Accepted Manuscript Pd-loaded magnetic mesoporous nanocomposites: A magnetically recoverable catalyst with effective enrichment and high activity for DDT and DDE removal under mild conditions Hua Tian, Jun Chen, Junhui He, Feng Liu PII: DOI: Reference:
S0021-9797(15)30045-X http://dx.doi.org/10.1016/j.jcis.2015.07.024 YJCIS 20582
To appear in:
Journal of Colloid and Interface Science
Received Date: Revised Date: Accepted Date:
8 April 2015 24 June 2015 12 July 2015
Please cite this article as: H. Tian, J. Chen, J. He, F. Liu, Pd-loaded magnetic mesoporous nanocomposites: A magnetically recoverable catalyst with effective enrichment and high activity for DDT and DDE removal under mild conditions, Journal of Colloid and Interface Science (2015), doi: http://dx.doi.org/10.1016/j.jcis.2015.07.024
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Pd-loaded magnetic mesoporous nanocomposites: A magnetically recoverable catalyst with effective enrichment and high activity for DDT and DDE removal under mild conditions
Hua Tian a, Jun Chen b, Junhui He a, *, Feng Liu a,c a
Functional Nanomaterials Laboratory, Center for Micro/Nanomaterials Laboratory, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100190, China b
c
Beijing Institute of Microchemistry, Beijing 100091, China
School of Chemical & Environmental Engineering, China University of Mining & Technology, Beijing 100083, China
*
Corresponding author: Junhui He Tel/Fax: +86-10-82543535. E-mail: [email protected].
1
ABSTRACT 1,1,1-trichloro-2,2-bis(4-chlorophenyl) ethane (DDT), an organochlorine pollutant, is highly persistent in environment and responsible for many ecological and health damages. Although remediation and degradation of DDT and its metabolites in soil and water by microorganisms and abiotic techniques can be accomplished, success is often accompanied by rigorous reaction conditions, such as anaerobic system, explosive gases, high pressure or temperature, and illumination. Here a triple-functional nanocomposite was prepared by integrating superparamagnetic Fe3O4 and palladium (Pd) nanoparticles onto mesoporous Fe3O4@nSiO2 @mSiO2 nanospheres. These magnetic mesoporous materials display excellent capabilities of capturing and catalytically degrading DDT in water. Over these nanocomposites, DDT and its metabolite, 1,1-dichloro-2,2-bis(4-chlorophenyl) ethylene (DDE) could be quickly enriched and completely degraded at as low as 150
o
C. The
nanocomposites can be magnetically separated from the dispersion after adsorption, and then be easily regenerated which is accompanied by catalytic reaction. The whole treatment process is convenient, energy-saving, and just requires ambient pressure and mild reaction conditions.
Keywords:
Nanocomposites;
Magnetic
degradation; Organochlorine pollutant
2
separation;
Adsorption;
Catalytic
1. Introduction The increasing agricultural and industrial use of organochlorine compounds, such as polychlorinated biphenyls (PCBs) and organochlorine pesticides (OCPs), had led to a widespread contamination of the environment. Even almost thirty years after they were banned in China and other numerous countries [1], high concentrations of organochlorine compounds still exist in soil, sediment, water and air [2-4]. Their persistence, toxicity and bioaccumulation greatly affect the environmental qualities, and animal and human health [5]. For instance, 1,1,1-trichloro-2,2-bis(4-chlorophenyl) ethane (DDT), banned in the U.S. in 1973 but frequently encountered in the environment, has been reported to be responsible for the thinning of bird eggshells [6]. It is also suggested that the derivatives and isomers of DDT are endocrine disrupters causing decreased reproduction in wildlife [7]. 1,1-dichloro-2,2-bis(4-chlorophenyl) ethylene (DDE) is a recalcitrant metabolite of DDT, and has been reported to be more persistent than DDT [8]. Its toxicity causes many serious environmental problems worldwide [9]. Several methods or approaches have been developed for remediation of DDT and DDE, mainly including phytoremediation, bioremediation [10], incineration and abiotic degradation [11]. The first two methods generally undergo for a long reaction time (several weeks or years), and very few cultures are capable of fully degrading DDE [12]. Abiotic degradation usually adopts zero-valent metals and bimetallic systems such as Fe0, Fe0/Pd and Mg0/Pd. Recently, other catalysts (Pd/C, Pd/C-Et3N, 3
TiO2, etc.) have been developed for degrading DDT and its metabolites [13]. These materials have been proven to be effective for the degradation of DDT, and even DDE. However, they also have drawbacks that greatly hinder their applications in the actual environment. First, their high activities are usually accompanied with the employment of harsh reaction conditions, such as high pressure [14], continuous bubbling of hydrogen [15], reductants or alkaline solution [16] or illumination [17]; second, these materials are difficult to be separated from dispersion solution due to their small particle size. Therefore, for the convenient remediation of organochlorine pollutants in environment, design and synthesis of new materials with good removal activity, simple separation character and mild treatment system are a very challenging issue but have great realistic meaning now. Multifunctional materials, which usually assemble different materials with only one unique property into a single nanostructure, have drawn much attention for their superior versatile properties toward different types of applications [18]. One of the most popular and widely investigated multifunctional materials is metal oxide core-shell heterostructures with magnetic core and metal oxide or transition metal oxide shell. Because of their well-defined shapes, abundant exposed surface planes and synergistic effect between the core and shell, those core-shell materials exhibit promising properties and have been adopted in optics, biophysics, and catalysis [19-22]. Recently, core-shell photocatalytic Fe3O4@TiO2 nanocomposites [23], yolk-shell FexOy/Pd@mesoporous SiO2 [24], and Fe3O4/sepiolite composite [25] have been synthesized and used to adsorb or catalytically degrade organic pollutants, such 4
as 4-nitrophenol, bisphenol A and atrazine. The magnetic performance significantly simplifies the entire process of enrichment, and also facilitates the recycling of used materials [26]. Mesoporous nanostructures can not only offer high surface area, but also allow the fast diffusion of reactants or adsorbates [27]. In earlier reports, excellent adsorption performance and high catalytic activity of magnetic mesoporous silica nanocomposites had been shown for DDT. These Fe3O4@nSiO2 @mSiO2 materials exploited the magnetic property of Fe3O4 core, effective protection of nonporous silica shell from corroding in reaction solution, and high surface area of mesoporous silica outer layer [28]. It was found that over 90 % of DDT, DDD and DDE can be removed from water within 1 h, and completely dechlorinated at 450 oC. Further study found that the deposition of iron oxide on the outer layer of Fe3O4@nSiO2@mSiO2 could decrease the temperature of complete dechlorination of DDT and DDE from 450 oC to 350 oC [29]. However, this temperature is still too high for practical applications and low energy consumption. In this work, we synthesized a triple-functional nanocomposite based on the deposition of Pd nanoparticles onto mesoporous Fe3O4@nSiO2 @mSiO2 nanospheres, and investigated their adsorption performances and catalytic activities toward DDT and DDE. This noble metal-loaded magnetic nanocomposite was firstly adopted for the remediation and degradation of organochlorine compounds. It was found that at relatively low temperature (150 oC) and short time (2 h), DDT and DDE could be completely degraded. More importantly, the treatment process is eco-friendly, energy-saving and rapid, and just requires ambient pressure and mild reaction 5
conditions.
2. Materials and methods 2.1. Materials DDT was obtained from Aldrich, USA. n-Hexane (high-performance liquid chromatography (HPLC) grade) was acquired from the Tedia Company, USA. All other chemicals were of analytical grade and used as received. The water used was purified through a Millipore purification system. 2.2. Synthesis of Fe3O4@nSiO2 composites Fe3O4 nanospheres were synthesized according to the previous work [30]. In a typical synthesis of Fe3O4@nSiO2 nanocomposites, 0.2 g of as-prepared Fe3O4 nanospheres were dispersed in a solution containing 320 mL ethanol and 76 mL H 2O mixed with 4 mL aqueous ammonia solution (25 wt%). After ultrasonication vibration for 15 min, 4 mL of tetraethyl orthosilicate (TEOS) was added dropwise into the Fe3O4 mixture under violent stirring at room temperature. The reaction continued for 24 h. The products were separated by a magnet and washed with ethanol and water several times. The final product was collected, dried at 60 oC, and marked as FS. 2.3. Synthesis of Fe3O4@nSiO2@mSiO2 composites The mesoporous SiO2 shells were achieved by the following procedure: 0.05 g of the as-synthesized FS nanospheres were dispersed into a mixed solution of ethanol (90 mL), deionized water (120 mL), cetyltrimethyl ammonium bromide (CTAB, 0.25 g), and 1.5 mL of aqueous ammonia solution by ultrasonication for 30 min. Then 0.5 6
mL of TEOS was added into the above suspension under stirring. The reaction was maintained at room temperature for 24 h, the product was collected by magnetic separation, washed first with an ethanol/HCl solution (5mL of HCl (37%) in 95 mL of ethanol) to remove CTAB template [29], and then with ethanol and deionized water. The final product was dried at 60 oC, denoted as Fe3O4@nSiO2 @mSiO2 or FSS. 2.4. Synthesis of Fe3O4@nSiO2@ mSiO2/Pd nanocomposites Pd supported yolk-shell composites were synthesized by reduction of PdCl2 with ascorbic acid as reducing agent [31]. In a typical procedure, 0.032 g of PVP was dissolved in 90 mL of water. Under stirring at 80 oC in a water bath, 10 mL aqueous solution with 0.0026 g PdCl2 and 10 mL aqueous solution with 0.051 g ascorbic acid were added sequentially to the mixture solution. After reaction for 1 h, 0.075 g of FSS (dispersed in 50 mL of water) was added. The reaction was allowed to proceed for 4 h. Then the product was separated by using a magnet, washed with ethanol and water. Finally, the as-prepared Fe3O4@nSiO2@mSiO2/Pd composites were dried at 60 oC, marked as Pd/FSS. To get different Pd loading for Pd/FSS materials, the mass ratio of Pd and FSS was changed (0.5wt%, 1.0 wt% and 2.0 wt%), while the other reaction conditions were kept the same. The products were marked as 0.5Pd/FSS, 1.0Pd/FSS and 2.0Pd/FSS, respectively. 2.5. Characterization The phases of the microspheres were studied by a Bruker D8 Focus X-ray diffractometer with Cu Kα (λ= 0.154 nm) radiation. JEOL JEM-2100F transmission electron microscope (TEM) was employed to examine the nanostructure and shell 7
thickness of the materials. The Braunauer–Emmett–Teller (BET) surface area measurements were carried out at 77 K using a Quadrasorb SI automated surface area and pore size analyzer. Magnetization measurements were performed at 295 K on a Lake Shore 7307 Vibrating sample magnetometer. The contents of Pd loading on the yolk-shell
magnetic
nanocomposites
were
obtained
using
a
PerkinElmer
NexION300D inductively coupled plasma mass spectroscopy (ICP-MS). The concentrations of DDT and its metabolites before and after the tests were determined using a Shimazu QP2010 gas chromatograph-mass spectrometer (GC-MS). DB-5 capillary column (0.25 μ m × 0.25 mm × 30 m) was used with ultra high purity helium as the carrier gas. The column temperature was ramped as follows: The initial oven temperature was 95°C with holding time of 2 min, and then the temperature was increased to 220°C at 20°C min-1, followed by an increase of 10°C min-1 to 280°C, holding at 280°C for 5 min. Determination of DDT’s metabolites during the catalytic dechlorination reaction was performed by an Agilent 6890N gas chromatograph equipped with Waters GCT Premier time of flight mass spectrometer (GC-TOFMS). 2.6. DDT adsorption and dechlorination DDT adsorption tests were carried in water-acetone (9:1, v/v) phase at room temperature. 30 mg of magnetic mesoporous samples were placed into 50 mL glass vials and 40 mL of 1.5 μg mL-1 DDT stock solution was added in each vial. The vials were tightly sealed with Teflon-lined screw caps to prevent any loss of solution, and shaken continuously at room temperature. At predetermined time intervals, vials were removed from the shaker, and the mixture was separated with a magnet. 20 mL of 8
supernatant solution was removed, and extracted three times with n-hexane for GC-MS analysis to confirm the residual DDT concentration. The adsorbent at the bottom of vial that reacted for 1 h was collected for the later dechlorination reaction. For reliability of adsorption data, the blank tests were also carried out in the same way. The adsorption efficiency of DDTs in solution was calculated by Eq. (1):
Removal(%)
C0 Ce 100 C0
(1)
The DDTs concentration retained on the adsorbent phase (qe, μg mg-1) was calculated by Eq. (2):
q
V (C0 C ) W
(2)
where, C0 (μg mL-1) and Ce (μg mL-1) are the initial and equilibrium concentrations of DDTs, V (mL) is the volume of the reaction solution, and W (mg) is the adsorbent mass. DDTs dechlorination experiments were performed in sealed glass tubes (i.d.=4 mm, l=100 mm). For each degradation reaction, 8 mg of used adsorbent with adsorbed DDT was put into a glass tube heated at a certain temperature for 2 h. After cooled to room temperature, the glass tube was carefully crushed. 15 mL of 25 mM NaOH solution was added into the solid residue to dissolve the mesoporous silica of adsorbent. The mixture was stirred for 2 h and extracted with n-hexane three times to recover DDT and its metabolites. The combined extracts were diluted with n-hexane for GC-MS and GC-TOFMS analysis. For simplicity, the obtained GM-MS 9
chromatograms were named as of the corresponding catalysts.
3. Results and discussion 3.1. Synthesis and characterization of materials The synthesis procedure, illustrated in Fig. 1, begins with the synthesis of Fe 3O4 nanospheres through a solvothermal method. Then, the Fe 3O4 nanospheres were firstly coated with a thin nonporous silica shell by a modified Stöber procedure, to protect the magnetic cores from corroding by reaction solution. Subsequently, a mesoporous silica layer was covered on the outer surface of as-obtained Fe3O4@nSiO2 nanospheres using CTAB as organic template. For avoiding the damage of magnetic core, the CTAB template in the nanoparticles was removed by extraction method (acid ethanol solution), instead of traditional calcination method. The product is denoted as Fe3O4@nSiO2@mSiO2. Next, at 80 oC, Pd nanoparticles were deposited onto the surface of Fe3O4@nSiO2 @mSiO2 nanospheres with PdCl2 as a precursor and ascorbic acid as a reducing agent, and Fe3O4@nSiO2 @mSiO2/Pd nanocomposites were prepared. TEM images in Fig. 2 show that Fe3O4 nanoparticles have spherical shape with an average diameter of about 295 nm. TEM images of Fig. 2b and c show the weaker contrast of the two nanoshells than Fe3O4 core. It reveals that after the next two synthesis step of coating, the Fe3O4 nanospheres were completely coated by a thin nonporous silica layer and then a thicker mesoporous silica layer. The size of nanospheres increases from 295 nm to 378 nm, and then 643 nm, respectively. The high resolution TEM (HRTEM) image (Fig. 2d) clearly shows the mesopores on the 10
outer shell of FSS nanospheres. They are perpendicular to the surface and the average pore diameter was measured to be 2.8 nm using the BET measurement (Table S1). This made reactants and products diffuse easily in these mesoporous channels. Fig. 2e displays a typical TEM image of Pd/FSS nanocomposites. Clearly, a number of Pd nanoparticles formed, and were well dispersed on the mesoporous silica shell. Some Pd nanoparticles tended to agglomerate with each other, covering some surface area or mesopores of FSS nanospheres. In addition, it was found that some macroporous structures appeared in the inner part of mesoporous silica shell. This was mainly caused by the decomposition of the mesoporous silica shell during the reducing process of PdCl2 by heating treatment. These results are in good agreement with the data obtained from the BET measurement. From the data of pore structure (Table S1), it is noted that after loading Pd nanoparticles, the surface area of nanocomposite decreases from 813 m2 g-1 to 639 m2 g-1, with the decrease of pore volume from 0.56 cm3 g-1 to 0.43 cm3 g-1. Fig. 2f gives a HRTEM image of Pd/FSS, which clearly shows the formation of Pd nanoparticles with average size of about 4.0 nm. It also shows a lattice spacing of 2.29 Å, corresponding to (111) facet of Pd, indicating that the oxidation state of Pd species on the FSS surface is Pd0. ICP-MS measurements show that the weight percentage of Pd in the 2.0Pd/FSS nanocomposites was about 0.32% (Table S2). These results demonstrate successful synthesis of Pd-loaded magnetic mesoporous nanocomposites with good dispersity of Pd nanoparticles. To optimize the Pd loading, we prepared magnetic mesoporous nanocomposites 11
with varied Pd content under otherwise identical conditions to those of 2.0Pd/FSS. Notably, the average sizes of Pd nanoparticles on FSS nanocomposites increased with increasing the Pd content. This observation accords with the nucleation and growth mechanisms of metal nanocrystals [32]. X-ray diffraction patterns exhibit well-resolved diffraction peaks for the obtained magnetic materials (Fig. 3a). All the peaks can be exactly indexed to the diffractions of Fe3O4 phase, suggesting that the silica coating process and surface modification had no effect on the crystal structure of Fe3O4 core. The much weaker peak intensities of the core-shell materials can be attributed to their low Fe3O4 proportions. There are no diffraction peaks corresponding to Pd nanoparticles due to their high dispersity and low content in the samples, which can be confirmed by the TEM image (Fig. 2) and data of ICP-MS measurement (Table S2). The magnetic properties (Fig. 3b and Fig. S2) of the obtained materials were investigated using a Lake Shore 7307 Vibrating sample magnetometer. The zero coercivity and the reversible hysteresis behavior indicate the superparamagnetic nature of both Fe3O4 core and core-shell nanocomposites. At 295 K, the saturation magnetization (Ms) values of Fe3O4 nanoparticles, FS, FSS and 2.0Pd/FSS nanocomposites are 64.7, 50.5, 15.3, and 17.5 emu g-1, respectively. The decrease in Ms may be due to the presence of the nonmagnetic silica, and the increase in M s of Pd/FSS may be caused by the destruction of the mesoporous silica shell [29]. Although the Ms decreased after coating the Fe3O4 core with the silica shell, the core-shell nanocomposites inherit the strongly superparamagnetic property from the Fe3O4 core. Moreover, due to the 12
coating of nonporous silica shell, complete magnetic separation of Pd/FSS materials can not only be achieved in neutral water, but also in strong acid solution, upon applying an external magnetic field. The inset graph of Fig. 3b shows an excellent dispersion ability of the 1.0Pd/FSS nanocomposites in HNO 3 solution (3 mol L-1) and magnetic response ability. As an external magnet was applied, 1.0Pd/FSS nanoparticles were attracted in 1 min, leaving the acid solution transparent. After taking away the magnet later, the nanoparticles could be dispersed into the solution again. Though Pd nanoparticles could be partly decomposed in such strong acid solution, this test suggests that these Pd/FSS nanocomposites can not only be used in mild aqueous system, but also acid solution. In another control experiment, bare Fe3O4 nanospheres were added into the identical HNO3 solution, and then quickly dissolved within 30 min. 3.2. Adsorption and catalytic performances toward DDTs As commonly known, mesoporous silica is a good adsorbent for organic pollutants. Therefore, before studying the catalytic activity of FSS and Pd/FSS nanocomposites, we investigated the adsorption of DDT in the presence of these magnetic mesoporous nanocomposites. The kinetics results of DDT removal over these magnetic mesoporous materials suggested a rapid initial uptake and a subsequent saturation (Fig. 4a). Over FSS, the DDT removal increased quickly within the initial 30 min and reached an adsorption equilibrium at 60 min when a removal efficiency of ca. 97% was achieved. Pd/FSS has a similar adsorption process and good adsorption ability, but slightly slower and lower than that of FSS. In general, the adsorption process 13
takes place by two consecutive processes, fast diffusion and slow reaction. TEM measurements display that the Pd loading not only blocks some of surface pores of mesoporous silica shell, but also partly destroys the ordered mesoporous silica shell (as pointed by arrows in Fig. 2e) during the reducing process of PdCl2. The former decreases the rate of DDT diffusion and the later decreases the reaction sites for DDT adsorption. Fig. 4b shows the effect of Pd content on adsorption efficiency of DDT. The experimental results suggest that all of these materials reveal excellent adsorption ability in one isolation process within 60 min. A decrease in adsorption ability with the increase of Pd loading occurs. The reasons may be that the outer mesoporous shell was destroyed partially (see Fig. 2), and the surface area and pore volume were reduced as the Pd nanoparticle size increased (see Table S1 and 2). To study the catalytic activity of magnetic mesoporous nanocomposites, we investigated the variation in the residual amount of adsorbed DDT on nanocomposites and produced metabolites as a function of treatment temperature. The DDT decomposition process was surveyed by monitoring the GC/MS total ion chromatogram (Fig. S2). An intense peak for DDE was observed just at ambient temperature of 20 oC, meaning the decomposition reaction over 1.0Pd/FSS could occur without any heat treatment. Meanwhile, we could anticipate that the first step in the dechlorination reaction was a dehydrochlorination reaction. Quantitative analysis (Fig. 5) reveals that the mass balance between DDT and produced DDE was not achieved, meaning the formation of DDE was accompanied by its subsequent dechlorination. After treatment at 150 oC for 2 h, all of peaks for DDT and DDE were 14
no longer observed in the chromatogram of 1.0Pd/FSS. Differences in the DDTs dechlorination efficiency and the Pd content were also explored as a function of treatment temperature. As shown in Fig. 5, over these three catalysts, the residual amount of DDT decreases with increase in the treatment temperature, suggesting the enhanced activity of these catalysts with temperature. The produced DDE was degraded in a similar trend to DDT, although the reaction results appeared somewhat more complicated than that of DDT over 0.5Pd/FSS and 2.0Pd/FSS. The produced DDE was consumed quickly before 60
o
C, but increased when the reaction
temperature increased to 150 oC. It is attributed to the decrease of the dechlorination rate of DDE at 150 oC by comparison with that at 60 oC. As a whole, the Pd content greatly affects the dechlorination efficiency of DDT and DDE. The introduction of Pd specie creates the excellent efficiency for DDT dechlorination, but larger loading leads to a decrease of dechlorination efficiency. As displayed clearly in Fig. S3, the optimum Pd content is 1.0 wt% in the current experiment range. With this Pd content, all of DDT and DDE could be completely consumed at 150 oC. This temperature is, to our best knowledge, the lowest for the thermal decomposition of DDT and DDE as compared to previous reports. An important consideration for the application of nanomaterials in water remediation is their separation and reusability.
FSS and 1.0Pd/FSS were chosen to
investigate the multiple usages of these magnetic mesoporous catalysts. The catalyst nanoparticles were magnetically separated from the DDT solution, and then were reused for the removal of a freshly DDT solution after the catalytic decomposition of 15
DDT at 150 oC. The nanoparticles were again magnetically separated and the adsorption and decomposition of DDT were repeated. The results (Fig. 6) indicate that in 4 cycles, the adsorption ability of DDT over FSS remained highly stable. The efficiency over Pd/FSS had a slight decrease, but still maintained 90% of the first use. 3.3. Mechanism and pathways of DDT dechlorination over Pd/FSS The above results indicate that the decomposition of DDT over Pd/FSS just gave DDE. DDD, another first intermediate, was not found under the current experimental conditions. These results indicate that DDT underwent a dehydrochlorination pathway to form DDE, which agrees with the results of previous reports for DDT decomposition under aerobic conditions [33]. The dechlorination of the alkyl chlorines of DDT would involve a single electron transfer process (Fig. 7), where Pd(0) works as an electron donor as well as a HCl scavenger. It was reported that 98% of the electron density in the LUMO of p,p’-DDT is localized in aliphatic carbon-chlorine σ antibonding orbital [34]. Initial single electron transfers from Pd(0) to this orbital, and a chloride anion and an alkyl radical intermediate of DDT are obtained. This intermediate subsequently undergoes an abstraction of the hydrogen at the benzylic position by the chloride radical to form DDE. The produced hydrogen chloride is scavenged by water or heating evaporation, which also promotes the dehydrochlorination rate. Based on the GC-MS and GC-TOFMS spectra of the products arising from the decomposition of DDT over 1.0Pd/FSS, a number of intermediates and products were detected. The degradation could be proposed to go a cleaving pathway (Fig. 8). In the 16
current experimental systems, the first reaction step is the base promoted dehydrochlorination at the Pd/FSS surface, leading to formation of DDE. Two chlorine atoms attached with the alkene moiety of DDE are dechlorinated to generate DDNU, and subsequent addition reaction of the alkene affords DDOH. The resulting DDOH could be directly transformed to DBP, or undergo stepwise degradation to afford DBP. Different from most common pathways of DDT transformation described in previous reports [13, 35, 36], several metabolites of DBP were successfully observed with the help of GC-TOFMS, such as BP, diphenylmethane, bibenzene, and so on. Thus a more complete pathway of DDT degradation than previously proposed can be obtained, as shown in Fig. 8. In the subsequent degradation, DBP gives two degradation pathways. In one pathway DBP undergoes substitution and reduction reactions to form diphenylmethane, and in another case, DBP transforms 4,4’-dichlorobiphenyl,
4-(4-chlorophenyl)phenol
and
then
bibenzene
by
carbonyl-extrusion and reduction reactions.
4. Conclusion In summary, a triple-functional nanocomposite for the complete removal of DDT from water was synthesized. The remarkable features of magnetic separation, adsorption and catalytic abilities render the material a promising candidate for the removal of DDT and other organochlorines. Among these materials, 1.0Pd/FSS exhibits the best performance with a low temperature of 150 oC for the complete degradation of DDT and DDE, without dangerous gases, irradiation, anaerobic treatment and pressurization. These materials could be easily and quickly separated 17
from reaction solutions by an external magnetic field and exhibit good reusability. No significant decrease in adsorption capability and catalytic activity of Pd/FSS nanocomposites were detected, even after the removal experiments were consecutively repeated 4 times. In addition, the results indicate that Pd/FSS nanocomposites have high stability even in acid solutions.
Acknowledgements This work was supported by the National Natural Science Foundation of China (Grant No. 21007076 and 21271177), and Technical Institute of Physics and Chemistry, Chinese Academy of Sciences. The authors are very grateful to Mr. Binbin Jin for his help in the preparation of the schematic illustration of the preparation procedure of Fe3O4@nSiO2@mSiO2/Pd nanocomposites.
References [1] B. Lopes, J.P. Arrebola, A. Serafim, R. Company, J. Rosa, N. Olea, Chemosphere 114 (2014) 291-302. [2] P.A. Pavlova, P. Schmid, C. Bogdal, C. Steinlin, T.M. Jenk, M. Schwikowski, Environ. Sci. Technol. 48 (2014) 7842-7848. [3] Y. Yu, Y. Li, Z. Shen, Z. Yang, L. Mo, Y. Kong, I. Lou, Chemosphere 114 (2014) 136-143. [4] P. Gong, X.P. Wang, S.H. Li, W.S. Yu, J.L. Li, D.B. Kattel, W.C. Wang, L.P. Devkota, T.D. Yao, D.R. Joswiak, Environ. Pollut.192 (2014) 44-51. 18
[5] H. Bouwman, H. Kylin, B. Sereda, R. Bornman, Environ. Pollut. 170 (2012) 63-70. [6] T. Heberer, U. Dunnbier, Environ. Sci.Technol. 33 (1999) 2346-2351. [7] S. RM, Nature 375 (1995) 538-539. [8] H.L. Boul, M.L. Garnham, D. Hucker, D. Baird, J. Aislabie, Environ. Sci.Technol. 28 (1994) 1397-1402. [9] P.A. Nguyen, T.H. Trinh, Y. Fukumitsu, J. Shimodaira, K. Miyauchi, M. Tokuda, D. Kasai, E. Masai, M. Fukuda, J. Biosci. Bioeng. 116 (2013) 91-100. [10] I. Ortiz, A. Velasco, S. Le Borgne, S. Revah, Biodegradation 24 (2013) 215-225. [11] M. Cao, L. Wang, L. Wang, J. Chen, X. Lu, Chemosphere 90 (2013) 2303-2308. [12] J.e. Thomas, L.-T. Ou, A. Al-Agely, DDE Remediation and Degradation, 2008. [13]S. Sudharshan, R. Naidu, M. Mallavarapu, N. Bolan, Biodegradation 23 (2012) 851-863. [14] A. Ghattas, R. Abu-Reziq, D. Avnir, J. Blum, Green Chem. 5 (2003) 40-43. [15] S.S. Zinovyev, N.A. Shinkova, A. Perosa, P. Tundo, Appl. Catal. B 55 (2005) 39-48. [16] Y. Ukisu, J. Hazard. Mater. 152 (2008) 287-292. [17] X. Quan, X. Zhao, S. Chen, H. Zhao, J. Chen, Y. Zhao, Chemosphere 60 (2005) 266-273. [18] X. Zhang, Z. Bi, W. He, G. Yang, H. Liu, Y. Yue, Energy Environ. Sci. 7 (2014) 2285-2294. [19] D.C. Watson, R.V. Martinez, Y. Fontana, E. Russo-Averchi, M. Heiss, A. 19
Fontcuberta i Morral, G.M. Whitesides, M. Lončar, Nano Lett. 14 (2014) 524-531. [20] L. Wang, M. Cao, Z. Ai, L. Zhang, Environ. Sci. Technol. 48 (2014) 3354-3362. [21] W.P. Li, P.Y. Liao, C.H. Su, C.S. Yeh, J. Am. Chem. Soc. 136 (2014) 10062-10075. [22] C. Li, R. Younesi, Y. Cai, Y. Zhu, M. Ma, J. Zhu, Appl. Catal. B 156-157 (2014) 314-322. [23] R.C.a.S. Vasudevan, ACS Nano 7 (2013) 4093-4104. [24] T. Yao, T. Cui, X. Fang, F. Cui, J. Wu, Nanoscale 5 (2013) 5896-5904. [25] H. Liu, W. Chen, C. Liu, Y. Liu, C. Dong, Micropor. Mesopor. Mater. 194 (2014) 72-78. [26] Q. Song, F. Wang, J. Cai, Y. Wang, J. Zhang, W. Yu, J. Xu, Energy Environ. Sci. 6 (2013) 994. [27] B. Liu, L.-M. Liu, X.-F. Lang, H.-Y. Wang, X.W. Lou, E.S. Aydil, Energy Environ. Sci. 7 (2014) 2592. [28] F. Liu, H. Tian, J. He, J. Colloid Interf. Sci. 419 (2014) 68-72. [29] H. Tian, F. Liu, J. He, J. Colloid Interf. Sci. 431 (2014) 90-96. [30] H. Chen, C. Deng, X. Zhang, Angew. Chem. Int. Edit. 49 (2010) 607-611. [31] X. Yu, J. He, D. Wang, Y. Hu, H. Tian, Z. He, J. Phys. Chem. C 116 (2012) 851-860. [32] Y. Xia, Y. Xiong, B. Lim, S.E. Skrabalak, Angew. Chem. Int. Edit. 48 (2009) 60-103. [33] T. Eggen, A. Majcherczyk, Chemosphere 62 (2006) 1116-1125. 20
[34] F.A. Beland, S.O. Farwell, P.R. Callis, R.D. Geer, J. Electroanal. Chem. 78 (1977) 145-159. [35] G.D. Sayles, G. You, M. Wang, M.J. Kupferle, Environ. Sci. Technol. 31 (1997) 3448-3454. [36] C. Lin, T.C. Chang, Chemosphere 66 (2007) 1003-1011.
21
Fig.1. Schematic illustration of the preparation procedure of Fe3O4@nSiO2 @mSiO2/Pd composites.
Fig.2. TEM images of magnetic mesoporous nanoparticles: (a) Fe3O4, (b) FS; (c) FSS, (e) 1.0Pd/FSS; (d) and (f) are high-resolution TEM images of (c) and (e), respectively. Red circles in (e) highlight Pd nanoparticles.
22
Fig. 3. (a) XRD patterns and (b) magnetic hysteresis loops of magnetic nanocomposites. The inset of (b) is a separation-redispersion process of 1.0Pd/FSS sample in HNO3 solution.
23
Fig. 4. (a) Kinetics of DDT adsorption over magnetic mesoporous nanocomposites and (b) effect of Pd content on the adsorption efficiency of DDT.
24
DDE
DDT 1.0
27
2.0Pd/FSS
1.0Pd/FSS
0.5Pd/FSS
24
0.8
18
0.6
15 12
0.4
qe/q0 (DDE)
qe/q0 (DDT)
21
9 6
0.2
3 0.0 0
30
60
90
120 150 0 o
Reaction temperature ( C)
30
60
90
120 150 0 o
Reaction temperature ( C)
0 30
60
90
120 150 o
Reaction temperature ( C)
Fig. 5. Effect of Pd content of the Pd/FSS materials on the degradation efficiencies of DDTs.
Fig. 6. Adsorption efficiencies of DDT in successive cycles of treatment process with 30 mg of catalysts. 25
Fig. 7. Proposed mechanisms for DDT degradation on Pd/FSS.
26
Fig. 8. Proposed degradation pathways of DDT over magnetic mesoporous catalysts. An underlined word indicates that the chemical was detected in the current work.
27
Graphical abstract for
Pd-loaded magnetic mesoporous nanocomposites: A magnetically recoverable catalyst with effective enrichment and high activity for DDT and DDE removal under mild conditions Hua Tian a, Jun Chen b, Junhui He a, *, Feng Liu a,c a
Functional Nanomaterials Laboratory, Center for Micro/Nanomaterials Laboratory, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100190, China b c
Beijing Institute of Microchemistry, Beijing 100091, China
School of Chemical & Environmental Engineering, China University of Mining & Technology, Beijing 100083, China
*
Corresponding author. Fax: +86 10 82543535. E-mail address: [email protected]
28
29
S0021-9797(15)30045-X http://dx.doi.org/10.1016/j.jcis.2015.07.024 YJCIS 20582
To appear in:
Journal of Colloid and Interface Science
Received Date: Revised Date: Accepted Date:
8 April 2015 24 June 2015 12 July 2015
Please cite this article as: H. Tian, J. Chen, J. He, F. Liu, Pd-loaded magnetic mesoporous nanocomposites: A magnetically recoverable catalyst with effective enrichment and high activity for DDT and DDE removal under mild conditions, Journal of Colloid and Interface Science (2015), doi: http://dx.doi.org/10.1016/j.jcis.2015.07.024
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Pd-loaded magnetic mesoporous nanocomposites: A magnetically recoverable catalyst with effective enrichment and high activity for DDT and DDE removal under mild conditions
Hua Tian a, Jun Chen b, Junhui He a, *, Feng Liu a,c a
Functional Nanomaterials Laboratory, Center for Micro/Nanomaterials Laboratory, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100190, China b
c
Beijing Institute of Microchemistry, Beijing 100091, China
School of Chemical & Environmental Engineering, China University of Mining & Technology, Beijing 100083, China
*
Corresponding author: Junhui He Tel/Fax: +86-10-82543535. E-mail: [email protected].
1
ABSTRACT 1,1,1-trichloro-2,2-bis(4-chlorophenyl) ethane (DDT), an organochlorine pollutant, is highly persistent in environment and responsible for many ecological and health damages. Although remediation and degradation of DDT and its metabolites in soil and water by microorganisms and abiotic techniques can be accomplished, success is often accompanied by rigorous reaction conditions, such as anaerobic system, explosive gases, high pressure or temperature, and illumination. Here a triple-functional nanocomposite was prepared by integrating superparamagnetic Fe3O4 and palladium (Pd) nanoparticles onto mesoporous Fe3O4@nSiO2 @mSiO2 nanospheres. These magnetic mesoporous materials display excellent capabilities of capturing and catalytically degrading DDT in water. Over these nanocomposites, DDT and its metabolite, 1,1-dichloro-2,2-bis(4-chlorophenyl) ethylene (DDE) could be quickly enriched and completely degraded at as low as 150
o
C. The
nanocomposites can be magnetically separated from the dispersion after adsorption, and then be easily regenerated which is accompanied by catalytic reaction. The whole treatment process is convenient, energy-saving, and just requires ambient pressure and mild reaction conditions.
Keywords:
Nanocomposites;
Magnetic
degradation; Organochlorine pollutant
2
separation;
Adsorption;
Catalytic
1. Introduction The increasing agricultural and industrial use of organochlorine compounds, such as polychlorinated biphenyls (PCBs) and organochlorine pesticides (OCPs), had led to a widespread contamination of the environment. Even almost thirty years after they were banned in China and other numerous countries [1], high concentrations of organochlorine compounds still exist in soil, sediment, water and air [2-4]. Their persistence, toxicity and bioaccumulation greatly affect the environmental qualities, and animal and human health [5]. For instance, 1,1,1-trichloro-2,2-bis(4-chlorophenyl) ethane (DDT), banned in the U.S. in 1973 but frequently encountered in the environment, has been reported to be responsible for the thinning of bird eggshells [6]. It is also suggested that the derivatives and isomers of DDT are endocrine disrupters causing decreased reproduction in wildlife [7]. 1,1-dichloro-2,2-bis(4-chlorophenyl) ethylene (DDE) is a recalcitrant metabolite of DDT, and has been reported to be more persistent than DDT [8]. Its toxicity causes many serious environmental problems worldwide [9]. Several methods or approaches have been developed for remediation of DDT and DDE, mainly including phytoremediation, bioremediation [10], incineration and abiotic degradation [11]. The first two methods generally undergo for a long reaction time (several weeks or years), and very few cultures are capable of fully degrading DDE [12]. Abiotic degradation usually adopts zero-valent metals and bimetallic systems such as Fe0, Fe0/Pd and Mg0/Pd. Recently, other catalysts (Pd/C, Pd/C-Et3N, 3
TiO2, etc.) have been developed for degrading DDT and its metabolites [13]. These materials have been proven to be effective for the degradation of DDT, and even DDE. However, they also have drawbacks that greatly hinder their applications in the actual environment. First, their high activities are usually accompanied with the employment of harsh reaction conditions, such as high pressure [14], continuous bubbling of hydrogen [15], reductants or alkaline solution [16] or illumination [17]; second, these materials are difficult to be separated from dispersion solution due to their small particle size. Therefore, for the convenient remediation of organochlorine pollutants in environment, design and synthesis of new materials with good removal activity, simple separation character and mild treatment system are a very challenging issue but have great realistic meaning now. Multifunctional materials, which usually assemble different materials with only one unique property into a single nanostructure, have drawn much attention for their superior versatile properties toward different types of applications [18]. One of the most popular and widely investigated multifunctional materials is metal oxide core-shell heterostructures with magnetic core and metal oxide or transition metal oxide shell. Because of their well-defined shapes, abundant exposed surface planes and synergistic effect between the core and shell, those core-shell materials exhibit promising properties and have been adopted in optics, biophysics, and catalysis [19-22]. Recently, core-shell photocatalytic Fe3O4@TiO2 nanocomposites [23], yolk-shell FexOy/Pd@mesoporous SiO2 [24], and Fe3O4/sepiolite composite [25] have been synthesized and used to adsorb or catalytically degrade organic pollutants, such 4
as 4-nitrophenol, bisphenol A and atrazine. The magnetic performance significantly simplifies the entire process of enrichment, and also facilitates the recycling of used materials [26]. Mesoporous nanostructures can not only offer high surface area, but also allow the fast diffusion of reactants or adsorbates [27]. In earlier reports, excellent adsorption performance and high catalytic activity of magnetic mesoporous silica nanocomposites had been shown for DDT. These Fe3O4@nSiO2 @mSiO2 materials exploited the magnetic property of Fe3O4 core, effective protection of nonporous silica shell from corroding in reaction solution, and high surface area of mesoporous silica outer layer [28]. It was found that over 90 % of DDT, DDD and DDE can be removed from water within 1 h, and completely dechlorinated at 450 oC. Further study found that the deposition of iron oxide on the outer layer of Fe3O4@nSiO2@mSiO2 could decrease the temperature of complete dechlorination of DDT and DDE from 450 oC to 350 oC [29]. However, this temperature is still too high for practical applications and low energy consumption. In this work, we synthesized a triple-functional nanocomposite based on the deposition of Pd nanoparticles onto mesoporous Fe3O4@nSiO2 @mSiO2 nanospheres, and investigated their adsorption performances and catalytic activities toward DDT and DDE. This noble metal-loaded magnetic nanocomposite was firstly adopted for the remediation and degradation of organochlorine compounds. It was found that at relatively low temperature (150 oC) and short time (2 h), DDT and DDE could be completely degraded. More importantly, the treatment process is eco-friendly, energy-saving and rapid, and just requires ambient pressure and mild reaction 5
conditions.
2. Materials and methods 2.1. Materials DDT was obtained from Aldrich, USA. n-Hexane (high-performance liquid chromatography (HPLC) grade) was acquired from the Tedia Company, USA. All other chemicals were of analytical grade and used as received. The water used was purified through a Millipore purification system. 2.2. Synthesis of Fe3O4@nSiO2 composites Fe3O4 nanospheres were synthesized according to the previous work [30]. In a typical synthesis of Fe3O4@nSiO2 nanocomposites, 0.2 g of as-prepared Fe3O4 nanospheres were dispersed in a solution containing 320 mL ethanol and 76 mL H 2O mixed with 4 mL aqueous ammonia solution (25 wt%). After ultrasonication vibration for 15 min, 4 mL of tetraethyl orthosilicate (TEOS) was added dropwise into the Fe3O4 mixture under violent stirring at room temperature. The reaction continued for 24 h. The products were separated by a magnet and washed with ethanol and water several times. The final product was collected, dried at 60 oC, and marked as FS. 2.3. Synthesis of Fe3O4@nSiO2@mSiO2 composites The mesoporous SiO2 shells were achieved by the following procedure: 0.05 g of the as-synthesized FS nanospheres were dispersed into a mixed solution of ethanol (90 mL), deionized water (120 mL), cetyltrimethyl ammonium bromide (CTAB, 0.25 g), and 1.5 mL of aqueous ammonia solution by ultrasonication for 30 min. Then 0.5 6
mL of TEOS was added into the above suspension under stirring. The reaction was maintained at room temperature for 24 h, the product was collected by magnetic separation, washed first with an ethanol/HCl solution (5mL of HCl (37%) in 95 mL of ethanol) to remove CTAB template [29], and then with ethanol and deionized water. The final product was dried at 60 oC, denoted as Fe3O4@nSiO2 @mSiO2 or FSS. 2.4. Synthesis of Fe3O4@nSiO2@ mSiO2/Pd nanocomposites Pd supported yolk-shell composites were synthesized by reduction of PdCl2 with ascorbic acid as reducing agent [31]. In a typical procedure, 0.032 g of PVP was dissolved in 90 mL of water. Under stirring at 80 oC in a water bath, 10 mL aqueous solution with 0.0026 g PdCl2 and 10 mL aqueous solution with 0.051 g ascorbic acid were added sequentially to the mixture solution. After reaction for 1 h, 0.075 g of FSS (dispersed in 50 mL of water) was added. The reaction was allowed to proceed for 4 h. Then the product was separated by using a magnet, washed with ethanol and water. Finally, the as-prepared Fe3O4@nSiO2@mSiO2/Pd composites were dried at 60 oC, marked as Pd/FSS. To get different Pd loading for Pd/FSS materials, the mass ratio of Pd and FSS was changed (0.5wt%, 1.0 wt% and 2.0 wt%), while the other reaction conditions were kept the same. The products were marked as 0.5Pd/FSS, 1.0Pd/FSS and 2.0Pd/FSS, respectively. 2.5. Characterization The phases of the microspheres were studied by a Bruker D8 Focus X-ray diffractometer with Cu Kα (λ= 0.154 nm) radiation. JEOL JEM-2100F transmission electron microscope (TEM) was employed to examine the nanostructure and shell 7
thickness of the materials. The Braunauer–Emmett–Teller (BET) surface area measurements were carried out at 77 K using a Quadrasorb SI automated surface area and pore size analyzer. Magnetization measurements were performed at 295 K on a Lake Shore 7307 Vibrating sample magnetometer. The contents of Pd loading on the yolk-shell
magnetic
nanocomposites
were
obtained
using
a
PerkinElmer
NexION300D inductively coupled plasma mass spectroscopy (ICP-MS). The concentrations of DDT and its metabolites before and after the tests were determined using a Shimazu QP2010 gas chromatograph-mass spectrometer (GC-MS). DB-5 capillary column (0.25 μ m × 0.25 mm × 30 m) was used with ultra high purity helium as the carrier gas. The column temperature was ramped as follows: The initial oven temperature was 95°C with holding time of 2 min, and then the temperature was increased to 220°C at 20°C min-1, followed by an increase of 10°C min-1 to 280°C, holding at 280°C for 5 min. Determination of DDT’s metabolites during the catalytic dechlorination reaction was performed by an Agilent 6890N gas chromatograph equipped with Waters GCT Premier time of flight mass spectrometer (GC-TOFMS). 2.6. DDT adsorption and dechlorination DDT adsorption tests were carried in water-acetone (9:1, v/v) phase at room temperature. 30 mg of magnetic mesoporous samples were placed into 50 mL glass vials and 40 mL of 1.5 μg mL-1 DDT stock solution was added in each vial. The vials were tightly sealed with Teflon-lined screw caps to prevent any loss of solution, and shaken continuously at room temperature. At predetermined time intervals, vials were removed from the shaker, and the mixture was separated with a magnet. 20 mL of 8
supernatant solution was removed, and extracted three times with n-hexane for GC-MS analysis to confirm the residual DDT concentration. The adsorbent at the bottom of vial that reacted for 1 h was collected for the later dechlorination reaction. For reliability of adsorption data, the blank tests were also carried out in the same way. The adsorption efficiency of DDTs in solution was calculated by Eq. (1):
Removal(%)
C0 Ce 100 C0
(1)
The DDTs concentration retained on the adsorbent phase (qe, μg mg-1) was calculated by Eq. (2):
q
V (C0 C ) W
(2)
where, C0 (μg mL-1) and Ce (μg mL-1) are the initial and equilibrium concentrations of DDTs, V (mL) is the volume of the reaction solution, and W (mg) is the adsorbent mass. DDTs dechlorination experiments were performed in sealed glass tubes (i.d.=4 mm, l=100 mm). For each degradation reaction, 8 mg of used adsorbent with adsorbed DDT was put into a glass tube heated at a certain temperature for 2 h. After cooled to room temperature, the glass tube was carefully crushed. 15 mL of 25 mM NaOH solution was added into the solid residue to dissolve the mesoporous silica of adsorbent. The mixture was stirred for 2 h and extracted with n-hexane three times to recover DDT and its metabolites. The combined extracts were diluted with n-hexane for GC-MS and GC-TOFMS analysis. For simplicity, the obtained GM-MS 9
chromatograms were named as of the corresponding catalysts.
3. Results and discussion 3.1. Synthesis and characterization of materials The synthesis procedure, illustrated in Fig. 1, begins with the synthesis of Fe 3O4 nanospheres through a solvothermal method. Then, the Fe 3O4 nanospheres were firstly coated with a thin nonporous silica shell by a modified Stöber procedure, to protect the magnetic cores from corroding by reaction solution. Subsequently, a mesoporous silica layer was covered on the outer surface of as-obtained Fe3O4@nSiO2 nanospheres using CTAB as organic template. For avoiding the damage of magnetic core, the CTAB template in the nanoparticles was removed by extraction method (acid ethanol solution), instead of traditional calcination method. The product is denoted as Fe3O4@nSiO2@mSiO2. Next, at 80 oC, Pd nanoparticles were deposited onto the surface of Fe3O4@nSiO2 @mSiO2 nanospheres with PdCl2 as a precursor and ascorbic acid as a reducing agent, and Fe3O4@nSiO2 @mSiO2/Pd nanocomposites were prepared. TEM images in Fig. 2 show that Fe3O4 nanoparticles have spherical shape with an average diameter of about 295 nm. TEM images of Fig. 2b and c show the weaker contrast of the two nanoshells than Fe3O4 core. It reveals that after the next two synthesis step of coating, the Fe3O4 nanospheres were completely coated by a thin nonporous silica layer and then a thicker mesoporous silica layer. The size of nanospheres increases from 295 nm to 378 nm, and then 643 nm, respectively. The high resolution TEM (HRTEM) image (Fig. 2d) clearly shows the mesopores on the 10
outer shell of FSS nanospheres. They are perpendicular to the surface and the average pore diameter was measured to be 2.8 nm using the BET measurement (Table S1). This made reactants and products diffuse easily in these mesoporous channels. Fig. 2e displays a typical TEM image of Pd/FSS nanocomposites. Clearly, a number of Pd nanoparticles formed, and were well dispersed on the mesoporous silica shell. Some Pd nanoparticles tended to agglomerate with each other, covering some surface area or mesopores of FSS nanospheres. In addition, it was found that some macroporous structures appeared in the inner part of mesoporous silica shell. This was mainly caused by the decomposition of the mesoporous silica shell during the reducing process of PdCl2 by heating treatment. These results are in good agreement with the data obtained from the BET measurement. From the data of pore structure (Table S1), it is noted that after loading Pd nanoparticles, the surface area of nanocomposite decreases from 813 m2 g-1 to 639 m2 g-1, with the decrease of pore volume from 0.56 cm3 g-1 to 0.43 cm3 g-1. Fig. 2f gives a HRTEM image of Pd/FSS, which clearly shows the formation of Pd nanoparticles with average size of about 4.0 nm. It also shows a lattice spacing of 2.29 Å, corresponding to (111) facet of Pd, indicating that the oxidation state of Pd species on the FSS surface is Pd0. ICP-MS measurements show that the weight percentage of Pd in the 2.0Pd/FSS nanocomposites was about 0.32% (Table S2). These results demonstrate successful synthesis of Pd-loaded magnetic mesoporous nanocomposites with good dispersity of Pd nanoparticles. To optimize the Pd loading, we prepared magnetic mesoporous nanocomposites 11
with varied Pd content under otherwise identical conditions to those of 2.0Pd/FSS. Notably, the average sizes of Pd nanoparticles on FSS nanocomposites increased with increasing the Pd content. This observation accords with the nucleation and growth mechanisms of metal nanocrystals [32]. X-ray diffraction patterns exhibit well-resolved diffraction peaks for the obtained magnetic materials (Fig. 3a). All the peaks can be exactly indexed to the diffractions of Fe3O4 phase, suggesting that the silica coating process and surface modification had no effect on the crystal structure of Fe3O4 core. The much weaker peak intensities of the core-shell materials can be attributed to their low Fe3O4 proportions. There are no diffraction peaks corresponding to Pd nanoparticles due to their high dispersity and low content in the samples, which can be confirmed by the TEM image (Fig. 2) and data of ICP-MS measurement (Table S2). The magnetic properties (Fig. 3b and Fig. S2) of the obtained materials were investigated using a Lake Shore 7307 Vibrating sample magnetometer. The zero coercivity and the reversible hysteresis behavior indicate the superparamagnetic nature of both Fe3O4 core and core-shell nanocomposites. At 295 K, the saturation magnetization (Ms) values of Fe3O4 nanoparticles, FS, FSS and 2.0Pd/FSS nanocomposites are 64.7, 50.5, 15.3, and 17.5 emu g-1, respectively. The decrease in Ms may be due to the presence of the nonmagnetic silica, and the increase in M s of Pd/FSS may be caused by the destruction of the mesoporous silica shell [29]. Although the Ms decreased after coating the Fe3O4 core with the silica shell, the core-shell nanocomposites inherit the strongly superparamagnetic property from the Fe3O4 core. Moreover, due to the 12
coating of nonporous silica shell, complete magnetic separation of Pd/FSS materials can not only be achieved in neutral water, but also in strong acid solution, upon applying an external magnetic field. The inset graph of Fig. 3b shows an excellent dispersion ability of the 1.0Pd/FSS nanocomposites in HNO 3 solution (3 mol L-1) and magnetic response ability. As an external magnet was applied, 1.0Pd/FSS nanoparticles were attracted in 1 min, leaving the acid solution transparent. After taking away the magnet later, the nanoparticles could be dispersed into the solution again. Though Pd nanoparticles could be partly decomposed in such strong acid solution, this test suggests that these Pd/FSS nanocomposites can not only be used in mild aqueous system, but also acid solution. In another control experiment, bare Fe3O4 nanospheres were added into the identical HNO3 solution, and then quickly dissolved within 30 min. 3.2. Adsorption and catalytic performances toward DDTs As commonly known, mesoporous silica is a good adsorbent for organic pollutants. Therefore, before studying the catalytic activity of FSS and Pd/FSS nanocomposites, we investigated the adsorption of DDT in the presence of these magnetic mesoporous nanocomposites. The kinetics results of DDT removal over these magnetic mesoporous materials suggested a rapid initial uptake and a subsequent saturation (Fig. 4a). Over FSS, the DDT removal increased quickly within the initial 30 min and reached an adsorption equilibrium at 60 min when a removal efficiency of ca. 97% was achieved. Pd/FSS has a similar adsorption process and good adsorption ability, but slightly slower and lower than that of FSS. In general, the adsorption process 13
takes place by two consecutive processes, fast diffusion and slow reaction. TEM measurements display that the Pd loading not only blocks some of surface pores of mesoporous silica shell, but also partly destroys the ordered mesoporous silica shell (as pointed by arrows in Fig. 2e) during the reducing process of PdCl2. The former decreases the rate of DDT diffusion and the later decreases the reaction sites for DDT adsorption. Fig. 4b shows the effect of Pd content on adsorption efficiency of DDT. The experimental results suggest that all of these materials reveal excellent adsorption ability in one isolation process within 60 min. A decrease in adsorption ability with the increase of Pd loading occurs. The reasons may be that the outer mesoporous shell was destroyed partially (see Fig. 2), and the surface area and pore volume were reduced as the Pd nanoparticle size increased (see Table S1 and 2). To study the catalytic activity of magnetic mesoporous nanocomposites, we investigated the variation in the residual amount of adsorbed DDT on nanocomposites and produced metabolites as a function of treatment temperature. The DDT decomposition process was surveyed by monitoring the GC/MS total ion chromatogram (Fig. S2). An intense peak for DDE was observed just at ambient temperature of 20 oC, meaning the decomposition reaction over 1.0Pd/FSS could occur without any heat treatment. Meanwhile, we could anticipate that the first step in the dechlorination reaction was a dehydrochlorination reaction. Quantitative analysis (Fig. 5) reveals that the mass balance between DDT and produced DDE was not achieved, meaning the formation of DDE was accompanied by its subsequent dechlorination. After treatment at 150 oC for 2 h, all of peaks for DDT and DDE were 14
no longer observed in the chromatogram of 1.0Pd/FSS. Differences in the DDTs dechlorination efficiency and the Pd content were also explored as a function of treatment temperature. As shown in Fig. 5, over these three catalysts, the residual amount of DDT decreases with increase in the treatment temperature, suggesting the enhanced activity of these catalysts with temperature. The produced DDE was degraded in a similar trend to DDT, although the reaction results appeared somewhat more complicated than that of DDT over 0.5Pd/FSS and 2.0Pd/FSS. The produced DDE was consumed quickly before 60
o
C, but increased when the reaction
temperature increased to 150 oC. It is attributed to the decrease of the dechlorination rate of DDE at 150 oC by comparison with that at 60 oC. As a whole, the Pd content greatly affects the dechlorination efficiency of DDT and DDE. The introduction of Pd specie creates the excellent efficiency for DDT dechlorination, but larger loading leads to a decrease of dechlorination efficiency. As displayed clearly in Fig. S3, the optimum Pd content is 1.0 wt% in the current experiment range. With this Pd content, all of DDT and DDE could be completely consumed at 150 oC. This temperature is, to our best knowledge, the lowest for the thermal decomposition of DDT and DDE as compared to previous reports. An important consideration for the application of nanomaterials in water remediation is their separation and reusability.
FSS and 1.0Pd/FSS were chosen to
investigate the multiple usages of these magnetic mesoporous catalysts. The catalyst nanoparticles were magnetically separated from the DDT solution, and then were reused for the removal of a freshly DDT solution after the catalytic decomposition of 15
DDT at 150 oC. The nanoparticles were again magnetically separated and the adsorption and decomposition of DDT were repeated. The results (Fig. 6) indicate that in 4 cycles, the adsorption ability of DDT over FSS remained highly stable. The efficiency over Pd/FSS had a slight decrease, but still maintained 90% of the first use. 3.3. Mechanism and pathways of DDT dechlorination over Pd/FSS The above results indicate that the decomposition of DDT over Pd/FSS just gave DDE. DDD, another first intermediate, was not found under the current experimental conditions. These results indicate that DDT underwent a dehydrochlorination pathway to form DDE, which agrees with the results of previous reports for DDT decomposition under aerobic conditions [33]. The dechlorination of the alkyl chlorines of DDT would involve a single electron transfer process (Fig. 7), where Pd(0) works as an electron donor as well as a HCl scavenger. It was reported that 98% of the electron density in the LUMO of p,p’-DDT is localized in aliphatic carbon-chlorine σ antibonding orbital [34]. Initial single electron transfers from Pd(0) to this orbital, and a chloride anion and an alkyl radical intermediate of DDT are obtained. This intermediate subsequently undergoes an abstraction of the hydrogen at the benzylic position by the chloride radical to form DDE. The produced hydrogen chloride is scavenged by water or heating evaporation, which also promotes the dehydrochlorination rate. Based on the GC-MS and GC-TOFMS spectra of the products arising from the decomposition of DDT over 1.0Pd/FSS, a number of intermediates and products were detected. The degradation could be proposed to go a cleaving pathway (Fig. 8). In the 16
current experimental systems, the first reaction step is the base promoted dehydrochlorination at the Pd/FSS surface, leading to formation of DDE. Two chlorine atoms attached with the alkene moiety of DDE are dechlorinated to generate DDNU, and subsequent addition reaction of the alkene affords DDOH. The resulting DDOH could be directly transformed to DBP, or undergo stepwise degradation to afford DBP. Different from most common pathways of DDT transformation described in previous reports [13, 35, 36], several metabolites of DBP were successfully observed with the help of GC-TOFMS, such as BP, diphenylmethane, bibenzene, and so on. Thus a more complete pathway of DDT degradation than previously proposed can be obtained, as shown in Fig. 8. In the subsequent degradation, DBP gives two degradation pathways. In one pathway DBP undergoes substitution and reduction reactions to form diphenylmethane, and in another case, DBP transforms 4,4’-dichlorobiphenyl,
4-(4-chlorophenyl)phenol
and
then
bibenzene
by
carbonyl-extrusion and reduction reactions.
4. Conclusion In summary, a triple-functional nanocomposite for the complete removal of DDT from water was synthesized. The remarkable features of magnetic separation, adsorption and catalytic abilities render the material a promising candidate for the removal of DDT and other organochlorines. Among these materials, 1.0Pd/FSS exhibits the best performance with a low temperature of 150 oC for the complete degradation of DDT and DDE, without dangerous gases, irradiation, anaerobic treatment and pressurization. These materials could be easily and quickly separated 17
from reaction solutions by an external magnetic field and exhibit good reusability. No significant decrease in adsorption capability and catalytic activity of Pd/FSS nanocomposites were detected, even after the removal experiments were consecutively repeated 4 times. In addition, the results indicate that Pd/FSS nanocomposites have high stability even in acid solutions.
Acknowledgements This work was supported by the National Natural Science Foundation of China (Grant No. 21007076 and 21271177), and Technical Institute of Physics and Chemistry, Chinese Academy of Sciences. The authors are very grateful to Mr. Binbin Jin for his help in the preparation of the schematic illustration of the preparation procedure of Fe3O4@nSiO2@mSiO2/Pd nanocomposites.
References [1] B. Lopes, J.P. Arrebola, A. Serafim, R. Company, J. Rosa, N. Olea, Chemosphere 114 (2014) 291-302. [2] P.A. Pavlova, P. Schmid, C. Bogdal, C. Steinlin, T.M. Jenk, M. Schwikowski, Environ. Sci. Technol. 48 (2014) 7842-7848. [3] Y. Yu, Y. Li, Z. Shen, Z. Yang, L. Mo, Y. Kong, I. Lou, Chemosphere 114 (2014) 136-143. [4] P. Gong, X.P. Wang, S.H. Li, W.S. Yu, J.L. Li, D.B. Kattel, W.C. Wang, L.P. Devkota, T.D. Yao, D.R. Joswiak, Environ. Pollut.192 (2014) 44-51. 18
[5] H. Bouwman, H. Kylin, B. Sereda, R. Bornman, Environ. Pollut. 170 (2012) 63-70. [6] T. Heberer, U. Dunnbier, Environ. Sci.Technol. 33 (1999) 2346-2351. [7] S. RM, Nature 375 (1995) 538-539. [8] H.L. Boul, M.L. Garnham, D. Hucker, D. Baird, J. Aislabie, Environ. Sci.Technol. 28 (1994) 1397-1402. [9] P.A. Nguyen, T.H. Trinh, Y. Fukumitsu, J. Shimodaira, K. Miyauchi, M. Tokuda, D. Kasai, E. Masai, M. Fukuda, J. Biosci. Bioeng. 116 (2013) 91-100. [10] I. Ortiz, A. Velasco, S. Le Borgne, S. Revah, Biodegradation 24 (2013) 215-225. [11] M. Cao, L. Wang, L. Wang, J. Chen, X. Lu, Chemosphere 90 (2013) 2303-2308. [12] J.e. Thomas, L.-T. Ou, A. Al-Agely, DDE Remediation and Degradation, 2008. [13]S. Sudharshan, R. Naidu, M. Mallavarapu, N. Bolan, Biodegradation 23 (2012) 851-863. [14] A. Ghattas, R. Abu-Reziq, D. Avnir, J. Blum, Green Chem. 5 (2003) 40-43. [15] S.S. Zinovyev, N.A. Shinkova, A. Perosa, P. Tundo, Appl. Catal. B 55 (2005) 39-48. [16] Y. Ukisu, J. Hazard. Mater. 152 (2008) 287-292. [17] X. Quan, X. Zhao, S. Chen, H. Zhao, J. Chen, Y. Zhao, Chemosphere 60 (2005) 266-273. [18] X. Zhang, Z. Bi, W. He, G. Yang, H. Liu, Y. Yue, Energy Environ. Sci. 7 (2014) 2285-2294. [19] D.C. Watson, R.V. Martinez, Y. Fontana, E. Russo-Averchi, M. Heiss, A. 19
Fontcuberta i Morral, G.M. Whitesides, M. Lončar, Nano Lett. 14 (2014) 524-531. [20] L. Wang, M. Cao, Z. Ai, L. Zhang, Environ. Sci. Technol. 48 (2014) 3354-3362. [21] W.P. Li, P.Y. Liao, C.H. Su, C.S. Yeh, J. Am. Chem. Soc. 136 (2014) 10062-10075. [22] C. Li, R. Younesi, Y. Cai, Y. Zhu, M. Ma, J. Zhu, Appl. Catal. B 156-157 (2014) 314-322. [23] R.C.a.S. Vasudevan, ACS Nano 7 (2013) 4093-4104. [24] T. Yao, T. Cui, X. Fang, F. Cui, J. Wu, Nanoscale 5 (2013) 5896-5904. [25] H. Liu, W. Chen, C. Liu, Y. Liu, C. Dong, Micropor. Mesopor. Mater. 194 (2014) 72-78. [26] Q. Song, F. Wang, J. Cai, Y. Wang, J. Zhang, W. Yu, J. Xu, Energy Environ. Sci. 6 (2013) 994. [27] B. Liu, L.-M. Liu, X.-F. Lang, H.-Y. Wang, X.W. Lou, E.S. Aydil, Energy Environ. Sci. 7 (2014) 2592. [28] F. Liu, H. Tian, J. He, J. Colloid Interf. Sci. 419 (2014) 68-72. [29] H. Tian, F. Liu, J. He, J. Colloid Interf. Sci. 431 (2014) 90-96. [30] H. Chen, C. Deng, X. Zhang, Angew. Chem. Int. Edit. 49 (2010) 607-611. [31] X. Yu, J. He, D. Wang, Y. Hu, H. Tian, Z. He, J. Phys. Chem. C 116 (2012) 851-860. [32] Y. Xia, Y. Xiong, B. Lim, S.E. Skrabalak, Angew. Chem. Int. Edit. 48 (2009) 60-103. [33] T. Eggen, A. Majcherczyk, Chemosphere 62 (2006) 1116-1125. 20
[34] F.A. Beland, S.O. Farwell, P.R. Callis, R.D. Geer, J. Electroanal. Chem. 78 (1977) 145-159. [35] G.D. Sayles, G. You, M. Wang, M.J. Kupferle, Environ. Sci. Technol. 31 (1997) 3448-3454. [36] C. Lin, T.C. Chang, Chemosphere 66 (2007) 1003-1011.
21
Fig.1. Schematic illustration of the preparation procedure of Fe3O4@nSiO2 @mSiO2/Pd composites.
Fig.2. TEM images of magnetic mesoporous nanoparticles: (a) Fe3O4, (b) FS; (c) FSS, (e) 1.0Pd/FSS; (d) and (f) are high-resolution TEM images of (c) and (e), respectively. Red circles in (e) highlight Pd nanoparticles.
22
Fig. 3. (a) XRD patterns and (b) magnetic hysteresis loops of magnetic nanocomposites. The inset of (b) is a separation-redispersion process of 1.0Pd/FSS sample in HNO3 solution.
23
Fig. 4. (a) Kinetics of DDT adsorption over magnetic mesoporous nanocomposites and (b) effect of Pd content on the adsorption efficiency of DDT.
24
DDE
DDT 1.0
27
2.0Pd/FSS
1.0Pd/FSS
0.5Pd/FSS
24
0.8
18
0.6
15 12
0.4
qe/q0 (DDE)
qe/q0 (DDT)
21
9 6
0.2
3 0.0 0
30
60
90
120 150 0 o
Reaction temperature ( C)
30
60
90
120 150 0 o
Reaction temperature ( C)
0 30
60
90
120 150 o
Reaction temperature ( C)
Fig. 5. Effect of Pd content of the Pd/FSS materials on the degradation efficiencies of DDTs.
Fig. 6. Adsorption efficiencies of DDT in successive cycles of treatment process with 30 mg of catalysts. 25
Fig. 7. Proposed mechanisms for DDT degradation on Pd/FSS.
26
Fig. 8. Proposed degradation pathways of DDT over magnetic mesoporous catalysts. An underlined word indicates that the chemical was detected in the current work.
27
Graphical abstract for
Pd-loaded magnetic mesoporous nanocomposites: A magnetically recoverable catalyst with effective enrichment and high activity for DDT and DDE removal under mild conditions Hua Tian a, Jun Chen b, Junhui He a, *, Feng Liu a,c a
Functional Nanomaterials Laboratory, Center for Micro/Nanomaterials Laboratory, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100190, China b c
Beijing Institute of Microchemistry, Beijing 100091, China
School of Chemical & Environmental Engineering, China University of Mining & Technology, Beijing 100083, China
*
Corresponding author. Fax: +86 10 82543535. E-mail address: [email protected]
28
29
