Combination of cool plasma and collision-reaction ...

Spectrochimica Acta Part B 66 (2011) 389–393

Contents lists available at ScienceDirect

Spectrochimica Acta Part B j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s a b

Analytical note

Combination of cool plasma and collision-reaction interface for correction of polyatomic interferences on copper signals in inductively coupled plasma quadrupole mass spectrometry Lucimar L. Fialho, Catarinie D. Pereira, Joaquim A. Nóbrega ⁎ Grupo de Análise Instrumental Aplicada, Departamento de Química, Universidade Federal de São Carlos, PO Box 676, São Carlos, SP, 13560–970, Brazil

a r t i c l e

i n f o

Available online 8 April 2011 Keywords: ICP-MS Copper Cool plasma Matrix effects Polyatomic interferences

a b s t r a c t Inductively coupled plasma mass spectrometry (ICP-MS) is an important instrumental technique for elemental analysis. However, some elements suffer from spectral interferences caused by ions derived from argon plasma gas and matrix components. The determination of copper isotopes is affected by 40Ar23Na+ and 40 Ar25Mg+. The performance of an ICP-MS with a collision reaction interface (CRI) and cool plasma conditions for correction of spectral interferences was evaluated here. The efficiency of the CRI was studied introducing H2 or He through sampler and skimmer cones. Gas introduction through the sampler cone was ineffective. Complete elimination of spectral interferences was reached when introducing 60 or 80 mL min−1 of H2 in the skimmer cone, but sensitivity losses were as large as 99%. Further, the effect of interferences was checked when the argon plasma was operated under cool plasma conditions. The effects of the applied radiofrequency (0.6, 0.8, 0.9, and 1.0 kW), sampling depth (5.5, 8.5 and 11.5 mm), and dwell time (25 and 50 ms) were studied considering interference reduction and sensitivities. Best conditions were reached at 0.8 kW. Subsequently, both CRI and cool plasma conditions were combined to evaluate their performance on reduction of polyatomic Na and Mg argide interferences. Spectral interferences were eliminated using a CRI with 20 mL min−1 H2 introduced through the skimmer cone, cool plasma conditions at 0.8 kW and sampling depth of 8.5 mm. This work demonstrated the feasibility of combining CRI and cool plasma for circumventing some spectral interferences on Cu determination by ICP-QMS. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Inductively coupled plasma mass spectrometry (ICP-MS) is a powerful technique mainly resulting from its large sample throughput associated with multi-elemental capacity and low limits of detection. Therefore, ICPMS is well suited to the determination of several elements in various matrices such as biological fluids, food, beverages and geological samples [1,2]. However, like other sensitive instrumental techniques, ICP-MS has limitations, and when a quadrupole-ICP-MS (ICP-QMS) is used spectral interferences, such as those caused by doubly charged and polyatomic ions occur, and they cannot be solved because of the typical unitresolution of the quadrupole mass analyzer [2]. The polyatomic interferences occur when ions formed by reactions among precursors in the argon plasma, sample matrix, the ambient air, water or analyte ions have the same mass-to-charge ratio (m/z) as the isotope(s) of the analyte(s). To remove these interferences some alternatives have been researched such as the use of alternative analyte isotopes, mathematical correction equations [3], cool plasma conditions [4–6], collision cell technology ⁎ Corresponding author. Fax: +55 16 3351 8088. E-mail address: [email protected] (J.A. Nóbrega). 0584-8547/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.sab.2011.04.001

(CCT) [1], dynamic reaction cell (DRC) [7], and collision-reaction interface (CRI) [8,9]. Among these strategies, the CRI works by introducing He or H2 as collision/reaction gasses into the plasma through one or both interface cones (sampler and skimmer) [10]. The introduced gas collides and/or reacts with potentially interfering ions, removing them from the plasma before any ions are extracted by the focusing lens system. Generally, ions that would cause spectral interferences are more susceptible to such processes than analyte ions, consequently, interfering ions can be selectively removed from the plasma [8]. An alternative approach for certain elements, applicable to ICP-QMS, is to operate the plasma source under conditions of lower applied power, higher argon gas flow rate and a sampling depth farther than used at normal operating conditions [4,11]. This condition is called “cool plasma”. When cool plasma conditions are used compared to normal conditions ionization of Ar is decreased, leading to a significant reduction in argon-based interferences, which in turn leads to lower limits of detection for many elements. Cool plasma is particularly effective for determining of the elements Fe and Ca and it may be applicable for Cu determination owing to its low first ionization potential (Cu: 7.73 eV). The accurate determination of isotopes 63Cu+ and 65Cu+ (abundance of 69.17 and 30.83%, respectively) using ICP-MS in samples containing

390

L.L. Fialho et al. / Spectrochimica Acta Part B 66 (2011) 389–393

high amounts of alkaline and alkaline earth elements, e.g. sea water, food, clinical, soil and plant samples, may be influenced by the formation of polyatomic ions, such as 40Ar23Na+ and 44Ca18OH+ and 40 Ar25Mg+ and 48Ca16OH+ on the m/z 63 and 65, respectively [12]. Usually these isotopes do not have equations for mathematical correction, therefore other strategies are necessary to remove these interferences. Either conventional chemical strategies may be used to remove these ions (Na+, Mg+ and Ca2+) before the analyses, such as precipitation resulting in Mg(OH)2 formation and ion exchange resin, or these interferences may be solved adopting an instrumental approach, such as high resolution magnetic sector inductively coupled plasma mass spectrometry (HR-ICP-MS) [13–15]. The use of collision reaction cell, cool plasma and their combination has not been studied for Cu determination in ICP-QMS. Therefore the goal of the work here described was to study the performance of a collision reaction interface (CRI), with introduction of H2 or He through sampler or skimmer cones, cool plasma and the combination of both to remove polyatomic interferences caused by 40Ar23Na+ and 40Ar25Mg+ on 63Cu+ and 65Cu+, respectively. This work shows results for cool plasma conditions where only the radiofrequency applied power and sampling depth operating conditions were optimized with or without using a CRI. 2. Materials and methods 2.1. Instrumentation All measurements were performed using an ICP-QMS (820-MS, Varian, Mulgrave, Australia) equipped with a CRI and a double off-axis arrangement in a 90° array for extraction and focusing of the ions. The CRI interface is composed of modified cones (sampler and skimmer), which allow direct introduction of H2 or He gasses to the plasma expansion zone located on vacuum region [16]. Instrumental setting and data acquisition parameters are summarized in Table 1. Employing these conditions, the effect of H2 and He gasses introduced through sampler or skimmer cones of the CRI, the cool plasma and their combination were evaluated for eliminating the interferences caused by 40 Ar23Na+ and 40Ar25Mg+ on 63Cu+ and 65Cu+, respectively.

10, and 50 mg L−1). Solutions of Na or Mg in HNO3 1% v v−1 without Cu were used as analytical blanks. Polypropylene containers were soaked in 10% v v−1 HNO3 for 24 h and rinsed thoroughly with ultrapure water before use. A multi-element tuning solution (SpecSol, Jacareí, SP, Brazil) containing Ba, Be, Ce, Co, In, Mg, Pb, Th and Tl was used to prepare 5 μg L−1 of these elements in HNO3 1% v v−1. This solution was used daily for torch alignment and mass calibration. Liquid argon, and H2 and He gasses were 99.999% pure (White Martins, Sertãozinho, SP, Brazil). 2.3. Sample preparation A closed vessel microwave digestion system (ETHOS-1600, MilestoneMLS, Sorisole, Italy) was used for standard reference material digestion. Sample aliquots of 250 mg of apple leaves (NIST 1515, National Institute of Standards and Technology, Gaithersburg, MD, USA), Dolt-4 (NRC National Research Council Canada, Ottawa, Ontario, Canada) and spinach leaves (NIST, 1570a) were microwave-assisted digested using 5 mL of 7.0 mol L−1 HNO3 and 3 mL of 30% v v−1 H2O2 in closed vessels. The microwave oven heating program was performed in three steps: (1) 10 min at 1000 W and 200 °C; (2) 10 min at 1000 W and 200 °C and, (3) 15 min of cooling. Digested solutions were transferred to volumetric flasks and diluted with water to 13 mL. Further dilution was performed to ensure a maximum of 0.1% m v−1 dissolved solids. 2.4. Evaluation of the collision reaction interface Solutions containing increasing concentrations of Na or Mg (0.5, 1, 5, 10, and 50 mg L−1) were used to evaluate the performance of the CRI. The effect of H2 and He gas flow rates through the sampler (0, 200, 400, 600, 800 and 1000 mL min−1) and skimmer (0, 40, 60 and 80 mL min−1) cones, to the elimination or reduction of polyatomic interferences caused by 40Ar23Na+ and 40Ar25Mg+ was studied. 2.5. Cool plasma The effect of the polyatomic interferences (40Ar23Na+ and Ar25Mg+) when the argon plasma was operated under cool plasma conditions (applied radio-frequency power: 0.6, 0.8, 0.9 and 1.0 kW and sampling depth: 5.5, 8.5 and 11.5 mm) was studied. The dwell time (25 and 50 ms) was optimized considering interference reduction and sensitivities. 40

2.2. Reagents and standard solutions All solutions were prepared in ultrapure water (resistivity higher than 18.2 MΩ cm) obtained from a Milli-Q water purification system (Millipore, Billerica, MA, USA). Purified nitric acid prepared by subboiling quartz distillation DuoPur (Milestone-MLS), stock solutions of 1000 mg L−1 Cu (Qhemis, São Paulo, SP, Brazil) and 1000 mg L−1 Na and Mg (Merck, Darmstadt, Hessen, Germany) were used. Copper reference solutions (10 μg L−1) were prepared in HNO3 1% −1 v v together with increasing concentrations of Na or Mg (0.5, 1.0, 5,

2.6. Combination of cool plasma and collision reaction interface Both CRI and cool plasma conditions were combined to evaluate their performance on reduction of polyatomic interferences. These studies

Table 1 Instrumental operating conditions of ICP-QMSa.

a

Parameter

Standard Mode

RF generator frequency (MHz) RF applied power (kW) Plasma gas flow rate (L min−1) Nebulizer gas flow rate (L min−1) Auxiliary gas flow rate (L min−1) Sheath gas flow rate (L min−1) Dwell time (ms) Sampler (mL min−1) Skimmer (mL min−1) Nebulizer Spray chamber Isotopes monitored Sampling depth (mm)

27 1.4 18 0.95 1.8 0.15 10 0 0 Seaspray Scott type 63 Cu+, 65Cu+ 5.5

Cool plasma

CRI

CRI + cool plasma

10 200, 400, 600, 800 and 1000 40, 60, 80 and 100

50

5.5

5.5, 8.5 and 11.5

0.6, 0.8, 0.9 and 1.0

50

5.5, 8.5 and 11.5

A specific parameter is shown only if it was changed compared to the standard mode.

20, 40 and 60


0.0 mg L-1 Na 1.0 mg L-1 Na 5.0 mg L-1 Na -1 10.0 mg L Na -1 50.0 mg L Na

Signal intensity (cps)

2.1x105 2.0x105 1.8x105 1.6x105 4.0x104 3.0x104 2.0x104

105


a

391

10 µg L-1 Cu + 50 mg L-1 Mg 10 µg L-1 Cu + 10 mg L-1 Mg 10 µg L-1 Cu + 5 mg L-1 Mg 10 µg L-1 Cu + 1 mg L-1 Mg 10 µg L-1 Cu + 0.5 mg L-1 Mg

104

103

102

1.0x104 0.0 0

200

400

600

800

101

1000

0

10

20

-1


50

60

H2flow rate (mL min ) Fig. 3. Effect of H2 gas introduced through the skimmer cone of CRI and cool plasma conditions on net analytical signals of copper solutions (isotope 65Cu) in different Mg media.

7.2x104 0.0 mg L-1 Na 1.0 mg L-1 Na 5.0 mg L-1 Na -1 10.0 mg L Na -1 50.0 mg L Na

6.8x104

6.4x104 1.0x104 3

8.0x10

6.0x103 4.0x103 2.0x103 0.0 0

20

40

60

80

100

H2flow rate (mL min-1) Fig. 1. Effect of H2 gas flow rates introduced through the sampler (a) and skimmer (b) cones of the CRI on the minimization of isobaric interferences on m/z 63.

were carried out using applied radio frequency power of 0.8 kW, sampling depth of 5.5, 8.5 and 11.5 mm and H2 gas flow rates of 20, 40, and 60 mL min−1.

3. Results and discussion 3.1. Performance of the collision reaction interface The introduction of H2 though the sampler cone was effective to reduce polyatomic interferences caused by 40Ar25Mg+ only at flow rates above 1000 mL min−1. A decrease around 96% of the background signal compared to the signal obtained without using the CRI was observed. However, this behavior was not observed for the interference caused by 40 Ar23Na+ (Fig. 1a). The introduction of H2 gas through the skimmer cone was more effective than through the sampler cone for interference reduction, and a significantly lower flow rate was needed when compared with the sampler cone performance (Fig. 1b). On the other hand, sensitivity losses for 63Cu+ as large as 99.7 and 99.9% were observed when using H2 flow rates of 60 or 80 mL min−1, respectively (Fig. 1b). Similar losses were observed for 65Cu+ signals. Experiments were carried out by introduction of either H2 or He gasses. However, the lowest values of background equivalent concentration (BEC) were obtained for 63Cu+ and 65Cu+ isotopes when using H2 gas. Consequently, further experiments were made using H2 gas introduction through the skimmer cone. The best performance of the H2 gas to control interferences may indicate the predominance of reactive processes for destruction of ArNa+

1.0x105

3.0x105

5

2.0x10


Blank solution (Na) 10 µg L-1 Cu + 50 mg L-1 Na Blank solution (Mg) 10 µg L-1 Cu + 50 mg L-1 Mg

2.5x105


40 -1

H2flow rate (mL min )

b

30

1.5x105 1.0x105

Na Na Na Na Na

1.0x104

1.0x103

1.0x102

5.0x104 1.0x101

0.0 0.6

0.7

0.8

0.9

1.0

Applied RF power (kW) Fig. 2. Effect of applied power on m/z 63 for solutions containing Na and on m/z 65 for solutions containing Mg.

0

10

20

30

40

50

60

H2 gas flow rate (mL min-1) Fig. 4. Effect of H2 gas introduced through the skimmer cone of CRI and cool plasma conditions on net analytical signals of copper solutions (isotope 63Cu) in different Na media.

392


Table 2 Slopes of analytical calibration curves using different ICP-MS operational conditions for controlling interferences. Solutions were prepared using the standard additions method for apple leave digestate (CRM NIST 1515). Isotope

63 65

Cu+ Cu+

Sensitivity (cps L μg−1) Standard mode

80 mL min−1 H2

Cool plasma (8.5 mm)

Cool plasma (8.5 mm) + 20 mL min−1 H2

6435 2638

18.2 7.9

252 86.7

53.6 24.3

and ArMg+ [16]. However, reactional and collisional processes did not only occur between gas and interference species, but addition of H2 may also cause a defocusing effect of the analyte ions [16] which may explain the pronounced losses of sensitivity. Considering the critical loss of sensitivity to analyte, an alternative to reduce or eliminate interferences based on the use of cool plasma was investigated. 3.2. Cool plasma The ionization of argon decreases by decreasing the plasma applied power, and results in a significant reduction in the occurrence of argonbased interferences, such as ArNa+ and ArMg+. The cool plasma studies were performed employing the same solutions previously used. The effect of the applied power on the interference signal intensities for ArMg+ was lower than that observed for ArNa+. This behavior can be explained by different first ionization energies of Na (1.54 eV) and Mg (7.65 eV). The best compromise conditions to decrease background signals without pronounced losses of sensitivity were reached at 0.8 kW of applied power (Fig. 2). A study varying the dwell time was made (25 ms and 50 ms). Using 50 ms dwell time the signal intensities were about three-fold higher than those obtained with 25 ms dwell time. 3.3. Evaluation of the combination cool plasma and collision reaction interface Based on these experiments, the combination of CRI and cool plasma conditions was evaluated as an alternative for reducing interferences. Using the CRI at lower H2 gas flow rates introduced through the skimmer

cone should lead to a reduction of interferences without accentuated sensitivity losses, and eventually these reduced flow rates may be more effective when working at cool conditions compared to normal plasma conditions. Spectral interference caused by 40Ar25Mg+ was eliminated using CRI with 20 mL min−1 of H2 introduced through the skimmer cone and cool plasma conditions at 0.8 kW applied power (Fig. 3). The combination of cool plasma and CRI showed best compromise between sensitivity and minimization of the spectral interference caused by ArMg+. However, a similar effect was not observed for ArNa+. This interference process was corrected only when introducing a H2 flow rate at 60 mL min−1 through the skimmer cone, but again severe losses of analytical signal were observed (Fig. 4). For the combination of cool plasma conditions and a flow rate of 60 mL min−1 H2 through the skimmer, the analytical signal losses were higher than those observed when using only the CRI. The losses of sensitivities expressed as slopes of analytical calibration curves obtained using different strategies for controlling interferences are shown in Table 2. All strategies caused accentuated losses of sensitivity and the best compromise between control of interferences and sensitivity was reached by combining CRI at lower flow rates and cool plasma conditions. Hydrogen gas introduced through the skimmer cone promotes several reactive processes, such as charge transfer, electron-ion reaction and ion-molecule interactions, resulting in the destruction of polyatomic species and formation of new ones [16]. However, the worst effects observed are related to the defocalization of analyte ions [10] and consequent implications on sensitivity. Based on the optimized conditions using synthetic solutions, the determination of copper was studied in three certified reference materials, i.e. NIST 1515 (apple leaves), NIST 1570a (spinach leaves) and Dolt-4 (dogfish). The ICP-MS instrument was operated using the combination of CRI at 20 mL min−1 H2 introduced through the skimmer cone and cool plasma conditions (0.8 kW of applied power, 5.5, 8.5 and 11.5 mm of sampling depth and 50 ms of dwell time) The determination of Cu in apple leaves was affected by positive errors when external calibration was used. This CRM has elevated concentrations of calcium when compared to sodium and magnesium contents (100 mg L−1 of Ca, 18 mg L−1 of Mg and 0.18 mg L−1 Na in the digestate solution according to the sample preparation protocol adopted). A hypothesis for explaining positive bias may be based on the generation of polyatomic species 44Ca18OH+ and 48Ca16OH+ at m/z 63 and 65, respectively, formed when H2 was introduced through the skimmer. To

Table 3 Determination of copper (m/z 63, mg kg−1, mean and standard deviations, n = 3) in three certified reference materials (CRM) using different ICP-QMS operational conditions for controlling interferences and respective recoveries values. 63

Cu+

CRM

Certified value (mg kg−1)

Observed Value Standard mode

CRI 60 mL min−1

NIST 1515 5.6 ± 0.2 2.9 ± 0.2 2.2 ± 0.3 Recovery 52 ± 4 39 ± 5 (%) NIST 12.2 ± 0.6 9.1 ± 0.3 6.7 ± 0.8 1570a Recovery 75 ± 2 55 ± 7 (%) Dolt-4 31.2 ± 1.1 23.3 ± 2.9 37.7 ± 1.0 Recovery 77 ± 9 121 ± 3 (%) a—0.8 kW and 5.5 mm. b—0.8 kW and 8.5 mm. c—0.8 kW and 11.5 mm. d—0.8 kW, 5.5 mm and 20 mLmin−1 H2. e—0.8 kW, 8.5 mm and 20 mLmin−1 H2. f—0.8 kW, 11.5 mm and 20 mL min−1 H2. nd—not determined.

CRI 80 mL min−1

Cool plasmaa Cool plasmab Cool plasmac Cool plasma + CRId Cool plasma + CRIe Cool plasma + CRIf

1.9 ± 0.5 34 ± 9

5.5 ± 0.7 95 ± 13

3.7 ± 0.5 65 ± 9

3.1 ± 0.7 55 ± 12

4.7 ± 0.3 83 ± 4

4.8 ± 0.4 85 ± 7

4.6 ± 0.1 81 ± 2

6.4 ± 0.5

20.8 ± 1.9

26.6 ± 1.4

17.4 ± 0.4

17.9 ± 1.3

13.9 ± 1.3

14.5 ± 0.3

52 ± 4

171 ± 15

218 ± 12

142 ± 3

147 ± 10

114 ± 10

119 ± 3

nd nd

22.8 ± 1.8 73 ± 6

17.0 ± 1.3 54 ± 4

17.9 ± 1.7 57 ± 6

26.9 ± 0.2 86 ± 1

23.3 ± 1.1 75 ± 3

23.1 ± 2.3 74 ± 7


correct this matrix effect it was used standard addition in CRMs which contain different concentrations of Na, Mg and Ca. The best results for both isotopes were obtained when the combination of cool plasma (0.8 kW, 8.5 mm and 50 ms dwell time) and CRI was used (Table 3, results are shown for the most abundant isotope). Table 3 shows that best recoveries were obtained when 8.5 mm sampling depth was used. This can be explained because positioning the sampling orifice relatively far from the load coil combined with use of low applied power has improved cool plasma conditions. This strategy was efficient to solve polyatomic interferences even in a matrix with elevated concentrations of interferences as spinach leaves diluted digestate (Ca 53 mg L−1, Mg 30 mg L−1 and Na 63 mg L−1). However, for Dolt-4 best recoveries were obtained using a sampling depth of 5.5 mm. The diluted digestate solution of this CRM contains low concentration of interferences (0.7 mg L−1 of Ca, 15 mg L−1 of Mg and 7 mg L−1 of Na), so only by using a low applied power 0.8 kW was enough to solve polyatomic interferences. 4. Conclusion This work showed that among all strategies studied for correcting polyatomic interferences caused by ArNa+ and ArMg+ in the determination of Cu isotopes, the combination of cool plasma and CRI has shown the best compromise between both minimum losses of sensitivity and minimum background signals. This interface must be carefully used because it shows a major effect on sensitivity and new interferences may be formed and transferred to the quadrupole mass spectrometer since there is no other device for mass filtering or energy discrimination. It may be concluded that the combination of cool plasma and CRI together with the standard addition method is effective for correcting matrix effects on determination of Cu by ICP-QMS. The work here discussed seems to be a preliminary study for inducing further research works combining cool plasma and CRI or any other strategy based on collision and reaction devices. Acknowledgements The authors express their gratitude to Fundação de Amparo à Pesquisa do Estado de São Paulo for research grant (Process 2006/

393

59083-9). All authors also would like to thank the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for fellowships.

References [1] V. Dufailly, L.N. Laurent, T. Guérin, Determination of chromium, iron and selenium in foodstuffs of animal origin by collision cell technology, inductively coupled plasma mass spectrometry (ICP-MS), after closed vessel microwave digestion, Anal. Chim. Acta 565 (2006) 214–221. [2] D. Pick, M. Leiterer, J.W. Einax, Reduction of polyatomic interferences in biological material using dynamic reaction cell ICP-MS, Microchem. J. 95 (2010) 315–319. [3] M. Leiterer, J.W. Einax, C. Löser, A. Vetter, Trace analysis of metals in plant samples with inductively coupled plasma—mass spectrometry Fresenius' J, Anal. Chem. 359 (1997) 423–426. [4] S.D. Tanner, Characterization of ionization and matrix suppression in inductively coupled ‘cold’ plasma mass spectrometry, J. Anal. At. Spectrom. 10 (1995) 905–921. [5] K.Y. Patterson, C. Veillon, D. Hill, P.B. Moser-Veillon, T.C. O'Haver, Measurement of calcium stable isotope tracers using cool plasma ICP-MS, J. Anal. At. Spectrom. 14 (1999) 1673–1677. [6] J. Fietzke, A. Eisenhauer, N. Gussone, B. Bock, V. Liebetrau, Th.F. Nägler, H.J. Spero, J. Bijma, C. Dullo, Direct measurement of 44Ca/40Ca ratios by MC–ICP–MS using the cool plasma technique, Chem. Geol. 206 (2004) 11–20. [7] S. D'Ilio, F. Petrucci, M. D'Amato, M. Di Gregorio, O. Senofonte, N. Violante, Method validation for determination of arsenic, cadmium, chromium and lead in milk by means of dynamic reaction cell inductively coupled plasma mass spectrometry, Anal. Chim. Acta 624 (2008) 59–67. [8] I. Kalinitchenko, X.D. Wang, B. Sturman, Simple and effective control of spectral overlap interferences in ICP-MS, Spectroscopy (2008) 38–46 Suppl. [9] V. Romarís-Hortas, A. Moreda-Piñeiro, P. Bermejo-Barrera, Microwave assisted extraction of iodine and bromine from edible seaweed for inductively coupled plasma-mass spectrometry determination, Talanta 79 (2009) 947–952. [10] C.D. Pereira, E.E. Garcia, F.V. da Silva, A.R.A. Nogueira, J.A. Nóbrega, Behaviour of arsenic and selenium in an ICP-QMS with collision and reaction interface, J. Anal. At. Spectrom. 25 (2010) 1763–1768. [11] S.J. Jiang, R.S. Houk, M.A. Stevens, Alleviation of overlap interferences for determination of potassium isotope ratios by Inductively Coupled Plasma Mass Spectrometry, Anal. Chem. 60 (1988) 1218–1221. [12] V. Chrastný, M. Komárek, Copper determination using ICP-MS with hexapole collision cell, Chem. Pap. 63 (2009) 512–519. [13] L. Leentvaar, M. Rebhun, Effect of magnesium and calcium precipitation on coagulation flocculation with lime, Water Res. 16 (1982) 655–662. [14] E.W.E. Strelow, F. Von, S. Toerien, C.H.S.W Weinert, Accurate determination of traces sodium and potassium in rocks by ion exchange followed by atomic absorption spectroscopy, Anal. Chim. Acta 50 (1970) 399–405. [15] A. Milne, W. Landing, M. Bizimis, P. Morton, Determination of Mn, Fe, Co, Ni, Cu, Zn, Cd and Pb in seawater using high resolution magnetic sector inductively coupled mass spectrometry (HRICPMS), Anal. Chim. Acta 665 (2010) 200–207. [16] X.D. Wang, I. Kalinitchenko, Principles and performance of the Collision Reaction Interface for the Varian 820-MS, Varian,Inc. Advantage Note 1, 2005.