Journal of Cleaner Production 175 (2018) 456e466

Contents lists available at ScienceDirect

Journal of Cleaner Production

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

Degradation of TCE, PCE, and 1,2eDCE DNAPLs in contaminated

groundwater using polyethylenimine-modified zero-valent iron
nanoparticles
Kuen-Song Lin*, Ndumiso Vukile Mdlovu, Chung-Yu Chen, Chao-Lung Chiang,
Khalilalrahman Dehvari
Department of Chemical Engineering and Materials Science/Environmental Technology Research Centre, Yuan Ze University, ChungeLi District, Taoyuan
City 32003, Taiwan, ROC

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

Article history: Remediation of dense non-aqueous phase liquid (DNAPL), which consists primarily of chlorinated sol-
vents, is considered a top priority in the field of groundwater decontamination. Downward migration of
DNAPL can lead to formation of impermeable strata due to low solubility and high density. Remediation
Keywords: is therefore one of the most complex technical challenges faced by environmental engineers. In the
Zero-valent iron nanoparticles present work, remediation of trichloroethylene (TCE), perchloroethene (PCE), and 1,2-dichloroethene
DNAPLs
(1,2-DCE) DNAPL-contaminated groundwater was studied by a reductive reaction with poly-
Polyethylenimine
ethylenimine (PEI) surface-modified zero-valent iron nanoparticles (PEIenZVI). Compared with fresh
Decontamination
Groundwater remediation
nZVI, PEIenZVI exhibited smaller spherical particles of 20e80 nm and a greater surface area of 53.4 m2/g.
Furthermore, slow desorption of the PEI indicated its potential application as a protective shell layer for
efficient delivery of active nZVI to the water/DNAPL interface. Laboratory batch remediation results
indicate that both nZVI and PEIenZVI can remove 99% of TCE, PCE, and 1,2-DCE. The rate of reaction for
fresh nZVI was higher in the early stage. Comparatively, PEIenZVI had a higher removal rate and effi-
ciency after 2 h. The kinetic studies also revealed that the removal rate for 1,2-DCE was greater than that
for TCE and PCE. Additionally, X-ray absorption near edge structure (XANES) and extended X-ray ab-
sorption fine structure (EXAFS) spectroscopy studies indicated that the nZVI and PEIenZVI have two
central Fe atoms coordinated by primarily FeeO and FeeFe with bond distances of 1.87 Å and 3.05 Å,
respectively. Furthermore, after the reductive reaction, nZVI and PEIenZVI were oxidized to Fe3O4, and
bond distance values for the reacted samples were 1.94 Å and 1.96 Å, respectively.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction
Chlorinated volatile organic compounds (CVOCs), such as TCE,
PCE, and 1,2eDCE, that have low aqueous solubility and a greater
density than water are known as DNAPLs (Baker et al., 2016; Chen
and Wu, 2017; Pe
rez-de-Mora et al., 2014) and are widely used in
the cleaning, extraction, foam, spray, and solvent manufacturing
industries. The DNAPL contaminants are a type of recalcitrant
compounds prevalent at contaminated sites (Dong et al., 2016; He
et al., 2015). These DNAPL compounds have a stable chemical
structure and are degraded very slowly in environmental systems
* Corresponding author.
E-mail address: kslin@saturn.yzu.edu.tw (K.-S. Lin).
(Fu et al., 2014; Hiortdahl and Borden, 2013; Tang et al., 2015). Due
to low solubility and high density, they can be quickly deposited as
an impermeable layer at the bottom of an aquifer. Contaminated
zones can serve as a persistent source of groundwater pollution,
posing grave risks to humans and ecosystems (Heron et al., 2016;
Orozco et al., 2015; Power et al., 2014). Over time, DNAPL sources
dissolve in groundwater and form plumes, leading to the closure of
downgradient water supply wells and creating vapor intrusion is-
sues in buildings located above the plume (Baker et al., 2016).
Conventional remediation strategies such as pump and treat
involve excavation and plume management via either advanced
oxidation processes to degrade the chlorinated compounds or air
stripping to transfer contaminants to other media, both of which
pose high operational costs (Parker et al., 2012; Trellu et al., 2016;
Wang et al., 2012). Therefore, in-situ remediation techniques for

https://doi.org/10.1016/j.jclepro.2017.12.074
0959-6526/© 2017 Elsevier Ltd. All rights reserved.
 K.-S. Lin et al. / Journal of Cleaner Production 175 (2018) 456e466
the treatment of DNAPL bring substantial economic benefits by
allowing soil or groundwater to be treated without being excavated
and transported (Baker et al., 2016; Hiortdahl and Borden, 2013;
Obiri-Nyarko et al., 2014; Sheng et al., 2016; West and Kueper,
2012).
Over the last two decades, many laboratory and field studies
have demonstrated that through the emplacement of permeable
reactive barriers (PRB), zero-valent iron nanoparticles (nZVI) can
effectively transform chlorinated contaminants at much faster rates
than other in-situ technologies (Baker et al., 2016 . Fu et al., 2014).
Nevertheless, PRB treatment still relies on dissolution of DNAPLs
and transport of the dissolved chlorinated solvents to the PRB,
which poses long-term monitoring and remediation costs (Ma
et al., 2015; Quinn et al., 2005; Su et al., 2013; Wang et al., 2015).
In order to overcome the limitations of PRB, the viable approach is
improving the dissolution of pooled DNAPLs using surfactant
solutions (Ma et al., 2017; Wang et al., 2012). Alternatively, reactive
agents such as nZVI can be transported to the DNAPL sources
(Goswami et al., 2017; Huang et al., 2016; Ma et al., 2017).
Zero-valent iron nanoparticles have increasingly gained atten-
tion from researchers and practitioners recently due to its highly
reactivity, cost effectiveness, and potential to treat a various range
of contaminants (Mukherjee et al., 2016). Zero-valent iron nano-
particles exhibit greater degradation rates, because it has a greater
surface area and reactivity from faster corrosion (Goswami et al.,
2017; He et al., 2015; Oprckal et al., 2017).
Numerous studies confirmed nZVI powder can degrade a wide
range of contaminants, even more than that of granular form. It is
attributed to their extremely high specific surface areas, resulting in
high reaction rates by orders of magnitude than those of millimetric
iron (Ghosh et al., 2017; Lien and Zhang, 2001). Moreover, nano-
scale zero-valent iron is well known as an electron-source material
for the in-situ remediation of contaminants in groundwater. How-
ever, nZVI is high susceptible to forming large aggregates that is
ascribed to the magnetic properties and speedy oxidation of
metallic iron. Thus, these new-formed aggregates with less surface
active sites and reaction rate, restricting the stability and mobility
of nZVI in subsurface environment (Huo et al., 2017; Kim et al.,
2014; Phenrat et al., 2007; Sun et al., 2007; Teng et al., 2017;
Tosco et al., 2014). Owing to these limitations, there is a need to
evade aggregation and oxidation by developing cost-effective ap-
plications of nZVI. For an effective utilization of in-situ DNAPLs
remediation, nZVI should remain long-time suspended in slurry for
the efficient injection into the contaminated site (Comba et al.,
2011; He and Zhao, 2005). Furthermore, efficient delivery of nZVI
to the water/DNAPL interface without flocculation and oxidation is
the key for achieving maximum efficiency (Dong and Lo, 2013; Lin
et al., 2013; Lubphoo et al., 2016; Tang et al., 2015). Surface modi-
fication of nZVI with a protective polymer layer keeps nZVI sepa-
rated from dissolved oxygen and hydrogen while creating an
affinity for the water/DNAPL interface, therefore enhancing the
destruction of chlorinated DNAPL in source zones (Golshirazi et al.,
2017; Noubactep et al., 2012). On the other hand, the disadvantages
of nZVI such as facile aggregation/oxidation, poor mobility, and low
stability can be solved by coating PEI on the surfaces of nZVI
particles, forming PEIenZVI composite material (Bennett et al.,
2010; Ghosh et al., 2017; Goswami et al., 2017; He et al., 2010;
Henn and Waddill, 2006; Opr ckal et al., 2017). Comparison of
nZVI, composite PEIenZVI can be fast injected into the source of
contamination through subsoil. Apart from PEI, there are another
polymeric materials can well-suspended the nZVI in water (Comba
et al., 2011; He and Zhao, 2005). However, PEI exhibits remarkable
content of functional groups, good water solubility, and chemical
stability for a great affinity with metals. Polyelectrolyte character
and larger macromolecular size of PEI make them very useful in
457
sequestering metal ions from dilute solutions (Chaufer and
Deratani, 1988; Kobayashi et al., 1987). In addition, PEI offers
good hydrophilicity, high rate, and capacity of adsorption for
contaminants (Chen et al., 2010). Kim et al. (2017) have confirmed
that the triphenylphosphine (TPP)-coated nZVI also enhanced the
stability of nZVI, and inhibited the agglomeration/sedimentation of
nZVI. Furthermore, the reactivity of nZVI is controlled by the nature
of the ligand and coating procedure. For instance, high concentra-
tions of sodium dodecyl sulfate and alcohols (Huo et al., 2017;
Loraine, 2001; Wang et al., 2012) decrease nZVI reactivity, while
Triton X-a00 (Loraine, 2001) and biodegradable vegetable oil
(Quinn et al., 2005) can enhance TCE reduction. Furthermore,
surface modification reduces nZVI activity due to uncontrolled
sequestering of DNAPLs in mobile micelles (Dong and Lo, 2013; Fu
et al., 2014; Noubactep et al., 2012; Wang et al., 2015).
Therefore, the present work aimed to study the treatment of
TCE, PCE, and 1,2eDCE contaminated wastewater using PEIenZVI.
In this context, the effects of different concentrations and time
scales on the degradation of contaminants were evaluated. The
removal rate and efficiency of PEIenZVI was compared with those
of fresh nZVI. Furthermore, Xeray photoelectron spectroscopy
(XPS) and Xeray Absorption near edge spectroscopy (XANES) was
also used to determine the corrosion products and the mechanisms
of DNAPL reduction on nZVI and PEIenZVI surfaces.
2. Materials and methods
2.1. Preparation of nZVI and PEIenZVI
The zero-valent iron nanoparticles were prepared using a well-
known chemical reduction method that employs sodium borohy-
dride as a reducing agent. Ponder et al. (2000), first proposed the
use of NaBH4, a strong reducing agent for Fe (0) precipitation. The
reaction can be described as:
FeSO4 þ Rðstrong reducing agentÞ/Feð0ÞðnanometalÞþRSO4
(1)
The reaction of a ferrous sulfate aqueous solution with this
strong reducing agent takes place in two steps:

FeðH2 OÞ2þ
7 þ 2e /Feð0ÞY þ 7H2 O (2)
2BH 
4 þ 6H2 O/2BðOHÞ3 þ 2e þ 7H2 O (3)
By merging the two half-reactions, the complete reaction
becomes:

FeðH2 OÞ2þ
7 þ 2BH4 /Feð0ÞY þ 2BðOHÞ3 þ H2 O þ 7H2 (4)
Following this method, 10 g of FeSO4$7H2O were dissolved in
100 mL of 30% ethanol, and 70% deionized (DI) water. The pH value
was adjusted to 6.8 with 3.8 N NaOH(aq). After the addition of 1.8 g
NaBH4 powder, the mixture was stirred for 20 min, and the residues
were then washed several times with ethanol. The resulting nZVI
were vacuum-dried overnight. The PEI solution was prepared by
dissolving an appropriate amount of the polymer into DI water
followed by sonication for 2 h. Upon mixing the polymer solution
with the resulting particles, the resultant mixture was sonicated for
1 h and then dried in an oven at 80  C for 12 h.
2.2. Characterization of nZVI and PEIenZVI
The morphology and microstructure of the nZVI were deter-
mined by FEeSEM/EDS (Hitachi, Se4700 Type II) and TEM (Hitachi,
He7500) with a resolution of 0.1 nm. The crystal structure of nZVI
458 K.-S. Lin et al. / Journal of Cleaner Production 175 (2018) 456e466

was measured by Xeray diffraction (XRD) using monochromatic a high performance hood at 60 fpm (ft/s) of gas emission rate with a
CuKa radiation (MAC Science, MXP18). The pore volume and fixed-bed activated carbon absorption column.
specific surface area of the nZVI were obtained by N2 adsorption
(Micromeritics ASAP, 2010) at 77 K. The BETeLangmuir surface area Headspace concentrations of VOCs
and pore volume were evaluated using BarreteJoynereHalenda Saturated Vapour Pressures of VOCs
(BJH) model. Average pore diameter was calculated by density ¼  106 ðppmÞ (5)
Atmospheric Pressure
functional theory (DFT) method using ASAP 2010 analyzer's built-in
software. The surface areas of nZVI (37 m2/g) and PEIenZVI (53 m2/ Additionally, the concentrations of TCE, PCE, and 1,2eDCE dur-
g) were much larger than that of the nonporous bulk Fe (0) particles ing decholorination reactions were derived by mass balance of
(3e7 m2/g) (Adhikari and Lin, 2016). The magnetic strengths of chloride species (Cl-species), and then confirmed with the instan-
nZVI and PEIenZVI have been measured at 300 K using a super- taneous concentrations which were measured using GCeTCD. For a
conduction quantum interference device (SQUID, Quantum Design, batch reactor, the mass balance of chloride species can be described
MPMS7) with a magnetic field at ±30 kOe. X-ray photoelectron as Eq. (6)
spectroscopy (XPS) spectra of the nanoparticle surfaces were
obtained using a Fison ESCA 210 (VG Scientific) spectrometer with Mass balance of Cl  species : Input  Output þ Generation
an Mg Ka Xeray (1253.6 eV) excitation source and a cylindrical ¼ Accumulation
mirror analyzer (CMA). In order to avoid further oxidation, the nZVI (6)
samples were prepared by drying in a small nitrogen-purged hood
at room temperature and then packed into the sample cell directly. In a steady-state batch reactor with constant volume, the output
XANES spectra were collected at the Wiggler Beamline BL 17C1 at and accumulation terms are zero, and the Cl-species are consumed
the National Synchrotron Radiation Research Center (NSRRC) of (Generation < 0). Thus Eq. (6) can be simplified to Eq. (7):
Taiwan. The electron storage ring was operated with 1.5 GeV of
energy and a current range of 100e200 mA. A Si (111) DCM was dðCCl VÞ
 0  RdV ¼ 0 (7)
used for providing highly monochromatized photon beams with dt
energies of 5e15 keV and a resolving power (E/DE) of up to 7000. The decholorination reactions of TCE, PCE, and 1,2eDCE were
Data were collected in fluorescence or transmission modes with a assumed as a first-order reaction that Eq. (7) can be derived to Eq.
Lytle ionization detector for the K-edge of Fe (7112 eV) at room (8):
temperature (Conradson, 1998; Lytle, 1999). XANES spectra were
analyzed using the IFEFFIT package (Ravel and Newville, 2005). dCCl
¼ R ¼ kCCl (8)
dt
2.3. Batch reductive dechlorination experiments over nZVI and Subsequently, Eq. (8) was integrated from initial to final state, as
PEIenZVI displayed in Eqs. (9)e(11):

Reductive dechlorination experiments were performed using ZCCl Zt

both fresh nZVI and PEIenZVI solutions at a concentration of 2 g/L. dCCl
¼ k t (9)
Kinetic studies for each set of experiments were conducted by CCl
0
CCl t0
adding nZVI to different concentrations (50 and 100 mg/L) of TCE,
PCE, and 1,2eDCE solutions in 500 mL glass beakers. All of the so-
lutions were stirred at the rate of 120 rpm (rate per minute) using a CCl
ln ¼ kðt  t0 Þ (10)
shaker with a temperature controller, and the dechlorination re- 0
CCl
actions (reductions of TCE, PCE, and 1,2eDCE) were carried out at
25  C for 24 h. Liquid samples were taken by a syringe every 10 min, 0
20 min, 30 min, 1 h, 10 h, and 24 h and subsequently filtered CCl ¼ CCl exp½  kðt  t0 Þ (11)
through a 0.22 mm PVDF syringe filter. To determine the rates and If the t0 ¼ 0, Eq. (11) can be further simplified to Eq. (12):
kinetics of DNAPL degradation, LC/MS/MS (LC 1100 series/MSD
TRAP XCT, Agilent) and GC/MS (TurboMassGold, Perkin Elmer) 0
CCl ¼ CCl expðktÞ (12)
were employed. FTIR experiments were recorded using Digilab FTIR
spectrometer (FTSe40). Standard calibration lines of TCE, PCE, and in which
1,2-DCE were performed for converting the GC peak areas into
concentrations for the evaluation of TCE, PCE, and 1,2-DCE reduc- C0Cl: Initial concentration of chloride species (ppm)
tion efficiencies. That is, few volumes of these pure VOCs (volatile CCl: Final concentration of chloride species (ppm)
organic compounds) solvents in a closed sample glass at 30  C to V: Volume of batch reactor (L)
acquire the high concentration vapor gases. Afterward, these vapor R: Consumption reaction rate of chloride species (ppm/min)
gases at the top of bottom were diluted with an appropriate volume k: Rate constant of decholorination reaction (1/min)
of nitrogen gas to form the diluted VOCs gases with different t: Reaction time (min)
concentrations (20, 40, 60, 80, and 100 ppm). These 0.5 mL of these
nitrogen-diluted TCE, PCE, and 1,2-DCE gases were collected by a
plunger-in-needle syringe to inject into GCeTCD analyzer with a 3. Results and discussion
Porapak Q stainless column (80/100, 9.15 mL  3.175 mm (O.D.)) at
50  C for 11 min, and then heated up to 100  C at 5  C min1. 3.1. Morphologies, configurations, and physiochemical properties of
Notably, the headspace concentrations of TCE, PCE, and 1,2-DCE nZVI/PEIenZVI
vapor gases at a specific temperature were derived from Eq. (5) to
obtain the nitrogen volumes for VOCs dilution. In order to control As shown in Fig. 1(a)e(d), the fresh ultrafine nZVI exhibited a
VOCs emission, the preparations of TCE, PCE, and 1,2-DCE solutions spherical shape with sizes of 20e50 nm, whereas the PEIenZVI
and batch reductive dechlorination experiments were conducted in exhibited a nearly monodisperse size of 60e70 nm. Since the strong
 K.-S. Lin et al. / Journal of Cleaner Production 175 (2018) 456e466
Fresh nZVI
200 nm
PEI-nZVI
100 nm
Fig. 1. (a) (c) FEeSEM microphotos and (b) (d) HReTEM images of fresh nZVI and PEIenZVI, respectively.
magnetic property of nZVI, an aggregation between the particles
was observed. This can be attributed to that the nZVI without any
surface modification by a polymer possesses smaller surface area
which easily causes agglomeration. However, with the coating of
PEI on the surface of nZVI having steric effect, the particles were
evenly dispersed, and thus providing large surface area with highly
reactive sites to enhance the removal of more contaminates. The
coating of nZVI particles with PEI reduced the magnetic in-
teractions between the nZVI particles. XRD patterns confirmed the
formation of elemental iron and the phase purity of the nano-
particles. As shown in Fig. 2(a), peaks assigned to the crystalline
structure of Fe (0) (JCPDS No. 87e0721) were detected in both
samples. As a consequence, the surface modification process had no
significant effect on iron oxidation. Interestingly, the nZVI exhibited
high magnetism and could easily be retained and separated from
treated water with the aid of a magnet. As shown in Fig. 1(a)e(d),
the magnetic attraction between the particles led to unavoidable
agglomeration, which in turn reduced the surface available for
reduction reactions. Having a notably higher surface area, nZVI also
undergoes redox reactions in aqueous media that have an adverse
impact on the reactivity and total reduction capacity of the nZVI
(Baltazar et al., 2014; Freyria et al., 2017). The amphiphilic nature of
the polymer shell structures of nZVI does have some advantages,
such as good dispersion and high stability against oxidation, which
enhances the delivery of active nZVI to the contaminant surface
(Dehvari et al., 2014).
The interaction between PEI and nZVI were examined through
comparison of FTIR measurements of dried samples. As shown in
Fig. 2(b), the characteristic peak of 3380 cm1 corresponding to
n(NH) stretching vibrations can be clearly observed in the IR spectra
of PEIenZVI samples. The NeH bending vibration was also
discernible at 1620 cm1, which is an indication of PEI presence on
the nZVI surface. Due to the high reactivity of nZVI, the PEI reaction
with nZVI may reduce some PEI molecules on the surface, resulting
459
(a) Fresh nZVI (b)
200 nm
(c) PEI-nZVI (d)
200 nm
in the appearance of other nitrogen and CH functional groups in the
FTIR spectra. Moreover, a schematic of the PEIenZVI structure is
represented in Fig. 3(a). NH groups of the branched PEI chains
interacted electrostatically with carboxyl groups on the nZVI sur-
face to form a cationic core/shell structure (Ku et al., 2013; Teng
et al., 2017). The XPS studies of the PEIenZVI surface composition
confirmed the core/shell structure. The EDS results (Table 1) indi-
cate that the carbon and oxygen content on the PEIenZVI is greater
than on the nZVI. The EDS results (Table 1) indicate that the carbon
(23.42 at%) and oxygen (39.21 at%) content on the PEIenZVI is
greater than that on the nZVI (C: 8.62 at%; O: 54.06 at%). As shown
in Fig. 3(a), the PEI is an amino (NH2) group-rich polymer that
coated on the surface of nZVI nanoparticles, leading to the
increasing of carbon and nitrogen ratios in EDS analysis. In addition,
the nZVI nanoparticles were oxidized to form a thin iron oxide
layer, preventing the further internal oxidation before coating PEI
polymer. Based on these reasons, the iron content was relatively
decreasing from 52.17 at% to 21.05 at%, implying the successful
coating onto the nZVI surfaces. Likewise, the XPS was used to probe
the carbon element backbone of the polymer structure, and it was
found that nanoparticles were coated with a layer of polymer as
represented in Fig. 3(b). From these results, the PEI shell thickness
calculated to be around 2 nm approximately.
Non-specific physical adsorption experiments were performed
for the nZVI and PEIenZVI to measure the total surface area and
pore size distribution (Fig. 4(a) and (b)). The surface areas of nZVI
(37 m2/g) and PEIenZVI (53 m2/g) were much higher than that of
the nonporous bulk Fe (0) particles (3e7 m2/g). The method of
Barrett, Joyner, and Halenda was employed to analyze the pore size
distribution. The distribution clearly indicates that the nZVI and
PEIenZVI specimens are mesopores. The adsorption-desorption
isotherms exhibit hysteresis behavior, with a Type IV hysteresis
isotherm, indicating that the specimens were mainly mesopores.
The porous structures of nZVI and PEIenZVI are supported by the
460 K.-S. Lin et al. / Journal of Cleaner Production 175 (2018) 456e466

(100) (a) H (a)

N
(200) Fe(0) H
H + H
44.47 N N
Intensity (a.u.)

H + +H
34.36 PEI-nZVI
61.99 H + +H
44.59 N N
H + H
H
Fresh nZVI N
64.62 H

20 30 40 50 60 70 80
2 (degree)

90 (b)
1.4
(b) 80 20 Fe
1.3 3380 NH
Atomic percent (%)

1340 CH3 70
1.2
60
Absorbance (a.u.)

1.1
1620 NH
CH 50
1.0
CH 40
0.9
30 C
0.8
20
0.7
10
0.6
0 50 100 150 200 250 300
0.5
Etch time (sec)
0.4
4000 3500 3000 2500 2000 1500 1000 500 Fig. 3. (a) A schematic of PEIenZVI core-shell microstructure and (b) surface probing
depth analyses of PEIenZVI using XPS/Auger measurement.
Wavenumber (cm-1)
Fig. 2. (a) XRD patterns of fresh nZVI, PEIenZVI, and Fe (0) standard; (b) FTIR spectra Table 1
of as-synthesized PEIenZVI. EDS analyses of fresh nZVI and PEI-nZVI.

nitrogen adsorption-desorption isotherm suggesting a Type IV
curve with a H3-like hysteresis loop and occurrence of mesopore-
characteristic capillary condensation of nitrogen. Types IV hyster-
esis isotherm is considered to reflect capillary condensation
phenomena in that it levels off before the saturation pressure is
reached. It may show hysteresis effects the characteristic Type IV
isotherm shape is generally associated with configurations of nZVI
and PEIenZVI. The mesoporous capacity is the amount adsorbed at
the plateau and the mesoporous volume is then obtained by
assuming the condensate density to be that of liquid nitrogen
(Fig. 4, insets) (Adhikari and Lin, 2016; Lin et al., 2013). According to
the data listed in Table 2, the synthesis procedure and ingredients
can govern the physicochemical properties of nZVI. The higher
surface area of PEIenZVI facilitates suspension of nZVI in ground-
water as well as the dissolution of contaminants, providing the
catalyst with a rich surface and redox chemistry, as well as
potentially unique properties for contaminant removal and
transformation.
Nevertheless, the magnetic strengths of nZVI and PEIenZVI are
also the key factor relating to the practical application in the in-situ
Elementa O Fe C N
Fresh nZVI 39.21 52.17 8.62 Not detectable
PEI-nZVI 54.06 21.05 23.42 3.47
a
Ratios of each element are expressed in atomic percent (atom%).
contaminant decomposition. In general, the saturation magneti-
zation of nZVI nanoparticles decreased since the polymer coated
over the outer shell of nZVI. (Namgung et al., 2010; Shi et al., 2012).
It is an important principle to distinguish the polymer is coated
over nZVI or not. As displayed in Fig. 5, the nZVI has no remanence
and magnetic hysteresis, inferring that the nZVI is a super-
paramagnetic material. Accordingly, nZVI nanoparticles can be
regarded as several tiny magnetic bars which are self-magnetic
polarized along the magnetic field line. It means that when the
outer magnetic field is removed (0 kOe), the remanence of nZVI
disappears. In addition, the saturation magnetization of nZVI and
PEIenZVI were 82.1 and 70.7 emu/g, respectively. These results
indicate that both nZVI and PEIenZVI with superparamagnetism
can be directed to the contaminant positions, and then recycled
from the treated ground water with the aid of magnet field.
 K.-S. Lin et al. / Journal of Cleaner Production 175 (2018) 456e466 461

100
80 0.016 (a) nZVI
Pore volume (cm3/g) 0.014 80
70 PEI-nZVI
Volume adsorbed (cm3/g)

0.012
0.010
60
60

Magnetization (emg/g)
0.008 40
50 0.006
0.004 20
40 0.002
0.000 0
30 -0.002
10 100 1000 -20
20 Pore diameter (Å)
-40
10 Adsorption -60
Desorption
0 -80
0.0 0.2 0.4 0.6 0.8 1.0 -100
-30 -20 -10 0 10 20 30
Relative pressure (P/P0) Magnetic field (kOe)
Fig. 5. Magnetic hysteresis curves of as-synthesized nZVI and PEIenZVI.

90
(b)
80 0.010
Pore volume (cm3/g)

nZVI+TCE 50 ppm (a)

Volume adsorbed (cm3/g)

0.008 50
70 PEI-nZVI+TCE 50 ppm
0.006
60 nZVI+PCE 50 ppm
0.004 40
Concentration (ppm)

50 0.002 PEI-nZVI+PCE 50 ppm

40 0.000 nZVI+1,2-DCE 50 ppm

30
10 100 1000 PEI-nZVI+1,2-DCE 50 ppm
30 Pore diameter (Å)
20 20

10 Adsorption
Desorption 10
0
0.0 0.2 0.4 0.6 0.8 1.0
Relative pressure (P/P0) 0
0 20 40 60 80 100 120
Fig. 4. N2 adsorption/desorption isotherms of (a) nZVI and (b) PEIenZVI. Inserts are Time (min)
pore size distribution curves.

Table 2
100 nZVI +TCE 100 ppm (b)
The specific surface area of as-synthesized nZVI and PEIenZVI.
PEI-nZVI +TCE 100 ppm
Fe (0) type Specific surface areaa (m2/g) References
80 nZVI +PCE 100 ppm
Concentration (ppm)

Bulk Fe (0) powder 3e7 This work

nZVI from FeCl3.6H2O 24.7 (Son et al., 2012) PEI-nZVI +PCE 100 ppm
nZVI from FeSO4.7H2O 43.0 (Su et al., 2013)
nZVI from FeSO4.7H2O 34.7 This work
60 nZVI +1,2-DCE 100 ppm
PEIenZVI 53.4 This work
PEI-nZVI +1,2-DCE 100 ppm
a
Note: Specific surface areas were measured by BJH method.
40
3.2. Degradation kinetics of PCE, TCE, and 1, 2eDCE
20
Batch studies were performed to determine the potential for
reductive elimination of PCE, TCE, and 1,2eDCE by dehalogenation
at different concentrations. The results of dechlorination studies for 0
three DNAPLs using both fresh and surface modified nZVI are 0 20 40 60 80 100 120
summarized. Careful observation of kinetic studies revealed that Time (min)
reaction rates varied for the different chemicals (PCE, TCE, and
Fig. 6. (a) 50 and (b) 100 ppm of TCE, PCE or 1,2eDCE over nZVI and PEIenZVI
1,2eDCE) and also for different initial concentrations. As shown in
respectively via reductive reaction within initial 120 min (concentration of nZVI was
Fig. 6(a) and (b), over 95% of 1,2eDCE at a concentration of 50 and 2 g/L).
100 mg/L were removed within 10 min in reaction with fresh nZVI,
462 K.-S. Lin et al. / Journal of Cleaner Production 175 (2018) 456e466

whereas the number for PCE and TCE was 93% and 90%, respec-
100 nZVI +TCE 100 ppm (a)
tively. However, the degradation of a higher concentration (100 mg/
L) was different, and only 90% of 1,2eDCE, 85% of TCE, and 75% of nZVI +PCE 100 ppm
PCE were removed (Fig. 6(b)). Higher concentrations may elevate 80 nZVI +1,2-DCE 100 ppm
the passivation of nZVI due to the formation of iron oxides and

Concentration (ppm)
contaminant immobilization on the nZVI surface (Lin et al., 2008).
The comparison of these results with the degradation rate and 60
removal efficiency of DNAPLs by PEIenZVI is enlightening. In the
beginning, fresh nZVI acted actively and degraded DNAPLs quickly,
as displayed in Figs. 7(a)e(b) and Figs. 8(a)e(b). However, after 2 h, 40
both reached the same values of 97%, 96%, and 96% for TCE, PCE, and
1,2eDCE, respectively. In fact, PEI played a protective role for the
nZVI catalyst until it could reach the polluted area. In the duration
20
of nanoparticle flowing, the PEI shell of composite nZVI may
dissolve with increasing immersing time in groundwater. The thin 0
film layer of PEI is able to enhance the mobility and stability of nZVI
until it reaches the DNAPLs contaminated pools. To investigate the 0 200 400 600 800 1000 1200 1400
role of PEI in nZVI delivery and mobility in contaminated ground- Time (min)
water, dissolution of PEIenZVI in aqueous solution was determined
by a total organic carbon (TOC) analyzer (Fig. 9). Results indicate 100 PEI-nZVI +TCE 100 ppm (b)
that the PEI concentration in solution increased quickly and leveled
off after 90 min with a total concentration of 250 ppm for organic PEI-nZVI +PCE 100 ppm
carbon. The initial concentration of organic carbon in PEI coated on 80 PEI-nZVI +1,2-DCE 100 ppm
nZVI was 800 ppm. Consequently, due to the presence of this Concentration (ppm)

60
(a)
50 nZVI+TCE 50 ppm
nZVI+PCE 50 ppm 40
40 nZVI+1,2-DCE 50 ppm
Concentration (ppm)

20
30
0
20 0 200 400 600 800 1000 1200 1400
Time (min)
10 Fig. 8. 100 ppm of TCE, PCE or 1,2eDCE over (a) nZVI and (b) PEIenZVI respectively via
reductive reaction at different time (concentration of nZVI was 2 g/L).

0
Total organic compound (ppm)

0 200 400 600 800 1000 1200 1400 250

Time (min)
200
PEI-nZVI+TCE 50 ppm (b)
50
PEI-nZVI+PCE 50 ppm 150
40 PEI-nZVI+1,2-DCE 50 ppm
Concentration (ppm)

100
30
50
20
0
10 0 20 40 60 80 100 120 140 160 180
Time (min)
0 Fig. 9. Dissolution rate of PEI segment of PEIenZVI in the aqueous solution.

0 200 400 600 800 1000 1200 1400

protective layer, the PEIenZVI could remain active and potentially
Time (min) mobile for a longer time. The longevity and dissolution rate of PEI
Fig. 7. 50 ppm of TCE, PCE or 1,2eDCE over (a) nZVI and (b) PEIenZVI respectively via depended on the number of decorated polymers on the nZVI sur-
reductive reaction at different time (concentration of nZVI was 2 g/L). faces and the grafting density (Shin et al., 2008).
 K.-S. Lin et al. / Journal of Cleaner Production 175 (2018) 456e466 463

3.3. Reductive dechlorination mechanism and nZVI/PEIenZVI 100000

corrosion products (a)

80000 Fe 2p3/2 O 1s
The main drawback for nZVI application in groundwater
remediation is its tendency to react with non-target substances. In
60000

Intensity (a.u.)
fact, high reactivity correlates with low selectivity (Kim et al., 2017;
Noubactep et al., 2012; Shin et al., 2008). It is well known that the O 1s
core-shell structure of reacted nZVI consists of zero-valent iron in
the nucleus and either iron oxide or hydroxide in the outer shell 40000
layers, which is consistent with the progress of nZVI oxidation
(Torrey et al., 2015). Comparative XPS studies in Figs. 10(a)e(b) and Fe 3p
20000
Figs. 11(a)e(b) indicate that surface modification occurs through an
enhanced reduction reaction of the nZVI surface, in which FeO
species are replaced with magnetite with a higher oxidation state. 0
Possibly, the dissolution of the chlorinated compound on the sur-
face of PEIenZVI improves contact of nZVI with the contaminant. 1400 1200 1000 800 600 400 200 0
The nature of the iron oxide surface is one of the dominant factors Binding energy (eV)
in the progress of the reductive reaction. The outer layer of iron
oxides mediates the reduction reaction through electron transfer at
the iron surface (Liang et al., 2017; Zhao and Reardon, 2012). That is,
since TCE particles dissociate, chloride ions are produced. Chloride
ions act as an electron donor, releasing electrons to metallic species
and generating new reactive sites over the metal surfaces (USEPA, 70000 (b)
FeSO4 Fe3O4

60000
120000
(a)
50000
Intensity (a.u.)

100000 Fe 2p3/2

80000 40000
Intensity (a.u.)

O 1s

60000 30000
C 1s
40000 20000

20000 Fe 3p 10000
716 714 712 710 708 706 704
Binding energy (eV)
0
Fig. 11. (a) XPS full spectrum and (b) fitting results of PEIenZVI reacted with 100 ppm TCE.
1400 1200 1000 800 600 400 200 0
Binding energy (eV) 2005; Johnson et al., 1998; Padmanabhan et al., 2008;
Ramamurthy and Eglal, 2014). Zhang et al. (2012) and Kaifas et al.
(2014) have indicated a chain dechlorination reaction of TCE into
1,1-dichloroethene (1,1eDCE), cise1,2edichloroethene (ciseDCE),
and then vinyl chloride (VC) analyzed using GC/MS. Moreover, the
degradation mechanism occurring over nZVI and polymer-
Fe3O4 (b)
FeO modified nZVI were also presented by many researchers (Liang
25000 FeSO4 et al., 2017;Padmanabhan et al., 2008; Yan et al., 2013, 2017;
USEPA, 2005). Based on these proposed models of TCE and PCE
dechlorination, the mechanism of the DNAPLs degradation
Intensity (a.u.)

20000
occurred was assumed as shown in Fig. 13. It exhibited the degra-
dation of PCE into ethylene, TCE degradation into 1,2eDCE and
15000 1,2eDCE degradation into vinyl chloride and ethylene. Further-
more, the porous structure of the iron oxides may contribute to
contaminant removal through an adsorption mechanism.
10000 In order to thoroughly examine the nature of nZVI products, the
XANES spectra were also studied. Fig. 12(a) and (b) show the XANES
spectra of an Fe atom in reacted nZVI and PEIenZVI samples,
5000
exhibited as an absorbance feature (Fe ¼ 7114 eV) for the 1s to 3d
transition. The sharp feature at 7134 eV arose from the dipole-
716 714 712 710 708 706 704
allowed 1s to 4pxy electron transition, indicated the existence of
Binding energy (eV) Fe(III) (Dehvari et al., 2014; Lin et al., 2013). The intensity of the
Fig. 10. XPS spectra of (a) full spectrum and (b) fine section of Fe 2p3/2 for fitting re-
transition was proportional to the population of Fe(III) in the nZVI
sults of fresh nZVI reacted with 100 ppm TCE. samples. In comparison to standard spectra, the XANES spectra of the
464 K.-S. Lin et al. / Journal of Cleaner Production 175 (2018) 456e466

1.6 Table 3
Fe foil (a) Fe XANES K-edge absorption of nZVI and PEIenZVI compared to that of
1.4 FeO STD standard samples.
Normalized absorbance (a.u.)

Fe2O3 STD
Iron compound XANES absorption (eV)
1.2 Fe3O4 STD
Fe (0) foil 7112.0
nZVI+O2 FeO standard 7113.7
1.0
Fe3O4 standard 7115.1
Fe2O3 standard 7117.9
0.8 0.2 nZVI 7114.7
PEIenZVI 7115.0
0.6
0.4
rich crystalline inner layer to an Fe(III)-rich outer layer that is more
0.2 amorphous, which is in good agreement with the XPS data. Addi-
0.0
7110 7120
tionally, EXAFS analyses were further performed to understand the
0.0 fine structure and atomic arrangement regarding the bond distance
7100 7110 7120 7130 7140 7150 7160 7170 and coordination number. The EXAFS fitting results for the oxygen
shell listed in Table 4 suggest that nZVI and PEIenZVI have two
Photon energy (eV)
center Fe atoms coordinated primarily by FeeO and FeeFe bonding.
The fine structural parameter of the fresh nZVI catalyst indicates the
Fe atom is in formation FeeFe with a bond distance of 2.25 Å and the
corresponding coordination number (CN) of 9.66 for the first shell. In
all EXAFS data, the DebyeeWaller factors (s2) are less than 0.017 (Å2).
1.6 However, the FeeO distance and coordination numbers of reacted
(b) samples are slightly different. The higher coordination number, for
Fe foil
1.4
Normalized absorbance (a.u.)

FeO STD example in the PEIenZVI sample, associated with the upper oxygen
Fe2O3 STD density and oxidation state (Dehvari et al., 2016; Fagerlund et al.,
1.2
Fe3O4 STD 2012; Kim et al., 2017; Son et al., 2012). Accordingly, a porous layer
1.0 PEI-nZVI of corrosion products formed onto the active sites of the nZVI surface
that governs mass transfer and consequently the reduction of con-
0.8 taminants by the nZVI core (Goswami et al., 2017; Jamieson-Hanes
0.36
0.32
et al., 2014; Liang et al., 2017). The pore structure on the surface di-
0.6 0.28
0.24 vides the nZVI into smaller particles, providing pathways for con-
0.20
0.4 0.16 taminants to react with fresh iron in the core.
0.12
0.08
0.2 0.04
3.4. Engineering assessment of PEIenZVI injection system for
0.00
7108 7110 7112 7114 7116 7118 7120
remediation of DNAPLs contaminated plume
0.0
7100 7110 7120 7130 7140 7150 7160 7170
Most known in-situ applications of nZVI have applied direct
Photon energy (eV) injection techniques. A wide range of direct injection methods
exists, however their main purpose is to deliver slurry of nZVI at a
Fig. 12. Fe K-edge XANES spectra of (a) fresh nZVI and (b) PEIenZVI reacted with
100 ppm TCE.
certain depth and in a definite amount directly into soil and/or
aquifer materials (Comba et al., 2011; Henn and Waddill, 2006; Kim
et al., 2017; Mueller et al., 2012). The slurry is expected to be
pumpable, which permits injection; and also to have a known
behavior in groundwater, such as to disperse through the aquifer
carried by groundwater flow and finally reach the DNAPL pools
(Bardos et al., 2011). As exhibited in Fig. 14, an engineering design of
in-situ DNAPLs removal using PEIenZVI material is proposed in this
study. In general, the dry cleaning, metal degreasing, pharmaceu-
tical production, pesticide formulation and petrochemical industry
emitted numerous DNAPLs that polluted soil and ground water
significantly (Chen and Wu, 2017; Lerner et al., 2003). These
emitted DNAPLs diffused with flowing groundwater to deeper sites
and then accumulated to DNAPLs pools until it reached to a low
permeability sites (Mo et al., 2017). In order to solve the problem,

Table 4
EXAFS structural parameters of the fresh and degradation products of nZVI and
PEIenZVI.
Fig. 13. The core-shell structure of PEIenZVI and reaction mechanism of TCE, PCE, and
1,2eDCE degradation. First Shell CN (±0.05) a
R (±0.01 Å)b Ds2 (Å2)c
Fresh nZVI FeeFe 6.53 2.36 0.016486
products are similar to the Fe3O4 standard. The XANES results of Fresh PEIenZVI FeeFe 5.88 2.51 0.012693
reacted samples are listed in Table 3. It is evident that after the re- Used nZVI FeeO 4.09 1.96 0.012914
action that PEIenZVI contains more Fe3O4 than nZVI. As a result, Used PEIenZVI FeeO 4.31 1.94 0.010282
corrosion might lead to a laminated structure, shifting from an Fe(II)- Notes: a
Coordination number; b
Bond distance; c Debye-Waller factor.
 K.-S. Lin et al. / Journal of Cleaner Production 175 (2018) 456e466
Fig. 14. Design concept of in-situ remediation of DNAPLs contamination using
PEIenZVI injection.
the as-prepared PEIenZVI particles can be injected into soil, and it
moves with flowing groundwater to DNAPLs pools. Most specially,
the configuration of composite PEIenZVI can be changed for
effective DNAPLs degradation with a more feasible engineering
design basis. In the duration of flowing, the PEI shell of composite
nZVI may dissolve with increasing immersing time in groundwater.
Afterward, highly active nZVI cores flows with groundwater, and
then decomposed the DNAPLs into low-toxic vinyl chloride, ethane,
and ethylene at deeper position for environmental protection. In
this case, the nZVI cores are exposed at DNAPL pools for degrada-
tion after a gradual dissolving of PEI from the surface of nZVI.
4. Conclusions
To address the limitation of nZVI application on wastewater
treatment, PEIemodified nZVI were synthesized, characterized, and
applied for dechlorination of TCE, PCE, and 1,2eDCE in a laboratory-
scale experiment. Results from this study indicate that nZVI and
PEIenZVI effectively remove the majority of three DNAPLs (TCE,
PCE, and 1,2eDCE) within 10 min. Notably, it was found that most of
TCE (90%), PCE (93%), and 1,2eDCE (95%) could be removed from
50 ppm DNAPLs solutions. At higher concentration (100 ppm), the
removal percent of TCE, PCE, and 1,2eDCE were 85, 75, and 90%,
respectively. After 2-h dechlorination reaction, the removal percent
of TCE, PCE, and 1,2eDCE could respectively reach to 97, 96, and 96%
in both of 50 and 100 ppm DNAPLs solution. There was a chain
dechlorination reaction of PCE into ethylene. At same time, TCE was
degraded into 1,2eDCE and then be dechlorinated into less toxic
vinyl chloride and chloride-free ethylene. Most interestingly,
PEIenZVI can enhance the nZVI delivery to the water/DNAPLs
interface, thereby improving nZVI activity and durability in aqueous
media. In addition, a postulated concept of engineering assessment
of nZVIs injection system for remediation of DNAPLs contaminated
plume was proposed to confirm the feasibility for such a treatment
practice and determine if further development would be war-
ranted. The high reaction rates and complete removal of DNAPLs,
suggest that the PEIenZVI remediation technology is an economical
and environmentally friendly method for in-situ decontamination
and detoxification of chlorinated groundwater or waste streams.
Acknowledgements
The financial support of the Ministry of Science and Technology
(MOST), Taiwan (MOST 103-2621-M-155-001) is gratefully
acknowledged. We also thank Prof. Y. W. Yang, Dr. J. F. Lee, and Dr.
Jeng-Lung Chen from Taiwan National Synchrotron Radiation
Research Center (NSRRC) for their help in the XANES/EXAFS ex-
periments or data analyses.
465
References
Adhikari, A.K., Lin, K.S., 2016. Improving CO2 adsorption capacities and CO2/N2
separation efficiencies of MOF-74(Ni, Co) by doping palladium-containing
activated carbon. Chem. Eng. J. 284, 1348e1360.
Baker, R.S., Nielsen, S.G., Heron, G., Ploug, N., 2016. How effective is thermal
remediation of DNAPL source zones in reducing groundwater concentrations?
Ground Water Monit. Remed. 36 (1), 38e53.
Baltazar, S.E., García, A., Romero, A.H., Rubio, M.A., Arancibia-Miranda, N., Altbir, D.,
2014. Surface rearrangement of nanoscale zero-valent iron: the role of pH and
its implications in the kinetics of arsenate sorption. Environ. Technol. 35 (18),
2365e2372.
Bardos, P., Bone, B., Boyle, R., Ellis, D., Evans, F., Harries, N.D., Smith, J.W., 2011.
Applying sustainable development principles to contaminated land manage-
ment using the SuRF-UK framework. Remed. J. 21 (2), 77e100.
Bennett, P., He, F., Zhao, D., Aiken, B., Feldman, L., 2010. In situ testing of metallic
iron nanoparticle mobility and reactivity in a shallow granular aquifer.
J. Contam. Hydrol. 116 (1), 35e46.
Chaufer, B., Deratani, A., 1988. Removal of metal ions by complexation-
ultrafiltration using water-soluble macromolecules: perspective of application
to wastewater treatment. Nucl. Chem. Waste Manag. 8 (3), 175e187.
Chen, H.M., Wu, M.T., 2017. Residential exposure to chlorinated hydrocarbons from
groundwater contamination and the impairment of renal function-an ecolog-
ical study. Sci. Rep. 7, 40283, 9 pages.
Chen, Y., Pan, B., Li, H., Zhang, W., Lv, L., Wu, J., 2010. Selective removal of Cu(II) ions
by using cation-exchange resin-supported polyethyleneimine (PEI) nano-
clusters. Environ. Sci. Technol. 44 (9), 3508e3513.
Comba, S., Di Molfetta, A., Sethi, R., 2011. A comparison between field applications
of nano-, micro-, and millimetric zero-valent iron for the remediation of
contaminated aquifers. Water Air Soil Pollut. 215 (1), 595e607.
Conradson, S.D., 1998. Application of Xeray absorption fine structure spectroscopy
to materials and environmental science. Appl. Spectrosc. 52 (7), 252Ae279A.
Dehvari, K., Chen, Y., Tsai, Y.H., Tseng, S.H., Lin, K.S., 2016. Superparamagnetic iron
oxide nanorod carriers for paclitaxel delivery in the treatment and imaging of
colon cancer in mice. J. Biomed. Nanotechnol. 12 (9), 1734e1745.
Dehvari, K., Lin, K.S., Wang, S.S.S., 2014. Structural characterization and adsorption
properties of Pluronic F127 onto iron oxides magnetic nanoparticles. J. Nanosci.
Nanotechnol. 14 (3), 2361e2367.
Dong, H., Lo, I.M., 2013. Influence of humic acid on the colloidal stability of surface-
modified nano zero-valent iron. Water Res. 47 (1), 419e427.
Dong, H., Zhao, F., Zeng, G., Tang, L., Fan, C., Zhang, L., Zeng, Y., He, Q., Xie, Y., Wu, Y.,
2016. Aging study on carboxymethyl cellulose-coated zero-valent iron nano-
particles in water: chemical transformation and structural evolution. J. Hazard
Mater. 312, 234e242.
Fagerlund, F., Illangasekare, T.H., Phenrat, T., Kim, H.J., Lowry, G.V., 2012. PCE
dissolution and simultaneous dechlorination by nanoscale zero-valent iron
particles in a DNAPL source zone. J. Contam. Hydrol. 131 (1), 9e28.
Freyria, F.S., Esposito, S., Armandi, M., Deorsola, F., Garrone, E., Bonelli, B., 2017. Role of
pH in the aqueous phase reactivity of zero-valent iron nanoparticles with acid
orange 7, a model molecule of azo dyes. J. Nanomater. 2017, 2749575, 13 pages.
Fu, F., Dionysiou, D.D., Liu, H., 2014. The use of zero-valent iron for groundwater
remediation and wastewater treatment: a review. J. Hazard Mater. 267, 194e205.
Ghosh, I., Mukherjee, A., Mukherjee, A., 2017. In planta genotoxicity of nZVI: in-
fluence of colloidal stability on uptake, DNA damage, oxidative stress and cell
death. Mutagenesis 32 (3), 371e387.
Golshirazi, A., Kharaziha, M., Golozar, M.A., 2017. Polyethylenimine/kappa carra-
geenan: micro-arc oxidation coating for passivation of magnesium alloy. Car-
bohydr. Polym. 167, 185e195.
Goswami, L., Kim, K.H., Deep, A., Das, P., Bhattacharya, S.S., Kumar, S.,
Adelodun, A.A., 2017. Engineered nano particles: nature, behavior, and effect on
the environment. J. Environ. Manag. 196, 297e315.
He, F., Zhao, D., 2005. Preparation and characterization of a new class of starch-
stabilized bimetallic nanoparticles for degradation of chlorinated hydrocar-
bons in water. Environ. Sci. Technol. 39 (9), 3314e3320.
He, F., Zhao, D., Paul, C., 2010. Field assessment of carboxymethyl cellulose stabi-
lized iron nanoparticles for in situ destruction of chlorinated solvents in source
zones. Water Res. 44 (7), 2360e2370.
He, Y.T., Wilson, J.T., Su, C., Wilkin, R.T., 2015. Review of abiotic degradation of
chlorinated solvents by reactive iron minerals in aquifers. Ground Water Monit.
Remed. 35 (3), 57e75.
Henn, K.W., Waddill, D.W., 2006. Utilization of nanoscale zero-valent iron for source
remediation-a case study. Remed. J. 16 (2), 57e77.
Heron, G., Bierschenk, J., Swift, R., Watson, R., Kominek, M., 2016. Thermal DNAPL
source zone treatment impact on a CVOC plume. Ground Water Monit. Remed.
36 (1), 26e37.
Hiortdahl, K.M., Borden, R.C., 2013. Enhanced reductive dechlorination of tetra-
chloroethene dense nonaqueous phase liquid with EVO and Mg(OH)2. Environ.
Sci. Technol. 48 (1), 624e631.
Huang, L., Liu, Q., Zhou, S., Xue, J., Liu, G., 2016. Laboratory investigation of a
magnetic system used for enhancing the migration of nanoscale zero-valent
iron (NZVI). J. Taiwan Inst. Chem. Eng. 61, 205e210.
Huo, L., Zeng, X., Su, S., Bai, L., Wang, Y., 2017. Enhanced removal of as (V) from
aqueous solution using modified hydrous ferric oxide nanoparticles. Sci. Rep. 7,
40765, 12 pages.
466 K.-S. Lin et al. / Journal of Cleaner Production 175 (2018) 456e466
Jamieson-Hanes, J.H., Lentz, A.M., Amos, R.T., Ptacek, C.J., Blowes, D.W., 2014. Ex-
amination of Cr (VI) treatment by zero-valent iron using in situ, real-time
Xeray absorption spectroscopy and Cr isotope measurements. Geochim. Cos-
mochim. Acta 142, 299e313.
Johnson, T.L., Fish, W., Gorby, Y.A., Tratnyek, P.G., 1998. Degradation of carbon tet-
rachloride by iron metal: complexation effects on the oxide surface. J. Contam.
Hydrol. 29 (4), 379e398.
Kaifas, D., Malleret, L., Kumar, N., Fe
timi, W., Claeys-Bruno, M., Sergent, M.,
Doumenq, P., 2014. Assessment of potential positive effects of nZVI surface
modification and concentration levels on TCE dechlorination in the presence of
competing strong oxidants, using an experimental design. Sci. Total Environ.
481, 335e342.
Kim, H.S., Ahn, J.Y., Kim, C., Lee, S., Hwang, I., 2014. Effect of anions and humic acid
on the performance of nanoscale zero-valent iron particles coated with poly-
acrylic acid. Chemosphere 113, 93e100.
Kim, H.H., Kim, M.S., Kim, H.E., Lee, H.J., Jang, M.H., Choi, J., Hwang, Y., Lee, C., 2017.
Nanoparticulate zero-valent iron coupled with polyphosphate: the sequential
redox treatment of organic compounds and its stability and bacterial toxicity.
Environ. Sci. Nano 4 (2), 396e405.
Kim, H.J., Leitch, M., Naknakorn, B., Tilton, R.D., Lowry, G.V., 2017. Effect of emplaced
nZVI mass and groundwater velocity on PCE dechlorination and hydrogen
evolution in water-saturated sand. J. Hazard Mater. 322, 136e144.
Kobayashi, S., Hiroishi, K., Tokunoh, M., Saegusa, T., 1987. Chelating properties of
linear and branched poly (ethylenimines). Macromolecules 20 (7), 1496e1500.
Ku, B., Kim, J.E., Chung, B.H., Chung, B.G., 2013. Retinoic acid-polyethyleneimine
complex nanoparticles for embryonic stem cell-derived neuronal differentia-
tion. Langmuir 29 (31), 9857e9862.
Lerner, D.N., Kueper, B.H., Wealthall, G.P., Smith, J.W.N., Leharne, S.A., 2003. An
Illustrated Handbook of DNAPL Transport and Fate in the Subsurface, pp. 1e63. .
(Accessed 5 July 2017).
Liang, Y., Min, X., Chai, L., Wang, M., Liyang, W., Pan, Q., Okido, M., 2017. Stabilization
of arsenic sludge with mechanochemically modified zero-valent iron. Chemo-
sphere 168, 1142e1151.
Lien, H.L., Zhang, W.X., 2001. Nanoscale iron particles for complete reduction of
chlorinated ethenes. Colloids Surf. A Physicochem. Eng. Asp. 191 (1), 97e105.
Lin, K.S., Chang, N.B., Chuang, T.D., 2008. Fine structure characterization of zero-
valent iron nanoparticles for decontamination of nitrites and nitrates in
wastewater and groundwater. Sci. Technol. Adv. Mater. 9 (2), 025015.
Lin, K.S., Dehvari, K., Hsien, M.J., Hsu, P.J., Kuo, H., 2013. Degradation of TNT, RDX,
and HMX explosive wastewaters using zero-valent iron nanoparticles. Propell.
Explos. Pyrot. 38 (6), 786e790.
Loraine, G.A., 2001. Effects of alcohols, anionic and nonionic surfactants on the
reduction of PCE and TCE by zero-valent iron. Water Res. 35 (6), 1453e1460.
Lubphoo, Y., Chyan, J.M., Grisdanurak, N., Liao, C.H., 2016. Influence of PdeCu on
nanoscale zero-valent iron supported for selective reduction of nitrate.
J. Taiwan Inst. Chem. Eng. 59, 285e294.
Lytle, F.W., 1999. The EXAFS family tree: a personal history of the development of
extended X-ray absorption fine structure. J. Synchrotron Rad. 6, 123e134.
Ma, B., Yu, W., Jefferson, W.A., Liu, H., Qu, J., 2015. Modification of ultrafiltration
membrane with nanoscale zero-valent iron layers for humic acid fouling
reduction. Water Res. 71, 140e149.
Ma, L., Dong, X., Chen, M., Zhu, L., Wang, C., Yang, F., Dong, Y., 2017. Fabrication and
water treatment application of carbon nanotubes (CNTs)-based composite
membranes: a review. Membranes 7 (1), 16.
Mo, L., Ye, S., Wu, J., Stahl, R.G., Grosso, N.R., Wang, J.C., 2017. Field application at a
DNAPL-contaminated site in Nanjing and discussion of a source search algo-
rithm based on stochastic modeling and kalman filter. Environ. Earth Sci. 2 (76),
1e15.
Mukherjee, R., Kumar, R., Sinha, A., Lama, Y., Saha, A.K., 2016. A review on synthesis,
characterization, and applications of nano zero-valent iron (nZVI) for environ-
mental remediation. Crit. Rev. Environ. Sci. Technol. 46 (5), 443e466.

Mueller, N.C., Braun, J., Bruns, J., Cerník, M., Rissing, P., Rickerby, D., Nowack, B.,
2012. Application of nanoscale zero-valent iron (nZVI) for groundwater reme-
diation in europe. Environ. Sci. Pollut. Res. 19 (2), 550e558.
Noubactep, C., Care
, S., Crane, R., 2012. Nanoscale metallic iron for environmental
remediation: prospects and limitations. Water Air Soil Pollut. 223 (3),
1363e1382.
Namgung, R., Singha, K., Yu, M.K., Jon, S., Kim, Y.S., Ahn, Y., Park, I.K., Kim, W.J., 2010.
Hybrid supermaramagnetic iron oxide nanoparticle-branch polyethylenimine
magnetoplexes for gene transfection of vascular endothelial cells. Biomaterials
31, 4204e4213.
Obiri-Nyarko, F., Grajales-Mesa, S.J., Malina, G., 2014. An overview of permeable
reactive barriers for in situ sustainable groundwater remediation. Chemosphere
111, 243e259.
Oprckal, P., Mladenovi c, A., Vidmar, J., Pranji

c, A.M., Milaci 
c, R., S can
car, J., 2017.
Critical evaluation of the use of different nanoscale zero-valent iron particles for
the treatment of effluent water from a small biological wastewater treatment
plant. Chem. Eng. J. 321, 20e30.
Orozco, A., Velimirovic, M., Tosco, T., Kemna, A., Sapion, H., Klaas, N., Sethi, R.,
Bastiaens, L., 2015. Monitoring the injection of microscale zero-valent iron
particles for groundwater remediation by means of complex electrical con-
ductivity imaging. Environ. Sci. Technol. 49 (9), 5593e5600.
Padmanabhan, A.R., Bergendahl, J., Sharma, S., 2008. Simultaneous reduction/
oxidation process for destroying an organic solvent. U.S. Patent Application No.
12/593,011.
Parker, J., Kim, U., Kitanidis, P., Cardiff, M., Liu, X., Beyke, G., 2012. Stochastic cost
optimization of DNAPL remediation method description and sensitivity study.
Environ. Model. Software 38, 74e88.
Perez-de-Mora, A., Zila, A., McMaster, M.L., Edwards, E.A., 2014. Bioremediation of
chlorinated ethenes in fractured bedrock and associated changes in dechlori-
nating and nondechlorinating microbial populations. Environ. Sci. Technol. 48
(10), 5770e5779.
Phenrat, T., Saleh, N., Sirk, K., Tilton, R.D., Lowry, G.V., 2007. Aggregation and
sedimentation of aqueous nanoscale zero-valent iron dispersions. Environ. Sci.
Technol. 41 (1), 284e290.
Ponder, S.M., Darab, J.G., Mallouk, T.E., 2000. Remediation of Cr (VI) and Pb (II)
aqueous solutions using supported, nanoscale zero-valent iron. Environ. Sci.
Technol. 34 (12), 2564e2569.
Power, C., Gerhard, J.I., Karaoulis, M., Tsourlos, P., Giannopoulos, A., 2014. Evaluating
four-dimensional time-lapse electrical resistivity tomography for monitoring
DNAPL source zone remediation. J. Contam. Hydrol. 162, 27e46.
Quinn, J., Geiger, C., Clausen, C., Brooks, K., Coon, C., O'Hara, S., Krug, T., Major, D.,
Yoon, W.S., Gavaskar, A., Holdsworth, T., 2005. Field demonstration of DNAPL
dehalogenation using emulsified zero-valent iron. Environ. Sci. Technol. 39 (5),
1309e1318.
Ramamurthy, A.S., Eglal, M.M., 2014. Degradation of TCE by TEOS coated nZVI in the
presence of Cu(II) for groundwater remediation. J. Nanomater. 2014, 606534, 9
pages.
Ravel, B., Newville, M.A.T.H.E.N.A., 2005. ATHENA, ARTEMIS, HEPHAESTUS: data
analysis for X-ray absorption spectroscopy using IFEFFIT. J. Synchrotron Radiat.
12 (4), 537e541.
Sheng, G., Yang, P., Tang, Y., Hu, Q., Li, H., Ren, X., Hu, B., Wang, X., Huang, Y., 2016.
New insights into the primary roles of diatomite in the enhanced sequestration
of UO2þ 2 by zero-valent iron nanoparticles: an advanced approach utilizing XPS
́ and EXAFS. Appl. Catal. B 193, 189e197.
Shi, Y., Du, J., Zhou, L., Li, X., Zhou, Y., Li, L., Zhang, X., Zhang, X., Pan, F., Zhang, H.,
Wang, Z., Zhu, X., 2012. Size-controlled preparation of magnetic iron oxide
nanocrystals within hyperbranched polymer and their magnetofection in vitro.
J. Mater. Chem. 22, 355e360.
Shin, M.C., Choi, H.D., Kim, D.H., Baek, K., 2008. Effect of surfactant on reductive
dechlorination of trichloroethylene by zero-valent iron. Desalination 223 (1e3),
299e307.
Son, Y.H., Lee, J.K., Soong, Y., Martello, D., Chyu, M.K., 2012. Heterostructured zero-
valent ironemontmorillonite nanohybrid and their catalytic efficacy. Appl. Clay
Sci. 62, 21e26.
Su, Y.F., Cheng, Y.L., Shih, Y.H., 2013. Removal of trichloroethylene by zero-valent
iron/activated carbon derived from agricultural wastes. J. Environ. Manag.
129, 361e366.
Sun, Y.P., Li, X.Q., Zhang, W.X., Wang, H.P., 2007. A method for the preparation of
stable dispersion of zero-valent iron nanoparticles. Colloids Surf. A Phys-
icochem. Eng. Asp. 308 (1), 60e66.
Tang, S., Wang, X.M., Mao, Y.Q., Zhao, Y., Yang, H.W., Xie, Y.F., 2015. Effect of dis-
solved oxygen concentration on iron efficiency: removal of three chloroacetic
acids. Water Res. 73, 342e352.
Teng, W., Fan, J., Wang, W., Bai, N., Liu, R., Liu, Y., Deng, Y., Kong, B., Yang, J., Zhao, D.,
Zhang, W.X., 2017. Nanoscale zero-valent iron in mesoporous carbon (nZVI@C):
stable nanoparticles for metal extraction and catalysis. J. Mater. Chem. A 5 (9),
4478e4485.
Torrey, J.D., Killgore, J.P., Bedford, N.M., Greenlee, L.F., 2015. Oxidation behavior of
zero-valent iron nanoparticles in mixed matrix water purification membranes.
Environ. Sci.: Water Res. Technol. 1 (2), 146e152.
Tosco, T., Papini, M.P., Viggi, C.C., Sethi, R., 2014. Nanoscale zero-valent iron particles
for groundwater remediation: a review. J. Clean. Prod. 77, 10e21.
Trellu, C., Mousset, E., Pechaud, Y., Huguenot, D., Van Hullebusch, E.D., Esposito, G.,
Oturan, M.A., 2016. Removal of hydrophobic organic pollutants from soil
washing/flushing solutions: a critical review. J. Hazard Mater. 306, 149e174.
USEPA, 2005. Nanotechnology and the Environment: Applications and Implications
Progress Review Workshop III: Applications and Implications Progress Review
Workshop III, vol. 51. United States Environmental Protection Agency (EPA),
Arlington, VA. www.epa.gov. (Accessed 5 July 2017).
Wang, Q., Jeong, S.W., Choi, H., 2012. Removal of trichloroethylene DNAPL trapped
in porous media using nanoscale zero-valent iron and bimetallic nanoparticles:
direct observation and quantification. J. Hazard Mater. 213, 299e310.
Wang, W., Li, S., Lei, H., Pan, B., Zhang, W.X., 2015. Enhanced separation of nanoscale
zero-valent iron (nZVI) using polyacrylamide: performance, characterization
and implication. Chem. Eng. J. 260, 616e622.
West, M.R., Kueper, B.H., 2012. Numerical simulation of DNAPL source zone reme-
diation with in situ chemical oxidation (ISCO). Adv. Water Resour. 44, 126e139.
Yan, J., Qian, L., Gao, W., Chen, Y., Ouyang, D., Chen, M., 2017. Enhanced fenton-like
degradation of trichloroethylene by hydrogen peroxide activated with nano-
scale zero-valent iron loaded on biochar. Sci. Rep. 7, 43051, 9 pages.
Yan, W., Lien, H.L., Koel, B.E., Zhang, W.X., 2013. Iron nanoparticles for environ-
mental clean-up: recent developments and future outlook. Environ. Sci. Process
Impacts 15 (1), 63e77.
Zhao, C., Reardon, E.J., 2012. H2 gas charging of zero-valent iron and TCE degra-
dation. J. Environ. Prot. 3 (03), 272.
Zhang, W., Li, L., Lin, K., Xiong, B., Li, B., Lu, S., Guo, M., Cui, X., 2012. Synergetic
degradation of Fe/Cu/C for groundwater polluted by trichloroethylene. Water
Sci. Technol. 65 (12), 2258e2264.