- Email: [email protected]
Simple and ultrasensitive microplate method for spectrofluorimetric determination of trace resorcinol
Microchemical Journal 122 (2015) 5–9
Contents lists available at ScienceDirect
Microchemical Journal journal homepage: www.elsevier.com/locate/microc
Simple and ultrasensitive microplate method for spectrofluorimetric determination of trace resorcinol Fabien Robert-Peillard ⁎, Claire Chottier, Bruno Coulomb, Jean-Luc Boudenne Aix-Marseille Université, CNRS, LCE FRE 3416, 13331 Marseille cedex 3, France
a r t i c l e
i n f o
Article history: Received 27 February 2015 Received in revised form 1 April 2015 Accepted 2 April 2015 Available online 9 April 2015 Keywords: Resorcinol Spectrofluorimetric Microplate Coumarin Wastewater Cosmetics
a b s t r a c t In this paper, a new spectrofluorimetric method is proposed for the fast, simple, and accurate determination of trace resorcinol in the microplate format (high-throughput screening). The analytical method is based on the formation of a coumarin derivative with methyl acetoacetate via a Pechmann condensation reaction. Experimental conditions have been optimized, and excellent analytical performances have been achieved with a limit of detection of 0.46 μg L−1, a wide linear range between 1.5 and 1000 μg L−1 and a relative standard deviation of 2.01% (n = 10). This method is remarkably specific for resorcinol compared to other phenols and hydroxyphenols and suffers from low interferences from other potential matrix components. The proposed protocol was finally applied and validated on real samples such as wastewater or pharmaceutical/cosmetic samples. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Phenolic compounds are a class of organic contaminants that are of great concern for environmental or human health issues. Several spectroscopic methods are available for the determination of total phenolic compounds, including Prussian Blue, o-phenanthroline, or 4aminoantipyrine [1,2]. Nevertheless, methods for the individual determination of these compounds can be very valuable tools to study or understand their environmental behavior, evaluate their toxic effect, or quantify a specific compound in different matrices. Among these phenolic compounds, resorcinol (benzene-1,3-diol) has received special attention due to its toxicity and significant occurrence in the environment. Resorcinol has indeed natural sources, being a monomeric by-product of the reduction, oxidation, and microbial degradation of humic substances [3]. It is thus often found in natural waters used for tap water production, contributing to trihalomethane potential formation when these waters are chlorinated [4,5]. Apart from these natural origins, resorcinol is produced worldwide at increasing levels (from 40 000 tons in 1990 to 45 000 tons in 2000) [6] and is used in various industrial applications (rubber industry, adhesives, dyes, speciality chemicals, pharmaceuticals) [7]. Resorcinol is a proven endocrine disruptor and has been listed as a category 1 substance (substance of high concern) in the European Union priority list of substances for further evaluation of their role in endocrine disruption [8]. Toxicity for human health include dermatitis, ⁎ Corresponding author. Tel.: +33 413551030; fax: +33 413551060. E-mail address: [email protected] (F. Robert-Peillard).
http://dx.doi.org/10.1016/j.microc.2015.04.004 0026-265X/© 2015 Elsevier B.V. All rights reserved.
catarrh, convulsion, or cyanopathy, and strict limitations has been set for resorcinol concentration in pharmaceutical products such as shampoos and hair lotions (0.5% limit) [9]. Serious implications of the deleterious impacts of resorcinol exposure on human health and the corresponding emergency measures to be adopted have been discussed in a document by World Health Organization (WHO) [10]. Moreover, it has been proven that resorcinol has the lowest biodegradation rates with aerobic biodegradation processes among all dihydroxybenzene isomers, resulting in increased threats for final environmental impacts [11]. Therefore, development of simple analytical methods for specific resorcinol determination has a great interest for several types of applications, such as industrial waters, wastewaters, and hair lotions. The most common and reliable methods for the determination of resorcinol are high-performance liquid chromatography (HPLC) [12–14] and gas chromatography [15]. These methods are efficient but require expensive instruments; therefore, cheaper and simpler alternatives are highly desirable. In recent years, several methods have been described for the alternative determination of resorcinol in different samples, such as electrochemistry [16,17], spectrophotometry with nanoparticles [18] or first-derivative methods [19], direct spectrofluorimetry [20] or inhibition-based spectrofluorimetry [21,22]. Although all these methods have interesting properties for detection of resorcinol, they all suffer from relatively low sensitivity and high limit of detection of about 0.1 to 1 mg L−1, a narrow linear dynamic range (less than 2 orders of magnitude) and poor or undefined specificity towards other phenols and other hydroxyphenols isomers. Thus, the development of a simple, specific, ultrasensitive, and fast method for the detection of resorcinol still remains a great challenge.
6
F. Robert-Peillard et al. / Microchemical Journal 122 (2015) 5–9
In this work, a very simple and sensitive spectrofluorimetric method is described. The method is based on the reaction of resorcinol with a β-keto ester in acidic medium which yields a fluorescent coumarin (Pechmann condensation). Fluorescence is measured after 5 min at an excitation and emission wavelength of 332 nm and 431 nm, respectively. The simplicity of this protocol enables its implementation in microplate wells, providing high-throughput analysis with low volumes of samples, reagents, and waste. Development of the global procedure and application on real wastewater and pharmaceutical/cosmetic products are presented herein. 2. Experimental 2.1. Reagents and solutions All chemicals were of analytical reagent grade and used without further purification. Resorcinol, methyl acetoacetate, and ethyl acetoacetate were purchased from Alfa Aesar (Schiltigheim, France), catechol, hydroquinone, phloroglucinol, phenol, 4-nitrophenol, 2-chlorophenol, and ascorbic acid were from Sigma-Aldrich (SaintQuentin Fallavier, France), Suprapur sulfuric acid from Merck (Darmstadt, Germany). Resorcinol and other phenolic compounds stock standard solutions and working solutions were prepared in ultrapure water (Millipore, resistivity N 18 MΩ cm).
3. Results and discussion 3.1. Formation of the fluorescent product Pechmann condensation is a well-known organic reaction used for synthesis of various coumarins. Reaction scheme with resorcinol is depicted on Fig. 1, with formation of 7-hydroxy-4-methylcoumarin (4-methylumbelliferone). Application of this reaction for analytical purposes appeared very interesting to us since the formation of highly fluorescent products from non- or poorly fluorescent analytes/reagents is a method of choice for attaining high sensitivity. The selection of optimum excitation and emission wavelengths was carried out on the fluorescent product formed under the reaction conditions. 7-Hydroxy-4-methylcoumarin is a well-known fluorophore with reported optimum excitation and emission wavelengths of 340–365 nm and 430–450 nm, respectively [23]. For optimum fluorescent measurements in black microplates, one has to investigate not only the maximum of intensity for a given standard, but also the intensity of a blank sample (with no standard, which takes also into account the autofluorescence of the microplate material). Fig. 2 displays the best set of wavelengths by measuring the optimum signal/blank ratio, with final wavelengths set at 332 nm and 431 nm for excitation and emission, respectively. 3.2. Optimization of experimental conditions
2.2. Microplate instrumentation Microplate fluorescence measurements were carried out on a microplate reader (Infinite M200, Tecan France SAS, Lyon, France) equipped with an excitation and emission double monochromator (bandwidths of 9 nm and 20 nm for excitation and emission monochromator, respectively) and controlled by i-control™ software (Tecan). Fluorescence detection was performed from above the microplate wells (top configuration), at λex = 332 nm and λem = 431 nm. Operating temperature was set to 25 °C. Other parameters were as follows: gain: 90; number of flashes: 25; integration time: 20 μs. Fluorescence intensities were expressed in arbitrary units (a.u.). Polystyrene black 96 flat-bottom well microplates (Fisher Scientific, Illkirch, France), with a maximum capacity of 375 μL for each wells were used. In order to avoid degradation of the microplate reader due to sulfuric acid use, microplates were shaken on an external vortex mixer (Vortex Genie 2, Scientific Industries, New-York, USA), placed in the microplate reader only for fluorescent measurements and then quickly removed and stored under a fume hood. 2.3. Optimized protocol for resorcinol determination in aqueous samples Seventy-five microliters of sample or resorcinol standard solution was dispensed into the wells of the microplate, followed by 75 μL of sulfuric acid and 10 μL of methyl acetoacetate. The plate was shaken for 5 min at room temperature on the vortex mixer, and fluorescence was subsequently measured at λex = 332 nm and λem = 431 nm in the microplate reader. Resorcinol concentrations were determined using the linear calibration curve obtained with standards. All experiments were performed in duplicate.
3.2.1. Sulfuric acid concentration Classical conditions for coumarin formation via Pechmann condensation require sulfuric acid to reach good yields in a reasonable time. Many alternatives have been developed to try to avoid the use of strong acid. Catalysts such as ZrOCl2.8H2O/SiO2 [24], zeolite [25] or boron trifluoride [26] have been used in organic solvents or in solvent-free procedures, but they are clearly unsuitable for our analytical purpose in aqueous medium (aqueous samples containing resorcinol). One catalyzed procedure in water has been described with acidic ionic liquids [27], but it requires heating at 80–100 °C for 1 h to reach coumarin formation. These reaction conditions are also clearly inapplicable with plastic microplates, and reaction times are too long for fast analyses. Considering drawbacks of these alternatives, we chose to keep sulfuric acid to promote fast coumarin formation. A waste disposal protocol has to be set up for sulfuric acid, but it is a routine procedure for an analytical laboratory and potentially less constraining than toxic elements as catalysts. Moreover, low volumes used due to microplate format enable strong reduction of acidic waste generated. Various volumes of sulfuric acid were tested (with a fixed volume of sample) in order to optimize the signal/blank ratio on a resorcinol standard solution (Fig. 3A). Best results were obtained for H2SO4/resorcinol volume ratio close to 1. Lower ratios resulted in incomplete coumarin formation, while higher ratios implied higher blank values due to residual fluorescence in the sulfuric acid, with no change in the resorcinol standard response. 3.2.2. Methyl acetoacetate concentration Methyl acetoacetate is added directly without any dilution, and influence of the addition of increasing volumes on the fluorescence
Fig. 1. Schematic depiction of the formation of a fluorescent coumarin with resorcinol.
F. Robert-Peillard et al. / Microchemical Journal 122 (2015) 5–9
7
Fig. 2. Influence of (A) excitation; (B) emission wavelength on the ratio between a 10 μM resorcinol standard and a blank sample.
Fig. 3. (A) Influence of H2SO4 volume added on the ratio between a 10 μM resorcinol standard and a blank sample. Condition: 75 μL of 10 μM resorcinol, 10 μL methyl acetoacetate, reaction time: 10 min. (B) Evolution of the fluorescence response with time for different H2SO4/resorcinol volume ratios. Condition: 75 μL of 10 μM resorcinol, 50/65/75 μL H2SO4, 10 μL methyl acetoacetate.
response of a 10 μM resorcinol standard was studied (with 75 μL resorcinol standard + 75 μL H2SO4). No significant change was observed for volumes between 4 and 20 μL. The addition of 10 μL methyl acetoacetate was selected as the lowest reproducible volume for the final procedure. Ethyl acetoacetate was also assessed for the coumarin formation, with lowest efficiency, and its methyl analogue was therefore preferred.
optimum conditions for detection of resorcinol. The increase of reaction temperature would probably further reduce reaction time, but it was deemed unnecessary considering the very acceptable optimized reaction time.
3.3. Comparative responses of different phenols 3.2.3. Effect of reaction time It is generally preferable for analytical purposes that an analytical protocol proceeds in a few minutes, so as to provide potential for high-throughput analysis. Reaction kinetics were studied for the three volume ratios that provided the best results for the preliminary experiments on Fig. 3A. The evolution of the fluorescence response of a standard solution is displayed on Fig. 3B. Reaction was still ongoing after 8 min for ratios of 0.67 and 0.87, while fluorescence response reached maximum value after only 4 min for a H2SO4/resorcinol volume ratio of 1. Therefore, we chose this ratio value with 5 min reaction time as Table 1 Comparative responses of various phenolic compounds (10 μM, 75 μL standard + 75 μL H2SO4 + 10 μL methyl acetoacetate; 5 min reaction time). Blank fluorescence was subtracted for each sample. Phenolic compound
Fluorescence response (a.u.)
Phenol Catechol Hydroquinone m-Cresol Phloroglucinol 1,3-Dihydroxynaphthalene Resorcinol
8 7 26 7 48 108 27907
The selectivity of this detection method for resorcinol was assessed by application of the optimized protocol on various phenolic compounds (Table 1). Pechmann reaction is known to be more efficient on activated phenols (hydroxyphenols) than on deactivated ones. Quite surprisingly, only resorcinol resulted in a strong fluorescence emission, although phenols like phloroglucinol could be anticipated as more activated compounds. The 3D fluorescence scanning of other possible phenols-derived coumarins showed no other fluorescence peak, suggesting exclusive reaction with resorcinol under the experimental conditions and very high selectivity of the proposed method.
Table 2 Figures of merit of the proposed method for the determination of resorcinol. Limit of detection (μmol L−1) Limit of detection (μg L−1) Limit of quantification (μg L−1) Linear range RSD (%, n = 10, 100 μg L−1) Calibration curve equation R2
4.1 × 10−3 0.46 1.52 1.5–1000 μg L−1 2.01 Y = 3047.1X − 65.1 0.9996
8
F. Robert-Peillard et al. / Microchemical Journal 122 (2015) 5–9
Table 3 Comparison of our method with some other methods proposed in the literature for the determination of resorcinol. Analytical technique
Sample
LDRa (mg L−1)
LODb (μg L−1)
Reference
HPLC/chemiluminescence Nanoparticles resonance Electrochemically modified GCEc Spectrofluorimetry Spectrofluorimetry Spectrofluorimetry
Water samples Cosmetic samples Wastewaters Wastewaters Pharmaceutical samples Wastewaters + cosmetic samples
0.05–1 0.22–1.21 0.11–6.05 0.024–0.28 0.53–17.9 0.0015–1
4 130 82 12 315 0.46
[8] [12] [10] [16] [15] This work
a b c
Linear dynamic range. Limit of detection. Glassy carbon electrode.
3.4. Analytical performance Analytical features of the proposed method are summarized in Table 2. Limit of detection (LOD) and quantification (LOQ) were calculated as 3σb/s and 10σb/s, respectively (σb the standard deviation of the blank signals with n = 10, and s the slope of the calibration curve). The method showed very good linearity over a wide range of concentration, with excellent sensitivity (LOD below 1 μg L−1) and reproducibility (RSD of 2.01%). Analytical performances of our protocol were compared with other published methods for the determination of resorcinol (Table 3). Our spectrofluorimetric method is the most sensitive method described so far for resorcinol detection (LOD = 0.46 μg L−1), most other methods exhibiting LOD above 10 μg L−1. Linear dynamic range is also the widest with almost three orders of magnitude. Moreover, specificity is also a lot higher for our method than others which assessed the response of other phenolic compounds [21,22], with non-negligible interference of at least one or two other compounds in these other studies.
3.5. Effect of interfering ions The influence of other matrix components that can potentially interfere for our resorcinol determination method was assessed on a 1 μM resorcinol standard solution. Table 4 displays ratios of some matrix components to resorcinol that cause less than 5% change in the fluorescence response of the resorcinol standard. Metallic cations were tested as possible complexing agents of resorcinol or methyl acetoacetate, with only Fe3+ exhibiting potential interference for ratios over 500. Nitrate ions were also suspected as possible interfering agents for our method, as resorcinol is used for the spectrophotometric determination of nitrate via nitration of resorcinol [28]. Experiments proved that no
Table 4 Influence of various matrix components on resorcinol determination. Matrix components
Ratioa
Na+, K+, Cl− Ca2+, Mg2+, Al3+, Cu2+, Mn2+, Zn2+, Pb2+ NO− 3 Fe3+ NO− 2
N10000 N1000 N1000 500 10 or N1000b
a Ratio of the compound compared to resorcinol that causes less than 5% change in the fluorescence response of a 1 μM resorcinol standard. b Ratio of the compound compared to resorcinol that causes less than 5% change in the fluorescence response of a 1 μM resorcinol standard with the addition of sulfamic acid.
interference occurred, possible explanation being that nitration of resorcinol requires high chloride concentrations. Finally, nitrite ions exhibited interference for low ratios compared to other compounds, presumably because of diazotization of resorcinol with nitrite in acidic conditions. Nevertheless, this interference could be easily suppressed by addition of 10 μL of a 10 mM sulfamic acid solution prior to other reagents, as sulfamic acid is known to reduce nitrite to nitrogen [29]. 3.6. Application on real samples The optimized protocol described in Section 2.3 was applied on real samples for validation purposes. Two different types of samples with potentially very different matrix composition were tested: wastewater (Table 5) and pharmaceutical/cosmetic samples (Table 6). As no resorcinol was found in wastewater samples A and B (inlets of wastewater treatment plants), standard addition method was applied to assess recovery after addition of a known amount of analyte. As can be seen, very good recoveries were obtained for both samples. Application to pharmaceutical (sample 3, liquid for treatment of muscular pain) and cosmetic (samples 1 and 2, hair lotions) samples that contained resorcinol was validated by comparison with the HPLC determination of resorcinol following published protocol for phenolic compounds separation [30]. Samples were appropriately diluted in water to match calibration ranges of both methods. Results by our microplate method were well correlated with HPLC measurement, with only slight underestimation potentially due to other matrix components. 4. Conclusions In this paper, we have developed a new spectrofluorimetric method for resorcinol determination, based on the formation of a coumarin derivative. This method is very fast and simple and is superior to other published methods in terms of specificity to other phenolic compounds and selectivity, with a limit of detection below 1 μg L−1. Application in the microplate format also enables high-throughput analysis if screening of numerous samples is required. The validation of the proposed method on real samples (wastewater and pharmaceutical/cosmetic samples) demonstrated its applicability for specific resorcinol determination. One of the other potential application of this analytical method could be the determination of resorcinol in water resources before chlorination treatment to prevent production of disinfection by-products.
Table 6 Application of the method to pharmaceutical/cosmetic products. Table 5 Application of the method to wastewater samples (standard addition). −1
Sample
Found (μg L
A B
bLOQ bLOQ
)
Added (μg L 110 110
−1
)
Found (μg L 103.5 104.6
−1
)
Recovery (%) 94.1 95.1
Sample
Microplate measurement (mg L−1)
HPLC measurement (mg L−1)
ΔC (%)
1 2 3
12.74 8.32 158.76
13.32 8.82 171.21
4.3 5.6 7.2
F. Robert-Peillard et al. / Microchemical Journal 122 (2015) 5–9
References [1] M. Gonzalez, B. Guzman, R. Rudyk, E. Romano, M.A.A. Molina, Spectrophotometric determination of phenolic compounds in propolis, Lat. Am. J. Pharm. 3 (2003) 243–248. [2] Standard methods for the examination of water and wastewater, 16American public health association, Washington, DC, 1985. 510. [3] R.Q. Wang, L. Gutierrez, N.S. Choon, J.P. Croué, Hydrophilic interaction liquid chromatography method for measuring the composition of aquatic humic substances, Anal. Chim. Acta 853 (2015) 608–616. [4] F. Zhao, X. Li, G. Nigel, Treatment of a model HA compound (resorcinol) by potassium manganate, Sep. Sci. Technol. 91 (2012) 52–58. [5] H. Li, J. Ni, Electrogeneration of disinfection byproducts at a boron-doped diamond anode with resorcinol as a model substance, Electrochim. Acta 69 (2012) 268–274. [6] Study on the scientific evaluation of 12 substances in the context of endocrine disrupter priority list of actions. Produced, by WRc-NSF, Medmenham, for the European Commission, 613 pp. (WRc-NSF Reference: UC 6052 http://europa.eu. int/comm/environment/endocrine/documents/wrc_report.pdf, 2002). [7] S.K. Pardeshi, A.B. Patil, Solar photocatalytic degradation of resorcinol a model endocrine disruptor in water using zinc oxide, J. Hazard. Mater. 163 (2009) 403–409. [8] C. Groshart, P.C. Okkerman, Towards the establishment of a priority list of substances for further evaluation of their role in endocrine disruption. BKH Consulting engineers, http://ec.europa.eu/environment/archives/docum/pdf/bkh_main.pdf2000. [9] European Communities, Council directive 76/768/EEC of 27 July 1976 on the approximation of the laws of the member states relating to cosmetic products, 2003. 1–83. [10] S. Hahn, J. Kielhorn, J. Koppenhöfer, A. Wibbertmann, I. Mangelsdorf, Concise International Chemical Assessment Document, vol. 71, WHO Press, World Health Organization, Switzerland, 2006. [11] L. Pramparo, M.E. Suarez-Ojeda, J. Perez, J. Carrera, Kinetics of aerobic biodegradation of dihydroxybenzenes by a p-nitrophenol-degrading activated sludge, Bioresour. Technol. 110 (2012) 57–62. [12] O. Fiehn, M. Jekel, Analysis of phenolic compounds in industrial wastewater with high-performance liquid chromatography and post-column reaction detection, J. Chromatogr. A 769 (1997) 189–200. [13] C. Hua, Z. Jian, X. Feng, L. Chun-Ze, W. Guo-Hui, Determination of phenolic compounds using high-performance liquid chromatography with Ce4+-Tween chemiluminescence detection, Anal. Chim. Acta 511 (2004) 273–279. [14] H.W. Wu, M.L. Chen, D. Shou, Y. Zhu, Determination of resorcinol and phloroglucinol in environmental water samples using ion chromatography with chemiluminescence detection, Chin. J. Anal. Chem. 40 (2012) 1747–1751. [15] R.L. Cooper, K.C. Wheatstone, The determination of phenols in aqueous effluents, Water Res. 7 (1973) 1375–1384.
9
[16] H. Yin, Q. Zhang, Y. Zhou, Q. Ma, T. Liu, L. Zhu, S. Ai, Electrochemical behavior of catechol, resorcinol and hydroquinone at graphene–chitosan composite film modified glassy carbon electrode and their simultaneous determination in water samples, Electrochim. Acta 56 (2011) 2748–2753. [17] S.M. Ghoreishi, M. Behpour, E. Hajisadeghian, M. Golestaneh, Voltammetric determination of resorcinol on the surface of a glassy carbon electrode modified with multiwalled carbon nanotube, Arab. J. Chem. (2012) 1–6. [18] B. Zargar, A. Hatamie, Colorimetric determination of resorcinol based on localized surface plasmon resonance of silver nanoparticles, Analyst 137 (2012) 5334–5338. [19] F. Onur, N. Acar, Simultaneous determination of methylene blue, hexamethylene tetramine and resorcinol in pharmaceutical formulations by first-derivative UV spectrophotometry, Int. J. Pharm. 78 (1992) 89–91. [20] M.F. Pistonesi, M.S. Di Nezio, M.E. Centurion, A.G. Lista, W.D. Fragoso, M.J.C. Pontes, M.C.U. Araujo, B.S. Fernandez Band, Simultaneous determination of hydroquinone, resorcinol, phenol, m-cresol and p-cresol in untreated air samples using spectrofluorimetry and a custom multiple linear regression-successive projection algorithm, Talanta 83 (2010) 320–323. [21] W. Xiao, D. Xiao, Aminopyrene functionalized mesoporous silica for the selective determination of resorcinol, Talanta 72 (2007) 1288–1292. [22] J. Fan, T. Zhang, J. Sun, M. Fan, Kinetic fluorimetric measurement of trace resorcinol in phenol mixtures, J. Fluoresc. 17 (2007) 113–118. [23] Z. Wu, Z. Li, L. Yang, J. Han, S. Han, Fluorogenic detection of hydrogen sulfide via reductive unmasking of o-azidomethylbenzoyl-coumarin conjugate, Chem. Commun. 48 (2012) 10120–10122. [24] B. Karami, M. Kiani, ZrOCl2.8H2O/SiO2: an efficient and recyclable catalyst for the preparation of coumarin derivatives by Pechmann condensation reaction, Catal. Commun. 14 (2011) 62–67. [25] A. Hegedus, Z. Hell, Zeolite-catalyzed Pechmann synthesis of coumarins, Catal. Lett. 112 (2006) 105–108. [26] E.V. Stoyanov, J. Mezger, Pechmann reaction promoted by boron trifluoride dihydrate, Molecules 10 (2005) 762–766. [27] F. Dong, C. Jian, G. Kai, S. Qunrong, L. Zuliang, Synthesis of coumarins via Pechmann reaction in water catalyzed by acyclic acidic ionic liquids, Catal. Lett. 121 (2008) 255–259. [28] J.Z. Zhang, C.J. Fischer, A simplified resorcinol method for direct spectrophotometric determination of nitrate in seawater, Mar. Chem. 99 (2006) 220–226. [29] J. Granger, D.M. Siglan, Removal of nitrite with sulfamic acid for nitrate N and O isotope analysis with the denitrifier method, Rapid Commun. Mass Spectrom. 23 (2009) 3753–3762. [30] A. Zhang, L. Wan, C. Wu, Y. Fang, G. Han, H. Li, Z. Zhang, H. Wang, Simultaneous determination of 14 phenolic compounds in grape canes by HPLC-DAD-UV using wavelength switching detection, Molecules 18 (2013) 14241–14257.
Contents lists available at ScienceDirect
Microchemical Journal journal homepage: www.elsevier.com/locate/microc
Simple and ultrasensitive microplate method for spectrofluorimetric determination of trace resorcinol Fabien Robert-Peillard ⁎, Claire Chottier, Bruno Coulomb, Jean-Luc Boudenne Aix-Marseille Université, CNRS, LCE FRE 3416, 13331 Marseille cedex 3, France
a r t i c l e
i n f o
Article history: Received 27 February 2015 Received in revised form 1 April 2015 Accepted 2 April 2015 Available online 9 April 2015 Keywords: Resorcinol Spectrofluorimetric Microplate Coumarin Wastewater Cosmetics
a b s t r a c t In this paper, a new spectrofluorimetric method is proposed for the fast, simple, and accurate determination of trace resorcinol in the microplate format (high-throughput screening). The analytical method is based on the formation of a coumarin derivative with methyl acetoacetate via a Pechmann condensation reaction. Experimental conditions have been optimized, and excellent analytical performances have been achieved with a limit of detection of 0.46 μg L−1, a wide linear range between 1.5 and 1000 μg L−1 and a relative standard deviation of 2.01% (n = 10). This method is remarkably specific for resorcinol compared to other phenols and hydroxyphenols and suffers from low interferences from other potential matrix components. The proposed protocol was finally applied and validated on real samples such as wastewater or pharmaceutical/cosmetic samples. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Phenolic compounds are a class of organic contaminants that are of great concern for environmental or human health issues. Several spectroscopic methods are available for the determination of total phenolic compounds, including Prussian Blue, o-phenanthroline, or 4aminoantipyrine [1,2]. Nevertheless, methods for the individual determination of these compounds can be very valuable tools to study or understand their environmental behavior, evaluate their toxic effect, or quantify a specific compound in different matrices. Among these phenolic compounds, resorcinol (benzene-1,3-diol) has received special attention due to its toxicity and significant occurrence in the environment. Resorcinol has indeed natural sources, being a monomeric by-product of the reduction, oxidation, and microbial degradation of humic substances [3]. It is thus often found in natural waters used for tap water production, contributing to trihalomethane potential formation when these waters are chlorinated [4,5]. Apart from these natural origins, resorcinol is produced worldwide at increasing levels (from 40 000 tons in 1990 to 45 000 tons in 2000) [6] and is used in various industrial applications (rubber industry, adhesives, dyes, speciality chemicals, pharmaceuticals) [7]. Resorcinol is a proven endocrine disruptor and has been listed as a category 1 substance (substance of high concern) in the European Union priority list of substances for further evaluation of their role in endocrine disruption [8]. Toxicity for human health include dermatitis, ⁎ Corresponding author. Tel.: +33 413551030; fax: +33 413551060. E-mail address: [email protected] (F. Robert-Peillard).
http://dx.doi.org/10.1016/j.microc.2015.04.004 0026-265X/© 2015 Elsevier B.V. All rights reserved.
catarrh, convulsion, or cyanopathy, and strict limitations has been set for resorcinol concentration in pharmaceutical products such as shampoos and hair lotions (0.5% limit) [9]. Serious implications of the deleterious impacts of resorcinol exposure on human health and the corresponding emergency measures to be adopted have been discussed in a document by World Health Organization (WHO) [10]. Moreover, it has been proven that resorcinol has the lowest biodegradation rates with aerobic biodegradation processes among all dihydroxybenzene isomers, resulting in increased threats for final environmental impacts [11]. Therefore, development of simple analytical methods for specific resorcinol determination has a great interest for several types of applications, such as industrial waters, wastewaters, and hair lotions. The most common and reliable methods for the determination of resorcinol are high-performance liquid chromatography (HPLC) [12–14] and gas chromatography [15]. These methods are efficient but require expensive instruments; therefore, cheaper and simpler alternatives are highly desirable. In recent years, several methods have been described for the alternative determination of resorcinol in different samples, such as electrochemistry [16,17], spectrophotometry with nanoparticles [18] or first-derivative methods [19], direct spectrofluorimetry [20] or inhibition-based spectrofluorimetry [21,22]. Although all these methods have interesting properties for detection of resorcinol, they all suffer from relatively low sensitivity and high limit of detection of about 0.1 to 1 mg L−1, a narrow linear dynamic range (less than 2 orders of magnitude) and poor or undefined specificity towards other phenols and other hydroxyphenols isomers. Thus, the development of a simple, specific, ultrasensitive, and fast method for the detection of resorcinol still remains a great challenge.
6
F. Robert-Peillard et al. / Microchemical Journal 122 (2015) 5–9
In this work, a very simple and sensitive spectrofluorimetric method is described. The method is based on the reaction of resorcinol with a β-keto ester in acidic medium which yields a fluorescent coumarin (Pechmann condensation). Fluorescence is measured after 5 min at an excitation and emission wavelength of 332 nm and 431 nm, respectively. The simplicity of this protocol enables its implementation in microplate wells, providing high-throughput analysis with low volumes of samples, reagents, and waste. Development of the global procedure and application on real wastewater and pharmaceutical/cosmetic products are presented herein. 2. Experimental 2.1. Reagents and solutions All chemicals were of analytical reagent grade and used without further purification. Resorcinol, methyl acetoacetate, and ethyl acetoacetate were purchased from Alfa Aesar (Schiltigheim, France), catechol, hydroquinone, phloroglucinol, phenol, 4-nitrophenol, 2-chlorophenol, and ascorbic acid were from Sigma-Aldrich (SaintQuentin Fallavier, France), Suprapur sulfuric acid from Merck (Darmstadt, Germany). Resorcinol and other phenolic compounds stock standard solutions and working solutions were prepared in ultrapure water (Millipore, resistivity N 18 MΩ cm).
3. Results and discussion 3.1. Formation of the fluorescent product Pechmann condensation is a well-known organic reaction used for synthesis of various coumarins. Reaction scheme with resorcinol is depicted on Fig. 1, with formation of 7-hydroxy-4-methylcoumarin (4-methylumbelliferone). Application of this reaction for analytical purposes appeared very interesting to us since the formation of highly fluorescent products from non- or poorly fluorescent analytes/reagents is a method of choice for attaining high sensitivity. The selection of optimum excitation and emission wavelengths was carried out on the fluorescent product formed under the reaction conditions. 7-Hydroxy-4-methylcoumarin is a well-known fluorophore with reported optimum excitation and emission wavelengths of 340–365 nm and 430–450 nm, respectively [23]. For optimum fluorescent measurements in black microplates, one has to investigate not only the maximum of intensity for a given standard, but also the intensity of a blank sample (with no standard, which takes also into account the autofluorescence of the microplate material). Fig. 2 displays the best set of wavelengths by measuring the optimum signal/blank ratio, with final wavelengths set at 332 nm and 431 nm for excitation and emission, respectively. 3.2. Optimization of experimental conditions
2.2. Microplate instrumentation Microplate fluorescence measurements were carried out on a microplate reader (Infinite M200, Tecan France SAS, Lyon, France) equipped with an excitation and emission double monochromator (bandwidths of 9 nm and 20 nm for excitation and emission monochromator, respectively) and controlled by i-control™ software (Tecan). Fluorescence detection was performed from above the microplate wells (top configuration), at λex = 332 nm and λem = 431 nm. Operating temperature was set to 25 °C. Other parameters were as follows: gain: 90; number of flashes: 25; integration time: 20 μs. Fluorescence intensities were expressed in arbitrary units (a.u.). Polystyrene black 96 flat-bottom well microplates (Fisher Scientific, Illkirch, France), with a maximum capacity of 375 μL for each wells were used. In order to avoid degradation of the microplate reader due to sulfuric acid use, microplates were shaken on an external vortex mixer (Vortex Genie 2, Scientific Industries, New-York, USA), placed in the microplate reader only for fluorescent measurements and then quickly removed and stored under a fume hood. 2.3. Optimized protocol for resorcinol determination in aqueous samples Seventy-five microliters of sample or resorcinol standard solution was dispensed into the wells of the microplate, followed by 75 μL of sulfuric acid and 10 μL of methyl acetoacetate. The plate was shaken for 5 min at room temperature on the vortex mixer, and fluorescence was subsequently measured at λex = 332 nm and λem = 431 nm in the microplate reader. Resorcinol concentrations were determined using the linear calibration curve obtained with standards. All experiments were performed in duplicate.
3.2.1. Sulfuric acid concentration Classical conditions for coumarin formation via Pechmann condensation require sulfuric acid to reach good yields in a reasonable time. Many alternatives have been developed to try to avoid the use of strong acid. Catalysts such as ZrOCl2.8H2O/SiO2 [24], zeolite [25] or boron trifluoride [26] have been used in organic solvents or in solvent-free procedures, but they are clearly unsuitable for our analytical purpose in aqueous medium (aqueous samples containing resorcinol). One catalyzed procedure in water has been described with acidic ionic liquids [27], but it requires heating at 80–100 °C for 1 h to reach coumarin formation. These reaction conditions are also clearly inapplicable with plastic microplates, and reaction times are too long for fast analyses. Considering drawbacks of these alternatives, we chose to keep sulfuric acid to promote fast coumarin formation. A waste disposal protocol has to be set up for sulfuric acid, but it is a routine procedure for an analytical laboratory and potentially less constraining than toxic elements as catalysts. Moreover, low volumes used due to microplate format enable strong reduction of acidic waste generated. Various volumes of sulfuric acid were tested (with a fixed volume of sample) in order to optimize the signal/blank ratio on a resorcinol standard solution (Fig. 3A). Best results were obtained for H2SO4/resorcinol volume ratio close to 1. Lower ratios resulted in incomplete coumarin formation, while higher ratios implied higher blank values due to residual fluorescence in the sulfuric acid, with no change in the resorcinol standard response. 3.2.2. Methyl acetoacetate concentration Methyl acetoacetate is added directly without any dilution, and influence of the addition of increasing volumes on the fluorescence
Fig. 1. Schematic depiction of the formation of a fluorescent coumarin with resorcinol.
F. Robert-Peillard et al. / Microchemical Journal 122 (2015) 5–9
7
Fig. 2. Influence of (A) excitation; (B) emission wavelength on the ratio between a 10 μM resorcinol standard and a blank sample.
Fig. 3. (A) Influence of H2SO4 volume added on the ratio between a 10 μM resorcinol standard and a blank sample. Condition: 75 μL of 10 μM resorcinol, 10 μL methyl acetoacetate, reaction time: 10 min. (B) Evolution of the fluorescence response with time for different H2SO4/resorcinol volume ratios. Condition: 75 μL of 10 μM resorcinol, 50/65/75 μL H2SO4, 10 μL methyl acetoacetate.
response of a 10 μM resorcinol standard was studied (with 75 μL resorcinol standard + 75 μL H2SO4). No significant change was observed for volumes between 4 and 20 μL. The addition of 10 μL methyl acetoacetate was selected as the lowest reproducible volume for the final procedure. Ethyl acetoacetate was also assessed for the coumarin formation, with lowest efficiency, and its methyl analogue was therefore preferred.
optimum conditions for detection of resorcinol. The increase of reaction temperature would probably further reduce reaction time, but it was deemed unnecessary considering the very acceptable optimized reaction time.
3.3. Comparative responses of different phenols 3.2.3. Effect of reaction time It is generally preferable for analytical purposes that an analytical protocol proceeds in a few minutes, so as to provide potential for high-throughput analysis. Reaction kinetics were studied for the three volume ratios that provided the best results for the preliminary experiments on Fig. 3A. The evolution of the fluorescence response of a standard solution is displayed on Fig. 3B. Reaction was still ongoing after 8 min for ratios of 0.67 and 0.87, while fluorescence response reached maximum value after only 4 min for a H2SO4/resorcinol volume ratio of 1. Therefore, we chose this ratio value with 5 min reaction time as Table 1 Comparative responses of various phenolic compounds (10 μM, 75 μL standard + 75 μL H2SO4 + 10 μL methyl acetoacetate; 5 min reaction time). Blank fluorescence was subtracted for each sample. Phenolic compound
Fluorescence response (a.u.)
Phenol Catechol Hydroquinone m-Cresol Phloroglucinol 1,3-Dihydroxynaphthalene Resorcinol
8 7 26 7 48 108 27907
The selectivity of this detection method for resorcinol was assessed by application of the optimized protocol on various phenolic compounds (Table 1). Pechmann reaction is known to be more efficient on activated phenols (hydroxyphenols) than on deactivated ones. Quite surprisingly, only resorcinol resulted in a strong fluorescence emission, although phenols like phloroglucinol could be anticipated as more activated compounds. The 3D fluorescence scanning of other possible phenols-derived coumarins showed no other fluorescence peak, suggesting exclusive reaction with resorcinol under the experimental conditions and very high selectivity of the proposed method.
Table 2 Figures of merit of the proposed method for the determination of resorcinol. Limit of detection (μmol L−1) Limit of detection (μg L−1) Limit of quantification (μg L−1) Linear range RSD (%, n = 10, 100 μg L−1) Calibration curve equation R2
4.1 × 10−3 0.46 1.52 1.5–1000 μg L−1 2.01 Y = 3047.1X − 65.1 0.9996
8
F. Robert-Peillard et al. / Microchemical Journal 122 (2015) 5–9
Table 3 Comparison of our method with some other methods proposed in the literature for the determination of resorcinol. Analytical technique
Sample
LDRa (mg L−1)
LODb (μg L−1)
Reference
HPLC/chemiluminescence Nanoparticles resonance Electrochemically modified GCEc Spectrofluorimetry Spectrofluorimetry Spectrofluorimetry
Water samples Cosmetic samples Wastewaters Wastewaters Pharmaceutical samples Wastewaters + cosmetic samples
0.05–1 0.22–1.21 0.11–6.05 0.024–0.28 0.53–17.9 0.0015–1
4 130 82 12 315 0.46
[8] [12] [10] [16] [15] This work
a b c
Linear dynamic range. Limit of detection. Glassy carbon electrode.
3.4. Analytical performance Analytical features of the proposed method are summarized in Table 2. Limit of detection (LOD) and quantification (LOQ) were calculated as 3σb/s and 10σb/s, respectively (σb the standard deviation of the blank signals with n = 10, and s the slope of the calibration curve). The method showed very good linearity over a wide range of concentration, with excellent sensitivity (LOD below 1 μg L−1) and reproducibility (RSD of 2.01%). Analytical performances of our protocol were compared with other published methods for the determination of resorcinol (Table 3). Our spectrofluorimetric method is the most sensitive method described so far for resorcinol detection (LOD = 0.46 μg L−1), most other methods exhibiting LOD above 10 μg L−1. Linear dynamic range is also the widest with almost three orders of magnitude. Moreover, specificity is also a lot higher for our method than others which assessed the response of other phenolic compounds [21,22], with non-negligible interference of at least one or two other compounds in these other studies.
3.5. Effect of interfering ions The influence of other matrix components that can potentially interfere for our resorcinol determination method was assessed on a 1 μM resorcinol standard solution. Table 4 displays ratios of some matrix components to resorcinol that cause less than 5% change in the fluorescence response of the resorcinol standard. Metallic cations were tested as possible complexing agents of resorcinol or methyl acetoacetate, with only Fe3+ exhibiting potential interference for ratios over 500. Nitrate ions were also suspected as possible interfering agents for our method, as resorcinol is used for the spectrophotometric determination of nitrate via nitration of resorcinol [28]. Experiments proved that no
Table 4 Influence of various matrix components on resorcinol determination. Matrix components
Ratioa
Na+, K+, Cl− Ca2+, Mg2+, Al3+, Cu2+, Mn2+, Zn2+, Pb2+ NO− 3 Fe3+ NO− 2
N10000 N1000 N1000 500 10 or N1000b
a Ratio of the compound compared to resorcinol that causes less than 5% change in the fluorescence response of a 1 μM resorcinol standard. b Ratio of the compound compared to resorcinol that causes less than 5% change in the fluorescence response of a 1 μM resorcinol standard with the addition of sulfamic acid.
interference occurred, possible explanation being that nitration of resorcinol requires high chloride concentrations. Finally, nitrite ions exhibited interference for low ratios compared to other compounds, presumably because of diazotization of resorcinol with nitrite in acidic conditions. Nevertheless, this interference could be easily suppressed by addition of 10 μL of a 10 mM sulfamic acid solution prior to other reagents, as sulfamic acid is known to reduce nitrite to nitrogen [29]. 3.6. Application on real samples The optimized protocol described in Section 2.3 was applied on real samples for validation purposes. Two different types of samples with potentially very different matrix composition were tested: wastewater (Table 5) and pharmaceutical/cosmetic samples (Table 6). As no resorcinol was found in wastewater samples A and B (inlets of wastewater treatment plants), standard addition method was applied to assess recovery after addition of a known amount of analyte. As can be seen, very good recoveries were obtained for both samples. Application to pharmaceutical (sample 3, liquid for treatment of muscular pain) and cosmetic (samples 1 and 2, hair lotions) samples that contained resorcinol was validated by comparison with the HPLC determination of resorcinol following published protocol for phenolic compounds separation [30]. Samples were appropriately diluted in water to match calibration ranges of both methods. Results by our microplate method were well correlated with HPLC measurement, with only slight underestimation potentially due to other matrix components. 4. Conclusions In this paper, we have developed a new spectrofluorimetric method for resorcinol determination, based on the formation of a coumarin derivative. This method is very fast and simple and is superior to other published methods in terms of specificity to other phenolic compounds and selectivity, with a limit of detection below 1 μg L−1. Application in the microplate format also enables high-throughput analysis if screening of numerous samples is required. The validation of the proposed method on real samples (wastewater and pharmaceutical/cosmetic samples) demonstrated its applicability for specific resorcinol determination. One of the other potential application of this analytical method could be the determination of resorcinol in water resources before chlorination treatment to prevent production of disinfection by-products.
Table 6 Application of the method to pharmaceutical/cosmetic products. Table 5 Application of the method to wastewater samples (standard addition). −1
Sample
Found (μg L
A B
bLOQ bLOQ
)
Added (μg L 110 110
−1
)
Found (μg L 103.5 104.6
−1
)
Recovery (%) 94.1 95.1
Sample
Microplate measurement (mg L−1)
HPLC measurement (mg L−1)
ΔC (%)
1 2 3
12.74 8.32 158.76
13.32 8.82 171.21
4.3 5.6 7.2
F. Robert-Peillard et al. / Microchemical Journal 122 (2015) 5–9
References [1] M. Gonzalez, B. Guzman, R. Rudyk, E. Romano, M.A.A. Molina, Spectrophotometric determination of phenolic compounds in propolis, Lat. Am. J. Pharm. 3 (2003) 243–248. [2] Standard methods for the examination of water and wastewater, 16American public health association, Washington, DC, 1985. 510. [3] R.Q. Wang, L. Gutierrez, N.S. Choon, J.P. Croué, Hydrophilic interaction liquid chromatography method for measuring the composition of aquatic humic substances, Anal. Chim. Acta 853 (2015) 608–616. [4] F. Zhao, X. Li, G. Nigel, Treatment of a model HA compound (resorcinol) by potassium manganate, Sep. Sci. Technol. 91 (2012) 52–58. [5] H. Li, J. Ni, Electrogeneration of disinfection byproducts at a boron-doped diamond anode with resorcinol as a model substance, Electrochim. Acta 69 (2012) 268–274. [6] Study on the scientific evaluation of 12 substances in the context of endocrine disrupter priority list of actions. Produced, by WRc-NSF, Medmenham, for the European Commission, 613 pp. (WRc-NSF Reference: UC 6052 http://europa.eu. int/comm/environment/endocrine/documents/wrc_report.pdf, 2002). [7] S.K. Pardeshi, A.B. Patil, Solar photocatalytic degradation of resorcinol a model endocrine disruptor in water using zinc oxide, J. Hazard. Mater. 163 (2009) 403–409. [8] C. Groshart, P.C. Okkerman, Towards the establishment of a priority list of substances for further evaluation of their role in endocrine disruption. BKH Consulting engineers, http://ec.europa.eu/environment/archives/docum/pdf/bkh_main.pdf2000. [9] European Communities, Council directive 76/768/EEC of 27 July 1976 on the approximation of the laws of the member states relating to cosmetic products, 2003. 1–83. [10] S. Hahn, J. Kielhorn, J. Koppenhöfer, A. Wibbertmann, I. Mangelsdorf, Concise International Chemical Assessment Document, vol. 71, WHO Press, World Health Organization, Switzerland, 2006. [11] L. Pramparo, M.E. Suarez-Ojeda, J. Perez, J. Carrera, Kinetics of aerobic biodegradation of dihydroxybenzenes by a p-nitrophenol-degrading activated sludge, Bioresour. Technol. 110 (2012) 57–62. [12] O. Fiehn, M. Jekel, Analysis of phenolic compounds in industrial wastewater with high-performance liquid chromatography and post-column reaction detection, J. Chromatogr. A 769 (1997) 189–200. [13] C. Hua, Z. Jian, X. Feng, L. Chun-Ze, W. Guo-Hui, Determination of phenolic compounds using high-performance liquid chromatography with Ce4+-Tween chemiluminescence detection, Anal. Chim. Acta 511 (2004) 273–279. [14] H.W. Wu, M.L. Chen, D. Shou, Y. Zhu, Determination of resorcinol and phloroglucinol in environmental water samples using ion chromatography with chemiluminescence detection, Chin. J. Anal. Chem. 40 (2012) 1747–1751. [15] R.L. Cooper, K.C. Wheatstone, The determination of phenols in aqueous effluents, Water Res. 7 (1973) 1375–1384.
9
[16] H. Yin, Q. Zhang, Y. Zhou, Q. Ma, T. Liu, L. Zhu, S. Ai, Electrochemical behavior of catechol, resorcinol and hydroquinone at graphene–chitosan composite film modified glassy carbon electrode and their simultaneous determination in water samples, Electrochim. Acta 56 (2011) 2748–2753. [17] S.M. Ghoreishi, M. Behpour, E. Hajisadeghian, M. Golestaneh, Voltammetric determination of resorcinol on the surface of a glassy carbon electrode modified with multiwalled carbon nanotube, Arab. J. Chem. (2012) 1–6. [18] B. Zargar, A. Hatamie, Colorimetric determination of resorcinol based on localized surface plasmon resonance of silver nanoparticles, Analyst 137 (2012) 5334–5338. [19] F. Onur, N. Acar, Simultaneous determination of methylene blue, hexamethylene tetramine and resorcinol in pharmaceutical formulations by first-derivative UV spectrophotometry, Int. J. Pharm. 78 (1992) 89–91. [20] M.F. Pistonesi, M.S. Di Nezio, M.E. Centurion, A.G. Lista, W.D. Fragoso, M.J.C. Pontes, M.C.U. Araujo, B.S. Fernandez Band, Simultaneous determination of hydroquinone, resorcinol, phenol, m-cresol and p-cresol in untreated air samples using spectrofluorimetry and a custom multiple linear regression-successive projection algorithm, Talanta 83 (2010) 320–323. [21] W. Xiao, D. Xiao, Aminopyrene functionalized mesoporous silica for the selective determination of resorcinol, Talanta 72 (2007) 1288–1292. [22] J. Fan, T. Zhang, J. Sun, M. Fan, Kinetic fluorimetric measurement of trace resorcinol in phenol mixtures, J. Fluoresc. 17 (2007) 113–118. [23] Z. Wu, Z. Li, L. Yang, J. Han, S. Han, Fluorogenic detection of hydrogen sulfide via reductive unmasking of o-azidomethylbenzoyl-coumarin conjugate, Chem. Commun. 48 (2012) 10120–10122. [24] B. Karami, M. Kiani, ZrOCl2.8H2O/SiO2: an efficient and recyclable catalyst for the preparation of coumarin derivatives by Pechmann condensation reaction, Catal. Commun. 14 (2011) 62–67. [25] A. Hegedus, Z. Hell, Zeolite-catalyzed Pechmann synthesis of coumarins, Catal. Lett. 112 (2006) 105–108. [26] E.V. Stoyanov, J. Mezger, Pechmann reaction promoted by boron trifluoride dihydrate, Molecules 10 (2005) 762–766. [27] F. Dong, C. Jian, G. Kai, S. Qunrong, L. Zuliang, Synthesis of coumarins via Pechmann reaction in water catalyzed by acyclic acidic ionic liquids, Catal. Lett. 121 (2008) 255–259. [28] J.Z. Zhang, C.J. Fischer, A simplified resorcinol method for direct spectrophotometric determination of nitrate in seawater, Mar. Chem. 99 (2006) 220–226. [29] J. Granger, D.M. Siglan, Removal of nitrite with sulfamic acid for nitrate N and O isotope analysis with the denitrifier method, Rapid Commun. Mass Spectrom. 23 (2009) 3753–3762. [30] A. Zhang, L. Wan, C. Wu, Y. Fang, G. Han, H. Li, Z. Zhang, H. Wang, Simultaneous determination of 14 phenolic compounds in grape canes by HPLC-DAD-UV using wavelength switching detection, Molecules 18 (2013) 14241–14257.