1. Introduction
Currently, continued economic growth, widespread urbanization and rapid population growth are leading to ever-increasing environmental pollution [
1]. Among the most acute problems, it should be noted the depletion of water resources due to their contamination by various pollutants, including inorganic and organic chemicals. Synthetic dyes are one of the main groups of water and wastewater pollutants, since they are widely used in textile, printing, paper, leather, plastic, rubber, carpet, food, pharmaceutical, and cosmetic industries [
2,
3]. These dyes create one of the most acute environmental problems due to their potential adverse effects on human health [
4,
5]. In addition, many organic dyes contain chemicals that are toxic or carcinogenic to mammals and other living organisms [
6,
7]. An important problem is also the non-biodegradable nature of some dyes and their resistance to light and oxidizing agents [
8,
9]. Methylene blue (MV), Congo red (CR) and methyl violet (MV) are some of the most common organic dyes present in wastewater and industrial effluents. Therefore, it is very important to remove dyes before discharging wastewater into water resources, in order to reduce environmental damage.
A variety of biological, chemical and catalytic methods are used to remove dyes from wastewater [
10,
11,
12,
13]. Among them, the adsorption of dyes on solids with a high specific surface area is widely used due to its high efficiency, reliability, low cost, versatility and ease of operation [
14,
15,
16]. The most common adsorbents are activated carbon [
17,
18,
19], polymers [
20], zeolites [
21] and biomaterials [
22,
23,
24]. Among their shortcomings, low adsorption capacity and separation difficulties should be noted. Therefore, an important task is the development of effective and cost-effective adsorbents for the removal of organic dyes from wastewater, which have a large capacity, high adsorption rate and easy separation.
In this regard, the attention of researchers was attracted to the coordination polymers or metal-organic frameworks (MOFs), which consist of metal-oxo clusters or metal ions and organic linkers, and are characterized by a diverse crystalline architecture, large specific surface areas and good porosities [
25,
26]. Their physico-chemical characteristics can be easily controlled by adjusting the structure and functionality, as well as using post-synthetic modification to improve the adsorption properties. Studies have shown the effectiveness of using MOF to remove pollutants, in particular, organic dyes, from wastewater [
27,
28,
29,
30]. A typical example is the adsorption of organic dyes on titanium-based MOFs, MIL-125, which improves significantly after NH
2-functionalization [
31]. In another interesting study, nickel-based MOF was used to remove MB as a cationic dye [
32]. It should also be noted that this study looks at the influence of the size of MOF mesopores with hierarchical pores on the adsorption capacity for various organic pollutants [
33]. Zirconium-based MOFs showed high structural resistance to water, which made it possible to use them for adsorption removal of anionic and cationic dyes from aqueous solutions [
34]. It is of interest to use the four cobalt (II) coordination polymers [
35] and siloxane-based MOF [
36] for the degradation removal of CR azo dye. Two copper (I) cyanide coordination polymers were used to decompose and remove MB [
37].
MOFs attract considerable attention as precursors of nanostructured materials, since they are not only starting materials, but also stabilizers of the resulting nanoparticles [
38,
39,
40,
41,
42]. One of the simplest and most rational methods for the synthesis of new nanostructured materials is solid-phase thermolysis of MOFs with various structures and compositions. Thermal decomposition of MOFs under various conditions allows one to obtain a variety of nanomaterials, for example, carbon, metals, metal oxides, etc., with the desired size and morphology.
Among the various MOFs, MOF, such as Cu
3(BTC)
2(H
2O)
3⋅xH
2O (where BTC = 1,3,5-benzenetricarboxylate) containing the binuclear Cu(II) paddlewheel structures are of particular interest for testing the adsorption capacity. This MOF was first synthesized in 1999 [
43] and designated as HKUST-1 (Hong Kong University of Science and Technology) or MOF-199. It has a high specific surface area and an interconnected 3D pore system with pore sizes of 9 Å × 9 Å [
44,
45]. Various Cu–BTC complexes were studied, including Cu
3(BTC)
2⋅3H
2O [
46,
47], Cu
2(OH)(BTC)(H
2O)
2·nH
2O [
48], and Cu(BTC–H
2)
2(H
2O)
2⋅3H
2O [
49]. The synthesis of Cu-BTC is mainly based on conventional hydrothermal processes, with long reaction times and relatively high temperatures, as well as a number of processes with a continuous flow [
50,
51,
52]. As an example, it is possible to obtain uniform hollow Cu–BTC microspheres under conditions of an interfacial reaction using a continuous droplet microreactor [
50]. A high-rate of synthesis of Cu–BTC was achieved based on uniform mixing of the reactant streams using a continuous impinging jet reactor under high pressure [
51]. Hollow nanocrystalline MOFs were obtained in a continuous process using a special nozzle for spraying precursor solutions to create atomized droplets [
52]. High-rate synthesis of Cu–BTC MOFs with a BET surface area of more than 1600 m
2 g
−1 and with a yield of 97% was carried out in a continuous-flow microreactor-assisted solvothermal system with a total reaction time of 5 min [
53,
54]. It should be noted the preparation of Cu–BTC MOF by green synthesis with water [
55]. Carboxylate-based MOFs are characterized by a regular structure and permanent porosity after thermal/vacuum activation, which allows them to encapsulate a variety of guest substrates into the bulk structure, which makes them promising solid materials for a number of applications, such as gas storage and separation.
Although considerable work has been done on the synthesis of HKUST-1, very few studies of HKUST-1 have indeed produced high-quality crystals with a large surface area. In particular, some studies reported very large surface areas, while others reported very small values using the solvothermal synthesis method. An attempt was made to improve the yield and surface area of HKUST-1 using the atmospheric pressure method [
56]. However, the typical octahedral crystalline structure of HKUST-1 was not obtained. Recently, HKUST-1 has also received significant attention in wastewater treatment due to its excellent properties [
57,
58]. However, the effect of water adsorption on the structure of MOFs was shown [
59]. For example, it was found that Cu
3(BTC)
2 is characterized by a low surface area, a change in crystallinity, and damage to the surface of the crystal. In addition, the stretching of the Cu
2C
4O
8 cage occurs as a result of the absorption of water molecules near Cu sites [
60]. To improve the adsorption properties of HKUST-1, nanoparticles of graphene, graphite, and carbon nanotubes were included in this MOF [
61,
62,
63,
64,
65]. Although a number of works have been devoted to the adsorption of dyes on HKUST-1, only a few reports on HKUST-1 produced large surface areas and pore volumes. Therefore, additional studies on the synthesis and characterization of HKUST-1 are needed.
In this work, several important synthesis pathways for MOFs were applied to determine how the synthesis method and conditions affect the structure and adsorption capacity of the resulting samples. Three different synthesis routes were used: Slow evaporation (A), solvothermal synthesis using a PEG-1500 modulator (B), and green synthesis in water (C). All MOFs were characterized by various analytical methods: X-ray powder diffraction, elemental analysis, thermogravimetric analysis, scanning electron microscopy and volumetric nitrogen adsorption/desorption. In addition, the obtained MOFs were used to remove aqueous MB, CR, and MV, as well as to obtain nanostructured carbon materials.
2. Materials and Methods
2.1. Starting Materials
Methanol (MeOH, 99.9%), ethanol (EtOH, 98%), ethyl acetate, methylene chloride and dimethylformamide (DMF) were supplied by Sigma-Aldrich (Moscow, Russia); copper(II) sulfate pentahydrate (CuSO4·5H2O, ≥99.0%), copper(II) acetate monohydrate (Cu(CH3COO)2·H2O, ≥99.0%), 1,3,5-benzenetricarboxylic acid (H3BTC, 95.0%), triethylamine and a PEG-1500 modulator were purchased from Sigma-Aldrich (Moscow, Russia) and used without further purification. The purity of the solvents was determined by selective gas chromatography.
2.2. Adsorbates
Dyes methylene blue (МВ), Congo red (CR) and methyl violet (MV) having the molecular formulas C16H18N3SCl, C32H22N6Na2O6S2, and C24H28N3Cl, respectively, were chosen as adsorbates. They were purchased from Sigma-Aldrich with a solubility in water of 50, 10 and 50 g L−1 (20 °C) and a molecular weight of 319.85, 696.66, and 407.99 g mol−1, respectively. These dyes were chosen in this study because of their known strong adsorption onto solids. A stock dye solution was prepared by dissolving a precisely weighed dye in distilled water to a concentration of 200 mg L−1. Experimental solutions were obtained by diluting the stock dye solution in exact proportions to the required initial concentrations.
2.3. Synthesis of MOF
2.3.1. Synthesis Using Slow Evaporation (A)
The following synthesis procedure is an optimization of the literary recipe [
66]. 1,3,5-Benzenetricarboxylic acid (1 g, 4.8 mmol) was dissolved in 30 mL of a mixture of DMF/EtOH/H
2O (1:1:1 ratio by volume) and added dropwise to Cu(CH
3CO
2)
2·H
2O (1.72 g, 8.6 mmol), which was dissolved in 30 mL of the same solvent mixture. Then, triethylamine (1.2 mL, 8.6 mmol) was added to completely deprotonate the linker and the resulting mixture was stirred for 30 min and slowly evaporated at room temperature for several weeks to obtain a blue precipitate. The precipitate was separated by centrifugation at 4500 rpm and washed several times with DMF. Then, the obtained sample was soaked in ethanol for several days with periodic replacement of the solvent and dried in vacuum (10
−3 Torr, 80 °C, 10 h).
2.3.2. Solvothermal Synthesis Using a PEG-1500 Modulator (B)
Copper trimesinate was also prepared using a modified solvothermal method based on the described procedure [
67]. The modulator (PEG-1500), dissolved in DMF at a concentration of 0.7 mol L
−1 under rapid stirring (900 rpm), was added to the same mixture of reagents and solvents that was used for procedure A. The resulting mixture was stirred for 30 min and placed in a stainless-steel autoclave with a Teflon liner. The autoclave was heated for 20 h at 120 °C, then was naturally cooled to room temperature. The resulting blue crystals were collected and purified as described above.
2.3.3. Synthesis in Water (C)
NaOH (1.2 g, 3 mmol) was dissolved in a bidistilled water (50 mL) and 1,3,5-benzenetricarboxylic acid (2.1 g, 1 mmol) was added with heating to 80 °C and at constant stirring until a clear solution was obtained. CuSO4·5H2O (3.74 g, 1.5 mmol) was dissolved in water (20 mL) and slowly poured into the first solution, without stopping mixing. The blue precipitate was left in the vessel until self-cooling, and then it was separated in a centrifuge at 8000 rpm for 10 min, washed in a centrifuge under the same conditions with a bidistilled water. The precipitate was transferred to a flask with a tight stopper, into which dry methanol (50 mL) was added, stirred for 2 h and kept at room temperature for 12 h. The last procedure was repeated twice, separating the precipitate by centrifugation. The crystals separated from methanol were poured into dehydrated ethyl acetate (50 mL), stirred for 2 h, incubated for 12 h, repeating the procedure twice and the precipitate was also separated by centrifugation. The crystals, aged in ethyl acetate, were poured into anhydrous, stirred for 2 h and left for 12 h. The procedure was repeated twice, each time separating the target product by centrifugation. The precipitate was dried in air at 120 °C for 5 h, and then in vacuum at a residual pressure of 10−6–10−8 Torr at 150 °C for 6 h. The obtained violet crystals were stored in vacuum or in an atmosphere of dry argon. The weight of the collected and dried crystals is 4.86 g (95.5% based on H3BTC).
2.4. Characterization
Elemental analysis was performed on a Vario Micro cube analyzer (Elementar GmbH, Langenselbold, Germany), and Сu was determined on an AAS-3 atomic absorption spectrometer (Zeiss, Jena, Germany). X-ray powder diffraction (XRD) analysis was performed using an ARL™ X’TRA powder diffractometer (Thermo Fisher Scientific, Waltham, MA, USA) with CuKα radiation (λCu = 1.54184 Å) in the range of 2θ = 5–80° with a scanning rate of 5°/min and temperature of 25 °C. Thermogravimetric analysis (TGA), performed simultaneously with differential scanning calorimetry (DSC) was carried out on a STA 409CLuxx synchronous thermal analyzer coupled to a QMS 403CAeolos quadrupole mass spectrometer (NETZSCH, Selb, Germany) in air at a rate of 10°/min in the range of 20–500 °C (powders, m = 0.3–0.4 g). Fourier-transform infrared (FTIR) spectra were performed on a Perkin-Elmer Spectrum 100 FTIR spectrometer using KBr pellets and Softspectra data analysis software (Shelton, CT, USA). Scanning electron microscope (SEM) analysis was conducted on a ZEISS Crossbeam 340 with an accelerating voltage of 3 kV (Jena, Germany). The detection of secondary electrons was carried out using an Everhart-Thornley detector (SE2), increasing the samples from 1.92 to 50 thousand times. The structure of thermolysis products was studied using a Tecnai G2 Spirit BioTWIN FEI high-resolution transmission microscope (Eindhoven, the Netherlands). The samples for high-resolution transmission electron microscopy (HRTEM) were prepared as follows: A powder suspension in hexane was applied onto a carbon-coated copper grid and the solvent was dried in air.
2.5. Determination of Kinetic and Thermodynamic Parameters of Thermal Decomposition
The activation energy was calculated using the Freeman-Carroll equation [
68],
where
w is the weight loss of the substance over time t,
Wf is the weight loss at the end of the process,
Wr is calculated as
Wr =
Wf −
w,
Z is the frequency factor,
Ea is the activation energy.
The entropy of activation (∆
S), enthalpy of activation (∆
H) and free energy of activation (∆
G) were calculated using the following equations:
2.6. Measurement of Sorption Equilibrium Properties
The nitrogen adsorption/desorption isotherms were obtained at 77 K (liquid N2) using the AUTOSORB-1 system (Quantachrome, Boynton Beach, FL, USA) by the static volumetric method; before analysis the samples were degassed by heating at 150 °C for 12 h in vacuum. The Brunauer-Emmett-Teller surface area (SBET) was obtained from the amount of N2 physically sorbed at various relative pressures (P/P0), based on the linear part of the 6-point adsorption data at P/P0 = 0.02–0.10. The total pore volume (Vtotal) was calculated by the Horvath Kawazoe method at P/P0 = 0.99. The micropore volume (Vmicro) was obtained by the Barrett-Joyner-Halenda adsorption and the t-plot methods, respectively. For assess the adsorption capacity, ultrahigh purity gases (99.995%) were used. The equilibrium values of the adsorbed gas volume were calculated as a function of the relative pressure in the system taking into account the weight of the sample and the volumes of various parts of the instrument.
2.7. Dye Equilibrium Adsorption Experiments
For equilibrium studies, a batch method was used because of its simplicity. A solution of the corresponding dye with a volume of 200 mL was placed in a 300 mL temperature-controlled beaker, thermostatically controlled at 283, 293, and 308 K on a magnetic stirrer, adjusting the rotation speed so that mixing was effective, but the air was not drawn into the liquid phase. When the solution reached a pre-determined temperature, an adsorbent (copper trimesinate synthesized by a method C, 0.1 g) was introduced and a timer was started. Every 15 min, 10 mL of a sorbent suspension in a dye solution was pipetted and centrifuged quickly; the concentration of residual dye was determined in the filtrate using a UV-visible spectrophotometer (Varian, Cary50) at λ
max = 492 nm (CR) [
69], 664 nm (MB) [
70], 584 nm (MV) [
71], respectively.
The amount of the dye uptake by MOF in each flask was calculated by the following equations,
where
qt and
qe are the quantities (mg g
−1) of the dye adsorbed on the adsorbent at time
t and equilibrium, respectively;
C0,
Ct, and
Ce are the dye concentrations in solution (mg L
−1) initially, at time
t and at equilibrium, respectively; m (g) and
V (L) represent the amount of adsorbent and the volume of the dye solution, respectively.
The percentage removal of the dye (%) was calculated by the following equation:
2.8. Studying the Kinetics of Adsorption
An adsorption model that describes the adsorption of a solute on a solid surface can be expressed as [
72],
where
k1 (min
−1) is the rate constant of the pseudo-first order model.
After a definite integration from
t = 0 to
t =
t and from
q = 0 to
q =
qe, Equation (8) takes the form:
The constant k1 can be determined experimentally from the slope of the linear plots ln(qe − qt) vs. t.
2.9. pH Experiments
To study the effect of pH on dye adsorption, a solution (80 mg L−1) of copper trimesinate was added to solutions (10 mg L−1) of dye. The initial pH was adjusted to 2–12 using HCl and NaOH. After shaking the suspensions for 120 min for an equilibrium time at a temperature of 300 K, they were filtered through 0.2 μm membrane filters and analyzed for the concentration of residual dye.
2.10. Study of Adsorption Isotherms
Two well-known isotherm equations, Langmuir and Freundlich, were used to interpret the obtained adsorption data.
2.10.1. Langmuir Isotherm
The Langmuir adsorption isotherm suggests that adsorption occurs in certain homogeneous sites within the adsorbent, and has found successful application in many sorption processes of monolayer adsorption [
73]. The Langmuir isotherm can be written in the form [
74],
where
KL is the Langmuir constant (L mg
−1) associated with the affinity of the binding sites and the free energy of sorption;
qm is the adsorption capacity expressing the dye concentration when a monolayer is formed on the sorbent (mg g
−1).
For the Langmuir equation, the favorable nature of adsorption can be expressed in terms of the dimensionless separation coefficient of the equilibrium parameter, which is determined as follows:
The
RL values indicate that the type of isotherm is irreversible (
RL = 0), favorable (0 <
RL < 1), linear (
RL = 1) or unfavorable (
RL > 1) [
75].
2.10.2. Freundlich Isotherm
The Freundlich isotherm is an empirical equation used to describe heterogeneous systems [
73]. The Freundlich equation is as follows,
where
KF is a constant showing the adsorption capacity of the adsorbent (mg
1−1/n L
1/n g
−1), and
n is an empirical constant related to the magnitude of the adsorption driving force.
The value of 1/
n quantitatively determines the favorable adsorption and the degree of heterogeneity of the surface of the MOF. According to Halsey [
76]:
To determine the maximum adsorption capacity (qm), it is necessary to work with a constant initial concentration C0 and a variable adsorbent weights.
2.11. Determination of Thermodynamic Parameters of Dye Adsorption
Assuming that the activity coefficients are equal to unity at low concentrations (the meaning of Henry’s law), thermodynamic parameters (∆
G0, ∆
H0 ∆
S0) were calculated using the following relations [
77],
where
KD is the adsorbate distribution coefficient.
The parameters ∆H0 and ∆S0 can be calculated from the slope and intersection of the plot ln KD vs. 1/T, respectively.
2.12. Thermolysis Technique
In a typical experiment, a portion of the substance (0.6–0.8 g) is placed in a quartz tube with a minimum height of 10 cm and a diameter of 1 cm. The assembled unit is evacuated, filled with argon, heated at a rate of 50/min until a temperature of 400 °C is reached and kept at this temperature for 1 h. After the specified time, the unit is evacuated again, continuing to maintain the indicated temperature. Vacuuming is carried out until the walls of the external vessel are freed from volatile decomposition products. The resulting substance is cooled in vacuum to room temperature; the target product is removed in the form of a porous column with a height of 20–25 mm and crushed.