Photoreactivity of metal-organic frameworks in the decolorization of methylene blue in aqueous solution

G Model CATTOD-9701; No. of Pages 8

ARTICLE IN PRESS Catalysis Today xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Catalysis Today journal homepage: www.elsevier.com/locate/cattod

Photoreactivity of metal-organic frameworks in the decolorization of methylene blue in aqueous solution Ekaterina A. Kozlova a,b,c , Valentina N. Panchenko a,b , Zubair Hasan d,1 , Namzul Abedin Khan d , Maria N. Timofeeva a,b,e,∗ , Sung Hwa Jhung d,∗∗ a

Boreskov Institute of Catalysis SB RAS, Prospekt Akad. Lavrentieva 5, 630090, Novosibirsk, Russian Federation Novosibirsk State University, Pirogova Str. 2, 630090, Novosibirsk, Russian Federation c Research and Educational Center for Energy Efficient Catalysis in Novosibirsk National Research University, Pirogova Str. 2, 630090, Novosibirsk, Russian Federation d Department of Chemistry and Green-Nano Materials Research Center, Kyungpook National University, Sankyuck-Dong, Buk-Ku, Daegu 702-701, Republic of Korea e Novosibirsk State Technical University, Prospekt K. Marksa 20, 630092, Novosibirsk, Russian Federation b

a r t i c l e

i n f o

Article history: Received 7 April 2015 Received in revised form 26 June 2015 Accepted 17 July 2015 Available online xxx Keywords: Metal-organic frameworks Photocatalysis Band gap Methylene blue

a b s t r a c t The effect of metal ions and functional groups in organic ligands on the photocatalytic properties of metal-carboxylates, such as isostructural MIL-100(M) (M = Al, Fe, V, and Cr), isoreticular UiO-66-R (R = H, NH2 , and NO2 ), MIL-47(V), and MIL-125(Ti), was investigated using the decolorization of methylene blue in aqueous solution at pH 7.0 and 20 ◦ C under UV radiation. The metal-carboxylates were characterized by diffuse reflectance UV/vis measurements. The determination of the band gap (Eg ) on the basis of DRUV/vis measurements showed that Eg values correlate with ionic covalent parameters, i.e., the strength of the interaction between metal and oxygen in the “M+n -O2− pair. According to DR-UV/vis spectroscopic and catalytic investigations, the reactivity of metal-carboxylates correlated with the band-gap value (Eg ), which could be adjusted by varying the nature of the metal ions in SBUs or functional groups in the organic ligands. © 2015 Published by Elsevier B.V.

1. Introduction Metal-organic frameworks (MOFs) are currently attracting attention from the point of view of their application in adsorption and catalysis. Catalytic applications of MOFs are related to their unique structural, textural, and physicochemical properties, which can be adjusted by changing the nature of the metal clusters and organic ligands in the framework [1,2]. Photocatalytic degradation of organic pollutants represents one of the fields of catalytic application [3–6]. Thus, the photocatalytic decolorization of methylene blue (MB) (Fig. 1) has been studied under solar radiation in the presence of MIL-53(Al, Fe,

∗ Corresponding author at: Boreskov Institute of Catalysis SB RAS, Prospekt Akad. Lavrentieva 5, 630090, Novosibirsk, Russian Federation. Tel.: +7 383 330 7284; fax: +7 383 330 8056. ∗∗ Corresponding author. Tel.: +82 53 950 5341; fax: +82 53 950 6330. E-mail addresses: [email protected] (M.N. Timofeeva), [email protected] (S.H. Jhung). 1 Present address: Department of Environment and Energy, Sejong University, Seoul 143-747, Republic of Korea.

and Cr) [7]; Fe3 O4 /MIL-100(Fe) core–shell microspheres [8]; MOFs formed by d10 metals (Cd and Ag); triangular ligands [6]; 3D framework structures [Co2 (C10 H8 N2 )]·[C12 H8 O(COO)2 ]2 , [Ni2 (C10 H8 N2 )2 ][C12 H8 O(COO)2 ]2 ·H2 O, and [Zn2 (C10 H8 N2 )]· [C12 H8 O(COO)2 ]2 [4]; ZIF-8 [9]; Co2 (dcpcpb)(␮3 -OH)(H2 O)2 and Cu4 (dcpcpb)2 (␮3 -OH)2 (CH3 OH)2 (H2 O) [10,11]; [Cu5 (H2 tmbtmp)2 (btb)2 (OH)2 ]·3H2 O [12]; and a titanium(IV)-based MOF (NTU-9, Ti2 (Hdhbdc)2 (H2 dhbdc), H4 dhbdc-2,5dihydroxyterephathalic acid) [13]. Application of these MOFs in the photocatalytic decolorization of MB clearly revealed their great potential as photocatalysts. In the presence of these materials, the time taken for complete decolorization of MB under visible-light irradiation was 20–60 min. Unfortunately, the comparison of photocatalytic activities of MOFs is practically absent in spite of the large amount of investigations. The high activity has been explained by the photocatalytic reaction. At the same time adsorption of MB on MOFs was not taken into account sometimes, despite the presence of a large amount of MOF in the reaction mixture (0.5–2 g/L) compared with the MB concentration (40 ppm–0.1 mmol/L). We can assume that this does not allow correlations between the catalytic activity of MOFs and their

http://dx.doi.org/10.1016/j.cattod.2015.07.026 0920-5861/© 2015 Published by Elsevier B.V.

Please cite this article in press as: E.A. Kozlova, et al., Photoreactivity of metal-organic frameworks in the decolorization of methylene blue in aqueous solution, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.07.026

G Model

ARTICLE IN PRESS

CATTOD-9701; No. of Pages 8

E.A. Kozlova et al. / Catalysis Today xxx (2015) xxx–xxx

2

2. Experimental

N

2.1. Materials

H3C

N

S

CH3

N

CH3

Cl

-

CH3

Fig. 1. Structure of methylene blue dye.

physicochemical properties such as band gap value (Eg ) and inductive effect of the functional groups in the organic linkers of MOFs. A predictive understanding of the relationship between the activity of MOFs and their corresponding physicochemical properties may be important when choosing MOFs for a particular application. Nowadays, experimental and computational techniques are rarely used to establish structural/physicochemical/photocatalytic property relationships, despite numerous studies on the photocatalytic properties of MOFs. It is well known that the electronic properties of MOFs depend on their chemical composition and structure [14–17], i.e., the nature of the organic ligands (linkers) and metal ions, or metal-oxide clusters, in the secondary building units (SBUs). The explanation of the electronic properties of MOFs is based on considering the SBUs of a metal-oxide cluster as a discrete quantum dot analogue, stabilized and interconnected by the conjugated organic linkers acting as the photon antenna [18,19]. Thus, the effect of the linker length (one, two, and three aromatic rings) on the Eg value has been demonstrated for UiO-6x, which was substituted with dicarboxylate (x = 6), biphenyl dicarboxylate (x = 7), and terphenyl dicarboxylate (x = 8), with a Zr6 O4 (OH)4 cluster as the inorganic fragment [20]. According to theoretical calculations, the Eg value decreases with increasing linker length for the hydroxylated cluster in the following order: 2.8 eV (x = 6) > 2.6 eV (x = 7) > 2.4 eV (x = 8) The Eg value can also be tuned by changing the dimensions of the metal cluster in the SBUs. According to Lin et al. [19], the size of SBU clusters of Zn-biphenyl dicarboxylates (Zn-BPDCs) such as IRMOF-9, Zn5-BPDC, and CPO-7 decreases in the following order: CPO − 7 > Zn5 − BPDC > IRMOF − 9 The Eg values also increase in the same order. This increase in Eg value with decreasing cluster size was attributed to the increasing quantum confinement effect of the SBUs, which was similar to those demonstrated for quantum dots. Effect of the functional groups in linker of MOFs on Eg value was found for UiO-66-R [20] and MIL-125-R (R = H, NH2 , CH3 , OH, or Cl) [21]. However, correlations between Eg values and catalytic properties of MOFs were not demonstrated. In this investigation, we wanted to estimate the effect of the nature of metal ions and functional groups in organic ligands on the band gap (Eg ) and photocatalytic properties of metal-carboxylates, based on experimental DR-UV/vis spectroscopic and catalytic data. The nature of the metal ions and the functional groups in organic ligands was studied using isostructural MIL-100(M) (M = Al, Fe, Cr, and V) and isoreticular UiO-66-R (R = H, NH2 , and NO2 ). We predicted that the variation in the chemical composition of MOFs should affect the Eg values, and, in turn, affect their photocatalytic properties. The catalytic properties were investigated in the photocatalytic degradation of MB by oxygen in air in an aqueous solution at pH 7.0 and 20 ◦ C under solar radiation. Note that MB is widely used as an organic dye to examine the photocatalytic activities of MOFs.

Zirconium(IV) chloride and terephthalic acid (TPA, H2 BDC), were purchased from Sigma–Aldrich (St. Louis, MO, USA). 2-aminoterephthalic acid (H2 N-H2 BDC) was procured from Alfa-Aesar (Ward Hill, MA, USA). UiO-66-R (R = H, NH2 and NO2 ) samples were synthesized following previously reported procedures [22,23], as were MIL100(Cr, Fe and V) [24,25], MIL-125(Ti) [27], MIL-100(Al) [28], and MIL-47(V) [29]. Textural properties and XRD patterns of MOFs are shown in Table S1 and Fig. S1 (supporting information). 2.2. Instrumental measurements The porous structure of the materials was determined from the adsorption isotherm of N2 at −196 ◦ C using Micromeritics ASAP 2400 equipment. The Brunauer–Emmett–Teller specific surface area (SBET ) was calculated from adsorption data over the relative pressure range 0.05–0.20. The total pore volume (V ) was calculated from the amount of nitrogen adsorbed at a relative pressure of 0.99. The X-ray diffraction patterns were measured with an X-ray ˚ radiation. diffractometer (ThermoARL) with Cu-K␣ ( = 1.5418 A) The DR-UV/vis spectra were recorded on a UV-2501 PC Shimadzu spectrometer with an IRS-250A accessory in the range 190–900 nm, with a 2 nm resolution. BaSO4 was used as a standard. 2.3. Photocatalytic activity tests The photocatalytic activity of the MOFs was studied in terms of the degradation of MB by dissolved oxygen in aqueous solution at pH 7.0 and 20 ◦ C, in a batch temperature-controlled photoreactor with a quartz window, under continuous stirring for 3 h. The distance between the light source and the reactor was 10 cm. In a typical experiment, 5 mg of the catalyst was added into 100 mL of MB aqueous solution (0.2 mmol/L) in a 200 mL reactor. The pH value of the suspension was adjusted to neutral (pH 7.0) by addition of NaOH solution (0.1 M). The reaction mixture was stirred in the dark for 60 min to ensure the establishment of an adsorption–desorption equilibrium before the suspension was illuminated by the full irradiation of a high-pressure mercury lamp (1000 W, USSR, main peaks at 310, 365 (maximum), 380, 405, 440, 555, and 585 nm, 21 mV/cm2 , predominantly wavelength 365 nm). The selected radiation wavelength of the lamp (365 nm) accounted for the fact that the width of the photon energy expected is higher than the region of fundamental absorption. The concentration of MB was analyzed by UV/vis spectroscopy at  = 660 nm using a UV/vis spectrophotometer (Cary 100 Varian) and a spectrometric quartz cell with a path length of 1 cm. Additionally, an experiment with a 365 nm LED (30 W, China, 25 mV/cm2 ) was carried out. The total organic carbon (TOC) content was determined in filtered aliquots of the reaction mixture using TOC/Nb Analyzer (Multi N/C 2100 S, Analytik Jena AG, Germany). 3. Results and discussion 3.1. DR-UV/vis investigation of MOFs The Eg values of metal-carboxylates were calculated from their DR-UV/vis spectra (Figs. S2–S3, supporting information) according to the well-known energy exponential relationship (Eq. (1)) [30]: (˛h)

1/2

= B · (h − Eg )

(1)

Please cite this article in press as: E.A. Kozlova, et al., Photoreactivity of metal-organic frameworks in the decolorization of methylene blue in aqueous solution, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.07.026

G Model CATTOD-9701; No. of Pages 8

ARTICLE IN PRESS E.A. Kozlova et al. / Catalysis Today xxx (2015) xxx–xxx

where ˛ is the absorbance, h is photon energy, B is an independent parameter of the photon energy for the respective transitions, and Eg is the band gap. Eq. (1) can then be rewritten as follows if h is equal to E (Eq. (2)): (˛ · E)1/2 = B · (E − Eg )

(2)

If (˛ · E)1/2 is plotted against E value, the Eg value can be found at the intersection, (˛ · E)1/2 = 0, on the corresponding graphic by extrapolating the linear region. DR-UV/vis spectra and correlations between the (˛h)1/2 and h of studied MOFs are shown in Figs. S2–S4 (supporting information). The Eg values determined from the steep absorption edge of the studied metal-carboxylates are shown in Table 2. 3.1.1. Effect of metal ions in the MOF framework on Eg value As can be seen from the experimental evidence (Table 2), the Eg value for isostructural MIL-100(M) (M = Al, Fe, Cr, and V) depended on the nature of the metal ions and decreased in the following order: MIL − 100(Al) > MIL − 100(Cr) > MIL − 100(Fe)∼MIL − 100(V) The explanation of this order can be based on the interaction between the SBUs of the metal-oxide cluster, as a discrete quantum dot, and the conjugated organic linkers. We decided to use the ionic covalent parameter (ICP) to estimate the extent of this interaction, because ICP takes into account not only the type (ionic to covalent) of Me–O bond (i.e., the interaction between the metal ion of the SBUs and the carboxylic groups of the organic linker), but also the extent (through the polarizability) of the negative charge borne by oxygen [31–33]. ICP can be calculated with Eq. (3): ICP = log P − 1.38 ·  + 2.07

(3)

where  is electronegativity (in Pauling-type unit) [34] and P is the polarizing power of the cation (P = e/r2 ; e is the formal charge and r is the Shannon ionic radius). In the case of MIL-100(V) and MIL-100(Fe) we used the e and r values of V3+ and Fe3+ , because their values were negligible [26,35]. However, considerable contributions would have been from these ions after calcination of samples at temperatures higher than 100 ◦ C. The ICP values for the MOFs studied are given in Table 2. Fig. 2A shows the correlation between ICP and Eg values for isostructural MIL-100(Al, Cr, Fe and V); the higher the ICP value, the higher the Eg value. This trend could be caused by the changing polarization of the M–O bond. The higher ICP value for MIL-100(Al) compared with that for MIL-100(Cr) indi˚ [36] than that of cated a smaller Al–O bond length (1.815–1.875 A) ˚ [37]. Cr–O (1.81–2.18 A) The decrease in the length of the M–O bond should strengthen the extent of the interaction between the SBUs and organic linkers. In other words, MIL-100(Al), with a high ICP value, needs more energy to transfer the charge from the carboxylic groups of the organic linker to the Al3+ ion, resulting in a larger Eg value than that of MIL-100(Cr, Fe and V). The effect of the nature of the metal ion on Eg value has previously been demonstrated for MIL-53(Al, Cr and Fe) [7]. Eg values decreased in the order: MIL − 53(Al) (3.87 eV) > MIL − 53(Cr) (3.20 eV) > MIL − 53(Fe) (2.7 eV) Based on the approach that we suggested above, we also found a correlation between ICP and Eg for MIL-53(Al, Fe, and Cr) (Fig. 2A). The possibility of tuning the Eg value to the range 1.88–2.88 eV by exchanging Zn atoms in pure MOF-74 (CPO-27) with varying amounts of Mg, Cu, or Co in the inorganic cluster was demonstrated by Botas et al. [16]. In addition, computational methods have pointed to the weak effect of the nature of the metal ion on the Eg

3

value of M-IRMOF (M = Be, Mg, Ca, Zn, and Cd) in comparison with the effect of the linker [38,39]. Moreover, we tried to use another approach for the qualitative estimation of the band gap of MOFs. According to prior research [40,41], the electronegativity of transition metal ions can be used to estimate the Eg value because the energy of the electronic level of an element is inversely proportional to its electronegativity. Thus, a linear relationship between Eg and 1/ was demonstrated for the transition metal oxides Ba(In1/3 Pb1/3 M 1/3 )O3 (M = Nb and Ta) with a perovskite structure [42]. We also tried to find a connection between Eg and  for metal-carboxylates with SBUs formed by transition metal ions. The relatively good correlation between the Eg value and 1/ is shown in Fig. 2B. This trend indicated that electronegativity could also be used for the qualitative estimation of the effect of nature of the transition metal on the Eg value. 3.1.2. Effect of functional group in the MOF framework on Eg value The possibility of tuning the Eg value of a MOF by changing the nature of the functional groups in the organic linker was investigated using isoreticular UiO-66-R (R = H, NH2 , and NO2 ) materials. According to experimental data, after the insertion of an NH2 -group into the organic linker, the Eg value decreased from 3.88 to 2.91 eV. Interestingly, the variation in the NH2 - group content of UiO-66-NH2 , due to changing the molar ratio of the synthetic mixtures of benzenedicarboxylic acid (H2 BDC) and 2aminobenzenedicarboxylic acid (H2 N-H2 BDC), led to a reduction in the Eg value (Fig. 3A). The decrease in the Eg value was also found for NO2 substitutions, where the changes were comparable to that of the unsubstituted structure (Fig. 2A). Investigations into the effect of the functional groups of UiO-66-R [20], UiO-68-R [42], MIL-88-R [43], and MIL-125-R [21] also showed a reduction in the Eg value. Thus, Eg for MIL-125-R (R = H, NH2 , CH3 , OH, or Cl) decreased in the following order [21]: H(3.9 eV) > CH3 , Cl(3.5 eV) > OH(2.8 eV) > NH2 (2.4 eV) Based on the previously reported data [17,44], the effect of resonance in the functional groups in the linkers on the Eg value can be revealed by Hammett analysis. Fig. 3B shows the correlation between the Eg values and resonance effects ( R ) of functional groups, characterizing the resonance effect of the groups on the distribution of p-electrons in the benzene ring. The linear correlation between the Eg and  R values indicated that Eg values, i.e., the semiconductor properties of the isoreticular UiO-66-R, depended on resonance effects in the organic linker. 3.2. Decolorization of methylene blue over metal-carboxylates The photocatalytic properties of metal-carboxylates were studied in terms of the degradation of MB. MB was selected as a model dye pollutant in aqueous solution. Experimental conditions of the catalytic test were selected based on previous reports. Firstly, the reaction rate depended on the pH value of the reaction medium [45–48], getting faster with increasing pH. Secondly, one should take into account that studied metal-carboxylates have a limited region of pH stability. Therefore, the choice of the pH value of the reaction medium should lead to the optimum combination between catalytic activity and stability. Thus, we investigated the photocatalytic performance of metal-carboxylates at pH 7.0. 3.2.1. Experimental verification of photocatalytic MB decolorization Adsorption of MB on the surface of metal-carboxylates should also be taken into account in the course of photocatalytic MB decolorization. The adsorption capacities of metal-carboxylates for MB

Please cite this article in press as: E.A. Kozlova, et al., Photoreactivity of metal-organic frameworks in the decolorization of methylene blue in aqueous solution, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.07.026

G Model

ARTICLE IN PRESS

CATTOD-9701; No. of Pages 8

E.A. Kozlova et al. / Catalysis Today xxx (2015) xxx–xxx

4

0,7

1,2

1,2

B

A

C

MIL-100(V)

ICP

MIL-100(Al) MIL-100(Cr)

0,8 MIL-100(V) MIL-100(Fe) MIL-53(Fe)

MIL-100(V) MIL-100(Fe)

0,8

0,6 MIL-100(Cr) UiO-66(Zr)

MIL-125(Ti)

0,4

-1 k 104, min

MIL-53(Al) MIL-53(Cr)

1/Χi

1,0

-1 k 104, min

0,8

MIL-100(Fe)

0,6

MIL-100(Cr) UiO-66(Zr)

0,5

0,6

MIL-125(Ti)

0,4

0,4

0,0

2

3

4

2

5

3

4

2,12

5

2,16

Eg, eV

Eg, eV

2,20

Q = lo g (

2,24 IP ⋅ Z ) r

2,28

Fig. 2. (A) Correlation between ICP and Eg values for isostructural MIL-100(Al, Cr, Fe and V) and MIL-53(Al, Fe and Cr) [7]. (B) Correlation between Eg value and the inverse electronegativity of metal ion and reaction constant of MB decolorization over metal-carboxylates under the irradiation. (C) Correlation between parameter Q and reaction constant.

Fig. 3. (A) Correlations between  R and Eg value: UiO-66 – (1), UiO-66-NH2 (50) – (2), UiO-66-NH2 – (3), UiO-66-NH2 (100) – (4), UiO-66-NO2 – (5), Zn-IRMOF-2-Br – (6) [13], Zn-MOF-5 – (7) [13]. (B) Correlation between  R and Eg of metal-carboxylates and reaction constant in MB decolorization over isoreticular UiO-66-R under the irradiation.

were studied at pH 7.0. In the reaction mixture, the concentrations of MB and metal-carboxylate were 0.2 mmol/L and 0.05 g/L, respectively. The adsorption kinetics of MB on MIL-100(Cr), MIL100(V), and UiO-66-NH2 are shown in Fig. 4. As can be seen from the

Time, min

Time, min

0,10

-60

UiO-66-NH2 0

60

120

Time, min

180

240

180

Time, min

240

0

60

120

180

1 2

2

0,10

-60

60

120

Time, min

180

MIL-100(Cr)

0,10

MIL-100(V)

0

0,15

UV-light

0,15

240

0,20

1

Ci, mmol/l

2

UV-light

0,15

120

0,20

1

Adsorption

Ci, mmol/l

0,20

240

Adsorption

180

Ci, mmol/l

120

60

UV-light

60

0

Adsorption

0

data, the adsorption of MB over MIL-100(Cr), MIL-100(V), and UiO66-NH2 was practically complete within 1 h. Similar trends were observed in the presence of other MOFs. The adsorption values of MB on metal-carboxylates are shown in Table 1. Based on both the

240

-60

0

60

120

180

240

Time, min

Fig. 4. (1) The adsorption kinetics of the MB on UiO-66-NH2 , MIL-100(V) and MIL-100(Cr) (experimental conditions: 100 mL of 0.2 mmol/L aqueous solution, 0.05 g/L of catalyst, pH 7.0 and 20 ◦ C); (2) the kinetics curves of photocatalytic MB decolorization over UiO-66-NH2 , MIL-100(V) and MIL-100(Cr) under the irradiation at pH 7.0 and 20 ◦ C.

Please cite this article in press as: E.A. Kozlova, et al., Photoreactivity of metal-organic frameworks in the decolorization of methylene blue in aqueous solution, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.07.026

G Model

ARTICLE IN PRESS

CATTOD-9701; No. of Pages 8

E.A. Kozlova et al. / Catalysis Today xxx (2015) xxx–xxx

5

Table 1 Textural data, adsorption and catalytic properties of M-BDCs and M-BTCs (M – V, Al, Cr, Fe, Ti and Zr). Sample

1 2 3 4 5 6 7 8 9

Textural data

MIL-125(Ti) UiO-66 UiO-66-NH2 UiO-66-NO2 MIL-100(Cr) MIL-100(V) MIL-100(Fe) MIL-100(Al) MIL-47(V)

ABET (m2 /g)

V (cm3 /g)

V␮ (cm3 /g)

912 670 682 560 1377 1525 1714 1450 838

0.54 0.30 0.33 0.34 0.64 0.75 0.79 0.68 0.43

0.30 0.29 0.26 0.18 0.40 0.43 0.38 0.37 0.38

aMB (mmol/g)

k·104 (min−1 )

0.40 0.16 0.08 0.16 0.21 0.08 0.32 – –

0.44 0.52 1.05 0.35 0.56 0.86 0.67 – –

aMB – adsorption value of MB (100 mL of 0.2 mmol/L aqueous solution, 5 mg of sample, pH 7.0 and 25 ◦ C).

Table 2 Band gap values of metal-carboxylates. Metal-carboxylates formed by H3 -BTC as organic linker MOF

MIL-100

Metal Eg (eV)

V3+ 2.75

Fe3+ 2.80

Cr3+ 3.86

Al3+ 4.06

3.2.3. Effect of metal ions in the framework of MOF on photocatalytic MB decolorization The results of the catalytic performance of isostructural MIL100(M) (M = Al, Fe, Cr, and V) are presented in Fig. S5 (supporting information). According to experimental data the photocatalytic activity of MIL-100(M) depends on the nature of the metal ions of SBUs and occurs in the following order:

Metal-carboxylates formed by H2 -BDC as organic linker MOFs

MIL-125

MIL-47

MIL-101

UiO-66

Metal Eg (eV)

Ti3+ 2.70

V3+ 2.76

Cr3+ 3.66

Zr4+ 3.88

adsorbed amount of MB and the rapid kinetics of adsorption within 1 h, irradiation by a high-pressure mercury lamp (1000 W, predominantly at a wavelength of 365 nm) was started after storing the reaction mixture in the dark for 1 h. A control experiment, without any catalyst, indicated that a small amount of MB degradation was observed under UV irradiation (Fig. S5, supporting information). For MIL-100(V), the degradation of MB was carried out under a high-pressure mercury lamp and 365-nm LED irradiation. In both cases, in the presence of MIL-100(V) reaction rates were similar, with the rate under 365 nm LED without catalyst lower than under the high-pressure mercury lamp (Fig. 5A). This difference may be caused by negligible absorption of light by MB at 365 nm [49]. The high-pressure mercury lamp had an emission peak at 585 nm, where MB shows a good absorption [49]; thus, the rate of MB degradation in a blank experiment was noticeable. For experimental verification of photocatalytic MB decolorization, we compared the rates of change of MB concentration and total organic content (TOC) during the reaction. The decrease in TOC (Fig. 5B) pointed to a photocatalytic process in the presence of MIL-100(V).

3.2.2. Kinetics of photocatalytic MB decolorization The kinetic curves of photocatalytic MB decolorization over metal-carboxylates are illustrated in Figs. 4 and S5 (supporting information). The kinetics curves can be described with a pseudofirst-order equation (Eq. (4)) (Fig. S6, supporting information): ln

C  i

C0

= −kt

(4)

where C0 and Ci are initial and current concentrations of MB, respectively; t is time of reaction; and k is the rate constant. Eq. (4) was used to obtain the rate constants of the reactions in the presence of metal-carboxylates (Table 1).

MIL − 100(V)∼MIL − 100(Fe) > MIL − 100(Cr) This order is in agreement with the reverse order of the Eg values (Fig. 2B). A high Eg value provokes low photocatalytic activity. Note that similar trend also is observed for porous metal-carboxylates without functional groups (MIL-125(Ti) and UiO-66(Zr)) (Fig. 2B), which agrees with the order of 1/ (Fig. 2B). It is well known that the photocatalytic activity of different materials is linked to their acidic and redox properties. Chung suggested parameter Q (Eq. (5)) for the estimation of their joint action [48]: Q = log

 IP · Z  r

(5)

where IP is the ionization potential (V), Z is the valency of the metal ion, and r is the radius of the metal ion (Å). Parameter Q allows to take both the acidic (Z/r) and redox (IP) properties of a metal ion into account, simultaneously. The acidity is an important factor for high photocatalytic activity, because an acid metal site tends to attract electrons with the formation of a covalent bond. Metal ions with strong acidity may increase their electron withdrawing power or ionization potential. At the same time, smaller metal ions attract electrons more strongly and become more acidic. Parameter Q takes all of these factors into consideration. We used parameter Q for prediction of the photocatalytic activity of metal-carboxylates. According to calculations the Q value decreased in the following order (Fig. 2C): MIL − 125(Ti) > UiO − 66(Zr) > MIL − 100(Cr) > MIL − 100(Fe) > MIL − 100(V)

At the same time the reaction rate increases in inverse order (Fig. 2C). Note that the activity of metal-carboxylates correlates better with the Q parameter than with Eg , which can be explained by the different parameters taken into account in the calculation of these characteristics. At present, the interpretation and understanding of this correlation needs further investigation with catalytic and theoretical methods.

Please cite this article in press as: E.A. Kozlova, et al., Photoreactivity of metal-organic frameworks in the decolorization of methylene blue in aqueous solution, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.07.026

G Model

ARTICLE IN PRESS

CATTOD-9701; No. of Pages 8

E.A. Kozlova et al. / Catalysis Today xxx (2015) xxx–xxx

6

A

B

1,1

0,20

0,16

0 C/C

UV-light

0,18

Adsorption

Ci, mmol/l

1,0

0

1

0,8 blank mercury lamp MIL-V mercury lamp MIL-V LED blank LED

0,14 -60

0,9

60

120

180

2

0,7 240

0

60

Time, min

120

180

240

Time, min

Fig. 5. (A) Kinetic curves of photocatalytic MB decolorization over MIL-100(V) under full irradiation of the high pressure mercury lamp and 365-nm LED. (B) The changes of MB concentration (1) and TOC (2) in the course of photocatalytic MB decolorization over MIL-100(V) under irradiation of the 365-nm LED. (Experimental conditions: 100 mL of 0.2 mmol/L aqueous solution, 0.05 g/L of catalyst, pH 7.0 and 20 ◦ C).

3.2.4. Effect of functional groups in the framework of MOF on photocatalytic MB decolorization The effect of the functional groups on the catalytic activity was studied for isoreticular UiO-66-R (R = H, NO2 , and NH2 ). It was found that the catalytic activity correlated with the resonance effect of the functional groups ( R ). One can see from Fig. 3B that the reaction rate decreased in the order of UiO-66-NH2 > UiO-66 > UiO66-NO2 . This order can be explained by the resonance effect of functional groups ( R ) in the ligands. This order is in agreement with the correlation between band gap values and  R (Fig. 3B). We can assume that the electron-donating NH2 group adjacent to the p-system activates the aromatic ring by increasing the electron density on the ring through a resonance-donating effect. Such electron redistribution favors the formation and stability of intermediates formed during the course of the reaction. At the same time, an electron-withdrawing NO2 group adjacent to the p-system deactivates the aromatic ring by decreasing the electron density on the ring through a resonance-withdrawing effect that leads to a decrease in the reaction rate. Therefore, the nature of functional groups in the organic linkers of MOFs is also a main factor influencing their photocatalytic activity. We believe that correlations found in this investigation are in agreement with a mechanism of the photocatalytic degradation of MB in the presence of MOFs, where the direct excitation of metal-oxocluster is the main contributor to the photocatalytic performance of MOFs [5]. The initial process of photocatalysis consists of the generation of electron–hole pairs. The electrons and holes migrate to the surface of the MOF, then the photoinduced

energy transfers to the adsorbed species: electrons reduce the oxygen (O2 ) to oxygen radicals (• O2 − ), while the photogenerated holes (h+ ) oxidize the hydroxyl (H2 O) to hydroxyl radicals (• OH) which react easily with the MB adsorbed on MOF to produce CO2 [9]. The change in type metal in SBU or functional group into organic linker affects the Eg , and, in turn, affects the rate of the generation of electron–hole pairs, and, finally, photocatalytic properties of MOFs. 3.2.5. Comparison of the efficiencies of different MOFs It is interesting to compare the efficiencies of the most active MOFs investigated in this study with those reported in the literature for photocatalytic MB decolorization. Because specific surface areas of MOFs are different, we estimated their efficiency using specific average degradation rate (Ws ): Ws =

Wm SBET

where Wm is the average degradation rate. Wm and Ws were calculated using the linear section of the MB degradation kinetic curves. However, the comparison of efficiencies is relative because experimental conditions were different. The main results are shown in Table 3. In general, efficiencies of MIL-100(Cr), MIL-100(V), and UiO-66 are higher than those for ZIF-8 [9], Co2 (dcpcpb)(␮3 OH)(H2 O)2 [10,11], Cu4 (dcpcpb)2 (␮3 -OH)2 (CH3 OH)2 (H2 O) [10,11], and [Cu5 (H2 tmbtmp)2 (btb)2 (OH)2 ]·3H2 O [11,12]. The insertion of an NH2 group into the framework of UiO-66 results in a two-fold increase in the efficiency (Table 3, runs 3–4). The comparison of the efficiencies of MIL-100(V) and UiO-66-NH2 with that of TiO2

Table 3 The activities of different photocatalysts in the degradation of MB in aqueous solutions under UV-irradiation. No.

1 2 3 4 5 6 7 8 9 10 a

Photocatalyst

MIL-100(Cr) MIL-100(V) UiO-66 UiO-66-NH2 ZIF-8 Co2 (dcpcpb)(␮3 -OH)(H2 O)2 Cu4 (dcpcpb)2 (␮3 -OH)2 (CH3 OH)2 (H2 O) [Cu5 (H2 tmbtmp)2 (btb)2 (OH)2 ]·3H2 O TiO2 Degussa P25 TiO2 Degussa P25

Photocatalyst (mg)

(g/L)

5 5 5 5 25 70 70 30 50 375

0.05 0.05 0.05 0.05 0.5 0.35 0.35 0.15 2.5 0.5

Initial MB concentration (mg/L) 64 64 64 64 10 16 16 16 23 27

Degradation rate a

Ref. a

Wm (mg MB/h)

Ws (mg MB/(h g)

0.21 0.31 0.20 0.40 1.00 1.55 1.36 0.94 0.42 50.4

0.042 0.062 0.040 0.080 0.040 0.022 0.020 0.031 0.008 0.134

This work This work This work This work [9] [10,11] [10,11] [11,12] [44] [50]

Wm – average degradation rate, Ws – specific average degradation rate. Wm and Ws were calculated using linear section of the MB degradation kinetic curves.

Please cite this article in press as: E.A. Kozlova, et al., Photoreactivity of metal-organic frameworks in the decolorization of methylene blue in aqueous solution, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.07.026

G Model CATTOD-9701; No. of Pages 8

ARTICLE IN PRESS E.A. Kozlova et al. / Catalysis Today xxx (2015) xxx–xxx

Degussa P25, the most commonly used in commercial photocatalysis [45,50], shows that their efficiencies are better than TiO2 with respect to the concentration of MB and amount of catalyst (Table 3, runs 2, 4, 9–10).

7

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cattod.2015.07. 026

4. Conclusions References In this work, nine metal-carboxylates were prepared and characterized by physicochemical methods, namely isostructural MIL-100(M) (M = Al, Fe, V, and Cr), isoreticular UiO-66-R (R = H, NH2 and NO2 ), MIL-47(V), and MIL-125(Ti). The semiconductor behavior of MOFs was studied by DR-UV/vis spectroscopy. It was found that the nature of metal ions in SBUs allowed us to adjust the Eg value of isostructural MIL-100(M) owing to the change in the extent of interaction between the SBUs and organic linkers. Correlation between ICP and Eg also indicated that variation of the metal ion leads to changing M–O bond lengths, and, therefore, changing energy for the transfer of charge from the carboxylic groups of the organic linker to the metal ion. Correlation between the Eg and 1/ values was suggested for use in the qualitative estimation of the effect of nature of the transition metal on the Eg value. Hammett analysis revealed the effect of the nature of functional groups in the organic linker on the Eg value of isoreticular UiO66-R. It was demonstrated that the semiconductor properties of isoreticular UiO-66-R mainly depend on resonance effects ( R ) in the organic linker. The relationships between the semiconductor and photocatalytic properties of metal-carboxylates have been investigated in terms of the decolorization of methylene blue (MB) in an aqueous solution at pH 7.0 and 20 ◦ C under solar radiation. It was demonstrated that the activity of metal-carboxylates depends on the Eg value. Parameter Q, which takes into consideration both acidic (Z/r) and redox (IP) properties of the metal ion, was suggested for the prediction of the activity of metal-carboxylates. The Q value decreases in the order: MIL − 125(Ti) > UiO − 66(Zr) > MIL − 100(Cr) > MIL − 100(Fe) > MIL − 100(V)

The reaction rate increases in the reverse order. The photoreactivity of isoreticular UiO-66-R depends on the resonance effect of the functional groups ( R ), decreasing in the order: UiO − 66 − NH2 > UiO − 66 > UiO − 66 − NO2 We can assume that correlations obtained in this investigation can be useful not only for the choice of catalysts based on MOFs in the decolorization of dyes, but these results may be used in other photocatalytic processes.

Acknowledgments This work was supported by SB RAS project V.44.2.12 and RFBR (Grants 14-03-00854, 14-05-00297, and 15-33-20458) and was supported by Kyungpook National University Research Fund, 2014. We also gratefully acknowledge the support of the Ministry of Education and Science of the Russian Federation (MK-3141.2015.3 and NSh-1183.2014.3). The work was performed with the partial support of the Skolkovo Foundation (Grant Agreement for Russian educational organization No 1 on 28.11.2013). We wish to thank Dr. A.A. Rastorguev for his thorough and critical reading of the manuscript and many helpful suggestions and valuable advices.

[1] A. Corma, H. Garcia, F.X. Llabresi Xamena, Chem. Rev. 110 (2010) 4606–4655. [2] J. Liu, L. Chen, H. Cui, J. Zhang, L.I. Zhang, C.-Y. Su, Chem. Soc. Rev. 43 (2014) 6011–6061. [3] J.-L. Wang, C. Wang, W. Lin, ACS Catal. 2 (2012) 2630–2640. [4] P. Mahata, G. Madras, S. Natarajan, J. Phys. Chem. B 110 (2006) 13759–13768. [5] M.C. Das, H. Xu, Z. Wang, G. Srinivas, W. Zhou, Y.-F. Yue, V.N. Nesterov, G.B. Qian, Chem. Commun. 47 (2011) 11715–11717. [6] W.-Q. Kan, B. Liu, J. Yang, Y.-Y. Liu, J.-F. Ma, Cryst. Growth Des. 12 (2012) 2288–2298. [7] J.-J. Du, Y.-P. Yuan, J.-X. Sun, F.-M. Peng, X. Jianga, L.-G. Qiu, A.-J. Xie, Y.-H. Shen, J.-F. Zhu, J. Hazard. Mater. 190 (2011) 945–951. [8] C.-F. Zhang, L.-G. Qiu, F. Ke, Y.-J. Zhu, Y.-P. Yuan, G.-S. Xu, X. Jiang, J. Mater. Chem. A 1 (2013) 14329–14334. [9] H.-P. Jing, C.-C. Wang, Y.-W. Zhang, P. Wang, R. Li, RSC Adv. 4 (2014) 54454–54462. [10] P. Du, Y. Yang, Y.-Y. Liu, Y.-C. He, H.-M. Zhang, J.-F. Ma, Polyhedron 70 (2014) 180–187. [11] C.-C. Wang, J.-R. Li, X.-L. Lv, Y.-Q. Zhang, G. Guo, Energy Environ. Sci. 7 (2014) 2831–2867. [12] S. Zhou, Z.-G. Kong, Q.-W. Wang, C.-B. Li, Inorg. Chem. Commun. 25 (2012) 1–4. [13] J. Gao, J. Miao, P.-Z. Li, W.Y. Teng, L. Yang, Y. Zhao, B. Liu, Q. Zhang, Chem. Commun. 50 (2014) 3786–3788. [14] M. Alvaro, E. Carbonell, B. Ferrer, F.X. Llabres i Xamena, H. Garcia, Chem. Eur. J. 13 (2007) 5106–5112. [15] Y. Horiuchi, T. Toyao, M. Saito, K. Mochizuki, M. Iwata, H. Higashimura, M. Anpo, M. Matsuoka, J. Phys. Chem. C 116 (2012) 20848–20853. [16] J.A. Botas, G. Calleja, M. Sanchez-Sanchez, M.G. Orcajo, Int. J. Hydrogen Energy 36 (2011) 10834–10844. [17] J. Gascon, M.D. Hernandez-Alonso, A.R. Almeida, G.P.M. van Klink, F. Kapteijn, G. Mul, ChemSusChem 1 (2008) 981–983. [18] S. Bordiga, C. Lamberti, G. Ricchiardi, L. Regli, F. Bonino, A. Damin, K.-P. Lillerud, M. Bjorgen, A. Zecchina, Chem. Commun. 20 (2004) 2300–2301. [19] C.-K. Lin, D. Zhao, W.-Y. Gao, Z. Yang, J. Ye, T. Xu, Q. Ge, S. Ma, D.-J. Liu, Inorg. Chem. 51 (2012) 9039–9044. [20] E. Flage-Larsen, A. Royset, J.H. Cavka, K. Thorshaug, J. Phys. Chem. C 117 (2013) 20610–20616. [21] C.H. Hendon, D. Tiana, M. Fontecave, C. Sanchez, L. D’arras, C. Sassoye, L. Rozes, C. Mellot-Draznieks, A. Walsh, J. Am. Chem. Soc. 135 (2013) 10942–10945. [22] S.J. Garibay, S.M. Cohen, Chem. Commun. 46 (2010) 7700–7702. [23] G. Akiyama, R. Matsuda, H. Sato, M. Takata, S. Kitagawa, Adv. Mater. 23 (2011) 3294–3297. [24] G. Férey, C. Serre, C. Mellot-Draznieks, F. Millange, S. Surblé, J. Dutour, I. Margiolaki, Angew. Chem. Int. 43 (2004) 6296–6301. [25] P. Horcajada, S. Surblé, C. Serre, D.-Y. Hong, Y.-K. Seo, J.-S. Chang, J.-M. Grenèche, I. Margiolaki, G. Férey, Chem. Commun. 27 (2007) 2820–2822. [26] A. Lieb, H. Leclerc, T. Devic, C. Serre, I. Margiolaki, F. Mahjoubi, J.S. Lee, A. Vimont, M. Daturi, J.-S. Chang, Micropor. Mesopor. Mater. 157 (2012) 18–23. [27] M. Dan-Hardi, C. Serre, T. Frot, L. Rozes, G. Maurin, C. Sanchez, G. Férey, J. Am. Chem. Soc. 131 (2009) 10857–10859. [28] N.A. Khan, J.S. Lee, J. Jeon, C.-H. Jun, S.H. Jhung, Micropor. Mesopor. Mater. 152 (2012) 235–239. [29] K. Barthelet, J. Marrot, D. Riou, G. Férey, Angew. Chem., Int. 41 (2002) 281–284. [30] J. Tauc, R. Grigorovici, A. Vancu, Phys. Status Solidi B 15 (1966) 627–637. [31] E. Bordes-Richard, P. Courtine, in: J.L.G. Fierro (Ed.), Chemistry and Applications, 2006, pp. 319–352. [32] J.A. Duffy, Geochim. Cosmochimic. Acta 57 (1993) 3961–3970. [33] J.A. Duffy, J. Solid State Chem. 62 (1986) 145–157. [34] K. Li, D. Xue, J. Phys. Chem. A 110 (2006) 11332–11337. [35] H. Leclerc, A. Vimont, J.-C. Lavalley, M. Daturi, A.D. Wiersum, P.L. Llwellyn, P. Horcajada, G. Ferey, C. Serrec, Phys. Chem. Chem. Phys. 13 (2011) 11748–11756. [36] C. Volkringer, D. Popov, T. Loiseau, G. Ferey, M. Burghammer, C. Riekel, M. Haouas, F. Taulelle, Chem. Mater. 21 (2009) 5695–5697. [37] G. Ferey, C. Serre, C. Mellot-Draznieks, F. Millange, S. Surbleґ, J. Dutour, I. Margiolaki, Angew. Chem. Int. 43 (2004) 6296–6301. [38] M. Fuentes-Cabrera, D.M. Nicholson, B.G. Sumpte, M. Widom, J. Chem. Phys. 123 (2005), 124713-1-124712-5. [39] L.-M. Yang, P. Ravindran, P. Vajeeston, M. Tilset, Chem. Chem. Phys. 14 (2012) 4713–4723. [40] S.G. Hur, T.W. Kim, S.-J. Hwang, H. Park, W. Cho, S.J. Kim, J.-H. Choy, J. Phys. Chem. B 109 (2005) 15001–15007. [41] H.J. Huheey, E.A. Keiter, R.L. Keiter, Inorganic Chemistry: Principles of Structure and Reactivity, HarperCollins, New York, 1993, pp. 171. [42] E. Flage-Larsen, K. Thorshaug, Inorg. Chem. 53 (2014) 2569–2572.

Please cite this article in press as: E.A. Kozlova, et al., Photoreactivity of metal-organic frameworks in the decolorization of methylene blue in aqueous solution, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.07.026

G Model CATTOD-9701; No. of Pages 8 8

ARTICLE IN PRESS E.A. Kozlova et al. / Catalysis Today xxx (2015) xxx–xxx

[43] P. Horcajada, F. Salles, S. Wuttke, T. Devic, D. Heurtaux, G. Maurin, A. Vimont, M. Daturi, O. David, E. Magnier, N. Stock, Y. Filinchuk, D. Popov, C. Riekel, G. Férey, C. Serre, J. Am. Chem. Soc. 133 (2011) 17839–17847. [44] M.N. Timofeeva, V.N. Panchenko, J.W. Jun, Z. Hasan, M.M. Matrosova, S.H. Jhung, Appl. Catal. A: Gen. 471 (2014) 91–97. [45] A. Houas, H. Lachheb, M. Ksibi, E. Elaloui, C. Guillard, J.-M. Herrmann, Appl. Catal. B: Environ. 31 (2001) 145–157.

[46] W.S. Kuo, P.H. Ho, Chemosphere 45 (2001) 77–83. [47] A. Franco, M.C. Neves, M.M.L. Ribeiro Carrott, M.H. Mendonca, M.I. Pereira, O.C. Monteiro, J. Hazard. Mater. 161 (2009) 545–550. [48] J.S. Chung, J. Macromol. Sci. Chem. A27 (1990) 479–490. [49] X. Liu, Yu. Yang, Xi. Shi, K. Li, J. Hazard. Mater. 283 (2015) 267–275. [50] H. Lachheb, E. Puzenat, A. Houas, M. Ksibi, E. Elaloui, Ch. Guillard, J.-M. Herrmann, Appl. Catal. B: Environ. 39 (2002) 75–90.

Please cite this article in press as: E.A. Kozlova, et al., Photoreactivity of metal-organic frameworks in the decolorization of methylene blue in aqueous solution, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.07.026