Chlorine determination via MgCl molecule in environmental samples using high resolution continuum source graphite furnace molecular absorption spectrometry

Author’s Accepted Manuscript Chlorine determination via MgCl molecule in environmental samples using high resolution continuum source graphite furnace molecular absorption spectrometry Rina Lourena da S. Medeiros, Sidnei de Oliveira Souza, Rennan Geovanny Oliveira Araujo, Djalma Ribeiro da Silva, Tatiane de A. Maranhão

PII: DOI: Reference:

www.elsevier.com/locate/talanta

S0039-9140(17)30845-7 http://dx.doi.org/10.1016/j.talanta.2017.08.026 TAL17822

To appear in: Talanta Received date: 13 April 2017 Revised date: 8 July 2017 Accepted date: 7 August 2017 Cite this article as: Rina Lourena da S. Medeiros, Sidnei de Oliveira Souza, Rennan Geovanny Oliveira Araujo, Djalma Ribeiro da Silva and Tatiane de A. Maranhão, Chlorine determination via MgCl molecule in environmental samples using high resolution continuum source graphite furnace molecular absorption spectrometry, Talanta, http://dx.doi.org/10.1016/j.talanta.2017.08.026 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Chlorine determination via MgCl molecule in environmental samples using high resolution continuum source graphite furnace molecular absorption spectrometry Rina Lourena da S. Medeiros1, Sidnei de Oliveira Souza2, Rennan Geovanny Oliveira Araujo2, Djalma Ribeiro da Silva1, Tatiane de A. Maranhão 1,3* 1

Instituto de Química, Universidade Federal do Rio Grande do Norte, Núcleo de

Processamento Primário e Reuso de Águas e Resíduos - NUPPRAR, 59078-970, Natal RN, Brazil 2

Instituto de Química, Universidade Federal da Bahia, Salvador-BA, Brazil.

3

Departamento de Química, Universidade Federal de Santa Catarina, 88040-900,

Florianópolis-SC, Brazil. *Corresponding Author. Tel./fax.: + 554837213635. [email protected]

Abstract

This paper describes a method development for chlorine determination through the formation of MgCl molecule, applied for the first time for Cl quantification, by high resolution continuum source graphite furnace molecular absorption spectrometry (HRCS GF MAS) in environmental samples. Pyrolysis and vaporization temperatures were optimized as well as the use of chemical modifier. Determinations were carried out at the wavelength of 377.010 and the compromise conditions of the graphite furnace temperature program were 500 °C and 2500 °C for pyrolysis and vaporization, respectively, using 10 µg of chemical modifier Pd. The concentration of reactants for the generation of MgCl molecule was optimized through Box-Behnken experimental design, using MgCl2 solution as source of chlorine. The optimum values according to the surface response were 5 g L-1 Mg, 25 mg L-1 of chlorine and 2% v v-1 of HNO3, condition in which the amount of Mg is at least 200 times higher than that of chloride. This excess of the forming agent ensures the complete formation of MgCl molecular species, since Cl is the limiting reactant. Certified reference materials, BCR 182 and NIST 8414, and addition and recovery tests were used to evaluate the accuracy of the method and good results were achieved at a 95% confidence level. The method was applied to direct determination of Cl in five produced water samples from offshore oil

wellbore, high complex matrix, whose conventional methods require tedious treatment before the analysis.

Keywords: Chlorine determination; MgCl; Chemical modifier; HR-CS MAS,

1. Introduction

Chlorine is

among the twenty most

abundant

elements

comprising

approximately 0.0314 % (w w-1) of Earth’s crust. A major reserve of chlorine is the ocean, with a portion of 2 % (w w-1) of chloride in seawater as sodium and magnesium salts, but it is also found in the atmosphere and soil [1], [2], [3]. Due to the constant presence of this element in industrial processes and in environment or even in numerous raw materials and manufactured products, it is usually evaluated as contaminant. Thus, quality control and more rigorous monitoring are necessary, requiring reliable analytical methods for its determination in different types of matrices [4], [5].

Chlorine is usually determined by potentiometric techniques [6], such as, ionselective electrode [7][8], ion exchange chromatography [8]. Beside this, lately microwave-induced

combustion

inductively coupled

plasma

optical

emission

spectrometry (MIC-ICP OES) was applied to the determination of chlorine [9]. However, all these instrumental techniques require pretreatment of the samples, increasing the analytical frequency and the risk of contamination. Several studies reporting techniques used for chorine determination have been published, such as cold vapor molecular absorption spectrometry and electro thermal vaporization inductively coupled plasma mass spectrometry (ETV-ICP-MS) [9], energy dispersive X-ray fluorescence spectrometry (EDXRF) [12] and laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) [13], but all of them are susceptible to high risk of contamination and also have as a disadvantage the need of certified reference materials (CRM) for calibration. Besides, these high-cost techniques and are not easily available for routine analysis. Atomic absorption spectrometry (AAS) is not commonly used for non-metals such as chlorine, since the resonance lines of absorption of halogens are below 190 nm, i.e., outside the range of conventional AAS [14], [15]. In the mid of 1970s, Haraguchi and Fuwa investigated absorption spectra of halogen diatomic molecules in u. v. region and opened the way for determination of non-metals by molecular absorption spectrometry (MAS), whose resonance lines could not be detected due to the low resolution of the instruments available that time [16]. The introduction of high resolution continuum source atomic absorption spectrometry (HR-CS AAS) in 2000s made it possible the determination of non-metals through their vaporization as diatomic molecules. Different from atoms, which are subject only to electronic transitions presenting a line spectrum, molecules are also liable to rotational and vibrational transitions, resulting in more complex spectra, composed of bands of absorption [15], [18]. The determination of Cl via diatomic molecules such as AlCl, InCl and SrCl by HR-CS MAS has been well reported in the literature [3,17-21].

Fechetia et al.

developed a method for chlorine determination as AlCl by MAS at wavelength of 261.418 nm in rye flour after digestion in HNO3 at room temperature for several hours [2]. Heitmann et al. also carried out the determination of chlorine via AlCl at the wavelength of 261.420 nm after acid digestion in certified reference material (CRM) rye flour [17]; The isotope determination of chlorine was performed by Nakadi et al.

through the monitoring of the AlCl molecule and direct analysis of mineral water samples [19]; InCl was the molecule chosen by Huang et al. to determine chlorine at wavelength of 267.240 nm in CRMs of sediment after acid extraction [3]; Pereira et al. determined chlorine using molecular absorption of SrCl at wavelength of 635.862 nm, by direct analysis of biological and coal samples [20], [21]. Ozbek and Akman determined chlorine in milk via SrCl using HR CS AAS [22]. Investigations on the use of MgCl molecule by MAS were carried out in early 1980s by Ditrich and Vorberg, but no further studies or application to quantify chlorine in real samples were taken into effect up to now [23]. The aim of this work was develop a method for Cl determination via MgCl by HR-CS MAS in environmental samples, whose matrices are highly complex.

2. Experimental 2.1. Instrumentation All measurements for the development of the analytical method for chlorine determination via MgCl molecule were carried out in a high resolution continuum source atomic absorption spectrometer (HR-CS AAS) model ContrAA 700 (AnalytikJena, Jena, Germany), equipped with two options of atomization, graphite furnace and flame. All experiments were performed using transversal heating graphite tubes without integrated platform (Part. No. 407-A81.012, AnalytikJena). Solutions and samples were directly introduced into the graphite tube by an auto-sampler model MPE 60 (Analytik Jena). Measurements of the MgCl molecule intensity were carried out on the wavelength range between 376.80 nm and 377.20 nm. The acquiring of the analytical signal was achieved as peak volume absorbance integrated over three pixels at the most intense band at 377.010 nm. Argon gas 99.998% purity (White Martins, São Paulo, Brazil) was used for the graphite furnace protection and for purge during the temperature program execution.

2.2. Reagents and solutions

All reagents used were of analytical grade. Deionized water was produced using a Milli-Q system (Millipore, Beadford, MA, USA) to a resistivity of 18.2 M cm-1.

Nitric acid 65% (m m-1) (Vetec, Rio de Janeiro, Brazil) and hydrochloric acid 37% (m m-1) (Chemistry, São Paulo, Brazil) were distilled using a sub-boiling system, Distillacid, Berghof (Eningen, Germany), operating with an infrared lamp. Four chloride salts were used as source of Cl, MgCl2.6H2O (Cód. 01C2014.01.AG, Labsynth, São Paulo , Brazil) FeCl3.6H2O (Cód. AN09710RA, Exodus Scientific), CaCl2 (Cód. CC09654RA, Exodus Scientific) and NaCl (Cód. 3132, Vetec, Duque de Caxias, Brazil). These compounds were used to prepare standard solutions by dissolution of the salt in deionized water to a final concentration of 1 g L-1 of Cl without addition of acid. A magnesium nitrate standard solution, containing 5 g L-1 of Mg, was prepared by dissolution of 0.05 g of metallic magnesium (SPEX Industries, Edison, USA) in 0.5 mL of concentrated HNO3, and addition of high pure water to a final volume of 10 mL. Palladium solution containing 10 g L-1 was prepared from the solid Pd (NO3)2 (Merck Darmstadt, Germany) and used as chemical modifier. A 1.0 g L-1 W solution (Merck Darmstadt, Germany) was used in the graphite furnace coating previously described.

2.3. Environmental samples and quality control The developed method was applied to five samples of produced water from offshore oil wellbore. Certified reference materials, steam coal (BCR 182) and bovine muscle powder (NIST 8414), were acquired from Institute for Reference Materials and Measurements (IRMM, Geel, Belgium) and National Institute of Standards and Technology (NIST, USA), respectively, and used for evaluation of the accuracy and precision of the analytical method.

2.4. Procedures and sample preparation

To extend the life time of the graphite tube a coating procedure, adopted from literature [24], was carried out. A mass of 1000 µg of W was deposited by 25 consecutives injections of 40 µL of a 1.0 g L-1 W solution and subsequent undergoing to the temperature program shown in Table 1. Pyrolysis and vaporization temperatures were optimized using different standard solutions, FeCl3.6H2O, CaCl2, NaCl and MgCl2, as source of Cl, and Mg(NO3)2 solution

as forming agent. Volumes of 10 µL of the 1.0 g L-1 Cl standard solution and 20 µL of the 5.0 g L-1 Mg nitrate solution were introduced into the graphite furnace at each cycle of the optimization. Investigation using palladium as chemical modifier was also carried out by adding 10 µL of a 1.0 g L-1 Pd solution. The optimized temperature program for Cl determination is presented in Table 2. The procedure for CRMs steam coal (BCR 182) and bovine muscle powder (NIST 8414) sample treatment was achieved through microwave-assisted digestion using an equipment model MarsX Press (CEM Corporation, USA) in a two stage program, as shown in Table 3. A mass of approximately 0.5 g of CRMs was weighed directly into the PTFE flasks. Volumes of 3 mL of HNO3 65% (m m-1), 2 mL of H2O2 30% (m m-1) and 5 mL of deionized water were added into the flasks and after that they were submitted to the microwave oven program. After digestion, the solutions were transferred to polyethylene flasks and filled up to a final volume of 50 mL with deionized water. All samples were analyzed in triplicate, including blank solutions of reagents used for quality control. Addition and recovery tests were performed by addition of 5 mg L-1 of Cl to the reference material and produced water samples in order to evaluate the accuracy and precision of the method.

3. Results and Discussion 3.1. investigation on the MgCl molecule formation The use of molecules such as AlCl, InCl e SrCl [2], [3], [17], [19], [20], [21] for Cl determination by HR-CS MAS has been well reported. Thereby an investigation was carried out in an attempt to find a different molecule that could be used for analytical purpose. Several compounds were used to generate different Cl diatomic molecules, based on data from literature [25], in different wavelength ranges and to select the one with assured stability and intensity that would make it possible the optimization, validation and application of the analytical method for Cl determination in different types of environmental samples. Success was obtained when Mg was introduced as forming agent to generate the molecule MgCl at the wavelength of 377.010 nm as shown by the characteristic bands of the system A2 - X2+ [26]. The obtained

spectrum presents two strong (or intense) band heads, typical for MgCl molecule as reported by Pearse and Gaydon [25]. Figure 1 shows the absorption spectrum for the MgCl molecule.

3.3. Optimization of temperature program Pyrolysis temperature is a critical parameter for the development of an analytical method in HR-CS GF MAS [18]. In this work, an evaluation of the thermal stability of the MgCl molecule was carried out using four salt solutions containing Cl: FeCl3.6H2O, MgCl2.6H2O, and anhydrous salts CaCl2 and NaCl. Magnesium solution prepared from the metal and one of the Cl solution were simultaneously introduced into the graphite furnace to generate the MgCl molecule. Low temperature during the pyrolysis stage may not promote the total burning of the matrix and cause background interferences through the spreading of radiation by particles which is considered as absorption. On the other hand, high temperature may promote loss of analyte before the vaporization stage. Figure 2 presents pyrolysis and vaporization curves for the MgCl molecule obtained using different salts. The profile of the pyrolysis curves in Fig. 2A shows that the analyte has considerable thermal stability up to 800 °C, for all different salts, indicating the effectiveness of the molecule formation, during the vaporization step, and its good thermal stability even in absence of modifier. However, it was also observed that the sensitivities were not the same for the different solutions. Dittrich and Vorberg showed that the absorbance of molecules is affected by different elements. Some examples are Na, K and In that cause a strong reduction in the signal, Cu and Zn improve it greatly, while the effect of Ca and Sr are not so significant. They stated that the use of some elements could be used to improve the formation of MgCl molecule, making it possible the determination of chlorine free of interferences [23]. Vaporization step provides enough energy to generate the vapor cloud containing the analyte as MgCl molecules, which will be subject to vibrational and rotational transitions when interacting with the radiation.

Figure 2B shows the

vaporization curves obtained at a pyrolysis temperature fixed at 500 °C. The profile of all curves was very similar except for NaCl. Nevertheless higher intensity was achieved for MgCl2.6H2O. This fact is quite justifiable since the addition of solution containing Mg nitrate to the Mg salt solution allows the complete saturation of Cl that acts as the

limiting reagent. It is also noteworthy that the Mg-Cl dissociation energy is 318 kJ mol1

, lower the bond energy of the other compounds, FeCl, CaCl2 and NaCl. The two

factors, Cl saturation by Mg combined with dissociation energy, favored the increase of signal intensity for the MgCl2.6H2O salt. Based on this study, the use of MgCl2 solution with the addition of Mg nitrate solution was adopted to evaluate the thermal stability of MgCl molecular species through pyrolysis and vaporization curves, in the absence and in presence of Pd chemical modifier. Volumes of 20 µL of 5 g L-1Mg solutions and 0,5 g L-1 of Cl solution 10 µL of 1.0 g L-1 Pd solution were introduced into the graphite furnace each cycle of measurements to obtain the pyrolysis and vaporization curves. The use of Pd as chemical modifier promoted an increase in the sensitivity, probably due to the holding of Cl up to higher vaporization temperature promoting higher levels of molecular vibrational and rotational transitions. Therefore, the use of Pd as chemical modifier was adopted in the subsequent studies, and the compromise pyrolysis and vaporization temperature were 500 °C and 2500 °C, respectively.

3.4. Evaluation of the molecule formation

To establish the best ration between Cl and the forming agent Mg, an optimization using response surface methodology with Box-Behnken design involving three variables was used, encompassing a total of 15 experiments performed in duplicates. Three variables direct affecting the MgCl molecule formation were simultaneously optimized: Mg nitrate solution concentration (1,0 – 5,0 g L-1), chlorine concentration (5,0 – 25,0 g L-1) and nitric acid concentration (2,0 – 10,0 v v-1). The Acid nitric concentration was also optimized since the samples are submitted to digestion in HNO3 medium. Box-Behnken design for MgCl molecule formation and the responses obtained for the test through the integrated absorbance measurement are shown in Table 4. The design considered the use of MgCl2 as a source of chlorine. The Pareto chart obtained for the proposed design is shown in Fig. 3. It can be observed that all evaluated parameters are significant for the formation of the MgCl molecule. However, chlorine concentration is the most significant factor. This observation meets the original idea to establish the forming agent ration for the

formation of the molecule and chlorine concentration was found to be the main variable of the experimental design, also the reagent that will limit formation of MgCl molecular species. The interaction between chlorine and magnesium was also statistically significant for a 95% confidence level, an evidence that the formation of the molecule is favored, however dependent on the ratio between chlorine and magnesium concentrations. Box-Behnken design allows a mathematical model to assess the analytical response as a function of the variables studied. Through a quadratic equation a critical point is determined, which may be of maximum, minimum or saddle [27], [28], as shown by the response surface. Relationship between the variables, magnesium concentration (Mg), as nitrate, chloride concentration (Cl), nitric acid concentration (HNO3) and integrated absorbance per second (int. abs. / s) is described by Equation 03, as following: (Int. abs. / s) = 0.182645 + 0.034295 (Mg) + 0.008894 (Mg)2 + 0.160975 (Cl) + 0.044829 (Cl)2 + 0.012400 (HNO3) + 0.026104 (HNO3)2 + 0.056950 (Mg)(Cl) + 0.014150 (Mg)( HNO3) + 0.019450 (Cl)( HNO3)

(03)

The critical point was obtained through Equation 03 for the condition in which the partial derivative is null value. For MgCl molecular species the experimental conditions were of maximum integrated absorbance, since the surfaces obtained by the general equations indicate maximum points. The critical point value for this design, as shown in Table 4, indicates surfaces whose maximum points are not observed due to the interval of the chosen levels. The conditions obtained as critical point by Equation 03 was 16.6 g L-1 for standard Mg nitrate, 47.3 mg L-1 for chlorine and 12.5 % v v-1 for HNO3. In this condition, the amount of Mg added is approximately 350 times greater than the amount of chlorine. Thus, for the formation of MgCl molecule, chlorine is limiting reagent and all Cl containing in the solution introduced in the graphite furnace was converted in molecular specie. Figure 4 presents the response surfaces obtained for the MgCl, through BoxBehnken design, showing maximum surfaces with well-defined regions for the optimal conditions [29]. In this study, the conditions adopted for the method were those of

maximum points obtained for the interval evaluated, i.e Mg nitrate solution 5 g L-1 in a HNO3 2% v v-1 medium and a Cl concentrations at least up to 200 times lower than the concentration of Mg. Box-Behnken design was used as a chemometric tool to identify the optimal ratio between the forming agent and the analyte in order to ensure that all chlorine present in the standard solutions and the samples will react with Mg, added in excess, making it possible the use of the analytical method for analytical purposes.

3.5. Figures of merit

3.5.1. Calibration curve

External calibration was used for the determination of the Clorine. The calibration curve, consisting of one blank and six aqueous solutions, was prepared in a Cl concentration range between 5.0 and 25.0 mg L-1 in HNO3 2% (v v-1) medium, at the conditions described in Section 2. The optimized conditions for graphite furnace were 500 °C and 2500 °C for pyrolysis and vaporization temperature, respectively. Evaluation of the data set used in the preparation of the calibration curve was carried out in order to identify potential deviations called Grubbs’ Outlier. No discrepant values were observed, thus the data set was used in the calibration curve [30]. Linearity, usually evaluated as correlation coefficient (r), represents the dependence of the response in concentration (mg L-1) as a function of the integrated absorbance (s). The correlation coefficient value (r) obtained from the data of the calibration curve was 0.9986, therefore 99.86% of the relation described above can be explained [31]. The coefficient of determination of 99.72% (R2 = 0.9972) obtained ensures the low dispersion of the residuals. Since under the optimized conditions, for concentrations of Cl higher than 25 mg L-1 a loss in linearity was observed, the linear calibration range was considered up to 25 mg L-1 of chlorine. This concentration is lower than those reported in other papers, but still high enough to not require large dilution of the sample prior the determination. Assessment using F Snedecor’s test was applied to the obtained data regarding the lack of fit and the significance of the linear regression and the Fcalculated values of 3.35 and 2.51, respectively, confirm the good linearity of the analytical method for chlorine determination by HR-CS MAS. Limits of detection (LoD) and quantification (LoQ) were calculated as three and ten times the standard deviation of 10 measurements of the blank solutions divided by the slope of the calibration curve, respectively [32]. Characteristic mass (m0), defined as the analyte mass corresponding to an integrated absorbance of 0.0044 s, obtained for Cl in 377.010 nm wavelength was 7.1 ng. These results are comparable to those reported in the literature [33]. Since MgCl molecule was applied for the first time for analytical purposes and considering that the molecule formation conditions were optimized through an experimental design, the proposed method are of significance since the

figures of merit achieved values similar to those reported by others papers. Figures of merit are shown in Table 5.

3.5.2. Accuracy, precision and analytical application

After the evaluation of the figures of merit, seven samples of environmental interest were submitted to the method for Cl determination, two certified reference materials BCR 182 and NIST 8414 and five samples of produced water from an offshore oil platform. The certified samples were prepared following the digestion procedure described in Experimental section. For NIST 8414, biological material, the procedure accomplished the total dissolution of the sample, but for BCR 182, steam coal, the digestion was not complete remaining solid particles at the end of the process. Nevertheless the total extraction of the analyte was achieved as confirmed by the obtained results. For produced water samples no treatment was applied, only dilution. Due to the high content of Cl in the samples a 100 times dilution was necessary, but for one of them (AM-2) a 1000 times dilution was required. The results of the determination of Cl via MgCl by HR-CS MAS in certified reference material and produced water samples are presented in Table 6. The obtained values for the certified samples, coal steam (BCR 182) and bovine muscle powder (NIST 8414), were in good concordance with the certified values, 88 ± 11% and 81 ± 5% of agreement, respectively. Precision, expressed as relative standard deviation (RSD), was 12.3% for BCR 182 and 6.6 % for NIST 8414 (n=3), assuring the precision and accuracy of the analytical method. Addition and recovery tests were applied to the CRMs by the addition of 5 mg L1

of Cl since the concentration in these samples are lower than in real samples of

produced water. Good recoveries were achieved with values between 102 and 112%, considering a range within the 80-120% interval as an acceptable criterion [34].

4. Conclusion Determination of Cl through diatomic molecule MgCl at the wavelength of 377.010 nm by HR-CS MAS in environmental samples is the main goal of this work. Investigations were carried out and the use of Mg nitrate solution was effective as

forming agent to promote the generation of MgCl molecule. Graphite furnace without platform was used and tungsten coating promoted an increase in the lifetime of the tube, enabling almost twice more burning cycles, when compared with graphite furnace with integrated platform. The use of 10 g of Pd as chemical modifier was effective in increasing the signal intensity and it was adopted for the determination of analyte in the samples. Box-Behnken design was applied to evaluate the optimum concentration of the reactants used to generate the molecule MgCl and it was established that the chlorine was the limiting reagent and the minimum ratio necessary between Cl and Mg is 1 Cl : 200 Mg. This ratio ensures sufficient excess of the agent forming to generate the molecular species of the total content of chlorine. Accuracy of the method was evaluated through submission of two certified reference material BCR 182 and NIST 8414 and addition and recovery tests and good agreements were achieved. The method was applied to five samples of produced water, considered high complex samples, .The analytical method proposed in this work is fast, effective with good accuracy and precision for chlorine determination via MgCl molecule by the HR-CS MAS in for environmental samples with high complex matrices.

Acknowledgements

The study was financially supported by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), which provided fellowships, financial support and infrastructure.

References [1] B. Welz, F. G. Lepri, R. G. O. Araujo, S. L. C. Ferreira, M. D. Huang, M. Okrussc, H. Becker-Ross, Determination of phosphorus, sulfur and the halogens using high-temperature molecular absorption spectrometry in flames and furnaces - A review, Anal. Chim. Acta, 647, 2 (2009) 137-148. [2] M. Fechetia, A. L. Tognon, M. A. M. S. da Veiga, Determination of chlorine in food samples via the AlCl molecule using high-resolution continuum source molecular absorption spectrometry in a graphite furnace, Spectrochim. Acta Part B, 71 (2012) 98-101. [3] M. D. Huang, H. Becker-Ross, S. Florek, U. Heitmann, M. Okruss, Determination of halogens via molecules in the air-acetylene flame using highresolution continuum source absorption spectrometry, Part II: Chlorine, Spectrochm. Acta Part B, 61 (2006) 959-964. [4] M. A. Fender, D. J. Butcher, Comparison of deuterium arc and Smith-Hieftje background correction for graphite furnace molecular absorption spectrometry of fluoride and chloride, Anal. Chim. Acta, 315, 1 (1995) 167-176. [5] E. L. M. Flores, J. S. Barin, E. M. M. Flores, V. L. Dressler, A new approach for fluorine determination by solid sampling graphite furnace molecular absorption spectrometry, Spectrochim. Acta Part B, 62, 9 (2007) 918-923. [6] A. P. Soldatkin, D. V. Gorchkov, C. Martelet, N. Jaffrezic-Renault, New enzyme potentiometric sensor for hypoclorite species detection, Sensor Actuar B, 43 (1997) 99-104. [7] P. Salazar, M. Martín, J. L. González-Mora, A. R. González-Elipe, Application of Prussian Blue electrodes for amperometric determination of free chlorine in water samples in water samples using Flow Injection Analysis, Talanta, 146 (2016) 410-416. [8] M. Jović, F. Cortés-Salazar, A. Lesch, V. Amstutz, H. Bi, H. H. Girault, Electrochemical detection of free chlorine at inkjet printed silver electrodes, J. Electroanal. Chem. 756 (2015) 171-178.

[9] E. L. C. Silveira, L. B. de Caland, M. Tubino, Simultaneous quantitative analysis of the acetate, formate, chloride, phosphate and sulfate anions in biodiesel by ion chromatography, Fuel, 124 (2014) 97-101. [10]

J. S. de Gois, É. R. Pereira, B. Welz, D. L. G. Borges, Simultaneous

determination of bromine and chlorine in coal using electrothermal vaporization inductively coupled plasma mass spectrometry and direct solid sample analysis, Anal. Chim. Acta, 852 (2014) 82-87. [11]

J. S. F. Pereira, P. A. Mello, D. P. Moraes, F. A. Duarte, V. L. Dressler,

G. Knapp, É. M. M. Flores, Chlorine and sulfur determination in extra-heavy crude oil by inductively coupled plasma optical emission spectrometry after microwave-induced combustion, Spectrochm. Acta Part B, 64 (2009) 554-558. [12]

A. Doyle, A. Saavedra, M. L. B. Tristão, M. Nele, R. Q. Aucélio, Direct

chlorine determination in crude oils by energy dispersive X-ray fluorescence spectrometry: An improved method based on a proper strategy for sample homogeneization and calibration with inorganic standards, Spectrochm. Acta Part B, 66 (2011) 368-372. [13]

J. H. Seo, M. Guillong, M. Aerts, Z. Zajacz, C. A. Heinrich,

Microanalysis of S, Cl and Br in fluid inclusions by LA-ICP-MS, Chem. Geol. 284 (2011) 35-44. [14]

B. Welz, H.Becker-Ross, S. Florek, U. Heitmann, High-Resolution

Continuum Source Atomic Absorption Spectrometry – The Better Way to do Atomic Absorption Spectrometry, Weinhein: Wiley – VCH, 2005. [15]

M. Resano, E. García-Ruiz, High-resolution continuum source graphite

furnace atomic absorption spectrometry: Is it as good as it sounds? A critical review, Anal. Bioanal. Chem. 399, 1 (2011) 323-330. [16]

H. Haraguchi, K. Fuwa, Atomic and molecular absorption spectra of

indium in air-acetylene flame, Spectrochim. Acta Part B, 30 (1975) 535-545. [17]

U. Heitmann, H. Becker-Ross, S. Florek, M-D. Huang, M. Okruss,

Determination of non-metals via molecular absorption using high-resolution

continuum source absorption spectrometry and graphite furnace atomization, J. Anal. Atom. Spectrom. 21, 11 (2006) 1314-1320. [18]

D. J. Butcher, Molecular absorption spectrometry in flames and furnaces:

A review, Anal. Chim. Acta, 804 (2013) 1-15. [19]

F. V. Nakadi, M. A. M. S. da Veiga, M. Aramendia, E. García-Ruiz, M.

Resano, Chlorine isotope determination via the monitoring of the AlCl molecule by high-resolution continuum source graphite furnace molecular absorption spectrometry- a case study, J. Anal. Atom. Spectrom. 30 (2015) 1531-1540. [20]

E. R. Pereira, B. Welz, A. H. D Lopez, J. S. Gois, G.F. Caramori, D.

L.G. Borges, E. Carasek, J. B. Andrade, Strontium mono-chloride — A new molecule for the determination of chlorine using high-resolution graphite furnace molecular absorption spectrometry and direct solid sample analysis, Spectrochim. Acta Part, B 102 (2014) 1-6. [21]

E. R. Pereira, L. M. Rocha, H. R. Cadorim, V. D. Silva, B. Welz, E.

Carasek, J. B. de Andrade, Determination of chlorine in coal via the SrCl molecule using high-resolution graphite furnace molecular absorption spectrometry and direct solid sample analysis, Spectrochm. Acta Part B: Atomic Spectroscopy (2015), doi: 10.1016/j.sab.2015.10.001 [22]

Ozbek N and Akman S, Determination of Chlorine in Milk via

Molecular Absorption of SrCl Using High-Resolution Continuum Source Graphite Furnace Atomic Absorption Spectrometry, J. Agric. Food. Chem., 64 (2016) 5767-5772. [23]

K. Dittrich,

Elektrothermischer

B.ernd Vorberg, Molekülabsorptionsspektrometrie Verdampfung

in einer

bei

Graphitrohrküvette, Analytica

Chimica Acta, 140 (1982) 237-243 [24]

M. Resano, J. Briceño, M. Aramendía, M. A. Belarra, Solid sampling-

graphite furnace atomic absorption spectrometry for the direct determination of boron in plant tissues, Anal. Chim. Acta, 582, 2 (2007) 214-222.

[25]

R. W. B. Pearse, A. G. Gaydon, The Identification of Molecular Spectra,

Fourth Edition, Imperial College, London, Chapman and Hall, (1976) 217-219. [26]

T. Hirao, P. F. Bernath, C. E. Fellows, R. F. Gutterres, M. Vervloet,

High-resolution Fourier transform study of MgCl: The A2Π-X2Σ+ band system, J. Mol. Spectrosc. 212 (2002) 53-56. [27]

M. A. Bezerra, R. E. Santelli, E. P. Oliveira, L. S. Villar, L. A. Escaleira,

Response surface methodology (RSM) as a tool optimization in analytical chemistry, Talanta 76 (2008) 965-977. [28]

S. L. C. Ferreira, R. E. Bruns, H. S. Ferreira, G. D. Matos, J. M. David,

G. C. Brandão, E. G. P. da Silva, L. A. Portugal, P. S. dos Reis, A. S. Souza, W. N. L. dos Santos, Anal. Chim. Acta 597 (2007) 179-186. [29]

B. B. Neto, I. S. Scarminio, R. E. Bruns, Como fazer experimentos, 4 ed.

Editora Bookman (2010). [30]

J. N. Miller and J. C. Miller, Statistics and chemometrics for analytical

chemistry, 5th ed. Pearson Education Limited (2005). [31]

INMETRO. DOQ-CGCRE-008, Orientação sobre validação de métodos

analíticos.

3ª

revisão,

February,

2013.

Available

in:

http://www.inmetro.gov.br/Sidoq/Arquivos/CGCRE/DOQ/DOQ-CGCRE8_03.pdf. (accessed in: october 15, 2015). [32]

IUPAC Analytical Chemistry Division, Spectrochim. Acta Part B, 33

(1978) 247-269. [33]

B. Welz, H. Becker-Ross, S. Florek, U. Heitmann, High-Resolution

Continuum Source AAS; Wiley-VCH: Weinheim, 2005. [34]

International Organization for Standardization (ISO), ISO 5725-1.

Accuracy (trueness and Precision) of measurement method and results. Part 1. General principles and definitions, ISO, Geneva, Switzerland, 1994.

Table Captions

Table 1 . Temperature program for the graphite furnace coating with W.

Step

Temperature (°C)

Ramp (°C s-1)

Hold (s)

1

120

5

25

2

150

10

60

3

600

20

15

4

1000

300

15

5

1500

1500

5

6

2000

500

5

Table 2. Graphite furnace temperature program for Cl determination via MgCl molecule and Pd as chemical modifier in environmental samples by HR-CS GF MAS.

Step

Temperature (°C)

Ramp (°C/s)

Hold (s)

Drying

90

3

10

Drying

110

5

15

Pyrolysis

500

300

10

Vaporization

2500

1500

5

Cleaning

2650

500

4

Table 3. Microwave program for acid digestion of the reference materials BCR 182 and NIST 8414 for chlorine determination by HR-CS MAS.

Step

Temperature (°C)

Power (W)

Hold (min)

1

175

900

04:30

2

-

0

01:00

Table 4 - Box-Behnken experimental design for optimization of the conditions for MgCl molecule formation.

Exp.

Mg

Cl

HNO3

1

-1

-1

2

1

3

Mg conc. -1

Cl conc. -1

HNO3 conc. -1

Abs.

Abs.

(g L )

(mg L )

(% v v )

Int.

Int.

0

1.0

5.0

6

0.1054

0.1082

-1

0

5.0

5.0

6

0.0952

0.1062

-1

1

0

1.0

25.0

6

0.2064

0.2068

4

1

1

0

5.0

25.0

6

0.3076

0.3212

5

-1

0

-1

1.0

15.0

2

0.1980

0.2018

6

1

0

-1

5.0

15.0

2

0.2051

0.2020

7

-1

0

1

1.0

15.0

10

0.1764

0.1916

8

1

0

1

5.0

15.0

10

0.2131

0.2188

9

0

-1

-1

3.0

5.0

2

0.0784

0.0792

10

0

1

-1

3.0

25.0

2

0.2119

0.2372

11

0

-1

1

3.0

5.0

10

0.0856

0.0862

12

0

1

1

3.0

25.0

10

0.2532

0.2879

13

0

0

0

3.0

15.0

6

0.2199

0.2653

14

0

0

0

3.0

15.0

6

0.2333

0.2338

15

0

0

0

3.0

15.0

6

0.2119

0.2511

Table 5. Figures of merit of the proposed analytical method for Cl determination via MgCl by HR-CS MAS.

Samples

Certified Values

Obtained Values by

RSD

Rec

HR-CS MAS

(%)

(%)

Working Range (mg L-1)

5 - 25

Correlation Coefficient

0.9986

Slope (s L mg-1)

0.00616

LOD (mg L-1)

1.7

LOQ (mg L-1)

5.0

RSD (%)

< 5%

Table 6 . Chlorine concentration in certified samples and real samples.

BCR 182

3.70 ± 0.07 g Kg-1

3.25 ± 0.40 g kg-1

12

107

NIST 8414

0.188 ± 0.015 %

0.152 ± 0.010 %

7

81

Produced Water 1

-

0.204 ± 0.010 % m/v

5

91

Produced Water 2

-

2.157 ± 0.013 % m/v

0.6

100

Produced Water 3

-

0.213 ± 0.002 % m/v

1

81

Produced Water 4

-

0.160 ± 0.003 % m/v

2

95

Produced Water 5

-

0.139 ± 0.012 % m/v

8

nd

Figure Captions Fig. 1. 3D Spectrum obtained through HR-CS GF MAS for the MgCl molecule.

Fig. 2. Pyrolysis and vaporization curve for MgCl molecule specie with the addition of Mg nitrate solution and different salts solutions as source of Cl. A) Pyrolysis curves, vaporization temperature 2200 oC. B) Vaporization curves, pyrolysis temperature 500 o

C.

Fig. 3. Pareto chart for the 23 statistical planning of the MgCl molecular formation for the parameter of the significance analysis.

Fig. 4. Surface responses obtained through Box-Behnken experimental design for MgCl molecule formation.

Highlights  Application of MgCl molecule for chlorine determination by HR-CS MAS  Chemometric tools to optimize the conditions for MgCl formation  Chemical modifier Pd to improve the molecule stabilization  Environmental samples