Nano-Al₂O₃-based micro solid-phase filter membrane extraction for simultaneous determination of tartrazine and sunset yellow in food

Abstract

A new micro-solid-phase extraction method based on a prestructured filter membrane was developed to facilitate simultaneous analysis of tartrazine (TA) and sunset yellow (SY) with high performance liquid chromatography (HPLC). A filter membrane impregnated with nano-Al₂O₃ particles was proposed as a new micro-solid-phase extraction (μ-SPE) device for the preconcentration of colorants in food. The method integrated isolation and concentration in one step, and the adsorbed colorants could be easily eluted from the filter membrane for HPLC analysis. This process does not involve any additional sample treatment. The method is facile and easy to operate. The method showed good linearity, with correlation coefficients >0.9982 in the range of 1.0–200.0 ng mL⁻¹, and low limits of detection (0.10–0.20 ng mL⁻¹). Finally, the recovery values were 86.9–124.0% for the colorants in various food samples. The results indicated that the developed method would be a novel method for rapid determination of TA and SY at trace levels in real samples.

---

1. Introduction

Tartrazine (TA) and sunset yellow (SY) are widely used synthetic dyes because of their low cost, excellent dyeing property, and good stability.^1–3 Generally, colorants contain azo (N=N) functional groups and aromatic ring structures,⁴ and must be controlled strictly by laws because of their potential risk to human health.^5–7 More and more evidence indicates that the abuse of synthetic pigments may cause cancer.^8–10 TA and SY (Fig. 1) are two of these azo colorants,^11,12 and permitted to be used as food additives in some common food products, such as sugar candies, sugar-coated pills, nutrient-enhanced sports beverages, cold drinks, cakes, and gelatins.^13–15 But they are banned as colorants in fisheries and cooked meat products. So the use of TA and SY in food products must be strictly controlled and supervised.¹⁶ Nowadays, it is found that they are used to attract consumers and enhance or mimic food color intensity in croakers and roast chickens. Consequently, considering both the potential effects on human health and the need for knowledge of the components of food, reliable methods for determining synthetic colorants are required to ensure food safety.^17,18

Fig. 1 The chemical structure of TA (A) and SY (B).

Today, high-performance liquid chromatography (HPLC) method is widely used because of its high sensitivity, selectivity, and capability of multi-component validation. Due to the complex matrix effect and the low level concentration of synthetic colorants in food samples, direct determinations are often difficult.¹⁹ Hence, the pretreatment and concentration step is inevitable before the determination of synthetic food colorants to enhance the sensitivity and selectivity.

Solid-phase extraction (SPE)^20,21 and solvent extraction (SE)^22,23 as clean-up techniques have been used for the determination of some synthetic dyes. However, both SPE and SE required a relatively large amount of solvents, and these protocols are laborious and time consuming. Miniaturization has become an important issue in the development of sample preparation techniques.²⁴ In recent years, some new modes of microextraction, which were termed as micro-solid-phase extraction (μ-SPE),²⁵ dispersive micro-solid-phase extraction (d-μ-SPE)²⁶ and dispersive liquid–liquid microextraction (d-μ-LLE)²⁷ have been developed for sample preparation. μ-SPE has the advantages of easy operation, short time-consuming, synchronous separation, purification and enrichment for determination of trace target analytes.

The selection of sorbent is the first and the most important process during μ-SPE. Nano-materials including carbon nanotubes (CNTs),²⁸ TiO₂ nanotube array²⁹ and MOFs³⁰ have attracted much attention due to their high adsorption capacity, simple operation, rapid adsorption process, etc. The hexagonal arrays of carbon atoms in graphene sheets of CNT surfaces are ideal for strong interactions with other molecules.³¹ In addition, the large surface area gives CNTs excellent sorption capacity in the preconcentration of organic compounds and heavy-metal ions.^32–34 TiO₂ nanotube array offer several advantages over the traditional microsize sorbents such as larger surface area, higher adsorption capacity, more easily to prepare large enrichment equipment, and higher possibility for cycle use.³⁵ MOFs have found many applications in the preconcentration of organic and/or inorganic substances in SPE due to their large adsorption capacity and uniform pore structure.^36,37 However, instability and limited functional binding sites make it difficult to apply MOFs into complicated matrix.

In this work, low-cost nano-Al₂O₃ combined with filter membrane was applied to the rapid colorants analysis. Nano-Al₂O₃-based filter membrane offers some advantages such as simplicity, low disposal costs, short time of extraction, high enrichment factor, and rapid sample clean-up. This method was used for the analysis of two colorants (TA and SY) in food samples. The main parameters affecting extraction efficiency were investigated. Under the optimized conditions, the performance of the present method for analysis of real samples was examined.

2. Experimental

2.1 Reagents and materials

All chemicals were analytical grade unless noted otherwise. Deionized water purified using a Sartorius Arium 611 system (Sartorius, Göttingen, Germany) was used throughout the experiment. HPLC grade methanol was purchased from Fisher Corporation (Pittsburgh, PA, USA). TA and SY were purchased from Aladdin Chemistry Co. Ltd. (Shanghai, China). The stock solutions of TA and SY (1.0 mg mL⁻¹) were prepared with deionized water and stored at 4 °C, and working solutions were obtained by appropriately diluting the stock standard solutions to the required concentrations.

Nano-Al₂O₃ was purchased from Beijing Nachen Nano-Meter Material Co., Ltd, China and its particle size was in 10–15 nm range. Nano-SiO₂ was purchased from Shanghai Xinzhong Chemical Reagent Co., China. Anatase nano-TiO₂ (nano-TiO₂(A)) and rutile nano-TiO₂ (nano-TiO₂(R)) were purchased from Zhoushanmingri Nanometer Material Co., China, and its particle size was about 10–15 nm. C₁₈ (SKR-10-5120; 50 μm, 120 Å) was purchased from Jinan Bona biological technology Co., Ltd., China. The nylon 66 filter membrane (13 mm × 0.22 μm) used throughout the experiment was purchased from Tianjin Jinteng Ltd., China.

2.2 Apparatus and operating conditions for HPLC

HPLC analysis was performed with an Agilent 1100 HPLC system (Palo Alto, CA, USA) equipped with an automatic sampler and diode array detector (DAD). Chromatographic separation of target analytes was performed on a ZOABAX Eclipse XDB-C₁₈ (150 mm × 4.6 mm, 5 μm) column (Agilent, Palo Alto, CA, USA) and the injection volume was 10 μL. A mixture of water and methanol (90 : 10, v/v) at a flow rate of 0.8 mL min⁻¹ was used as a mobile phase in isocratic elution mode. The detection wavelength was set at 254 nm.

2.3 Micro-solid-phase extraction procedure

Briefly, 5.0 mg nano-Al₂O₃ and 10.0 mL deionized water were mixed to form a homogeneous suspension. Then the nano-Al₂O₃ suspension was homogeneously loaded on a nylon 66 filter membrane by a syringe driver. The membrane impregnated with nano-Al₂O₃ was pre-conditioned with 1.0 mL methanol and 1.0 mL deionized water, respectively. Then, 150.0 mL of the aqueous samples containing TA and SY were loaded on the membrane by a syringe. The analytes were desorbed by 3.0 mL 0.2 M NaCl. Finally, the eluent with TA and SY was dried under nitrogen and dissolved with 0.5 mL deionized water for further HPLC analysis (shown in Fig. 2). The used membrane was washed with methanol and deionized water for several times to remove the residual colorants and NaCl and dried at 70 °C for the next recycle use.

Fig. 2 Schematic procedure of μ-SPE.

2.4 Sample treatment

The croakers, corn flavored candy, orange flavored candy and green tea flavored candy were purchased from a local supermarket (Shenyang, China) and stored at a refrigerator.

For croaker samples, after being homogenized in a high-speed food blender, they were stored below −20 °C in a freezer until the time of analysis. A 2.0 g of homogenized sample were accurately weighed. Then 50 mL of methanol–water (1 : 1 v/v) were added and the sample was placed into an ultrasonic bath for 15 min and centrifuged at 4000 rpm for 10 min. The supernatant was separated and the solid-state sample was extracted twice again with a further 100 mL of the solvents mixture. The extracts were collected and subjected to the μ-SPE process.

For candy samples, a 1.0 g sample pounded to a homogenate was accurately weighed and mixed with 50 mL water. The mixture was dissolved in a water bath at 60 °C for 15 min, followed by ultrasonication for 15 min, and then centrifuged at 4000 rpm for 10 min. The supernatant was separated and the lower sediment was extracted twice as mentioned above. Finally the sample solution was used for the following μ-SPE step described as 2.3.

3. Results and discussion

3.1 Optimization of the adsorption experiments

To achieve the best extraction efficiency of the target colorants with the filter membrane impregnated by nano-Al₂O₃, several experimental parameters, such as the adsorption efficiency of nano-Al₂O₃ compared with other sorbents, nano-Al₂O₃ amount dispersed on filter membrane, sample pH and ionic strength were investigated using standard solutions containing TA and SY.

3.1.1 Selection of sorbent. To find the best sorbent for extraction of TA and SY, several sorbents including nano-Al₂O₃, nano-SiO₂, nano-TiO₂(A), nano-TiO₂(R) and C₁₈ were applied to the dispersive solid-phase extraction for 10.0 mL of 5.0 μg mL⁻¹ colorant solution. As shown in Fig. 3A, it was found that nano-Al₂O₃ provided the highest adsorption efficiency of more than 96%. So nano-Al₂O₃ was chosen as solid-phase extractant to be dispersed on filter membrane for extraction of TA and SY in the following experiments. This filter membrane-based device has the advantage of low cost, short time-consuming, synchronous separation and easy recycling.

Fig. 3 The optimization of adsorption experiments: (A) the selection of sorbent (sorbent amount: 50.0 mg; shake time: 40 min); (B) the effect of nano-Al₂O₃ amount dispersed on the filter membrane; (C) the effect of pH; (D) the effect of ionic strength.

3.1.2 Effect of nano-Al₂O₃ amount dispersed on the filter membrane. Different amounts of the nano-Al₂O₃ in a range of 0–10.0 mg were dispersed on the filter membrane to extract colorants from the sample solutions. The results (Fig. 3B) indicated that 5.0 mg nano-Al₂O₃ was enough for the extraction, and further increasing the amount of the adsorbent gave no significant improvement for the extraction efficiency of colorants. Therefore, the filter membrane containing 5.0 mg nano-Al₂O₃ was selected to be applied to μ-SPE process for the following experiments. High adsorption efficiency of colorants from aqueous phase could be attributed to the large contact area of nanoparticles and analytes on the membrane filter.

3.1.3 Effect of sample pH. The influence of pH on extraction based on nano-Al₂O₃ was investigated. The pH values of the sample solution were adjusted over the range of 3.0–10.0 to investigate the effect. As could be seen in Fig. 3C, initially the adsorption efficiency remained high when pH was raised over 3.0–6.0. However, when pH value was ranged from 6.0–10.0, both the analytes showed a decrease in adsorption efficiency. In consideration of these observations, a pH of 6.0 was considered optimal in μ-SPE.

3.1.4 Effect of ionic strength. In order to investigate the influence of the ionic strength, the colorant adsorption experiments were carried out by varying NaCl concentration from 0.02 to 1.0 M. Theoretically, when the electrostatic forces between the adsorbent surface and adsorbate ions were attractive, as in this system, an increase in ionic strength will decrease the adsorption capacity. Conversely, when the electrostatic forces were repulsive, an increase in ionic strength will increase adsorption.³⁸ Results showed that the adsorption efficiency of TA and SY decreased as NaCl concentration increased (Fig. 3D), which was in accordance with the theory described above. Thus in the adsorption experiment, the colorant solution without adjusting ionic strength was adopted.

3.2 Adsorption mechanism of colorants on nano-Al₂O₃

The adsorption mechanism was discussed mainly according to the point of zero charge (pzc) of nano-Al₂O₃ (Fig. 4) and pH experiments (Fig. 3C). The point of zero charge (pzc) of nano-Al₂O₃ obtained was 9.0. At pH < pH_pzc, the nano-Al₂O₃ surface carried positive charges, and this will enhance its electrostatic attraction with anionic dyes TA and SY. For this reason, the adsorption efficiency of TA and SY was high. While at pH > pH_pzc the nano-Al₂O₃ surface was negatively charged, so anionic dyes TA and SY couldn't be adsorbed by nano-Al₂O₃ sufficiently. The adsorption mechanism of nano-Al₂O₃ towards TA and SY as shown in Fig. 5 may be proposed on the basis of above experimental results.

Fig. 4 Zeta potential of nano-Al₂O₃ suspensions at different pH.

Fig. 5 The adsorption mechanism of TA and SY on nano-Al₂O₃.

3.3 Optimization of desorption condition

3.3.1 Effect of eluent type. Different organic solvents, including ethanol and methanol were investigated after loading the membrane filter with 10.0 mL of 5.0 μg mL⁻¹ colorant solution. Results (Fig. 6A) showed that these organic solvents couldn't quantitatively elute the colorants. From Fig. 3D, it could be concluded that NaCl was unfavourable to the interaction between nano-Al₂O₃ and the colorants. So NaCl solution was investigated to be eluent. As expected, it could be seen that the highest desorption efficiency was achieved when using NaCl solution as eluent. For further validation, different concentrations of NaCl solutions varying from 0.02 M to 0.3 M were employed as eluents, and desorption capacities were compared with each other. As shown in Fig. 6B, the recovery of colorants increased with increasing NaCl concentration and 0.2 M NaCl provided the best desorption efficiency. The result is accordant to that of the ionic strength experiment.

Fig. 6 The selection of desorption condition: (A) the effect of eluent type; (B) the effect of concentration of eluent; (C) the effect of eluent volume.

3.3.2 Effect of eluent volume. During the desorption process, the effect of eluent volume was studied ranging from 1.0 to 4.0 mL. As seen from Fig. 6C, a volume of 3.0 mL of eluent was found to be sufficient to desorb the analytes from the filter membrane containing nano-Al₂O₃. So 3.0 mL 0.2 M NaCl was chosen as the eluent. To further lower the LOD, the eluent with analytes in it was dried under nitrogen and redissolved with 0.5 mL deionized water for HPLC analysis.

3.4 Effect of sample volume

The effect of sample volume on the colorant adsorption was studied in the range of 10.0–200.0 mL; 10.0 mL sample containing 5.0 μg mL⁻¹ of colorants was diluted to 10.0, 25.0, 50.0, 75.0, 100.0, 150.0 and 200.0 mL with deionized water. Then the extraction process was performed under the optimum conditions. The results (Fig. 7) showed that the sample in the volumes up to 150.0 mL was quantitatively adsorbed by the nano-Al₂O₃-based membrane, but there was a decrease in the amount adsorbed at higher volumes. Therefore, for the determination of trace quantities of the colorants, a sample volume of 150.0 mL was selected for a high preconcentration factor.

Fig. 7 Effect of initial sample volume.

3.5 Recycling of the filter membrane

To investigate the reusability of filter membrane, the filter was reused for the extraction of TA and SY after desorption with eluent and the treating process described in 2.3. The results (Fig. 8) indicated that above 87% recovery for the target analytes was obtained after a 6 cycle run of filter membrane for the adsorption and desorption of colorants. It could be seen that the filter membrane impregnated with nano-Al₂O₃ has the advantage of reuse stability, which can reduce the cost in sample preparation.

Fig. 8 Recycling of the membrane filter.

3.6 Method validation

The μ-SPE method was evaluated by characterizing its analytical performance in terms of linearity, precision, and limits of detection (LODs) at the optimized working conditions determined earlier. The obtained results were listed in Table 1. The linearity of the method was tested over a range 1.0–200.0 and 2.0–200.0 ng mL⁻¹ for TA and SY, respectively. The calibration plots of each analyte prepared at eight concentration levels were linear, with correlation coefficients (R²) between 0.9982 and 0.9999. The precision of the method was evaluated by calculating six individual standards prepared at a concentration of 4.5 ng mL⁻¹. The relative standard deviation (RSD) ranged from 3.25% to 3.58% (n = 6). The LODs and LOQs for the colorants ranged from 0.10 to 0.20 ng mL⁻¹ and from 0.32 to 0.65 ng mL⁻¹, respectively.

Table 1 Linearity, precision and limits of detection (LODs) and quantification (LOQs) of the developed μ-SPE method

Analyte	Linear range (ng mL^−1)	Linearity equation	R^2	LODs (ng mL^−1)	LOQs (ng mL^−1)	Precision (RSD, n = 6)
						Intra-day	Inter-day
TA	1.0–200.0	y = 5.3899x − 1.8606	0.9982	0.10	0.32	3.25	4.98
SY	2.0–200.0	y = 3.1775x − 0.8027	0.9999	0.20	0.65	3.58	5.12

---

3.7 Application to real samples

The optimized μ-SPE method was applied to the extraction of colorants in various food samples including croaker, corn flavored candy, orange flavored candy and green tea flavored candy. Fig. 9 presents representative chromatograms of the croaker extraction solution and sample spiked with 67.5 ng mL⁻¹ under the determined optimal conditions. The average recoveries of the analytes in the samples ranged from 86.9% to 124.0% at different spiked levels with relative standard deviations (RSD) less than 7.8% (Table 2). The results of the experiments confirmed that the method was validated with good reproducibility, satisfactory precision and accuracy for the determination of colorants from real samples.

Fig. 9 Representative chromatograms of the croaker extraction solution and sample spiked with 67.5 ng mL⁻¹ under the determined optimal conditions.

Table 2 Summary of results from analysis of colorants in croaker by μ-SPE method

Sample^a	Colorants contained in ingredients	Added (ng mL^−1)	TA			SY
			Found (ng mL^−1)	Recovery (%)	RSD (%)	Found (ng mL^−1)	Recovery (%)	RSD (%)
a Candy 1, corn flavored candy; candy 2, orange flavored candy; candy 3, green tea flavored candy.
Croaker	—	0.0	17.52	—	3.3	nd	—	—
		22.50	40.81	103.5	2.9	20.34	90.4	4.5
		67.50	83.68	98.0	5.6	63.59	94.2	7.8
Candy 1	Tartrazine	0.0	15.63	—	7.5	nd	—	—
		11.25	28.51	114.5	1.5	12.61	112.1	5.9
		45.00	69.01	118.6	6.3	55.80	124.0	7.7
Candy 2	Tartrazine and sunset yellow	0.0	52.52	—	2.2	4.68	—	5.2
		11.25	63.90	101.2	4.4	15.44	95.6	4.8
		45.00	93.73	91.6	6.6	52.26	105.7	5.9
Candy 3	Tartrazine and brilliant blue	0.0	23.15	—	4.4	nd	—	—
		11.25	35.40	108.9	1.4	9.77	86.8	3.9
		45.00	73.96	112.9	7.2	43.41	96.5	2.0

---

3.8 Comparison with other methods

The proposed method based on μ-SPE (Al₂O₃)-HPLC-DAD was compared with other methods as listed in Table 3. As could be seen from the table, the LODs of TA and SY obtained by the proposed method were lower than that obtained by SPE sample pretreatment methods. Besides, compared with other SPE methods, this method owned some other advantages. Firstly, the sorbent concentration of this method was two orders of magnitude lower than other SPE methods. In addition, the device of this method was simpler. Although the LODs of this method were higher than that obtained by LLE-HPLC, the cost of the proposed method is lower because the LLE methods involved in the use of ionic liquids which was very expensive.

Table 3 Comparison with other methods

Method^a	Sorbent^b	Sorbent mass	Solvent	LODs (ng mL^−1)	Ref.
a d-μ-SPE, dispersive solid-phase microextraction; CE, capillary electrophoresis; M-SPE, magnetic solid-phase extraction; d-LLE, dispersive liquid-phase extraction; μ-LLE, liquid-phase microextraction; IVMD, infinity variable wavelength detector; PMME, polymer monolith microextraction; MWD, multiple wavelength detector.b dASNPs, diamino moiety functionalized silica nanoparticles; NH2-LDC-MP, amino-functionalized low degrees of cross-linking magnetic polymer.
d-μ-SPE-CE	dASNPs	2.0 mg mL^−1	1.0 mL PBS	30–72	18
M-SPE-LC-MS	NH2-LDC-MP	10.0–20.0 mg mL^−1	—	0.45–1.25	10
d-LLE-HPLC-UV	—	—	[C8MIM][BF4]	0.015–0.15	7
μ-LLE-HPLC-IVWD	—	—	[C4MIM]Br	0.051–0.053	17
PMME-HPLC-MWD	Al2O3	87.5 mg mL^−1	15.0 mL NH3·H2O/ethanol	9.5–9.9	15
μ-SPE-HPLC-DAD	Al2O3	0.03 mg mL^−1	3.0 mL 1.0 M NaCl	0.10–0.20	This work

---

4. Conclusions

In the present work, we propose a micro-solid-phase extraction (μ-SPE) based on the dispersing of nano-Al₂O₃ on a filter membrane with HPLC to determine TA and SY in food samples. The present procedure could be applied to the analysis of large volumes of sample with a very low concentration of sorbent. Additionally, in the developed method, the device which could shorten analysis time by integrating the extraction and isolation process in one step was very simple. Moreover, during the desorption process, 0.2 M NaCl could desorb the colorants from the filter membrane without the need of using toxic organic solvent or high-cost ionic liquids. The entire process was thus efficient and economical. This developed μ-SPE method was demonstrated successfully for its practicality and applicability to the analysis of TA and SY in food samples and good recoveries were obtained. There is, therefore, great potential to further expand the applicability of the developed procedure.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (NSFC51178212), Liaoning Provincial Department of Education Innovation Team Projects (LT2012001), Liaoning Provincial Department of Education Innovation Team Projects (LT2015012), Shenyang Science and Technology Plan Project (F13-289-1-00), the Foundation of 211 Project for Innovative Talent Training, Liaoning University, and the Foundation for Young Scholars of Liaoning University (No. 2013LDQN13).

References

1. T. Gan, J. Sun, S. Cao, F. Gao, Y. Zhang and Y. Yang, Electrochim. Acta, 2012, 74, 151–157.
2. Y. Shen, X. Zhang, W. Prinyawiwatkul and Z. Xu, Food Chem., 2014, 157, 553–558.
3. Q. G. Liao, W. H. Li and L. G. Luo, Anal. Chim. Acta, 2012, 716, 128–132.
4. P. L. López-de-Alba, L. López-Martínez and L. M. de-León-Rodríguez, Electroanalysis, 2002, 14, 197–205.
5. T. Tanaka, Food Chem. Toxicol., 2006, 44, 179–187.
6. X. Ye, Y. Du, D. Lu and C. Wang, Anal. Chim. Acta, 2013, 779, 22–34.
7. H. Wu, J.-b. Guo, L.-m. Du, H. Tian, C.-x. Hao, Z.-f. Wang and J.-y. Wang, Food Chem., 2013, 141, 182–186.
8. F. Feng, Y. Zhao, W. Yong, L. Sun, G. Jiang and X. Chu, J. Chromatogr. B: Biomed. Sci. Appl., 2011, 879, 1813–1818.
9. M. Watanabe, M. Ishidate Jr and T. Nohmi, Mutat. Res., Environ. Mutagen. Relat. Subj., 1990, 234, 337–348.
10. X.-H. Chen, Y.-G. Zhao, H.-Y. Shen, L.-X. Zhou, S.-D. Pan and M.-C. Jin, J. Chromatogr. A, 2014, 1346, 123–128.
11. D. D. Stevenson, R. A. Simon, W. R. Lumry and D. A. Mathison, J. Allergy Clin. Immunol., 1986, 78, 182–191.
12. K. S. Rowe and K. J. Rowe, J. Pediatr., 1994, 125, 691–698.
13. F. I. de Andrade, M. I. Florindo Guedes, Í. G. Pinto Vieira, F. N. Pereira Mendes, P. A. Salmito Rodrigues, C. S. Costa Maia, M. M. Marques Ávila and L. de Matos Ribeiro, Food Chem., 2014, 157, 193–198.
14. Y. Ni, Y. Wang and S. Kokot, Talanta, 2009, 78, 432–441.
15. W.-J. Li, X. Zhou, S.-S. Tong and Q. Jia, Talanta, 2013, 105, 386–392.
16. L. Zhao, F. Zhao and B. Zeng, Electrochim. Acta, 2014, 115, 247–254.
17. O. Sha, X. Zhu, Y. Feng and W. Ma, Food Chem., 2015, 174, 380–386.
18. F.-J. Liu, C.-T. Liu, W. Li and A.-N. Tang, Talanta, 2015, 132, 366–372.
19. K. Ridgway, S. P. D. Lalljie and R. M. Smith, J. Chromatogr. A, 2007, 1153, 36–53.
20. X. Luo, Y. Zhan, X. Tu, Y. Huang, S. Luo and L. Yan, J. Chromatogr. A, 2011, 1218, 1115–1121.
21. P. Qi, T. Zeng, Z. Wen, X. Liang and X. Zhang, Food Chem., 2011, 125, 1462–1467.
22. H. Hu, M. Yang and J. Dang, Sep. Purif. Technol., 2005, 42, 129–136.
23. M. de los Santos, R. Batlle, J. Salafranca and C. Nerín, J. Chromatogr. A, 2005, 1064, 135–141.
24. W. Cao, S.-S. Hu, L.-H. Ye and J. Cao, J. Agric. Food Chem., 2014, 62, 9683–9689.
25. O. Sae-Khow and S. Mitra, J. Chromatogr. A, 2009, 1216, 2270–2274.
26. J. M. Jiménez-Soto, S. Cárdenas and M. Valcárcel, J. Chromatogr. A, 2012, 1245, 17–23.
27. E. S. Emídio, C. P. da Silva and M. R. R. de Marchi, J. Chromatogr. A, 2015, 1391, 1–8.
28. L. Wang, Z. Zhang, X. Xu, D. Zhang, F. Wang and L. Zhang, Talanta, 2015, 142, 97–103.
29. Q. Zhou, Y. Huang and G. Xie, J. Chromatogr. A, 2012, 1237, 24–29.
30. A. Bagheri, M. Taghizadeh, M. Behbahani, A. Akbar Asgharinezhad, M. Salarian, A. Dehghani, H. Ebrahimzadeh and M. M. Amini, Talanta, 2012, 99, 132–139.
31. B. Hu, M. He and B. Chen, Anal. Bioanal. Chem., 2015, 407, 2685–2710.
32. H. Huang, T. Chen, X. Liu and H. Ma, Anal. Chim. Acta, 2014, 852, 45–54.
33. M. Tuzen, K. O. Saygi and M. Soylak, J. Hazard. Mater., 2008, 152, 632–639.
34. M. Moazzen, R. Ahmadkhaniha, M. E. h. Gorji, M. Yunesian and N. Rastkari, Talanta, 2013, 115, 957–965.
35. Y. Huang, Q. Zhou and G. Xie, J. Hazard. Mater., 2011, 193, 82–89.
36. G. Wang, Y. Lei and H. Song, Talanta, 2015, 144, 369–374.
37. S.-H. Huo and X.-P. Yan, Analyst, 2012, 137, 3445–3451.
38. Y. S. Al-Degs, M. I. El-Barghouthi, A. H. El-Sheikh and G. M. Walker, Dyes Pigm., 2008, 77, 16–23.

---