Article pubs.acs.org/IC
Metal−Organic Framework Based on Isonicotinate N‑Oxide for Fast and Highly Efficient Aqueous Phase Cr(VI) Adsorption Leila Aboutorabi,† Ali Morsali,*,† Elham Tahmasebi,† and Orhan Büyükgüngor‡ †
Department of Chemistry, Faculty of Basic Sciences, Tarbiat Modares University, P.O. Box 14115-175, Tehran, Iran Department of Physics, Faculty of Arts and Sciences, Ondokuz Mayis University, 55139 Kurupelit, Samsun, Turkey
‡
S Supporting Information *
ABSTRACT: Synthesis of new porous materials has been developed for efficient capture of pollutants in environmental sciences. Here, the application of a new metal−organic framework (TMU-30) has been reported based on isonicotinate N-oxide as an adsorptive site for fast and highly efficient aqueous phase adsorption of Cr(VI). The adsorption process showed no remarkable effect over a pH range of 2−9. The maximum capacity of the adsorption was reached in just less than 10 min and followed the pseudo-second-order kinetics. The maximum capacity of 2.86 mol mol−1 (145 mg/g) was obtained according to Langmuir model at 298 K. The spontaneous adsorption and an endothermic process were controlled by positive entropy changes. XPS analysis revealed electrostatic interactions between N-oxide groups of TMU-30 and Cr(VI) species, which were responsible for the adsorption process.
■
increase the flexibility and enrich coordination modes to form fascinating structures.26,27 Newly synthesized cationic MOFs have been applied for exchange and removal of oxo-anion contaminants, such as ReO4−, CrO42, and TcO4−,21,22,28 but to the best of our knowledge, there has been no report of MOFs with inherent potential positive charge. The main aim of this research is to prepare a MOF with sufficient water stability that can be used as an adsorbent for highly efficient Cr(VI) adsorption from aqueous solutions. Here, we designed and synthesized a new 3D framework, [Pb(INO)2]2·DMF (TMU-30), (TMU stands for Tarbiat Modares University), based on isonicotinate N-oxide (INO) consisting of 1D channels. N-Oxide groups with a potential positive charge can be an appropriate site for efficient adsorption of chromate species and generate electrostatic interactions with them. The effects of various experimental parameters like pH, initial Cr(VI) concentration, contact time, and temperature are examined on Cr(VI) adsorption. Fast adsorption kinetics, large adsorption capacity, and excellent water stability over a wide range of pH (2−9) make TMU-30 a high potential adsorbent to adsorb Cr(VI) from water.
INTRODUCTION Metal−organic frameworks (MOFs) have been developed greatly in recent years. Porous frameworks of MOFs can be tuned for their characteristics such as pore sizes, surface areas, organic functional groups, thermal stability, and available metal sites.1−4 These types of structures have attracted intensive interest in gas adsorption,5 storage,6 sensing,7 drug delivery,8 catalysis,9 and photocatalysis,10 but they are less investigated for pollutant adsorption and removal.11−14 MOFs are suitable candidates for liquid phase adsorption due to their pore size distributions.15 Stability of MOFs in water has become a challenging issue in aqueous phase separations.16,17 MOF behavior in the presence of water is a topic of significant importance when considering these materials for adsorption applications.18 Cr(VI), which is produced by some sources related to the genesis and development of the human, is identified as a common pollutant because of its high toxicity and mobility.19,20 The Environmental Protection Agency’s (EPA) legal limit for total chromium is 100 ppb in drinking water.15,16 Investigation of new materials for efficient adsorption and removal of Cr(VI) is extremely important because conventional operation technologies and materials possess insufficient capacities, slow rates, and low selectivity.21−23 The presence of particular functional groups on the surface may enhance the adsorption selectivity and efficiency of MOFs toward certain contaminants.24,25 The isonicotinic acid N-oxide ligand contains two different coordinating groups in which the carboxylate groups act in different modes and N-oxide groups possess opposite charges on nitrogen and oxygen atoms. Such intrinsic features © XXXX American Chemical Society
■
EXPERIMENTAL SECTION
Materials and Instrumentations. Lead(II) nitrate, potassium chromate, and isonicotinic acid N-oxide (HINO) were purchased from Merck and Aldrich. An Electrothermal 9100 apparatus was applied for melting point measurements. IR spectra were measured on a Thermo Received: March 1, 2016
A
DOI: 10.1021/acs.inorgchem.6b00522 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
initial Cr(VI) concentrations varying from 5 to 100 mg L−1) was poured into a Falcon conical tube with 10.0 mg of TMU-30 and shaken for 2 h at 298 K. The mixture was then centrifuged at 12000 rpm for 5 min, and the concentration of remaining Cr(VI) was determined by ICP analysis. The equilibrium adsorption capacity was determined using eq 3,
Scientific Nicolet IR100 (Madison, WI) Fourier-transform infrared (FT-IR) spectrometer. The thermal behavior was investigated by a PLSTA 1500 apparatus at a temperature ramp of 10 °C min−1 up to 600 °C under a continuous flow of dry nitrogen. Powder X-ray diffraction (XRD) patterns were measured with a Philips X’pert diffractometer by using monochromated Co Kα radiation (generator tension and current, 40 kV and 40 mA, respectively). After adsorption, TMU-30 was characterized by field-emission scanning electron microscopy (FESEM) on a Hitachi S4160 instrument with a gold coating. Inductively coupled plasma optical emission spectrometry (ICP-OES) was used for simultaneous determination of the target elements on a Varian Vista-PRO instrument (Springvale, Australia) with a charge-coupled detector (CCD). X-ray photoelectron spectroscopy (XPS) of the samples was conducted with a Bestec instrument (Germany). Single-Crystal X-ray Analysis. The single-crystal X-ray data were collected on a STOE IPDS II image plate diffractometer with graphite monochromated Mo Kα radiation (λ = 0.71073 Å) at 296 K by using the ω-scan technique. The structure was solved by direct methods and refined through the full-matrix least-squares method using SHELXL97:30,29 Data collection, Stoe X-AREA;31 cell refinement, Stoe XAREA;31 data reduction, Stoe XRED.32 Analysis of the structure and presentation of the results were carried out by PLATON33 and ORTEP-331 software. Preparation of TMU-30. TMU-30 single crystals suitable for Xray diffraction were synthesized by vial method (Figure S1 in the SI). Pb(NO3)2 (0.01 g, 0.3 mmol) and HINO (0.084 g, 0.6 mmol) were mixed in 20 mL of DMF. The mixture was stirred until complete dissolution, and then it was dispensed into several vials, and the vials were kept in an oven with the temperature of 353 K. After 72 h, yellow crystals of TMU-30 (Figure S1 in the SI) were separated from mother solution (yield 80% based on Pb, bp >573 K). FT-IR (KBr, cm−1): 635 cm−1 (s, v (Pb−O)), 783 cm−1 (m, voop (C−H)ring), 860 cm−1 (s, v (C−C)), 1134 cm−1 (s, v (C−H)ring), 1210 cm−1 (vs, v (N−O)N‑oxide), 1376 cm−1 (vs, vsym (CO2−Pb)), 1480 (m, v (CC)ring), 1543 and 1585 cm−1 (s, vasym (CO2−Pb)), 1655 cm−1 (m, v (CO)DMF), 3101 cm−1 (w, vsym (C−H)). Elemental analysis (%) calculated: C 30.38, H 1.95, N 6.83. Found: C 30.4, H 2.1, N 7.2. Batch Adsorption Experiments. Potassium chromate (K2CrO4, 373.5 mg) was dissolved in 100 mL of deionized water to prepare a Cr(VI) solution with the concentration of 1000 mg L−1. The other solutions were obtained by appropriate dilution of the stock solution. The amount of solid adsorbent was adjusted to be 10 mg in all experiments. Effect of pH. Evaluation of pH effects on Cr(VI) adsorption was done at a pH range of 2.0−9.0. The adjustment of pH was performed by adding either 0.1 M HCl or 0.1 M NaOH while the initial concentration of Cr(VI) was kept at 30 mg L−1 for all samples. The Cr(VI) adsorption efficiency (% adsorption) was obtained using eq 1,
⎛ C − Ce ⎞ %adsorption = ⎜ 0 ⎟ × 100 ⎝ C0 ⎠
qe =
⎛ C0 − C t ⎞ ⎜ ⎟V ⎝ m ⎠
(3)
where qe is the equilibrium amount of Cr(VI) adsorbed per unit mass of adsorbent (mg g−1) and C0 and Ce are initial and equilibrium concentrations of Cr(IV), respectively.
■
RESULTS AND DISCUSSION
Characterization of TMU-30. The single-crystal X-ray diffraction analysis revealed that TMU-30 crystallizes in the monoclinic space group C2/c (Table S1 in the SI). Selected bond lengths and angles are listed in Table S2 (SI). The experimental PXRD of the synthesized TMU-30 is exactly comparable in pattern to the simulated one (Figure S2 in the SI). Each lead(II) center is surrounded by seven INO ligands in two different coordination modes forming a nine-coordinated sphere around lead(II): five oxygen atoms from carboxylate groups and four oxygen atoms from N-oxide groups (Figure S3a in the SI). Carboxyl and N-oxide groups bind in modes chelating (μ2), chelating−bridging (μ−η2:η1),34 and μ2−O, respectively (Figure S3b in the SI). This pattern is repeated infinitely to form 1D Pb−O chains along the b-axis and faceshared PbO9 polyhedra (Figure 1a). The Pb−O distances are 2.452−2.981 and 2.637−2.854 Å for oxygen atoms of carboxylate and N-oxide groups, respectively. The PbO9 nodes are linked by C(carboxylate) and N(N‑oxide) atoms, which connect each node to four neighboring rods in the a and c directions, generating 7.1 × 6.8 Å2 1D rhombic channels in the b axis (Figure 1b). DMF molecules are located in the 1D channels comprising 20.4% of the crystal volume (2993.0 Å3 per unit cell) calculated by PLATON (Figure 1c). The framework can be simplified by using 3-connected nets instead of metal centers and INO ligands. The most common 3connected nets are the (10,3) nets [Well’s (10,3)].35,36 Additionally, N and C(carboxylate) atoms around lead(II) are arranged so that 3-connected nets with (10,3)-d37 topology are constructed (Figure 2a). In TMU-30, connection of C1−N2 and C7−N1 from INO ligands gives rise to 4-fold cubic helices with opposite handedness (red and green linkages in Figure 2a,b). Stability of TMU-30. The weight loss for TMU-30 according to thermogravimetric analysis (TGA) data was 6.12% at ∼465 K corresponding to 1 mol of DMF per mole of TMU-30. This framework was stable up to 592 K (Figure S4 in the SI). In order to evaluate the water stability of TMU-30, it was soaked in water for 24 h. The PXRD patterns before and after the immersion indicated no differences in the patterns except changing two peaks of (002) and (402)̅ located at 14.16° and 14.97° to a peak at 14.51° (Figure S5 in the SI). This alteration in the pattern can be attributed to decrease in lattice plane spacing due to exchange of DMF (kinetic diameter of 5.5 Å) with water (kinetic diameter of 2.6 Å). TGA (Figure S6 in the SI) and FT-IR (Figure S7 in the SI) analyses also confirmed this justification.38 Cr(VI) Adsorption Studies. Since Cr(VI) ions exist in the form of oxo-anions in aqueous solutions, TMU-30 containing
(1)
where C0 and Ce are the initial and the equilibrium Cr(VI) concentrations (mg L−1), respectively. Kinetics for the Cr(VI) Adsorption. Ten samples were prepared by adding a fixed concentration of Cr(VI) (30 mg L−1) to Falcon 50 mL conical tubes containing 10 mg of TMU-30 at 298 K. The adsorption process was stopped at different times from 1 to 120 min. Then, the mixture was centrifuged at 12000 rpm for 5 min and sampled for ICP analysis. The amount of Cr(VI) adsorbed was calculated using eq 2,
qt =
⎛ C0 − Ce ⎞ ⎜ ⎟V ⎝ m ⎠
(2)
where qt and Ct are the amount of Cr(VI) adsorbed per unit mass of adsorbent (mg g−1) and the Cr(VI) concentration (mg L−1) at time t (min), respectively; m is the adsorbent mass (g), and V is the volume (L) of the sample. Adsorption Isotherms and Thermodynamic Evaluations. To study the adsorption isotherm, 50 mL of Cr(VI) aqueous solution (the B
DOI: 10.1021/acs.inorgchem.6b00522 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
determine existing species of adsorbate in solution.39 The distribution of Cr(VI) species is dependent on both pH and total Cr(VI) concentration as shown in Figure S8 (SI).40 Since total concentration of Cr(VI) species was changed between 5 and 100 mg L−1 in all specimens here, HCrO4− and CrO42− are the most predominant species at the experimental pH range (2−9). Evaluation of pH effect on Cr(VI) adsorption by TMU30 revealed that the process is almost independent of the pH parameter. Unlike most typical adsorbents in which the maximum adsorption is reached at a particular pH toward special species,41 TMU-30 adsorbed all existing forms of Cr(VI) over a wide range of pH values (2−9) (Figure 3A).
Figure 1. (a) Pb−O chain formed by face-shared PbO9 polyhedra running along the b-axis in TMU-30; (b) the rhombic channels along the b-axis; free DMF molecules are located in the channels; (c) 3D Connolly surface representation of porous TMU-30 along the b direction. Color code: pink, Pb; red, O; blue, N; gray, C atoms. In the Connolly surface representation, gray and blue areas represent outside and inside channel surfaces, respectively.
Figure 3. (A) The effect of pH on adsorption of Cr(VI) by TMU-30. Initial Cr(VI) concentration, 30 mg L−1; amount of adsorbent, 10 mg; sample volume, 10 mL; T, 298 K. (B) PXRD patterns of TMU-30 in difference pH values: (a) before adsorption; (b, c, d, e) after adsorption at pH = 3, 6, 7, and 9, respectively.
Figure 2. (a) Schematic representation of 3D net, the C/N framework of (10,3)-d topology; (b) 3-connected nets that create 4-fold helices with opposite handedness; carbon, nitrogen, and lead atoms are represented by black, blue, and pink spheres. C1−N2 and C7−N1 connections are depicted by red and green sticks, respectively.
Also, the crystalline structure of TMU-30 did not change after the adsorptions were accomplished at various pH values up to 9 according to PXRD patterns (Figure 3B). The crystalline pattern of TMU-30 began to change over pH of 9 (Figure 3B,e). Therefore, subsequent tests were performed without any pH adjustment (pH = 5.6). Adsorption Kinetics. Determination of kinetic parameters is of great importance because they can be used to define the
N-oxide groups with the potential positive charge was applied for Cr(VI) oxo-anion adsorption. Effect of pH. The role of pH in an adsorption process is to influence charge density on the adsorbent surface and to C
DOI: 10.1021/acs.inorgchem.6b00522 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
was used to fit the kinetic results. The model can be expressed as eq 4,
adsorption process and efficiency. Fast kinetics is very important, acceptable, and useful in aqueous phase adsorption. The adsorption capacity of Cr(VI) onto TMU-30 increased with an increase in contact time and about 90% of final adsorption capacity was obtained in 5 min. After less than 10 min, 95% of Cr(VI) can be adsorbed from the aqueous solution, and the solution became colorless (Figure 4a,b). This time for Cr(VI) adsorption is much faster than traditional adsorbents such as activated carbon and biomass, which need several hours to reach better removal percentage.24−44 In order to define the adsorption rate, a pseudo-second-order model45 with the highest value of correlation coefficient (R2 = 0.999)
t 1 t = + qt qe kadqe 2
(4)
−1
where qt (mg g ) is the adsorption capacity of TMU-30 at time t (min), qe (mg g−1) is the adsorption capacity at equilibrium, kad (g mg−1 min−1) is the rate constant of adsorption. The values of qe and kad were calculated from the slope and intercept of t/qt vs t plot (Figure S9a in the SI). The obtained rate constant, kad, was 0.193 g mg−1 min−1 at 298 K, which is the highest one for the Cr(VI) adsorption so far.40,46−49 Adsorption Isotherm. The effect of initial Cr(VI) concentration on adsorption was evaluated to determine the correlation between the adsorbent and the adsorbate as well as maximum adsorption capacity of the adsorbent. The adsorption on TMU-30 increased rapidly with increase in initial Cr(VI) concentration until the plot reached a plateau indicating saturation of the adsorbent sites (Figure 4c). In order to derive an appropriate correlation between the experimental and theoretical data, the Langmuir model49 was applied (R2 = 0.999), which is represented by eq 5, Ce C 1 = e + qe qm kLqm
(5)
where qm (mg g−1) and kL (L mg−1) are the maximum adsorption capacity of the adsorbent and Langmuir constant, respectively. The values of qm and kL were calculated from the slope and intercept of the Ce/qe vs Ce plot (Figure S9b in the SI). The Langmuir isotherm is applicable to monolayer adsorption with all identical and energetically equivalent adsorption sites. The q m was obtained 2.86 mol Cr(VI) molTMU‑30−1 (145 mg g−1), which can be assumed one of the highest adsorbed amount on porous materials reported so far. Maximum adsorption capacity and time of equilibrium have been compared with other reported MOFs in Table 1. Table 1. Adsorption Capacities and Time of Equilibrium for Cr(VI) on Various MOFs MOF adsorbent
maximum capacity (mol mol−1)
ABT·2ClO4 3-D DyMOFs Zn0.5Co0.5SLUG-35 FIR-53 FIR-54 Ag-SLAG-21 TMU-30
1.2 0.85
maximum capacity (mg g−1)
time of equilibrium
ref
62.88
3h 6h
23 50
0.43
68.5
4h
22
0.71 0.5 0.4 2.86
74.2 103 60 145
1h 1h 48 h
