Method development for the determination of sulfur by high resolution continuum source atomic absorption spectrometry

Sulfur, one of the most abundant elements on earth, is a necessary and important mineral for people, plants and animals. Sulfur is used in many forms of life, such as food, petrochemical, environment (air, water, soil), agriculture (soil fertility, medicines) and biological samples. In addition, the use of various additives is inherently or indirectly involved in the structure of many food nowadays. The determination of sulfur content, concentration and compliance with the limit values determined by the regulations are great importance all over the world. Therefore, the most practical and appropriate method for precise, accurate and reliable sulfur determination must be selected and applied. Sulfur can not be directly determined by atomic absorption spectrometer (AAS) because the main atomic absorption resonance line is located in the vacuum ultraviolet (UV) region where the air components are also absorbed. However, total sulfur can be indirectly determined by means of diatomic sülfür molecules formed in the gas phase using high resolution continuum source atomic absorption spectrometry (HR-CS AAS) developed and marketed in recent years. For this purpose, the first experimental studies started with the optimization of flame conditions for sulfur analysis with diatomic carbon monosulfite (CS) molecules, which are tried to be formed in the gas phase using the flame HR-CS AAS. For this purpose, the necessary instrumental and experimental parameters were investigated and analytical parameters were determined. Firstly, formation of CS molecules in different flame types such as air-acetylene and nitrous oxide-acetylene was achieved. The effect of gas flow rate and flame height on two different flame types, were investigated at 257.593 nm, 257.959 nm and 258.056 nm. In the scope of experimental studies on the flame atomizer, firstly different inorganic and organic sulfur compounds (ammonium sulfate, sodium sulfite, potassium disulfate, thiourea, L-cysteine and L-cystine) were used for the selection of the wavelengths in the sulfur analysis using the CS molecules. In the air acetylene flame, analytical parameters such as linear boundaries, detection limit (LOD) and characteristic mass (m0) were determined for three different wavelengths for each sulfur compound and their different pixels. The result of achieving low LOD values is chosen as the optimum working pixel range of the center pixel ± 1 (CP ± 1). Similarly, analytical results were also obtained in the nitrous oxide-acetylene flame only at CP ± 1 pixel per wavelength for each of different 3 wavelengths. When the results obtained from different flames are evaluated, air-acetylene is selected as the most suitable flame type because of the fact that lower detection limits are reached. Experimental optimization studies were continued by investigating the sensitivities of the various sulfur compounds at different concentrations and CS wavelengths. There was no significant difference between the sensitivities of 6 different sulfur compounds at a certain concentration. At optimum conditions, sulfur concentrations of four different certified reference materials (CRM) such as spinage (NCS ZC73013 Spinage), milk powder (NCS ZC73015 Milk powder), chicken (NCS ZC73016 Chicken) and whole egg powder (NIST RM 8415 Whole egg powder) were determined at 3 different absorption wavelengths. The results obtained at 258,056 nm were found to be consistent with the certificate values in the 95 % confidence interval, while their measurements obtained with some calibration standards at other wavelengths were found to be outside of certificate values for the SRM. Therefore, in the case of the flame HR-CS AAS conditions, 258,056 nm was determined to be the most suitable wavelength and the accuracy of the method was revealed. Using the developed method, sulfur concentrations of various brands of dried fruits (fig, mulberry, apricot, plum and their fumigated with SO2) and vegetables such as garlic and onion supplied from the markets in Istanbul were determined with the aid of diatomic CS molecules formed in gas phase under the optimum conditions in flame HR-CS AAS. In the second part of the thesis, studies were made on HR-CS AAS using graphite furnace (electrothermal) atomizer. Unlike the flame atomizer studies, sulfur determinations were carried out in graphite furnace HR-CS AAS (HR-CS GFAAS) by taking advantage of the simultaneous measurement of absorbances at 2 different wavelengths of the CS diatomic molecule (257.959 nm and 258.056 nm). Optimization studies have been started with the investigation of the thermal behavior of different sulfur compounds such as potassium disulfate, thiourea, L-cystine and sodium sulphate and different SRMs such as tea, milk powder and spinach for selecting of suitable calibrator. From the pyrolysis and molecular formation curves obtained, the thiourea as calibration standard was chosen to show similar behavior and sensitivity to SRMs. The pyrolysis temperature was set at 800 °C, since graphite tubes and platforms are coated with a permanent modifier and the high sensitivity at between from 400 to 900 °C for CS molecules does not show any significant change. In addition, the molecular formation step and the effect of heating rate in this step on the sensitivity of CS have been investigated. Thus, it was observed that the highest peak area was obtained at temperature of 2200 °C and the heating rate of 3000 ° C sec-1. It was investigated that the coating of graphite tubes/platforms with tungsten (W) and zirconium (Zr) and the use of chemical modifiers such as palladium (Pd), Pd / Mg, Pd / citric acid, nano silver, nano gold particles effect on absorbance signals. For both the thiourea and GBW 07605 Tea SRM sample, the signals with and without modifier were evaluated. It was obtained that smooth, reproducible and stable peaks, W were added under conditions the permanent modifier coated graphite tube / platform added 40 μg Pd and 300 μg citric acid as chemical modifier. In the HR-CS GFAAS studies, the linearity (r>0.998) obtained with the thiourea aqueous standard solution for the CS molecule continued to 2500 ng of sulfur. In addition, the effect of simultaneous evaluation of different CS absorption lines on the LOD and m0 values of the method was also investigated. The LOD and m0 values obtained by summing the absorbances at wavelengths of 257.959 nm and 258.056 nm are 21.6 ng and 7.4 ng sulfur, respectively, which are lower than the values obtained at a single wavelength. Using this method developed with the CS diatomic molecule formed in the gas phase of the graphite furnace, the sulfur concentrations in some dried fruits and nuts were determined as well as different SRM samples such as tea, milk powder and spinach. A standard addition to each dry apricot and hazelnut sample has been proven by quantitatively recovering (>95%) the amount of sulfur added to specimens without any matrix interference effects. Moreover, the accuracy of the method has been demonstrated by the fact that the certified values of the sulfur concentrations of the SRM samples are within the uncertainty limits. Finally, sulfur concentrations of different brands of dried fruit (grape, apricot, plum, and their fumigated with SO2) and dried nuts such as hazelnut and almond obtained from various markets in Istanbul have been determined in HR-CS GFAAS using linear calibration method. In the last part of the thesis works, a new method has been developed for direct determination of sulfur concentrations of various vegetables, without any digestion process, by combining advantageous of solid sampling technique with diatomic CS molecules formed in gas phase in graphite furnace using solid sampling HR-CS GFAAS. In the HR-CS GFAAS studies with solid sampling, measurements were made using molecular absorption of the CS molecule at a wavelength of 258.056 nm. Pyrolysis and molecular formation curves were obtained using solid spinach SRM in addition to thiourea and thioacetamide sulfur aqueous calibration standards, and thus thermal behaviors of the samples were investigated. When the modifier coated graphite furnace / solid platform and the chemical modifier were used, the pyrolysis temperature was determined as 1000 ºC and CS molecules were formed at 2400 ºC (heating rate 3000 ºC sn-1 at the molecular formation step). Since standards and SRM show similar thermal behavior, different from the second part of the thesis, calibration standards can not be choosen in this step. In addition, graphite tubes and platforms in solid sampling HR-CS GFAAS were coated with suitable temperature programs for various refractive metals such as W, Zr, iridium (Ir) and ruthenium (Ru) and the effects of the using various modifiers such as Pd, Pd + Ru and Pd + citric acid on the signals have been investigated. Thus, the most stable and reproducible signals were achieved with solid platforms coated with Ru permanent modifier and with the use of 40 μg Pd and 300 μg citric acid as modifiers. While the linearity of the thioacetamide aqueous solution for the CS molecule was proceed up to 1500 ng of sulfur, the solid samples which were dried and pulverized were placed directly on graphite furnace platforms in the range of 0.05 mg to 0.50 mg linear working range. In addition, various certified powdered samples such as tea, spinach, chicken and bushes & leaves (NCS DC73349) were placed directly into the graphite furnace platform and sulfur concentrations were determined using two different standards, thiourea and thioacetamide. According to the results, it has been decided that thioacetamide should be selected as the calibrant to be used in the study, as both the accuracy was better and the low LOD (7.5 ng S) and m0 (8.7 ng S) were obtained. Using this method developed for solid sampling in graphite furnace conditions, homogeneity, homogeneity factor (HE) was determined at micro scale. The sulfur concentrations of vegetable samples such as powdered spinach, leeks, zucchini, lettuce, chard, radish, brussel sprouts were directly determined without any digestion process. Thus, the direct analysis of solid samples with the solid sampling method was eliminated some of the risks associated with sample digestion process, and a new rapid and reliable method for the determination of sulfur was developed.