Visible light assisted photodegradation of thimerosal by high performance ZnFe2O4/poly(o-phenylenediamine) composite

Article abstract of DOI:10.1016/j.materresbull.2019.04.008

Thimerosal is a mercury-based preservative that is used in pharmaceuticals, vaccines and health-care products. However, thimerosal toxicity has been well explored and hence it should be properly treated for avoiding its occurrence in the environment. Hence, we synthesized visible light active ZnFe₂O₄/poly(o-phenylenediamine) composite as photocatalyst for degradation of thimerosal. The well characterized ZnFe₂O₄ and composite effectively degraded thimerosal and subsequently reduced Hg(II) into Hg(0) under visible light irradiation. Thimerosal degradation by-products and generation of Hg(0) were analyzed by high performance liquid chromatography and atomic fluorescence spectroscopy. The composite showed better photocatalytic activity than the pure ZnFe₂O₄ nanoparticles. Under the optimum conditions, 90.2% degradation of thimerosal was achieved within 6 h of irradiation. An efficient charge separation ability of poly(o-phenylenediamine) contributes to the high photocatalytic performance of the composite. This work provides a new photocatalytic degradation pathway of thimerosal and thus will stimulate further studies in the removal of organometallic contaminants.

Full text of DOI:10.1016/j.materresbull.2019.04.008

Materials Research Bulletin 116 (2019) 8–15
Contents lists available at ScienceDirect
Materials Research Bulletin
journal homepage: www.elsevier.com/locate/matresbu
Visible light assisted photodegradation of thimerosal by high performance
ZnFe O /poly(o-phenylenediamine) composite
2
4
a
a
a
b
Claudio Sandoval , Suresh Ranganathan , Eimmy Ramírez , Héctor D. Mansilla ,
Robinson Dinamarca , Gina Pecchi , Jorge Yáñeza,⁎
Laboratory of Trace Elements and Speciation, Department of Analytical and Inorganic Chemistry, University of Concepción, Concepción, Chile
Laboratory of Environmental Organic Chemistry, Department of Organic Chemistry, University of Concepción, Concepción, Chile
Department of Physical Chemistry, Faculty of Chemical Sciences, University of Concepción. Concepción, Chile
c
c,d
a
b
c
d
Millenium Nuclei on Catalytic Processes Towards Sustainable Chemistry (CSC), Chile
A R T I C L E I N F O
A B S T R A C T
Keywords:
Composite materials
ZnFe O
2 4
Poly(o-phenylenediamine)
Photocatalysis
Thimerosal is a mercury-based preservative that is used in pharmaceuticals, vaccines and health-care products.
However, thimerosal toxicity has been well explored and hence it should be properly treated for avoiding its
occurrence in the environment. Hence, we synthesized visible light active ZnFe
composite as photocatalyst for degradation of thimerosal. The well characterized ZnFe
2
O
4
/poly(o-phenylenediamine)
and composite ef-
2 4
O
fectively degraded thimerosal and subsequently reduced Hg(II) into Hg(0) under visible light irradiation.
Thimerosal degradation by-products and generation of Hg(0) were analyzed by high performance liquid chro-
matography and atomic fluorescence spectroscopy. The composite showed better photocatalytic activity than the
Thimerosal
2 4
pure ZnFe O nanoparticles. Under the optimum conditions, 90.2% degradation of thimerosal was achieved
within 6 h of irradiation. An efficient charge separation ability of poly(o-phenylenediamine) contributes to the
high photocatalytic performance of the composite. This work provides a new photocatalytic degradation
pathway of thimerosal and thus will stimulate further studies in the removal of organometallic contaminants.
1. Introduction
physical separation but they do not achieve degradation of TMS.
Therefore, development of an efficient method to degrade the organic
Nowadays, one of the most alarming issues is the presence of
matter in TMS simultaneous with a safe final disposition of Hg, is es-
sential for avoiding its occurrence in the environment.
In the last few decades, great attention has been paid to synthesis
semiconductor photocatalysts for decontamination of polluted water.
emerging pollutants like organometallic compounds in the environ-
ment. Organometallic pollutants are mainly discharged by industrial,
agricultural activities and used massively in different products [1–3].
They can have harmful impact in the balance of ecosystem, as the
process known as bioaccumulation [4]. A potential organometallic
emerging contaminant is thimerosal (TMS, sodium ethylmercury thio-
salicylate) which is widely used to inhibit the growth of fungi and
bacteria in vaccines, pharmaceutical and personal care products.
However, there have been reported potential negative health effects in
humans due to TMS which are still a topic of scientific controversy [5].
Therefore, disposal of TMS containing products could seriously con-
taminate the environment by release of mercury or ethylmercury. Some
treatment methods for TMS elimination such as filtration, biological
treatments (aerobic and anaerobic), solvent extraction, phytoremedia-
tion, adsorption in activated carbon have been reported [6–10]. How-
ever, these methods are generally non-destructive and mainly based on
Generally, TiO
However, considerable attention has been given to zinc ferrite
(ZnFe ), due to its visible light harvesting ability, easy preparation,
stability and magnetic property [11]. So far, ZnFe has been utilized
2
and ZnO photocatalysts have been widely studied.
2 4
O
2 4
O
as photocatalyst for dyes [12,13]. Nevertheless, owing to the fast re-
combination rate of photo-generated charges, the photocatalytic ability
of pure ZnFe O is poor. In order to improve its activity, compositing
2 4
with conducting polymer like polyaniline have been adopted [14].
Among the conducting polymers, poly(o-phenylenediamine) (PoPD), a
derivative of polyaniline, has also been studied due to its electro-
chemical and optical properties including high stability. It also acts as
an efficient electron donor and good hole transporter upon visible-light
illumination [15]. With this background, we developed a strategy to
⁎
Corresponding author at: Department of Analytical and Inorganic Chemistry, Faculty of Chemical Sciences, University of Concepción, 4070371, Edmundo Larenas
129, Concepción, Chile.
E-mail address: jyanez@udec.cl (J. Yáñez).
https://doi.org/10.1016/j.materresbull.2019.04.008
Received 16 October 2018; Received in revised form 4 April 2019; Accepted 4 April 2019
Available online 05 April 2019
0025-5408/ © 2019 Published by Elsevier Ltd.

C. Sandoval, et al.
Materials Research Bulletin 116 (2019) 8–15
couple ZnFe
2
O
4
and PoPD, which would develop a synergetic phe-
recorded using a Nicolet Nexus 470 FT-IR (Thermo Scientific, USA) in
−1
nomenon that will increase its adsorption in the visible range and
photocatalytic properties. To the best of our knowledge, synthesis of
ZnFe O /PoPD composite with photocatalytic activity, especially to-
2 4
the range of 4000–50 cm . BET analysis was performed by Tristar II
3020 (Micrometrics, USA) at 77 K. Diffuse reflectance spectra were
recorded using a UV-2700 spectrophotometer (Shimadzu, Japan) with
wards organometallic compound such as TMS, has not been reported
yet.
4
integration sphere (ISR-2600 Plus), using BaSO as a reference (Merck,
Germany). SEM micrographs were obtained using a JSM-6380 LV
(JEOL, Germany) scanning electron microscopy. TEM analysis was
performed by JEM-1200EX II (JEOM, Germany) transmission electron
microscopy with a resolution of 5 Å and 120 kV with a camera F82
(Gatan, USA).
2 4
Therefore, we have attempted to synthesis the ZnFe O /PoPD
composite and have been examined its TMS photocatalytic degradation
activity. The synthesized photocatalysts were characterized by XRD,
FTIR, DRUV visible spectroscopy, SEM, TEM and BET method. The
photocatalytic degradation of TMS was achieved under visible light
illumination. Furthermore, photocatalytic degradation pathway of TMS
has also been proposed.
2.5. Photocatalytic experiment
All the photocatalytic experiments were conducted in a borosilicate
glass reactor with an effective volume of 100 mL, equipped with inlet
and outlet to modify the reaction atmospheric conditions. The con-
2
. Experimental procedure
−1
−1
2.1. Materials
centration of TMS and photocatalyst was 10 mg L
and 100 mg L
respectively. The pH of the solution was adjusted to 4 with hydrochloric
acid (0.01 M) (Merck, Germany). Photodegradation experiment was
3 2
Hydrated zinc nitrate (ZnNO ·4H O), ferrous nitrate (Fe
−2
(
NO
3
)
2
·6H
2
O), glycine (C
H
2 5
NO
2
) and potassium persulfate (K
S
2 2
O
8
)
performed at (Xe lamp) visible light irradiation (2.2 mWcm ). The
distance between the lamp source and the reactor was maintained at
15 cm. To monitor the reaction, samples were taken periodically using
syringe and subsequently subjected to filtration by nitrocellulose
membrane filter with a pore diameter of 0.22 μm (Merck, Germany). A
LAMBDA 25 UV–vis spectrophotometer (PerkinElmer Ltd., USA) was
used for monitoring the UV–vis spectra of TMS during the photo-
catalysis. The equipment operated with a band pass of 1 nm and with
1 cm of optical path using quartz cuvettes.
were obtained from Sigma-Aldrich, US. 1,4-phenylenediamine, malonic
acid, thimerosal (≥98.5%), salicylic acid, dithiosalicylic acid, thiosa-
−
1
licylic acid, mercury(II) nitrate (standard 1000 mg L
CertiPUR,
Merck), methanol, ortho-phosphoric acid, sodium borohydride, hydro-
chloric acid and potassium hydroxide were purchased from Merck,
Germany. Ultra-pure water was obtained from EASY water purification
device (Millipore, US). Argon (99.998%) was purchased from Air-
Liquide (Chile).
2
.2. Auto combustion synthesis of ZnFe
2
O
4
nanoparticles
·6H O and glycine as
2.6. Analytical methods
Aqueous solution of ZnNO ·4H O, Fe(NO
3
2
3
)
2
2
The identification of intermediates and degradation products was
performed by HPLC (KNAUER Smartline, Germany) with a UV–vis de-
tector (KNAUER Smartline UV detector 2500) operating at 254 nm and
25 °C. Reverse-phase chromatography was performed w®ith
ignition promoting agent were taken in the ratio of 1:2:1 and mixed
together [16]. The resulting mixture was slowly heated to evaporate the
water and promote the formation of the gel. Then the gel formed was
brought to the auto-combustion temperature of the glycine (263 °C).
Once the combustion process was finished, a calcination step was car-
ried out in an oxidizing atmosphere at 600 °C in order to remove or-
a
150 mm × 4.6 mm LiChroCART column (θ = 5 μm) Purospher STAR
RP-18 endcapped, with pre-column, plus a 20 μL loop (for sample in-
jection). A mobile phase composed of methanol (Merck, Germany),
ultrapure water (Merck, USA) and ortho-phosphoric acid (Merck,
Germany) in the ratio 66: 33: 1 at pH 3 was used. The flow rate was
ganic contents and thus we obtained ZnFe
2 4
O powder.
−1
2.3. Synthesis of ZnFe /PoPD composite
2
O
4
1 mL min . TMS (Merck, Germany) concentration was determined
−1
using an external calibration curve between 1–30 mg L . The mercury
−1
Oxidative polymerization method for the synthesis of ZnFe
PoPD composite was used [17]. The typical synthetic procedure is as
follows: In a 250 mL beaker, 0.48 g of ZnFe in 100 mL of water was
2
O
4
/
(II) nitrate (1000 mg L CertiPUR, Merck, Germany), salicylic acid
(Merck, Germany), thiosalicylic acid and dithiosalicylic acid (Merck,
Germany) were used as standards.
2 4
O
sonicated for 15 min. To this dispersion, 1.046 g malonic acid is added
and the reaction set-up is kept in an ice bath (−4 °C) with constant
stirring. After that, 1.081 g of o-phenylenediamine was added. Finally,
To determine the generation of elemental mercury (Hg(0)) during
photocatalysis, the photocatalytic reactor was connected to the Aurora
Lumina 3300 (Canada) atomic fluorescence spectrometer (AFS) where
the generated Hg(0) was released in to the gas phase by using Ar
2 2 8
100 mL aqueous solution (0.45 g) of K S O was slowly added with
−
1
constant stirring. The resulting reaction mixture was allowed to stand in
cold for 24 h, and then the obtained precipitate was filtered and dried in
an oven at 45 °C for 4 h.
100 mL min as carrier gas. This equipment used an additional make
−
1
up gas in the torch consisting of argon 400 mL min and a shield gas of
−1
200 mL min . For calibration, mercury cold vapor was generated by
reduction of Hg(II) in a 500 mL volumetric flask using a mixture of 20%
2
.4. Characterization techniques
by volume of ultra-pure hydrochloric acid and a 1.5% (w/v) NaBH
4
basic solution (7.5 g of NaBH
3. Results and discussion
3.1. Characterization
4
, 0.05% w/v NaOH to 500 mL of water).
The X-ray diffraction patterns were obtained using a Bruker D4
(
Endeavor, Germany) with Fe Kα1 radiation (λ = 1.9360 Å). Mn filter
for K radiation was used. Data were collected over the 2θ range of
0°–80° with a velocity of 2° min . The average crystallite size (d) was
β
−
1
2
calculated using the Debye-Scherrer equation:
The structure of pure ZnFe
determined by XRD analysis (Fig. 1). All the diffraction peaks of pure
2 4
ZnFe (Fig. 1 pattern a) can be perfectly indexed with spinel ZnFe O
(JCPDS No.: 01-086-0507). The predominant diffraction peaks ob-
served at 2θ = 29.9°, 35.2°, 42.9°, 56.6° and 62.2° can be assigned to
2 4
(220), (311), (400), (511) and (440) crystal planes of spinel ZnFe O
2 4 2 4
O and ZnFe O /PoPD composite was
Kλ
d =
βcosθ
(1)
2
O
4
Where, K is the Scherrer constant (0.9), λ is the wavelength of the X-ray
radiation (λ = 1.9360 Å), β is the full width of half maximum peak and
θ is diffraction angle of most intense peak. The FTIR spectra were
9

C. Sandoval, et al.
Materials Research Bulletin 116 (2019) 8–15
band gap of the samples, Kubelka-Munk function was utilized (Sup-
plementary Fig. S1) [26]. From the Fig. 3, the band gap of pure ZnFe
and ZnFe /PoPD composite was calculated as 2.13 eV and 1.51 eV
respectively. The observed 0.62 eV difference indicated the reduction of
gap between conduction and valence band of ZnFe by PoPD. This
observation clearly infers that the PoPD have strong interaction with
ZnFe . As indicated in reference [24,27,28], the interaction between
2 4
O
2 4
O
2 4
O
2 4
O
the metal oxide nanoparticles and a polymer would develop a syner-
getic phenomenon that will increase its adsorption and photocatalytic
properties.
2 4
SEM micrograph (Fig. 4a) with 5 μm resolution of pure ZnFe O
powder shows agglomerated particles. When the resolution increased
up to 1 μm (Fig. 4b), large number of cavities with diameter in micro-
meter sizes can be observed. These results are expected, due to its
3 3 2
synthetic method, i.e., when the gel of ZnNO , Fe(NO ) and glycine
was calcined at 600 °C, oxides of nitrogen and carbon will be liberated
as by-products with leaving cavities in the sample [17].
In order to examine the shape of ZnFe
was performed. As can be seen in Fig. 5a and b, ZnFe
2 4
quired plate like morphology with different sizes. In addition, ZnFe O
2
O
4
particles, TEM analysis
2 4 2 4
Fig. 1. XRD pattern of (a) pure ZnFe O and (b) ZnFe O /PoPD composite in
the range of 2θ = 20–80°. Insert Figure: XRD patterns in the range of
2 4
O
particles ac-
2
θ = 34–37°.
plates are agglomerated either one over the other or head-to-head style.
The average particle size was calculated as 50 nm. EDX spectrum
respectively [18]. Absence of other impurity phases such as ZnO and α-
in the XRD pattern, demonstrates the high purity of the
synthesized samples. Furthermore, sharp peaks with relatively high
intensity also suggest the good crystallinity of the sample [18]. The
(
Supplementary Fig. S2) of pure ZnFe
Fe, Zn and O only [28]. Unlike pure ZnFe
powder (Fig. 4c) shows the agglomerates without any cavity. This is
because the growth of PoPD on ZnFe surface. The formation of
ZnFe /PoPD composite can be explained as follows: when sub-
sequent addition of malonic acid (MA) and o-phenylenediamine (o-PD)
monomer to ZnFe dispersion, a salt will be formed by the reaction of
MA with o-PD which will be adsorbed on the surface of ZnFe par-
2
O
4
powder shows peaks due to
2 3
or γ-Fe O
2
O
4
, ZnFe /PoPD composite
2 4
O
2 4
O
XRD pattern of ZnFe
milar to that of pure ZnFe
chain are not visible. It might be due to the dominance of ZnFe
diffraction peaks. However, there is a difference in peak intensity and
position. The diffraction peaks of ZnFe /PoPD composite have lower
intensity and shifted (1°) towards higher diffraction angle than that of
pure ZnFe (inset in Fig. 1). This observation indicates that the PoPD
has some effect on the structure of ZnFe . This result is well con-
2
O
4
/PoPD composite (Fig. 1 pattern b) is also si-
2 4
O
2 4
O . The periodicity parallel peaks of PoPD
2 4
O
2 4
O
2 4
O
2 4
O
ticles through hydrogen bonding. Then, the potassium persulfate as an
oxidizing agent was added to complete the polymerization process and
2 4
O
thus a layer of PoPD was grown on the surface of ZnFe
above discussion can also be evidenced by TEM images (Fig. 5c–f).
Amorphous PoPD can be easily seen along with ZnFe particles. In-
terestingly, ZnFe core with PoPD shell particles can also be found
Fig. 5e). These results can be corroborated with series of XRD, FTIR
2 4
O [29,30]. The
2 4
O
sistent with other reports [19–21]. By using full width of half maximum
of high intensity peak in Debye-Scherrer equation [22], the average
2 4
O
2 4
O
crystallite size was calculated as 58.2 nm for ZnFe
ZnFe /PoPD composite.
The functional groups and interaction happened between ZnFe
and PoPD were ascertained by FTIR spectral analysis. Fig. 2(a and b)
and (c and d) shows the FTIR spectrum of ZnFe and ZnFe /PoPD
and 600 to 50 cm
is due to the stretching vi-
2 4
O and 46.5 nm for
(
2 4
O
and UV visible results. The co-existence of Zn, Fe, O along with C in
EDX spectrum of composite (Supplementary Fig. S3) sample also sup-
ports the above discussion.
2 4
O
2
O
4
2 4
O
−
1
−1
The BET surface area of pure ZnFe
2
O
4
and ZnFe
site was found to be 18 and 10 m g respectively (Supplementary Fig.
S4). The determined BET surface area of ZnFe is similar to the report
27]. However, ZnFe /PoPD composite has lesser surface area than
that of pure ZnFe . As discussed in earlier, the growth of PoPD layer
on ZnFe particles covered the available pores. The pore volume of
composite sample (0.011 cm³ g ) is significantly lower than pure
2 4
O /PoPD compo-
composite within the range of 4000–400 cm
2
−1
−
1
respectively. A weak band at 3400 cm
2 4
O
bration of OeH bond, which would indicate the presence of surface
eOH group in the sample (Fig. 2a) [23]. FTIR spectrum (Fig. 2b) of
[
2 4
O
2 4
O
2 4
pure ZnFe O shows characteristic bands in the region of 600 to
−
1
2+
2 4
O
4
00 cm . In the far IR spectrum, the intrinsic vibration of Zn eO
−
1
−1
3+
bond can be observed at 525 cm . The band due to Fe eO in tet-
−1
2 4
ZnFe O (0.024 cm³ g ).
rahedral (T
h
) and octahedral (O
h
) site is also clearly observed [24]. In
Fig. 2c FTIR spectrum of ZnFe
2 4
O /PoPD composite shows characteristic
3
.2. Photocatalytic degradation of thimerosal
bands of PoPD along with ZneO and FeeO bonds (MeO). The band at
−
1
3220 cm
corresponds to the stretching vibration of NeH bond. The
−
1
Photodegradation of TMS under visible light illumination was
peaks at 1571 and 1506 cm correspond to the stretching of C]N and
−
1
2 4 2 4
achieved by using pure ZnFe O and ZnFe O /PoPD composite as
C]C of quinoid and benzonid ring of PoPD. The peak at 1278 cm can
be assigned to the stretching of CeN bond. The band at 813 cm
associated with the out-of-plane bending for the benzene ring [25]. As
expected, there is no change in the number of band in far IR region
−1
photocatalyst. Based on the DRS UV visible analysis, visible light in-
duced photocatalytic activity toward TMS has been performed. Fig. 6a
displays the characteristic deconvoluted UV-absorption spectrum of
blank TMS solution. It shows absorption bands at 202, 235 and 265 nm,
attributed to aromatic ring, thiol (Hg-S) and carboxyl group (eCOOH)
attached with an aromatic ring, respectively [31]. Visible light irra-
diation for 6 h leads to slight decrease in absorbance with respect to
blank TMS (Fig. 6b). Noticeably, a band at 235 nm was disappeared
which indicates that the visible light has some effect on TMS. Fur-
thermore, about 43% of decrease in absorbance can be obtained in
is
(
Fig. 2d). However, significant changes are observed in the band in-
tensity of metaleoxygen bonds. This is due to the formation of inter-
action between PoPD and ZnFe that would affects the vibration
energy of the metaleoxygen bonds. These results suggest that the
presence of PoPD on ZnFe particles.
Fig. 3 shows the UV visible absorption spectrum of pure ZnFe
and ZnFe /PoPD composite. Pure ZnFe (Fig. 3a) shows strong
absorption from UV to visible region (up to 500 nm), while composite
Fig. 3b) sample strongly absorbs in visible region. To calculate the
2 4
O
2 4
O
2 4
O
2
O
4
2 4
O
presence of pure ZnFe
in absorbance was observed in presence of ZnFe
Fig. 6d). In both Fig. 6c and d, the peaks correspond to SeHg and
2
O
4
nanoparticles (Fig. 6c) whereas 50% decrease
2 4
O /PoPD composite
(
(
10

C. Sandoval, et al.
Materials Research Bulletin 116 (2019) 8–15
2 4 2 4
Fig. 2. FTIR spectra of (a and b) pure ZnFe O and (c and d) ZnFe O /PoPD composite.
3.3. TMS degradation pathway
As discussed earlier, TMS degradation in presence of ZnFe
2 4
O is
similar to that of photolysis. It might be due to the faster recombination
rate of photocharges and therefore possibility for excitation of electrons
from VB to CB of ZnFe
photocatalyst shows best activity which clearly reveals that the pre-
sence of PoPD along with ZnFe plays a major role in TMS de-
2 4
O can be neglected [32]. However, composite
2 4
O
gradation. According to UV visible result (Fig. 3b), PoPD can absorbs
visible light and consequently electronic excitation from highest occu-
pied molecular orbital (HOMO) to lowest unoccupied molecular orbital
(
LUMO) will takes place. Therefore, holes will be left at HOMO level of
PoPD [33]. Photo-generated electrons in LUMO can jump to the con-
duction band (CB) of ZnFe [34]. At the same time, electrons in
valence band (VB) of ZnFe can shift to the HOMO and leaving holes
at the VB of ZnFe [15]. Dissolved oxygen (O ) can absorbs photo-
generated electrons from CB of ZnFe and becomes superoxide ra-
dical. Simultaneously water molecule can reacts with hole at VB of
ZnFe and produce hydroxyl radical. Both superoxide and hydroxyl
2 4
O
O
2 4
2
O
4
2
2 4
O
Fig. 3. Diffuse reflectance UV visible absorption spectrum of (a) pure ZnFe
and (b) ZnFe /PoPD composite.
2 4
O
2 4
O
2 4
O
radical react with TMS to produce degradation by-products.
HPLC was used to identify some oxidation intermediates, using
expected standards. Previously, some intermediates have been reported
as by-products from the degradation of TMS (Supplementary Fig. S5)
eCOOH was disappeared. It is important to mention that the decrease
in absorbance of TMS cannot be considered to calculate its degradation
efficiency. Because, our previous experience [31] suggests that the
degradation of TMS leads to derivatives of benzene carboxylic acid such
as thiosalicylic acid and dithiosalicylic acid that also absorbs in the
same wavelength. Hence, HPLC was used to monitor the degradation
profile of TMS (Fig. 7). After 6 h of visible light irradiation, only 26.3%
[
5,31]. Fig. 8 shows HPLC chromatograms of intermediates by using
different treatments as a function of reaction time. As can be seen in
Fig. 8a, blank TMS shows an unsymmetrical peak at 6.9 min. Visible
light illumination for 6 h leads notorious changes in TMS (Fig. 8b).
Apart from 20% decrease in TMS peak intensity, two new peaks at
of TMS degradation was observed. Pure ZnFe
TMS degradation. It suggests that pure ZnFe
2
O
4
also shows 24.0% of
doesn’t show photo-
2
min and 8.1 min appeared. A high intensity signal at 2 min corre-
2
O
4
sponds to Hg(II) ions. A peak at 8.1 min is attributed dithiosalicylic
acid. These results indicate that the visible light has sufficient energy to
cleave HgeS and HgeC bonds in TMS (Scheme 1).The observed results
are also well consistent with Tan et al. [35] who reported that the
degradation of TMS leads thiosalicylic acid and ethylmercury ion.
Moreover, thiosalicylic acid undergoes dimerization to forms dithiosa-
licylic acid.
catalytic activity towards TMS degradation which might be due to its
faster recombination rate of photocharges [32]. However, composite
photocatalysis shows 90.2% of TMS degradation efficiency within 6 h of
visible light irradiation. The enhanced photocatalytic activity of com-
posite photocatalyst must be due to the charge separation ability of
2 4
PoPD on ZnFe O .
11

C. Sandoval, et al.
Materials Research Bulletin 116 (2019) 8–15
2 4 2 4
Fig. 4. SEM micrographs of (a, b) pure ZnFe O and (c) ZnFe O /PoPD composite.
2 4 2 4
Fig. 5. TEM images of (a, b) pure ZnFe O and (c–f) ZnFe O /PoPD composite.
12

C. Sandoval, et al.
Materials Research Bulletin 116 (2019) 8–15
Fig. 6. UV visible absorption spectrum in liquid phase. (a) TMS at time 0 h (control spectrum), (b) photolysis, (c) photocatalysis with pure ZnFe
photocatalysis with ZnFe /PoPD composite after 6 h of reaction.
2 4
O and (d)
2 4
O
Fig. 8. HPLC Chromatogram of photodegradation by-products of TMS. (a) At
−
1
0
h 10 mg L , (b) photolysis at 6 h, (c) pure ZnFe
ZnFe /PoPD composite at 6 h. TMS: thimerosal, SA: salicylic acid, DTSA:
dithiosalicylic acid.
2 4
O at 8 h and (d) with
Fig. 7. Plot of C/C
0 2 4
versus time. (a) Photolysis, (b) pure ZnFe O and (c)
2 4
O
ZnFe /PoPD composite. Concentration was measured by HPLC.
2 4
O
The fact that the CeHg bond in ethylmercury ion is less stable and
hence its decomposition is faster [36]. Therefore, the as formed ethyl-
mercury ion undergoes decomposition and leave Hg(II) ions in solution.
However, remaining Hg(II) ions in solution is highly undesirable since
its toxicity in the environment. As photolysis, TMS behaves similarly in
2 4
presence of pure ZnFe O with visible light irradiation (Fig. 8c). In
contrast to photolysis, peak intensity of Hg(II) was significantly sup-
pressed which might be due to reduction of Hg(II) into Hg(0). Yepsen
et al. [31] and Khalil et al. [37] have reported the photocatalytic re-
duction of Hg(II) salts using TiO
They found that the Hg(II) ions adsorbed at the surface of TiO
undergo reduction. Herein, adsorbed Hg(II) ions on the surface of pure
2
photocatalyst under UV irradiation.
2
could
2 4
ZnFe O was reduced as Hg(0) (Eq. (2)).
>
420 nm
Hg2 + H
+
0
+
Scheme 1. Proposed TMS degradation pathway by visible light.
2
O⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ ⎯→ Hg +2H +1/2O
2
ZnFe
2
O
4
/PoPD
(2)
13

C. Sandoval, et al.
Materials Research Bulletin 116 (2019) 8–15
Fig. 9. Atomic fluorescence of Hg° evolution in gas phase. (a) Standard of Hg
(
II) reduced with NaBH
4
, (b) photodegradation of TMS with ZnFe
2
O
4
/PoPD
Fig. 10. Variation of concentration of TMS as a function of time. (a) With
composite and (c) photolysis of TMS.
ZnFe O /PoPD/Ar, and (b) with ZnFe O /PoPD/O .
2
4
2
4
2
Further evidence for generation of Hg(0) has discussed later.
Interestingly, by the action of the photocatalysis with ZnFe /PoPD
following degradation pathway (Scheme 2) can be suggested.
2
O
4
Furthermore, the photocatalytic degradation of TMS in presence
and absence of oxygen atmosphere was also investigated. For the
former, the reactor was used under normal conditions with the presence
of atmospheric condition (air as the O2 source). Photocatalysis in the
composite the signal of TMS disappears (Fig. 8d). Along with increase
of peak intensity at 8.1 min, appearance of new peak at about 5 min due
to salicylic acid is also observed. Furthermore, the peak of Hg(II) was
also suppressed which indicates the removal of Hg through reduction
process (Hg(II) into Hg(0)). Hg(0) can be easily trapped by copper
through the formation of amalgam [31]. Otherwise, Hg(0) vapors can
also be adsorbed by activated carbon [38]. This is the great advantage
of these photocatalysts, since removal of Hg from solution is highly
absence of O was achieved by bubbling argon gas through the gas inlet
2
−
1
of the photocatalytic reactor with a flow rate of 400 mL min
(con-
ditions necessary to be able to evaluate the generation of Hg(0) in the
atomic fluorescence equipment). In Fig. 10, we can observe the varia-
tion in the concentration of TMS as a function of time for two different
experiments that were carried out using composite photocatalyst. In
desirable. From these results, it can determine that ZnFe
composite is the best photocatalyst for removal of TMS.
2 4
O /PoPD
presence of O , TMS degradation is greater and achieved around 90%
2
The generation of Hg(0) during the photocatalysis of TMS with
ZnFe /PoPD composite was confirmed by using CV-AFS technique.
efficiency whereas in the absence of O , TMS degradation rate was
2
2
O
4
significantly suppressed. It clearly indicated that the molecular O plays
2
For this purpose, the photocatalytic reactor was connected with AFS
equipment (Supplementary Fig. S6) and three different experiments
were performed. The first one is the reduction of a standard of Hg(II) to
an important role in the photocatalysis i.e., oxygen adsorbed on the
surface of composite nanoparticles will generate highly oxidizing and
−
%
reactive species (O2 , H O , HO , to point out some) that promote the
2
2
Hg(0) using NaBH
4
as reducing agent in hydrochloric acid (Fig. 9a). It
photocatalytic degradation of TMS.
can be observe that a very definite signal of Hg(0) appears in the first
minutes of reaction. Fig. 9b corresponds to the photocatalytic de-
gradation of TMS with ZnFe O /PoPD composite in suspension, where
2 4
the reduction of Hg(II) to Hg(0) occurs approximately after 17 min of
reaction which remains constant until 4 h. Fig. 9c corresponds to the
photolysis of TMS, where it is evidenced that there is negligible for-
mation of Hg(0). This result shows that the photocatalysis of TMS using
In order to know the stability of ZnFe O /PoPD composite after
2
4
photocatalytic degradation reaction, TEM analysis was performed and
the obtained results are shown in Fig. S7. It shows that the PoPD layer
was still attached on the surface of ZnFe O nanoparticles. Hence, it can
2
4
be concluded that ZnFe O /PoPD has good stability.
2
4
4
. Conclusion
2 4
ZnFe O /PoPD composite achieves the reduction of Hg(II) into Hg(0)
after the degradation of TMS.
From the results obtained with the different analytical techniques
2 4
for the photocatalysis process with ZnFe O /PoPD composite, the
In this work, we achieved ZnFe
photocatalyst for degradation of TMS assisted by visible light. ZnFe
PoPD composite has been characterized by various techniques. The
2 4
O /PoPD composite as an efficient
2 4
O /
2 4
Scheme 2. TMS degradation pathway in presence of ZnFe O /PoPD composite photocatalyst assisted with visible light.
14

C. Sandoval, et al.
Materials Research Bulletin 116 (2019) 8–15
direct photolysis and photocatalysis with pure ZnFe
thiosalicylic acid and salicylic acid along with Hg(II) as major products.
The results suggest that PoPD on the surface of ZnFe enhanced the
photocatalytic activity. Analytical experiment results reveal that the
TMS photocatalytic degradation occurs through cleavage of HgeS and
HgeC bond. Further, reduction of Hg(II) into Hg(0) also takes place.
The as-formed Hg(0) can be removed from the solution and easily
trapped in copper as amalgam.
2
O
4
produces di-
enhanced photocatalytic performance under solar light, J. Environ. Chem. Eng. 1
(2013) 620–627.
[
[
15] X. Zhao, L. Lv, B. Pan, W. Zhang, S. Zhang, Q. Zhang, Polymer-supported nano-
composites for environmental application: a review, Chem. Eng. J. 170 (2011)
381–394.
2 4
O
16] F. Liu, X. Li, Q. Zhao, Y. Hou, X. Quan, G. Chen, Structural and photovoltaic
properties of highly ordered ZnFe
2 4
O nanotube arrays fabricated by a facile sol–gel
template method, Acta Mater. 57 (2009) 2684–2690.
[17] A. Anand, N. Rani, P. Saxena, H. Bhandaria, S.K. Dhawan, Development of poly-
aniline/zinc oxide nanocomposite impregnated fabric as an electrostatic charge
dissipative material, Polym. Int. 64 (2015) 1096–1103.
[
18] L. Jing, Y. Xu, C. Qin, J. Liu, S. Huang, M. He, H. Xu, H. Li, Visible-light-driven
ZnFe /Ag/Ag VO photocatalysts with enhanced photocatalytic activity under
Acknowledgments
O
2 4
3
4
visible light irradiation, Mater. Res. Bull. 95 (2017) 607–615.
[
2 4
19] H. Kim, P. Borse, J. Jan, E. Jeong, O. Jung, Y. Suh, J. Lee, Fabrication of CaFe O /
The authors thank National Commission for Scientific and
Technological Research (CONICYT), Chile, for providing projects as
FONDECYT1151296 andN°3160499; Center for Optics and Photonics,
Grant CONICYT-PFB-0824 and CONICYT/FONDAP/15110019. Claudio
Sandoval thanks the M.Sc. fellowship of the Postgraduate Direction,
Universidad de Concepción, Chile and CONICYT Doctoral Fellowship
MgFe
2 4
O bulk heterojunction for enhanced visible light photocatalysis, Chem.
Commun. 39 (2009) 5889–5891.
[20] B. Lafuente, R.T. Downs, H. Yang, N. Stone, The power of databases: the RRUFF
project, in: T. Armbruster, R.M. Danisi (Eds.), Highlights in Mineralogical
Crystallography, W. De Gruyter, Berlin, Germany, 2015, pp. 1–30 http://rruff.info/.
21] X. Sun, H. Zhang, L. Zhou, X. Huang, C. Yu, Polypyrrole-coated zinc ferrite hollow
spheres with improved cycling stability for lithium-ion batteries, Small 12 (2016)
[
[
3
732–3737.
21180934.
22] P.K. Kannan, R. Saraswathi, Impedimetric detection of alcohol vapours using na-
nostructured zinc ferrite, Talanta 129 (2014) 545–551.
Appendix A. Supplementary data
[23] H. Zhao, R. Liu, Q. Zhang, Q. Wang, Effect of surfactant amount on the morphology
and magnetic properties of monodisperse ZnFe nanoparticles, Mater. Res. Bull.
5 (2016) 172–177.
2 4
O
7
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.materresbull.2019.04.
[
[
[
[
[
[
[
[
24] M.M. Lakouraj, E.N. Zare, P.N. Moghadam, Synthesis of novel conductive poly(p-
phenylenediamine)/Fe O nanocomposite via emulsion polymerization and in-
3
4
008.
vestigation of antioxidant activity, Adv. Polym. Technol. 33 (2014) 21385.
25] N. Song, Y. Wu, W. Wang, D. Xiao, H. Tan, Y. Zhao, Layer-by-layer in situ growth
2
flexible polyaniline/graphene paper wrapped by MnO nanoflowers for all-solid-
state supercapacitor, Mater. Res. Bull. 111 (2019) 267–276.
References
26] Y. Abouliatim, Y.E. Hafiane, A. Smith, M. Mesnaoui, T. Chartier, A. Benhammou,
A. Abourriche, L. Nibou, Study of borosilicate glaze opacification by phosphates
using Kubelka-Munkmodel, Ceram. Int. 43 (2017) 5862–5869.
27] P. Magdziarz, P. Bober, M. Trchová, Z. Morávková, M. Bláha, J. Prokes, J. Stejskal,
Conducting composites prepared by the reduction of silver ions with poly(p-phe-
nylenediamine), Polym. Int. 64 (2015) 496–504.
[
1] G. Acosta, A. Spisso, L. Fernández, L. Martinez, P. Pacheco, R. Gil, Determination of
thimerosal in pharmaceutical industry effluents and river waters by HPLC coupled
to atomic fluorescence spectrometry through post-column UV-assisted vapor gen-
eration, J. Pharm. Biomed. Anal. 106 (2015) 79–84.
[
[
[
2] I. Sirés, E. Brillas, Remediation of water pollution caused by pharmaceutical re-
sidues based on electrochemical separation and degradation technologies: a review,
Environ. Int. 40 (2012) 212–229.
3] D.J. Lapworth, N. Baran, M.E. Stuart, R.S. Ward, Emerging organic contaminants in
groundwater: a review of sources, fate and occurrence, Environ. Pollut. 163 (2012)
2 4
28] B. Jiang, C. Han, B. Li, Y. He, Z. Lin, In-situ crafting of ZnFe O nanoparticles im-
pregnated within continuous carbon network as advanced anode materials, ACS
Nano 10 (2016) 2728–2735.
29] F. Liu, L. Wang, Q. Yin, B. Jiang, Optimization of thermoelectric figure of merit in
poly(p-phenylenediamine)/exfoliated grapheme nanosheets composites, RSC Adv.
2
87–303.
4
(2014) 51558–51568.
4] E. Avigliano, N.F. Schenone, A.V. Volpedo, W. Goessler, A.F. Cirelli, Heavy metals
and trace elements in muscle of silverside (Odontesthes bonariensis) and water from
different environments (Argentina): aquatic pollution and consumption effect ap-
proach, Sci. Total Environ. 506–507 (2015) 102–108.
30] R. Suresh, K. Giribabu, R. Manigandan, S.P. Kumar, S. Munusamy, S. Muthamizh,
V. Narayanan, Characterization and dopamine sensing property of V
nanohybrid, Synth. Met. 196 (2014) 151–157.
31] O. Yepsen, D. Contreras, P. Santander, J. Yáñez, H.D. Mansilla,
2 5
O @polyailine
[
[
[
[
5] S. Bernard, A. Enayati, L. Redwood, H. Roger, T. Binstock, Autism: a novel form of
D. Amarasiriwardena, Photocatalytic degradation of thimerosal in human vaccine’s
residues and mercury speciation of degradation by-products, Microchem. J. 121
mercury poisoning, Med. Hypotheses 56 (2001) 462–471.
6] S. Babel, T.A. Kurniawan, Low-cost adsorbents for heavy metals uptake from con-
taminated water: a review, J. Hazard. Mater. B97 (2003) 219–243.
(2015) 41–47.
[
[
32] J. Xie, Q. Wu, D. Zhao, Electrospinning synthesis of ZnFe O /Fe O /Ag nano-
2 4 3 4
7] H. Ali, E. Khan, M.A. Sajad, Phytoremediation of heavy metals—concepts and ap-
particle-loaded mesoporous carbon fibers with magnetic and photocatalytic prop-
erties, Carbon 50 (2012) 800–807.
33] C. Yang, W. Dong, G. Cui, Y. Zhao, X. Shi, X. Xia, B. Tang, W. Wang, Highly-efficient
plications, Chemosphere 91 (2013) 869–881.
8] M. Hua, S. Zhang, B. Pan, W. Zhang, Q. Zhang, Heavy metal removal from water/
wastewater by nanosized metal oxides: a review, J. Hazard. Mater. 211–212 (2012)
photocatalytic degradation of methylene blue by PoPD-modified TiO
composites due to photosensitization-synergetic effect of TiO with PoPD, Sci. Rep.
(2017) 3973 12 pages.
2
nano-
3
17–331.
9] M.A. Fox, M.T. Dulay, Heterogeneous photocatalysis, Chem. Rev. 83 (1995)
41–357.
10] F.S. Zhang, J. Nriagu, H. Itoh, Photocatalytic removal and recovery of mercury from
water using TiO -modified sewage sludge carbon, J. Photochem. Photobiol. A 167
2004) 223–228.
11] R. Liu, M. Lv, Q. Wang, H. Li, P. Guo, X.S. Zhao, Solvothermal synthesis of size-
tunable ZnFe colloidal nanocrystal assemblies and their electrocatalytic activity
towards hydrogen peroxide, J. Magn. Magn. Mater. 424 (2017) 155–160.
12] P. Dai, M. Wu, T. Jiang, H. Hu, X. Yu, Z. Bai, Silver-loaded ZnO/ZnFe
2
[
7
3
[
34] J. Feng, Y. Hou, X. Wang, W. Quan, J. Zhang, Y. Wang, L. Li, In-depth study on
adsorption and photocatalytic performance of novel reduced graphene oxide-
ZnFe O -polyaniline composites, J. Alloys Compd. 681 (2016) 157–166.
2 4
35] M. Tan, J.E. Parkin, Route of decomposition of thiomersal (thimerosal), Int. J.
[
[
[
[
[
2
(
[
[
Pharm. 208 (2000) 23–34.
2 4
O
36] J. Qvarnstrom, L. Lambertsson, S. Havarinasab, P. Hultman, W. Frech,
Determination of methylmercury, ethylmercury, and inorganic mercury in mouse
tissues, following administration of thimerosal, by species-specific isotope dilution
GC-inductively coupled plasma-MS, Anal. Chem. 75 (2003) 4120–4124.
37] L.B. Khalil, M.W. Rophael, W.E. Mourad, The removal of the toxic Hg(II) salts from
water by photocatalysis, Appl. Catal. B: Environ. 36 (2002) 125–130.
38] R.D. Vidic, J.B. McLaughlin, Uptake of elemental mercury vapors by activated
carbons, Air Waste Manage. Assoc. 46 (1996) 241–250.
2 4
O meso-
porous hollow spheres with enhanced photocatalytic activity for 2,4-dichlorophenol
degradation under visible light irradiation, Mater. Res. Bull. 107 (2018) 339–346.
13] R. Tao, C. Zhao, C. Shao, X. Li, X. Li, J. Zhang, S. Yang, Y. Liu, Bi WO /ZnFe O
2 6 2 4
[
[
heterostructures nanofibers: enhanced visible-light photocatalytic activity and
magnetically separable property, Mater. Res. Bull. 104 (2018) 124–133.
14] P. Muthirulan, C.K. Nirmala Devi, M. Meenakshi Sundaram, Facile synthesis of
novel hierarchical TiO @poly(o-phenylenediamine) core–shell structures with
2
15