Assessment of semi-permanent hair dyes in wash water from beauty salons by liquid chromatography-tandem mass spectrometry-selected reaction monitoring (LC-MS/MS-SRM)†

Jefferson Honorio Franco , Bianca F. da Silva and Maria V. B. Zanoni *
Institute of Chemistry, State University “Julio de Mesquita Filho” – UNESP, Avenida Professor Francisco Degni, 55, Quitandinha, 14800-900, Araraquara, SP, Brazil. E-mail: mvboldrin@gmail.com; Tel: +55-16-3301-9619

First published on 2nd October 2020

---

---

1 Introduction

Hair dyes represent an important aspect of cosmetology,¹ with a constant increase in consumption enabling outstanding developments of the hair dye industry, and resulting in an income of US$ 7.7 billion a year.²

The type of interaction and persistence of the dye on hair are responsible for their classification, divided into three groups: temporary, semi-permanent and permanent.¹ A semi-permanent hair dye is defined as a synthetic dye interacting in the hair shaft with moderate resistance. It is relatively more stable compared to temporary dyes, since the dye molecules adhere strongly to the outside of the strand,³ but less persistent than permanent hair dyes that penetrate the cuticle layer. Usually, the dyes have a chromophore group such as an azo group, anthraquinones, triphenylmethanates, nitrophenylenediamines, or nitroaminophenols and a charged group to interact by chemical forces with the amino acids present in the hair keratin.⁴

The production of 22 hair dyes has been banned since 2006,⁵ due mainly to the allergies caused by them, and their toxic, mutagenic and carcinogenic potential.^6,7 Hair dyes can be absorbed by the scalp after the dyeing process, since the hair is connected to the dermis, where the hair follicles are in contact with blood vessels.^8–10 Some of these dyes have tested positive for gene mutation in prokaryotic cells and they are capable of inducing frameshift mutations, based on the results with Salmonella typhimurium in the presence of reductive metabolic activation.¹¹ Semi-permanent hair dyes are usually applied directly in hair dye formulations in high concentrations up to 20 mg mL⁻¹ and form aromatic amines that could be toxic, mutagenic or carcinogenic after metabolization mainly in semi-permanent hair dyes bearing azo groups as chromophores.^11–15

Occupational skin and respiratory disorders and disputable reproductive and genotoxic effects have been linked to chemical exposure of beauty workers.^16,17 Salon wastewater can be a potential public and environmental health hazard.¹⁸ The literature reports that most physicochemical parameters (biochemical oxygen demand, dissolved oxygen, total dissolved solids, and chemical oxygen demand levels) and the major pathogens Pseudomonas aeruginosa and Staphylococcus aureus are over the World Health Organization's regulatory limits.¹⁹ The determination of semi-permanent hair dyes by high performance liquid chromatography (HPLC) coupled with diode array detection (DAD) has been published for their determination in hair coloring formulations by using 1-butyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide as an ionic liquid.²⁰ However, the method is not sensitive enough to determine very low levels of dyes diluted in wastewater from hair salons. Thus, there is a lack of studies dealing with effective analytical methods to identify the major source of this wastewater pollution or methods to treat hair dye effluents.

The wastewater from hair salons is very complex due to the great diversity of chemicals and types of dyes present, and its treatment is difficult.^21,22 To the best of the authors' knowledge, no convenient and efficient method or product for removing semi-permanent coloring from hair is available or known. Taking into account that the presence of hair dyes in the environment poses serious risks to human health and aquatic biota²³ and that the ingestion of dyes based on azo groups by improper wastewater treatment can generate metabolites more toxic than the original dye,^24,25 studies dealing with their presence in wastewater and wash samples containing the dye lost from the dyed hair are important to understand how they are released into the environment.

The present work investigates the use of Liquid Chromatography tandem Mass Spectrometry (LC-MS/MS) to detect semi-permanent hair dyes in commercial formulations. In the literature, important and widely used analytical methods, such as HPLC-DAD,²⁵ HPLC coupled with chemiluminescence detection,²⁶ HPLC coupled with electrochemical detection,²⁷ capillary electrophoresis,²⁸ and inductively coupled plasma-mass spectrometry,²⁹ have been described for the identification and quantification of permanent hair dyes in environmental samples,^30,31 as well as aromatic amines released from azo dyes.^{24,25,32–34} Additionally, liquid chromatography-tandem mass spectrometry (LC-MS/MS) has shown high versatility, specificity and selectivity in the analysis of hair dyes in real samples, thus achieving popularity as a fundamental analytical technique for the analysis of compounds in complex environmental and wash water samples.^{14,15,20,33,34} The SRM (selected reaction monitoring) assay is an excellent method of the LC-MS technique, which has high sensitivity and reproducibility in the quantification of dyes, pharmaceutical products, specific peptides in complex mixtures, proteins, and human growth hormones.^32,35–37 There are no studies in the literature related to hair dyes using the SRM method. Therefore, LC-MS/MS-SRM has been successfully applied in several studies involving hair samples as the procedure of choice. The detection, identification and quantification of anabolic steroid residues,³⁸ nicotine,³⁹ and drugs of abuse,⁴⁰ such as cocaine, morphine, amphetamine, and methamphetamine,⁴¹ in hair samples have been developed using the SRM method. Thus, it is of utmost interest, from both economical and environmental perspectives, to quantitatively examine hair dyes to determine their fixation in hair or their release in wash water.

Herein, the aim of the present work is to validate an analytical method for detection, identification and quantification of some selected semi-permanent hair dyes: Basic Blue 99 (BB 99), Basic Brown 16 (BB 16), Basic Red 76 (BR 76), Basic Yellow 57 (BY 57), and Acid Violet 43 (AV43) with different chromophoric groups and charged functional groups, widely used in semi-permanent hair dyes. In order to obtain more details about lixiviation of semi-permanent hair dyes during the washing process, LC-MS/MS-SRM was applied here to monitor the loss of hair dyes during washing in a beauty salon and trough to the sewage.

2 Materials and methods

2.1 Reagents and solutions

All five semi-permanent hair dyes were purchased from LCW Dyes, Arianor (São Paulo, Brazil) as individual standards. The purity of each individual standard dye based on HPLC assays was as follows: 58–70% BB 99, > 97% BB 16, > 98% BR 76, > 98% BY 57 and > 54% AV 43.

The commercial sample HF 65 from Arianor (120 g) is a hair dye product which contains water, algae extract, sorbitol, benzyl alcohol, EDTA, citric acid, triethanolamine, 1,2-propanediol, glucose, hydroxyethyl cellulose, hydroxypropyltrimonium chloride, diazolidinyl urea, cocamidopropyl betaine, Basic Brown 17, Blue HC 2, BB 99, BB 16, BR 76, BY 57, and AV 43. For analytical purposes, solutions of the standard hair dyes (1.0 mg mL⁻¹) were prepared, both individually and as a mixture, by direct dissolution in high-purity water (Milli-Q system; Millipore, São Paulo, Brazil). They were identified and quantified by the LC-MS/MS-SRM chromatographic technique. The chromatographic eluents acetonitrile (HPLC grade) and formic acid were purchased from Merck and the Strata-X cartridge was purchased from Phenomenex. All of the solutions were prepared with high-purity water from a Millipore Milli-Q system (Millipore, São Paulo, Brazil).

2.2 LC-MS/MS-SRM analysis

The validation methods of the standard hair dyes were analyzed by LC-MS. The concentration of the standard hair dyes used in the method was 1.0 mg mL⁻¹.

The liquid chromatographic (LC) analyses were performed using a 1200 Agilent Technologies HPLC (Palo Alto, CA, USA) and a Luna C18 column (250 mm x 4.6 mm; 5 μm) from Phenomenex. A mixture of water and acetonitrile (20:80, v/v) both containing 0.1% formic acid in isocratic mode at 1 mL min⁻¹ was employed.

The MS data acquisition was performed in full scan mode (EMS, enhanced mass scan – using a linear ion trap mass analyzer) using a 3200 QTRAP mass spectrometer (linear ion trap quadrupole mass spectrometer, AB SCIEX (Framingham, MA, USA)) with an auto injector, quaternary pump, and column oven from Agilent 1200. The QTRAP was operated under optimized conditions: gas 1 50 psi, gas 2 50 psi, collision gas 12 psi and curtain gas 30 psi. The gas employed for the analysis was nitrogen. All measurements were carried out in the negative and positive ion mode by using a TurboIonSpray ionization (ESI) source with a capillary potential of 5500 V and source temperature of 550 °C. Table 1 shows the declustering potential, entrance potential and collision energy for each analysed dye.

---

The occurrence of the hair dyes in the wash water of a beauty salon was proven analyzing the collected wash water before and after the application of the commercial dye product, HF 65 Arianor, by LC-MS/MS-SRM. SRM (selected reaction monitoring) experiments were performed to selectively detect and quantify each hair dye in the wash water (real samples). SRM was used to monitor a specific transition with high sensitivity. In this case, the first analyzer is configured to select ions with a specific mass. The collision energy is optimized to produce the fragment that will be diagnosed on the second analyzer. Quantitative analysis by LC-MS/MS-SRM is a strategy employed by using the two transitions (m/z transitions between the precursor ions and their product ions generated) of each compound analyzed. This is used to confirm the correct identification of the target analytes in real samples with higher sensitivity and selectivity. The elution method used was in gradient mode using the following conditions: from 0 to 0.50 min – 10% ACN, from 0.50 to 15.0 min – 10–70% ACN, from 15.0 to 16.0 min – 70% ACN, from 16.0 to 17.0 min – 70–10% ACN, followed by 3 min conditioning. The solvents varied according to the ionization mode. The gradient elution with 0.01% formic acid in water was used in the positive mode while the gradient elution with 0.01 mM ammonium formate in acetonitrile was used in the negative mode.

The injection volume and temperature of the column were 20 μL and 25 °C, respectively. The flow rate used was 800 μL min⁻¹. The total analysis time was 10 min. All measurements were performed in triplicate (n = 3).

2.3 Analytical curves

The stock solution (c = 1.0 mg mL⁻¹) of each of the standard hair dyes was analyzed by LC-MS/MS to confirm their precursor ions and the corresponding fragment ions. Analytical calibration curves were obtained for each of the five hair dyes (BB 99, BB 16, BR 76, BY 57, and AV 43) individually. Thus, 11 different concentrations (0.5, 1.0, 2.0, 5.0, 10, 20, 50, 100, 200, 500, and 1000 ng mL⁻¹) were prepared by successive dilution of the stock solution for each standard hair dye.

The analytical calibration curves were constructed by LC-MS/MS-SRM taking into consideration the peak area ratios, which were plotted against the concentration of hair dyes. The concentration of each hair dye in the wash water sample was obtained by linear regression of the analytical curve and confirmed by the standard addition method for each isolated dye. All measurements were performed in triplicate (n = 3).

The lowest level of the analyte in the sample solution when providing a measurable response at a 3:1 signal-to-noise (S/N) ratio corresponds to the detection limit (LOD).⁴² Meanwhile, the lowest analyte concentration confirmed by an accurately quantifiable response, at S/N equal to 10:1, corresponds to the limit of quantification (LOQ).

2.4 Strata-X solid phase extraction (SPE)

Before performing the LC-MS/MS analysis, the collected dye wash water was submitted to a clean-up procedure by vacuum filtration (using a vacuum manifold processing station) with a 0.22 μm PES membrane filter (HCS Scientific & Chemical Pte Ltd, Singapore). To remove the other substances present in the commercial HF 65 hair dye product, the SPE (Solid Phase Extraction) clean-up procedure was applied using Strata-X cartridges with a polymeric reversed-phase sorbent (33 μm, 200 mg/3 mL; Phenomenex, USA). The Strata-X cartridges were conditioned with 3 mL of methanol, followed by 3 mL of sample. After washing, the elution was carried out with 1 mL of methanol and acetonitrile (50:50 v/v) both containing 0.1% acetic acid. The SPE product was dried with a nitrogen flow and resuspended in a 10 mL flask with ultrapure Milli-Q water and then analyzed in triplicate.

2.5 Method validation for the commercial hair dye

1.00 g of the commercial dye HF 65 was weighed and diluted with 10 mL of water and subjected to solid phase extraction by using a Strata-X cartridge. After this, we added selected concentrations of the standard hair dyes into the solution. The results of the fortified sample were evaluated by comparing the chromatographic peak area value with the linear equation obtained from the calibration curve of the standard hair dyes. Thus, it was possible to calculate the concentration of each hair dye in the fortified solution, and determine the accuracy and percentage of recovery of the analytical method by SPE.

All system performance calculations were based on three replicate injections.

2.6 Analysis of wash water from beauty salons

The hair dye paste, prepared following the manufacturer's recommendations, was applied on the hair and then evenly distributed by massage. After 30 minutes, the hair was washed with water until the excess of the hair dye paste was completely removed. The monitoring of hair dyes lost in each wash was done by collecting 2 L of dye wash water from each hair wash in the 5 days of successive washing. These samples of dye wash water were stored in a dark glass flask and maintained at a temperature of −5 °C. All procedures were performed in triplicate.

3 Results and discussion

3.1 Analysis of the selected standard semi-permanent hair dyes by LC-MS/MS-SRM

From the LC-MS/MS-SRM analysis of the selected standard semi-permanent hair dyes, the LC-MS/MS chromatographic profiles obtained in EMS mode showed for the standard hair dyes the following retention times and m/z values corresponding to the protonated species [M + H]⁺: 6.41 min and m/z 415.0 for BB 99 (Fig. 1A), 8.46 min and m/z 321.3 for BB 16 (Fig. 1B), 9.04 min and m/z 336.4 for BR 76 (Fig. 1C), 9.38 min and m/z 336.3 for BY 57 (Fig. 1D), and 9.99 min and m/z 408.3 for AV 43 (Fig. 1E). All analysis parameters of the standard hair dyes, including retention times, molecular mass, precursor ions (m/z) and fragment ions (m/z), are summarized in Table 1. In order to obtain the quantification and confirmation of each dye, individual infusion of the standard hair dyes was carried out in the mass spectrometer equipped with an electrospray ionization source with scan acquisition mode SRM. Since one of the objectives of this work was to develop a reliable analysis method, the best chromatographic separation of hair dyes was required in order to (1) avoid a significant loss in detectability and (2) obtain the minimum number of recommended points for each chromatographic peak as a function of time for monitoring each transition. Thus, an analytical calibration curve was constructed with the values obtained from LC-MS/MS-SRM for the quantification of all standard hair dyes.

Table 2 shows the characteristics of the analytical calibration curve of each analyzed hair dye, such as the linear equation (y = ax + b) and the correlation coefficient, as well as the limit of detection and the limit of quantification. The excellent sensitivity of the analytical calibration curve was confirmed by the value of the correlation coefficient, which was close to one. The results outlined in Table 2 show that the BB 99 dye has the highest values of detection and quantitation limits in relation to the other dyes analyzed, while the AV 43 dye presents the lowest values. Through the results of the calibration curve with standard hair dyes it is possible to determine the amounts of these dyes in the hair wash waters after dyeing.

---

3.2 Determination of hair dyes in the commercial formulation and validation method by SPE

Firstly, we calculated the percentage of dye recovery in the commercial formulation HF 65 fortified with BB 99, BB 16, BR 76, BY 57, and AV 43 dyes subjected to solid phase extraction with Strata-X cartridges. The recoveries ranged from 79% to 104%, which confirms the outstanding accuracy of the proposed extraction method (Table 3), validating the applicability of the method to the analysis of dyes in commercial hair dye formulations.

---

We employed the method developed to analyze the mass and percentage of each hair dye in the commercial formulation HF 65. Through the results obtained in Table 4, it was proven that the dyes described on the product label are really present in the commercial formulation. The amount of each hair dye varied from 2.0% to 32% (w/w) of the total product composition (120 g). Remarkably, the method allows the determination of all the hair dyes of the present work, constituting a simple methodology with useful application in the development of analytical methods for dye validation.

---

3.3 Determination of hair dyes lost in the wash water in the 5 successive washing days

The LC-MS/MS approach was fully employed to detect the hair dyes BB 99, BB 16, BR 76, BY 57 and AV 43 in wash water from the beauty salon after the use of the commercial hair dye formulation HF 65 Arianor.

Despite the matrix effect, very common in the analysis of real samples, the retention time of the chromatographic peak obtained from the wash water samples was very close to the retention time obtained after the addition of the analytical standard of each hair dye, confirming that the chromatographic peaks identified in the wash water were actually from the released hair dyes.

The LC-MS chromatographic profile was obtained in the EMS mode (scanning ions) for hair wash water collected on the first (Fig. 2A), second (Fig. 2B), third (Fig. 2C), and fourth (Fig. 2D) day after dyeing hair. No dye was detected in the hair washing water on the fifth day by this method. BY 57 dye was detected and identified in the first four days in the wash water as m/z 336.4 ([M + H]⁺) (Fig. 2a₃, b₃, c₂ and d₁). Meanwhile, BB 99 was detected in the first three days in the hair washing water as m/z 417.1 ([M + H]⁺) (Fig. 2a₁, b₁ and c₁). During the five days of analysis, BB 99 and BY 57 were the most detected dyes in the wash water compared to the other dyes studied. In addition, BB 16 was the third most abundant dye detected in hair washing waters as m/z 321.3 ([M + H]⁺) (Fig. 2a₂ and b₂).

Based on the obtained results, it can be stated that BR 76 and AV 43 dyes adhere more strongly to the cuticle of the hair, requiring a greater number of washes to achieve their release, as confirmed by them not being detected in the dye-wash water during the five days of analysis. On the other hand, BB 99, BB 16 and BB 57 showed low fixation on the scalp and hair, since they were detected from the first day of washing. The other observed bands confirm the presence of other compounds, which may have resulted from different dyes or substances present in the commercial hair dye HF 65 Arianor. It is important to note that the retention time of each hair dye found in the wash water is slightly different from the retention time of its corresponding standard (Table 1) due to the presence of other components in the commercial hair dye HF 65 which cause a small displacement of the retention times of these hair dyes in the real samples compared to the values shown by the standard dyes separately. The dye wash water samples daily collected for 5 successive washings on different days were analyzed by LC-MS/MS and the results are shown in Table 5.

---

Although some of the dyes were detected in the wash water, a more effective method was necessary for the quantification of these dyes in the real sample. Therefore, upon detection of hair dyes in the dye wash water, their quantification was performed by LC-MS/MS-SRM (Fig. S1†) in order to quantify the hair dyes in each sample of the dye wash water during the five days of analysis. Thus, other hair dyes contained in the commercial dye HF 65 could be identified with this new method proposed.

Table 6 indicates the concentration of each of the hair dyes in wash water after washing hair during five successive days of analysis. The results reveal that all dyes were being released gradually as the hair was washed. BB 99 presented the highest concentration in the wash water both on the first day, 866.20 ng mL⁻¹, and last day of analysis, 128.60 ng mL⁻¹, indicating a low fixation of this dye on the hair and scalp. BY 57 also showed high concentration on the first day in the wash water, 145.10 ng mL⁻¹, indicating its low fixation on the hair and scalp. The results indicated that BB 99 and BY 57 release the highest dye concentration during the first day of washing. BB 16, BR 76, BY 57 and AV 43 were not detected on the last day of analysis. In addition, BB16, BR 76 and AV 43 dyes were not detected on the third and fourth days of analysis in the wash water, only on the first two days of analysis, possibly due to their greater fixation to the hair. These results highlight that the hair dyes have different fixing power to the scalp and hair. Even though their concentration in the dye wash water is low (ng mL⁻¹), the sewage receives both the hair dye residues and some of the dyes superficially adhered to the scalp and hair.

---

The difficulty of achieving a satisfactory hair coloring has led consumers to prefer semi-permanent hair coloring instead of permanent hair coloring products. Another positive point is the ease of removing the dye, since semi-permanent hair dyes can be removed by shampooing the hair 6–8 times.^8,9 According to IARC,⁴³ BB 99, BB 16, BB 17, BR 76, BY 57, and AV 43 are among the most produced dyes worldwide by the dyestuff industry, ranging from 1 to 10 tons annually,⁴³ which are then used to prepare the commercial products utilized in beauty salons.

The presented results underline that the hair dye industry should be cautious about the selection of semi-permanent hair dyes utilized in their commercial formulations. Specifically, focus should be placed on rethinking commercial dyes with the aim of replacing those that are not well fixed on the scalp and hair with more efficient dyes in the dyeing process. Accordingly, a search for easily biodegradable hair dyes in the environment is required in order to completely avoid or minimize contact with humans. Rigorous legislation should be applied to improve the monitoring of dyes in commercial formulations, since recent results have indicated that dye contamination has severely affected environmental safety, especially aquatic environments.⁴⁴

4 Conclusions

The LC-MS/MS-SRM technique has proven to be an important tool to characterize, identify and quantify the semi-permanent hair dyes in the dye wash water that characterizes the effluents of beauty salons, even when they are present in low concentrations.

The analytical method provided good linearity and satisfactory quantification and detection limits at low concentrations for all five semi-permanent hair dyes. These results confirm the outstanding sensitivity obtained for the SRM method for analysing commercial formulation Arianor HF 65 in wash waters.

Hair dye wash water samples were collected for five successive days and analyzed by LC-MS/MS-SRM according to the developed method. The results indicated that some of the analyzed hair dyes are gradually released during hair washing. The method is simple and economical and could potentially be applied in the monitoring of dye wash waters.

Through the results obtained by LC-MS/MS-SRM associated with the negative effects after exposure to semi-permanent hair dyes, it can be concluded that extreme caution should be required regarding their use, especially in beauty salons and dedicated new regulation should be put in effect in order to avoid serious risks to human health and the environment.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors are grateful for the financial support from CNPq, CAPES and FAPESP (2008/10449-7 and 2015/18109-4).

References

1. R. G. Harry, Harry's Cosmeticology, ed. J. B. Wilkinson and R. J. Moore, Chemical Publishing Company, Inc., New York, 7th edn, 1982, p. 521 Search PubMed .
2. K.-H. Kim, E. Kabir and S. A. Jahan, Environ. Int., 2016, 89–90, 222–227 CrossRef CAS .
3. V. Andrisano, R. Gotti, P. Roveri and V. Cavrini, Chromatographia, 1997, 44, 431–437 CrossRef CAS .
4. M. G. De Navarre, The Chemistry and Manufacture of Cosmetics, Van Nostrand, Princeton, N. J., 2 nd edn, 1962 Search PubMed .
5. International Agency for Research on Cancer (IARC), IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Human Papillomaviruses, Lyon, France, 2007, vol. 90 Search PubMed .
6. K. Czene, S. Tiikkaja and K. Hemminki, Int. J. Cancer, 2003, 105, 108–112 CrossRef CAS .
7. M. Kujawa, Food/Nahrung, 1994, 38, 351 CrossRef .
8. J. Lademann, H. Richter, U. Jacobi, A. Patzelt, F. Hueber-Becker, C. Ribaud, F. Benech-Kieffer, E. K. Dufour, W. Sterry, H. Schaefer, J. Leclaire, H. Toutain and G. J. Nohynek, Food Chem. Toxicol., 2008, 46, 2214–2223 CrossRef CAS .
9. T. B. Zanoni, M. Tiago, F. Faião-Flores, S. B. de Moraes Barros, A. Bast, G. Hageman, D. P. de Oliveira and S. S. Maria-Engler, Toxicol. Lett., 2014, 227, 139–149 CrossRef CAS .
10. H. Liu, J. Chen, J. Jiang, J. P. Giesy, H. Yu and X. Wang, Environ. Toxicol. Pharmacol., 2008, 26, 309–314 CrossRef CAS .
11. SCCS (Scientific Committee on Consumer Safety), Opinion on Basic Red 51 (COLIPA no. B116), 20 September 2011, (SCCS/1436/11).
12. T. B. Zanoni, T. M. Lizier, M. das D. Assis, M. V. B. Zanoni and D. P. de Oliveira, Food Chem. Toxicol., 2013, 57, 217–226 CrossRef CAS .
13. G. J. Nohynek, R. Fautz, F. Benech-Kieffer and H. Toutain, Food Chem. Toxicol., 2004, 42, 517–543 CrossRef CAS .
14. J. C. de Souza, B. F. da Silva, D. A. Morales, G. de A. Umbuzeiro and M. V. B. Zanoni, J. Hazard. Mater., 2020, 387, 122000 CrossRef .
15. J. C. de Souza, B. F. da Silva, D. A. Morales, G. de A. Umbuzeiro and M. V. B. Zanoni, Sci. Total Environ., 2019, 685, 911–922 CrossRef .
16. T. Leino, L. Tammilehto, M. Hytönen, E. Sala, E. Sala, H. Paakkulainen and L. Kanerva, Scand. J. Work. Environ. Health, 1998, 24, 398–406 CrossRef CAS .
17. M. P. Galiotte, P. Kohler, G. Mussi and G. J. F. Gattás, Ann. Occup. Hyg., 2008, 52, 645–651 CAS .
18. M. A. Nkansah, F. Opoku, J. H. Ephraim, D. D. Wemegah and L. P. M. Tetteh, Environ. Health Insights, 2016, 10, 147–154 Search PubMed .
19. World Health Organization, Guidelines for Drinking-Water Quality: Recommendations, World Health Organization, Geneva, 2004, vol. 1 Search PubMed .
20. J. H. Franco, B. F. Silva and M. V. B. Zanoni, Anal. Methods, 2015, 7, 1115–1122 RSC .
21. American Public Health Association, Standard Methods for the Examination of Water & Wastewater, American Public Health Association, 2005, vol 21 Search PubMed .
22. J. H. Franco, A. B. Aissa, G. G. Bessegato, L. M. Fajardo, M. V. B. Zanoni, M. I. Pividori and M. Del, Pilar Taboada Sotomayor, Environ. Sci. Pollut. Res., 2017, 24, 4134–4143 CrossRef CAS .
23. D. M. Lewis, Rev. Prog. Color. Relat. Top., 1999, 29, 23–28 CrossRef CAS .
24. J. H. Franco, B. F. da Silva, R. V. Oliveira, G. Meireles, D. P. de Oliveira, A. A. de Castro, T. C. Ramalho and M. V. B. Zanoni, Sci. Total Environ., 2018, 613–614, 1093–1103 CrossRef CAS .
25. J. H. Franco, B. F. da Silva, A. A. de Castro, T. C. Ramalho, M. I. Pividori and M. V. B. Zanoni, Environ. Pollut., 2018, 242, 863–871 CrossRef CAS .
26. J. Zhou, H. Xu, G.-H. Wan, C.-F. Duan and H. Cui, Talanta, 2004, 64, 467–477 CrossRef CAS .
27. T. M. Lizier and M. V. Boldrin Zanoni, Molecules, 2012, 17, 7961–7979 CrossRef CAS .
28. Y. Masukawa, J. Chromatogr. A, 2006, 1108, 140–144 CrossRef CAS .
29. M. A. Gondal, Y. W. Maganda, M. A. Dastageer, F. F. Al Adel, A. A. Naqvi and T. F. Qahtan, Appl. Opt., 2014, 53, 1636–1643 CrossRef CAS .
30. M. Shelke, S. K. Sanghi, A. Asthana, S. Lamba and M. Sharma, J. Chromatogr. A, 2005, 1089, 52–58 CrossRef CAS .
31. M. Akyüz, Talanta, 2007, 71, 486–492 CrossRef .
32. J. H. Franco, B. F. da Silva, E. F. G. Dias, A. A. de Castro, T. C. Ramalho and M. V. B. Zanoni, Ecotoxicol. Environ. Saf., 2018, 160, 114–126 CrossRef CAS .
33. L.-H. Ahlström, J. Raab and L. Mathiasson, Anal. Chim. Acta, 2005, 552, 76–80 CrossRef .
34. M. Bhaskar, A. Gnanamani, R. J. Ganeshjeevan, R. Chandrasekar, S. Sadulla and G. Radhakrishnan, J. Chromatogr. A, 2003, 1018, 117–123 CrossRef CAS .
35. D. B. Barr, J. R. Barr, V. L. Maggio, R. D. Whitehead Jr, M. A. Sadowski, R. M. Whyatt and L. L. Needham, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2002, 778, 99–111 CrossRef CAS .
36. M. A. Aramendía, V. Borau, F. Lafont, A. Marinas, J. M. Marinas, J. M. Moreno, J. M. Porras and F. J. Urbano, Food Chem., 2006, 97, 181–188 CrossRef .
37. J. L. Little, M. F. Wempe and C. M. Buchanan, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2006, 833, 219–230 CrossRef CAS .
38. L. Rambaud, F. Monteau, Y. Deceuninck, E. Bichon, F. André and B. Le Bizec, Anal. Chim. Acta, 2007, 586, 93–104 CrossRef CAS .
39. L. Cashman and J. Nutt, Toxicol. Anal. et Clin., 2019, 31, S83 Search PubMed .
40. D. Di Corcia, F. D'Urso, E. Gerace, A. Salomone and M. Vincenti, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2012, 899, 154–159 CrossRef CAS .
41. H. Miyaguchi, M. Kakuta, Y. T. Iwata, H. Matsuda, H. Tazawa, H. Kimura and H. Inoue, J. Chromatogr. A, 2007, 1163, 43–48 CrossRef CAS .
42. L. R. Snyder and J. J. Kirkland, Practical HPLC Method Development, John Wiley & Sons, Inc., New York, 2nd edn, 1997 Search PubMed .
43. International Agency for Research on Cancer (IARC), IARC monographs on the evaluation of carcinogenic risks to humans, Occupational Exposures of Hairdressers and Barbers and Personal Use of Hair Colourants; Some Hair Dyes, Cosmetic Colourants, Industrial Dyestuffs and Aromatics Amines, Lyon, France, 1993, vol. 57 Search PubMed .
44. A. Tkaczyk, K. Mitrowska and A. Posyniak, Sci. Total Environ., 2020, 717, 137222 CrossRef CAS .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d0ay01395a

This journal is © The Royal Society of Chemistry 2020