclave lined with Teflon. The resulting nano-
particles were collected using magnetic filtration, and then ethanol
was used to wash them several times. The resultant nanoparticles
were vacuum-dried overnight. To prepare the PEI coated ZVIN,
500 mL of DI water was used to dissolve a suitable volume of the
PEI. The polymer was mixed with ZVIN particles followed 1 h

sonication and then 12 h oven drying at 80 C.
2.2. Identification of ZVIN and PEIeZVIN properties
By using the Xeray diffraction (XRD) (monochromatic CuKa
radiation (MAC Science, MXP18)), the crystal structure of zero-
valent iron was determined. The specific pore volume and surface
area of the ZVIN particles were measured using nitrogen adsorption
(Micromeritics ASAP 2010) at 77 K. The BarretteJoynereHalenda
(BJH) model was used to determine the BETeLangmuir surface area
and pore volume. Field-emission scanning electron microscopy
(FEeSEM) (Hitachi, S-4700 Type II) and HReTEM (Hitachi, He7500)
with a resolution of 0.1 nm were done to study the morphology and
distribution of the size of nanoparticle. The samples were purged
with nitrogen in a fume hood at room temperature to evade the
oxidation of ZVIN. The X-ray absorption spectroscopies (XANES and
EXAFS spectra) were experimented at the Wiggler Beamline BL
17C1 at the National Synchrotron Radiation Research Center
(NSRRC) of Taiwan. The energy and current range for operating the
electron storage ring were 1.5 GeV and 100e200 mA, respectively.
To provide highly monochromatized photon beams with the en-
ergies of 5e15 keV and a resolving power (E/DE) of up to 7000, a Si
(111) double crystal monochromator (DCM) was used. Data
collections were achieved in transmission mode by a Lytle detector
in the regions of the Fe K-edges at room temperature (Lytle, 1999).
By using the IFEFFIT package, XANES spectra were successfully
analyzed (Ravel and Newville, 2005).
2.3. Study area
The cause of contamination in the in-situ location emanated
from discharges by a Taiwan producer of vinyl chloride monomer
(VCM), a petrochemical company situated in Miaoli County in
northern Taiwan. The pilot research project was conducted in an
area of (120 m  60 m) located at the southern side of the VCM
plant, which was the direction of the outbound groundwater flow.
The major pollutants emitted by the factory were PCE, 1,2eDCE,
TCE, and other hazardous compounds. Owing to the high level of
contaminants in that area, the facility was forced to shut down and
was later relocated to another area. The two-dimensional lines
from north to south (cabling points D1, D2, D18, and D19), occu-
pying all the area contaminated by DNAPLs, were separated into
numerous sampling wells denoted A1, A14, A16, and A24 as pre-
sented in Scheme 1.
2.4. Resistivity image profiling
The Resistivity Image Profiling (RIP) technique was used to
characterize the complex structures beneath the surface for sites
where traditional DC signaling and other techniques are impossible
to use (Griffiths and Barker, 1993; Mdlovu et al., 2018). Several
electrodes were implanted into the ground, adjacent to the targeted
survey line. The control unit of the resistivity meter was used
switch two current-injected electrical conductors and other two
electrodes for measuring the resistivity or conduction. Using the
apparent resistivity, a rough image of structures beneath the sur-
face was then produced before digital data processing. A 2D finite-
element inversion was used to obtain the resistivity depth profile
(Yang et al., 2002; Chowdhury et al., 2015). In this in-situ study, a
pole-pole electrode setting and multifunctional resistivity meter
called McOhm-21 manufactured by OYO Company were utilized. A
remote monitoring system for current and potential conductors
were positioned at a distance 10 times the anticipated depth of
study, distant from the line of survey. To make sure that the in-
jection and flow of electricity through the conductor into the
Wastewater
treatment plant
1 2
N A24
Sampling wells A16
D1 D18
D2 D19
A1
A14
On-site treatment area
Scheme 1. An illustration of the sampling wells layout.
 N.V. Mdlovu et al. / Chemosphere 224 (2019) 816e826 819

stratum via groundwater was successful, the electrodes were made

(a)
to be contact with the groundwater (Mdlovu et al., 2018). 3.0

2.5. Injection of PEIeZVIN nanocomposites and field sampling 2.5

g/L)
1,2-Dichloroethane
In addition, to understand the range of the contaminated zone
2.0 c-1,2--Dichloroethene

Concentration x10-6 (
and the effectiveness of the dechlorination process, 54 wells were t-1,2--Dichloroethene
drilled for pumping and monitoring purposes around the Tetrachloroethene
contaminated area, 42 of these wells being on the factory premises. 1.5
Trichloroethane
The wells were drilled to a depth range of 10e15 m. The PEIeZVIN
1,1,2-Trichloroethene
composite material was injected using the direct injection method. 1.0
The injected and pumpable slurry was disseminated over the Vinyl Chloride
plume transported by flowing groundwater to reach the DNAPL 0.5
plume. Nonstop pumping and treatment of polluted groundwater
marked the initial phase of the long-term groundwater remedia-
tion process. The efficiency of the remediation process was evalu- 0.0
ated by regularly collecting samples from the wells every three
months for testing. After treatment, to collect the solution, the 0 3 6 9 12 15 18 21 27 30 33
liquidesolid separation by pressure filtration was used. The sam- Sampling date (months)
ples collected in liquid form were transferred into VOC bottles. The
samples were then dried by a heat lamp and then kept in an oxygen
(b)
free environment to enable further instrumental characterization. 25 1,2-Dichloroethane
c-1,2--Dichloroethene
3. Results and discussion t-1,2--Dichloroethene
g/L) 20 Tetrachloroethene
3.1. Morphologies of ZVIN/PEIeZVIN and concentration profile of Trichloroethane
Concentration x10-4 (

DNAPLs
The concentration profiles of seven major DNAPLs taken from Al
and A14 sampling wells were investigated on site and the results
are displayed in Fig. 1(aeb). It can be observed in Fig. 1(a) that the
concentrations of 1,2eDCE and TCE were 2,800,000 and
940,000 mg/L, respectively. Moreover, in Fig. 1(b) the concentrations
of 1,2eDCE, TCE, and vinyl chloride were 40,000 mg/L, 200,000, and
50,000, respectively. It can be seen that the concentrations of
DNAPL pollutants were very high, especially those of 1,2eDCE
(2,800,000 mg/L) and TCE (940,000 mg/L) as shown in Fig. 1(a) and
(b). This is mainly due to the long-term dissolution of DNAPL
contaminants. Moreover, during the first feeding, the concentra-
tions of DNAPL contaminants were very high since PEI polymer was
still covering the surface of ZVIN. Interestingly, a significant
reduction in DNAPLs concentration was observed during the sec-
ond feeding. The decrease in the concentration suggests the com-
plete dissolution of PEI and the suspension of ZVIN solution, which
had slowly dissolved, allowing the highly reactive surface of ZVIN
to reduce the DNAPLs (Mdlovu et al., 2018). Furthermore, a signif-
icant increase of TCE contaminant was observed between months
15 and 27 on well A14. This could be attributed to the prolonged
dissolution of DNAPLs over time, which led the concentration to
become high.
In addition, Fig. 2(a) and (b) present the concentration profiles of
DNAPLs at sampling wells A1, A14, A16, and A24. It can be seen that
at the beginning of the sampling process, the concentration of
DNAPLs in wells A16 and A24 were very high. Fortunately, the
concentrations were gradually reduced after the injection of ZVIN.
Nevertheless, due to the continuous flow of groundwater, the high
concentration of these contaminants, and the limited amount of
ZVIN, the contaminant concentration levels increased and could
not be completely removed within a short period. Moreover, during
the injection and pumping of the ZVIN surfactant, ZVIN could be
driven to other positions/locations by the pumping, causing the
ZVIN to leave the vicinity of the wells or contaminated zones over
time. These phenomena might have led to the increase in
contaminant concentrations. The comparative results of the
decontamination efficiency before and after the injection of ZVIN
1,1,2-Trichloroethene
15
Vinyl Chloride
10
5
0
0 3 6 9 12 15 18 21 24 27 30 33 36
Sampling date (months)
Fig. 1. Concentration profiles of (a-b) seven different DNAPL contaminants at sampling
wells of (a) A1, and (b) A14. (c) trichloroethylene and (d) 1,2edichloroethene at
different sampling wells of A1, A14, A16, and A24 over the years, respectively.
and PEI-coated ZVIN are shown in Table 1. The reductive studies for
DNAPLs using both ZVIN and PEI stabilized ZVIN are stated and all
samples were determined in triplicates and the average was used to
obtain the data. The concentration profiles of seven major DNAPL
samples were taken from sampling wells A1 and A14 after a one-
day reaction with ZVIN and PEI-coated ZVIN, respectively. The
concentration of DNAPLs from sampling well A1 before injection
were 1,2eDCE (506 mg/L), cis-1,2eDCE (27 mg/L), trans-1,2eDCE
(9.52 mg/L), PCE (not detected), TCE (7.11 mg/L), vinyl chloride
(133 mg/L), and 1,2-trichloroethane (60.0 mg/L). However, after the
injection of ZVIN the concentration was reduced to 1,2eDCE
(70.6 mg/L), cis-1,2eDCE (1.35 mg/L), trans-1,2eDCE (not detected),
PCE (not detected), TCE (not detected), vinyl chloride (1.55 mg/L),
and 1,2-trichloroethane (6.50 mg/L). On the other hand after the
injection of PEI-coated ZVIN, there was a significant reduction to
the concentrations of DNAPLs to 1,2eDCE (22.8 mg/L), cis-1,2eDCE
(not detected), trans-1,2eDCE (not detected), PCE (not detected),
TCE (not detected), vinyl chloride (not detected), and 1,2-
trichloroethane (2.11 mg/L). Similarly, for the DNAPL concentration
profile for contaminants from sampling well A14, the initial con-
centrations were 1,2eDCE (2.80 mg/L), cis-1,2eDCE (3.80 mg/L),
820 N.V. Mdlovu et al. / Chemosphere 224 (2019) 816e826

2.5 Table 2
(a) Fe XANES K-edge absorption of ZVIN and PEIeZVIN after one-day reac-
tion with DNAPL contaminants from sampling wells A1 and A14
compared to that of standard samples.
2.0
A1 Iron compound XANES absorption (eV)
Concentration x10-4 ( g/L)

A14 Fe(0) foil 7111.99

FeO standard 7113.74
1.5 A16 Fe3O4 standard 7115.10
A24 Fe2O3 standard 7117.88
ZVIN þ A1 7114.81
PEIeZVIN þ A1
1.0 ZVIN þ A14
7114.73
7114.55
PEIeZVIN þ A14 7115.13

0.5

0.0
0 3 6 9 12 15 18 21 27 30 33 36 39 42 45 48 51
Sampling date (months)

3.0
(b)

2.5 Concentration
g/L)

2.0 A1
A14
Concentration x10-4 (

A16
1.5
A24
1.0 One-day reaction

After
Before

0.5
0.0
0 3 6 9 12 15 18 21 27 30 33 36 39 42 45 48 51
Sampling date (months)
Fig. 2. Concentration profiles of (a) trichloroethylene and (b) 1,2edichloroethene at
different sampling wells of A1, A14, A16, and A24 over the years, respectively.
trans-1,2eDCE (0.65 mg/L), PCE (0.62 mg/L), TCE (2.05 mg/L), vinyl
chloride (64.8 mg/L), and 1,2-trichloroethane (14.7 mg/L). However,
after the injection of ZVIN the concentration was reduced to
1,2eDCE (0.73 mg/L), cis-1,2eDCE (not detected), trans-1,2eDCE
(not detected), PCE (not detected), TCE (not detected), vinyl chlo-
ride (0.75 mg/L), and 1,2-trichloroethane (not detected). Conversely,
after the injection of PEIeZVIN, the concentration was reduced to
1,2eDCE (1.00 mg/L), cis-1,2eDCE (not detected), trans-1,2eDCE
Fig. 3. The result of a one day reaction of unmodified ZVIN and PEIeZVIN with
1,2edichloroethene obtained from sampling wells A1 and A14.
(0.79 mg/L), PCE (not detected), TCE (0.71 mg/L), vinyl chloride
(7.61 mg/L), and 1,2-trichloroethane (not detected). Moreover, as
shown in Fig. 3 after the injection of ZVIN the concentration of
1,2eDCE was significantly reduced. The data from sampling well A1
shows that 1,2eDCE was reduced from 506 mg/L to 70.6 mg/L.
Likewise, from sampling well A14, the concentration of 1,2eDCE
was reduced from 2.80 mg/L to 0.73 mg/L. Moreover, for the PEI-
coated ZVIN, the 1,2eDCE concentration profile for contaminants
from sampling wells A1 and A14 was reduced from 506 mg/L to
22.8 mg/L and 2.80 mg/L to 1.00 mg/L, respectively.
It is apparent that the concentrations of the other seven major
DNAPLs decreased significantly after one-day of treatment. Most
importantly, PEI-coated ZVIN displayed the highest degradation
efficiency when compared to ZVI with some contaminants not

Table 1
Contaminants concentration from each well before and after one-day treatment with ZVIN and PEIeZVIN suspension solution (unit: mg/L).

Before ZVIN/PEIeZVIN injection After ZVIN/PEIeZVIN injection

DNAPL contaminants ZVIN þ A1 PEIeZVIN þ A1 ZVIN þ A14 PEIeZVIN þ A14 ZVIN þ A1 PEIeZVIN þ A1 ZVIN þ A14 PEIeZVIN þ A14

1,2-dichloroethene 506 506 2.80 2.80 70.6 22.8 0.73 1.00

cis-1,2-dichloroethene 27 27 3.80 3.80 1.35 N.D. N.D. N.D.
trans-1,2-dichloroethene 9.52 9.52 0.65 0.65 N.D. N.D. N.D. 0.79
Perchloroethylene N.D. N.D. 0.62 0.62 N.D. N.D. N.D. N.D.
Trichloroethylene 7.11 7.11 2.05 2.05 N.D. N.D. N.D. 0.71
Vinyl chloride 133 133 64.8 64.8 1.55 N.D. 0.75 7.61
1,1,2-trichloroethane 60.0 60.0 14.7 14.7 6.50 2.11 N.D. N.D.

Note: “N.D.” denotes “not detected”.

 N.V. Mdlovu et al. / Chemosphere 224 (2019) 816e826 821

detectable after the injection of PEI-coated ZVIN suspension into
the DNAPLs polluted plume. This phenomenon could be ascribed to
the well dispersed and large surface area PEI-coated ZVIN which
escalates its reactivity and enable more rapid degradation of a vast
amount of contaminants. Nevertheless, because PEI-coated ZVIN is
a newly innovated remediation technique, cost and efficiency re-
sults collected from site studies of in-situ reduction are still limited
(Huling and Weaver, 1996; Lin et al., 2018; Mdlovu et al., 2018).
Samples of contaminated water from the sampling of wells A1
and A14 after a one-day reaction with ZVIN and PEIeZVIN were
retained to conduct further laboratory tests. FEeSEM images
(Fig. 4(a), (c), (e), and (g)) and HReTEM images (Fig. 4(b), (d), (f) and
(h)) of ZVIN and PEIeZVIN after the reaction with DNAPLs polluted
water samples are displayed in Fig. 4(a)e(h). The uncoated ZVIN
displays ball-shaped particles with sizes of 60e80 nm. Due to the
robust magnetic properties of ZVIN, agglomeration between the
nanoparticles occurs. This occurrence can be credited to the fact
that the ZVIN without any surface coating by polymeric material
contains a smaller surface area, which easily causes particle ag-
gregation (Mukherjee et al., 2016). The Fe(0) particles modified by
PEI polymer are predominantly monodisperse with sizes of
30e70 nm. In addition, the PEI coating results in a less aggregated
individual ZVIN particles as demonstrated in Fig. 4(c), (d), (g), and
ZVIN+A1 (a) ZVIN+A1
(h). Because positively charged PEI exhibits electrostatic repulsion,
the distinct PEI-coated ZVIN particles remain dispersed as separate
ZVIN particles and magnetic interaction between the particles is
avoided. The evenly dispersed particles also provide a wide surface
area and extremely reactive sites to facilitate the degradation of
high amounts of DNAPLs. Moreover, by modifying the surface of
ZVIN particles using PEI polymer, we can avoid the magnetic in-
teractions between the nanoparticles.
To distinguish the nature of the obtained products from the
standard characteristic peaks of Fe3O4, the XRD patterns of ZVIN
and the PEI-coated ZVIN reacted with contaminated water samples
from wells A1 and A14 are shown in Fig. 5(a) and (b), respectively.
From the results, the phase purity and formation of elemental iron
are confirmed. With or without surface modification by polymer,
after the one-day reaction was completed, the signal for iron was
still at about 2q ¼ 45 . The peaks associated with the crystal-like
structure of Fe(0) (JCPDS No. 87e0721) were identified in both
311 (a)
440
333
220 400
Fe3O4
Intensity (a.u.)
(b)

ZVIN

PEI–ZVIN
PEI-ZVIN+A1 (c) PEI-ZVIN+A1 (d)

20 30 40 50 60 70 80
2 (degree)

311 (b)
ZVIN+A1 (e) ZVIN+A1 (f) 440
333
220 400
Intensity (a.u.)

Fe3O4

PEI-ZVIN+A14 (g) PEI-ZVIN+A14 (h) ZVIN

PEI–ZVIN

20 30 40 50 60 70 80
2 (degree)
Fig. 4. (a), (c), (e), (g) FEeSEM microphotos and (b), (d), (f), (h) HReTEM images of Fig. 5. XRD patterns of Fe3O4, unmodified ZVIN and PEIeZVIN, for one-day reaction
unmodified ZVIN and PEIeZVIN, after one-day reaction with contaminants from with contaminated aqueous solutions from sampling wells of (a) A1, and (b) A14,
sampling wells of A1 and A14, respectively. respectively.
822 N.V. Mdlovu et al. / Chemosphere 224 (2019) 816e826

specimen. As shown in Fig. 5, the XRD patterns displayed several Table 3

characteristic peaks at 2q ¼ 30 , 36 , 43 , 53 , 57, and 63 after The specific surface area of unmodified ZVIN and PEIeZVIN before and after
one-day reaction with DNAPL contaminants from sampling wells of A1 and
reaction, which indicates the presence of magnetite on the particles A14.
(Yazdanbakhsh et al., 2016). Thus, the coating on ZVIN has no
substantial effect on nanoparticle crystallization. Fe(0) type Specific surface areaa (m2/g)

Fig. 6(a)e(d) shows N2 adsorption isotherms of samples ob- Unmodified ZVIN 34.7
tained from contaminated water sampled from wells A1 and A14 PEIeZVIN 53.4
ZVIN þ A1 29.4
after a one-day reaction with ZVIN and PEIeZVIN. The surface areas
PEIeZVIN þ A1 45.6
of ZVIN and PEIeZVIN before reaction with contaminants were 34.7 ZVIN þ A14 29.9
and 53 m2/g, respectively. After the reaction, the surface areas were PEIeZVIN þ A14 47.4
29.4 (ZVIN þ A1), 45.6 (PEIeZVIN þ A1), 29.9 (ZVIN þ A14), and Note.
47.4 m2/g (PEIeZVIN þ A14), as listed in Table 3. For the analyses of a
Specific surface area was measured by BJH method.
the pore size distribution of the samples, the Bar-
retteJoynereHalenda method was used. The results indicate that
the ZVIN and PEIeZVIN samples were mesoporous. A hysteresis sharp feature at 7134 eV which arises from the dipole-allowed
performance of Type IV isotherm can be inferred from the 1se4pxy electron transition, indicating the existence of Fe(III) (Lin
adsorption-desorption isotherms, suggesting the mesoporous na- et al., 2013). When compared to iron standards (Fe, FeO, Fe2O3,
ture of the specimens. Moreover, a Type IV isotherm is character- and Fe3O4), Fig. 7(b) indicates that the dried sample of the reacted
ized by its hysteresis loop, which is linked with capillary ZVIN with contaminated water from sampling well A14 displays an
condensation that happens in the mesoporous materials. The transition intensity comparable to that of Fe3O4. Overall, chlori-
mesoporous volume is the total adsorbed at the plateau and it can nated compounds were directly reduced onto ZVIN, which aided
be attained by presuming that the condensate is of liquid nitrogen the reaction through disintegrating away FeO or Fe2O3. The results
(Adhikari and Lin, 2016). also direct the corrosion products of iron might be a combination of
iron(III) oxide and iron(II,III) oxide after the degradation of the
contaminants (Fig. 7(a)) (Lin et al., 2008). Similarly, Table 2 displays
3.2. Mechanism for reduction reaction and ZVIN/PEIeZVIN
the XANES results of reacted samples; the PEIeZVIN contains more
corrosion products
Fe3O4 than ZVIN does. Consequently, oxidation could result in a
layered structure, shifting from a crystalline inner layer rich in
The XANES technique was used to observe the nature of the
iron(II) oxide to a more amorphous outer layer rich in ferric oxide.
zero-valent iron products in samples obtained after the reaction of
EXAFS indicates the physical and chemical nature of the ZVIN uti-
ZVIN and PEIeZVIN with contaminated water samples from wells
lized for the chemical reduction of chlorinated compounds in
A1 and A14 (Fig. 7(a)e(b)). The results indicate the existence of an
Fig. 8(a)e(b). In terms of the EXAFS results presented in Table 4, the
absorbance characteristic (Fe ¼ 7115 eV) for the 1se3d shift and a

100 (a) 0.05 (b) 495.2 Å

Pore volume (cm3/g)

Pore volume (cm3/g)

80 0.04
adsorption
60 desorption 0.03

40 0.02

20 0.01 36.3 Å
0 0.00
0.0 0.2 0.4 0.6 0.8 1.0 10 100 1000
Relative pressure (P/P0) Pore diameter (Å)

200 0.10
180 (c) (d) 302.7 Å
160 0.08
Pore volume (cm3/g)

Pore volume (cm3/g)

140 adsorption
120 desorption 0.06
100
80 0.04
60
40 0.02 90.1 Å
20
0.00
0
0.0 0.2 0.4 0.6 0.8 1.0 10 100 1000
Relative pressure (P/P0) Pore diameter (Å)

Fig. 6. N2 adsorption/desorption isotherms of (a) unmodified ZVIN and (c) PEIeZVIN reaction with contaminants from sampling well A1, respectively. Pore size distribution curves
of (b) unmodified ZVIN and (d) PEIeZVIN, reaction with contaminant from sampling well A1, respectively.
 N.V. Mdlovu et al. / Chemosphere 224 (2019) 816e826 823

6
Fe–O (a) 4
(b)

FT magnitude (a.u.)
5 3
(a) 2
4
1
3 ZVIN+A1
0
Normalized absorbance (a.u.)

k3
2 -1

1
-2
ZVIN+A1
-3
0 -4
0 2 4 6 8 10 0 2 4 6 8 10 12 14 16
Bond distance (Å) k (Å-1)
6 3
Fe–O (c) (d)

FT magnitude (a.u.)
5 2

4 1
PEI–ZVIN+A1
3 0

k3
2
-1
1
-2
PEI–ZVIN+A1
0
-3
0 2 4 6 8 10 0 2 4 6 8 10 12 14 16

Bond distance (Å) k (Å-1)

7
3
Fe–O (e) (f)

FT magnitude (a.u.)
6
Photon energy (eV) 2
5
1
4
ZVIN+A14 0
3
(b)

k3
-1
2
1 -2 ZVIN+A14
Normalized absorbance (a.u.)

0 -3

0 2 4 6 8 10 0 2 4 6 8 10 12 14 16

Bond distance (Å) k (Å-1)

8
4
Fe–O (g) (h)
FT magnitude (a.u.)

7 3
6 2
5 1
4 PEI–ZVIN+A14 0
3 -1

k3
2 -2
-3
1
-4 PEI–ZVIN+A14
0
-5
0 2 4 6 8 10 0 2 4 6 8 10 12 14 16
Bond distance (Å) k (Å-1)

Fig. 8. Fe K-edge EXAFS (a, c, e, and g) Fourier transformed spectra, and (b, d, f, and h)
oscillation k3c(k) obtained after one-day reaction of unmodified ZVIN and PEIeZVIN
with contaminants from sampling wells of A1 and A14, respectively.

Photon energy (eV)

Table 4
Fig. 7. Fe K-edge XANES spectra for one-day reaction of unmodified ZVIN and Fine-structure parameters of ZVIN and PEIeZVIN after one-day reaction with DNAPL
PEIeZVIN with DNAPL contaminants from sampling wells of (a) A1 and (b) A14, contaminants from sampling wells of A1 and A14.
respectively.
Sample name First shell CN (±0.05) a
R (±0.01 Å)b Ds2 (Å2)c
ZVIN þ A1 FeeO 3.78 1.98 0.011,957
used ZVIN contains center Fe atoms linked by FeeO bonding. The PEIeZVIN þ A1 FeeO 4.14 2.00 0.014,487
FeeO standard bonding distances observed for ZVIN and PEIeZVIN ZVIN þ A14 FeeO 4.09 1.96 0.012,914
PEIeZVIN þ A14 FeeO 4.31 1.94 0.010,282
were 1.98, 2.00, 1.96, and 1.94 Å with coordination numbers (CN) of
3.78, 4.14, 4.09, and 4.31, respectively. The DebyeeWaller factor (s2) Notes.
a
is smaller than 0.014 (Å2) in all samples. Likewise, some correlation b
Coordination number.
Bond distance.
with maghemite species was witnessed for k3-weighted and least c
Debye-Waller factor.
squares-fitted maghemite K-edge fine structure inverse Fourier-
transform oscillation spectra of the samples (Fig. 8(b), (d), (f), and
(h)). As revealed in Fig. 8, the bonding distances and the coordi- groundwater treatment is its susceptibility to reacting with other
nation numbers for the different samples have some significant non-targeted elements (Chen and Wu, 2017). Therefore, to measure
differences. This phenomenon can be attributed to the different the elemental composition of ZVIN and PEIeZVIN after a one-day
degree of oxidation of each sample after reaction with the reaction with sampling wells A1 and A14 contaminants, compara-
contaminated water. Normally, iron atoms contain larger CN in tive XPS studies were conducted as depicted in Figs. 9 and 10. The
oxides when compared to metals, which is attributed to the atoms results direct that surface stabilization happen via elevated
of oxygen occupation close the main atom of iron present in the reduction reactions occurring on ZVIN surface, whereby Fe3O4 at a
crystal structure. In particular, PEIeZVIN had a larger CN than ZVIN more oxidation state replaces the FeO types. In addition, the sus-
samples, signifying that PEIeZVIN contained more surrounding pension of DNAPL compounds onto PEI-coated ZVIN probably fa-
atoms of oxygen around the central iron atoms (Hsueh et al., 2017). cilitates the interaction of ZVIN with the pollutants. The feature of
The principal disadvantage of the utilization of ZVIN in the surface of the iron oxide is very important for the reduction
824 N.V. Mdlovu et al. / Chemosphere 224 (2019) 816e826

on the surface of the metal (Wu et al., 2005).

Fe 2p3/2 A degradation mechanism for ZVIN and PEI-coated ZVIN with
pollutants has been reported in many studies (US EPA, 1996; Liang
et al., 2017; Lin et al., 2018). Several studies have indicated the
O 1s
existence of a chain reduction reaction whereby TCE is transformed
into 1,1eDCE, and then vinyl chloride (VC) and ethylene (Yan et al.,
Intensity (a.u.)

Fe 3p 2017; Lin et al., 2018). An illustration of the reaction mechanism

1400 1200 1000 800
Binding energy (eV)
Fig. 9. Full scan of XPS spectra for one-day reaction of (a) and (c) unmodified ZVIN and
(b) and (d) PEIeZVIN with contaminants from sampling wells of A1 and A14,
respectively.
mechanism of the contaminants. The iron oxide shell plays a pivotal
role in facilitating the reduction reaction via the transfer of elec-
trons on the surface of iron (Goswami et al., 2018; Su et al., 2018).
For example, chloride ions are formed from the dissociation of TCE
particles. These chloride ions become electron donors, discharging
electrons to metallic elements thus new reactive sites are produced
and engineering representation of in-situ DNAPL degradation using
C 1s the PEI-coated ZVIN suspension is depicted in Fig. 11. A majority of
(a) reported in-situ applications of ZVIN in groundwater treatment
have used direct injection methods. Several direct injection
(b) methods are available; nevertheless, their main objective is the
(c) delivery of certain amounts of ZVIN suspension at designated
depths straight into the contaminated site (Kim et al., 2017). The
(d) pumpable ZVIN suspension, which is administered through an in-
jection process, should allow us to predict its behavior in ground-
water and its ability to disperse through the water and maintain
600 400 200 0 stability and mobility until it reaches the DNAPL pools (Bardos et al.,
2011). Many studies have mentioned about the cytotoxicity of PEI
(Du et al., 2016; Mdlovu et al., 2019), however, its cytotoxicity
cannot be compared to that of DNAPLs when they enter the system
of a human being and such toxicity is incomparable to the advan-
tages brought by PEI throughout the remediation process. Most
DNAPLs are very toxic even at very trace concentrations in
groundwater and can lead to severe risk to the health of human
being (Lerner et al., 2003). However, PEI can be easily be decom-
posed by ecosystem, but these chlorinated compounds cannot be
easily degraded except through ZVIN. Therefore, in this case, PEI
might not course a significant second pollution that can have severe
effects on human being as DNAPLs; hence, the cytotoxicity and
further pollution of PEI can be easily ignored. Moreover, the coated

Fe3O4 (a) FeSO4 (b)

FeSO4 Fe3O4
Intensity (a.u.)
Intensity (a.u.)

FeO FeO

716 714 712 710 708 706 704 716 714 712 710 708 706 704
Binding energy (eV) Binding energy (eV)

(c) (d)
FeSO4 FeSO4 Fe3O4
Fe3O4
Intensity (a.u.)

Intensity (a.u.)

FeO FeO

716 714 712 710 708 706 704 716 714 712 710 708 706 704
Binding energy (eV) Binding energy (eV)
Fig. 10. Fine section of Fe 2p3/2 fitting results for (a) and (c) unmodified ZVIN; (b) and (d) PEIeZVIN with contaminants from sampling wells of A1 and A14, respectively.
 N.V. Mdlovu et al. / Chemosphere 224 (2019) 816e826 825

(a) 48.0
16.0 32.0
-
0.0

-5.0 -

-10.0 -

Elevation (M)
-15.0 -

-20.0 -
N S
-25.0 -

-30.0 -
Unit Electrode Spacing = 1.00 m
(b) 48.0
16.0 32.0
0.0 -
-5.0 -

Elevation (M)
-10.0 -

-15.0 -

-20.0 -
Fig. 11. Systematic illustration of in-situ remediation and reaction mechanism of TCE,
PCE, and 1,2eDCE DNAPLs through the injection of PEIeZVIN with a core-shell
N S
-25.0 -
microstructure.

-30.0 -
PEI is not at high concentrations, thus after dissolution, it is diluted
in the water without causing any serious water contamination. Unit Electrode Spacing = 1.00 m
However, further studies are necessary to evaluate the exact effect (c) 0.0 48.0
16.0 32.0
of the dissolved PEI shell in as far as the cytotoxicity and the pos- -
0.0
sibility to cause a second pollution in groundwater are concerned.
-5.0 -
Elevation (M)

3.3. Resistivity image profiling -10.0 -

Resistivity image profiling is an excellent method to locate and -15.0 -

investigate the location of plumes of DNAPLs. The RIP method
displayed that the conductivity information for the PEIeZVIN sus-
pension and groundwater samples similarly establish a lower
conductivity of FeO types. The purple rectangle in Fig. 12 indicates
the pumping well while the black one indicates the feeding well. As
exhibited in Fig. 12(aec), the change of color from bright red
(Fig. 12(a)) to light orange (Fig. 12(c)), showing that the DNAPL
contaminants were successfully degraded. The use of the direct
well injection technique enhanced the reduction of contaminants.
From the results, it can be seen that high resistivity zones are
indicative of high concentrations of DNAPLs. Finally, the RIP studies
helped to offer an understanding of the sensitivity of RIP technique
for detecting DNAPL pools within the shallow subsurface. It further
aided in providing useful information regarding the monitoring and
potential migration path of DNAPL plume.
4. Conclusions
In this study, the efficacy of the PEI-stabilized ZVIN in the
reduction of chlorinated DNAPLs in groundwater was evaluated in-
situ. Through direct injection, PEIeZVIN and ZVIN suspensions
successfully degraded 1,2eDCE, PCE, and TCE, whereby TCE was
transformed into 1,1eDCE, followed by vinyl chloride (VC) and
ethylene. PEI improved the movement and firmness of ZVIN during
its flow with groundwater to reach the contaminated plume,
thereby improving its reactivity in groundwater. In addition, XANES
and EXAFS studies indicated that Fe3O4 at a higher oxidation state
replaced the FeO species. However, because PEI-coated ZVIN is a
newly innovated remediation technique, cost and performance
data derived from field applications of in-situ reduction are still few
-20.0 -
N S
-25.0 -
-30.0 -
Unit Electrode Spacing = 1.00 m
Fig. 12. Resistivity image profiling results from line 1 (a) before, after (b) 5, and (c) 17
days of PEIeZVIN suspension feeding for the contaminated site.
in number. More studies to understand the total amount of PEI-
coated ZVIN required completely remove all the contaminants
from the site would be necessary. Further studies are necessary to
evaluate the exact effect of the dissolved PEI shell in as far as the
cytotoxicity and the possibility to cause a second pollution in
groundwater are concerned. Consolidated experiments of
PEIeZVIN direct injection and resistivity image profiling would be
economically and ecologically attractive for future studies.
Acknowledgements
We would like to extend our thanks to the Ministry of Science
and Technology (MOST), Taiwan (MOST 103-2621-M-155-001) for
the financial support. We are also thankful to Prof. Y. W. Yang, Dr. J.
F. Lee, Dr. Jeng-Lung Chen, and Dr. Chao-Lung Chiang from Taiwan
National Synchrotron Radiation Research Center (NSRRC) for their
assistance in the XANES/EXAFS experiments or data analyses. We
finally thank Prof C.H. Yang from National Central University (NCU)
who helped us with the Resistivity Image Profiling (RIP)
experiments.
826 N.V. Mdlovu et al. / Chemosphere 224 (2019) 816e826

References 323e329.
Lin, K.S., Chang, N.B., Chuang, T.D., 2008. Decontamination of nitrates and nitrites in
wastewater by zero-valent iron nanoparticles. Nano 3 (04), 291e295.
Adhikari, A.K., Lin, K.S., 2016. Improving CO2 adsorption capacities and CO2/N2
Lin, K.S., Dehvari, K., Hsien, M.J., Hsu, P.J., Kuo, H., 2013. Degradation of TNT, RDX,
separation efficiencies of MOF-74 (Ni, Co) by doping palladium-containing
and HMX explosive wastewaters using zero-valent iron nanoparticles. Pro-
activated carbon. Chem. Eng. J. 284, 1348e1360.
pellants, Explos. Pyrotech. 38, 786e790.
Andreev, M., Prabhu, V.M., Douglas, J.F., Tirrell, M., de Pablo, J.J., 2018. Complex
Lin, K.S., Mdlovu, N.V., Chen, C.Y., Chiang, C.L., Dehvari, K., 2018. Degradation of TCE,
coacervation in polyelectrolytes from a coarse-grained model. Macromolecules
PCE, and 1, 2eDCE DNAPLs in contaminated groundwater using
51 (17), 6717e6723.
polyethylenimine-modified zero-valent iron nanoparticles. J. Clean. Prod. 175,
Bahrami, H., Eslami, A., Nabizadeh, R., Mohseni-Bandpi, A., Asadi, A., Sillanp€ a€a, M.,
456e466.
2018. Degradation of trichloroethylene by sonophotolytic-activated persulfate
Litter, M.I., Quici, N., Meichtry, M. (Eds.), 2018. Iron Nanomaterials for Water and
processes: optimization using response surface methodology. J. Clean. Prod.
Soil Treatment. CRC Press, New York: Pan Stanford, p. 338.
198, 1210e1218.
Liu, H.C., Lin, C.P., Yang, C.H., Wang, T.P., 2016. Geoelectrical mapping of the soil and
Bardos, P., Bone, B., Boyle, R., Ellis, D., Evans, F., Harries, N.D., Smith, J.W., 2011.
groundwater contaminated site: case study from Taiwan. In: Proceedings of the
Applying sustainable development principles to contaminated land manage-
7th International Conference on Environment and Engineering Geophysics
ment using the SuRF-UK framework. Remed. J. 21 (2), 77e100.
(ICEEG) & Summit Forum of Chinese Academy of Engineering on Engineering
Busch, J., Meißner, T., Potthoff, A., Bleyl, S., Georgi, A., Mackenzie, K., Trabitzsch, R.,
Science and Technology, vol. 71, pp. 408e411.
Werban, U., Oswald, S.E., 2015. A field investigation on transport of carbon-
Lucattini, L., Poma, G., Covaci, A., de Boer, J., Lamoree, M., Leonards, P., 2018.
supported nanoscale zero-valent iron (nZVI) in groundwater. J. Contam.
A review of semi-volatile organic compounds (SVOCs) in the indoor environ-
Hydrol. 181, 59e68.
ment: occurrence in consumer products, indoor air and dust. Chemosphere 201,
Chaturvedi, S., Dave, P.N., 2019. Water purification using nanotechnology an
466e482.
emerging opportunities. Chem. Methodol. 3, 115e144.
Lytle, F.W., 1999. The EXAFS family tree: a personal history of the development of
Chen, H.M., Wu, M.T., 2017. Residential exposure to chlorinated hydrocarbons from
extended X-ray absorption fine structure. J. Synchrotron Radiat. 6, 123e134.
groundwater contamination and the impairment of renal function-An ecolog-
Mdlovu, N.V., Lin, K.S., Dwitya, S.S., Chen, C.Y., Chiang, C.L., 2018. Decontamination
ical study. Sci. Rep. 7, 9. ID 40283.
of 1,2-dichloroethane DNAPL in contaminated groundwater by polymer-
Chowdhury, A.I., Krol, M.M., Kocur, C.M., Boparai, H.K., Weber, K.P., Sleep, B.E.,
modified zero-valent iron nanoparticles. Top. Catal. 61, 1653e1664.
O'Carroll, D.M., 2015. nZVI injection into variably saturated soils: field and
Mdlovu, N.V., Chen, Y., Lin, K.S., Hsu, M.W., Wang, S.S.S., Wu, C.M., Lin, Y.S.,
modeling study. J. Contam. Hydrol. 183, 16e28.
Ohishi, K., 2019. Multifunctional nanocarrier as a potential micro-RNA delivery
Crane, R.A., Scott, T.B., 2012. Nanoscale zero-valent iron: future prospects for an
vehicle for neuroblastoma treatment. J. Taiwan Inst. Chem. E. 96, 526e537.
emerging water treatment technology. J. Hazard Mater. 211, 112e125.
Mo, L., Ye, S., Wu, J., Stahl, R.G., Grosso, N.R., Wang, J.C., 2017. Field application at a
Du, J., Zhu, W., Yang, L., Wu, C., Lin, B., Wu, J., Jin, R., Shen, T., Ai, H., 2016. Reduction
DNAPL-contaminated site in Nanjing and discussion of a source search algo-
of polyethylenimine-coated iron oxide nanoparticles induced autophagy and
rithm based on stochastic modeling and Kalman filter. Environ. Earth Sci. 76 (2),
cytotoxicity by lactosylation. Regen. Biomater. 3 (4), 223e229.
92e106.
EPA, J., 1996. Pump-and-Treat Ground-Water Remediation: A Guide for Decision
Mukherjee, R., Kumar, R., Sinha, A., Lama, Y., Saha, A.K., 2016. A review on synthesis,
Makers and Practitioners. EPA/625/R-95/005. Office of Research and Develop-
characterization, and applications of nano zero valent iron (nZVI) for environ-
ment, Washington DC. http://www.epa.gov/ORD/WebPubs/pumptreat.
mental remediation. Crit. Rev. Environ. Sci. Technol. 46 (5), 443e466.
Ezzatahmadi, N., Ayoko, G.A., Millar, G.J., Speight, R., Yan, C., Li, J., Li, S., Zhu, J., Xi, Y.,
Phenrat, T., Saleh, N., Sirk, K., Kim, H.J., Tilton, R.D., Lowry, G.V., 2008. Stabilization of
2017. Clay-supported nanoscale zero-valent iron composite materials for the
aqueous nanoscale zerovalent iron dispersions by anionic polyelectrolytes:
remediation of contaminated aqueous solutions: a review. Chem. Eng. J. 312,
adsorbed anionic polyelectrolyte layer properties and their effect on aggrega-
336e350.
tion and sedimentation. J. Nanoparticle Res. 10, 795e814.
Fu, F., Dionysiou, D.D., Liu, H., 2014. The use of zero-valent iron for groundwater
Qu, X., Alvarez, P.J., Li, Q., 2013. Applications of nanotechnology in water and
remediation and wastewater treatment: a review. J. Hazard Mater. 267, 194e205.

s, E., Gallego, J.R., Lobo, M.C., 2017. wastewater treatment. Water Res. 47 (12), 3931e3946.
Gil-Díaz, M., Alonso, J., Rodríguez-Valde
Ravel, B., Newville, Mathena, 2005. ATHENA, ARTEMIS, HEPHAESTUS: data analysis for
Comparing different commercial zero valent iron nanoparticles to immobilize
X-ray absorption spectroscopy using IFEFFIT. J. Synchrotron Radiat. 12, 537e541.
as and Hg in brownfield soil. Sci. Total Environ. 584, 1324e1332.
Ravikumar, K.V.G., Argulwar, S., Sudakaran, S.V., Pulimi, M., Chandrasekaran, N.,
Golshirazi, A., Kharaziha, M., Golozar, M.A., 2017. Polyethylenimine/kappa carra-
Mukherjee, A., 2018. Nano-Bio sequential removal of hexavalent chromium
geenan: micro-arc oxidation coating for passivation of magnesium alloy. Car-
using polymer-nZVI composite film and sulfate reducing bacteria under
bohydr. Polym. 167, 185e195.
anaerobic condition. Environ. Technol. Innov. 9, 122e133.
Goswami, C., Hazarika, K.K., Bharali, P., 2018. Transition metal oxide nanocatalysts
San Rom
an, I., Galdames, A., Alonso, M.L., Bartolome, L., Vilas, J.L., Alonso, R.M.,
for oxygen reduction reaction. Mater. Sci. Energy Technol. 1, 117e128.
2016. Effect of coating on the environmental applications of zero valent iron
Griffiths, D.H., Barker, R.D., 1993. Two-dimensional resistivity imaging and model-
nanoparticles: the lindane case. Sci. Total Environ. 565, 795e803.
ling in areas of complex geology. J. Appl. Geophys. 29, 211e226.
Schrick, B., Hydutsky, B.W., Blough, J.L., Mallouk, T.E., 2004. Delivery vehicles for
Heller, L., 2015. The crisis in water supply: how different it can look through the lens
zerovalent metal nanoparticles in soil and groundwater. Chem. Mater. 16 (11),
of the human right to water? Cad. Saúde Pública 31, 447e449.
2187e2193.
Hsueh, Y.H., Hsueh, Y.H., Tsai, P.H., Lin, K.S., Ke, W.J., Chiang, C.L., 2017. Antimicrobial
Sheu, Y.T., Chen, K.F., Cao, W.Z., Ou, J.H., Kao, C.M., 2018. Remediation of contami-
effects of zero-valent iron nanoparticles on gram-positive Bacillus strains and
nated soil and groundwater using nanoscale zero-valent iron (nZVI) coupled
gram-negative Escherichia coli strains. J. Nanobiotechnol. 15 (1), 77. https://doi.
with anaerobic bioremediation: a review. Recent Pat. Eng. 12 (2), 84e91.
org/10.1186/s12951-017-0314-1.
Soukupova, J., Zboril, R., Medrik, I., Filip, J., Safarova, K., Ledl, R., Mashlan, M.,
Hu, Y.B., Zhang, M., Qiu, R., Li, X.Y., 2018. Encapsulating nanoscale zero-valent iron
Nosek, J., Cernik, M., 2015. Highly concentrated, reactive and stable dispersion
with a soluble Mg (OH) 2 shell for improved mobility and controlled reactivity
of zero-valent iron nanoparticles: direct surface modification and site appli-
release. J. Mater. Chem. A 6 (6), 2517e2526.
cation. Chem. Eng. J. 262, 813e822.
Huling, S.G., Weaver, J.W., 1996. Dense nonaqueous phase liquids. In: Boulding, J.R.
Su, L., Liu, C., Liang, K., Chen, Y., Zhang, L., Li, X., Han, Z., Zhen, G., Chai, X., Sun, X.,
(Ed.), 1996. EPA Environmental Engineering Sourcebook. Ann Arbor Sci. Publ.,
2018. Performance evaluation of zero-valent iron nanoparticles (NZVI) for high-
Chelsea, MI, pp. 73e103.
concentration H 2 S removal from biogas at different temperatures. RSC Adv. 8
Huo, L., Zeng, X., Su, S., Bai, L., Wang, Y., 2017. Enhanced removal of as (V) from
(25), 13798e13805.
aqueous solution using modified hydrous ferric oxide nanoparticles. Sci. Rep. 7,
Tasharrofi, S., Hassani, S.S., Taghdisian, H., Sobat, Z., 2018. Environmentally friendly
12. ID 40765.
stabilized nZVI-composite for removal of heavy metals. In: New Polymer
Kavanaugh, M.C., Suresh, P., Rao, C., 2003. The DNAPL Remediation Challenge: Is
Nanocomposites for Environmental Remediation, pp. 623e642. https://doi.org/
There a Case for Source Depletion? (No. EPA/600/R-03/143). ENVIRONMENTAL
10.1016/B978-0-12-811033-1.00024-X.
PROTECTION AGENCY WASHINGTON DC, USA.
Teng, W., Fan, J., Wang, W., Bai, N., Liu, R., Liu, Y., Deng, Y., Kong, B., Yang, J., Zhao, D.,
Kim, H.J., Leitch, M., Naknakorn, B., Tilton, R.D., Lowry, G.V., 2017. Effect of emplaced
Zhang, W.X., 2017. Nanoscale zero-valent iron in mesoporous carbon (nZVI@ C):
nZVI mass and groundwater velocity on PCE dechlorination and hydrogen
stable nanoparticles for metal extraction and catalysis. J. Mater. Chem. A 5,
evolution in water-saturated sand. J. Hazard Mater. 322, 36e144.
4478e4485.
Lee, N., Choi, K., Uthuppu, B., Jakobsen, M.H., Hwang, Y., Broholm, M.M., Lee, W.,
Wu, L., Shamsuzzoha, M., Ritchie, S.M.C., 2005. Preparation of cellulose acetate
2014. Synthesis of iron nanoparticles with poly (1-vinylpyrrolidone-co-vinyl
supported zero-valent iron nanoparticles for the dechlorination of trichloro-
acetate) and its application to nitrate reduction. Adv. Environ. Res. 3 (2),
ethylene in water. J. Nanoparticle Res. 7, 469e476.
107e116.
Yan, J., Qian, L., Gao, W., Chen, Y., Ouyang, D., Chen, M., 2017. Enhanced fenton-like
Lerner, D.N., Kueper, B.H., Wealthall, G.P., Smith, J.W.N., Leharne, S.A., 2003. An
degradation of trichloroethylene by hydrogen peroxide activated with nano-
Illustrated Handbook of DNAPL Transport and Fate in the Subsurface. Research
scale zero-valent iron loaded on biochar. Sci. Rep. 7, 9. ID 43051.
Report. Environment Agency, Bristol, UK.
Yang, C.H., You, J.I., Tao, M.C., 2002. High resistivities associated with DNAPL plume
Liang, Y., Min, X., Chai, L., Wang, M., Liyang, W., Pan, Q., Okido, M., 2017. Stabilization
images by RIP technique. In: 2002 SEG Annual Meeting. Society of Exploration
of arsenic sludge with mechanochemically modified zero-valent iron. Chemo-
Geophysicists, p. 4. ID SEG-2002-1587.
sphere 168, 1142e1151.
Yazdanbakhsh, A.R., Daraei, H., Rafiee, M., Kamali, H., 2016. Performance of iron
Lin, C.C., Chen, S.C., 2016. Enhanced reactivity of nanoscale zero-valent iron pre-
nano particles and bimetallic Ni/Fe nanoparticles in removal of amoxicillin
pared by a rotating packed bed with blade packings. Adv. Powder Technol. 27,
trihydrate from synthetic wastewater. Water Sci. Technol. 73 (12), 2998e3007.