A thesis submitted for the degree of Doctor of Philosophy in the Division of Health Science at the University of South Australia

Transcription

1 Selenium speciation in environmental and biological systems using liquid chromatographic techniques hyphenated with Inductively Coupled Plasma Mass Spectrometry (ICP-MS) by Ms Wei Hong Wang B. Eng., in Industrial Analysis (Chengdu University of Technology, China) M. Sc., in Analytical Chemistry (Sichuan University, China) A thesis submitted for the degree of Doctor of Philosophy in the Division of Health Science at the University of South Australia School of Pharmacy and Medical Sciences City East Campus, South Australia, 5000 CRC for Contamination Assessment and Remediation of the Environment (CRC-CARE) Centre for Environmental Risk Assessment and Remediation (CERAR) Mawson Lakes Campus, South Australia, 5095 University of South Australia March 2010

2 Table of Contents List of Figures... v List of Tables... viii List of Publications and Manuscripts... ix Abbreviations... x Declaration... xii Acknowledgments... xiii Abstract... xiv Chapter 1 Introduction... 1 Chapter 2 Selenium speciation The Role of Selenium in our World Speciation Analytical characterisation of Se species Sample extraction The continuing search for improved trace level detection Hydride atomic spectroscopic methods ICP-MS sensitive detection Separation techniques for Se species and applications Electrophoresis Gas chromatography Size-exclusion chromatography Ion-exchange chromatography (IEC) Ion-pair reversed phase chromatography (IP-RPC) Conclusions References Chapter 3 Implementation of instrumentation i

3 3.1 Introduction ICP-MS and Collision/Reaction Cell technology Interfacing of Liquid Chromatography (LC) and ICP-MS Optimising the ICP-MS ICP-MS parameters Tuning reaction gas (H 2 and He) under ORS References Chapter 4 Inorganic selenium speciation using ion chromatography with ICP-MS and an octopole reaction system Introduction Experimental Chemicals and solutions IC-ICP-MS conditions Results and discussion ICP-MS detection Evaluation of no reaction gas, H 2 and He added to the cell Ion chromatographic separation of Se(IV) and Se(VI) Sample applications Conclusions References Chapter 5 Ion-exclusion chromatographic separation of inorganic selenium species in water and detection using ICP-MS Introduction Experimental Chemicals and Solutions Instrument and Analytical Conditions Results and discussion Ion-exclusion chromatographic separation of SeO 2-3 and SeO ii

4 5.3.2 Sample applications Conclusions References Chapter 6 Extraction of selenium species in pharmaceutical tablets using enzymatic and chemical methods Introduction Experimental Chemicals and solutions Total Se determination Extraction Instrumentation Results and Discussion Determination of total selenium content by ICP-MS IC-ICP-MS for the speciation of selenium The extraction of selenium using SDS solution The extraction of selenium using enzymes Comparing single and dual enzymatic hydrolysis Effect of Tris-HCl buffer Se species in pharmaceutical tables Conclusions References Chapter 7 Selenium speciation in nutritional tablets and biofortified foods using ion chromatography with ICP-MS Introduction Experimental Chemicals and solutions Total Se determination Extraction iii

5 7.2.4 Instrumentation Results and discussion Protonation state of Se species in aqueous solution Anion-exchange chromatographic separation of selenium ICP-MS detection of 78 Se and 80 Se with an ORS Analytical performance characteristics and application Conclusions References Chapter 8 Selenium speciation of biofortified foods using ion-pair reversed phase chromatography with ICP-MS Introduction Experimental Chemicals and solutions Total Se determination Extraction Instrumentation Results and discussion Comparing ion-pairing reagents Effect of mobile phase concentration ph effect Applications Conclusions References Chapter 9 Conclusions and Future Trends Conclusions Future trends References Appendix iv

6 List of Figures Figure 1.1 A diagrammatic overview of the thesis structure... 5 Figure 2.1 Pathways of Se metabolism, following (a) Toxic and (b) Sub-toxic doses Figure 2.2 Cation exchange HPLC separation of mixture of 10 Se standard compounds.. 29 Figure 2.3 Ion-pair reverse phase chromatogram of a mixture of 5 Se standards with ICP- MS (5 ng/ml of each compound; 78 Se monitored) Figure 3.1 Schematic of ICP-MS processes Figure 3.2 Schematics of an Agilent ICP-MS with Figure 3.3 Schematics of DIHEN Figure 3.4 (a) Babington nebuliser; (b) Scott double pass spray chamber as used in the Agilent 7500c ICP-MS instrument Figure 3.5 Tuning the ORS for Se78 and Se 80. Influence of H2 (a) and He (b) cell gas flow rates. Detector response (CPS) variation: Blank in blue, spiked 10 μg/l Se standard in red and background equivalent concentration (BEC) μg/l in green Figure 4.1 Chromatogram obtained from no reaction gas, H 2 (4.0 ml/min), He (4.5 ml/min) as the collision/reaction gas and detection at m/z 78. The mobile phase: 20 mm NH 4 NO 3, 10 mm NH 4 H 2 PO 4 at ph 7.0. Concentration of both Se(IV) and Se(VI) is 200 µg/l Figure 4.2 Chromatogram obtained from (a) no reaction gas and detection at m/z 80, (b) H2 (4.0 ml/min) and He (4.5 ml/min) as the collision/reaction gas, other conditions as in Fig Figure 4.3 Effect of the eluent ph on the retention time of Se(IV) and Se(VI). Mobile phase: 20 mm NH 4 NO 3, 10 mm NH 4 H 2 PO 4, H 2 (4.0 ml/min) as the reaction gas Figure 4.4 Dissociation diagram for selenous acid on the ph range from 0 to Figure 4.5 (a) Water solution, (b) soil solution. Mobile phase: 20 mm NH 4 NO 3, 10 mm NH 4 H 2 PO 4 at ph 6.5, H 2 (4.0 ml/min) as the reaction gas, detection at m/z Figure 5.1 Ion exclusion chromatogram with a polystyrene-divinylbenzene resin (HPX-87 H 30 cm X 7.8 mm i.d., 5µm. eluent at 1.0 ml/min and 30ºC. Sample, 50 µl of Se(IV) and Se(VI) standards (500 µg/l each). H 2 flow rate of 2.0 ml/min, detection at m/z (a) Using 5 mm acetic acid as mobile phase. (b) Using 5 mm sulphuric acid as the mobile phase Figure 5.2 Calibration plots for Se(VI) and Se(IV) using ion-exclusion chromatography with 5 mm sulphuric acid v

7 Figure 5.3 Ion exclusion chromatogram obtained for contaminated water from Bangladesh using aqueous sulphuric acid as the mobile phase. H 2 flow rate of 3.0 ml/min, detection at m/z 80. Other conditions were the same as in Fig. 5.1 (b) Figure 6.1 Anion-exchange HPLC-ICP-MS profile for a 200 µg L -1 Se standard mixture containing SeCys 2 (1), SeMC(2), SeMet(3), selenite(4) and selenate(5), 20 mm NH 4 H 2 PO 4 at ph 6.5 eluent, m/z 78 monitored Figure 6.2 Extraction of Se species by 4% SDS from the SeMC TM tablet Figure 6.3(a) Extraction of Se species by 20 mg protease XIV and 10 mg lipase VII from the SeMC TM tablet, Se detected at m/z Figure 6.3(b) A 150 µg L -1 mixture of each of five Se species standards were mixed with the extracted protease XIV and lipase VII solution from the SeMC TM tablet (1:1), Se detected at m/z Figure 7.1 Species distribution diagram for SeCys 2 in the ph range from 0 to Figure 7.2 Species distribution diagram for SeMet in the ph range from 0 to Figure 7.3 Species distribution diagram for cysteine in the ph range from 0 to Figure 7.4 The influence of the concentration of NH 4 H 2 PO 4 on the retention of Se species at ph Figure 7.5 The influence of mobile phase ph on the retention of Se species Figure 7.6 Comparison of H 2, He gas added in reaction cell and no reaction gas (Ar alone) for detecting 78 Se. Octopole conditions: H 2 flow rate of 3.0 ml/min and He flow rate of 3.5 ml/min. Injection of mixture of five Se species standards each at level of 500 µg/l in Se containing SeCys 2 (1), SeMC(2), SeMet(3), selenite(4) and selenate(5) Figure 7.7 The impact of reaction cell gas selection on sensitivity for 80 Se. Flow rates and chromatographic conditions were the same as in Fig Reaction cell gas was H 2 or He Figure 7.8 Comparison of tuning methods for detecting Se species by Ion chromatography ICP-MS: Argon alone no reaction gas ; H 2 ; and He. Chromatographic conditions as in Fig Each Se species is at level 500 g/l in Se Figure 7.9 A typical chromatogram for separation of selenium species, detection at m/z 80. Octopole H 2 flow rate of 3.0 ml/min. Chromatographic conditions and species numbering as in Fig Figure 7.10 Calibrations for Se five species vi

8 Figure 7.11 Chromatograms obtained from commercial products. Chromatographic conditions as in Fig Samples were: (a) nutritional tablets (Se-no-yeast TM ); (b) biofortified food (Wafer grains biscuits) Figure 8.1 Chromatogram of reversed phase chromatogram on C18 column (250 mm x 4.6 mm, eluent with 1% methanol at 1.0 ml/min and 25ºC. sample, 30µL 1µg/mL in each of SeCys 2 (1), SeMC(2), SeMet(3), Se(IV) (4)and Se(VI)(5)). Octopole H 2 flow rate of 2.5 ml/min. Detection Se at m/z Figure 8.2 Chromatogram of 1 mm TEA-HFOS as ion-pairing reagent in 1% methanol as mobile phase on short C18 column (150 mm x 4.6 mm). Other conditions were the same as in Fig Figure 8.3a Using 1mM SOS in 1% MeOH as the ion-pairing reagent, ph Figure 8.3b Using 1 mm PHFOS in 1% MeOH as the ion-pairing reagent, ph Figure 8.3c Using 0.1% HFBA in 1% MeOH as the ion-pairing reagent, ph Figure 8.3d Using 0.1% TFA in 1% MeOH as the ion-pairing reagent, ph Figure 8.3e Using 1mM TBAH in 1% MeOH as the ion-pairing reagent, ph Figure 8.3f Using 1mM TEA-HFOS in 1% MeOH as the ion-pairing reagent, ph Figure 8.3 Comparing chromatograms of different ion-pairing reagents and 1% methanol as mobile phase on a 25 cm C18 column, Se detection at m/z 78. Other conditions were the same as in Fig Figure 8.4 The influence of the concentration of TEA-HFOS on the retention of Se species Figure 8.5 The influence of mobile phase ph on the retention of Se species. Mobile phase: 1 mm TEA-HFOS and 1% MeOH Figure 8.6 Chromatogram obtained for biofortified Laucke flour sample using TEA-HFOS as ion-pairing reagents on C18 column (250 mm x 4.6 mm), detected as 80 Se. Other conditions were the same as in Fig vii

9 List of Tables Table 2.1 Principle Selenium species in environmental and biological systems Table 2.2 Extraction efficiency of Se from biological samples Table 2.3 Applications of ion-exchange chromatography for Se speciation Table 2.4 Applications of ion-pairing reverse-phase chromatography for Se speciation Table 3.1 Reagent gases used in collision and reaction cell Table 3.2 Instrument ICP-MS optimised parameters Table 4.1 IC conditions Table 4.2 Comparison of three tuning methods for the detection of Se Table 4.3 Comparison of three tuning methods for the intensity of background Table 4.4 The characteristics for Se species by the proposed method Table 6.1 Se species found in tablet samples (µg g -1 ) Table 7.1 Concentration (µg g -1 ) found for Se species in biological samples by ICP-MS Table 8.1 The retention times (min) of five Se species on a short C18 column (150 mm x 4.6 mm), flow rate at 1mL/min Table 8.2 The retention times (min) of five Se species on a longer C18 column (250 mm x 4.6 mm), flow rate at 1mL/min Table 8.3 Se species in biofortified foods (µg g -1 ) viii

10 List of Publications and Manuscripts Wang, W.H., Chen, Z.L., Davey, D.E., and Naidu, R. (2010) Speciation of selenium in biological samples by ion chromatography with inductively coupled plasma mass spectrometry. Journal of Liquid Chromatography & Related Technologies, 33, Wang, W.H., Chen, Z.L., Davey, D.E., and Naidu, R. (2009) Extraction of selenium species in pharmaceutical tablets using enzymatic and chemical methods. Microchim Acta, 165, Chen, Z.L., Wang, W.H., Mallavarapu, M., and Naidu, R. (2008) Comparison of no gas and He/H 2 cell modes used for reduction of isobaric interferences in selenium speciation by ion chromatography with inductively coupled plasma mass spectrometry. Spectrochimica Acta, Part B 63, Wang, W.H., Chen, Z.L., Davey, D., Rahman, M.M., and Naidu, R. (under review) Selenium Speciation Analysis in Nutritional Tablets and Biofortified Foods by Anionexchange Chromatography with Inductively Coupled Plasma Mass Spectrometry. Wang, W.H., Sun, Q., Davey, D., Chen, Z.L., and Naidu, R. (under review) Ion-exclusion chromatographic separation of inorganic selenium species in water and detection by ICP- MS with an octopole reaction system. ix

11 Abbreviations AAS AFS BEC CE CID CRM EI-TOF-MS ESI-MS ETV DRC DMSeP FI GC GC-MS GPX HPLC HFBA HG HEX IC IEC IP-RP ICP-AES ICP-MS KED LA LC LOD MALDI MIP NIST Atomic absorption spectrometry Atomic fluorescence spectrometry Background equivalent concentration Capillary electrophoresis Collision induced dissociation Certified reference materials Electro impact time of flight mass spectrometry Electrospray ionisation mass spectrometry Electrothermal vaporisation Dynamic reaction cell Dimethylselenonium propionate Flow injection Gas chromatography Gas chromatography-mass spectrometry Glutathione peroxidase High performance liquid chromatography Heptafluorobutyric acid Hydride generation Hexapole reaction cell Ion chromatography Ion exchange chromatography Ion pair reverse phase Inductively coupled plasma atomic emission spectrometry Inductively coupled plasma mass spectrometry Kinetic energy discrimination Laser ablation Liquid chromatography Limit of detection Matrix assisted laser desorption ionisation Microwave induced plasma National Institute of Standards and Technology x

12 ORS PAGE PFPA PHFOS RP RSD SDS Se SEC SELDI SOS SRM SPE SPME S/N TBAA TBAHS TEA-HFOS TFA TMAH TR Tris Octopole reaction system Polyacrylamide gel electrophoresis Pentafluoropropionic acid Potassium heptadecafluorooctylsulphonate Reversed phase Relative standard deviation Sodium dodecyl sulphate Selenium Size exclusion chromatography Surface enhanced laser desorption ionisation Sodium octylsulphonate Standard reference materials Solid phase extraction Solid phase microextraction Signal/noise Tetrabutylammonium acetate Tetrabutylammonium hydrogensulphate Tetraethyl ammonium heptadecafluorooctyl sulphonate Trifluoracetic acid Tetramethylammonium hydroxide Thioredoxin reductase Tris(hydroxymethyl)aminomethane xi

13 Declaration I declare that: This thesis presents work carried out by myself and does not incorporate without acknowledgment any material previously submitted for a degree or diploma in any university; To the best of my knowledge it does not contain any materials previously published or written by another person except where due reference is made in the text; and all substantive contributions by others to the work presented, including jointly authored publications, is clearly acknowledged.. Date Weihong Wang xii

14 Acknowledgments I would like to thank the following: My principle supervisor, Assoc. Prof. David Davey, for guiding me through my research and for always being patient and supportive. From the very beginning to the end of my PhD candidacy, he helped me to achieve my goal. Prof. Ravi Naidu, co-supervisor, for providing the opportunity to do research in the laboratory of Centre for Environmental Risk Assessment and Remediation (CERAR) and for offering support throughout. Assoc. Prof. Zuliang Chen, my co-supervisor, for his strong technical guidance on my experiment designs and for discussing details of data and results. He showed me how to develop a scientific career. He taught me much about research skills and publishing academic papers. The people in CERAR for their friendship and camaraderie, especially my friends, Dr. Cobus Gerber, Ms. Jeanette Lunnie and others who have always supported and inspired me. I want to gratefully thank my loving family, my parents and my brother. They constantly, strongly encourage me to pursue my goal. I gratefully acknowledge the Australia Postgraduate Award (APA) and thank the Cooperative Research Centre for Contamination Assessment and Remediation of Environments (CRC CARE), Australia, for a top-up scholarship. xiii

15 Abstract It is widely recognised that trace elements are associated with vital functional properties in environmental and biological systems. An important example is selenium (Se). It is toxic at high concentrations in animals and humans, but possesses beneficial properties at low levels. A better understanding of the role of particular Se compounds in such health issues is required. For that reason, improved analytical techniques to determine and identify selenium-containing species in foods and dietary supplements are crucial. Methods for determining the major selenium species in various environmental and biological matrices using liquid chromatographic and other separation techniques coupled with inductively coupled plasma mass spectrometry (ICP-MS) have occupied researchers over the last two decades. To this end, this thesis has investigated liquid chromatographic separation methodologies (ion-exclusion, ion-exchange and ion-pairing reverse phase chromatography) for selenium speciation. These methods have been applied to a variety of aqueous environmental samples, selected foodstuffs (health biscuits and wheat flour), and nutritional/ pharmaceutical tablets. Measurement of selenite, selenate and seleno-amino acids at trace levels is emphasised in this work. The compatibility of ICP-MS with ion-exchange chromatography has provided the opportunity to develop some innovative methods for the selective measurement of analytes at trace levels. Two different columns, an ion-exclusion column (Chapter 5) and a newly developed commercial anion-exchange column (Chapter 4), were employed for the first time to examine Se speciation. Both methods have been applied to environmental samples. In applying the anion-exchange column, it was found that aqueous eluent containing 10 mm NH 4 H 2 PO 4 and 20 mm NH 4 NO 3 at ph 6.5 can resolve selenite and selenate within 5 minutes, thus providing a high throughput for samples containing inorganic Se species. However, a single buffer 20 mm NH 4 H 2 PO 4 at ph 6.5 led to improved resolution when Se organic species (seleno amino acids) as well as inorganic species were present in samples. During the above ion chromatographic investigations, it was discovered that incorporation of an octopole reaction system (ORS) into the ICP-MS provided markedly higher sensitivity for Se. The ORS is described in detail in Chapter 3. Selenium detection xiv

16 following chromatography and the ORS was improved to trace levels by measuring the abundant isotope 80 Se. Prior to the application of the ORS, serious polyatomic interferences by 40 Ar 40 Ar + and 38 Ar 40 Ar + originating from the Ar fuel in the ICP plasma prohibited use of 80 Se for sensitive analysis. It was discovered that the argon species are efficiently removed by reaction with He and H 2 gas in the ORS reaction cell. H 2 provided a measurement with a superior signal to noise ratio to that observed for He. Although Se containing amino acids had been successfully separated by anion exchange chromatography together with Se inorganic species, the research then aimed to improve the separation of three seleno-amino species using ion-pairing reverse phase chromatography combined with ICP-MS. Therefore, a search for a suitable ion pairing reagent was systematically conducted and applied to important organic and inorganic Se species. A combination of the amphoteric ion-pairing reagent tetraethyl ammonium heptadecafluorooctyl sulphonate and 1% methanol as mobile phase provided better performance for organic species at low concentrations. This procedure successfully achieved the detection of selenium species in foodstuffs. The effective release of free seleno-amino acids from protein bound biological matrices is an issue of paramount importance in selenium studies. Sample pre-treatment was thus investigated by comparing enzyme hydrolysis with the use of sodium dodecyl sulphate in 1 mm HCl extraction. Based upon Se recovery tests the enzymatic approach proved to be the more effective, particularly when using dual enzymes, protease XIV and lipase VII, in a buffer containing 75 mm Tris-HCl at ph 7.5. In conclusion, the methods developed for quick and precise simultaneous analysis of organic and inorganic Se species, and the information obtained for individual Se species in environmental and biological systems, provided a better illustration of how Se is distributed in its various molecular forms. Sound analytical methods for such species are essential tools to better understand the potential animal and human health benefits provided by selenium. This thesis paves the way for future investigations which should focus on identification and structure information of unknown Se compounds such as that observed in Chapter 6. Mass spectrometric methods, such as ESI-MS or other instruments such as matrix assisted xv

17 laser desorption ionisation (MALDI) and quadrupole time of flight mass spectrometry (TOF-MS) etc., have a role to play here. Moreover, the anion exchange column and procedure studied in Chapters 4 and 7 should be explored further for the rapid simultaneous speciation studies of multiple elements (Se and As etc.) in a variety of matrices. The dynamic reaction/collision cell technology applied throughout this work reduces spectral interference of Se significantly and the procedure will certainly continue to be applied in the future. xvi

18 Chapter 1 Introduction 1

19 It is widely recognised that trace elements are associated with vital functional properties in environmental and biological systems (Hill, 1997; Szpunar and Lobinski, 2002). Elements such as phosphorus, sulphur, arsenic, vanadium, chromium, zinc and cobalt, etc., in various physiochemical forms have been identified as exhibiting either toxicological or beneficial properties in human nutrition, the environment, biochemical and physiological processes in plants and animals and in clinical medicine. Interacting species can arise as metabolites (endogenous or bio-induced molecules) in plants and animals or can be discharged from anthropogenic activities such as the use of pesticides or herbicides. Trace metal and metalloid-containing species in various physiochemical forms have therefore attracted scientific interest in recent years. Their behaviour in terms of mobility, bioavailability, biological functions, accumulation, nutritional and health benefits decide their fate in environment and in biological systems (Caruso and Montes-Bayon, 2003). In order to understand the health and environmental impact of an element, characterisation of each form as a species has become a key challenge of modern analytical chemistry. Research has stimulated the development of analytical techniques for the identification, determination and structural characterisation of different forms in biological and environmental niches. Such an appraisal is termed elemental speciation (Templeton et al., 2000). Coupled techniques, involving the hyphenation of separation methods with element/molecule specific detection, have become powerful tools in metal (ioid) speciation (Montes-Bayón et al., 2003). Hyphenation of separation techniques opens up various possibilities for the on-line coupling of liquid chromatography, gas chromatography and electrophoresis, depending primarily on the physiochemical properties of the target species (Michalke, 2002a). Inductively coupled plasma mass spectrometry (ICP-MS) has become a favoured detection technique (Zoorob et al., 1998). It provides high sensitivity, sample throughput, and has multi-element detection capability with little interference. Other mass spectrometric techniques such as electrospray mass spectrometry (ESI-MS) and matrixassisted laser desorption ionisation (MALDI), coupled with electron-impact time of flight mass spectrometry (EI-TOF-MS), are usually employed in parallel to determine the identity of eluted species (Szupnar and Lobinski, 2003). 2

20 The trace element selenium (Se) is well known as both an essential and toxic element in animals and humans (Cornelis et al., 2005). Se has been recognised as having many important roles in plant and animal physiology and beneficial for human health. The environmental and biological interest in selenium covers a large variety of compounds, ranging from small volatiles, inorganic ions and non-volatile amino acids to macromolecules such as selenoproteins (Cornelis et al., 2005). There is strong evidence that diets supplemented with Se can reduce the incidence of certain types of cancer, such as prostate and breast cancer (Infante et al., 2005). Through metabolic processes, Se dietary supplements can be converted into anti-carcinogenic metabolites such as methylselenol (CH 3 SeH) (Ip et al., 2000). As a consequence, Se is increasingly used in food and supplements. This has resulted in a need for high performance analytical methods that characterise selenium species. In most cases, Se species constitute very small portions of the total sample. Thus, sample preparation, analyte extraction and pre-concentration play key roles in practical method development. Since target species in complex matrices are found at trace to ultratrace (sub-ppb) levels, they require robust analytical techniques with remarkable specificity and sensitivity to be developed (Cornelis et al., 2005). The purpose of this dissertation is to identify major selenium species in various environmental and biological matrices using liquid chromatographic separation coupled with inductively coupled plasma mass spectrometry (LC-ICP-MS). To this end, the present study has investigated liquid chromatographic separation methodologies (ion-exclusion, ion-exchange and ion-pairing reverse phase chromatography) for inorganic species and mixtures of both inorganic and organic species at trace levels. These methods have been applied to a variety of aqueous environmental samples, selected foodstuffs (health biscuits and wheat flour), and nutritional/pharmaceutical tablets. The study has three main attributes: extraction of selenium species using enzymatic and chemical methods; species characterisation using LC-ICP-MS; and, removal of isobaric interferences in selenium analysis by applying octopole reaction system within inductively coupled plasma mass spectrometry. The thesis is structured as follows (see Fig. 1.1 below): Following the introduction, a literature review of Se speciation follows in Chapter 2, then Chapter 3 discusses the implementation of instrumentation. Chapters 4 and 5 describe ion- 3

21 exchange and ion-exclusion chromatography to separate the inorganic Se species selenite and selenate at trace levels in environmental samples, respectively. The focus then moves to the analysis of biological samples for mixtures of both inorganic and organic species, selenite, selenate and seleno-amino acids at trace levels. Hence, Chapter 6 presents the development of high recovery sample extraction techniques for selenium species by enzymatic and chemical means. Finally, efficient liquid chromatographic approaches are examined to improve removal of interferences is outlined in the subsequent chapters. Chapters 7 and 8 focus on the use of anion-exchange and ion-pair reverse chromatography for sepeciation of inorganic and organic selenium species in biological samples. Detection via the less abundant isotopes 82 Se (8.73%) and 77 Se (7.63%) has been the most commonly accepted procedure, resulting in relatively poor sensitivity for Se. Discussion in Chapters 4 and 7 focuses on an exploration of an octopole reaction system (ORS) combined with the ICP-MS. The question is whether the ORS chemistry can improve sensitivity by removing interferences in the plasma gas and thus allow the major selenium isotopes, 78 Se and 80 Se to be detected. A central theme of this dissertation is method development, the purpose of which is the study of Se species. The work reflects the need for fast, reliable and sensitive separation methods for determination of active Se species. Any information obtained for individual Se species in environmental and biological systems provide an improved knowledge of the way Se distribution in its various molecular forms. Simultaneous analyses of such species are essential for better understanding of the potential animal and human health benefits that selenium can provide. 4

22 Figure 1.1 A diagrammatic overview of the thesis structure 5

23 Chapter 2 Selenium speciation 6

24 2.1 The Role of Selenium in our World Selenium (Se) is positioned between sulphur and tellurium in Group VIA and between arsenic and bromine in Period 4 of the periodic table. Thus, it is classified as a metalloid. Se is found in four different oxidation states: -2, 0, +4 and +6. The most common inorganic forms are selenite (SeO 2-3 ) and selenate (SeO 2-4 ). Organic Se, the selenide which can be formally described as Se (II) is the most bioavailable form. In terms of the toxicity of Se species, it is known that organic forms are more bioavailable and less toxic than inorganic forms, with the toxicity of the inorganic forms following the order Se (IV) <Se (VI) <hydrogen selenide (H 2 Se). Natural sources of Se are mainly sulphur containing minerals (Pyrzynska, 2002). Biological or dietary sources of Se include kidney, liver, other meat products, seafood, milk, cheese, and whole grain. Fruit and vegetables are poor sources of the element (Frankerberger and Engberg, 1998). Se compounds are widely used in industry as inorganic pigments, in glass manufacture, electronic applications, photocopy machines, rubbers, ceramics, plastics, lubricants, etc (Pyrzynska, 2002). Se may be released in the environment through both natural processes and anthropogenic activities, such as the combustion of fossil fuels, glass and electronic industry processes, and in agriculture. Se has been recognised as having many important roles in plant and animal physiology as well as being beneficial to human health. Se is both essential and toxic (Whanger, 2002). Se concentration in food varies with the source of agricultural products originate and the Se content of the soil (Combs, 2001). Problems occur regarding both deficiency and excess of the element. Se deficiency leads to adverse effects on animal and human health. It was reported that diseases, such as alkali disease in horses and Keshan and Kashin-Beck disease, in China are closely related to low Se levels in local geoecosystems; food grains in disease-affected areas are found to be low in Se compared to those in unaffected areas (Tan et al., 2002). Se deficiency can also result in an increased risk of immune dysfunction and development of cancer (Rayman, 2007a). In contrast, soils rich in Se (e.g. in excess of 5 mg/kg) are found in the San Joaquin Valley in California, USA. High Se contamination and bioaccumulation in water has been linked to the death and deformity of waterfowl and fish in the region (Amweg et al., 2003). The nutritional range of Se between toxic and deficient level is therefore very narrow ( µmol/l for a human) (Pyrzynska, 1996). 7

25 The optimal human intake of Se ranges from 55 to 70 µg/day, which is the recommended value by the Food and Agricultural Organisation/World Health Organisation (Rayman, 2007b). In early 1970s, a key metabolic function was found for Se, as an essential component of the antioxidant, enzyme glutathione peroxidase (GPX) (Pyrzynska, 2002). This enzyme protects cell membranes from damage caused by the peroxidation of lipids. Several additional Se containing enzymes and proteins have since been identified. The nutritional essentiality of Se appears to be due to three factors: formation of the active seleno group in cystein (SeCys-proteins), antioxidant protection by the GPX and gene expression by thioredoxin reductase (TR) (Tinggi, 2003). Some additional health benefits for Se are now recognized. Firstly, it has anticarcinogenic properties (Larsen et al., 2004), and secondly, it is capable of providing prevention against heavy metal toxic effects. Selenium has also been found in 5'-deiodinase where it played a role by forming unreactive complexes with heavy metals, such as cadmium, mercury, and silver (Jamba et al., 1997). Se is present in low levels in most plant-derived foods as compounds with reasonably good bioavailability. Plants metabolise the inorganic selenite and selenate species to generate seleno-amino acids predominantly as selenomethionine (SeMet) and selenocysteine (SeCys) derivatives, as well as selenocystathionine and SeMC. SeCys is functionally present in biochemical processes and in active Se proteins. It has been classified as the 21 st amino acid (B Hymer and Caruso, 2006). SeMet is usually incorporated non-selectively into proteins as a substitute for sulphur containing methionine. The allium family whose sulphur chemistry is well-defined, exhibit parallel selenium chemistry (Uden et al., 2004). After enrichment with Se, garlic, onion or ramp was found to contain Se-methylselenocysteine and γ-glutamyl-se methylselenocysteine (Uden et al., 2004). It has been hypothesised that the latter serves primarily as a carrier of SeMC. SeMC, supported by metabolites of Se, has shown promise in mammary cancer prevention. The metabolite, methylselenol (CH 3 SeH), appears to be a key anti-carcinogenic compound (Ogra et al., 2005a). Dietary Se is available from nutritional supplements. Se-enriched yeast is widely available as an organically-bound Se source. Recently, the species Se-methyl-selenocysteine (SeMC) became available to the public as a dietary supplement (Dumont et al., 2006b; 8

26 Infante et al., 2005). It is considered to have greater efficacy as a cancer chemopreventative agent than other organic or inorganic analogues. Fertilisation of crops to increase Se intake in the food chain is now practiced, leading to products such as biofortified bread flour and biscuits. Much attention has now focused on increasing the Se content of foods, or by nutritional supplements to diets. Increased bioavailability and absorption of Se is expected to provide better animal and human nutrition. Increased tissue distribution should result in health benefits such as cancer prevention. The metabolic cycle in the human diet is illustrated in Fig Figure 2.1 Pathways of Se metabolism, following (a) Toxic and (b) Sub-toxic doses. Adapted from (Infante et al., 2005; Ip et al., 2000). Trace Se plays many important health roles as well as being toxic and creating deficiency in health and nutrition. Se species in foods and supplements have become the subject of our large research. A key aspect in this thesis involves a Se speciation (characterisation) study. 9

27 2.2 Speciation In 2000, the IUPAC presented the following variants in its definition of speciation (Templeton et al., 2000): Chemical species; species (chemical elements): specific form of a chemical element defined as molecular, complex, electronic or nuclear structure. Speciation analysis: (elemental analysis): measurement of the quantities of one or more individual chemical species in a sample. Speciation of an element; speciation: distribution of defined chemical species of an element in a system. Se exists naturally in different oxidation states and as various compounds (forms) in environmental and biological systems. Inorganic Se can exist in water and soil as the colloidal element Se, selenide, selenite and selenate. Selenates (SeO 2-4 ) are easily leached from soils, transported to ground waters and are readily taken up by plants. Selenites (SeO 2-3 ) occur in mildly oxidizing environments and are less soluble than selenates. Comprehensive reviews of the environmental chemistry of Se, including sampling, sample preparation and storage strategies have been reported previously (Pyrzynska, 2002; Wake et al., 2004). Organic forms of Se in biological systems (animals, plants and microorganisms) are reported as volatile methylated compounds, seleno amino acids, selenoproteins or their derivatives. Se can biologically form covalent C-Se and Se-S bonds, and is found as enzyme or gene components. It can be incorporated into such substances by enzymatic actions including reduction and methylation, which lead to seleno-amino acid synthesis or Se gene products. These, for example, include UGA codon encodes for the selenocysteinyl residue. Identified Se species of major interest to date are summarised in Table

28 Table 2.1 Principle Selenium species in environmental and biological systems IUPAC name Abbrev. Formula Sample type Selenous acid(selenite) Se(IV) 2- SeO 3 Natural water, soil Selenic acid(selenate) Se(VI) 2- SeO 4 Natural water, soil Methylselenol CH 3 SeH Urine Dimethylselenide DMSe CH 3 -Se-CH 3 (volatile) Breath, Se-enriched broccoli, garlic, onions Dimethyldiselenide DMDSe CH 3 -Se-Se-CH 3 (volatile) Breath, Se-enriched broccoli, garlic, onions Dimethylseleniumsulphide CH 3 -Se-S-CH 3 (volatile) Breath, Se-enriched broccoli, garlic, onions Diethylselenide DESe CH 3 -CH 2 -Se-CH 2 -CH 3 Air, sediment Trimethylselenonium ion TMSe + (CH 3 ) 3 Se + Urine Selenocysteine SeCys HSe-CH 2 CH(NH 2 )-COOH Seagull eggs, Se-enriched broccoli, garlic, onions Selenocystine SeCys 2 HOOC-CH(NH 2 )CH 2 -Se-Se-CH 2 CH(NH 2 )-COOH Se-enriched yeast, broccoli, garlic, onions Se-Methylselenocysteine SeMC CH 3 Se-CH 2 CH(NH 2 )-COOH Se-enriched yeast, broccoli, garlic, onions, chives Selenomethionine SeMet CH 3 Se-CH 2 CH 2 CH(NH 2 )-COOH Se-enriched yeast, broccoli, garlic, onions, seleniferous corn, wheat, soybeans, Brazil nuts, mushrooms, sesame seeds, chives Se-Methylselenomethionine MeSeMet (CH 2 ) 2 Se + -CH 2 -CH 2 -CH(NH 2 )-COOH Se-enriched yeast, broccoli, garlic, onions Γ-Glutamyl-Se-methylselenocysteine Γ-Glu-SeMC H 2 NCH 2 CH 2 -CO-NHCH(COOH)CH 2 -Se-CH 3 Se-enriched yeast, broccoli, garlic, onions Selenohomocysteine HSe-CH 2 -CH 2 -CH(NH 2 )-COOH Se-enriched yeast, broccoli, garlic, onions Selenocystathionine HOOC-CH(NH 2 )-CH 2 -CH 2 -Se-CH 2 -CH(NH 2 )COOH Se-enriched yeast, broccoli, garlic, onions Selenocystamine SeCystm H 2 N-CH 2 -CH 2 -Se-Se-CH 2 -CH 2 -NH 2 Human urine, seagull eggs, milk Se-adenosylselenohomocysteine AdoSeHcy H 2 N-CH(COOH)-CH 2 -CH 2 -Se-CH 2 -C 4 H 5 O 3 C 5 N 4 -NH 2 Se-enriched yeast Selenourea SeUr Se=C(NH 2 ) 2 Urine Selenosugars Various sugar structures Urine Selenoproteins Various proteins and enzymes Milk, soybean, yeast, plasma, tissues 11

29 To better understand the roles of selenium in metabolic processes (Dumont et al., 2006b), nutrition (Infante et al., 2005), cancer prevention efficacy (Ip et al., 2000), environment (Pyrzynska, 2002), biology (Whanger, 2002) and clinical chemistry (Francesconi and Pannier, 2004), it is pertinent to describe inorganic and organoseleno compounds in environmental and biological systems. A number of authors have reviewed Se speciation, including Se species stability, sample preparation (extraction, matrix separation, preconcentration), selective separation of inorganic and organic species, and sensitive detection in environmental and biological samples (Uden et al., 2004; Polatajko et al., 2006; B Hymer and Caruso, 2006). Several researchers have been examining the above issues with increasingly sophisticated means in recent years. In the next section, the key questions which have been the focus of attention will be presented. These are summarised in three major categories: Assessment of different sample extraction methods The continuing search for improved trace level detection method o Hydride atomic spectroscopic methods o ICP-MS sensitive detection Separation techniques for Se species and applications 2.3 Analytical characterisation of Se species Sample extraction Thorough preparation of the samples is crucial for successfully assessment of the concentration and range of an analyte species. An important requirement for reliable speciation is to retain the concentration and structure of the original chemical forms in the sample. The direct analysis of Se species by either atomic spectroscopic or separation techniques is difficult because of the relatively low Se concentrations when compared to other metals or metalloids and the complex matrices of biological samples, which can interfere in the determination of low level Se species. Extraction of the species of interest with the highest recovery is the next challenge prior to analysis using the chosen techniques. The successful extraction of Se compounds will rely on their aqueous solubility and protein or polysaccharide binding (Ebdon, 2001). A conventional approach to extracting Se species is to use liquid extraction (Ebdon, 2001). A variety of different 12

30 solutions have been used to extract species for selected matrices. The extraction methods and their efficiencies from biological samples are summarised in Table 2.2 at the end of the subsection. The main methods may be categorised as: Aqueous extraction; Enzymatic hydrolysis; Ultrasonic probe proteolytic hydrolysis; In vitro digestion; and Sequential treatment. Aqueous extraction Low recoveries (10-40%) from protein material have generally been observed when using simple methods such as hot water extraction. This has stimulated interest and promoted the development of more efficient methods for Se species. Acid addition is one such improvement, including mineral acid as well as organic acid attack. Methanesulphonic acid was used for selenomethionine extraction from yeast and Brazil nuts when heated under reflux (Wrobel et al., 2003). Since the SeMet found in these matrices is in a bound form, the acid liberates the bound analyte, thus proving more efficient than hot water. However, acid digestion was less favoured for examining biological systems because some important organic forms may not remain intact under such conditions. Enzymatic hydrolysis Enzymatic hydrolysis focuses on the extraction of Se protein bound compounds, either mechanically or via coordination complexes in biological matrices. A number of procedures have been developed. Driselase is a commercial enzyme preparation originating from Basidiomycetes sp., which is usually used to destroy cell-wall components (Infante et al., 2005). Proteinase K or proteolytic enzymes (Protease XIV) have often been used for water-insoluble Se fractions in various complicated matrices, including human blood serum, cod muscle, wheat flour, etc. (Huerta et al., 2004). The efficiency of enzymatic hydrolysis is relatively high, but variable (40-90%). The efficiency of extraction depended on some key parameters, such as the choice of enzymes, hydrolysed solution ph, temperature and hydrolysis time. For Se based food supplements, the main extraction problems are caused by bulking agents and binders used in the manufacture of food supplements. 13

31 Ultrasonic probe proteolytic hydrolysis Traditional, enzymatic hydrolysis is time-consuming and requires incubation (Capelo et al., 2004). Ultrasonic probe proteolysis has been described as an efficient alternative to reduce extraction time for inducing the breakdown of Se containing proteins (peptide bonds) into seleno-amino acids. For example, the release of selenomethionine from selenised yeast using a ultrasonic probe was completed in only 30 seconds (Capelo et al., 2004). In vitro digestion In vitro gastrointestinal digestion was used to assess the bioaccessibility of Se compounds from fish samples (swordfish, sardine and tuna) (Cabañero et al., 2004). The digestion method was studied in gastric juice at 37 C in a water bath and shaken for a few hours, and then intestinal juice was added for a few more hours under the same conditions. Selenium solubility in the gastrointestinal supernatants was higher in swordfish and sardine (76% and 83%, respectively) than in tuna (50%). Simulated human gastric and intestinal digestion led to the identification of selenomethionine (SeMet) in the three digested fish. Sequential treatment Sequential treatment combining enzymes with conventional techniques may also improve extraction efficiency. This strategy has been applied to Se-enriched yeast material (Chassaigne et al., 2002) using Tris-HCl buffer for water-soluble fractions. This was followed by Tris-HCl buffer with SDS for solubilisation of the protein fraction and, finally, used concentrated HNO 3 for dissolving the resulting solid residue. Following these three steps, total Se recovery of 91.5% was achieved. In another example, mushrooms were treated with the sequential enzymatic processes with the highest efficiency (89%) being obtained by using a 3-step procedure with lysing enzyme and pronase (Dernovics et al., 2002). A number of such procedures, including up to fourteen extraction methods, were compared previously by B'Hymer and Caruso (2000) and Yang et al. (2004b). However, the true efficiency obtained from these different methods was difficult to compare due to the absence of certified reference materials (CRM) for the different Se species and the sample matrices of varying complexity. The extraction methods and their efficiencies applied to biological samples are summarized in Table 2.2. However, the mechanism of hydrolysis or 14

32 extraction and comparison of efficiency using different extraction methods need to be established. 15

33 Table 2.2 Extraction efficiency of Se from biological samples Sample Extraction method Extraction efficiency (% of References total Se) Se-enriched shiitake mushroom Natural & selenised Agaricus mushrooms Enriched green onions (Allium fistulosum) Root, stem, leaf of Dill (Anethum graveolens L.) Brazil nuts (1) Water in an ultrasonic bath for 30 min at 25ºC, then incubated for 3 h at 37 or 95ºC (2) 20mg pronase Tris-HCl (ph 7.2) at 37ºC overnight 68 (water) 77 (pronase) (Ogra et al., 2004) Water (37 or 85ºC) (Huerta et al., 2006) (1) Alkaline extraction: 0.1 M NaOH for 15min (1) 55 (Kápolna and (2) Enzymatic hydrolysis: 5 mg pronase E shaken at 25ºC for 24 h Fodor, 2006) (2) 78 (1) 0.1 M HCl were stirred for 24 at 25ºC (Cankur et al., 2006) (1) water microwave (1) 9 (Vonderheide et (2) 0.5 M HCl microwave (2) 37 al., 2002b) (3) 25 mg Proteinase K incubated at 37ºC for 20 h (3) 83 Brazil nuts 4 M methanesulphonic acid at reflux for 8 h 75 (found as SeMet) (Wrobel et al., 2003) Six different brands of Water with microwave for 2 h (water) (B'Hymer and Se food supplements 2 M HCl with microwave for 2 h (HCl) Caruso, 2000) Edible mushroom(agaricus bisporus) Step 1: water at 37ºC for 3 h Step 2: 45mg pepsin in Tris-HCl(pH 2.1) stirred at 37ºC for 20 h Step 3: 45 mg trypsin in phosphate buffer (ph 7.6) stirred at 37ºC for 20 h 75 (Stefánka et al., 2001) 16

34 Yeast supplements (1) In vitro, simulated gastric fluid (SGF) (Dumont et al., (Saccharomyces (2) Simulated Intestinal fluid (SIF) 2004) cerevisiae) Se-enriched fresh and In Vitro gastrointestinal digestion (gastric juice) (Pedrero et al., dried radish (intestinal juice) 2006) Laboratory reference 15 mg Pronase-E in phosphate buffer (ph 7.4) stirred at 37ºC for 24 h 98 (Bodó et al., material from Brazil nuts 2003) Brassica juncea (Indian mustard) (1) 1M HCl; (2) Tri-HCl, ph 8; (3) proteinase K; (4) protease XIV (1) 54 (2) 46.5 (3) 73 (4) 70 (Montes-Bayón et al., 2002) Antarctic krill 20 mg Pronase-E in Tris-HCl (ph 7.5) incubated at 37ºC for 24 h and 87 (without ultrasound) (Siwek et al., with ultrasound-assisted for 15 min 98 (ultrasound-assisted) 2006) Yeast certified reference 20 mg pronase and 10mg lipase at 37ºC for 24 h (obtained for SeMet) (Yang et al., material (CRM) 2004b) 77 Se-enriched yeast (1) 30 mg β-glucosidase in a water bath for 3h, followed by 15 mg (1) 95 (Larsen et al., mixture of endo- and exopeptidases at 50ºC for 24 h 2003) (2) 20 mg Protease XIV at 37ºC for 24 h (2)90 Yeast-based supplements (1) 1 mg protease XIV in Tris-HCl at ph 7 for 20 h at 37ºC (2) 4% driselase in Tris-HCl for 4h in shaking bath, then 1mg protease XIV for 20h incubated at 37ºC (3) in vitro gastrointestinal model: 100 µl gastric juice in a shaking bath at 37 C for 4 h and 100 µl gastrointestinal juice incubated for further 4 h (1) 87 (2) 92 (3) (Reyes 2006) et al., 17

35 Selenised yeast Se-enriched plants(garlic and Indian mustard) Chicken muscle, liver and kidney Sequential leaching: Water 3h, then the residue in 4% Driselase in Tris- HCl (ph 7.5) for 18h, and the last residue in 4% SDS in Tris-HCl (ph 7.5) for 1 h 32 (water) 60 (Driselase) 8 (SDS) (Połatajko et al., 2004) 25 mm Ammonium acetate (ph 5.6) by ultrasonic probe 3 min 80 (Montes-Bay on et al., 2006) Method 1: Pronase-E incubation at 37ºC for 48 h in water and Tris-HCl at Method 1: in water and (Cabanero et al., ph 7.5) in buffer 2005) Method2: ProteaseXIV and Tris-HCl at ph 7.5, probe sonication Method 2: extraction for 2 min Dietary supplement Sequential extraction: 10 mm Tris-HCl (ph 8) for water soluble fraction, 4% SDS (ph 7) for the residue 1, 5% driselase on the residue 2, last step for adding protease inhibitors in supernatant. 85 (Ayouni, 2007) 18

36 2.3.2 The continuing search for improved trace level detection The following subsection describes detection and separation of Se species. Total Se concentration is already low in biological and environmental samples. Individual Se species becomes much lower when fractionated at trace or ultra-trace levels. The search for sensitive detection techniques of Se species in different environmental niches and biological systems is becoming important. The detection techniques that are currently available include: electrochemical and UV/vis spectroscopic methods; atomic spectrometry; GC-MS; electrospray ionisation mass spectrometry (ESI-MS); inductively coupled plasma optical emission spectrometry (ICP-OES); and inductively coupled mass spectrometry (ICP-MS). Electrochemical and simple spectroscopic methods are seldom used in the field of Se chemistry. ESI-MS is more commonly employed to obtain structural information about Se compounds Hydride atomic spectroscopic methods Atomic spectroscopic techniques as element-specific detection including atomic absorption, emission and fluorescence spectrometry (AAS, AES and AFS) have been developed for selenium speciation. Heated quartz tube AAS and nitrogen cold trapping AFS (Sturgeon et al., 1985; Aeungmaitrepirom et al., 1999; Stripeikis et al., 2004; Capelo et al., 2006) are two such procedures applied to Se speciation. The detection limits obtained with flame AAS (FAAS) at a few mg/l for Se were too high to be useful for most practical applications and, consequently, efforts and energies have been put into sample pre-treatment and for improving sample introduction efficiencies (Wang, 2007). Since Se is one of the hydride-forming elements (As, Sb, Bi, Ge, Pb, Se, Te and Sn), hydride generation (HG) makes an ideal gas-phase sample-introduction technique for Se determination at trace levels in a wide range of matrices. It is commonly combined with instruments, such as AAS, AES and AFS. The methods are well-established (Cobo Fernandez et al., 1983; He et al., 1998; Wake et al., 2004). In general, the detection limit is less than 2-3 µg/l. However, the HG methods are limited by Se chemistry; only Se(IV) can be converted to gaseous hydride (H 2 Se) after reaction with NaBH 4 and HCl (Hill et al., 1995) or in alkaline mode HG (Xu et al., 2002). Se(VI) must be indirectly determined by the pre-reduction of Se(VI) to Se(IV) and calculated from results found with and without a 19

37 pre-reduction step. The matrix may also affect both hydride generation and reduction of Se(VI). The methods have been broadly applied for detection of inorganic Se species, but were less successful with organically bound Se. Most organoseleno compounds are inert and unable to form volatile Se compounds with NaBH 4 (De-qiang et al., 1997). Thus, the phases involved in mineralisation (UV irradiation or wet digestion) are mandatory in order to convert Se compounds into selenic acid and the subsequent reduction of Se(IV) to H 2 Se, prior to HG. It has been recently stated that some organic Se compounds (SeMet, SeEt and TMSe) can form volatile analytes with NaBH 4 and HCl (Chatterjee et al., 2001b). consequently there is room for further development in this context (Ipolyi et al., 2001; Chatterjee et al., 2001a) ICP-MS sensitive detection With the main limitations of element specific detectors (AAS, AES, AFS) for Se speciation being trace levels of Se compounds in samples, complex matrix composition, chemical lability and oxidative degradation, more advanced, sensitive analytical methods for characterisation are required. Mass spectrometry-based approaches have become a powerful instrumental tool for determination of selenium species (Uden, 2002; Zoorob et al., 1998). Plasma mass spectrometry (MS) has the advantage of simplifying the data obtained during elemental analysis. Instead of complex emission spectra, sample data is reduced to mass-to-charge ratio measures of elemental ions. State-of-the-art ICP-MS was developed in the late 1980's to combine the easy sample introduction and quick analysis of ICP technology with the accurate and low detection limits of a mass spectrometer. This technique has several advantages for speciation, such as multi-element and multi-isotope detection, excellent detection limits with a wide linear dynamic range, high selectivity and sensitivity. Instrumental development in mass spectrometric techniques over the last two decades has provided the possibility of determining compounds at trace (ppb) or ultratrace (ppt) levels. Therefore, the use of ICP- MS is showing the most rapid growth amongst the tools developed to harness MS capabilities (Vanhaecke and Moens, 1999). 20

38 The most widely used quadrupole mass analysers have resolutions between amu. which is sufficient for most routine applications (Wang, 2007). However, in some cases they do not have sufficient resolution to separate the target isotope from interfering polyatomic ions having similar nominal mass to charge ratio. To resolve this problem, high resolution double focusing sector field mass analysers have been adapted for ICP. Time of Flight (TOF) and sector field (SF) mass spectrometers has been used for ICP as well. Increasing the resolution has inherently compromised performance in other ways, resulting in a loss of instrument sensitivity and high cost (Bandura et al., 2001). In ICP-MS, spectral and non-spectral interferences remain a persistent analytical hindrance. Polyatomic ion interferences are the main spectral interferences. They derive from plasma gas (typically Ar, but occasionally He or other mixed gases), solution (water, organic solvents), matrix (acids, buffers, salts, etc.) and plasma entrained atmospheric gas (O 2, N 2, etc.) sources. The most affected elements are P, S, Ca, Fe, Cr, As and Se. Spectral interference is a serious problem when detecting Se by conventional ICP-MS. For this reason, the less abundant 82 Se and 77 Se isotopes are often monitored since they experience less interference. The need for improvement in ICP-MS detection power for Se is becoming more important. Thus, new technological developments are required that enable efficient removal of interferences, resulting in improved selenium detection by monitoring its most abundant isotope, 80 Se. A collision/reaction cell such as octopole reaction system (ORS) cell or dynamic reaction cell (DRC) used in ICP-MS has been explored (see Chapter 3). The gases used in the collision/reaction cell are hydrogen, helium, dinitrogen oxide, methane (Warburton and Goenaga-Infante, 2007) and deuterium (Ogra et al., 2005b). For example, Reyes et al. (2003) determined Se in biological reference materials and serum samples using H 2 reaction gas in ORS. Huerta et al. (2003) determined Se species in yeast and wheat flour samples using the same system. Marchante-Gayón et al. (2001) used H 2 and He in a hexapole collision cell in urine samples before and after supplementation. In their study, He proved especially useful in improving the ion transmission, while H 2 was useful in reducing polyatomic argon ions and increased sensitivity by orders of magnitude. Darrouzès et al. (2005) reported Se detection limit was doubly improved by monitoring 80 Se when using a collision/reaction cell. Using a DRC system, Larsen et al. (2001) 21

39 determined seleno-amino acids in yeast and algal extracts at trace levels after enzymatic digestion. Se speciation in human plasma, urine and faeces was also performed by DRC- ICP-MS Separation techniques for Se species and applications Se species in environmental and biological samples are at very low levels and the concentration of each species needs to be established instead of only the total Se concentration. After sample extraction, and before instrumental detection of Se, separation/pre-concentration of each species in the extraction solution is necessary. Coprecipitation (Hudnik and Gomiscek, 1984), living bacteria (Robles et al., 1999), hydride generation (Pohl, 2004), capillary electrophoresis (CE) (Gilon and Potin-Gautier, 1996; Sun et al., 2004), and solid phase extraction (SPE) (Camel, 2003) have been used for separation of Se species. For development of hyphenated technique, the on-line coupling of separation with an element-specific detector has become a highly effective method for Se speciation (Pitts et al., 1995; Capelo et al., 2006; Uden, 2002). Liquid chromatography (LC) is most commonly used for separating complex Se mixtures into individual components. Gas chromatography (GC) and electrophoresis have also been applied to Se speciation, albeit to a lesser extent. Based on the nature of Se compounds (inorganic species, small metabolites, amino acids, or even polypeptides containing Se), the main forms of separation for Se speciation are: Electrophoresis; GC; Size-exchange chromatography; Ion-exchange chromatography; and Ion-pair reverse phase chromatography. The separation methods are described in the following sub-sections Electrophoresis Electrophoretic techniques have received much attention in speciation analysis because of their outstanding resolving power. Electrophoretic techniques have thus been applied to Se 22

40 speciation analysis. The separation mechanism is based on differences in the electrophoretic mobilities of ions, which are largely determined by their mass-to-charge ratio, physical dimensions and interactions with buffer components. Capillary electrophoresis (CE) is used to analyse different sample types with high resolution achieved for cations, anions and small metal ions, metal organic ligand complexes, organometallic and biomacromolecules. CE coupled ICP-MS is useful for Se speciation with detection limits in the low ppb range. CE has often been used as part of twodimensional separation strategies regarding Se speciation. Mounicou et al. (2002) used CE in conjunction with ICP-MS via a self-aspirating total consumption nebuliser after SEC separation. Detection limits were in the range 7-18 µg L -1 for selenate, selenite, SeCys 2, SeMet and SeEt in aqueous extracts of selenised yeast. The difficulty experienced with CE-ICP-MS was that larger molecules such as selenopolypeptides did not migrate fast enough to leave the column in a reasonable time. Enzymatic digestion of selenoprotein fractions may be necessary prior to CE-ICP-MS. For larger biomolecules in the molecular weight range above 1000 Da, electrophoresis (1 or 2 D gel) is the most powerful tool, which can separate more than a thousand proteins in a complex protein mixture within a single separation. Gel electrophoresis provides better resolution than HPLC for high molecular weight selenoproteins. Chassaigne et al. (2002) studied a Se-enriched yeast candidate reference material after a sequential extraction of water-soluble and insoluble Se fractions. The procedure involved firstly isolating Se containing fractions by SEC on Superdex 75 and 200 HiLoad columns, then low molecular mass compounds (SeCys 2, SeMet, selenite and selenate) were further separated and monitored by anion-exchange HPLC-ICP-MS using gradient ammonium phosphate and acetate mobile phases at ph 8.0. Finally, separation of extracts in water and SDS-soluble high molecular mass fractions (proteins) were achieved by SDS-polyacrylamide gel electrophoresis (PAGE). The gel was subsequently silver-stained and cut into bands to be determined by electrothermal vaporisation (ETV)-ICP-MS with detection limits of about 50 ng g -1. Furthermore, Chery et al. (2003) employed SDS-PAGE for separate of selenoproteins in red blood cells extracts and proteins in yeast using laser ablation (LA) with ICP-MS to achieve improved detection limits for 80 Se to 0.07 µg g -1. Laser ablation (LA) for direct gel interrogation has opened up new opportunities for interfacing with ICP- MS for rapid characterisation of protein-bound trace element containing compounds (Renli Ma, 2004). Although electrophoresis can provide high resolution for the separation 23

41 of selenium compounds, its effective combination with ICP-MS detectors has not been resolved Gas chromatography Apart from electrophoresis, gas chromatography (GC) is another popular separation technique for determining Se species (Wilber, 1980). These have to be either volatile or convertible to volatile forms by derivatisation. Interfacing the GC with ICP-MS, ICP-atomic emission spectrometry (AES) and microwave induced plasma atomic emission spectrometry (MIP-AES) have been used. GC-ICP-MS is the most sensitive hyphenated method compared to HPLC and CE-ICP-MS in terms of seleno-compounds speciation. Since samples are introduced to the plasma in gaseous form, atomisation and ionisation of the sample is more complete. The mobile phase in GC is usually helium gas, which not only enables the quantitative transport of the sample (nearly 100%) to the detector without nebulisation, but also accounts for a lowering of background in the detector itself. This process contributes generally to lowering the method s detection limits. The most abundant volatile Se species found in environmental and biological samples are the methylselenides (see Table 2.1). GC-ICP-MS was used to determine dimethylselenide in human breath after ingestion of 77 Se-enriched selenite (Kremer et al., 2005). Volatile species in B. juncea seedlings were measured by GC-ICP-MS and GC-MS. The primary species detected from the headspace above the plants were DMSe and DMDSe (Wilber, 1980). For other non-volatile Se species, derivatisation was required. Two derivatisations were often required for SeMet and selenoethionine (SeEt) in pharmaceutical preparations of Se nutritional supplements by GC-ICP-MS (Pelaez et al., 2000). One method involved esterification of the carboxylic acid group via propan-2-ol/ acetylchloride, followed by acylation of the amino group (derivatisation 1). Drying time in the first step limited this method. Alternative derivatisation entailed using ethanol-ethyl chloroformate in a single step (derivatisation 2). 24

42 Derivatisation 1 Derivatisation 2 More derivatisation reagents were also investigated. For example, Vonderheide et al. (2002a) used initial derivatisation of SeMet, selenoethionine and SeCys 2 standards with isobutylchloroformate for analyte volatility, then pre-concentration by solid phase microextraction (SPME). The derivatised seleno-amino acids were detected using ICP-MS. Organoselenium compounds in the allium family (garlic, onion and broccoli grown in Se rich soil) were detected via GC-MIP-AES. The species detected were usually mixed Se/S compounds (Kahakachchi et al., 2004). Free seleno-amino acids such as SeCys, SeMC, and SeMet in normal or selenised plants were ethylated before detection by GC-AES (Cai et al., 1995). Species-specific isotope dilution ID GC/MS allowed accurate determination of Met and SeMet in yeast (Yang et al., 2004a). Although GC interfacing with ICP-MS usually provides better resolved peaks and lower detection limits than LC-ICP-MS, GC-ICP-MS is generally more suited for Se containing volatile compounds. Due to the need for derivatisation of non-volatile Se species when using GC means, determination of seleno-amino acids using LC-ICP-MS is still much more preferable. 25

43 In LC-ICP-MS methods, size exclusion chromatography (SEC), ion exchange (IC), and ion pair reversed phase (IP-RP) have been described in terms of separating each species from mixtures. Separation power and compatibility with ICP-MS are the most important selection criteria for developing a separation method Size-exclusion chromatography Size exclusion chromatography (SEC) is often used to separate various selenoproteins from different sample matrices and provides preliminary information about the distribution of Se in a sample. This is because the retention of Se species in SEC depends on molecular size. It is ideal for high molecular weight compounds (1000 Da to 10K Da) such as polymers, peptides and proteins. In addition, due to its high matrix tolerance, SEC can be used for sample purification prior to other types of HPLC (such as IP-RP and IC) and electrophoresis. Multidimensional approaches combining SEC are often employed in characterising particular Se species by ESI-MS (Moreno et al., 2004). The enzymatic digests of both water-soluble and insoluble Se protein fractions in biological samples were analysed using SEC-ICP-MS and UV spectrophotometry (Moreno et al., 2004; Siwek et al., 2005). Samples range from marine organisms, for instance tuna, trout, krill, oyster, and mussel; to foods, such as white clover, wheat flour and selenised yeast. Two different SEC columns were used in combination with a cation-exchange column. A Biosep-SEC-2000 column for the effective separation of 300 Da to 1 kda peptides and a Superdex Peptide HR 10/30 column were employed for high molecular weight selenoproteins (150Da to 50 kda) present in the samples (Moreno et al., 2004). Another separation scheme combined size exclusion and anion-exchange chromatographies coupled with ICP-MS for Se speciation in dietary supplements (Ayouni, 2007). Additional unidentified, high molecular weight Se compounds (22%) in the samples were observed as resulting from the sequential extraction steps and separation by SEC. Affinity chromatography combined with SEC-ICP-MS have been used to separate three major Se-containing proteins (albumin, glutathione peroxidise and selenoprotein P) from human plasma (Koyama et al., 1999). Incorporating selenium into cyanobacterial metallothionein induced under heavy metal stress has been investigated using SEC-ICP- MS (Takatera et al., 1994). Similarly, selenite and selenate metabolism in rats were 26

44 successfully studied using SEC, with exchange of endogenous and dietary selenium examined in brain, liver and kidneys (Kobayashi et al., 2001). SEC advantages include approximating the molecular mass for any unknown high molecular weight Se species and it normally does not result in a change of species during separation owing to the gentle nature of the technique. Currently, SEC is used to separate labile and weak metal-complexed biopolymers (Michalke, 2002b; B Hymer and Caruso, 2006; Infante et al., 2005). However, the main limitation of SEC in separating selenium species is that the separation efficiency of multi-selenium species is poor, subsequently leading to poor resolution between species Ion-exchange chromatography (IEC) Ion exchange chromatography (IEC) has often been used for the separation of Se species owing to its high separation efficiency compared to SEC. In this approach, ions or easily ionisable analytes are separated, such as inorganic ions (cations and anions) and low molecular weight organic acids and bases (Haddad et al., 2004). The technique includes ion-exchange and ion-exclusion chromatography. Ion-exclusion chromatography relies on exclusion of inorganic anions being excluded by a strong anion- or cation-exchange resin stationary phase through ionic repulsion. It is suited for separating weak acids, weakly ionisable or neutral compounds. In this mode, the charge on the ion-exchange resin is the same as that of the weakly ionized species (Haddad and Jackson, 1990). It has been applied to arsenic (As) speciation in seawater (Nakazato et al., 2002). Ion exchange chromatography includes cation and anion-exchange. The mechanism is based on the exchange equilibria between charged solute ions and the oppositely charged surface of the stationary phase. Solute ions and ions of equivalent charge in the mobile phase compete for oppositely charged spaces on the stationary phase. The difference in interaction between analytes with the charged stationary phase brings about their separation. Selectivity in separation can be achieved by optimizing ionic strength of solute or mobile phase, ph of mobile phase, temperature, and flow rate (Haddad et al., 2004). 27

45 Fritz (2005) proposed a mechanism of ion-exchange chromatography resulting from enforced pairing effects (EP). Enforced pairing effects resulted when stronger ion pairing occurs within the stationary phase compared to that in the aqueous solution. This suggested it was a combination of effects that include hydrophobic attraction, hydrogen bonding, lower dielectric constant and water-structure induced ion pairing. Both cation and anion-exchange chromatography have been employed in Se speciation. Isocratic and gradient elution were used to separate inorganic and organic Se compounds (Kápolna and Fodor, 2006, McSheehy et al., 2002). Inorganic species (selenite and selenate) was easily separated on IEC, which could not be easily accomplished with reversed-phase or ion-pair reverse phase chromatography (IP-RPC). For example, inorganic selenium (Se) species selenite, selenate and selenocyanate in sea and rain water were separated by anion exchange chromatography using a gradient of 0.1 M NaOH as the mobile phase and determined by hydride generation-inductively coupled plasma-dynamic reaction cell-mass spectrometry (HG-ICP-DRC-MS). Detection limits of 0.15, 0.27 and 0.19 ng Se L -1 were achieced for selenite, selenate and selenocyanate, respectively (Wallschlager and London, 2004). Cation exchange chromatography was to analyse 77 Se-enriched yeast in a human absorption study (Larsen et al., 2003). Using a range of pyridinium formate buffer from 0.75 to 8 mm with 3% methanol as mobile phase in a gradient elution, ten organic Se compounds were successfully separated (Fig. 2.2). However, this method was not suitable for separating inorganic selenite and selenate. 28

46 Figure 2.2 Cation exchange HPLC separation of mixture of 10 Se standard compounds (AllSeCys: Se-allylselenocysteine, PrSeCys: Se-propylselenocysteine, other compound names see Table 2.1). (Larsen et al., 2003) The possibility of simultaneous separation of seleno-amino acids and inorganic species has been investigated. Ion chromatography with microwave assisted on-line speciesconversion hydride generation (HG) by ICP-MS enabled of the quantification of selenite, selenate, SeMet and SeCys in white clover (Johansson et al., 2000). Detection limits were ng g -1. Another research group (Juresa et al., 2006) investigated three chromatographic systems (anion, cation and reversed-phase chromatography) with hydride generation ICP-MS for examining Se metabolites in human urine. Though hydride generation was employed to obtain better sensitivity and lower detection limits, it had a more complicated interface and has not been frequently used in conjunction with ICP-MS compared to AAS/AFS. Larsen et al. (2001) subsequently combined cation and anion exchange chromatography for separation of mixture of 12 selenium species, comprising seleno-amino acids, selenonium ions and inorganic Se. Cationic species were separated using a cationexchange column and gradient elution with aqueous pyridinium formate at ph 3 as the mobile phase. The anionic species were separated by anion-exchange column and isocratic elution with an aqueous salicylate-tris mobile phase at ph 8.5. Two additional species - 29

47 selenomethionine-se-oxide (oxidative degradation of SeMet) and dimethylselenonium propionate (DMSeP) - were also detected in selenised yeast and an algal extract. Compatibility of mobile phase used for IC with ICP-MS is a serious consideration. Mobile phases that have been used include ammonium citrate, pyridinium formate, ammonium phosphate and salicylate-tris (Table 2.3). For ICP-MS, excessive salt (sodium or potassium phosphate) or organic solvent (methanol) introduced into the plasma is undesirable because high concentrations of both salt and solvent can change the sensitivity over the experimental period when salt or carbon deposits build up on the sampler and skimmer cones. High levels of salt and solvent also alter the ionisation characteristics of the Ar plasma and cause plasma instability which can eventually extinguish the plasma. Therefore, lower concentration substance such as ammonium salts which can be converted to volatile components under high ICP torch temperatures are suggested for Se speciation when using ion-exchange chromatography (IEC) with ICP-MS (B Hymer and Caruso, 2006). The applications of IC-ICP-MS to the analysis of Se are listed in Table 2.3 at the end of this section. Despite the regular use of IC-ICP-MS analysis of selenium species, this technique has some limitations which should be addressed. One major limitation is that organic modifiers (methanol) at concentrations lower than 10% may also influence the selectivity of the separation by affecting the mechanism controlling the hydrophobic interaction of the solutes with the matrix (B Hymer and Caruso, 2006). Another is that the employed eluents often have high ionic strength. Consequently, the high salt (e.g. sodium or potassium phosphate) load causes severe problems when interfacing IC with ICP-MS. Besides eluent, resolution of selenium species by ion-exchange depends on the type of column. As can be seen in Table 2.3, the Hamilton PRP-X100 and Dionex AS are commonly used and reported. Not many other columns (e.g. Bio-Rad HPX-87H and G3154A/101) have been used for selenium speciation. 30

48 Table 2.3 Applications of ion-exchange chromatography for Se speciation Sample Se target molecules Mobile phase Detection Reference Human urine after Se(IV) and Se(VI) Anion exchange Dionex Ionpac AG11-HC column in series ICP-MS (Gammelgaard supplemented with an AS11-HC column, 25 mm NaOH and 2% MeOH and Jrns, 2000) with SeMet Yeast and wheat SeCys 2, SeMC, SeMet, Se(IV) Hamilton PRP-X100 column, 5 mm ammonium citrate in Isotope dilution (Huerta et al., flour and Se(VI) 2% MeOH by gradient and isocratic ICP-MS 2003) 77 Se-enriched SeMC, Se-allylselenocysteine, A silica-based strong cation exchange HPLC column, Dynamic (Larsen et al., yeast Se-propylselenocysteine, SeMet, gradient elution by mobile phase mm pyridinium reaction cell 2003) SeEt, SeCys 2, formate with 3% MeOH (DRC)-ICP-MS dimethylselenoniumpropionic acid, methylselenomethionine, TMSe, Selenohomocystine Yeast and tablets Methaneseleninic acid, Se(IV), Hamilton PRP-X100, mm parahydroxybenzoic acid ICP-MS (Ayouni et al., Se(VI), SeMC, SeCys 2 and SeMet (PHBA) in 1% MeOH at ph8, gradient 2006) Wheat-based food SeCys 2, SeMC, SeMet, Se(IV) Hamilton PRP-X100, (1) gradient of acetic acid and ICP-MS (Warburton and and Se(VI) triethylamine, ph 4.7 Goenaga- (2) 5 mm ammonium hydrogen citrate (ph 5.9) containing Infante, 2007) 2% MeOH Enriched green Se(IV), SeMC, SeMet, SeCys 2 (1) Silica-based strong cation exchange column, 2 mm ICP-MS (Kápolna and onions (Allium and Se(VI) pridinium formate, ph 2.8 as mobile phase, isocratic elution Fodor, 2006) fistulosum) (2) Anion exchange column, Hamilton PRP-X100, mm NH 4 H 2 PO 4 (ph 5.0) in 1% MeOH by gradient elution 31

49 Se-enriched yeast Low molecular mass Anion exchange column Dionex AS10, mobile phase A: 5 ICP-MS (Chassaigne et material selenocompounds ( 10000) mm ammonium phosphate, ph 8.0; B: 50 mm ammonium acetate, ph 8.0 gradient elution al., 2002) Certified DMSe, DMDSe, three (1) Anion-exchange PRP-X100 column, 10 mm citric acid Hydride (Juresa et al., reference material selenosugars, TMSe, SeCys 2, at ph 4.8 generation 2006) (NIES CRM 18 SeMet, Se(IV) and Se(VI) (2) Reversed-phase C18 column, 20 mm ammonium (HG)-ICP-MS human urine) SeCystm, Methaneseleninate, formate at ph3.0 containing 3% MEOH (3) Cation exchange PRP-X200 column, 10 mm pyridine at ph 5.0 Se-enriched SeCys 2, Se(IV), SeMet and Hamilton PRP-X100, 5 mm ammonium hydrogen citrate Ultrasonic and (Infante et al., supplements Se(VI) (ph 5.9) in 2% MeOH pneumatic nebulisation ICP-MS 2004) Se-enriched yeast SeMet PRP-X100 column, gradient elution mm phosphate buffer (ph 7) in 2% MeOH Isotope dilution analysis-icp- MS (Reyes et al., 2004) yeast Seleno compounds (1) PRP-X100, mm acetic acid, mm ICP-MS (McSheehy et triethylamine (ph 4.7) gradient elution (2) Cation-exchange Supelco SCX column, 20 mm pyridine formate (ph 3), isocratic al., 2002) Broccoli (Brassica oleracea) SeMet, SeMC, Se(IV) and Se(VI) PRP-X100 column, 10 mm citric acid, 2% MeOH, ph5 ICP-MS (Pedrero et al., 2007) 32

50 Ion-pair reversed phase chromatography (IP-RPC) IP-RPC is a useful technique for the simultaneous separation of mixtures of anionic and cationic ions, as well as neutral molecules. It is a variation on reversed phase chromatography. The separation mechanism depends on analyte ions in solution being paired or neutralized using low concentrations of appropriate ion-pairing reagents also in solution, for separation on a reversed phase column. Selectivity and retention of the analytes can be increased by counter-ions of the ion pairing reagents. The resolution of Se species depends on the concentration of the ion-pairing reagent, flow rate, ionic strength and ph of the mobile phase, as well as the properties of the stationary phase (Michalke, 2002b; Uden, 2002; B Hymer and Caruso, 2006; Polatajko et al., 2006). Anion-pairing (e.g. alkylsulphonates) and cation-pairing (e.g. tetraalkylammonium salts) reagents and long-chain carboxylate ions (e.g. perfluorinated carboxylic acid derivatives) have often been used as ion-pairing reagents (Uden et al., 2004). A C-18 column is typically used for seleno-amino acid analysis, while C-8 has been described for selenoprotein or large peptide analysis (B Hymer and Caruso, 2006). Hexanesulphonic acid has been used as anion-pairing reagent for the analysis of Se in Brazil nut extracts on a C8 column (Vonderheide et al., 2002b). SeMet emerged as the most abundant compound with some unidentified Se containing species. Organic species were separated by ion-pairing, but the technique proved less sensitive for selenite and selenate. Difficulties were experienced with poor separation or co-elution of inorganic Se and organic species. Marchante-Gayón et al. (2000) developed tetrabutylammonium acetate (TBAA) as cation-pairing reagent to separate the anionic selenite and selenate from other seleno compounds (SeCys 2, SeMet and SeEt) (Fig. 2.3). 33

51 Figure 2.3 Ion-pair reverse phase chromatogram of a mixture of 5 Se standards with ICP- MS (5 ng/ml of each compound; 78 Se monitored) (Marchante-Gayón et al., 2000). Trifluoracetic acid (TFA) and perfluoroaliphatic acids were used as pairing reagents for the separation of seleno-amino acids, anionic, cationic and neutral Se compounds in Se enriched yeast, garlic, mushrooms and ramps (Kotrebai et al., 2000). Heptafluorobutanoic acid (HFBA) was used as a superior ion-pairing reagent to improve the resolution of many organoselenium species compared to either pentafluoropropionic acid (PFPA) or TFA (Kotrebai et al., 2000). For example, a mixture of 20 Se compounds was separated on a C- 8 column using isocratic 1% methanol and 0.1% HFBA as the mobile phase, obtaining good resolution of the individual species (Kotrebai, Tyson et al., 2000). A mobile phase containing HFBA was also used in parallel studies with ESI-MS in order to obtain the structural identification of Se Species (Wrobel and Caruso, 2002). IP-RPC has similarly been used for Se speciation in animal tissues and serum (Stürup et al., 2005). Mixed ion-pairing reagents (butanesulphonic acid and tetramethylammonium hydroxide) were also used to simultaneously separate inorganic and organic species with good resolution of seven species (selenite, selenate, TMSe +, SeCys 2, SeMet, selenoethionine and selenocyatamine) (Zheng and Kosmus, 2000). To obtain high efficiency, resolution and speed of separation of selenocompounds, ultra-performance liquid chromatography was utilised (Bendahl et al., 2005). The main IP-RPC-ICP-MS methods and techniques currently being utilised for selenium speciation using ion-pairing reagents are summarised in Table

52 Table 2.4 Applications of ion-pairing reverse-phase chromatography for Se speciation Sample Se target molecules Mobile phase Detection Reference Se-accumulating Brassica Se(IV), SeCys 2, SeMet Four columns: XTerra RP-C18 (3.5 and 5µm), ICP-MS (Kahakachchi juncea (Indian mustard) and and SeEt Symmetry Shield RP-C18 and C8, 0.1% HFBA, 1% et al., 2004) selenised yeast MeOH Dill (Anethum graveolens SeCys 2, MeSeMet, L.) root, stem, leaf SeMC, TMSe, SeMet, Se(IV) and Se(VI) Nutritional supplements and Se(IV), Se(VI), SeCys 2, Chinese tea SeMet and SeEt Human Urine TMSe, SeCys 2, Se(IV), SeUr, SeMet and SeEt Yeast-based supplements S- (methylseleno)cysteine, SeMet and Se-oxide Biofortified products SeMet, SeCys 2, Se(IV) and Se(VI) Allima C8 column, mobile phase: 0.15% HFBA, 5% MeOH ph 2.1 ICP-MS (Cankur et al., 2006) Nucleosil 120 A, C18, 30 mm ammonium formate, ph ICP-MS (hexapole (Marchante- 5.0, 5% MeOH, 10 mm TBAA, isocratic collision/reaction cell Gayón et al., with meinhard, microconcentric 2000) & hydraulic high pressure nebulisers) C8 column, 13 mm tetrabutylammonium hydroxide, ICP-MS (Pan et al., 1.3% MeOH, ph ) RP-C8 column, 0.1% HFBA, 1% MeOH ICP-MS (Amoako et al., 2007) C8, 0.1% HFBA, 10% MeOH Species-unspecific (Kirby et al., isotope dilution ICP- 2008) MS 35

53 Human urine Se(VI), SeUr, SeMet, RP C18, mixed ion-pairing mobile phase containing 2.5 SeEt and TMSe + mm sodium 1-butanesulphonate and 8 mm Urine from before and during supplementation with SeMet Brazil nuts SeMet, MeSeMet,SeMC, TMSe and selenogammaaminobutyr ic acid (SeGaba) SeMet, SeEt, SeCys 2 and unidentified Se species tetramethylammonium hydroxide Luna C8, mobile phase: (1) 10 mm HFBA, 20% MeOH (2) 3 mm nonafluoropentanoic acid (NFPA), 20% MeOH (3) 0.2 mm tridecafluoroheptanoic acid (TFHA), 20% MeOH Alltima C8, 5mM citric acid and 5mM hexanesulphonic acid, ph 3.5, 10% MeOH ICP-MS ICP-MS using a laboratory made direct injection nebuliser for allowing high MeOH concentration ICP-MS (Zheng et al., 2002) (Gammelgaard et al., 2002) (Vonderheide et al., 2002b) Se-enriched garlic (Allium SeCystm, SeMet, SeMC, XTerra MS C18, ICP-MS and ESI-MS- (Dumont et sativum) γ-glutamyl-semc and (1) 0.01% tetraethylammonium chloride (TEACl), 2% MS al., 2006a) two unknown Se MeOH, ph 4.5 compounds (2) 2% MeOH Se-enriched chives(allium Se(IV), Se(VI), SeCys 2, (1) Sperdex Peptide HR 10/300, 30 mm Tris buffer, ICP-MS (Kápolna et schoenoprasum) SeMC and SeMet ph7.5 al., 2007) (2) Alltima C8, 0.1%heptafluorobutyric acid, 5% MeOH, ph2.5 (3) Daicel Crownpak CR(+), 0.1M perchloric acid, ph

54 Se-enriched onion leaves Se(IV), SeCys 2, SeMC Alltima C8, 5mM citric acid, 5 mm hexanesulphonic ICP-MS (Wróbel et al., and SeMet acid, ph 4.5, 5% MeOH 2004) Se-enriched yeast, ramp, Se compounds, Symmetry Shield RP C8, ICP-MS and ESI-MS (Kotrebai et garlic, onion, Astragalus unidentified species (1) 0.1%HFBA, 1% MeOH al., 2000) praleongus and Brassica (2) 0.1%TFA, 1%MeOH juncea 37

55 The advantages of IP-RPC are that this mode incorporates flexibility and versatility. Efficient separation is possible for a wide range of selenium species, including both inorganic and organic Se. However, Se speciation has limitations when hyphenated techniques are used in that organic solvents and ion-pairing reagents used in the mobile phase are not compatible with the ICP plasma. This is a consequence of the alteration of the ionisation characteristics of the Ar plasma and production of polyatomic ions, as well as changes in structure of selenium species. Therefore, concerns about analyte stability during separation require more consideration of IP-RPC than of RPC alone. As mentioned before, organic solvents used in RP or IP-RPC can cause serious interferences in the ICP plasma. Gradient elution can reduce sensitivity in ICP-MS. To overcome these problems, micro and nano-flow nebulisers have been introduced which have low solvent load (Marchante-Gayón et al., 2000). Alternatively, post-column dilution or post-column addition of standards has been described. Oxygen addition has been recommended and Pt cones may be used instead of the common Ni cones (Michalke, 2002b; Michalke, 2002a). Apart from organic solvent, selenium speciation by IP-RPC is dependent on the selected ion-pairing reagents. As seen in Table 2.4, TFA and HFBA are the most widely applied ion-pairing reagents particularly for organic selenium species in plant material. Mixed ionpairing reagents, (e.g. tetramethylammonium salts) used in IP-RPC are becoming of increasing interest for both inorganic and organic selenium species. 2.4 Conclusions The determination of Se speciation is of great importance in understanding the biological and physiological functions of Se, as well as its potential health benefits. Hence, Se speciation analysis will continue to rapidly grow in importance in the near future. Along with hyphenated techniques they provide maximum sensitivity, low detection limits and identification of element specific species. Despite recent advances in speciation analysis, direct determination of trace Se in complexes matrix is still very difficult and therefore separation/pre-concentration techniques are frequently required prior to any analysis. In addition, various chromatographic separation modes can be coupled with ICP-MS, but factors such as separation mechanisms, mobile phase, ph and the steps involved in sample 38

56 preparation must be carefully considered to prevent the inter-conversion of species and to ensure accurate characterisation of the sample. Furthermore, the interference from matrix and argon gas can be improved by using dynamic reaction/collision cell technology and this area will probably continue to develop and grow more rapidly in the near future. The following chapters will firstly describe implementation of instrumentation and then fully explore the behaviour of major selenium species in various environmental and biological matrices using liquid chromatographic separation coupled with inductively coupled plasma mass spectrometry (LC-ICP-MS). Initially, inorganic species will be investigated using ion-exchange and ion-exclusion chromatography for trace level detection. The application of separation methods will then be described for a variety of aqueous environmental samples, such as water and soil extracts. 39

57 2.5 References Aeungmaitrepirom, W., Hagcge, A., Asfari, Z., Bennouna, L., Vicens, J. and Leroy, M. (1999) Selenium speciation and preconcentration by a novel diammoniumcalix arene. Tetrahedron Letters, 40, Amoako, P. O., Kahakachchi, C. L., Dodova, E. N., Uden, P. C. and Tyson, J. F. (2007) Speciation, quantification and stability of selenomethionine, S-(methylseleno) cysteine and selenomethionine Se-oxide in yeast-based nutritional supplements. Journal of Analytical Atomic Spectrometry, 22, Amweg, E. L., Stuart, D. L. and Weston, D. P. (2003) Comparative bioavailability of selenium to aquatic organisms after biological treatment of agricultural drainage water. Aquatic Toxicology 63, Ayouni, L. (2007) Speciation of Selenium in a Commercial Dietary Supplement by Liquid Chromatography Coupled with Inductively Coupled Plasma Mass Spectrometry (ICP-MS). Journal of Toxicology and Environmental Health, Part A. 70, Ayouni, L., Barbier, F., Imbert, J., Gauvrit, J., Lantéri, P. and Grenier-Loustalot, M. (2006) New separation method for organic and inorganic selenium compounds based on anion exchange chromatography followed by inductively coupled plasma mass spectrometry. Analytical and Bioanalytical Chemistry. 385, B'Hymer, C. and Caruso, J. A. (2000) Evaluation of yeast-based selenium food supplements using highperformance liquid chromatography and inductively coupled plasma mass spectrometry. Journal of Analytical Atomic Spectrometry, 15, B Hymer, C. and Caruso, J. A. (2006) Selenium speciation analysis using inductively coupled plasma-mass spectrometry. Journal of Chromatography A, 1114, Bandura, D. R., Baranov, V. I. and Tanner, S. D. (2001) Reaction chemistry and collisional processes in multipole devices for resolving isobaric interferences in ICP MS. Fresenius' journal of analytical chemistry, 370, Bendahl, L., Sturup, S., Gammelgaard, B. and Hansen, S. H. (2005) UPLC-ICP-MS a fast technique for speciation analysis. Journal of Analytical Atomic Spectrometry, 20, Bodó, E. T., Stefánka, Z., Ipolyi, I., Sörös, C., Dernovics, M. and Fodor, P. (2003) Preparation, homogeneity and stability studies of a candidate LRM for Se speciation. Analytical and Bioanalytical Chemistry. 377,

58 Cabanero, A. I., Madrid, Y. and Camara, C. (2005) Enzymatic probe sonication extraction of Se in animal-based food samples: a new perspective on sample preparation for total and Se speciation analysis. Analytical and Bioanalytical Chemistry. 381, Cabañero, A. I., Madrid, Y. and Cámara, C. (2004) Selenium and mercury bioaccessibility in fish samples: an in vitro digestion method. Analytica Chimica Acta. 526, Cai, X. J., Block, E., Uden, P. C., Quimby, B. D. and Sullivan, J. J. (1995) Allium Chemistry: Identification of Natural Abundance Organoselenium Compounds in Human Breath after Ingestion of Garlic Using Gas Chromatography with Atomic Emission Detection. Journal of Agricultural and Food Chemistry, 43, Camel, V. (2003) Solid phase extraction of trace elements. Spectrochimica Acta Part B: Atomic Spectroscopy, 58, Cankur, O., Yathavakilla, S. K. V. and Caruso, J. A. (2006) Selenium speciation in dill (Anethum graveolens L.) by ion pairing reversed phase and cation exchange HPLC with ICP-MS detection. Talanta, 70, Capelo, J. L., Fernandez, C., Pedras, B., Santos, P., Gonzalez, P. and Vaz, C. (2006) Trends in selenium determination/speciation by hyphenated techniques based on AAS or AFS. Talanta, 68, Capelo, J. L., Ximenez-Embun, P., Madrid-Albarran, Y. and Camara, C. (2004) Enzymatic Probe Sonication: Enhancement of Protease-Catalyzed Hydrolysis of Selenium Bound to Proteins in Yeast. Analytical Chemistry-Washington DC-, 76, Chassaigne, H., Chery, C. C., Bordin, G. and Rodriguez, A. R. (2002) Development of new analytical methods for selenium speciation in selenium-enriched yeast material. Journal of Chromatography A, 976, Chatterjee, A., Shibata, Y. and Morita, M. (2001a) Determination of selenomethionine by high performance liquid chromatography-direct hydride generation-atomic absorption spectrometry. Microchemical Journal, 69, Chatterjee, A., Shibata, Y., Yoneda, M., Banerjee, R., Uchida, M., Kon, H. and Morita, M. (2001b) Identification of Volatile Selenium Compounds Produced in the Hydride Generation System from Organoselenium Compounds. Analytical Chemistry- Washington DC-, 73, Chery, C. C., Gunther, D., Cornelis, R., Vanhaecke, F. and Moens, L. (2003) Detection of metals in proteins by means of polyacrylaminde gel electrophoresis and laser 41

59 ablation-inductively coupled plasma-mass spectrometry: Application to selenium. Electrophoresis 24, Cobo Fernandez, M. G., Palacios, M. A. and Camara, C. (1983) Flow-injection and continuous-flow systems for the determination of Se (IV) and Se (VI) by hybride generation atomic absorption spectrometry with on-line prereduction of Se (IV) to Se (IV). Analytica Chimica Acta, 283, Combs, G. F. (2001) Global aspects of selenium in nutrition. British Journal of Nutrition 85, Darrouzès, J., Bueno, M., Lespès, G. and Potin-Gautier, M. (2005) Operational optimisation of ICP octopole collision/reaction cell MS for applications to ultratrace selenium total and speciation determination. Journal of Analytical Atomic Spectrometry, 20, De-qiang, Z., Han-wen, S. and Li-li, Y. (1997) Determination of trace inorganic selenium in organoselenium (selenosugar) oral nutrition liquids by graphite furnace atomic absorption spectrometry with hydride generation. Fresenius'Journal of Analytical Chemistry, 359, Dernovics, M., Stefánka, Z. and Fodor, P. (2002) Improving selenium extraction by sequential enzymatic processes for Se-speciation of selenium-enriched Agaricus bisporus. Analytical and Bioanalytical Chemistry, 372, Dumont, E., Ogra, Y., Vanhaecke, F., Suzuki, K. T. and Cornelis, R. (2006a) Liquid chromatography mass spectrometry (LC MS): a powerful combination for selenium speciation in garlic (Allium sativum). Analytical and Bioanalytical Chemistry, 384, Dumont, E., Vanhaeche, F. and Cornelis, R. (2006b) Selenium speciation from food source to metabolites: a critical review. Analytical and Bioanalytical Chemistry, 385, Dumont, E., Vanhaecke, F. and Cornelis, R. (2004) Hyphenated techniques for speciation of Se in in vitro gastrointestinal digests of Saccharomyces cerevisiae. Analytical and Bioanalytical Chemistry, 379, Ebdon, L. (2001) Trace element speciation for environment, food and health, Royal Society of Chemistry. Francesconi, K. A. and Pannier, F. (2004) Selenium Metabolites in Urine: A Critical Overview of Past Work and Current Status. Clinical Chemistry, 50,

60 Frankerberger, J. W. T. and Engberg, R. A. E. (1998) Environmental chemistry of selenium, Marcel Dekker Inc., New York, USA. Fritz, J. S. (2005) Factors affecting selectivity in ion chromatography. Journal of Chromatography A, 1085, Gammelgaard, B., Bendahl, L., Sidenius, U. and Jøns, O. (2002) Selenium speciation in urine by ion-pairing chromatography with perfluorinated carboxylic acids and ICP- MS detection. Journal of Analytical Atomic Spectrometry, 17, Gammelgaard, B. and Jrns, O. (2000) Determination of selenite and selenate in human urine by ion chromatography and inductively coupled plasma mass spectrometry. Journal of Analytical Atomic Spectrometry, 15, Gilon, N. and Potin-Gautier, M. (1996) Capillary electrophoresis applied to the determination of some seleno compounds. Journal of Chromatography A, 732, Haddad, P. R. and Jackson, P. E. (1990) Ion Chromatography: Principles and ApplicationsJournal of Chromatography, Library Vol. 46. Elsevier, Amsterdam. Haddad, P. R., Mattew, J. S., Madden, J. E. and Dicinoski, G. W. (2004) A computerbased undergraduate exercise Using internet-accessible simulation software for the study of retention behaviour and optimization of separation conditions in ion chromatography. Journal of Chemical education 81, He, Y., Moreda-piñeiro, J., Cervera, M. L. and Guardia, M. (1998) Direct determination of dissolved selenium (iv) and selenium (vi) in sea-water by continuous flow hydride generation atomic fluorescence spectrometry. Journal of Analytical Atomic Spectrometry, 13, Hill, S. J., Pitts, L. and Worsfold, P. (1995) Investigation into the kinetics of selenium (VI) reduction using hydride generation atomic fluorescence detection. Journal of analytical atomic spectrometry, 10, Hudnik, V. and Gomiscek, S. (1984) The atomic absorption spectrometric determination of arsenic and selenium in mineral waters by electrothermal atomization. Analytica Chimica Acta, 157, Huerta, V. D., Reyes, L. H., Marchante-Gayon, J. M., Sanchez, M. L. F. and Sanz-Medel, A. (2003) Total determination and quantitative speciation analysis of selenium in yeast and wheat flour by isotope dilution analysis ICP-MS. Journal of Analytical Atomic Spectrometry, 18,

61 Huerta, V. D., Sa nchez, M. L. F. and Sanz-Medel, A. (2004) Quantitative selenium speciation in cod muscle by isotope dilution ICP-MS with a reaction cell: comparison of different reported extraction procedures Journal of Analytical Atomic Spectrometry, 19, Huerta, V. D., Sánchez, M. L. F. and Sanz-Medel, A. (2006) An attempt to differentiate HPLC-ICP-MS selenium speciation in natural and selenised Agaricus mushrooms using different species extraction procedures. Analytical Bioanalytical Chemistry, 384, Infante, H. G., Hearn, R. and Catterick, T. (2005) Current mass spectrometry strategies for selenium speciation in dietary sources of high-selenium. Analytical Bioanalytical Chemistry, 382, Infante, H. G., O Connor, G., Rayman, M., Wahlen, R., Entwisle, J., Norris, P., Hearn, R. and Catterick, T. (2004) Selenium speciation analysis of selenium-enriched supplements by HPLC with ultrasonic nebulisation ICP-MS and electrospray MS/MS detection. Journal of Analytical Atomic Spectrometry, 19, Ip, C., Birringer, M., Block, E., Kotrebai, M., Tyson, J. F., Uden, P. C. and Lisk, D. J. (2000) Chemical Speciation Influences Comparative Activity of Selenium- Enriched Garlic and Yeast in Mammary Cancer Prevention. Journal of Agricultural and Food Chemistry, 48, Ipolyi, I., Stefánka, Z. and Fodor, P. (2001) Speciation of Se (IV) and the selenoamino acids by high-performance liquid chromatography direct hydride generation atomic fluorescence spectrometry. Analytica Chimica Acta, 435, Jamba, L., Nehru, B. and Bansal, M. P. (1997) Redox modulation of selenium binding proteins by cadmium exposures in mice Molecular and Cellular Biochemistry, 177, Johansson, M., Bordin, G. and Rodriguez, A. R. (2000) Feasibility study of ionchromatography microwave assisted on-line species conversion hydride generation atomic absorption spectrometry for selenium speciation analysis of biological material. Analyst 125, 273. Juresa, D., Kuehnelt, D. and Francesconi, K. A. (2006) Consequences of Vapor Enhancement on Selenium Speciation Analysis by HPLC/ICPMS. Analytical chemistry -Washington, DC- 78, Kahakachchi, C., Boakye, H. T., Uden, P. C. and Tyson, J. F. (2004) Chromatographic speciation of anionic and neutral selenium compounds in Se-accumulating Brassica 44

62 juncea (Indian mustard) and in selenized yeast. Journal of Chromatography A, 1054, Kápolna, E. and Fodor, P. (2006) Speciation analysis of selenium enriched green onions (Allium fistulosum) by HPLC-ICP-MS. Microchemical Journal, 84, Kápolna, E., Shah, M., Caruso, J. A. and Fodor, P. (2007) Selenium speciation studies in Se-enriched chives (Allium schoenoprasum) by HPLC-ICP MS. Food Chemistry, 101, Kirby, J. K., Lyons, G. H. and Karkkainen, M. P. (2008) Selenium Speciation and Bioavailability in Biofortified Products Using Species-Unspecific Isotope Dilution and Reverse Phase Ion Pairing- Inductively Coupled Plasma- Mass Spectrometry. Journal of Agricultural and Food Chemistry, 56, Kobayashi, Y., Ogra, Y. and Suzuki, K. T. (2001) Speciation and metabolism of selenium injected with 82Se-enriched selenite and selenate in rats. Journal of Chromatography B: Biomedical Sciences and Applications, 760, Kotrebai, M., Tyson, J. F., Uden, P. C., Birringer, M. and Block, E. (2000) Selenium speciation in enriched and natural samples by HPLC-ICP-MS and HPLC-ESI-MS with perfluorinated carboxylic acid ion-pairing agents. The Analyst, 125, Koyama, H., Omura, K., Ejima, A., Kasanuma, Y., Watanabe, C. and Satoh, H. (1999) Separation of Selenium-Containing Proteins in Human and Mouse Plasma Using Tandem High-Performance Liquid Chromatography Columns Coupled with Inductively Coupled Plasma-Mass Spectrometry. Analytical Biochemistry, 267, Kremer, D., Ilgen, G. and Feldmann, J. (2005) GC ICP MS determination of dimethylselenide in human breath after ingestion of 77 Se-enriched selenite: monitoring of in-vivo methylation of selenium. Analytical and Bioanalytical Chemistry, 383, Larsen, E. H., Hansen, M., Fan, T. and Vahl, M. (2001) Speciation of selenoamino acids, selenonium ions and inorganic selenium by ion exchange HPLC with mass spectrometric detection and its application to yeast and algae. Journal of Analytical Atomic Spectrometry, 16, Larsen, E. H., Hansen, M., Paulin, H., Moesgaard, S., Reid, M. and Rayman, M. (2004) Speciation and Bioavailability of Selenium in Yeast-Based Intervention Agents Used in Cancer Chemoprevention Studies. Journal of AOAC International, 87,

63 Larsen, E. H., Sloth, J., Hansen, M. and Moesgaard, S. (2003) Selenium speciation and isotope composition in 77 Se-enriched yeast using gradient elution HPLC separation and ICP-dynamic reaction cell-ms. Journal of Analytical Atomic Spectrometry, 18, Marchante-Gayón, J. M., Feldmann, I., Thomas, C. and Jakubowski, N. (2001) Speciation of selenium in human urine by HPLC-ICP-MS with a collision and reaction cell. Journal of Analytical Atomic Spectrometry, 16, Marchante-Gayón, J. M., Thomas, C., Feldmann, I. and Jakubowski, N. (2000) Comparison of different nebulisers and chromatographic techniques for the speciation of selenium in nutritional commercial supplements by hexapole collision and reaction cell ICP-MS. Journal of Analytical Atomic Spectrometry, 15, McSheehy, S., Pannier, F., Szpunar, J., Potin-Gautier, M. and Lobinski, R. (2002) Speciation of seleno compounds in yeast aqueous extracts by three-dimensional liquid chromatography with inductively coupled plasma mass spectrometric and electrospray mass spectrometric detection. The Analyst, 127, Michalke, B. (2002a) The coupling of LC to ICP-MS in element speciation Part II: Recent trends in application. Trends in analytical chemistry, 21, Michalke, B. (2002b) The coupling of LC to ICP-MS in element speciation: I. General aspects. Trends in analytical chemistry, 21, Montes-Bay on, M., Molet, M. J. D., Gonz alez, E. B. and Sanz-Medel, A. (2006) Evaluation of different sample extraction strategies for selenium determination in selenium-enriched plants (Allium sativum and Brassica juncea) and Se speciation by HPLC-ICP-MS. Talanta, 68, Montes-Bayón, M., Yanes, E. G., León, C. P., Jayasimhulu, k., Stalcup, A., Shann, J. and Caruso, J. L. (2002) Initial Studies of Selenium Speciation in Brassica juncea by LC with ICPMS and ES-MS Detection: an Approach for Phytoremediation Studies. Analytical Chemistry, 74, Moreno, P., Quijano, M. A., Gutierrez, A. M., Perez-Conde, M. C. and Camara, C. (2004) Study of selenium species distribution in biological tissues by size exclusion and ion exchange chromatography inductively coupled plasma-mass spectrometry. Analytica Chimica Acta 524, Mounicou, S., McSheehy, S., Szpunar, J., Potin-Gautier, M. and Lobinski, R. (2002) Analysis of selenized yeast for selenium speciation by size-exclusion 46

64 chromatography and capillary zone electrophoresis with inductively coupled plasma mass spectrometric detection (SEC-CZE-ICP-MS). Journal of Analytical Atomic Spectrometry, 17, Nakazato, T., Tao, H., Taniguchi, T. and Isshiki, K. (2002) Determination of arsenite, arsenate, and monomethylarsonic acid in seawater by ion-exclusion chromatography combined with inductively coupled plasma mass spectrometry using reaction cell and hydride generation techniques. Talanta, 58, Ogra, Y., Ishiwata, K., Encinar, J. R., Łobin ski, R. and Suzuki, K. T. (2004) Speciation of selenium in selenium-enriched shiitake mushroom, Lentinula edodes. Analytical and Bioanalytical Chemistry, 379, Ogra, Y., Ishiwata, K., Iwashita, Y. and Suzuki, K. T. (2005a) Simultaneous speciation of selenium and sulfur species in selenized odorless garlic (Allium sativum L. Shiro) and shallot (Allium ascalonicum) by HPLC-inductively coupled plasma-(octopole reaction system)-mass spectrometry and electrospray ionization-tandem mass spectrometry. Journal of Chromatography A, 1093, Ogra, Y., Ishiwata, K. and Suzuki, K. T. (2005b) Effects of deuterium in octopole reaction and collision cell ICP-MS on detection of selenium in extracellular fluids. Analytica Chimica Acta, 554, Pan, F., Tyson, J. F. and Uden, P. C. (2007) Simultaneous speciation of arsenic and selenium in human urine by high-performance liquid chromatography inductively coupled plasma mass spectrometry. Journal of Analytical Atomic Spectrometry, 22, Pedrero, Z., Elvira, D., Cámara, C. and Madrid, Y. (2007) Selenium transformation studies during Broccoli (Brassica oleracea) growing process by liquid chromatography inductively coupled plasma mass spectrometry (LC ICP-MS). Analytica Chimica Acta, 596, Pedrero, Z., Madrid, Y. and Cámara, C. (2006) Selenium species bioaccessibility in enriched radish (Raphanus sativus): a potential dietary source of selenium. Journal of Agricultural and Food Chemistry, 54, Pelaez, M. V., Bayon, M. M., Alonso, J. I. G. and Sanz-Medel, A. (2000) A comparison of different derivatisation approaches for the determination of selenomethionine by GC-ICP-MS. Journal of Analytical Atomic Spectrometry, 15, Pitts, L., Fisher, A., Worsfold, P. and Hill, S. J. (1995) Selenium speciation using highperformance liquid chromatography-hydride generation atomic fluorescence with 47

65 on-line microwave reduction. Journal of analytical atomic spectrometry, 10, Pohl, P. (2004) Hydride generation recent advances in atomic emission spectrometry. Trends in analytical chemistry, 23, Polatajko, A., Jakubowski, N. and Szpunan, J. (2006) State of the art report of selenium speciation in biological samples. Journal of Analytical Atomic Spectrometry, 21, Połatajko, A., Sliwka-Kaszynska, M., Dernovics, M., Ruzik, R., Encinar, J. R. and Szpunar, J. (2004) A systematic approach to selenium speciation in selenized yeast. Journal of Analytical Atomic Spectrometry, 19, Pyrzynska, K. (1996) Speciation analysis of some Organic Selenium Compounds. Analyst 121, 77R-83R. Pyrzynska, K. (2002) Determination of Selenium Species in Environmental Samples. Michrochimica Acta, 140, Rayman, M. P. (2007a) Selenium in cancer prevention: a review of the evidence and mechanism of action. Proceedings of the Nutrition Society, 64, Rayman, M. P. (2007b) The use of high-selenium yeast to raise selenium status: how does it measure up? British Journal of Nutrition, 92, Renli Ma, Mcleod, C. W., Tomlinson, K. and Poole, R. K. (2004) Speciation of proteinbound trace elements by gel electrophoresis and atomic spectrometry. Electrophoresis, 25, Reyes, L. H., Gayón, J. M. M., Alonso, J. I. G. and Sanz-Medel, A. (2003) Determination of selenium in biological materials by isotope dilution analysis with an octapole reaction system ICP-MS. Journal of Analytical Atomic Spectrometry, 18, Reyes, L. H., Marchante-Gayon, J. M., Alonso, J. I. G. and Sanz-Medel, A. (2006) Application of Isotope Dilution Analysis for the Evaluation of Extraction Conditions in the Determination of Total Selenium and Selenomethionine in Yeast- Based Nutritional Supplements. Journal of Agricultural and Food Chemistry, 54, Reyes, L. H., Sanz, F. M., Espílez, P. H., Marchante-Gayón, J. M., Alonso, J. I. G. and Sanz-Medel, A. (2004) Biosynthesis of isotopically enriched selenomethionine: application to its accurate determination in selenium-enriched yeast by isotope dilution analysis-hplc-icp-ms. Journal of Analytical Atomic Spectrometry, 19,

66 Robles, L. C., Feo, J. C., de Celis, B., Lumbreras, J. M., Garci a-olalla, C. and Aller, A. J. (1999) Speciation of selenite and selenate using living bacteria. Talanta, 50, Siwek, M., Bari Noubar, A., Bergmann, J., Niemeyer, B. and Galunsky, B. (2006) Enhancement of enzymatic digestion of Antarctic krill and successive extraction of selenium organic compounds by ultrasound treatment. Analytical and Bioanalytical Chemistry, 384, Siwek, M., Galunsky, B. and Niemeyer, B. (2005) Isolation of selenium organic species from antarctic krill after enzymatic hydrolysis. Analytical and Bioanalytical Chemistry, 381, Stefánka, Z., Ipolyi, I., Dernovics, M. and Fodor, P. (2001) Comparison of sample preparation methods based on proteolytic enzymatic processes for Se-speciation of edible mushroom (Agaricus bisporus) samples. Talanta, 55, Stripeikis, J., Pedro, J., Bonivardi, A. and Tudino, M. (2004) Determination of selenite and selenate in drinking water: a fully automatic on-line separation/pre-concentration system coupled to electrothermal atomic spectrometry with permanent chemical modifiers. Analytica Chimica Acta, 502, Sturgeon, R. E., Willie, S. N. and Berman, S. S. (1985) Preconcentration of selenium and antimony from seawater for determination by graphite furnace atomic absorption spectrometry. Analytical Chemistry, 57, 6-9. Stürup, S., Hayes, R. B. and Peters, U. (2005) Development and application of a simple routine method for the determination of selenium in serum by octopole reaction system ICPMS. Analytical and Bioanalytical Chemistry, 381, Sun, B., Macka, M. and Haddad, P. R. (2004) Speciation of arsenic and selenium by capillary electrophoresis. Journal of Chromatography A, 1039, Takatera, K., Osaki, N., Yamaguchi, H. and Watanabe, T. (1994) HPLC/ICP Mass Spectrometric Study of the Selenium Incorporation into Cyanobacterial Metallothionein Induced under Heavy-Metal Stress. Analytical Sciences, 10, Tan, J. A., Zhu, W., Wang, W., Li, R., Hou, S., Wang, D. and Yang, L. (2002) Selenium in soil and endemic diseases in China. The Science of The Total Environment 284, Templeton, D. M., Ariese, F., Cornelis, R., Danielsson, L. G., Muntau, H., Van leeuwen, H. P. and Lobinski, R. (2000) Pure Application Chemistry, 72,

67 Tinggi, U. (2003) Essentiality and toxicity of selenium and its status in Australia: A review. Toxicology Letters, 137, Uden, P. C. (2002) Modern trends in the speciation of selenium by hyphenated techniques. Analytical and Bioanalytical Chemistry, 373, Uden, P. C., Boakye, H. T., Kahakachchi, C. and Tyson, J. F. (2004) Selective detection and identification of Se containing compounds-review and recent developments. Journal of Chromatography A, 1050, Vanhaecke, F. and Moens, L. (1999) Recent trends in trace element determination and speciation using inductively coupled plasma mass spectrometry. Fresenius'Journal of Analytical Chemistry, 364, Vonderheide, A. P., Montes-Bayon, M. and Caruso, J. A. (2002a) Solid-phase microextraction as a sample preparation strategy for the analysis of seleno amino acids by gas chromatography-inductively coupled plasma mass spectrometry. The Analyst, 127, Vonderheide, A. P., Wrobel, K., Kannamkumarath, S. S., B'Hymer, C., Montes-Bayon, M., de Leon, C. P. and Caruso, J. A. (2002b) Characterization of Selenium Species in Brazil Nuts by HPLC-ICP-MS and ES-MS. Journal of Agricultural and Food Chemistry, 50, Wake, B. D., Bowie, A. R., Butler, E. C. V. and Haddad, P. R. (2004) Modern preconcentration methods for the determination of selenium species in environmental water samples. Trends in Analytical Chemistry 23, Wallschlager, D. and London, J. (2004) Determination of inorganic selenium species in rain and sea waters by anion exchange chromatography-hydride generationinductively-coupled plasma-dynamic reaction cell-mass spectrometry (AEC-HG-ICP-DRC-MS). Journal of Analytical Atomic Spectrometry, 19, Wang, T. (2007) Liquid Chromatography-Inductively Coupled Plasma Mass Spectrometry (LC-ICP-MS). Journal of Liquid Chromatography & Related Technologies, 30, Warburton, E. and Goenaga-Infante, H. (2007) Methane mixed plasma improved sensitivity of inductively coupled plasma mass spectrometry detection for selenium speciation analysis of wheat-based food. Journal of Analytical Atomic Spectrometry, 22,

68 Whanger, P. D. (2002) Selenocompounds in Plants and Animals and their Biological Significance. Journal of the American College of Nutrition, 21, Wilber, C. G. (1980) Toxicology of selenium: A review. Clinical Toxicology, 17, Wrobel, K. and Caruso, J. A. (2002) Selenium speciation in low molecular weight fraction of Se-enriched yeasts by HPLC-ICP-MS: detection of selenoadenosylmethionine. Journal of Analytical Atomic Spectrometry, 17, Wróbel, K., Kannamkumarath, S. S., Caruso, J. A., Wysocka, I. A., Bulska, E., Swia tek, J. and Wierzbicka, M. (2004) HPLC ICP-MS speciation of selenium in enriched onion leaves a potential dietary source of Se-methylselenocysteine. Food Chemistry, 86, Wrobel, K., Kannamkumarath, S. S., Wrobel, K. and Caruso, J. A. (2003) Hydrolysis of proteins with methanesulfonic acid for improved HPLC-ICP-MS determination of seleno-methionine in yeast and nuts. Anaytical Bioanalytical Chemistry, 375, Xu, F., Ni, Z., Qiu, D., You, J., Hu, W. and Yang, P. (2002) Alkaline fusion-alkaline mode HG-ICP spectrometry for selenium determination. Journal of Analytical Atomic Spectrometry, 17, Yang, L., Mester, Z. and Sturgeon, R. E. (2004a) Determination of Methionine and Selenomethionine in Yeast by Species-Specific Isotope Dilution GC/MS. Analytical Chemistry-Washington DC-, 76, Yang, L., Sturgeon, R. E., McSheehy, S. and Mester, Z. (2004b) Comparison of extraction methods for quantitation of methionine and selenomethionine in yeast by species specific isotope dilution gas chromatography mass spectrometry. Journal of Chromatography A, 1055, Zheng, J. and Kosmus, W. (2000) Retention study of inorganic and organic selenium compounds on a silica-based reversed phase column with mixed ion-pairing reagents. Chromatographia 51, Zheng, J., Ohata, M. and Furuta, N. (2002) Reversed-phase liquid chromatography with mixed ion-pair reagents coupled with ICP-MS for the direct speciation analysis of selenium compounds in human urine. Journal of Analytical Atomic Spectrometry, 17, Zoorob, G. K., McKiernan, J. W. and Caruso, J. A. (1998) ICP-MS for elemental speciation studies. Microchimica Acta, 128,

69 Chapter 3 Implementation of instrumentation 52

70 3.1 Introduction In this chapter instrumentation details will be presented. The principles and common features of ICP-MS as referred to in the experimental chapters will be described. However, the actual method development using different separation modes and experimental specifics have been deferred to the relevant chapters. Inductively coupled plasma mass spectrometry (ICP-MS) is a powerful technique for detecting elements. This very sensitive and selective method enables detection of selenium species in a wide variety of samples due to its species specificity, wide linear range, and low detection limits when coupled with liquid chromatography (LC). A number of authors have discussed the use of LC-ICP-MS for Se speciation. A number of groups have focused on organic Se species in plants, food and various supplements (Pedrero and Madrid, 2009; Dumont et al., 2006; Polatajko et al., 2006; B Hymer and Caruso, 2006; Uden, 2002; Michalke, 2002b). ICP-MS is the detector of choice for trace element analysis. However, a conventional ICP- MS (quadrupole filter without collision/reaction cell system) is not sufficient when dealing with selenium determinations at trace levels in complicated matrices. Difficulties are mainly due to its high first ionisation potential (9.75 ev) and, as a consequence, its low ionisation in Ar plasma. Another major problem is caused by the most abundant selenium isotopes 80 Se (49.6%) and 78 Se (23.8%), suffering from Ar plasma interferences caused by 40 Ar 40 Ar + and 38 Ar 40 Ar +. Hence, the less abundant Se isotopes ( 77 Se and 82 Se) are generally monitored with less interference. To solve this issue, ICP with high-resolution MS (HRMS) can differentiate between these interferences (B Hymer and Caruso, 2006), although it has less commercial application than a conventional ICP with low resolution MS owing to its high cost. For this reason, ICP-MS equipped with a reaction/collision cell has been developed as a technique that complements ICP-HRMS (B Hymer and Caruso, 2006). In this application, various gases, including hydrogen (H 2 ), helium (He), ammonia (NH 3 ) and oxygen (O 2 ), can be used as reaction/collision gas to reduce spectral interferences (B Hymer and Caruso, 2006; Polatajko et al., 2006). In practice, only He and H 2 options are commonly available for commercial ICP-MS analyses. 53

71 This chapter describes the overall instrumental layout. Emphasis is placed on the interface between the LC system and ICP-MS, the design and chemistry of the reaction/collision cell, as well as all the important MS detection process. Particular attention will be given to the application of ORS technology in ICP-MS with regard to interferences. 3.2 ICP-MS and Collision/Reaction Cell technology Element-specific ICP-MS is based on the ionisation of elements in argon plasma produced at temperatures in the ,000 K range and detecting element ions by mass spectrometry. The main sequences are as follows: Sample introduction Aerosol generation (A) Ionisation by Argon plasma (B) Ion focussing (C) Mass discrimination and detection (D) Data analysis (Fig. 3.1) Figure 3.1 Schematic of ICP-MS processes (adapted from Tanner, Sample introduction is accomplished with a nebuliser and spray chamber (A). The nebuliser creates an aerosol at its tip which is carried into the plasma torch by carrier argon gas. The plasma is formed in a quartz torch (B) by coupling radiofrequency and electromagnetic fields at 27 to 40 MHz through a copper load coil. Free electrons within the plasma are accelerated by the induced magnetic field and bombard other gas atoms, which causes further ionisation and sustains the plasma. Analytes in the aerosol are then vaporised, desolvated, atomised, and finally ionised in ICP plasma. Once ions are formed 54

72 in the plasma, they enter the mass spectrometer through sampler and skimmer cones. The ions are focused using a set of lenses (extraction and omega lens) (C) held at different potentials in the mass spectrometer. The optimum values are determined by maximizing the S/N at the selected mass-to-charge ratio. These lenses are located before the gate valve separating the atmospheric pressure region of the interface from the high vacuum region of the mass spectrometer. Ions may then enter an octopole reaction system (ORS) (Fig. 3.2), which is coaxial or positioned at an angle with the mass analyser and detector. This enhances the high ion transmission and consequently increases the sensitivity, enabling an analysis to be undertaken at trace and ultra-trace levels (Leonhard et al., 2002). Figure 3.2 Schematics of an Agilent ICP-MS with octopole reaction system (ORS) (adapted from Leonhard et al., 2002). The focused ions enter the quadrupole mass analyser (D), which acts as a filter allowing only ions of one mass to charge ratio to be transmitted at a time depending on the applied voltages to the quadrupole rods, while other ions are electrostatically directed from their original path. Quadrupole mass analysers have unit mass resolution. Here, polyatomic interferences can prevent the determination of some elements in some matrices as, for example, 75 As in sea water through interference from 40 Ar 35 Cl + (Thomas, 2002; McCurdy and Woods, 2004; Chen et al., 2007). Koppenaal et al. (2004) reviewed collision/reaction cell technologies in terms of spectral simplicity and improvement in sensitivity and resolution. A collision and reaction cell typically consists of six (hexapole) or eight (octopole) rods to which RF only voltage is applied. While there are various designs from different manufacturers, all use the same 55

73 basic principle to prevent the interfering species entering the cell or reaching the detector (Koppenaal et al., 2004; Yip and Sham, 2007). An octopole reaction system (ORS) (a cell angled relative to the axis of the ion beam) is incorporated in Agilent devices 7500c, 7500cs and GV instruments platform, and a hexapole (HEX) reaction cell (a chicane lens downstream of the cell) is favoured by Thermo Electron X-Series. Alternatively, a dynamic reaction cell (DRC) (an on-axis beam stop) is used in Perkin-Elmer Sciex instruments. The collision/reaction cell is operated by pressurizing the cell interior with gases in reaction and collision mode, respectively. The useful reaction gases in ICP-MS are defined as collision, reaction, oxidation, reduction and other reagent gases (see Table 3.1). In reaction mode, the interferent or the analyte ions undergo chemical reactions with a gas (McCurdy and Woods, 2004; Yamada et al., 2002; Bandura et al., 2001). This technique helps prevent interfering ions from spectrally overlapping with the target isotopes. Helium is an inert gas, and is often used in the collision cell regardless of element (Bandura et al., 2001). Table 3.1 Reagent gases used in collision and reaction cell Collision gases He, Ar, Ne, Xe Reaction gases H 2, NH 3, N 2, CH 4 Oxidation gases O 2, N 2 O, CO 2 Reduction gases H 2, CO Other gases C 2 H 6, C 2 H 4, CH 3 F, CH 3 OH Hydrogen is the most commonly used reaction gas for element analysis in reaction cells because it provides lower detection limits compared to He and dinitrogen oxide (Yamada et al., 2002; Bandura et al., 2001). The reaction of the argon dimer with hydrogen is assumed to be interference association, where hydrogen associates with argon dimers by charge transfer, proton transfer or atom transfer. Charge transfer: ArH + + H 2 H Ar (Removes interference with 39 K + ) Proton transfer: Ar H 2 Ar 2 H + + H (Removes interferences with 78 Se + and 80 Se + ) 56

74 Atom transfer: ArO + + H 2 H 2 O + + Ar (Removes interference with 56 Fe + ) While in collision mode, helium gas can remove polyatomic interferences by one of two mechanisms: (1) Collision induced dissociation; and (2) Kinetic energy discrimination. Dissociation of polyatomic ions into individual components occurs when the energy associated with a collision is greater than the bond energy. In most cases, the second mechanism is followed. Kinetic energy discrimination is based on the fact that polyatomic ions undergo more energy reducing collisions with He gas than analyte ions owing to their larger cross-section (Yamada et al., 2002; Koppenaal et al., 2004). Stopping potential is then applied between the octopole and quadrupole to reject the lower energy interfering ions, while allowing the higher energy analyte ions to enter the quadrupole mass analyser (Bandura et al., 2001; McCurdy and Woods, 2004; Yamada et al., 2002). Spectral interferences are serious problems when detecting Se with a conventional ICP- MS, which only uses a quadrupole filter. These arise from isobaric and polyatomic, as well as non-spectral interferences originating mostly from the sample matrix (such as carbon and chloride in the case of Tris-HCl). Se has six isotopes: 80 Se(49.61%), 78 Se(23.77%), 76 Se(9.37%), 82 Se(8.73%), 77 Se(7.63%), 74 Se(0.89%). Polyatomic spectral interferences 40 Ar 37 Cl + and 81 Br 1 H + affect 77 Se and 82 Se when samples contain traces of chloride or bromide. In argon plasma, several dimer interferences of 40 Ar 36 Ar +, 40 Ar 38 Ar + and 40 Ar 40 Ar + have identical mass as the abundant isotopes 76 Se, 78 Se and 80 Se, respectively. This clearly poses a sensitivity problem. In most cases, 40 Ar 40 Ar + is unavoidable and is reflected in higher background counts leading to higher detection limits, so other less abundant isotopes such as 82 Se(8.73%) or 77 Se(7.63%) are often monitored and detection limits are compromised, making Se determination more difficult (Qiu et al., 2006). As mentioned previously, such interferences can be minimised. The collision/reaction cell technology achieves the removal of polyatomic interferences using physical or chemical means without significant loss in sensitivity. It has helped to increase the signal-to-noise ratio for those isotopes that cannot possibly be monitored by low resolution quadrupole mass analysers. Increasing application of the octopole reaction system (ORS) has resulted in relatively interference-free responses and lower detection limits for trace elements (B Hymer and Caruso, 2006; Uden et al., 2004). Therefore ORS-ICP-MS techniques are 57

75 being effectively employed in a wide range of matrices, especially in biological applications. 3.3 Interfacing of Liquid Chromatography (LC) and ICP-MS Hyphenated technique development has resulted in the on-line coupling of liquid chromatography with an element-specific detector for more effective Se speciation (Pitts et al., 1995; Capelo et al., 2006; Michalke, 2002). Because there is only a single connection between the LC discharge of eluent from the separation column to the ICP-MS, nebulisation is required. Flow rates from 0.1 to 1 ml/min are suited for a regular ICP-MS nebuliser. As mentioned in Chapter 2, under certain circumstances, organic solvents or high salts mobile phases may have to be used, and this can result in compatibility issues with ICP-MS. Nebulisers varying in their design may be effective in reducing mobile phase introduced into the ICP plasma. The development of micro-nebulisers such as direct injection nebuliser (DIN), high efficiency nebuliser (HEN), DIHEN (Fig. 3.3), etc., allows low flow rates of µl/min and, hence less solvent is introduced to the plasma (Jensen et al., 2003; Hettipathirana and Davey, 1998). Figure 3.3 Schematics of DIHEN (Jensen et al., 2003) 3.4 Optimising the ICP-MS ICP-MS parameters ICP-MS conditions were optimised to obtain good results in the experiment chapters, but common parameters are summarised here. An Agilent 7500c ICP-MS (Tokyo, Japan) was made available for use in this study by the centre of environmental risk assessment and remediation (CERAR) at University of South Australia. The liquid chromatography module was an Agilent 1100 equipped with different types of column. Samples were injected using an Agilent 1100 auto-sampler. The mobile phase flow rate and temperature were set according to optimised values. The column effluent was passed via a length of 58

76 PEEK tubing (50 cm, I.D.= 0.13 mm) to a Babington nebuliser and a Scott double pass chamber attached to an Agilent 7500c ICP-MS (Tokyo, Japan) (Fig. 3.4a and 3.4b). This Babington nebuliser which has two holes for sample and carrier gas lines allows the introduction of higher matrix samples. This arrangement is not as efficient as the DIHEN nebuliser (Fig. 3.3). However it is better for environmental samples with generally high dissolved solids contents to prevent blockage of the sampling cone and contamination of the MS detection system (Walton and Goulter, 1985). Fig. 3.4 (a) Fig. 3.4 (b) Figure 3.4 (a) Babington nebuliser; (b) Scott double pass spray chamber as used in the Agilent 7500c ICP-MS instrument. Data was collected using single-ion monitoring (SIM) to ensure maximum sensitivity. When using liquid chromatography (LC), LC-ICP-MS system was controlled and the data was processed using Agilent s Chemstation software package. ICP-MS conditions are presented in Table 3.2. A Babington nebuliser and a cooled double-pass spray chamber were used for sample introduction more suitable for high contaminated environmental samples. The sample flow rate was regulated at 0.4 ml/min with a peristaltic pump. Nickel 59

77 sampler and skimmer cones were used. When the reaction cell was pressurised, the octopole bias was set at -13 V with a quadrupole bias of typically -10 V (see Table 3.2). This has been found to be optimal for elemental sensitivity and a low background. In the standard (no reaction gas) mode the optimal voltage settings for cell exit, octopole and quadrupole bias were -6 and -3V, respectively. Table 3.2 Instrument ICP-MS optimised parameters (specific details in experimental section) ICP-MS Agilent 7500c ICP parameters- RF power 1430 W Plasma gas (Ar) flow 15 L/min Auxiliary gas (Ar) flow 1.0 L/min Carrier gas (Ar) flow 1.15 L/ min Reaction parameters- H 2 gas flow 0 to 5 ml/min He gas flow 0 to 5 ml/min Octopole bias Quadrupole bias Detection parameters- Monitoring mass m/z 78, 80 Integration time 1 s Dwell time 0.5 s. Calibration External Tuning reaction gas (H 2 and He) under ORS The octopole reaction system (ORS) of the ICP-MS was tuned to reach minimum background at optimum sensitivity. The flow rate for He or H 2 is the most significant parameter affecting the background and analytical signals due to increasing gas flow to reduce the interfering 40 Ar 40 Ar + and 38 Ar 40 Ar +. Therefore flow rate must be optimized (McCurdy and Woods, 2004; Yamada et al., 2002). The solution containing 0.1% (v/v) HNO 3 and a spiked solution containing 10 μg/l Se was aspirated into the nebuliser and 60

78 then into the plasma. The gas flow rate varied from 0 to 5 ml/min. One tuning example is shown in Fig. 3.5 (a) for H 2 gas and Fig. 3.5 (b) for He. Fig. 3.5 (a) Fig. 3.5 (b) Figure 3.5 Tuning the ORS for Se78 and Se 80. Influence of H2 (a) and He (b) cell gas flow rates. Detector response (CPS) variation: Blank in blue, spiked 10 μg/l Se standard in red and background equivalent concentration (BEC) μg/l in green (Image taken from the instrument s computer screen). 61

79 By using the optimum flow of H 2, the spectral background was reduced by about five orders of magnitude, from 10 7 CPS to 5x10 2 CPS at m/z 80 using ICP-MS fitted with an ORS as shown in Fig. 3.5 (a). However, at m/z 78 the background count was reduced by two orders of magnitude. The analytical and background signals decreased with increasing gas flow. Flow rate of H 2 and He were adjusted to give a maximum difference between background and analytical signals and, therefore these values were used as the optimised flow rates in the subsequent experiments. 62

80 3.5 References B Hymer, C. and Caruso, J. A. (2006) Selenium speciation analysis using inductively coupled plasma-mass spectrometry. Journal Chromatography A, 1114, Bandura, D. R., Baranov, V. I. and Tanner, S. D. (2001) Reaction chemistry and collisional processes in multipole devices for resolving isobaric interferences in ICP MS. Fresenius' journal of analytical chemistry, 370, Capelo, J. L., Fernandez, C., Pedras, B., Santos, P., Gonzalez, P. and Vaz, C. (2006) Trends in selenium determination/speciation by hyphenated techniques based on AAS or AFS. Talanta, 68, Chen, Z. L., Khan, N. I., Owens, G. and Naidu, R. (2007) Elimination of chloride interference on arsenic speciation in ion chromatography inductively coupled mass spectrometry using an octopole collision/reaction system. Microchemical Journal, 87, Dumont, E., Vanhaeche, F. and Cornelis, R. (2006) Selenium speciation from food source to metabolites: a critical review. Analytical Bioanalytical Chemistry, 385, Hettipathirana, T. D. and Davey, D. E. (1998) Evaluation of a microconcentric nebuliser cyclonic spray chamber for flow injection simultaneous multielement inductively coupled plasma optical emission spectrometry. Journal of Analytical Atomic Spectrometry, 13, Jensen, B. P., Gammelgaard, B., Hansen, S. H. and Andersen, J. V. (2003) Comparison of direct injection nebulizer and desolvating microconcentric nebulizer for analysis of chlorine-, bromine-and iodine-containing compounds by reversed phase HPLC with ICP-MS detection. Journal of Analytical Atomic Spectrometry, 18, Koppenaal, D. W., Eiden, G. C. and Barinaga, C. J. (2004) Collision and reaction cells in atomic mass spectrometry: development, status, and applications. Journal of Analytical Atomic Spectrometry, 19, Leonhard, P., Pepelnik, R., Prange, A., Yamada, N. and Yamada, T. (2002) Analysis of diluted sea-water at the ng L 1 level using an ICP-MS with an octopole reaction cell. Journal of Analytical Atomic Spectrometry, 17, McCurdy, E. and Woods, G. (2004) The application of collision/reaction cell inductively coupled plasma mass spectrometry to multi-element analysis in variable sample 63

81 matrices, using He as a non-reactive cell gas. Journal of Analytical Atomic Spectrometry, 19, Michalke, B. (2002b) The coupling of LC to ICP-MS in element speciation: I. General aspects. Trends in analytical chemistry, 21, Pedrero, Z. and Madrid, Y. (2009) Novel approaches for selenium speciation in foodstuffs and biological specimens: a review. Analytica Chimica Acta, 634, Pitts, L., Fisher, A., Worsfold, P. and Hill, S. J. (1995) Selenium speciation using highperformance liquid chromatography-hydride generation atomic fluorescence with on-line microwave reduction. Journal of analytical atomic spectrometry, 10, Polatajko, A., Jakubowski, N. and Szpunan, J. (2006) State of the art report of selenium speciation in biological samples. Journal Analytical Atomic Spectrometry, 21, Qiu, J., Wang, Q., Ma, Y., Yang, L. and Huang, B. (2006) On-line pre-reduction of Se (VI) by thiourea for selenium speciation by hydride generation. Spectrochimica Acta Part B: Atomic Spectroscopy, 61, Thomas, R. (2002) A beginner's guide to ICP-MS: Part XII-A review of interferences. Spectroscopy, 17, Uden, P. C. (2002) Modern trends in the speciation of selenium by hyphenated techniques. Analytical and Bioanalytical Chemistry, 373, Uden, P. C., Boakye, H. T., Kahakachchi, C. and Tyson, J. F. (2004) Selective detection and identification of Se containing compounds-review and recent developments. Journal Chromatography A, 1050, Walton, S. J. and Goulter, J. E. (1985) Performance of a commercial maximum dissolved solids nebuliser for inductively coupled plasma spectrometry. The Analyst, 110, Yamada, N., Takahashi, J. and Sakata, K. (2002) The effects of cell-gas impurities and kinetic energy discrimination in an octopole collision cell ICP-MS under nonthermalized conditions. Journal of Analytical Atomic Spectrometry, 17, Yip, Y. and Sham, W. (2007) Applications of collision/reaction-cell technology in isotope dilution mass spectrometry. Trends in Analytical Chemistry, 26,

82 Chapter 4 Inorganic selenium speciation using ion chromatography with ICP-MS and an octopole reaction system 65

83 4.1 Introduction The presence of 38 Ar 40 Ar + and 40 Ar 40 Ar + dimers generated in the ICP torch when using argon (Ar) gas interfere with the most abundant selenium isotopes 78 Se (23.8%, abundance) and 80 Se (49.6%, abundance). Since sensitivity proved to be an issue, it made sense to investigate a means of using these isotopes. Therefore, ICP-MS with an octopole reaction system (ORS) was explored for detecting selenite and selenate in water and soil. The toxicity of inorganic selenium depends on its chemical form. It is present in natural water and soil as selenite anion (SeO 2-3 ) and selenate (SeO 2-4 ). Selenates are easily leached from soils, transported to ground waters, and most readily sequestered by plants (Frankenberger and Engberg, 1998). Hence, it can cause greater risk to plant health. The ability to determine selenium in environmental samples is important for reducing uncertainty in selenium estimation in selected matrices (Frankenberger and Engberg, 1998). A number of analytical techniques can be used for determining selenium as mentioned in Chapter 2. These include hydride generation by flow injection, atomic spectroscopic methods, electrophoresis, and gas chromatography. However, liquid chromatography (LC) coupled with ICP-MS detection (Michalke, 2002b; Michalke, 2002a; Montes-Bayón et al., 2003) is frequently used for selenium speciation analysis at trace levels due to the combination of a separation technique with a highly specific and sensitive detector. As discussed in Chapter 2, liquid chromatography is most used separation technique for speciation studies. As pointed out in section 2.3.3, ion-pairing reverse-phase (IP-RPLC) and ion exchange chromatography (IEC) are commonly used for separating selenium species (Uden et al., 2004; B Hymer and Caruso, 2006; Michalke, 2002a). In an IP-RPLC approach, a C 18 or C 8 column is coated with an ion-pairing reagent added to the mobile phase and selenium species subsequently exchange on the coated column surface (Michalke, 2002b; Michalke, 2002a; Montes-Bayón et al., 2003). Ion-pairing chromatography is useful for simultaneous separation of ionic and non-ionic selenium species (Uden et al., 2004). However, the organic modifier which is a major constituent of the mobile phase is often not compatible with the ICP plasma. The modifier increases the reflected power, alters the ionisation characteristics of the Ar plasma (see subsections and ), and a reduction in sensitivity for element species is often experienced. 66

84 Under gradient elution conditions, the sensitivity changes with modifier concentration are particularly pronounced (Michalke, 2002b; Michalke, 2002a; Montes-Bayón et al., 2003). These limitations can be addressed by IEC since its mobile phase does not require ionpairing reagents or organic solvents (Michalke, 2002b; Michalke, 2002a; Montes-Bayón et al., 2003). Therefore, IEC has been commonly selected for the separation of ionic species when coupled with ICP-MS, and a number of studies have applied this approach to assessing ionic selenium species (Martinez-Bravo et al., 2001; Wallschläger and Roehl, 2001; Gammelgaard and Jrns, 2000). A number of publications discuss the use of IEC-ICP-MS for selenium speciation analysis. However, the most abundant selenium isotopes, which are 80 Se (49.6%) and 78 Se (23.8%), suffer from 40 Ar 40 Ar + and 38 Ar 40 Ar + interferences formed in the Ar plasma gas, and the detection limits suffer (see section 3.2). To address this issue, ICP with high-resolution MS has been used, although the procedure can just barely separate 80 Se from 40 Ar 40 Ar + (Jakubowski et al., 1998). Owing to its considerably high cost, the use of high-resolution MS is not commercially viable when compared to a conventional ICP with low resolution MS. For this reason, ICP-MS equipped with a reaction/collision cell has been developed as an alternative technique to ICP with high-resolution MS (Tanner et al., 2002; McCurdy and Woods, 2004). Various gases, including hydrogen (H 2 ), helium (He), ammonia (NH 3 ) and oxygen (O 2 ), can be used as reaction/collision to reduce spectral interference (Yamada et al., 2002; Takahashi and Yamada, 2004; see also section 3.2 of this thesis). Although a number of reports concerning ICP-MS detection of 78 Se and 80 Se using a reaction cell have been published, the differences between the detection of 78 Se and 80 Se using an octopole reaction system (ORS) have not been fully explored (Huerta et al., 2004; Huerta et al., 2005; Huerta et al., 2006; Darrouzès et al., 2005; Ogra et al., 2005). For example, detection of 77 Se and 78 Se using H 2 as the reaction gas was the focus of work by Huerta et al. (2004, 2005, 2006). It is also necessary to have information using gases other than H 2 (Darrouzès et al., 2005; Ogra et al., 2005). He and H 2 collision/reaction gases are available with the ICP-MS instrument used in the present study, providing the opportunity to collect Se data with both H 2 and He as reaction gases. It is also worthwhile exploring the possibility of using no reaction gas in ICP-MS while detecting 78 Se, at 23.8%, the less important Se isotope. 67

85 On the basis of the points mentioned, the effect of collision/reaction gas on the detection of 78 Se and 80 Se has been investigated. From the inferred information, this may allow reclassification of previous reports (Darrouzès et al., 2005; Yamada et al., 2002; Takahashi and Yamada, 2004; Huerta et al., 2004; Huerta et al., 2005; Huerta et al., 2006; Ogra et al., 2005). This study further aimed to develop a fast IC method for inorganic Se species using a new anion-exchange column. Use of the selected column to separate selenium species has not been reported previously. Mobile phase condition was varied to provide a reasonable chromatographic resolution. Finally, the proposed method was applied to inorganic selenium analysis in water and soil samples. 4.2 Experimental Chemicals and solutions All chemicals used in this study were of analytical grade reagent obtained from Sigma- Aldrich (Sydney, Australia). Milli-Q water (a specific resistance of 18.2 MΩ cm, Millipore, Bedford, MA, USA) was used for the preparation of all solutions and standards. Stock solutions: 1000 mg/l of Se(IV) were prepared by dissolving g of analytical reagent Na 2 SeO 3 (BDH Chemicals Ltd., Poole, United Kingdom) in a 1000 ml volumetric flask and made up to mark with MQ water. A 1000mg/L solution of Se(VI) was similarly prepared with g Na 2 SeO 4 10H 2 O (BDH Chemicals Ltd., Poole, United Kingdom) in 1000 ml MQ water. Se standards were stored in a refrigerator below 4 C (Pedrero et al., 2006). Other standard solutions were prepared daily (diluted from stock solutions). The eluent containing 20 mm ammonium nitrate and 10 mm ammonium dihydrogen phosphate required for IC-ICP-MS was prepared by dissolving g ammonium nitrate or g ammonium dihydrogen phosphate in 1000 ml of Milli-Q water, and ph was adjusted with 0.1 M ammonium hydroxide to the desired values. Solutions were filtered through disposable 0.45 µm cellulose acetate membrane filters and degassed in an ultrasonic bath prior to use. Waters and soil solutions were collected from contaminated sites and samples were filtered through a 0.45 µm membrane filter prior to analysis. 68

86 4.2.2 IC-ICP-MS conditions An Agilent 1100 liquid chromatography module (Tokyo, Japan) equipped with a guard column (G3154A/102), and a separation column (G3154A/101) were used. The samples were injected using an 1100 auto-sampler using an injection volume of 50 µl. The eluent flow-rate was 1.0 ml/min. The column eluent was passed via a length of PEEK tubing (50 cm) to a Babington nebuliser of an Agilent 7500c ICP-MS (Tokyo, Japan), which served as an element-specific mass detector. The ICP-MS operating parameters are shown in subsection 3.4. IC conditions are listed in Table 4.1. Table 4.1 IC conditions IC Agilent 1100 Column Anion-exchange (G3154A/101), 10µm, 150 mm x 4.6 mm Mobile phase 10 mm NH 4 H 2 PO 4 and 20 mm NH 4 NO 3 at selected ph values, isocratic elution Flow rate 1.0 ml/min Injection volume 50 µl Column temperature 25ºC 4.3 Results and discussion ICP-MS detection The primary selenium isotopes 78 Se and 80 Se suffer from polyatomic interferences such as 40 Ar 40 Ar + and 40 Ar 38 Ar + arising from the argon plasma source [ 38 Ar(0.336%) and 40 Ar(99.60%)]. This interference may be reduced by using a collision/reaction cell with the addition of He or H 2 gas (Tanner et al., 2002; McCurdy and Woods, 2004). In a preliminary study, the flow rate of He or H 2 into the cell was the most significant parameter affecting the background and analytical signals. Therefore, the gas was initially optimised to maximise the analyte signal and reduce background noise. Similarly, as described in Chapter 3 for tuning H 2 or He gas, the test solution, containing a 10 mm NH 4 H 2 PO 4 and 20 mm NH 4 NO 3 at ph 6.5, spiked with a 10 μg/l Se, was 69

87 aspirated into the nebuliser and then into the plasma. The reaction gas flow rate varied from 0 to 5 ml/min, and the background and Se signal were subsequently monitored. With He added to the collision cell, the Se and background signals both decreased when the He flow increased. The desired reduction of the interference by either collision induced dissociation (CID) or kinetic energy discrimination (KED) was not significant, but reached its maximum at a He flow rate of 4.5 ml/min. When hydrogen was introduced directly into the reaction cell, the background and Se signals decreased with increasing H 2 flow rate. However, a more pronounced decrease in the background signal relative to the Se signal was observed. A maximum difference between background and Se signals was obtained when H 2 flow rate was at 4.0 ml/min, which gave an acceptable detection limit of about 0.2 µg/l (detection limit will be further examined) Evaluation of no reaction gas, H 2 and He added to the cell As shown in Fig. 4.1, no reaction gas, He and H 2 were explored to detect 78 Se after separation by ion chromatography to determine whether interference had fallen. A mixture of 200 µg/l Se(IV) and Se(VI) was injected into the IC-ICP-MS system with a mobile phase containing 20 mm NH 4 NO 3 and 10 mm NH 4 H 2 PO 4 at ph 7.0. A He flow rate of 4.5 ml/min or a H 2 flow rate of 4.0 ml/min was introduced to the reaction/collision cell. 70

88 Abundance Se(IV) Se(VI) No reaction gas 8000 H He Time/Minutes--> Figure 4.1 Chromatogram obtained from no reaction gas, H 2 (4.0 ml/min), He (4.5 ml/min) as the collision/reaction gas and detection at m/z 78. The mobile phase: 20 mm NH 4 NO 3, 10 mm NH 4 H 2 PO 4 at ph 7.0. Concentration of both Se(IV) and Se(VI) is 200 µg/l. The sensitivity for each mode, as calculated by counts, was in the following order: no reaction gas > H 2 > He, while the background was in the next order: no reaction gas > He H 2. The observed counts and the relative sensitivities indicated by counts normalised to the no reaction gas mode for each analyte is given in Table 4.2. Table 4.2 Comparison of three tuning methods for the detection of Se m/z = 78 m/z = 80 Mode Se (IV) Se (VI) Se (IV) Se (VI) (counts x10 6 (relative counts)) No reaction gas 1.37 (100%) 1.27 (100%) 4.54 (100%) 2.42 (100%) H (49.6%) 0.77 (60.5%) 1.46 (32.2%) 1.71 (70.5%) He 0.13 (9.4%) 0.13 (10.1%) 0.41 (9.0%) 0.51 (21.2%) 71

89 Fig. 4.1 shows that since 38 Ar only has 0.336% abundance, the 40 Ar 38 Ar + did not significantly interfere with the detection of 78 Se using the no reaction gas mode. The baseline was at 1000 (Abundance) and the highest sensitivity was obtained. This indicated that the detection of 78 Se using a no reaction gas mode was possible. However, the background intensity was reduced by adding He (20, abundance) or H 2 (60, abundance) to the cell because the 40 Ar 38 Ar + ion was effectively removed or reduced by either CID and by KED, or by chemical reaction (Tanner et al., 2002; McCurdy and Woods, 2004). This is evident in the results given in Table 4.3. Table 4.3 Comparison of three tuning methods for the intensity of background (Abundance, percentage relative to no reaction gas abundance) Mode m/z = 78 m/z = 80 No reaction gas 980 (100%) (100%) H 2 60 (6.1%) 60 (0.06%) He 20 (2.1%) (1.14%) In addition, compared to He as a collision gas, a higher sensitivity was obtained with H 2 + added to the cell. This is due to the fact that H 2 reacted with Ar 2 and removed the 40 Ar 38 Ar + with a consequential improvement of the S/N ratio (Tanner et al., 2002; McCurdy and Woods, 2004). Furthermore, compared to the no reaction gas mode, the sensitivity decreased when using He or H 2 in the cell. This could be due to H 2 reacting with Se ions, or Se experiencing collisions with He in the cell and consequently the Se signal is lost (Tanner et al., 2002; McCurdy and Woods, 2004). In the present study, it was concluded that detecting 78 Se using a no reaction gas mode is preferred since 40 Ar 38 Ar + does not significantly interfere with the determination of Se. This observation differs from other reports (Huerta et al., 2004; Huerta et al., 2005; Huerta et al., 2006; Darrouzès et al., 2005; Ogra et al., 2005). As mentioned previously, a different picture for the most abundant isotope 80 Se (49.6%) emerges. The detection of 80 Se is of concern because 40 Ar (99.60%) is the principal isotope of the plasma gas. Consequently 40 Ar 40 Ar + formed in the plasma source can significantly interfere with the determination of Se (McCurdy and Woods, 2004; Yamada et al., 2002). Fig. 4.2 (a) illustrates the chromatogram obtained when using the no reaction gas mode and 72

90 detecting 80 Se. The background and noise were very high (9.2x10 5, abundance), which resulted from the formation of 40 Ar 40 Ar + in the plasma source. This dimer therefore interfered with detection of Se. However, as shown in Fig. 4.2 (b), the background noise was significantly reduced when He or H 2 was added to the reaction/collision cell, where background noise was (abundance) and 60 (abundance) using He and H 2, respectively. Compared to the no reaction gas mode, the background decreased from 100% to 1.14% using He in the cell, while the background fell from 100% to 0.06% when H 2 was added the cell (see Table 4.3). Figure 4.2 (a) Abundance Se(IV) Se(VI) No reaction gas Time/Minutes--> 73

91 Figure 4.2 (b) Abundance Se(IV) Se(VI) He H Time/Minutes--> Figure 4.2 Chromatogram obtained from (a) no reaction gas and detection at m/z 80, (b) H2 (4.0 ml/min) and He (4.5 ml/min) as the collision/reaction gas, other conditions as in Fig.4.1 This indicated that most of the 40 Ar 40 Ar + was removed by using He or H 2 as the gas, and these two modes can be effectively used to reduce 40 Ar 40 Ar + when detecting 80 Se. The improved detection was based on mechanisms such as CID, KED, and chemical reaction (see section 3.2; Tanner et al., 2002; McCurdy and Woods, 2004). However, comparing these two modes, a higher sensitivity and a lower background were achieved using H 2 as the gas. As can be seen in Table 4.2, a similar drop in sensitivity was experienced by using an ORS system for detecting 78 Se. It is also noted in Table 4.2 that the relative signals for a given Se species at 78 Se and 80 Se differ between H 2 and He. This could be the result of different collision/reaction mechanisms (Tanner et al., 2002; McCurdy and Woods, 2004). As a compromise between sensitivity and background, the use of H 2 as the reaction gas was considered most effective for reducing 40 Ar 40 Ar +. A H 2 flow rate of 4.0 ml/min as the reaction gas together with a detection of 80 Se was used in the following section. 74

92 4.3.3 Ion chromatographic separation of Se(IV) and Se(VI) Anion exchange chromatography is based on the exchange equilibrium between charged analyte ions and the oppositely charged site on the stationary phase. The retention of the selenium anions are determined by the nature of the competing ion, its concentration, and the mobile phase ph (B'Hymer and Caruso, 2002; Sarzanini and Bruzzoniti, 2001). Examples of anion-exchange columns commonly used and reported are given in Table 2.3. However, aqueous sodium hydroxide used as the mobile phase with the Dionex AS 11 column with a UV detector is not suited to ICP-MS (Gammelgaard and Jrns, 2000). Here the Agilent, G3154A/101 column (150 mm x 4.6 mm) was employed with ph rear neutrality for Se speciation. The stationary phase is based on porous polymethacrylate resin particles and ion exchange capacity of 50 µeq g -1. Ammonium eluents are more compatible with ICP-MS (see section ) and different ammonium eluents have been examined for arsenic speciation (Chen et al., 2006). In the case of Se speciation, using a mobile phase containing a single buffer, NH 4 H 2 PO 4, initially resulted in a long retention time for Se(VI). Further phosphate eluent studies will be described in Chapter 7. For fast separation of the two inorganic species, Se(IV) and Se(VI), another salt was needed in the mobile phase system to reduce analysis time. Examples described previously include ammonium citrate (Pedrero et al., 2006) and ammonium acetate (Chassaigne et al., 2002). For this reason, a mobile phase containing 20 mm NH 4 NO 3 to the eluent was chosen. Another factor affecting the retention of selenium species is eluent ph. The effect is shown in Fig

93 Fraction of Species (Alpha n) Retention time (min) 6 5 Se(IV) Se(VI) Eluent ph Figure 4.3 Effect of the eluent ph on the retention time of Se(IV) and Se(VI). Mobile phase: 20 mm NH 4 NO 3, 10 mm NH 4 H 2 PO 4, H 2 (4.0 ml/min) as the reaction gas. Changes in the eluent ph directly influence the charges of competing ions and the disassociation of selenous acid (pka 1 : 2.46; pka ) and selenic acid (pka 2 : 1.92). The dissociation of selenous acid is indicated in a diagram in Fig Alpha0 _H 2 SeO 3 Alpha1_HSeO - 3 Alpha2 _SeO ph Figure 4.4 Dissociation diagram for selenous acid on the ph range from 0 to 14 76

94 Since the pka 2 of selenic acid is much lower than that of selenous acid, the former exists as a dianion and would be expected to adhere strongly to the stationary phase compared to selenous acid which is present as HSeO - 3. In practice, selenic acid eluted after selenous acid over the experimental ph range. The retention time of selenium species decreased with increasing eluent ph from 5.0 to 7.0 (Fig. 4.3). For example, the retention times of Se(IV) and Se(VI) were 2.80 and 5.39 min, respectively when the eluent ph was 5.5, while at ph 7.0 the retention times were 2.17 and 3.56 min, respectively. It is evident from the above figure that changing mobile phase ph had little effect on Se(IV) species distribution. The two acids in aqueous solution remained present as HSeO - 3 and SeO 2-4. The key was the mobile phase solute ion which changed from monoionic (H 2 PO - 4 ) to dianionic (HPO 2-4 ), leading to an increase in eluting power (Sarzanini and Bruzzoniti, 2001; Weiss and Haddad, 1995). Consequently, the retention time was reduced. The electrolyte ph at 6.5 was chosen in subsequent studies. The separation of Se (IV) and Se (VI) was achieved within 5 min with a mobile phase containing 20 mm NH 4 NO 3 and 10 mm NH 4 H 2 PO 4 at ph 6.5. Calibration curves for quantification were obtained by plotting peak area versus the concentration of the corresponding target ion. As listed in Table 4.4, calibrations were linear over a concentration range of 1.0 to 200 µg/l with correlation coefficients greater than when a 50 µl was injected. Table 4.4 The characteristics for Se species by the proposed method Species Regression line Coefficient Detection limit Repeatability ( g/l) (%) 2- SeO y x SeO y x Detection limit- signal/noise = 3; Repeatability- a mixed Se standard solution (20 g/l, n=5) 77

95 Detection limits (S/N=3) ranged from µg/l. The repeatability when injecting 20 µg/l (n=5) of a standard solution containing a mixture of selenium species showed a RSD less than 2.2 %. In addition, little residue built up on both the sample and skimmer cones after long-term use (48 hours) when ammonia salt buffer was used, which is similar to previous reports (B'Hymer and Caruso, 2002). In order to test the method s applicability for determination of selenium, water samples were spiked with a mixture of 20 µg/l standards of each selenite and selenate. Recoveries for selenium species ranged from 95.3 to % (n=3) Sample applications The proposed method was tested using field samples to determine if it could be used to monitor selenium speciation in contaminated waters and soils. Fig. 4.5 shows two typical chromatograms from water (Fig. 4.5 (a)) and soil solution (Fig. 4.5 (b)). It can be seen that the background signals were lower (100, intensity) using H 2 tuning and detection at m/z 80. The contaminated water was found to contain Se(IV) 1.56± 0.1 µg/l (n=3) and Se(VI) 1.78 ± 0.1 µg/l (n=3), whereas only Se(IV) 1.95 ±0.15 µg/l (n=3) was detected in soil solution when the samples were diluted 20 fold. This indicates that the proposed method can be used to determine Se species in environmental samples, and the interferences were eliminated by using H 2 gas in this study. 78

96 Fig. 4.5 (a) Abundance 220 Se(IV) 2.45 Se(VI) Time/Minutes--> Figure 4.5 (b) Abundance 220 Se(IV) Time/Minutes--> Figure 4.5 (a) Water solution, (b) soil solution. Mobile phase: 20 mm NH 4 NO 3, 10 mm NH 4 H 2 PO 4 at ph 6.5, H 2 (4.0 ml/min) as the reaction gas, detection at m/z

97 4.4 Conclusions This chapter has described the use of anion-exchange chromatography for separation and detection of selenite and selenate in water and soil. In addition, the sensitivity was optimised using an octopole reaction system (ORS) to deliver the analytes to an ICP-MS. This work has shown that an ORS with H 2 as the reaction gas at a flow rate of 4.0 ml/min can be used to effectively remove the 40 Ar 40 Ar + interference when detecting 80 Se. While reduction of background is at the expense of sensitivity, no reaction gas mode can be used directly to detect 78 Se with an acceptable background to provide the best sensitivity. A new anion exchange column was successfully used for selenium speciation with a mobile phase which contained 20 mm NH 4 NO 3 and 10 mm NH 4 HPO 4 at ph 6.5. Inorganic Se speciation can be performed within 5 min by adding NH 4 NO 3 to the buffer, reducing the throughput and analysis cost per sample. The proposed method has reasonable resolution and low detection limits (µg/l) without extensive sample pretreatment. Detection limits of less than 0.4 g/l for 80 Se (IV) and 0.6 g/l for 80 Se (VI) were achieved by using H 2 as the reaction gas. 80

98 4.5 References B Hymer, C. and Caruso, J. A. (2002) Evaluation of HPLC systems for the separation and quantification of arsenic compounds from apple extracts. Journal of Liquid Chromatography & Related Technologies, 25, B Hymer, C. and Caruso, J. A. (2006) Selenium speciation analysis using inductively coupled plasma-mass spectrometry. Journal of Chromatography A, 1114, Chassaigne, H., Chery, C. C., Bordin, G. and Rodriguez, A. R. (2002) Development of new analytical methods for selenium speciation in selenium-enriched yeast material. Journal of Chromatography A, 976, Chen, Z. L., Akter, K. F., Rahman, M. M. and Naidu, R. (2006) Speciation of arsenic by ion chromatography inductively coupled plasma mass spectrometry using ammonium effluents. Journal of High Resolution Chromatography, 29, Darrouzès, J., Bueno, M., Lespès, G. and Potin-Gautier, M. (2005) Operational optimisation of ICP octopole collision/reaction cell MS for applications to ultratrace selenium total and speciation determination. Journal of Analytical Atomic Spectrometry, 20, Frankenberger, W. T. and Engberg, R. A. (1998) Environmental chemistry of selenium, Marcel Dekker Inc., New York, USA. Gammelgaard, B. and Jrns, O. (2000) Determination of selenite and selenate in human urine by ion chromatography and inductively coupled plasma mass spectrometry. Journal of Analytical Atomic Spectrometry, 15, Huerta, V. D., Sa nchez, M. L. F. and Sanz-Medel, A. (2004) Quantitative selenium speciation in cod muscle by isotope dilution ICP-MS with a reaction cell: comparison of different reported extraction procedures Journal of Analytical Atomic Spectrometry, 19, Huerta, V. D., Sánchez, M. L. F. and Sanz-Medel, A. (2005) Qualitative and quantitative speciation analysis of water soluble selenium in three edible wild mushrooms species by liquid chromatography using post-column isotope dilution ICP MS. Analytica Chimica Acta, 538, Huerta, V. D., Sánchez, M. L. F. and Sanz-Medel, A. (2006) An attempt to differentiate HPLC-ICP-MS selenium speciation in natural and selenised Agaricus mushrooms using different species extraction procedures. Analytical and Bioanalytical Chemistry, 384,

99 Jakubowski, N., Moens, L. and Vanhaecke, F. (1998) Sector field mass spectrometers in ICP-MS. Spectrochimica Acta Part B: Atomic Spectroscopy, 53, Martinez-Bravo, Y., Roig-Navarro, A. F., Lopez, F. J. and Hernandez, F. (2001) Multielemental determination of arsenic, selenium and chromium (VI) species in water by high-performance liquid chromatography-inductively coupled plasma mass spectrometry. Journal of chromatography, 926, McCurdy, E. and Woods, G. (2004) The application of collision/reaction cell inductively coupled plasma mass spectrometry to multi-element analysis in variable sample matrices, using He as a non-reactive cell gas. Journal of Analytical Atomic Spectrometry, 19, Michalke, B. (2002a) The coupling of LC to ICP-MS in element speciation Part II: Recent trends in application. Trends in analytical chemistry, 21, Michalke, B. (2002b) The coupling of LC to ICP-MS in element speciation: I. General aspects. Trends in analytical chemistry, 21, Montes-Bayón, M., DeNicola, K. and Caruso, J. A. (2003) Liquid chromatography inductively coupled plasma mass spectrometry. Journal of Chromatography A, 1000, Ogra, Y., Ishiwata, K. and Suzuki, K. T. (2005) Effects of deuterium in octopole reaction and collision cell ICP-MS on detection of selenium in extracellular fluids. Analytica Chimica Acta, 554, Pedrero, Z., Madrid, Y. and Cámara, C. (2006) Selenium species bioaccessibility in enriched radish (Raphanus sativus): a potential dietary source of selenium. Journal of Agriculture and Food Chemistry, 54, Sarzanini, C. and Bruzzoniti, M. C. (2001) Metal species determination by ion chromatography. Trends in analytical chemistry, 20, Takahashi, J. and Yamada, N. (2004) Development of collision/reaction cell for reduction of spectral interference in ICP mass spectrometry. Bunseki Kagaku, 53, Tanner, S. D., Baranov, V. I. and Bandura, D. R. (2002) Reaction cells and collision cells for ICP-MS: a tutorial review. Spectrochimica Acta Part B: Atomic Spectroscopy, 57, Uden, P. C., Boakye, H. T., Kahakachchi, C. and Tyson, J. F. (2004) Selective detection and identification of Se containing compounds-review and recent developments. Journal of Chromatography A, 1050,

100 Wallschläger, D. and Roehl, R. (2001) Determination of inorganic selenium speciation in waters by ion chromatography-inductively coupled plasma-mass spectrometry using eluant elimination with a membrane suppressor. Journal of Analytical Atomic Spectrometry, 16, Weiss, J. and Haddad, P. R. (1995) Ion chromatography, VCH Weinheim. Yamada, N., Takahashi, J. and Sakata, K. (2002) The effects of cell-gas impurities and kinetic energy discrimination in an octopole collision cell ICP-MS under nonthermalized conditions. Journal of Analytical Atomic Spectrometry, 17,

101 Chapter 5 Ion-exclusion chromatographic separation of inorganic selenium species in water and detection using ICP-MS 84

102 5.1 Introduction Selenium (Se) is a potentially toxic trace element. However, the fact that it is also essential has been recognised. In the natural environment, selenium exists in oxidation states +6 and +4 as dissolved selenates (SeO 2-4 ) and selenites (SeO 2-3 ). The total Se levels in natural waters range between 0.1 to 400 µg L -1 (Munoz Olivas et al., 1994). To understand the selenium biogeochemical cycling, it is important to determine the concentration of each species. Since the composition of samples can be complex and concentrations are at trace levels, a separation technique combined with a specific and sensitive detection system is required in order to reduce interferences and improve detection limits. In recent years, the coupling of ion chromatography to accomplish analyte separation with highly sensitive detectors such as inductively coupled plasma-mass spectrometry (ICP-MS) has proven to be an effective hyphenated system for elemental speciation analysis (B Hymer and Caruso, 2006). However, the critical issues for LC-ICP-MS are: firstly, the choice of suitable separation modes; and secondly, overcoming polyatomic ion interference in the detection of Se. Using ion chromatography, various separation modes can be used for the speciation of SeO 2-4 and SeO 2-3. These include anion-exchange and ion-pairing chromatography. In the ion-exchange chromatographic approach, the selenium species are initially retained on the column by anion exchange and subsequently eluted by a competitive anion included in the mobile phase (Uden et al., 2004; Michalke, 2002; Montes-Bayón et al., 2003). Recently, inorganic selenium speciation was achieved by using anion exchange chromatography with various mobile phases, including the mobile phase containing NH 4 H 2 PO 4 and NH 4 NO 3 at ph 6.5 (Chen et al., 2008), gradient elution using (NH 4 ) 2 CO 3 and 2% methanol at ph 9.0 (Wang et al., 2007), ammonium citrate in 2% methanol (Huerta et al., 2003) and pyridinium formate in 3% methanol (Larsen et al., 2003). Since there was a high salt concentration in the mobile phase for these approaches, the ICP-MS experienced signal suppression due to increased space-charge effects which defocused the ion beam (Larsen, 1998). Another approach used for the separation of inorganic selenium species was ionpairing chromatography where a C 18 or C 8 column was coated with an ion-pairing reagent present in the mobile phase and the ionic selenium species was subsequently exchanged on the coated column (Michalke, 2002; Montes-Bayón et al., 2003). However, an organic modifier such as methanol and the ion-paring reagent used in the mobile phase 85

103 (Vonderheide et al., 2002; Zheng et al., 2000) are not compatible with the ICP plasma due to an increase in the reflected power. This results in plasma extinction even at relatively low organic modifier concentrations. Another undesired effect is the altered ionisation characteristics of the Ar plasma and a reduction in detection sensitivity for elemental species (Michalke, 2002; Montes-Bayón et al., 2003). Hence, there is a need to explore a new separation mode which is compatible with ICP-MS detection of Se species. Ion-exclusion chromatography is a widely accepted technique particularly suited to the separation of weak acids. The separation mechanism relies on inorganic anions being excluded by the resin phase through ionic repulsion (Weiss and Haddad, 1995; Chen and Adams, 1999; Ng et al., 2001; Fischer, 2002). Ion-exclusion chromatography with ICP-MS has been successfully used to separate and detect Arsenic (As) species in seawater because it effectively excluded the matrix with its high Cl - level. Consequently, the Cl - ion was separated from As species (Nakazato et al., 2002) and the determination of arsenic species in seawater was possible. However, ion-exclusion chromatography with ICP-MS has not been used for the speciation of inorganic selenium. The detection of 78 Se and 80 Se using ion-exchange chromatography (IEC) by ICP-MS with ORS has been investigated in Chapter 4. This study is based on ion-exclusion chromatography, which to the author s knowledge has not been applied to inorganic Se species. One aim was to develop a method for the speciation of inorganic selenium species in water and soil using this technique. Therefore, this study mainly focuses on combining ion-exclusion chromatography with ICP-MS to determine whether detection of inorganic selenium can be accomplished both procedural simplicity and fast separation. Finally, the effectiveness of proposed method was demonstrated for environmental water samples. 5.2 Experimental Chemicals and Solutions Milli-Q water (a specific resistance of 18.2 MΩ cm, Millipore, Bedford, MA, USA) was used for preparing all solutions and standards. Se stock solutions (see subsection 4.2.1) and other standard solutions were prepared daily (diluted from stock solutions). All solutions were stored at below 4 C. 86

104 All acids were analytical grade and purchased from Biolab (Aust.) Ltd. Mobile phases required for IEC were prepared as follows (5 mm): dissolution of 140 µl concentrated sulphuric acid (d=1.84, 98%); 160 µl concentrated nitric acid (D=1.42, 69%); 143 µl acetic acid (d=1.049, 99.8%); and 94 µl formic acid (d=1.22, 90%) in 500 ml volumetric flasks to line with Milli-Q water, respectively. Eluents were filtered through a disposable 0.45 µm cellulose acetate membrane filter (Millipore) and degassed in an ultrasonic bath prior to use. Water samples collected from Mawson Lakes, South Australia and from arsenic contaminated ground water from Bangladesh were filtered through a disposable 0.45 µm cellulose acetate membrane filter (Millipore). Aqueous extracts from soil samples were provided by CERAR, UniSA, South Australia. The soil was extracted with MQ water and the mixture centrifuged and filtered through a 0.45 µm membrane filter prior to analysis Instrument and Analytical Conditions An Agilent 1100 liquid chromatography module was equipped with an ion-exclusion column containing polystyrene-divinylbenzene resin (HPX-87H, 30 cm x 7.8 mm i.d., 5 µm) purchased from Bio-Rad (Richmond, USA). Samples (50 µl) were injected using an Agilent 1100 auto-sampler. The mobile phase flow rate and temperature were set at 1.0 ml/min and 30ºC respectively. The column effluent was passed via a length of PEEK tubing (50 cm, I.D.= 0.13 mm) to a Babington nebuliser attached to an Agilent 7500c ICP- MS (Tokyo, Japan), which served as the element-specific mass detector. The ICP-MS conditions were set as listed in subsection 3.4. The IEC-ICP-MS system was controlled and the data was processed using Agilent s Chemstation software package. 5.3 Results and discussion Ion-exclusion chromatographic separation of SeO 3 2- and SeO 4 2- In ion-exclusion chromatography, solute retention depends mainly on the degree of ionisation of the solute and, hence on solute pka and the eluent ph (Weiss and Haddad, 87

105 1995; Chen and Adams, 1999; Ng et al., 2001; Fischer, 2002). The theoretical dissociation diagram is shown in Fig On the basis of the ion-exclusion characteristics, it was thus expected that the retention of selenic acid (pka2: 1.92) and selenous acid (pka 1 : 2.62; pka 2 : 8.32) would be affected by the eluent ph. Generally, strong acids such as sulphuric acid are frequently used as the eluent in ionexclusion chromatography owing to their effective control of the degree of ionisation of the solute, leading to improved resolution (Fischer, 2002; Chen and Adams, 1999; Ng et al., 2001). To examine the resolution between SeO and SeO 3 in this study, chromatograms were obtained firstly by using organic acids such as acetic, and formic acid, followed by nitric and sulphuric acid as the mobile phase, all at 5 mm concentration in water. Fig. 5.1 (a) and (b) represent the chromatograms using acetic and sulphuric acid as the eluents, respectively. Elution order was opposite those observed in anion exchange chromatography in Chapter 4. Fig. 5.1 (a) Abundance Se(VI) Se(IV) Time/Minutes--> 88

106 Fig. 5.1(b) Abundance Se(VI) Se(IV) Time/Minutes--> Figure 5.1 Ion exclusion chromatogram with a polystyrene-divinylbenzene resin (HPX-87 H 30 cm X 7.8 mm i.d., 5µm. eluent at 1.0 ml/min and 30ºC. Sample, 50 µl of Se(IV) and Se(VI) standards (500 µg/l each). H 2 flow rate of 2.0 ml/min, detection at m/z 80. (a) Using 5 mm acetic acid as mobile phase. (b) Using 5 mm sulphuric acid as the mobile phase. It can be seen that the retention times for SeO 2-4 and SeO 2-3 were significantly different using acetic or sulphuric acid as the mobile phase. The retention time for SeO and SeO 3 were reasonably similar at 3.37 min and 3.87 min when acetic acid (ph 3.40) was used as the mobile phase, whereas the retention times were 3.63 min for SeO 2-4 and 6.49 min for 2- SeO 3 using sulphuric acid (ph 2.18) as the eluent. The difference was 2.86 min. This is because sulphuric acid controls more effectively the degree of ionisation of selenic and selenous acid than acetic acid by lowering the ph. As a result, it causes the species to be retained more strongly on the ion-exclusion column (Fischer, 2002; Chen and Adams, 1999; Ng et al., 2001). As expected, similar chromatograms displaying resolved peaks were obtained using acetic and formic acid. With 5 mm organic acids, the ph of the solution was in the range which meant selenite (HSeO - 3 ) (see Fig. 4.4) was singly and selenate (SeO 2-4 ) doubly ionised, meaning their behaviour towards the stationary phase was unaffected by change in organic acid. The strong acids gave a higher resolution since 2- at ph 2.2 sulphuric acid causes SeO 4 to form while selenous acid remains fully 89

107 protonated (H 2 SeO 3 ). Unionised H 2 SeO 3 thus adheres to the column, increasing its elution 2- time compared to SeO 4 which is repelled. Changing sulphuric acid for nitric acid essentially led to the same result, albeit with peak tailing. It is conceivable that 5 mm nitric acid at a ph of 2.3 resulted in a higher proportion of ionised selenous acid which caused peak broadening (see Fig. 4.4). The ICP-MS signal obtained for SeO 2-4 and SeO 2-3 comparing acetic or sulphuric acid as the mobile phase was also noteworthy. Fig. 5.1 (b) shows a higher signal for SeO 2-4 and SeO 2-3 using sulphuric acid as the mobile phase. For example, the intensity of SeO 2-4 and 2- SeO 3 was 2.8 x10 4 and 1.4 x10 4 abundance units using acetic acid (Fig. 5.1 (a)), while the intensity was 8.0x10 4 for SeO 2-4 and 1.9 x10 4 for SeO 2-3 using sulphuric acid eluent (Fig. 5.1 (b)), which suggested species Se(VI) as SeO 2-4 ionisation in the plasma was more efficient than that of Se(IV) as H 2 SeO 3 compared with the other acid eluents. In addition to the lower intensity, using formic or acetic acid as eluent results in a carbon residue being left on the sampler and skimmer cones of the ICP-MS. This causes considerable drift during prolonged use and loss of the sensitivity (Montes-Bayón et al., 2003; McCurdy and Woods, 2004). It was also noted that the peaks for the two analytes at the same concentrations were not of equal area. A similar observation was previously ascribed to the different efficiency of nebulisation and ionisation of SeO 2-4 and SeO 2-3 in ICP-MS (Stewart and Olesik, 1998). A sulphuric acid mobile phase was thus selected for further experiments due to its good resolution, high detection sensitivity, and extended operating time with the ICP-MS Sample applications As shown in Fig. 5.1 (b), the separation of SeO and SeO 3 using ion-exclusion chromatography was accomplished within 7 min using 5 mm sulphuric acid. Calibration curves were obtained by plotting peak area versus the concentration of the corresponding selenium species. Linear models were satisfactory over a concentration range of 1 to 1000 µg/l with correlation coefficients greater than observed for a 50 µl sample injection (Fig. 5.2). Detection limits (S/N=3) were 0.2 µg/l for Se(VI) and 0.8 µg/l for Se(IV). 90

108 Abundance y = 8055.x R² = Se(VI) Se(IV) y = 4444.x R² = Se Conc. (µg/l) Figure 5.2 Calibration plots for Se(VI) and Se(IV) using ion-exclusion chromatography with 5 mm sulphuric acid. The proposed method was used for the speciation of selenium species in water samples collected from various sources. For contaminated ground water from Bangladesh, 148.6±1.1 µg/l Se(VI) and 19.2±0.8 µg/l Se(IV) were found (n=4) as shown in Fig To validate the method, both 20 and 200 µg/l mixtures of both standards were spiked into water samples, with recoveries for SeO 2-4 and SeO 2-3 found to range from 95% to 115% (n=4). 91

109 Abundance 8000 Se(VI) Se(IV) Time/Minutes--> Figure 5.3 Ion exclusion chromatogram obtained for contaminated water from Bangladesh using aqueous sulphuric acid as the mobile phase. H 2 flow rate of 3.0 ml/min, detection at m/z 80. Other conditions were the same as in Fig. 5.1 (b). Water samples from a local natural source, Mawson Lakes in South Australia were collected and tested. It was found that Se(VI) was present at an average concentration of 23.7 ±0.3µg/L (n=4), while Se (IV) was not observed. The repeatability from the injection of Mawson Lakes water samples (n=4) showed a RSD of less than 1.5 %. The Se(VI) value of 23.7 ± 0.3 µg/l in local lake water is of a similar magnitude to the Se content observed in another published water sample (0.123 µg/l and µg/l) (Yamada et al., 2002). Therefore, the dominant form of Se in Mawson Lakes water is SeO 2-4, whereas SeO 2-3 is only found in the more acidified waters. These facts are consistent with the behaviour of selenium species in the environment at different phs (Bujdos et al., 2005). In addition, the method was similarly applied to soil samples. Soil analysis by another group in our research centre was achieved using the developed method. Standards of Se(IV) and Se(VI) at different concentrations were added to the soil and incubated. Then the Se species were extracted with water and analysed using ion-exclusion chromatographies with ICP-MS. Forty samples were detected and the results were in good accordance with the amount spiked into soil. 92

110 5.4 Conclusions The results presented show that ion-exclusion chromatography using an acid eluent, coupled with ICP-MS, can be used for the separation and detection of inorganic selenium. Sulphuric acid was selected over 3 other acids as it gave good chromatographic resolution because peaks were well separated, no tailing and long retention. Best detection sensitivity was achieved and avoided the possibility of MS skimmer deposits. As demonstrated, ionexclusion chromatography coupled to ICP-MS with an ORS provides a simple and fast method for the speciation of inorganic selenium in waters and soil. Ion-exclusion chromatography provided improved resolution over the ion-exchange procedures shown in Chapter 4, while retaining good analysis time. Sensitivity was also improved by using sulphuric acid as eluent. A linear range was obtained from µg/l for both isotopic masses. The repeatability from the injection of water samples showed that RSD was less than 1.5% (n=4) and recoveries ranged from 95% to 115%. Finally, selenite and selenate were determined at trace levels in water and soil samples with the proposed method. 93

111 5.5 References B Hymer, C. and Caruso, J. A. (2006) Selenium speciation analysis using inductively coupled plasma-mass spectrometry. Journal of Chromatography A, 1114, Bujdos, M., Mulova, A., Kubova, J. and Medved, J. (2005) Selenium fractionation and speciation in rocks, soils, waters and plants in polluted surface mine environment. Environmental Geology, 47, Chen, Z. L. and Adams, M. A. (1999) Simultaneous determination of aliphatic and aromatic acids in plant tissue extracts by ion-exclusion chromatography. Analytica Chimica Acta, 386, Chen, Z. L., Wang, W., Mallavarapu, M. and Naidu, R. (2008) Comparison of no gas and He/H 2 cell modes used for reduction of isobaric interferences in selenium speciation by ion chromatography with inductively coupled plasma mass spectrometry. Spectrochimica Acta Part B: Atomic Spectroscopy, 63, Fischer, K. (2002) Environmental analysis of aliphatic carboxylic acids by ion-exclusion chromatography. Analytica Chimica Acta, 465, Huerta, V. D., Reyes, L. H., Marchante-Gayón, J. M., Sánchez, M. L. F. and Sanz-Medel, A. (2003) Total determination and quantitative speciation analysis of selenium in yeast and wheat flour by isotope dilution analysis ICP-MS. Journal of Analytical Atomic Spectrometry, 18, Larsen, E. H. (1998) Method optimization and quality assurance in speciation analysis using high performance liquid chromatography with detection by inductively coupled plasma mass spectrometry. Spectrochimica Acta Part B: Atomic Spectroscopy, 53, Larsen, E. H., Sloth, J., Hansen, M. and Moesgaard, S. (2003) Selenium speciation and isotope composition in 77 Se-enriched yeast using gradient elution HPLC separation and ICP-dynamic reaction cell-ms. Journal of Analytical Atomic Spectrometry, 18, McCurdy, E. and Woods, G. (2004) The application of collision/reaction cell inductively coupled plasma mass spectrometry to multi-element analysis in variable sample matrices, using He as a non-reactive cell gas. Journal of Analytical Atomic Spectrometry, 19, Michalke, B. (2002) The coupling of LC to ICP-MS in element speciation: I. General aspects. Trends in analytical chemistry, 21,

112 Montes-Bayón, M., DeNicola, K. and Caruso, J. A. (2003) Liquid chromatography inductively coupled plasma mass spectrometry. Journal of Chromatography A, 1000, Munoz Olivas, R., Donard, O. F. X., Camara, C. and Quevauviller, P. (1994) Analytical techniques applied to the speciation of selenium in environmental matrices. Analytica Chimica Acta, 286, Nakazato, T., Tao, H., Taniguchi, T. and Isshiki, K. (2002) Determination of arsenite, arsenate, and monomethylarsonic acid in seawater by ion-exclusion chromatography combined with inductively coupled plasma mass spectrometry using reaction cell and hydride generation techniques. Talanta, 58, Ng, K. L., Glód, B. K., Dicinoski, G. W. and Haddad, P. R. (2001) Retention modelling of electrostatic and adsorption effects of aliphatic and aromatic carboxylic acids in ion-exclusion chromatography: II. Calculations of adsorption coefficients in unbuffered eluents. Journal of Chromatography A, 920, Stewart, IIand Olesik, J. W. (1998) Steady state acid effects in ICP-MS. Journal of Analytical Atomic Spectrometry, 13, Uden, P. C., Boakye, H. T., Kahakachchi, C. and Tyson, J. F. (2004) Selective detection and identification of Se containing compounds-review and recent developments. Journal of Chromatography A, 1050, Vonderheide, A. P., Wrobel, K., Kannamkumarath, S. S., B'Hymer, C., Montes-Bayon, M., de Leon, C. P. and Caruso, J. A. (2002) Characterization of Selenium Species in Brazil Nuts by HPLC-ICP-MS and ES-MS. Journal of Agricultural and Food Chemistry, 50, Wang, R. Y., Hsu, Y. L., Chang, L. F. and Jiang, S. J. (2007) Speciation analysis of arsenic and selenium compounds in environmental and biological samples by ion chromatography inductively coupled plasma dynamic reaction cell mass spectrometer. Analytica Chimica Acta, 590, Weiss, J. and Haddad, P. R. (1995) Ion chromatography, VCH Weinheim. Yamada, N., Takahashi, J. and Sakata, K. (2002) The effects of cell-gas impurities and kinetic energy discrimination in an octopole collision cell ICP-MS under nonthermalized conditions. Journal of Analytical Atomic Spectrometry, 17, Zheng, J., Ohata, M., Furuta, N. and Kosmus, W. (2000) Speciation of selenium compounds with ion-pair reversed-phase liquid chromatography using inductively 95

113 coupled plasma mass spectrometry as element-specific detection. Journal of Chromatography A, 874,

114 Chapter 6 Extraction of selenium species in pharmaceutical tablets using enzymatic and chemical methods 97

115 6.1 Introduction As mentioned in the review (Chapter 2), selenium is an essential micronutrient trace element that both animals and humans require. It is toxic at high concentrations, but possesses beneficial properties at low levels. For example, recent evidence shows that diets supplemented with Se can reduce the incidence of prostate and breast cancer (Ip et al., 2000; Cornelis et al., 2005). Insufficient Se intake can lead to an increased risk of thyroid and immune dysfunction, infections, cancer and diseases, such as Keshan and Kashin-beck disease (Tan et al., 2002; Tinggi, 2003). With the increased evidence of organic-based Se health benefits, research interest is shifting from a focus on environmental analysis to biological systems. This is necessary to better understand the role of particular Se compounds in health. In recent years, increased research attention has focused on understanding the relationships between the Se species content in foods and supplement tablets and their nutritional benefits, such as absorption, tissue distribution, bioavailability and cancer preventative properties (Wrobel et al., 2003). Therefore, it is highly desirable to determine the identity and concentration of Se species present in foods and supplements (B Hymer and Caruso, 2006). Some analytical methods have been developed (Weixiao et al., 2006; Ajtony et al., 2005; Bidari et al., 2008), including sample preparation from yeasts and pharmaceutical tablets (Yang et al., 2004a; McSheehy et al., 2005; Mester et al., 2006; Goenaga-Infante et al., 2008). The preparation of samples is important for successfully assessment of the concentration and range of selenium species. Liquid extraction is the most conventional procedure (Yang et al., 2004a) where different solutions can be used to extract selenium species from various matrices. Such solutions include aqueous HCl (Yang et al., 2004b), tetramethylammonium hydroxide (TMAH) (Casiot et al., 1999), sodium dodecyl sulphate (SDS) leaching (Huerta et al., 2004), hot water (Marchante-Gayón et al., 2000) and methanesulphonic acid (Wrobel et al., 2003). Extraction efficiencies with average yields of 66 67% selenium in the form of selenomethionine were obtained from yeast after digestion using a 4 M methanesulphonic acid solution under reflux conditions (Wrobel et al., 2003; Yang et al., 2004b). Alternatively, sequential extraction was employed to separate water-soluble and insoluble selenium fractions from enriched yeast. Extracts were 98

116 fractionated by preparative columns such as Superdex 75 and 200 (Chassaigne et al., 2002). Enzymatic hydrolyses have often been used in selenium speciation analysis. The right choice of enzymes, hydrolysed solution ph and temperature were key parameters for successful enzymatic extraction (B Hymer and Caruso, 2001). Proteinase K, a general protein hydrolysing enzyme, was used to recover selenium from yeast food supplements (B'Hymer and Caruso, 2000). Mixtures of various enzymes were also found to be highly efficient. Proteinase K, aminopeptidase, and carboxypeptidase Y were successfully used to extract selenium compounds from yeast (B Hymer and Caruso, 2001). Simultaneous use of proteinase K and aminopeptidase M gave the highest extraction of selenomethionine from yeast, 68% of the total selenium level (B Hymer and Caruso, 2001). In the method of using enzyme I β-glucosidase with mixtures of endo- and exopeptidases or enzyme II protease XIV, both enzymatic procedures liberated in excess of 90% of the total selenium (Larsen et al., 2003). The suggestion was made that selenomethionine in yeast was bound physically as part of the cell wall rather than being incorporated chemically into the yeast s protein (Połatajko et al., 2004). The enzyme driselase was therefore chosen to increase Se extraction. Human serum (Encinar et al., 2004), cod muscle (Huerta et al., 2004) and wheat flour (Huerta et al., 2003) were also treated enzymatically for selenium. Combined treatments using enzymes and traditional techniques, including enzymatic digestion and ultrasonic baths or ultrasonic probes, have also been reported (Bermejo et al., 2004; Capelo et al., 2004). As an example, mushrooms have been treated by sequential and/or systematic enzymatic processes; the highest efficiency (89%) was obtained using a 3-step procedure with lysing enzyme and pronase (Dernovics et al., 2002). This chapter deals with selenium species from biological systems and presents the development of efficient sample extraction methods for selenium species by enzymatic and chemical methods. The samples selected include both inorganic and organic species (selenite, selenate and seleno-amino acids) at trace levels. Extraction of Se species from pharmaceutical tablets using either enzymatic hydrolysis or chemical solution is investigated since an effective extraction method depends on a few factors including sample matrix (B Hymer and Caruso, 2006). The Se species were determined by ion chromatography with inductively coupled plasma mass spectrometry (IC-ICP-MS) using ammonium phosphate solution as eluent. 99

117 6.2 Experimental Chemicals and solutions All reagents, including Se-(methyl) selenocysteine (SeMC), seleno-l-methionine (SeMet) and seleno-l-cystine (SeCys 2 ) were of analytical reagent grade and obtained from Sigma- Aldrich (Sydney, NSW, Australia). Stock solutions at 500 mg L -1 (for Se) of each of the above solution were prepared separately by dissolving g SeMC, g SeMet and g SeCys 2 in a 100 ml volumetric flask and filling with MQ water to the line. Se(IV) and Se(VI) stock solutions (see subsection 4.2.1). All Se standards were stored in a refrigerator below 4 C. Working standard solutions were prepared daily from their stock solutions. Milli-Q water (a specific resistance of 18.2 MΩ cm, Millipore, Bedford, MA, USA) was used to prepare all solutions and standards. The enzymes, protease XIV and lipase VII were also purchased from Sigma-Aldrich (Sydney, Australia). Tris-(tris(hydroxymethyl)aminomethane)-HCl buffer (75 mm, adjusted ph=7.5) and 20 mm NH 4 H 2 PO 4 (ph=6.5) were prepared daily as required. For this purpose, g tris(hydroxymethyl)aminomethane was dissolved in a beaker with about 90 ml and using 1M HCl to adjust the ph to 7.5. The solution was quantitatively transferred to a 100 ml volumetric flask and diluted to the line with MQ water. A 2.30 g ammonium dihydrogen phosphate was dissolved with water (about 950 ml) in a volumetric flask, adjusted to ph 6.5 and diluted to 1000 ml. Other concentrations of Tris buffer were prepared by dissolving appropriate amounts of ammonium salts in MQ water. Four different sources of Se samples were investigated. Included were commercial tablets, Se-no-yeast TM, Se-high-yeast TM and SeMC TM from Wholesale Nutrition (Saratoga, CA, USA, Total Se determination Commercial pharmaceutical Se-enriched tablets were ground in a mortar. A 0.5 g portion of sample was accurately weighed into PTFE vials and dissolved in a mixture of 2.0 ml of conc. HNO 3 and 2.0 ml of 30% H 2 O 2. After 17 h of contact time, total digestion was performed at 170ºC for 15 min at 300 W in a MARSX microwave oven (CEM, NC, USA). The final solutions were made up to 25 ml in volumetric flasks filled with MilliQ water. 100

118 6.2.3 Extraction In order to extract the protein-bound Se species in all samples, enzymatic hydrolysis and SDS extraction were performed as summarized below. Tris-HCL buffer 15 ml adjusted to ph 7.5 was added to 0.2 g of the sample in a 50 ml centrifuge tube, followed by the addition of 20 mg protease XIV and 10 mg lipase VII. The mixture was sonicated for 2 h, and then incubated in the dark for 12 h at 25ºC. Hydrolysed samples were centrifuged at 4000 rpm for 25 min and the supernatants filtered through a 0.45 µm membrane filter prior to analysis. Alternatively, a 15 ml solution of 1 M HCl was added to 0.2 g of the sample in a 50 ml centrifuge tube, followed by the addition of 0.6 g SDS. Further processing was the same as for Tris-HCL buffer. All solutions were immediately analysed after extraction or stored at -20ºC (Ogra et al., 2004) Instrumentation See subsection 3.4 for ICP-MS parameters. IC conditions are described in subsection Aqueous ammonium dihydrogen phosphate was the only mobile phase used. 6.3 Results and Discussion Determination of total selenium content by ICP-MS All samples were digested in a microwave oven, and then total selenium content was measured by ICP-MS. A certified reference material from NIST (SRM rice 1568a) with a certified Se content of 0.38±0.04 µg g -1 was used for evaluation of the accuracy of the procedure along with samples. The result of the reference rice sample was 0.36 ± 0.07 µg g -1, which agreed well with the certified value. The total selenium content obtained from samples (n=3) was: Se (no yeast) tablets 131±12 µg g -1 ; SeMC TM tablets 245.9±3.4 µg g -1 ; Se (high Se yeast) tablets 65.4 ± 3.9 µg g IC-ICP-MS for the speciation of selenium In the LC-ICP-MS approach, ion-exchange chromatography has been used to separate both inorganic and organic selenium (Wrobel et al., 2003; B Hymer and Caruso, 2006; Yang et 101

119 al., 2004a). The selenium species are initially retained on the column by anion exchange and subsequently eluted by a competitive anion included in the mobile phases (Yang et al., 2004a). Cation or anion-exchange chromatography has been used in the speciation of selenium. However, in this chapter, the anion-exchange column (G3154A/101) was investigated for the separation of both inorganic and organic selenium. A mobile phase containing 20 mm NH 4 H 2 PO 4 at ph 6.5 was used to give the complete resolution of different Se containing compounds as shown in Fig In the figure the elution order was SeCys 2, SeMC, SeMet, selenite and selenate and total separation was achieved within 12 minutes. Here, the ammonium nitrate employed to increase the ionic strength as described in Chapter 4 was not included. Its use led to poor resolution of the organic Se compounds (phosphate eluent studies will be described in Chapter 7). Abundance Time/Minutes--> Figure 6.1 Anion-exchange HPLC-ICP-MS profile for a 200 µg L -1 Se standard mixture containing SeCys 2 (1), SeMC(2), SeMet(3), selenite(4) and selenate(5), 20 mm NH 4 H 2 PO 4 at ph 6.5 eluent, m/z 78 monitored. The collision cell method discussed in Chapter 3 was not used in this chapter. Se specific detection was achieved using the isotope 78 Se by ICP-MS. Detection limits obtained for Se 102

120 species ranged from µg L -1. This is sufficiently low for the analysis of samples in this study. Calibration details and detection limits will be discussed later The extraction of selenium using SDS solution Since SDS can solubilise proteins and render them water soluble by forming ion pairs, it is used to extract selenoproteins from various matrices. Se containing amino acids can exist in free form or covalently bound in proteins. To release and characterize selenocompounds from proteins, a 15 ml solution of 1 M HCl was added to 0.2 g of the SeMC TM in a 50 ml centrifuge tube, followed by the addition of 0.6 g SDS. The data showed only SeMC and SeMet were extracted with a recovery of 60% of the total selenium (Fig. 6.2). Figure 6.2 Extraction of Se species by 4% SDS from the SeMC TM tablet Low yields indicated SDS could not completely solubilise the protein. Hence, the effect of SDS concentration on the extraction was examined in the range of 2%- 8% (w/w). The data shows the recovery was slightly reduced at high concentration with the best yields at 4%. The results indicate that SDS can be used to extract the organic selenium species such as SeMC and SeMet due to the formation of ion-pairs between SDS and such selenium species (Huerta et al., 2004). However, SDS cannot form ion pairs with inorganic selenium such as Se(IV) and Se(VI). Hence, Se(IV) and Se(VI) were not observed in this case. 103

121 6.3.4 The extraction of selenium using enzymes Comparing single and dual enzymatic hydrolysis Enzymatic extractions are often used in the literature for selenium speciation analysis. The type of enzyme chosen is important for successful enzymatic extraction (Wrobel et al., 2003). In this work, a single enzyme (protease XIV) and two-enzyme system (protease XIV/lipase VII) were employed to extract selenium and also to compare their extraction efficiency. When protease XIV enzyme was used for the leaching of the SeMC TM tablets, only SeCys 2, SeMC and SeMet were found. The extraction percentage for SeCys 2, SeMC and SeMet were 28%, 32% and 30%, respectively. No indication of the inorganic species Se(IV) and Se(VI) was observed. These results indicate that protease XIV may be useful to leach organic selenium species from a sample and confirmed its reported use for extracting selenium from yeast (Wrobel et al., 2003; Larsen et al., 2003). A dual mixture of protease XIV and lipase VII was used to extract selenium species in the same SeMC TM tablets. Lipase VII breaks down lipids in biological matrices, which extract additional membrane-bound Se fractions. The extraction yield percentages for SeCys 2, SeMC and SeMet improved to 31%, 37% and 32%, respectively. The mixed enzymatic extraction also gave an additional unidentified Se species as shown in Fig. 6.3(a), probably a selenium containing protein fragment (Połatajko et al., 2004). The unidentified compound had a peak retention time of 3.4 min close to the peak of Se(IV). The peak was found not to be Se(IV) by spiking with a mixture of standards of five Se species, including Se(IV), as shown in Fig. 6.3(b). The unknown compound may perhaps be further identified by other techniques such as ESI-MS-MS (B Hymer and Caruso, 2006). 104

122 Abundance ? Time/Minutes--> Figure 6.3(a) Extraction of Se species by 20 mg protease XIV and 10 mg lipase VII from the SeMC TM tablet, Se detected at m/z 78 Abundance ? Time/Minutes--> Figure 6.3(b) A 150 µg L -1 mixture of each of five Se species standards were mixed with the extracted protease XIV and lipase VII solution from the SeMC TM tablet (1:1), Se detected at m/z

123 The improved extraction efficiency using a dual enzyme system suggests that Se containing proteins are hydrolysed to release free seleno-amino acids more efficiently following the cleavage of peptide bonds. Additional information needs to be gathered to establish whether the enzymes operate synergistically or simply target a broader range of peptide bonds (Wrobel et al., 2003; Larsen et al., 2003). On the basis of these data, dual protease XIV and lipasevii was used for further studies. The effect of the amount of enzyme was tested. Recoveries for Se varied, probably due to different rates of hydrolysis of the protein (B Hymer and Caruso, 2006). A mixture consisting of 20 mg protease XIV and 10 mg lipase VII gave consistent results and was chosen for further experiments Effect of Tris-HCl buffer In order to increase the scope of the study, buffering of the enzyme matrix solution with Tris-HCl was investigated. Variable extraction efficiency for the selected species was obtained. For example, the recovery of SeMet was obtained in greater amounts at lower levels of buffer (50 mm), while a higher concentration of buffer resulted in improved recovery of SeMC and SeMet. Thus, a solution of 75 mm buffer was chosen for all further experiments. In summary, a mixture consisting of 20 mg protease XIV and 10 mg lipase VII with 75 mm Tris-HCl provided the best recovery method for extracting Se species Se species in pharmaceutical tables The quantification of Se species was based on the calibration plots using known standards. Calibrations were obtained by plotting peak area against the concentration of the respective Se target species or total Se. Linear calibrations were obtained over a concentration range of µg L -1 with correlation coefficients ranging from when 50 µl was injected. Detection limits obtained for Se species ranged from µg L -1, which proved to be adequate for analysis at trace levels. The repeatability of five replicates from a standard solution containing a mixture of the Se species was tested at 50 µg L -1 of each component. The results gave a RSD in the range %. The method was used to quantify Se species in the extracts. 106

124 The proposed extraction method was used to leach Se species from three samples. Selenium species distribution is given in Table 6.1. It was evident that the major Se species in the three types of tablets were SeMet and SeCys 2. The lowest level of SeMet, comprising 19% of the extract total Se content, was in Se-no-yeast TM tablets. Tablets of SeMC TM gave 37% SeMC which is an important organically-bound Se source. Recently, SeMC has exhibited greater efficacy as a chemopreventative agent than any other Secompound in experimental models of breast cancer (Tan et al., 2002; Tinggi, 2003; Wrobel et al., 2003). To validate the method, the five Se compounds selected were spiked to a level of 150 µg L -1 extracts. A total recovery in the range of % was achieved. Table 6.1 Se species found in tablet samples (µg g -1 ) SeCys 2 SeMC SeMet Se(IV) Se(VI) Extract total Se Total Se E (%) Se a 33.6 ±1.2 n/a 20.1±0.8 n/a 53.1± ± ± SeMC 76.3± ± ±2.4 n/a n/a 245.9± ± Se b 40.5±1.9 n/a 11.9±0.8 n/a n/a 52.4± ± The sample was determined with triplicates; n/a is not available. (a - no yeast; b - high yeast; E (%) - extract percentage recoveries based on total Se) 6.4 Conclusions The study in this chapter has shown that enzymatic hydrolysis and sodium dodecyl sulphate (SDS) digestion can be used for the extraction of organic Se species in tablets. However, higher extraction efficiency was obtained using enzymatic methods. In this case, a mixed proteolytic enzyme (protease XIV and lipase VII) extraction approach in a buffer containing 75 mm Tris-HCl at ph 7.5 gave significantly improved recovery of more than 80% compared to SDS at 60% (Section 6.3.3, Figure 6.2) In the enzymatic system, selenocystine (SeCys 2 ), Se-methyl-selenocysteine (SeMC) and selenomethionine (SeMet) in the samples were extracted. In contrast, chemical extraction of Se using SDS gave only two species, SeMC and SeMet. In addition, the new type of anion-exchange column was used successfully to separate both inorganic and organic Se species with ICP-MS 107

125 detection, while reasonable resolution and detection sensitivity was obtained. The information obtained for individual Se species supplements provides a better understanding for potential animal and human health benefits. 108

126 6.6 References Ajtony, Z., Szoboszlai, N., Bella, Z., Bolla, S., Szakál, P. and Bencs, L. (2005) Determination of total selenium content in cereals and bakery products by flow injection hydride generation graphite furnace atomic absorption spectrometry applying in-situ trapping on iridium-treated graphite platforms. Microchimica Acta, 150, 1-8. B'Hymer, C. and Caruso, J. A. (2000) Evaluation of yeast-based selenium food supplements using highperformance liquid chromatography and inductively coupled plasma mass spectrometry. Journal of Analytical Atomic Spectrometry, 15, B Hymer, C. and Caruso, J. A. (2001) Canadian soil quality guidelines selenium environmental and human health effects. Canadian Journal Analytical Science Spectroscopy, 46, 136. B Hymer, C. and Caruso, J. A. (2006) Selenium speciation analysis using inductively coupled plasma-mass spectrometry. Journal of Chromatography A, 1114, Bermejo, P., Capelo, J. L., Mota, A., Madrid, Y. and Camara, C. (2004) Enzymatic digestion and ultrasonication: a powerful combination in analytical chemistry. Trends in analytical chemistry, 23, Bidari, A., Hemmatkhah, P., Jafarvand, S., Milani Hosseini, M. R. and Assadi, Y. (2008) Selenium analysis in water samples by dispersive liquid-liquid microextraction based on piazselenol formation and GC ECD. Microchimica Acta, 163, Capelo, J. L., Ximenez-Embun, P., Madrid-Albarran, Y. and Camara, C. (2004) Enzymatic Probe Sonication: Enhancement of Protease-Catalyzed Hydrolysis of Selenium Bound to Proteins in Yeast. Analytical Chemistry-Washington DC-, 76, Casiot, C., Szpunar, J., obi ski, R. and Potin-Gautier, M. (1999) Sample preparation and HPLC separation approaches to speciation analysis of selenium in yeast by ICP- MS. Journal of Analytical Atomic Spectrometry, 14, Chassaigne, H., Chery, C. C., Bordin, G. and Rodriguez, A. R. (2002) Development of new analytical methods for selenium speciation in selenium-enriched yeast material. Journal of Chromatography A, 976, Cornelis, R., Crews, H., Caruso, J. A. and Heumann, K. G. (2005) Handbook of elemental speciation II: Species in the environment, Food, Medicine & Occupational Health. John Wiley & sons, Ltd, ISBN: (HB)

127 Dernovics, M., Stefánka, Z. and Fodor, P. (2002) Improving selenium extraction by sequential enzymatic processes for Se-speciation of selenium-enriched Agaricus bisporus. Analytical and Bioanalytical Chemistry, 372, Encinar, J. R., Schaumloffel, D., Ogra, R. and Lobinski, R. (2004) Determination of Selenomethionine and Selenocysteine in Human Serum Using Speciated Isotope Dilution-Capillary HPLC-Inductively Coupled Plasma Collision Cell Mass Spectrometry. Analytical Chemistry, 76, Goenaga-Infante, H., Sturgeon, R., Turner, J., Hearn, R., Sargent, M., Maxwell, P., Yang, L., Barzev, A., Pedrero, Z. and Cámara, C. (2008) Total selenium and selenomethionine in pharmaceutical yeast tablets: assessment of the state of the art of measurement capabilities through international intercomparison CCQM-P86. Analytical and Bioanalytical Chemistry, 390, Huerta, V. D., Reyes, L. H., Marchante-Gayón, J. M., Sánchez, M. L. F. and Sanz-Medel, A. (2003) Total determination and quantitative speciation analysis of selenium in yeast and wheat flour by isotope dilution analysis ICP-MS. Journal of Analytical Atomic Spectrometry, 18, Huerta, V. D., Sa nchez, M. L. F. and Sanz-Medel, A. (2004) Quantitative selenium speciation in cod muscle by isotope dilution ICP-MS with a reaction cell: comparison of different reported extraction procedures. Journal of Analytical Atomic Spectrometry, 19, Ip, C., Birringer, M., Block, E., Kotrebai, M., Tyson, J. F., Uden, P. C. and Lisk, D. J. (2000) Chemical Speciation Influences Comparative Activity of Selenium- Enriched Garlic and Yeast in Mammary Cancer Prevention. Journal of Agricultural and Food Chemistry, 48, Larsen, E. H., Sloth, J., Hansen, M. and Moesgaard, S. (2003) Selenium speciation and isotope composition in 77 Se-enriched yeast using gradient elution HPLC separation and ICP-dynamic reaction cell-ms. Journal of Analytical Atomic Spectrometry, 18, Marchante-Gayón, J. M., Thomas, C., Feldmann, I. and Jakubowski, N. (2000) Comparison of different nebulisers and chromatographic techniques for the speciation of selenium in nutritional commercial supplements by hexapole collision and reaction cell ICP-MS. Journal of Analytical Atomic Spectrometry, 15,

128 McSheehy, S., Yang, L., Sturgeon, R. and Mester, Z. (2005) Determination of methionine and selenomethionine in selenium-enriched yeast by species-specific isotope dilution with liquid chromatography-mass spectrometry and inductively coupled plasma mass spectrometry detection. Analytical Chemistry, 77, Mester, Z., Willie, S., Yang, L., Sturgeon, R., Caruso, J. A., Fernández, M. L., Fodor, P., Goldschmidt, R. J., Goenaga-Infante, H. and Lobinski, R. (2006) Certification of a new selenized yeast reference material (SELM-1) for methionine, selenomethinone and total selenium content and its use in an intercomparison exercise for quantifying these analytes. Analytical and Bioanalytical Chemistry, 385, Ogra, Y., Ishiwata, K., Encinar, J. R., Lobin ski, R. and Suzuki, K. T. (2004) Speciation of selenium in selenium-enriched shiitake mushroom, Lentinula edodes. Analytical and Bioanalytical Chemistry, 379, Polatajko, A., Sliwka-Kaszynska, M., Dernovics, M., Ruzik, R., Encinar, J. R. and Szpunar, J. (2004) A systematic approach to selenium speciation in selenized yeast. Journal of Analytical Atomic Spectrometry, 19, Tan, J., Zhu, W., Wang, W., Li, R., Hou, S., Wang, D. and Yang, L. (2002) Selenium in soil and endemic diseases in China. Science of the Total Environment, 284, Tinggi, U. (2003) Essentiality and toxicity of selenium and its status in Australia: A review. Toxicology Letters, 137, Weixiao, W., Jingyin, L., Huimin, D., Jing, H. and Shufang, L. (2006) Determination of Selenium in Biological Materials Using an Ion-Selective Electrode. Microchimica Acta, 154, Wrobel, K., Kannamkumarath, S. S., Wrobel, K. and Caruso, J. A. (2003) Hydrolysis of proteins with methanesulfonic acid for improved HPLC-ICP-MS determination of seleno-methionine in yeast and nuts. Analytical and Bioanalytical Chemistry, 375, Yang, L., Mester, Z. and Sturgeon, R. E. (2004a) Determination of Methionine and Selenomethionine in Yeast by Species-Specific Isotope Dilution GC/MS. Analytical Chemistry-Washington DC-, 76, Yang, L., Sturgeon, R. E., McSheehy, S. and Mester, Z. (2004b) Comparison of extraction methods for quantitation of methionine and selenomethionine in yeast by species specific isotope dilution gas chromatography mass spectrometry. Journal of Chromatography A, 1055,

129 Chapter 7 Selenium speciation in nutritional tablets and biofortified foods using ion chromatography with ICP- MS 112

130 7.1 Introduction Only total Se concentrations are indicated on the labelling of market products. The Se species distribution for each product is also necessary. Not only does it provide useful information to the consumer, government departments like the Therapeutic Goods Administration (TGA) can use such information to assess the nutritional worth of products to set guidelines. Selenium (Se) is an essential micronutrient and also a potential toxic trace element in animals and humans. The element s toxicological and biological effects as well as cancer chemopreventative activity depend on its concentration and chemical form or species in the original sample (Infante et al., 2005; Kápolna and Fodor, 2006). Selenocystine (SeCys 2 ), selenomethionine (SeMet), selenite and selenate are commonly encountered active compounds in Se-enriched supplements. The species Se-methyl-selenocysteine (SeMC) has recently become available to the public as a dietary supplement. It is considered to have greater efficacy as a chemopreventative agent than other organic or inorganic analogues (Kápolna and Fodor, 2006; Infante et al., 2005). Furthermore, recent research has focused on understanding the relationship between Se-enriched supplements and biofortified foods and their nutritional benefits for humans (Kirby et al., 2008). Hence, the speciation of organic and inorganic selenium is of interest. A variety of methods are used for the determination of selenium species (Uden et al., 2004; B Hymer and Caruso, 2006). The LC-ICP-MS approach using ion-exchange chromatography has been used to separate some organic and inorganic selenium (Kirby et al., 2008; Uden et al., 2004; B Hymer and Caruso, 2006). Larsen et al. (2003) developed a gradient elution cation-exchange method for the separation of organic selenium species in the hydrolysates of selenium enriched yeast. The eluent consisted of 0.75 to 8 mm pyridinium formate and 97% v/v in water/methanol. Various mobile phases such as citrate (Huerta et al., 2003), phosphate (Chassaigne et al., 2002; Reyes et al., 2004) and formate (Larsen et al., 2003; Sloth and Larsen, 2000) have been used to resolve the Se species by anion-exchange chromatography. For example, anion-exchange chromatographic separation of selenomethionine, selenite and selenate from yeast extracts was achieved by Hueta et al. (2003) using an ammonium citrate buffered mobile phase with 2% methanol as organic modifier (Larsen et al., 2003). However, modifiers such as methanol are not 113

131 compatible with the ICP plasma due to an increase in the reflected power (Larsen, 1998). This can result in plasma extinguishing even at relatively low organic modifier concentrations. Another undesired effect is the altered ionisation characteristics of the Ar plasma leading to a reduction in detection sensitivity for elemental species (Sadi et al., 2004; Michalke, 2002; Montes-Bayón et al., 2003). In addition, sodium or potassium phosphate buffers are generally not desirable owing to the residue left on the sampler and skimmer cones of the ICP-MS (B Hymer and Caruso, 2006). Therefore, ammonium salts buffer systems, which can be converted into volatile compounds under high plasma temperature, are more popularly used in selenium speciation analysis when employing ionexchange chromatography with ICP detection (B Hymer and Caruso, 2006). However, optimal resolution of Se species depends not only on the eluent used but also on the column used. Anion-exchange columns (Hamilton PRP-X100 and Dionex AS, etc.) are commonly utilised and reported (see Table 2.3). The use of the polymethacrylate column has not been reported to date. An Agilent G3154A/101 column has already been employed successfully for inorganic selenium in Chapter 4 (Chen et al., 2008); however it was not used for both organic and inorganic selenium species. Hence, it was selected for consideration in this study and its performance tested for separating such mixtures. Previously, it was noted that polyatomic ions in the argon plasma interfere with measurement of selenium isotopes in ICP-MS (subsection 3.2). As demonstrated in Chapters 3 and 4, the use of collision/reaction cell technology can remove these interferences, resulting in improved ICP-MS detection power for Se by permitting monitoring of its most abundant isotope, 80 Se. Therefore Chapter 4 provided the details of comparing no reaction gas, H 2 and He with the ORS for selenite and selenate species and showed the possibility of using a no reaction gas in ICP-MS for detecting 78 Se. Hence, it is required to broaden the study to obtain more complete information about sensitivity for each of five Se species with or without ORS. In this chapter, the performance of a new type of anion-exchange column designed to separate organic and inorganic selenium species will be investigated for considering this separation method. One reason was to reduce interference in the ICP-MS through mobile phase selection. Thus, an organic modifier will be avoided by using an ammonium salt in the eluent. Finally, the influence of an octopole reaction system (ORS) on detection sensitivity was explored for five Se species. In summary, the following aspects of Se 114

132 chromatographic analysis by ICP-MS were investigated: (1) the separation using anion exchange; (2) ICP-MS detection conditions; and (3) the application of the optimized method for the analysis of selenium in nutritional tablets and biofortified foods. 7.2 Experimental Chemicals and solutions All reagents were of analytical reagent grade and obtained from Sigma-Aldrich (Sydney, Australia). Milli-Q water, specific resistance of 18.2 MΩ cm, prepared by Millipore equipment, (Bedford, MA, USA), was used to prepare all solutions and standards. Stock solutions were prepared at 500 mg L -1 (for Se) of Se-methylselenocysteine (SeMC), seleno- L-methionine (SeMet), seleno-l-cystine (SeCys 2 ), 1000 mg L -1 (for Se) of Se(IV) and Se(VI) by dissolving the appropriate amount of the corresponding compound in Milli-Q water (see subsections and 4.2.1). The enzymes, protease XIV and lipase VII were purchased from Sigma-Aldrich (Sydney, Australia). Tris-(tris(hydroxymethyl)aminomethane)-HCl buffer (75 mm, adjusted ph=7.5) and 20 mm NH 4 H 2 PO 4 (ph 6.5) were prepared daily as required (see 6.2.1). Five different sources of Se samples were accessed. These included commercial tablets of Se-no-yeast TM, Se-high-yeast TM and SeMC TM from Wholesale Nutrition (Saratoga, CA, USA, and biofortified Se foods (Bio-Fort TM ) Laucke wheat flour and Wafer grains TM biscuits purchased from a local Woolworths store in Adelaide, South Australia Total Se determination Fortified supplements were prepared for analysis by grinding tablets in a mortar and biscuits using a coffee mixer. A 0.5 g portion of sample was accurately weighed into PTFE vials and dissolved in a mixture of 2.0 ml of conc. HNO 3 and 2.0 ml of 30% H 2 O 2. After 17 h of contact time, total digestion was performed at 170ºC for 15 min at 300 W in a microwave oven. The final solutions were made up to 25 ml in a volumetric flask with 115

133 MilliQ water. A digestion blank was also performed containing water, acid and peroxide solution with the same procedures Extraction A 15 ml solution of Tris-HCl buffer, adjusted to ph 7.5, was added to 0.2 g of the sample in a 50 ml centrifuge tube. This was followed by adding 20 mg protease XIV and 10 mg lipase VII. The mixture was then sonicated for 2 h, and incubated in the dark for 12 h at 25ºC. Hydrolysed samples were centrifuged at 4000 rpm for 25 min and the supernatant filtered through a 0.45 µm membrane filter prior to analysis. For the other samples see section Instrumentation Instrument of ICP-MS parameters used in this study were the same as provided in section 3.4. IC instrumental conditions used were the same as list in subsection The mobile phase was 20 mm ammonium dihydrogen phosphate in water, adjusted ph to selected values. 7.3 Results and discussion Protonation state of Se species in aqueous solution The behaviour of inorganic Se anions, selenite and selenate, in ion chromatography was previously discussed in Chapter 4 (4.3.3). Since selenoamino-acids are weak organic acids, their behaviour in IC is expected to be different from anions such as SeO 2-3 and SeO 2-4. This raises the question: what is the protonation state of each selenoamino-acid as the ph is varied? A useful approach to the above considerations is provided by viewing the master variable or deprotonation diagram for each selenoamino-acid as ph is altered (Harris, 2007). The diagrams display fractional composition values, α HnA, derived using the following equations. 116

134 Fraction of Species Alpha n α HnA =[H + ] n /D α Hn-1A =Ka 1 [H + ] n-1 /D α Hn-2A = Ka 1 Ka 2 [H + ] n-2 /D α Hn-jA =Ka 1 Ka 2.Ka j [H + ] n-j /D D=[H + ] n + Ka 1 [H + ] n-1 + Ka 2 [H + ] n Ka 1 Ka 2.Kan The relative distribution of SeCys 2, SeMet and cysteine will be discussed below. For SeCys 2 five species are evident in the range ph 1-14 (Fig 7.1). In the ph range from 2.0 to 8.0, the zwitterionic species (H 2 SeCys 2, alpha2) is the major form. At ph 4.8 it represents almost 100% of SeCys 2 and continues to be dominant in the range The totally deprotonated form, SeCys 2-2 (alpha4) has a significant concentration only beyond ph 8.0. The other ions such as the monoprotonated form, HSeCys - 2 (alpha3), constitutes about 60% at ph 8.5, while the triprotic ion, H 3 SeCys + 2 (alpha1), and fully protonated cation, H 4 SeCys 2+ 2 (alpha0), are dominant below ph 2.0. Both species are absent at ph levels above Alpha0_H 4 SeCys 2+ 2 Alpha1_H 3 SeCys + 2 Alpha2 _H 2 SeCys 2 Alpha3 _HSeCys - 2 Alpha4 _SeCys ph Figure 7.1 Species distribution diagram for SeCys 2 in the ph range from 0 to

135 Fraction of Species Alpha n Equilibria shift to the right in accordance with their pka values as ph increases: H 4 SeCys 2 H 3 SeCys 2 H 2 SeCys 2 HSeCys 2 SeCys 2 - The relative levels of species such as H 2 SeCys 2 and HSeCys 2 chromatographic considerations. may play a role in The outcome is similar for SeMet (Fig. 7.2). The zwitterionic species (HSeMet, alpha1) exists over a wide ph range from 2.0 to Between ph 4 and 7 it is the only species present in solution at any appreciable level. At higher ph levels between 7.0 and up to 12.0, the completely deprotonated species, SeMet - (alpha2), dominates, while at a ph lower than 5.0 the completely protonated form, H 2 SeMet + (alpha0), is abundant. The equilibrium equations according to pka values are: H 2 SeMet + HSeMet SeMet Alpha0_H 2 SeMet + Alpha1_HSeMet Alpha2_SeMet ph Figure 7.2 Species distribution diagram for SeMet in the ph range from 0 to 14 To the author s knowledge, no experimental dissociation constants (pka) are available for SeMC. However, its behaviour would be expected to be similar to that of selenocysteine, which in turn is closely related to cysteine in acid-base terms. For cysteine (Fig. 7.3), the zwitterionic species is the dominant species in solution between ph 4 and

136 Fraction of Species Alpha n Alpha0 _H 2 Cys + Alpha1_HCys Alpha2_Cys - Alpha3_Cys ph Figure 7.3 Species distribution diagram for cysteine in the ph range from 0 to 14 Hence, we can predict that the same would apply to SeMC, the zwitterionic species, HSeMC, as the dominate SeMC species in solution between ph 4 and Anion-exchange chromatographic separation of selenium With conventional anion exchange chromatography, the separation mechanism is based on exchange equilibria between surface ions on the stationary phase and similarly charged ions in the mobile phase (Uden et al., 2004; B Hymer and Caruso, 2006). While this is true for the ions with a higher charge, seleno-amino acids are weak organic acids and exist mainly as zwitterionic species (ammonium and carboxylate groups) in the ph range 4-7 as dicussed in section In this instance, the mechanism of ion-exchange chromatography is more likely the result of enforced pairing effects (EP) (Fritz, 2005). Enforced pairing effects result when stronger ion pairing occurs within the stationary phase compared to that in the aqueous solution. This was suggested to be a combination of effects that include hydrophobic attraction, hydrogen bonding, lower dielectric constant and water-structure induced ion pairing. The equilibrium of exchange reaction in IC can be expressed by the equation: R 4 N + E - + A - R 4 N + A - + E - 119

137 where E - is the eluent anion and A - a sample anion. The fixed ammonium ion (R 4 N + ) on the ion-exchanger surface excludes cations from the mobile phase entering the stationary phase. Thus, cations have little effect on the exchange equilibrium, while the nature of the counter-ions within the ion exchanger has a major impact. Accordingly, the ph of the mobile phase and the concentration of competing ions in the mobile phase can significantly influence the retention of the species under study. In experiments outlined in Chapter 4, a combination of ammonium nitrate and ammonium phosphate was used in the mobile phase. The nitrate salt significantly increased the ionic strength of the mobile phase, causing the rapid elution of analytes. Since the present aim is to resolve organic species, the mobile phase in this experiment was simplified to just aqueous ammonium phosphate. This caused both a delay in elution times of all analytes and at the same time assisted in their separation. The concentration of competing ions and the ph were found to be the main factors in determining whether selenium species were successfully eluted from the column. Thus, it was necessary to optimize these parameters. Ammonium phosphate was also selected as buffer in the mobile phase since it converts to volatile products under ICP-MS conditions (Weiss and Haddad, 1995). The concentration of competing ions such as H 2 PO - 4 and HPO 2-4 will exert a major effect on successful ion exchange within a sample. Hence, the concentration of phosphate ions in the eluent was investigated in the range of 5-25 mm at ph 7.0. The effect of phosphate anions on the retention for the selenium species is shown in Fig The retention time generally decreased with increasing concentration of NH 4 H 2 PO

138 Log Retention time SeCys2 SeMC SeMet Se(IV) Se(VI) NH 4 H 2 PO 4 (mm) Figure 7.4 The influence of the concentration of NH 4 H 2 PO 4 on the retention of Se species at ph 7.0. The retention of Se (VI) retention time was reduced significantly when the concentration of the phosphate ions increased. The higher the concentration of competing ions in the eluent, the more effectively the eluent displaced target ions from the stationary phase and the more rapid the elution of the target ions from the column (Weiss and Haddad, 1995). Good retention and resolution of all five selenium species was obtained using mm NH 4 H 2 PO 4. Efficient chromatography was observed beyond 20 mm NH 4 H 2 PO 4. This concentration was chosen for further experimentation. The relationship between the mobile phase ph and retention time using 20 mm NH 4 H 2 PO 4 in the mobile phase is shown in Fig

139 Log retention time SeCys2 SeMC SeMet Se(IV) Se(VI) ph Figure 7.5 The influence of mobile phase ph on the retention of Se species Retention of the selenium species decreased with increasing eluent ph. This results from the marked increase in the concentration of the doubly charged phosphate species over the ph range 5.5 to 7.0. The HPO 2-4 anion competes very effectively with doubly charged inorganic selenium ions or zwitterions of seleno-amino acids for sites on the stationary phase (Weiss and Haddad, 1995). Inorganic selenite and selenate ions in aqueous solution are affected by the eluent ph, where the species HSeO - 3 and SeO 2-4, mainly present as strongly charged ions, are subject to electrostatic attraction (ES) as in conventional anion-exchange mechanisms. Here, tight bonding results between the positively charged stationary phase and ions of HSeO - 3 and SeO 2-4, causing prolonged retention times as shown in Fig However, seleno-amino acids in this ph range exist essentially as zwitterions with ionised ammonium and carboxylate groups. These neutral species are not retained as a result of anion-exchange mechanisms. Instead, factors like hydrophobic character, hydrogen bonding and polarity of the compounds affect the retention time. Therefore, the lowest retention time was observed for SeCys 2, which has the highest carbon content (six carbons). The retention time for SeMC with five carbons was increased, with the longest retention time being for SeMet with four carbons. Considering retention times, a mobile phase ph at 6.5 was used for further work. 122

140 7.3.3 ICP-MS detection of 78 Se and 80 Se with an ORS An ORS technique was used to determine whether interferences such as 40 Ar 40 Ar + and 38 Ar 40 Ar + could be reduced during detection of 78 Se and 80 Se in this study. Chapter 4 investigated the effect of an ORS on two inorganic Se species. In the present case, it is necessary to find the sensitivity for each of the five Se species. Separation of the selenium species was performed on the chosen anion-exchange column with a mobile phase containing 20 mm NH 4 H 2 PO 4 at ph 6.5. Chromatograms obtained using IC-ICP-MS with Ar alone, i.e. no reaction gas, as well as H 2 and He in the ORS are shown in Fig Each set of conditions was applied to the ORS when a mixture of 500 μg/l of each selenium species was injected into the IC-ICP-MS system. Abundance He Ar Gas H Time/ Minutes--> Figure 7.6 Comparison of H 2, He gas added in reaction cell and no reaction gas (Ar alone) for detecting 78 Se. Octopole conditions: H 2 flow rate of 3.0 ml/min and He flow rate of 3.5 ml/min. Injection of mixture of five Se species standards each at level of 500 µg/l in Se containing SeCys 2 (1), SeMC(2), SeMet(3), selenite(4) and selenate(5) Despite the background being near 3000 (abundance), detection of 78 Se using no reaction gas was possible. However, the background intensity was reduced when H 2 (1000, abundance) or He (60, abundance) was added to the cell because the 40 Ar 38 Ar + species was 123

141 effectively removed or reduced by their different mechanisms (Tanner et al., 2002; McCurdy and Woods, 2004). In contrast to the detection of 78 Se, Fig. 7.7 shows the chromatogram obtained when monitoring 80 Se. The separation and detection conditions were the same as described for 78 Se in Fig No observable signal was obtained when no reaction gas was used for 80 Se, only baseline noise. The background noise was very high ( , abundance), resulting from the formation of 40 Ar 40 Ar +. Abundance He H Time/Minutes--> Figure 7.7 The impact of reaction cell gas selection on sensitivity for 80 Se. Flow rates and chromatographic conditions were the same as in Fig Reaction cell gas was H 2 or He. Since 40 Ar(99.60%) is the principle isotope, 40 Ar 40 Ar + formed in the plasma source can significantly interfere with the determination of Se (Darrouzès et al., 2005; Ogra et al., 2005). Therefore, signals for the five Se compounds were masked at m/z 80. However, as shown in Fig. 7.7, the background noise was significantly reduced when either H 2 or He gas was used, where background noise reduced to 2300 (abundance) and 7000 (abundance), respectively. Compared to no reaction gas, this translated into a decrease in background noise from 100% to 0.02% using H 2 gas at flow rate of 3.0 ml/min, and from 100% to 0.07% using He at flow rate of 3.5 ml/min. In summary, the sensitivity of H 2 gas was about ten times higher than that of He gas for detection of 80 Se. 124

142 m/z=78 m/z=80 m/z=78 m/z=80 m/z=78 m/z=80 m/z=78 m/z=80 m/z=78 m/z=80 Peak Area The sensitivity for each mode was calculated by peak area for detection of 78 Se and 80 Se as shown in Fig Ar H2 He Comparing 78 Se/ 80 Se with Ar/H 2 /He gas SeCys2 SeMC SeMet Se(IV) Se(VI) Figure 7.8 Comparison of tuning methods for detecting Se species by Ion chromatography ICP-MS: Argon alone no reaction gas ; H 2 ; and He. Chromatographic conditions as in Fig Each Se species is at level 500 g/l in Se Detection of 78 Se decreased in the order: no reaction gas > H 2 > He, the same order as the decrease in background noise (Fig. 7.6). Since 38 Ar is only a minor isotope, the 40 Ar 38 Ar + did not interfere significantly with the detection of 78 Se using no reaction gas. It was concluded that the detection of 78 Se using no reaction gas is possible since 40 Ar 38 Ar + does not interfere significantly with the determination of Se. Furthermore, compared to 78 Se when using no reaction gas, sensitivity of detecting 80 Se using H 2 gas was nearly two times higher. H 2 as the reaction gas was more effective in removing 40 Ar 40 Ar + background while providing higher sensitivity (signal/noise) and lower background. Thus, H 2 added at 3.0 ml/min to the ORS and monitoring of 80 Se was used in the following section. 125

143 7.3.4 Analytical performance characteristics and application Fig. 7.9 illustrates the chromatogram of Se species detected using 80 Se, where the separation of all Se species was observed within 15 min with a mobile containing 20 mm NH 4 H 2 PO 4 at ph 6.5. Abundance Time/Minutes--> Figure 7.9 A typical chromatogram for separation of selenium species, detection at m/z 80. Octopole H 2 flow rate of 3.0 ml/min. Chromatographic conditions and species numbering as in Fig The elution order was SeCys 2, SeMC, SeMet, selenite and selenate with reasonable resolution and detection sensitivity. Calibrations were obtained by plotting peak area against the concentration of the respective Se target species. Linear calibrations for mixed standards were obtained over a concentration range of µg/L -1 with correlation coefficients ranging from when 50 µl samples were injected (Fig. 7.10). Detection limits (S/N=3) obtained for Se species ranged from µg/l -1. The repeatability of five replicates from a standard solution containing a mixture of the Se species at 50 µg L -1 of each was tested. The results gave RSD values in the range %. 126

144 Abundance SeCys2 SeMC SeMet Se(IV) Se(VI) y = x R² = y = x R² = y = x R² = y = x R² = y = x R² = Se Conc. in µg/l Figure 7.10 Calibrations for Se five species A reference sample NIST SRM rice 1568a with a certified Se content of 0.38±0.04 µg/g, was used for evaluation of the accuracy of the procedure. The ICP-MS method gave a value of 0.36 ±0.07 µg/g for the reference rice sample, which agreed well with the certified value. The proposed method was applied to the determination of replicate (n=3) samples in the nutritional tablets and biofortified Se foods. Total Se was measured by ICP-MS after microwave and acidic digestion (see section 7.2.2). The total Se concentrations in tablets of Se-no-yeast TM, Se-high-yeast TM, SeMC TM, were found to contain 131±12 µg/g, 245.9±3.4 µg/g, and 65.4±3.9 µg/g, while biofortified foods Laucke flour and Wafer grains biscuits contained 1.00±0.03 µg/g and 3.50± 0.04 µg/g of Se, respectively. These results are listed in Table

145 Table 7.1 Concentration (µg g -1 ) found for Se species in biological samples by ICP-MS Sample Name SeCys 2 SeMC SeMet Se(VI) Total Se Se(no yeast) 33.6±1.2 n/a 20.1± ± ±12 SeMC 76.3± ± ±2.4 n/a 245.9±3.4 Se(high yeast) 40.5±1.9 n/a 11.9±0.8 n/a 65.4±3.9 Laucke flour n/a n/a 0.80±0.05 n/a 1.00±0.03 Wafer grains biscuits n/a n/a 3.0±0.1 n/a 3.50±0.04 The samples were determined as triplicates; n/a is not available or not detected. The Se species were extracted using enzymatic hydrolysis as described previously in Chapter 6 (Wang et al., 2009) and followed by determining the species with the developed method. To validate this method, a mixed standard containing the five Se compounds was added to the extracts of tablets (each Se species at a level of 250 µg/l) and to biofortified foods (25 µg/l of each Se species). Recoveries for standards of each after spiking ranged from %. Species concentrations are listed in Table 7.1, and typical chromatograms obtained from nutritional tablets (Se-no-yeast TM ) and biofortified foods (Wafer grains biscuits) are shown in Fig (a) and (b). Figure 7.11 (a) Abundance Time/Minutes--> 128

146 Figure 7.11 (b) Abundance Time/Minutes--> Figure 7.11 Chromatograms obtained from commercial products. Chromatographic conditions as in Fig Samples were: (a) nutritional tablets (Se-no-yeast TM ); (b) biofortified food (Wafer grains biscuits) The chromatogram in Fig (a) shows that SeCys 2, SeMet and selenate species were present in the Se-no-yeast TM tablet, while in Wafer Grains biscuits SeMet was the main species observed (Fig (b)). In general, the major Se species in the three tablets types were SeMet and SeCys 2, the brand with the lowest SeMet content comprising 15% of the total Se was Se-no-yeast TM, and species SeMC was only present in the SeMC TM tablet. While in two biofortified foods, SeMet was the main Se species observed. SeMC, an important organically-bound Se source, comprising 37% of the total Se was found in tablets of SeMC TM. It was not observed in the other tablets. In recent research, SeMC was found to exhibit a greater efficacy as a chemopreventative agent than other Secompounds for breast cancer (Larsen et al., 2003; Huerta et al., 2003). These results suggest that of all supplements, tablet SeMC TM containing high SeMC species has the greatest nutritional worth for cancer prevention. It is hoped that the information obtained for individual Se species in supplements and biofortified foods will provide a better understanding of potential animal and human health benefits. 129

147 7.4 Conclusions This Chapter has focused on the separation conditions using anion-exchange chromatography for five Se species in biological samples selected from foodstuffs (health biscuits and wheat flour) and nutraceutical/pharmaceutical tablets. The speciation of selenium, including inorganic selenite and selenate and organic selenocystine (SeCys 2 ), Semethyl-selenocysteine (SeMC) and selenomethionine (SeMet) in nutritional tablets and biofortified foods, was successfully achieved using ion chromatography (IC) with inductively coupled plasma mass spectrometry (ICP-MS). A newly developed anion-exchange column (G3154A/101) with porous polymethacrylate resin was demonstrated to perform well in complete separation of selenium species, including organic and inorganic selenium compounds. The mobile phase containing 20 mm NH 4 H 2 PO 4 at ph 6.5 gave reasonable resolution. The selection of H 2 as the reaction gas in an ORS led to the effective removal of Ar interference and provided excellent sensitivity for 80 Se. Compared to the no reaction gas mode for directly detecting 78 Se, sensitivity of detection 80 Se using H 2 gas was nearly two times higher. The analysis of Se in nutritional supplements and biofortified foods demonstrated the adequacy of the developed separation method for complex matrices. The study indicates that its application to a range of complicated biological samples is feasible. 130

148 7.5 References B Hymer, C. and Caruso, J. A. (2006) Selenium speciation analysis using inductively coupled plasma-mass spectrometry. Journal of Chromatography A, 1114, Chassaigne, H., Chery, C. C., Bordin, G. and Rodriguez, A. R. (2002) Development of new analytical methods for selenium speciation in selenium-enriched yeast material. Journal of Chromatography A, 976, Chen, Z. L., Wang, W., Mallavarapu, M. and Naidu, R. (2008) Comparison of no gas and He/H 2 cell modes used for reduction of isobaric interferences in selenium speciation by ion chromatography with inductively coupled plasma mass spectrometry. Spectrochimica Acta Part B: Atomic Spectroscopy, 63, Darrouzès, J., Bueno, M., Lespès, G. and Potin-Gautier, M. (2005) Operational optimisation of ICP-octopole collision/reaction cell-ms for applications to ultratrace selenium total and speciation determination. Journal of Analytical Atomic Spectrometry, 20, Feldmann, I., Jakubowski, N. and Stuewer, D. (1999) Application of a hexapole collision and reaction cell in ICP-MS Part I: Instrumental aspects and operational optimization. Fresenius'Journal of Analytical Chemistry, 365, Fritz, J. S. (2005) Factors affecting selectivity in ion chromatography. Journal of Chromatography A, 1085, Harris, D. C. (2007) Quantitative Chemical Analysis (7th Edition), New York, W. H. Freeman. Huerta, V. D., Reyes, L. H., Marchante-Gayón, J. M., Sánchez, M. L. F. and Sanz-Medel, A. (2003) Total determination and quantitative speciation analysis of selenium in yeast and wheat flour by isotope dilution analysis ICP-MS. Journal of Analytical Atomic Spectrometry, 18, Infante, H. G., Hearn, R. and Catterick, T. (2005) Current mass spectrometry strategies for selenium speciation in dietary sources of high-selenium. Analytical and Bioanalytical Chemistry, 382, Kápolna, E. and Fodor, P. (2006) Speciation analysis of selenium enriched green onions (Allium fistulosum) by HPLC-ICP-MS. Microchemical Journal, 84, Kirby, J. K., Lyons, G. H. and Karkkainen, M. P. (2008) Selenium Speciation and Bioavailability in Biofortified Products Using Species-Unspecific Isotope Dilution 131

149 and Reverse Phase Ion Pairing Inductively Coupled Plasma Mass Spectrometry. Journal of Agricultural and Food Chemistry, 56, Larsen, E. H. (1998) Method optimization and quality assurance in speciation analysis using high performance liquid chromatography with detection by inductively coupled plasma mass spectrometry. Spectrochimica Acta Part B: Atomic Spectroscopy, 53, Larsen, E. H., Sloth, J., Hansen, M. and Moesgaard, S. (2003) Selenium speciation and isotope composition in 77 Se-enriched yeast using gradient elution HPLC separation and ICP-dynamic reaction cell-ms. Journal of Analytical Atomic Spectrometry, 18, McCurdy, E. and Woods, G. (2004) The application of collision/reaction cell inductively coupled plasma mass spectrometry to multi-element analysis in variable sample matrices, using He as a non-reactive cell gas. Journal of Analytical Atomic Spectrometry, 19, Michalke, B. (2002) The coupling of LC to ICP-MS in element speciation: I. General aspects. Trends in analytical chemistry, 21, Montes-Bayón, M., DeNicola, K. and Caruso, J. A. (2003) Liquid chromatography inductively coupled plasma mass spectrometry. Journal of Chromatography A, 1000, Ogra, Y., Ishiwata, K. and Suzuki, K. T. (2005) Effects of deuterium in octopole reaction and collision cell ICP-MS on detection of selenium in extracellular fluids. Analytica Chimica Acta, 554, Reyes, L. H., Sanz, F. M., Espílez, P. H., Marchante-Gayón, J. M., Alonso, J. I. G. and Sanz-Medel, A. (2004) Biosynthesis of isotopically enriched selenomethionine: application to its accurate determination in selenium-enriched yeast by isotope dilution analysis-hplc-icp-ms. Journal of Analytical Atomic Spectrometry, 19, Sadi, B. B. M., Vonderheide, A. P. and Caruso, J. A. (2004) Analysis of phosphorus herbicides by ion-pairing reversed-phase liquid chromatography coupled to inductively coupled plasma mass spectrometry with octapole reaction cell. Journal of Chromatography A, 1050, Sloth, J. J. and Larsen, E. H. (2000) The application of inductively coupled plasma dynamic reaction cell mass spectrometry for measurement of selenium isotopes, 132

150 isotope ratios and chromatographic detection of selenoamino acids. Journal of Analytical Atomic Spectrometry, 15, Tanner, S. D., Baranov, V. I. and Bandura, D. R. (2002) Reaction cells and collision cells for ICP-MS: a tutorial review. Spectrochimica Acta Part B: Atomic Spectroscopy, 57, Uden, P. C., Boakye, H. T., Kahakachchi, C. and Tyson, J. F. (2004) Selective detection and identification of Se containing compounds-review and recent developments. Journal of Chromatography A, 1050, Wang, W., Chen, Z. L., Davey, D. E. and Naidu, R. (2009) Extraction of selenium species in pharmaceutical tablets using enzymatic and chemical methods. Microchimica Acta, 165, Weiss, J. and Haddad, P. R. (1995) Ion chromatography, VCH Weinheim. 133

151 Chapter 8 Selenium speciation of biofortified foods using ion-pair reversed phase chromatography with ICP-MS 134

152 8.1 Introduction The retention times for seleno amino species are such that their resolution may be lost in real samples. As mentioned in section of Chapter 2, ion pairing reversed phase chromatography has been successfully used to separate seven selenium species containing compounds (Zheng and Kosmus, 2000). Since literature reports utilise an extensive range of ion pairing reagents, a search for a suitable combination needs to be conducted. Selenium is an essential micronutrient for humans and animals found in many foods. Concentrations vary according to geographical origin of the agricultural products and the Se content of the soil (Girling, 1984; Dumont et al., 2006b). Dietary intake of Se is recommended at 55 µg/day (Gosetti et al., 2007). To increase Se intake, the enrichment of food crops and plants with Se has become available to improve nutritional requirements (Ellis and Salt, 2003). Vegetables from the Allium family, such as garlic, onion or ramp after enrichment yield Se-methyl-selenocysteine and γ-glutamyl-se methylselenocysteine as the two primary species. These are further metabolised to methylselenol and could be a critical factor in Se chemoprevention (Shah et al., 2004). To understand the relationships between the Se species content in foods and their nutritional benefits, such as absorption, tissue distribution, bioavailability and cancer preventative properties, speciation of major Se species (seleno-amino acids and inorganic species, etc) in complicated foods is therefore very relevant. State-of-the-art methods have been applied for Se speciation including combined separation techniques with highly sensitive detectors, such as HPLC-ICP-MS (B Hymer and Caruso, 2006) or HPLC-ESI-MS (Dumont et al., 2005). Electrospray mass spectrometry (ESI-MS) is used to confirm the structure of the synthesised and purified compounds. In the case of ESI-MS, due to matrix effects, samples needed to be extensively concentrated or cleaned up, while the mobile phase composition is limited. Therefore HPLC-ICP-MS has been proved to be a powerful tool for Se speciation with its high sensitivity, minimal matrix effects and wide compatibility with a range of mobile phase compositions. The critical issue for HPLC-ICP-MS is the choice of separation modes to separate organic and inorganic Se species. In HPLC, size-exclusion, anion exchange, reverse phase and ion 135

153 pair reversed-phase chromatography have been employed for separating Se compounds in biological materials (Goenaga-Infante et al., 2008; Mester et al., 2006). In the anion exchange chromatographic approach, the selenium species are initially retained on the column by anion exchange and subsequently eluted by a competitive anion included in the mobile phase. Those which have been used include ammonium carbonate (Wang et al., 2007); ammonium citrate (Huerta et al., 2003; Warburton and Goenaga-Infante, 2007); ammonium phosphate (Reyes et al., 2004); pyridinium formate (Larsen et al., 2003; Kápolna and Fodor, 2006); and ammonium salts (Chen et al., 2008). Ammonium salts are popular with Se speciation using ICP-MS (B Hymer and Caruso, 2006; Huerta et al., 2003; Warburton and Goenaga-Infante, 2007; Chen et al., 2008), because ammonium salts can decompose into volatile substances under a high ICP plasma temperature. However, the chromatographic baseline can be high since there is a high salt concentration in the mobile phase resulting in signal suppression. This is due to increased space-charge effects which defocuses the ion beam (Larsen, 1998). In Chapter 7, baseline counts were often about In recent years, with increased demands for organic Se speciation and improvement of background noise being raised, the application of ion-pair reverse phase chromatography with appropriate reagents for separation of mixtures of anionic and cationic ions, as well as neutral molecules in complicated matrices has increased. The stationary phase, C8 or C18, is coated with an ion-pairing reagent with a lipophilic tail present in the mobile phase. The ionic selenium species can subsequently exchange on the coated column by pairing with the polar head (Zheng and Kosmus, 2000). Therefore, the separation mechanism depends on the ability of ion-pair formation between target compounds and counter ions. Selectivity and retention of the analytes are controlled by varying ion pairing reagents. Anionic reagents, such as 5 mm hexanesulphonic acid (HSA) and methanesulphonic acid (MSA) at ph 3.5 have been reported for separating selenomethionine (SeMet), selenoetnionine (SeEt) and selenocystine (SeCys 2 ) in Brazil nuts (Vonderheide et al., 2002). Heptafluorobutanoic acid (HFBA) (Kotrebai et al., 2000; Wróbel et al., 2004; Cankur et al., 2006) and trifluoroacetic acid (TFA) as general purpose system have also been used in ICP-MS or ESI-MS. Using HFBA, it allowed more than 75% of the total eluting Se compounds to be determined in extraction solutions of garlic, onion, ramp and yeast (Gosetti et al., 2007; Kotrebai et al., 2000). The use of other ion-pairing agents was also 136

154 investigated. Cationic counter ions, such as 13 mm tetrabutylammonium hydroxide (TBAH) have been employed for the separation of six Se species in human urine at ph (Pan et al., 2007). A mixture of 8 mm TBA with 2.5 mm butanesulphonate at ph 4.5 was used to separate nine Se species (Zheng et al., 2003), while 10 mm at ph 5.0 achieved separation of Se species in supplements (Marchante-Gayón et al., 2000). Tetrabutylammonium acetate (TBAA) at a concentration of mm mixed with 4.0 mm Na 2 HPO 4 at ph were able to separate arsenic and Se compounds (Do et al., 2001). A solution of 0.1% tetraethylammoniumchloride (TEA) as cationic counterion at ph was used to separate seven Se species in tablets and detected by microwave assisted hydride generation-afs (Dumont et al., 2004). However, difficulties have been reported in the effort to completely separate Se organic species, as well as inorganic species using ion-pair reversed phase chromatography (Zheng and Kosmus, 2000; Dumont et al., 2006a). Some methods are not suitable for maintaining their separation performance in complicated food samples. Furthermore, due to lack of sensitivity, some failed to provide sufficiently low detection limits (less than 5µg/g in dry food). To address the low levels of Se in food, even in enriched samples, advances in ICP-MS using an octopole reaction system (ORS) have been utilised in Chapter 4 and 7 (Chen et al., 2008). The purpose of this chapter was to separate and identity Se species in biofortified foods using ion-pair reverse phase chromatography with inductively coupled plasma mass spectrometry (IP-RP-ICP-MS) with ORS. Six ion-pairing agents were systematically examined and optimized for ph of the mobile phase and concentration of ion-pairing reagents. 8.2 Experimental Chemicals and solutions All reagents were of analytical reagent grade and obtained from Sigma-Aldrich (Sydney, Australia). Milli-Q water (a specific resistance of 18.2 MΩ cm, Millipore, Bedford, MA, USA) was used to prepare all solutions and standards. Stock solutions were prepared by mixing appropriate amounts of standards (in terms of Se) to give 500 mg/l of Se-(methyl) 137

155 selenocysteine (SeMC) (C 4 H 9 O 2 NSe), seleno-l-methionine (SeMet) (C 5 H 11 O 2 NSe) and seleno-l-cystine (SeCys 2 ) (C 6 H 12 O 4 N 2 Se 2 ) (see subsection 6.2.1). Standard solutions of SeO 3 2- and SeO 4 2- (for Se) were freshly prepared on the day from their stocks (see subsection 4.2.1). Ion-pairing reagents included tetraethylammonium heptadecafluorooctyl sulphonate (TEA- HFOS), sodium octylsulphonate (SOS), heptafluorobutyric acid (HFBA), tetrabutylammonium hydrogensulphate (TBAHS), trifluoroacetic acid (TFA), potassium heptadecafluorooctylsulphonate (PHFOS). A g portion of either TEA-HFOS or g of SOS, g of TBAHS or g of PHFOS with 10 ml MeOH were dissolved and diluted to 1000 ml given aqueous solutions of 1 mm in 1% methanol, respectively. Solutions of 1.00 ml each of HFBA and TFA and 10 ml MeOH were diluted to 1000 ml to give aqueous solutions of 0.1%. Each was filtered through a disposable 0.45 µm cellulose acetate membrane filter and degassed in an ultrasonic bath prior to use. The enzymes, protease XIV and lipase VII were purchased from Sigma-Aldrich (Sydney, Australia); Tris-(tris(hydroxymethyl)aminomethane)-HCl buffer (75mM, adjusted ph=7.5) were prepared daily as required (see subsection 6.2.1). Commercially available Se-biofortified products Wafer grains biscuits and Laucke Flour (Adelaide, South Australia) were investigated Total Se determination A 0.5 g portion of sample was weighed into PTFE vials and dissolved in a mixture of 2.0 ml of conc. HNO 3 and 2.0 ml of 30% H 2 O 2. After 17 hrs of contact time, total digestion was performed at 170ºC for 15 min at 300 W in a microwave oven. The final solutions were made up to 25 ml in a volumetric flask with MilliQ water. A reference rice sample from NIST SRM 1568a with a certified Se content of 0.38 ±0.04 µg/g was used for evaluation of the accuracy of the procedure. The result of the reference rice sample was 0.36 ±0.07 µg/g, which was in good agreement with the certified value. A digestion blank was also performed. 138

156 8.2.3 Extraction Tris-HCL buffer 15 ml adjusted to ph 7.5 was added to 0.2 g of the sample in a 50 ml centrifuge tube, followed by the addition of 20 mg protease XIV and 10 mg lipase VII. The mixture was sonicated for 2 h, and then incubated in the dark for 12 h at 25ºC. Hydrolysed samples were centrifuged at 4000 rpm for 25 min and the supernatants filtered through a 0.45 µm membrane filter prior to analysis. For the others see section Instrumentation An Agilent 1100 liquid chromatography module was equipped with a Phenomenex 250 mm x 4.6 mm C18 reversed-phase column. A 150 mm x 4.6 mm was also used. Samples of 30 µl were injected using an Agilent 1100 auto-sampler. The mobile phase flow rate and temperature were set at 1.0 ml/min and 25ºC respectively. Instrument conditions were described in subsection Results and discussion Comparing ion-pairing reagents In ion-pairing reverse phase chromatography, an ion-pairing reagent is introduced to improve retention and separation power by increasing solute-stationary phase interactions. The role of the reagents is to provide counter ions for binding anions, cations, or associate with neutral molecules to achieve separation in particularly difficult matrices (Dumont et al., 2006a; Zheng and Kosmus, 2000; Marchante-Gayón et al., 2000). Without any ionpairing reagents, Se compounds are hydrophilic and insufficiently retained and separated on C18 reverse phase columns. A typical chromatogram is shown in Fig The two inorganic species virtually co-eluted (0.13 min difference in retention time). The organic compounds were separated according to their hydrophobicity in the order of SeCys 2, SeMC and SeMet. 139

157 Abundance Time/Minutes--> Figure 8.1 Chromatogram of reversed phase chromatogram on C18 column (250 mm x 4.6 mm, eluent with 1% methanol at 1.0 ml/min and 25ºC. sample, 30µL 1µg/mL in each of SeCys 2 (1), SeMC(2), SeMet(3), Se(IV) (4)and Se(VI)(5)). Octopole H 2 flow rate of 2.5 ml/min. Detection Se at m/z 80. This study systematically investigated the performance of anionic, amphoteric and cationic ion pairing reagents on Se speciation. Six ion pairing agents were selected and mixtures of five selenium species standards (1 µg/ml each) were separated on both a short (150 x 4.6 mm) and long (250 x 4.6 mm) C18 column with 1% aqueous methanol at a flow rate of 1 ml/min. Retention times are listed in Tables 8.1 and

158 Table 8.1 The retention times (min) of five Se species on a short C18 column (150 mm x 4.6 mm), flow rate at 1mL/min Se(VI) Se(IV) SeCys 2 SeMC SeMet SOS HFBA TFA TBAH TEA-HFOS Table 8.2 The retention times (min) of five Se species on a longer C18 column (250 mm x 4.6 mm), flow rate at 1mL/min Se(VI) Se(IV) SeCys 2 SeMC SeMet SOS PHFOS HFBA /8.63 TFA TBAH > TEA-HFOS %MeOH The chromatogram for the short column is presented in Fig. 8.2 using TEA-HFOS. The path length was insufficient to achieve separation of the five species as Se(VI) and Se(IV) co-eluted as did SeCys 2 and SeMC. It shows that it was not possible to resolve all species individually by this approach. This scenario also occurred when using the ion pairing reagents TFA and SOS. 141

159 Abundance Time/Minutes--> Figure 8.2 Chromatogram of 1 mm TEA-HFOS as ion-pairing reagent in 1% methanol as mobile phase on short C18 column (150 mm x 4.6 mm). Other conditions were the same as in Fig. 8.1 The effect of ion pairing reagent was more evident from the chromatograms when a longer column was used (Fig. 8.3(a-f)). Separation and sensitivity of inorganic species were strongly dependent on the ion-pairing reagent selected (Dumont et al., 2006a). 142

160 Figure 8.3a Using 1mM SOS in 1% MeOH as the ion-pairing reagent, ph Abundance Time/Minutes--> Figure 8.3b Using 1 mm PHFOS in 1% MeOH as the ion-pairing reagent, ph 5.70 Abundance Time/Minutes--> 143

161 Figure 8.3c Using 0.1% HFBA in 1% MeOH as the ion-pairing reagent, ph Abundance Time/Minutes--> Figure 8.3d Using 0.1% TFA in 1% MeOH as the ion-pairing reagent, ph 1.93 Abundance Time/Minutes--> 144

162 Figure 8.3e Using 1mM TBAH in 1% MeOH as the ion-pairing reagent, ph 3.08 Abundance Time/Minutes--> Figure 8.3f Using 1mM TEA-HFOS in 1% MeOH as the ion-pairing reagent, ph 5.24 Abundance Time/Minutes--> Figure 8.3 Comparing chromatograms of different ion-pairing reagents and 1% methanol as mobile phase on a 25 cm C18 column, Se detection at m/z 78. Other conditions were the same as in Fig

163 Even on a longer column, negatively charged ion-pairing reagents such as octanesulphonate (SOS) and heptadecafluorooctanesulphonate (HFOS) caused selenite and selenate to remain co-eluting. Since all these compounds are anionic, they cannot form stable ion pairs with selenite and selenate. Therefore, the retention time was short and coelution occurred. Three organic species were separated, however, and eluted in the order SeCys 2, SeMC and SeMet (Fig. 8.3a and 8.3b). Using HFBA and TFA as ion-pairing reagents (Fig. 8.3c and 8.3d), the intensities of the analytes dropped. Peak shapes were also poor and a high baseline (1000 abundance units) was observed with these eluents. Inorganic species were separated from organic selenoamino acids. This is probably owing to hydrogen bonding effects of the anionic analytes with the protonated form of the pairing agents. However, hydrophobic SeMet was retained too strongly on the column with HFBA or TFA as the mobile phase. At the operating ph of 2-2.5, seleno-amino acids were protonated to achieve pairing, which delayed their elution (retention time was about 10 min), similar to published reports (Uden et al., 2004; Kotrebai et al., 2000; Dumont et al., 2004). The origin of the split peak for SeMet in HFBA at 8.3 min is interesting. It is due to the mobile phase ph which matches the pka 1 (2.18) of SeMet, the two species then being partially resolved. With the positively charged ionic reagent, TBAHS (Fig. 8.3e), the inorganic species Se(IV) eluted last. Se(VI) was not eluted due to being strongly retained. Cations as counter ions were clearly good for resolving inorganic Se species by forming strong ion pairs (Pan et al., 2007). To overcome the overly strong affinity for the column and ion-paired Se(VI), the reagent TEA-HFOS was considered (Fig. 8.3f). The positive ion component (TEA) afforded reasonable separation of the inorganic species with good peak shape and satisfactory retention time. The level of the baseline was also improved compared to HFBA and TFA (Fig. 8.3c and 8.3d). The negative ion component (PHFOS) caused excellent separation of the organic species. It also resulted in an improved intensity of species. Compared to the previously described ion-pairing reagents (Fig. 8.3a, 8.3b, 8.3c, 8.3d and 8.3e), the amphoteric ion-pairing reagent TEA-HFOS (CF 3 (CF 2 ) 7 SO - 3 N + (C 2 H 5 ) 4 ) (Fig. 8.3f) showed that species intensities improved and good resolution was achieved for the organic species. The short elution times also fit with the overall aim of the thesis to achieve efficient chromatography (see Chapter 4). The resolution for the inorganic species was about 0.9. Since the peaks were symmetrical, the software dealt satisfactorily with the 146

164 Log Retention Time data for those species too. Therefore, TEA-HFOS was chosen as ion pairing reagent for the further experiments Effect of mobile phase concentration The retention time of the three seleno-amino acids and two inorganic species was sensitive to TEA-HFOS concentration in the range 0.1 to 1.0 mm (Fig. 8.4). At levels between 0.1 to 0.5 mm, retention times reduced, while at concentration ranges between 0.5 to 1.0 mm, the shifts were negligible. For this reason, 1mM TEA-HFOS in 1% MeOH was chosen to separate inorganic from organic species Se(VI) Se(IV) SeCys2 SeMC SeMet TEA-HFOS Concentration(mM) Figure 8.4 The influence of the concentration of TEA-HFOS on the retention of Se species ph effect The distribution of Se species depends on deprotonation constants (pka). Therefore, the mobile phase ph can have an effect on the ionic character of inorganic Se and organic seleno animo acids in aqueous solution, as discussed in In brief, amino acids will be positively charged (ammonium group) at relatively low ph, zwitterionic (ammonium group, carboxylate group) at intermediate ph, and anionic (carboxylate group) at higher ph. In the presence of TEA-HFOS (Fig. 8.5), the retention time of the two inorganic Se species increased between ph 3 and 4. Retention times of the organic species decreased. 147

165 Log Retention Time Se(VI) Se(IV) SeCys2 SeMC SeMet ph Figure 8.5 The influence of mobile phase ph on the retention of Se species. Mobile phase: 1 mm TEA-HFOS and 1% MeOH. Retention times were stable for all species in the ph range 4 to 6, since the organic compounds mainly consisted of zwitterionic ions which were effectively paired with the amphoteric character ion-pairing reagent TEA-HFOS (Zheng et al., 2003; Do et al., 2001; Dumont et al., 2004). The ph range of 4-6 provided a robust condition. Therefore, a ph of 5.0 was selected for further experiments. Fig. 8.5 indicates that the elution order for anions differed from conventional ion-exchange chromatography where cations are excluded from the stationary phase and have little effect on separation in IC. The lowest retention time for the two inorganic species was observed, with prolonged retention times for SeCys 2, SeMC and SeMet which increased in this order Applications With a mobile phase consisting of 1mM TEA-HFOS at ph of 5 and 1% methanol on a C18 column, all species eluted in less than 5 min in the order Se(VI), Se(IV), Secys 2, SeMC, SeMet. All calibrations for mixed standards were linear over a concentration range of 5 to 1000 µg/l with correlation coefficients greater than when a 30 µl sample was injected. Detection limits (S/N=3) ranged from µg/l. 148

166 The proposed method was used to determine selenium species in two biofortified foods. Fig. 8.6 shows a biofortified flour sample using TEA-HFOS as ion-pairing reagents. The retention time of SeMet shifted somewhat possibly owing to the buffer (Tris-HCl) and the sample hydrolysate compared to standards in Milli-Q water. The peak identity was confirmed by spiking the sample solution with SeMet standard. The peaks at minutes were of unknown origin. The background noise in sample had also increased, possibly due to matrix effect. Abundance SeMet Time/Minutes--> Figure 8.6 Chromatogram obtained for biofortified Laucke flour sample using TEA-HFOS as ion-pairing reagents on C18 column (250 mm x 4.6 mm), detected as 80 Se. Other conditions were the same as in Fig. 8.1 Samples were extracted by enzymatic hydrolysis (protease XIV and lipase VII in Tris-HCl buffer in Chapter 6 (Wang et al., 2009) and determined by the current method. The results showed that the major species present in both products was SeMet, which represented between 80-86% of the total Se in biofortified flour and biscuits (Table 8.3). The information obtained proved biofortified products provide a good source of organic selenium. To validate the method, a 30 µg/l mixture of standards were spiked into samples with recoveries for SeMet ranging from 90% to 115% (n=3). 149

167 Table 8.3 Se species in biofortified foods (µg g -1 ) Sample Name SeMet Other Se Compounds Total Se Claimed Se Laucke flour 0.86± ± ± Wafer grains 3.14± ± ± Conclusions In this chapter, a systematic investigation of ion-pairing reagents for separation of selenoamino acids and inorganic selenium (Se) species was conducted. The method chosen was based on an ion-pair reverse phase chromatographic system with a C18 column using inductively coupled plasma mass spectrometry (ICP-MS) and an octopole reaction system (ORS). The results showed selectivity and efficiency of the ion-pairing reagents varied, with the best separation achieved for both inorganic and organic Se compounds within 5 minutes by using 1 mm tetraethylammonium heptadecafluorooctylsulphonate (TEA-HFOS) and 1% methanol as mobile phase at ph 5.0. The method was found to be robust and withstood the matrix effects due to undiluted Tris-HCl buffer proteolytic enzymatic extraction. Different ion-pairing reagents were explored for the separation of five Se species (organic and inorganic compounds). The amphoteric ion-pairing reagent, TEA-HFOS, containing both cationic and anionic ions, proved useful in separating organic and inorganic Se species on ion-pair reverse-phase chromatography. This method was successfully applied to Se biofortified foodstuffs with low Se concentration (between 1-5 µg/g of Se in dry samples) by ICP-MS with ORS. Due to its high sensitivity and matrix tolerance, application to more complicated environmental and biological samples could prove feasible. 150

168 8.5 References B Hymer, C. and Caruso, J. A. (2006) Selenium speciation analysis using inductively coupled plasma-mass spectrometry. Journal of Chromatography A, 1114, Cankur, O., Yathavakilla, S. K. V. and Caruso, J. A. (2006) Selenium speciation in dill (Anethum graveolens L.) by ion pairing reversed phase and cation exchange HPLC with ICP-MS detection. Talanta, 70, Chen, Z. L., Wang, W., Mallavarapu, M. and Naidu, R. (2008) Comparison of no gas and He/H 2 cell modes used for reduction of isobaric interferences in selenium speciation by ion chromatography with inductively coupled plasma mass spectrometry. Spectrochimica Acta Part B: Atomic Spectroscopy, 63, Do, B., Robinet, S., Pradeau, D. and Guyon, F. (2001) Speciation of arsenic and selenium compounds by ion-pair reversed-phase chromatography with electrothermic atomic absorption spectrometry Application of experimental design for chromatographic optimisation. Journal of Chromatography A, 918, Dumont, E., Cremer, K. D., Hulle, M. V., Chéry, C. C., Vanhaecke, F. and Cornelis, R. (2004) Separation and detection of Se-compounds by ion pairing liquid chromatography-microwave assisted hydride generation-atomic fluorescence spectrometry. Journal of Analytical Atomic Spectrometry, 19, Dumont, E., Cremer, K. D., Hulle, M. V., Chéry, C. C., Vanhaecke, F. and Cornelis, R. (2005) Identification of the major selenium compound, Se-Methionine, in three yeast (Saccharomyces cerevisiae) dietary supplements by on-line narrowbore liquid chromatography electrospray tandem mass spectrometry. Journal of Chromatography A, 1071, Dumont, E., Ogra, Y., Vanhaecke, F., Suzuki, K. T. and Cornelis, R. (2006a) Liquid chromatography mass spectrometry (LC MS): a powerful combination for selenium speciation in garlic (Allium sativum). Analytical and Bioanalytical Chemistry, 384, Dumont, E., Vanhaeche, F. and Cornelis, R. (2006b) Selenium speciation from food source to metabolites: a critical review. Analytical and Bioanalytical Chemistry, 385, Ellis, D. R. and Salt, D. E. (2003) Plants, selenium and human health. Current Opinion in Plant Biology, 6,

169 Girling, C. A. (1984) Selenium in agriculture and the environment. Agriculture, Ecosystems and Environment, 11, Goenaga-Infante, H., Sturgeon, R., Turner, J., Hearn, R., Sargent, M., Maxwell, P., Yang, L., Barzev, A., Pedrero, Z. and Cámara, C. (2008) Total selenium and selenomethionine in pharmaceutical yeast tablets: assessment of the state of the art of measurement capabilities through international intercomparison CCQM-P86. Analytical and Bioanalytical Chemistry, 390, Gosetti, F., Frascarolo, P., Polati, S., Medana, C., Gianotti, V., Palma, P., Aigotti, R., Baiocchi, C. and Gennaro, M. C. (2007) Speciation of selenium in diet supplements by HPLC MS/MS methods. Food Chemistry, 105, Huerta, V. D., Reyes, L. H., Marchante-Gayón, J. M., Sánchez, M. L. F. and Sanz-Medel, A. (2003) Total determination and quantitative speciation analysis of selenium in yeast and wheat flour by isotope dilution analysis ICP-MS. Journal of Analytical Atomic Spectrometry, 18, Kápolna, E. and Fodor, P. (2006) Speciation analysis of selenium enriched green onions (Allium fistulosum) by HPLC-ICP-MS. Microchemical Journal, 84, Kotrebai, M., Tyson, J. F., Uden, P. C., Birringer, M. and Block, E. (2000) Selenium speciation in enriched and natural samples by HPLC-ICP-MS and HPLC-ESI-MS with perfluorinated carboxylic acid ion-pairing agents. The Analyst, 125, Larsen, E. H. (1998) Method optimization and quality assurance in speciation analysis using high performance liquid chromatography with detection by inductively coupled plasma mass spectrometry. Spectrochimica Acta Part B: Atomic Spectroscopy, 53, Larsen, E. H., Sloth, J., Hansen, M. and Moesgaard, S. (2003) Selenium speciation and isotope composition in 77 Se-enriched yeast using gradient elution HPLC separation and ICP-dynamic reaction cell-ms. Journal of Analytical Atomic Spectrometry, 18, Marchante-Gayón, J. M., Thomas, C., Feldmann, I. and Jakubowski, N. (2000) Comparison of different nebulisers and chromatographic techniques for the speciation of selenium in nutritional commercial supplements by hexapole collision and reaction cell ICP-MS. Journal of Analytical Atomic Spectrometry, 15, Mester, Z., Willie, S., Yang, L., Sturgeon, R., Caruso, J. A., Fernández, M. L., Fodor, P., Goldschmidt, R. J., Goenaga-Infante, H. and Lobinski, R. (2006) Certification of a 152

170 new selenized yeast reference material (SELM-1) for methionine, selenomethinone and total selenium content and its use in an intercomparison exercise for quantifying these analytes. Analytical and Bioanalytical Chemistry, 385, Pan, F., Tyson, J. F. and Uden, P. C. (2007) Simultaneous speciation of arsenic and selenium in human urine by high-performance liquid chromatography inductively coupled plasma mass spectrometry. Journal of Analytical Atomic Spectrometry, 22, Reyes, L. H., Sanz, F. M., Espílez, P. H., Marchante-Gayón, J. M., Alonso, J. I. G. and Sanz-Medel, A. (2004) Biosynthesis of isotopically enriched selenomethionine: application to its accurate determination in selenium-enriched yeast by isotope dilution analysis-hplc-icp-ms. Journal of Analytical Atomic Spectrometry, 19, Shah, M., Kannamkumarath, S. S., Wuilloud, J. C. A., Wuilloud, R. G. and Caruso, J. A. (2004) Identification and characterization of selenium species in enriched green onion (Allium fistulosum) by HPLC-ICP-MS and ESI-ITMS. Journal of Analytical Atomic Spectrometry, 19, Uden, P. C., Boakye, H. T., Kahakachchi, C. and Tyson, J. F. (2004) Selective detection and identification of Se containing compounds-review and recent developments. Journal of Chromatography A, 1050, Vonderheide, A. P., Wrobel, K., Kannamkumarath, S. S., B'Hymer, C., Montes-Bayon, M., de Leon, C. P. and Caruso, J. A. (2002) Characterization of Selenium Species in Brazil Nuts by HPLC-ICP-MS and ES-MS. Journal of Agricultural and Food Chemistry, 50, Wang, R. Y., Hsu, Y. L., Chang, L. F. and Jiang, S. J. (2007) Speciation analysis of arsenic and selenium compounds in environmental and biological samples by ion chromatography inductively coupled plasma dynamic reaction cell mass spectrometer. Analytica Chimica Acta, 590, Wang, W., Chen, Z. L., Davey, D. E. and Naidu, R. (2009) Extraction of selenium species in pharmaceutical tablets using enzymatic and chemical methods. Microchimica Acta, 165, Warburton, E. and Goenaga-Infante, H. (2007) Methane mixed plasma improved sensitivity of inductively coupled plasma mass spectrometry detection for selenium speciation analysis of wheat-based food. Journal of Analytical Atomic Spectrometry, 22,

171 Wróbel, K., Kannamkumarath, S. S., Caruso, J. A., Wysocka, I. A., Bulska, E., Swia tek, J. and Wierzbicka, M. (2004) HPLC ICP-MS speciation of selenium in enriched onion leaves a potential dietary source of Se-methylselenocysteine. Food Chemistry, 86, Zheng, J. and Kosmus, W. (2000) Retention study of inorganic and organic selenium compounds on a silica-based reversed phase column with mixed ion-pairing reagents. Chromatographia 51, Zheng, J., Shibata, Y. and Furuta, N. (2003) Determination of selenoamino acids using two-dimensional ion-pair reversed phase chromatography with on-line detection by inductively coupled plasma mass spectrometry. Talanta, 59,

172 Chapter 9 Conclusions and Future Trends 155

173 9.1 Conclusions The review of the literature (Chapter 2) demonstrated that selenium species in environmental and biological systems possess both beneficial and toxic effects (Hill, 1997, Infante et al., 2005). With the increasing use of Se-enriched supplements and food, an understanding of health benefits and metabolism of Se species is required (Pyrzynska, 2009). Hence, one area of interest and challenge is elemental speciation in complex matrices. ICP-MS hyphenated with various separation techniques provides an opportunity for improving our understanding of Se speciation (Pedrero and Madrid, 2009). This thesis examined the challenge of Se speciation in environmental and biological systems using liquid chromatographic techniques hyphenated with ICP-MS. The analytical performance of ICP-MS must address some technological questions for the complete speciation of selenium, including organic and inorganic forms at trace levels. The following outcomes were achieved in this study: A successful investigation of sample pre-treatment methods using enzyme hydrolysis and comparison with a sodium dodecyl sulphate (SDS) surfactant method; Development of advanced chromatographic methods, such as anion-exchange, ionexclusion and ion-pairing coupled to plasma mass spectrometry for fast and sensitive Se analysis; Improvement of the detection of 78 Se and 80 Se and, hence seleno species through application of an octopole reaction system (ORS) to the ICP-MS detection process; Application of resulting methods to environmental and biological samples. Sample pre-treatment was an important realisation in this thesis. In particular, extraction of analytes of interest from complex biological matrices was addressed prior to method development using the hyphenated technique, LC-ICP-MS. The soap-like surfactant, sodium dodecyl sulphate (SDS), was initially investigated for extraction of protein-bound Se. Unfortunately the reagent could not completely solubilise Se-containing protein in the chosen samples. Enzymatic attack using protease XIV was found to be more efficient in releasing free seleno amino acids by cleavage of peptide bonds. When protease XIV was combined with lipase VII which is also able to hydrolyse lipids, the dual enzyme system 156

174 contributed to further efficiency in Se release and speciation. Such examples of releasing agents provide a guide to future investigations and the continuing improvement in the extraction of protein bound Se from biological samples. A new type of Agilent anion exchange column based on porous polymethacrylate resin was trialled in Chapter 4 for the first time for Se speciation. An ammonium 2- phosphate/nitrate based mobile phase enabled fast separation of SeO 3 and SeO 2-4. Detection limits for Se(IV) and Se(VI) were also good achieving levels of µg L -1. These experiments were followed in a later chapter (Chapter 7) by additional analyses using the same column to achieve resolution of seleno-amino acids with a range of physicochemical properties in Chapter 7. Furthermore, by using aqueous NH 4 H 2 PO 4 as single mobile phase and optimising separation conditions, both organic and inorganic species were successfully resolved and detected precisely at levels found in pharmaceutical tablets and biofortified foods. Chapter 5 took a different approach as it examined the application of ion-exclusion chromatography to separate selenite and selenate in water, with sulphuric acid performing best over 3 other acids (nitric, acetic and formic acids). As an eluent, sulphuric acid provided good chromatographic resolution, and enhanced detection sensitivity for Se(VI), achieving a detection limit of 0.2 µg L -1. Chapter 8 took up another challenge, namely ion pairing reverse-phase chromatography. The amphoteric ion-pairing reagent, tetraethylammonium heptadecafluorooctylsulphonate (TEA-HFOS), in 1% methanol was employed as mobile phase. It proved useful for separating Se organic and inorganic species. Even though anion exchange chromatography best separated the mixtures containing both organic and inorganic species, ion-pairing reverse phase led to improved performance for low concentration of organic species, as demonstrated with the detection of selenium levels in flour. Chromatographic studies were carried out in parallel to an investigation into the removal of spectral interferences within the MS detection system. Detection of 78 Se and 80 Se in the presence of Ar dimers and other polyatomic interferences can be difficulty. The Ar dimers of mass 80 have been an obstacle to the efficient detection of the most abundant Se isotope, 80 Se. By using an octopole reaction system (ORS) with H 2 gas as reactant the 157

175 Ar + Ar + interference was largely removed. 78 Se detection without reaction gas was of course possible. However, it emerged that the sensitivity for all Se species studied improved nearly two-fold by detecting 80 Se using H 2 gas in the ORS compared with conventional detection of 78 Se. Many products are labelled just with total Se levels. Interest has been raised in examining species and their concentrations. Fertilisation of crops to increase Se intake in the food chain is now practiced, leading to products such as biofortified bread flour and biscuits. On the other hand, unusual species such as Se-methyl-selenocysteine (SeMC) have recently been added to dietary supplements because it is believed that such compounds have health benefits. The methods developed here were applied to the characterisation of a range of selenium species. Samples included contaminated water, nutritional tablets containing seleno-proteins, and biofortified foods. Results from this study indicate that supplements and biofortified foods could be a good source of Se in human diets. The Se content was mainly identified as seleno-amino acids. Of these, SeMC TM tablets containing high SeMC levels may have cancer prevention qualities (Ogra et al., 2005; Infante et al., 2005; Rayman, 2007). In summary, contributions of this study to analytical practice included a highly efficient extraction method, and the development of three separation modes for Se species. ICP-MS combined with optimised ORS technology gave excellent sensitivity for Se detection. The study has led to rapid and precise simultaneous analysis of organic and inorganic selenium species with demonstrated applications to trace levels in water, biofortied food and health supplements. 9.2 Future trends Se speciation studies have increasingly shifted from environmental to biological applications. In particular characterisation of Se molecules in food and supplements are being researched for their efficacy. Further studies will be required to fully establish links between health benefits, metabolism and particular Se species in animals and humans. Sample extraction remains a priority if a true picture of Se speciation is to be gained. Enzymatic processes are increasingly applied to biological systems. Specific enzymes have 158

176 to be developed to completely hydrolyse Se proteins to the seleno-amino acids (Infante et al., 2005). The ICP-MS instrument will continue to play an important role in Se speciation. Presented in this study, the interference from matrix and argon gas was reduced by use of dynamic reaction/collision cell technology. This area will certainly continue to develop and grow more rapidly in the future. The design and development of more efficient interfaces between ICP-MS and separation technologies (HPLC, GC, CE) is an ongoing scientific endeavour (Hettipathirana and Davey, 1998, Wang, 2007). Direct injection nebulisation promises a further advance in detection limits in the future. With increasing interest in more organic and inorganic Se species, multiple columns or column switching may be investigated to see if different chromatography limitations can be overcome (Haddad and Jackson, 1990). Ease of on-line coupling with efficient multiple columns with ICP-MS will be an additional improvement. On the other hand, a single column may still be investigated as a means for simultaneously separating more species in complicated matrices. For example, future LC studies should explore further the multelement capability of the ICP-MS system as Welz hoped in 1998 (Welz, 1998). Examples would include arsenic and selenium speciation in water, soil and sediments, including mixed species, such as arsenite, arsenate, monomethylarsonic acid (MMA), dimethylarsinic acid (DMA), selenite, selenate, SeCys 2 and SeMet (Wang et al., 2007; Shah et al., 2008). Laser ICP-MS can also be used for solid samples. Multidimensional separation approaches combining gel electrophoresis for selenoproteins determination by laser ablation-icp-ms have much more room to develop (Renli Ma, 2004). It should be mentioned that several unidentified Se compounds were also reported previously by various research groups (Shah et al., 2004; Pedrero et al., 2006). In this study, unknown compounds were also observed and need to be identification. Although ICP-MS is powerful for detecting selenium compounds, it does not give any structural information. Therefore, ESI-MS, matrix assisted laser desorption ionisation (MALDI) and quadrupole time of flight ionisation techniques can be employed to provide structural characterisation (De Hoffmann and Stroobant, 2007). Se species with low molecular mass 159

177 have already been identified and quantified. However, it is quite likely that many Se compounds still remain unidentified; in particular high molecular weight selenoproteins have not been fully characterised. In the future, Se containing species will be characterised in more complex matrices (body fluids and tissues). New Se species involved with metabolism in biochemical processes will be described. Examples include characterisation of selenosugar metabolites (Francesconi and Pannier, 2004; Letsiou et al., 2007). Identification of these and selenoproteins may benefit from ESI or surface enhanced laser desorption ionisation (SELDI) MS. Improvements in the accuracy of quantitative determination, the quality of Se speciation and identification of unknown Se species, will follow as species-specific isotope dilution techniques become more commonly combined with ICP-MS (Reyes et al., 2004); (McSheehy et al., 2005). The increased availability of suitable standards and certified reference compounds is necessary to achieve validation of measurements of target Se species (Mester et al., 2006). Continued development of advanced analytical methodologies and strategies applied to chromatographic separation and mass spectrometry for Se research in more complicated biological samples promises an exciting future. 160

178 9.3 References De Hoffmann, E. and Stroobant, V. (2007) Mass spectrometry: principles and applications, Wiley-Interscience. Francesconi, K. A.and Pannier, F. (2004) Selenium Metabolites in Urine: A Critical Overview of Past Work and Current Status. Clinical Chemistry, 50, Haddad, P. R. and Jackson, P. E. (1990) Ion Chromatography: Principles and ApplicationsJournal of Chromatography, Library Vol. 46. Elsevier, Amsterdam. Hettipathirana, T. D. and Davey, D. E. (1998) Evaluation of a microconcentric nebuliser cyclonic spray chamber for flow injection simultaneous multielement inductively coupled plasma optical emission spectrometry. Journal of Analytical Atomic Spectrometry, 13, Hill, S. J. (1997) Speciation of trace metals in the environment. Chemical Society Reviews. 26, Infante, H. G., Hearn, R. and Catterick, T. (2005) Current mass spectrometry strategies for selenium speciation in dietary sources of high-selenium. Analytical and Bioanalytical Chemistry, 382, Letsiou, S., Nischwiz, V., Traar, P., Francesconi, K. A.and Pergantis, S. A. (2007) Determination of selenosugars in crude human urine using high-performance liquid chromatography/atmospheric pressure chemical ionization tandem mass spectrometry. Mass Spectrom., 21, McSheehy, S., Yang, L., Sturgeon, R. and Mester, Z. (2005) Determination of methionine and selenomethionine in selenium-enriched yeast by species-specific isotope dilution with liquid chromatography-mass spectrometry and inductively coupled plasma mass spectrometry detection. Analytical Chemistry, 77, Mester, Z., Willie, S., Yang, L., Sturgeon, R., Caruso, J. A., Fernández, M. L., Fodor, P., Goldschmidt, R. J., Goenaga-Infante, H. and Lobinski, R. (2006) Certification of a new selenized yeast reference material (SELM-1) for methionine, selenomethinone and total selenium content and its use in an intercomparison exercise for quantifying these analytes. Analytical and Bioanalytical Chemistry, 385, Ogra, Y., Ishiwata, K., Iwashita, Y.and Suzuki, K. T. (2005) Simultaneous speciation of selenium and sulfur species in selenized odorless garlic (Allium sativum L. Shiro) and shallot (Allium ascalonicum) by HPLC-inductively coupled plasma-(octopole 161

179 reaction system)-mass spectrometry and electrospray ionization-tandem mass spectrometry. Journal of Chromatography A, 1093, Pedrero, Z., Madrid, Y.and Cámara, C. (2006) Selenium species bioaccessibility in enriched radish (Raphanus sativus): a potential dietary source of selenium. Journal of Agricultural and Food Chemistry, 54, Pedrero, Z. and Madrid, Y. (2009) Novel approaches for selenium speciation in foodstuffs and biological specimens: a review. Analytica Chimica Acta, 634, Pyrzynska, K. (2009) Selenium speciation in enriched vegetables. Food Chemistry, 114, Rayman, M. P. (2007) Selenium in cancer prevention: a review of the evidence and mechanism of action. Proceedings of the Nutrition Society, 64, Renli Ma, Mcleod, C. W., Tomlinson, K. and Poole, R. K. (2004) Speciation of proteinbound trace elements by gel electrophoresis and atomic spectrometry. Electrophoresis, 25, Reyes, L. H., Sanz, F. M., Espílez, P. H., Marchante-Gayón, J. M., Alonso, J. I. G. and Sanz-Medel, A. (2004) Biosynthesis of isotopically enriched selenomethionine: application to its accurate determination in selenium-enriched yeast by isotope dilution analysis-hplc-icp-ms. Journal of Analytical Atomic Spectrometry, 19, Shah, M., Kannamkumarath, S. S., Wuilloud, J. C. A., Wuilloud, R. G. and Caruso, J. A. (2004) Identification and characterization of selenium species in enriched green onion (Allium fistulosum) by HPLC-ICP-MS and ESI-ITMS. Journal of Analytical Atomic Spectrometry, 19, Shah, P., Strezov, V., Prince, K.and Nelson, P. F. (2008) Speciation of As, Cr, Se and Hg under coal fired power station conditions. Fuel, 87, Wang, R. Y., Hsu, Y. L., Chang, L. F. and Jiang, S. J. (2007) Speciation analysis of arsenic and selenium compounds in environmental and biological samples by ion chromatography inductively coupled plasma dynamic reaction cell mass spectrometer. Analytica Chimica Acta, 590, Wang, T. (2007) Liquid Chromatography-Inductively Coupled Plasma Mass Spectrometry (LC-ICP-MS). Journal of Liquid Chromatography & Related Technologies, 30, Welz, B. (1998) Speciation analysis. The future of atomic absorption spectrometry. Journal of Analytical Atomic Spectrometry, 13,

180 Appendix Papers published: Wang, W.H., Chen, Z.L., Davey, D.E., and Naidu, R. (2010) Speciation of selenium in biological samples by ion chromatography with inductively coupled plasma mass spectrometry. Journal of Liquid Chromatography & Related Technologies, review paper, 33, Wang, W.H., Chen, Z.L., Davey, D.E., and Naidu, R. (2009) Extraction of selenium species in pharmaceutical tablets using enzymatic and chemical methods. Microchim Acta, 165, Chen, Z.L., Wang, W.H., Mallavarapu, M., and Naidu, R. (2008) Comparison of no gas and He/H 2 cell modes used for reduction of isobaric interferences in selenium speciation by ion chromatography with inductively coupled plasma mass spectrometry. Spectrochimica Acta, Part B 63, Conference paper: Wang, W.H., David Davey, Ravendra Naidu (2006) The Comparative Performance of Redox Processes during Flow Analysis of Selenium Species by ICP-AES, awarded the best conference poster at the Interact RACI Conference in Perth, Western Australia. (Preliminary work was done on The comparative performance of redox processes during flow analysis of selenium species by ICP-AES. The most significant finding was that thiourea proved to be a far more efficient pre-reduction reagent compared to thioacetic acid, ascorbic acid, hydrochloric acid, potassium iodide and sodium bromide in hydride generation. Since it was not possible to measure continuous output signals using ICP-AES, this particular research could not be extended to include organic Se species. It was therefore deemed appropriate to shift the research focus to the hyphenated technique LC-ICP-MS for Se speciation.) 163

181 164

182 165

183 166

184 167

185 168

186 169

187 170

188 171

189 172

190 173

191 174

192 175

193 176

194 177

195 178

196 179

197 180

198 181

199 182

200 183

201 184

202 185

203 186

204 187

205 188

206 189

207 190

208 191

209 192

210 193

211 194

212 195

213 196

214 197

215 198

216 199

217 200