1 Introduction

Color is used in foodstuffs to give them charming and attractive visibility and to give them allure features (Rovina et al. 2016). Allura red (AR), brilliant blue (BB) and sunset yellow (SY) are the synthetic food dyes which are manufactured from petroleum and its derivatives and they have been known as food additives with E129, E133 and E110 food codex numbers for AR, BB and SY, respectively (Khorshidi and Niazi 2018). These are the most used and preferred dyes in foodstuffs because each of mentioned dyes have an impressive and attractive appearances. In addition, these synthetics dyes are among the third group chemicals which are listed in International Agency for Research on Cancer (IARC) (https://monographs.iarc.fr/list-of-classifications). The substances which belong to list of IARC 3th Group chemicals are not classifiable as to its carcinogenicity to humans. Although these food dyes have not carcinogenic effect on human, the use of food dyes in foodstuffs could lead to some drawbacks for human health and adverse effect on human. The use and overuse of these food dyes in foodstuffs could cause to diverse health problem in adults and children who consumes excessive amount of these foodstuffs (Wang et al. 2015). The health problems including eczema, skin irritation and asthma in the adults and hyperactive behavior disorders in children could be observed (Gosetti et al. 2013). For this reason, WHO has defined the limited concentration for each food dye which can be consumed for per day. The identified limited concentration value has been known as acceptable daily intake (ADI). These ADI values are 7, 6 and 4 mg kg−1 per day on a basis of human body weight for AR, BB and SY, respectively (ECFA 2016; ECFA 2017). For these reasons detection and determination of food dyes in foodstuffs are necessary and have a great importance.

Various scientific analytical instruments including UV–Vis spectrophotometer (El-Sheikh and Al-Degs 2013), high-performance liquid chromatography (Vidal et al. 2018), spectrofluorometer (Fu et al. 2015), electrochemical voltammetric analyzer (Gao et al. 2016) and mass spectrometer (Tamim et al. 2020) were used to detect and determine dye contents of foodstuffs. Nevertheless, during the measurements of dye concentrations various limitations and obstacles could be observed. One of the problems is that food samples can contain at a very low-level dye concentration which is below than dye detection limit of the instrument. Instrument cannot detect and determine dye content of food samples. Second one is the either positive or negative interference effect of matrix components (Soylak et al. 2011). Because of interfering species dye concentrations of foodstuffs can be measured higher or lower than they contain. Therefore, dye concentrations of the foodstuffs cannot be detected and determined accurately (Bişgin et al. 2015a). Due to these problems, separation and preconcentration methods are needed.

In order to solve these obstacles and limitations, analytical chemists have developed various separation, purification and preconcentration methods including solid-phase extraction (SPE) (Oymak et al. 2020), cloud point extraction (Shokrollahi and Ahmadi 2017), dispersive liquid–liquid microextraction (Unsal et al. 2015), deep eutectic solvent extraction (Uzcan and Soylak 2020), dispersive solid-phase microextraction (Dil et al. 2016) and liquid–liquid microextraction (Rovina et al. 2017). Among these sample extraction and preconcentration methods SPE is simple, consistent, robust, precision, stable and very sensitive for organic dyes which have high adsorption affinity to solid phase (Behbahani et al. 2014a). It is possible to operate with samples which have high salinity matrices because resin can be washed with water after adsorption process of target analyte on the resin. Hence, foreign ions can be eliminated from working medium. Moreover, the SPE method provides working possibility at higher sample volume than other separation methods. Thus, higher preconcentration factors and low detection limits can be obtained. In addition, solid-phase material has high adsorption capacity and can be used many times by regenerating with appropriate solvent (Behbahani et al. 2014b).

SPE combined UV–Vis spectrophotometry is very simple, suitable, fast and practical for extraction and determination of the food dyes among these instrumental techniques and methods. In order to operate a UV–Vis spectrophotometer, high operation experience does not require and it is not needed highly experienced and skilled analyst. On the other hand, UV–Vis spectrophotometer price is lower than other instruments and it is available in lots of laboratory. These features make the SPE combined UV–Vis spectrophotometry more attractive and preferable compared with other methods and techniques (Behbahani et al. 2015).

Dowex Optipore V493 highly cross-linked synthetic styrenic polymer which has excellent adsorption capacity for organic dyes was used as a solid-phase material. The polymeric adsorbent V493 was chosen because it has hydrophobic considerably high adsorbent surface, unique pore size and spherical beads with good physical strength. It is insoluble in strong acids, strong bases and organic solvents. V 493 is powerful adsorbent that can be recycled through a regeneration cycle prior to use to extract organic dyes from aqueous solution. The usage of highly cross-linked polymeric adsorbent in the solid-phase application for extraction, determination, preconcentration and adsorption of food dyes is of great importance. Therefore, there are many SPE applications on adsorption of food dyes onto hypercrosslinked polystyrene resin in the literature (Alkahtani et al. 2017; Tikhomirova et al. 2018).

In this study, SPE method was developed for single determination of AR, BB and SY and simultaneous determination of AR–BB and BB–SY. V493 highly cross-linked synthetic styrenic polymer was used for this purpose. Extraction parameters were separately examined for each of the dyes and optimized in detail. Foodstuff samples containing AR, BB and SY were subjected to the developed method and their dye contents were determined by UV–Vis spectrophotometry. Finally, analytical features of the method for each of the dyes were compared with each other and recent reported studies.

2 Materials and Methods

2.1 Chemicals

Dowex optipore V493 (Dow Chemical Company) commercially available polymeric adsorbent is highly cross-linked synthetic styrenic polymer resin that is insoluble in strong acid, strong base and organic solvents. It has considerably high bet surface area which is 1100 m2 g−1. It has 20–50 mesh size and orange to brown spheres physical form (https://www.lenntech.com/Data-sheets/Dowex-Opticore-L493-L.pdf). It was used throughout the extraction and preconcentration studies.

All commercial available analytical grade chemicals used in the experiments were purchased from Merck (Merck, Darmstadt, Germany) and Sigma (Sigma–Aldrich, St. Louis, MO, USA) and they were used without purification process.

In the preparation of buffer solutions, H3PO4–NaH2PO4 for pH 2, 3, HAc–NaAc for pH 4, 5 and NaH2PO4–Na2HPO4 for pH 6–8 were used to fix the aqueous media pH.

Each of 100 µg mL−1 of AR, BB and SY stock dye solutions were prepared and stored in refrigerator at 4 °C. More diluted dye solutions were prepared daily from these stock dye solutions.

2.2 Instruments

A Shimadzu Model UV-160A double-beam UV–Vis spectrophotometer (Shimadzu Corporation, Kyoto, Japan) was used to measure of absorbance and absorption spectra of AR, BB and SY. pHs of the buffer and aqueous sample solutions were measured and controlled with Hanna Model HI-2211 digital pH meter (Hanna Instruments, Woonsocket, RI, USA). The chromatographic mini glass column (Çalışkan Laboratuvar Ürünleri Ltd. Şti., Ankara, Turkey) has 1 cm in diameter, 10 cm in length and 500 mL reservoir with Teflon stopcock was used in extraction and preconcentration procedures. In order to provide stabile sample and eluent flow rate, the LongerPump YZ1515x BT100-2J model peristaltic pump (Longer Precision Pump Co., Ltd., Hebei, China) was combined with mini chromatographic glass column. Olympus SZX7 (Olympus Corporation, Tokyo, Japan) model industrial stereo microscope was used to take micrographs of the dye adsorbed form and natural form of V493 resin.

2.3 Solid-Phase Extraction Procedure

The chromatographic glass separation column was loaded with 350 mg of V493 resin. The resin was washed with water after each organic solvent was passed through the column. Acetone and water, methanol and water were sequentially passed through the column to remove impurities and other contaminants. Each of solvents volumes used in the experiments to remove impurities from the column was 25 mL.

Each of 25 mL of sample solutions containing 10 µg AR, 3.5 µg BB and 10 µg SY dyes were prepared separately. These solutions were buffered to pH 4 which has 0.2 mol/L of buffer concentration and 0.3 mol/L ionic strength concentration by adding HAc//NaAC binary buffer chemicals and NaCl. The solutions were separately poured into reservoir of the column. Sample and eluent flow rates were controlled only using peristaltic pump and by disabling stopcock of the chromatographic column. It was permitted that solution passed through the column under vacuum by using the peristaltic pump by applying at 2 mL/min sample flow rate each of adsorbed dye molecules on the resin was eluted to 4 mL of final volume with pure methanol solution by controlling flow rate with the peristaltic pump and applying at 2 mL/min eluent flow rate. AR, BB and SY concentrations of the eluent solutions were determined separately at 506 nm, 630 nm and 483 nm, respectively.

2.4 Adsorption Experiments

Batch adsorption experiments were performed separately for each of target dyes to investigate adsorption behavior between dyes and resin and to explain adsorption mechanism of the method. Experiments were conducted in a 15 mL polyethylene plastic centrifuge tube with cap, at room temperature, by using pH 4 buffer solution and contacting with 25 mg of DOV493 resin for 4 h. Adsorption equilibrium between dyes and resin was reached after shaking by a shaker for 4 h. The suspensions were centrifuged and dye concentrations of homogenous aqueous phases were determined at 506, 630 nm and 483 nm for AR, BB and SY, respectively.

2.5 Pretreatment of Foodstuff Samples

The certain amounts of solid foodstuff samples were dissolved in pure water. Solutions were subjected to the optimized method after filtration process and necessary dilutions.

The certain volumes of liquid foodstuff samples were directly applied to the method after necessary dilutions.

3 Results and Discussion

In the complete study, in order to well-understand and see easily, all figures are given according to original color of the target dyes. Red, blue and yellow-orange colors symbolize AR, BB and SY, respectively. Any interference effects were observed between AR–BB and BB–SY binary dyes. The UV–Vis spectra of AR, BB, SY, mixture of AR–BB and mixture of BB–SY are given in Fig. 1 with their molecule structures. Each of experimental parameters of the method were investigated and optimized separately for each of the dyes by applying three replicates. First, whole optimization processes were performed separately to obtain quantitative extraction for each of the dyes and then simultaneous preconcentration and determination of AR–BB and BB–SY were examined and performed.

Fig. 1
figure 1

UV–Vis spectra of AR, BB, SY, mixture of AR–BB and mixture of BB–SY; molecule structures of AR (a), SY (b) and BB (c)

Full size image

3.1 Effect of pH

In column SPE studies, pH is the most important parameter in view of adsorption of organic dyes onto resin. Applied pH must provide complete adsorption of dye molecules onto resin. Effect of pHs on the extraction were separately investigated for the each of the dyes between pH 2 and pH 8 by applying and keeping constant the optimum values of the other parameters. Results are given in Fig. 2a with bar graphic and standard deviations. Quantitative extractions were obtained between pH 2 and pH 6 for AR, pH 3–6 for BB and pH 4–7 for SY dyes. Therefore, optimum pH values were selected 5 for all the three target dyes.

Fig. 2
figure 2

Effect of pH (a), sample flow rate (b), eluent flow rate (c) and type of eluent (d) on the single extractions of AR, BB and SY, N = 3

Full size image

Hydrophobic interactions between resin and dyes are the main leading force which is affected the adsorption. As mentioned previously in V493 characteristics, the resin has hydrophobic feature. Therefore, the resin has no pHpzc thus; there are no electrostatic interactions between resin and the dyes. The reason of difference in optimum pH ranges between AR and SY is difference of hydrophobic interactions between resin and the dyes. AR and SY dyes have the almost similar molecular structures except one methoxy and a methyl groups which belong to AR molecule. These alkoxy and alkyl groups make the AR molecule more hydrophobic than SY molecule. In addition, molecular weight of AR is slightly higher than SY molecule. Therefore, the resin has higher adsorption affinity to AR than SY. Because of all these reasons, difference in optimum pH ranges between AR and SY was observed (Bişgin 2019; Rukosueva et al. 2020).

3.2 Effect of Sample Flow Rate

In column solid-phase extraction studies, sample flow rate is very important parameter to provide complete retention of analyte onto adsorbent. In order to investigate effect of sample flow rate on the extraction of AR, BB and SY, sample flow rate was applied between 1 and 10 mL min−1 by keeping constant the optimum values of the other extraction parameters. Results are comparatively given in Fig. 2b for the dyes with their standard deviations. Recoveries of the examined dyes decreased with increasing sample flow rate. Quantitative recoveries were obtained between 1 and 2 mL min−1 for AR, 1 and 3 mL min−1 for both BB and SY dyes. For this reason, 2 mL min−1 of sample flow rate was applied for three dyes in all further extraction studies to save time and to apply fast extraction.

3.3 Effect of Eluent Flow Rate

The eluent flow rate is important parameter in column SPE studies to obtain quantitative results because each of dyes must be eluted one by one or simultaneously as quantitative. To examine the effect of eluent flow rate on the extraction of AR, BB and SY, eluent flow rate was separately investigated for the each of the dyes between 1 and 10 mL min−1 by keeping constant the optimum values of the other extraction parameters. Results are depicted in Fig. 2c with standard deviations. Recoveries of the dyes decreased with increasing eluent flow rate. Quantitative recoveries were obtained between 1 and 2 mL min−1 for AR, 1 and 3 mL min−1 for both BB and SY dyes. For this reason, 2 mL min−1 of sample flow rate was applied for the three dyes in all further extraction studies to save time and to apply fast extraction.

3.4 Effect of Eluent Type

Type of eluent solution is another important parameter to obtain high extraction efficiency and to desorb complete dye molecules from the resin. Effects of various eluents on the extraction were separately investigated for the each of the dyes by keeping constant the optimums of the other parameters. In order to desorb AR, BB and SY molecules adsorbed on the resin, different pure organic solvents including acetonitrile, acetone, ethanol and methanol were used. Spectrophotometric absorbance measurements of AR, BB and SY were performed separately for each solvent in presence of used own elution solvent. Recovery values were given in Fig. 2d with bar graphics. Quantitative recovery results were obtained for all three target dyes when used ethanol and methanol as an elution solvent. Therefore, methanol was selected to be optimum eluent and all further preconcentration studies were performed by using methanol.

3.5 Effect of Dowex Optipore V493 Amount

Adsorbent amount in the column solid-phase extraction is very important to obtain both quantitative retention and quantitative elution of the dyes. Optimum adequate amount of adsorbent must be loaded to the column for retention of complete dye molecules onto resin by adsorption and to provide stabile flows through the chromatographic column for both sample and eluent solutions. Amount of V493 was examined between 50 and 400 mg by keeping constant the optimums of the other variables. Resin amount experiments were performed separately for each of the target dyes. Results are given comparatively for the three dyes in Fig. 3a. The axis of resin amount values are given in reverse order to understand and see recovery values in detailed. Extraction efficiency increased with increasing resin amount up to 350 mg and then remained constant. Consequently, extraction column was loaded 350 mg of V493 resin.

Fig. 3
figure 3

Effects of resin amount (a), ionic strength (b), sample volume (c) and eluent volume (d) on the single extractions of AR, BB and SY, N = 3

Full size image

3.6 Effect of Ionic Strength

Impact of ionic strength on the single extraction of AR, BB and SY was performed separately for AR, BB and SY dyes by keeping constant the other optimum variables with using NaCl at different concentration ranges between 100 and 500 mM. Results are given in Fig. 3b. The axis of ionic strength values are given in reverse order to understand and see recovery values in detailed. Recovery values of the all three dyes increased with increasing ionic strength concentration up to 300 mM and reached plateau and remained constant. Therefore, 300 mM ionic strength was kept optimum and the optimum value of 300 mM was applied in all further extraction studies.

3.7 Sample Volume

Sample volume is one of the important parameters for preconcentration studies to obtain high preconcentration factor. Detection limits of the method for the dyes decrease with increasing preconcentration factor by depending on sample volume and eluent volume. Therefore, samples containing AR, BB and SY dyes which have very low-level dye contents can be determined as more sensitive. Influence of sample volume on the simultaneous extraction of BB and SY was examined between 25 and 300 mL of sample volume by keeping constant the optimum values of the other parameters. Results are given in Fig. 3c with standard deviations. Single quantitative extraction values were observed up to 400 mL of sample volume for the three analyte dyes. At higher than 400 mL sample volume, sample solution behaved to be an eluent. Therefore, recovery results decreased with increasing sample volume at an amount higher than 400 mL. As a result of these experiments and according to 4 mL of eluent volume, 100 preconcentration factor was obtained for the three analyte dyes.

3.8 Eluent Volume

Eluent volume is as important as sample volume parameter in order to gain higher preconcentration factor (PF). Optimum eluent methanol was used to optimize eluent volume. For this purpose eluent volume was examined between 1 and 5 mL by keeping constant the other optimum extraction parameters. Investigation of eluent volume was performed separately for each of the dyes. Results are presented in Fig. 3d by giving methanol volume in reverse order to be shown more understandable recovery values. Quantitative extractions of AR, BB and SY dyes were obtained by applying 4 and 5 mL of eluent volume. In order to obtain higher PF, 4 mL methanol was selected as optimum elution volume.

3.9 Adsorption Behavior of the Method

Adsorption isotherms help to describe the interaction between sorbent and dye molecules and are useful to determine adsorption behaviors of the method. Adsorption isotherms with related to retentions of AR, BB and SY onto V493 resin were separately investigated for each of the dyes.

Obtained experimental results and calculations have been separately applied to fit Langmuir and Freundlich adsorption isotherm models for each of the dyes. Perfect compatibilities with high correlation coefficients were observed with Freundlich model. The Freundlich model was found to be the most appropriate to describe the adsorption process of AR, BB and SY dyes on V493 with suggesting that multilayer adsorption process occurs on the heterogeneous distribution of active sites onto V493 surface. Adsorption isotherms with related to adsorption of AR, BB and SY are given comparatively in Fig. 4a–c.

Fig. 4
figure 4

Adsorption isotherms of the method for AR (a), BB (b) and SY (c); adsorbed AR (d), adsorbed BB (e), adsorbed SY (f), pure resin (g), adsorbed AR–BB (h) and adsorbed BB–SY (i) on the resin

Full size image

In order to support adsorption of the all three dyes and mixtures of the dyes on the surface of V493, surface micrographs of the resin was also taken with industrial stereo microscope. AR, BB and SY adsorbed resins are given in Fig. 4d–f, respectively. Pure resin micrograph is given in Fig. 4g. Mixtures of AR–BB and BB–SY adsorbed resins are given in Fig. 4h, i, respectively. The dye adsorbed colorful resin micrographs showed the quantitative applicability of the method to real samples containing AR, BB, SY, mixtures of AR–BB and BB–SY.

3.10 Effect of Matrix Components

Potentially interfering species including dyes and ions, which could be found in foodstuffs, effect the accurate determination of dye concentration. Interference effects of commonly used dyes and naturally found ions in foodstuffs were investigated by applying optimum conditions. The examined concentrations of the interfering species were also given in Table 1. Investigated interfering ions and dyes at given concentrations in Table 1 have no interference effects on the determination of target dyes. These results suggested the applicability of the developed method for the single and simultaneous extraction and determination of AR, BB and SY in food samples.

Table 1 Matrix effect of the interfering species, N = 3
Full size table

3.11 Application of Foodstuff Samples

In order to prove applicability of the method red energy drink, bear jelly and red candy samples containing AR dye were subjected to the method. Analyte addition/recovery tests were also applied to the samples by varying added amounts of AR between 2.00 and 4.00 µg. Results are given in Table 2 with standard deviations. Recovery results were obtained between 95 and 101%.

Table 2 Analyte addition/recovery tests for determination of AR in foodstuffs, N = 3
Full size table

In order to show applicability of the method on determination of foodstuff samples containing BB dye, optimized parameters of the method were applied to blue energy drink, citrus flavored cocktail syrup and blue candy. Analyte addition/recovery tests were also applied to the samples by adding 1.00 µg and 2.00 µg of BB dye. Results are presented in Table 3 with standard deviations. Quantitative satisfactorily recovery values were obtained in the range between 95 and 97%.

Table 3 Analyte addition/recovery tests for determination of BB in foodstuffs, N = 3
Full size table

The determinations of SY concentrations of orange flavored cocktail syrup, syrup medicine and peach juice powdered drink were performed by applying optimized parameters of the developed method. In order to test applicability and repeatability of the method analyte addition/recovery technique was also use. For this purpose, 2.00 µg SY and 4.00 µg SY were added to foodstuff samples. Results are given in Table 4 with standard deviations. Quantitative recovery values were obtained between 96 and 99%.

Table 4 Analyte addition/recovery tests for determination of SY in foodstuffs, N = 3
Full size table

Finally, simultaneous extractions and determinations of AR–BB and BB–SY were also tested by adding various amounts of AR, BB and SY dyes to diverse matrices samples such as tap water and natural spring water. Results are depicted in Table 5 with standard deviations. Recovery values were obtained between 97 and 102%. These satisfactorily quantitative results showed the applicability and repeatability of the method on the simultaneous extraction and determination of AR–BB and BB–SY in different matrix.

Table 5 Analyte addition/recovery tests on the simultaneous extraction of AR–BB and BB–SY, N = 3
Full size table

The determined AR, BB and SY concentrations of foodstuff samples are given as µg g−1 and µg mL−1 in Table 6 with standard deviations according to their solid and liquid forms, respectively. AR contents of foodstuffs were determined between 6.85 and 21.25 µg g−1 and 20.40 and 27.21 µg mL−1 for solid and liquid samples, respectively. BB concentrations were between 3.15 and 10.05 µg g−1 for solid samples and 5.06 and 43.30 µg mL−1 for liquid samples. SY levels of solid and liquid foodstuff samples were determined between 18.55–121.46 µg g−1 and 3.04–29.67 µg mL−1, respectively.

Table 6 AR, BB and SY contents of foodstuff samples
Full size table

3.12 Analytical Features of the Method

Optimum parameters of the method were determined separately for AR, BB and SY. Under the optimum conditions analytical characteristics including limit of detection (LOD), preconcentration factor (PF), linear dynamic ranges (LDR) were determined separately for each of the dyes. Analytical characteristics and optimum parameters of the method are given comparatively in Table 7 for AR, BB and SY dyes. Reusability study for the V493 resin was also performed. A 350 mg resin was used at least 34 cycles by applying optimum parameters of the method and using methanol as the regeneration chemical.

Table 7 Analytical characteristics and optimums of the method
Full size table

Analytical specifications of the method was determined and also compared in detail with reported recent studies in Table 8. LOD values of the method were determined 0.55, 0.20 and 0.58 µg L−1 for AR, BB and SY, respectively. LOD values of the dyes are lower than reported studies. PF was obtained to be 100 for the all three target dyes. PF was higher than those reported studies. 0.055–6.0 µg mL−1, 0.020–4.2 µg mL−1 and 0.058–10.0 µg mL−1 linear dynamic ranges (LDR) were determined with correlation coefficient 0.9996, 0.9995 and 0.9993 for AR, BB and SY, respectively. LDR values of the method for the target dyes was comparable with previously reported studies. Relative standard deviations (%RSD) of the method were calculated below than 3%, 4% and 3% for AR, BB and SY, respectively. These RSD values were lower than previously reported studies in Table 8.

Table 8 Comparison between analytical features of the method and recently reported methods on the extraction and determination of AR, BB and SY
Full size table

4 Conclusions

The proposed SPE method is easy, safe, simple, consistent, robust, precision, stable, fast and economic methodology for the either single preconcentration and determination of AR, BB and SY and simultaneous preconcentration and determination of AR–BB and BB–SY in diverse foodstuff samples. The method offers broad application pH range between 3 and 6 for both single and simultaneous preconcentration of target dyes. Commercially available simple and a little toxic eluents such as ethanol and methanol were used. The experimental quantitative satisfactorily results and data demonstrate that single and simultaneous, reliable determination of AR, BB and SY in different food samples are applicable. Colorful surface micrographs prove that detecting and determining AR, BB and SY dyes could be performed simultaneously and separately. The mentioned advantages and sufficient features of the method will make the method more useful for future food quality and control applications. Due to its safety, simplicity and practicality, the method can be applied to the either determination of AR, BB and SY or simultaneous determination of AR–BB and BB–SY in diverse foodstuffs.