The Use of Copper Terephthalate for the Determination and Separation of Organic Dyes via Solid-Phase Extraction with Spectrophotometric Detection

Abstract

In this work, a metal–organic framework (MOF), copper benzene dicarboxylate (Cu-MOF), was tested for the adsorptive recovery of organic dyes (Sunset Yellow FCF, Tartrazine, Orange II, and Methyl Orange) from aqueous solutions. Studies were also carried out to determine the effects of various parameters, and isothermal and kinetic models were proposed. The adsorption capacity of Cu-MOF was much higher than that of activated carbon. The experimental data are best described by the Langmuir isotherm model (R² > 0.997) and show the ability of Cu-MOF to adsorb 435 mg/g of the dye under optimal conditions. The study of the kinetics of the dye adsorption process followed a pseudo-second-order kinetic model indicating the coexistence of physical and chemisorption, with diffusion within the particles being the rate-limiting step. Thermodynamic studies were also carried out, and they led to the conclusion that the adsorption of the dye was a feasible, spontaneous, and exothermic process (−25.53 kJ mol⁻¹). The high organic dye recovery shows that Cu-MOF can be used as an efficient and reusable adsorbent for the extraction of dyes from aqueous solutions. These studies may lead to economic interest in this adsorbent material for environmental purposes.

1. Introduction

Environmental pollution is increasing all over the world due to population growth and industrial development. Dyes are considered one of the leading pollutants, and they have every chance of affecting the surrounding environment in several ways [1]. They have a profound and lasting effect on exposed animals, depending on the concentration and duration of exposure, stimulating the evolution of bacteria that interfere with photosynthesis, absorbing and reflecting light from incoming water, and preventing the growth of aquatic plants [2]. In addition, dyes are used in the food industry to improve the appearance, smell, taste, color, texture, calorie content and shelf life of food. Most dyes are considered mutagenic and carcinogenic due to the presence of aromatic rings in their structure. The presence in food and the appearance of these dyes in surface and groundwater are considered unsafe for human well-being. High doses can cause a variety of health problems, including allergic reactions, skin irritation, gastrointestinal problems, and overactivity in children. Consequently, the presence of dyes in foods is highly regulated, and scientists are developing simple and inexpensive methods for detecting and extracting dyes to minimize their contamination of wastewater.

The problem of detecting and identifying the content of organic dyes in food products is relevant, and, therefore, various methods are used for this purpose, including UV spectrophotometry, high-performance liquid chromatography, spectrofluorometry, electrochemical voltammetry, and mass spectrometry [3,4,5,6,7,8,9,10,11,12,13,14]. However, when determining the concentration of the dye, various limitations and obstacles are observed, such as the low concentration of the dye in food samples, which cannot be detected using instruments, and the positive or negative interference effect of the matrix components. Because of this, the concentrations of dyes in foods cannot be accurately determined. Therefore, separation and pre-concentration methods are needed.

There are several methods most suitable for the extraction, separation, and concentration of dyes. One of them is the solid-phase extraction (SPE) method. This method is common and is used to isolate and concentrate synthetic food colorings. Metal–organic frameworks (MOFs), which are used as sorbents in the SPE method, are new porous compounds, which are also called hybrid porous coordination polymers [15]. MOFs are types of adsorbents that exhibit unique physical and chemical properties by combining metal-containing units with organic linkers, using strong bonds to create open crystalline frameworks with a constant porosity [16,17,18,19,20,21,22,23]. A consistent porosity, low density, high surface area, suitable pore size, metal node functionality, and structural flexibility make MOFs unique and well suited for a wide range of applications. In addition, the high surface area and porosity of MOFs can facilitate the diffusion of contaminants through the framework and the presence of an adsorption site. Thus, MOFs are among the most suitable structures in the field of adsorption and separation [24,25,26,27].

The terephthalate ion is used as a ligand in MOF chemistry because functional groups in the para-position contribute to the formation of polymer chains that form both monometallic and heterometallic two- or three-dimensional coordination compounds. Based on the possibilities of the structure of the terephthalate ion, it can be assumed that it can function as a bi-, tri-, or tetra-dentate ligand [28]. For example, copper terephthalate exhibits a high adsorption capacity with respect to various dyes and metal ions [29,30,31,32].

An important task today is the study and development of simple and efficient methods for the synthesis of these compounds. The aim of this study was to develop a simple sorption–spectrophotometric method for the determination of food additives using MOF based on copper terephthalate as an adsorbent. To achieve this goal, the adsorption on the sorbent was determined (the dependence of adsorption on time, concentration, and pH); the mechanism of adsorption was studied; and food additives were determined separately and together in real objects (the drink “Orange Fresh”), as well as their analytical characteristics.

2. Materials and Methods

2.1. Initial Materials

A sorbent was prepared with benzene dicarboxylic acid (H₂BDC, chemically pure), hydrochloric acid (chemically pure), sodium hydroxide (analytically pure), n-octanol, and copper(II) sulfate purchased from Sigma-Aldrich.

2.2. Dyes

Organic dyes were purchased from Sigma-Aldrich. Figure 1 shows the chemical structures of the dye molecules.

Stock solutions of dyes (1000 mg/L) were prepared by dissolving the required amount of dye in bidistilled water, and a test solution was prepared daily with the required dilution.

2.3. Characterization

Elemental analyses were carried out on a CHNOS vario EL cubic analyzer (Elementar Analysensysteme GmbH, Langenselbold, Germany), and their results are in good agreement (±0.3%) with the calculated values. Copper was determined on an X-Art M energy-dispersive X-ray fluorescence spectrometer (Comita, Russia) or an MGA-915 atomic absorption spectrometer (Lumex, Russia). IR spectra were recorded on a Nicolet 380 FTIR spectrometer using KBr beads and Softspectra data analysis software. An X-ray diffraction analysis (XRD) was performed on a DRON-UM-2 diffractometer with CuKa radiation (λ_Cu = 1.54184 Å) in the range 2θ = 5–80° of 2θ angles at a scanning rate of 5°/min and a temperature of 25 °C. Scanning electron microscopy (SEM) images were obtained on a Zeiss LEO SUPRA 25 scanning electron microscope.

2.4. Synthesis of MOF

The synthesis of copper terephthalate took place in a medium containing a surfactant (n-octanol) to give it a bulk structure. A total of 2.4 g (0.06 mol) of sodium hydroxide was dissolved in 150 mL of hot water, and 5 g (0.03 mol) of terephthalic acid was added, achieving the complete dissolution of the latter. In another vessel, a solution was prepared from 120 mL of hot water and 15 g of CuSO₄·5H₂O. Both solutions were filtered using the hot filtration method. Then, 15 mL of n-octanol was added to a solution of sodium terephthalate with constant stirring, and at the same time, the number of revolutions of the magnetic stirrer was increased to form an emulsion. Next, a hot solution of CuSO₄·5H₂O was added dropwise, while the solution was stirred. As a result of this reaction, a voluminous precipitate of a bright blue color formed. The crystals were filtered on a porous glass plate and dried in air and then at 60 °C for 6 h. This method allowed us to obtain 6.25 g of an almost colorless substance with a slight bluish tint, which corresponds to a yield of 91%, calculated as terephthalic acid. Elemental analysis: Found (%) C—28.9; H—2.5; Cu—40.2. Calculated for Cu(BDC)·2H₂O (%): C—29.26; H—2.4; Cu—39.

2.5. Dye Adsorption Study

For the analysis of all the dyes, three series of 5 solutions of 40 mL, each with different concentrations of the additive, were prepared: 20; 10; 5; 2.5; and 1.25 mg/L. Next, the initial solutions were placed in 100 mL beakers, and adsorption was studied at temperatures of 5, 20, and 35 °C. To do this, after reaching the required temperature, a weighed portion of the sorbent (20 mg) was placed in each of the analyzed solutions, and the solutions were placed on a magnetic stirrer, adjusting the rotation speed for good mixing. After 3 h of active stirring, 10 mL of the resulting suspension was taken from each solution and centrifuged for 10 min. The equilibrium concentration of the additive was studied using a spectrophotometer at certain wavelengths. The wavelengths were determined individually for each dye.

The adsorption capacity (q_e) of Cu-MOF was calculated using Equation (1):

$$q_e=\frac{\left(C_0-C_e\right)·V}{m}$$

(1)

where C₀ is the initial concentration of the dye (mg/L), C_e is the equilibrium concentration of the dye (mg/L), V is the volume of the dye solution, and m is the weight of the sorbent.

Recovery factors (R, %) were calculated using Formula (2):

$$\mathrm{R}, \%=\frac{C_0-C}{C_0}$$

(2)

where C is the concentration of the dye solution after sorption (mg/L).

2.6. Study of Dye Adsorption on a Real Object

To carry out adsorption on real objects, it is necessary to detect the studied dyes in the analyzed objects. For the analysis of the content of synthetic dyes, the method of thin-layer chromatography (TLC) was used. STX-1VE Sorbfil silica gel plates were used as the stationary phase in the experiment, and the mobile phase was a mixture of n-butanol–ethanol–water in a ratio of 60:60:150. The carbonated drink “Orange Fresh” was used as the analyte.

To determine the presence of synthetic dyes in a beverage, they were separated from the matrix solution into an organic layer, which was toluene, using an extraction method. To do this, 40 mL of toluene solution and 5 mL of the test drink were mixed in a separating funnel with a capacity of 100 mL. Standard dye solutions and a solution of the analyzed object were applied with a microtool, 1 μL each, onto an STX-1VE silica gel plate. Chromatographic separation occurs due to the transfer of the components of the mobile phase along the layer of the stationary phase at different speeds in accordance with the distribution coefficients of the separated components and zones of different colors.

To separate the dyes and determine their concentrations in the drink, a column 1.7 cm in diameter and 15 cm in height was used. A sealant layer was placed in the column, and then the column was filled with a sorbent (Cu-MOF), densely filling the sorbent layer until the layer height was 7 cm. Then, 20 mL of the analyzed object (orange juice drink) was poured into the column. After the complete adsorption of the test solution, the eluent (n-butanol–ethanol–water in a ratio of 60:60:150) was passed through the sorbent layer, and fractional samples of 10 mL were taken. Dye concentrations were determined using the above method.

3. Results and Discussion

3.1. Synthesis and Characterization of the Sorbent

The preparation of Cu-MOF was carried out by mixing a metal salt and an acid in water in the presence of a surfactant to impart a bulk structure to MOF (Figure 2). Cu-MOF was analyzed using XRD, SEM, and IR spectroscopy.

X-ray diffraction was performed to identify the obtained compounds. The observed sharp and distinct diffraction peaks indicate that the prepared samples of copper terephthalates have good crystallinity and phase purity. All positions of the diffraction peaks are in good agreement with the results of the simulation performed in the “MatH-3”program used for processing XRD data. Figure 3 shows the diffraction pattern of the synthesized Cu-MOF; the lower part of the figure shows lines corresponding to the calculated peaks, which satisfactorily coincide with the values on the experimental curve and are in good agreement with previously published data [29,30,31,32].

An SEM image is shown in Figure 4, which clearly shows prismatic crystals with a length of 30 to 100 µm and a thickness of 10 µm.

The IR spectrum of Cu-MOF (Figure 5) contains absorption bands corresponding to the crystallization water, carboxylate ions, C=C bonds of the aromatic ring, and Me-O bonds. In the region of 1720 cm⁻¹, there is no absorption band characteristic of the carboxyl group. The difference between the asymmetric and symmetric vibrations of the carboxyl groups indicates that carboxyl groups can function as both monodentate and bidentate ligands.

3.2. Solid-Phase Extraction of Dyes

In the course of this work, studies were carried out that made it possible to evaluate the ability of copper terephthalate to adsorb dyes in aqueous solutions. A study was made of the effect of the initial concentration, pH, contact time on the adsorption of dyes, and the kinetics of the process.

With an increase in the initial concentration, the degree of adsorption increased significantly, and with an increase in temperature, it decreased (Figure 6).

Figure 7 shows photographs of the dye solutions before and after adsorption.

The initial pH of the dye solution is a key factor controlling the adsorption process. The pH was studied in the range from 3.0 to 11.0, as shown in Figure 8. The degree of the extraction of dyes increases with increasing pH from 3.0 to 7.0. However, when the pH was between 7.0 and 11.0, the recovery decreased. It can be concluded that the surface of copper terephthalate can acquire a positive charge, which can interact with anionic dyes via various binding mechanisms and pathways, such as electrostatic attraction or hydrogen bonding. Dyes are best extracted in an environment close to neutral. However, in an acidic environment, extraction processes are better than in an alkaline one.

The results obtained in the study of the adsorption of various azo dyes on copper terephthalate were analyzed using various adsorption models, particularly the Langmuir and Freundlich models.

The scheme of the Langmuir isotherm is based on the hypothesis of monolayer adsorption, where all sorption centers are dynamically and identically equivalent [33]. The linearized Langmuir isotherm can be expressed using Equation (3):

$$\frac{C_e}{q_e}=\frac{1}{q_{max}{K}_L}+\frac{C_e}{q_{max}}$$

(3)

where K_L is the Langmuir adsorption constant (mg/L), and q_max is the maximum adsorption capacity of the monolayer (mg/g).

The Freundlich isotherm shows adsorption on a heterogeneous surface with the same energy and in the presence of interactions between adsorbed molecules. This means that it is not limited to building a monolayer. The linearized Freundlich isotherm can be expressed using Equation (4):

$$\log q_e=\frac{1}{n}\log C_e+\log K_F$$

(4)

The parameters of the Langmuir and Freundlich adsorption isotherms are summarized in

Table 1. Calculated parameters of the Langmuir and Freundlich isotherms for the adsorption of dyes by copper terephthalate.

Dye	T, °C	Langmuir Model			Freundlich Model
		qmax	KL	R2	KF	1/n	R2
SY	5	3.73	0.67	0.739	1.22	0.554	0.922
	20	3.27	0.53	0.742	1.82	0.675	0.907
	35	2.25	0.63	0.756	2.12	0.649	0.828
TAR	5	1.60	4.81	0.837	1.82	0.384	0.879
	20	2.25	2.13	0.757	1.22	0.466	0.892
	35	3.28	1.18	0.735	1.82	0.488	0.878
Orange II	5	5.15	0.97	0.911	2.01	0.554	0.884
	20	3.73	0.64	0.913	1.16	0.675	0.878
	35	4.69	0.41	0.871	1.22	0.649	0.909
Methyl Orange	5	4.35	0.82	0.868	1.65	0.601	0.883
	20	2.75	1.01	0.835	1.22	0.577	0.878
	35	2.91	0.82	0.905	1.22	0.625	0.907

.

High K_F and 1/n values were obtained. This indicates that the adsorption of dyes by copper terephthalate promotes positive cooperative binding and its inherent heterogeneity [34]. The correlation coefficient (R²) shows that the adsorption of the Sunset and Tartrazine dyes is better described by the Freundlich isotherm than by the Langmuir model, which indicates that adsorption occurs on a heterogeneous surface with the same energy, interactions between adsorbed molecules are possible, and adsorption is not limited to building a monolayer. However, it can be concluded that the adsorption of the Orange II dye is better described by the Langmuir model, which indicates that all sorption sites are energetically equivalent and identical and that adsorption is based on monolayer adsorption. It was also concluded that both models are applicable for Methyl Orange.

To determine the spontaneity of the adsorption process and verify the experimental data, three main thermodynamic parameters were used: the change in enthalpy (ΔH⁰), entropy (ΔS⁰), and Gibbs free energy (ΔG⁰).

Adsorption as a spontaneous process of the accumulation of dye molecules on the surface of a sorbent is accompanied by a decrease in the entropy of the system. Since the criterion for the spontaneity of the process is

∆H − T·∆S = ∆G < 0,

then adsorption is possible only at ∆H < 0 (exothermic process). Equilibrium is determined by the condition ∆H = T·∆S. With an increase in temperature, the equilibrium shifts towards the endothermic process, i.e., desorption.

The Gibbs equation was used to study the effect of temperature on equilibrium adsorption (5):

$$\Delta G^0=-RTlnK$$

(5)

where K is the thermodynamic equilibrium constant, which can be determined using Equation (6):

$$K=\frac{C_t}{C_e}$$

(6)

Considering the change in concentrations in the real-time interval and depending on the dose of the sorbent, Equation (6) can be written as

$$\mathrm{K}=\frac{\left(C_0-C_e\right)V}{m{C}_e}$$

(7)

The Gibbs equation can also be expressed as follows:

$$\Delta G^0=\Delta H^0-\mathrm{T}\text{∆}{S}^0$$

(8)

Combining the above equations, we obtain

$$lnK=\frac{\Delta S^0}{R}-\frac{\Delta H^0}{RT}$$

(9)

The obtained thermodynamic parameters are presented in

Table 2. Calculated thermodynamic parameters.

Dye	T, K	lnK	ΔG, kJ/mol	ΔH, kJ/mol	ΔS, kJ/mol K
SY	278	2.55	−5.89	−0.700	2.02
	293	2.28	−5.56
	308	2.11	−5.40
TAR	278	3.21	−7.41	−0.753	2.80
	293	2.95	−7.18
	308	2.89	−7.40
Orange II	278	4.01	−9.26	−0.674	3.30
	293	3.75	−9.13
	308	3.36	−8.60
Methyl Orange	273	2.55	−5.89	−0.809	2.405
	293	2.47	−6.02
	308	2.43	−6.21

.

The data presented in Table 2 indicate that the adsorption process occurs spontaneously. The process is exothermic because the enthalpy change is negative. A positive entropy change value indicates that the number of degrees of freedom at the solid–liquid interface increases during dye adsorption and reflects the sorbate affinity for sorbent molecules. The values of the Gibbs free energy change also increase. This suggests that an increase in temperature indicates a higher adsorption efficiency at moderately low temperatures.

To achieve equilibrium, it is necessary to provide an appropriate contact time between the dye and the sorbent, as this is one of the main factors in the SPE process based on equilibrium. The results are shown in Figure 9. It can be seen that the degree of adsorption is higher at a lower temperature. However, all three temperatures can be considered satisfactory since the degree of adsorption for each of them exceeds 90%. The dye recovery rates during adsorption turned out to be quantitative: SY ≥ 80%; TAR ≥ 90%; Orange II ≥ 93%; Methyl Orange ≥ 85%.

The mechanism of adsorption, particularly the chemical reaction and mass transfer, was analyzed using the pseudo-first-order rate equation and the pseudo-second-order rate equation. The linear form of the pseudo-first-order kinetic model can be expressed using the following equation:

$$\log \left(q_e-q_t\right)=\log q_e-(\frac{k_1}{2303})t$$

(10)

where q_e (mg/g) is the amount of dye adsorbed at equilibrium; q_t (mg/g) is the amount of dye adsorbed at any given time (t); and k₁ (1/min) is the pseudo-first-order absorption rate constant, which can be calculated from the plot of the logarithm (q_e − q_t) versus time.

The pseudo-second-order kinetic model can be expressed in linear form using Equation (11):

$$\frac{t}{q_t}=\frac{1}{k_2{q}_e^2}+\frac{t}{q_e}$$

(11)

where k₂ (g/(mg/min)) is the pseudo-second-order sorption rate constant, which can be calculated from a linear plot of t/q_t versus time.

The kinetic parameters under different conditions are given in

Table 3. Estimated kinetic parameters.

Dye	T, °C	qe	Pseudokinetics of the First Order		Pseudokinetics of the Second Order
			k1	R2	k2 × 103	R2
SY	5	34.6	0.554	0.965	0.696	0.886
	20	33.3	0.600	0.946	0.622	0.873
	35	32.2	0.624	0.935	0.603	0.800
TAR	5	37.0	0.965	0.987	2.435	0.983
	20	36.2	0.700	0.992	0.848	0.972
	35	36.1	0.554	0.989	0.698	0.978
Orange II	5	38.6	0.624	0.865	0.584	0.917
	20	38.2	0.600	0.883	0.548	0.890
	35	37.5	0.509	0.879	0.508	0.767
Methyl Orange	5	34.6	0.674	0.994	1.044	0.968
	20	34.3	0.445	0.889	0.607	0.855
	35	34.0	0.383	0.902	0.558	0.849

. The applied pseudo-first-order equation satisfactorily describes the patterns of dye adsorption at the initial stages of the adsorption process when the film diffusion phenomenon has a significant effect on the process. However, the high values of the correlation coefficients make it possible to judge in favor of the applicability of both the pseudo-first- and pseudo-second-order models for describing the chemical stage of the sorption process, as well as for the possibility of considering intermolecular interactions in the analyzed systems.

The adsorption of organic dyes by MOFs can occur for various reasons:

• The adsorption of the dye may be due to the electrostatic interaction between MOF and the adsorbate. The neutrality of the studied sorbent means the absence of any active functional groups but does not exclude the occurrence of an electrostatic charge on the surfaces of the sorbent and sorbate. For example, well-vacuum-dried copper terephthalate, placed in a glass storage vessel, easily covers the walls of this vessel, spreading in a relatively thin layer, which we repeatedly observed.
• A sufficiently high oxygen content in the MOF composition allows for the formation of hydrogen bonds with dye molecules.
• The formation of π-π stacking between the benzene rings of the sorbent and sorbate is possible.
  Cation and anion π interactions are possible.
• Similar examples of interactions between sorbents of similar classes and organic dyes are quite often discussed in the works [35,36].

The sorbent can be used repeatedly. For this, a study was made of the number of working cycles (Figure 10). The results show that the extraction of the dyes from the aqueous solutions was quantitative over four cycles. The average values of extraction with the standard deviations of the dyes for four cycles were as follows: SY—79.3 ± 3.1%; TAR—89.1 ± 2.3%; Orange II—92.1 ± 3.2%; Methyl Orange—85.4 ± 1.0%. These results suggest that the sorbent was mechanically stable and had excellent recyclability.

3.3. Analysis of a Real Object

A chromatographic analysis confirmed the presence of dyes in the carbonated drinks analyzed in this study. The distribution coefficients R_f calculated as the ratio of the distance l traveled by the dye in a thin sorbent layer to the distance L = 9.3 cm traveled by the solvent from the start to the start line are shown in

Table 4. R_f values for synthetic dyes.

Object Under Study	Synthetic Dye	Rf Values
Drink “Orange Fresh” (1)	SY	0.65
	TAR	0.30
Standard dye solution (2)	TAR	0.39
Standard dye solution (3)	SY	0.87

.

The analyzed object was passed through the column, and the following results were obtained: when the real object was passed through the sorbent, a transparent solution was obtained at the output—the sorbent absorbed both dyes (Figure 11). No dye was detected in this sample, which means that objects such as sugar were present in the clear solution. When the eluent was passed through the sorbent, the dyes were successively washed out. First, the yellow-green dye, TAR, was washed off, followed by the orange dye, SY. When eluting the Orange Fresh drink applied to a chromatographic column, it was possible to clearly separate the dyes. During further elution, the individual dyes were washed out with the eluent in the following sequence: 1—TAR and 2—SY. As a result of the elution, 5 mL of TAR dye and 7 mL of SY dye were collected. The concentration of each dye in a separate sample was also determined, and it was determined that the content of dyes in the original sample was as follows: TAR 10.75 mg/L and SY 12.34 mg/L.

4. Conclusions

The resulting MOF based on Cu-containing terephthalic acid was characterized using XRD, SEM, and IR spectroscopy. In Cu-MOF, the existence of prismatic crystals 30 to 100 µm long and 10 µm thick was found. Copper terephthalate was used for the adsorptive removal of azo dyes from aqueous solutions (SY, TAR, Orange II, and Methyl Orange), as well as for the separation of food additives in real systems. The kinetic data are consistent with both first- and second-order pseudokinetic models. The Langmuir and Freundlich isotherm models were used to describe the adsorption isotherms. According to the correlation coefficients (R²), the Freundlich model fits the data better than the Langmuir model for SY and TAR dyes. The adsorption of the Orange II dye is better described by the Langmuir model. For Methyl Orange, both models are acceptable. Copper terephthalate showed a satisfactory ability to adsorb azo dyes. The values of thermodynamic parameters (ΔG⁰, ΔH⁰, and ΔS⁰) show that the dye adsorption process is spontaneous and exothermic; in the process of sorption, the number of degrees of freedom increases, which characterizes the good affinity of the sorbate with the sorbent. When studying real objects, it was determined that the sorbent under study makes it possible to separate food additives that can be used to detect these dyes and determine their concentrations.