Chemosphere 138 (2015) 726–734

Contents lists available at ScienceDirect

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

The removal of chromium (VI) and lead (II) from groundwater using
sepiolite-supported nanoscale zero-valent iron (S-NZVI)
Rongbing Fu a,⇑, Yingpin Yang b, Zhen Xu a, Xian Zhang b, Xiaopin Guo a, Dongsu Bi b
a
Shanghai Academy of Environmental Sciences, Shanghai 200233, China
b
Department of Chemical Engineering, Shanghai Institute of Technology, Shanghai 201418, China

h i g h l i g h t s

 NZVI was immobilized on sepiolite to overcome agglomeration.

 Heavy metals were rapidly and extensively remediated by sepiolite-supported NZVI.
 The remediation of metals occurred in a stepwise manner (adsorption–reduction).
 Sepiolite-supported NZVI could be applied to in situ heavy metal remediation.

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

Article history: In this study, the synthesis and characterization of sepiolite-supported nanoscale zero-valent iron parti-
Received 7 February 2015 cles (S-NZVI) was investigated for the adsorption/reduction of Cr(VI) and Pb(II) ions. Nanoscale
Received in revised form 20 June 2015 zero-valent iron (NZVI) supported on sepiolite was successfully used to remove Cr(VI) and Pb(II) from
Accepted 19 July 2015
groundwater with high efficiency. The removal mechanism was proposed as a two-step interaction
including both the physical adsorption of Cr(VI) and Pb(II) on the surface or inner layers of the
sepiolite-supported NZVI particles and the subsequent reduction of Cr(VI) to Cr(III) and Pb(II) to Pb(0)
Keywords:
by NZVI. The immobilization of the NZVI particles on the surface of sepiolite could help to overcome
Sepiolite
Chromium (VI)
the disadvantage of NZVI particles, which have strong tendency to agglomerate into larger particles,
Lead (II) resulting in an adverse effect on both the effective surface area and reaction performance. The techniques
Nanoscale zero-valent iron of XRD, XPS, BET, Zeta potential, and TEM were used to characterize the S-NZVI and interaction between
Groundwater S-NZVI and heavy metals. The appropriate S-NZVI dosage was 1.6 g L1. The removal efficiency of Cr(VI)
and Pb(II) by S-NZVI was not affected to any considerable extent by the presence of co-existing ions, such
2 2+
as H2PO 4 , SiO3 , Ca and HCO3 . The Cr(VI) and Pb(II) removal kinetics followed a pseudo-first-order
rate expression, and both Langmuir isotherm model and Freundlich isotherm model were proposed.
The results suggested that supporting NZVI on sepiolite had the potential to become a promising
technique for in situ heavy metal-contaminated groundwater remediation.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction and toxic to many different aquatic organisms even at low concen-
trations (Mohan and Pittman Jr., 2006). Therefore, chromium (VI) is
Land and water pollution by heavy metals is a worldwide issue. regarded as the priority controlled pollutant in many countries
Chromium (VI) and lead (II) are two common toxic metals found in (Hizal and Apak, 2006). Pb(II) is widely used in many industrial
both soil and groundwater. Cr(VI) is an industrial contaminant in applications, such as storage battery manufacturing, painting pig-
both soil and groundwater from such industries as leather tanning, ment, fuels, photographic materials, coatings, and the automotive
electroplating, metal processing, film processing, and mining of and aeronautical industries (Lalhruaitluanga et al., 2010). This
chromium ores (Sarin et al., 2006). Chromium (VI) engenders envi- heavy metal is a highly toxic and cumulative poison that accumu-
ronmental and health problems because of its potent toxicity, lates mainly in the bones, brain, kidneys and muscles. Lead poison-
mutagenicity and carcinogenicity. It is highly soluble in water ing in humans causes severe damage to the kidneys, nervous
system, reproductive system, liver and brain (S ß ölener et al.,
2008). Therefore, it is very important to remove heavy metals
⇑ Corresponding author.
before disposal (Liu et al., 2014).
E-mail address: rongermmfu@163.com (R. Fu).

http://dx.doi.org/10.1016/j.chemosphere.2015.07.051
0045-6535/Ó 2015 Elsevier Ltd. All rights reserved.
 R. Fu et al. / Chemosphere 138 (2015) 726–734 727

Nanoscale zero-valent iron (NZVI) has been demonstrated to be dispersion of 40 g of sepiolite in 200 mL of a 4 M HCl solution
a good alternative for the remediation of a wide variety of organic and stirring the suspension for 48 h at room temperature. The solid
and inorganic pollutants in soil and groundwater, showing a high was then separated from the solution by centrifugation and
efficiency and low economic and environmental costs (Cundy washed with deionized water until it was Cl ion free. The
et al., 2008). The NZVI has high reactivity and surface energy due obtained acid-activated sample was dried at 110 °C for 2 h and cal-
to its large SSA (Sun et al., 2006), more reactive surface sites, and cinated at 300 °C for 4 h and then sieved through a 160 mesh
high reaction rate (Li et al., 2006). Nevertheless, there are still some screen prior to use.
technical challenges in NZVI applications that affect the use of this The S-NZVI with a sepiolite/iron mass ratio of 9:1 was prepared
material in in-situ remediation. For instance, the aggregation of by the following method (Fig. 1): Asp (7.5 g) was added to 150 mL
NZVI limits its mobility, dispersivity, durability, and mechanical of a Fe3+ solution (0.1 M) and stirred overnight using a mechanical
strength, and the oxidization of NZVI can significantly decrease stirrer at room temperature. One-hundred fifty millilitres of 0.4 M
its reactivity. To address these issues, technologies have attempted NaBH4 was added drop-wise into the suspension under vigorous
to develop material (kaolin (Zhang et al., 2010a), bentonite (Shi agitation and a nitrogen atmosphere. The sepiolite particles in
et al., 2011a), montmorillonite (Zhang et al., 2013), zeolite (Kim the solution turned black immediately and the mixture was
et al., 2013), and rectorite (Luo et al., 2013) ), attapulgite sequentially stirred for 0.5 h. The solid was separated from the sus-
(Guixiang Quan et al., 2014) to support NZVI to decrease aggrega- pension by vacuum filtration and washed several times with
tion and improve its dispersivity. In addition, (Fei et al., 2012) had deionized water followed by absolute ethyl alcohol. Finally,
reported the use of sepiolite as support of NZVI to degrade S-NZVI was dried under vacuum at 60 °C overnight. The deoxy-
bromamine acid. genated water used in the experiment was boiled for 0.5 h and
Sepiolite is a clay mineral with a wide range of applications then purged with nitrogen for another 0.5 h to remove dissolved
based on its physicochemical properties and especially surface oxygen.
properties. It is chemically stable, inexpensive and an ideal support Conventionally synthesized NZVI was prepared by the Fe3+
material for NZVI (Suarez and García-Romero, 2012). It is a hydrous reduction method. Briefly, 1.6 M freshly prepared NaBH4 aqueous
magnesium silicate (Si12O30Mg8(OH)4(H2O)48H2O) characterized solution was carefully added drop-wise to the 0.4 M FeCl36H2O
by its fibrous morphology that is due to its crystalline structure solution at room temperature with continuously stirring under a
(Marjanović et al., 2013). It presents a structure of needle-like par- gentle stream of nitrogen (Shi et al., 2011a).
ticles and has talc-like layers that consist of two layers of tetrahe-
dral silica and a central octahedral magnesium layer (Lazarević 2.3. Characterizations
et al., 2012). Due to its sorptive, rheological and catalytic proper-
ties, sepiolite is widely used in a variety of industrial applications The morphological analysis of NZVI, S-NZVI and sepiolite was
(Lazarević et al., 2012). However, its relatively low surface acidity, performed using a transmission electron microscope (TEM,
narrow channels, small surface area, and poor thermal stability TECNAI G2 F30, America FEI, USA). The X-ray diffraction (XRD)
seriously limit applications of the natural sepiolite (Liang, 2008). patterns of sepiolite, S-NZVI, and NZVI were determined using an
Therefore, it is essential to modify the sepiolite before synthesizing X-ray diffractometer (PANalytical-X’Pert PW3040/60, the
sepiolite–supported materials. Netherlands) with a high-power Cu Ka radioactive source (40 kV,
In this study, we investigated the feasibility of using 40 mA), and X-ray photoelectron spectroscopy (XPS, PHI5000C
sepiolite-supported nanoscale zero-valent iron particles (S-NZVI) ESCA System) analysis was carried out to characterize the S-NZVI
to remove Cr(VI) and Pb(II) from groundwater. We developed a particles before and after exposure to Cr(VI) and Pb(II). All of the
two-step reduction technique to successfully synthesize S-NZVI samples were scanned from 3° to 90° (2h) and The 2h-scanning rate
particles at normal pressure and temperature. The synthesized was 0.2089°s1. The specific surface areas (SSA) of sepiolite, Asp,
material before and after reaction was characterized by X-ray NZVI, and S-NZVI were determined by the Brunauer–Emmet–Tell
diffraction (XRD), a BET-N2 adsorption technique, transmission er (BET) technique with an accelerated surface area and porosime-
electron microscopy (TEM), Zeta potential analysis and X-ray try analyser (Micromeritics’ ASAP 2020 Instrument Crop., USA).
photoelectron spectroscopy (XPS). The reaction mechanism was The zeta potentials were measured by a Zetasizer Nano zeta poten-
studied and proposed. In addition, the effects of pH, S-NZVI dosage, tial analyser (Malvern Instruments, UK, motion range: ±10 lm/vs
initial heavy metal concentration and coexisting ions on Cr(VI) and conductivity Range: 0–200 ms/cm, minimum volume: 0.75 mL,
Pb(II) removal were also evaluated. temperature: 2–90 °C).

2. Materials and methods 2.4. Batch experiments

2.1. Materials and chemicals The concentration of Pb(II) in solution was determined using an
inductively coupled plasma emission spectrometer ICP (Optima
Iron (III) chloride hexahydrate (FeCl36H2O), sodium borohy- 7000DV, Perkin–Elmer, USA). The concentration of Cr(VI) was mea-
dride (NaBH4), sodium hydroxide (NaOH), hydrochloric acid sured by the 1,5-diphenylcarbazide method at a wavelength of
(HCl), sulphuric acid (H2SO4), phosphoric acid (H3PO4), nitric acid 540 nm on a UV–VIS spectrophotometer (TU-1810, Beijing
(HNO3), calcium chloride (CaCl2), sodium dihydrogen General Instrument Co., Ltd. China).
phosphate (NaH2PO4), sodium bicarbonate (NaHCO3), sodium All of the experiments were conducted in duplicate. Each batch
metasilicate (Na2SiO3), potassium dichromate (K2Cr2O7), and lead reaction took place in a 50 mL centrifuge tube containing a given
nitrate (Pb(NO3)2) were purchased from the Sinopharm Chemical amount of S-NZVI and 25 mL of a solution with a certain concen-
Reagent Co., Ltd. (Shanghai, China). All chemicals were of analytical tration of Cr(VI) and Pb(II) solution. These tubes were placed in a
reagent grade and used without further purification. steam-bath vibrator (Model ZD-85, China) at a speed of 200 rpm
at 28 °C. After a given time interval, several vials were taken out
2.2. Preparation of NZVI and S-NZVI of the vibrator. The initial pH values of the suspensions were
adjusted using 0.1 M HCl or NaOH. The particles and an aqueous
Sepiolite was purchased from Nanyang Boxing Mining Co., Ltd. phase were separated by centrifugation at 8000 rpm for 3 min.
(Hebei, PRC). Acid-activated sepiolite (Asp) was prepared by the The supernatant samples were filtered through a 0.22-lm
728 R. Fu et al. / Chemosphere 138 (2015) 726–734

Fig. 1. Schematic illustration of the immobilization of nanoscale zero-valent iron particles on sepiolite (a). A proposed mechanism for Cr (VI) and Pb(II) remediation by S-NZVI
(b).

membrane filter. The comparison tests of acid-activated sepiolite, the conventionally synthesized NZVI particles (Fig. 2a) appeared
NZVI and S-NZVI were also conducted. Water to prepare the solu- to be spherical in the size range of 20–60 nm. They exist as promi-
tion was boiled for 0.5 h and subsequently purged with N2 for 0.5 h nent chain-like aggregates with lengths of 50 nm–2 lm because of
to remove dissolved oxygen. their high surface energies and magnetic interactions. The NZVI
particles (D = 10–50 nm) immobilized on sepiolite were clearly
2.5. Adsorption isotherms discrete and well dispersed on the sepiolite support that itself
had a needle-like morphology (Fig. S1) without aggregation
Adsorption equilibrium of S-NZVI with Cr(VI) and Pb(II) can be (Fig. 2b). Therefore, the nanoparticles of iron were in the form of
described by Langmuir and Freundlich. The Langmuir isotherms spherical particles, which were very different from conventionally
expressed in Eq. (1) and the Freundlich isotherms is given by synthesized NZVI. Other materials, such as resin, starch, and zeo-
Eq. (2). lite, were also useful in enhancing the dispersibility of NZVI parti-
cles (Shu et al., 2010; Shi et al., 2011a; Kim et al., 2013; Mosaferi
Ct 1 Ct et al., 2014) because of a decrease in aggregation and an increase
¼ þ ð1Þ
Q t Q m kL Q m in mechanical strength.
lgCt Fig. 2b and c show the changes in the morphology of S-NZVI
lgQ t ¼ þlgkF ð2Þ before and after reaction. Before reaction (Fig. 2b), NZVI particles
n
were distributed uniformly on the surface of sepiolite with an ele-
where Qt is the amount of adsorbate per mass of adsorbent at equi- mental composition of Fe, O, Si (Fig. 2e). However, after S-NZVI
librium (mg g1), Ct is the equilibrium concentration of adsorbate in reacted with Cr(VI) and Pb(II), as shown in Fig. 2c and d, flocculent
aqueous solution (mg L1), Qm is the monolayer adsorption capacity clusters with diffuse edges formed on the surface of reacted S-NZVI
at equilibrium (mg g1), kL is the Langmuir equilibrium constant composites indicating the presence of elemental Pb and Cr and
(L mg1), kF is a Freundlich constant (index of adsorption capacity), their oxides via redox reactions. This changed the morphology of
and n is the Freundlich constant (index of adsorption intensity or S-NZVI (Xi et al., 2010). This could be confirmed by the elemental
surface heterogeneity). distribution of S-NZVI before and after reaction with Cr(VI) and
Pb(II) using SEM-EDX (Fig. 2e, f and g). The EDX result showed that
3. Results and discussion S-NZVI before reaction mainly consisted of Fe, O, and Si. The O and
Si were from sepiolite. The Fe came from NZVI (Su et al., 2011).
3.1. Characterization and interaction mechanism After reaction with Cr(VI) and Pb(II) (Fig. 2f and g), Cr and Pb were
introduced, respectively. In addition, the Fe content in S-NZVI
The removal mechanism of Cr(VI) and Pb(II) by S-NZVI is dropped from 63.88% to 56.82% and 42.71% (w:w), while the O con-
outlined in Fig. 1. A two-step interaction mechanism is proposed tent increased from 15.26% to 26.61% and 32.34% (w:w) compared
(Li et al., 2012). First, Cr(VI) and Pb(II) in aqueous solution with the blank S-NZVI. This could be explained by the oxidation of
were absorbed on the S-NZVI surface. Second, Cr(VI) and Pb(II) Fe0 as a reductant to form the Fe2+ and Fe3+ iron oxide or hydroxide
were reduced by Fe0 and co-precipitated or precipitated on the sur- in the S-NZVI (Su et al., 2011).
face of S-NZVI. Cr(VI) was reduced to the less toxic Cr(III) by S-NZVI The XRD patterns of sepiolite, Asp and S-NZVI before and after
and immobilized by precipitation as Cr(OH)3 or by incorporation reaction with Cr(VI) and Pb(II) are illustrated in Fig. 3. The typical
into the iron oxide/iron hydroxide shell forming alloy-like Cr3+– peak of MgO (2h at 43.277°, d = 0.20889 nm) became too weak to
Fe3+ hydroxides (Ponder et al., 2000; Shi et al., 2011a), but the for- be visible in Asp, and other typical peaks of sepiolite remained,
mation of these Cr3+–Fe3+ hydroxides may inhibit electron transfer which indicated that Mg2+ was replaced by H+ under acidic condi-
from the Fe0 to the surface, thus stopping the redox process and tions. Before reaction, a diffraction peak at 44.67° (d = 0.22077 nm)
causing a decrease of Cr(VI) removal, especially at high Cr(VI) con- (Fig. 3c) corresponding to a formation of zero-valent iron (Fe0)
centrations (Hu et al., 2010; O’Carroll et al., 2013). could be observed, indicating that Fe0 nanoparticles were incorpo-
The surface morphology of the sepiolite, NZVI and S-NZVI that rated with Asp. After activation of sepiolite particles by HCl solu-
were analysed by TEM is presented in Fig. 2. It was observed that tion their crystalline structure was significantly decreased and
 R. Fu et al. / Chemosphere 138 (2015) 726–734 729

Fig. 3. The X-ray diffractograms of (a) sepiolite; (b) Asp; (c) S-NZVI; (d) S-NZVI after
reaction with Pb(II) and (e) S-NZVI after reaction with Cr(VI).

The SSA of sepiolite, Asp, NZVI and S-NZVI were measured by

the BET-N2-adsorption–desorption method. Certain amounts of
structural Mg2+ ions were removed (Lazarević et al., 2007), and
the SSA and total pore volume increased to 230.37 m2 g1 and
0.29 cm3 g1 (Table S1) after an acidic treatment of sepiolite.
Previous studies also confirmed that acidic treatment increased
the surface area of sepiolite. Solids were obtained that had high
porosity and a high number of acidic centres (Lazarević et al.,
2007).
When NZVI was loaded onto sepiolite, a mean SSA of
141.42 m2 g1 was obtained, which was much higher than that of
NZVI (47.74 m2 g1) and other results (Table S2). Compared to
NZVI, the increased SSA of S-NZVI was due to the
non-aggregation of NZVI particles (Kim et al., 2013). The TEM
image of S-NZVI (Fig. 2c) showed that the NZVI particles were dis-
persed onto the surface of sepiolite, which indicated that sepiolite
played a role in dispersing and stabilizing NZVI and that the sup-
ported nanoparticles might exhibit higher reaction rates with con-
taminants compared to unsupported particles. This was confirmed
by previous studies (Fei et al., 2012). Arshadi et al. synthesized
S-NaOH-NZVI with a lower SSA but higher reaction activity than
that of other synthesized iron nanoparticles (Arshadi et al.,
2014). The supported NZVI was immobilized on the active sites
of sepiolite, which would decrease the oxidation of the NZVI
Fig. 2. The TEM image of NZVI (a), S-NZVI (b), S-NZVI after reaction with Cr(VI) (c) surface (Arshadi et al., 2014).
and (d) the TEM image of S-NZVI after reaction with Pb(II); the EDX survey for S- Generally, particles with a zeta potential from 15 mV to
NZVI (e), S-NZVI after reaction with Cr(VI) (f) and (g) EDX survey for S-NZVI after
+15 mV are expected to be unstable from electrostatic considera-
reaction with Pb(II).
tions (Hiemenz, 1977). The zeta potentials of sepiolite, Asp, NZVI
and S-NZVI were 26.1 mV, 22.7 mV, +14 mV and 44 mV under
did not affect the performance of NZVI particles. Both the EDX and neutral condition (Fig. S2), respectively, indicating that the S-NZVI
the XRD analysis demonstrate that the metallic iron products have was most stable among these materials. After loading NZVI on Asp,
been obtained and the iron nanostructures are very stable under the zeta potential changed to 44.0 mV. However, the zeta poten-
ambient conditions (Tong et al., 2008)_ENREF_21. However, after tial of S-NZVI after reaction with Cr(VI) (Fig. S2e) and Pb(II)
S-NZVI reacted with Cr(VI) (Fig. 3d) and Pb(II) (Fig. 3e), the typical (Fig. S2f) were +0.76 mV and +7.9 mV, respectively, because the
peak of Fe0 (2h at 44.67°) became too weak to be visible, while the surface charge of the S-NZVI was dominated largely by Cr(III)
peaks of Fe2O3 (2h at 35.6° and 41.2°), FeO(OH) (2h at 29.5°), and Pb(II) owing to the negative surface charges of S-NZVI which
Fe3O4 (2h at 35.4°/56.6°/62.2°), Cr2FeO4, Cr2O3 (2h at 33.4°, attracted the cations (Xu et al., 2013).
d = 0.26806 nm and 42.699°, d = 0.21158 nm) and Pb0, PbO2, The results of the XPS characterization are shown in Fig. 4. The
PbOxH2O (2h at 42.537°, d = 0.21235 nm and 32.723°, new peaks at the binding energy (BE) of 580 eV and 140 eV appear-
d = 0.27244 nm) appeared more obvious. These observations ing after the reduction of the chromium and lead (Fig. 4a) were not
demonstrated the adsorption or co-precipitation of the products obvious, which might be due to a part of heavy metal adsorbed by
on the surface of the S-NZVI particles. These products resulted sepiolite entering the internal channels of sepiolite. The intense
from the redox reaction between Fe0, Fe2+ or H+ and Cr(VI) or signal C 1s might be due to the fiber of filter paper or carbon diox-
Pb(II) where the supported NZVI acted as a reductant in aqueous ide contamination from the air, water or other containers during
solution. the sample preparation and transfer (Li et al., 2008).
730 R. Fu et al. / Chemosphere 138 (2015) 726–734

Fig. 4. Typical wide scan XPS spectra for S-NZVI before and after Cr(VI) and Pb(II) reactions (a), high-resolution XPS survey of iron 2p (b), oxygen 1s (c), chromium 2p (d) and
lead 4f (e).

Detailed XPS surveys of the region of Fe 2p, O 1s, Cr 2p and Pb 4f Pb(0) was observed on the particle surface with a peak at
are presented in Fig. 4b, c and d. In Fig. 4b, the photoelectron peaks 136.6 eV, and these values are only a little higher than those
approximately 711–724 eV corresponded to the binding energies obtained in a previous study (Xi et al., 2010). It can be concluded
of 2p3/2 and 2p1/2 of Fe2+ and Fe3+. It is very difficult to detect the that the presence of a portion of Cr(III) and Pb(0) on the surface
presence of Fe° and it seems to be background noise. The photo- of the S-NZVI can be solely attributed to the reduction by Fe0 but
electron peak for the oxygen 1s region resolved into three curves the generation of Fe(III) can be caused by the synergetic oxidation
with peaks of approximately 529, 531 and 532 eV, which repre- of Cr(VI), Pb(II) and by a simultaneous redox reaction between
sented the binding energies of oxygen in OH, O2 and chemically water and O2 in the experimental system (Geng et al., 2009).
or physically adsorbed water (H2O) (Li and Zhang, 2007; Lv et al.,
2012), respectively. In the Fig. 4c, there was a significant shift of 3.2. Effect of pH, S-NZVI dosage, initial concentration and co-existing
peak positions before and after the S-NZVI reacted with Cr(VI) ions
and Pb(II). The Cr(VI) and Pb(II) sorption was accompanied by a
change in oxygen binding, providing evidence that the The pH value, S-NZVI dosage, initial concentration and
oxygen-containing functional groups on the surface of the S-NZVI co-existing ions are important factors affecting the removal effi-
take part in the sorption of Cr(VI) and Pb(II) (Xu et al., 2008). The ciency of Cr(VI) and Pb(II) (Fig. 5).
oxygen peak shifted obviously due to the pellets of Cr(OH)3 and The pH value of natural groundwater commonly ranges from
Pb(OH)2 begin to occur and cover the S-NZVI surface. In 5.0 to 9.0. We chose 4.0–9.0 to assess the pH effect on the removal
Fig. 4d and e, the photoelectron peaks for chromium 2p3/2, 2p1/2, of Cr(VI) and Pb(II) by S-NZVI, and the result is shown in
and lead 4f5/2, 4f7/2 centres at 580.1 eV, 589.8 eV and 146.1 eV, Fig. 5a and b. The removal efficiency of Cr (VI) decreased with
141.3 eV after 60 min of treatment, respectively. In Fig. 4d, the Cr the increase in pH (Fig. 5a). From pH 4.0 to 9.0, the removal rate
2p3/2 region could be decomposed into two peaks at 576.6 and decreased from 98.5% to 30.7%, which indicated that the removal
577.6 eV, implying that Cr(III) and Cr(VI) coexisted on these of Cr(VI) by NZVI was strongly pH-dependent, and an acidic envi-
extracted particles. The binding energies of the Cr 2p peaks are ronment favoured Cr(VI) removal by the S-NZVI. Fig. 5b showed
in good agreement with those reported for FeCr2O4 and Cr2O3 or the effect of pH on Pb(II) removal by the S-NZVI. It was observed
Cr(OH)3 (Nahuel Montesinos et al., 2014). This result demonstrated that by increasing the pH from 4.0 to 9.0, the Pb(II) sorption
that both adsorption of Cr(VI) and reduction of Cr(VI) on the sur- decreased from 99.9% to 80.3%. With an increase of pH from 4.0
face of the S-NZVI occurred simultaneously (Geng et al., 2009). to 9.0, the pH value did not significantly affect the Pb(II) removal.
As shown in Fig. 4e, Pb 4f dominated with two peaks at 136 eV Both the removal of Cr(VI) and Pb(II) decreased with an increase in
and 141 eV, which can be assigned to Pb(0) and Pb(II), respectively, pH. The kobs (Table 1) of Cr(VI) and Pb(II) indicated that the highest
according to previous studies (Huang et al., 2013). In this study, kobs of Cr(VI) and Pb(II) were 0.754 min1 and 1.26 min1,
 R. Fu et al. / Chemosphere 138 (2015) 726–734 731

Fig. 5. Effect of different conditions on Cr(VI) and Pb(II) removal by S-NZVI; (a), (b) effect of pH; (c), (d) effect of S-NZVI dosage; (e), (f) effect of the initial concentration; (g),
(h) effects of co-existing ions.
732 R. Fu et al. / Chemosphere 138 (2015) 726–734

Table 1
Pseudo-first-order adsorption kinetics constants for Cr(VI), Pb(II) removal by S-NZVI composites under different pH, S-NZVI dosage and initial concentration of Cr(VI) and Pb(II)
(S-NZVI = 1.6 g L1, t = 60 min).

pH
4.0 5.0 6.0 7.0 8.0 9.0
Cr kobs (min1) 0.652 0.711 0.754 0.022 0.005 0.004
R2 0.93 0.98 0.98 0.98 0.97 0.93
Pb kobs (min1) 0.11 0.165 1.26 0.026 0.015 0.006
R2 0.98 0.99 0.99 0.97 0.94 0.98
S-NZVI Dosage (g L1)
0.05 0.1 0.4 0.8 1.6 3.2
Cr kobs (min1) 0.002 0.009 0.036 0.054 0.52 0.56
R2 0.90 0.91 0.91 0.91 0.94 0.92
Pb kobs (min1) 0.004 0.008 0.013 0.028 0.076 0.082
R2 0.94 0.95 0.98 0.99 0.98 0.96
Initial concentration (mg L1)
20 40 50 60 70 80 100 120
1
Cr kobs (min ) 0.082 0.29 0.094 0.0042 0.012 0.0048 0.0035 0.006
R2 0.97 0.86 0.98 091 0.99 0.90 0.95 0.96
Pb kobs (min1) 1.82 1.91 1.32 1.27 1.05 0.014 0.016 0.008
R2 0.93 0.99 0.99 0.99 0.99 0.99 0.99 0.99

respectively, at pH 6.0. In Fe0 treatment systems, the removal was not affected significantly (Cr(VI) 0.52–0.56 min1; Pb(II)
mechanisms of Cr(VI) and Pb(II) are generally believed to involve 0.076–0.082 min1). Therefore, it is not necessary to use a much
the adsorption of Cr(VI) and Pb(II) on the iron surface where elec- higher S-NZVI dosage. A dosage of 1.6 g L1 S-NZVI was chosen
tron transfer takes place, and then, Cr(VI) is reduced to Cr(III) and as the optimal dosage.
Pb(II) is reduced to Pb(0) with the oxidation of Fe0 to Fe3+ or Fe2+ The effect of the initial concentration of Cr(VI) and Pb(II) on
under acidic conditions (O’Carroll et al., 2013). At the same time, heavy metal (Cr(VI), Pb(II)) removal is shown in Fig. 5e and f. It
precipitates formed by iron hydroxides alone or co-precipitates can be observed that the removal rate of Cr(VI) and Pb(II)
of chromium (or lead) have been observed by other researchers decreased with an increase in the initial concentrations of Cr(VI)
at higher pH values (Powell et al., 1995). In the range of pH and Pb(II). The pseudo-first-order model provided the best fit
4.0–6.0, the removal of Pb and Cr remains constant and reaches a based on the experimental data as shown in Table 1. The removal
maximum. The removal of Pb(II) and Cr(VI) is mainly accomplished kobs of Cr(VI) and Pb(II) decreased with the increase of the initial
by sorption reactions (Xu et al., 2008). At high pH values concentration of Cr(VI) and Pb(II). It has been reported that both
(pH 7.0–9.0), the adsorption of Cr(VI) was decreased due to electro- Cr(VI) and Pb(II) are oxidants and passivators of Fe0; as more
static repulsion caused by the negative NZVI surface charge above Cr(VI) or Pb(II) approached Fe0, more Fe0 would be oxidized and
the pH of 7.8 (O’Carroll et al., 2013). The higher the valence of would lose its activity leading to a decrease in the removal rate.
adsorbed anions, the more negative the surface becomes, hence A precipitate forming on the surface of the NZVI would reduce
the inhibition of the further adsorption of anions (Hu et al., the electron transfer from the NZVI to Cr(VI) and Pb(II) and would
2009). The decrease of Pb(II) sorption on S-NZVI was attributed accordingly retard the reduction of Cr(VI) or Pb(II). Furthermore,
to the increase in the hydroxyl(–OH) concentration, which leads for a fixed adsorbent dosage, the total available active sites are lim-
to a negatively charged S-NZVI surface and caused precipitation ited, thus leading to a decrease in the percentage removal of adsor-
of Pb(OH)2 on the S-NZVI surface as well (Zhang et al., 2010a). In bate corresponding to an increased initial adsorbate concentration.
addition, the Fe (oxy) hydroxide coating on the NZVI surface at a The higher removal efficiency of the S-NZVI at low Cr(VI) and Pb(II)
pH above 7.0 would also occupy the active sites on the NZVI initial concentrations could be related to the high ratio of initial
surfaces and decrease reactivity. mole numbers of Cr(VI) and Pb(II) to the available active sites on
The impact of the S-NZVI dosage (0.05–3.2 g L1) on the the surface area. Therefore, the Cr(VI) and Pb(II) removal is depen-
removal reaction was determined (Fig. 5c and d). The removals dent on the initial concentration.
of Cr(VI) and Pb(II) of 98.8% and 97.31% were obtained within There are various types of co-existing ions in the groundwater,
10 min when the S-NZVI addition was 3.2 g L1. As the S-NZVI such as calcium, bicarbonate, phosphate and silicate (Lv et al.,
dosage increased from 0.05 g L1 to 3.2 g L1, the removal effi- 2013). As shown in Fig. 5g and h, the effect of the common
ciency of Cr(VI) increased significantly from 45.1% to 99.2% and co-existing ions on Cr(VI) and Pb(II) removal were investigated.
that of Pb(II) increased from 56.2% to 99.9%, respectively. From figure, it is evident that the Cr(VI) and Pb(II) removal effi-
Meanwhile, kobs (Table 1) increased as the S-NZVI dosage ciency of S-NZVI was not obviously affected by the presence of
increased. In light of the previously published data (Shu et al., co-existing ions. However, the existence of HCO 3 did slightly affect
2007; Chen et al., 2011; Shih et al., 2011), the removal efficiency Cr(VI) removal by the S-NZVI; the removal efficiency decreased
or rate constants increase with the increase of material addition. from 99.80% to 88.18%. According to the principle of oxidation
According to the results, the highest removal capacity for Cr(VI) and mass balance, the reaction was conducive to the formation
and Pb(II) were 609.5 mg Cr/g Fe and 756.7 mg Pb/g Fe which of Cr(III). However, the flocculent precipitate contributed to the
was much higher than a few reports (25–100 mg Cr/g Fe (Shi formation of a precipitate and Cr(III) decreased in the reduction
et al., 2011b), 50–180 mg Cr/g Fe (Li et al., 2008), 344.8 mg Pb/g reaction when HCO 3 was in the system. The competition between
Fe (the highest removal capacity) (Kim et al., 2013), 241.75 mg HCO 3 and Cr(VI) for active sites on the S-NZVI affected the Cr(VI)
Pb/g Fe (Zhang et al., 2010b)). However, as the dosage of S-NZVI removal; the HCO 3 could form inner-sphere surface complexes
increased from 1.6 g L1 to 3.2 g L1, the kobs of Cr(VI) and Pb(II) with iron (oxy)hydroxides and reduce the adsorption capacity
 R. Fu et al. / Chemosphere 138 (2015) 726–734 733

(Giasuddin et al., 2007; Tanboonchuy et al., 2012) as 4. Conclusions

well_ENREF_26. Moreover, the Fe2+ that is generated could react
with HCO 3 to form FeCO3, and it would drop out of solution as a In this study, S-NZVI was successfully prepared with a high
precipitate due to its limited solubility (Heuer and Stubbins, capacity for and efficiency of removal of Cr(VI) and Pb(II) from
1999). Besides, with the increasing of HCO 3 concentration, the groundwater. The NZVI particles with a diameter of 10–50 nm
hydrolyzation reaction of metal aggravated (Du et al., 2008) and were distributed uniformly on the surface of sepiolite, demonstrat-
the pH increased, which could inhibit electron transfer in the ing that modified sepiolite was effective in preventing NZVI parti-
system. Therefore the removal efficiency was affected. cles from agglomerating. The removal rates of Cr(VI) and Pb(II)
decreased with the increasing initial pH value and initial concen-
3.3. Comparison of Cr(VI) and Pb(II) removal by Asp, NZVI and S-NZVI tration of Cr(VI) and Pb(II). The removal efficiency of Cr(VI) and
Pb(II) by S-NZVI was not affected to considerable extent by the
The comparison of Cr(VI) and Pb(II) removal by sepiolite, acid presence of co-existing ions. Optimal removal occurred at pH 6.0
activated sepiolite, NZVI and S-NZVI was show in Fig. S3. The and a S-NZVI dosage of 1.6 g L1. The adsorption equilibrium data
removal efficiency of Cr(VI) and Pb(II) from 25 mL aqueous solu- for both heavy metals were well fitted by both Langmuir and
tion containing an initial concentration of 20 mg L1 using acid Freundlich isotherm models. The kinetic data were closely fitted
activated sepiolite (0.09 g), NZVI (0.01 g) and S-NZVI (0.1 g) after to the pseudo-first-order model. The results of the XRD, XPS, BET,
1 h incubation at pH 6.0. In Fig. S3, S-NZVI displayed the highest Zeta potential, and TEM techniques indicated that the removal
Cr(VI) and Pb(II) removal efficiency in the solution. The removal mechanism was a two-step interaction, including both physical
efficiency of Cr(VI) and Pb(II) were 8.147% and 26.713% by acid adsorption of Cr(VI) and Pb(II) on the surface or internal layers of
activated sepiolite which was lower than that of Cr(VI) and S-NZVI particles and the subsequent reduction of Cr(VI) to Cr(III)
Pb(II). This indicated that the removal of Cr(VI) and Pb(II) from and Pb(II) to Pb(0) by the S-NZVI. The results indicated that
solution using S-NZVI was much higher than that of acid activated S-NZVI was effective in the removal of various heavy metals from
sepiolite and NZVI. groundwater, in terms of high efficiency, stability, and reactivity.
Such a material can be a potential candidate for the in situ remedi-
3.4. Adsorption isotherm analysis ation of heavy metals.

The equilibrium isotherms for the adsorption of Cr(VI) and

Acknowledgments
Pb(II) ions by the S-NZVI are presented in Table 2. It can be seen
that the adsorption capacity increases with an increase in the equi-
This research was supported by the National Nature Science
librium metal ion concentration for both metal ions in the solution.
Foundation of China (Grant No. 41372262), the National
The equilibrium data are fitted by Langmuir and Freundlich
Nature Science Foundation of China (Grant No. 41301344),
adsorption isotherms. The isotherm parameters and related corre-
the Nature Science Foundation of Shanghai (Grant No.
lation coefficients (R2) were calculated from the slope. As displayed
13ZR1435100), the Talent Development Fund Project of Shanghai
in Table 2, the results indicated that the adsorption isotherm data
(Grant No. 201325), and the Shanghai Environmental Protection
of both metals were well fitted to both the Langmuir isotherm
Scientific Research Grand Project (Huhuanke2013-04).
model and the Freundlich model. Based on the Langmuir iso-
therms, the maximum adsorption capacity of the S-NZVI toward
Cr(VI) and Pb(II) is 43.86 mg g1 (0.84 mmol g1) and 44.05 mg g1 Appendix A. Supplementary material
(0.16 mmol g1), respectively. This is higher than the adsorption
capacity of most reported adsorbents: activated alumina (Cr(VI): Supplementary data associated with this article can be found, in
25.57 mg g1) (Bhattacharya et al., 2008), iron oxide nanoparticles the online version, at http://dx.doi.org/10.1016/j.chemosphere.
(Cr(VI): 34.1 mg g1; Pb(II) 36.0 mg g1) (Nassar, 2010; Kim et al., 2015.07.051.
2012), oxidized multiwalled carbon nanotubes (Pb(II):
9.92  103 mmol g1) (Xu et al., 2008), granular activated carbon References
(Cr(VI): 3.9 mg g1; Pb(II) 21.5 mg g1) (Rivera-Utrilla et al., 2003),
montmorillonite (Pb(II): 33 mg g1) (Bhattacharyya and Gupta, Arshadi, M., Soleymanzadeh, M., Salvacion, J., SalimiVahid, F., 2014. Nanoscale Zero-
2007). The higher R2 of the Freundlich model implied a greater ten- Valent Iron (NZVI) supported on sineguelas waste for Pb (II) removal from
aqueous solution: kinetics, thermodynamic and mechanism. J. Colloid Interface
dency of the heterogeneous surface of S-NZVI (Lv et al., 2012) for
Sci. 426, 241–251.
the removal of Cr(VI) and Pb(II). The low value of 1/n (<0.36) in Badruddoza, A.Z.M., Shawon, Z.B.Z., Rahman, M.T., Hao, K.W., Hidajat, K., Uddin,
the Freundlich isotherm suggested that any large change in the M.S., 2013. Ionically modified magnetic nanomaterials for arsenic and
equilibrium concentration of chromium and lead would not result chromium removal from water. Chem. Eng. J. 225, 607–615.
Bhattacharya, A.K., Naiya, T.K., Mandal, S.N., Das, S.K., 2008. Adsorption, kinetics and
in a significant change in the amount of chromium and lead equilibrium studies on removal of Cr(Vi) from aqueous solutions using different
adsorbed by S-NZVI (Nguyen et al., 2010). The Freundlich constant low-cost adsorbents. Chem. Eng. J. 137, 529–541.
n is found to be greater than 1 which indicates a favourable Bhattacharyya, K.G., Gupta, S.S., 2007. Adsorptive accumulation of Cd(II), Co(II),
Cu(II), Pb(II), and Ni(II) from water on montmorillonite: influence of acid
condition for adsorption (Badruddoza et al., 2013). activation. J. Colloid Interface Sci. 310, 411–424.
Chen, Z.X., Jin, X.Y., Chen, Z., Megharaj, M., Naidu, R., 2011. Removal of methyl
Table 2 orange from aqueous solution using bentonite-supported nanoscale zero-valent
iron. J. Colloid Interface Sci. 363, 601–607.
Langmuir and Freundlich isotherm parameters for Cr(VI) and Pb(II) removal by
Cundy, A.B., Hopkinson, L., Whitby, R.L., 2008. Use of iron-based technologies in
S-NZVI.
contaminated land and groundwater remediation: a review. Sci. Total Environ.
Langmuir isotherm Freundlich isotherm 400, 42–51.
Du, C.W., Li, X.G., Chen, X., Liang, P., Guo, H., 2008. Crevice corrosion behavior of X70
Cr(VI) kL (L mg1) 0.052 kF 2.61 steel in HCO 61 3 solution under cathodic polarization. Acta Metall. Sin. 21,
Qm (mg g1) 43.86 1/n 0.31 235–244.
R2 0.9523 R2 0.9772 Fei, X., Cao, L., Zhou, L., Gu, Y., Wang, X., 2012. Degradation of bromamine acid by
Pb(II) kL (L mg1) 0.073 kF 3.76 nanoscale zero-valent iron (nZVI) supported on sepiolite. Water Sci. Technol.
Qm (mg g1) 44.05 1/n 0.34 66, 2539–2545.
R2 0.9504 R2 0.9575 Geng, B., Jin, Z., Li, T., Qi, X., 2009. Kinetics of hexavalent chromium removal from
water by chitosan-Fe0 nanoparticles. Chemosphere 75, 825–830.
734 R. Fu et al. / Chemosphere 138 (2015) 726–734

Giasuddin, A.B., Kanel, S.R., Choi, H., 2007. Adsorption of humic acid onto nanoscale Nahuel Montesinos, V., Quici, N., Beatriz Halac, E., Leyva, A.G., Custo, G., Bengio, S.,
zerovalent iron and its effect on arsenic removal. Environ. Sci. Technol. 41, Zampieri, G., Litter, M.I., 2014. Highly efficient removal of Cr (VI) from water
2022–2027. with nanoparticulated zerovalent iron: Understanding the Fe (III)–Cr (III)
Quan, Guixiang, Zhang, Jing, Guo, Jing, Lan, Y., 2014. Removal of Cr(VI) from aqueous passive outer layer structure. Chem. Eng. J. 244, 569–575.
solution by nanoscale zero-valent iron grafted on acid-activated attapulgite. Nguyen, T.V., Vigneswaran, S., Ngo, H.H., Kandasamy, J., 2010. Arsenic removal by
Water, Air, Soil Pollut. 225, 1–11. iron oxide coated sponge: experimental performance and mathematical
Heuer, J., Stubbins, J., 1999. An XPS characterization of FeCO3 films from CO2 models. J. Hazard. Mater. 182, 723–729.
corrosion. Corros. Sci. 41, 1231–1243. Nassar, N.N., 2010. Rapid removal and recovery of Pb(II) from wastewater by
Hiemenz, P.C., 1977. Principles of colloid and surface chemistry, third ed. New York, magnetic nanoadsorbents. J. Hazard. Mater. 184, 538–546.
Marcel Dekker. O’Carroll, D., Sleep, B., Krol, M., Boparai, H., Kocur, C., 2013. Nanoscale zero valent
Hizal, J., Apak, R., 2006. Modeling of copper (II) and lead (II) adsorption on kaolinite- iron and bimetallic particles for contaminated site remediation. Adv. Water Res.
based clay minerals individually and in the presence of humic acid. J. Colloid 51, 104–122.
Interface Sci. 295, 1–13. Ponder, S.M., Darab, J.G., Mallouk, T.E., 2000. Remediation of Cr (VI) and Pb (II)
Hu, C.Y., Lo, S.L., Liou, Y.H., Hsu, Y.W., Shih, K., Lin, C.J., 2010. Hexavalent chromium aqueous solutions using supported, nanoscale zero-valent iron. Environ. Sci.
removal from near natural water by copper–iron bimetallic particles. Water Technol. 34, 2564–2569.
Res. 44, 3101–3108. Powell, R.M., Puls, R.W., Hightower, S.K., Sabatini, D.A., 1995. Coupled iron corrosion
Hu, J., Chen, C., Zhu, X., Wang, X., 2009. Removal of chromium from aqueous and chromate reduction: mechanisms for subsurface remediation. Environ. Sci.
solution by using oxidized multiwalled carbon nanotubes. J. Hazard. Mater. 162, Technol. 29, 1913–1922.
1542–1550. Rivera-Utrilla, J., Bautista-Toledo, I., Ferro-Garcı´a, M.A., Moreno-Castilla, C., 2003.
Huang, P., Ye, Z., Xie, W., Chen, Q., Li, J., Xu, Z., Yao, M., 2013. Rapid magnetic Bioadsorption of Pb(II), Cd(II), and Cr(VI) on activated carbon from aqueous
removal of aqueous heavy metals and their relevant mechanisms using solutions. Carbon 323-330 (328).
nanoscale zero valent iron (nZVI) particles. Water Res. 47, 4050–4058. S
ß ölener, M., Tunali, S., Özcan, A.S., Özcan, A., Gedikbey, T., 2008. Adsorption
Kim, J.H., Kim, J.H., Bokare, V., Kim, E.J., Chang, Y.Y., Chang, Y.-S., 2012. Enhanced characteristics of lead (II) ions onto the clay/poly (methoxyethyl) acrylamide
removal of chromate from aqueous solution by sequential adsorption– (PMEA) composite from aqueous solutions. Desalination 223, 308–322.
reduction on mesoporous iron-iron oxide nanocomposites. J. Nanoparticle Sarin, V., Singh, T.S., Pant, K., 2006. Thermodynamic and breakthrough column
Res. 14 (1010), 721–726. studies for the selective sorption of chromium from industrial effluent on
Kim, S.A., Kamala-Kannan, S., Lee, K.J., Park, Y.J., Shea, P.J., Lee, W.H., Kim, H.M., Oh, activated eucalyptus bark. Bioresource Technol. 97, 1986–1993.
B.T., 2013. Removal of Pb (II) from aqueous solution by a zeolite–nanoscale Shi, L.N., Lin, Y.M., Zhang, X., Chen, Z.L., 2011a. Synthesis, characterization and
zero-valent iron composite. Chem. Eng. J. 217, 54–60. kinetics of bentonite supported nZVI for the removal of Cr (VI) from aqueous
Lalhruaitluanga, H., Jayaram, K., Prasad, M., Kumar, K., 2010. Lead (II) adsorption solution. Chem. Eng. J. 171, 612–617.
from aqueous solutions by raw and activated charcoals of Melocanna baccifera Shi, L.N., Zhang, X., Chen, Z.L., 2011b. Removal of chromium (VI) from wastewater
Roxburgh (bamboo)—a comparative study. J. Hazard. Mater. 175, 311–318. using bentonite-supported nanoscale zero-valent iron. Water Res. 45, 886–892.
Lazarević, S., Janković-Častvan, I., Jovanović, D., Milonjić, S., Janaćković, D., Petrović, Shih, Y.H., Hsu, C.Y., Su, Y.F., 2011. Reduction of hexachlorobenzene by nanoscale
R., 2007. Adsorption of Pb2+, Cd2+ and Sr2+ ions onto natural and acid-activated zero-valent iron: Kinetics, pH effect, and degradation mechanism. Sep. Purif.
sepiolites. Appl. Clay Sci. 37, 47–57. Technol. 76, 268–274.
Lazarević, S., Janković-Častvan, I., Potkonjak, B., Janaćković, D., Petrović, R., 2012. Shu, H.Y., Chang, M.C., Chen, C.C., Chen, P.E., 2010. Using resin supported nano zero-
Removal of Co2+ ions from aqueous solutions using iron-functionalized valent iron particles for decoloration of Acid Blue 113 azo dye solution. J.
sepiolite. Chem. Eng. Process Process Intensification 55, 40–47. Hazard. Mater. 184, 499–505.
Liang, K., 2008. Present situation of sepiolite activated and modified and its Shu, H.Y., Chang, M.C., Yu, H.H., Chen, W.H., 2007. Reduction of an azo dye Acid
application prospect. Ind. Minerals Process. 4, 5–9. Black 24 solution using synthesized nanoscale zerovalent iron particles. J.
Li, X.Q., Cao, J., Zhang, W.X., 2008. Stoichiometry of Cr (VI) immobilization using Colloid Interface Sci. 314, 89–97.
nanoscale zerovalent iron (nZVI): a study with high-resolution X-ray Su, J., Lin, S., Chen, Z., Megharaj, M., Naidu, R., 2011. Dechlorination of p-
photoelectron spectroscopy (HR-XPS). Ind. Eng. Chem. Res. 47, 2131–2139. chlorophenol from aqueous solution using bentonite supported Fe/Pd
Li, X.Q., Elliott, D.W., Zhang, W.X., 2006. Zero-valent iron nanoparticles for nanoparticles: synthesis, characterization and kinetics. Desalination 280, 167–
abatement of environmental pollutants: materials and engineering aspects. 173.
Crit. Rev. Solid State Mater. Sci. R 31, 111–122. Suarez, M., García-Romero, E., 2012. Variability of the surface properties of sepiolite.
Li, X.Q., Zhang, W.X., 2007. Sequestration of metal cations with zerovalent iron Appl. Clay Sci. 67, 72–82.
nanoparticles a study with high resolution X-ray photoelectron spectroscopy Sun, Y.P., Li, X.Q., Cao, J., Zhang, W.X., Wang, H.P., 2006. Characterization of zero-
(HR-XPS). J. Phys. Chem. C. 111, 6939–6946. valent iron nanoparticles. Adv. Colloid Interface Sci. 120, 47–56.
Li, Y., Li, J., Zhang, Y., 2012. Mechanism insights into enhanced Cr (VI) removal using Tanboonchuy, V., Grisdanurak, N., Liao, C.H., 2012. Background species effect on
nanoscale zerovalent iron supported on the pillared bentonite by macroscopic aqueous arsenic removal by nano zero-valent iron using fractional factorial
and spectroscopic studies. J. Hazard. Mater. 227, 211–218. design. J. Hazard. Mater. 205, 40–46.
Liu, T., Wang, Z.L., Yan, X., Zhang, B., 2014. Removal of mercury (II) and chromium Tong, G., Guan, J., Xiao, Z., Mou, F., Wang, W., Yan, G., 2008. In situ generated H2
(VI) from wastewater using a new and effective composite: Pumice-supported bubble-engaged assembly: a one-step approach for shape-controlled growth of
nanoscale zero-valent iron. Chem. Eng. J. 245, 34–40. Fe nanostructures. Chem. Mater. 20, 3535–3539.
Luo, S., Qin, P., Shao, J., Peng, L., Zeng, Q., Gu, J.D., 2013. Synthesis of reactive Xi, Y., Mallavarapu, M., Naidu, R., 2010. Reduction and adsorption of Pb2+ in aqueous
nanoscale zero valent iron using rectorite supports and its application for solution by nano-zero-valent iron—A SEM, TEM and XPS study. Mater. Res. Bull.
Orange II removal. Chem. Eng. J. 223, 1–7. 45, 1361–1367.
Lv, X., Hu, Y., Tang, J., Sheng, T., Jiang, G., Xu, X., 2013. Effects of co-existing ions and Xu, D., Tan, X., Chen, C., Wang, X., 2008. Removal of Pb (II) from aqueous solution by
natural organic matter on removal of chromium (VI) from aqueous solution by oxidized multiwalled carbon nanotubes. J. Hazard. Mater. 154, 407–416.
nanoscale zero valent iron (nZVI)-Fe3O4 nanocomposites. Chem. Eng. J. 218, 55– Xu, J., Wang, W., Wang, A., 2013. Influence of anions on the electrokinetic and
64. colloidal properties of palygorskite clay via high-pressure homogenization. J.
Lv, X., Xu, J., Jiang, G., Tang, J., Xu, X., 2012. Highly active nanoscale zero-valent iron Chem. Eng. Data. 58, 764–772.
(nZVI)–Fe3O4 nanocomposites for the removal of chromium (VI) from aqueous Zhang, X., Lin, S., Lu, X.Q., Chen, Z.l., 2010a. Removal of Pb (II) from water using
solutions. J. Colloid Interface Sci. 369, 460–469. synthesized kaolin supported nanoscale zero-valent iron. Chem. Eng. J. 163,
Marjanović, V., Lazarević, S., Janković-Častvan, I., Jokić, B., Janaćković, D., Petrović, R., 243–248.
2013. Adsorption of chromium (VI) from aqueous solutions onto amine- Zhang, X., Lin, S., Lu, X.Q., Chen, Z.l., 2010b. Removal of Pb(II) from water using
functionalized natural and acid-activated sepiolites. Appl. Clay Sci. 80, 202–210. synthesized kaolin supported nanoscale zero-valent iron. Chem. Eng. J. 163,
Mohan, D., Pittman Jr., C.U., 2006. Activated carbons and low cost adsorbents for 243–248.
remediation of tri-and hexavalent chromium from water. J. Hazard. Mater. 137, Zhang, Y.Y., Jiang, H., Zhang, Y., Xie, J.F., 2013. The dispersity-dependent interaction
762–811. between montmorillonite supported nZVI and Cr (VI) in aqueous solution.
Mosaferi, M., Nemati, S., Khataee, A., Nasseri, S., Hashemi, A.A., 2014. Removal of Chem. Eng. J. 229, 412–419.
Arsenic (III, V) from aqueous solution by nanoscale zero-valent iron stabilized
with starch and carboxymethyl cellulose. J. Environ. Health Sci. Eng. 12, 74.