摘要:

Influence of plasma gas composition on analyte ionization in furnace atomization plasma emission spectrometry† Fusheng Sun‡ and Ralph E. Sturgeon Institute for National Measurement Standards, National Research Council of Canada, Ottawa, Ontario, Canada, K1A 0R6 Received 8th January 1999, Accepted 29th March 1999 Temporally and spatially integrated measurements of the intensities of a number of atomic and ionic lines of several elements (Cr, Mn, Mg, Co, Fe, Cd and Zn) were used to calculate their degree of ionization in He, Ar and He–Ar mixed gas plasmas in a radiofrequency (rf ) furnace atomization plasma emission spectrometry (FAPES) source.The eVects of rf power and plasma gas flow rate were also investigated. The emission intensities of ionic and atomic lines and the degree of analyte ionization are enhanced as Ar is added to the He plasma. Maximum degrees of ionization of 99, 80, 81, 72, 76, 12 and 38% for Cr, Mn, Mg, Co, Fe, Cd and Zn, respectively, were obtained in pure Ar.The mechanism of analyte ionization and excitation is discussed. gas composition of the plasma discharge.12–20 Collisional Introduction exchange of internal energy amongst various levels of each Furnace atomization plasma emission spectrometry (FAPES) plasma gas species subsequently plays a significant role in provides a new approach for multi-element ultra-trace analysis determining the extent of excitation and ionization of analytes, based on a combined source.A heated graphite tube serves as primarily as a result of charge exchange and Penning an atomizer and an atmospheric-pressure radiofrequency (rf ) ionization. rare gas plasma supported on a graphite electrode centered In this study, the influence of plasma gas composition, along the axis of the tube provides the excitation medium.1–11 including He, Ar and their mixtures, on the degree of ionization Owing to the high eYciencies of vaporization, atomization of a number of analyte elements was investigated.The eVects of and excitation,5 only a few microliters of analyte solution are rf power and plasma gas flow rate were also examined and needed, resulting in detection limits comparable to those of mechanisms for ionization and excitation of analytes are graphite furnace atomic absorption spectrometry.4,5 Although proposed. the intensities are weaker than their corresponding resonance atomic lines, signals emitted by the ionic lines of several Experimental elements have been measured in the FAPES source, indicative of analyte ionization.5 Indeed, FAPES also serves as an ion Apparatus source for elemental mass spectrometry,9 as the plasma con- The FAPES workhead has been fully described in earlier tains suYciently energetic species (metastable He, He ions and papers2,3,5 and will be only briefly summarized here.A Model vibrationally excited molecular impurities) to ionize eYciently HGA-2200 graphite furnace workhead (Perkin-Elmer, elements having ionization potentials as high as 10.45 eV Norwalk, CT, USA) was used as the atomizer.A Model RF (iodine). 10L 40 MHz rf generator (RF Power Products, Voorhees, Recently, temporally and spatially integrated intensities for NJ, USA) provides 20–100 W of power to a 1 mm a number of atomic and ionic lines emitted by several thermodiameter ×30 mm long pyrolytic graphite-coated graphite elec- metric species introduced into an rf He FAPES source were trode (RingsdorV Werke, Bonn, Germany) which is centered used to calculate the degree (a) of analyte ionization.11 For along the axis of a non-grooved pyrolytic graphite-coated elements having ionization potentials below 8 eV, a is >50% graphite tube (Perkin-Elmer, U� berlingen, Germany).The rf and generally increases with increasing rf power. At 50W power supply was connected to the workhead through an forward power, a varied from 94% for Cr to 7% for Zn. An AM-5 RF antenna tuner (RF Power Products) to permit ionization temperature of 6025±350 K, derived from the use impedance matching. A 0.5 m monochromator (Varian, of several thermometric species, characterized the ionization Springvale, Australia) equipped with a Hamamatsu R446 process. It is necessary to gain additional insight into the PMT operated at -550 V was calibrated for wavelength ionization phenomena in FAPES if it is to be further developed response in the 200–700 nm range.6 The FAPES source was as an eYcient ion source for elemental MS or as a source for imaged on to the spectrometer entrance slit with an f/2 lens.optical emission (atomic or ionic). An IBM AT PC with in-house software written in Turbo One means of improving the sensitivity of FAPES for MS Pascal version 4.0 (Borland International, Scotts Valley, CA, application is to enhance a. It is recognized that the emission USA) was used to collect and manipulate data. intensities from singly ionized species in a Grimm-type glow discharge source are drastically changed by varying the support Reagents Stock standard solutions were prepared by dissolution of the †Canadian Crown Copyright.high-purity metals (Cd, Co, Cr, Fe, Mg, Mn and Zn) in ultra- ‡Permanent address: Suzhou Institute of Urban Construction and Environmental Protection, Postcode 215011, Suzhou, China. pure acids. Working standard solutions were obtained by J. Anal. At. Spectrom., 1999, 14, 901–912 901Table 1 Furnace and plasma operating conditions Mass/ng Temperature/ °C (time/s) Element Atomic line Ionic line Dry Char Atomize Cr 200 2a 110 (10) 500 (10) 2700 (5), MPc Mn 1 1 110 (10) 500 (10) 2600 (5), MP Mg 0.05 0.05 110 (10) 500 (10) 2600 (5) Co 100 50 110 (10) 500 (10) 2700 (5), MP Fe 10 10 110 (10) 500 (10) 2600 (5), MP Cd 2 5 —b Room temp.(10) 1800 (5) Zn 0.4 2 — Room temp. (10) 2300 (5) aMass was 20 ng for pure He. bNo dry step. cMaximum furnace power.dilution of the stock standard solutions with high-purity, emission intensity of the excited atom or ion of the element, Z(T) is the partition function (calculated from the tables of distilled, de-ionized water (DDW) acidified to 1% v/v with quartz sub-boiling distilled nitric acid. Industrial grade He de Galan et al.28), Ei and Ea are the excitation energies (eV ) for the atom and ion, respectively, k is Boltzmann’s constant and Ar (99.995% pure) (Air Products, Mississauga, ON, Canada) were used as the plasma gases.(eV K-1) and T is the common excitation temperature (K) characterizing the relative population levels of both atomic and ionic species. Procedure The principle of line selection for use with eqn. (1) has been Internal plasma gas and external Ar sheath gas flow rates of described in earlier work.11 A temperature of 5000 K was 1100 and 1000 ml min-1, respectively, were selected for all adopted for use in eqn. (1) as detailed earlier.11 The primary studies and regulated with use of ball float meters.All wave- experimental uncertainty arises from the precision of measurelengths were set by using the appropriate hollow cathode ment of the emitted intensity of the spectral lines because all lamps. A 25 mm entrance slit on the monochromator, corre- other terms in eqn. (1) are constants. The major absolute sponding to a nominal spectral bandpass of 0.08 nm, was used uncertainties in the derived a values arise from the Einstein for all elements except Fe, for which a 15 mm slit (0.05 nm transition probability terms, whose accuracies are listed in bandpass) was used.A 2 ml volume of working standard Table 2. solution for a given element was manually pipetted into the furnace using an adjustable pipette. Contrary to the ‘normal’ Degree of ionization of analyte sequence of operation,2–7 the rf power was then immediately applied and the plasma ignited (the pure Ar plasma was EVect of plasma gas composition on analyte signal intensity.diYcult to ignite and, as a consequence, a plasma was first Fig. 1 illustrates typical blank corrected emission transients established in He and then converted to Ar by gradually obtained during the atomization of Mg into a 50 W plasma increasing the Ar–He mixture). The furnace program was then under the experimental conditions summariz in Table 1. started (programs are summarized in Table 1), the emission Increasing the fraction of Ar in the He plasma enhances the transient recorded and the plasma extinguished. The mass of signal intensity from both atomic and ionic lines owing to, each element introduced into the furnace, summarized in amongst other factors, decreased diVusional loss rates.The Table 1, must be such as to result in emission intensities within emission pulses are higher and broader and the peak time the linear response of the system, thereby ensuring that an occurs later as the Ar fraction is increased.The reason for the optically thin plasma exists, i.e., self-absorption is negligible ‘sharper’ response in emission from the atomic lines of Mg for all lines. The eVect of forward power was investigated over [Fig. 1(A)] compared with the ionic lines [Fig. 1(B)] is not the range 20–100 W for several elements. Although the known at this time. Integrated emission intensities of the ionic reflected power could be tuned to less than 1 W during the and atomic lines are 13- and 6-fold greater in the pure Ar char step, it increased sharply during the atomization step plasma than in the pure He plasma, respectively.Trends in when the temperature exceeded 1800 °C (the matching network the data for the atomic and ionic lines of the other elements was not permitted to auto-tune). Measurements for each line studied are similar to those for Mg and Fig. 2 and 3 illustrate were repeated at least three times. Blank signals were obtained the eVects of plasma gas composition on the relative enhancefor each analyte line, whereby 2 ml of 1% nitric acid in DDW ment of integrated intensities for the atomic and ionic lines of was injected into the furnace and subjected to the same each element.For ease of comparison, intensities are norprocedure as stated above. malized to unity when pure He is used. Relative intensity The physical characteristics of the analyte atom and ion enhancements increase with the increase in percentage of Ar lines are listed in Table 2; spectroscopic data were obtained in the He–Ar mixture and reach their highest values in pure from a number of sources.21–27 Ar, i.e., 4–29- and 1–7-fold higher for ionic and atomic lines, respectively, compared with pure He.Selective enhancement of the intensities of the ionic lines occurs as the composition Results and discussion of the plasma shifts to being richer in Ar and the magnitude of The degree of ionization, a, was estimated from the ratio of such increases appears element dependent.the number density of excited atoms and ions of a given element using the following equations:11 EVect of plasma gas composition on a. The degree of ionization [calculated from eqns. (1) and (2)] for each analyte ni/na=[Il Z(T)/gA]i [gA/Il Z(T)]a exp[(Ei-Ea)]/kT (1) under the condition of constant total internal plasma gas flow a=ni/(ni+na)=(ni/na)/[1+(ni/na)] (2) rate (1100 ml min-1) is summarized in Table 3.The data presented for Ar at 50 W provide an estimate of the typical where na and ni are the number densities of the excited atoms and ions, respectively, A is the Einstein transition probability reproducibility arising from the precision of replicate measurement of the emission intensities of the atomic and ionic lines (s-1), g is the statistical weight of the excited level, I is the 902 J. Anal. At. Spectrom., 1999, 14, 901–912Table 2 Selected line parameters Species l/nm Energya/eV Spectroscopic terms gA×108/s-1 Uncertaintyb Cr I 259.2 5.81 4p u 5P03–4s2 a 5D4 4.55 C Cr II 283.6 12.69 (6.77) 4p z 6F011/2–4s a 5D9/2 24 C Mn I 279.5 4.43 4p y 6P07/2–4s2 a 6S5/2 29.6 C Mn II 259.4 12.21 (7.43) 4p z 7P03–4s a 7S3 18.2 C Mg I 285.2 4.35 3p 1P01–3s2 1S0 15.9 D Mg II 279.6 12.07 (7.64) 3p 2P03/2–3s 2S1/2 10.4 C Co I 235.8 5.48 4p w 4F05/2–4s2 a 4F3/2 0.84 D Co II 237.9 13.50 (7.87) 4p z 5F04–4s a 5F5 17.1 B Fe I 254.1 4.99 4p x 5D02–4s2 a 5D1 4.6 C Fe II 259.9 12.67 (7.90) 4p z 6D09/2–4s a 6D9/2 22 B Cd I 228.8 5.42 5p 1P01–5s 1S0 12 C Cd II 226.5 14.46 (8.99) 5p 2P01/2–5s 2S1/2 99 C Zn I 213.9 5.80 4p 1P01–4s 1S0 19 B Zn II 202.6 15.51 (9.39) 4p 2P03/2–4s 2S1/2 21 C aAtom excitation energy and the sum of first ionization potential (the values in parentheses) and ion excitation energy for the atomic and ionic lines, respectively.bEstimated uncertainties: B<10%; C<25%; D<50%.25 Fig. 3 EVect of plasma gas composition on emission intensity of ionic lines for a 50 W plasma. 0, Cr II 283.6 nm; +, Mn II 259.4 nm; #, Fig. 1 EVect of plasma gas composition on emission transients from Mg II 279.6 nm; (, Co II 237.9 nm; -, Fe II 259.9 nm; 2, Cd II Mg atomized at 2600 °C into a 50W plasma. (A) Mg I at 285.2 nm, 226.5 nm; L, Zn II 202.6 nm. 0.05 ng; (B) Mg II at 279.6 nm, 0.05 ng. Traces: (a) He flow of 1100 ml min-1; (b) He flow of 650 and Ar flow of 450 ml min-1; (c) He flow of 240 and Ar flow of 860 ml min-1; (d) Ar flow of Table 3 EVect of plasma gas composition on aa 1100 ml min-1.He5Ar/ml min-1 Element 110050 6505450 2405860 051100b Cr 92 99 100 100±6 Mg 57 79 80 80±4 Mn 61 79 80 81±5 Co 41 59 64 72±7 Fe 52 59 63 76±8 Cd 3.0 7.0 9.0 12±1 Zn 11 19 33 38±3 aa given in %. bMean and standard deviation (n=3); 50 W plasma. suspect, as critically evaluated data for the atomic and ionic lines used were not available in the NIST website database.24 It is likely for this reason that the degree of ionization of Cd is substantially lower than that of Zn despite its lower ionization potential.The data in Table 3 suggest that, as the percentage of Ar in the He–Ar mixture is increased, the degree Fig. 2 EVect of plasma gas composition on emission intensity of atomic lines for a 50 W plasma. 0, Cr I 259.2 nm; +, Mn I 279.5 nm; of ionization increases. Fig. 4 illustrates the eVect of plasma #, Mg I 285.2 nm; (, Co I 235.8 nm; -, Fe I 254.1 nm; 2, Cd I gas composition on the relative enhancement of a at 50W 228.8 nm; L, Zn I 213.9 nm.power where all values have been normalized to unity for a pure He plasma. Data for Cd and Zn are especially remarkable (4.4- and 3.3-fold enhancements, respectively) and that for Cd of each analyte. Similar precision was obtained with other plasma gas compositions. The absolute accuracy in the degree will be discussed at greater length later (the characteristics for Zn appear to be similar).It was speculated earlier11 that the of ionization is primarily dependent on the uncertainty in the Einstein transition probabilities (summarized in Table 2). use of Ar in the He plasma may lead to a reduction in a; it is clear from the present studies that the eVect noted earlier11 Transition probabilities for Cd and Zn may be particularly J. Anal. At. Spectrom., 1999, 14, 901–912 903EVect of plasma gas flow rate on a. The influence of the gas flow rate on the signal intensities and calculated a values for all the elements atomized into a 50W Ar plasma was evaluated.As expected, residence times are enhanced at lower flow rates, maintaining the excited analyte atoms and ions for a longer time in the plasma, thereby increasing both the integrated and peak intensities. Consistent with data reported earlier11 is the slight enhancement of a at the higher gas flow rates; this is probably occurring as a consequence of the decreased concentration of molecular impurities within the plasma (such as CO, NO, NH, N2, O2 and OH arising from ingress of ambient air) which serve to quench collisional excitation/ionization processes.EVect of rf power on a. It has been reported earlier that the degree of ionization of analytes generally increases as the forward rf power increases.11 Le Blanc30 also confirmed this, suggesting that complete ionization of Mg could be achieved in a FAPES source at a forward power as low as 225 W.In Fig. 4 EVect of plasma gas composition on degree of analyte ionization for a 50 W plasma. 0, Cr; +, Mn; #, Mg; (, Co; -, Fe; this work, only Co and Cd were selected as representative 2, Cd; L, Zn. elements for the investigation of the eVect of forward rf power on a for various plasma compositions. In pure Ar, a for Co increases from 70 to 77% as the rf power increases from 20 to was due to the ingress of ambient atmosphere into the He 60 W; for Cd, a increases from 3% (20W) to 12% (50 W), plasma and was not associated with the Ar sheath gas.but as the power is raised to 60 W, a decreases to 7%. Similar eVects were observed for the intermediate plasma gas mixtures. EVect of ionization potential on a. The Saha equation, Since Ar plasmas increase the temperature of the center presented below, quantitatively expresses the relationship electrode by several hundred degrees compared with that in between the degree of ionization and the ionization potential He, it is not unexpected that decreases in a may occur for of the analyte, provided ionization occurs via thermal volatile elements as the forward power is raised beyond the processes: point at which the temperature of the electrode becomes too high to permit condensation of analyte vaporized in the early log [a/(1-a)]=15.684+log(Zion/Zat) stages of the heating cycle.As a consequence, the spatial +1.5 log T-5040 (IP/T)-log ne (3) distribution of the analyte in the plasma is altered.11 Analyte, which would normally condense and be re-released from the where ne is the plasma electron density (cm-3) and IP is the center electrode into the most intense region of the discharge ionization potential (eV ).A plot of log [a/(1-a)] vs. IP should as the electrode is radiatively heated by the tube wall, is unable yield a straight line from the slope of which (-5040/T) the initially to condense on the electrode when the latter is heated ionization temperature can be derived.Fig. 5 illustrates this either by the use of Ar plasma gas or when elevated rf power relationship for both a pure He and a pure Ar plasma. Electron is used. As a consequence, the distribution of the analyte is densities of 8×1013 and 1×1015 cm-3 were assumed for the altered during the ‘atomization’ stage and may not be exposed He and Ar plasma, respectively.29 Linear regression fits of the to the same extent to the annular region of highest plasma data yield a temperature of 5910±410 K for He and density around the center electrode. 5220±680 K for Ar. The former is in agreement with the ionization temperature reported earlier for the He plasma.11 Excitation and ionization of analytes The lower ionization temperature for Ar is intuitively inconsistent with the experimental results. Additionally, the much The enhancement of a which ensues on changing the plasma poorer linear fit obtained with Ar raises the question of gas composition is related to the fundamental excitation/ whether the assumption that thermal processes account for ionization mechanisms of the element which, in turn, are based the ionization phenomenon is correct.This point will be on the collisional and radiational exchange of energy amongst discussed at greater length below. the various plasma gas species and with atoms and ions of the Fig. 5 EVect of ionization potential on degree of analyte ionization. Fig. 6 Partial energy level diagram for excited levels of He, Ar and analyte species.+, He=1100 ml min-1; $, Ar=1100 ml min-1. 904 J. Anal. At. Spectrom., 1999, 14, 901–912analyte. Only three analytes (in the interests of brevity) have element that responds strongly to the change in plasma gas composition, particularly with respect to ionic excitation, been selected for detailed discussion of potential excitation/ ionization reactions in the mixed gas plasmas, viz., Cr, Cd and which is thus reflected in the substantial increase in a in the presence of Ar.For Mg, both the atom and ion line intensities Mg. As is evident from Fig. 2–4, Cd is exemplary of an Fig. 7 Partial term diagram for (A) Cr I and (B) Cr II. J. Anal. At. Spectrom., 1999, 14, 901–912 905Fig. 8 Partial term diagram for (A) Cd I and (B) Cd II. 906 J. Anal. At. Spectrom., 1999, 14, 901–912Fig. 9 Partial term diagram for (A) Mg I and (B) Mg II. are enhanced in a similar manner and therefore a is not Fig. 6 is a partial energy level diagram for the most significant rare gas species and the two energy levels of the changed appreciably by the presence of Ar. Chromium displays marked enhancement in ionic excitation only but undergoes monitored excited atomic and ionic states of each analyte. This figure provides a convenient energy perspective of the substantial ionization throughout. J. Anal. At. Spectrom., 1999, 14, 901–912 907entire system. The relative internal energy levels of the various are metastable atoms (2s 3S1, 19.82 eV, and 2s 1S0, 20.61 eV), ground state ions (1s 2S1/2, 24.59 eV), metastable molecules rare gas plasma species are compared with the partial grotrian energy level diagrams for the atomic and ionic states for these (2s 3S+u, 13.3–15.9 eV), molecular ions (2s 2S+u, 18.3–20.5 eV), excited atoms occupying a range of energy elements in Fig. 7–9. In addition to the interactions of analytes with rare gas plasma species, there are a number of (excited) levels (20.61 eV<He*<24.59 eV), an electron density29 of approximately 8×1013 cm-3, photons and molecular molecular impurities in the FAPES plasma (N2, N2+, CO, CO+, OH, CN, C2, NO, NH), as evidenced by their intense impurities similar to those found in the Ar plasma.band emission,5,7,31 which also serve as collision partners in quenching and excitation/ionization reactions. Although no General considerations. Although little is known of the excitation and ionization processes in the FAPES source, experimental data have been included here, the presence of these excited molecular species could be clearly discerned in mechanisms and models for other plasmas, such as the ICP,36–39 MIP40,41 and glow discharge source,42 have been the plasmas under study, despite the external flow of Ar sheath gas and internal plasma gas.Direct measurement of the reported. A cursory examination of these models highlights the importance of several general analyte excitation/ionization ambient atmosphere diVusing into the graphite tube against a convective flow of sheath and plasma gas has been presented schemes involving the participation of plasma gas species: collisional excitation by electrons, Penning ionization/exci- earlier.32,33 Table 4 summarizes many of the common band systems and energy levels for a number of these molecular tation with metastable rare gas species (although evidence also suggests that, in the ICP, argon atoms excited to levels above species.Data for the latter were taken from Pearse and Gaydon34 and Huber and Herzberg.35 the metastable states also participate in Penning reactions43,44) and charge transfer ionization and excitation collisions (DuVendack reaction) involving asymmetric charge transfer Plasma species. The introduction of atoms in their ground state into the FAPES plasma occurs during the atomization between plasma gas ions and analyte atoms which leads to the production of excited state analyte ions.De-excitation pro- step; subsequent excitation and ionization of the atom depend on the nature of the plasma gas. The energy distribution of cesses appear to be dominated by radiative decay and electron recombination. primary electrons, energy levels and densities of metastable atoms as well as density and kinetic energy of ionized gases With respect to the charge transfer reaction, the excess energy is generally considered to be dissipated as the kinetic play important roles in determining the excitation and ionization of analytes. It is therefore necessary to identify the key energy of the colliding partners; while the process need not be resonant37 (energy defect, DE=0), a large energy mismatch species produced in He and Ar plasmas.In an Ar FAPES plasma at atmospheric pressure there are is unfavorable.14 Turner-Smith et al.45 have shown that the cross-section for charge transfer is optimum when DE is metastable atoms (4s 3P02, 11.55 eV, and 4s¾ 3P00, 11.72 eV,), metastable molecules (10.2 eV), ground state ions (3p5 2P03/2, exothermic by 0.1–0.4 eV, although reactions may occur with decreasing eYciency to levels of higher energy defect up to at 15.76 eV), metastable state ions (3p5 2P01/2, 15.94 eV), molecular ions (14.0 eV), excited atoms occupying a range of energy least 1 eV.It has been conservatively assumed by Bauer and Skogerboe46 that processes of these types are possible when levels (11.83 eV<Ar*<15.76 eV), approximately29 1×1015 electrons cm-3 and photons, in addition to the molecular DE is exothermic by as much as 0.8 eV.Green and Webb47 concluded that, for levels oV-resonance by a positive (exother- impurities noted earlier. In the corresponding He plasma there Table 4 Partial molecular emission systems34,35 Molecule Band system Term symbol Energy range/eV N2 Second positive system C 3Pu–B 3Pu 4.42–2.49 Fourth positive system D 3Su+–B 3Pg 5.50–4.27 Vegard–Kaplan bands A 3Su+–X 1Sg+ 5.31–2.45 Lyman–Birge–Hopfield system a 1Pg–X 1Sg+ 6.18–4.77 Gaydon–Herman singlet systems —a–a 1Pg 5.57–3.39 Fifth positive (van der Ziel ) system x 1Sg-–a¾ 1Su- 6.10–4.46 Kaplan’s first system y 1Pg–a¾ 1Su- 5.97–5.03 Kaplan’s second system y 1Pg–w 1Du 5.48–4.34 Herman–Kaplan system E 3Sg+–A 3Su+ 5.80–4.54 CO Fourth positive system A 1P–X 1S+ 6.18–4.43 Third positive and 5B bands b 3S+–a 3Pr 4.38–3.24 3A bands c 3P–a 3P 5.40–4.57 OH 3064 A° system A 2S+–X 2Pi 5.07–3.08 Ultraviolet system 2550–2250 A° C 2S+–A 2S+ 5.51–4.62 CN Douglas and Routly’s systems —b 6.05–4.08 C2 Fox–Herzberg system B 3Pg–X¾ 3Pu 5.21–3.78 Mulliken system, 2313 A° d 1Su+–x 1Sg+ 5.13–3.57 Freymark bands e 1Sg+–b 1Pu 5.98–5.59 NO b System B 2Pr–X 2Pr 6.14–2.35 c System A 2S+–X 2P 6.34–3.58 d System C 2P–X 2P 6.78–5.13 e System D 2S+–X 2Pr 6.37–5.68 NH 2530 A° system d 1S–c 1P 4.95–4.62 CO+ First negative system B 2S+–X 2S+ 6.19–3.93 Marchand–D’incan–Janin system C 2Dr–A 2Pi 5.36–4.42 N2+ Second negative system C 2Su+–X 2Sg+ 6.48–6.02 Janin–D’incan system D 2Pg–A 2Pu 6.03–4.24 aTransitions for this band are all to a 1Pg state.bThis system includes four transitions which include H 2P–B 2S, F 2D–A 2P, D 2P–A 2P and D 2P–X 2S. 908 J. Anal. At. Spectrom., 1999, 14, 901–912mic) defect of up to 2 eV, such reactions may be strongly considered in this table as the energy defects are far too large to be of interest).The lowest energy defect is 4.39 eV excited, but levels of much greater defect are seldom directly excited. It has also been suggested that charge exchange for the reaction between Cr I (259.2 nm, 5.81 eV) and Ar2m (10.2 eV). As a consequence, it may be assumed that should not be excluded as a potential reaction when the energy defect is endothermic;48 provided momentum is excitation of all measured analyte atomic lines occurs primarily by (energetic) electron impact with ground state conserved, the required additional energy, in the amount of kT, can be supplied in the form of kinetic energy from the atoms or via radiative recombination, in which capture of an electron by an analyte ion results in a highly excited collision partners.Contrary to the case for DuVendack reactions, Penning reactions do not selectively excite levels of state of the atom which subsequently undergoes radiative cascade or collisional thermalization to populate the level small energy defect.47 Similar arguments regarding collisional energy transfer from excited state molecular species to atomic in question. As noted earlier, there are a variety of molecular species in systems have also been made.49 A general note must be raised regarding the Wigner spin the plasma and they may play some role in the excitation of atoms of the analytes. Fahey et al.50 presented evidence for conservation rule when considering the probability of such collisional energy exchanges.The Wigner conservation rule the direct collisional excitation of Cd, Zn and Sr metal vapors by electronically and vibrationally excited N2 molecules (A requires that the resultant spin of the reactants be the same as that of the products. Exceptions to the rule are possible 3Su+) extracted as an active nitrogen beam from an N2 glow discharge. The resulting analyte excitation temperature was in when one of the atoms shows j, j coupling [for heavier atoms (Z>30) where spin–orbit interactions are large].The Wigner accord with the N2 vibrational temperature. Similarly, Baltayan et al.51 reported that the weak emission from Zn spin rule does seem to hold in at least one case documented by Steers and Thorne,20 however, namely that for charge atom resonance lines observed in experiments in a flowing afterglow apparatus was due to energy transfer from meta- transfer reactions involving Ar+ and Cr I.stable N2* (A 3Su+) molecules present as an impurity in the He carrier gas. It is evident from the data in Fig. 2 that the Excitation of analyte atoms. The intensity of an emission line from an analyte atom is proportional to the state relative intensities of all of the measured atomic lines of the analytes are enhanced as the volume fraction of Ar increases population of the excited level from which the transition originates. The upper energy levels for all of the excited states in the mixed gas plasma.This is in accord with the increasing electron density likely to occur on moving from pure He to of the atomic lines of the analytes studied here are <6 eV (Fig. 6, range 4.43–5.81 eV). It is clear from inspection of pure Ar. The extent of the emission intensity enhancement should then be related to the excitation energy of the upper Fig. 7–9 that direct population of the excited states of any of the analyte atoms is unlikely to occur via eYcient exchange level involved in the transition (Table 2); this is approximately the case, in that emission from Mg I (3p 1P01, 285.2 nm, of internal energy during collisions with any of the He or Ar plasma gas species.The energy defects for the various 4.35 eV) is enhanced the most (660%) and Cr the least [Cr I (4p u 5P03, 259.2 nm, 5.81 eV) only 10%]. As noted earlier, collisional excitation scenarios, as summarized by the data in Table 5, are too large for all of the elements studied collisions with excited state molecular impurities may also give rise to excited analyte atoms.In the case of Mg, there are (interactions with excited ionic states of Ar and He are not Table 5 Energy defects for reactions of analyte atoms with plasma species leading to collisional ionization/excitation Energy defect from species in plasma/eV Analyte l/nm Energya/eV Arm Ar2m Ar2+ Ar* Ar+ Ar+m He2m He2+ Hem He* He+ Cr I 259.2 5.81 5.74 4.39 8.19 6.02< 9.95 10.13 7.49– 12.49– 14.01 14.80< 18.78 5.91 <9.95 10.09 14.69 14.79 <18.78 Cr II 283.6 12.69 -1.14 -2.49 1.31 -0.75< 3.07 3.25 0.61– 5.61– 7.13 7.92< 11.90 -0.97 <3.07 3.21 7.81 7.91 <11.90 Mg I 285.2 4.35 7.20 5.85 9.65 7.48< 11.41 11.59 8.95– 13.95– 15.47 16.26< 20.24 7.37 <11.41 11.55 16.15 16.25 <20.24 Mg II 279.6 12.07 -0.52 -1.87 1.93 -0.24< 3.69 3.87 1.23– 6.23– 7.75 8.54< 12.52 -0.35 <3.69 3.83 8.43 8.53 <12.52 Mn I 279.5 4.43 7.12 5.57 9.37 7.40< 11.33 11.51 8.87– 13.87– 15.09 15.88< 20.16 7.29 <11.33 11.47 16.07 15.87 <20.16 Mn II 259.4 12.21 0.66 -2.01 1.79 -0.38< 3.55 3.73 1.09– 6.09– 7.61 7.61< 12.38 0.83 <3.55 3.69 8.29 8.39 <12.38 Fe I 254.1 4.99 6.56 5.21 9.01 6.84< 10.77 10.95 8.31– 13.31– 14.83 15.62< 19.60 6.73 <10.77 10.91 15.51 15.61 <19.60 Fe II 259.9 12.67 -1.12 -2.47 1.33 -0.84< 3.09 3.27 0.63– 5.63– 7.15 7.94< 11.92 -0.95 <3.09 3.23 7.83 7.93 <11.92 Co I 235.8 5.48 6.07 4.72 8.52 6.35< 10.28 10.46 7.82– 12.82– 14.34 15.13< 19.11 6.24 <10.28 10.42 15.02 15.12 <19.11 Co II 237.9 13.50 -1.95 -3.30 0.50 -1.67< 2.26 2.44 -0.20– 4.80– 6.68 7.47< 11.09 -1.78 2.26< 2.40 7.00 7.46 <11.09 Cd I 228.8 5.42 6.13 4.78 8.58 6.41< 10.34 10.52 7.88– 12.88– 14.40 15.19< 19.17 6.30 10.34 10.48 15.08 15.18 <19.17 Cd II 237.9 14.46 -2.91 -4.26 -0.46 -2.63< 1.30 1.48 -1.16– 3.84– 5.36 6.15< 10.13 -2.74 1.30< 1.44 6.04 6.14 <10.13 Zn I 213.9 5.80 5.75 4.40 8.20 6.03< 9.96 10.14 7.50– 12.50– 14.02 14.81< 18.79 5.92 9.96< 10.10 14.70 14.80 <18.79 Zn II 202.6 15.51 -3.96 -5.31 -1.51 -3.68< 0.25 0.43 -2.21– 2.79– 4.31 5.10< 9.08 -3.79 0.25< 0.39 4.99 5.09 <9.08 aEnergy for ionic lines is the sum of ionization potential and excitation energy.J. Anal. At. Spectrom., 1999, 14, 901–912 909several such reactions which could be considered wherein the reactions.The energy levels for Cd II shown in Fig. 8(B) suggest that population of the 5p 2P01/2 level in He plasma energy defect is less than 0.4 eV: may occur by radiative/collisional cascade from upper levels N2* (C 3Pu, 282.0 nm, 11.79 eV)+Mg (3s2 1S0, 0 eV)A which are excited in collisions with excited state He atoms, where a manifold of overlaps appears accessible. These energy N2* (B 3Pg, 7.39 eV)+Mg* (285.2 nm, 3p 1P1, 4.35 eV) transfers cannot be considered viable as they all violate the +DE=0.05 eV (4) Wigner spin conservation rule.Collision with He2m is also an unlikely scenario; both Collins52 and Green and Webb47 CO* (b 3S+, 283.3 nm, 10.42 eV)+Mg (3s2 1S0, 0 eV)A suggest that such reactions are not significant. Charge transfer CO* (a 3Pr, 6.04 eV)+Mg* (285.2 nm, 3p 1P1, 4.35 eV) from He+ may be considered to populate the upper Cd II levels,52,53 e.g., +DE=0.03 eV (5) He+ (1s 2S1/2, 24.59 eV)+Cd (5s 1S0, 0 eV)A C2* (B 3Pg, 277.2 nm, n¾ 1)+Mg (3s2 1S0, 0 eV)A He (1s2 1S0, 0 eV)+Cd+* (8d 2D3/2, 24.44 eV) C2* (X¾ 3Pu n 3)+Mg* (285.2 nm, 3p 1P1, 4.35 eV) +DE=0.15 eV (8) +DE=0.12 eV (6) OH* (A 2S+,Q2, 282.9 nm, 4.38 eV)+Mg (3s2 1S0, 0 eV)A Approximately half the product Cd+ ions excited are in the five high-lying 9p 2P, 8d 2D, 6g 2G, 6f 2F and 9s 2S states, OH (X 2Pi, 0 eV)+Mg* (285.2 nm, 3p 1P1, 4.35 eV) the remaining being distributed between other levels of Cd+ down to a few eV below the energy of the He+ ion.53 +DE=0.03 eV (7) Collisional equilibration will serve to populate the measured For Cd I (5p 1P01, 228.8 nm, 5.42 eV), the number of available 5p 2P01/2 level.molecular states with similarly small energy defect is consider- As the Ar content of the plasma is increased, emission from ably less than that for Mg I. Consideration of such molecular the Cd II 5p 2P01/2 level is enhanced (Fig. 3), possibly as a energy transfer reactions for the excitation of Cr I shows them result of direct collisional population of the measured 5p 2P01/2 to be spin forbidden.level, i.e., Formation and excitation of analyte ions. It is evident from Ar+ (3p 2P3/2, 15.76 eV)+Cd (5s 1S0, 0 eV)A Fig. 3 that the intensities of the ionic lines of all of the analytes Ar (3p 1S0, 0 eV)+Cd+* (5p 2P01/2, 226.5 nm, 14.46 eV) studied are enhanced as the composition of the plasma is changed to favor Ar. For Cr II (4p z 6F011/2, 283.6 nm, +DE=1.30 eV (9) 12.69 eV), intensities are increased nearly 30-fold in pure Ar as well as via a Penning ionization reaction with Ar* such as compared with pure He, while emission from Cd II (5p as43,44 2P01/2, 226.5 nm, 14.46 eV) and Mg II (3p 2P03/2, 279.6 nm, 12.07 eV) is enhanced more than 10-fold. Ar* (5p 3D2, 14.51 eV)+Cd (5s 1S0, 0 eV)A Two processes generally dominate the production of excited state ions in analytical plasmas: Penning and DuVendack Ar (3p 1S0, 0 eV)+Cd+* (5p 2P01/2, 226.5 nm, reactions.Energy considerations for these reactions have been 14.46 eV)+e-+DE=0.05 eV (10) discussed earlier. It is evident from the energy level diagrams presented in Fig. 7–9 and the data in Table 5 that collisional (the Ar 5p 3D2 level provides an intense emission line in the transfer of the internal energy of any rare gas species to FAPES source at 430.0 nm). populate directly the measured excited levels of the ions of Cr, Direct collisional population of the measured level in the Cd and Mg is impossible in the He plasma owing to the large Ar plasma would account for both the substantially enhanced energy defects.Excitation of high-lying states followed by intensity of the 226.5 nm Cd II emission line and the radiative cascade may account for some fraction of the level significantly increased degree of ionization (Fig. 3 and 4). population, although this process would be expected to be Substantial increases in emission intensity from both the ineYcient owing to the many branching ratios.52 atomic and ionic lines of Mg occur when the plasma gas For Cr II, collisional population of any high-lying levels by composition is altered to contain more Ar.Ionization of Mg interaction with various He species is spin forbidden by the in the He plasma may arise as a result of Penning reactions Wigner rule. Electron impact ionization/excitation thus with both Hem (into the 6s 2S1/2 Mg II level ) and He* (into appears to be one of the few routes available to account for the 9, 8 and 7 s, p and d levels of Mg II ) species (spin emission from the measured Cr II (4p z 6F011/2, 283.6 nm, conserved with the liberated electron), followed by collisional 12.69 eV) level in He plasmas.Increases in the intensity of the equilibration to the measured 3p 2P03/2 level (albeit through a Cr II line with increased Ar content of the plasma could be large manifold of transitions which makes the overall process linked to the 10-fold greater electron density in this plasma. ineYcient): Penning ionization/excitation to populate directly the Cr II 4p Hem (2s 1S0, 20.61 eV)+Mg (3s2 1S0, 0 eV)A z 6F011/2 level (283.6 nm, 12.69 eV) where the collision partner is an excited state Ar atom (3P1, 13.33 eV), as proposed by He (1S0, 0 eV)+Mg+* (6s 2S1/2, 20.43 eV)+e- Boumans,43,44 is also an attractive possibility, wherein spin +DE=0.18 eV (11) conservation is achieved through the release of a free electron in the ionization process.Such a scenario would account for and the rise in emission intensities from the atom and ion lines in Ar plasmas and also the corresponding increase in degree of He*(3s 3S1, 22.72 eV)+Mg (3s2 1S0, 0 eV)A ionization. Participation of excited state Ar atoms in such He (1S0, 0 eV)+Mg+* (10s 2S1/2, 21.99 eV)+e- processes may occur despite their short lifetime because of their continual overpopulation (with respect to the ground +DE=0.73 eV (12) state) via interaction with Ar metastables. Ionization and excitation of Cd atoms may be accounted However, in the presence of Ar, charge transfer reactions have been shown to overpopulate directly the Mg II 4s 2S1/2 and for from consideration of a limited number of potential 910 J.Anal. At. Spectrom., 1999, 14, 901–9129 R. E. Sturgeon and R. Guevremont, Anal.Chem., 1997, 69, 2129. 3d 2D5/2 levels:48,54 10 G. C. Y. Chan and W. T. Chan, J. Anal. At. Spectrom., 1998, Ar+ (3p5 2P03/2, 15.76 eV)+Mg (3s2 1S0, 0 eV)A 13, 209. 11 R. E. Sturgeon and R. Guevremont, J. Anal. At. Spectrom., 1998, Ar (3p6 1S0, 0 eV)+Mg+* (4s 2S1/2, 16.29 eV) 13, 229. 12 K. Wagatsuma and K. Hirokawa, Spectrochim. Acta, Part B, +DE=-0.53 eV (13) 1991, 46, 269. 13 K. Wagatsuma and K. Hirokawa, Anal. Chem., 1985, 57, 2901. which can collisionally or radiatively equilibrate with the 14 E. B. M. Steers and R. J. Feilding, J. Anal. At. Spectrom., 1987, measured 3p 2P3/2 level.Although the endothermic energy 2, 239. defect can be overcome in the Ar ICP48 by considering the 15 K. Wagatsuma and K. Hirokawa, Spectrochim. Acta, Part B, thermal energy (kT) of the heavy particles to be of the order 1987, 42, 523. of 0.5 eV (6000 K), such an excess is not available in the 16 K. Wagatsuma and K. Hirokawa, Anal. Chem., 1988, 60, 702. FAPES source even with the furnace is at its maximum 17 K. Wagatsuma and K. Hirokawa, J. Anal. At. Spectrom., 1989, 4, 525.atomization temperature. It is not surprising, then, that no 18 D. Serxner, R. L. Smith and K. R. Hess, Appl. Spectrosc., 1991, emission is detected from the Mg II at 292.9, 293.7, 279.8 or 45, 1656. 279.1 nm, arising from these 3d and 4s levels.48 19 E. B. M. Steers and F. Leis, Spectrochim. Acta, Part B, 1991, Alternatively, the Mg II 3p 2P03/2 level may be populated 46, 527. by Penning reaction with excited Ar atoms: 20 E. B. M. Steers and A. P. Thorne, J. Anal. At. Spectrom., 1993, 8, 309.Ar* (4p 3P1, 13.33 eV)+Mg (3s2 1S0, 0 eV)A 21 Atomic Energy Levels as Derived from the Analyses of Optical Spectra: Volume I. The Spectra of 1H–23V, ed. C. E. Moore, US Ar (3p6 1S0, 0 eV)+Mg+* (3p 2P03/2, Department of Commerce, Washington, DC, 1949. 12.07 eV)+e-+DE=1.26 eV (14) 22 Atomic Energy Levels as Derived from the Analyses of Optical Spectra: Volume II. The Spectra of 24Cr–41Nb, ed. C. E. Moore, and via electron impact processes [the Ar* transition produces US Department of Commerce,Washington, DC, 1952.an intense line at 696.5 nm (4p 3P1, 13.33 eV)in the FAPES 23 Atomic Energy Levels as Derived from the Analyses of Optical source]. Such reactions may account for the enhanced ionic Spectra: Volume III. The Spectra of 42Mo–57La, 72Hf–89Ac, ed. C. E. Moore, US Department of Commerce, Washington, DC, emission and increase in the calculated degree of ionization 1958. with Ar-based plasmas. 24 NIST Website, NIST Atomic Spectroscopic Database Version 1.1, http://aeldata.phy.nist.gov/nist-atomic-spectra.html. Conclusions 25 Wavelengths and Transition Probabilities for Atoms and Atomic Ions, ed.J. Reader, C. H. Corliss, W. L. Wiese and G. A. Martin, Emission intensities from both atomic and ionic lines of the US Department of Commerce,Washington, DC, 1980. analytes investigated, as well as their calculated degrees of 26 Atomic Transition Probabilities: Volume II. Sodium Through ionization, were enhanced as the Ar content of the He plasma Calcium, a Critical Data Compilation, ed.W. L. Wiese, M. W. Smith and B. M. Miles, US Department of Commerce, was increased. The larger electron density and specific Penning Washington, DC, 1969. and charge transfer reactions with analyte and energetic argon 27 Experimental Transition Probabilities for Spectral Lines of Seventy species probably account for this trend. Apparent temperatures Elements, ed. C. H. Corliss and W. R. Bozman, US Department of 5910 and 5220 K for He and Ar characterize the ionization of Commerce, Washington, DC, 1962.processes in these two plasmas. The degree of ionization is 28 L. de Galan, R. Smith and J. D. Winefordner, Spectrochim. Acta, >70% in the Ar plasma for elements measured having an Part B, 1968, 23, 521. ionization potential below 8 eV. The application of such a 29 R. E. Sturgeon, V. T. Luong and R. K. Marcus, paper presented at the Federation of Analytical Chemistry and Spectroscopy mixed gas plasma in an eVort to enhance the capabilities for Societies Meeting, St.Louis, MO, October 2–7, 1994, paper 272. use of this source for mass spectrometry11 is attractive. The 30 C. W. Le Blanc, PhD Thesis, University of British Columbia, excitation/ionization pathways remain to be explored in this Vancouver, 1996. source and any speculations must be based on a more compre- 31 C. W. Le Blanc and M. W. Blades, Spectrochim. Acta, Part B, hensive examination of multiple lines from each analyte 1995, 50, 1395.element in order to obtain a more complete and clear picture. 32 R. E. Sturgeon and H. Falk, J. Anal. At. Spectrom., 1988, 3, 27. 33 R. E. Sturgeon and H. Falk, Specrochim. Acta, Part B, 1988, 43, 421. Acknowledgements 34 The Identification of Molecular Spectra, ed. R. W. B. Pearse and A. G. Gaydon, Chapman and Hall, London, 4th edn., 1976. The authors thank S. Mitchel (NRCC) for helpful discussions 35 Molecular Spectra and Molecular Structure, IV.Constants of and V. T. Luong and V. Boyko (NRCC) for technical Diatomic Molecules, ed. K. P. Huber and G. Herzberg, Van assistance. F.S. thanks the NRCC for financial support while Nostrand Reinhold, New York, 1979. working in Canada. 36 D. C. Schram, I. J. M. M. Raaymakers, B. van der Sijde, H. J. W. Schenkelaars and P. W. J. M. Boumans, Spectrochim. Acta, Part B, 1983, 38, 1545. References 37 A. Goldwasser and J. M. Mermet, Spectrochim. Acta, Part B, 1986, 41, 725. 1 D. C. Liang and M. W. Blades, Spectrochim. Acta, Part B, 1989, 38 M. Li and Z. Zhang, Spectrochim. Acta, Part B, 1998, 53, 1391. 44, 1059. 39 M. W. Blades, in Inductively Coupled Plasma Emission 2 R. E. Sturgeon, S. N. Willie, V. T. Luong, S. S. Berman and Spectroscopy, Part II, ed. P. W. J. M. Boumans, Wiley, New York, J. G. Dunn, J. Anal. At. Spectrom., 1989, 4, 669. 1987, p. 387. 3 R. E. Sturgeon, S. N. Willie, V. T. Luong and S. S. Berman, 40 C. I. M. Beenakker, Spectrochim. Acta, Part B, 1977, 32, 173. J. Anal. At. Spectrom., 1990, 5, 635. 41 P. Brassem, F. J. M. J. Maessen and L. de Galan, Spectrochim. 4 D. L. Smith, D. C. Liang, D. Steel and M. W. Blades, Spectrochim. Acta, Part B, 1976, 31, 537. Acta, Part B, 1990, 45, 493. 42 A. Bogaerts and R. Gijbels, J. Anal. At. Spectrom., 1996, 11, 841. 5 R. E. Sturgeon, S. N. Willie, V. T. Luong and S. S. Berman, 43 P. W. J. M. Boumans, Spectrochim. Acta, Part B, 1982, 37, 75. Anal.Chem., 1990, 62, 2370. 44 P. W. J. M. Boumans and F. J. De Boer, Spectrochim. Acta, Part 6 R. E. Sturgeon, S. N. Willie and V. T. Luong, Spectrochim. Acta, B, 1977, 32, 365. Part B, 1991, 46, 1021. 45 A. R. Turner-Smith, J. M. Green and C. E. Webb, J. Phys. B: At. 7 R. E. Sturgeon, S. N. Willie, V. T. Luong and J. G. Dunn, Appl. Mol. Phys., 1973, 6, 114. Spectrosc., 1991, 45, 1413. 46 C. F. Bauer and R. K. Skogerboe, Spectrochim. Acta, Part B, 8 C. W. Le Blanc and M. W. Blades, Appl. Spectrosc., 1997, 51, 1715. 1983, 38, 1125. J. Anal. At. Spectrom., 1999, 14, 901–912 91147 J. M. Green and C. E. Webb, J. Phys. B: At. Mol. Phys., 1974, 52 G. J. Collins, J. Appl. Phys., 1973, 44, 4633. 7, 1698. 53 P. Baltayan, J. C. Pebay-Peyroula and N. Sadeghi, J. Phys. B: At. 48 P. B. Farnsworth, B. W. Smith and N. Omenetto, Spectrochim. Mol. Phys., 1985, 18, 3615. Acta, Part B, 1991, 46, 843. 54 L. L. Burton and M. W. Blades, Spectrochim. Acta, Part B, 1991, 49 Introduction to Molecular Energy Transfer, ed. J. T. Yardley, 46, 819. Academic Press, New York, 1980. 50 D. W. Fahey, W. F. Parks and L. D. Schearer, J. Chem. Phys., 1979, 71, 2840. Paper 9/00273A 51 P. Baltayan, J. C. Pebay-Peyroula and N. Sadeghi, J. Phys. B: At. Mol. Phys., 1986, 19, 2695. 912 J. Anal. At. Spectrom., 1999, 14, 901–912

ISSN:0267-9477    

DOI:10.1039/a900273a

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Determination of ions in individual fluid inclusions by laser ablation optical emission spectroscopy: development and applications to natural fluid inclusions Ce�cile Fabre,*a Marie-Christine Boiron,a Jean Dubessya and Alain Moissetteb aCREGU–UMR G2R 7566, BP 23, 54501, Vandoeuvre-le`s-Nancy Cedex, France. E-mail: cecile.fabre@g2r.u-nancy.fr; Fax:+33 (0) 3 83 91 38 01 bLASIR, UER de Chimie C8, USTL , 59655 Villeneuve d’Ascq and CREGU–UMR 7566, BP 23, 54501 Vandoeuvre-le`s-Nancy Cedex, France Received 30th November 1998, Accepted 1st April 1999 Chemical data on the composition of individual fluid inclusions are required to model paleofluid–rock interactions, as these inclusions are the direct remains of the former fluid circulation in the earth’s crust.The proposed technique of laser ablation coupled with optical emission spectrometry (LA-OES) determines ion ratios in individual fluid inclusions. The radiations of the elements present in the plasma are analysed with a spectrometer equipped with a pulsed and gated multichannel detector.The lateral resolution is around 6 mm and the depth of the quartz crater obtained for one laser shot is around 1 mm. For fluid inclusions, the laser beam is initially focused on the surface of the sample, to drill until the inclusion is reached, then the liquid is analysed. Plasma temperature studies showed that various kinds of standards (synthetic glasses, fluid inclusions and minerals) can be used for the establishment of calibration curves for Na/K, Na/Ca and Na/Li ratios.As the emission line intensity is a function of the ablated mass, only emission line ratios between two elements are used. In addition, a self-absorption process was considered for the calibration strategy. The relative standard deviations of the calibration curves vary from 5% (glasses) to 25% (fluid inclusions) and the detection limits are those required for the determination of ions in individual inclusions (Na and Li 10, Ca 20 and K 750 ppm).Natural fluid inclusions were analysed using the LA-OES technique and it is now possible to determine the ratios of major elements in individual fluid inclusions. This paper reports recent developments using LA-OES, Introduction previously described by Boiron and co-workers14,15 and Quantitative chemical analyses of fluid inclusions are a Moissette et al.16 and applied to individual fluid inclusions. It prerequisite for understanding and modelling fluid–rock inter- is now possible to determine the ratios of major elements.This actions as only these inclusions contain direct evidence of the work consisted in establishing the analytical procedure and composition of fluids responsible for past geological processes. the calibration curves. First applications to major ion ratios Fluid inclusions with diVerent ages and compositions (size, to natural fluid inclusions are also described. abundance and diversity) within a single crystal complicate the interpretation of the bulk chemical analysis of fluids Principle and experimental procedure extracted using crush–leach techniques.The determination of ion contents in individual fluid inclusions is a new challenge Principle to achieve this aim. During the last 10 years, various techniques have been The eVects of absorption of laser radiation on a material are developed in an attempt to determine ion ratios in individual heating, melting, vaporisation, atomisation, excitation and inclusions. Some modern analytical methods are based on the ionisation, depending on the energy deposited on the sample.analysis of electromagnetic radiation produced by excitation Lines emitted from excited atoms and ions are used in laser with accelerated particles or electrons, e.g., X-ray microanal- microanalysis based on plasma emission spectroscopy.17–19 ysis of frozen inclusions,1 X-ray emission (PIXE) or gamma- The incident laser radiation generates plasma of high temperaray emission (PIGE) spectroscopy2–4 and synchrotron X-ray ture containing atoms and ions in an excited state.When fluorescence spectroscopy.5–8 PIXE, PIGE and SXRF atoms and ions transfer from an excited electronic level to a methods, although non-destructive, are ‘heavy’ methods and lower electronic level, a radiatial transition may occur with the analytical results depend on the shape and depth of the emission of photons.inclusions in the host mineral. Other methods under develop- The intensity of an emission line during such electronic ment are destructive at the scale of a focused laser beam and transition can be written as: are based either on ICP-MS9–13 or optical emission spectroscopy (OES).14–16 Both techniques are coupled with laser Iij= VBcNgiAij 4plijZ(T ) exp A-Ei kT B (1) ablation (LA) at the microscopic scale (<10 mm) to extract or to reach the liquid to be analysed within the inclusions. where I is the intensity of the emission line corresponding to LA-ICP-MS9–13 seems more appropriate for the determination the transition i�j, i and j are the indices of the higher and of trace elements, although LA-OES14–16 gives better results lower energetic quantum state, respectively, V is the collection for major elements, especially Na and Ca (better detection limits).solid angle for the plasma emission, B is Planck constant, c is J. Anal. At. Spectrom., 1999, 14, 913–922 913Fig. 1 Schematic diagram of the experiment set-up (built thanks to EU project program MAT1-CT-93–002924) showing the Nd5YAG laser coupled with an optical emission spectrometer. M, mirror; PM, parabolic mirror; L, lens; D, diaphragm; V-C, video camera; P-D, pulsed detector, X-Y, microscope stage; C-O, Cassegrain objective. the speed of light, N is the number of free atoms of the studied of the elements present in the plasma are analysed directly with a spectrometer (40 mm working distance) (Dilor) element in the plasma, gi is the statistical weight of the quantum state i, Aij is the transition probability for spon- equipped with a pulsed and gated multichannel detector. An adjustable time delay (between 50 and 1000 ns) is used to taneous emission from i to j, lij is the wavelength of the emission line, Z(T ) is the partition function of the quantum remove the continuous emission occurring in the first tens of nanoseconds after the laser shot and to optimise the signal- state, T is the electronic excitation temperature (K) and Ei is the energy level of excited electronic level i.to-background ratio. The best delay is 150 ns for a temporal window of 500 ns. The monochromator has a 280 nm focal length (Spex-280; Spex, Edison, NJ, USA). It is equipped with Instrumentation two gratings with 300 grooves mm-1 blazed at 250 and 600 nm An Nd5YAG laser (Continuum, Minilite, Evry, France) (Jobin-Yvon, Longjumeau, France).A 200 nm spectral range coupled with an optical emission spectrometer (Fig. 1) delivers is simultaneously recorded, utilising the 300 grooves mm-1 a laser pulse (5 ns). The operating conditions and the charac- holographic grating. The whole system was assembling by teristics of the instrument are summarised in Table 1. The laser Dilor. The spectral resolution of the spectrograph is 1.2 nm. beam is focused on to the sample through a Cassegrain-type microscope objective (Dilor, Lille, France). Plasma is created Spatial resolution and ablated mass during the laser–matter interaction.The typical emission lines Lateral resolution was measured for diVerent types of mineral (quartz, halite, sylvite, calcite, fluorite) and metallic matrices. Table 1 LA-OES operating conditions for solids and individual fluid inclusions The lateral resolution (Fig. 2) is around 6 mm for each type of material even if the depth of the crater hole obtained for Laser system — one laser shot can be very diVerent, e.g., from 1 mm for quartz Laser mode Q-switched up to 6 mm for copper.Hence the ablated mass is diVerent for Laser type Nd5YAG (Minilite, Continuum) each matrix and depends on various physical parameters of Energy per pulse Maximum output 2.8 mJ (266 nm) the solid (thermodynamic parameters, optical and transport Wavelength 266 nm (quadrupled frequency) Pulse width 5 ns properties).17,18,20 Thus the type of crater depends on the Acquisition time 500 ns ablated mass and consequently on the heat diVusion in the Crater size 6–10 mm in quartz for one shot matrix, the melting temperature of the material and its Spectroscopic system — crystalline state. Spectrograph Spex 270 M Wavelength 200–800 nm Spectral range 200 nm Experimental procedure Slit width 50 mm Spectral resolution 1.2 nm The laser beam is initially focused on the surface of the sample Grating 300 grooves mm-1 progressively ablated until the inclusion is reached.The pro- Gas Argon gress of the ablation is monitored using the matrix lines, Si Flow rate 0.8 l min-1 for a quartz matrix. Once the inclusion is opened, the Si 914 J. Anal. At. Spectrom., 1999, 14, 913–922Fig. 2 Microphotographs of ablation craters made on (A) metallic copper (10 shots) and (B) quartz (100 shots). emission lines decrease and the characteristic lines of the ions Fig. 3 Optical emission spectra obtained by LA-OES on synthetic fluid inclusion (one laser shot) for the determination of the three present in the liquid phase appear.calibration curves, using the emission lines of sodium (588.9 nm), It is worth noting that at least 10 laser shots are required lithium (670.7 nm), potassium (766.5 nm) and calcium (787.2 nm). to empty an inclusion of 20×20 mm. With larger inclusions, 30–50 shots can readily be made. This eVect is due to the short laser pulse duration (5 ns) and to the supersonic ejection eVect of plasma temperature in order to estimate its influence of plasma (10 mm/10 ns) which prevents heat transfer to on the calibration curves.The two emission lines and their non-ablated liquid and so its vaporisation after one laser shot. electronic transitions22–24 are: 515.32 nm (3d104p1�3d104d1) The concentrations of the ions in the liquid diVer widely; and 510.55 nm (3d94s2�3d104p1) (Fig. 3). sodium is often the main cation. Thus, filters are used to Experiments were carried out under an argon flow on pure reduce the intensity of the Na emission doublet to prevent metallic copper, silicate glass (Si, K, Cu, Na) containing 5% saturation of the intensified photodiode array.Two types of m/m of copper, copper solution (6000 ppm) and copper-rich filters have been used: (i) the first filter cuts the emission lines synthetic fluid inclusions (5000 ppm) to estimate the plasma with wavelength lower than 400 nm (Ca) and reduces the temperature.This temperature is an average temperature over signal intensity of sodium (around 590 nm) to 40%; (ii) the time and space. Temperature is time averaged because the second filter permits the recording of 6% Na intensity, 14% signal is integrated over 500 ns (T decreases slightly with for Li and 92% for K and Ca. time). Temperature is also space averaged because all the Sweeping the sample with an argon flow has been proved plasma is imaged on the entrance slit of the spectrometer and to be very eYcient in enhancing the plasma temperature and on the detector. then the line intensities.21 Thus, a cell was specially adapted A study of the depth eVect was carried out on copper metal.to work under a constant argon atmosphere. The argon flux The results show that the average plasma temperature does was optimised in order to obtain the best reproducibility of not change until a depth of the laser focus of around 80 mm. the signal. It has been demonstrated that low (<0.5 l min-1) Therefore, the fact that the plasma originating from fluid or high (>1.5 l min-1) argon fluxes are not adequate for a inclusions located at depths less than 80 mm is expelled through correct and repeatable signal intensity.The optimised value of a crater of a few tens of micrometres will not have any the argon flux was estimated to be around 0.8 l min-1. influence on the average plasma temperature. Only the first shot creates a plasma with a temperature (9300 K) lower than Calibration standards those following and the temperature RSD calculated on 30 shots performed in the same crater is around 2.4%.Temperature of plasma The calculated plasma temperatures are comparable for the For a constant ablated mass of an element, the excitation diVerent materials and around 11 000±500 K (solution, temperature of the plasma controls the intensity of emission 11 500±320 K; glasses, 10 400±540 K). The small variation lines.19 Consequently, the emission line ratio of two distinct of temperature that can occur between glasses and aqueous elements is a function of plasma temperature.solutions is insuYcient to achieve a variation of a two-element As shown by eqn. (1), the intensity of the optical emission emission lines ratio for the diVerent temperature conditions line is a function of excitation temperature. This temperature [as shown in eqn. (1), the ratio of two emission lines is a is of great importance and can be calculated using the following function of temperature].Using Student’s t-test, it can be equation (considering that the plasma is in thermodynamic demonstrated that the temperature calculated for the metallic equilibrium): copper (11 500±230 K) is the same than those found for the solution. These results show that the calibration of atomic emission kT= E1-E2 lnAg1A1l1 g2A2l2B-lnAI1 I2B (2) lines of the diVerent elements versus their concentration can be performed using synthetic glasses and applied for fluid inclusion analysis.where k is Boltzmann’s constant, T is the plasma temperature (K), E1 and E2 are the excitation energy levels for the two Standards considered emission lines, g1 and g2 are degeneration degrees of the higher level, A1 and A2 are transition probabilities and The determination of the ion content in fluid inclusions requires the calibration of the method with standards that l1 and l2 are the wavelengths of the emission lines.Using eqn. (2) and two Cu emissions lines, it is possible to study the could be analysed under the same conditions as the natural J. Anal. At. Spectrom., 1999, 14, 913–922 915Table 4 Compositions (molal ) of synthetic fluid inclusions in quartz samples. Standards consist of synthetic glasses, minerals and used for the establishment of the calibration curves synthetic fluid inclusions in quartz (Tables 2–4). Synthetic glasses were prepared from carbonates (CaCO3, Synthetic fluid inclusion in quartz Na2CO3, K2CO3) and oxide (Li2O) in an SiO2 matrix (Table 2).After decarbonation, the powders were heated twice Sample Na K Li Ca in platinum crucibles in an oven at around 1300 °C for 10 min. FI 1 0.05 0.025 0.006 0.05 Between the two heatings, the glasses were crushed to homog- FI 2 0.1 0.01 0.005 0.05 enise the standard. The compositions of the synthetic glasses FI 3 0.1 0.2 0.02 0.02 FI 4 1 — — 1 FI 5 0.5 — — 0.5 Table 2 Compositions (in % m/m) of the 30 synthetic glasses used for FI 6 1 — 0.1 — the establishment of the calibration curves FI 7 0.5 — 0.5a — FI 8 2 0.2 — — Sample SiO2 Al2O3 Na2O K2O CaO Li2 O FI 9 0.66 0.33 — — FI 10 2 2 — — SG1 71.25 11.15 5.68 3.82 — — FI 11 0.08 0.02 0.005 0.014 SG2 71.99 11.10 5.69 4.03 — — FI 12 0.10 0.02 0.008 0.03 SG3 69.59 11.28 6.26 3.80 — — FI 13 0.10 — — 1 SG4 65.35 10.81 8.72 3.18 — — FI 14 1 — 0.10 — SG5 65.69 11.30 9.50 3.20 — — aCorresponds to the concentrations of lithium that could not be used SG6 76.96 — 20.47 4.07 — — for the Na/Li calibration curve owing to the self-absorption of the Li SG7 71.93 10.89 4.70 4.05 — — emission line at 670.7 nm. SG8 66.28 — 12.67 7.28 — — SG9 71.04 — 18.51 4.59 3.89 — SG10 72.47 — 15.52 5.29 4.88 — SG11 62.64 17.44 1.19 10.87 4.68 4.13b were then analysed by electron microprobe analysis to check SG12 58.05 15.73 10.06 — — 18.54b the element ratios.SG13 74.33 — 20.78 2.80 3.06 — In addition, homogeneous Li minerals (an Li mica and two SG14 69.64 &md 18.17 3.28 — — SG15 62.00 16.34 11.37 3.49 4.17a — muscovites; Table 3) of known compositions (analysed by SG16 62.20 16.07 11.80 5.53 2.65a — ICP-AES) were used to calibrate the Na/Li and Na/K ratios. SG17 59.60 15.40 11.75 3.99 6.21a — These minerals were chosen for their homogeneity in lithium SG18 73.98 — 15.68 5.08 3.11 — and major elements.SG19 75.97 — 12.75 4.98 4.34 — Synthetic fluid inclusions in quartz (Table 4) were prepared SG20 75.68 — 9.74 4.97 7.59 — by hydrothermal synthesis following the methodology of pre- SG21 78.67 — 8.05 8.29 2.24 — SG22 71.12 — 15.20 4.24 7.82 — vious workers.25 Synthetic quartz devoid of fluid inclusions SG23 78.47 — 12.25 4.92 2.28 — was used as the raw material.After fracturing, prisms of SG24 69.60 10.65 5.52 3.93 — — quartz were loaded into gold tubes with the standard solutions. SG25 70.66 10.96 5.90 4.13 — — The capsules were heated in autoclaves at 650 °C and 250 MPa SG26 72.94 15.57 4.56 4.14 0.57a 0.08 for 1 week.A series of synthetic fluid inclusions were kindly SG27 50.00 14.00 5.50 0.20 7.20a 0.17 provided by K. Schmulovich, who performed the synthesis at SG28 81.98 7.42 0.15 — 0.07a 13.16b SG29 50.00 14.00 5.00 0.20 7.20a 2.09 800 °C and 900 MPa for several days. The compositions of the SG30 68.19 — 16.36 4.85 6.97 — solutions before and after experiments were checked by atomic aCorresponds to concentrations where emission line of calcium could absorption spectrometry.Pieces of the synthetic quartz with not be used for the Na/Ca calibration curve owing to the interference inclusions were analysed by the crush–leach technique to with an emission line of aluminium. bCorresponds to the concen- validate the compositions of the trapped fluids. In addition, trations of lithium which could not be used for the Na/Li calibration microthermometry was performed on the synthetic inclusions curve owing to the self-absorption of the Li emission line at 670.7 nm.to check their salinity using a Chaix–Meca stage.26 Table 3 Compositions (in % m/m) of minerals used for the establishment of the calibration curves Component M1 (Li mica) M2 (muscovite) M3 (muscovite) SiO2 50.3 45.73 73.04 Al2O3 25.74 34.89 15.32 FeO (total ) 0 2.45 0.74 Fe2O3 (total ) — — 0.17 MnO 0.4 0.02 <0.01 MgO 0 0.48 0.2 CaO 0 0 0.85 Na2O 0.32 0.48 3.85 K2O 9.94 10 4.96 TiO2 0 0.05 0.13 P2O5 nd nd 0.14 B2O nd nd 0.005 F 7.36 0.19 0.086 H2O (total ) nd nd nd Rb2O nd — nd Li2O 11.24a 0.23 0.073 Sum 97.94 94.33 99.48 Formula [SiO3AlO10|F2]KLi1.5Al1.5 [SiO3AlO10|(OH)2]KAl2 [SiO3AlO10|(OH)2]KAl2 aCorresponds to the concentrations of lithium which could not be used for the Na/Li calibration curve owing to the self-absorption of the Li emission line at 670.7 nm. 916 J. Anal. At. Spectrom., 1999, 14, 913–922dimensions of the crater is around 2×10-9 g per shot. Ablation Calibration curves of a fluid inclusion of 10×10×10 mm requires five shots to This study was focused on the 580–780 nm spectral range, empty all the inclusion, so the calculated ablated mass (only covering the main lines of major elements present in inclusions liquid ) is around 0.5×10-9 g.This value is based on the (Na, K, Ca, Mg, Li, etc.). Na (588.9 and 589.5 nm non- hypothesis that only liquid is ablated. In the case of sodium, resolved by our spectrometer; for the two electronic transitions self-absorption starts from concentrations above 7000 ppm 3p�4d and 3s�3p; resonance transitions), Li (670.7 nm; (Fig. 4), which represents >0.05 mol kg-1 for aqueous solu- 1s22s�1s22p; resonance transition), K [766.5 nm; tions in fluid inclusions. The concentrations of the diVerent 3p6(1s)4s�3p6(1s) 4p; resonance transition] and Ca standards (Tables 2–4) and those found in the natural fluid [786.6 nm, second order of 393.3 nm; 3p6(1s) 4s�3p6(1s) 4p] inclusions are always above this calculated value.Hence for of ionic calcium (Ca II ) emission lines (Fig. 3) were used to fluid inclusion analyses, the self-absorption process is expected establish the calibration curves for the diVerent ion ratios to occur for Na. It is worth noting that the calibration curve (Na/Li, Na/Ca and Na/K) using synthetic glasses, minerals of the emission line of the Na doublet seems to show linear and synthetic fluid inclusions in quartz (Tables 2–4).behaviour with respect to concentration in the range of concentrations where the self-absorption process occurs. Self-absorption Concerning lithium, the shape of the calibration curve is similar to that of sodium and consists of two branches It must be noted that the three emission lines selected for above and below around 3000 ppm in glasses. This threshold sodium, lithium and potassium are the most intense but are for self-absorption process corresponds to 0.12 mol kg-1 H2O also resonance lines.27 It is necessary to determine the consefor fluid inclusions.The concentration of lithium is usually quence of the resonance eVect on the calibration curves. The below this limit, as shown from the analysis of geothermal major problem which can occur with the resonance lines is the fluids.29 It is worth noting that Li-rich fluids can be easily self-absorption process.28 Self-absorption corresponds to the identified from low eutectic temperatures determined by micro- reabsorption of the emitted photons by atoms at the fundathermometric data.30–33 As for Na, the calibration curve mental electronic level, hence the emission intensity is reduced remains linear with concentration (Fig. 5) when the self- (and no longer proportional to the measured concentration). absorption process occurs. For the usual fluid inclusions, the Consequently, during the self-absorption process, the intensity self-absorption process is expected to occur for Na and not of the emission increases linearly with the concentration of for Li.the element in the matrix and the probability of the electronic For calcium (Fig. 6), the intensity data do not show any transition. Hence it is necessary to quantify the element self-absorption: the calibration curve displays a single straight concentration threshold from which self-absorption is a limitline passing through the origin. This result is in complete ing process before any analytical application. The net intensity agreement with the fact that the emission line of calcium is of the diVerent emission lines (used for the establishment of not a resonance line, but a II line coupled with the the calibration curves) was measured versus the element confundamental state.centration in glasses. Self-absorption can be demonstrated The calibration curve for potassium shows a regression line when the intensity of an emission line versus concentration is passing near the origin (Fig. 7). For potassium, the experimen- not a linear function and when the slope decreases with tal data are more diYcult to interpret because the photoca- increase in concentration. In addition, the part of the calithode of the intensifier has a low quantum eYciency at 766 nm. bration curves established for high concentrations does not It is six times less sensitive than the emission intensity for Na pass through the origin. This kind of variation is illustrated at 589 nm.This can explain the fact that the self-absorption in Fig. 4 for the self-absorption of the Na doublet at 589 nm. eVect for potassium cannot be identified even if it really occurs The estimated ablated mass in glasses determined from the Fig. 4 Calibration of the emission line of sodium at 589 nm, to Fig. 5 Calibration of the emission line of lithium at 670.7 nm. An demonstrate the eVect of self-absorption for this line and the limit of this eVect. The arrow indicates the threshold of the self-absorption expansion for values <800 ppm is also represented.The arrow indicates the threshold of the self-absorption eVect. eVect. An expansion for values <6000 ppm is also shown. J. Anal. At. Spectrom., 1999, 14, 913–922 917Before each study, a test is aled out under the same conditions: use of the same optical filter, no argon flow and the record of the spectrum obtained at the second laser shot. This allows the identification of some variations in the focusing of the laser beam or a change in the optical coupling between the microscope and the spectrometer.In addition, this intensity measurement is used as an intensity reference for any comparison of calibration curves. However, in the case of fluid inclusions the repeatability is very poor because the absolute intensity of the emission lines depends on the ablated mass, which can change from shot to shot. Indeed, the focusing of the laser beam on the liquid phase is diYcult to optimise inside the fluid inclusion.Hence the absolute intensity of an emission line of a given element from a fluid inclusion cannot be calibrated versus its concentration (or with respect to water). In addition, no oxygen or hydrogen emission lines are detected which could be ratioed with the intensity of an element and linked with a direct calibration on the molality scale. Therefore, only intensity line ratios of dissolved elements are relevant and standardisation is made with respect to sodium.It can be noted that RSDs found for emission lines ratios are lower in synthetic glasses (between 5 and 15% depending on the concentration of the elements) owing to the good Fig. 6 Calibration of the second order of the emission line of calcium reproducibility of the emission line intensity from shot to shot at 393.3 nm, showing no self-absorption eVect. and the high intensity peaks. The repeatability of ratios even in fluid inclusions is often close to that obtained with glass samples (15–20% against 10–15%), which is acceptable.Reproducibility and repeatability limitations are strongly associated with signal-tobackground (S/B) values. It was demonstrated on glasses that an emission line with S/B <5 leads systematically to a repeatability worse than 20%. This can explain the high RSD for the intensity ratio when one of the two elements has a weak emission line, either due to its low concentration or because the sensitivity of the detector is poor, especially for the resonance line of potassium at 765 nm.Detection limits have been calculated for major elements in fluid inclusions, considering an optimum signal-to-background ratio of 5. The detection limits are for Na and Li 10, Ca 20 and K 750 ppm. These values are divided by a factor of two in glasses. Such detection limits are those required for the determination of ions in individual fluid inclusions. Calibration strategy for the determination of cation ratios For glasses and minerals, 10 series of five shot accumulations are recorded for each standard.The ratio used for the calibration is the average value corresponding to 50 shots. For each fluid inclusion, an average ratio is obtained after around Fig. 7 Calibration of the emission line of potassium at 766.5 nm. 10 shots. The final ratio for each standard is the mean value obtained on 10–12 fluid inclusions (corresponding to 100 with our apparatus.Therefore, samples with low concenspectra). The self-absorption process has been discussed pre- trations of potassium (<3% m/m) could not be studied. viously. The determination of the calibration curves takes into account the possibility of self-absorption for one element or Reproducibility both elements. The calibration curves for Na/Ca, Na/Li and Na/K are For solid samples (glasses and minerals), the shot-to-shot repeatability of the emission line absolute intensities is around given in Fig. 8, 9 and 10, respectively. For the three calibration curves, data from synthetic glasses, minerals and synthetic 10% for glasses (up to 20% for potassium) and in the range 4–10% for the emission line for a metal (e.g., copper). These fluid inclusions plotted along the same curve, attesting that emission line ratios are independent of the studied material. values are in agreement with those found by Geertsen et al.34 with the LA-OES technique.It is worth noting that the noise The RSDs are in the range 5–25% depending on the standards; high RSDs are found for the lowest concentrations of potass- of the electronic detection system limits the reproducibility to about 3%. Hence the repeatability of this technique is close to ium owing to the low intensity of the signal, especially in fluid inclusion standards. the RSD found using the electron microprobe technique for the analysis of solid matrices. The reproducibility of the ablation In Fig. 8, three curves are presented for Na/Ca ratios. The first (D1) is the regression line established from the standards. conditions is controlled with a standard, in our case copper metal. The intensity of the line at 520 nm was selected for its The second (D2) is calculated for low sodium concentrations for which the self-absorption process does not occur. The shot-to-shot stability (RSD#5%) and its good response. 918 J. Anal. At. Spectrom., 1999, 14, 913–922Fig. 10 Calibration curve obtained for Na/K ratios on synthetic Fig. 8 Calibration curve obtained for Na/Ca ratios on synthetic glasses glasses, synthetic inclusions and minerals. and synthetic inclusions. Therefore, a limit of calibration for Na/K ratios of Na (ppm)/K (ppm) <4 had to be used. Analytical strategy for natural samples Two types of calibration curve were established for the determination of the Na/Li, Na/Ca and Na/K ratios: D1 lines for which self-absorption process occurs for Na and D2 lines with no self-absorption process.Therefore, it is necessary to determine which calibration curve is relevant. In the previous section, the threshold of concentration above which the selfabsorption process occurs was established. It corresponds to a given intensity of the Na doublet emission lines. Consequently, below this intensity limit the relevant calibration curves are D2 lines and above it D1 lines are used. Application to natural fluid inclusions Natural fluid inclusions were analysed using LA-OES to check the validity of the method.Samples well characterised by microthermometric, Raman spectroscopic and crush–leach data were chosen for the first application of this technique (Pierre-Joseph clefts, French Alps). In addition, they were selected for their large size (30–50 mm) so as to be able to make several shots on each fluid inclusion. They are two phase Fig. 9 Calibration curve obtained for Na/Li ratios on synthetic glasses, inclusions, without any of solids (Fig. 11). The selection of synthetic inclusions and minerals. these natural inclusions was also based on moderate salinity (7% m/m equiv. NaCl), which is commonly observed in geological fluids. slope of line D2 is four times higher than that of D1, corresponding to high Na concentrations. This calibration Figure 12 shows spectra obtained by several shots on the same fluid inclusion. Na, K, Li and Ca emission lines were curve could be used for materials with low sodium levels (0.05 mol kg-1 H2O).observed. The emission line intensity varies during the ablation progress. However, it is important to note that the variations In Fig. 9, the calibration curve D1 for Na/Li was established from standards for which the self-absorption process occurs of the line intensity ratios shot to shot are within 25%; this indicates that the phenomenon has no influence on the plasma only for Na. The second line, D2, corresponds to the case of a low sodium concentration, below its self-absorption thresh- temperature.It was not possible for all the inclusions analysed to obtain old. It can be noted that its slope is three times greater than that of line D1. the three element ratios because (i) the detection limit for K is around 750 ppm, and the emission line intensity must be For calibration of Na/K (Fig. 10), the calibration curve shows a strong decrease in slope for Na/K >4. This eVect suYciently high to obtain a good SBR which could be then used for the ion ratio; (ii) filters are necessary to decrease the results mainly from the strong self-absorption process of Na, because the samples used have high Na concentrations (>20% Na emission line intensity to prevent saturation of the detector, but these filters simultaneously prevent Ca emission; hence in m/m Na2O).These high Na concentrations were necessary to have a measurable potassium signal for high Na/K ratios.some cases Ca emission lines could not be recorded; and J. Anal. At. Spectrom., 1999, 14, 913–922 919limit of detection for lithium in fluid inclusions, as the salinity is only 7% m/m equiv. NaCl. The RSD is around 30%. However, the RSD for the Na/Li ratio can be better (around 15–20%) for higher Li concentrations. Such an Na/Li determination is accurate enough since lithium concentration cannot be used as marker of fluid–mineral equilibrium. The Na/Li ratio is generally used as an indicator of fluid sources and only its order of magnitude is geochemically relevant.The Na/Ca ratio calculated for 26 fluid inclusions is 9 with an RSD of 20%. For the Na/Ca ratio we have found in some cases a relatively high RSD (close to 40%), which may be due to contamination of calcium. Despite the high RSD values, the technique gives satisfactory results, taking into account that the salinity of the fluid inclusions is around 7% m/m equiv.NaCl. Although the crush–leach technique analyses populations of thousands of fluid inclusions and LA-OES works on individual inclusions, the ratios for the four elements studied are in good agreement (Table 5). Such a result is encouraging for the validation of the LA-OES technique. Fig. 11 Microphotograph of fluid inclusions from alpine quartz, used for the first application of the LA-OES technique. Conclusions (iii) although Li displays a good signal response, it is not always detected in all the inclusions analysed, owing to the This work has shown that LA-OES can be used for localised analysis and especially for the quantification of ion ratios in too small size of the inclusion (<15 mm).Around 30 fluid inclusions from the same piece of wafer individual fluid inclusions. The laser beam is used to drill up to the inclusion and to produce plasma from the trapped were analysed by LA-OES. For each inclusion, the same procedures as those used for synthetic fluid inclusions were liquid.An optical spectrometer analyses directly the radiation emissions from the plasma. Plasma temperature studies showed carried out. The Na/K, Na/Li and Na/Ca ratios were calculated using the calibration curves (Table 5). that several standards (synthetic glasses, fluid inclusions and minerals) can be used for the establishment of calibration The Na/K value found with the calibration curve is 3 with an RSD of 33% for all the analyses. This result seems to be curves.Calibration curves for three element ratios (Na/K, Na/Li and Na/Ca) were established in the 580–790 nm spectral realistic for this geological context. For Na/K ratios, the reproducibility is satisfactory although the SBR is relatively range. The RSDs calculated from the calibration curves range from 5% (glasses) to 25% (fluid inclusions) and the detection small (often <5) as the potassium concentration is too low or the ablated mass of the liquid is too small.The Na/K ratio limits are those required for the determination of ions in individual fluid inclusions (Na and Li 10, Ca 20 and K allows the estimation of the fluid trapping temperature based on the equilibrium between fluid and K and Na feldspars.35 750 ppm). LA-OES is probably one of the techniques that can be used The RSD of the Na/K analysis results in a temperature uncertainty of around±50 °C. Such a temperature determi- to solve the challenge of determining major ion ratios in an average volume of liquid of mass <10-9 g.It is now possible nation is not accurate enough for geological interpretation. However, it could probably be optimised by using a to determine the ratios of major elements (Na, K, Li, Ca) in individual fluid inclusions. A recently available echelle spec- red-sensitive detector. The Na/Li ratio is close to 55 and it corresponds to the trometer has been used in LA-OES and allowed the collection Fig. 12 Optical emission spectra obtained by LA-OES on one natural individual fluid inclusion (three laser shots).The emission lines of Na, Li, K and Ca are used for the determination of the cation ratios via the calibration curves. 920 J. Anal. At. Spectrom., 1999, 14, 913–922Table 5 LA-OES data for 30 fluid inclusions from natural sample (Alpine clefts), their relative standard deviations (RSDs) and their cation ratios calculated from calibration curves No. of RSD RSD RSD Na/K Na/Li Na/Ca Inclusion shots I(Na)/I(K) (%) I(Na)/I(Li) (%) I(Na)/I(Ca) (%) (ppm) ± (ppm) ± (ppm) ± 71–47A 12 166 12 10 24 3 0.3 — 11 2.5 71–47B 11 120 25 21 21 12 24 2 0.4 65 13.7 — 71–47C 9 163 13 17 10 26 3 0.3 — 10 2.7 71–47D 6 126 24 22 26 9 20 2 0.4 69 17.9 9 1.9 71–47E 14 120 23 17 18 16 29 2 0.4 51 9.4 — 71–47F 15 180 9 18 8 15 3 0.3 — 8 1.3 71–47G 18 147 16 20 18 7 40 2 0.4 62 11.4 7 2.7 71–47H 8 190 9 12 25 4 0.3 — 12 2.9 71–47I 4 16 26 8 38 — 45 11.6 8 2.9 71–47J 1 110 18 18 8 8 — 52 9.1 8 0.6 71–47a 5 183 22 9 29 3 — 69 — 9 2.7 71–47b 14 182 31 21 42 10 31 3 1.0 64 26.4 10 3.0 71–47c 1 7 — —7— 71–47d 15 199 7 16 36 10 32 4 0.3 48 17.0 10 3.1 71–47e 4 18 11 19 — 55 — 11 2.1 71–47f 4 16 2 9 11 — 47 1.1 9 1.0 71–47g 2 27 19 8 7 — 85 15.9 8 0.6 71–47h 9 7 17 — — 7 1.1 71–47i 5 7 11 25 — — 11 2.6 71–47j 5 8 11 8 44 — 19 2.0 8 3.7 71–47k 3 14 49 10 18 — 41 19.9 10 1.8 71–47l 6 9 42 — — 9 3.9 71–47m 4 107 6 24 — — 6 1.4 71–47n 6 11 17 — — 11 1.9 71–47o 5 103 11 16 27 — 30 — — 71–47p 4 19 31 9 21 — 58 17.7 9 2.0 71–47q 3 8 37 — — 8 3.1 71–47r 4 15 20 — — — 71–47s 7 5 30 — — 4 1.3 71–47t 7 231 9 22 5 — — 9 2.1 LA-OES 3 54 9 s 1.0 16.1 2 RSD (%) 32.9 29.9 19 Crush–leach 3 69 7 J.Anal. At. Spectrom., 1999, 14, 913–922 92111 T. J. Shepherd and S. R. Chenery, Geochim. Cosmochim. Acta, of the emission spectrum from 200 to 800 nm.36 Such a spectral 1995, 59, 3997. range permits the emission lines of most chemical elements to 12 A.Aude�tat, D. Gu�nther and C. A. Heinrich, Science, 1998, 279, be obtained and will make LA-OES as a multi-element analyt- 2091. ical technique. Its use for fluid inclusion analysis will be 13 D. Gu� nther, A. Aude�tat, R. Frischknecht and C. A. Heinrich, checked in the near future. J. Anal. At. Spectrom., 1998, 13, 263. 14 M. C. Boiron, J. Dubessy, N. Andre�, A. Briand, J. L. Lacour, P. Mauchien and J. M. Mermet, Geochim. Cosmochim. Acta, 1991, Acknowledgements 55, 917. 15 M.C. Boiron, A. Moissette, C. Fabre, J. Dubessy, D. Banks and The EU program MAT1-CT-93–0029 supported this work B. Yardley, in Proceedings of the XIV ECROFI Conference, ed. and P. Mauchien (CEA, LSLA, Saclay, France) is warmly M. C. Boiron and J. Pironon, Nancy, 1997, p. 44. acknowledged for managing the project and coordinating the 16 A. Moissette, J. Dubessy, M. C. Boiron, C. Fabre, P. Mauchien teams to achieve success. J. L. Lacour (CEA, LSLA) is and J. L. Lacour, in Proceedings of the XIV ECROFI Conference, acknowledged for assistance during the advancement of the ed.M. C. Boiron and J. Pironon, Nancy, 1997, p. 211. 17 L. Moenke-Blankenburg, Laser Microanalysis. Chemical Analysis, experimental procedure. The authors thank C. PeiVert a Series of Monographs on Analytical Chemistry and its (CREGU), S. Decitre and L. Tissandier (CRPG) for their Applications, vol. 105, Wiley-Interscience, New York, 1989. help in the synthesis of the glasses which were used for the 18 L.J. Radziemski and D. A. Cremers, Laser Induced Plasma and calibration of the technique, and K. Schmulovich for providing Its Applications,Marcel Dekker, New York, 1989. a series of synthetic fluid inclusions. T. Lhomme (CREGU) 19 U. Paisch, M. Clara and R. Niessner, Spectrochim. is thanked for her help during the experiments. The authors Acta, Part B, 1998, 53, 1957. 20 C. Chale�ard, J. Anal. At. Spectrom., 1997, 12, 183. thank D.Banks (Leeds University) for the crush–leach analysis 21 CEA, Fr. Pat., No. 93-13855, 1993. of fluid inclusions and M. Cathelineau (CREGU) for fruitful 22 N. Andre�, unpublished thesis, Universite� Paris Sud, 1995. discussions during the progress of this study. 23 C. Geertsen, unpublished thesis, Universite� Paris Sud, 1996. 24 P. Mauchien, A. Bengston, R. DeMarco, E. DaSilva, J. Dubessy, F. Noronha, C. Prieto, G. Tomandl and B. W. Yardley, in Final References Report, European Program MAT1-CT-93–0029, 1996. 1 C. Ayora, J. Garcia-Veigas and J. J. Pueyo, Geochim. Cosmochim. 25 S. Sterner and R. J. Bodnar, Geochim. Cosmochim. Acta, 1984, Acta, 1994, 58, 43. 48, 2659. 2 A. J. Anderson, A. H. Clark, P. Max, G. R. Palmer, 26 B. Poty, J. Leroy and L. Jackimowicz, Bull. Miner., 1976, 99, 182. J. D. Macarthur and E. Roedder, Econ. Geol., 1989, 84, 924. 27 Handbook of Chemistry and Physics, ed. R. C. Weast, CRC Press, 3 C. G. Ryan, D. R. Cousens, C. A. Heinrich, W. L. GriYn, Boca Raton, FL, 60th edn., 1980. S. H. Sie and T. P. Mernagh, Nucl. Instrum. Methods B, 1991, 28 M. Autin, unpublished thesis, Universite� Lyon I, 1990. 54, 292. 29 A. J. Ellis, in Geochemistry of Hydrothermal Ore Deposits, ed., 4 M. Volfinger, C. Ramboz, M. Aissa and C. G. Choý�, in H. L. Barnes, John Wiley, New York, 2nd edn., 1979, p. 632. Proceedings of the XIV ECROFI Conference, ed. M. C. Boiron 30 W. F. Linke, Solubilities of Inorganic and Metal–Organic and J. Pironon, Nancy, 1997, p. 344. Compounds, vols. I and II, American Chemical Society, 5 J. D. Frantz, H. K. Mao, Y. G. Zhang, Y. Wu, A. C. Thomson, Washington, DC, 1965. J. H. Underwood, R. D. Giauque, K. W. Johns and M. L. Rivers, 31 A. S. Borisenko, Geol. Geophys., 1977, 8, 15. Chem. Geol., 1988, 69, 235. 32 In Solubility of Inorganic Substances in Water, ed. 6 D. A. Vanko, S. R. Sutton, M. L. Rivers and R. J. Bodnar, Chem. A. N. Kirguintsev, L. N. Trouchnikova and V. G. Lavrentieva, Geol., 1993, 109, 125. 1972. 7 J. A. Mavrogenes, R. J. Bodnar, A. J. Anderson, S. Bajt, 33 M. Cathelineau, Ch. Marignac, M. C. Boiron, G. Gianelli and S. R. Sutton and M. L. Rivers, Geochim. Cosmochim. Acta, 1995, M. Puxeddu, Geochim. Cosmochim. Acta, 1994, 58, 1083. 59, 3987. 34 C. Geertsen, J. L. Lacour, P. Mauchien and L. Pierrard, 8 P. Philippot, B. Menez, P. Chevallier, F. Gibert, F. Legrand and Spectrochim. Acta, Part B, 1996, 51, 1403. P. Populus, Chem. Geol., 1998, 144, 121. 35 S. Verma and E. Santoyo, J. Volcanol. Geotherm. Res., 1997, 79, 9. 9 D. Gu� nther, A. Aude�tat, R. Frischknecht and C. Heinrich, in 36 C. Haisch, U. Panne and R. Niessner, Spectrochim. Acta, Part B, Proceedings of the XIV ECROFI Conference, ed. M. C. Boiron 1998, 53, 1657. and J. Pironon, Nancy, 1997, p. 144. 10 A. Moissette, T. J. Shepherd and S. R. Chenery, J. Anal. At. Spectrom., 1996, 11, 177. Paper 8/09338E 922 J. Anal. At. Spectrom., 1999, 1

ISSN:0267-9477    

DOI:10.1039/a809338e

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

A comparison of lamp control parameters in both bulk and depth profile analysis by glow discharge optical emission spectroscopy Kim A. Marshall LECO Corporation, 3000 Lakeview Avenue, St. Joseph, MI 49085–2396, USA Received 22nd January 1999, Accepted 15th April 1999 The impact of the glow discharge lamp control parameters (voltage, current and pressure) on the quality of the multi-matrix calibration curves, bulk check standard results and quantitative depth profile analysis is investigated.The lack of current and voltage stability is shown to have a larger influence on the quality of these results than do similar changes in pressure. Lamp control modes that achieve both constant current and constant voltage at the expense of constant pressure are found to be best for both multi-matrix bulk analysis and quantitative depth profile analysis. then utilized by the analyst for the characterization of both Introduction and background bulk and layered samples.The temporal variation of these Elemental analysis is an essential part of the industrial analyst’s emission signals is utilized to provide analysis of layered job. It is frequently utilized for determinations required during samples as the layers are sputtered sequentially from the both quality and process control. In the industrial laboratory, sample surface.2,6 bulk elemental concentrations of both raw materials and The so-called plasma parameters (lamp voltage, current and finished products are typically required.More recently, the pressure) control, either directly or indirectly, all of these determination of surface depth concentration profiles of coated processes. Clearly, accurate control of the appropriate plasma and layered materials, also known as quantitative depth pro- parameters is essential to good glow discharge analysis. In files (QDP), are becoming increasingly more important. Glow practice, one parameter, the voltage (E), current (i) or power discharge optical emission spectroscopy (GD-OES) is an (W), can be controlled by the lamp power supply.The pressure accepted technique for providing the information necessary (P) of the lamp is the other variable that can be directly for both types of analyses.1–5 The glow discharge lamp emission controlled. This pressure will then determine the third lamp source is the heart of this technique. In the work presented parameter. Given any two parameters, there is only one value here, diVerent approaches to glow discharge lamp control are of the third parameter that will produce a stable plasma with studied and compared. These comparisons are made using a given sample.This is because these parameters are essentially multi-matrix methods similar to those required for the related by Ohm’s law: calibration of quantitative depth profile analysis. The relative simplicity of the glow discharge source makes E=i R(P,mx) (1) it well suited for both bulk and depth profile analysis.In glow discharge, a low-pressure inert gas environment (usually argon) where E is the lamp voltage, i is the current and R(P,mx) is is maintained over the sample. A large negative voltage (e.g. the plasma resistance which is a function of the lamp pressure 500–1500 V) is applied to the sample, making the sample the P and the sample material mx being sputtered. The plasma cathode. The anode of the glow discharge lamp is grounded gas composition, composed of various amounts of argon gas, and held at a small fixed distance from the sample surface.In sample material and electrons, most likely determines the modern commercial instruments, this anode is often cylindrical plasma resistance. The particular sample material being sputtin geometry with a concentric annular orifice for support gas ered is one significant factor influencing this composition. flow. This common design is called a Grimm lamp.2 Electrons Therefore, it is not surprising that the plasma resistance can emitted from the sample under the influence of the electric change dramatically for diVerent sample materials.This field are accelerated across the sample anode gap. These high observation is reported both in this work and elsewhere.6,7 velocity electrons collide with fill gas molecules. These colli- Although it is true that only one power parameter (E, i or sions cause plasma formation and the production of argon W) and pressure can be directly controlled, this does not mean ions and more electrons.The argon ions produced in the that the other parameters cannot be controlled indirectly. It is plasma are in turn accelerated towards the sample surface. On possible to control any one of the other power parameters the way to the sample, the argon ions also collide with other through feedback controlled pressure. For example, one might gas molecules. The process is self-sustaining over a wide range choose to control the lamp voltage with the lamp power of sample to anode voltages and gas pressures.supply, and then the lamp current by varying the lamp pressure Argon ions, accelerated by the lamp potential, strike the such that the current remains at some predetermined constant sample. Here they sputter material from the sample surface. value. Alternatively, the lamp current could be controlled by Some of this material migrates into the plasma where it is the power supply and the lamp voltage held constant by further dissociated, vaporized, atomized and finally excited.feedback controlled pressure. Power could also be controlled Some materials may also be ionized and excited-state ions by the power supply and either voltage or current fixed by may also be produced. It is this process of sampling and feedback to the pressure controller. Obviously, all of these approaches have a similar result—that of holding both lamp excitation which gives rise to the emission signals which are J.Anal. At. Spectrom., 1999, 14, 923–928 923voltage and current constant while allowing the lamp pressure accepted2,6,11,16 that Boumans equation17 expressed as: to vary in order to provide this control. q=Sy i [E-St] (4) Although there are many possible lamp parameter control is a good description of the relationship between current (i), schemes, two approaches have emerged as the most common voltage (E) and the resulting sputter rate (q).Here Sy and St modes of operation.6 These are voltage–current and power– are material-specific constants that determine the sputter yield pressure. In the voltage–current (VC) approach, the lamp and sputter threshold respectively. Note that this equation power supply is set to deliver a constant voltage (e.g. 1000 V) would imply that, in order to provide a constant sputtering and the pressure in the lamp is then maintained at the value rate, one must keep both voltage and current constant.necessary to keep the current at some fixed value, e.g. 30 mA. Constant power would not necessarily provide the same con- This is accomplished by either a software or hardware feedback trol. The eVective voltage [E-St] in eqn. (4) is that portion loop. In a closely related mode, current–voltage (CV), the of the applied voltage, above the sputter threshold, which is parameters are switched. Constant current is supplied by the active in the sputtering process.Therefore, a doubling of the lamp power supply and the pressure controller is used to hold applied voltage, e.g. from 500 V to 1000 V, at constant power, the voltage constant. While the overall eVect of these two would produce a sputter rate increase of 1.75 (assuming a modes is the same, the CV mode tends to stabilize faster on typical value of 300 V for St). plasma startup. This is important for thin layer QDP analysis. The emission yield, Ynm, is also a function of the discharge In the power–pressure (PP) approach, the lamp power supply parameters.Unfortunately, there is no general agreement as is directed to supply a desired fixed power to the lamp (e.g. to how Ynm responds to changes in E, i or P. In a recent study, 30W) and the pressure in the lamp is fixed at a specified value, Bengtson10 concluded that, while all three plasma parameters e.g. 10 Torr. In this approach, the power will be supplied at the have some impact on Ynm, the emission yield is about four particular combination of voltage and current that not only times more sensitive to changes in voltage or current than it produces the power level requested, but is also compatible with is to changes in pressure. With minor variations, an expression the lamp pressure and sample type being analyzed.This might similar to: be 30 mA at 1000 V or 40 mA at 750 V or any similar combination that provides 30 W of power. The specific combination Ynm=k ia (E-St)b f(P) (5) that is stable is dependent not only on the pressure, but also has been used by various researchers6,11 to describe the on the sample matrix being sputtered at a given time.Therefore, relationship between emission yield and the lamp parameters. it can vary with time on a layered sample where diVerent In this expression, k is a proportionality constant and the materials are sputtered at diVerent depths in the specimen.2 factor f(P) is a function of pressure.The exact nature of this The PP control mode can be utilized in a dc glow discharge, pressure function is still in question.10,13 as it was in this study. However, it is more frequently applied It is apparent from the above discussion that, in order to on radiofrequency glow discharge (rf-GD) systems. This is achieve the best analytical results from glow discharge, one because many commercially available rf-GD systems are must have control of the appropriate discharge parameters.In strictly limited to this control mode due to the diYculty of the work presented here, we try to answer the question of controlling both voltage and current in rf-GD. While, in this which control parameters provide the best analytical results. work, only dc glow discharge is studied, it is probable that many of the observations made here could be extrapolated to Experimental rf-GD as well. Obvious questions arise from this situation. Which of the The data in this study were obtained on a GDS-750A (LECO lamp control modes provides the best possible results in glow Inc.St. Joseph, MI, USA). This is a 0.75 m vacuum spec- discharge analysis? Should diVerent modes be used for diVerent trometer. The standard 4 mm Grimm type dc glow discharge purposes? While there has been some disagreement on this lamp was used in this work. This instrument can control the subject,8–13 it is still useful to review the prevailing theories. lamp discharge parameters in any of the modes chosen for The experimentally measurable quantity in all emission this study, as well as some that were not investigated.The spectroscopy is the emission intensity of a given element n for three modes utilized were PP, VC and CV. The operation of a given analytical line m denoted as Inm. In glow discharge, these modes is as described above. All analyses were run at this quantity is determined by:1 an applied power of 30 W. In the VC mode, the voltage was Inm=cn q Ynm+Ibk (2) held at 1000 V by the lamp power supply, and the gas pressure was controlled to hold the current constant at 30 mA.The where cn is the concentration of the element n in the sputter pressures required to achieve these values on diVerent materials sample, q is the sputter rate, Ynm is the emission yield and Ibk are listed in Table 1. In the PP mode, the power was controlled is the intensity of the background signal. The emission yield by the lamp power supply to 30 W, and the pressure was held is essentially the emission flux (e.g.photons s-1) per atom of at 9 Torr in the lamp. This pressure was chosen, as it is analyte n at the wavelength m. Obviously, instrumental consistent with the pressure required in the VC mode to run collection eYciencies and transducer constants are also both the copper and iron samples. Table 1 also contains the involved in determining the measured signal intensities. These applied current and voltages observed in the PP mode on these instrumental constants are not represented in eqn.(2). If selfsame materials, as well as the signals observed for four pure absorption becomes significant, then the model must be materials in both control modes. changed to include a self-absorption factor.14,15 In the work A multi-matrix calibration was performed for both modes presented here, an inverse relationship was used: of operation. This calibration was performed with 32 standards cn q=(Inm-Ibk)/Ynm (3) in four matrices: iron, copper, zinc and aluminum.Table 2 lists the certified standards used, categorized by major compo- which infers the analyte flux (cnq in g s-1) from the background corrected intensity (Inm-Ibk) and emission yield. A second nent. In order to keep the calibration of both modes of lamp control consistent, the following calibration procedure was order fit between the quantities (cnq) and (Inm-Ibk) can be used as well. Such a quadratic approximation is another followed. Three signal integrations were collected on each standard at the conditions prescribed above for each mode of approach to providing a better fit in cases where self-absorption is significant.14 operation.The concentration for each standard, multiplied by the standard’s known sputter rate, was then plotted against Of the three independent factors in eqn. (2), both q and Ynm are functions of the lamp control parameters. It is fairly the average signal intensity observed for that standard.Table 3 924 J. Anal. At. Spectrom., 1999, 14, 923–928Table 1 Condition comparison Intensity in VC mode/V Intensity in PP mode/V (at 1000 V, 30 mA) (at 30 W, 9 Torr) Pressure VC mode/ Voltage Current Matrix Torr Fe Cu Al Zn PP mode/V PP mode/mA Fe Cu Al Zn Pure iron 8.9 0.743 1018 29.8 0.753 Pure copper 8.9 2.78 998 30.3 2.82 Pure aluminum 6.2 1.46 692 43.7 2.38 Zinc (96%) 6.7 2.24 784 38.7 4.56 Table 2 Certified standardsa Iron Check Copper Check Aluminum Check Zinc Check BS50B BCS197f BCS198f NIST 626 X NIST 1761 X IARM 72a SUS CAL3–1 NIST 1765 IARM 89a Alcoa SQ11-h05 NIST 1767 MBH17867-t Alcoa SQ12-kt NIST 1768 MBH17868-t Alcan 5000AC X IARM 28b MBH17869-s SUS RA14–4 IARM 29b MBH17870-r SUS RA14–7 IARM 30b CKD 304a IARM 31b CKD 305a IARM 32b CKD 306a IARM 34b CKD 307a CKD 308b CKD 309b X aBCS—British Chemical Standards, Bureau of Analyzed Samples Ltd., Newham Hall, Middlesbrough, UK.BS—Brammer Standard, Houston, TX, USA.CKD—CKD Research Institute, Prague, Czech Republic. IARM—Analytical Reference Materials International, Evergreen, CO, USA. NIST—National Institute of Standards and Technology, Gaithersburg, MD, USA. Alcoa—Alcoa Specialty Metals, New Kensington, PA, USA. Alcan—Alcan International Ltd., Jonquiere, Quebec, Canada. MBH—MBH Analytical Ltd., Barnet, Hertfordshire, UK. SUS—Ulrich Nell, Siegmundstr., 8, D-46149 Oberhausen, Germany. lists the RMS errors of the linear least-squares curve fit in Table 5 contains a list of five diVerent layered materials which were analyzed by QDP under both the PP and CV both modes of operation for the 15 elements included in this study. For each element, the table indicates the order of fit, if modes.This table also contains the depths of certain known features observed with these two methods. The materials 1/concentration weighting was used to improve the lower end curve fit and any inter-element corrections (IECs) used.Except designated as Kocour Standards are coated reference materials obtained from Kocour, Chicago, IL, USA. for the NIST 626 standard, used as the high standard on the Zn curve, the four standards designated as checks in Table 2 were not included in the least squares fit of the calibration Results and discussion curves. Thus, they are independent checks of the fitted curves. These check standards were chosen to be roughly representa- An examination of Table 1 reveals some obvious diVerences between the PP and VC modes of lamp parameter control.In tive of the four diVerent material types used in this calibration. NIST 1761 is a relatively high alloy piece from the NIST 1760 the PP mode, the power and pressure are held at 30W and 9 Torr respectively, but the current and voltage are clearly not series low alloy steel set. CKD 309b is a Cu/Al brass. The Alcan 5000AC piece is an Al/Cu standard. The high zinc piece constant.The voltage changes from a high of 1020 V on iron to a low of 690 V on aluminum. The relative standard deviation in the calibration set, NIST 626, is also used as a check for the other elements it contains. The results of the check standard of the applied voltage is almost 20%. The variation is even greater if one considers the eVective voltage changes (the analyses are tabulated in Tables 4(a)–4(d). Table 3 Regression parameters Element Wavelength Order Weighted IECs VC rms error PP rms error PP/VC Al 396.15 2 Yes 1.0 4.7 4.7 C 165.70 1 No 0.0044 0.0087 2.0 Cr 425.43 1 No 0.017 0.072 4.2 Cu 327.39 2 Yes 0.29 2.6 9.0 Fe 371.99 2 Yes Si 1.9 2.2 1.2 Mg 383.82 1 Yes 0.16 0.22 1.4 Mn 403.44 2 Yes 0.090 0.20 2.2 Mo 386.41 1 Yes Al 0.0021 0.0029 1.4 Ni 341.48 2 No 0.075 0.097 1.3 Pb 405.78 1 No 0.019 0.025 1.3 Si 288.15 1 No 0.021 0.074 3.5 Sn 317.50 2 Yes Fe 0.027 0.032 1.2 Ti 337.27 1 Yes 0.0092 0.0093 1.0 V 318.54 2 Yes 0.0010 0.020 20.0 Zn 334.50 1 Yes Fe 0.72 2.3 3.2 Average 3.8 J.Anal. At. Spectrom., 1999, 14, 923–928 925Table 4(a) CKD 309b Table 4(d) NIST 626 PP/VC VC mode PP mode PP/VC error VC mode PP mode error Certified factor Certified factor Element (%) Average Error Average Error (9 reps) Element (%) Average Error Average Error (6 reps) Al 3.56 3.24 0.33 6.71 3.15 27.4 Al 12.7 13.14 0.44 11.56 1.14 1.8 C C Cr 0.0395 0.05 0.010 0.09 0.047 0.17 Cr Cu 82.6 82.38 0.22 83.30 0.70 1.5 Cu 0.056 0.05 0.01 0.11 0.06 0.33 Fe 0.103 0.14 0.037 0.19 0.085 0.11 Fe 0.86 0.90 0.038 0.89 0.033 0.0 Mg Mg 0.02 0.01 0.008 0.01 0.007 0.00 Mn 0.048 0.06 0.008 0.10 0.048 0.23 Mn 1.02 1.07 0.047 1.07 0.048 0.0 Mo Mo Ni 0.047 0.06 0.008 0.07 0.024 0.05 Ni 1.71 1.73 0.022 1.75 0.044 0.0 Pb 0.050 0.06 0.010 0.05 0.001 -0.2 Pb 0.0022 0.00 0.002 0.00 0.002 0.00 Si 0.042 0.08 0.033 0.14 0.099 0.20 Si 0.13 0.13 0.004 0.13 0.003 0.0 Sn 0.53 0.50 0.032 0.48 0.049 0.0 Sn 0.0012 0.00 0.001 0.00 0.001 0.00 Ti 0.039 0.00 0.037 0.00 0.035 0.00 Ti V V 0.012 0.00 0.12 0.00 0.012 0.00 Zn (96.1) 96.32 92.60 Zn 0.22 0.12 0.097 0.76 0.542 2.5 Check standard average 2.4 Check standard average 0.6 Overall average 5.3 Table 4(b) NIST 1761 PP/VC VC mode PP mode error current is also about 20%, but in the opposing direction.Using Certified factor eqn. (4), we can calculate the relative sputter rate for the Element (%) Average Error Average Error (6 reps) environment for the conditions observed in the two control modes (Table 1).This relationship indicates that q for the Al 0.055 0.05 0.01 0.09 0.03 0.1 same material would be lowered by about 20% for the con- C 1.03 1.03 0.001 1.10 0.072 10.2 Cr 0.22 0.21 0.011 0.21 0.013 0.0 ditions applied to the aluminum sample in the PP mode vs. Cu 0.30 0.26 0.04 0.36 0.06 0.0 the conditions applied to the iron sample using the same Fe (95.0) 92.65 94.60 control mode. Note that this does not mean that aluminum Mg sputters just 20% slower than iron. In fact, aluminum sputters Mn 0.678 0.68 0.006 0.69 0.007 0.0 much more slowly (qFe=3.4 mg s-1, qAl=1.36 mg s-1 at 700 V Mo 0.103 0.09 0.009 0.09 0.009 0.0 and 20 mA), but this is due to the material diVerence not the Ni 1.99 2.37 0.377 2.45 0.462 0.1 Ob sputtering environment.The Boumans relationship [eqn. (4)] Si 0.18 0.19 0.005 0.20 0.022 0.1 indicates that, in the VC control mode, the sputtering environ- Sn ment is unchanged. Both voltage and current are constant in Ti 0.18 0.12 0.062 0.09 0.091 0.0 this mode.However, to achieve this constant voltage and V 0.053 0.04 0.009 0.03 0.025 0.0 current environment, the pressure stability must be sacrificed. Zn In the VC mode, the relative standard deviation of the pressure Check standard average 1.06 is 20%, as observed for voltage and current in the PP mode. The intensities observed in Table 1 indicate that more is Table 4(c) Alcan 5000AC occurring than just sputter rate changes.The intensities for both the pure iron and copper samples are very similar in PP/VC both control modes. This is because the plasma is running VC mode PP mode error under very similar conditions in both control modes (~1000 V, Certified factor Element (%) Average Error Average Error (9 reps) 30 mA at 9 Torr). This is not true for either the pure aluminum or the 96% zinc sample. Here, the lower pressure in the VC Al (88.7) 87.78 69.77 mode and the lower voltage and higher current in the PP C mode have dramatic impacts on the observed intensities.The Cr 0.042 0.04 0.003 0.05 0.009 0.0 intensities measured in the PP mode are 1.6 and 2.0 times Cu 10.32 11.27 0.95 26.50 16.18 100.0 higher for the Al and Zn pieces respectively than for these Fe 0.33 0.47 0.136 0.61 0.276 0.3 Mg 0.042 0.04 0.002 0.04 0.004 0.0 same pieces in the VC control mode. These observed increases Mn 0.034 0.00 0.034 0.00 0.034 0.0 are in opposition to the predicted sputter rate changes dis- Mo cussed above.This is, however, consistent with the observation Ni 0.045 0.02 0.025 0.03 0.016 0.0 of Bengtson.11 In this study, Bengtson found that the emission Pb 0.042 0.04 0.001 0.09 0.053 1.9 yield is almost inversely proportional to voltage (E-0.93) and Si 0.2 0.25 0.051 0.45 0.250 1.0 ‘the emission yield decreases with pressure’. These findings Sn 0.042 0.06 0.018 0.09 0.045 0.1 Ti 0.039 0.04 0.001 0.04 0.005 0.0 were for the Si 288.1 nm line, but the above observations V 0.012 0.00 0.012 0.00 0.009 0.0 indicate somewhat similar behavior for both the Zn 334.5 nm Zn 0.044 0.00 0.044 2.34 2.292 100.0 and Al 396.1 nm lines used in the present study.Check standard average 16.9 The large pressure drop in the VC control mode and the higher current at lower voltages in the PP mode, observed when running an aluminum or zinc sample as opposed to an iron or copper piece, indicate that there are large plasma portion above the sputter threshold).The relative standard deviation of the eVective voltage is almost 30%. To keep the resistance implications when sputtering these diVerent materials. It is beyond the scope of this paper to speculate on the applied power constant, current must change in the opposite manner. Therefore, the relative standard deviation of the causes of these resistance changes. 926 J. Anal. At. Spectrom., 1999, 14, 923–928Table 5 Depth profile analysis Feature depth Sample Description (previous) CV mode depth PP mode depth 1037 m-inch Zn/steel Zn on steel 26.3+/-1.3 mm 24.7–25.6 mm 18–18.2 mm Kocour Standard Cu/Ni/steel Kocour Cu on Ni on Cu on steel 22.8+/-2.3 mm 20.4 mm 17.5 mm Standard (to steel interface) (to steel interface) (to steel interface) Al coat Al/Si coating on steel (24 mm) 24 mm 21mm GalvAneal Zn/Fe coating on steel (6.9 mm) 7.6–7.8 mm 4.3–4.4 mm GalvAlum ~50550 Zn/Al on steel (30 mm) 29 mm 30mm The issue under consideration here is which control mode apparent in the VC calibration.This is certainly due to selfabsorption at the higher Al concentrations. As the fit is provides the most accurate analysis. The calibration results excellent and the curve never comes close to becoming double compiled in Table 3 shed some light on this question. This valued, this does not pose an analytical problem. While this table contains the root mean square errors (rms errors) for line almost certainly possesses the same behavior with the PP the calibrations obtained in both the PP and VC control control mode, it is not obvious from the resulting calibration modes.The last column of the table shows the ratio of the curve. The PP mode’s calibration curve shows no evidence of rms error for the PP mode to that of the VC mode for the 15 self-absorption. In fact, the fit is poor, but it exhibits the elements studied in this work. While the diVerences are not opposite behavior. How can this be? The aluminum standards large, the VC mode does appear to have a clear advantage by (see Fig. 1) all have higher intensities than one would expect. this measure. Smaller rms errors are desirable and, on average, This is, however, consistent with an emission yield increase as the VC fitting errors are almost four times smaller than those the voltage drops and the current increases, as is seen in the obtained in the PP mode. Additionally, these values ranged Al standards (intensity=2.28 at 730 V, 41 mA) as compared from unity (equal ) to a high of 20.This ratio was never found with the Cu/Al standards (intensity=1.48 at 1020 V, 30 mA). to be less than unity in this investigation, indicating that the This is in agreement with the observed emission yield behavior linear fits obtained with the VC mode were always at least as already discussed above for the pure Al standard. good and usually better than those obtained with the PP mode. Closer examination of the PP mode calibration curve for Al Fig. 1 provides a comparison of the Al 396.15 nm calibration [Fig. 1(b)] reveals that it is made up of at least two diVerent curves for the two diVerent control modes. Because of the calibration curves. The iron and copper materials form one particular make-up of the standards used in this calibration, line. In the PP mode, these materials run at plasma conditions this line provides a good deal of information about the of about 1000 V and 30 mA.This is similar to the conditions behavior of the two diVerent control modes. First, the fit for used in the VC mode. Another line can be imagined running Al in the VC mode is mathematically 4.7 times better than the through the one 96% zinc standard and then through the fit in the PP mode (see Table 3). In appearance the comparison aluminum standards. These materials run at lower voltages is perhaps even greater. Both modes were fitted with 29 of the and higher currents (approximately 750 V and 30 mA).This 32 standards, only three non-certified for Al, pure copper suggests that it is the lack of voltage and current control that standards were excluded. A very distinct second order fit is causes the poor fit in the PP control mode. Calibration results are one thing, but what impact does this have on the actual analytical results? Tables 4(a)–4(d) contain the analytical results obtained from the four check standards in both the VC and PP control modes.These results are averages of either six or nine replicate analyses as indicated in the tables. Tables 4(a)–4(d) also contain the certified values and errors in the measured values for each control mode. Missing values indicate non-certified elements in that particular standard. The last column is an error factor, which indicates the relative error between the PP mode vs. the VC mode. This factor is weighted by the magnitude of the error. This value is positive if the PP error is larger, negative if the VC error is larger and zero if the magnitudes of the errors in both modes are essentially the same.In this way, the average of these values is an indication of just how the methods compare on a given standard. Note that, with the exception of the CKD 309b standard, the matrix element is non-certified. For these standards, the nominal matrix concentration is shown for comparison, but not used in the calculated averages. The results presented in Tables 4(a)–4(d) indicate that the VC mode obtains significantly more accurate results on three of the four check standards.Only on the iron standard is there little diVerence. This is to be expected as both control methods use similar voltage and current conditions on iron based materials. Conversely, the largest diVerences are observed on the aluminum check standard. This material runs at very diVerent conditions in the two control modes. Of particular note are the extremely inaccurate results obtained in the PP mode for Cu and Zn (as well as the matrix, Al ) in the Alcan 5000AC Fig. 1 Aluminium calibration curves. standard. The PP mode also has significant diYculties with the J. Anal. At. Spectrom., 1999, 14, 923–928 927analysis of Al and Zn in the CKD 309b copper material, as to accurately deconvolute into a usable concentration vs. depth map of the analyzed material. well as Al in the NIST 626 Zn material. Both methods have In the CV (or VC) control mode, the voltage and current some problem with Ni in the NIST 1761 low alloy steel.It are held constant. Thus, in this mode, the sputtering conditions should be noted, however, that this standard and one other remain constant with time. The sputter rate changes only in similar (2% Ni) steel material were obvious outliers on both Ni response to changes in the material being sputtered. Also, the calibration curves. No inter-element correction or other eVect emission yield should remain relatively constant.While it is could be found that accounted for this displacement. These true that pressure changes occurring in the CV control mode points were excluded from both curves and would have been will have some eVect on the emission yield, these changes are even if, as in the case of NIST 1761, they were not being used small compared to the fluctuations produced by similar as independent check standards. These errors, while not underchanges in voltage and/or current.11 In QDP analysis using stood, are not considered to be related to the method of plasma the CV control mode, the measured emission signal vs.time control used during the analysis. For Cu and Fe in the Alcan profiles can be directly mapped into a sputter 5000AC standard somewhat inaccurate results are observed rate×concentration vs. time relationship with the use of using both control modes. Even so the VC mode is still more appropriate calibration curves.As all emission signals collected accurate for these and most other elements in this standard. at a particular time must be a result of a layer with the same The average error factor for all four check standards is 5.3, sputter rate, it is easy to determine this sputter rate (the sum indicating that the VC mode is much more accurate than the of all elements must equal 100%). This ease of calculation PP mode on these materials. gives the CV control mode a distinct advantage, particularly It is important to note that the results obtained on these during QDP analysis.check standards were generated using very broad multi-matrix calibrations. Therefore, these results should not be used to Conclusions judge the ultimate capability of glow discharge on bulk applications. In a typical bulk application, such extremely varied Clearly, this comparison of the VC (CV) and PP plasma control modes shows significant diVerences. The large variation materials, i.e.iron, zinc, aluminum and copper based alloys, of voltage and current observed when using the PP mode in a would not be analyzed on the same calibration curve. This is multi-matrix analysis was determined to be a very big disadvan- an enormous challenge, not even possible by most other tage, as compared to the less critical pressure fluctuations analytical techniques. However, when analyzing coated mateoccurring with the CV control mode for the same analysis. rials (e.g. GalvAneal or GalvAlum), this kind of multi-matrix This is particularly important in quantitative depth profile calibration is more typical.The same extreme accuracy necesanalysis where the multi-matrix nature of the samples makes sary in a bulk analysis is not usually desired or required in the avoidance of this problem impossible. However, even in such QDP analysis. bulk material analysis, this weakness severely limits the The eVect of the plasma parameters on the analysis of applicability of the PP mode to very narrowly defined coated materials by QDP is exhibited in both the calculated matrix-matched applications. elemental analysis of a given coating and the depth measurements provided during these analyses.Table 5 lists the diVerent Acknowledgements samples analyzed in both CV and PP modes and the calculated depths of distinct interface features in these materials. The The author wishes to thank the LECO Corporation for support feature depths in parentheses are not certified, but indicate of this project, and both Joel Mitchell and Dr.Arne Bengtson previously run results and therefore may be biased towards for their comments and suggestions. the CV mode as that approach was used in the previous determinations. The depths of the two Kocour Standards are References ‘known’ values determined by an independent chemical method 1 W. W. Harrison, C. M. Barshick, J. A. Klingler, P. H. RatliV and by the vendor.Y. Mei, Anal. Chem., 1990, 62, 943A. There seems to be a small but significant bias between the 2 A. Bengtson, Spectrochim. Acta, Part B, 1994, 49, 411. depths determined in the PP mode and the ‘known’ or pre- 3 W.W. Harrison, J. Anal. At. Spectrom., 1992, 7, 75. viously determined feature depths. In particular, in the PP 4 A. Bogaerts and R. Gijbels, Spectrochim. Acta, Part B, 1998, mode, the analyzed feature depths for both Kocour Standards 53B, 1. 5 Z.Weiss and K. A. Marshall, Thin Solid Films, 1997, 308/309, 382.are on the order of 25–30% lower than the ‘known’ value. On 6 R. Payling, in Glow Discharge Optical Emission Spectrometry, ed. the other hand, the CV mode agrees quite well with the R. Payling, D. G. Jones and A. Bengtson, Wiley, Chichester, Kocour ‘known’ values as well as previously run depth values 1997, pp. 20–47. from the other materials. These depths are again a function 7 A. Bengston, in Glow Discharge Optical Emission Spectrometry, of the calibration accuracy in the two control modes. ed. R. Payling, D. G. Jones and A. Bengtson, Wiley, Chichester, 1997, pp. 192–197. There are major ramifications of the lack of voltage and 8 A. Bengtson, A. Eklund, M. Lundholm and A. Saric, J. Anal. At. current stability in the PP control mode during a QDP analysis. Spectrom., 1990, 5, 563. Remember that the sputter rate of any material is controlled 9 Z. Weiss, J. Anal. At. Spectrom., 1994, 9, 351. by the current and eVective voltage [eqn. (4)]. Also remember 10 R. Payling and D. Jones, Surf. Interface Anal., 1993, 20, 787. 11 A. Bengston, J. Anal. At. Spectrom., 1998, 13, 437. that these parameters change in the PP mode as diVerent 12 R. Payling, Surf. Interface Anal., 1995, 23, 12. materials are sputtered (see Table 1). Finally, the emission 13 R. Payling, in Glow Discharge Optical Emission Spectrometry, ed. yield also changes as the plasma conditions vary. All of these R. Payling, D. G. Jones and A. Bengtson, Wiley, Chichester, 1997, changes are convoluted in the PP control mode to make the pp. 483–496. measured intensity vs. time relationship extremely complicated. 14 Z. Weiss, J. Anal. At. Spectrom., 1997, 12, 159. 15 R. Payling and D. G. Jones, in Glow Discharge Optical Emission While the signals measured at a particular time must by Spectrometry, ed. R. Payling, D. G. Jones and A. Bengtson, Wiley, definition come from a layer with the same sputter rate, the Chichester, 1997, pp. 460–471. emission yield may have changed dramatically. It is therefore 16 Z. Weiss, Spectrochim. Acta, Part B, 1993, 48, 1247. diYcult to use an appropriate calibration curve to calculate 17 P. W. J. M. Boumans, Anal. Chem., 1972, 44, 1219. the (sputter rate×concentration) for each element. In the PP control mode, these relationships are diYcult, if not impossible, Paper 9/00628A 928 J. Anal. At. Spectrom., 1999, 14, 923–928

ISSN:0267-9477    

DOI:10.1039/a900628a

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Direct determination of selenium in serum by electrothermal atomization laser-induced fluorescence spectrometry D. J. Swart and J. B. Simeonsson* Department of Chemistry, University of Iowa, Iowa City, IA 52242–1294, USA Received 8th January 1999, Accepted 29th March 1999 A technique for the determination of serum selenium by electrothermal atomization laser-induced fluorescence (ETA-LIF) spectrometry is discussed. Far-UV radiation used for Se excitation at 196.026 nm was generated as the second anti-Stokes stimulated Raman shift of the output of a frequency doubled dye laser operating near 234 nm.Owing to the high spectral selectivity and high sensitivity of the ETA-LIF technique, direct determination is possible by using a simple aqueous dilution (1510 or 1520) of the serum samples and a Pd–Mg chemical modifier to reduce pre-atomization losses of the analyte. The accuracy of the method was demonstrated by analysis of NIST SRM 1598 Inorganic Constituents in Bovine Serum (experimental value 42±3 ng g-1; certified value 42.4±3.5 ng g-1).Analysis of 128 serum samples drawn from bone marrow transplant patients yielded values of 89±20 and 92±23 ng g-1 for pre- and post-operative samples, respectively. of standard additions. This procedure is time intensive, as it Introduction requires the analysis of several spiked solutions for each Selenium (Se) is an essential trace element for many living sample, and reduces the precision of the measurements.A organisms and it is unusual in that it is often toxic at levels second method is calibration by use of matrix matched stanthat are only slightly higher than those considered healthy.1 dards, which reduces the time requirement of standard While having a low activity in atomic form, mechanisms addition, but does not address the possibility of variations in involving its oxidized and reduced forms are now being the matrix from sample to sample. A diVerent approach to uncovered as vital contributors to a wide range of biological this problem is the removal of the analyte from the sample functions.Selenium is well known to be an integral component matrix prior to measurement. This is most often accomplished of the enzyme glutathione peroxidase, which reduces hydrogen by hydride generation after a rigorous sample digestion. These peroxides and organic lipoperoxides, and is an important digestions are time intensive and require the use of strong, factor in cellular defenses against oxidative damage.2 It has high purity reagents, such as hydrogen peroxide and nitric also been found to exhibit anticarcinogenic properties3 and to and perchloric acid, that generate hazardous products.be strongly related to immunity and resistance to infections in Although hydride generation can be coupled with both ETAAS animals and humans.4 and ICP-MS, its high degree of sample treatment makes it In recent years, it has been shown that nutritional Se is a prone to contamination and less attractive for large numbers significant factor in determining the severity of infections by of serum samples.Coxsackie virus in mice, suggesting that nutritional deficiencies Laser-induced fluorescence (LIF) spectrometry is a powerful of Se could also be an important factor in the mutation and ultra-trace method that has extraordinarily high sensitivity development of highly virulent forms of other diseases, e.g., and selectivity.When LIF is utilized for elemental analysis, it influenza, hepatitis and HIV.5,6 Epidemiological studies have is also referred to as laser-excited atomic fluorescence specalso indicated an increased risk of prostate cancer with low trometry (LEAFS). The combination of electrothermal atomiserum Se7,8 and reduced risk of prostate cancer when Se is zers and laser-induced fluorescence (ETA-LIF or ETAprovided as a dietary supplement.9 Nevertheless, the relation- LEAFS) in particular has been shown to provide unparalleled ships of Se status to a range of health outcomes remain performance for ultra-trace elemental analysis measurements, diYcult to assess, in part owing to inadequate measurement particularly when the sample amounts are limited.17–19 The methods for assaying biological Se species.limits of detection for this approach are typically 100–1000 Measurements of Se in mammalian sera have often been times lower than those achieved by ETAAS and compare performed by electrothermal atomization atomic absorption favorably with those obtained by ICP-MS.Absolute mass spectroscopy (ETAAS)10–12 and in some cases by inductively detection limits for many elements are in the femtogram and coupled plasma mass spectrometry (ICP-MS).13 Direct deter- even sub-femtogram range, including Au,20 Tl,21 Pb,22,23 Bi,24 minations of Se by these approaches, via aqueous calibration, Sb,25 Te,26 Se,27,28 In29 and Au, Ir, Rh, Pd and Pt.30 It is are often hindered by the complex nature of the sera and worth noting that these exceptionally low detection limits associated spectral and physical interferences. Species such as often can be achieved directly in the sample materials, which iron and phosphate in human serum have been shown to makes the technique extremely powerful and well suited to depress significantly the analytical signal of Se in ETAAS,14,15 direct measurements of trace metals in biological samples at and quantification of the four isotopes of Se by ICP-MS has the pg g-1 and ng g-1 levels.been shown to suVer substantially from spectral interferences An important advantage of the ETA-LIF approach for real by argon adducts and other molecular ions.16 The severity of samples is the high selectivity which results from excitation these interferences is such that trace Se determinations in and detection at diVerent wavelengths. With this selectivity, biological samples are at all times vulnerable to inaccuracies.the probability of spectral interferences due to the matrix decreases substantially. It was the intention in this study to One way of accounting for matrix eVects is by utilization J. Anal. At. Spectrom., 1999, 14, 929–932 929Table 1 Parameters for PE HGA-400 graphite furnace develop and evaluate the analytical utility of an ETA-LIF technique for the determination of Se in human serum that Ar flow rate requires little sample pre-treatment, small sample volumes and Step Temperature/ °C Ramp/s Hold/s (30 ml min-1) a minimum of measurement time.Calibration by aqueous standards fulfils these requirements. Dry 120 10 20 On Ash 900 5 10 On The ETA-LIF technique has been noted for its high Atomize 2000 0 5 OV sensitivity, which often results from the ability to saturate Clean 2650 1 5 On fully the analyte excitation transition and the long residence time of the graphite tube atomizer. Saturation ensures that the maximum possible fluorescence emission is generated and Nominal pulse energies for the second anti-Stokes shifts were minimizes the dependence of the signal on variations in the 5–10 mJ.quantum eYciency or the laser source intensity. The current The 196 nm beam was directed into a graphite tube furnace ETA-LIF approach uses stimulated Raman scattering (SRS) (HGA-400; Perkin-Elmer, Norwalk, CT, USA) that to produce the far UV radiation for Se excitation,27 as opposed employed pyrolytic graphite-coated graphite tubes with L’vov to the sum frequency generation technique which is instrumen- platforms.Table 1 outlines the temperature program used for tally more complex and expensive.28 The limit of detection for the Se determinations. To reduce scatter from the incident Se by the current approach is 20 ppt,27 which is over two laser beam, the furnace housing was fitted with quartz windows orders of magnitude lower than ETAAS and three orders of mounted at 56°.Fluorescence was collected using a 2 in magnitude lower than the concentration of Se normally found diameter mirror set at 45° to the furnace housing, which in human serum, and allows a high dilution to be used. allowed the incident beam to pass through a 4 mm hole drilled Dilution of the serum sample is beneficial as it lowers the through the center of the mirror and into the tube. matrix concentration of the serum and reduces matrix inter- Fluorescence was imaged by a 2 in diameter lens on to the ferences while maintaining adequate sensitivity for the determi- 100 mm slit of a monochromator (Spectrapro-275, f/3.5; Acton nation.It therefore provides a simple approach to removing Research, Acton, MA, USA). errors which may result from spectral and non-spectral inter- Fluorescence was detected using a solar blind photomultiferences (e.g., atomization interferences). An added benefit of plier tube (R166UH; Hamamatsu, Bridgewater, NJ, USA) dilution is that only a very small amount of serum is necessary which minimized background signals that resulted from therfor each determination, which is critically important in studies mal emission of the graphite tube.To protect against saturation where sample volumes may be limited. of the detector at concentrations in the range 1–20 ng ml-1, the fluorescence was attenuated by neutral density filters. The anode output of the PMT was terminated into 50 V and Experimental sampled by a boxcar averager using a gatewidth of 20 ns Measurement system (SR250; Stanford Research Systems, Sunnyvale, CA, USA).At a repetition rate of 30 Hz and an average of 10 laser pulses, The experimental set-up for beam generation and coupling to the measurement system had an eVective time constant of the ETA has been discussed previously.17 A simple schematic 0.33 s. Atomization events were recorded for a period of 6.7 s. diagram of the apparatus is shown in Fig. 1. An XeF excimer The boxcar output was delivered to an analog to digital laser operating at 351 nm (LPX 200; Lambda Physik, Fort converter/computer interface for data acquisition and pro- Lauderdale, FL, USA) provides up to 340 mJ at repetition cessing. Time-integrated measurements of the atomization rates of up to 100 Hz. For data collection, a rate of 30 Hz was profile were used for quantification. Background emissions used to produce 180 mJ per pulse. The output of the excimer were accounted for by averaging the signal measured in the was directed into a dye laser (Scanmate 2; Lambda Physik) last 5 s of the ashing step and subtracting it from the signal utilizing coumarin 460 dye (Lambda Physik) to produce a integrated during the atomization step.primary beam near 468 nm. This primary beam was then directed into a BBO I crystal for frequency doubling. The Serum samples resulting radiation near 234 nm propagated collinearly with the primary beam and both were directed into a Raman Human serum was obtained from bone marrow transplant converter (0.3 m, Light Age, Somerset, NJ, USA) filled with patients taking part in a study at Bowman-Gray Medical approximately 100 psi of H2. School (Winston-Salem, NC, USA).Samples were taken at By use of the 4155 cm-1 shift of H2, the second anti-Stokes two times, typically 10 d prior to and 60 d after the transplant shift of the frequency doubled dye laser produced radiation procedure.The serum samples were stored at -30 °C on-site near 196 nm. This wavelength was subsequently separated and were shipped overnight packed in dry-ice. Upon receipt, from all other wavelengths with a quartz Pellin–Broca prism. the samples were again stored at -30 °C until measurement. The average density of a representative sample of the serum was determined to be 1.004±0.007 g ml-1 at 23 °C (n=9) and this value was used to determine their Se content while using a volumetric dilution procedure.Possible contamination of the collection and storage vessels was investigated by acid extraction of a representative sample of the vacutainers and microcentrifuge tubes. A 25% HNO3 (Trace Metal Grade; Fisher Scientific, Fair Lawn, NJ, USA) solution was used to extract contaminants from the vessels for 48 h. At the end of this period, the Se concentrations of the extractant solutions were measured by ETA-LIF. The Se levels in the extractant solutions [ x=0.2±0.3 ppb (n=5); x= 0.1±0.3 ppb (n=6)] were found not to be statistically diVerent from the blank [ x=0.2±0.3 ppb (n=6)], which indicated that Se contamination from these materials was unlikely. Fig. 1 Schematic diagram of pierced-mirror fluorescence collection apparatus. The accuracy of the ETA-LIF approach was verified using 930 J. Anal. At. Spectrom., 1999, 14, 929–932Table 2 Analyses of NIST SRM 1598 a bovine serum standard reference material obtained from the National Institute of Standards and Technology (NIST) (SRM Concentration/ng g-1 1598 Inorganic Constituents in Bovine Serum), which was certified to contain 42.4 ng g-1 of Se (reported serum density Certified value 42.4±3.5a of 1.029 g ml-1 at 23 °C). 1520 dilution 43±7b (n=4) Sample preparation 1510 dilution 42±3b (n=5) All glassware was initially cleaned with 18MV de-ionized 1510 standard addition 45±1b water (Millipore, Bedford, MA, USA) and placed in a surfac- (n=6) tant bath (Alconox, New York, USA).All plasticware was aUncertainty approximates 95% confidence level. bUncertainty correcleaned with de-ionized water and placed in an approximately sponds to ± one standard deviation. 1:6 HNO3–H2O bath. Both container types were stored submerged until used (minimum of 48 h). All ware was quadruply rinsed with de-ionized water before use. profile of the diluted bovine serum was not significantly Calibration standards (20, 60, 100, 140 and 200 ppb) were diVerent from that of aqueous standards, which indicates that produced by volumetric dilution of a 1000 ppm stock standard the atomization characteristics of Se are not significantly solution of Se (Perkin-Elmer) with de-ionized water. Dilution altered by the serum matrix at this dilution (Fig. 2). was performed in volumetric glassware and the resulting Measurements of the diluted bovine serum at an excitation solutions were placed in polyethylene bottles. In order to wavelength slightly oV resonance were performed and were minimize analyte losses during atomization, a chemical modi- found not to be statistically diVerent from those of the aqueous fier consisting of 1.5 g l-1 Pd and 1.0 g l-1 Mg in the form of blank, which indicated that the sample matrix did not the nitrates was used.31 contribute any additional background signals.Prior to dilution, serum samples were thawed and equilibrated To ascertain further whether aqueous calibration was in a water-bath (23 °C) (Model 180; Precision Scientific, adequate to determine the Se content in the serum matrix, Winchester, VA, USA), followed by vortex mixing to ensure experiments were conducted in which samples of the bovine sample homogeneity.Dilution of solutions was carried out by serum standard containing standard additions were measured micropipetting either 80 or 160 ml of serum or calibration concurrently with aqueous calibrations. In this way, a direct standard into a 2 ml polyethylene microcentrifuge tube, followed comparison of the two methods could be performed.by a pipetting of de-ionized water to increase the final volume Determination of Se concentration by standard addition to 1.6 ml. This dilution method produces either a 1520 dilution yielded a value (45±1 ng g-1) that was the same within error of the serum and 1–10 ppb calibration set or a 1510 dilution (one standard deviation) as the aqueous calibration and certiof the serum and a 1–20 ppb calibration set.Solutions were fied values (Table 2). Moreover, the slopes of the calibration then stored at -30 °C until used (storage time typically less curve and spike recoveries were found to be within 10% of than 48 h). Measurements of the diluted serum samples were one another, which confirmed that the serum matrix did not carried out in random order between the two dilution schemes. aVect the sensitivity. It is noteworthy that upon freezing, volumes greater than 1.6 ml expanded enough to damage the walls of the microcentrifuge Determination of selenium in human serum tubes.Initial Se measurements of the human serum samples were conducted using a 1520 dilution ratio. Calibration was Results and discussion accomplished using 1, 3, 5, 7 and 10 ppb calibration solutions, Selenium excitation and fluorescence scheme and was verified by an analysis of the NIST bovine serum standard. Determination of the Se concentration in the human The LIF scheme employed for these investigations utilized the serum samples was based on the average of 4–5 individual 196.026 nm transition of Se from the lowest level of the ground measurements.For quality control, a calibration solution was state triplet, 4p4 3P2, to the 4p35s 3S01 excited state at interposed after every four determinations and required to be 50966 cm-1. Fluorescence emissions were detected at within 15% of the linear regression value. 203.989 nm, corresponding to the 4p35s 3S01 to 4p4 3P1 trans- Although ppb and sub-ppb level uncertainties are obtained ition.Owing to the relatively small separation between the at the 1–20 ppb level where calibrations are performed, these excitation and fluorescence wavelengths, it was found that uncertainties are amplified into relatively high uncertainties in noise due to scattered laser radiation was significant. While it the undiluted serum samples. For this reason, it was of interest is usually considered advantageous to carry out analytical measurements under fully saturated conditions, it was necessary to limit the laser energy in order to minimize the contributions of laser scatter to the overall signals.Laser scatter noise was reduced by using pulse energies not greater than those corresponding to the region in which saturation first manifested itself (~2 mJ ). Noise from laser scatter was further reduced by replacement of the standard flat quartz windows with tilted windows.The windows were rotated until a minimum of scattering was achieved. Analysis of bovine serum standard A NIST certified bovine serum standard was used to evaluate the accuracy of the ETA-LIF approach. By utilizing a 1520 dilution ratio and the Pd–Mg chemical modifier, determinations by aqueous calibration of the serum standard were well within certified error (see Table 2). Dilution of the sample reduces the probability of spectral and atomization inter- Fig. 2 Temporal atomization profiles: aqueous calibration standard (upper trace) and 1520 diluted serum sample ( lower trace). ferences. It was noted that the time dependent atomization J. Anal. At. Spectrom., 1999, 14, 929–932 931reported for whole blood include the contribution by erythrocytes and are high compared with those reported here, but are still within the range that is considered normal.33 Acknowledgments The authors gratefully acknowledge funding from the Carver Scientific Research Initiative Grants Program (University of Iowa), Dr.Kevin High, Bowman Gray School of Medicine, Wake Forest University, for providing access to the serum samples and Mr. Rick Frost of Light Age Inc. for providing support and assistance in the use of the Raman cell. References 1 C. G. Wilber, Clin. Toxicol., 1980, 17(2), 171. 2 P. Garberg and J. Ho�gberg, Ambio, 1986, 15(6), 354. 3 K. E. Burke, G. F. Combs, Jr., E. G. Gross, K. C. Bhuyan and J. Abu-Libdeh, Nutr.Cancer, 1992, 12, 123. Fig. 3 Selenium concentrations in human serum samples as determined 4 A. Dhur, P. Galan and S. Hercberg, Comp. Biochem. Physiol., by ETA-LIF. Circles are for samples diluted 1520 and squares are for 1990, 96C, 271. samples diluted 1510; open and filled symbols are samples obtained 5 M. A. Beck, Q. Shi, V. C. Morris and O. A. Levander, Nature before and after the transplant procedure was performed, respectively. Med., 1995, 1(5), 433. The error bars correspond to ±one standard deviation. 6 M. A. Beck, P. C. Kohlbeck, Q. Shi, V. C. Morris and O. A. Levander, J. Med. Virol., 1994, 43, 166. 7 L. Hardell, A. Degerman, R. Tomic, S. L. Marklund and to lower the overall uncertainty of the procedures. The results M. Bergfors, Eur. J. Cancer Prev., 1995, 4, 91. 8 K. Yoshizawa, W. C. Willett, S. J. Morris, M. J. Stampfer, of the calibration measurements indicated that the uncertainty D. Spiegelman, E. B. Rimm and E. Giovannucci, J.Natl. Cancer did not scale with increasing Se concentration. Since manual Inst., 1998, 90(16), 1219. injections are used in these studies and are likely to contribute 9 G. F. Combs, Jr., L. C. Clark and B. W. Turnbull, Biomed. significantly to the irreproducibility of the measurements, it Environ. Sci., 1997, 10, 227. was hypothesized that there would be an advantage to using 10 J. M. Marchante-Gayo� n, J. E. Sa�nchez-Urý�a and A. Sanz-Medel, J. Trace Elem. Med. Biol., 1996, 10, 229.a 1510 dilution ratio, provided that no spectral or atomization 11 P. H. E. Gardiner, D. Littlejohn, D. J. Hall and G. S. Fell, J. Trace interferences were introduced by this procedure. Direct com- Elem. Med. Biol., 1995, 9, 74. parison of the results for seven samples selected randomly 12 Z. Jin, J. Shougui, C. Shikun, J. Desen and D. Chakraborti, using both dilution procedures yielded the same results at the Fresenius’ J. Anal. Chem., 1990, 337, 887. 95% confidence level [tcalc=1.12, ttable=2.447 (95%, six degrees 13 H.T. Delves and C. E. Sieniawska, J. Anal. At. Spectrom., 1997, 12, 387. of freedom)], indicating that a 1510 dilution ratio was also 14 I. Martinsen, B. Radziuk and Y. Thomassen, J. Anal. At. feasible and did not reduce the accuracy. Spectrom., 1988, 3, 1013. Another important result of using 1510 dilutions was that 15 B. Sampson, J. Anal. At. Spectrom., 1987, 2, 447. the overall uncertainties of the results were reduced by half 16 M.P. Rayman, F. R. Abou-Shakra and N. I. Ward, J. Anal. At. relative to that observed using 1520 dilutions. These results Spectrom., 1996, 11, 61. 17 N. Omenetto, B. W. Smith and J. D. Winefordner, Spectrochim. support the hypothesis that the manual injection technique Acta, Part B, 1988, 43(9), 1111, and references cited therein. limits the uncertainty of the measurements and suggests that 18 D. J. Butcher, J. P. Dougherty, F. R. Preli, A. P. Walton, G. Wei, the uncertainty would be improved by using an autosampler R.L. Irwin and R. G. Michel, J. Anal. At. Spectrom., 1988, 3, device. The increase in the variance with increased sample 1059 and references cited therein. dilution has been noted recently by Zhou et al.32 in ETA-LIF 19 S. Sjo� stro�m and P. Mauchien, Spectrochim. Acta Rev., 1993, 15(3), 153 and references cited therein. determinations of Cd, Co, Pb, Mn and Tl in BuValo River 20 G. A. Petrucci, H. Beissler, O. Matveev, P.Cavalli and sediments. Results for aqueous calibration of all of the human N. Omenetto, J. Anal. Atom. Spectrom., 1995, 10, 885. sera samples using a 1510 or 1520 dilution ratio are shown 21 R. L. Irwin, D. J. Butcher, J. Takahashi, G.-T. Wei and in Fig. 3. R. G. Michel, J. Anal. At. Spectrom., 1990, 5, 603. Since Se is known to be important for an eVective immune 22 V. Cheam, J. Lechner, I. Sekerka, R. Desrosiers, J. Nriagu and G. Lawson, Anal. Chim. Acta, 1992, 269, 129. response, it was of interest to compare the serum Se content 23 E.P. Wagner, III, B. W. Smith and J. D. Winefordner, Anal. in the bone marrow transplant patients at 10 d (-10) prior to Chem., 1996, 68, 3199. and 60 d (+60) following treatment, which can be a period of 24 M. A. Bolshov, S. N. Rudnev, J.-P. Candelone, C. F. Boutron and compromised immunity. The means and standard deviations S. Hong, Spectrochim. Acta, Part B, 1994, 49, 1445. for the patients at these two time points were m-10= 25 J. Enger, A. Marunkov, N. V. Chekalin and O. Axner, J. Anal. At. Spectrom., 1995, 10, 539. 89±20 ng g-1 for n-10=75 and m+60=92±23 ng g-1 for 26 Z. Liang, R. F. Lonardo and R. G. Michel, Spectrochim. Acta, n+60=53. Comparison of means yields a t value of 0.817, Part B, 1993, 48, 7. which is significantly less than the tabulated t value of 1.96 27 D. J. Swart, M. Ezer, H. L. Pacquette and J. B. Simeonsson, Anal. (95% confidence for 126 degrees of freedom). The results Chem., 1998, 70, 1324. demonstrate no diVerence in Se status of the two groups as 28 U. Heitmann, T. Sy, A. Hese and G. Schoknecht, J. Anal. At. Spectrom., 1993, 3, 437. indicated by total serum Se levels. These values are consistent 29 R. Q. Aucelio, B. W.. D. Winefordner, Appl. with the range reported as typical for serum Se (approximately Spectrosc., 1998, 52(11), 1457. 50–150 ng g-1).33 30 E. Masera, P. Mauchien and Y. Lerat, Spectrochim. Acta, Part B, It is worth noting the only other report of ETA-LIF 1996, 51, 543. measurements of Se in blood or serum. Using a sum frequency 31 B. Welz, G. Schlemer and J. R. Mudakavi, J. Anal. At. Spectrom., 1988, 3, 695. generation technique for producing the laser excitation at 32 J. X. Zhou, X. Hou, K. X. Yang and R. G. Michel, J. Anal. At. 196.026 nm, Heitmann et al.28 reported Se concentrations of Spectrom., 1998, 13, 41. 122.0±17.3 mg l-1 for n=200 in frozen whole blood and 33 J. Versieck, Crit. Rev. Clin. Lab. Sci., 1985, 22(2), 97. 114.2±15.6 mg l-1 for n=103 in fresh whole blood sampled from a population of healthy Berliners. The Se concentrations Paper 9/00271E 932 J. Anal. At. Spectrom., 1999, 14, 929–932

ISSN:0267-9477    

DOI:10.1039/a900271e

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Determination of long-lived radionuclides by inductively coupled plasma quadrupole mass spectrometry using diVerent nebulizers J. Sabine Becker,*a Rajiv S. Soman,b Karen L. Sutton,c Joseph A. Carusoc and Hans-Joachim Dietzea aZentralabteilung fu�r Chemische Analysen, Forschungszentrum Ju�lich GmbH, D-52425 Ju�lich, Germany bDepartment of Chemical Technology, University of Cincinnati, Cincinnati, OH 45206, USA cDepartment of Chemistry, University of Cincinnati, Cincinnati, OH 45221-0172, USA Received 26th November 1998, Accepted 24th March 1999 DiVerent nebulizers (cross-flow, ultrasonic and two microconcentric nebulizers) were used for sample introduction of radioactive solutions into a quadrupole-based inductively coupled plasma mass spectrometer (ICP-QMS).The best sensitivity (from 420 to 850 MHz, which is about one order of magnitude higher in comparison with the cross- flow nebulizer) for long-lived radionuclides (226Ra, 230Th, 237Np, 238U and 241Am) was observed using the ultrasonic nebulizer. However, using the ultrasonic nebulizer, a significantly higher sample size (26-fold) in comparison with the micronebulizers is required.Sample introduction by micronebulization with a small sample size in the low picogram range is the method of choice for the determination of long-lived radionuclides. The precision of determination of a 10 ng l-1 concentration was in the low-% range (and sub-% range) for all measurements using diVerent nebulizer types.The detection limits for the determination of long-lived radionuclides in aqueous solutions applying the diVerent nebulizers were 0.01–0.6 ng l-1. The flow injection analysis approach was optimized for isotope dilution analysis of 232Th (using 20 ml of 5 mg l-1 230Th) by ICP-QMS. The isotopic abundance ratios of 230Th–232Th isotope mixtures (230Th/232Th=0.01, 0.001 and 0.0001) were determined using a microconcentric nebulizer and 1 mg l-1 Th solutions with a relative external standard deviation of long-term stability measurements (over 20 h) of 0.17, 0.62 and 2.66%, respectively.The determination of long-lived radionuclides at ultratrace for the determination of the long-lived radionuclides at high mass resolution where isobaric interferences with molecular or concentration levels in diVerent types of materials (such as environmental samples, for evidence of contamination from atomic ions would be expected, as demonstrated for the determination of 79Se or 129I, where mass spectrometric inter- radioactive waste in biological samples, waters or geological materials, for age dating of geological materials in geochronol- ferences of analyte ions with 38Ar40ArH+ molecular ions12 or 129Xe+ atomic ions of plasma gas contaminant13 were ogy and for the characterization of radioactive wastes from nuclear reactors for their recycling and final storage) is a observed.In comparison with sample introduction using a conventional Meinhard or cross-flow nebulizer, the sample challenging task for analytical chemistry.1–8 Conventional radiochemical methods such as alpha spectrometry, for the introduction into the ICP for the determination of selenium was improved by hydride generation (using NaBH4 as the determination of long-lived radionuclides at low concentration levels, often requires many time-consuming chemical separa- reducing agent)12 and for iodine by oxidizing iodide to iodine (using HClO4 as the oxidation reagent).13 Desolvation using tion steps for selected analytes.In contrast, ICP-MS possesses a high isotopic selectivity in comparison with classical radio- a special solution introduction apparatus was used for both applications when coupled to ICP-MS. analytical measurements and sample preparation is often easier. Owing to the excellent sensitivity of ICP-MS, fewer A problem that is diYcult to solve in the determination of long-lived radionuclides is the isobaric interference of analyte separation steps for the determination of long-lived radionuclides at ultratrace concentration levels are necessary.ions with stable atomic ions at the same mass number (e.g., 107Pd and 107Ag, 151Sm and 151Eu, 93Mo and 93Zr and 93Nb, Therefore, since the establishment of ICP-MS in ultratrace analysis,9,10 this technique has been applied to an increasing 99Tc and 99Ru, 126Sn and 126Te and 126Xe). For the determination of such long-lived radionuclides a chemical separation extent as an eYcient and sensitive analytical mass spectrometric method for the determination of ultratrace levels of long-lived of interfering chemical elements [on-line by HPLC14,15 or capillary electrophoresis (CE) or oV-line separation] is required radionuclides and their isotopic compositions in aqueous solution.The possibility of determining long-lived radionuclides before mass spectrometric analysis is applied. HPLC-ICP-MS on-line coupling was applied in our laboratory for the separa- in the low pg l-1 concentration range was demonstrated using a double-focusing sector field ICP mass spectrometer with tion of long-lived radionuclides and stable isotopes of lanthanides in a tantalum target from a spallation neutron source ultrasonic nebulization by Kim et al.1 [with a PlasmaTrace, ICP-MS instrument (Fisons, Loughborough, UK)] and in our irradiated with 800 MeV protons16,17 in order to determine the spallation nuclides.Recently, Sutton et al.18 described the laboratory11 [with an ELEMENT ICP-MS system (Finnigan MAT, Bremen, Germany)]. Such extremely low detection coupling of CE to ICP-MS for the separation of a very small sample size (nl ) of lanthanides. This technique is of great limits for long-lived radionuclides were observed owing to the low background (<0.2 counts s-1) and the high sensitivity of interest for application to long-lived radionuclides in order to reduce operator exposure and contamination of the ICP-MS double-focusing sector field ICP-MS at low mass resolution.The detection limits are several orders of magnitude higher instrument. J. Anal. At. Spectrom., 1999, 14, 933–937 933Furthermore, a study of long-term stability over 24 h7 Measurement procedures demonstrated that isotope ratio measurements of uranium and The optimization of the experimental parameters (see Table 1) thorium using a ICP-quadrupole MS (ICP-QMS) system with was performed using the maximum ion intensity of 238U+ as a cross-flow nebulizer and Scott spray chamber yielded a the figure of merit.relative external standard deviation (RESD) of about In this work the total integration time per replicate for the 0.05%. ICP-QMS can compete favorably with isotope ratio measurement of all long-lived radionuclides was 60 s (with measurements using thermal ionization mass spectrometry three replicates, i.e., a total measuring time of 3 min for each (TIMS).sample). In agreement with the results of Crain,4 the peak The aim of this work was to study various commercial hopping mode was used, which results in lower detection limits nebulizer types [cross-flow nebulizer (CFN) with Scott spray compared with the scanning mode. In the case of the cross- chamber, ultrasonic nebulizer (USN) and diVerent micronebu- flow nebulizer, ultrasonic nebulizer and micronebulizers, the lizers with Scott spray chamber and minicyclonic spray sample size (10 ng l-1 concentration) was 30, 66 and 2.55 pg, chamber] coupled to a quadrupole-based ICP-MS system for respectively, of each long-lived radionuclide in the investigated the determination of concentrations, detection limits, precision aqueous solution.The element concentrations used to deter- and sensitivity of long-lived radionuclides in the ultra-low mine the detection limits were 0.1, 1, 5 and 10 ng l-1. The concentration range. detection limits were determined using calibration curves applying the 3s criterion.Fig. 1 shows the calibration curve Experimental of 237Np using the micronebulizer (MicroMist) and minicyclonic spray chamber (Cinnabar) for the determination of the Instrumentation detection limit in ICP-QMS. The determination of the pre- A Perkin-Elmer ELAN 6000 quadrupole ICP-MS (Perkin- cision (short-term stability) and sensitivity of ICP-QMS using Elmer SCIEX, Thornhill, ON, Canada) equipped withiVerent nebulizer types was performed using a standard original cross-flow nebulizer was used to measure the sensi- solution of 10 ng l-1 of each radionuclide. tivity, detection limits and precision of long-lived radionuclides in aqueous solution at low concentration levels.In addition, a U-6000AT ultrasonic nebulizer (Cetac Technologies, Omaha, NE, USA), an MCN-100 microconcentric nebulizer (Cetac Technologies) and a MicroMist micronebulizer (Glass Expansion, Hawthorn, Victoria, Australia) equipped with a Cinnabar minicyclonic spray chamber (Glass Expansion) were also evaluated.Solution aspiration was performed with a peristaltic pump (Perimax 12, Spetec, Erding, Germany). The experimental parameters are summarized in Table 1. Materials For the characterization of diVerent nebulizer types for the determination of long-lived radionuclides, a mixture of 226Ra, 230Th, 237Np, 238U and 241Am was investigated. Thorium isotope mixture solutions (230Th/232Th # 0.01, 0.001 and 0.0001%) using the isotopic reference material IRMM-060 (99.85% 230Th and 0.15% 232Th) and ca. 99.99% of NIST3159 spectrometric solution (natural abundance 232Th) were used for isotope ratio measurements with a microconcentric nebulizer. Furthermore, diVerent radioactive Fig. 1 Calibration curve for 237Np using a micronebulizer (MicroMist) waste samples (unknown origin) were characterized by and a minicyclonic spray chamber (Cinnabar) for the determination of detection limits.ICP-MS. Table 1 Experimental parameters for the ELAN 6000 ICP-QMS CFN USN Microconcentric nebulizers Nebulizer type Standard ELAN U-6000 AT MCN-100 MicroMist (AR30–1F02) Spray chamber Scott type — Scott type Minicyclonic, Cinnabar Rf power/W 1250 1050 1150 1250 (1375) Sample uptake rate/ml min-1 1.00 2.2 0.085 0.085 (0.2) Cylindrical lens potential/V 7.00 6.55 7.25 7.20 Nebulizer gas flow rate/l min-1 0.92 0.95 0.92 0.98 Sample size/pg (10 ng l-1 solution) 30 66 2.5 2.5 (6) For all nebulizer types— Auxiliary gas flow rate/l min-1 0.8 Coolant gas flow rate/l min-1 14 Dwell time/ms 50 No.of sweeps per replicate 30 Integration time per replicate/s 60 No. of replicates 3 Detection system dead time (calibrated)/ns 53 Mass resolution (m/Dm) 300 Scanning mode Peak hopping 934 J. Anal. At. Spectrom., 1999, 14, 933–937Flow injection for the introduction of small sample volumes in For the micronebulizers (MCN-100 with Scott spray chamber and MicroMist with minicyclonic Cinnabar spray ICP-MS chamber), the sensitivity for the long-lived radionuclides at a In order to minimize the contamination of ICP-MS, small solution uptake rate of 0.085 ml min-1 (in contrast to the sample volumes are required.For the injection of small sample USN, the total analyzed sample volume was only 0.255 ml ) is volumes of a solution of long-lived radionuclides, a commercial in the 22–70 MHz (ppm)-1 range. An increase in sensitivity HPLC injection valve was coupled to a micronebulizer for (by a factor of approximately 2) was observed for a higher small droplet formation.sample uptake rate (0.2 instead of 0.085 ml min-1) for For flow injection analysis, a small sample volume (20 ml ) measurements using the MicroMist with minicyclonic of 232Th or an enriched 230Th spike (for the isotope dilution Cinnabar spray chamber. technique) was introduced into a continuous flow of 2% nitric acid using the HPLC injection valve.11 Detection limits for the determination of long-lived radionuclides using diVerent nebulizers Results and discussion All detection limits for the determination of the long-lived The results of measuring the sensitivity, detection limits and radionuclides were in the sub-ng l-1 concentration range precision of the long-lived radionuclides of interest (226Ra, (Table 3).The lowest detection limits were observed using the 230Th, 237Np, 238U and 241Am) are summarized in Tables 2–4.USN for solution introduction. The detection limits were determined in the present work (3s criterion) using calibration Sensitivity of ICP-QMS for the determination of long-lived curves in the appropriate concentration range. As an example, radionuclides using diVerent nebulizers the calibration curve for 237Np in the ng l-1 concentration range for the determination of the detection limit is shown in The results in Table 2 can be discussed as follows.Fig. 1. The measured detection limits using the micronebulizers For the CFN, under the conditions described above, the demonstrated the good possibility of utilizing them for the sensitivity for diVerent long-lived radionuclides was analysis of small sample volumes. approximately 47–93 MHz (ppm)-1. The application of the USN demonstrated the highest sensitivity for the radionuclides investigated. The sensitivity Precision of long-lived radionuclide determination by ICP-QMS was about one order of magnitude higher [between 420 and using diVerent nebulizers 850 MHz (ppm)-1] in comparison with the CFN.However, the USN requires the highest sample size; for each The precision (RSD) of long-lived radionuclide determination at a 10 ng l-1 concentration of each analyte, which was measurement, 6.6 ml sample volume was used. Table 2 Comparison of sensitivity [MHz (ppm)-1] for the determination of long-lived radionuclides by ICP-QMS using diVerent nebulizers MicroMist Long-lived CFN USN MCN-100 radionuclide (1 ml min-1)a (2.2 ml min-1)a (0.085 ml min-1)a 0.085 ml min-1a 0.2 ml min-1a 226Ra 47 499 23 22 47 230Th 52 420 54 34 80 237Np 84 853 38 39 84 238U 50 667 70 45 90 241Am 93 848 39 42 93 aSample uptake rate.Table 3 Comparison of detection limits (ng l-1) for the determination of long-lived radionuclides by ICP-QMS using diVerent nebulizers MicroMist Long-lived CFN USN MCN-100 radionuclide (1 ml min-1)a (2.2 ml min-1)a (0.085 ml min-1)a 0.085 ml min-1a 0.2 ml min-1a 226Ra 0.420 0.194 0.171 0.217 0.433 230Th 0.390 0.011b 0.157 0.567 0.186 237Np 0.429 0.186 0.080 0.106 0.257 238U 0.350 0.026 0.350 0.342 0.330 241Am 0.536 0.196 0.315 0.480 0.612 aSample uptake rate.b232Th. Table 4 Comparison of precision (%) for the determination of long-lived radionuclides (for a 10 ng l-1 solution) by ICP-QMS using diVerent nebulizers MicroMist Long-lived CFN USN MCN-100 radionuclide (1 ml min-1)a (2.2 ml min-1)a (0.085 ml min-1)a 0.085 ml min-1a 0.2 ml min-1b 226Ra 1.32 0.65 2.73 3.97 2.81 230Th 1.79 0.70 2.67c 3.85 3.67 237Np 1.87 0.44 1.84 0.78 1.10 238U 1.17 0.98 0.55 5.76 2.47 241Am 0.89 0.16 3.17 3.40 2.83 aSample uptake rate. bScott spray chamber.c232Th. J. Anal. At. Spectrom., 1999, 14, 933–937 935Fig. 2 (a) Transient signals for the determination of 232Th (concentration, 1, 2 and 4 mg l-1; sample volume, 20 ml ) using flow injection with a micronebulizer (MicroMist) and minicyclonic spray chamber (Cinnabar).(b) Application of flow injection for the isotope dilution technique for the determination of 232Th (continuous flow), spiked with 20 ml of 5 mg l-1 230Th. determined under the described experimental conditions, was volume with a uranium concentration of 6 ng ml-1 and an RSD of approximately 3%.11 in the low-% range or even less than 1% (Table 4). Flow injection for the introduction of small sample volumes in Formation of molecular ions in ICP-MS ICP-MS The formation of hydride and oxide molecular ions of thorium and uranium in ICP-MS, using a cross-flow nebulizer with In order to minimize radioactive contamination of the ICP-MS system, small sample volumes are required.To handle micro- Scott spray chamber and the MicroMist microconcentric nebulizer with minicyclonic spray chamber, is compared in Table 5. liter amounts of an aqueous solution of long-lived radionuclides a commercial HPLC injection valve was coupled to With both nebulizer types, similar ion intensities were observed, namely 2–3% as the oxide formation rate and about a micronebulizer for small droplet formation.The small sample volume (the concentration of thorium was 1, 2 and 4 mg l-1) 2×10-5 as the hydride formation rate of thorium and uranium. Similar ion formation rates for both molecular ions of Th and was introduced in a continuous flow of 2% nitric acid by flow injection. Fig. 2(a) shows the transient signals of diVerent U were measured by Crain.4 232Th solutions using a sample loop of 20 ml solution. We applied the flow injection method for isotope dilution analysis Isotope ratio measurements on thorium isotope mixture solutions using ICP-QMS.In this experiment a 2 mg l-1 232Th solution (continuous flow) was spiked with 20 ml of 5 mg l-1 230Th, and In a recent study,7 we determined the relative external standard this is demonstrated in Fig. 2(b). The flow injection isotope deviation (RESD) for 1.25 mg l-1 solutions (230Th/232Th= dilution technique is of special interest for the accurate determi- 0.22) by ICP-QMS using an ELAN cross-flow nebulizer.The nation of a radionuclide concentration using a small sample RESD of long-term stability measurements (over about 10 h) volume of a highly enriched spike (demonstrated here) or a was 0.067%, which, under these conditions, is more typical of highly radioactive sample in order to avoid contamination of TIMS than ICP-QMS.the mass spectrometer. Considering the approach of Table 6 summarizes the thorium data, which were obtained Beauchemin and Specht,19 using the sample solution as a carrier and injecting the enriched 230Th isotope spike, the 232Th concentration can be calculated using the known equa- Table 6 Results of long-term isotope ratio measurements of 1 mg l-1 tion for isotope dilution analysis.20 The dispersion coeYcient thorium with diVerent isotope ratios was determined as described in reference 19 using a Isotope mixture Isotope ratio Measured isotope ratio RESD (%) 2 mg l-1 solution of 205Tl.As demonstrated for the determination of uranium using 230Th–232Th ~10-2 0.010 209±0.000 019 0.17 double-focusing sector field ICP-MS,11 the sample volume 230Th–232Th ~10-3 0.001 148±0.000 007 0.62 may be reduced to 1.5 ml. The concentrations of long-lived 230Th–232Th ~10-4 0.000 128±0.000 003 2.66 radionuclides were determined using a sample loop of 5 ml Table 5 Relative intensities of molecular ions of long-lived radionuclides in ICP-MS Relative ion intensity (MX+/M+; X=O, H) Possible interference Half-life of Molecular ion CFN MicroMist Crain4 with analyte ion actinide nuclide/yr 232Th1H+ 2.0×10-5 1.6×10-5 3.9×10-5 233U+ 1.6×105 238U1H+ 2.1×10-5 1.9×10-5 3.2×10-5 239Pu+ 2.4×104 232Th16O+ 2.5×10-2 2.8×10-2 2.6×10-2 248Cm+ 3.4×105 238U16O+ 2.7×10-2 3.1×10-2 1.6×10-2 — — 936 J.Anal. At. Spectrom., 1999, 14, 933–937Table 7 Results of isotope analysis of thorium in diVerent waste to the low-pg range.Future work will include the application samples of high-eYciency micronebulizers [e.g., direct injection highe Yciency nebulizer (DIHEN)] which, in principle, should Sample No. 230Th/232Th RSD (%) allow for more eYcient sample introduction into the ICP, to yield improved sensitivities. Preliminary data suggest an 1 0.000 212 4.82 2 0.000 125 6.12 improvement in sensitivity by a factor of 4–6 compared with 3 0.002 218 7.03 the MicroMist microconcentric nebulizer for the determination 4 0.001 200 5.11 of long-lived radionuclides.22 The results of this ongoing study will be reported elsewhere.Table 8 Results of isotope analysis of uranium in diVerent waste samples Acknowledgement Sample RSD RSD RSD The authors thank Mr. M. Girnus (Zentralabteilung fu� r No. 234U/238U (%) 235U/238U (%) 236U/238U (%) Chemische Analysen, Forschungszentrum Ju� lich) for his technical assistance. 1 0.000 044 1 3.25 0.006 321 0.15 0.000 112 9 2.24 2 0.000 062 0 2.70 0.008 216 0.17 0.000 105 5 2.19 3 0.000 020 3 1.66 0.003 531 0.24 0.000 066 0 2.51 References Table 0.000 055 — 0.007 25 — — — value21 1 C.K. Kim, R. Seki, S. Morita, S. Yamasaki, T. Tsumura, Y. Igarashi and M. Yamamoto, J. Anal. At. Spectrom., 1991, 6, 205. 2 G. P. Russ and J. M. Bazan, Spectrochim. Acta, Part B, 1987, 42, 49. by analyzing a thorium isotope mixture with significantly 3 J.S. Becker and H.-J. Dietze, Adv. Mass Spectrom., 1998, 14, 681. lower isotope ratios (230Th/232Th=10-2, 10-3 and 10-4 at 4 J. S. Crain, Spectroscopy, 1996, 11, 30. 1 mg l-1) with the MicroMist micronebulizer with minicyclonic 5 J. S. Becker and H.-J. Dietze, J. Anal. At. Spectrom., 1997, 12, 881. spray chamber over a 24 h period. The RESD for a 1 mg l-1 6 P. D. P. Taylor, P. De Bievre, A. J. Walder and A. Entwistle, Th isotope mixture solution with 230Th/232Th isotope ratios of J.Anal. At. Spectrom., 1995, 10, 395. 10-2, 10-3 and 10-4 was 0.17, 0.6 and 2.7%, respectively. 7 I. T. Platzner, J. S. Becker and H.-J. Dietze, At. Spectrosc., 1999, 20, 6. 8 W. Kerl, J. S. Becker, H.-J. Dietze and W. Dannecker, Fresenius’ Isotope ratio measurement of thorium and uranium in J. Anal. Chem., 1997, 359, 40. radioactive waste samples 9 Inductively Coupled Plasma Mass Spectrometry, ed. A. Montaser, Wiley-VCH, New York, Chichester, Weinheim, 1998. For the analysis of real radioactive waste samples the 10 J. S.Becker and H.-J. Dietze, J. Anal. At. Spectrom., 1998, 13, acquisition time was 3 min. The precision of isotope ratio 1057. measurements of thorium and uranium in radioactive waste 11 W. Kerl, J. S. Becker, H.-J. Dietze and W. Dannecker, Adv. Mass samples (see Tables 7 and 8) at a concentration of 1 mg l-1 Spectrom., 1998, 14, MoPo143. was determined to be in the low-% and sub-% range. The 12 K. Hoppstock, J. S. Becker and H.-J.Dietze, At. Spectrosc., 1997, 18, 180. mass bias correction for thorium and uranium isotope ratio 13 W. Kerl, J. S. Becker, H.-J. Dietze and W. Dannecker, J. Anal. At. measurements was performed, as described previously,7 using Spectrom., 1996, 11, 723. isotope standard reference materials. In all radioactive waste 14 J. I. G. Alonso, F. Sena, P. Arbore, M. Betti and L. Koch, J. Anal. solutions 236U and 230Th of non-natural origin were detected. At. Spectrom., 1991, 6, 335. 15 J. M. B. Moreno, J. I. G. Alonso, P. Arbore, G. Nicolaou and L. Koch, J. Anal. At. Spectrom., 1996, 11, 929. Conclusions 16 J. S. Becker, G. Ku�ppers, F. Carsughi, W. Kerl, W. Schaal, F. Ullmaier and H.-J. Dietze, Ber. Forschungszentrums Ju� lich, It has been shown that an ICP-QMS system can be applied 1996, 3272, 143. successfully for the sensitive and precise determination of long- 17 W. Kerl, J. S. Becker, W. Dannecker and H.-J. Dietze, Fresenius’ lived radionuclides at low concentration levels. The application J. Anal. Chem., 1998, 362, 433. of all the diVerent nebulizers studied yielded detection limits 18 K. L. Sutton, C. B’Hymer and J. A. Caruso, J. Anal. At. in the pg l-1 concentration range, one to two orders of magni- Spectrom., 1998, 13, 885. 19 A. Beauchemin and A. A. Specht, Anal. Chem., 1997, 69, 3183. tude higher than those found using double-focusing sector 20 K. G. Heumann, Int. J. Mass Spectrom. Ion Processes, 1992, field ICP-MS with USN.11 The best sensitivity, which was 118/119, 575. observed using an ultrasonic nebulizer, was combined with a 21 IUPAC, Pure Appl. Chem., 1991, 63, 991. significantly higher sample size (26-fold) in comparison with 22 J. S. Becker and H.-J. Dietze, presented at the 4th Symposium the micronebulizers examined. Sample introduction by mic- Massenspektrometrische Verfahren der Elementspurenanalyse, ronebulization, which also shows good sensitivity and pre- Mainz, 1998, Fresenius’ J. Anal. Chem., in the press. cision, is the method of choice for the determination of longlived radionuclides, because the sample size could be reduced Paper 8/09252D J. Anal. At. Spectrom., 1999, 14, 933&ndash

ISSN:0267-9477    

DOI:10.1039/a809252d

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Microwave digestion of plant and grain standard reference materials in nitric and hydrofluoric acids for multi-elemental determination by inductively coupled plasma mass spectrometry Xinbang Feng,* Shaole Wu, Angela Wharmby and Adolph Wittmeier Alberta Research Council, P.O. Bag 4000, Vegreville, Alberta, Canada T9C 1T4 Received 22nd June 1998, Accepted 29th March 1999 A microwave-assisted HNO3–HF digestion system was explored for the total dissolution of biological plant and grain materials followed by multi-elemental determination using ICP-MS, in order to improve the low recoveries of several elements observed in a previous study using a microwave-assisted nitric acid digestion system.NIST standard reference materials (SRMs), including Apple Leaves (1515), Peach Leaves (1547), Wheat Flour (1567a), Rice Flour (1568a), Tomato Leaves (1573) and Pine Needles (1575), were analyzed. Approximately 0.5 g of sample was digested in 5 ml of HNO3 and 0.1 ml of HF, with or without a subsequent digestion stage with boric acid.The matrix eVect for boron was evaluated for an ICP-MS system and signal enhancement was observed for all the elements tested. Potential spectral interferences in ICP-MS with HNO3–HF, boron and biological matrices are discussed and the spectral interferences on Co, As and Se are tabulated. The ICP-MS system was calibrated using external standards prepared in undigested reagent blanks with In as an internal standard.It was found that with a low but suYcient amount of HF in the digestion, the possible precipitation of metal fluorides in the digestate (without boric acid) was not significant. The recoveries for some silicon-bound elements, such as Al, Co, Cr, Ni, Th, U and V, were significantly improved compared with those from digestions with HNO3 alone. Using the HNO3–HF digestion procedure, the ICP-MS results for 30 elements, Al, As, B, Ba, Ca, Cd, Co, Cr, Cu, Fe, Hg, K, Mg, Mn, Mo, Na, Ni, P, Pb, S, Sb, Se, Si, Sr, Th, Ti, Tl, U, V and Zn, agreed well with the certified values in leaf and grain SRMs.The recoveries were mostly within the range 85–115%. Hence, the use of boric acid in the digestion was not necessary, which simplified the procedure, minimized the content of the total dissolved solids in solution for ICP-MS analysis and allowed the determination of boron. The microwave-assisted nitric acid digestion of biological plant HNO3–HF alone without the addition of boric acid, has been reported for the dissolution of atmospheric aerosol samples tissue and grain materials was studied in previous work for and analysis by ICP-MS.12 trace multi-elemental determination using inductively coupled In this study, the microwave-assisted HNO3–HF digestion plasma mass spectrometry (ICP-MS).1 The method is simple system, with and without the addition of boric acid, was and convenient to use.The analytical results for most SRMs explored for the total dissolution of biological plant and grain tested were satisfactory except for some silicon-bound material, followed by the multi-elemental determination using elements, such as Al, Ni, Th and U, with recoveries in the ICP-MS.The spectral characteristics produced from the range 40–80%, and V, Co and Cr, with recoveries varying HNO3–HF digestion and the HNO3–HF–H3BO3 digestion from 70 to 100%.1 If these elements are of interest, an alternative were compared.Potential spectral interferences, especially with digestion method for plant and grain materials would be regard to Co, As and Se, are discussed. NIST standard required. Complete digestion of silicon-containing materials reference materials (SRMs) were analyzed for 30 elements. with hydrofluoric acid is well established for the dissolution The ICP-MS results were compared with the certified or of environmental, biological and geological solid samples such reference values to evaluate the accuracy and precision of as peat, leaves, oyster tissue, bovine liver, fly ash, coal, soils, the methods.sediments, ores and rocks.2–10 Studies of the use of a mixture of HNO3 and HF with microwave heating for the dissolution of some plant tissues followed by ICP-AES and ICP-MS Experimental analysis2 and HF digestion of dry-ashed orchard leaves for Standard reference materials, reagents and standard solutions the determination of Th and U using ICP-MS3 have been conducted.After HF digestion, any free HF can be eliminated The NIST SRMs used for this study were Apple Leaves either by evaporating the solution2,11 or by adding boric (1515), Peach Leaves (1547), Tomato Leaves (1573), Pine acid4,10 or water-soluble tertiary amines4 to neutralize excess Needles (1575), Wheat Flour (1567a) and Rice Flour (1568a). fluoride ion. However, the evaporation procedure is time Both high-purity HNO3 (68–71%) and HF (45–50%) (double consuming and suVers from partial losses of some volatile sub-boiling distilled in quartz; Seastar Chemicals, Sidney, BC, elements such as Si, B, Ge and Hg.2 The addition of boric Canada) were used for sample digestion.A 5% H3BO3 solution acid, which masks excess HF by forming tetrafluoroboric acid, was prepared using analytical-reagent grade reagent (BDH, makes the determination of boron impossible and increases Poole, Dorset, UK). Distilled, de-ionized water (DDW) was the total dissolved solids (TDS) content in the sample obtained from a three-column de-ionizing system with 18 digestate.In addition, the presence of boric acid results in mV cm specific resistivity capability. more complicated spectral interferences, matrix eVects and A set of undigested standard solutions (1, 10 and 100 mg l-1) poor detection limits for certain elements due to high back- including about 30 elements were prepared in the corresponding undigested reagent blank solutions. In most cases, a 100 ml ground signals.10 A simple digestion method, utilizing J.Anal. At. Spectrom., 1999, 14, 939–946 939reagent blank or standard solution contained 2 ml of concen- with In as the internal standard. A 100 ml aliquot of In solution containing 3.5 mg l-1 of In was added into each 10 ml of trated HNO3 and 0.04 ml of HF for the HNO3–HF system and an additional 2.4 ml of 5% H3BO3 for the standard or sample solution prior to ICP-MS analysis. Each calibration curve was constructed linearly through zero after HNO3–HF–H3BO3 system. Reagent blanks were processed for background subtraction in order to subtract any contami- subtraction of the reagent blank.nation contributions especially from the use of an analyticalreagent grade reagent such as the 5% H3BO3 solution. Results and discussion Microwave digestion system and digestion procedures Evaluation of matrix eVect All sample digestions were accomplished using a QWAVE-1000 Matrix eVect problems in ICP-MS can be divided into two microwave sample preparation system (Questron, Mercerville, basic categories: matrix induced signal intensity changes and NJ, USA), equipped with temperature and pressure regulation matrix induced spectral overlaps.A high concentration of a (through a sensor vessel ) and controlled by a personal com- matrix element is known to alter the analyte sensitivity and puter. The details of the system and the procedure used for analyte signal suppression is the most commonly observed cleaning the PFA liners of the digestion vessels were described eVect in ICP-MS.10,13–22 In the HNO3–HF digestion of biologiin a previous paper.1 cal plant materials, the matrix induced signal change was Two digestion procedures were tested: digestion with minimal since the sample matrix from carbohydrates generally HNO3–HF only and digestion with HNO3–HF followed by a did not contribute, having been converted to CO2 during second-stage digestion using boric acid.Approximately 0.5 g digestion and subsequently released. The concentrations of (dry mass) of plant or grain SRM was weighed directly into major elements, such as K, Na, Ca, Mg and P, were not high the PFA inner liners, to which 5.0 ml of concentrated HNO3 enough (<10–30 mg l-1) in the analyzed solutions to produce and 0.1 ml of HF were added. Samples were pre-digested significant matrix eVects. In the two-stage HNO3–HF–H3BO3 overnight in a clean fume hood at room temperature.Twelve digestion, H3BO3 was added after the dissolution to mask the vessels including a reagent blank vessel and a sensor vessel free HF and to dissolve the fluoride precipitate in the digested were sealed and digested using the following procedures. The solution by forming tetrafluoroboric acid. Hence the matrix temperature was ramped to 165 °C within 10 min with the eVect of the boric acid and spectral interferences from the application of 1000 W power, followed by a dwell time of boron matrix cannot be avoided in the ICP-MS procedure.10 20 min at 165 °C.The temperature limit was 175 °C and the In general, lower mass elements are subject to more serious pressure limit was 15.2 bar (220 psi). After cooling, each vessel matrix eVects and heavier matrix elements cause more severe was vented. For the HNO3–HF digestion, the sample digestate suppression.14–16 was diluted to 100 ml (by mass) with DDW. For the For the boron matrix, both suppression and enhancement HNO3–HF–H3BO3 digestion, there was a post-digestion stage, of signal intensities have been observed and reported.10,17,20 i.e., 6 ml of 5% boric acid were added to each sample digestion The mechanisms of the matrix eVects depend on the changes vessel and warmed to 100 °C in the closed vessels for 10 min in analyte ionization in the ICP, in sampling processes in the by microwave heating.After cooling, each sample was diluted interface region and in ion transmission in the mass specto 100 ml with DDW.All diluted digestates were transferred trometer.22 In this study, instead of suppression, the presence to high-density polyethylene (HDPE) bottles. Prior to ICP-MS of the boron matrix enhanced the signal intensities of elements analysis, each sample solution was further diluted 2.5-fold. at both low and high m/z regions (Table 1). The extent of the Hence the overall dilution factor was 500 (v/m) and the signal enhancement was related to the concentration of the contents of the digestion reagents in the final solution were boron matrix and the m/z values.In Table 1, the signal was about 2% HNO3, 0.04% HF (45–50%) and 0.12% H3BO3. enhanced by about 16–30% for the lighter elements such as The amounts of HF used in the digestion and the amounts of H3BO3 used in the post-digestion were also tested. In one Table 1 Enhancement of signal intensities of elements in HNO3– HF–H3BO3 at diVerent concentrations of boron matrix compared digestion batch, 0.01, 0.05 and 0.1 ml of HF were used with with that in HNO3–HF 5 ml of HNO3, followed by addition of 1.2 ml of 5% H3BO3 for the post-digestion.In another batch, 0.1, 0.3 and 0.5 ml of Enhancement (%) HF were used with 5 ml of HNO3, and 6 ml of 5% H3BO3 were added for the post-digestion. The procedures for the Element m/z 210 mg l-1 boron 350 mg l-1 boron digestion, post-digestion and dilution were the same as Li 7 22.2 23.2 described above.Be 9 11.8 17.8 Al 27 25.9 33.5 ICP-MS system Sc 45 16.4 25.5 Ti 47 8.58 11.6 All measurements were carried out using a Perkin-Elmer V 51 10.7 13.5 SCIEX (Thornhill, ON, Canada) Elan Model 5000 ICP quad- Mn 55 8.95 11.0 rupole mass spectrometer. The details of the ICP-MS system Co 59 9.12 13.1 and the operation parameters were described in a previous Ni 60 9.20 12.4 paper.10 Briefly, a GemTip cross-flow nebulizer, a Ryton spray Cu 65 8.27 10.1 Se 77 6.15 12.6 chamber and an ICP torch with a quartz injector (2.0 mm id) Sr 88 8.4 10.1 were used.The sample uptake rate was set at 1 ml min-1, the Y 89 8.13 9.29 plasma forward power was 1000 W, the outer gas and inter- Mo 98 10.2 11.9 mediate gas flow rates were 15 and 0.8 l min-1 and a central Cd 114 6.33 10.2 gas flow rate of 0.9 l min-1 was used to aspirate sample In 115 9.94 9.29 solutions. The oxide ratio of 140CeO+/140Ce+ and doubly Ho 165 11.0 9.94 Pb 208 7.15 7.21 charged species ratio of 138Ba2+/138Ba+ were maintained below Bi 209 11.2 12.0 0.03 and 0.02, respectively.‘Omni’ ranges (i.e., reduced deflec- Tl 205 7.03 8.53 tor voltage) were used for the determination of several elements Th 232 9.34 7.57 such as Na, Mg, Al, Si, K and Mn. U 238 7.86 5.4 The ICP-MS system was calibrated using external standards 940 J. Anal. At. Spectrom., 1999, 14, 939–946Table 2 Spectral interferences on Co, As and Se in ICP-MS for biological samples Element m/z Chloride/fluoride Oxide/hydride Doubly charged ionsa Co 59 40Ar19F+, 40Ca19F+ 43Ca16O+, 42Ca16OH+ As 75 40Ar35Cl+, 40Ca35Cl+ 59Co16O+, 43Ca16O2+ 150Nd2+ (5.64), 150Sm2+ (7.4), 151Eu2+ (47.8) Se 77 40Ar37Cl+, 40Ca37Cl+ 60Ni16OH+, 45Sc16O2+ 154Sm2+ (22.7), 154Gd2+ (2.18), 153Eu2+ (52.2) Se 82 12C35Cl2+ 66Zn16O+, 81BrH+ 164Dy2+ (28.2), 164Er2+ (1.61), 165Ho2+ (100.0) aPercentage natural abundance in parentheses Li, Be, Al and Sc and by about 5–9% for the heavier elements and 42CaOH+ on m/z 59, while the slope of the line represents the contribution of 40CaF+ when the percentage concentration such as Tl, Th and U.Boron is a light element with a relatively high ionization potential of 8.3 eV. Usually, a matrix consisting of HF (45–50%) is unity. Derived from the linear regression equation in Fig. 1, the interference of 40CaF+ at m/z 59 in a of a light element with a low degree of ionization leads to a minimal space charge environment.15,19 As a lighter and less solution containing 100 mg l-1 Ca and 0.04% HF solution was a similar contribution to that of the sum of 43CaO+ and energetic ion, B+ ions are defocused to a greater extent than Ar+.Little or no space charge in the beam current downstream 42CaOH+. A correction equation, 59M=59M-0.002543M, should be applied for the determination of 59Co+, where the of the skimmer is expected, so the signal enhancement may be attributable to the decrease in the space charge.The matrix coeYcient should be measured for each analytical batch.1 The spectral interferences on 75As+ and 77Se+ are generally induced signal suppression or enhancement could be compensated by using internal standardization18–22 and/or matrix from the high content of chlorides in biological samples.17–20 Both the 75As+ and 77Se+ signals suVer from the isobaric matching of calibration standards.20,22 overlap of argon chlorides (40Ar35Cl+ and 40Ar37Cl+). Actually, the spectral interferences on 75As+, 77Se+ and 82Se+ Evaluation of potential spectral interferences are very complicated.In addition to the argon chlorides, there In addition to the matrix induced signal enhancement, matrix are other interfering species, such as hydrides, oxides and induced spectral interferences were observed in the ICP-MS doubly charged species of some rare earth elements in the of these biological samples. Molecular background species in biological samples. As listed in Table 2, 75As+ may also be ICP-MS are derived from components of the overall sample aVected by 59Co16O+ and/or by the doubly charged species of solution matrix.Typical spectral interferences in biological the rare earth elements such as 150Nd2+, 150Sm2+ or 151Eu2+, samples digested in HNO3 were well documented in previous and 77Se+ may be aVected by 60NiOH+, 154Sm2+, 154Gd2+ or work.1 Spectral interferences originating from major elements 153Eu2+. Also, 82Se+ may be overlapped by the peaks of such as Ca, Cl, P, K, C, Na and S were usually found in the 81BrH+, 164Dy2+, 164Er2+ or 165Ho2+.These interferences are ICP-MS of biological samples.23–26 The most serious spectral not easily corrected and may cause biased results especially interferences for the HNO3–HF digestion system involve the when they dominate the interferences. monoxide, hydroxide, fluoride and argide species of Ca, Cl With the HNO3–HF–H3BO3 digestion system, the formation and C.These species are CaO+, CaOH+, CaF+, ArF+, ClO+, of boron oxides, boron hydroxides, boron hydrates and boron ArCl+ and ArC+, which aVect the results for 57Fe+, 59Co+, argides, such as BO+, BOH+, BH2 O+, BO+2, BO2 H+, 60Ni+, 63Cu+, 51V+, 52Cr+, 75As+ and 77Se+. As shown in B(OH)+2 and BAr+,10 aVect the determination of several Table 2, Co at m/z 59 suVers from interferences from 40ArF+, elements such as 27Al+, 43Ca+, 45Sc+, 47Ti+ and 51V+. All 43CaO+, 40CaF+ and 42CaOH+.The only diVerence between the isotopes of Ti and V from m/z 46 to 51 suVer from isobaric using HF–HNO3 and HNO3 alone is the interference of overlaps from boron argide species such as 10B36Ar+, 40CaF+ and 40ArF+ on m/z 59. The interference from 40ArF+ 11B36Ar+, 10B38Ar+, 11B38Ar+, 10B40Ar+ and 11B40Ar+. was easily corrected by subtracting reagent blanks. The inter- Because of the presence of the boron argides, the formation ference from 40CaF+ was more severe than that from 40ArF+, of boron–argon oxides and hydroxides is possible.These 43CaO+ and 42CaOH+, and was directly related to the concen- molecular species (10B40ArO+ and 11B40ArO+) may aVect the tration of HF in the solution. Fig. 1 shows that the signal determination of zinc. On the other hand, boron fluoride ratios of m/z 59 to m/z 43 in solutions containing 100 mg l-1 species such as BF+, BF2 +, BF3 + and BF4+ in the plasma Ca and various amounts of HF increased linearly with increase may also exist, which overlap with other peaks such as NO+, in HF concentration.The intercept of the linear regression 11B38Ar+ and 11BAr40OH+ peaks at m/z 30, 49 and 68, curve shown in Fig. 1 represents the contribution of 43CaO+ respectively. However, the boron matrix related interferences can be simply corrected through reagent blank subtraction.10 Analytical results The analytical results are given in Tables 3–6, in which the uncertainties are shown at the 95% confidence level.The matrix induced signal enhancement was compensated by using In as an internal standard. The matrix induced background interferences were corrected using reagent blank subtraction. Spectral interferences originating from the biological sample matrix were corrected by either selecting alternative analytical isotopes or using correction equations. The related correction equations for 51V+, 53Cr+, 57Fe+, 59Co+, 60Ni+, 75As+ and 77Se+ and the assigned equation numbers were given in a previous paper.1 Briefly, the default eqn. (0) indicates that no correction was applied, while the eqn.(1) or (2) relates to the results after interference correction. With eqn. (1) or (2), the Fig. 1 Signal ratios of m/z 59 to m/z 43 from solutions containing 100 mg l-1 of Ca and various concentrations of HF. results for 51V+, 53Cr+, 75As+ and 77Se+ were those after the J. Anal. At. Spectrom., 1999, 14, 939–946 941Table 3 Comparison of recoveries of elements in 0.5 g of NIST SRM 1547 Peach Leaves with HNO3 digestion and HNO3–HF–H3BO3 digestion.Recoveries of 59Co+ [eqn. (0)] are listed to demonstrate interferences from CaO+, CaOH+ and CaF+ Recovery (mean±uncertainty) (%) HNO3–HF–H3BO3 1.2 ml 5% H3BO3 6.0 ml 5% H3BO3 Eqn Certified HNO3 a Element m/z No. value/mg g-1 (no HF) 0.01 ml HF 0.05 ml HF 0.1 ml HF 0.1 ml HF 0.3 ml HF 0.5 ml HF Al 27 0 249±8 75.5±4 102±2 99.3±3 93± 6 88.9±2 90.0±3 94.5±0.3 Ba 137 0 124±4 102±8 102±1 101±1 97.4±4 101±2 101±0.5 102±2 Ca 43 0 15600±200 109±6 113±0.4 110±3 104±5 94.3±6 101±4 107±3 Cd 114 1 0.026±0.003 105±6 114±2 104±4 89± 5 95.6±17 87.8±15 86.2±11 Co 59 0 0.07b 118±10 143±5 332±18 503±15 147±11 408±9 935±75 Co 59 1 0.07b 79.2±9 112±5 102±9 105±4 115±9 107±4 109±19 Cr 53 0 1b 94.8±7 101±1 100±4 92.5±4 97.0±2 91± 3 101±3 Cu 65 0 3.7±0.4 105±3 105±2 101±2 98.1±3 93.2±1 95.9±3 101±2 Mn 55 0 98±3 97.7±5 104±1 99.6±3 93.7±4 94.5±4 97.8±1 100±3 Mo 98 0 0.06±0.008 94.3±11 88.9±2 105±9 92.3±5 90.2±5 95.4±4 90± 4 Ni 60 1 0.69±0.09 56.3±13 101±4 97± 1 94.6±11 98.5±6 89.6±3 104±9 Pb 208 1 0.87±0.03 102±7 94.5±0.7 94.9±0.4 96.9±0.8 93.2±0.5 89.8±2 90.7±2 Sb 121 0 0.02b 96.8±17 138±3 155±17 137±14 104±8 105±6 104±8 Sr 86 0 53±4 104±4 113±0.2 109±1 107±3 103±2 105±2 107±0.3 Th 232 0 0.05b 60.9±8 105±2 111±2 119±1 118±2 114±5 113±4 U 238 0 0.015b 65.1±8 87.8±9 99.5±9 105±4 97.9±14 98.6±16 104±10 V 51 0 0.37±0.03 99.0±5 113±4 114±5 104±7 99.3±5 106±7 115±2 Zn 66 0 17.9±0.4 108±4 110±2 106±3 103±2 97.8±1 99.9±3 103±3 aData from ref. 1.bNIST reference value. Table 4 Comparison of recoveries of elements in NIST SRM 1547 Peach Leaves and NIST SRM 1573 Tomato Leaves with HNO3–HF digestion and HNO3–HF–H3BO3 digestion. Results for 75As+, 59Co+, 52Cr+, 57Fe+, 60Ni+ and 51V+ [eqn. (0)] are listed to demonstrate interferences NIST SRM 1547 NIST SRM 1573 Recovery (mean±uncertainty) Recovery (mean±uncertainty) (%) (%) Eqn Certified HNO3–HF Certified HNO3–HF Element m/z No.value/mg g-1 –H3BO3 HNO3–HF value/mg g-1 –H3BO3 HNO3–HF Al 27 0 249±8 88.9±2.3 103±1.8 1200a —e 97.1±8 As 75 0 0.06±0.018 176±3 221±12 0.27±0.05 188±8 174±12 As 75 2 0.06±0.018 168±3b 190±14b 0.27±0.05 96.3±12 100±0.4 Ba 137 0 124±4 101±2, 3 100±2 57± 9d 112±0.3 116±4 Ca 43 0 15600±200 94.3±6.9 103±1, 30000±300 112±0.7 106±1 Cd 114 1 0.026±0.003 95.6±19 122±7 2.5±0.2d 104±0.4 104±0.1 Co 59 0 0.07a 147±12 189±4 0.525±0.046d 116±3 108±1 Co 59 1 0.07a 115±10b 99.1±3.5b 0.525±0.046d 105±4 92.6±1 Cr 52 0 1a 134±8 205±5 4.5±0.5 117±4 111±1 Cr 53 0 1a 97.0±2 104±4 4.5±0.5 109±6 104±2 Cr 53 1 1a 95.6±0.9 97.3±2.7 4.5±0.5 92.3±3 88.7±0.6 Cu 65 0 3.7±0.4 93.2±1.1 109±6 11±1 91.0±2 91.5±2 Fe 57 0 218±14 124±5 128±2 690±25 109±1 114±1 Fe 57 1 218±14 95.4±3 94.9±2 690±25 91.3±1 94.3±1 Mn 55 0 98±3 94.5±4 97.3±2 238±7 —e 95.8±0.7 Mo 98 0 0.06±0.008 90.2±6b 84.7±5b 0.53±0.09d 101±2 100±2 Ni 60 0 0.69±0.09 134±7 189±5 1.3±0.2d 152±3 161±1 Ni 60 1 0.69±0.09 98.5±7 95.1±6 1.3±0.2d 94.5±3 91.9±2 Pb 208 1 0.87±0.03 93.2±0.6 101±3 6.3±0.3 97.1±1 96.4±5.6 Sb 121 0 0.02a 104±9 114±4 0.036±0.007d 98.0±12 88.6±4 Se 77 0 0.12±0.009 <0.15 —c 0.054±0.006d <0.15 —c Se 82 0 0.12±0.009 <0.2 —c 0.054±0.006d <0.2 —c Sr 86 0 53±4 103±2 108±1 44.9±0.3 99.4±1 100±0.6 Th 232 0 0.05a 118±2 112±6 0.17±0.03 100±4 86.9±7 U 238 0 0.015a 97.9±16 95.3±6 0.061±0.003 94.1±9 80.0±10 V 51 0 0.37±0.03 99.3±5.7 105±3 1.2±0.2d 148±6 140±3 V 51 2 0.37±0.03 99.3±5.7 103±1 1.2±0.2d 127±2 124±1 Zn 66 0 17.9±0.4 97.8±1 97.8±4 62± 6 97.1±2 91.5±1 aNIST reference values.bBelow MDL but above DL; refer to Table 7. cBelow DL; refer to Table 7. dConsensus values.27 eOut of the linear calibration range; ‘Omni’ range or further dilution required. 942 J. Anal. At. Spectrom., 1999, 14, 939–946Table 5 Analytical results (mg g-1) and recoveries of elements in NIST SRM 1515 Apple Leaves and 1575 Pine Needles with HNO3–HF digestion.Results for 75As+, 59Co+, 52Cr+, 57Fe+, 60Ni+ and 51V+ [eqn. (0)] are listed to demonstrate interferences NIST SRM 1515 NIST SRM 1575 Eqn Certified Found Recovery Certified Found Recovery Element m/z No. (mean±uncertainty) (mean±uncertainty) (mean±uncertainty) (%) (mean±uncertainty) (mean±uncertainty) (mean±uncertainty) (%) Al 27 0 286±9 293±3.9 102±14 545±30 576±6.8 106±1.2 As 75 0 0.038±0.007 0.129±0.011b 339±29b 0.21±0.04 0.212±0.014 101±6.7 As 75 2 0.038±0.007 0.107±0.010b 282±26b 0.21±0.04 0.202±0.012 96.2±5.7 B 10 0 27± 2 28.0±0.6 104±2.2 17±2d 16.6±0.3 97.6±1.8 Ba 137 0 49±2 50.0±0.5 102±10 7.2±0.8d 7.46±0.11 104±1.5 Ca 43 0 15260±150 16797±123 110±0.8 4100±200 4522±67 108±1.6 Cd 114 0 0.013±0.002 0.0277±0.018b 213±138b 0.22±0.06d 0.211±0.018 95.9±8.2 Cd 114 1 0.013±0.002 0.0126±0.0032b 96.9±25b 0.22±0.06d 0.2±0.017 90.9±7.7 Co 59 0 0.09a 0.117±0.014b 130±16b 0.122±0.014d 0.121±0.003b 99.2±2.5b Co 59 1 0.09a 0.0751±0.014b 83.4±16b 0.122±0.014d 0.11±0.0028b 90.2±2.3b Cr 52 0 0.3a 1.58±0.050 527±17 2.6±0.2 3.59±0.1 138±3.8 Cr 53 0 0.3a 0.495±0.028 165±9.0 2.6±0.2 2.8±0.086 108±3.3 Cr 53 1 0.3a 0.422±0.033 141±11 2.6±0.2 2.76±0.088 106±3.4 Cu 63 0 5.64±0.24 5.52±0.17 97.9±3.0 3±0.3 3.00±0.19 100±6.3 Cu 65 0 5.64±0.24 5.51±0.24 97.7±4.3 3±0.3 2.97±0.18 99.0±6.0 Fe 57 0 83±5 147±2.8 177±3.4 200±10 236±21 118±10 Fe 57 1 83±5 75.4±2.3 90.8±2.8 200±10 217±21 109±10 Hg 202 0 0.044±0.004 0.057±0.011b 130±25b 0.15±0.05 0.133±0.012 88.7±8.0 K 39 0 16100±200 17668±2132 110±13 3700±200 3086±380 83.4±10 Mg 25 0 2710±80 2924±44 108±1.6 1220±160d 1145±30 93.9±2.5 Mn 55 0 54±3 53.1±0.41 98.3±0.8 675±15 672±9 99.6±1.3 Mo 98 0 0.094±0.013 0.086±0.010b 91.5±11b 0.15±0.05d 0.115±0.010 76.7±6.7 Na 23 0 24.4±1.2 29.9±3.8 123±16 50±30d 29.6±5.3 59.2±10 Ni 60 0 0.91±0.12 1.33±0.051 146±5.6 2.5±0.3d 2.56±0.09 102±3.6 Ni 60 1 0.91±0.12 0.859±0.046 104±5.1 2.5±0.3d 2.43±0.09 97.2±3.6 P 31 0 1590±110 1806±25 114±1.6 1200±200 1332±21 111±1.8 Pb 208 0 0.47±0.024 0.462±0.001 98.3±0.2 10.8±0.5 11.3±0.38 105±3.5 Pb 208 1 0.47±0.024 0.458±0.001 97.4±0.2 10.8±0.5 11.2±0.38 104±3.5 S 34 0 1800a 2267±122 126±6.8 1320±111d 1370±173 104±13 Si 29 0 944±200 814d 1119±282 137±35 Sb 121 0 0.013a 0.0126±0.0024 96.9±18 0.197±0.017d 0.179±0.013 90.9±6.6 Se 77 0 0.05±0.009 <0.1 —c 0.047±0.005d <0.1 —c Se 77 1 0.05±0.009 <0.15 —c 0.047±0.005d <0.15 —c Se 82 0 0.05±0.009 <0.2 —c 0.047±0.005d <0.2 —c Sn 118 0 <0.2 4.38±6.1 3.35±0.32 Sr 86 0 25±2 26.4±0.4 106±1.6 4.8±0.2 4.75±0.03 99.0±0.6 Th 232 0 0.03a 0.026±0.004b 86.7±13b 0.037±0.003 0.035±0.0003 94.6±0.8 Ti 47 0 17.8±0.7 13.7d 14.7±0.8 107±5.8 Tl 205 0 0.014±0.001 0.05a 0.046±0.002 92.0±4.0 U 238 0 0.006a 0.0070±0.0012b 117±20b 0.02±0.004 0.0178±0.001 89.0±5.0 V 51 0 0.26±0.03 0.28±0.018 108±6.9 0.39±0.07d 0.419±0.01 107±2.6 V 51 2 0.26±0.03 0.258±0.019 99.2±7.3 0.39±0.07d 0.409±0.01 105±2.6 Zn 66 0 12.5±0.3 11.3±0.043 90.4±0.3 67±9d 78.1±21 116±31 aNIST reference values.bBelow MDL, but above DL; refer to Table 7. cBelow DL; refer to Table 7. dConsensus values.27 J. Anal. At. Spectrom., 1999, 14, 939–946 943Table 6 Analytical results (mg g-1) and recoveries of elements in NIST SRM 1567a Wheat Flour and 1568a Rice Flour with HNO3–HF digestion NIST SRM 1567a NIST SRM 1568a Eqn Certified Found Recovery Certified Found Recovery Element m/z No. (mean±uncertainty) (mean±uncertainty) (mean±uncertainty) (%) (mean±uncertainty) (mean±uncertainty) (mean±uncertainty) (%) Al 27 0 5.7±1.3 4.85±0.004 85.0±0.1 4.4±1 4.69±0.01 107±0.2 As 75 0 0.006a <0.01 —c 0.29±0.03 0.298±0.012 103±4 Ca 43 0 191±4 205±1 107±0.7 118±6 125±0.5 106±0.4 Cd 114 1 0.026±0.002 0.0282±0.0003 108±1 0.022±0.002 0.0241±0.002 109±9 Co 59 1 0.006a <0.015 —c 0.018a <0.015 —c Cu 65 0 2.1±0.2 2.07±0.01 98.8±0.6 2.4±0.3 2.47±0.008 103±0.3 Fe 57 1 14.1±0.5 14.1±0.6 100±4 7.4±0.9 7.77±0.43 105±6 K 39 0 1330±30 1263±3 95.0±0.2 1280±8 1228±9 95.9±0.7 Mg 25 0 400±20 403±4 101±1 560±20 572±5 102±0.9 Mn 55 0 9.4±0.9 9.57±0.08 102±0.8 20±1.6 20.5±0.12 103±0.6 Mo 98 0 0.48±0.03 0.487±0.001 102±0.1 1.46±0.08 1.50±0.03 103±2 P 31 0 1340±60 1497±3 112±0.2 1530±80 1713±14 112±0.9 Pb 208 1 <0.02a 0.014±0.001b — <0.01a 0.011±0.002b — S 34 0 1650±20 1671±46 101±3 1200±20 1312±23 109±2 Se 77 0 1.1±0.2 1.18±0.18 107±17 0.38±0.04 0.369±0.066 97.1±17 Sn 118 0 0.003a <0.005 —c 0.005a <0.005 —c U 238 0 0.0003a <0.002 —c 0.0003a <0.002 —c V 51 1 0.011a <0.05 —c 0.007a <0.05 —c Zn 66 0 11.6±0.4 11.5±0.1 98.9±0.6 19.4±0.5 18.5±0.2 95.3±1.0 aNIST reference value.bBelow MDL, but above DL; refer to Table 7. cBelow DL; refer to Table 7. 944 J. Anal. At. Spectrom., 1999, 14, 939–946Table 7 Comparison of ICP-MS detection limits (mg g-1, directly in correction for ArCl+ and ClO+ interferences, while the results dry solid) using diVerent digestion procedures for the primary isotopes for 57Fe+, 59Co+ and 60Ni+ were those after the correction selected in this study for CaF+ and/or CaO+ and CaOH+ interferences.1 The HNO3–HF–H3BO3 digestion system was tested for HNO3–HF–H3BO3 HNO3–HF 0.5 g of NIST SRM 1547 Peach Leaves, using 0.01, 0.05 and Eqn Element m/z No.DLa MDLb DLa MDLb 0.1 ml of HF, each with 1.2 ml of 5% H3BO3, or using 0.1, 0.3 and 0.5 ml of HF, each with 6 ml of 5% H3BO3.As shown Al 27 0 0.4 1.5 0.1 2c in Table 3, the use of 0.1 ml of HF (45–50%) per gram of dry As 75 0 0.025 0.2 0.01 0.05 sample was found to be suYcient to improve the recoveries As 75 1 0.1 0.5 0.1 0.2 (within 85–115% in most cases) for the silicon-bound elements B 10 0 0.4 1 0.4 2 Al, Co, Ni, Th and U, compared with those obtained using Ba 137 0 0.02 0.3 0.01 0.5 Be 9 0 0.3 3 0.04 0.1 the HNO3 digestion system.1 The use of 0.25–0.5 ml HF per Ca 43 0 8 15 5 12 gram dry mass has been reported to give a complete recovery Cd 114 1 0.025 0.06 0.01 0.04 of Si in food samples.4 The silicon content in most of the Cl 35 0 5 450 80 360 plant and grain samples is about 0.1–0.3%.9 The stoichiometric Co 59 0 0.005 0.1 0.015 0.2 amount of HF required for 0.3% of Si is about 0.03 ml of 50% Co 59 1 0.05 0.5 0.015 0.3 HF per gram of sample.To ensure complete digestion, 0.2 ml Cr 53 0 0.05 0.4 0.1 0.2 Cu 65 0 0.1 0.25 0.05 0.25 of HF (45–50%) per gram of dry sample was selected for Fe 57 0 1.5 4 1.5 4 subsequent use in this study, being about 3–10-fold in excess Fe 57 1 1.5 4 1.5 4 of the amount required. The amount of HF could be increased Hg 202 0 0.02 0.05 0.02 0.05 to a maximum of 1 ml if the Si content in plant materials is K 39 0 50 100c 50 100c considerably higher.In Table 3, the recoveries of 59Co+ with Li 9 0 0.05 0.15 0.05 0.3 eqn. (0) (i.e., without interference correction) are listed to Mg 25 0 0.5 1.5c 0.2 1c Mn 55 0 0.015 0.025 0.015 1c demonstrate the interferences from CaO+, CaOH+ and especi- Mo 98 0 0.01 0.1 0.01 0.1 ally CaF+.They were biased high and increased with increase Na 23 0 2 4c 2 4c in the amount of HF used. The recovery of 59Co+ with eqn. Ni 60 0 0.04 0.4 0.1 0.2 (1) was significantly improved after the interference coeYcient Ni 60 1 0.04 0.4 0.1 0.2 from CaF+, CaO+ and CaOH+ was derived and applied in P 31 0 — — 1.5 10 the correction equation, as was done in previous work.1 Pb 208 1 0.01 0.04 0.02 0.08 S 34 0 — — 50 100 As demonstrated in Table 4 for NIST SRM 1547 and 1573, Sb 121 0 0.005 0.008 0.0025 0.008 there were no significant diVerences between the ICP-MS Se 77 0 0.1 0.15 0.15 0.2 results using the HNO3–HF or the HNO3–HF–H3BO3 diges- Se 82 0 0.2 0.5 0.2 0.4 tion systems.With the HNO3–HF digestion system, the Si 29 0 — — 12 800c recoveries for Al, As, B, Ba, Ca, Co, Cd, Cr, Cu, Fe, Hg, Mg, Sn 118 0 0.025 0.06 0.01 0.15 Mn, Mo, Ni, Pb, Sb, Se, Sr, Th, Ti, Tl, U, V and Zn in NIST Sr 86 0 0.025 0.1 0.05 0.15 Th 232 0 0.0025 0.05 0.003 0.03 SRM 1547, 1573, 1515, 1575, 1567a and 1568a were mostly in Ti 47 0 0.5 1 0.15 0.3 the range 85–115% (Tables 4–6).By applying the ‘Omni’ Tl 205 0 0.0035 0.008 0.002 0.005 range, this method can also be used to determine major U 238 0 0.0025 0.006 0.002 0.008 elements such as K, Na, Mg, P, S and Si in plant and grain V 51 0 0.15 0.5 0.01 0.15 materials (Tables 5 and 6).V 51 1 0.15 0.5 0.05 0.15 Similarly to the HNO3 digestion, carbon residue still Zn 66 0 0.3 0.15 0.1 0.3 remained in solution after microwave digestion with aDetection limits derived from three times the standard deviation for HNO3–HF. The problem of the spectral interferences from blank solutions containing 2% HNO3 and 0.04% HF (45–50%) or polyatomic species involving carbon remained the same, as containing 2% of HNO3, 0.04% of HF (45–50%) and 0.12% of H3BO3 (n=10), assuming an overall dilution factor of 500 (v/m).bMethod seen from the biased high recovery of 52Cr+ arising from the detection limits derived from within-run standard deviations of dupli- interferences from 40Ar12C+. The biased high recovery of cate digestion and analysis of SRMs and digestion blanks (n7) with 82Se+ mainly came from the interferences of 81BrH+, whose an overall dilution factor of 500 (v/m). c‘Omni’ range used. contributions could be significant, but were not corrected. After correcting for ArCl+ interference, the recovery for 75As+ remained high for SRM 1515 and 1547 (Tables 4 and 5), because of the low As content and the non-corrected inter- pared with the HNO3–HF digestion system, mainly owing to the spectral interferences from the boron matrix.ferences (Table 2) which might dominate the signal at m/z 75. Satisfactory results for Se were obtained for grain materials SRM 1567a and 1568a in Table 6. The Se levels in the plant Conclusions materials in Tables 4 and 5 are all below the detection limits. The instrument detection limits (DLs) and the method For plant tissue and grain materials in which the Si content is usually low, the closed-vessel microwave-assisted HNO3–HF detection limits (MDLs) for both digestion systems are presented in Table 7.The MDLs were derived from the within- digestion is favorable for elemental analysis by ICP-MS. The amount of HF used in the digestion is so low (0.1 ml per 0.5 g run standard deviation of duplicate digestions of standard reference materials and digestion blanks10 (n7), and were of sample) that the coprecipitation of metal fluorides in the solution is negligible and clogging of the ICP-MS nebulizer or 2–10-fold greater than the corresponding DLs.The ‘Omni’ ranges that were applied reduced the sensitivities for Al, K, the sampling cone will not occur. Hence the use of the boric acid in the post-digestion stage was not necessary.The sample Mg, Mn and Na and resulted in poorer MDLs for these elements. With the exception of Co, which suVered from the dissolution was more complete with the HNO3–HF digestion system than that with the HNO3 digestion system. Using the high background signal from ArF+, the DLs with the HNO3–HF digestion system were similar to those reported HNO3–HF digestion system, the previously reported low recoveries for silicon-bound elements such as Al, Co, Ni, Th, with the HNO3 digestion system.1 The DLs for 9Be+, 27Al+, 43Ca+, 65Cu+, 47Ti+, 51V+ and 66Zn+ with the U and V in SRMs digested with HNO3 were significantly improved.In only one instance, that of U for SRM 1573, was HNO3–HF–H3BO3 digestion system were relatively poor com- J. Anal. At. Spectrom., 1999, 14, 939–946 94512 L. M. Jalkanen and E. K. Ha�sa�nen, J. Anal. At. Spectrom., 1996, the recovery low (80%) for the HNO3–HF digestion, whereas 11, 365–369. it was adequate (94%) with the HNO3–HF–H3BO3 digestion 13 C.Vandecasteele, H. Vanhoe and R. Dams, J. Anal. At. Spectrom., (Table 4). Biased high recoveries of V (148%) and Cr (141%) 1993, 8, 781. in Tables 4 and 5 may be due to the results being compared 14 G. Horlick, Spectroscopy, 1992, 7, 22. with a reference value and a consensus value.27 Similarly to 15 S. H. Tan and G. Horlick, J. Anal. At. Spectrom., 1987, 2, 745. 16 M. A. Vaughan and G. Horlick, J. Anal. At. Spectrom., 1989, the HNO3 digestion procedure, the HNO3–HF digestion pro- 4, 45. cedure is simple and rapid, and allows the determination of 17 D. Beauchemin, J. W. McLaren and S. S. Berman, Spectrochim B, long known to be an essential element for plant growth.28 Acta, Part B, 1987, 42, 467. 18 J. J. Thompson and R. S. Houk, Appl. Spectrosc., 1987, 41, 801. 19 G. R. Gillson, D. J. Douglas, J. E. Fulford, K. W. Halligan and S. D. Tanner, Anal. Chem., 1988, 60, 1472. References 20 S. J. Stotesbury, J. M. Pickering and M. A. GriVerty, J. Anal. At. Spectrom., 1989, 4, 457. 1 S. Wu, X. Feng and A. Wittmeier, J. Anal. At. Spectrom., 1997, 21 X. Feng and G. Horlick, J. Anal. At. Spectrom., 1994, 9, 823. 12, 797. 22 I. Rodushkin, T. Ruth and D. Klockare, J. Anal. At. Spectrom., 2 B. Madeddu and A. Rivoldini, At. Spectrosc., 1996, 17, 148. 1998, 13, 159. 3 K. Shiraishi, Y. Takacu, K. Yoshimizu, Y. Igarashi, K. Masuda, 23 M. A. Vaughan and G. Horlick, Appl. Spectrosc., 1986, 40, 434. J. F. McInroy and G. Tanaka, J. Anal. At. Spectrom., 1991, 6, 335. 24 H. Vanhoe, J. Goossens, L. Moens and R. Dams, J. Anal. At. 4 A. Krushevska, A. La�sztity, M. Kotrebai and R. M. Barnes, Spectrom., 1994, 9, 177. J. Anal. At. Spectrom., 1996, 11, 343. 25 Y. Shao and G. Horlick, Appl. Spectrosc., 1991, 45, 143. 5 L. Xu and W. Shen, Fresenius’ J. Anal. Chem., 1989, 333, 108. A. Vaughan and D. M. Templeton, Appl. Spectrosc., 1990, 6 C. S. E. Papp and L. B. Fischer, Analyst, 1987, 112, 337. 44, 1685. 7 L. B. Fischer, Anal. Chem., 1986, 58, 261. 27 Compilation of Elemental Concentration Data for NBS Clinical, 8 P. J. Lamothe, T. L. Fries and J. J. Consul, Anal. Chem., 1986, Biological, Geological, and Environmental Standard Reference 58, 1881. Materials, National Bureau of Standards, Gaithersburg, MD, 1987. 9 R. A. Nadkarni, Anal. Chem. 1984, 56, 2233. 28 S. Evans and U. Kra�henbu� hl, J. Anal. At. Spectrom., 1994, 9, 10 S.Wu, Y. Zhao, X. Feng and A. Wittmeier, J. Anal. At. Spectrom., 1249. 1996, 11, 287. 11 C. F. Wang, W. H. Chen, M. H. Yang and P. C. Chiang, Analyst, 1995, 120, 1681. Paper 8/04683B 946 J. Anal. At. Spectrom., 1999, 14, 939&ndash

ISSN:0267-9477    

DOI:10.1039/a804683b

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Capabilities of fast protein liquid chromatography coupled to a double focusing inductively coupled plasma mass spectrometer for trace metal speciation in human serum Marý�a Montes Bayo�n, A. Bele�n Soldado Cabezuelo, Elisa Blanco Gonza�lez, J. Ignacio Garcia Alonso and Alfredo Sanz-Medel* Department of Physical and Analytical Chemistry, University of Oviedo, 33006 Oviedo, Spain Received 18th November 1998, Accepted 19th April 1999 The analytical potential of fast protein liquid chromatography (FPLC) coupled with a double focusing inductively coupled plasma mass spectrometer (DF-ICP-MS) was evaluated for the multielemental speciation of trace elements in human serum.The separation of the main serum proteins was performed in a MonoQ (HR5/5) anion exchange column using an ammonium acetate gradient (0–0.5 mol l-1) at the physiological pH of 7.4 (0.05 mol l-1 TRISHCl buVer) as the mobile phase. Proteins were first detected on-line spectrophotometrically at 295 nm and specific detection of metals was also carried out on-line by using a double focusing ICP-MS working at both low (m/Dm= 300) and medium resolution (m/Dm=3000) in order to avoid possible polyatomic interferences.The use of variable resolution to carry out multielement speciation studies allowed the detection of Ca, Sr, Fe, Cu, Zn, Se, Mn, Cr, Pb, Al and Sn in diVerent serum samples and introduces a new dimension to this area of investigation. From the multielemental study it was observed that detectable levels of Cr, Al, Pb and Sn were present in uraemic sera while those elements could not be detected in healthy subjects.Using resolution settings (m/Dm) up to 3000, many well known Introduction polyatomic ions overlapping the masses of the analytes can be Over the last decade, inductively coupled plasma mass resolved.16 Although this setting implies a decrease in sensispectrometry (ICP-MS) has become a most powerful technique tivity, the detection limits can still be adequate as we have for trace and ultratrace elemental analysis of biological mate- recently reported17 for aluminium speciation studies in human rials, especially body fluids.Rapid analysis, excellent detection serum, where the coupling of anion exchange fast protein limits, multielemental capabilities and the ability to measure liquid chromatography (FPLC) with DF-ICP-MS provided a isotopic ratios explains why ICP-MS applications are rocketing most reliable and sensitive hybrid technique for studying the these days for the determination of the total elemental concen- binding of Al to serum proteins.tration in serum,1–3 blood,4–6 urine,7–9 etc. However, in recent In the present work, the suitability of the previously reported years it has become recognised world-wide that both the FPLC-DF-ICP-MS methodology16 for multielemental trace toxicological and the essential function of a trace element are element speciation in human serum is investigated.Taking primarily understood only by studying all the diVerent chemi- advantage of the capability, provided by the ICP-MS detector, cal forms in which the sought element can be present in the of measuring many elements almost at the same time, a sample. In other words, speciation information rather than multielemental survey of those elements present in serum total metal content is needed for adequate decision making suVering spectral interference and those which do not has been after analytical results have been obtained in the laboratory.conducted. Metal speciation studies in biological materials can be The use of low resolution (m/Dm=300) allows metal performed by several techniques,10,11 although coupling a detection in the separated serum fractions at extremely low concentration levels (due to the combination of a high ion powerful separation technique (e.g., a chromatographic techtransmission and low background given by the DF instru- nique) with an element specific detector (e.g., ICP-MS) is the ments), while at m/Dm=3000 most isotopes of interest can be most promising approach.The use of quadrupole ICP-MS separated from their spectral interferences.16 These studies (Q-ICP-MS) as a detector for liquid chromatography (e.g., were carried out in basal human serum from both uremic size exclusion chromatography, anion exchange, etc.), has patients and healthy subjects and the serum proteins were provided satisfactory results for studying protein binding monitored by UV absorption.biometals in biological fluids.10–15 Unfortunately, the literature on this issue is still rather limited. The chromatographic separation of a compound from its matrix should avoid the formation of most interfering poly- Experimental atomic ions from matrix elements. However, isobaric inter- Instrumentation ferences originating from polyatomics from plasma and atmospheric gases or from reagents used and the mobile phases The chromatographic systems used for the separation of the still can pose a serious threat to accurate species determination.proteins were: a Shimadzu (Kyoto, Japan) LC-10 A HPLC An eVective solution to overcome these remaining interference pump with a Rheodyne (Cotati, CA, USA)Model 7125 sample problems is the use of double focusing instruments injection valve fitted with a 100 mL loop, a 10 mm anionexchange Mono-Q HR 5/5 FPLC analytical column (DF-ICP-MS) as LC detectors instead of a Q-ICP-MS.11 J.Anal. At. Spectrom., 1999, 14, 947–951 947Table 1 Optimum operating conditions of the DF-ICP-MS (50×5 mm id) (Pharmacia Biotech, Uppsala, Sweden), a UV/VIS absorption detector from LKB (Bromma, Sweden), Instrument Element (Finnigan MAT) Model 2151, and a Shimadzu C-R6A recording integrator. A Resolution (m/Dm) 300 and 3000 scavenger column (25×0.5 cm id) was placed between the Rf power 1290 W pump and the injection valve to prevent exogenous metallic Nebulizer Meinhard Spray chamber Scott-type, double pass, contamination coming (mainly) from the mobile phases used.room temperature This column was packed with Kelex-100 impregnated silica C18 material (20 mm particle size, prepared as described Gas flow rates Ar outer 14 l min-1 elsewhere18). Ar intermediate 0.85 l min-1 Multielemental detection was carried out using a Ar carrier 0.95 l min-1 DF-ICP-MS (Element, Finnigan MAT, Bremen, Germany), Ion lens setting which permits operation in three diVerent resolution modes Extraction -2000 V (300, 3000 and 7500, 10% valley definition).The instrument Focus quadrupole -800 V was fitted with a conventional Meinhard nebulizer and a X-deflection quadrupole -6 V double-pass spray chamber. Y-deflection quadrupole -10 V Shape -125 V Reagents High resolution lensesa Rotation quadrupole 1 3.98 V Ammonium acetate and tris(hydroxymethyl )aminomethane Rotation quadrupole 2 0.12 V (TRIS) of highest purity available were obtained from Merck Focus quadrupole 1 -3.14 V (Darmstadt, Germany). Kelex 100 (Schering Espan� a, Madrid, Focus quadrupole 2 -1.36 V Spain) was used as received and the bonded silica reversedaOptimised when using m/Dm=3000. phase sorbent with octadecyl functional groups (C18) was purchased from Waters (Milford, MA, USA).Human protein standards (albumin, apotransferrin and immunoglobulin G) were obtained from Sigma (St.Louis, MO, USA) of diVerent compounds, and even when this solution is frac- Uremic and normal human serum samples were kindly tionated by FPLC each fraction is still a very complex mixture. provided by the Hospital Central of Asturias and the Centro However, some studies19 have shown the identification of the de Transfusiones (Asturias, Spain). All serum samples were main components of each fraction and confirmed its applicainjected into the chromatographic system without any previous bility for comparative analysis of serum proteins during treatment or dilution.therapy. A multielemental standard solution containing Li, Sc, Co, Good separation of the main serum proteins was achieved Y, Rh, In, Tb, Tl, Th and U at 10 ng g-1 was prepared from in previous studies15 carried out in our group using a Mono the 1000 mg ml-1 standard solutions of each element. These Q (HR 5/5) anion exchange column with a gradient standardsined from Merck (Li, In, Ba and Tb) and (0–0.25 mol l-1) of sodium chloride in TRIS-HCl buVer from J.T.Baker (Phillipsburg, NJ, USA) (Sc, Co, Y, Rh, Th (pH 7.4) as the mobile phase. However, the high level of and U), respectively. In order to reduce the risk of contami- sodium used there provided salt deposits on the sampling cone nation all work was carried out under clean room conditions. of the DF-ICP-MS. In order to avoid cone blocking, several reagents were tested as alternative mobile phases, including Procedures lithium chloride and ammonium acetate.In both cases the separation of serum proteins was successfully achieved, but A standard solution of human serum proteins containing long-term results have shown a better performance of the cone 0.5 g l-1 apo-transferrin, 5 g l-1 albumin and 1 g l-1 immunoorifice when using ammonium acetate as mobile phase.17 In globulin G was prepared in 0.01 mol l-1 TRIS-HCl buVer this case, no deposits on the cones were observed, and no (pH 7.4).The separation of serum proteins by anion exchange variations on the ICP-MS analytical signal during the gradient FPLC in both standard solution and serum samples was were appreciated for the elements under study. In the case of performed using a linear gradient of ammonium acetate Mn and Pb only, the background was noticeably increased (0–0.25 mol l-1) in a 0.05 mol l-1 TRIS-HCl buVer (pH 7.4) when increasing the percentage of ammonium acetate in the in 15 min.The injected volume was 100 ml. mobile phase, probably due to contamination eVects not The mobile phases employed were de-gassed with helium eliminated by the scavenger column. It has also to be pointed for 15 min before use. The eluate from the FPLC column was out that, owing to the diameter of the central channel of the passed through a UV detector, set at 295 nm for protein torch used in the DF-ICP-MS (1 mm), daily cleaning was monitoring, and a DF-ICP-MS detector for multi-element necessary to avoid its clogging probably due to the plasma detection and the corresponding chromatograms were proteins of the serum matrix.recorded. In order to prevent changes in the in-vivo association of A multielemental solution containing Li, Sc, Co, Y, In, Tb, metals to serum components, the separation was performed at Tl, Th and U at 10 ng g-1 was used to optimise the operation the physiological pH of 7.4 (0.05 mol l-1 TRIS-HCl buVer). conditions of the DF-ICP-MS instrument, as well as to carry Using such chromatographic conditions the observed elution out mass calibration at R=300 and 3000.The final operating profile of the main proteins in a normal serum is illustrated conditions of the ICP-MS system are detailed in Table 1. in Fig. 1. Three peaks, identified as immunoglobulin G, transferrin Results and discussion and albumin,15 were confirmed by spiking standards of the three proteins to the serum sample. A similar profile of the The method of choice for many authors performing the separation and identification of protein bound to trace proteins was obtained in the case of uraemic serum samples.As is shown in Fig. 1, immunoglobulin G eluted at 2 min elements appears to be FPLC. This technique oVers the major advantage that the separations occur in a short time (e.g., while transferrin eluted in two incompletely separated peaks at 6.8 and 7.8 min (which could be ascribed to the presence of 15 min), preventing possible interactions between the column beads and the sample components, which may cause false diVerent transferrin molecular forms17–19). Albumin, on the other hand, elutes at approximately 13 min from the column. results.Serum is a very complex mixture containing thousands 948 J. Anal. At. Spectrom., 1999, 14, 947–951Fig. 2 Elution profile of calcium, iron and copper (multiplied by 10) in an undiluted uraemic human serum using the coupling FPLC-DFICP- MS at R=3000.As expected, 56Fe eluted in two peaks that clearly overlapped Fig. 1 Elution profiles of the main serum proteins in an undiluted normal serum by UV absorption at 295 nm (immunoglobulin G, IgG, with the transferrin peaks in the UV absorption profile. This transferrin, Tf, and albumin, Alb). result is in agreement with the well known fact that about 97% of the total Fe content in serum is bound to transferrin, forming three diVerent complexes which diVer with respect to Selection of measurement conditions the amount of bound iron.18,20 The chromatogram for 63Cu In a previous paper17 aluminium binding to serum proteins in Fig. 2 shows that the main signal appears at 15.7 min in was studied using a DF-ICP-MS, as an aluminium-specific protein fractions corresponding to those eluting from the detector, coupled on-line to FPLC for protein separation. In column as a shoulder on the albumin peak (Fig. 1). Although the present work the multielement detection capabilities of the this protein fraction was not further investigated, it could be DF-ICP-MS instrument have been investigated to establish ascribed to the protein ceruloplasmin, according to previously metal–protein simultaneous profiles of several essential and reported19,21 publications where the fractions eluting from the toxic element in human serum.Table 2 indicates the elements FPLC column were identified. It is also noteworthy that a for which ICP-MS detection would not be hampered by small amount of 63Cu appears in the dead volume of the spectral interferences and so they were detected using a reso- column, as shown in Fig. 2 (i.e., positive free ion unretained lution setting of 300.The other element (upper part of Table 2) by the anion exchange column or weakly bound). required a resolution setting of at least 3000 to separate the Calcium is an essential element needed for protein sythesis analyte signal from the potential interfering polyatomic ions and has been monitored at m/z 44 (Fig. 2). It was found that (also shown in the third column of Table 2). Arsenic-75 would both for healthy subjects and uraemic patients 44Ca elutes need even higher resolution to be separated from 40Ar35Cl from the column in the dead volume at approximately 1 min polyatomics. (i.e., free ion or weakly bound to biomolecules). The DF-ICP-MS elution profiles for 56Fe, 63Cu and 44Ca Metal elution profiles using a resolution setting of 3000 from healthy persons serum samples were very similar to those obtained from uraemic serum (Fig. 2), except for the case of As a result of the spectral interferences detailed in Table 2, iron where the ICP-MS intensity signal of the two Fe peaks R=3000 was necessary to avoid polyatomics in the signal in normal serum was much higher than uraemic serum (in corresponding to 44Ca, 63Cu, 56Fe, 52Cr, 55Mn and 64Zn in agreement with previous studies22), probably because dialysis serum samples.Fig. 2 shows the elution profile of 44Ca, 56Fe patients suVer from iron deficiency. and 63Cu in an undiluted uraemic human serum (similar profile Zinc showed a diVerent chromatographic elution profile was obtained for normal serum) using the chromatographic comparing normal and uraemic serum samples. This can be conditions detailed in the procedures and the ICP-MS appreciated in the chromatograms shown in Fig. 3. Zinc operating conditions shown in Table 1.(monitored at mass 64) was found both in normal and uraemic serum, in fractions corresponding to the transferrin Table 2 Polyatomic interferences of the elements under study and peak at 7.8 min (Fig. 1) and also in other two serum fractions, resolution used one eluting at approximately 6 min (between immunoglobulin G and transferrin) and the other eluting at 11–12 min (between Element Mass Interfering species Resolution used transferrin and albumin). This last 64Zn fraction at 12 min could correspond to the protein a2-macroglobulin (720 kDa) Ca 44 12C16O2 3000 which, according to the literature, elutes from the FPLC Fe 56 40Ar16O 3000 Cu 63 40Ar23Na 3000 column at that retention time19 (this protein seems to be the Zn 64 32S16O2 3000 main Zn distributor throughout the organism21).Again, some Cr 52 40Ar12C, 35Cl16O1H 3000 64Zn is also detected eluting from the column in the dead Mn 55 40Ar14N1H 3000 volume. As 75 40Ar35Cl 7500 However, the distribution of 64Zn in the human serum Se 82 — 300 fractions is diVerent depending on the type of serum: in Sr 88 — 300 Cd 111 — 300 uraemic serum (Fig. 3A) the amount of 64Zn associated with Sn 120 — 300 the fraction eluting at 11 min decreases significantly, while the Hg 200 — 300 signal intensity of the peak eluting at 6 min increases when Pb 208 — 300 compared to normal human serum (Fig. 3B). Both normal Bi 209 — 300 and uraemic sera showed peaks corresponding to transferrin J. Anal.At. Spectrom., 1999, 14, 947–951 949results confirm this finding, as shown by Fig. 4A. A single chromium peak was observed, corresponding to just one of the transferrin forms. In Fig. 4B, the elution profile of 55Mn is shown. This element was found to elute from the column mainly as a peak after 5 min. This peak overlapped with the elution peak of unidenti- fied UV absorbent serum fraction, as shown in Fig. 1. According to the literature,20 diVerent chains of immunoglobulin and low-molecular-weight components of serum could be present in this unidentified serum fraction eluting at approximately 6 min, containing also 64Zn .Previous work carried out by Bra�tter et al.24 by size exclusion chromatography-ICPAES found 55Mn in serum fractions corresponding to diVerent molecular weight proteins, but no positive identification of these proteins has been published. A small amount of 55Mn also eluted from the column in the dead volume, probably as free (solvated) ion (Fig. 4B). Metal elution profiles using a resolution setting of 300 On line DF-ICP-MS detection of 88Sr, 82Se, 120Sn and 208Pb at R=300 after FPLC separation of uraemic serum samples shows the elution profiles given in Fig. 5. In living systems, strontium behaviour tends to follow calcium metabolism and uptake, internal organs distribution and excretion. As Fig. 3 Zinc elution profiles by FPLC-ICP-MS at R=3000 for: A, expected, 88Sr was found to be present in the serum following uraemic human serum; B, normal serum.similar behaviour to Ca2+ and therefore eluting from the column in the fraction of the electrolytes at about 1 min, as and free ions fraction with similar size. This result suggests shown by Fig. 5A. A relatively small Sr peak was also observed further work to elucidate possible 64Zn speciation modification at a retention time of 14.7 min, overlapping the albumin peak. linked to renal failure disease. Total strontium levels found in uraemic sera seemed to be For chromium, the sensitivity of the DF-ICP-MS instrument elevated in comparison with normal sera,22 although no sig- at R=3000 was insuYcient for on-line Cr detection after nificant diVerences were detected in the chromatographic FPLC separation of serum samples from healthy people (total profiles of both types of sera.concentration of Cr reported to be very low, 0.05 ng g-1).22 The 82Se and 120Sn elution profiles are observed in Fig. 5A. However, chromium speciation studies were possible in ura- Related to 82Se speciation in serum, three selenium-containing emic serum where the concentration of the element was much plasma proteins have been identified in the literature: seleno- higher (2.3 ng g-1).22,23 Fig. 4 (A and B) shows the chromatoprotein P, glutathione peroxidase and albumin, using aYnity gram of a uraemic serum obtained under the selected working chromatography25 in an estimated ratio of 50540510. Using conditions.As can be seen, the chromium elution profile FPLC and the gradient detailed in the procedure section the showed a single peak (Fig. 4A), which clearly overlaps one of separation of the fractions corresponding to selenoprotein P the transferrin peaks obtained by UV absorption (see Fig. 1), and glutathione peroxidase cannot be observed. However, a so confirming the previously reported association of the metal small 82Se peak appears at the retention time of albumin with this serum protein.23 According to Cornelis and De accordingly to previous results26 using this column.For Sn Kimpe,23 patients with terminal renal failure treated with haemodialysis become iatrogenically loaded with Cr. Our Fig. 5 Elution profile of a uraemic serum by FPLC-DF-ICP-MS Fig. 4 Chromatograms obtained at R=300 in a uraemic serum for: using a resolution setting of 300 for: A, selenium, strontium and tin; B, lead. A, chromium; B, manganese. 950 J. Anal. At. Spectrom., 1999, 14, 947–951bound to serum proteins, it can be appreciated in Fig. 5A that at physiological element levels (e.g., Se in normal serum, Sn and Pb in uraemic serum). a single peak appears at 13.7 min when m/z 120 is monitored (a similar behaviour to 82Se), corresponding to the albumin This multielement capacity to analyse any single chromatographic peak along with the DF-ICP-MS detector ability to elution profile. Tin was only found in serum obtained from uraemic patients and no signal was noticed for healthy using higher resolution (if needed) or lower resolution/higher sensitivity (when possible) open up new avenues to trace subjects¡� sera.Fig. 5B also shows the observed Pb elution profile at mass element analytical speciation in biological systems and its future application to health and disease diagnosis. 208. Several Pb speciation studies in blood have been published using quadrupole ICP-MS coupled to size exclusion chromatography. 27,28 These studies showed that lead seems to elute References in a fraction ascribed to ceruloplasmin27 and ferritin.28 We 1 J.Riondato, F. Vanhaecke, L. Moens and R. Dams, J. Anal. At. observed, with the separation of the proteins carried out by Spectrom., 1997, 12, 933. FPLC, that the main 208Pb peak appeared at approximately 2 J. Szpunar, J. Bettmer, M. Robert, H. Chassaigne, K. Cammann, 6 min (with a similar elution profiles to that shown previously R. Lobinski and O. F.X. Donard, Talanta, 1997, 44, 1389. for 55Mn, Fig. 1B). Moreover, Pb peaks were only observed 3 C. Sariego Mun. iz, J. M. Marchante Gayo¢¥ n, J. I. Garc©¥¢¥a Alonso when the speciation was performed in uraemic serum. Again, and A. Sanz-Medel, J. Anal. At. Spectrom., 1997, 13, 283. 4 J. Begerow and L. Dunemann, J. Anal. At. Spectrom., 1996, 11, no 208Pb signal was detected in the analysis of the serum of 303. healthy subjects. The total amount of lead in uraemic sera 5 E. Barany, I.A. Bergdahl, A. Schutz, S. Skerfving and analysed was evaluated by external calibration using internal A. Oskarsson, J. Anal. At. Spectrom., 1997, 13, 283. standards, as described elsewhere22 in previous work carried 6 D. R. Wiederin, 1996 Winter Conference on Plasma out by this group, and mean values about 1.9 ng g.1 were Spectrochemistry, Fort Lauderdale, FL, USA, January 8.13, obtained. 1996. 7 F. Ko and M. Yang, J. Anal. At. Spectrom., 1996, 11, 413. Serum samples of healthy people were also directly analysed 8 V.Poluzzi, B. Cavalchi, A. Mazzoli, G. Alberini, A. Lutman, using the FPLC-DF-ICP-MS method at R=300. The results P. Coan, I. Ciani, P. Trentini, M. Ascanelli and V. Davoli, J. Anal. obtained for Sr and Se were very similar to those described At. Spectrom., 1996, 11, 731. above for dialysis patients. However, the concentration of Sn 9 K.-L. Lee, S.-H. Liu and S. Jiang, Analyst, 1998, 123, 1557. and Pb in normal serum, as in the case of Cr, is too low and, 10 A.Sanz-Medel, Spectrochim. Acta, Part B, 1998, 53, 197. consequently, the proposed speciation methodology was not 11 ed. R. �©obin¢¥ ski, Analusis, 1998, 6, 21. 12 X. C. Le, W. R. Cullen and K. J. Reimer, Talanta, 1994, 41, 495. sensitive enough for direct speciation measurements. 13 S. C. K. Shum, H. Pang and R. S. Houk, Anal. Chem., 1992, Similarly we investigated the presence of Bi, Cd, Hg and As 64, 1284. in the same sera previously described.Unfortunately, the 14 R. Cornelis, F. Borguet and J. DeKimpe, Anal. Chim. Acta, 1992, sensitivity of the proposed FPLC-DF-ICP-MS method, even 283, 183. at R=300, was insuYcient to detect those elements in either 15 A. Soldado Cabezuelo, E. Blanco Gonza¢¥lez and A. Sanz-Medel, normal or uraemic serum (thus, possible direct speciation of Analyst, 122, 573. 16 N. M. Reed, R. O. Cairns, R. C. Hutton and Y. Takaku, J. Anal. such elements cannot be carried out unless enriched samples, At.Spectrom., 1994, 9, 881. or intoxicated specimens, are selected for such studies). 17 A. Soldado Cabezuelo, M. Montes Bayo¢¥ n, E. Blanco Gonza¢¥lez, J. I. Garc©¥¢¥a Alonso and A. Sanz-Medel, Analyst, 1998, 123, 865. 18 J. R. Strahler, B. B. Rosenblum, S. Hanash and R. Butkunas, Conclusions J. Chromatogr., 1983, 266, 281. 19 T. Tomono, H. Ikeda and E. Tokunaga, J. Chromatogr., 1983, Although further work is required to gain a more detailed 266, 39. knowledge of the behaviour of the diVerent isotopes in serum 20 B. R. Martin, J. Savory, S. Brown, R. L. Bertholf and M. R. Wills, and of trace elements actual binding to the serum proteins, Clin. Chem., 1987, 33/3, 405. 21 P. Gardiner, J. M. Ottaway, G. S. Fell and R. R. Burns, Anal. the multielemental capabilities of FPLC coupled to Chim. Acta, 1981, 124, 281. DF-ICP-MS have been illustrated for human serum samples. 22 C. Sariego Mun. iz, J. M. Marchante Gayo¢¥ n, J. I. Garc©¥¢¥a Alonso First of all, the capability of measuring at m/Dm=3000 and A. Sanz-Medel, J. Anal. At. Spectrom., 1999, 14, 193. allows the identification of potential spectroscopic inter- 23 R. Cornelis and De Kimpe, J. Anal. At Spectrom., 1994, 9, 945. ferences, which would jeopardise measurements carried out at 24 P. Bra¡§tter, B. Ribas and P. Schramel, Trace Elem. Anal. Chem. m/Dm=300. Moreover, the use of this setting m/Dm=3000 Med. Biol., 1994, 6, 1. 25 I. Harrison, D. Littlejohn and G. S. Fell, Analyst, 1996, 2, 189. allows an accurate detection of such elements aVected for 26 J. M. Marchante Gayo¢¥ n, J. E. Sa¢¥nchez Ur©¥¢¥a and A. Sanz-Medel, polyatomic interferences coming from the HPLC mobile J. Trace Elem. Med. Biol., 1996, 10, 229. phases and/or plasma and atmospheric gases. 27 B. Gercken and R. M. Barnes, Anal. Chem., 1991, 63, 283. Of course, there are important bioelements for which the 28 L. M. W. Owen, H. M. Crews, R. C. Hutton and A. Walsh, use of low resolution mode (m/Dm=300) is possible and that J. Anal. At. Spectrom., 1992, 117, 649. possibility can be advantageous to detect extremely low concentration in serum and so direct speciation becomes possible Paper 8/09026B J. Anal. At. Spectrom., 1999, 14, 947.951 951

ISSN:0267-9477    

DOI:10.1039/a809026b

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Quantitative determination of platinum complexes in human plasma generated from the oral antitumour drug JM216 using directly coupled high-performance liquid chromatography-inductively coupled plasma mass spectrometry without desolvation Peter Galettis, Jocelyn L. Carr, James W. Paxton and Mark J. McKeage* Department of Pharmacology and Clinical Pharmacology, Faculty of Medicine and Health Sciences, University of Auckland, Private Bag 92019, Auckland, New Zealand. E-mail: m.mckeage@auckland.ac.nz Received 6th January 1999, Accepted 19th April 1999 A new method was developed and validated for measuring platinum species generated from the clinical antitumour agent JM216 in methanol extracts of human plasma using high-performance liquid chromatography-inductively coupled plasma mass spectrometry (HPLC-ICP-MS).Good separation of JM216 and three of its biotransformation products (JM118, JM518 and JM383) was achieved with a run time of 20 min using a C8 column (4.6×150 mm) and a gradient methanol–water mobile phase (pH 2.5) at a flow rate of 1 ml min-1.The presence of methanol in the mobile phase and in the sample matrix suppressed the platinum counts and the gradient step was associated with some base-line drift. However, the quantitation of JM216 and its biotransformation products (JM118, JM518 and JM383) was achieved with good intra-assay precision (range 1–12% RSD), inter-assay precision (range 2.3–11% RSD), accuracy (range 89–103%) and limits of quantitation (range 1–2 ng ml-1) without having to use a desolvation device.This new HPLC-ICP-MS technique has the advantages of greater sensitivity and eYciency compared to existing methods that use HPLC, fraction collection and the oV-line detection of platinum by AAS. biotransformation products by using ICP-MS detection Introduction systems directly coupled to a HPLC. Platinum compounds have been used in cancer chemotherapy Cairns et al.6 directly coupled an HPLC and ICP-MS using since the 1970s.Up to now, the concentrations of platinum a desolvation device in order to allow the use of HPLC containing species in blood and urine have been measured by solvents that would otherwise destabilise the argon plasma high-performance liquid chromatography with on-line ultra- and to permit the use of a HPLC solvent gradient that could violet or electrochemical detection or the oV-line analysis of be associated with baseline drift.Using this technique they HPLC fractions by graphite furnace atomic absorption spec- achieved good separation and limits of detection for JM216. trometry.1 Studies employing these techniques have demon- Before the method can be used for the quantitative analysis strated that the concentrations of low molecular weight of clinical samples, however, it must be validated and proven platinum species determine the likelihood of tumour response to be precise, accurate and sensitive in the determination of and toxicity during platinum-based cancer therapy.1 JM216 in the matrix of interest.Moreover, information on JM216 [Pt(NH3) (NHC6H11)(OCOCH3)2Cl2] is a new the plasma concentrations of the biotransformation products platinum complex currently in phase III clinical trials and is generated from JM216 is also of interest. The platinum species showing promise in the treatment of prostate cancer.2 JM216 were expected to form adducts with plasma proteins with low biological activity, and free platinum species were extracted diVers from the existing clinical platinum agents in that it is from plasma using methanol.administered by mouth rather than by intravenous injection, In an extension of the studies of Cairns et al.,6 we attempted with the potential advantages of increased patient convenience, to develop and validate a quantitative assay for JM216 and better quality of life and lower cost of delivering the treatment. three of its biotransformation products (JM118, JM518 and To date, studies of the clinical pharmacokinetics on JM216 JM383), using directly-coupled HPLC-ICP-MS with diVerent have been diYcult because of the low concentrations of chromatography conditions that made it unnecessary to platinum present in plasma ultrafiltrate (<100 ng ml-1)3 and desolvate the mobile phase before it entered the ICP-MS.the extensive biotransformation of this drug in patients.4 Although the presence of methanol in the mobile phase and In order to obtain information on the pharmacokinetic in the sample matrix suppressed the platinum counts, the assay behaviour of JM216 in cancer patients, a new analytical achieved adequate precision, accuracy, sensitivity and sample technique is required that is capable of separating the various throughput for quantitative pharmacokinetic studies.7 platinum species generated from JM216 and detecting these individual species at concentrations down to 1 ng ml-1.A HPLC-AAS technique has been developed for this purpose4 Description of experimental procedures but it has a sensitivity limit of 10–50 ng ml-1 and poor sample Instrumentation throughput due to the oV-line analysis of HPLC fractions.5 We and others6 have attempted to develop more sensitive and Samples were introduced into a HP4500 inductively coupled plasma mass spectrometer (Hewlett-Packard, Yokowaga, eYcient methods for analysing clinical samples for JM216 J.Anal. At. Spectrom., 1999, 14, 953–956 953Table 1 HPLC-ICP-MS conditions for quantifying platinum species Japan) using a Babington-type (V-groove) nebulizer and a in human plasma Scott double-pass spray chamber. The HPLC consisted of a HP1100 binary pump, a rheodyne injector and a 20 ml sample Chromatography— loop (Hewlett-Packard, Wilmington, DE, USA) and a Prodigy Stationary phase Phenomenex Prodigy C8 C8 column (4.6×150 mm) (Phenomenex, Auckland, New (4.6×150 mm) Mobile phase A 25% methanol–0.01% Zealand).The HPLC was connected to the ICP-MS nebulizer orthophosphoric acid pH 2.5 by a 10 cm piece of 0.25 mm diameter PEEK tubing. B 100% Methanol Chromatograms were processed using HP chromatographic Gradient 0–10 min 100% A analysis software. 10–20 min 80% A 20% B Flow rate 1.0 ml min-1 Injection volume 20 ml Reagents ICP-MS— HPLC grade methanol (Labscan, Dublin, Ireland), ortho- Forward power 1350 W phosphoric acid (Riedel-de Ha�en, Seelze, Germany), 0.9% Reflected power <5 W sodium chloride (v/v) (Baxter Healthcare, Old Toongabbie, Gas flow rates: Australia), Milli-Q water (Millipore, Bedford, MA, USA) and Plasma 15 l min-1 Auxiliary 1 l min-1 fresh frozen human plasma (New Zealand Blood Service, Nebuliser 1 l min-1 Auckland, New Zealand) were used for preparing standards, Sampling depth 8 mm samples and the mobile phase.JM216 [Pt(NH3)- Sample uptake rate 1 ml min-1 from HPLC (NHC6H11)(OCOCH3)2Cl2] and its biotransformation prod- Sampler (nickel ) orifice 1 mm ucts JM383 [Pt(NH3) (NHC6H11)(OCOCH3)2(OH)2], JM118 Skimmer (nickel ) orifice 0.4 mm [Pt(NH3) (NHC6H11)Cl2] and JM518 [Pt(NH3)(NHC6H11)- Spray chamber temperature 1 °C Mass range 194–195 (OCOCH3)2(OH)Cl ] were generously supplied by the Johnson Acquisition time 1 s Matthey Technology Centre (Sonning, Oxfordshire, UK).Sample preparation Results and discussion Methanolic extracts of plasma were prepared by adding 100 ml The chromatographic conditions shown in Table 1 achieved of ice-cold methanol to 100 ml of plasma and mixing the good separation of JM216 and its biotransformation products sample before leaving it to sit at -20 °C.After 18 h, the (JM118, JM518 and JM383) with a run time of about 20 min sample was centrifuged at 20 000g for 20 min and the super- (Fig. 1). A change in the mobile phase composition at 10 min natant was removed for analysis. JM216 and its biotransformto an increased methanol content was associated with suppres- ation products JM118, JM518 and JM383 were stable in sion of platinum counts by approximately 70% and smaller methanol extracts of plasma at -20 °C for at least 28 d.peaks for compounds JM216 and JM518 compared with compounds JM383 and JM118. Platinum-194 and -195 were Assay validation used in the determination of the platinum species and there was excellent agreement between the isotopes, consistent Stock solutions of JM216, JM118, JM518 and JM383 were with a lack of mass interference.made up in methanol at 1, 10 and 100 mg ml-1. Standard Calibration standards were made up in human plasma for solutions of JM216, JM118, JM518 and JM383 in plasma were made by adding the stock solutions to plasma at 8 diVerent concentrations ranging from 1 to 120 ng ml-1. Quality control samples were made up from separate stock solutions at 7.5 and 75 ng ml-1 in plasma. The standards and quality control samples were prepared for analysis by methanol extraction.Precision was defined as the relative standard deviation (RSD) on the same day (intra-assay precision) and diVerent days (inter-assay precision) of repeated measurements of JM216, JM118, JM518 and JM383 in plasma at 7.5 and 75 ng ml-1.7 Accuracy was calculated by comparing the measured and expected concentrations of JM216, JM118, JM518 and JM383 at 7.5 and 75 ng ml-1 in plasma.7 The limit of quantitation was defined as the lowest concentration that could be measured with precision and accuracy within the limits (20% RSD) that are acceptable for analytical methods validation.7 Selection of chromatography conditions The starting point was an isocratic methanol–water (10% v/v) mobile phase (pH 2.6) because these conditions have previously been used to separate platinum species generated from cisplatin.8 Methanol was chosen as an HPLC solvent because of its lower vapour pressure and lower carbon loading compared with acetonitrile since these factors could influence the stability of the argon plasma.The HPLC conditions produced Fig. 1 Chromatogram of a methanol extract of human plasma contain- a run time of over 2 h. Changing to a C8 column, increased ing 75 ng ml-1 of JM383 [Peak A, Pt(NH3) (NHC6H11)(OCOCH3)2- methanol content and a step gradient at 10 min reduced the (OH)2], JM118 [Peak B, Pt(NH3) (NHC6H11)Cl2], JM518 [Peak C, run time and achieved separation of the platinum compounds Pt(NH3) (NHC6H11)(OCOCH3)2(OH)Cl ] and JM216 [Peak D, Pt(NH3) (NHC6H11)(OCOCH3)2Cl2].at a flow rate of 1 ml min-1. 954 J. Anal. At. Spectrom., 1999, 14, 953–956each of the platinum compounds at 8 concentrations ranging from 1 to 120 ng ml-1. The standards were processed before analysis by carrying out a methanolic extraction. The calibration curves were linear over the concentration range with correlation coeYcients of greater than 0.99. Standards made up in human plasma and then processed by methanolic extraction had platinum counts that were 10–20% lower than the counts of standard solutions, which were made up in 0.9% sodium chloride (v/v) and analysed without further processing, consistent with a significant matrix eVect.The accuracy and precision of the assay were determined by making up quality control solutions of the four platinum compounds in human plasma at 7.5 and 75 ng ml-1 and analysing methanolic extracts of these solutions. The methanolic extracts of the quality control solutions were stable in storage (-20 °C for 9 d).The accuracy, intra-assay precision and inter-assay precision (Table 2) were within the limits that have been stated to be acceptable for quantitative pharmacokinetic studies.7 The limit of quantitation, defined as the lowest concentration that could be measured with acceptable accuracy and precision, was 2 ng ml-1 for compounds JM216 and JM518, and 1 ng ml-1 for compounds JM118 and JM383.The platinum compounds were unstable in human plasma at room temperature with the loss of 11% and 22% of compounds JM118 and JM216, respectively, within 2 h. The compounds were more stable in methanol extracts of blood plasma and when stored at 4 or -20 °C. When methanol extracts of plasma were stored at -20 °C all four compounds (JM216, JM118, JM518 and JM383) were stable for at least 28 d. Measurement errors related to the poor stability of the compounds could be avoided by processing samples immediately after blood has been collected by the preparation of methanol extracts of the plasma and then storing the methanol extracts at -20 °C until analysis.The other analytical method currently available for quantitating platinum species generated from JM216 in human plasma involves high-performance liquid chromatography, fraction collection and oV-line detection of platinum using atomic absorption spectrometry.4 The HPLC-ICP-MS method described here has improved eYciency compared with the HPLC-AAS technique because the detection system is directly coupled to an HPLC and there is no requirement for the collection of fractions or for the oV-line analysis of fractionated samples.The HPLC-ICP-MS method also has lower limits of detection for platinum compounds generated from JM216 compared with the HPLC-AAS technique.4 Despite the suppression of platinum counts by the presence of methanol in the mobile phase and in the sample matrix, the HPLC-ICP-MS Fig. 2 Chromatograms of methanol extracts of plasma taken from a method has the advantages of greater sensitivity and eYciency patient 40 min (a) and 4 h (b) after taking a 270 mg oral dose of JM216. over the HPLC-AAS technique. To demonstrate the feasibility of detecting platinum species in plasma from cancer patients given JM216 by mouth, samples At 40 min, JM216 appears in plasma at 307 ng ml-1 in association with a metabolite (JM118) at 844 ng ml-1. By four taken during a clinical trial5 were prepared for analysis.Fig. 2 shows chromatograms of methanol extracts of plasma taken hours the concentrations of JM216 and JM118 have fallen to 9.3 ng ml-1 and and 38 ng ml-1, respectively. At this time a from a patient 40 min and 4 h after an oral dose of JM216. Table 2 Performance of the assay Platinum compounds JM118 JM383 JM518 JM216 Plasma concentration 7.5 ng ml-1 Intra-assay precision (RSD) (n=6) 6.2 6.9 6.7 11.9 Inter-assay precision (RSD) (n=3) 4.8 11.0 2.3 4.4 Accuracy (%) 95 99 102 94 75 ng ml-1 Intra-assay precision (RSD) (n=6) 5 1 5 2 Inter-assay precision (RSD) (n=3) 9 3 7 10 Accuracy (%) 103 99 89 102 Limit of quantitation/ng ml-1 1 1 2 2 J.Anal. At. Spectrom., 1999, 14, 953–956 955D. Vaughn, C. Brassard, D. Lebwohl and R. Bukowski, Proc. Am. new platinum containing species has appeared, eluting with a Soc. Clin. Oncol., 1998 17, 314. retention time of 1.2 min.The chemical structures of this and 3 M. J. McKeage, F. Raynaud, J. Ward, C. Berry, D. O’Dell, several other uncharacterised platinum containing species L. R. Kelland, B. A. Murrer, P. Santabarbara, K. R. Harrap and found in the plasma of patients given JM216 are currently I. R. Judson, J. Clin. Oncol., 1997 15, 2691. under investigation. 4 F. I. Raynaud, P. Mistry, A. Donaghue, G. Poon, L. R. Kelland, C. F. J. Barnard, B. Murrer and K. R. Harrap, Cancer Chemother. Pharmacol., 1996, 38, 155. Acknowledgments 5 M. J. McKeage, P. Mistry, J. Ward, F. E. Boxall, S. Loh, C. O’Neill, P. Ellis, L. R. Kelland, S. E. Morgan, B. Murrer, The authors would like to thank Dr. F.I. Raynaud for P. Santabarbara, K. R. Harrap and I. R. Judson, Cancer providing the blood samples and the Johnson Matthey Chemother. Pharmacol., 1995 36, 451. Technology Centre for providing the platinum complexes. The 6 W. R. L. Cairns, L. Ebdon and S. J. Hill, Fresenius’ J. Anal. Chem., 1996, 355, 202. project was funded by The Wellcome Trust, Lottery Science 7 V. P. Shah, K. K. Midha, D. Dinge, I. J. McGilveray, J. P. Skelly, and The University of Auckland. A. Yacobi, T. LayloV, C. T. Viswanathan, C. G. Cook, R. D. McDowall, K. A. Pittman and S. Spector, Eur. J. Drug Metab. Pharmacokinet., 1991, 16, 249. References 8 Z. Zhao, K. Tepperman, J. G. Dorsey and R. C. Elder, J. Chromatogr., 1993, 615, 83. 1 A. H. Calvert, I. Judson and W. J. F. Van Der Vijgh, Cancer Surveys, 1993 17, 189. 2 D. Peereboom, L. Wood, C. Connell, J. Spisak, D. Smith, Paper 9/00199I 956 J. Anal. At. Spectrom., 1999, 14, 953–956

ISSN:0267-9477    

DOI:10.1039/a900199i

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Flow injection on-line sorption preconcentration of trace amounts of copper and manganese in a knotted reactor precoated with 1-phenyl- 3-methyl-4-benzoylpyrazol-5-one coupled with electrothermal atomic absorption spectrometry Karima Benkhedda, Elissaveta Ivanova† and Freddy Adams* Department of Chemistry, University of Antwerpen (UIA), B-2610 Wilrijk, Belgium. E-mail: adams@uia.ua.ac.be Received 22nd January 1999, Accepted 26th March 1999 A new scheme was developed for the flow injection on-line sorption preconcentration of trace elements in a knotted reactor (KR) precoated with the chelating reagent 1-phenyl-3-methyl-4-benzoylypyrazol-5-one. It oVers several advantages in comparison with the conventional preconcentration scheme involving on-line merging of the sample and chelating reagent solution: higher sensitivity owing to the more favourable conditions of analyte preconcentration on the ‘immobilized’ reagent, better optimization of the separated processes of reagent sorption on the KR and analyte preconcentration, no analyte losses due to adsorption of chelate complexes on the tubing outside the KR and no need of a prefill step between samples of diVerent analyte concentration.The performance characteristics of the method for copper and manganese, respectively, are as follows: detection limits (3s), 5.7 and 5.0 ng l-1; minimum detectable masses, 0.014 and 0.013 ng; precision (RSD), 2.4% for 0.1 mg l-1 copper and 1.9% for 0.1 mg l-1 manganese; enhancement factors, 33.6 and 20.8 for a 30 s preconcentration time; and concentration eYciencies, 12.1 and 7.5.The sampling frequency was 21.6 h-1. The accuracy of the method was demonstrated by the analysis of certified reference materials. The chelating reagent 1-phenyl-3-methyl-4-benzoylpyrazol- Introduction 5-one (PMBP) was used, which forms complexes with more On-line separation and preconcentration using flow injection than 40 metal ions and has found numerous applications in (FI ) techniques have been shown to be eYcient in enhancing trace element separation and preconcentration by solvent the sensitivity and selectivity of electrothermal atomic absorp- extraction.17–20 There are no data about its use for FI on-line tion spectrometry (ETAAS).1–3 Various kinds of separation sorption preconcentration.As model analytes, the environmentechniques have been adapted to FI on-line separation and tally and toxicologically relevant elements copper and mangapreconcentration ETAAS systems, including solvent extrac- nese were used.The accuracy of the method was checked by tion,4 ion exchange,5 sorbent extraction,6–8 coprecipitation9 the analysis of biological certified reference materials. and sorption on a knotted reactor (KR).10–16 KR sorption systems proved to be a promising alternative to FI on-line Experimental preconcentration systems using sorbent columns. The lower back-pressure produced by the open-tube KR permits the use Apparatus of higher sample loading rates, thus achieving higher precon- The ETAAS determination was performed on a Perkin-Elmer centration eYciencies; KRs also oVer almost unlimited lifetime (Norwalk, CT, USA) Model AAnalyst 300 atomic absorption and stability in sorption properties.The key to the successful spectrometer equipped with a deuterium background corrector coupling of the KR sorption preconcentration system with and a Model HGA-800 graphite furnace and computer- ETAAS has been the FI manifold proposed by Sperling et al.,10 controlled using AAWinlab software, version 2.44.Hollow- used in subsequent work11–15 with minor changes. The manicathode lamps (Z-tek, Amsterdam, The Netherlands) were fold has recently been modified16 to permit thorough cleansing used as light sources at 324.8 nm for copper and 279.5 nm for of the KR and pertinent conduits of the FI system by the manganese with a spectral bandpass of 0.70 and 0.20 nm, eluent to avoid carry-over between individual samples instead respectively.Pyrolytic graphite-coated standard tubes (Z-tek) of the prefill step formerly used.10–15 The general scheme for were employed. The graphite furnace temperature–time pro- analyte preconcentration employed so far involved on-line gramme for the determination of copper and manganese in merging of the sample solution with the chelating reagent the methanolic eluate was as described previously.15 Peak solution and subsequent sorption of the formed analyte chelate height (absorbance), peak area (integrated absorbance) and on the inner walls of the KR.statistical data were printed out using a Hewlett-Packard In this work, a new preconcentration scheme was developed (Avondale, PA, USA) Laser Jet 5L printer. for enhancing the sensitivity and eYciency of the FI on-line The on-line separation and preconcentration were performed KR preconcentration ETAAS system.For this purpose, prewith a Perkin-Elmer Model FIAS-200 accessory equipped with coating of the KR with the chelating reagent was employed. a Perkin-Elmer 2×4-channel, 16-port double-layer rotary injector valve and controlled by a separate computer using the †On leave from the Institute of General and Inorganic Chemistry, Bulgarian Academy of Sciences, 1113 Sofia, Bulgaria. Perkin-Elmer FIAS VI.5 software. The scheme of the manifold J. Anal. At. Spectrom., 1999, 14, 957–961 957reagent solution [Fig. 1(a)]; (5) with the valve still in the inject position and pump 2 active, an air flow expelled the residual solution from the EL, KR and DT before elution [Fig. 1(a)]; (6) with pump 2 still active and the valve in the fill position, the EL was filled with the eluent while the DT mounted on the autosampler arm and connected with the FI manifold was manually inserted into the dosing hole of the graphite tube of the ETA unit [Fig. 1(b)]; (7) with the valve in the inject position and pump 2 still running, an air flow drove the eluent from the EL through the KR and DT to elute the analyte and deliver the eluate into the graphite tube [Fig. 1(a)]; (8) with the two pumps stopped the DT was withdrawn from the ETA. The preconcentration procedure was performed in parallel with the ETAAS determination of the previous preconcentrated sample. For calibration aqueous standard solutions were subjected to the preconcentration procedure. The linear least-squares regression method was used for establishing the calibration function.pH adjustment was performed with a Schott (Hofheim, Germany) CG820 pH meter using a combined glass–calomel electrode. Reagents All reagents were of the highest available purity. Doubly de-ionized water (18.2 MV cm) obtained from a Milli-Q water system (Millipore, Bedford, MA, USA) was used throughout. Suprapur HNO3 (65% m/m), HClO4 (70% m/m) and aqueous Fig. 1 FI manifold for on-line sorption preconcentration–ETAAS NH3 (25% m/m) and GR-grade potassium hydrogenphthalate using a KR precoated with the chelating reagent.P1, P2, peristaltic (all from Merck, Darmstadt, Germany) were used for sample pumps; S, sample; R, chelating reagent; E, eluent; KR, 100 cm knotted digestion and for the adjustment of sample acidity. The reactor; W, waste; EL, eluent loop (30 ml for Cu, 40 ml for Mn); V, chelating reagent PMBP was purchased from Fluka (Buchs, valve; DT, 35 cm delivery tube; ETA, electrothermal atomizer.Valve Switzerland). A 0.05% m/v solution of PMBP was prepared positions: (a) inject; (b) fill. daily by dissolving a weighed portion of the reagent in the minimum amount of aqueous ammonia, diluting with water is presented in Fig. 1. Knotted reactors of 100, 150 and 200 cm and adjusting the pH of the solution to 4.5–5.0 with dilute length were made of PTFE tubing (0.5 mm id) by tying nitric acid. Working standard solutions of the analytes were interlaced knots with ca. 5 mm diameter loops. Eluent loops prepared by stepwise dilution of 1000 mg l-1 stock standard of 20, 30, 40, 50 and 60 ml volume were made of PTFE tubing solutions (Merck) in 0.1% m/v potassium hydrogenphthalate (0.5 mm id). The connections and conduits were made of and adjusted to pH 4.0±0.2 for copper and pH 8.5±0.2 for PTFE tubing (0.35 mm id). Ismaprene pump tubes (Ismatec, manganese with aqueous ammonia. Methanol (Merck) was Wertheim, Germany) were employed for propelling the sample, used as the eluent.The glassware used was cleaned by boiling reagent, eluent and air. A 35 cm delivery tube was used to for several hours in 20% v/v HNO3 and rinsing with Milli-Q connect the FI system with the ETA unit. The operation water before use. sequence is presented in Table 1. The preconcentration procedure consisted of the following steps: (1) with pump 1 active Sample pre-treatment and the valve in the inject position, the chelating reagent solution was directed to the KR while the sample solution A 0.1 g amount of the biological sample was weighed in a 50 ml Erlenmeyer flask and gently heated on a hot-plate with filled the tubing before the valve and went to waste [Fig. 1(a)]; (2) with the valve still in the inject position and pump 2 4 ml of (65% m/m) HNO3 for 15 min. To prevent contamination during digestion, the flask was covered with a PTFE running, the reagent solution was removed from the KR by an air flow [Fig. 1(a)]; (3) with the valve in the fill position and cover with a small hole. After cooling, 2 ml of (70% m/m) HClO4 were added and the sample was further heated until a pump 1 active, the sample solution was directed to the KR where the analyte complex with the ‘immobilized’ reagent was moist residue was obtained. The latter was dissolved in 10 ml of water and diluted to 100 ml. A blank sample was prepared formed [Fig. 1(b)]; (4) with pump 1 still running and the valve in the inject position, the KR was washed with the in the same way.An aliquot of the sample solution with an Table 1 Operating sequence of the FI on-line preconcentration–ETAAS system for determination of copper and manganese sorbed on a KR precoated with PMBP Valve Pump Pumped Flow rate/ Step Fig. position active medium ml min-1 Time/s Function 1 1(a) Inject 1 Reagent solution 2.5 30 Coat KR with the reagent 2 1(a) Inject 2 Air 5.0 15 Remove reagent solution from KR 3 1(b) Fill 1 Sample solution 5.0 30 Preconcentrate 4 1(a) Inject 1 Reagent solution 2.0 (Cu) 15 Wash KR 2.5 (Mn) 5 1(a) Inject 2 Air 5.0 35 Remove sample solution from KR 6 1(b) Fill 2 Eluent 0.7 10 Fill EL and insert DT into ETA 7 1(a) Inject 2 Air 1.6 30 Elute and introduce eluate into ETA 8 — Inject — — — 2 Withdraw DT 958 J.Anal. At. Spectrom., 1999, 14, 957–961analyte concentration within the linear calibration range (0.02–0.4 mg l-1 for copper and 0.01–0.2 mg l-1 for manganese) was mixed with the hydrogenphthalate buVer solution and adjusted to the optimum pH with aqueous ammonia.Method development A univariate approach was employed for optimization starting with previously used conditions.15 The integrated absorbance (peak area) was taken as the output with simultaneous consideration of precision (aiming at <3% RSD) and eYciency. The parameters studied were pH of the reagent solution and of the sample solution, reagent concentration, flow rate and time of KR coating, preconcentration and washing, KR length, eluent volume and elution flow rate.Results and discussion Optimization of chemical and FI variables Sample preconcentration. The eVect of pH of complexation on the integrated absorbance of copper and manganese precon- Fig. 3 EVect of pH of ($) the sample solution and (&) the chelating centrated on the KR precoated with PMBP solution of pH 4.5 reagent solution on the integrated absorbance of 0.2 mg l-1 manganese.is depicted in Fig. 2 and 3, respectively. The pH of complexation was virtually identical with that of the buVered sample solution since no reagent solution was present in the KR during complexation. The optimum pH range of preconcentration was 3.8–4.2 for Cu and 8.3–9.8 for Mn. The eVect of pH of the chelating reagent solution used for KR coating on the integrated absorbance of copper and manganese preconcentrated at pH 4.0 or 8.5, respectively, is also shown in Fig. 2 and 3. The highest signals for both analytes were registered with a PMBP solution of pH 4.5–5.0. Hence it may be assumed that PMBP is better sorbed on the KR in a weakly acidic medium. PMBP solutions of pH <4.5 could not be employed for coating the KR since the reagent gradually precipitated on the tubing of the R line after pump 1 and caused valve blockage. The optimum PMBP concentration in the solution used for KR coating was found to be 0.04–0.06% m/v: at lower reagent concentrations lower integrated absorbance signals of the analytes were obtained and at higher reagent concentrations a precipitate appeared at pH >4.5.The eVect of time of coating and preconcentration and flow rate of coating and preconcentration on the analytical signal was studied. Copper and manganese displayed the same dependences, shown in Fig. 4 and 5 for copper. The optimum Fig. 4 EVect of the time of ($) coating and (&) preconcentration on the integrated absorbance of 0.2 mg l-1 copper.coating time was considered to be 30 s since a twice longer coating time yielded a <15% gain in sensitivity. Similarly, a flow rate of 2.5 ml min-1 was chosen for KR coating. The integrated absorbance of the analyte increased almost linearly with preconcentration time up to 60 s, after which it gradually levelled. As a compromise between sensitivity and sample frequency, a preconcentration time of 30 s was chosen for both analytes. Since the higher preconcentration flow rate favoured the analytical signal, a value of 5.0 ml min-1 was chosen in further work.The reagent solution was removed from the KR by sucking air at a flow rate of 5.0 ml min-1 for 15 s. Washing the KR. With the present preconcentration scheme, washing of the EL, KR and DT with the reagent solution used for coating the KR was only possible as both operations proceeded through the same manifold line. The integrated absorbance of copper and manganese as a function of the washing time is presented in Fig. 6. There is a considerable increase in sensitivity with only a 1 s washing time, a steady Fig. 2 EVect of pH of ($) the sample solution and (&) the chelating reagent solution on the integrated absorbance of 0.2 mg l-1 copper. signal being obtained with washing times between 5 and 20 s. J. Anal. At. Spectrom., 1999, 14, 957–961 959manganese was eluted with 55 ml of methanol. With the longer KRs (150 and 200 cm), eluent volumes of above 65 ml were needed and the higher sensitivity achieved was accompanied by a worse precision of the ETAAS determination due to the irreproducible spreading of the larger methanolic eluates in the graphite tube.The flow rate of elution was varied in the range 1.6–3.8 ml min-1. A value at the lower limit of this range was chosen in further work since the slower eluent flow favoured both the recovery of the sorbed analyte from the inner walls of the KR and the accommodation of the eluate in the graphite tube.Using this flow rate, a 30 s elution time was suYcient for the complete transfer of the eluent from the EL through the KR and the DT to the ETA. Interferences The tolerance towards metal ions encountered as matrix components in environmental and biological samples was studied. For better selectivity, the interference studies were performed at the lower limit of the optimum pH range of preconcentration. This was of particular significance for manganese, since Fe3+, Al3+ and Ca3+ form complexes with Fig. 5 EVect of the flow rate of ($) coating and (&) preconcentration PMBP in the same pH range.17 The results obtained are given on the integrated absorbance of 0.2 mg l-1 copper. in Table 2. Heavy metal ions up to an interferent-to-analyte ratio of about 100 did not interfere with the preconcentration of copper and manganese with the present system. The tolerated interferent concentrations are higher for copper than for manganese.This could be related to the lower pH range of copper preconcentration and to the fact that no masking of aluminium with fluoride or of iron with fluoride or Tiron was eYcient in the manganese–PMBP system. Nevertheless, the high sensitivity of the method permitted the successful determination of copper and manganese in biological samples. Performance characteristics The sensitivity of the FI on-line KR sorption preconcentration of copper and manganese with PMBP according to the conventional scheme was about 60% of that using the PMBPprecoated KR, which could be related to an improved adsorption eYciency in the latter case.Characteristic data for the performance of the FI on-line sorption preconcentration– ETAAS determination of copper and manganese using a PMBP-coated KR are compiled in Table 3. The performance characteristics of the present system are superior to those of the conventional preconcentration on the KR using APDC Fig. 6 EVect of the washing time on the integrated absorbance of ($) or HQ.15 0.1 mg l-1 manganese and (&) 0.2 mg l-1 copper. The accuracy of the method was checked by the analysis of biological certified reference materials (CRMs): Community This eVect could be related to the stabilization of the sorbed Bureau of Reference (BCR) CRM 274 single cell protein, analyte complex by the reagent solution flowing through the BCR CRM 422 cod muscle and International Atomic Energy KR during the wash step.11,12 It is worth mentioning that no (IAEA) CRM A-13 freeze-dried animal blood. The results are destruction or desorption of the sorbed manganese chelate given in Table 4.All values obtained are in good agreement takes place after prolonged washing with a reagent solution with the certified concentrations. of pH 4.5, which is far from the optimum pH range of manganese preconcentration (8.2–9.8). Table 2 EVect of interferents on the recovery of 0.1 mg l-1 copper or Changing the flow rate of the washing solution in the range manganese preconcentrated on a KR precoated with PMBP 1.4–2.5 ml min-1 had no eVect on the integrated absorbance Recovery (%) of manganese and slightly aVected that of copper. The maxi- Concentration/ Interferent-to- mum signal for the latter was obtained with a washing flow Interferent mg l-1 analyte ratio Copper Manganese rate of 1.4 ml min-1 and decreased by about 8% with a washing flow rate of 2.5 ml min-1.Since washing at lower NaCl 500 5×106 92.5 flow rates did not eYciently remove matrix components, a 10000 1×108 100.0 Ca2+ 0.5 5×103 100.4 flow rate of 2.0 ml min-1 was used for copper and 2.5 ml min-1 100 1×106 100.3 for manganese. The complete removal of the washing solution Fe3+ 0.1 1×103 98.3 92.0 from the EL, KR and DT before elution was eVected by an 5.0a 5×104 96.0 air flow (5.0 ml min-1) for 35 s. Al3+ 0.5 5×103 99.7 96.2 2.0a 2×104 97.7 Elution. Copper was eluted from the 100 cm KR with 45 ml aIn the presence of 0.2 mol l-1 F-.of methanol (30 ml EL+15 ml dead volume of the valve) and 960 J. Anal. At. Spectrom., 1999, 14, 957–961Table 3 Performance of the FI on-line preconcentration–ETAAS system for the determination of copper and manganese using a KR precoated with PMBP Analyte Parameter Copper Manganese Working concentration range/mg l-1 0.02–0.4 0.01–0.2 Calibration function (six standards, n=3, Canalyte in mg l-1) 1.279CCu+0.0016 1.500CMn+0.0015 Correlation coeYcient 0.9999 0.9990 Sample volume per assay/ml 5.9 6.3 Sampling frequency ( f )/h-1 21.6 21.6 Precision (RSD, n=11) (%) 2.4 (0.1 mg l-1) 1.9 (0.1 mg l-1) Detection limit (3s)/ng l-1 5.7 5.0 Enhancement factor (EF)a 33.6 20.8 Concentration eYciency (CE) (CE=EF×f/60) 12.1 7.5 Adsorption eYciencyb (%) 60.1 46.5 aCompared with the direct injection of aqueous solution of a volume corresponding to that of the eluate.bCompared with the total analyte mass passed through the KR.Table 4 Results (mean ± 3s based on three replicate measurements) of the determination of copper and manganese in BCR CRM 274 (single cell protein), BCR CRM 422 (cod muscle) and IAEA CRM A-13 (freeze-dried animal blood) Copper/mg g-1 Manganese/mg g-1 Sample Certified Found Certified Found BCR CRM 274 13.1±0.4 13.25±0.09 51.9±1.2 51.5±1.0 BCR CRM 422 1.05±0.07 1.09±0.05 0.543±0.028 0.555±0.036 IAEA CRM A-13 4.3±0.7 4.04±0.10 — — 2 Z. Fang, S.Xu and G. Tao, J. Anal. At. Spectrom., 1996, 11, 1. Conclusions 3 Z.-L. Fang, Spectrochim. Acta, Part B, 1998, 53, 1371. 4 G. Tao and Z. Fang, Spectrochim. Acta, Part B, 1995, 50, 1747. The results obtained in this work indicate that PMBP is a 5 Z. Fang, J. Ruzicka and E. H. Hansen, Anal. Chim. Acta, 1984, suitable chelating reagent for trace element preconcentration 164, 23. in a KR. The new preconcentration scheme using a KR 6 L. C. Azeredo, R. E. Sturgeon and A. J. Curtius, Spectrochim. precoated with the chelating reagent oVers several advantages Acta, Part B, 1993, 48, 91.compared with the conventional scheme involving on-line 7 M. Sperling, X. Yin and B. Welz, J. Anal. At. Spectrom., 1991, merging of sample and reagent solutions: (i) the competition 6, 295. 8 R.-L. Ma, W. Van Mol and F. Adams, Anal. Chim. Acta, 1994, between the molecules of the reagent and the analyte complex 293, 251. for the inner walls of the reactor11 is avoided; (ii) the prelimi- 9 H.-W.Chen, S.-K. Xu and Z.-L. Fang, J. Anal. At. Spectrom., nary coating of the KR with the chelating reagent creates 1995, 10, 533. more favourable conditions for trace element preconcentration 10 M. Sperling, X.-P. Yan and B. Welz, Spectrochim. Acta, Part B, expressed in higher sensitivity; (iii) the separation of the 1996, 51, 1891. processes of reagent sorption on the KR and analyte precon- 11 X.-P. Yan, W. VanMol and F. Adams, Analyst, 1996, 121, 1061. 12 X.-P. Yan and F. Adams, J. Anal. At. Spectrom., 1997, 12, 459. centration permits their better optimization; (iv) there are no 13 E. Ivanova, X.-P. Yan, W. VanMol and F. Adams, Analyst, 1997, analyte losses due to adsorption of chelate complexes on the 122, 667. tubing before the valve12 since the sample solution contacts 14 E. Ivanova and F. Adams, Spectrochim. Acta, Part B, 1998, 53, the reagent only in the KR; and (v) no prefill step is required 1041. between samples of diVerent analyte concentration since the 15 E. Ivanova, K. Benkhedda and F. Adams, J. Anal. At. Spectrom., sample solution in the tubing before the valve changes during 1998, 13, 527. KR precoating. Owing to the high sensitivity achieved, the 16 S. Nielsen and E. H. Hansen, Anal. Chim. Acta, 1998, 366, 163. 17 Yu. A. Zolotov and N. M. Kuz’min, Ekstraktsiya Metallov method permits an accurate and precise determination of Acylpyrazolonami, Nauka, Moscow, 1977. copper and manganese in biological samples. 18 E. Ivanova, S. Mareva and N. Jordanov, Fresenius’ Z. Anal. Chem., 1977, 288, 62. This work was funded by the Fonds Wetenschappelijk 19 E. Ivanova, S. Mareva and N. Jordanov, Fresenius’ Z. Anal. Onderzoek Vlaanderen (FWO, Brussels). Chem., 1980, 303, 378. 20 E. Ivanova, I. Havezov, N. Vracheva and N. Jordanov, Fresenius’ Z. Anal. Chem., 1985, 320, 133. References 1 Z.-L. Fang, Flow Injection Atomic Absorption Spectrometry, Wiley, Chichester, 1995. Paper 9/00630C J. Anal. At. Spectrom., 1999, 14, 957–961 961

ISSN:0267-9477    

DOI:10.1039/a900630c

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Use of polytetrafluoroethylene slurry for silica matrix removal in ETAAS direct determination of trace cobalt and nickel in silicon dioxide powder Wang Fuyi, Jiang Zucheng* and Peng Tianyou Department of Chemistry, Wuhan University, Wuhan 430072, China Received 6th January 1999, Accepted 8th April 1999 A novel method for the direct determination of trace Co and Ni in SiO2 powder by slurry sampling ETAAS was developed. At high temperature of the graphite furnace, a PTFE slurry in HNO3 was used as a fluorinating reagent to convert the silica matrix into high-volatility fluoride, which was subsequently evaporated by selective vaporization prior to the atomization of analytes.In this case, the severe interference of the matrix on the vaporization and atomization of analytes was reduced significantly, and chemical attack of the excess of silica matrix on the graphite tube was also minimized. The proposed method was successfully applied to the determination of trace Co and Ni in SiO2 powder with aqueous calibration and minimum chemical pre-treatment.For the direct analysis of high purity SiO2 powder, the detect limits of Co and Ni were 18.8 and 35.0 ng g-1, respectively. The analysis of NIES CRM2 Pond Sediment confirmed the reliability of the approach. High purity siliceous materials are widely applied in optical fluorine content, moderate decomposition temperature, very low inorganic impurities, lack of corrosion and easy utilization.waveguide fibre, glass, semiconductor and ceramics manufac- In our more recent studies, a PTFE slurry was also used as a ture. Any trace impurities (such as Cu, Ni, Cr and Fe) in these chemical modifier for the determination of chromium and materials strongly influence the quality of the products.1,2 It cadmium by ETAAS.24–26 It was concluded that the addition is therefore desirable to be able to determine accurately and of PTFE could improve the sensitivity for the refractory quickly the contents of trace impurities in siliceous materials element chromium, increase the maximum ashing temperature for quality control.for the volatile element cadmium and eYciently reduce the Several techniques have been used to determine trace chemical and spectral interferences of co-existing components impurities in siliceous materials, such as inductively coupled in the determination of Cr and Cd. plasma atomic emission spectrometry (ICP-AES),3,4 induc- In this work, a PTFE slurry as a fluorinating reagent was tively coupled plasma mass spectrometry (ICP-MS)5,6 and employed to convert the silica matrix and analytes (Co and Ni) atomic absorption spectrometry (AAS).7,8 However, in most in SiO2 into fluorides of various volatilities at high temperature cases, chemical pre-treatment is necessary to remove the silica of the graphite furnace.The high-volatility silicon fluoride was matrix prior to the determination. Since chemical preevaporated prior to the vaporization and atomization of ana- treatment leads to a long analysis time and the risk of lytes.The action of PTFE not only minimized the deterioration contamination and loss of analyte, the direct analysis of solid of the graphite tube, but also reduced substantially the inter- siliceous materials has received increasing attention.9,10 ference of the silica matrix on the atomization of analytes. In In ETAAS, there are two approaches for the direct this case, the vaporization and atomization processes of analytes introduction of solid samples, i.e., direct solid sampling and were free from the matrix.Moreover, the eVect of particle size slurry sampling. Owing to its advantages, slurry sampling has on the analytical performance was also minimized. With cali- been widely applied for the practical analysis of solid bration using aqueous standards, the proposed method was samples.9,11 The main problems related to the direct analysis used to determine directly Co and Ni in SiO2 powder.Only of siliceous samples by slurry sampling ETAAS includes12–17 minimum chemical pre-treatment was necessary prior to the (1) the severe physical/chemical and/or spectral interferences measurements and the determined values were in good agree- resulting from the excess of silica matrix and (2) the rapid ment with those obtained by the standard additions method. deterioration of the graphite as a consequence of chemical NIES CRM2 Pond Sediment certified reference material with attack by silicon at high temperatures.The deleterious eVect a high silica content was used to evaluate this method. of silicon on the graphite tube could be minimized by adding graphite powder.18 In this case, the lifetime of the graphite tube was increased considerably. However, interference of the Experimental silica matrix on the vaporization and atomization of analytes Apparatus still exists and causes diYculties in the direct measurement of trace elements in siliceous samples.Therefore, during the A WFX-IF2 atomic absorption spectrometer with a deuterium preparation of slurry samples of siliceous materials, HF and/or background corrector and a WF-4A electrothermal graphite HNO3 are added to the slurries to remove partially the silica furnace system (Beijing Second Optics, Beijing, China) was matrices from samples.13–17 employed. The operating conditions for AAS and the graphite In recent years, polytetrafluoroethylene (PTFE) slurry as a furnace are given in Table 1.fluorination modifier has been successfully applied to electrothermal vaporization (ETV) ICP-AES for the determination Standard solutions and reagents of refractory elements (such as rare earth elements, Ti and Mo) in high purity materials and biological and environmental Stock standard solutions (1 mg mL-1) of cobalt and nickel samples.19–23 Compared with the general fluorinating reagent were National Reference Solutions (10% HCl, GSBG 62021–90 and GSBG 62022–90) from the National Center for HF, the PTFE slurry has the special advantages of high J.Anal. At. Spectrom., 1999, 14, 963–966 963Table 1 Operating condition for ETAAS Wavelength/nm Co 240.7; Ni 232.0 Spectral bandwidth/nm 0.4 HCL current/mA Co 2.5; Ni 3.5 Deuterium lamp current/mA (Co) 42; (Ni) 45 Graphite furnace— Temperature/ °C Ramp/s Hold/s Dry 120 20 30 Ashing 1200 10 20 Atomization 2700 0 10 Flow rate of argon 500 mL min-1; 100 mL min-1 during atomization Analysis and Testing of Steel Materials.Working standard solutions were prepared by diluting these stock standard Fig. 1 (I ) Ashing curves and (II ) atomization curves of Co and Ni solutions. A 60% m/v PTFE slurry (Shanghai Organic in aqueous solutions and PTFE slurries. Chemistry Institute, Shanghai, China) was commercially available. A 0.5% m/v aqueous solution of plant glue (kindly EVect of silica matrix on the analytical signals provided by Professor Zhang Xiangcha) was employed as a stabilization reagent during the preparation of slurry samples Fig. 2 shows the dependence of the Co and Ni absorption of SiO2 powder. Nitric acid (suprapure) and silicon dioxide signals in suspensions on the content of SiO2. It can be seen (suprapure) were supplied by Shanghai reagent factory that without PTFE the silica matrix produced an increasing (Shanghai, China) and silicon dioxide powder for analysis by suppression of the analytical signals when its concentration Wuhan Industry University (Wuhan, China).Doubly distilled was increased. However, this matrix eVect could be reduced water was used throughout. significantly by adding the PTFE slurry. In the presence of 6% m/v PTFE the analytical signals remained constant until Slurry sample preparation the content of SiO2 was higher than 15 mg mL-1. Typical signal profiles of Co and Ni in aqueous standard Portions (5–100 mg) of SiO2 powders (particle size 74–97 mm) solutions and in SiO2 slurries are shown in Fig. 3. The resultant were transferred into 5 ml calibrated flasks and 0.5 mL of 60% graphs indicated that (1) without PTFE, the silica matrix not PTFE slurry and 0.4 mL of nitric acid (1+1) were added in only delayed the appearance time of absorption peaks, but turn, then diluted to 5 mL with a 0.1% aqueous solution of also caused bi-peak profiles with lower absorption signals plant glue.The resulting mixtures were dispersed with an [Fig. 3(A) and (B)] and (2) the addition of PTFE greatly ultrasonic wave vibrator for 20 min. The flasks were shaken changed this situation, similar profiles and peak heights being vigorously prior to any analysis. obtained for Co and Ni either in aqueous standard solutions For a calibration with the standard additions method, the or in SiO2 slurries [Fig. 3(C) and (D)]. In other words, the slurries prepared as described above were spiked with approvaporization and atomization behaviours of Co and Ni in priate amounts of aqueous standard solutions.In the resulting both cases were coincident. This result provided the possibility standard series the content of SiO2 powder was 12 mg mL-1 of the determination of Co and Ni in SiO2 slurries with and spiked Co and Ni ranged from 0.05 to 0.3 and from 0.1 calibration using aqueous standards. to 0.6 mg mL-1, respectively. For calibration against aqueous standards, the aqueous standards, in which the concentrations Influence of ashing time on analytical signal and background of Co and Ni varied from 0.1 to 0.3 and 0.2 to 0.6 mg mL-1, respectively, were prepared as slurries containing 6% PTFE At an ashing temperature of 1200 °C, the influence of ashing and 2% HNO3 by the procedure mentioned above.time on the intensities of the analytical and background signals were investigated and the results are shown in Fig. 4. It can Recommended procedure be seen that the analytical signals slowly increase with increase in ashing time and reach a maximum up to 20 s and then A 20 mL volume of solution or slurry samples was deposited remain constant in the range 20–50 s, whereas the background on the wall of the pyrolytic graphite-coated graphite tube with intensities do not change when the ashing time is shorter than a micropipette.After being dried and ashed, the analyte was 50 s. Hence a 20 s ashing time was selected for the direct vaporized and atomized.The absorption signals of the analyte analysis of SiO2 powder. and background were collected simultaneously by the incorporated microcomputer and the dependence of the integrated absorbance on atomization time was then plotted graphically. The peak height was used for quantification. Results and discussion Influence of PTFE on the vaporization and atomization processes of Co and Ni The ashing curves and atomization curves of Co and Ni with and without 6% PTFE were investigated and are presented in Fig. 1. Under the same atomization conditions the introduction of PTFE in 2% HNO3 slightly reduced the absorption signals; PTFE had no obvious influence on the maximum ashing temperature and the minimum atomization temperature of Co and Ni. Therefore, similar ashing/atomization curves were observed in the absence and presence of PTFE. An ashing temperature of 1200 °C and an atomization temperature of Fig. 2 Influence of concentration of silicon dioxide in slurries on cobalt and nickel absorption. 2700 °C were selected for subsequent experiments. 964 J. Anal. At. Spectrom., 1999, 14, 963–966Fig. 3 Typical signal profiles of Co and Ni in (a) aqueous solutions and (b) SiO2 suspensions in the (A, B) absence and (C, D) presence of PTFE. The values in parentheses are the peak areas. Co, 3 ng; Ni, 5 ng; SiO2, 120 mg; PTFE, 6%. Mechanism of suppression of matrix interference by selective vaporization at an appropriate ashing temperature prior to the vaporization and atomization of the Co and Ni It has been reported12 that the silica matrix caused a very analytes.In this case, the analytes (Co and Ni) could be strong background at a wavelength of 240 nm when the atomized freely from the silica matrix, and their atomization atomization temperature was higher than 2350 °C. This back- behaviours are similar to those in aqueous solutions (Fig. 3). ground absorption may be caused by SiO molecules evaporat- On other hand, since most of the silica matrix is removed prior ing at high temperature. In the present study, the strong to atomization of the analytes, the deleterious eVect resulting background absorption mentioned above was also observed from chemical attack of the silica matrix at high temperature at 240.7 nm on evaporating an SiO2 suspension under the is also minimized considerably.conditions given in Table 1. However, the strong signal was almost eliminated on adding 6% PTFE to the slurries (Fig. 5). EVect of particle size on the analytical signals Previous work19,20 indicated that at the high temperatures oVered by a graphite furnace, a PTFE slurry could convert the The particle size of the solid materials used to prepare a slurry can influence the stabilization, deposition and atomization silica matrix and analytes into fluorides of various volatilities. Among the fluorides, silicon fluoride (bp -86 °C) is one of the eYciency of the sample, which in turn can influence both accuracy and precision.9 In this paper, the eVects of the particle most volatile compounds and could be evaporated prior to the vaporization and atomization of most of analytes by selective size of the SiO2 powder on the relative absorbance of Co and Ni in SiO2 slurries with PTFE were investigated, and the vaporization.The results presented in Fig. 5 again confirm that the silica matrix could be converted by the fluorinating reagent results are given in Table 2.The results obtained indicate that PTFE into high-volatility fluoride and subsequently removed Fig. 5 Signal profiles of silica background absorption in the (a) absence and (b) presence of PTFE. During the atomization step, the Fig. 4 Dependence of Co and Ni absorption and background signals flow of argon was stopped. SiO2, 120 mg. The operating conditions are given in Table 1. in SiO2 slurries on ashing time. J. Anal. At. Spectrom., 1999, 14, 963–966 965Table 2 EVect of particle size of SiO2 powder on analytical signals Particle size/mm >300 300–200 200–150 150–125 125–97 97–74 <74 Relative abs.a of Co 0.80±0.19b 0.85±0.12 0.89±0.069 0.91±0.059 0.96±0.041 0.98±0.028 1.00±0.025 Relative abs.a of Ni 0.63±0.14 0.60±0.10 0.86±0.072 0.82±0.054 0.90±0.020 0.95±0.030 0.99±0.024 aThe relative absorbances of Co and Ni in PTFE slurries are 1.00 without SiO2; the concentration of SiO2 in suspensions is 12 mg mL-1; Co and Ni are 0.15 and 0.25 mg mL-1, respectively.bMean±SD (n=3). Table 3 Results for the determination of Co and Ni in silicon dioxide the recoveries are higher than 90% with RSDs of 2.0–5.9% powder and NIES CRM2 when the average particle size of the sample is smaller than 125 mm. This tolerable particle size is as large as four times Cobalt/mg g-1a Nickel/mg g-1a that reported in the literature.13,14 Obviously, the decreased influence of particle size on the analytical signals is attributed Aqueous Standard Aqueous Standard calibration additions calibration additions to the chemical modification of PTFE.It is PTFE that converts the silica matrix into volatile fluoride, which makes the silica Silicon dioxide 2.92±0.18 3.13±0.15 8.39±0.32 8.10±0.52 matrix evaporate prior to the atomization of analytes and NIES CRM2 25.27±2.20 —b 40.87±1.84 —b allows the analytes to be atomized under the conditions of aMean±SD (n=6). bCertified values: Co 27±3 mg g-1; Ni less matrix.However, when the average particle size of the 40±3 mg g-1. sample is larger than 125 mm, the recoveries of the analytes gradually decrease owing to slower vaporization of the silica matrix as a consequence of incomplete fluorination. Meanwhile, the RSDs slowly increase to 14–19% because of References the poorer pipetting eYciency. Therefore, the particle size of 1 J.W. Mitchell, Pure Appl. Chem., 1982, 54, 819. SiO2 powder should be smaller than 125 mm for direct analysis, 2 J.Dolezal, J. Lenz and Z. Suleck, Anal. Chim. Acta, 1969, 47, 517. which is easily met in practical analysis. 3 M. T. Larrea, I. Grome-Pinilla and J. C. Farinas, J. Anal. At. Spectrom., 1997, 12, 1323. Sample analysis 4 S. Mann, D. Geilenberg, J. A. C. Brockaert and M. Jansen, J. Anal. At. Spectrom., 1997, 12, 975. With either aqueous standards or the standard additions 5 G. Tsoupras, Analusis, 1996, 24(9/10), M23. method for calibration, the contents of trace Co and Ni in 6 H.Naka and H. Kusayssu, Bunseki Kagaku, 1996, 45, 1139. SiO2 powder were determined and the results are given in 7 X. C. Zhou, F. Y. Wang and J. Y. Lihua, Huaxue Fence, 1997, 33(5), 207. Table 3. It is found that the results obtained by above methods 8 X. H. Wen, L. Z. Wu, Y. Zhang and Y. Chu, Fresenius’ J Anal. are identical with each other. The characteristic masses of Co Chem., 1997, 357, 1111. and Ni are 28.5 and 77.2 pg per 0.0044 absorbance, respect- 9 C.Bendicho and M. T. C. De Loos-Vollebregt, J. Anal. At. ively; the detection limits (3s) of Co and Ni in aqueous Spectrom., 1991, 6, 353. solutions are 6.08 and 14.1 pg, respectively; for direct determi- 10 J. S. Crighton, B. Fairman, J. Haines, M. W. Hinds, S. M. Nelms nation of Co and Ni in SiO2 powder by slurry sampling, the and D. M. Penny, J. Anal. At. Spectrom., 1997, 12, 509R. 11 S. C. Stephen, D. Littlejohn and J. M. Ottaway, Analyst, 1985, detection limits are 18.8 and 35.0 ng g-1, respectively. With 110, 1147.aqueous calibration, the NIES CRM2 Pond Sediment certified 12 J. Mierzwa and R. Dobrowolski, Fresenius’ J. Anal. Chem., 1994, reference material with a high silica content (45% m/m SiO2) 348, 422. was used to evaluate the proposed method, with satisfactory 13 C. Bendicho and M. T. C. De Loos-Vollebregt, Spectrochim. Acta, results (Table 3). Part B, 1990, 45, 679. 14 C. Bendicho and M. T. C. De Loos-Vollebregt, Spectrochim.Acta, Part B, 1990, 45, 695. Conclusions 15 I. Lo�pez Garcia, J. Arroyo Cortez and M. Hernandez Cordoba, J. Anal. At. Spectrom., 1993, 8, 103. At high temperature in a graphite furnace the fluorinating 16 I. Lo�pez Garcia, J. Arroyo Cortez and M. Hernandez Cordoba, reagent PTFE could convert the silica matrix into high- Anal. Chim. Acta, 1993, 283, 167. 17 I. Lo�pez Garcia, E. Navarro, P. Vinas and M. Hernandez volatility fluoride (SiF4), which allows the interfering matrix Cordoba, Fresenius’ J.Anal. Chem., 1997, 357, 642. to be eYciently separated by selective vaporization during the 18 T. Nakamura, H. Oka, H. Morikawa and J. Sato, Analyst, 1992, thermal pre-treatment. In this case, the severe interference of 117, 131. the matrix on the vaporization and atomization of the analytes 19 T. Y. Peng and Z. C. Jiang, Anal. Sci., 1997, 13, 595. (Co and Ni) was reduced significantly. 20 T. Y. Peng and Z. C. Jiang, Fresenius’ J. Anal. Chem., 1998, For direct analysis of siliceous materials, the removal of the 360, 43. 21 Z. C. Jiang, B. Hu, Y. Qin and Y. Zeng, Microchem. J., 1996, silica matrix prior to atomization could minimize the deleteri- 53, 326. ous eVect of silicon on the graphite, increasing the lifetime of 22 Y. Qin, Z. C. Jiang, Y. Zeng and B. Hu, J. Anal. At. Spectrom., the graphite tube. 1995, 10, 455. The developed method has been used successfully in the 23 Z. C. Jiang, B. Hu, M. Huang and Y. Zeng, Xitu, 1993, 15(6), 23. direct determination of trace Co and Ni impurities in SiO2 24 F. Y.Wang and Z. C. Jiang, J. Anal. At. Spectrom., 1998, 13, 539. powder and siliceous standard material. Only minimum chemi- 25 F. Y.Wang and Z. C. Jiang, Anal. Chim. Acta, in the press. 26 F. Y.Wang and Z. C. Jiang, Fenxi Kexue Xuebao, 1999, 15, 111. cal pre-treatment was necessary. The tolerable particle size of solid samples in this method is as large as four times the value in the literature. Paper 9/00198K 966 J. Anal. At. Spectrom., 1999, 14, 963&nda

ISSN:0267-9477    

DOI:10.1039/a900198k

出版商:RSC    

年代:1999

数据来源: RSC