Efficient degradation of high concentration azo-dye wastewater by heterogeneous Fenton process with iron-based metal-organic framework

Accepted Manuscript Title: Efficient degradation of high concentration azo-dye wastewater by heterogeneous Fenton process with iron-based metal-organic framework Author: Huanli Lv Hongying Zhao Tongcheng Cao Lin Qian Yanbin Wang Guohua Zhao PII: DOI: Reference:

S1381-1169(15)00055-2 http://dx.doi.org/doi:10.1016/j.molcata.2015.02.007 MOLCAA 9423

To appear in:

Journal of Molecular Catalysis A: Chemical

Received date: Revised date: Accepted date:

20-9-2014 6-2-2015 8-2-2015

Please cite this article as: Huanli Lv, Hongying Zhao, Tongcheng Cao, Lin Qian, Yanbin Wang, Guohua Zhao, Efficient degradation of high concentration azo-dye wastewater by heterogeneous Fenton process with iron-based metal-organic framework, Journal of Molecular Catalysis A: Chemical http://dx.doi.org/10.1016/j.molcata.2015.02.007 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.

Efficient

degradation

of

high

concentration

azo-dye

wastewater by heterogeneous Fenton process with ironbased metal-organic framework

HuanliLv, Hongying Zhao*, Tongcheng Cao, Lin Qian, Yanbin Wang, Guohua Zhao*

Shanghai Key Lab of Chemical Assessment and Sustainability, Department of Chemistry, Tongji University, 1239 Siping Road, Shanghai, 200092, China.



Corresponding author Phone: (86)-21-65981180. Fax.: (86)-21-65982287.

E-mail address: [email protected],[email protected]

1

Abstract A novel iron-based metal-organic framework, possessing high surface area and good catalytic activity, was proposed as a heterogeneous Fenton catalyst for degrading high concentration methylene blue (CMB=500 ppm). The morphology and physicochemical properties of prepared catalysts were characterized by SEM, XRD, XPS, FT-IR, Raman spectra, etc. The obtained results showed that MOF-based catalysts MIL100(Fe) and FeII@MIL-100(Fe) possess high surface area of 1646 and 1228 m2 g1

,respectively. The MB removal though adsorption by MIL-100(Fe) and FeII@MIL-

100(Fe) is respectively 27 and 6% in 30 min, due to the electrostatic interaction between negative (or positive) adsorbent and positive pollutant. However, FeII@MIL100(Fe) exhibited highest Fenton catalytic ability compared to MIL-100(Fe) and Fe2O3 catalysts. The catalytic activity of each active site, evaluated by the turn over frequency (TOF) value, varied in the order of FeII@MIL-100(Fe)>MIL100(Fe)>Fe2O3. The main role of determining the decomposition efficiency, i.e., hydroxyl radical (∙OH) generation, surface redox properties and surface reaction, was greatly enhanced by the synergistic effect between FeII and FeIII in FeII@MIL100(Fe). Moreover, iron-based metal-organic framework retained the catalytic performance in a wide pH range of 3-8, and had a relative low iron leaching even in acidic condition.

Keywords: Heterogeneous Fenton reaction; Metal-organic framework; Surface reaction; Azo-dye. 2

1. Introduction Nowadays, increasing global attention of water pollution has become a serious issue facing humanity in the current situation, which promoting increasing demand for clean water environment. Heterogeneous Fenton and related reactions encompass the reactions of hydrogen peroxide with solid iron-based catalysts to form powerful hydroxyl radicals (·OH) to remove organic pollutants.Importantly, the heterogeneous Fenton catalystshave already overcome the critical drawbacks of homogeneous Fenton reaction, i.e., rigorous operating pH range and the formation of iron sludge in the coagulation step, and been widely explored in wastewater treatment[1-3]. Additionally, in order to improve the activity of traditional Fenton catalysts[4, 5],such as α-FeOOH, Fe2O3, Fe3O4, transitional metal substitutions (Fe, Cr, Mn, Co,Nb and Ni) can be added as secondary active sites for both improving activation H2O2 and enhancing stability[2, 6]. Meanwhile, the structure and the morphology of catalysts can be also modified for enhancing catalytic activity [7].Another way is iron immobilization, which immobilized the active components iron-containing substances on carriers[8, 9] like activated carbon (AC),mesoporous silica (SBA-15), mesoporous molecular sieves, and so on. Although lot of efforts similar with what we mentioned above have already made to improve the catalytic activity of heterogeneous Fenton catalysts, to some extent, they still have the problem of poor H2O2utilization and catalytic activity[10, 11].Recently, iron complex got considerable attention for its application in Fenton-like systems, for example, T.J. Collins[12]designed a tetraamido macrocycle ligandFeIII-TAML, which can activate hydrogen peroxide with high 3

efficiency to rapidly decompose the persistent and nonbiodegradable sertralinefrom pH 7.25 to 10.25[13]. Butmost iron complexes, i.e.,FeIII-TAML, are always watersoluble and unable to avoid the limitation of homogeneous Fenton reaction. Therefore, design and fabrication of effective heterogeneous catalysts for H2O2 activation in broader working pH range is the key issue to realize the desirable activity and long-term stability. Metal-Organic Frameworks (MOFs)are an intriguing type of inorganic-organic hybrid porous crystalline materials, which possess a three-dimensional well defined structure and are constructed by metal-containing nodes connected by various organic bridges[14, 15]. Owing to its characteristic of high surfacearea, large pore volume and tunable topology, MOF-basedmaterials have been exploited diverse applications in separation, gas and vapor storage, catalysis, drug delivery, optical switching device and so on[16-20].Besides to this, it is very interesting to find that iron-based MOF system possesses not only active sites(FeIII) for heterogeneous Fenton reaction but also have the advantages of iron complex in homogeneous Fenton process due to its attractive properties of organometallic coordination. That is to say, MOFs-based samples can be potentially devoted as Fenton catalysts for removing organic contaminants. For example, MIL-53(Fe) was used to activate H2O2 for decomposing methylene blue dye with the help of visible light [21]. MOF1(organotin-polymer [(Me3Sn)4Fe(CN)6]) was investigated as photo-Fenton catalyst with UV-light for discoloration of organic dyes[22]. However, pure MOF catalyst possessesonly FeIII sites with weak Fenton activity and contains low iron concentration, which restricts 4

Fenton oxidation catalytic activity. That is why supplementary technique, i.e., UVlight, visible light, usually applied for enhancing the degradation efficiency with pure MOFs. So it is still essential to extend the fabrication of novel MOF catalysts by adding iron-based active sites with high Fenton catalytic ability and exploit its corresponding high catalytic performance. Herein, we demonstrated the efficient Fenton catalytic activity of MIL-100(Fe), with chemical formulaFe3(III)O(H2O)2(F){C6H3(CO2)3}2·nH2O (n~14.5) [23], for degradation high concentration azo-dye (500 ppm) wastewater. Apart from this, since the large surface area and porosity of MIL-100(Fe), it is very useful to fabricate novel MOF-based Fenton catalyst using MIL-100(Fe) as support to enlarge iron concentration and modify the surface charge in Fenton system. Therefore, novel FeII@MIL-100(Fe) heterogeneous Fenton catalyst was fabricated to enhance the catalytic performance by synergic effect between FeII and FeIII active sites in this work. Methylene blue (MB) was selected as a model target contaminate for assessing the catalytic activity because it is respective of wide range of azo-dye modern and low biodegradability in water systems. Moreover, the activation mechanism of FeII@MIL100(Fe) was investigated and proposed on the basis of experiment results of characterization of surface properties, detection of ∙OH, and revelation of redox properties.

2. Experimental 2.1.Preparation of samples 5

MIL-100(Fe) was synthesized according to literature reported by Yoon et al [24].Typically, iron powder (532 mg),H3BTC (1337 mg), hydrofluoric acid (40 wt%, 849μL), nitric acid (65wt%, 395 μL), and deionized water (47 mL) was mixed in a Teflon autoclave and held at 150 °C for 24 h. The as-synthesized MIL-100(Fe) was further purified in two steps using boiling water (stirring 1 g of MIL-100(Fe) in 350mL of water at 80°C for 5h) to remove residual unreacted ions and hot ethanol (stirring 1g of MIL-100(Fe) in 200mL of ethanol at 60°C for 3h) until no detection of colored impurities in the mother liquor solution. The light-orange solid product wasfinally dried at 70°C under vacuum overnight and then stored in a desiccator. FeII@MIL-100(Fe) was synthesized by modified hydrothermal reaction according to the previous literature[25]. Briefly, the mixture of prepared MIL-100(Fe) (0.1 g) , deionized water (13 mL ), polyvinylpyrrolidone (PVP, 600 mg), FeCl3 (26 mg) and KI (380 mg) were well stirred and then transferred to a 23 mL Teflon-lined stainless-steel autoclave, then heated at 180 °C for 1.5 h. The brown solid was recovered by centrifugation at 10 000 rpm for 5 min, and then washed with ethanol and deionized water for several times. Finally the obtained products were blow-dried with N2. α-Fe2O3was prepared by a typical hydrothermal process[26]. Typically, FeCl3 (163 mg) was dissolved into 50 mL deionized water to form a transparent solution. NaAc (248 mg) was added to the above solution for modifying and improving the size and morphology of catalyst. Then the mixture was transferred and sealed into a 75 mL Teflon-lined autoclave, and heated at 180 °C for 24 h. The precipitate was collected 6

by centrifugation at 10 000 rpm for 5 min, and then washed with deionized water and ethanol for several times, finally dried at 70 °C under vacuum overnight. 2.2.Characterization methods Power X-ray diffraction (XRD) measurements were performed on a Bruker D8 Advance X-Ray diffractometer using Cu Kα (λ = 1.540562 Å, 40 kV, 40 mA) as the X-ray source at a scanning rate of 3°/min in the range from 2.5° to 60°. The scanning electron microscopy (SEM) images were obtained on a Hitachi S-4800scanning electron microscope. N2adsorption-desorption analysis was measured on a Micrometrics TriStar 3000 instrument, Brunauer-Emmet-Teller (BET) special surface area was estimated from the relative pressure range from 0.05 to 0.20, pore volumes were determined using the adsorbed volume at a relative pressure of 0.99033. Fourier transform infrared spectra (FT-IR) were examined in the region from 400 to 2000 cm-1 at room temperature using a Nicolet 6700 spectroscopy on KBr pellets. Raman spectroscopy was performed by a Renishaw Raman micro spectrometer using an Ar+ ion laser (514.5 nm line) as the excitation source. The X-ray photoelectron spectroscopy was carried out on a Kratos ASIS-HS X-ray photoelectron spectroscope fittedwith a standard and monochromatic source (Al Kα) operatedat 150 W (15 kV, 10 mA).The iron content was determined by inductively coupled plasma-atomic emission spectrometry (ICP-AES, PE OPTIMA 2100DV).Zeta potentials of the catalysts in deionized water were measured on a zeta potential analyzer (Zetasizer 2000, Malvern, UK). H2-TPR measurements were performed using AutoChem II 2920 instrument (Micromeritics). Prior to the TPR run, the fresh sample was pretreated in situ for 2 h 7

at 150 °C in air flow. After cooling and a step of purge of the lines in Ar flow, an H2: Ar mixture (H2: Ar volume ratio of 10% and total flow of 50 mL min-1) was sent through the sample while increasing the temperature up to 800°C with a rate of 10 °C min-1. 2.3.Experimental procedure All the catalysts would be dried at 150 °C for 12 h before each use. Alldegradation experiments of MB were carried out in a beaker (100 mL) at room temperature. The initial concentration of MB was 500 mg L-1, and the total volume of reaction solution was 50 mL.The reaction suspension was prepared by adding the required amount of catalyst (1 g L-1) into 50 mL solution that had been adjusted to the desired pH valueby 0.1 M H2SO4 or 0.1 MNaOH. A known concentration (40 mM) of H2O2 was added to the solution to initiate the reaction after adsorption for 45 min. Samples were taken at set intervals using a 5 mL syringe, centrifuged at 100 000 rpm for 5min, then filtered the supernatantimmediately through a 0.22 μm filter film, and quenched with excess isopropyl.In order to test the adsorption of MB behave during the Fenton reaction, the desorption studies were conducted in the thermostated shaker using 0.5% HCl (v/v) ethanol solution as eluent for one night. 2.4.Sample analysis The concentration of MB during the reaction was analyzed by a UV-vis spectrophotometer (Agilent, 8453) with the maximum absorbance wavelength for MB at 665 nm. ESR spectra were obtained on a Bruker EMX Xplus-10/12 with Microwave Bridge (microwave frequency, 9.853 GHz; microwave power, 20 mW; modulation 8

amplitude, 1 Gauss; modulation frequency, 100 kHz). For ESR measurement, 25 μL of the sample was collected form the suspension system and immediately mixed with 25 μL of 0.22 mol L-1 DMPO to form DMPO-·OH adduct.The concentration of ion leaching were measured according to the 1,10-phenanthroline method[27,28], and the absorbance was measured at 510 nm using a UV-vis spectrophotometer (Agilent, 8453) with a 1 cm path length spectrometric quartz cell. TOCvalues were determined by a Multi 3100 TOC/TN analyzer. For TOC analysis, all the samples were immediately treated with scavenging reagent (0.1 M Na2SO3, 0.1 M KH2PO4, 0.1 M KI and 0.05 M NaOH) to obtain accurate TOC value [29].

3.

Results and Discussion

3.1 The morphology and physicochemical properties of MIL-100(Fe) and FeII@MIL-100(Fe) catalysts The crystallographic structure of obtained catalysts was examined by XRD. As presentedin Fig.1, sample MIL-100(Fe) possessed typical XRD patterns of MOF(Fe) as previously reported in literatures[23, 30]. Sample FeII@MIL-100(Fe) displayed the same XRD patterns as MIL-100(Fe), suggesting there is no change on the crystal structure after the loading of iron species. No peak belonging to FeII species in FeII@MIL-100(Fe) was detected possible due to the low amount of introduced iron species. Textural properties, such as specific surface area and porous structure, of catalysts were determined by N2 adsorption-desorption at 77 K. As exhibited in Table 1, sample MIL-100(Fe) possessed high BET surface area of 1646 m2 g-1 with a pore 9

volume of 0.76 cm3g-1. After incorporating ironspecies, slightly decreased BET surface of 1228m2 g-1 with a pore volume of 0.61cm3g-1 were obtained. Additionally, the N2 adsorption-desorption isotherms, shown in Fig.2, displayed an intermediate mode between type I and type IV, which is respectively associated with the presence of microporous windows and mesoporous cages in MIL-100(Fe) and FeII@MIL100(Fe) catalysts[23].

II

Intensity(a.u)

Fe @MIL-100(Fe)

MIL-100(Fe)

10

20 2Theta (deg.)

30

40

Fig.1. XRD patterns of the synthesized MIL-100(Fe) and FeII@ MIL-100(Fe).

10

400 0.10 0.08

3

300

Pore Volume (cm /g STP)

3

Quantity Adsorbed (cm /g STP)

500

200 100

MIL-100(Fe) II Fe @MIL-100(Fe)

0.06 0.04 0.02 0.00 0

20

40

60

80

100

Pore Diameter (nm)

0 0.0

0.2

0.4

0.6

0.8

1.0

Relative Pressure (P/P0) Fig.2. N2 adsorption–desorption isotherms measured at 77 K and pore sizedistributions from the desorption branches through the BJH method (inset) of MIL-100(Fe) (black line) and FeII@MIL-100(Fe) (red line).

The morphology of as-prepared products was investigated by SEM, as shown in Fig.3. MIL-100(Fe) exhibited large octahedral crystals with smooth surface (see Fig.3A1). After the loading of FeII species, the surface of octahedral crystal became obviously coarse (Fig.3B1).The FeII particles seemed to be uniformly dispersed on the surface of MIL-100(Fe). The presence of iron in two catalysts was confirmed with EDS analysis,and the corresponding EDS results were listed in Table 1, Fig.3A2 and 3B2. The surface Fe content for MIL-100(Fe) and FeII@MIL-100(Fe) was respectively 8.49 and 29.2 w.t.%, That is to say the Fe content was increased after the adding of FeII species. Additionally, the EDS elemental mapping further confirmed that all iron elements were homogeneously distributed in the obtained catalysts even for sample FeII@MIL-100(Fe) (see Fig.3A3 and 3B3). The bulk Fe content of MIL11

100(Fe) and FeII@MIL-100(Fe) catalysts, examined by ICP analysis, was around 14 and 21w.t.%, which is in good agreement with the observation of EDS results. So, from the above results, it is reasonable to say that the second active sites (FeII species) in this system were successfully interacted into MIL-100(Fe).

Fig.3. SEM images of the prepared MIL-100(Fe)(A1); FeII@MIL-100(Fe) (B1). EDS spectra of MIL-100(Fe) (A2); FeII@MIL-100(Fe) (B2).EDS elemental mappings of MIL-100(Fe) (A3); FeII@MIL-100(Fe)(B3).

12

Table1 The physicochemical properties of MIL-100(Fe) and FeII@MIL-100(Fe) catalysts Sample MIL-100(Fe)

SBET (m2 g-1) 1646

Vpore (cm3 g-1) 0.76

EDS(w.t.%) C O Fe 55.12 36.39 8.49

XPS Fe (w.t.%) 8.94

Zeta Potential (mV)a -22.0

FeII@MIL-100(Fe)

1228

0.61

40.43 30.37 29.2

20.31

5.3

a

the Zeta potential of different catalyst was examined at initial pH=3.

In order to deeply analysis the molecular structure and get further insight on structural variations of the two catalysts, FT-IR and Raman spectra were performed, as shown in Fig.4A and 4B, respectively. The IR spectrum of MIL-100(Fe) exhibited the clear adsorption at 1624, 1442, 1371, 760 and 707 cm-1. The peak at 1624 cm-1 is assigned to v(C=O) bond of carboxylate groups, while the bands at 1442 and 1371 cm-1 are respectively attributed to asymmetric and symmetric vibrational bands characteristic of the –O–C–O– group. The other two sharp peaks at 760 and 707 cm-1 corresponds to C–H bending vibrations of benzene [31-33]. After the introducing of FeIIspecies, all the obtained IR spectra remain similar with sample MIL-100(Fe) except that the spectra intensity is decreased, which is due to the surface coverage by iron species. Typical Raman spectra of MIL-100(Fe) were also obtained, which is identical to those of reported data in literature[33].

13

A

II

Transmittance(%)

Fe @MIL-100(Fe)

MIL-100(Fe)

760 1624

2000

1442 1371

707

1500 1000 Wavenumber(cm-1)

500

B

II

Intensity (a.u)

Fe @MIL-100(Fe)

MIL-100(Fe)

800

1200 1600 Raman Shift(cm-1)

2000

Fig.4. FT-IR (A) and Raman (B) spectra of MIL-100(Fe) and FeII@MIL-100(Fe) catalysts.

The chemical states of iron species in FeII@MIL-100(Fe), MIL-100(Fe) and

14

reference Fe2O3were examined by XPS analysis. Fig.5 shows the Fe 2p binding spectra of the three catalysts mentioned above, and the exposed Fe surface species can be quantified in Table 1. The Fe 2p1/2 and Fe 2p3/2 peak of MIL-100 (Fe) was respectively centered at 724.8 eV and 710.8 eV and their separation (Δ=2p1/2-2p3/2) was 14 eV. These characteristics were quite similar to reference Fe2O3[34].That is to say, all the obtained peaks were assigned to FeIII in MIL-100(Fe) shell. After loading iron species, the main peak for Fe 2p1/2 and Fe 2p3/2 were keep the same, nevertheless, two right-shoulder at 722.3 and 708.8 eV appeared for Fe 2p1/2 and Fe 2p3/2, respectively. In order to well understand what happened after loading iron species, the Fe 2pspectra for MIL-100(Fe) and FeII@MIL-100(Fe) were fit into multiplet peaks, which referenced as A, B, C, D, E, F, G and H have been labeled on the spectrum in Fig.5. Peak A ,B, C, D, E and F were assigned to FeIII in MIL-100(Fe), while the new generated peak G and H were corresponded to FeIImight be in Fe3O4 and/or FeO[35, 36], suggesting that the introduced iron species (FeII) would improve the catalytic efficiency for heterogeneous Fenton reaction.

15

710.7 eV

Fe 2p Fe 2p1/2

FeII@MIL-100(Fe)

Fe 2p3/2

713.2 eV 708.8 eV

A

B

C

717.3 eV

G

D

F E

Intensity (a.u.)

H

710.8 eV 713.1 eV

MIL-100(Fe)

717.2 eV

709.8 eV 712.8 eV 718.2 eV

Fe2O3

740

735

730

725

720

715

710

705

Binding Energy (eV) Fig.5. XPS spectra for Fe 2p regions of FeII@MIL-100(Fe), MIL-100(Fe) and Fe2O3 catalysts. 3.2 Efficient catalytic degradation performance on methylene blue and plausible mechanism The degradation efficiency of MB with very high initial concentration of 500 mg L-1 by various processes on different catalysts was evaluated in Fig.6. Fig.6A showed pure adsorption ability of catalysts. The removal of MB was rapidly increased to 27% 16

in the first 30 min by MIL-100(Fe), due to its high adsorption capacity with surface area of 1646 m2 g-1. Nevertheless, sample FeII@ MIL-100(Fe), possessing the surface area of 1228 m2 g-1, exhibited relative low adsorption ability that the removal of MB is only 6% in 30 min. This can be explained with an electrostatic interaction between dye MB and MOF-based materials (as presented in Scheme 1)[37]. As reported in the literatures, MIL-100(Fe) generally presents a negative charged surface in aqueous solution, while MB existes as a cationic guest molucule[38,39]. So, the adsorption process is mainly governed by the strong electrostatic attractionbetweenthe negatively charged surface of MIL-100(Fe) and the cationic dye MB[24, 40]. Note that the surface charge of MIL-100(Fe) is negative with the zeta potential of -22.0 mV (see Table 1), which is consistent with the reported results in the literature[40]. After introducing FeII species, the surface charge of sample FeII@MIL-100(Fe) would change to positive charge with zeta potential of 5.3 mV. That is to say, the increased positive charge of FeII@MIL-100(Fe) results in the decreased adsorption ability of MB. Although FeII@MIL-100(Fe) exhibited weak adsorption ability for cationic dye MB, it is reasonable to hypothesize that FeII@MIL-100(Fe) could possess high adsorption capacity for some anionic and non-chargedpollutants. To verify this, the adsorption between non-charged pollutant phenol and two MOF-based catalysts was further carried out. MIL-100(Fe) and FeII@MIL-100(Fe) catalysts respectively with negatively and positively charged surface exhibited similar adsorption removal efficiency for phenol due to the similar high BET surface area. From a fundamental standpoint, this work emphasized the electrostatic interaction on dye capture with 17

MOFs, and the obtained results may be useful for broadening the applicationfields of MOFs. However, there are still some questions need to be addressedin the future work for in-depth understanding the adsorption of organics over MOFs, such as acid-base interaction, hydrogen bonding and π-πstacking/interaction [41].

X-

XFeII

FeIII

FeIII

FeIII

FeIII

FeIII FeIII

FeIII

FeIII

FeIII

FeIII

FeIII

FeIII

FeIII

FeIII

FeIII

FeIII

FeIII FeIII

(X- = F- or OH-)

(X- = F- or OH-)

FeII@MIL-100(Fe)

MIL-100(Fe)

Scheme 1

Fig.6B exhibited the total degradation ability including adsorption and Fenton oxidation. Before adding H2O2, the pollutant solution was stirred for 45 min with catalyst in order to achieve adsorption equilibrium. Once introducing H2O2, FeII@MIL-100(Fe) exhibited a sharp increase for MB removal, revealing the efficient Fenton oxidation catalytic performance. Nevertheless, since the high adsorption capacity of MIL-100(Fe), the overall degradation ability including adsorption and Fenton oxidation varied in the order of MIL-100(Fe) > FeII@MIL-100(Fe) > Fe2O3. In order to investigate pure Fenton oxidation catalytic ability, the corresponding MB removal efficiency was calculated by getting rid of MB removal of adsorption (as 18

shown in Fig.6A) from the total MB removal including adsprotion and Fenton oxidation (as shown in Fig.6B). Additionally, considering the possible competition between MB, H2O2, and other generated interminates during the whole degradation process, the desorption experiment at the end of Fenton oxidation reaction was performed to test the MB adsorption behave. After Fenton reaction 240 min, the desorpted MB from MIL-100(Fe) was 46% of initial MB, which is similar with its adsorption capacity (240 min, MB removal: 49%) as shown in Fig.6A. That is to say, the adsorption of MB is relatively stable in the Fenton oxidation process and the calculated oxidation ability was reasonable to evaluate the pure Fenton catalytic ability.As presented in Fig.6C, FeII@MIL-100(Fe) catalyst exhibited highest Fenton oxidation catalytic ability. Fe2O3 catalyst presented higher Fenton catalytic degradation performance than MIL-100(Fe) catalyst. For example, at 105 min, MB removal achieved to respectively 53% and 42% for Fe2O3 and MIL-100(Fe), which was due to the high iron concentration of 70w.t.% (theoretical value) for Fe2O3 catalyst, while 14w.t.% iron for MIL-100(Fe) (see Table 1).

19

1.0 MIL-100(Fe) II Fe @MIL-100(Fe) Fe2O3

0.8

0.6

0.4

0.2

0.0 0

60

120

180

240

300

Time (min) 1.0

MB removal by adsorption and Fenton

MB removal by adsorption

A

B 0.8

0.6 adsorption

0.4 MIL-100(Fe) II Fe @MIL-100(Fe) Fe2O3

0.2

0.0 0

60

120

180

Time (min)

20

240

300

MB removal by Fenton

0.8

C

0.6

0.4

MIL-100(Fe) II Fe @MIL-100(Fe) Fe2O3

0.2

0.0 60

120

180

240

300

Time (min) Fig.6. The removal efficiency of MB with different catalysts during the reaction in 285 min, at initial pH 3 with 40 mM H2O2, 1 g L-1 catalyst dosage, and 500 mgL1 MB:(A) adsorption efficiency; (B) total degradation efficiency; (C) Fenton oxidation efficiency with the same catalyst dosage.

As we discussed above, it is reasonable to say that the total MB removal efficiency was related with adsorption capacity, iron concentration, and catalytic ability of each active site. Therefore, the TOF value[41, 42], estimated by considering the number of degraded MB molecules (mM) per iron atoms (mM) in the required time to achieve 45% and 50% MB degradation, was proposed to evaluate the catalytic activity of each active site (as depictured in Fig.7). The TOF value is related with the surface loading of active iron species. According to the literatures[41,42], the exposed Fe surface species can be quantified by XPS. The surface iron content for FeII@MIL100(Fe), MIL-100(Fe) and Fe2O3 20.3 wt%, 8.9 wt% and 52.6 wt%, respectively. FeII@MIL-100(Fe) catalyst presented the highest TOF value, suggesting thatthe introduced FeIIfavored oxidationability of Fenton reaction . Additionally, all MOF21

based catalysts presented much higher catalytic activity than traditional iron oxide catalysts, i.e., at 45% MB removal, the TOF value of FeII@MIL-100(Fe) and MIL100(Fe) was respectively 1.2 and 0.14 h-1, which was 12 and 1.4 times of sample Fe2O3 with its TOF value of 0.10 h-1.

1.25

F OT

(h1 )

1.00 0.75 0.50 0.25 0.00

50% M

(Fe) 100

II @MIL Fe

(Fe) -100 L I M

45% M

O Fe 2 3

B re mov al

B re mo v al

Fig.7.The turn-over frequency (TOF) for the different catalysts determined at respectively 45% and 50% MB removal. TOF has been estimated by considering the number of degraded MB molecules (mM) per iron atoms (mM) in the required time to achieve 45% and 50% MB removal. The total organic carbon (TOC) in reaction solution was measured as an indicator for the organic mineralization at different time intervals[43], as shown in Fig.8. The TOC removal of FeII@MIL-100(Fe) in the first 8 h was relative low (1329%), perhaps because of the strong scavenge of ∙OH by related intermediates with high concentration [42]. While for MIL-100(Fe), the TOC removal can be reached at 22

40% in 4 h since its high adsorption capacity. Nevertheless, the TOC removal was increased with increasing the reaction time, possibly due to the further decomposition of related intermediates. In order to ensure the H2O2 dosage would be enough for oxidation reaction in case of extending the reaction time to 24 h. We added another 40 mM H2O2to solution after oxidation for 12 h. This measurement showed that about 91% of the MB dye was degraded completely into its mineral component after about 25 h for FeII@MIL-100(Fe). This observation was consistent with the tendency of MB removal by adsorption and Fenton oxidation treatment.

100 MIL-100(Fe) II Fe @MIL-100(Fe)

TOC removal (%)

80

60

40

20

0 4

8

12

16

20

25

Time (h) Fig.8. TOC removal of MB with different catalysts at pH 3 with 40 mM H2O2, 1 g L-1 catalyst dosage, and 500 mg L-1MB. The amount of ∙OH, formed by the catalytic decomposition of H2O2 with active sites, is strongly depending on the nature of used catalysts, to some extent, can reflect the activity of catalyst. Therefore, spin trapped EPR with 5,5-dimethypylpyrroline-1-

23

oxide (DMPO) was used to examine the generation of ∙OH after reaction 60 min, as depicted in Fig.9. Due to the hyperfine interaction between the electron spin of ∙OH and the orbital spin of N atom in DMPO, the EPR signal was split into four single lines with an intensity ratio of 1:2:2:1[42]. Additionally, with the identical lineshape of DMPO-adduct among all tested samples, the intensity can be directly related to the amount of generated ∙OH[44]. Note that the productivity of ∙OH with MOFs-based catalysts was much higher than Fe2O3, indicating that MOFs samples possess high catalytic ability of decomposition H2O2. After introducing FeII active sites into pure MOF structure, the concentration of generated ∙OH was obviously enhanced. We conclude that the steady-state concentration of ∙OH over FeII@MIL-100(Fe) is highest since synergic catalytic effect between FeII and FeIII. Moreover, to investigate the contribution of homogeneous Fenton reaction catalyzed by leaching Fe, comparison experiment was carried out by removing catalysts after vigorous agitation for 1 h and then adding H2O2 into the filtrate (see Fig.9). The amount of formed ∙OH by leaching Fe from FeII@MIL-100(Fe) and MIL-100(Fe) was quite little especially for MIL100(Fe). While the leaching Fe from Fe2O3 generated relatively high concentration of ∙OH, which is similar with solid Fe2O3 catalyst. These results confirmed that the degradation reaction catalyzed by MOF-based catalysts was mainly heterogeneous even at pH=3, but with Fe2O3 it was homogeneous reaction.

24

II

Fe @MIL-100(Fe)

II

leaching Fe from Fe @MIL-100(Fe)

MIL-100(Fe) leaching Fe from MIL-100(Fe) Fe2O3 leaching Fe from Fe2O3

3480

3500

3520 3540 Magnetic Field (G)

3560

Fig.9. DMPO spin trapping ESR spectra over different catalysts with 1 g L-1catalyst dosage, 40 mM H2O2, and initial pH 3.

TPR profile in Fig.10 was used to study the redox properties of catalysts, which could affect the catalytic ability of decomposing H2O2. It is reasonable to correlate redox properties and catalytic ability of catalysts. Sample MIL-100(Fe) exhibited one peak maximized at 522°C with a left-shoulder centered at 439 °C and another peak maximized at 719°C, which was due to the reduction of FeIII to FeII and/or even to Fe[45-48]. After loading FeII species, the above mentioned reduction peaks for FeII@MIL-100(Fe) were shifted to the lower temperature respectively at 492, 387, and 674°C. These observation indicate that FeII@MIL-100(Fe) presented higher redox properties than MIL-100(Fe). In another word, FeII@MIL-100(Fe) possessed more efficient catalytic active sites in Fenton reaction, which is consistent with the observation of ∙OH generation ability and MB degradation efficiency. 25

o

492 C

o

674 C

TCD signal (a.u)

Fe@MIL-100(Fe) o

387 C o

522 C

o

o

MIL-100(Fe)

200

719 C

439 C

400 600 o Temperature ( C)

800

Fig.10.H2-TPR profiles of MIL-100(Fe) and FeII@MIL-100(Fe) catalysts (TPR conditions: 10% H2/Ar with heating rate 10oC/min).

To better understand the efficient catalytic performance of FeII@MIL-100(Fe) during Fenton oxidation reaction, we proposed a reasonable surface-based reaction mechanism, as displayed in Fig. 11[49-51]. Surface FeIII active sites in MIL-100 structure were interacted with H2O2 to generate HO2∙oxidants and FeII sites through Eq.(1)[52]. Then the formed FeIIwas oxidized by H2O2 to generate ·OH through Eq(2).In general, FeIImay play an important role as an electron donor to initiate the Fenton reaction through Eq (2) to produce·OH. The generation rate of ·OH over FeII was proven higher than that over FeIII by 3-4 orders of magnitudes[42]. So on one hand, the introduced FeIIin FeII@MIL-100(Fe) greatly increased the concentration of generated·OH radical through Eq(2). On the other hand, from the dynamics view, the 26

introduced FeIIwould accelerate the reaction rate of Eq (2) and then resulted in higher concentration of FeIII species. At the same time, the reaction rate of Eq(1) would be also accelerated with more FeIII species. This observation implied thepotential in-situ recycling of FeII and FeIII sites (FeIIFeIII FeII) in catalyst, which isthe synergistic effect of FeII and FeIII species in Fenton process to enhance the degradation efficiency.  Fe III  H 2 O 2   Fe II  O 2 H  H  (1)  Fe II  H 2 O 2   Fe III  OH  OH  (2)

So, in heterogeneous Fenton reaction, MB pollutants were firstly adsorbed on the surface of FeII@MIL-100(Fe) catalyst. Once H2O2 was added, ·OH radicals would be generated by surface active sites of catalyst in terms of FeII and FeIII. And then the formed ·OH reacted with MB pollutants involving that adsorbed on catalyst surface and dissolved in solution, leading to degradation products. In summary, although the electrostatic interaction between pollutant and catalyst would greatly influence adsorption capacity, high surface area of MOF-based sample and synergistic effect betweenintroduced FeII and FeIIIexisted in MOF structureare responsible for the efficient catalytic activity of FeII@MIL-100(Fe) in Fenton reaction.

27

Degradation products

H2O2

.OH

H2O2

+

FeII FeIII H 2O 2

FeIII

MB

FeIII

MB H 2O 2

FeIII H2O2

FeII@MIL-100(Fe)

FeIII

+

Fig.11. Proposed catalytic mechanism for the activation of H2O2 by FeII@MIL-100(Fe).

28

H2O2

FeIII .OH

3.3 Excellent chemical stability of iron-based metal-organic framework and itsstabilized catalytic activity over wide pH Range From the view of actual application of heterogeneous Fenton oxidation, the stability of catalyst is an important issue to be considered. Thus, leaching Fe from solid catalysts was detected to gain a better understanding of sample stability. As presented in Fig.12, theconcentration of Fe (CFe)increased with increasing the reaction time. Whereas, the leached CFe of MIL-100(Fe) and FeII@MIL-100(Fe) after 240 min respectively was 2.2 and 7.1 ppm, implying 1.6 w.t.% and 3.3 w.t.% Fe in catalyst was leached. The leaching Fefrom MIL-100(Fe) and FeII@MIL100(Fe)is respectively only 14% and 46% of Fe2O3 catalyst(CFe=15.2 ppm). Moreover, the XRD patterns of the MIL-100(Fe) and FeII@MIL-100(Fe) catalysts that after used in MB solution were as same as fresh samples (not shown), indicating again the high stability of MOF-based catalysts in solution. That is to say, MOFsbased heterogeneous Fenton catalyst could be a potential candidate for thelong-term industrial application.

29

18 MIL-100(Fe) II Fe @MIL-100(Fe) Fe2O3

16 14

CFe(ppm)

12 10 8 6 4 2 0 1

2

3

4

Time (h) Fig.12. The concentration of iron leaching in solution during MB removal in 4 h with different catalysts, at initialpH 3 with 40 mM H2O2, 1 g L-1 catalyst dosage.

It is well known that the actual wastewater always has variable pH values. Thus, it is very useful to understand the effect of initial pH on the degradation efficiency. As exhibited in Fig.13, the further increase of pH from 3 to 8 has little impact to the catalytic activity of MIL-100(Fe). For example, the MB removal was decreased 7%, 9% and 14% at 300min when increasing the pH value from 3 to 5, 6.5 and 8, respectively. It is claimed that the catalytic oxidation performance is relative stable in the wide pH range, which is of great significance in the actual treatment of textile printing and dyeing wastewater.

30

MB removal by adsorption and Fenton

1.0

0.8

0.6

0.4 300 min 165 min 90 min 60 min

0.2

0.0 3

4

5

6

7

8

pH Fig.13. The degradation efficiency of MB with MIL-100(Fe) at different initial pH with 40 mM H2O2, 1g L-1 catalyst dosage, and 500 mg L-1MB.

4. Conclusions

This study reported the fabrication of novel iron-based metal-organic framework catalyst and its application on degrading high concentration methylene blue (500 ppm) via heterogeneous Fenton oxidation process. The removal of MB through adsorption for MIL-100(Fe) was rapidly decreased 27% in the first 30 min since the high surface area (1646 m2 g-1) and strong electrostatic interaction between positive MB and negative adsorbent. Nevertheless FeII@MIL-100(Fe), with surface area of 1228 m2 g-1, showed relatively low adsorption ability due to the weak electrostatic interaction between positive MB and positive adsorbent.Moreover, the catalytic 31

activity of active sites in FeII@MIL-100(Fe), estimated by TOF value, was highest during the whole Fenton reaction.The amount of ·OH determined by EPR varied in the order of FeII@MIL-100(Fe)>MIL-100(Fe)>Fe2O3.The surface redox properties, revealed by TPR analysis, showed that FeII@MIL-100(Fe) had lower reduction peaks. Based on the above results, a plausible mechanism was proposed to explain thecorresponding high Fenton catalytic activity. The introduced FeII species not only increased iron content of pure MOF sample, but also favored the in-situ recycling of FeII and FeIII sites, providing synergistic effect between FeII and FeIII species to efficiently generate ·OH.MB pollutant adsorbed on the surface of catalyst, and then in-situ reacted with ·OH to form degradation products. Nevertheless, more study about improving the interaction between introduced FeII active sites and MOF sample, modifying catalyst structure is highly desired, which is important for the further design of MOFs-based Fenton catalysts.

Acknowledgements This work was supported jointly by the National Natural Science Foundation P.R. China (Project NO. 21277099, 21207101), the fundamental Research Funds for the Central Universities.

32

Reference [1] G.B. Ortiz de la Plata, O.M. Alfano, A.E. Cassano, Appl. Catal. B: Environ. 95 (2010) 1-13. [2] M. Barreto-Rodrigues, F.T. Silva, T.C. Paiva, J. Hazard. Mater. 165 (2009) 1224-1228. [3] H.H. Yan, P. Lu, Z.Q. Pan, X. Wang, Q.Y. Zhang, L.S. Li, J. Mol. Catal.AChem. 377 (2013) 57-64. [4] W. Luo, L.H. Zhu, N. Wang, H.Q. Tang, M.J. Cao, Y.B. She, Environ. Sci. Technol. 44 (2010) 1786-1791. [5] M. Das, K.G. Bhattacharyya, J. Mol. Catal.A-Chem. 391 (2014) 121-129. [6] A.D. Bokare, W. Choi, J. Hazard. Mater. 275 (2014) 121-135. [7] X.B. Hu, B.Z. Liu, Y.H. Deng, H.Z. Chen, S. Luo, C. Sun, P. Yang, S.G. Yang, Appl. Catal. B: Environ. 107 (2011) 274-283. [8] A. Kumar, D. Srinivas, J. Mol. Catal.A-Chem. 368 (2013) 112-118. [9] F.L. Cao, H. Li, Z.M. Xu, J. Zhang, Y. Zhang, Y.I. Huo, J. Mol. Catal.A-Chem. 353 (2012) 215-219. [10] J.L. Wang, L.J. Xu, Crit. Rev. Env. Sci. Technol. 42 (2012) 251-325. [11] A. Dhakshinamoorthy, S. Navalon, M. Alvaro, H. Garcia, Chemsuschem 5 (2012) 46-64. [12] A. Ghosh, D.A. Mitchell, A. Chanda, A.D. Ryabov, D.L. Popescu, E.C. Upham, G.J. Collins, T.J. Collins, J. Am. Chem. Soc. 130 (2008) 15116-15126. [13] L.Q. Shen, E.S. Beach, Y. Xiang, D.J. Tshudy, N. Khanina, C.P. Horwitz, M.E. Bier, T.J. Collins, Environ. Sci. Technol. 45 (2011) 7882-7887. [14] S.T. Meek, J.A. Greathouse, M.D. Allendorf, Adv. Mater. 23 (2011) 249-267. [15] Y.X. Zhou, J.L. Song, S.G. Liang, S.Q. Hu, H.Z. Liu, T. Jiang, B.X. Han, J. Mol. Catal.A-Chem. 308 (2009) 68-72. [16] L.J. Murray, M. Dinca, J.R. Long, Chem. Soc. Rev. 38 (2009) 1294-1314. [17] Z. Zang, W. Yang, J. Appl. Polym. Sci. 109 (2011) 103106. [18] E. Haque, J.W. Jun, S.H. Jhung, J. Hazard. Mater. 185 (2011) 507-511. [19] A. Dhakshinamoorthy, M. Opanasenko, J. Cejka, H. Garcia, Adv. Synth. Catal. 355 (2013) 247-268. [20] A. Dhakshinamoorthy, A.M. Asiri, H. Garcia, Chem. Commun., 2014, DOI: 10.1039/C4CC04387A [21] J.J. Du, Y.P. Yuan, J.X. Sun, F.M. Peng, X. Jiang, L.G. Qiu, A.J. Xie, Y.H. Shen, J.F. Zhu, J. Hazard. Mater. 190 (2011) 945-951. [22] M.T.H. Nguyen, Q.T. Nguyen, J. Photoch. Photobio. A 288 (2014) 55-59. [23] P. Horcajada, S. Surble, C. Serre, D.Y. Hong, Y.K. Seo, J.S. Chang, J.M. Greneche, I. Margiolaki, G. Ferey, Chem. Commun. (2007) 2820-2822. [24] J.W. Yoon, Y.K. Seo, Y.K. Hwang, J.S. Chang, H. Leclerc, S. Wuttke, P. Bazin, A. Vimont, M. Daturi, E. Bloch, P.L. Llewellyn, C. Serre, P. Horcajada, J.M. Greneche, A.E. Rodrigues, G. Ferey, Angew. Chem. Int. Ed. 49 (2010) 5949-5952. [25] M.D. Ye, J.J. Gong, Y.K. Lai, C.J. Lin, Z.Q. Lin, J. Am. Chem. Soc. 134 33

(2012) 15720-15723. [26] X.L. Wu, Y.G. Guo, L.J. Wan, C.W. Hu, J. Phys. Chem. C 112 (2008) 1682416829. [27] A.L.T. Pham, C. Lee, F.M. Doyle, D.L. Sedlak, Environ. Sci. Techn. 43 (2009) 8930–8935. [28] L.J. Xu, J.L. Wang, Appl. Catal. B: Environ. 123 (2012) 117–126. [29] D.N. Thi, H.P. Ngoc, H.D. Manh, T.N. Kim, J. Hazard. Mater. 185 (2011) 653661. [30] A. Dhakshinamoorthy, M. Alvaro, Y.K. Hwang, Y.K. Seo, A. Corma, H. Garcia, Dalton T. 40 (2011) 10719-10724. [31] Y.K. Seo, J.W. Yoon, J.S. Lee, U.H. Lee, Y.K. Hwang, C.H. Jun, P. Horcajada, C. Serre, J.S. Chang, Micropor. Mesopor. Mater. 157 (2012) 137-145. [32]H. Leclerc, A. Vimont, J.C. Lavalley, M. Daturi, A.D. Wiersum, P.L. Llwellyn, P. Horcajada, G. Ferey, C. Serre, Phys. Chem. Chem. Phys. 13 (2011) 11748-11756. [33] M.R. Lohe, M. Rose, S. Kaskel, Chem. Commun. (2009) 6056-6058. [34] Z.Y. Sun, H.Q. Yuan, Z.M. Liu, B.X. Han, X.R. Zhang, Adv. Mater. 17 (2005) 2993-2297. [35] H.Y. Zhao, Y.J. Wang, Y.B. Wang, T.C. Cao, G.H. Zhao, Appl. Catal. B: Environ. 125 (2012) 120-127. [36] A.P. Grosvenor, B.A. Kobe, M.C. Biesinger, N.S. McIntyre, Surf. Interface Anal. 36 (2004) 1564-1574. [37] E. Haque, J.W. Jun, S.H. Jhung, J. Hazard. Mater. 185 (2011) 507-511. [38] M. Tong, D. Liu, Q. Yang, S. Devautour-Vinot, G. Maurin, C. Zhong, J. Mater. Chem. A (1) 2013 8534-8537. [39] Z. Hasan, S.H. Jhung, J. Hazard. Mater. 283 (2015) 329-339. [40] S.H. Huo, X.P. Yan, J. Mater.Chem. 22 (2012) 7449-7455. [41] R. Martin, S. Navalon, J.J. Delgado, J.J. Calvino, M. Alvaro, H. Garcia, Chemistry 17 (2011) 9494-9502. [42] X.J. Yang, X.M. Xu, J. Xu, Y.F. Han, J. Am. Chem. Soc. 135 (2013) 1605816061. [43] L. Zhang, H. Zeng, Y. Zeng, Z. Zhang, X. Zhao, J. Mol. Catal.A-Chem.392 (2014) 202-207. [44] D. Tsukamoto, Y. Shiraishi, Y. Sugano, S. Ichikawa, S. Tanaka, T. Hirai, J. Am. Chem. Soc. 134 (2012) 6309-6315. [45] T. Zeng, X.L. Zhang, S.H. Wang, Y.R. Ma, H.Y. Niu, Y.Q. Cai, Chem.-Eur. J. 20 (2014) 6474-6481. [46] B.S. Wu, L. Tian, H.W. Xiang, Z.X. Zhang, Y.W. Li, Catal. Lett. 102 (2005) 211218. [47] H.H.P. Yiu, M.A. Keane, Z.A.D. Lethbridge, M.R. Lees, A.J. El Haj, J. Dobson, Nanotechnology 19 (2008). [48] R.C.C. Costa, F.C.C. Moura, J.D. Ardisson, J.D. Fabris, R.M. Lago, Appl. Catal. B: Environ.83 (2008) 131-139. [49] A.G. Marquez, A. Demessence, A.E. Platero-Prats, D. Heurtaux, P. Horcajada, C. Serre, J.S. Chang, G. Ferey, V.A. de la Pena-O'Shea, C. Boissiere, D. Grosso, C. 34

Sanchez, Eur. J. Inorg. Chem. (2012) 5165-5174. [50] A. Dhakshinamoorthy, M. Alvaro, H. Garcia, Chem. Commun. 48 (2012) 1127511288. [51] X. Wang, L.G. Wang, J.B. Li, J.J. Qiu, C. Cai, H. Zhang, Sep. Purif. Technol. 122 (2014) 41-46. [52] S.S. Lin, M.D. Gurol, Environ. Sci. Technol. 32 (1998) 1417-1423.

35