Simultaneous Preconcentration of Fast Green FCF and Rhodamine B Using Deep Eutectic Solvent and Determination via High-Performance Thin Layer Chromatography

,


Introduction
Synthetic dyes have drawn considerable attention in different industries including food, paper, medicine, textile, and cosmetic [1] because of low cost and favorable appearance.In general, synthetic pigments infuence psychological efects, which lead to loss of control, crying, increased sleeplessness, and other diseases [2].Of these approaches, the consumption of food dyes in diferent applications should be limited.Rhodamine B (Rh B) as one of the synthetic dyes is known as a harmful compound for human health and used in various industries such as food, leather, textile, paper, printing, and plastic [3].Due to the carcinogenic efects of Rh B on the eyes, skin, and liver, it is banned for use in the food industry [4,5].Fast green FCF (FG) as another synthetic dye is extensively used in the coating of candy, beverages, ice cream, health, and skin care products [6].Te toxicity and allergenic efects of FG had been published [7,8].It may also cause irritation of the eyes and skin [8].
Direct quantitation of FG and Rh B without specifc pretreatment is problematic because of the matrix efects of other components and their low amounts in real samples.Terefore, preconcentration methods have been applied when it is difcult to detect trace amounts of analytes.Te miniaturized form of liquid-liquid extraction entitled as liquid-phase microextraction (LPME) uses microliter volume to provide high extraction efciency.LPME has some advantages such as economic extraction, rapid phase transportation, high capacity of extraction, and the facility of direct injection of samples into analytical instruments [9].Numerous extraction methods including magnetic solid phase extraction [10,11], solid phase microextraction [12,13], hollow fber liquid phase microextraction [14,15], single drop microextraction [16], dispersive liquid-liquid microextraction [17], ultrasound-assisted LPME [18], and liquid-liquid microextraction [19,20] had been reported for the extraction and determination of synthetic dyes.In recent years, there has been an increasing focus on the utilization of deep eutectic solvents (DESs) as a novel green solvent.Tese solvents were frst introduced by Abbott et al. in 2004 [21].DESs are made via the combination of a hydrogen bond acceptor (HBA) and a hydrogen bond donor (HBD) due to the formation of hydrogen bonds after heating at a temperature of 60-90 C for 30 min to one hour [21].Te melting point of DES is lower than that of its constitutive component [22].Some properties of DES including the degree of afnity for target components, less solubility, and good dispersion in aqueous media infuence the extraction methodology [23].DESs have been extensively used for extraction, separation, and preconcentration processes [24].
Numerous compounds have been proposed for the synthesis of DES.For instance, choline chloride or vitamin B 4 is introduced as the most used HBA, whereas amino acids, carboxylic acids, urea, ethylene glycol, glycerol, and sugars have been used as HBD [24].DESs show the characteristic of high purity and eco-friendly [25], which have been used for the extraction and microextraction of synthetic dyes: simultaneous identifcation of eight synthetic dyes [26], determination of some red dyes [27], and determination of sunset yellow dye [28].
Various techniques have been proposed for the analysis of dyes such as spectroscopy [19,29], high-performance liquid chromatography [30,31], high-performance liquid chromatography-mass spectrometry [32], scanometry [33,34], and thin layer chromatography (TLC) [35].Among the diferent analytical techniques, high-performance thin layer chromatography (HPTLC) is known as a low cost, simple, and rapid technique that is appropriate for the comparison and separation of diferent components [36].HPTLC has numerous advantages over other methods: simultaneous study of diferent compounds, automation, less solvent compared to other liquid chromatography methods, easy pretreatment, fast, and cheap analysis [37,38].Terefore, it can be considered as a green analytical technique [39].
As mentioned above, it is vital to use extraction systems employing environmentally friendly solvents to reduce the consumption of hazardous organic materials.Tus, in this investigation, a DES based on the combination of decanoic acid (DA) as a HBD and tetrabutylammonium bromide (TBAB) as a HBA was used for the simultaneous preconcentration of two synthetic dyes followed by HPTLC technique as the modern analytical instrument of TLC for the determination of dyes, which is a reliable and cost-efective technique.Additionally, central composite design (CCD) was applied as an efcient method to study the interactions between efective parameters and system response.Te developed analysis is innovative in terms of the methodology for the simultaneous quantitation of artifcial dyes and the used solvent for the extraction process.
Agilent Technologies Cary Series 100 UV/VIS Spectrophotometer was used for spectra acquisition.Te chromatographic analyses were performed by Camag HPTLC, made in Switzerland, equipped with ATS4, scanner 3, and visualizer.

Preparation of DES.
DES was prepared according to our previous report [40].In brief, 2 mmol DA and 1 mmol TBAB were mixed and stirred at a temperature of 50 °C until a clear liquid was attained (about 30 min).

Preconcentration Process. Stock solutions of FG and Rh
B (1 × 10 − 2 mg•mL −1 ) were prepared in MeOH.To survey the preconcentration of dyes, diferent concentrations of sodium chloride, volume of DES, stirrer time, and pH were studied using CCD by the software package Design-Expert version 7.0.0trial.Te efective factors and condition of levels are presented in Tables 1 and 2, respectively.For each level, in a Falcon tube (15 mL), a defnite concentration of salt (according to Table 2, for example, 0.350 mol•L −1 ) was mixed with distinctive volume of dyes 1 × 10 − 2 mg•mL −1 (according to Table 2, for example, 390 µL) and the solution was made up to 13 mL via deionized water.Ten, adjusting pH (according to Table 2, for example, 4.50) was performed and an appropriate volume of DES (according to Table 2, for example, 375 µL) was added to the Falcon tube.In the following, the mixture was stirred at desired time (according to Table 2, for example, 15 min).After fnishing the stirrer time, the solution was centrifuged at 4000 rpm for 5 min.Next, the Falcon tubes were inserted in an ice beaker, resulting in the formation of a solid and thin layer on top of the solution, which was related to the preconcentrated dyes.Tis layer was separated using a spatula and diluted using MeOH before HPTLC analysis.Te schematic preparation of this procedure is presented in Scheme 1.

Preparation of Real Samples.
Lipstick and pastille were purchased from Shiraz City, Fars Province, Iran.Te preparation of lipstick was carried out based on Yilmaz ' s et al. investigation with a little modifcation [29].In brief, 0.899 g of lipstick was weighed in a 50 mL Falcon tube and 10 mL of ethanol was added.Te sample was vortexed for 1 min, ultrasonicated for 30 min, and centrifuged for 20 min at 4000 rpm.In the next step, the supernatant was collected, diluted with a dilution factor of 50, and kept in a refrigerator for further experiments.In the following, 1.3 g of pastille was dissolved in 6 mL of deionized water.Te solution was vortexed for 1 min, ultrasonicated for 30 min at a temperature of 70 °C, and fltered using a PTFE flter.

2
Journal of Analytical Methods in Chemistry 2.5.Preconcentration of Real Samples.To extract chosen dyes using DES from real samples, 390 µL of lipstick (diluted solution) and 600 μL of supernatant solution of pastille were transferred to a Falcon tube.Ten, 2.3 mL of NaCl (2 mol•L −1 ) was added and after dilution using deionized water (13 mL), the pH of solution was adjusted (pH 4.64).Afterward, 378 μL of synthesized DES was added to the tube and the solution was stirred for 15 min.For the separation of phases, centrifugation was done at 4000 rpm for 5 min.In the following, the Falcon tube was kept in an ice beaker, which resulted in the separation of organic phase containing extracted dyes.Due to the high viscosity of DES, the obtained extraction phase was diluted by MeOH (600 μL) before HPTLC analysis.

Instrumental Conditions
. Chromatographic analysis at room temperature and 20% humidity was accomplished by Camag HPTLC, made in Switzerland, equipped with ATS4, scanner 3, and visualizer.Before each analysis, the plate was washed using MeOH to remove any contamination.Te preconcentrated dyes, in the form of band, were spotted by ATS4 under N 2 gas (5 bar pressure) on an aluminum plate silica gel 60 F 254 (20 × 20 cm) with a band length of 9 mm and a distance of 13.5 mm.2.5 and 5 µL of preconcentrated and standard dyes were applied, respectively.Developing of the plates was done manually under the following conditions: MeOH-ammonia and ethyl acetate with a ratio of 20-20 and 60% v/v, respectively, as the mobile phase; chamber saturation time,

Optimization of Chromatographic Conditions.
It is vital to improve some signifcant parameters including the volume of spots, time of tank saturation, type of mobile phase, migrating distance of solvent, and detection wavelength of analytes using HPTLC technique [41].In the current research, among diferent spot volumes that had been loaded on an aluminum silica gel plate, the volumes of 2.5 and 5 µL for preconcentrated and standard dyes, respectively, were applied, which yielded the narrow bands.Also, among various band lengths (6, 8, and 9 mm), the band length of 9 mm was selected, which resulted in an appropriate band.In order to fnd a mobile phase that can separate two analytes with a good R f , seventeen diferent mobile phases were investigated, which are described in Table 3.Finally, the best simultaneous separation for dyes of FG and Rh B was achieved using MeOH (20% v/v)-ammonia (20% v/v) and ethyl acetate (60% v/v) that resulted in R f values of 0.11 and 0.70 for FG and Rh B, respectively.Before each analysis, tank saturation, as a main factor in planar chromatography, was performed for 20 min providing a uniform atmosphere to achieve equilibration for separation.Increasing the migration distance of solvent resulted in high resolution.Terefore, the solvent distance of 90 mm was chosen to obtain a good resolution.Te HPTLC images of preconcentrated dyes are shown in Figure 1.
UV-VIS spectra of selected dyes were obtained using a spectrophotometer for spectra acquisition of FG and Rh B. By doing so, the detection of FG was carried out in an absorption mode at the maximum wavelength of the dye (620 nm).For the detection of Rh B, at frst, scanner 3 was adjusted at the maximum wavelength of dye (545 nm) in an absorption mode, which formed an inappropriate signal.Hence, due to the fuorescence properties of Rh B, scanning of the plates was performed in fuorescence mode.Te densitograms of preconcentrated FG in absorption mode at 620 nm and Rh B in fuorescence mode at 365 nm are presented in Figure 2. As shown in Figure 1, FG and Rh B were distinctly detected as red and yellow colours under 365 nm, respectively.Similarly, under visible light, FG and Rh B were separately identifed as blue and pink spots, respectively.So, there is no interference between analytes.Ethyl acetate (6.1 mL), methanol (2 mL), and ammonia (0.925 mL) 3

Journal of Analytical Methods in Chemistry
Finally, a quadratic equation is achieved via the coefcient values, which is as follows: where R FG is the area of FG and A, B, C, D are considered as the pH of mixture, volume of DES, concentration of salt, and stirrer time, respectively.
Similarly, ANOVA data for Rh B were achieved as follows: values of p value less than 0.0500 indicate that the model terms are signifcant.In this case, A, B, C, AB, AC, BC, BD, CD, A 2 , B 2 , C 2 , and D 2 are signifcant terms.Te F value for Lack of Fit (0.64) implies that the Lack of Fit is not signifcant relative to the pure error.Te predicted R-Squared of 0.9909 is in agreement with the adjusted R-Squared of 0.9955 (Table 5).High amounts of R 2 (Table 5) approve that there is a high correlation between the expected and experimental data [42].A high value of adequate precision (89.016) indicates an adequate signal.Finally, a quadratic equation is achieved via the coefcient values, which is as follows: (2) In order to understand the acquired results, the graphs of three-dimensional response surface are shown in Figure 3.To fnd the best response, the desirability parameter of 1.0 was selected as the maximum response.Of these approaches, the pH value of 4.64, salt concentration of 0.35 mol•L −1 , DES volume of 378 μL, and stirrer time of 15 min were chosen as the optimum points.

Optimization of Variables.
As mentioned before, a CCD was applied for preconcentration of two dyes to acquire the best conditions and assay the main variables and their interactions.As depicted in Figure 3, the volume of DES in the preconcentration of selected dyes was studied in a range of 225-525 μL by CCD.Because of the satisfactory properties of DESs, they have been extensively applied instead of conventional organic solvents and ionic liquids in the extraction, separation, and preconcentration processes [43].An appropriate extraction agent in liquid phase microextraction processes plays a signifcant efect for the efective separation of analytes because the DES provides a stable phase (due to nonvolatile solvent), which improves the extraction efciency [22].Moreover, in liquid phase microextraction methods, the volume of extraction agent (solvent) is considered as one of the most signifcant factors, which extremely afected the extraction efciency and preconcentration factor [22].In the current investigation, high amounts of DES lead to a signifcant response because of increasing droplets of DES.Meanwhile, a low volume of DES reduces extraction efciency, probably due to the decrease in DES droplets making insufcient separation.It seems that when the volume of solvent is high, the available droplets are more resulting in an increase of interaction between two phases.
Te application of an inorganic salt afects the extraction efciency [44].Tis is because it causes variations in ionic strength and provides an inert medium that is suitable for the extraction process.In the current study, the concentration of sodium chloride (as the salt) was investigated in the range of 0.0-0.7 mol•L −1 .As presented in Figure 3, a low signal was detected, when no salt was used.Te reason for this is that the DES cannot be absolutely removed from the solution [44]; therefore, small amounts of analytes are collected and extraction efciency will decrease.Also, high concentration of sodium chloride can reduce the extraction efciency, maybe due to the decrease in distribution coefcients of the target compounds that can change the ionic strength of the solution [44].Tus, the concentration of 0.35 mol•L −1 was selected to achieve good separation.

Journal of Analytical Methods in Chemistry
In an extraction process, pH is an essential parameter, which can increase extraction efciency.In the current research, the efect of pH was investigated in a range of 2.10-6.91 by CCD.At acidic media, the carboxylic group of Rh B (pK a � 4.1) [45] is not ionizing and approximately forms a neutral molecule, which leads to pass from the aqueous phase to the extraction phase and results in high efciency [29].At low acidic pH for FG, the anionic groups of SO 3 -convert to the neutral form (SO 3 H) resulting in high extraction efciency.In addition, the presence of DA (pK a � 4.9) and salt makes approximately an inert medium, which is suitable for an efective extraction.
Another investigated parameter was stirrer time, which was studied in the range of 5-25 min.Te amount of stirrer time between DES and analyte solution infuences the extraction efciency due to providing a mass transfer of the target compound from aqueous media to the extraction phase and also formation of a good emulsifcation [22].When the phases of organic and inorganic are mixed at high stirrer time, a partial separation of two phases happens, which results in decreasing extraction efciency [42].Consequently, the stirrer time of 15 min was achieved as the best contact time.
According to the presented reasons above, the pH value of 4.64, salt concentration of 0.35 mol•L −1 , DES volume of 378 μL, and stirrer time of 15 min as the best conditions for simultaneous preconcentration of FG and Rh B were proposed by CCD.

Method Validation.
Validation of an analytical method is important to confrm the reliability and suitability of a designed method [46].Terefore, the fgures of merit such as linear range, calibration curve, correlation coefcient (r), limit of detection (LOD), preconcentration factor (PF) [34], enrichment factor (EF) [34], and repeatability were investigated under the optimized conditions as presented in Table 6.Calibration curves of the two dyes were plotted EF � Slope of DES calibration curve Slope of calibration curve (direct injection) . (4)

Efect of Interference Components. Interference efects of
Quinoline Yellow and tartrazine as synthetic dyes were studied under optimized conditions.Te percent relative error for FG-Quinoline Yellow, FG-tartrazine, Rh B-Quinoline Yellow, and Rh B-tartrazine was achieved as 3.79%, 6.62%, 8.79%, and 4.59%, respectively, that were less than 10%, indicating no serious efects on the separation and assay of dyes using DES [31].
3.6.Analysis of Real Samples.Pastille and lipstick were used as real samples.A standard addition method was applied to plot the calibration curves.Te achieved data, HPTLC images, and three-dimensional chromatograms for quantitation of analytes are shown in Table 7 and Figures 4 and 5, respectively.As shown in Figure 4, the value of R f for FG and Rh B in real samples is similar to those in preconcentrated dyes (Figure 1), which proves the existence of FG and Rh B in pastille and lipstick, respectively.In addition, the achieved data are compatible with three-dimensional chromatograms (obtained from scanner 3) of FG and Rh B, which are depicted in Figure 5.For each dye, the analysis was performed in triplet.Finally, the amounts of FG and Rh B in real samples were found to be 0.01 and 0.55 mg•g −1 , respectively.Te recovery was calculated using the equation of 5 [40], which showed that the matrix efect was not important in the extraction process.Also, the results revealed acceptable recoveries for the quantitation of both analytes (FG and Rh B) in real samples.

Conclusions
In the current research, an innovative microextraction method (DES) combined with the HPTLC technique for the preconcentration, separation, and determination of two synthetic dyes including FG and Rh B was established.Te selection of a mixture of MeOH (20%), ammonia (20%), and ethyl acetate (60%) as the mobile phase in HPTLC analysis displayed good separation (diferent R f values) between FG and Rh B. Te best conditions for the simultaneous preconcentration of FG and Rh B were investigated using CCD, achieving a pH value of 4.64, salt concentration of 0.35 mol•L −1 , DES volume of 378 μL, and stirrer time of 15 min.Te quantitative analysis for two dyes showed good repeatability (RSD < 5%), high correlation coefcient (r � 0.998 for two dyes), low interference, and excellent PF (21.66 for both dyes) and EF (7.43 for FG and 10.77 for Rh B) parameters, which can be suitable for simultaneous detection of FG and Rh B in food, medicine, textile, and cosmetic industries.Also, the analysis of real samples revealed a low serious matrix efect in this procedure.Te main advantage of the developed analysis is the application of a green and environmentally friendly extraction solvent.To the best of our knowledge, the proposed method is the frst study for the simultaneous preconcentration of FG and Rh B using DES and the determination of dyes via HPTLC as a rapid, green, and cheap analytical method.Finally, it can be concluded that due to the possible risk of synthetic dyes for the environment and human health, the developed process is favorable for routine analysis in diferent industries because of its sensitivity, simplicity, time saving, and expenditure.

20. 0
min; volume of mobile phase, 10 mL; and migration distance, 90 mm.Scanning of the plates was performed by scanner 3 to the following conditions: the wavelength of 620 nm and absorption mode for FG; the wavelength of 365 nm and fuorescence mode for Rh B; slit dimension, 6.00 mm × 0.40 mm, macro; scanning speed, 20 mm/s; data resolution, 100 µm/step; lamp W for FG; lamp Hg for Rh B. Finally, the image of plates was obtained by the visualizer at 366 nm and visible wavelengths.Recording of data was carried out via WinCATS software.

Figure 1 :
Figure 1: High-performance thin layer chromatography images of preconcentrated fast green FCF and rhodamine B using deep eutectic solvent.(A) Under UV 366 nm.(B) Under visible light.
r e r ti m e ( m in ) C o n c e n tr a ti o n o f s a lt ( m o l/ L ) r e r ti m e ( m in ) C o n c e n tr a ti o n o f s a lt ( m o l/ L )

Figure 3 :
Figure 3: Tree-dimensional response surface graphs for simultaneous preconcentration of fast green FCF and rhodamine B.

Figure 4 :
Figure 4: High-performance thin layer chromatography images of fast green FCF and rhodamine B. (A) Lipstick at the wavelength of 366 nm.(B) Pastille at visible light.

Figure 5 :
Figure 5: Tree-dimensional chromatograms of (a) fast green FCF in pastille and (b) rhodamine B in lipstick.(A), (B), and (C) are related to diferent concentrations of analytes based on the standard addition method.

Table 2 :
Experimental conditions from central composite design analysis for the simultaneous preconcentration of fast green FCF and rhodamine B.

Table 1 :
Efective factors and limit of levels for response surface quadratic model.

Table 1 ,
a CCD was applied to adjust four efective parameters including pH, concentration of salt, volume of DES, and stirrer time.30experimentsweredonebased on the experimental design.Characteristics of each level are shown in Table2.Analysis of variance (ANOVA) shown in Table4expresses the signifcance of designed model and the main interactions among variables.Te amounts of F value (Table4) show that the model for two dyes is signifcant.

Table 3 :
Various mobile phases for simultaneous separation of fast green FCF and rhodamine B.

Table 4 :
Analysis of variance results for simultaneous preconcentration of fast green FCF and rhodamine B.

Table 5 :
Te values of R 2 from central composite design analysis for fast green FCF and rhodamine B.

Table 6 :
Analytical parameters for simultaneous determination of fast green FCF and rhodamine B based on deep eutectic solvent.

Table 7 :
Determination of fast green FCF and rhodamine B in real samples via standard addition method.
� C found − C real  * 100 C added .(5) 3.7.Comparison of Method with Other Investigations.Te proposed method was compared to other investigations for the analysis of FG and Rh B. As shown in Table 8, the current research provided reasonable results.Compared with other studies, RSD % and PF amounts were higher than or comparable to values achieved using other analytical methods, indicating that the current study can be sensitive and accurate for the simultaneous analysis of FG and Rh B.

Table 8 :
Comparison of the proposed method with other analytical techniques for the determination of fast green FCF and rhodamine B.