摘要:

JOURNAL O F ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 679 Determination of Cadmium by Electrothermal Atomic Absorption Spectrometry Using Palladium and Tartaric Acid as a Mixed Chemical Modifier and a Tungsten-foil Platform with the Possibility of Standardless Analysis Ma Yizai Li Zhikun Wang Xiaohui Wang Jiazhen and Li Yongquan Institute of Analysis and Measurement Chinese Research Academy of Environmental Sciences Beijing 1000 12 China Palladium and tartaric acid as a mixed chemical modifier is a better modifier for the determination of Cd in complex environmental samples than Pd and tartaric acid (or ascorbic acid) used as separate modifiers. The mixed modifier can effectively eliminate matrix effects and the two peak signals observed in the atomization of complex environmental samples.It has also been found that a W-foil platform has a longer lifetime than a Ta-foil platform. A comparison of the calculated and experimental values obtained for the characteristic mass m, with a W-foil platform in a pyrolytic graphite coated graphite tube using a mixed modifier shows that an atomization temperature higher than T(set) = 1673 K and T(effective) = 1530 K can be used for the determination of Cd. Quantitative analytical data for Cd in sea-water and in solutions of reference materials (pork liver wheat powder peach leaf tea leaf tea tree leaf cabbage water sediments coal fly ash soils and rock) by standardless analysis are presented (LOD = 0.3pg; RSD < 5%). Keywords Nectrothermal atomic absorption spectrometry; chemical modifier; cadmium; environmental samples Ma et al.' have previously described the determination of Cd in environmental samples by electrothermal atomic absorption spectrometry (ETAAS) using a Ta-foil platform with the pos- sibility of standardless analysis.The highly resistant Ta-foil surface is preferable for the determination of Cd especially when large amounts of various mineral acids and perchloric acid are used for sample decomposition. Ma and Cheng developed a new atomizer a pyrolytic graphite coated graphite tube lined with tungsten (WPGT) for the determination of C O ~ Pd,3 Y,4 Ba In Pb Cu Ni Rb Au Cd Mn Li Cr and Sb.' The common oxides and carbides of W sublime at moderate temperatures making the W surface 'self-cleaning' during high-temperature treatment.In the present study a new W-foil platform is used for the determination of Cd in environmental samples. Shan and Ni6 have reported the use of Pd as a modifier for the determination of Hg Pb Sb Bi Se and Te. It has been ascertained that the reduction of Pd begins around a minimum temperature of 600"C.7 Shan et aZ.* found Pd and ascorbic acid to be a more effective modifier than Pd or ascorbic acid alone; ascorbic acid reduces Pd ions in solution to elemental Pd. Zhuang et a2.' used Pd and citric acid as a mixed chemical modifier for the determination of Zn and Cd using wall PGT atomization. The aim of this work was to utilize Pd and tartaric acid (TA) as a mixed chemical modifier and a W-foil platform for the determination of Cd in complex environmental samples with the possibility of standardless analysis.Experimental Apparatus A Hitachi Model 28000 d.c. magnet Zeeman-effect atomic absorption spectrometer was used for the determination of Cd at the resonance line of 228.8 nm with a spectral bandwidth of 0.4 nm. The Cd hollow cathode lamp was operated at a current of 2.5mA. Samples of 0.010ml were introduced into the furnace using Eppendorf microlitre pipettes fitted with dispos- able polypropylene tips. The temperature measurements were made using an electro-optical pyrometer and chart recorder. The argon sheath gas flow rate was 3 1 min-' in the 'gas stop' mode during the atomization step. The PGTs were made by the Beijing Research Institute of Materials and Technology Ministry of Astronautics. The dimensions of the PGTs are inner radius 2.95 outer radius 4.05 and length 28 mm.The W-foil platform (WFP) is 4 x 9 mm with a 1 mm rim a thickness of 0.22 mm and a mass of 230 mg. The WFP is inserted into the centre of a PGT. The furnace heating programme consisted of drying at 80-120°C €or 20 s ashing at 300°C for 20 s (all ashing times 20 s) atomization at 1400 "C for 20 s 1500 "C for 20 s (for Table l) 2000 "C for 10 s 2100 "C for 10 s and 2600 "C for 5 s and cleaning at 2400°C for 3 s. Reagents A stock solution of Cd (0.500 mg ml-l) was obtained from the Chinese National Environmental Monitoring Centre and stored in an ampoule. Tartaric acid (10 mg ml-l) was prepared by dissolving the materials in concentrated ammonium solution. A solution of Pd (as PdCl,); (5 mg ml-l) was prepared by dissolving a suitable amount of PdCl (spectroscopic-grade Beijing Chemical Company) in 0.3 mol 1-' HNO,. The 0.5 mg ml-' solution of Pd was obtained by diluting the above solution with concentrated ammonium solution.The mixed solution of 10 mg ml-1 TA+O.5 mg ml-I Pd was obtained by dissolving the TA in the solution of 0.5 mg ml-1 Pd. Sample Dissolution Procedures Geochemical and water sediment samples Samples of 100 mg of reference materials (RM) of geochemical samples Rock GSR1 soil samples ESS3 893 Tibetan soil and water sediments GSD1,2,4,5,6,9 10 11 and 81-101 (obtained from the Institute of Physical and Chemical Prospecting Institute of Analysis and Measurement of Minerals and Rocks Beijing China) and Coal Fly Ashes National Institutue of Standards and Technology (Gaithersburg MD USA) Standard Reference Material (SRM) 1633a and 82-201 Coal Fly Ash (Research Centre for Eco-Environmental Sciences Academia Sinica China) were decomposed under pressure in poly(tetrafluorethy1ene) (PTFE) crucibles with the addition of 4 ml of 67% HNO 6 ml of 35% HF and 4 ml of 72% HC10,680 JOURNAL O F ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 and placed in a heating vessel at 180°C for 8 h followed by evaporation of the solution to near dryness. The residues were dissolved in 5 ml of 0.8 mol 1-' HNO and the solutions filtered into 50 ml calibrated flasks. The concentrations of the final solutions were 2.00 mg ml-' of Cd. Biologicul sunzples Samples (1 g) of biological RMs were placed in PTFE crucibles and 5 ml of HNO,+ 1 ml of HF were added.The crucibles were capped and after being stored overnight the samples were further digested by the addition of 4 ml of HClO and heated in an oven. The solutions were then evaporated to near dryness followed by the addition of 5ml of 4mol 1-' HNO,. The solutions were transferred into 25 ml calibrated flasks and diluted with de-ionized water. The RMs used were cabbage pork liver (Ministry of Commerce China) tea tea tree leaf peach leaf (Research Centre of Eco-Environmental Sciences Academia Sinica) rice flour (Beijing Centre of Environmental Monitoring) and wheat flour (Ministry of Grain China). The concentrations of the final solutions were 40.0 mg ml-' of Cd. Results and Discussion Role of Mixed Modifier (Pd + TA) in WFP for Determination of Cd Palladium is a good retarding modifier of Cd because of the formation of the Pd-Cd alloy.Unfortunately Cd is a very volatile element so is lost easily at low temperatures before Pd can be reduced to the metallic form. Clearly to overcome the problem of loss of Cd Pd must be present in reduced metallic form as early as possible in the ashing stage. The addition of a reducing agent' is an essential and effective way to achieve the optimum result for Pd as a modifier. This has previously been shown by Shan et d8 and Zhuang et aL9 using ascorbic and citric acid respectively as reducing agents. In this paper a mixed modifier (Pd + TA) in a WFP is used for the determi- nation of Cd the mixed modifier solution is stable for several months without precipitation of the Pd metal.Fig. 1 illustrates the Cd absorbance profiles of 50pg of Cd for an aqueous solution in the absence of modifier and in the presence of 0.010 ml of 0.5 mg ml-' Pd in aqueous solution in 10 mg ml-' TA or 0.5 mg ml-' Pd+ 10 mg ml-' TA. Fig. 2 illustrates the absorbance profiles for 33.5 pg of Cd for a pork liver solution at a concentration of 50 mg ml-'. The ashing temperature is 50O0C which is the maximum ashing tempera- ture that can be used without significant loss of Cd in aqueous solutions in the absence of a modifier. The results reveal that in the presence of a Pd modifier the first peak which is thought to correspond to the dissociation of CdO is weak for 50 pg of Cd in aqueous solution and becomes stronger in the pork liver solutions. The second peak is believed to correspond to the atomization of the Pd-Cd species.The absorbance profiles for 50pg of Cd in aqueous and pork liver solutions for the mixed modifier (Pd + TA) and without modifiers reveal only one peak. The peak position without modifiers corresponds to the first peak (dissociation of CdO) and the peak position for the mixed modifier (Pd + TA) corresponds to the second peak (atomization of the Pd-Cd species). Evidently the addition of 10 mg ml-' of TA reduces all CdO species to Pd-Cd species so that the first peak disappears in atomization. Thus the reduction of Cd could also be promoted at lower temperatures in the presence of 10 mg ml-' of TA in a WFP. Theoretical Calculation of the Characteristic Masses and Atomization Effeciencies of Cd at Different Atomization Temperatures Calculation of the characteristic masses was performed as previously described' and a table of the calculated values mo (cal) is given as Table4 in ref.1. The calculated atomization 0.4 0.2 a 0 + a C m 2 0.4 0.2 0 (iai/'/+ A 0 kkx - ( 6 ) ' B' A' 0 5 ' 10' 16 Tirne/s Fig. 1 Atomization signals for 50 pg of Cd in H,O with W platform atomization. Injection volume 0.010 ml. Atomization temperature T(set) 1 2600; 2 2000; and 3 1400°C. Modifiers used (a) A 100 pg of tartaric acid; B 5 pg of Pd + 100 pg of tartaric acid; and (b) A' no modifier; and B' 5 pg of Pd 0.4 0.2 8 0 m n A' B' I 0 -7- 5 10 16 Tirne/s Fig. 2 Atomization signals for Cd in 50 mg ml-' SRM solution of Pork Liver (33.5 pg of Cd) with W platform atomization. Injection volume 0.010 ml.Atomization temperature T(set) 1 2600; 2 2000; and 3 1400°C. Modifiers used (a) A 100 pg of tartaric acid; and B 5 pg of Pd + 100 pg of tartaric acid; and (b) A' no modifier; and B' 5 pg of Pd modifierJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 68 1 efficiency (&IA) is defined as follows1o &IA= 100[rno(cal)/m,(exp)] where m,(exp) is the experimental characteristic mass. Ma et al.l have previously discussed the temperature dependence of atomization efficiencies of Cd in a Ta-foil platform inserted into a PGT &IA is near 100 when 0.5 mg ml-1 Pd is used as a modifier and atomization tempera- tures higher than T(set)= 1673 K and T(eff) = 1530 K are used where T(set) is set from the 28000 instrument and the effective atomization temperature T(eff) = 1/2[T(max) + T(end)].Atomization efficiencies can be used for the evaluation of the loss of Cd during ashing with the WFP and the dependence of the efficiency of Cd atomization on atomization temperature with the WFP. Ashing Temperature Dependence of Atomization Efficiencies of Cd with the WFP The dependence of atomization efficiencies of Cd with the WFP on ashing temperature are shown in Figs. 3-6. The high atomization temperature T(set) = 2373 K T(eff ) = 2100 K was used in order to obtain an &IA value near 100. T(set)= 1773 K T(eff) = 1550 K was used for sea-water in order to achieve low background absorption values. Fig. 3 shows the dependence of &IA on ashing temperature without a chemical modifier. It can be seen that the loss of Cd occurs at temperatures above 773 K.For selected sample solutions which have serious matrix effects the following values of &IA(%) can be reached 76 (sediment GSD10); 71 (1 mol 1-1 HClO,) 65 (tea tree leaf) and 47 (sea water) Fig.4 shows the dependence of &IA on ashing temperature when using 10 mg ml-1 of TA as modifier. The loss of Cd with the TA modifier occurs at temperatures above 773 K as observed in the absence of a modifier. Using 100 hq 50 0 x-x-x-x 500 1000 1500 2000 2500 Ashing temperature (T,,,) and atomization temperature (Teff)IK Fig. 3 Ashing (left-hand curves) and atomization (right-hand curves) temperature dependence of atomization efficiencies using a WFP without chemical modifier. Injection volume 0.010 ml. A 20 pg of Cd in H,O; B 20 pg of Cd in 1 mol 1-' HC104; C 20 pg of Cd in sea- water; D 40 mg ml-l solution of tea tree leaf RM (9.2 pg of Cd); and E 2.0 mg ml-' solution of water sediment RM ESDlO (22 pg of Cd) / 5--x-x-x-x I 0 2500 500 1000 1500 2000 Ashing temperature (T,,,) and atomization temperature (T,dIK Fig.4 Ashing (left-hand curves) and atomization (right-hand curves) temperature dependence of atomization efficiency using a WFP with 10 mg ml-' of tartaric acid as modifier. Injection volume 0.010 ml. Curves A-E as in Fig. 3 I L I I I j 500 1000 1500 2000 2500 Ashing temperature (TSet) and atomization temperature (T,,)/K 0 ' ' Fig. 5 Ashing (left-hand curves) and atomization (right-hand curves) temperature dependence of atomization efficiency using a WFP with 0.5 mg ml-' of Pd.Injection volume 0.010 ml. Curves A-E as in Fig. 3 I 100 1 hm 50 n " 500 1000 1500 2000 2500 Ashing temperature (T,,,) and atomization temperature (Teff)/K Fig. 6 Ashing (left-hand curves) and atomization (right-hand curves) temperature dependence of atomization efficiency using a WFP and 0.5 mg ml-' of Pd+ 10 mg ml-' of tartaric acid as mixed modifier. Injection volume 0.010 ml. Curves A-E as in Fig. 3 TA as modifier the values of &IA(%) are lower than those achieved without a modifier 59 (sediment GSD10); 48 (1 mol 1-1 HC104); and 41 (tea tree leaf). Using TA as a modifier the strong background absorption values can be depressed the over-corrected background correction in Zeeman AA is lower in TA than when no modifier is used and the value of &IA is enhanced to 79% for sea-water.Fig. 5 shows the depen- dence of &IA on ashing temperature using 0.5 mg ml-1 of Pd as modifier. The loss of Cd with the Pd modifier occurs at temperatures above 1073 K this value is 300 K higher than that with no Pd modifier. Using Pd as a modifier in the range 773-1073 K the values of &IA are near 100% and the matrix effects for 1 mol 1-1 HC104 sediment GSDlO and tea tree leaf can be eliminated effectively. However for an ashing tempera- ture lower than 773 K the matrix effects are still serious. Because of the over-corrected background correction for sea- water the &IA is low (71%). Fig. 6 shows the dependence of &IA on ashing temperature using 10 mg ml-1 TA + 0.5 mg ml-1 Pd as mixed modifier. Using a mixed modifier the values of &IA are near 100% in the range 473-1073 K and the matrix effect can be eliminated effectively. The over-corrected background correction for sea-water using a mixed modifier is the same as that using a TA modifier is 80%.From Figs. 3-6 it can be seen that the best modifier is 0.5 mg ml-' Pd+ 10 mg ml-' TA. Atomization Temperature Dependence of Atomization Efficiencies of Cd with the WFP The dependence of atomization efficiencies of Cd on atomiz- ation temperature with the WFP are shown in Figs. 3-6. The optimum ashing temperature used in Figs. 3-6 is 773 K. Fig. 3 shows that the value of &IA for 20 pg of Cd in H,O decreases below 1500 K (&IA= 91Y0) without a chemical modifier. A possible explanation for this is that some Cd is lost from the682 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 Table 1 Determination of Cd in solutions of RMs with the WFP and 0.5 mg ml- ' Pd+ 10 mg ml-' TA as mixed modifier. Injection volume 0.010 ml ashing temperature 573 K; atomization temperature T (set)= 1773 K T(eff)= 1550 K; m,(cal)=0.377 pg m,(exp)=0.377f0.010 pg dA = 100k 2.7%. Results based on eight measurements per sample Standard or sample Pork liver Cabbage Wheat flour Rice flour Peach leaf Tea Tea tree leaf Rock GSRl Tibetan Soil Soil ESS3 Soil 893 Coal Fly Ash NIST SRM 1633a Sediment 8 1 - I01 Sediment GSD 1 Sediment GSD2 Sediment GSD4 Sediment GSD5 Sediment GSD6 Sediment GSD9 Sediment GSDlO Sediment GSD 1 1 1 mol dm-3 HC104 1 mol dm H2S04 1 mol dm-3 HN03 1 rnol dm-3 HCl 82-201 Concentration of sample/ng ml- ' 50 40 40 40 40 40 40 10 10 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 10 Reference value (ppm) 0.067 0.029 0.03 1 0.020 0.020 0.032 0.023 0.030 0.08 1 0.044 1.24 0.16 1 .o 2.4 0.088 0.065 0.19 0.82 0.43 0.26 1.1 2.3 - - -.- . Concentration of Cd/ng ml - 3.35 1.16 1.24 0.80 0.80 1.28 0.92 0.30 0.162 0.44 2.48 0.32 2.0 4.8 0.176 0.13 0.38 1.64 4.3 0.52 2.2 4.6 5.0 5.0 5.0 5.0 Concentration of Cd determined*/ng ml - ' 3.52 k 0.090 1.22 0.046 1.30 f 0.067 0.84 f 0.049 0.85 f 0.01 7 1.30 f 0.049 0.90 f 0.030 0.298 f O.OO85 0.170 -t 0.0035 0.42 f 0.029 2.35 kO.061 0.34f0.018 1.90 f 0.063 4.84 f 0.155 0.187+0.0115 0.1 35 f 0.0044 0.40 f 0.01 2 1.63 + 0.096 4.41 f 0.184 0.54 & 0.039 2.05 f 0.097 4.67 f 0.188 5.0 f 0.15 4.48 f 0.25 5.08 f 0.07 5.0 f0.15 * Error limits are f 1 standard deviation. atomizer in a molecular form at low atomization temperature. The values of dA for Cd in solutions of 1 mol 1-' HC104 sediment RM GSDlO and tea tree leaf RM were low even when using a high atomization temperature [T(set)= 2873 K T(eff)=2500 K] the values of E ' ~ were 83 80 and 70% respectively and a sharp reduction in dA was obtained at lower temperatures. Because the 1 moll-' HC104 and the RM solutions contained perchloric acid the probable explanation is that Cd can be lost from the atomizer as gaseous cadmium chloride at low temperatures. The low dA for Cd in sea-water can be influenced by two factors firstly the formation of gaseous cadmium chloride during atomization and secondly the strong background absorption of the sea-water matrix at high atomization temperatures which causes over-correction even when using the Zeeman spectrometer. It can be seen from Fig.3 that the use of a chemical modifier is necessary for the determination of Cd in different environmental samples with a WFP. Fig. 4 shows the dependence of atomization efficiencies of Cd on the atomization temperature with the WFP using 10 mg ml- of TA as modifier. Using a TA modifier the matrix effects are more serious than those observed when no chemical modifier is used the values of dA(%) are 65 (Sediment RM GSDlO) 60 ( 1 mol 1-' HC104) and 52 (tea tree leaf RM) at T(eff ) = 2500 K. The over-correction of background absorption for sea-water is low in comparison to the case without a modifier dA is 30% for sea-water using the TA modifier.Figs. 5 and 6 show the dependence of atomization efficiencies of Cd on the atomization temperature using 0.5 mg ml-.' Pd as a modifier and 0.5 mg ml- ' Pd + 10 mg ml- TA as a mixed modifier respectively. The values of dA(%) are 80 [T(eff)= 1300 K] 90 [T(eff)= 1400 K]. 100-105 [T(eff)> 1550 K] for all sample solutions which have serious matrix effects. Because of over-correction of the strong background absorption of sea- water the values of dA are low for sea-water 72% (using Pd modifier) and 80% (using Pd + TA mixed modifier). From Figs. 3-6 it can be seen that the Pd modifier and the Pd + TA mixed modifier are the best modifiers for the determi- nation of Cd in environmental samples and that matrix effects can be eliminated effectively.Determination of Cd in Solutions of Environmental Sample RMs with the WFP using Pd + TA Mixed Modifier As has been shown the use of 0.5 mg ml-' of Pd + 10 mg ml-' of TA as a mixed modifier at high atomization temperatures with the WFP can eliminate matrix effects effectively. I n Table 1 the results are shown for the determination of Cd in solutions of various environmental RMs. The results obtained are in good agreement with the certified or reference values and indicate that the use of 0.5 mg ml-' of Pd+ 10 mg ml-' TA with the WFP is a practical and reliable procedure for routine analyses. The results in Table 1 were obtained by standardless analysis and provide strong evidence for the applicability of the standardless analysis approach. Stability of Integrated Absorbance Characteristic Mass Values over the Lifetime of a PGT with the WFP for the Determination of Cd Nitric and perchloric acids result in a signal depressing effect and corrosion of the surface of the pyrolytic graphite platform and PGT but have no effect on the surface of a WFP. This is particularly advantageous for the routine analysis of environ- mental biological and geological samples dissolved in nitric and perchloric acids. The values of rn,(exp) were stable over 200-300 firings.Because the protection of the PGT provided by argon was not good in the graphite furnace of the 28000 spectrometer atomization temperatures changed significantly after 200-300 firings and the values of mo were not stable.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 683 Conclusions Nitric and perchloric acids have no effect on the surface of a WFP. This is particularly advantageous for the routine analysis of environmental biological and geological samples dissolved in nitric and perchloric acids from the point view of stan- dardless or absolute analysis. Many sample solutions contain- ing perchloric acid suffer from serious matrix effects with W platform atomization and 0.5 mg ml-1 Pd+ 10 mg ml-I TA can be used to eliminate these matrix effects. The dependence of &IA on ashing temperature shows a plateau in the range 473-1073 K. The dependence of &IA on atomization tempera- ture reaches a plateau at temperatures higher then T(eff)= 1550 K T(set)= 1773 K and &IA is 100-105%. The possibility of developing a standardless analysis for the determination of Cd in environmental samples with a WFP and 0.5mgml-I Pd + 10 mg ml-I TA as mixed modifier was discussed. The authors are grateful to the National Natural Science Foundation of China for financial support of this research project. 1 2 3 4 5 6 7 8 9 10 11 References Ma Y.-z. Bai J. Wang J.-z. Li Z.-k. Zhu L. Li Y.-q. Zheng H. and Li B.-w. J. Anal. At. Spectrom 1992 7 425. Ma Y.-z. and Cheng J.-g. Fenxi Huaxue 1986 14 746. Ma Y.-z. and Cheng J.-g. Fenxi Huaxue 1988 16 225. Ma Y.-z. and Cheng J.-g. Guangpuxue Yu Guangpu Fenxi 1989 9 36. Ma Y.-z. and Cheng J.-g. Fenxi Huaxue 1990 13 266. Shan X.-q. and Ni Z.-m. Huaxue Xuebao 1981 39 575. Rettberg T.M. and Beach L.M. J. Anal. At. Spectrom. 1989 4 247. Shan X.-q. Ni Z.-m. and Yuan Z.-n. Anal. Chim. Acta 1985 171 269. Zhuang Z.-x. Yang P.-y. Luo J. Wang X. and Huang B.-l. Can. J. Appl. Spectrosc. 1991 36 9. Frech W. and Baxter D.C. Spectrochim. Acta Part B 1990 45 867. L'vov B.V. Spectrochim. Acta Part B 1990 45 633. Paper 3104065 H Received July 12 1993 Accepted January 12 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900679

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL O F ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 68 5 Studies on the Determination of the Metal Content of Airborne Particulates by Furnace Atomization Non-thermal Excitation Spectrometry* Christian Ludke Erwin Hoffmann and Jochen Skole lnstifut fur Spektrochemie und Angewandfe Spektroskopie (ISAS) Laboratorium fur Spektroskopische Methoden der Umweltanalytik (LSMU) Geb. 11. 1 Rudower Chaussee 5 12489 Berlin Germany Quantifying the metal content of airborne particulates is of outstanding relevance because of its potentially toxic and carcinogenic behaviour. The most widely used measuring strategy is based on filtration of particles from ambient air followed by analysis of the loaded filter tissue after wet digestion. In the present work an alternative approach was investigated which is based on drawing the air through the wall of a porous graphite tube in such a way that particulates are collected on the inner surface of the tube.Moreover the graphite tube acts not only as an efficient collector but can also be employed as an electrothermal atomizer when applying furnace atomization non-thermal excitation spectrometry. If combined with an echelle poly- chromator simultaneous multi-element determinations are possible. The amount of metals in air dust collected in this way were determined at the ng mP3 level with limits of detection calculated on the 3s criterion ranging between 0.1 and 1 ng m-3. The precision represented by the relative standard deviation varied from 0.12 to 0.26 depending upon the concentration of metals in the air.Keywords Furnace atomization non-thermal excitation spectrometry; airborne particulates; filters; multi- element analysis The increasing pollution of the atmosphere by natural and anthropogenic sources requires detailed investigation of a large variety of air polluting substances. Therefore measurements of atmospheric trace metals increasingly gain importance as they provide detailed information which can be used to identify sources of pollutants and follow their movements.' In order to study effects of major concern such as deposition mechan- isms or modifications of the climate the composition of particles of different sizes is of great interest. For these appli- cations it is desirable to measure the concentration of the elements in particles fractionated according to size.A pre- condition is that measurement methods having multi-element capability and high sensitivity are available. In most cases the presence of metals in the atmosphere is associated with airborne particulate matter. For the collection of airborne particulates many different sampling devices have already been developed.2 A conventional method is the collec- tion of particulates by pulling air through a filter tissue digestion of the filter and subsequent analysis of the solution preferably by atomic spectrometric techniques. Disadvantages of these methods are the time-consuming procedure and the fact that the chemical preparation of the samples involves sources of error^.^,^ These can result from blank values of the filters loss of certain elements during the digestion of filter material or contamination of the sample during the preparation procedure.A loss of particles is possible if they are smaller than the pore size of the filter. An alternative way of using filters is to filter the air directly through a porous graphite tube or cup which itself acts as the atomizer in an atomic spectrometric detection system. By applying such a sampling technique the disadvantages of filtering methods mentioned above are overcome. Drawing of air through a porous graphite tube followed by electrothermal atomic absorption spectrometry (ETAAS) has been described by Siemer and Wo~driff,~ as well as by Chakrabarti et aL6 Also combinations of electrostatic accumu- lation7 or single-stage impactions with ETAAS has been employed.Broekaertg reported on the direct determination of Pb and Cd by sampling the air through a porous graphite cup followed by excitation in a hot hollow cathode lamp. All these * Presented at the XXVIII Colloquium Spectroscopicurn Internationale (CSI) York UK June 29-July 4 1993. direct sampling methods except the last one have only single- element detection capability. The aim of the present work is to demonstrate a new method for the sensitive multi-element determination of trace metal contents in airborne particulates. In a simple sampling device air was pulled through the wall of a thermally cleaned porous graphite tube so that particulates were collected on the inner surface of the tube. After inserting the tube into the furnace atomization non-thermal excitation spectrometry (FANES) Source'O transient emission spectrometric signals were gener- ated by using an optimized working programme.For regis- tration of simultaneous spectrometric signals an tchelle polychromator was used." This method promises quick and sensitive multi-element analysis of collected particulates with- out any sample pre-treatment. Besides it is to be expected that the low limits of detection will allow short sampling times and analysis of the airborne particulates which are fractionated according to size. Experimental Sample Collection Sampling of airborne particulates was carried out by using a special porous graphite tube as filter. These graphite tubes were manufactured in-house from a block of pure graphite (28 mm long 6 mm i.d.and 1 mm thick). They had neither an injection hole nor any coating. According to the carbon products manufacturer (Ringsdorff-Werke Bonn Germany) the porosity was higher than that normally used for furnaces in analytical applications. The lower hardness of the graphite used is the reason for the shorter life of the tubes. The lifetime of a tube is about 60 firings. Before being used as a sampler each graphite tube was cleaned by three electrothermal heatings in the FANES source to 2600°C for 2 s. After this the furnace blanks were read in the normal heating programme (see Table 1). Cleaned tubes were stored separately in small polypropene containers before being used. For use as a filter the thermally cleaned graphite tube was held in a plastic vessel which was connected to a vacuum pump and had an opening for the entry of air.Inside the plastic vessel the graphite tube was fixed gas-tight on its ends686 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 Table 1 Working conditions for FANES source Element set 1 Ni 352.5 nm Fe 248.3 nm Cr 425.4 nm (:a 393.4 nm Sr 460.7 nm Mg 279.6 nm Temperature/ Ramp rate/ Step time/ Argon pressure/ Discharge current/ C "C s - I S h Pa mA 200 5 47 35 No power 53 800 200 5 2200 1600 3 35 No power 12 1013 50 50 50 1013 0 0 0 90 0 Element set 2 Na 586.9 nm Pb 405.8 nm Mn 403.1 nm Cu 324.8 nm T1 276.8 nm Cd 228.8 nm Temperature/ Ramp rate/ Step time/ Argon pressure/ Discharge current/ "C "C s - ' S hPa mA 200 5 47 35 No power 53 300 75 5 I800 lo00 4 35 No power 11 1013 20 20 20 1013 0 0 0 90 0 so that the air had to pass through the tube wall from inside to outside [see Fig.l(b)]. The system for size fractionated sampling of airborne par- ticulates consists of a combination of a standard sampling head for filter tissues of 50mm in diameter followed by the porous graphite tube in the plastic vessel. The standard sampling head was equipped with a poly( tetrafluoroethylene) 2 d I Fig. 1 Air sampling device (LI) scheme of set-up; and (h) detail A. 1 sampler; 2 magnetic valve; 3 timer; 4 pump; 5 porous graphite tube; 6 rubber seal; 7 plastic vessel; and 8 standard sampling head (PTFE) membrane filter of 5 pm pore size (Membrane Filter Type TE 38 Schleicher & Schiill Dassel Germany). Particles which have passed through the PTFE filter are collected in the graphite tube mounted behind it.An identical graphite tube without a pre-filter was used to collect the total air dust matter in the same sampling time. Six replicate samples were taken with the air sampling device as shown in Fig. 1. The sampling conditions were as follows. Samples (0.5-1.5 m3 of air) were taken at the site of the laboratory buildings in Berlin- Adlershof at a height of 3 m above the ground and at a distance of > 1.5 m from the buildings for a period of 2.5-4 h. A mechanical rotary pump suitable to maintain a pressure difference of 650 mbar (1 bar= lo5 Pa) at the sampling tube was used. Instrumentation To analyse the metal content of particulates collected in the graphite tube it was inserted into the FANES source.There the tube serves both as an electrothermal atomizer and as a cathode for the hollow cathode discharge. The computer assisted work programme of FANES involves temperature and time settings a pump-down step filling argon up to a low pressure level and switching the discharge voltage. Atoms generated by electrothermal atomization from the particulates collected were excited by impaction with the 4 r GraDhite tube Anode 1 Collimator Galvanometer Quartz Slits u k J mirror scanner prism I -e Pipette tip - - - - - Furnace ready for inserting the loaded graphite tube Fig. 2 Scheme of the experimental arrangementJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 687 discharge electrons. To measure the emission signal the FANES source was operated in conjunction with the Cchelle polychromator in a tetrahedral mounting designed by Falk et On the focal plane of the echelle polychromator about 120 optical fibres were fixed at the wavelength position of the most sensitive analytical lines.By connecting the appropriate optical fibres to the built-in photomultiplier tubes (PMTs) six elements can be measured simultaneously. A change in these connections allows the measurement of several sets of elements. For this study two sets of elements with six elements in each were chosen for simultaneous determination. With regard to their ashing and atomization temperatures the elements of each set were chosen so as to fit the compromise conditions best. The analytical signals decrease by < 10% when compro- mise rather than optimized conditions are used.For each set of elements separate tubes were taken for sampling and analy- sis. The working conditions of the FANES source (listed in Table 1) were adapted to the sets of elements also listed in the table. Analytical lines were selected on the basis of their providing the best detection limits as indicated by previous FANES reports.">' An oscillating quartz refractor plate 3 mm thick mounted on a galvanometer scanner behind the entrance slit of the polychromator (slit-width 150 pm) permits background correc- tion by wavelength modulation. A three-step square wave modulation was employed and the modulation frequency was 36Hz. All operational parameters of FANES as well as data acquisition and wavelength modulation were controlled by appropriate software installed in a PC.A schematic diagram of the experimental arrangement is given in Fig. 2. Both the FANES source and the Cchelle polychromator were manufac- tured in-house. Reagents and Calibration Reference solutions were prepared from inductively compled plasma (ICP) multi-element standard solution IV (Merck 11355.01) in 0.5 mol dmP3 HNO (sub-boiling quality). Neither the distilled water nor the HNO used contained measurable concentrations of the elements of interest here. These reagents were distilled sub-boiling in-house and the purity confirmed by FANES and ICP mass spectrometry measurements. For calibration a sample aliquot of 20 mm3 of reference solution was injected manually with a micropipette into the graphite tube still outside the furnace as seen in Fig.2. After this the tube was kept in a horizontal position and carefully inserted into the FANES source. The furnace was then closed and the 6 7 8 9 Fig. 3 Schematic sectioning of the digestion vessel 1 PTFE cap; 2 spring; 3 glassy carbon plate; 4 PTFE band; 5 graphite tube; 6 PTFE beaker; 7 glassy carbon vessel; 8 PTFE distance piece; 9 1 cm3 HNO f 3 cm3 HF concentrated heating programme started. The reproducibility of repeated injections of reference solutions in the uncoated tube ranges between 0.07 and 0.10 (relative standard deviation calculated from 12 measurements). For all elements studied linear cali- bration curves were determined. Two different methods were used to verify the accuracy of the calibration. (i) Isoformation by high-pressure vapour-phase digestion inside the graphite tube loaded with (a) airborne particulates and (b) dried reference solution.For this purpose a PTFE beaker containing the graphite tube was placed on top of a distance piece in the glassy carbon digestion vessel with only the distance piece standing in a mixture of concen- trated acids (1 cm3 HNO and 3 cm3 HF). A schematic diagram showing the sectioning of the digestion vessel is shown in Fig. 3. Six such glassy carbon vessels each with a spring-fixed lid were placed in a high pressure asher.I6 During the digestion procedure the temperature was maintained at 90°C and the pressure at 9 MPa for 1 h and 170 "C and 10 MPa respectively for another hour. After the gas-phase digestion the graphite tubes were analysed by FANES.(ii) Particulates collected at the inner surface of the tube were dissolved in 5rnoldm- HN03. For this purpose a small PTFE beaker containing the graphite tube was filled with HNO (1.5 cm3) till the tube was fully covered. After 8 h the graphite tube was removed and the remaining acid was analysed by electrothermal vaporization ICP-MSl7 according to the standard additions method. Results and Discussion Calibration The results of the different calibration procedures listed in Table 2 show acceptable agreement within the different methods. The results missing for Cd TI and Cu with the dissolution procedure were caused by diluting the analytes below the concentration which could be readily measured. Differences in calibration procedures checked by statistical tests were found to be random events.Although for Mn and Ni larger variations were observed a significance test at the Table2 Comparison of trace metal contents based on different calibration procedures; n = 5 Element Cd Pb Mg T1 c u Mn Ni Calibration with reference solution/ ng m-3 6.1 3t_ 0.4 212f 16 174+ 15 1.4f0.3 16+ 1 89+ 16 2 7 f 3 After isoformation/ ng m-3 6.6 f0.5 230+ 18 170f 14 2.5 f 0.5 15+ 1 64+6 36+7 After dissolution ng rn- 260 3t_ 70 120 f 40 - - - 100 f 30 50+7 Table 3 Metals contents determined in airborne particulates; n = 6 Element Ni Fe Cr Ca Sr Mg Cd Na TI Pb Mn c u Particulates unfractionated/ ng m-3 12$1 781 f 62 13+1 5200 f 420 27+2 580 & 29 7.2 2 514f 133 45$7 365 t- 72 174 +_ 21 64+_ 12 Particulates less than 5 pm/ ng m- 8.6 f 0.6 20+3 3 0.3 41 f 9 0.5$0.1 2.9 -t 0.3 1.9 -t 0.7 11.4f2 5.8 f 0.9 7.8 & 1 0.710.1 2.7 f0.9 LOD/ ng m-3 0.4 2 0.3 3 0.2 0.3 0.2 0.3 1 0.8 0.1 0.6688 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 Table 4 Composition of metal contents in airborne particulates measured by various techniques Sampling method Sample Sample Measuring volume/m3 pre-'treatment method Filtering by quartz 60-70 Wet digestion ETAAS * Filtering by porous 1.5 None FANES fibre filter tissues ICP-AESt graphite tube Metal content/ng m-3 Element Cd Cr Ni Sr c u Pb Fe Ca Ref. 184 2.21-6.45 8.51-25.17 4.3- 10.1 13.4-29.2 26.0-363.3 89-49 1 1097-625 1 2216-4598 Ref. 195 0.99-3.04 7.0-16.2 7.3-11.5 - 33.0-110 '73-303 - - This work (n = 9) 5.8 & 0.7 15-t 1.2 15 f 1.5 26f3 45f5 228 f 40 633 2 59 4520 2 500 * ETAAS used for Cd and Ni.t ICP-AES used for Cr Ni Sr Cu Pb Fe and Ca. 5 Mean values for the year. 99% level could not confirm their difference. The difference in TI values cannot be seen as real either as the measurements were made near the detection limit where higher errors are more likely. Measured Concentration in Air Measured concentrations of metal content in airborne particu- lates are given in the first column of Table 3. Flow rates through the graphite tubes range between 140 and 200 dm3 h-' and air volumes sampled range between 1.1 and 1.5 m3. The approximate sampling time of 8 h gave a concentration value in the worst case (Ni= 12 ng m-3) a factor 30 higher than its limit of detection (LOD). Provided that a reliable determi- nation is still possible at 8 times the LOD a total analysis times of about 2 h is sufficient as the measuring time by FANES is only 2min and to change tubes takes only a few seconds.The comparison with measurements of metals in airborne particulates carried out by established methods shows satisfac- tory agreement as seen in Table4. The comparative values were measured in 198818 and 199119 on selected sites in Berlin by the Association for Technical Inspection. Values for 1993 were not yet available at the time of writing but they should not differ widely from previous measurements. Size-fractionated Sampling The porous graphite tube collectors mounted behind the standard filter head were also analysed after loading them into the FANES source.To compare the metal content in total air dust with the metal content in particulates of less than 5 pm in size the values in both cases are summarized in Table 3. The LODs given in the last column are based on a 3s estimation of 15 furnace blanks related to a sample volume of 1 m3 of air. In most cases the LODs are one order of magnitude lower than the measured contents in particulates less than 5 pm in size. The elements with the highest contents in the air e.g. Ca Mg Nay Fe and Pb exist for the most part as particles greater than 5 pm in size. Elements with contents below 100 ng mV3 such as Ni Cry Cd T1 and Cu were generaly found as particles of less than 5 pm in size. This result is presented graphically in Fig. 4. 80 60 A + 2 40 E a 20 0 Fig. 4 71.7 26.3 Ni Ccl Cr TI Cu Sr Fe Na Ca Mg Mn Pb Element Percentage of given elements contained in particulates of less than 5 pm size Conclusion Collection of particulates by filtration of air through the wall of a porous graphite tube followed by simultaneous multi- element detection using a FANES-Cchelle system has proved to be a useful tool for quantifying the metal content of airborne dust. The procedure is quick and easy does not need any chemical preparatory steps and can be calibrated with refer- ence solutions.The detection power of the procedure presented was high enough to analyse the fraction of particulates smaller than 5 pm contained in 1 m3 of air. The financial support by the Senatsverwaltung fur Wissenschaft und Forschung des Landes Berlin and the Bundesministerium fur Forschung und Technologie is gratefully acknowledged.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 689 1 2 3 4 5 6 7 8 9 10 11 References Gordon G. E. in Air Pollutants and Their ESfects on the Terrestrial Ecosystem ed. Legge A. H. and Krupa S. V. Wiley New York 1986 138. Klockow D. Fresenius’ Z . Anal. Chem. 1987 326 5. Tolg G. and Tschopel P. Anal. Sci. 1987 3 199. Tschopel P. Kotz L. Schulz W. Veber M. and Tolg G. Fresenius’ Z . Anal. Chem. 1980 302 1. Siemer D. D. and Woodriff R. Spectrochim. Acta Part B 1974 29 269. Chakrabarti C. L. He X. Wu S. and Schroeder W. H. Spectrochim. Acta Part B 1987 42 1127. Torsi G. and Bergamini D. Ann. Chim. 1982 79 45. Lian Z.-w. Wei G.-t. Irwin R. L. Walton A. P. Michel R. G. and Sneddon J. Anal. Chem. 1990,62 1452. Broekaert J. A. C. Bull. Soc. Chim. Belg. 1976 85 755. Falk H. Hoffmann E. and Liidke C. Prog. Anal. At. Spectrosc. 1988 11 417. Schmidt K. P. Becker-ROB H. and Florek S. Spectrochim. Acta Part B 1990 45 1203. 12 13 14 15 16 17 18 19 Falk H. Becker-Rol3 H. Florek S. Hoffmann E. Ludke C. and Tischendorf R. in 4. Coll. Atomspektrom. Spurenanal. ed. Welz B. Perkin-Elmer Uberlingen 1987. Falk H. Hoffmann E. and Liidke C. Spectrochim. Acta Part B 1984 39 283. Naumann B. Knull B. Kerstan F. and Opfermann J. J. Anal. At. Spectrorn. 1988 3 1121. Baxter D. C. Nichol R. Littlejohn D. Liidke C. Skole J. and Hoffmann E. J. Anal. At. Spectrorn. 1992 7 727. Knapp G. Intern. J. Environ. Anal. Chem. 1985 22 71. Hoffmann E. Ludke C. and Scholze H. J. Anal. At. Spectrum. submitted for publication. Technical Report No. D-89,485 ed. TUV Berlin e.V. Berlin 1989 170. Technical Report No. U-92/279 ed. TUV Berlin-Brandenburg Berlin 1992 58. Paper 3/07135I Received December 2 1993 Accepted February 22 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900685

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 691 Flame Atomic Absorption Spectrometric Determination of Cadmium in Biological Samples Using a Preconcentration Flow System With an Activated Carbon Column and Dithizone as a Chelating Agent Yaneira Petit de Pefia Mercedes Gallego and Miguel Valcarcel* Department of Analytical Chemistry Faculty of Sciences University of Cordoba I4004 Cordoba Spain A combined flow atomic absorption spectrometric system was used to develop an efficient on-line preconcen- tration-solvent elution procedure for the determination of trace amounts of cadmium. The metal was preconcentrated as the dithizonate on a mini-column packed with activated carbon and eluted with isobutyl methyl ketone (4-methylpentan-2-one) into the nebulizer. A preconcentration factor of 40-1 30 equivalent to 3-12 ml of sample was achieved by using a time-based technique.The detection limit obtained (30) ranged between 0.3 and 1.3 ng ml-' and the relative standard deviation from 1.3 to 2.4% for a sample volume of 12 and 3 ml respectively. The results obtained for the determination of cadmium in reference materials testify to the applicability of the proposed procedure to the analysis of biological materials. Keywords On-line sorption-extraction; activated carbon mini-column; cadmium; biological material; flow injection flame atomic absorption spectrometry Knowledge of the detrimental effects of trace and ultra-trace amounts of metals in the environment and biological materials such as blood and animal tissues has greatly increased in recent years.Because of the extremely low concentrations of metals in these matrices a preliminary concentration step is usually necessary before their determination. There are cur- rently two major preconcentration methods used uiz. off- and on-line. The separation of metal ions from liquid samples in both methods can be obtained in several ways; however one of the most commonly used approaches involves allowing the sample to flow through a column packed with an active material. Preconcentration is accomplished by adjusting the sample pH to an appropriate value and using a chelating agent to interact with the metals of interest before they are retained or immobilized on the active material. A literature scan- search on enrichment of heavy metals on activated carbon by off-line procedures revealed that this is usually carried out after chelation with ammonium pyrrolidinedithiocarbamate (APDC ammonium pyrrolidin-1-yl dithioformate),' 8- hydroxyquinoline ( 8-HQ),2 potassium ethyl anth hate,^^^ dithi- zone,5 chrome azurol S6 or the ammonium salt of the dithi- ophosphoric acid 0,O-diethyl ester;7 also following desorption in a small volume of nitric acid the metal concentrations are measured by atomic absorption spectrometry (AAS) or (as a slurry) by inductively coupled plasma atomic emission spec- trometry (ICP-AES).Activated carbon has proved to be an excellent collector for ~admiurn~-~ after formation of neutral chelates. Dithizone has been used as a chelating agent for collection of trace amounts of cadmium on a thin layer of activated ~ a r b o n ; ~ the metal chelate is released from the activated carbon layer by digestion with 14 moll-' nitric acid then the nitric acid is evaporated and the released trace metal element is collected in a small volume of dilute acid.Recoveries above 90% can be obtained at hydrochloric and nitric acid concentrations below 0.005 and 0.002 moll- ' respectively; the detection limit is 0.3 ng ml-' of cadmium and the preconcen- tration factor is high (1 1 of water can be reduced to 1 ml). Two classes of column materials have been used for precon- centration of cadmium and matrix removal in flow injection analysis viz. ion exchanger^^'^ and sorbent materials such as C18;1s22 there is only one reference to the use of activated carbon impregnated with 8-HQ in continuous systems for preconcentration of cadmium by ICP-AES (detection limit 0.25 ng ml-' of cadmium).23 The present paper describes a new method whereby trace ~ * To whom correspondence should be addressed.amounts of cadmium are preconcentrated as a metal chelate using the chelating dye dithizone on an activated carbon mini- column included in the flow injection (FI) system. The chelate is eluted with a small volume of a water-immiscible solvent and the analyte does not disperse on transfer to the flame atomic absorption spectrometry (FAAS) instrument which increases the preconcentration factors. The system was used successfully for the determination of cadmium in biological reference materials. Experimental Apparatus A Perkin-Elmer 380 atomic absorption spectrometer equipped with a bead impact system in the burner chamber and a hollow cathode cadmium lamp was used.The wavelength and lamp current used were 229 nm and 4 mA respectively; deuterium arc background correction was employed throughout. The acetylene flow rate was 2.01 min-' and an air flow rate of 21.5 1 min-' was employed to obtain a clean blue flame. The flow system consisted of a Gilson-Minipuls-2 peristaltic pump furnished with poly(viny1 chloride) tubes two Rheodyne 5041 injection valves and a laboratory-made adsorption mini- column packed with 70mg of activated carbon. The mini- column (2.5 cm in length and 3 mm i.d) was made of poly (tetra- fluoroethylene) (PTFE) capillary and sealed at one end with a small glass-wool bead to prevent losses of material.The column was initially flushed with 0.1 moll-' nitric acid and the subsequent use of isobutyl methyl ketone (IBMK) as eluent in each operating cycle was sufficient to make it ready for re-use. Peak heights and areas were measured with a Merck-Hitachi D-2500 Chromato-Integrator. A Hetosicc Freeze Dryer Type CD-53-1 was also employed. Reagents and Standard Solutions All chemicals used were of analytical-reagent grade and the water was ultrapure (Milli-Q Water System Millipore Seville Spain). A 1000 mg 1-' cadmium stock solution was prepared by dissolving 1.000 g of the metal in a small volume of concentrated nitric acid and diluting to 1 1 with 1% (v/v) nitric acid. A saturated solution of dithizone (Riedel de Haen Hannover Germany) was prepared as follows 5 mg of dithi- zone were shaken electromagnetically in a 100 ml vessel con- taining 50 ml 0.4 moll-' ammonia solution for 3 min.The solution was then filtered and the filtrate diluted with water in a 100 ml calibrated flask. Isobutyl methyl ketone (Probus,692 JOURNAL OF ANAJLYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 Barcelona Spain) was also used. Darco 20-40 activated carbon (Aldrich Barcelona Spain) and polygosyl bonded silica reversed-phase sorbent with octadecyl functional groups (RP-C,,) 60-100 pm (Millipore) were employed as sorbents. Standard solutions (100 ml) containing 0.5-70 ng ml-' of cad- mium were all freshly prepared in 0.01 mol I-' sulfuric acid by appropriate dilution of the stock standard solution (1000 mg I-').Procedures Sample preparation The reference materials analysed were as follows Pig Kidney (Community Bureau of Reference BCR No. 186) City Waste Incineration Ash (BCR No. 176) Oyster Tissue [National Institute of Standards and Technology (NIST) Standard Reference Material (SRM) 1566a1 and Lobster Hepato- pancreas Marine (National Research Council Canada TORT-1). The reference materials were dried to constant mass by freeze drying at 6 Pa (0.04 mmHg) for 24 h after which an accurately weighed amount of 250-500 mg was digested24 with 10 ml of 65% nitric acid and 1 ml of 24.5% sulfuric acid in glass digestion tubes. The mixture was heated at 150-200 "C in a digestion block until the sample was completely dissolved and nitrogen fumes were given off. The tubes were allowed to cool for about 2min and the digestion was repeated (about six times) in the same way with multiple additions of 5 ml of nitric acid until a clear solution remained and appearance of nitrogen dioxide fumes had ceased.Once cold the solution was transferred quantitatively into calibrated flasks of 50 or 100ml capacity and made up to volume with ultrapure (Milli-Q) water. A reagent blank was prepared in parallel. Sub-samples of between 1 and 25 ml diluted to 50 or 100 ml and a pH of 2.0 adjusted with 2 moll-1 sulfuric acid were analysed immediately after preparation by introducing them into the manifold shown in Fig. 1. Continuous preconcentration-elution system The FI manifold for on-line preconcentration and elution of cadmium(I1) is illustrated in Fig.1 together with the optimized operating parameters. The sample or standard containing 0.5-70 ng ml-' of cadmium(11) in 0.01 moll-' sulfuric acid was continuously pumped through the manifold for 1-4 min and thoroughly mixed with the chelating reagent (saturated dithizone in 0.2 moll-' ammonia solution). After merging retention of the chelate took place on the activated carbon column placed in the loop of the injection valve and the sample matrix was sent to waste (W). During this step a water carrier was pumped to the instrument in order to flush the nebulizer after each measurement. The preconcentration step was termin- ated when the two injection valves were switched simul- taneously so elution of the adsorbed chelate took place when 150 pl of IBMK were passed through the adsorbed chelate to desorb it and sweep the cadmium to the detector. The duration of the elution step was set to 20 s.The peak height absorbance of the elution signal for quantification and a blank of 150 pl of IBMK injected prior to sample preconcentration were used (about 0.050 absorbance units). In this step the sample was also changed whereby the remainder of the previous sample in the pump tube was driven to waste and the next sample made ready for preconcentration. Results and Discussion Selection of Chelating Reagent and Eluting Solvent Activated carbon has been used as a trace collector for multi- element preconcentration by simply adjusting the pH to an appropriate value or using a chelating agent. Best results are obtained when the metals are complexed with organic chelating (AAS) Sample mI rnin-' 300cm Column )W2 H 2 0 4.0 Dithizone 0.3 Sample 3.0 Fig.1 Schematic diagram of the assembly used for on-line preconcen- tration of Cd. (a) and (b) adsorption and elution step respectively. Bold lines denote flows relevant to the individual stages. IV Injection valve; W waste; IBMK isobutyl methyl ketone agents prior to adsorption on the activated carbon. In order to test this approach a variety of chelating agents were assayed in an FI system similar to that shown in Fig. 1 namely 8-HQ dithizone 4-( 2-pyridylazo)resorcinol (PAR) ammonium diethyldithiocarbamate (NH,DDC) and APDC. The reagents at 0.1 % (m/v) concentration were prepared in ultrapure water except for dithizone (sparingly soluble in water) which was dissolved in ammonia solution (5 mg in 100 ml of 0.2 moll-' ammonia solution).Several calibration graphs were run for cadmium by using the above reagents; the sensitivity (slope of the calibration graph) achieved by using dithizone was about 20 40 60 and 100% higher than that obtained with APDC NH,DDC 8-HQ and PAR respectively. So dithizone was selected as the chelating agent. Elution of the adsorbed chelate from the column was investigated with different solutions. The solvents used for this purpose were 0.1 moll-' sulfuric acid ethanol acetone carbon tetrachloride chloroform and IBMK. By using the automatic configuration shown in Fig. 1 and 6 ml sample volumes containing 15 ng ml-' of cadmium(11) in 0.01 moll-' sulfuric acid and injecting 150 pl of the extractant the above mentioned eluting reagents were assayed in order to select the fastest.Solvent changeovers required flushing the column with IBMK in order to remove the residual adsorbed chelate. As can be seen in Fig. 2 the best results (difference between the sample and blank) were provided by IBMK and chloroform. The final choice was IBMK because it was less toxic than chloroform and released no hydrogen chloride in the flame; in addition it gave rise to a lower blank signal. Other extractants (sulfuric acid ethanol and acetone) provided worse results either because they were water miscible and the plug underwent dispersion or because the chelate could not be dissolved. Peak heights were selected as the analytical measurements because elution with IBMK was instantaneous.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 693 il 0.100 I - 1 v * LJL ( a ) I I I----I 30 s Time - Fig. 2 FI peaks obtained with different eluting solutions. (a) and (b) blank and sample signal respectively. Sample concentration 15 ngml-' of Cd. Peaks (peak areas in FV s are given in parentheses for sample and blank respectively) 1 = 0.1 mol I-' H,SO (715; 120); 2 = ethanol (2740; 3719); 3 = acetone (7227; 4392); 4 =carbon tetra- chloride (3950; 2952); 5 = IBMK (8227; 4217); 6 = chloroform (7050; 2946) Optimization of Chemical Variables Initially an attempt was made to dissolve dithizone which is moderately soluble in dilute ammonia solution ethanol ace- tone chloroform and carbon tetra~hloride.~~ Water immiscible organic solvents were discarded since the aqueous sample and chelating agent should be miscible.The best results were obtained by dissolving 5 mg of dithizone in 100 ml of dilute ammonia solution or 20 + 80 (v/v) ethanol-water. The effect of the sample pH was studied by introducing 6ml (sample flow rate 3.0 ml min-'; pumping time 2 min) of a solution contain- ing 15 ng ml-' of cadmium into the system the pH being adjusted to 1.0-5.0 with dilute sulfuric acid. As can be seen in Fig. 3 the maximum chelate adsorption was achieved at pH 2.0 and 4.0 for dithizone in 0.2moll-' ammonia solution and 20% ethanol respectively since the ammoniacal medium allowed the reagent to be acidified to a greater extent even though the pH obtained on mixing the sample with the reagent was similar in both cases (pH about 2.5).Ammonia solution was therefore selected to dissolve dithizone because it enabled work to be carried out at a lower pH than the ethanol solution which is an obvious advantage in terms of selectivity with a view to future applications to real samples. The influence of various acids on the preconcentration reaction was also studied I 0.05 1 d I - 0 1 2 3 4 5 PH Fig.3 Effect of pH on Cd absorbance as measured after on-line preconcentration with dithizone in A 0.2 moll-' ammonia solution or B 20% v/v ethanol-water. Sample 15 ng ml-I of Cd; dithizone 5 mg in 100 ml of solvent by using samples containing 15 ng ml-I of cadmium at pH 2.0 adjusted with nitric sulfuric hydrochloric or perchloric acids. The absorbance decreased by 60% with nitric and hydrochloric acids and 80% with perchloric acid relative to that in sulfuric acid.The effect probably arose from the cadmium-dithizone chelate being protonated at an acid pH and sulfate ion being involved in the coordination sphere thereby favouring adsorp- tion of the chelate. Thus the aqueous samples were all prepared in 0.01 moll-' sulfuric acid. The effect of the dithizone concentration was studied in the range 4 x 10-5-4 x mol 1-' by dissolving different amounts (1-10 mg) in 100 ml of ammonia solution at various concentrations (0.1-0.4 moll-'). The signal increased with increasing dithizone concentrations up to 1.5 x moll-' (4 mg of dithizone in 100 ml) above which it remained constant because the solution was saturated. The signal also increased with increasing ammonia concentrations up to 0.2 mol 1-' at any of the dithizone concentrations assayed; however above 0.25 moll-' of ammonia the signal started to decrease as a result of the reaction being hindered by the increased alkalinity in the sample-dithizone reaction coil.A dithizone concen- tration of 2 x 1 0 - ~ mo1 1-1 ( 5 mg in 100 ml) in 0.2 mol I-' ammonia solution was finally chosen. The influence of temperature from 20 to 60°C on the preconcentration of 15 ng ml-' of cadmium was studied by thermostating the preconcentration reactor located before the column and the activated carbon column. The absorbance decreased slowly with increasing temperature above 30 "C (by 10 and 30% at 40 and 50°C respectively relative to room temperature) because probably the cadmium-dithizone com- plex decomposed so measurements were made at room tem- perature.Replacing the sample stream with 0.01 mol 1-1 sulfuric acid (blank) resulted in a similar absorbance in the eluting step to that obtained by successively injecting 150 pl of IBMK solvent before the preconcentration step; no blank (0.01 moll-' sulfuric acid) was therefore required. FI Conditions for Preconcentration-Elution The FI system was optimized by the univariate approach. The flow variables studied for a sample of 15 ng ml-' were the sample and reagent flow rate the length of the preconcen- tration coil and the volume of IBMK in addition to the IBMK flow rate for the preconcentration and elutions step. Changes in the flow rate of the sample (6.0 ml of a solution containing 15 ngml-' of cadmium) between 0.6 and 4.0mlmin-' at a constant reagent flow rate of 0.3 ml min-' resulted in increas- ing the analytical signals up to 1.5 mlmin-' because the sample was less markedly diluted at the higher flow rates (Fig.4). Very small variations were observed in the range 1.5-3.0 ml min-'. The signal decreased at flow rates above 3.0 ml min-' because the dispersion increased and also decreased the residence time. Once the dithizone solution was saturated increasing the flow rate was equivalent to increasing the dithizone concentration; however the sample was diluted 0 1 2 3 4 Sample flow rate/mt min-' Fig. 4 Dependence of the efficiency of preconcentration on the sample flow rate. Sample and dithizone concentrations as in Fig. 3694 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 Table 1 Figures of merit of the determination of cadmium by preconcentration on activated carbon mini-column Detection limit/ Sampling frequency/ Enrichment factorf- Time/min Regression equation* Range/ng ml-I ng ml-' RSD (%) h-' 1 A = -0.002 + 0.006~ 3-70 1.3 2.4 40 40 2 A = 0.003 + 0.01 IX 1-40 0.6 1.7 25 80 2 A = 0.003 + 0.007~$ 2-60 1.1 2.2 25 50 4 A = 0.001 +0.019~ 0.5-22 0.3 1.3 113 130 * A absorbance and x cadmium concentration (ng ml-'). t Compared with conventional sample introduction of an aqueous solution (A = 0.006 + 1.42 x 10-4~). $ By using an RP-CI mini-column. simultaneously so the atomic absorption signals decreased with increase in the dithizone flow rate from 0.3 to 0.9 ml min-l. A reagent flow rate of 0.3 ml min-' and a sample flow rate of 3.0mlmin-' were chosen as a compromise.Maximum chelate adsorption was achieved at a length of the preconcentration coil (located before the activated carbon column in the loop of the injection valve) of 250cm as the chelate was completely formed at the resulting residence time (k 9 s). Longer lengths up to 430 cm gave rise to constant signals. With coils longer than 430 cm the chelate formed was carried along the system and probably adsorbed on the inner walls of the PTFE tubing (Le. spread over a large surface area) so it was incompletely dissolved by the injected IBMK and the analytical signal decreased as a result. A coil that was 300cm long and 0.5 mm in i.d. was selected for further experiments (residence time 11 s).The volume of IBMK played an important role in the chelate elution its effect being studied between 40 and 300 yl. The adsorbed chelate was found to be eluted throughout this range but some carryover was observed below 100 pl; at volumes above 150 pl the signal decreased through dilution of the analyte in the solvent. Accordingly experiments were carried out with 15 ngml-' of cadmium (the method can be applied to higher concentrations of cadmium) the volume of IBMK finally used being 150 pl in order to avoid carryover and allow usage of only one injection per sample. Since water acted as the carrier for injected IBMK it was essential to examine the effect of its flow rate. The peak area remained constant over the range 1-6mlmin-l but the peak height increased with increasing water flow rate up to 3.5 ml min-' because the nebulizer efficiency also increased in parallel.A water flow rate of 4.0 ml min-' was thus chosen. Determination of Cadmium Several calibration graphs for cadmium(11) were run by using sampling times of 1 2 and 4 min which is equivalent to using sample volumes of 3.0 6.0 and 12.0m1 respectively; higher enrichment factors were obtained at higher sample volumes. The correlation coefficients obtained ranged between 0.998 and 0.999 in all instances. The detection limit was calculated as three-fold the standard deviation of the peak height Table3 Results obtained for the determination of cadmium in reference materials. All values are given in pg g-' of cadmium Certified Found Reference material value (n=3) City Waste Incineration Ash (BCR No.176) 470 _+ 9 466 11 Pig Kidney (BCK No. 186) 2.71 k0.15 2.95f0.44 Oyster Tissue (SRM 1566a) 4.15k0.38 3.97L0.36 Lobster Hepatopancreas Marine (TORT-1) 26.3 & 2.1 28.6 & 2.1 absorbance for 15 injections of 150 pl of IBMK (blank). The precision of the method [expressed as the relative standard deviation (RSD)] was checked on 11 samples containing 5 or 20 ng ml-' of cadmium each at different time-based sampling of 2 and 4 or 1 min respectively. Preconcentration factors of up to 130 calculated as the ratio between the slopes of the calibration graphs provided by this method and by direct aspiration of cadmium(Ir) were achieved for a sample volume of 12 ml (sample flow rate 3.0 ml min-'; pumping time 4 min).A comparative study of the chelate retention on an activated carbon mini-column and a reversed-phase silica sorbent (RP-CI8) mini-column both of the same dimensions was carried out with the column placed after the 300 cm preconcen- tration PTFE coil in the loop of the injection valve. The characteristic parameters of the calibration graphs are listed in Table 1 from which the following conclusions can be drawn (a) at the same sample volume 6.0m1 the sensitivity (slope of the calibration graph) was 1.6 higher for activated carbon than for the RP-C18 sorbent; (b) the linear range was wider for the RP-Cl8 mini-column; and (c) such analytical features as the detection limit and precision were similar in both instances but the enrichment factor for the activated carbon mini-column was more favourable. Because of the non-specific character of the complexant dithizone the effect of the most common ions that react with it were investigated in order to identify potential interferences.The cations investigated included Cu2+ Ni2+ Co2+ Zn2+ Pb2+ Mn2+ Sn2+ Hg2+ F e 3+ B'3+ 1 and A13+ which were tested at concentrations up to 5 yg ml-l for a sample containing 5 ng ml-' of cadmium (sample loading time 2 min). Normally these cations are tolerated at high concen- trations owing to the high selectivity of the atomic spectro- Table 2 Tolerated concentrations of foreign cations in the determination of 5 ng ml-' of cadmium Ion cu2+ Bi3 + Sn2 + Fe3 Hg2+ Ni2 + Zn2 + co2+ Pb2+ Mn2+ ~ 1 3 + Metal concentration/ pg ml-I 0.30 0.50 0.50 0.75 0.75 1 .oo 1 .oo 1.50 2.50 5.00 5.00 Tolerated ratio [metal] [Cd] 60 100 100 150 150 200 200 300 500 1000 1000 Metal concentration/pg ml- ' 0.5 0.75 0.75 1 .o 1 .o 2.5 2.5 2.5 5.0 Signal supression (%) 56 14 55 10 16 56 12 39 27 -Table 4 Features of automatic-preconcentration methods for the determination of cadmium by atomic spectrometric techniques Detection FAAS FAAS FAAS FAAS FAAS FAAS FAAS ETAAS I/ ETAAS FAAS ETAAS ETAAS FAAS FAAS ETAAS ICP-AES ICP-AES ICP-AES ICP-AES Preconcentration method I-E* (Chelex-100) I-E (Resin-122) I-E (Chelex-100 S-HQ,? Resin-122) I-E (TriPEN)t$ I-E (Chelex-100) I-E (Chelex-100) I-E (Several resin) I-E (Amberlite XAD-2R Sorbent extraction (cis)** Sorbent extraction (c18)?? Sorbent extraction (c18)tt Sorbent extraction (C,,)$$ Sorbent extraction (CIS)?? Sorbent extraction@ Biosorption (Alga)? Co-precipitationv Co-precipitation 11 11 I-E (8-HQ)t I-E (IDAECE Sampling Eluting solvent Preconcentration factor frequency/h- RSD (%) Application Ref.2 moll-' HNO 2moll-' HNO 2 rnol I-' HN0 1 mol I-' HCl+0.1 moll-' HNOJ 6 mol I-' HNO 2moll-1 HNO 2 mol I-' HNO 2 rnol I-' HNO 0.2 moll-' HCl 2.0 mol I-' HCI+O.l mol 1-1 HNO Acetonitrile Ethanol Ethanol or methanol Methanol Ethanol 0.5 moll-' HCl 0.1 moll-' HCl IBMK IBMK 20 20-28 50-100 500 30 15 2-4 100 125 20 17 18-25 50 5-450 16-19 - 500 16 43-52 30-60 40 60 2 30 24 10 12 12 23 120 7 23 35 20 24 3-150 - - - 1.5-4.1 1.2-3.2 - - 3 < 1.7 0.8-1.2 1.7 5-10 4 3.3 1.4 2 2.0-3.3 1.1 1.5 1.5 5.9 Sea-water Water Tap water - - - Biological standard material Sea-water urine River water Antartic sea-water Sea-water Sea and river water SRM Sea and drinking water Sea-water SRM Sea-water and river water SRM Biological materials Biological samples Whole blood digests - - 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 26 27 28 * I-E ion exchange. t Immobilized on controlled pore glass.$ TriPEN N,N N'-tri( 2-pyridylmethy1)ethylene diamine. 9 IDAEC iminodiacetic-ethylcellulose. 7 XAD-2 resin functionalized with 1-( 2-thiazolylazo)-naphth-2-ol. 11 ETAAS electrothermal atomic absorption spectrometry. ** Chelation by pyrrolidin- 1-yl-dithioformate ( pyrrolidinedithiocarbamate) with subsequent adsorption. tt Chelation by DDC with subsequent adsorption. $$ Chelation by APDC with subsequent adsorption. & Active carbon impregnated with 8-HQ.M[ Coprecipitation on the iron@-hexahydroazepinium hexahydroazepine-1-dithiocarboxylate. )I /I Coprecipitation on the iron@)-hexahydroazepine-1-dithiocarboxylate.696 JOURNAL OF ANALYTICAL ATO'MIC SPECTROMETRY JUNE 1994 VOL. 9 metric technique (Table 2). Only copper@) interfered at concentration ratios below 100. In all instances the interferent decreased the cadmium signal because the dithizone concen- tration was inadequate for the chelates of both the interferent and the cadmiurn to be formed so formation of the foreign cation chelates at higher concentrations Was favoured or the IBMK volume used was inadequate for eluting all the chelates adsorbed. Manganese@) and aluminium(I1I) were both toler- ated at the highest concentrations assayed.Applications The accuracy of the proposed method for the analysis of biological samples was tested by determining cadmium in Pig Kidney Oyster Tissue Lobster Hepatopancreas Marine and City Waste Incineration Ash. Each sample was mineralized in duplicate as described previously (see Procedure) together with a similarly prepared blank. Each dissolved sample was analysed in triplicate. The blank absorbances corresponded to a cadmium concentration of less than 0.4 ng ml-I (this blank allowed the contribution of cadmium ion present in the reagents to the digestion sample used to be assessed). The results obtained are shown in Table 3. As can be seen consistent values with certified values were obtained. Conclusions The results obtained in this work testify to the applicability of on-line sorbent-extraction preconcentration to FAAS for the determination of cadmium in complex matrices such as biologi- cal materials. The proposed method is fairly fast and simple; also taking into account the lack of dithizone selectivity it could probably be applied to other cations such as lead zinc copper and mercury.The features of other automatic methods for the preconcentration and determination of cadmium by atomic spectrometry are summarized in Table 4. Different ion- exchange chelating resins have been used with a water-miscible eluent (dilute acids) which results in the analyte being dispersed on transfer to the detector. The preconcentration factors achieved range from 20 to 500; however an enrichment factor of 500 was achieved from 100 ml of sample by using a sampling time of 25 min'l but the immediate result of using longer sampling times was obviously a decreased sample throughput.The proposed system is clearly superior to existing continuous- flow alternatives using C18 mini-columns since (a) both the sample matrix and the eluent are sent to waste during the preconcentration step; (b) the preconcentration reactor and the column are located in the loop of the injection valve so the eluent sweeps any chelate potentially adsorbed by the reactor thereby avoiding any carry-over; and (c) the eluent (IBMK) is water immiscible and the analyte is not dispersed on transfer to the detector which results in higher preconcen- tration factors (about 20 for the methods shown in Table4) versus 40- 130 for the proposed method.The coprecipitation of cadmium with iron complexes provides lower preconcen- tration factors even though IBMK is also employed as the eluting solvent. The methods shown in Table4 have usually been applied to water samples (there are only four applications to biological materials). The Comision Interministerial de Ciencia y Tecnologia is acknowledged for financial support (Grant No. PB 93-0717). Y.P. de P. is also grateful to the University of Cordoba the University of Los Andes and Consejo Nacional de Investigaciones Cientificas y Tecnol6gicas (Venezuela) for additional financial support. Professor R. E. Santelli is also acknowledged for his invaluable suggestions. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 References Kimura M.and Kawanami K. Talanta 1979 26 901. Vanderborght B. M. and Van Grieken R. E. Anal. Chem. 1977 49 311. Kimura M. Talanta 1977 24 194. Devi P. R. and Naidu G. R. K. Analyst 1990 115 1469. Beinrohr E. Rojcek J. and Garaj J. Analyst 1988 113 1831. Ambrose A. J. Ebdon L. and Jones P. Anal. Proc. 1989,26,377. Monte V. L,. A. and Curtius A. J. J. Anal. At. Spectrom. 1990 5 21. Olsen S. Pessenda L. C. R. RiiiiEka J. and Hansen E. H. Analyst 1983 108 905. Fang Z. Xu S. and Zhang S. Anal. Chim. Acta 1984 164 41. Fang Z. RGiEka J. and Hansen E. H. Anal. Chim. Acta 1984 164 23. Malamas F. Bengtsson M. and Johansson G. Anal. Chim. Acta 1984 160 1. Bengtsson IM. Malamas F. Torstensson A. Regnell O. and Johansson G. Mikrochim. Acta 1985 3 209. Hartenstein S. D. REiEka J. and Christian G. D. Anal. Chem. 1985 57 21. Hirata S. Umezaki Y. and Ikeda M. Bunseki Kagaku 1986 35 106. Caroli S. Alimonti A. Petrucci F. and Horvath Z. Anal. Chim. Acta 1991 248 241. Purohit R. and Devi S. Analyst 1991 116 825. Porta V. Sarzanini C. Abollino O. Mentasti E. and Carlini E. J. Anal. At. Spectrom. 1992 7 19. Porta V. Abollino O. Mentasti E. and Sarzanini C. J. Anal. At. Spectrom. 1991 6 119. Sperling M. Yin X. and Welz B. J. Anal. At. Spectrom. 1991 6 295. Fang Z. Guo T. and Web B. Talanta 1991 38 613. Liu Z.-S. and Huang S.-D. Anal. Chim. Acta 1992 267 31. Welz B. Yin X. and Sperling M. Anal. Chim. Acta 1992 261 477. Okamoto Y. Murata A. and Kumamaru T. Anal. Sci. 1991 7 879. Adeloju S. B. Bond A. M. and Briggs M. H. Anal. Chem. 1984 56 2397. Burger K. Organic Reagents in Metal Analysis Pergamon Press Oxford 1973 pp. 118-124. Elmahadi H. A. M. and Greenway G. M. J. Anal. At. Spectrom. 1991 6 643. Welz B. Xu S. and Sperling M. Appl. Spectrosc. 1991 45 1433. Fang Z. and Dong L. J. Anal. At. Spectrom. 1992 7 439. Paper 3/05 704 F Received September 21 1993 Accepted February 14 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900691

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 697 Pulse Optimization Criteria for the Microcavity Hollow Cathode Discharge Emission Source Paul D. Mixon and Steven T. Griffin Department of Electrical Engineering Memphis State University Memphis TN 38 152 USA J. C. Williams Jr. Department of Anatomy Indiana University Medical Center Indianapolis IN 46202-5 120 USA Xiangjun J. Cai and J. C. Williams* Department of Chemistry Memphis State University Memphis TN 381 52 USA Criteria are presented that are important for optimizing the emission intensity in a microcavity hollow cathode (MCHC) discharge source operated in the pulsed mode. The d.c. component must be sufficiently large to ensure a continuous discharge inside the HC. It is shown that the pulse parameters that result in maximum light output for the spectral lines of the cathode material do not result in maximum light output from sodium as an analytical sample which has been mounted inside the HC.This lack of correlation may be due to either the physical and chemical properties of the species or to the manner in which the species enters the HC discharge plasma. Voltage-current characteristics are presented for the MCHC to illustrate the role of the d.c. component during pulsed mode operation. The d.c. current of the discharge must be >5 mA in order to maintain a discharge in the HC. At currents <5 mA for this system the discharge is restricted to the planar surface of the cathode and will not participate in the excitation of analytes thrown into the hollow space during the pulse.Keywords Glow discharge; hollow cathode discharge; atomic emission spectrometry The hollow cathode discharge (HCD) has been used extensively as a line source for atomic absorption spectrometry (AAS) atomic fluorescence spectroscopy (AFS) and as a radiation source for atomic emission spectrometry (AES). Djulgerova' has reviewed the application of the pulsed HCD to AAS AFS and AES. The HCD emission intensity is reported to be increased over that achieved in the d.c. mode by up to two orders of magnitude and instability and reproducibility prob- lems encountered in the analysis of dry residues from liquid samples are greatly reduced by operating the source in the pulse The idea of the pulsed HCD is to use short pulses (a few p) of high current (a few hundred mA) to spatially separate the vaporization and excitation.A few ps of diffusion time are needed for the analyte atoms in the HC to reach the negative The discharge current is reduced to several mA after the sputtered atoms have reached the negative glow to avoid possible discharge instability and self-reversal.6 Most of the work in this area has been directed toward improving the HCD as a source for AAS and AFS where narrow lines are an absolute requirement. Although line widths are less critical for AES sufficient self-absorption will affect the linearity of calibration curves. The Russian group,4 as reviewed by Djulgerova,' has investigated the analysis of dry residues in the HCD using pulses 0.02-2 A 5-1000 ps wide at frequencies of 0.02-10 kHz that were superimposed on a d.c.current of a few mA. The combined mode of d.c. and pulse discharge gave 10-100 times better detection limits than the d.c. discharge a10ne.~ Normally the high-current pulse is superimposed on a d.c. or pilot current that maintains a continuous discharge inside the HC. There are several parameters associated with pulsed mode operation that directly affect emission intensity. These include pulse amplitude Ip pulse width t pulse frequency f and the pilot (d.c.) component Id.c.. A widely used optimization technique for AAS and AFS applications maximizes the inten- sity gain ratio I / l o as a function of these four parameter^.^ In this technique I is the intensity of the spectral lines of the cathode material under pulsed excitation and lo is the intensity for the case of d.c.excitation at the same average value of current. The four pulse parameters (Ip Id,c. t andf) are selected and the emission intensity for these pulses is recorded over a specific period of time. The d.c. current is set to the average value of the pulsed excitation. Then the intensity is recorded over the same period of time for the case of d.c. excitation. The intensity gain ratio l / I o is calculated from the results. The pulse parameters are adjusted and the procedure repeated until an optimum is reached. Using the above procedure intensity gains of 50-800 have been reported for Ca Co Cu Pb Mg and Mn by applying pulsed mode operation to commercial lamps in absorption ~tudies.~ Djulgerova has reported intensity gains of 55 for Al,*?' 400 for Cu and 100 for Fe.l0 A number of elements were studied by Cordos and Malmstadt," who reported intensity gains of 40-200 when using pulsed excitation for AFS.An empirical formula for the ratio I / I o has been developed by Katskov et a1.I2 Good agreement was reported between pre- dicted and measured values of gain up to 70." Intensity gains over the d.c. mode of an order of magnitude have been reported in studies using a hot HCD source in pulsed mode.13 The I / I optimization technique has been used primarily to maximize the emission signal from HC lamps that are to be used in absorption and fluorescence techniques. It has been suggested that optimizing pulse parameters for maximum intensity gain ( I / I o ) may not yield the best conditions for analytical performance in the HCD as an emission s o ~ r c e .~ It has also been noted that physical or chemical properties of the analyte will cause unique optima for some ana1ytes.l' This is particularly the case when using the HCD as an emission source for AES where the analytical sample is mounted directly inside the HC. For some analytes such as the alkali metals the pulse parameters that result in maximum light output for the spectral lines of the cathode material may not maximize light output from the analytical sample. Provided that precision and sensitivity are maintained the maximum absolute emission intensity is desired in AES. This may or may not correspond to maximizing I/I@ Some aspects and character- istics of a particular HCD system can be studied by observing the emission intensity of the hollow material of the HC itself.This is much easier than observing the transient signals from samples deposited in the hollow space. However optimum pulse parameters must at least be verified for each analyte. The main thrust of the research in this laboratory is to698 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 develop a suitable method for the determination of several elements (Nay C1 K Cay Mg and P) of physiological interest in 1 nl of renal fluid. Suitable detection limits for most of these elements have been demonstrated here using the HCD operated in the d.c. rnode.l4 However the improved stability and precision promised by pulsed-mode operation is very attractive. Thus the behaviour of Na in a pulsed discharge is of primary interest and must be investigated in spite of difficulties caused by environmental contamination. Fortunately highly con- trolled sample handling techniques developed for handling small samples of ubiquitous elements allows competent deter- mination of such elements.15 Even though the concentration of the components of renal fluid is relatively high the nanolitre volumes usually available make the determinations very challenging.Experimental The instrumentation used in the pulsed HCD atomic emission experiments is shown schematically in Fig. 1. Zero grade argon was used in the experimental work described here. The gas pressure was monitored using a Veeco model TG-270 pressure gauge. The demountable HCD source used in these experi- ments has been described in detail pre~ious1y.l~~~"~~ Br iefly a water-cooled HC emission source was mounted on an optical rail.The light from the source was collimated by a lens and then directed along the optical rail to a neutral beam splitter that divided the light between two channels. Each channel had a monochromator one at the end of the main rail and the other mounted at a right angle to the main rail. A lens before each monochromator was used to focus an image of the HC on the slit. Each monochromator employed a Hamamatsu R928 photomultiplier tube (PMT) mounted just outside the exit slit. Appropriate wavelength settings were selected and data were collected in two channels simultaneously. The output of each of the PMTs was fed into a multichannel integrator box before being sent to an analogue-to-digital converter for processing by a computer.The integrator box included six integrator cards (Evans 41 30A) and three dual amplifier cards (Evans 4163A) under computer control through Metrabyte PIO-12 and Dash-16 boards. The electronic circuits used to drive the HCD in pulsed mode and d.c. mode have been described in detail previo~sly.l~*~~ For the work described here stainless-steel (type 304) HCs having dimensions 1.5 mm diameter by 5 mm depth were used. These dimensions were found to maximize the emission signal during previous studies using A1 and Cu e l e c t r o d e ~ . ~ ~ ' ~ ~ After sample deposition the cathode was dried under an infrared lamp or in a vacuum chamber heated to ~ 5 9 5 "C placed in the HC source chamber and discharged.In the case of microsamples the emission signal is transient. It builds up HCD To second channel enclosure PMT - I Monochromator Power supply I electronics Trigger I Multichannel gated integrator/amplifier Digitizing oscilloscope Fig. 1 intensity from the HCD Block diagram of the instrumentation used to observe emission rapidly and decreases quickly as the small sample dissipates. Most of the sample is sputtered from the hollow space in a few seconds producing the analytical signal. The remainder of the discharge period is used to clean the electrode for use with the next sample. Lapsed time between samples is 3-5 min. The argon passing to the HCD source was filtered to remove particulates and samples were prepared in a 'clean bench' by NUAIRE Model NU-201-S24 to reduce contamination.20 A 1000 point histogram of this transient emission signal was made by integrating the emission signal for 6-1 5 ms.The pulse width of 15-30 ps was much shorter than the integration periods used. Thus the emission signal collected during the integration period consisted of the signal from 30 to 75 pulses plus that from the d.c. pilot current during that time. Plastic [either polypropylene poly(tetrafluoroethy1ene) or polyethylene] containers were employed for the preparation and storage of all test solutions as recommended by Moody and Lindstrom.'" In order to minimize sample contamination the sample electrodes were thoroughly cleaned by HCD sput- tering handled only with stainless-steel tweezers and trans- ported in an enclosed container.Calibrated nanolitre-range pipettes are not available com- mercially and reproducible deposition of such small samples without contamination is difficult. Pipettes were fabricated according to the accepted practice developed by Bonventre et aZ.22 Williams and S ~ h a f e r ~ ~ give a more recent detailed description of the preparation and use of pipettes for handling nanolitre volumes. Because the volumes used in the electron probe method are less than 100~1 the technique used for handling samples is clearly adequate for the HCD with the present renal fluid samples. For delivering the sample to HC electrodes volumetric pipettes calibrated for 1- 10 nl were used. The pipettes weire fashioned on a microforge using capillary glass and consist of a chamber at the end of the capillary that lies between an opening in the drawn-out tip and a constriction a few hundred inicrometres from the tip. The opening at the tip and the narrowest part of the constriction are both about 20 pm in diameter.The pipette is silanized to reduce wetting of the glass surface by aqueous solutions and improve handling. The accepted procedure23 has been adapted to the HCD and successfully used in this laboratory for several years.14,16-18,24 Samples are delivered to the HC which is mounted on the stage of a low-power stereomicroscope using a micromanipulator upon which a calibrated glass micropip- ette is mounted. The pipette tip is bent at nearly 90" so that the sample can be deposited on the bottom of the HC.The sample is drawn up by syringe suction until its meniscus reaches the centre of the constriction and is immediately deposited on the HC bottom. In the recommended procedure for microsamples a small volume of oil is pulled into the tip after the sample to prevent evaporation following exposure of the pipette to air. However oil has not been used here as the small sample taken from a bulk standard solution is easily transferred from the silanized pipette tip to the more hydro- philic metal surface without loss of analyte. The volume of each pipette was determined by the use of a calibrated solution of a radioactive solute. Three to five samples of this solution were taken with each pipette and ejected into liquid scintillation counting fluid and counted to determine pipette volume.The coefficient of variation of the measure- ments on a pipette had to be less than 1% for a pipette to be considered acceptable for use. The measurement of the volume delivered by a pipette is not crucial to the technique as samples and standards were handled with the same pipette for a given experiment. The precision of sample handling is far more important and the calibration method provides a way to select only those pipettes that handle samples in a reproducible manner. Results and Discussion The results of intensity gain ( I / I o ) measurements for two major constituents of a 304 stainless-steel HC (Fe at 371.99 nm andJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 699 Ni at 341.48 nm) are shown in Fig. 2 The gain was measured for values of peak pulse current (I,) at 50 80 and 120 mA.The other pulse parameters were held constant at f= 5.1 kHz t = 15 ps and I,,.=6 mA. The pulse parameters used for these measurements do not represent an attempt at optimization; however the parameter values used are within the range of previously reported optimum val~es.~-ll The argon pressure was held constant at 5 Torr (1 Torr = 133.322 Pa). The results in Fig. 2 show that I / I increased monotonically for each constituent as the current was increased from 50 to 80 to 120 mA. Fig. 3 shows normalized emission intensities from Li (670.8 nm) Na (589.6 nm) K (766.5 nm) Fe (371.99 nm) and Ni (341.48 nm) as a function of current. A solution containing 100 pg each of Na Li and K in 2.5 nl was prepared from doubly de-ionized water (Millipore de-ionized H20 18 Mi2 cm resistivity) and Spex plasma emission standards. The data were normalized after subtracting the water blank.A 2.5 nl pipette was used to deposit an aliquot of solution on the bottom of the HC. The stainless-steel cathodes were conditioned in three steps using 6 kHz pulsed discharges of 100 mA pulses that were superimposed on a 7mA d.c. pilot current. First a discharge for 2.5 h with 60 ps pulses (47 mA average current) was used to produce a nearly spherical shape in the HC. Next the HC was conditioned for 1.5 h using a discharge of 30 ps pulses (25 mA average current). Finally the HC was discharged for 10 min using the exact parameters (15 ys pulses in this case) planned for the analysis.After sample deposition the cathodes were placed under vacuum for 2min in the HCD enclosure before discharging. The emission intensity from the alkali metals is relatively constant compared with that from 4.0 3.6 3.2 2.8 2.4 2.0' ' I I I I I I I I 50 60 70 80 90 100 110 120 Peak pulse current/mA Fig.2 Emission intensity ratio for A Fe (371.99 nm) and B Ni (341.48 nm) as a function of discharge current. The stainless-steel HC used was 1.5mm in diameter by 5mm deep. The RSD for three replicates was less than 1% for all points shown. Pulse parameters f 5.1 kHz; t 15 ps; and Io 6 mA 1 .o > 0.9 .I- .- 2 0.8 .'= 0.7 u .- 2 0.6 E 0.5 2 0.4 0.3 .I- - m I 40 60 80 100 120 140 160 180 200 Pulse current/mA Fig. 3 Normalized emission intensities from a 1.5 x 5 mm stainless- steel HC for A Fe (371.99 nm); €3 Ni (341.48 nm); C Na (589.6 nm); D Li (670.8 nm); and E K (769.8 nm).Pulse parameters:f=6.0 kHz; t 15 ps; and Id.c. 7 mA. The emission signals come from either solution residue deposited on the bottom of the HC (alkali metals) or the stainless-steel HC itself (Fe and Ni). The RSD for three replicates was less than 1 % for all points shown the transition metals of the cathode itself. The wider range of emission intensities from Fe and Ni tend to compress the alkali data on the plot thus minimizing the peak. However there is a distinct peak for each alkali metal emission intensity. This is consistent with previously reported results for Na Li K and Ca in which all showed an intensity maximum at a d.c. discharge current of 75 mA.I4 The emission intensity from the alkali metals shows little correlation with that from cathode components.This lack of correlation may be due to fundamental differences between alkali metals and transition metals. It could also be due in part to the manner in which the species of interest enters the HCD plasma. In both cases the sample is sputtered into the plasma. However sputtering rates may differ at different locations in the hollow space and sputtering rates for samples deposited on the HC may differ from the sputtering rates of the cathode material that is present uniformly in the hollow space. Thus excitation conditions sputtering rates vaporization temperatures and other parameters will differ for the two cases. Voltage-Current Characteristic for the Microcavity Hollow Cathode The term microcavity hollow cathode (MCHC) is used to describe electrodes having hollow diameters of 2 mm or less.25 The discharge parameters used for analysis in this laboratory place the MCHC discharge well to the right of the minimum point on the Paschen curve.The result is that the discharge initially breaks down over the shortest possible path. When performing analysis with the MCHC an insulating quartz disc is placed over the mouth of the HC. This quartz disc has an opening which is larger than the diameter of the hollow space. As a result there is a front planar surface present which results in a 'lip' effect. The presence of this front planar surface has been found to affect the evolution of the discharge.26 It is important to determine the effect of this front planar surface in terms of the evolution from planar discharge mode to HCD mode in the MCHC.An experiment was conducted to measure the I/-i characteristic for the MCHC over the range of current from 1 to 20mA. Fig. 4 shows a two-piece MCHC which was constructed of two different materials. The top section was made of 304 stainless steel which is identical to the cathodes used for the data shown in Figs. 2 and 3. The bottom section was composed of spectrally pure aluminium (99.9995%) and was designed to snap into the top section to form a one-piece electrode. The dimensions of the hollow were 1.5mm in diameter by 5mm deep. Fig. 5 shows the results of measuring the run voltage and emission intensity versus discharge (d.c.) current for the two- piece cathode.The solid line represents the voltage across the discharge and the broken line represents the emission intensity 6.35 mm 1.5 mm T 1 T x m Fig. 4 Drawing of a two-part HC. The bottom is A1 while the upper part is 304 stainless steel700 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 330 t A 320 310 300 290 - - 2 - 4- - 0 - 3 (r 4 9000 2 6 - 7000 u) 0 - 5000 .z 3000 - B b) c 280 0 2 4 6 8 10 12 14 16 18 20 Current/mA Fig. 5 Breakdown voltage as a function of discharge current for the two-part cathode shown in Fig.4. The solid line represents voltage across the discharge and the broken line represents the emission intensity of the Al 396.2 nm line of the A1 line at 396.2 nm. As is evident from the figure at low values of current (0-4 mA) both the voltage and the emission signal showed a steady increase with increasing current.The abrupt drop in run voltage that occurs at about 5 mA marks the point at which the discharge transfers from the planar face of the cathode down into the hollow. This is evidenced by the sharp rise in the emission intensity of the A1 line. Similar behaviour has been noted by White27 in a study of spherical cavity hollow cathodes. The sharp decrease in voltage is due to a combination of two separate effects. The increase in cathode surface area that occurs when the discharge transfers to the inside of the hollow results in a decrease in the required discharge voltage. In addition the on-set of the HC effect contributes to a decrease in discharge voltage.28 It is interesting to note that the discharge in this MCHC exhibits a slight but well-defined negative resistance characteristic over most of the range of current from 6 to 20 mA. The V-i characteristic for the MCHC yields valuable infor- mation with regard to the optimization of pulse parameters particularly the pilot current (Id.c.). The primary purpose of pulsed excitation is to separate temporally the processes of atomization and excitation in the discharge.When the pulse parameters are set correctly the relatively large pulse current does the job of atomization by sputtering the proper amount of atoms into the hollow space. The appearance of a large population of ions results in an increased sputtering rate. When the atomic cloud in the hollow space reaches the optimum density the high current is switched off and excitation continues as the pilot current excites the cloud.Thus the function of the pilot current is for both sample excitation and ionization maintenance. From Fig. 5 it can be seen that the sharp decrease in voltage across the discharge is accompanied by the on-set of emission from the aluminium bottom of the hollow. Apparently the pilot current for this particular MCHC geometry must be greater than 5 mA for the discharge to enter the hollow space. For values of pilot current less than approxi- mately 5 mA the discharge is planar and is restricted to the uninsulated flat cathode surface around the mouth of the hollow. If the d.c. component does indeed contribute signifi- cantly to analyte excitation Id.c must be sufficiently large to ensure a continuous discharge within the hollow space.Conclusion Analytically optimal pulse parameters for the alkali metals are clearly different from those that maximize light output when using the cathode metal as the analytical sample. The HCD plasma is known to interact non-homogeneously with the HC s ~ r f a c e . ~ ~ * ~ * Thus an analyte deposited entirely on the bottom of the HC may be subjected to very different sputtering conditions from the remainder of the HC surface. The alkali metals have a much lower ionization potential than Fe and Ni and also differ markedly from these constituents in terms of melting-point thermal conductivity and other factors which influence sputtering rates. A broader study is needed to deter- mine the optimum pulse parameters for analysis of samples deposited from solutions in the MCHC and to determine if the effect seen for alkali metals is generally observed.The d.c. current must be > 5 mA for this system in order for a discharge to be maintained in the hollow space between pulses. At currents < 5 mA the discharge is restricted to the planar surface of the cathode and will not participate in excitation of analyte that was deposited on the bottom of the hollow. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 References Djulgerova R. B. in Improved Hollow Cathode Lamps for Atomic Spectroscopy ed. Caroli S. Ellis Horwood Chichester 1985 pp. 52-73. Borkowska-Burnecka J. and Zyrnicki W. Spectrosc. Lett. 1987 20 795. Maksimov D.E. and Rudnevskii N. K. Zh. Prikl. Spektrosk. 1983 39 5. Rudnevskii N. K. Pichugin N. G. and Maksimov D. E. Zh. Prikl. Spektrosk 1975 25 921. Piepmeier E. H. and deGalan L. Spectrochim. Acta Part B 1975 30 263. Pichugin N. G. Rudnevskii N. K. and Maksimov D. E. Zh. Anal. Khim. 1977 32 12. Dawson J. B. and Ellis D. J. Spectrochim. Acta Part A 1967 23 565. Djulgerova R. B. Bulg. J. Phys. 1977 5 569. Djulgerova R. B. Bulg. J. Phys. 1980 7 91. Djulgerova R. B. Bulg. J. Phys. 1977 4 459. Cordos E. and Malmstadt H. V. Anal. Chem. 1973 45 27. Katskov D. A. Lebedev G. G. and L'Vov B. V. Zh. Prikl. Spektrosk. 1969 10 215. Rudnevsky N. K. Maksimov D. E. and Pichugin N. G. Zh. Prikl. Spektrosk. 1973 19 5. Ryu J. Y. Davis R. L. Williams J. C. and Williams J. C. Jr. Appl. Specrosc. 1988 42 1379. Tolg G. Talanta 1972 19 1489. Williams J. C. McDonald J. T. and Davis R. L. Anal. Instrum. 1987 16 241. Chen F. Y. and Williams J. C. Anal. Chem. 1990 62 489. Tseng J. L. Williams J. C. Bartlow R. B. Griffin S. T. and Williams J. C. Jr. Anal. Chem. 1991 63 1933. Mixon P. D. Bray C. W. Griffin S. T. and Williams J. C. Proc. IEEE Southeastcon 1992 2 586. McDonald J. T. Williams J. C. and Williams J. C. Jr. Appl. Spectrosc. 1989 43 697. Moody J. R. Lindstrom R. M. Anal. Chem. 1977 49 2264. Bonventre J. V. Blouch K. and Lechene C. in X-Ray Microscopy in Biology ed. Mazat M. A. University Park Press Baltimore 1981. Williams J. C. and Schafer J. A. Meth. Enzymol. 1990 191 232. Tseng J. L. Kung J. Y. Williams J. C. and Griffin S. T. Anal. Chem. 1992 64 1831. Czakow J. in Improved Hollow Cathode Lumps for Atomic Spectroscopy ed. Caroli S. Ellis Horwood Chichester 1985 p. 36. Mixon Paul Ph.D. Thesis Memphis State University Memphis 1993. White A. D. J. Appl. Phys. 1959 30 711. Slevin P. J. and Harrison W. W. Appl. Spec. Ret.. 1975 10 201. Puper 3104899C Received August 12 1993 Accepted March 9 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900697

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 701 Simulation of Nebulization Process in Inductively Coupled Plasma Atomic Emission Spectrometry with a Modified Model Using the Monte Carlo Technique* Hu Yanping and Zhang Zhanxiat Department of Chemistry Zhongshan University Guangzhou 5 70275 China Zheng Jianguo Guangdong Commodity Inspection Bureau Guangzhou China An improved Monte Carlo program based on a jet model for the simulation of various particle loss processes that occur in the spray chamber is presented. This program is evaluated by comparing the simulated mass transport efficiency and total mass transport rate with those found by experiment. The aerosol flow velocity distribution as a function of axial and radial distances of the chamber is outlined. The effect of carrier gas flow rate on the distribution of loss particles along the chamber due to various loss processes is studied.The results indicate that the predominant loss mechanism depends strongly on the magnitude of the carrier gas flow rate. Keywords Nebulization process; inductively coupled plasma atomic emission spectrometry; Monte Carlo simulation Generally the liquid samples are converted into aerosols before introduction into the inductively coupled plasma (ICP) through a pneumatic nebulization-spray chamber system. The solution is split into droplets under the influence of a high- speed gas flow and the aerosol produced is highly polydispers- ive with droplet diameters of up to l o o p . Nukiyama and Tanasawa’ have derived empirically the size of the primary aerosol droplets relative to the operating parameters of the nebulizer and the physical properties of the solution.However the equation is valid only for describing the aerosol formed at the nebulizer. Because of other modifying processes,2 namely gravitational settling impaction turbulence and centrifugation taking place in the spray chamber and in the transport of the aerosol between the nebulizer and the plasma the aerosol will have entirely different properties from the primary aerosol on reaching the plasma. These processes generally act to shift the aerosol size distribution to smaller droplets. The spray chamber can be regarded as a hypothetical filter having a cut-off diameter d through which the aerosol is passed i.e. droplets having diameters larger than d are retained and drained to waste while those droplets having diameters of d or less pass through unaffected.Different mechanisms have been suggested to describe the primary separation process in an aerosol chamber. Skogerboe and Olson3 studied gravitational settling and inertial deposition and concluded that gravitational set- tling imposed the primary limitation on aerosol transport for the chamber and condition used. Browner et a1.’ studied gravitational settling impaction turbulence and centrifugal loss and proposed that turbulence-induced loss is the predomi- nant mechanism for the separation of large droplets in an ICP aerosol chamber of the dual concentric type. Gustavsson4 considered that the most likely process causing separation of large droplets in an aerosol chamber is inertial dep~sition.~ Sharp5g6 summarized from the experimental data that for a fixed nebulizer geometry the liquid-to-gas ratio is the param- eter that most determines the quality of the aerosol produced.He further confirmed that the nature of the processes determin- ing droplet deposition rates is particle size dependent and that the inertial and turbulence losses are important whereas the gravitational and centrifugal losses are not important. The aerosol drop size distribution can give important infor- f To whom correspondence should be addressed. * Presented at the XXVIII Colloquium Spectroscopicum Internationale (CSI) York UK June 29-July 4 1993. mation on the sizes of the droplets entering the plasma prior to evaporation decomposition and atomization.The mass transport efficiency defined as the percentage of the mass of nebulized solution that actually reaches the plasma and the total analyte mass transport rate defined as the total mass of analyte reaching the plasma per second are important param- eters in the quantitative characterization of critical aerosol properties as well as in optimizing the experimental param- eters. A further study using a more powerful method is thus essential to understanding all the processes occurring in the spray chamber and to determining the predominant processes responsible for the loss of aerosol droplets. In a previous paper,’ the Monte Carlo technique was used for the first time to simulate nebulization processes in ICP atomic emission spectrometry (ICP-AES).The results were encouraging. The study was based on the fact that the drop size distribution is Gaussian and that the aerosol droplet loss due to gravitational settling impaction turbulence and centri- fugation are random in character. The operating conditions and the effect of the design of the nebulizer on the mass transport efficiency the analyte mass transport rate and the aerosol drop size distribution were obtained in a straight- forward manner since the Monte Carlo technique is especially useful for complex situations and has the ability to obtain information which cannot be extracted experimentally. However the model used previously was too simple. In this work a modified model namely the jet model is proposed with the aim of obtaining more realistic information of the nebulization process and discussing the loss mechanism taking place in the spray chamber.Theory A Monte Carlo simulation is based on the principle that any complex process can be broken down into a series of simpler independent events each represented by a probability distri- bution. Thus the various processes occurring within the spray chamber that generally act to shift the aerosol drop size distribution to smaller droplets can be considered indepen- dently when the time period being viewed is sufficiently small. According to Browner et aZ.? the gravitational turbulence and centrifugal loss processes responsible for the separation of large droplets in the spray chamber can be expressed as follows. (i) Impaction loss dci= 1.5 q g W P ~ V ( 1 )702 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 where dCi is the cut-off diameter (pm) of the aerosol droplet due to impaction loss q is the absolute gas viscosity ( P ) W is the width of the gas stream (cm) pa is the aerosol particle density (g crnp3) and V is the particle velocity (cm s-'). (ii) Gravitational loss 4 = 3 (?,D/Pa@)+ (2) where dcg is the cut-off diameter (pm) of the aerosol droplet due to gravitational settling D is the chamber diameter (cm) g is the gravitational constant ( 10-3N m2 kgV2) and r is the time spent in the chamber. (iii) Turbulence loss dCt = ( 3 Q / W * ( 3 ) where d is the cut-off diameter (pm) of the aerosol droplet due to turbulence loss K is a constant whose value is deter- mined by the degree of turbulence of the system Lis the tube length and Q is the total gas volumetric flow rate.(iv) Centrifugal loss (4) where d is the cut-off diameter of the aerosol droplet due to the centrifugal loss M. is the inlet width (cm) n is the number of rotations and aerosol path taken (dimensionless) Vi is the inlet gas velocity and ps is the gas density (g ~ m - ~ ) . As the various interacting forces between the gas flow the aerosol stream and spray chamber are examined the variation of the aerosol flow velocity as the aerosol passes the chamber to the plasma should not be ignored. Since the Monte Carlo technique yields information on a 'model system' more useful and accurate results will be available if a more realistic model is assigned to the system.A jet model8 is thus proposed. A schematic diagram of the model for a round jet nozzle is shown in Fig. 1. The aerosol flow area is divided into moving and diffusion parts. Inside the former are the core zone where the aerosol flow velocity is equal and the mixed zone where the aerosol velocity attenuates and finally goes to zero at the boundary. The length of the core zone is determined by the shape of the nebulizer nozzle and is approximately equal to five times the diameter of the round jet nozzle (W,). In the diffusion area the aerosol velocity decreases as it moves down along the central part of the chamber and finally goes to zero at the boundary with increasing aerosol diffusion angle 2 is the axial distance and R is the radial distance (cm); Vo is the aerosol velocity (cm s - l ) at the jet nozzle i.e.at Z=O and V is the aerosol particle velocity at the 2 and R coordinate axes. Thus within the moving part V= V for the core zone and ZoR-W,(Zo-Z)/2 "[' -( 1.34W,Z Moving area * Diffusion area I I_ Fig. 1 Schematic diagram of the jet model for a round jet nozzle W width of jet nozzle; 2 distance of moving area; Z axial direction and R radial direction for the mixed zone. Within the diffusion part Assuming that the diameter of the jet nozzle is equal to the inner diameter of the carrier gas outlet and that the jet primary velocity of the liquid sample leaving the nozzle is equal to the outlet velocity of the carrier gas then Q Vo %+ (7) AJ3 where Q is the carrier gas flow rate (mls-') and A is the cross-sectional area of the gas outlet.According to this model the velocity of the particles at any position of the spray chamber can be calculated. By inserting the calculated velocity into eqn. ( l ) the cut-off diameter of the aerosol droplet due to impaction at any vectorially displaced position can be found. The cut-off diameter of the droplet for gravitational settling turbulence and centrifugal loss can also be found from their corresponding equations. It is important to note that the proposed jet model is based on a gross assumption; strictly speaking it is not appropriate for the enclosed jet within the spray chamber. However the results obtained indicate that the Monte Carlo simulation based on this jet model is valid. To perform the simulation it is necessary to know the distribution of the primary aerosol droplet size.According to Nukiyama and Tanasawa,' the primary aerosol droplet diam- eter is expressed as follows 585 (s)'.' [ q l [ 1:Q,]'-5 +597 - (8) D- - ,--I/ p (sp)O.' where D is the Sauter median diameter (pm) V is the velocity difference between gas and liquid flows to the nebulizer (m s- ') s is the surface tension of the liquid ( 1OP3N m-') p is the liquid density (gcmd3) ql is the liquid viscosity ( P ) and Q and Q are the volume flow rate of liquid and gas (crnp3 s)' respectively. The Sauter median diameter is defined as where D is the droplet diameter and n(D) is the number of droplets of diameter D and Do and D are the lower and upper limits of the distribution respectively. Generally Do is assumed to be zero. Thus after obtaining Ds from eqn.(8) the DM value can be calculated from eqn. (9). The primary aerosol droplet size distribution [O,D,] which is Gaussian can thus be obtained. In this work the parameters of three different concentric nebulizers with Scott-type spray chambers are used for simu- lation. The time interval for each simulation is 0.01 ms. For every Ar each particle is monitored for a vectorial displace- ment. Since the chamber is axially symmetrical the movement of the aerosol particle in each of the rectilinear coordinates is given by (11) where d and d are the axial and radial movement (cm) respectively; R and Rg2. are the random Gaussian number distributed about zero with a standard deviation of 1.0; V is the aerosol flow velocity (cm s-'); and D T is the temperature dependent diffusion coefficient (DT = D0(T/T0)3/2 where Do is the diffusion coefficient at To=273 K).The vectorial sum of these motions produces the new location of the particle. For every Af to determine whether the particle is lost by gravitational settling centrifugation or turbulence the normal random number R within the interval of 0 and D is compared dZ = Rgl (2D-&)+ + I/& dR = Rg2 (2D,At)* (12)JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 703 with the corresponding cut-off diameter of the droplet due to the processes mentioned above. If R,3dc the particle is lost and drained to waste. If R,<dc the particle will then remain in the gas. From the cumulative movement it is possible to determine whether the particle collides with the chamber wall or with any other objects such as the paddle in the chamber.If impaction occurs the particle velocity can be calculated from eqns. (4) (5) and ( 6 ) depending on the position of the particle. Thence the cut-off diameter of the droplet dci due to the impaction loss can be obtained and compared with R,. Similarly if R 3 dci the particle is lost whereas if R < dci the particle will remain in the gas. When the movement Edz is greater than the chamber length the simulation is complete. Through the statistical cumulation the final particle loss number N1 and the particle number N2 that remain in the gas can be obtained. The mass transport efficiency E can be calculated from the folowing equation N N x 100% = A x 100% (13) N2 Nl+N2 En=- where N is the total particle number for simulation i.e.200 000 in this work. Hence the total analyte mass transport rate Kot can be derived from E provided that the Q1 is specified since Wot =~plQiC/100 where C is the analyte concentration (pg ml-I). r-1-1 Read in parameters Experimental The Monte Carlo simulation flow chart is depicted in Scheme 1. The program is written in FORTRAN 77 and is run on an M340 computer. It has a memory of 12 megabytes the hard disk has a memory of 4300 megabytes and the CPU compu- tation velocity is 2 500000 times s-'. To evaluate the validity of the program the simulation results are compared with those obtained experimentally by Browner et a1.' The dimensions of the nebulizer nozzles provided by Canals et aE." are used in the simulation work.They are listed in Table 1. The analyte solution used is 1000 pg ml-I Mn(NO,),. Results and Discussion Selection of Simulation Particle Number Generally the estimated value approaches the true value as the simulation particle number goes to infinity. However this would be too time consuming and is not realistic. Fig. 2 elucidates the relationship between the simulation particle number and the total mass transport rate. From the figure it can be seen that the total mass transport rate Wot becomes stable as the simulation particle number increases to 100000. In this work 200000 is selected as the particle number in order to guarantee higher precision. The total time required generated aerosol particles L v I Turbulence loss? increment + numberof loss particle t N,+N22N? Y I I N N I Increment of diffusion move (dZ dR)? N Increment number of particles to plasma N2 Parameters change? N End Scheme 1704 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 Table 1 Dimensions of the nebulizer nozzles" Concentric nebulizer Parameter Inner diameter of the inner sample tube IJmm Outer diameter of the inner sample tube I,/mm Inner diameter of the outer gas tube IJmm Cross-section area of the gas outlet A,/mm2 Cross-section area of the liquid outlet AJmm' Recess of the inner tuber IJmm 1 0.318 0.510 0.580 0.060 0.079 -I- 0.07 - 2 0.508 0.700 0.730 0.034 0.203 '0.06 - 3 0.424 0.538 0.574 0.03 1 0.141 0.06 " 0.12 I I 0) 1. -. 5 0.11 : 0.10 g 0.09 r F CI " t E I I I I 1 10 20 30 40 50 Particle number x 10' Fig.2 Relationship between simulation particle number and total mass transport rate to obtain a simulation is in the range 8-25 min depending on the carrier gas flow rate used. Evaluation of the Validity of the Monte Carlo Program To evaluate the validity of our Monte Carlo program the following simulations were performed using the dimensions of the nebulizer nozzle listed in Table 1. Influence of Q on E and W, Simulation data of E and W, are obtained by increasing Q1 from 0.63 to 1.90 ml min-' while keeping the nebulizer gas flow rate Q constant at 0.65 1 min-'. The data are compared with the experimental data of Browner et aL9 and presented in Table 2. The results show that on the whole the simulation data agree quite well with those given in ref. 9. Influence of Q on WIot The value of Q was increased from 0.39 to 1.10ml min-' while keeping QI constant at 0.63 ml min- for three different concentric nebulizers (Nos.1-3). The simulation Wot data obtained are compared with the experimental results obtained by Browner et aL9 in Table 3. These results show that the simulated data correspond well with Browner's experimental results especially for lower values of Q,. Some discrepancy between them is observed for high values of Q but the trend that Fot increases with an increase of carrier gas flow rate is the same. Therefore it can be concluded that the model proposed is reliable and the program is valid. This program can be used to evaluate the performance of the nebulizer as well as to optimize experimental conditions.Aerosol Droplet Loss Mechanism Aerosol frow velocity distribution along the spray chamber In the jet model depicted in Fig. 1 it has been assumed that the aerosol flow velocity varies in the Z axial and R radial directions as the aerosol particles move along the spray chamber to the plasma torch. To illustrate this variation the following simulations are performed. The dimension of the No.1 nebulizer nozzle is taken into consideration. The values of Q and QI are kept at 0.65 and 0.63 ml min-' respectively. The results obtained by increasing the Z axial distance at R = 0 are presented in Table 4. The data indicate that as 2 increases from 0.1 to 14.5 cm the aerosol flow velocity decreases signifi- cantly from 180.6 to 3.6m s-'. The velocity decreases nearly 50% as Z increases from 0.5 to 1.0 cm and from 1.0 to 2.0 cm.To study the variation of the flow velocity along the radial distance simulations of the variation of the aerosol flow as a function of radial distance at various values of Z are performed. The results are presented in Table 5. The results show that generally the aerosol flow velocity V decreases with an increase of radial distance R at any axial distance 2; but varies significantly with R near the nebulizer nozzle. The flow velocity decreases almost 90% as the aerosol moves radially from 0.05 to 0.10 cm at Z=OS cm and from 0.10 to 0.20 cm at Z= 1 cm. However as 2 increases to 5 cm the relationship between V and R is no longer significant. The phenomena described above indicate that the proposed model might be helpful to give an understanding of the various loss processes taking place in the chamber.It can be presumed that it is the significant attenuation of the aerosol flow velocity that causes gravitational settling and turbulence and centrifu- gal and impaction loss processes to occur more easily. Table3 Effect of Q on W, Table 2 Effect of Q on E and W, Nebulizer QJI min-' Exp.* Simt 1 0.63 0.101 0.094 1.90 0.107 0.098 2 0.63 0.116 0.124 1.90 0.129 0.105 3 0.63 0.154 0.158 1.90 0.132 0.114 EXP Sim 0.97 0.90 0.39 0.3 1 1.1 1 1.18 0.4 1 0.33 1.52 1.51 0.43 0.36 Q 1 min 0.39 0.52 0.65 0.84 1.10 No 1. nebulizer ' Exp.* Sim.? 0.020 0.028 0.040 0.051 0.086 0.094 0.150 0.185 0.225 0.255 No 2. nebulizer Exp. Sim. 0.020 0.031 0.049 0.058 0.122 0.410 0.220 0.250 0.340 0.390 No 3.nebulizer Exp. Sim. 0.020 0.027 0.070 0.071 0.154 0.158 0.240 0.278 0.340 0.366 * Exp = experimental data from Browner et aL9 t Sim =simulation data from this work. * Exp = experimental data from Browner9 t Sim = simulation data from this work.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 705 Table 4 Variation of aerosol flow velocity as a function of 2 axial displacement Z/cm 0.1 0.2 0.5 1 .o 2.0 3.0 4.0 6.0 10.0 12.0 14.5 V/m s-' 180.6 180.6 104.7 52.4 26.2 17.5 13.1 8.7 5.3 4.4 3.6 Table5 Variation of aerosol flow rate velocity V as a function of radial distance R at various values of axial distance 2 Z/cm 0.5 1 5 10 Rlcm 0.05 0.10 0.13 0.14 0.05 0.10 0.20 0.25 0.27 0.2 0.4 0.6 0.8 0.2 0.4 0.6 0.8 vlm s-' 24.30 3.29 0.04 0 23.75 22.15 1.65 0.11 0 5.41 3.21 1.81 0.89 3.54 2.71 2.1 1 1.61 Effect of Qg on the Distribution of Loss Particles in the Chamber To determine which of the loss processes predominantly causes the separation of large droplets in the aerosol chamber the simulation of the number of lost aerosol particles Ni due to the various loss processes as a function of Q are performed.The dimensions of the No.1 nebulizer are used. The liquid flow rate Q1 is kept at 0.63ml min-l while the carrier gas flow rate is varied from 0.39 to 1.10 1 min-l. The total particle number Ntot for simulation is 200 000. The percentage of particles lost as a result of any loss process can be calculated from Ni/NtOt x 100. The data obtained are presented in Table 6. It is interesting to note that the percentage of particles lost and their distribution due to the various processes are influ- enced greatly by the magnitude of Q,.For example at lower Q the percentage loss due to turbulence processes is highest (87.65) but it decreases abruptly as Q increases to 0.84 1 min-l. On the other hand the percentage of particles lost due to impaction increases with an increase of Q,. At a Q of Table 6 Effect of Q on the distribution of loss particles Q,/1 min - 0.39 0.52 0.65 0.84 1.10 Q,/1 min-l 0.39 0.52 0.65 0.84 1.10 Centrifugal loss Turbulence loss No. YO 15 520 7.76 1169 0.58 0 0 0 0 0 0 Gravitational loss No. YO 175 290 87.65 164 861 82.43 109 000 54.50 21 693 10.85 785 0.39 Impaction loss No. Yo 2 770 1.39 16 061 8.03 49 062 24.53 75 107 37.55 21 720 10.86 No. % 6 000 3.00 17 909 8.47 41 938 20.07 103 200 49.84 177 495 86.32 1.10 1 min-' the percentage loss reaches 86.32. Gravitational settling also plays a role in the separation of large droplets. At Q values of 0.65 and 0.841 rnin-l the percentage losses due to gravitational settling are 24.53 and 37.55 respectively.The role of centrifugal loss is insignificant. At low Q the percentage loss due to centrifugation is only 7.7 and it goes to zero with an increase of Q,. Evidently the predominanting loss process in the separation of the large droplets in the aerosol chamber depends strongly on the magnitude of Q,. When a lower Q is applied i.e. from 0.39 to 0.52 1 min-l it appears that the turbulence-induced loss is the predominant mechanism whereas when a higher Q is applied e.g.1.1 1 min - ' the impaction-induced loss might predominate. Conclusions A jet model is proposed for calculating the variation of the aerosol flow velocity in a spray chamber. The Monte Carlo program is evaluated by comparing the simulated and exper- imental data of E and KO,. Good agreement between them is obtained. This suggests that the proposed model is reliable. The program can be used to select and evaluate the perform- ance of nebulizers and to optimize the operating conditions for ICP-AES work. The aerosol flow velocity distribution is a function of the axial and radial distances of the chamber. The results indicate that the proposed velocity model might be useful in providing an understanding of the various loss processes that occur in the chamber.The effect of the carrier gas flow rate on the distribution of loss particles due to various loss processes along the chamber has been studied. The results indicate that the predominant loss mechanism depends strongly on the magnitude of Q,. However the results obtained are only preliminary. In order to obtain the relation between signal intensity and aerosol particle behaviour the vaporization atomization and ionization processes in the ICP have been studied and simulated. The results will be reported in a future paper. This work was supported by the National Natural Science Foundation of China. References 1 2 3 4 5 6 7 8 9 10 Nukiyama S. and Tanasawa Y. Trans. SOC. Mech. Eng. (Jpn.) 1939 4 68. Browner R. F. Boorn A. W. and Smith D. D. Anal. Chem. 1982,54 1411. Skogerboe R. K. and Olson K. W. Appl. Spectrosc. 1978,32,181. Gustavason A. Spectrochim. Acta Part B 1984,39 85. Sharp B. L. J. Anal. At. Spectrom. 1988 3 613. Sharp B. L. J. Anal. At. Spectrom. 1988 3 939. Zheng J. Zhang Z. and Qian H. Proceedings of the Fourth International Beijing Conference and Exhibition on Instrumental Analysis C. Spectroscopy 1991 p. 13. Seichi H. and Yotab O. Jet Engineering Science Publishing House China 1977 66. Browner R. F. and Canals A. and Hernandis V. Spectrochim. Acta Part B 1992 47 659. Canals A. Hernandis V. and Browner R. F. Spectrochim. Acta Part B. 1990 45 591. Paper 3103951 J Received July 7 1993 Accepted January 12 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900701

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 707 Halocarbon-assisted Slurry Vaporization in Inductively Coupled Plasma Atomic Emission Spectrometry for the Analysis of Silicon Nitride Powder" Gyula Zaray lmre Varga and Tibor Kantort Institute for Inorganic and Analytical Chemistry L. Eotvos University P. 0. Box 32 H- 15 18 Budapest 7 12 Hungary A conventional Babington-type nebulizer was applied for introduction of samples in the form of solutions and slurries (1% m/m) produced from a finely dispersed silicon nitride (Si3N4) powder (mean particle size 0.54 pm). Calibration with solution standards for the slurry method resulted in 20-40% negative deviations in the results for impurities of Al Fe Ca Mg and Ti compared with the dissolution-based analysis. The addition of Freon-12 (CCI,F,) as a possible halogenation agent to the plasma resulted in decreased negative deviations by factors of between 1.2 and 2.8 which suggests that the degree of evaporation of the slurry particles was increased by halogenation.The fact that the negative deviations could not be completely eliminated could perhaps be explained by a difference in the efficiency of nebulization (as the sample introduction process) for the solutions and the slurries which applies rigorously to the present sample type and nebulizer system. With the introduction of Freon no degradation in the linearity of the analytical curves was found in contrast to earlier observation by other workers. The line-to-background intensity ratios (wavelength >220 nm) were not decreased at the rate of halocarbon introduction eventually used.Keywords lnductively coupled plasma atomic emission spectrometry; silicon nitride slurry; halogenation; Freon- 72 Pneumatic nebulization of slurries has been found to be a fairly easy and readily accessible method of sample introduc- tion for the inductively coupled plasma atomic emission spec- trometric (ICP-AES) analysis of finely dispersed In principle accurate results can be expected if calibration is based on certified reference samples similar in physico-chemical properties (including particle size distribution) to the samples. The approach of replacing powdered standards by standards in solution has led to moderate success in general although small or large negative errors in the results are often observed for several sample types and different constituents.The high solid V-groove nebulizer was designed' specifically for slurry nebulization and resulted in a lowering of the negative errors but these could not be eliminated ~ompletely.~?~ One of the possible reasons for this is related to the lower degree of evaporation of slurry particles relative to the smaller particu- lates formed from solution droplets in the plasma.24 To decrease this potential source of error Ebdon and Goodal16 suggested halogenation and introduced a mixture of Freon-116 (C,F,) and argon as the aerosol carrier (nebulizer) gas. Although they reported an apparent improvement in 'analytical recovery' the overall result was discouraging in that increased curvature of the analytical graphs towards the concentration axis (for Al Fe Mg and Ti) was found for solution standards with the introduction of Freon.In the present work the applicability of a commercial Babington-type nebulizer to slurry nebulization was specifically investigated. The sample was a finely dispersed silicon nitride powder (mean particle size of 0.54 pm) analysed first by a conventional dissolution-based method the results of which were then used as reference data for the slurry method. Even applying standard additions in the form of solutions to the slurry as an attempt at matrix matching negative errors were found similar to those observed in the works cited above. It therefore became a vital question as to whether the halogen- ation suggested in ref. 6 can improve the results of the present slurry nebulization technique.The method of Freon introduction selected for use was * Presented at the XXVIII Colloquium Spectroscopicurn Internationale (CSI) York UK June 29-July 4 1993. t To whom correspondence should be addressed. similar to that used earlier in flame atomic absorption spec- trometry (AAS),7 in that it was supplied after the spray chamber into the injector flow rather than into the high- pressure gas that feeds the nebulizer.6 In AAS experiments (using an acetylene-air flame) the depression effect of A1 on the Mg signal (as a typical solute vaporization interference) could be eliminated at a Freon-12 concentration of 1.8% v/v in the total flame gases.7 Experimental Instrumentation and Operating Conditions Spectrometer.Labtam 8440 Plasmalab 1440 vacuum polych- romator with Paschen-Runge mounting of the grating ( 1440 grooves mm-' 1 m focal length) simultaneous detection of the lines A1 I 396.152 Ca I1 393.367 Fe I1 259.940 Mg I1 279.553 Si I 251.611 and Ti I1 334.940nm. Integration time 5 s and background correction by shifting the entrance slit. ICP source. A 27.12 MHz crystal controlled generator 1.2 kW output power demountable torch of medium size (outer quartz tube of 17.5 mm i.d.) sample injection tube of 1.8 mm i.d. argon flow rates (outer + intermediate + inner) of (14 + 1 + 0.8) dm3 min-' respectively and observation height in the plasma of 16 mm unless stated otherwise. Sample introduction system. Babington-type GMK nebulizer aerosol carrier argon flow of 0.7 dm3 min-l impact bead distance of 3 mm Gilson Minipuls-2 peristaltic pump and sample delivery of 3 ml min-'.Freon introduction system. A T-shaped glass junction 4 cm long was incorporated between the outlet of the nebulizer chamber and the inlet of the torch sample capillary in order to feed a mixture of 100 cm3 min-l of argon and 10 cm3 min-l of Freon-12 (Union Carbide Westerlo Belgium) unless stated otherwise. The flow rates of argon used for dilution of the Freon and for sample nebulization were kept constant (total 0.8 dm3 rnin-l see above). The flow rate of Freon was moni- tored by a 'pressure difference' flow meter which was filled with paraffin oil and calibrated by a soap-bubble gas volu- metric device. The argon+Freon mixture was streamed in a polyethylene tube and it was concluded that the adsorption- desorption equilibrium of Freon on the internal tube wall708 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 could be attained in 3-5 min depending on the flow rate. A favourable practice was to first set a higher flow rate of Freon than that to be selected for a particular experiment for 3 min and then set the required value and allow 5 min before sample introduction. The reflected power could also be used for checking the constancy of Freon introduction. Characteristics of Sample The silicon nitride (Si,N,) powder sample was manufactured by Starck Berlin Germany. The particle size distribution of the Type LC-10 powder was determined in the form of a suspension (0.05 YO Calgon liquid) using Micrometrics equip- ment.After ultrasonic treatment for 8 h the mean particle size was dS0=0.54 pm. Slurry Preparation A 1 f0.01% m/m concentration of silicon nitride powder was achieved in 1% m/m HCl (see below) the suspension was agitated in an ultrasonic bath for at least for 5 min before nebulization and stirred continuously during nebulization by a magnetic stirrer. Dissolution Procedure for Solution Analysis A CEM Model MDS-2100 microwave digestion apparatus was used for sample dissolution in an HF+HN03 mixture. Silicon nitride (150mg) was weighed into a 120ml volume poly(tetrafluoroethy1ene) (PTFE) vessel and 10 ml of HF (40% m/m) and 1 ml of HNO (65% m/m) were added to digest the samples at 170°C (for 30 min). After cooling the solution was transferred into a PTFE beaker and heated gently to dryness.Then 10 ml of HCl (18% m/m) were added and the evapor- ation was repeated. Again 10 ml of 1% m/m HCl were added for dissolution of the dry residue. Three solutions were pre- pared simultaneously as described above all were combined and transferred into a 50ml calibrated flask and made up to the mark with doubly distilled water. Calibration Standard additions Standard additions in the form of solution to the powdered sample was performed as follows. Four sample aliquots of 0.4 g were weighed in 50 ml glass beakers 10 ml of 1% m/m HCl were added to each and three of the samples were spiked with linearly increasing volumes (50 100 and 200 pl) of a multi-element (Al Fe Ca Mg and Ti) stock solution. The first addition approximated the impurity contents according to available information.For the 'zero standard addition' the stock solution was replaced with water. The liquid phases of the suspensions were adjusted to volumes of 40 ml again with the 1% m/m HCl. Stundard and blank solutions The standard and blank solutions were also made in the same manner as above using the same stock solution and diluent but without addition of the silicon nitride powder. Evaluation of intensity data The intensity data (measured with background correction) of triplicate runs was evaluated by linear regression analysis and the unknown concentrations were calculated on the basis of standard additions to slurries and also external calibration using matrix-free standard solutions. Blank concentrations were determined on the basis of solution standards and sub- tracted from the former results.Results and Discussion Linearity of the Analytical Curves With Introduction of Freon Ebdon and Goodal16 found increased curvature of the analyt- ical curves towards the concentration axis for the analytes of interest (Al Fe Mg and Ti) when an argon+Freon-116 mixture was used for the nebulization of solution standards. They added 4% v/v Freon to the aerosol carrier gas which at the maximum corresponded to a Freon flow rate of 52 cm3 min-'. As the linearity of the analytical curves is of prime importance for simplicity of calibration investigation of this parameter was given prominence in the present research. However a different method of Freon introduction (described under Experimental) was applied from that in earlier work.6 As demonstrated for two elements (A1 and Fe) in Fig.1 rectilinear analytical curves (logarithmic plots) were found in the 1-100 mg 1-' concentration range independent of the introduction of Freon which applies also to the other elements (Ca Mg and Ti) investigated. In this instance multi-element standard solutions were nebulized and the concentration range was similar to that investigated in ref. 6 except for Al where the concentration in the earlier work extended to 1000 mg 1-'. It should be noted that linear intensity versus concentration functions were also found by addition of solution standards to the slurries (see Experimental) and setting the observation heights to 20 and 25 mm (16 mm in Fig. 1). Other Spectral Characteristics Versus Freon Flow Rate By increasing the flow rate of Freon up to 40 cm3 min-' the net line intensities decreased for all constituents in the slurry and the solution by factors of between 2.5 and 3.2.Aluminium was selected for demonstration as shown in Fig. 2 because of the considerable separation of the 'slurry' and 'solution' curves (see also below). As is also shown in Fig. 2 the line-to- background intensity ratios (which are important from the point of view of detection limits) vary according to the maxi- mum of the curves peaking at a Freon flow rate of around 25 cm3 min-'. More general information on the variation of the spectral background on introduction of Freon is presented in Fig. 3 where the intensity ratio (with :without Freon) is plotted as a function of wavelength for 23 analytical lines that can be detected with the polychromator used.Data are shown for nebulization of high-purity water and for a blocked sample uptake tube i.e. for an almost dry plasma. The flow rate of Freon was 10cm3min-' which was also selected for the 5x10' I 5 ' I I I 1 x 10-1 1 10 1 ~ 1 0 2 1 ~ 1 0 3 Concentration/mg I-' Fig. 1 Analytical curves for A1 and Fe determined with solution standards A without Freon; B with 10 cm3 rnin-.' Freon; and C with 40 cm3 min- ' FreonJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 709 > L .- 1.6 Q) C c .- 1.4 3 2 E 1.2 n 2 1.0 m .- - U 0.8 i 0.6 I I I I I 0 10 20 30 40 Flow rate (Freon-12)/cm3 min-' Fig. 2 (a) Net line intensity and (b) normalized line-to-background intensity ratio for Al A in slurry and B in solution as a function of Freon flow rate ~ 0 150 200 250 300 350 400 450 500 Wavelengthhm Fig.3 Background intensity ratio with and without the introduction of Freon at a flow rate of 10 cm3 min-' for 23 analytical lines in the 180-460 nm wavelength range x without water nebulization and 0 with water nebulization analytical work (see below). As can be seen the back- ground intensity ratio (with without Freon) is higher for a dry plasma than for a wet plasma above 250nm and the opposite is valid below 250 nm. On the other hand for a wet plasma this ratio is lower than unity above 250nm which means that a lower background is found with halocarbon introduction than without it. The background is increased by introduction of Freon below 250 nm and this increase is higher for a wet plasma than for a dry plasma.These findings are probably due to the band emission of CO (and chemically related) species,' while the lower background above 250 nm (mostly the electron continuum) could be the result of a drop in excitation-ionization temperature on introduction of Freong (see below). By visual inspection of the plasma a green colouration of the lower zone of the inner core can be observed at and above a Freon flow rate of 20cm3 min-' which is much more intensive without nebulization of water. This phenomenon has also been observed when carbon tetrachloride was nebulized into the ICP and is related to the band emission of C2 species.' The decrease in intensity of C2 bands in the wet plasma is the consequence of oxidation of carbon along with the formation of CO (see above).In the work described in ref. 9 the 'excitation temperatures' of the ICP were compared when nebulizing aqueous and carbon tetrachloride solutions using a desolvation system. The rate of introduction of the solvent vapour could be varied by operating the condenser of the desolvation unit in the tempera- ture range from -10 to 20°C. At the lowest condenser temperature (the lowest rate of vapour introduction) the plasma temperature was lower by about 700 K for carbon tetrachloride compared with that for introduction of water at an observation height of 10mm. A considerable decrease in the ratio of the ionic to atomic line intensities (using the most sensitive lines of Mg) was also found with the introduction of carbon tetrachloride indicating a depression in the ionization.It could be expected that similar trends in changes of the excitation parameters prevail with introduction of Freon as those deter- mined for carbon tetrachloride introduction,' as outlined above. It was anticipated that the decomposition products of the fluorinated hydrocarbons generated in the plasma would react with the hot parts of the torch thus resulting in corrosion after long-term application. Indeed this corrosion was noted in ref. 6 on introducing 52cm3 min-' of Freon-116 at an intermediate argon flow rate of 0.4 dm3min-'. As is known the latter component of the argon supply to the torch plays a dominant role with respect to heating of the tip of the quartz sample introduction capillary by the plasma. Therefore the effect of the Freon flow rate was studied with increasing intermediate argon flow rates by monitoring the Si 251.61 1 nm line as an indicator of Si release from the torch.The results are depicted in Fig. 4 (curve B) when using a 1 dm3 min-' intermediate argon flow rate which was found to be sufficient to diminish corrosion of the tip of the sample introduction capillary. Also shown is the variation in the Si line intensity when a silicon nitride slurry is nebulized (curve A) which is similar to that found for the other slurry components (see Fig. 2 and more details are given below). Axial Intensity Profiles In Fig. 5 net line and background intensities and relative line- to-background intensity ratios are plotted as a function of plasma observation height (POH) for slurry nebulization 201 7 I " 0 10 20 30 40 Flow rate (Freon-12)/cm3 min-' Fig.4 Line intensity of Si (251.611 nm) as a function of Freon flow rate A slurry nebulization; and B water nebulization710 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 1 lo5 -5 2 1x103 & 2 1x102 Y m C .- - -a C m .- ,z 1x10’ w al C 4- - 1 1 x lo-’ A 4 8 12 16 20 24 28 Plasma observation height/mm Fig.5 Axial plasma profiles of some spectral parameters for slurry nebulization A Si line intensity; B Mg line intensity; C background intensity of Mg line; and D Mg line-to-background intensity ratio with the introduction of Freon (10 cm3 min-l) relative to that without Freon (0 argon; 0 Freon; and 0 Freon argon) without and with the use of Freon (10 cm3 min-l). The shape of the corresponding curves of the five analytes (note the logarithmic scale of the ordinate) were similar and Mg was however selected for demonstration purposes.The main characteristics of these curves correspond to those determined for a matrix-free solution; deviations in the finer details will be shown later. In summary the line intensity maxima are found at a POH of 16 mm for the analytes (14 mm for Si) the log of the background intensity decreases linearly with POH to a first approximation and all of these intensities are between 0.59 and 0.83 of those obtained when no Freon was present in the plasma. This last observation for line intensities again suggests a lower excitation temperature with introduction of Freon as is presumed above on the basis of the work cited in ref.9. Although this explanation is plausible it has not been proved so far. The curve representing the line-to-background intensity ratio with the use of Freon related to that in the absence of Freon shows a slight maximum in the 14-18 mm POH range which also applies to the other analytes. The maximum values were found to be close to unity (lower for Mg and Fe and higher for Ti Ca and Al). This means that no gross change in the detection limits are expected with the introduction of Freon for the elements investigated under optimum plasma conditions. The comparison of axial intensity profiles with and without introduction of Freon was also aimed at finding experimental evidence for halocarbon assisted vaporization of slurry par- ticles in addition to the analytical results discussed in the next section.As is known the main excitation parameters (tempera- ture and electron pressure) change dramatically along the vertical axis of the plasma and also these parameters vary because of the introduction of Freon. The latter variation can be related to a higher energy consumption and also to the increase in the total gas flow rate in the plasma channel. Therefore ratioing of line intensities is necessary to compensate for the changes in excitation parameters and thus to obtain information on changes in the vaporization of sample particles. The better this compensation is approximated the more reasonable is the expectation that the difference in line intensity ratios (with without Freon) is due to the difference in the vapour concentration of the emitting analyte.Two approaches of this ratioing are represented by Figs. 6 and 7 and will be discussed. To avoid possible problems in understanding the terms ‘vaporization’ and ‘atomization’ as they are used here will be clarified. By (high-temperature) vaporization a heterogeneous phase transition (from condensed phase to gas phase) process is implied which often also involves a chemical decomposition 0‘ I I I I I 1 0 ‘ I I I I I I I 16 20 24 28 0 4 8 12 Plasm a observation he ig ht/m m Fig. 6 Axial plasma profiles of intensity ratios measured with Freon (10 cm3 min-l) and without Freon (i.e. with argon) when nebulizing 0 a solution and 0 a slurry for (a) Si; (b) Ca; (c) Ti; and ( d ) Fe I I I I I I ~ 0 4 8 12 16 20 24 28 Plasma observation height/mm Fig.7 Analyte-to-Si intensity ratios for the slurry 0 wit1 out and 0 with the introduction of Freon (10 cm3 min-’) as a fui ction of plasma observation height for (a) Mg; (b) Ti; (c) Fe; and ( d ) t 1JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 71 1 and/or chemical transformation. For an unambiguous differ- entiation the term gas-phase atomization can be used which means a homogeneous phase chemical process (gas-phase dissociation of molecules). In the literature on analytical atomic spectrometry the term 'atomization' most often means both vaporization and gas-phase atomization together which is accepted terminology if no mechanistic interpretations are concerned. In the present work the possible effects of Freon on the vaporization and on the gas-phase atomization must be distinguished and thus the more specific meaning of the terms should be utilized.The best compensation for the changes in excitation param- eters is expected when ratioing is made for the same spectral line without and with a supply of Freon and this ratioing is made separately for slurry and solution nebulization. In addition to the curves seen in Fig. 6 for Fe Ti Ca and Si it should be noted that the corresponding curves for A1 and Mg are similar to those shown for Fe and Ti respectively. The Si curve (determined only for a slurry) shows a relatively high value at the lowest observation height (6 mm) and the same applies for Fe and A1 (not shown) for both slurries and solutions.On the other hand the curves for Ca Mg (not shown) and Ti start with the lowest value at an observation height of 6 mm. In this low-lying zone of the plasma desolv- ation is probably far from complete and desolvation can be different for solution droplets and slurry particles. It has been shown that the excitation parameters are strongly influenced by the degree of desolvation," so the interpretation of the change in line intensities at an observation height of 6 mm is complicated by too many factors. On considering the range of observation heights of between 10 and 22 mm slightly higher intensity ratios (with without Freon) were found for all analytes present in the slurry than those present in the solutions (demonstrated for three elements in Fig. 6).It was expected that evaporation of the residue particles formed from droplets of the matrix-free solution also readily takes place without Freon introduction. Therefore the slightly higher intensity ratios found for the slurry relative to the solution could be explained by the increase in the degree of evaporation of slurry particles under halogenation. The gas-phase atomization of the elements of interest could be influenced by the introduction of Freon predominantly through the formation of the most stable monofluorides as also discussed in ref. 6. The dissociation energies of these species" are as follows A1F 6.89 TiF 5.90 CaF 5.42 SiF 5.03 and MgF 4.77 eV. The data for FeF are not given in ref. 11 but they can be assumed to be similar to those of MnF (4.4eV). It is expected that the formation of fluorides takes place to a greater extent in the lower temperature regions of the higher plasma zones and also the value of the dissociation energy is reflected in the magnitude of the decrease in line intensities.Measurement points of up to an observation height of 26 mm (Fig. 6) however do not indicate a decrease in the intensity ratios (with without Freon) i.e. a decrease in gas- phase atomization owing to introduction of Freon. (This means that the formation of monofluoride could probably be studied at higher plasma zones not considered in the present work.) The measurement points of the analytes in Fig. 6 show a tendency to reach constant and equal values for both the solution and the slurry in the POH region of 22-26 mm.This tendency can be explained by supposing that either the analytes are vaporized from the slurry particles or the whole particles vaporized to completion also without the introduction of Freon up to this observation zone. It should be remembered that finely dispersed silicon nitride was used in the present experiments which was therefore less appropriate for a clear demonstration of halocarbon assisted vaporization. Also because of the use of a relatively high concentration of HC1 in the slurry (see Experimental) partial dissolution of certain slurry components could take place. This could be another reason for the apparent decrease in the difference in the degree of vaporization of slurry particles and the residue particles from droplets of solution.In Fig. 7 reference is made to Si when the slurry is nebulized without introduction of Freon and separately when Freon was supplied to the plasma. This ratioing corresponds to the use of the internal reference method. The Ca Si curve (not shown) was similar to that depicted for Ti Si. Here compensation for the change in excitation parameters is limited because of the difference in the ionization energies of the analytes and the Si and also in the excitation energies of the corresponding spectral lines. In addition the possible change in the degree of vaporiz- ation of the major Si component under the effect of halogen- ation influences the information that can be drawn for the analytes in this respect. In spite of these limitations the higher intensity ratios (analyte silicon) seen for the case of introduc- tion of Freon in the POH region of 14-22 mm can be explained by an increase in the degree of vaporization of the analytes relative to the major Si component.The declining tendency of the curves (the A1 Si curve is an exception) at higher obser- vations heights can be explained by supposing that the degree of vaporization increases only for the major element Si (POH=22-26 mm). On the other hand the almost constant values for the Al Si curves with and without introduction of Freon above a POH of 14mm can be considered as an indication that the selectivity of vaporization for this element from the silicon nitride matrix is lowest irrespective of the presence of Freon. The changes in line intensity ratios due to introduction of Freon shown in Figs. 6 and 7 are rather small and if the possible limitations in compensation for changes in the exci- tation parameters are considered one cannot make an unam- biguous statement about the assistance provided by introduction of Freon towards vaporization.Further evidence should be anticipated from the analytical results discussed below. Analytical Results As described under Experimental two calibration methods were selected for slurry analysis (i) using matrix-free solution standards and (ii) using solution-spiked slurries as an approxi- mation to matrix matching. The preparation of acidic solution standards of matched Si matrix is problematic and was not pursued. The results found by the conventional dissolution- based method were considered as reference values and these are shown in Table 1 together with results for the different versions of the slurry method.It was interesting to observe the effect of the slurry matrix Table 1 Results for analysis of a silicon nitride sample using solution and slurry nebulization methods without and with the use of Freon-12. Average relative standard deviation (RSD) for the slurry results was 3.2% Method Element concentration/mg kg - Sample dissolution* Slurry At no Freon Deviation (YO) Slurry BS no Freon Deviation (YO) Slurry At with Freon Deviation (YO) Slurry BS with Freon Deviation (YO) A1 392 239 233 268 - 31.6 253 - 35.5 - 39.0 - 40.6 Fe 56.6 47.7 - 15.7 45.3 - 20.0 52.9 - 6.5 52.6 -7.1 Ca 51.5 45.1 - 12.4 40.6 -21.2 48.7 - 5.4 46.5 - 9.7 Mg 51.3 40.8 -20.5 37.4 - 27.1 44.2 43.0 - 13.8 - 16.2 Ti 12.6 10.3 - 18.3 9.3 - 26.2 11.8 - 6.3 11.4 - 9.5 ~ ~~~ ~ ~~~ * Reference method. t Calibration based on matrix-free standard solutions.1 Calibration based on addition of solution standards to the slurry (approximate matrix matching).712 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 on the slopes of the linear analytical curves (expressed as the ratio of slopes found with solution-standard spiked slurries and with matrix-free solution standards). The averages of this ratio for the five components were 1.078+0.026 and 1.035 kO.019 without and with introduction of Freon respect- ively. It follows that the analytical sensitivities were enhanced slightly in the presence of the slurry matrix and this enhance- ment effect was smaller in the presence of Freon. Considering the negative errors found with the four versions of the slurry method (Table 1) it is interesting that these are slightly larger when calibration is based on the spiked slurries as a consequence of the noted matrix effect. It is clear that both of the calibration methods used are imperfect in several aspects and because of a reason as yet unknown better error compensation results when matrix-free standard solutions are used for calibration in the case of the present sample and under conditions used.The other important consideration from Table 1 is that on introduction of Freon the negative deviations are decreased significantly for Fe Ca Mg and Ti (by factors between 2.2 and 2.8); the improvement is smallest for A1 (by a factor of 1.14).Conclusions According to the data in Table 1 the negative deviations in the results found with the slurry nebulization method decreased with the introduction of Freon but were not completely eliminated. A comparison of the axial intensity profiles of the plasma (Figs. 6 and 7) and that of the results without and with introduction of Freon- 12 together indicate an enhance- ment in the degree of vaporization for the constituents of the slurry under the effect of this halocarbon as a halogenating agent. The question emerged as to whether the negative deviations could be further decreased by applying more favour- able conditions to the evaporation of slurry particles. Therefore measurements were conducted at higher POHs (20 and 25 mm) and also the flow rate of Freon was doubled (20 cm3 min-l).Eventually practically no further improvement could be obtained which suggests that the remainder of the negative deviations found with the introduction of Freon (Table 1) are probably predominantly because of the difference in nebuliz- ation efficiency of the solution and the slurry. In this respect both earlier opinions are in agreement according to whether the insufficient degree of evaporation or the less efficient nebulization was attributed to the negative errors obtained with the slurry methods (as disputed in ref. 5). In the present work the highest negative deviation was consist- ently found for Al with and without halogenation which could be explained by a higher loss for this element during nebuliz- ation if its concentration is higher in the coarser fraction of the sample.However from the A1 Si curve seen in Fig. 7 it can also be concluded that the degree of evaporation is the smallest for this element from a silicon nitride matrix and the halogenation has only a slight influence on this behaviour. The final recommendation as a result of the present work for the expedient use of the slurry-ICP method can be summarized as follows. If high accuracy is mandatory in a certain analytical task sample introduction by slurry nebuliz- ation must be calibrated with the use of certified reference samples matched to the samples in all possible respects. This can best be achieved in laboratories by quality control of well defined types of materials. The halogenation method investi- gated here can be beneficial in broadening the tolerance with respect of the ‘physical’ non-uniformity of samples and stan- dards. However unambiguous proof for this expectation is anticipated in the future. The authors are grateful to the National Scientific Research Foundation (Hungary) for the support under project numbers OTKA 2278 and OTKA 2786. 1 2 3 4 5 6 7 8 9 10 11 References Ebdon L. and Cave M. R. Analyst 1982 107 172. Raeymaekers B. Graule T. Broekaert J. A. C. Adams F. and Tschopel P. Spectrochim. Acta Part B 1988 43 923. Huang M. and Shen X.-E. Spectrochim. Acta Part B 1989 44 957. Gervais L. S. and Salin E. D. J. Anal. At. Spectrom. 1991 6,41. Halicz L. Brenner I. B. and Yoffe O. J. Anal. At. Spectrom. 1993 8 475. Ebdon L. and Goodall P. Spectrochim. Acta Part B 1992 47 1247. Kantor T. Atomic Spectroscopy Methods for Solid Sample Analysis and for Studying High Temperature Vaporization Processes Thesis (in Hungarian) Library of Hungarian Academy of Sciences 1985. Pearse R. W. B. and Gaydon A. G. The Identijcation of Molecular Spectra Chapman and Hall 4th edn. London 1976. Pan Ch. Zhu G. and Browner R. F. J. Anal. At. Spectrom. 1990 5 537. Hobbs S. E. and Olesik J. W. Spectrochim. Acta Part B 1993 48 817. CRC Handbook of Chemistry and Physics ed. Wheast R. C. Chemical Rubber Co. Cleveland OH 54th edn. 1974. Paper 31049260 Received August 13 1993 Accepted February 2 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900707

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 71 3 Determination of Transition Metals in the Primary Water of Pressurized Water Reactors by Inductively Coupled Plasma Mass Spectrometry Rolf J. Rosenberg and Riitta Zilliacus Technical Research Centre of Finland VTT Chemical Technology P. 0. Box 1404 FIN-02044 VTT Finland Pentti K. G. Manninen Health and Environmental Inspection Centre Control and Research Laboratory Niemenkatu 73 FIN-I5210 LAHTI Finland A method for the determination of the transition metals Fe Mn Ni and Co in the primary water of pressurized water reactors (PWRs) is described and applied to the analysis of water at the Loviisa power plants. Samples are collected via the standard sampling line or an isokinetic line. There is no difference in the results obtained.The water is filtered through a 0.45 pm membrane filter under oxygen-free conditions to separate the soluble and insoluble fractions. The transition metals are concentrated and separated from boric acid using cation chromatography. A quantitative yield is obtained and concentration factors of up to 500 are used. The samples are analysed using inductively coupled plasma mass spectrometry. The detection limits obtained are Fe 2 Mn 0.2 Ni 0.2 and Co 0.02 ng I-' when a 1 I sample is used. Keywords Inductively coupled plasma mass spectrometry; transition metals; primary water; pressurized water reactor The radiation field build-up on the primary circuit components in nuclear power plants is of concern because of the exposure of the workers to radiation. Most of the radiation is caused by activation products from transition metals mostly 6oCo from Co 5sC0 from Ni and 54Mn from Fe.In order to study the physico-chemical mechanisms of the production and trans- port of the radionuclides it is necessary to know the concen- trations of the source metals in the primary water. The concentration of Fe must also be known in order to understand the corrosion and corrosion product transport phenomena. Usually the soluble and insoluble metals are determined separ- ately in order to obtain information on the solubilities. Two different stages of reactor operations are normally studied. During full power the transition metal concentrations are very low and the variation moderate. Therefore the sampling frequency can be one sample per month or less.During shut- down the solubilities increase and fast transients in the trans- ition metal concentrations can be observed. Therefore shut- downs and sometimes the return to full power are followed with high sampling frequency up to one sample per hour over 3-4 days. The analysis is not without problems. The sampling is complicated by the need to reduce the pressure and tempera- ture the latter because the pH changes which affects the solubility.' The samples are transported through tens of metres of steel pipes. The problems concerning representative sampling have been extensively investigated and discussed in a number of papers.'-6 It has been shown that absorption and desorption phenomena affect the transport of radionuclides and transition metals in the pipelines in a way which is not yet fully understood.However it has also been shown that under appropriate conditions analytical results behave logically in a way that enables a comparison between different reactors and different years of the same reactor to be made. Bridle and co-~orkers'*~ introduced a so-called isokinetic sampling system in which the pressure drop is effected with a 50 m long capillary tube instead of the normal valves and heat exchangers. They indicate that the isokinetic sampling line gives a more represen- tative sample than the standard sampling line because the water flow is smooth without variations in flow rate. However experimental data verifying this statement have not been presented. Svoboda et a1.2 compared an isokinetic sampling line with a standard one.They did not observe any differences in the concentrations of Fe Ni and Co. The samples should be filtered before the water has been in contact with air. This prevents the oxidation of Fe which would drastically decrease its solubility. Usually 0.45 pm mem- brane filters are used although part of the insoluble material seems to be in colloidal form with particle sizes much below that. The lowest Co concentrations found in the insoluble fraction during full power are below 1 ng 1-l.' Therefore several tens or even hundreds of litres of water might be filtered for a sample. The lowest soluble Ni concentrations during full power are a few tens of ng 1-' the lowest Co concentrations a few ng 1-l.' Therefore most analytical techniques in use require a precon- centration step.Two methods have been used. Ion-exchange filters such as Acropor SA 6404 Whatman SA2 and Sumimoto Chemical Expapier F-2 have been ~ s e d . ~ ~ . ' ~ These have the benefits that fairly high sampling rates can be maintained 50-150mlmin-' and that they can be used in combination with a particle filter and an anion filter. The reported capacities can be 0.5 mmol per filter although experiment has shown the practical capacity to be one tenth of this.2 In combination with ion chromatographs preconcentration columns can be conveniently used.'*7 They have capacities of up to 0.45 mequiv. One drawback is the low sampling speed 1-2 ml min- '. Two different types have been used in this work. They are the Dionex MetPac CC-1 and Waters Guard Pak.A variety of analytical techniques have been used for the determination of the transition metals. The only technique capable of determining Co during full-power conditions in most reactors without preconcentration is ~oltammetry."-~~ The reported detection limits are 1-2 ng 1-' for Co and 8 ng 1-1 for Ni. All other methods require a degree of preconcentration if low concentrations are to be detected. Ion chromatography has been widely applied in European power plants mainly by the Winfrith Group in the UK.',7,14-17 Using preconcentration of a 1 1 sample a detection limit of 0.5 ng 1-1 of Co has been reported.' Atomic absorption spectrometry can in principle be used after preconcentration from a very large sample.2 The detection limit could be significantly improved by using electro- thermal vaporization but no actual results from determinations of low levels of Co have been reported.A few determinations using X-ray fl~orescence~,~ and flow injection te~hniques''~~~ have also been reported the latter without any details on the method. Neutron activation analyses have also been used but714 JOURNAL OF ANA.LYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 the detection limits are low requiring a considerable precon- centration.2 Inductively coupled plasma mass spectrometry (ICP-MS) is a very sensitive method for analysis of water but to date very few applications on the analysis of nuclear power plant (NPP) waters have been reported. There are a few references to work on loop tests and boiling water reactor (BWR) analysis but these contain very few The present paper describes the development of methods for sampling preconcentration and ICP-MS determination of soluble and insoluble Fe Mn Ni and Co in the primary water of a pressurized water reactor (PWR) results for the determi- nation of these metals in the primary water of the Loviisa 1 power plant during the refuelling shut-down in 1992 and Loviisa 1 and Loviisa 2 during full power in 1993 are presented.Only a few results are presented to show the applicability of the method. The full results together with a discussion will be published later. Experimental Sampling In old reactors the sampling lines are fixed and cannot be changed. In the Loviisa 1 NPP an isokinetic sampling by-line had also been installed and this was compared with the standard line.The basic principle of sampling is to have a steady continu- ous flow through the sampling line in order to avoid transients which change the physico-chemical parameters. Another prin- ciple is to prevent the oxidation of Fe before the filtering and the preconcentration steps. In the present work a 0.45 pm membrane filter was used mainly in order to facilitate a comparison with literature data. It was realized however that a smaller pore size filter should be used in order to separate all the smaller particles. This will be investigated in future work. During the shut-down of the Loviisa 1 plant (18-21 July 1992) 0.5-1 1 samples were filtered in closed Nalgene filter holders equipped with 11 receiver containers.Nitrogen gas was passed through the system to prevent contact with air. Part of each of the filtered samples was poured into 1OOml acid washed polyethylene bottles which were immediately frozen. The frozen samples were stored for up to 3 months before analysis. Most of the samples were taken from the purification circuit before the ion-exchange columns. This sampling line had an isokinetic by-line installed in 1989.7 Samples were taken both from the standard outlet and the isokinetic line. The isokinetic line was opened 9 d before the actual sampling started. The flow rate was 35 ml min-I. The standard sampling line was opened 5 d before the sampling started. The flow rate was about 1 1 min-I. During the shut-down of the Loviisa 2 plant (11-15 September 1992) sampling was made from the corresponding purification circuit but only using the standard sampling line which was opened 3 d before the sampling started. A Millipore filter holder was connected directly to the sampling line to prevent the access of air.A 101 bottle was used for measure- ment of the volume of the filtered water. From 10 to 201 samples were filtered. The water was preconcentrated immedi- ately after filtering but the preconcentration was made from water that had been poured from one container into another. Therefore it had been in contact with air. The above experiences resulted in the final sampling arrange- ment shown in Fig. 1. Both the filtering and the preconcen- tration are performed in a closed system which prevents the access of air. This system was used for the analysis of the water of the Loviisa 2 reactor during full power in February 1993.The continuous sampling flow was 1.6 1 min-'. About 30 1 of water were filtered. Preconcentration The reason for the preconcentration is two-fold. Firstly the samples have to be concentrated in order to improve the Sample in Q Sample to Three-way valve Peristaltic pump +- Vo I LI me me as u re m e nt flask Bottle Eluant and regeneration solution Fig. 1 a PWR Principles of the arrangement for sampling primary water of detection limits and thus the accuracy of the analysis. Secondly the high boric acid concentrations up to 12 g 1-' prevent the direct injection of the water into the ICP-MS system which is equipped with a normal nebulizer. Therefore the boric acid has to be removed.Two different preconcentration columns were tested. These were the Waters Guard Pak which is packed with a porous polystyrene gel with sulfonate functional groups and the Dionex MetPac CC-1 which contains a macroporous iminodiacetate chelating resin. The preconcen- tration procedures shown in Table 1 are those used although not necessarily the optimal ones. The reason for choosing these schedules was partly to be able to run the two preconcentration columns in parallel. The Waters column was tested in the laboratory using simulated reactor water containing 12 g 1-' of boric acid 6mg1-I of NH 2mg1-I of KOH 50ngml-I of Fe 13 ng ml-' of Mn 106 ng ml-' of Ni and 0.3 ng ml-I of Co. Different amounts of the water were run through the column which was eluted according to Table 1.The metal concen- trations in the 1Oml elution fraction were determined by ICP-MS to establish the yield. The capacity was determined by running 500ml of the water through the column in order to exceed the capacity significantly. The column was then eluted and the concentrations determined by ICP-MS. Because only preconcentration with the Waters Guard Pak was tested in the laboratory the behaviour of the Dionex MetPac CC-1 was tested by comparing it with the Waters Guard Pak. This was done by preconcentrating every second sample of the Loviisa 1 samples with each column. The Loviisa 1 samples were left to thaw in the refrigerator after which they were preconcentrated according to the above scheme. In the Loviisa 2 sampling a number of parallel samples were preconcentrated with each column immediately after sampling.The blank of the MetPac CC-1 column and the reagents was measured by regenerating the column with 20ml of the Table 1 Preconcentration scheme for the Waters Guard Pak and the Dionex MetPac CC-1. During the shut-downs 50-100 ml samples were preconcentrated while during full power 1-2 1 samples were used ~ Step Dionex MetPac CC-1 Waters Guard Pak Adsorption 100-2000 ml water 100 ml water Elution 2 ml 1 moll-' HNO 10 mlO.1 moll-' HNO Regeneration 10 ml 0.1 moll-' HNO 8 ml 1 mol I-' HNO 10 ml buffer 2 mol I-' NH,Ac pH 5.5JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 71 5 buffer solution cleaning the column with 10 ml of de-ionized water and eluting with 2 ml of 1 mol 1-' HNO,.Determination of the Transition Metals by ICP-MS The samples were analysed with a VG PlasmaQuad PQ2-t ICP-MS. The operating parameters are shown in Table 2. The samples are radioactive although the activity is low being mostly at a level permissible for free handling. In order to avoid unneccessary releases of radioactive material the gaseous effluents were filtered with an absolute filter and the liquid effluents collected and stored as radioactive waste. The samples taken during shut-down were diluted with water to 5-10m1 and run with the sample changer with the parameters shown in Table 2 using a 30 s dual mode counting time. The scanning mode was used. The samples taken during full power were diluted to 5 ml. The samples were run using the sample changer by injecting a 250 pl sample followed by a 1% HNO washing solution.After a 45 s delay when the sample reached the analyser a 10 s measurement in the peak jumping pulse counting mode was started. The best masses to be used proved to be 55Mn 56Fe 58Ni and 59C0. Calibration curves were produced from solutions of 1.0-50 ng ml-I prepared by dissolving pure metals in HN03. The measurements were checked by analysing suitably diluted SPEX multi-element plasma standard ICP-MS solutions. Blanks were prepared from the corresponding elution solutions. When using the normal Ni sampling cone the blanks caused by contamination from the analyser and the argon ions varied somewhat but were of the order of Fe 180 Ni 3-10 Mn 0.3 and Co 0.04-0.1 ng ml - '. Through using a sampler made from A1 at the work-shop of the Reactor Laboratory the Ni blank was lowered to 0.5 ng ml-'.Also the Fe blank was decreased to 35 ng ml-l. In future the skimmer will also be made of A1 and the solutions and other parts of the analyser checked once more to decrease the blanks further. The A1 sampler seems to be much more resistant to the plasma than the Ni sampler probably because of its excellent heat and electrical conductivity. The column was shown to give a considerable blank when left standing for a few hours both after the regeneration and elution. Therefore the column has to be cleaned by elution with 1 moll- ' HNO before it is taken into use. The purity of the reagents must be controlled. The similarity of the Ni and Co results and the great difference in concentration caused a suspicion of a Ni inter- ference on the Co peak.Therefore pure Ni solutions were analysed for Co in order to detect any overlapping. The reproducibility of the ICP-MS measurements were investigated by analysing the same samples several times at different time intervals. The long-term stability of the concen- trated samples which were stored in a refrigerator and the long-term stability of the ICP-MS instrumentation was investi- gated by analysing a set of samples twice with an interval of 3 months. The first time was in October 1992 and the second time in January 1993. The samples on the filters were first dissolved in HN03 but the dissolution proved to be incomplete. Later a procedure was adopted where the filter was soaked in 1 ml of 1 + 1 HCl with gentle heating.The sample was then diluted with water for the ICP-MS measurements. Table 2 Operating parameters for ICP-MS R.f. power Cooling gas Aerosol carrier gas Sample introduction Measurement time Dwell time Range 1.35 kW 14 1 min-' 0.8 1 min-' 1 ml min-' 10-30 s per pulse counting mode 0.64 ms 50-65 ?TI/Z Results and Discussion Sampling and Storage of Samples A comparison of the soluble metal concentrations obtained from samples taken through the standard sampling device (A) and the isokinetic sampling line (B) is shown in Fig. 2. Although there is some variation in the results there is no indication that there would be a systematic difference between the two sampling lines. This confirms the results of Svoboda et aL2 that the isokinetic sampling line is no better than the normal one.The results also indicate that no loss or contamination occurred during the 3 months' storage of the samples. Preconcentration The results of the recovery tests are shown in Table 3. The recoveries are satisfactory provided the concentrations are well below the capacity of the column. The results for Fe are poor probably because it occurs in colloidal form or is hydrolysed. In the real preconcentration situation the non-filterable Fe is kept in solution because of the lack of oxygen which would oxidize it. The results of the comparison of the two different column types are shown in Fig. 3. It can be seen that in general the results for all elements agree well although the Dionex results are systematically somewhat higher.The reason for this is not known. Although the Waters column contains metal parts there is no sign of contamination. The low concentrations also agree well. In practical use the Dionex columns have been 80 - (a) B 60 - 40 - 20 - 0 80 i 60 40 20 0 0.15 0.12 0.09 0.06 0.03 B 0 5 10 15 20 25 Time/h Fig. 2 Metal concentrations of samples obtained from the Loviisa 1 shut-down using A standard and B isokinetic sampling lines in parallel for (u) Mn; (b) Ni; and (c) Co. Time scale in hours from a reference time716 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 Table 3 Results of the concentrations tests with a Waters Guard Pak column the yield of the elements and the total capacity of the column Parameter Fe Mn Ni c o Yield from 10 ml of water (YO) 44 100 99 100 Yield from 20 ml of water (%) 34 100 100 100 Yield from 50 ml of water (YO) 7 89 97 97 Capacit y/ng 560 1210 9800 27 Corresponds to x ml of water 10 94 92 90 more reliable the capillary connections of the Waters columns tending to be clogged at times.ICP-MS Determinations The results of the interference of Ni on Co when an area of 0.6m/z is used for integration are shown in Table4. This means that 1 ng of Ni corresponds to 7 x lo-' ng of Co. A Ni to Co ratio of 14000 would cause a positive error of 10% in the Co result. In the results presented the greatest Ni to Co ratio is 400. When a 0.4 m/z integration area is used no effect can be detected. The reproducibility of the ICP-MS results is shown in Table 5. The results indicate that the long-term stability of the concentrated samples is good.The detection limits for the different metals depend mainly on the variation of the blanks. Using the aluminium sampler and the 10 s measurement technique detection limits can be as good as 1 ng ml-' for Fe 0.1 ng ml-' for Mn 0.1 ng ml-' for Ni and 0.010 ng ml-' for Co. With concentration of a 11 sample to 2 ml the sample primary water detection limits are correspondingly Fe 2 Mn 0.2 Ni 0.2 and Co 0.02ng1-I. Naturally accurate results can only be obtained with concen- trations at ten times these levels but even then the limits are clearly below what is normally found in these samples. The concentrations of the soluble metal ions during the refuelling shut-downs of Loviisa 1 in 1992 are shown in Fig. 4. Some values of the concentrations in the primary water of the Loviisa 1 and Loviisa 2 power plants in 1993 during full power are shown in Table 6.The concentrations of the elements show similar behaviour to results reported from other reac- t o r ~ . ' ~ ~ * ~ ~ ' ~ The TV04 samples the water in the purification circuit before the ion exchangers and TV08 after the water has passed the ion exchangers. The results for soluble Ni and 60 (a) A V 15 45 75 105 Table 4 Effects of Ni on the concentration of Co Ni/ng ml- Co/ng ml- ' 100 200 400 0.008 0.01 1 0.025 Table 5 Transition metal concentrations (ng ml-') of samples taken during the Loviisa 2 (L02) shut-down. Parallel measurements made with an interval of 3 months Sample LO216 L02/10 LO2113 L02/16 L02/19 L02/41 L02/44 LO2147 LO2150 LO2153 LO2156 LO2160 Fe 0.11 1.7 4.5 1.2 11.5 14 29 18.5 32 35 30 - Fe 0.21 1.5 4 0.45 13.6 13.7 28 12 44 22 19 - Mn 0.91 2 2.6 3.6 2.3 3.9 4.3 7.7 11.1 30 28 23 Mn 0.89 2.2 3 3.7 2.2 3.7 3.7 5.4 7.5 18 18 14 Ni Ni 0.13 0.13 0.36 0.43 0.71 0.87 1.3 1.4 0.75 0.83 3.6 3.8 3.5 3.2 7.8 6.2 7.5 6 27 25 43 52 41 49 c o 0.0045 0.007 0.009 0.017 0.01 0.019 0.02 0.032 0.023 0.069 0.11 0.11 c o 0.004 0.006 0.01 1 0.017 0.008 0.018 0.017 0.024 0.014 0.048 0.074 0.076 Co are of the same order of magnitude as the full power values for other reactors as reported in the l i t e r a t ~ r e .~ * ~ * ~ * ~ ~ ~ ' ~ * ~ ~ The range of results is as follows Ni 10-300 ngl-' with a few values above 500 ng 1-'; and Co 0.4-10 pg 1-' with a few values considerably above these.The work is continuing with further development of the sampling and analysis methods. More results together with a discussion of the significance of the results will be published later. Conclusions A new method has been developed that enables the sampling and determination of Fe Mn Ni and Co in the primary water of PWRs using ICP-MS. The soluble and insoluble fractions are analysed separately. The fast transients in the concen- trations of Mn Ni and Co occuring during shut-down can be followed. This fact together with the detected decrease of the transition metal concentrations in the water going through the purification system seems to indicate that the sampling and 0.20 0.15 0.10 0.05 0 15 45 75 105 Tirne/h Fig. 3 Comparison of the results obtained by preconcentration of parallel samples during the Loviisa 2 shut-down using the columns A Dionex MetPac CC-1 and B Waters Guard Pak for (a) Fe; (b) Ni; (c) Mn; and (d) CoJOURNAL OF ANALYTICAL ATOMIC SPECTROM 0.5 (C) I 00 06 12 18 00 06 12 18 00 06 12 18 00 12 18 Timeth Fig.4 Soluble transition metal concentrations (a) Mn (b) Ni and (c) Co in the primary water of the Loviisa 1 power plant during the 1992 re-fuelling shut-down. Time in hours from noon the day before shut-down (July 18 1992) to July 21 1992. The reactor reached sub- criticality on July 18 1993 at 6.46 hours Table 6 Soluble transition metals (in ng 1-I) in the Loviisa 1 and Loviisa 2 primary water during full power in 1993 ;TRY JUNE 1994 VOL. 9 71 7 Sampling date 22-23 June 1993 23-24 June 1993 Loviisa I TV04 Loviisa 2 TV04 3-4 February 1993 4-5 February 1993 11-12 February 1993 16-17 February 1993 28-29 April 1993 29-30 April 1993 Loviisa 2 TVO8 Fe Mn 460 54 270 48 2300 112 2400 103 2000 101 1700 71 1260 81 73 10 Ni 17 8 171 171 169 110 117 22 c o 0.9 0.8 7.8 3.8 3.7 2.7 3.4 0.55 analysis is representative enabling the follow-up of concen- tration gradients occuring in different parts of the primary circuit.The detection limits are low enough to enable the determination of the elements during full power in all reactors. This study was commissioned by the Finnish Centre for Nuclear Safety (STUK) and has been funded by the Imatran Voima Power Company (IVO). Seija Suksi (STUK) Magnus Halin Thomas Buddas Risto Jarnstrom and Vesa Talvitie (IVO) Marjo Lauren and Jaana Rantanen (Reactor Laboratory) assisted in the performance of the work.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 References Bridle D. A. Brown G. R. and Johnson P. A. V. AEA Technology Winfrith UK unpublished results. Svoboda R. Ziffermayer G. Romanelli S. Kaufmann W. Sozzi L and Schaefer M. E. Water Chemistry of Nuclear Reactor Systems 3 British Nuclear Energy Society (BNES) London 1983 p. 331. Polley M. V. and Anderson P.-O. Water Chemistry of Nuclear Reactor Systems 5 British Nuclear Energy Society (BNES) London 1989 p. 71. Larger N. R. Mead S. Nichols J. L. Patel N. M. Lawson D. and Becket N. A. Water Chemistry of Nuclear Reactor Systems 5 British Nuclear Energy Society (BNES) London 1989 p. 63. Large N. R. Harper A. Ashmore C. B. Beckett N.A. Nichols J. L. Johnson P. A. V Bridle D. A. and Cake P. Proceedings of the 1991 JAIF Conference on Water Chemistry in Nuclear Power Plants Fukui City April 22-25 1991 p. 672. Eley C. D. Thomas D. M. Libaert D. F. and Cattle R. A. Water Chemistry of Nuclear Reactor Systems 6 British Nuclear Energy Society (BNES) London 1992 p. 224. Bridle D. A. Brown G. R. Cake P. and Staunton G. M. AEA Technology Winfrith UK Report AEA-TRS-2043 1991. Bridle D. A. and Cake P. Water Chemistry of Nuclear Reactor Systems 6 British Nuclear Energy Society (BNES) London 1992 p. 324. Raymond A. De Murcia A. and Dhainaut S. Anal. Chim. Acta 1987 195 265. Takiguchi H. Morishita S. Kasahara K. Fukuda F. Sakai K. and Meguro Y. Proceedings of the International Conference on Water Chemistry in Nuclear Power Plants Japan Atomic Industrial Forum Tokyo Japan 1988 p.168. Torrance K. and Gatford C. Talanta 1985 32 273. Torrance K. and Gatford C. Analyst 1986 111 359. Ruhle W. VGB Kraftwerkstechnik 1985 65 252. Amey M. D. H. and Brown G. R. AEE Winfrith Report Bird E. J. Bridle D. A. Amey M. D. H. Roofthoof R. De Ranter K. and Philippe R. Water Chemistry for Nuclear Reactor Systems 4 British Nuclear Energy Society (BNES) London 1986 p. 21. Bridle D. A. Brown G. R. and Hamacher P. presented at VGB-Konferenz Chemie in Kraftwerk Essen Germany October 1989 p. 39. Bridle D. A. Cake P. Symons W. J. and Katona J. AEA Technology Winfrith UK internal report. Abe K. Mizusaki H. Ohta H. Hemmi Y. Umehara R. Ooshima S. Fukuda F and Kashara K. Proceedings of the 1991 JAIF International Conference on Water Chemistry in Nuclear Power Plants Japan Atomic Industrial Forum Fukui City Japan 1991 p. 599. Miyazaki S. Ohshima S. and Ojima Z Proceedings ofthe 1991 JAIF International Conference on Water Chemistry in Nuclear Power Plants Japan Atomic Industrial Forum Fukui City Japan 1991 p. 584. Monahan J. Mead A. P. and Lawson D. Water chemistry of Nuclear Reactor Systems 4 British Nuclear Energy Society (BNES) London 1986 p. 95. Buckley D. Water Chemistry of Nuclear Reactor Systems 6 British Nuclear Energy Society (BNES) London 1992 p. 199. Schenker E. Francioni W. and Degueldre C. Water Chemistry of Nuclear Reactor Systems 6 British Nuclear Energy Society (BNES) London 1992 p. 133. AEEW-R2044 1986. Paper 3/06147G Received October 14 1993 Accepted February 25 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900713

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL O F ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 719 Signal Enhancement and Reduction of Interferences in Inductively Coupled Plasma Mass Spectrometry With an Argon-Trifluoromethane Mixed Aerosol Carrier Gas Isaac Platzner* Jose V. Sala Francis Mousty and Pier R. Trincherini Environment Institute Joint Research Centre 1-2 1020 lspra (VA) Italy Alberto L. Polettinit SEA Marconi Technologies Collegno (TO) Italy c/o Joint Research Centre T.P. 460 1-27020 lspra (VA) Italy The addition of trifluoromethane (CHFJ to the aerosol carrier gas in inductively coupled plasma mass spectrometry was assessed as a method of improving detection limits (DL) for elements such as As Se Cu and Zn in matrices containing interfering species. It was observed that in each case the analyte response was significantly increased with a coincident decrease in the blank signal.The improved DL for As in CI- and Ca2+ matrices (interferences 40Ar%I+ and 43Ca'60z were 0.02 and 0.04 ng ml-' compared with 0.65 and 0.28 ng ml-' respectively without CHF,; for 78Se (interference 40Af'8Ar+) 0.032 compared with 0.88 ng ml-'; for 63Cu in Na2S0 or Na,HPO matrices (interferences ,'AP3Na+ and 31P1602+) 0.022 and 0.089 compared with 0.35 and 0.53 ng ml-' respectively; and for 64Zn in an Na,HPO matrix (interferences H3'P1602+ and 31P160170+) 0.011 compared with 0.42 ng ml-' The reduction of the interference is attributed to competitive reactions between the matrix species and the CHF or species derived from it in the plasma. The analyte enhancement effect is not yet clear. It has been suggested that this effect is related to elements with ionization potential (IP) in the 9-11 eV region and is affected by organic compounds added to the aerosol carrier gas stream.Copper (IP 7.73 ev) Al (5.99 eV) Br (1 1.30 eV) and I (1 0.44 ev) are exceptions to this assumption. Analytical curves of the studied elements at low ppb and sub-ppb levels (in the interfering matrices) further demonstrated the advantage of adding CHF in trace elemental analysis. Keywords Inductively coupled plasma mass spectrometry; trifluoromethane addition; interference reduction; analyte signal enhancement; detection limits Intensive studies have been carried out in the past few years and are still under way at various laboratories on ways of improving the analytical performance of inductively coupled plasma mass spectrometry (ICP-MS).Particular attention has been given to improving the signal-to-blank ratio for different analytical applications in the range of up to 80 m/z where strong interferences are observed which originate from isobaric masses of the aerosol carrier (nebulizer) gas matrix compounds and their mutual interaction products. The experimental approaches to overcoming or partially reducing the interfering polyatomic species have recently been extensively reviewed by Evans and Giglio.' One of the simplest and most widely used methods is to add minor amounts of different gases such as hydrogen nitrogen oxygen air helium or xenon to one of the three argon gas flows mainly mixing with the aerosol carrier stream.Recently the addition of methane to the aerosol carrier flow was studied by Hill et al. who observed that interfering ions such as ArCl' ArO' ClO' and CeO' were reduced relative to the unmodified plasma. Allain et aL3 has shown that the addition of methane moderately enhanced the analyte response for As Se and Te. These workers also reported matrix enhancement effects on As Se Te Hg and Au in glycerol and glucose solutions. Signal enhancement through the addition of nitrogen to the argon outer flow was also observed by Lam and H ~ r l i c k . ~ In the present paper results are reported for the mutual effect of reducing the interfering blank and the signal enhance- ment achieved for several elements by mixing trifluoromethane with the argon aerosol carrier flow in a conventional ICP-MS instrument.Experiments with this aerosol medium were also performed by adding glycerol to several solutions and their blanks. To our knowledge this is the first report where reduction in interference accompanied by simultaneous *Visiting scientist from NRCN PO Box 9001 Beer-Sheva Israel. t To whom correspondence should be addressed. enhancement of the analyte signal has been observed upon mixing gases in the aerosol stream. Experimental Instrurnenta tion The experiments were performed on an inductively coupled plasma mass spectrometer Model PQ2 (VG Elemental Winsford Cheshire UK) equipped with standard Meinhard nebulizer. The addition of trifluoromethane (CHF,) to the argon aerosol carrier gas stream was controlled with a 0-5 ml min-' gas mass-flow controller Tylan General Model FC-260 (Swindon Wiltshire UK).It was introduced directly to the nebulized analyte through a modified glass port prior to the plasma torch. Materials All solutions of the elements were prepared from 1 mgml-' stock solutions (Aldrich Chemicals Milwaukee WI USA) diluted with 1 % HNO (Suprapur Merck Darmstadt Germany) in de-ionized water (MilliQ Millipore Bedford MA USA). The other chemicals were HC1 (Suprapur Merck); redistilled glycerol (pro analysi Merck); Na,S04 (Suprapur Merck); Na,HPO Ca(NO,) and NaCl (Carlo Erba Milano Italy) and CHF (Halocarbon or Freon 23 Sol Monza Italy). Procedures The concentration of all the elements was 100 ng ml-' in 1% nitric acid. The element solutions and their solutions in different matrices were studied without and with the addition of CHF3 over a range of 1-5 ml min-l.The matrix solution at variable flow rates of the CHF was measured as the blank in each experiment. Data collection started 3 min after gas mixing. All experiments were performed at a fixed distance between the plasma torch and the sampling interface.720 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 The following isotopic species were studied in 1% HNO (unless otherwise stated) 'Be; 27Al; 63Cu and 65Cu (5% HNO,); 64Zn 66Zn 67Zn 68Zn and 70Zn (5% HNO,); 69Ga and "Ga; 72Ge and 74Ge; 75As; 78Se and 82Se; 79Br and 8'Br; '"In; 'I6Sn 'I7Sn 'I8Sn and "'Sn; "'Sb and lz3Sb; 124Te 126Te '"Te and More complex matrices were used for the following elements for Cu 5% HC1 0.02mol I-' Na2HP04 and 0.02 mol I-' Na2S04; for Zn 5% HCl 0.02moll-' Na2HP04 and 0.02 mol I-' Na2S04; for As 0.02 and 0.04 mol I-' NaCl 0.25 0.5 and 1.0 moll-' glycerol 0.01 moll-' Ca(NO,) and 0.02 moll-' Na2HP04; for Se 0.25 0.5 and 1.0 mol I-' glycerol; and for Te 0.25 0.5 and 1.0 moll-' glycerol.Blank measurements were also carried out at m/z values of 48 51 52 53 54 55 59 and 77. The linearity of the response in the interfering matrices for element concentration ranges from 5 to 50 (As) 1 to 10 (Zn) 0.5 to 5 (Se) and 0.25 to 1 ng ml-' (As and Cu) were tested at the optimal signal-to- blank ratio with addition of CHF,. Each result given is an average of at least three measure- ments. The operating conditions of ICP-MS instrument are summarized in Table 1.l 3 q e ; 1271; 202Hg; 208pb; and 2 3 8 ~ . Results and Discussion Arsenic Arsenic is a mono-isotopic element of mass 75. In most of the natural samples where chlorine is available (as C1- or organic chloride) a strong interference from 40Ar35Cl+ is always observed. Various procedures have been adopted to eliminate this interference hydride generation in the analysis of water;s modified dissolution procedures of marine sediment$ precipi- tation of chloride with Ag for protein sample^;^ matrix separ- ation by gel filtration for serum;8 liquid chromotographic separation (anion exchange) of As species in urine;g910 addition of nitrogen to argon for correction using elemental and mathematical calculations in l~bster,'~ oyster tissue bovine liver and kale.14 Interferences of 59C0160 + in nickel al10ys'~ and of iron oxides and hydroxides in steel on 75As have also been observed.16 The effect of addition of CHF to a 100 ng ml-' solution of As without and with 0.02 and 0.04 mol I-' NaCl and to the blank is shown in Table 2 and Fig.1. It can be seen that CHF enhances the "As+ signal in all the three experiments by a Table 1 Experimental and operating conditions ICP Nebulizer R.F. power/kW Outer gas flow rate/l min-' Intermediate gas flow rate/l min-' Aerosol carrier gas flow rate/l min-' CHF gas flow rate/ml min-' Solution uptake rate/ml min-' Spray chamber temperaturePC Mass spectrometer Instrument Sampler orifice (nickel)/mm Skimmer orifice (nickel)/mm Interface pressure/mbar* Analyser pressure/mbar Data acquisition parameters Mode Channels per m/z Interval between channels Dwell time/ms Acquisition time/s Measurements per sample Meinhard 13 1.35 0.5 0.70-0.90 0-5.0 2 0-10 VG/PQ2 1 0.5 1 1.8 x Peak jumping 3 5 10.24 3&60 3 * 1 mbar= 1 x 10' Pa. factor of 4-5 with a maximum at a flow of 2-3 ml min-'.The opposite effect is observed for the blank solutions the high blank counts in the presence of 0.02 and 0.04moll-' NaCl are reduced to a minimum at about the same CHF flow rate. It is not immediately obvious why the signal for As' increases but it is easier to understand the effect on the blank. Without CHF the blank counts are due to 40Ar35Clf ions. In the presence of CHF an efficient competitive reaction could take place increasing ArF+ relative to ArCl'. The net effect is that the relative enhancement for a 0.04 moll-' concentration NaCl is >9 times higher (37.4 to 4) than without NaCl (see Table 2).The 'relative enhancement' (RE) is defined as the ratio of the sample counts to the blank counts when CHF is added (ISIb)cF divided by the ratio of the sample counts to the blank counts when CHFJ is absent (IsI,JoY [RE= (IsIb)CF/(lsIb)O]. The blank at m/z 59 (ArF') and at m/z 77 (40Ar37C1+) was measured with the addition of CHF3. A continuous increase of ArF+ from 150 counts s-' (at 0 ml min-') to 3470 (at 5 ml min-') was observed supporting the decrease in the amount of ArCl' owing to a competitive mechanism. The blank at m/z 77 followed exactly the behaviour of the blank at m/z 75 except that the signal intensities were only one third of those at m/z 75 proving that the blank is ArC1' (the isotopic ratio 35C1:37C1 is approximately 3:l).The matrix effect of NaCl which reduces the As' signal at 0.04 moll-' to 80% (at a CHF flow rate of 2 ml min-') is of minor importance. The same combined effect of As+ signal enhancement and parallel reduction in the blank signal was also observed for the 43ca1602 + interference originating from calcium nitrate. It is possible that atomic fluorine in the plasma reacts with Ca-containing species to yield different products. In this case the combined effect is a steady relative enhancement of about 10 for CHF flow rates of 2mlmin-' and above as shown in Table 2. Selenium and Tellurium The most abundant Se isotope 80Se (49.7%) is completely hidden by the strong 40Ar2+ interference the presence of chloride (40Ar37Cl+) interferes with 77Se (7.6%) 40Ar38Ar+ interferes with 78Se (23.6%) while considering 82Se (9.2%) there was no evidence of interference from krypton which could be an impurity in the argon used. The interferences in biological materials were eliminated by generation of selenium hydride and determination of the element by isotope dilution analy~is,'~ in coal by slurry nebulization'8 and in urine by addition of nitrogen to any of the three gas streams." Addition of CHF reduces the blank at m/z 78 has no effects (as expected) on the blank at m/z 82 and causes a 3-4-fold increase in the 78Se+ and "Se+ signal; consequently a relative enhance- ment of up to 10.8 is observed for 78Se and 2.7 for 82Se.The reduction in the blank at m/z 78 is consistent with the assumption of CHF reacting with argon.Thus it can be concluded that the addition of CHF solves the interference problems in quantitative determination of Se when using the second most abundant isotope (78Se). The method is compar- able to hydride generation in sensitivity and preferable in simplicity of experiment. The data are summarized in Table 3 and the enhancement for 78Se is shown in Fig. 2. It should be noted that when nebulizing a 1% solution of HNO in the presence of 2 ml min-' of CHF an increase in the blank is observed at m/z 77 relative to a zero flow rate of CHF,. The effect is more evident when increasing the concentration of HN03 to 5%. This indicates the formation of an inter- ference owing to the simultaneous presence of CHF and nitrogen in the plasma an interference that is attributed to Tellurium does not suffer from interferences except for its minor isotopes from "'Sn and 124Sn. The presence of xenon as an impurity in the argon used would interfere with the 12CF14N160 + 2 -JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 Table 2 Addition of CHF to As plus matrix and matrix only solutions 72 1 Parameter 100 ng ml-' 75As in 1% HNO,/counts s-' Blank (1% HNO,)/counts s-' Signal-to-blank ratio Relative enhancement 100 ng ml-' 75As in 0.02 rnol 1-' NaCl solution/counts s-' Blank (0.02 mol 1-' NaCl)/counts s-' Signal-to-blank ratio Relative enhancement 100 ng ml-' 75As in 0.04 moll-' NaCl solution/counts s-' Blank (0.04 mol 1-' NaCl)/counts s-' Signal- to-blank ratio Relative enhancement 100 ng ml-' "As in 0.01 mol 1-' Ca(NO,) solution/counts S - I Blank (0.01 rnol 1-' Ca(NO,),/counts s-' Signal-to-blank ratio Relative enhancement 100 ng ml-' 75As in 0.02 mol 1-' Na,HPO solution/counts S-' Blank (0.02 mol 1-' Na,HPO,)/counts s-' Signal-to-blank ratio Relative enhancement CHF flow rate/ml min-' 18 1 0 37541 144 26 1 1 .o 3 1597 595 53 1 .o 26905 2043 13 1 .o 60823 4208 14 1 .o 39266 416 94 1 .o 1 93152 126 739 2.8 8098 1 229 3 54 6.7 82515 456 181 13.7 150149 1787 84 5.8 96340 675 143 1.5 2 167737 162 1035 4.0 147480 164 899 16.9 133620 27 1 493 37.4 206307 1522 136 9.4 211719 1589 133 1.4 3 168737 45 1 374 1.4 142546 249 572 10.8 130135 355 367 27.8 150258 1072 140 9.7 226500 1813 125 1.3 4 124150 589 21 1 0.8 112051 365 307 5.8 101451 520 195 14.8 929 18 64 1 145 10.0 193749 2119 91 1.6 5 86930 550 158 0.6 78213 457 171 3.2 72030 646 112 8.5 54083 403 134 9.3 161458 2476 65 0.7 x I /-A- - C - I I I I I 0 1 2 3 4 5 0 1 2 3 4 5 CHF flow rate/ml min-' Fig.1 Effect of addition of CHF (a) on 75As and (b) on the blank at m/z 75 for different concentrations of NaCl A 1% HNO,; B 0.02 moll-' NaCl; and C 0.04 mol I-' NaC1. most abundant Te isotopes ("'Te and l3'Te) but this was not observed. The addition of CHF as shown in Table 3 has only a slight enhancement effect on this element without affecting the blank. Copper The behaviour of Cu towards the addition of CHF was studied in solutions of 5% HNO 0.02 moll-' Na2S04 0.02 moll-' Na2HP04 and 5% HC1. All of the results for Cu are summarized in Table4.In HNO no interferences were expected and the addition of CHF follows the pattern as with elements such as In Pb and U i.e. a slight increase followed by a strong decrease when the CHF flow rate is increased. The situation is different for 63Cu (69.2%) in an Na2S04 matrix. At a CHF flow rate of zero a strong interference of 40Ar23Naf is observed which disappears even with low flow rates of CHF,. The count rate for the element also increased consequently at a 1-2mlmin-' CHF flow rate the relative enhancement is between 28 and 30. The observed decrease in the blank signal in this case is consistent with the competition reactions of argon with the matrix elements uersus the reactivity with CHF,. The removal of the 40Ar23Na+ interference in samples of waste water was achieved by preconcentration and matrix removal using an iminoacetate resin,Ig in serum by size-exclusion separation followed by 63Cu 65Cu isotope ratio determination2' and by subtraction of a synthetic blank con- taining the matrix elements.21 The blank interferences at m/z 65 in this experiment are mostly ,,SI6O 2 + 32S'60170+ 32S33S+ and H32S1602+ which probably exhibit low reactivity with CHF except for a possible abstraction of H from the H32S1602+.As the element count rate increase with addition of CHF is very moderate the maximum enhancement effect for this isotope is only 2.9. In the Cu-Na2HP0 s stem there are two interferences on the blank at m/z 63 40Ar Na+ and 31P1602+. Comparing this blank with the blank in the Cu-Na2S04 experiment it is evident that about 1 x lo4 counts s-l are contributed by 40Ar23Na+ and about 4 x lo3 counts s-l by 31P1602+.The first interference disap- pears almost completely with a 1 ml min-' addition of CHF whereas PO2+ is apparently not affected. A relative enhance- ment of 9.5 is calculated for this system. The blank level at m/z 65 is almost constant and the 65Cu+ signal is enhanced only by a factor less than 4 so that the relative enhancement is about this value. The addition of CHF to Cu-HCl solutions provides an interesting case. The Cu+ signal is enhanced to a maximum at a 1-2mlmin-' flow rate of CHF,. The blank HCl solution at a CHF flow rate of zero has no interferences at 63 and 65 m/z22 but addition of CHF increases the blank counts maintaining a blank (63) blank (65) ratio very close to the 35C1:37C1 ratio. It was therefore assumed that these interferences are C03'Clf and C037Cl+ their intensity depending on the availability of carbon i.e.on the CHF flow rate. 7 3722 Table 3 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 Addition of CHF3 to Se or Te in HNO and HNO solutions > r 10 G cr c CHF3 flow rate/ml min-' - 5 C F _ _ B t 1 I r3 " e Parameter 100 ng ml-l 78Se in 1% HNO,/counts s-' Blank (1% HNO,)/counts s-' Signal-to-blank ratio Relative enhancement 100 ng ml-1 82Se in 1% HNO,/counts s-' Blank (1% HNO,)/counts s-' Signal-to-blank ratio Relative enhancement 100 ng ml-' IZ4Te in 1% HNO,/counts s-l Blank (1% HNO,)/counts s-' Signal-to-blank ratio Relative enhancement 100 ng ml-l lZ6Te in 1% HNO,/counts s-' Blank ( 1 YO HNO,)/counts s - Signal-to-blank ratio Relative enhancement 100 ng ml-I IZ8Te in 1% HNO,/counts s-l Blank (1% HNO,)/counts s-' Signal-to-blank ratio Relative enhancement 100 ng ml-' 130Te in 1% HNO,/counts s-' Blank (1 % HNO,)/counts s-' Signal-to-blank ratio Relative enhancement 0 13601 3963 3.4 1 .o 3870 63 61.4 1 .o 4953 832 6.0 1 .o 16290 18 905.0 1.0 27856 49 568.5 1 .o 30351 59 514.4 1 .o 1 28237 2603 10.8 3.2 9567 100 95.7 1.6 11128 953 11.7 2.0 40574 23 1764.1 1.9 69259 87 796.1 1.4 75139 117 642.2 1.2 2 36372 985 36.9 10.8 13462 80 168.3 2.7 11763 69 5 16.9 2.8 43876 22 1994.0 2.2 75024 64 1172.3 2.1 8 1346 70 1162.1 2.3 3 26304 746 35.3 10.3 10003 114 87.7 1.4 7004 392 17.9 3.0 26466 12 2205.5 2.4 449 15 36 1247.6 2.2 49184 37 1329.3 2.6 4 18072 668 27.1 7.9 6826 96 71.1 1.2 3758 221 17.0 2.9 14026 13 1078.9 1.2 23980 28 856.4 1.5 25977 21 1237.0 2.4 5 13111 650 20.2 5.9 4802 125 38.4 0.6 2 143 126 17.0 2.9 8381 11 761.9 0.8 14210 18 789.4 1.4 15512 16 969.5 1.9 .- 40 1 I Fig.2 the blank in 1% HN03. Effect of addition of CHF on A 78Se in 1% HNO and B on Zinc This element was studied in matrices of 5% HCl 5% HNO 0.02 mol 1-1Na2S04 and 0.02 mol 1-lNa2PO4 The experi- ments with the 5% acid solutions which are slightly higher concentrations than normally used in ICP-MS applications were carried out to increase the formation probability of yet unknown interferences originating from HCl or HNO reacting with CHF in the 63-70 m/z range. Such interferences have been observed in the case of Cu with 5% HC1.For Zn isotopes no evidence for interferences of this type were observed. Only for 64Zn (the most abundant isotope 48.6%) in the Na2HPO4 solution has a relative enhancement owing to CHF been observed. There is an increase in the signal intensity for this element parallel to the decrease in the blank with an overall relative enhancement of about 9 for a CHF flow rate of between 2-3 ml min-l. The possible interferences at m/z 64 are H31P1602+ and 31P160170+. The results are presented in Table 5. Methods for reducing phosphate inter- ferences on 64Zn have not been considered in the literature. Concerning the Na2S04 matrix CHF did not affect the strong 32S1602+ and 34S1602+ interferences at m/z values of 64 and 66 and also no effect was observed on the 64Zn and %Zn signals.The various analytes the matrices used their concentrations and the consequent interferences for the elements discussed above are summarized in Table 6. Other Elements and Blanks The effect of addition of CHF on 100ngml-' solutions in 1% HNO of a series of elements listed under Experimental were studied. Taking the relative enhancement as a criteria for enhancement by CHF the following cases (apart from the already discussed As Se Te Cu and Zn) were identified. (i) An increase in the ion signal to a maximum at a CHF flow rate of about 2rnlmin-l followed by a decrease in the signal at higher flow rates. The blank intensity increased with increasing CHF flow rate. In this category are Be and Al. The sample counts for Be were 3 x lo5 4.1 x lo5 9.1 x lo5 6.2 x lo5 and 3.7 x lo5 counts s-l and the blank counts 100 200 900 1500 1700 and 1200 counts sK1 for the addition of 0 1 2 3 4 and 5 mlmin-1 CHF respectively. For A1 the sample counts were approximately the same as for Be but the blank increased from 5000 counts s-l at zero CHF flow rate to 55 000 counts s-l at 4 ml min-l CHF addition.The blank increase in both cases could be a result of stripping from the cones since both elements are present in the mass calibration solution. (ii) A continuous decrease in ion signal along the whole CHF flow rate range followed by a stable low or slightly increasing blank signal. In this category are Ga Ge Br In Sn Sb I Hg Pb and U. (iii). A strong increase in the blank counts in 1% HNO for 0-3mlminP1 additions of CHF were observed at m/z values of 48 (Ti) (800-2.2 x lo4 counts s-'); 51 (V) (200-2.4 x lo4 counts s-'); 52 (Cr) (1000-3 x lo5 counts s-'); 54 (Fe) (9000-9 x lo4 counts s-'); and 59 (Co) (500-104 counts s-I).Moderate blank increases under the same conditions were observed at m/z values of 53 (Cr) (100-4000 counts s-'); 55 (Mn) (700-3000 counts s-'); and 57 (Fe) (900-2000 counts s-I); The polyatomic interferencesJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 723 Table 4 Addition of CHF to Cu plus matrix and matrix only solutions CHF flow rate/ml min-' Parameter 100 ng ml-' 63Cu in 5% HNO,/counts s-' Blank (5% HNO,)/counts s-' Signal-to-blank ratio Relative enhancement 100 ng ml-' 65Cu in 5% HNO,/counts s-' Blank ( 5 % HNO,)/counts s - ' Signal-to-blank ratio Relative enhancement 100 ng ml-' 63Cu in 5% HCl/counts s-' Blank (5% HCl)/counts s-' Signal-to-blank ratio Relative enhancement 100 ng ml-' 65Cu in 5% HCl/counts s-' Blank (5% HCl)/counts s-' Signal-to-blank ratio Relative enhancement 100 ng ml-' 63Cu in 0.02 mol 1-' Na2S0 solution/counts s-' Blank (0.02 mol 1-' Na,SO solution)/counts s-' Signal-to-blank ratio Relative enhancement 100 ng ml-' 65Cu in 0.02 moll-' Na,HPO solution/counts s-' Blank (0.02 mol 1-' Na2HP04 solution)/counts s-' Signal-to-blank ratio Relative enhancement 100 ng ml-' 63Cu in 0.02 mol I-' Na2S04 solution/counts s-' Blank (0.02 moll-' Na2S04 solution)/counts s-' Signal-to-blank ratio Relative enhancement 100 ng ml-I 65Cu in 0.02 mol I-' Na2S0 solution/counts s-' Blank (0.02 mol 1-' Na2S0 solution)/counts s-' Signal-to-blank ratio Relative enhancement 0 226698 524 433 1 .o 102373 294 348 1 .o 216578 619 350 1 .o 96393 273 353 1 .o 102360 14644 7.0 1 .o 44578 1013 44 1 .o 176163 10419 17 1 .o 88129 8467 10.4 1 .o 1 318670 599 532 1.2 140265 332 422 1.2 333522 1466 228 0.7 145161 533 272 0.8 155568 4350 35.8 5.1 703 17 860 82 1.9 260796 543 480 28.4 121691 9466 12.9 1.2 2 310320 603 515 1.2 134219 304 442 1.3 295976 2679 110 0.3 128320 968 133 0.4 382601 5731 66.8 9.6 164782 1176 140 3.2 237874 47 1 505 29.9 109819 3685 29.8 2.9 3 183970 467 394 0.9 79458 210 378 1.1 166979 2750 61 0.2 73145 921 79 0.2 324351 5324 60.9 8.7 136422 896 152 3.5 145177 566 256 15.2 68041 2583 26.3 2.5 4 140840 379 372 0.9 45487 159 286 0.8 103656 2741 38 0.1 45348 930 49 0.1 196744 4295 45.8 6.6 84775 806 105 2.4 84264 878 96 5.7 40097 2714 14.8 1.4 5 69896 388 180 0.4 30424 166 183 0.5 73228 2605 28 0.1 32323 874 37 0.1 129092 3944 32.7 4.7 53502 968 55 1.3 52734 2918 18 1.1 25062 2664 9.4 0.9 Table 5 Addition of CHF to Zn-Na,HPO and Na2HP0 solutions CHF3 flow rate/ml min-' Parameter 0 1 2 3 4 5 100 ng ml-' 64Zn in 0.02 mol 1-' Na,HPO solution/counts s-' 39034 56640 125526 118492 83950 59523 Signal-to-blank ratio 11.1 62.2 96.7 100.3 82.1 63.1 Relative enhancement 1 .o 5.6 8.7 9.1 7.4 5.7 Blank (0.02 mol 1-' Na2HP04 solution)/counts s-l 3524 910 1298 1181 1023 943 which originate from CHF (in addition to those originating from HN0322) presumably are m/z 48 36Ar12C+; m/z 52 40Ar12C+; m/z 53 40Ar13C+; m/z 54 F2I60+; m/z 55 36ArF+; m/z 57 38ArF+; and m/z 59 40ArF+. Effect of Glycerol Following the results from Allain et al.on the enhancement effect of glycerol on As Se and Te sol~tions,~ glycerol solutions were studied with the addition of CHF,. No improvement of ion signal intensities was observed in the presence of CHF,. Analytical Applications of Addition of CHF It is immediately evident that the reduction of an interference and under the same conditions an increase in the analyte signal will improve the detection limits (DL). The DLs for As '%e 63Cu and 64Zn in various matrices are summarized in Table 7. The following expression was used DL=3 x (SD) x Cs/(Is-IB) where (SD) C Is and IB are the stan- dard deviation of the blank (in counts s-') concentration of the standard counts of the standard and counts of the blank (both in counts s-') respectively.It should be noted that the detection limits improved by factors of 7-38 except for As in 1% HNO where the factor is only 2. A further important point is the stability of the plasma with the addition of CHF,. The following cases have been studied at CHF flow rates of 0 and 2mlmin-I (a) As in the concentration range 5-50 ng m1-I; (b) As 0.25-1 ng m1-I; (c) 78Se and 82Se 0.5-5 ngml-I (total Se); (d) 63Cu 0.25-1 ng ml-1 (total Cu); and (e) 64Zn 1-5 ng ml-1 (total Zn). (a) Arsenic was determined in the National Institute of Standards and Technology (NIST) Standard Reference Material (SRM) 3172 Multielement Mix B Standard Solution which contains 200+ 1 jig ml-1 of As and nine other metallic elements (Ba Ca Co Cu Pb Se Ag Sr and Zn) in the range724 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 Table 6 Summary of analytes matrices and their interferences ~ Analyte 75As 77Se 78Se 82Se 63cu 65cu 66Zn Matrix and concentration NaCl(0-0.04 moll-') Ca(N03)2 (0.02 mol 1-' 1 Na2HP04 (0.02 mol 1- ') + CHF NaCl(0-0.04 moll-') HN03 ( 1 YO) + CHF - HCl(5Yo) + CHF3 Na2S04 (0.02 mol 1-') Na2HP04 (0.02 mol 1-l) Na,S04 (0.02 mol 1- ') HC1( 5%) + CHF3 Na2HP04 (0.02 mol 1 - '11 Na2S04 (0.02 mol I-') Na2HP04 (0.02 mol 1-':I Na2S04 (0.02 mol 1-') 1 Interference 40Ar35C1 + 43ca160 + 12c31p'602+ 40Ar37Cl + 40Ar38Ar + 12CF_14N1602+ - 12C16035c1+ 40Ar23Na + 4 0 ~ ~ 2 3 ~ ~ + 31p1602+ 2 ~ 1 6 0 3 7 ~ 1 + 12 c 18 0 35 c1+ 3 3 ~ 1 6 0 + 32 16 17 + ~ 3 2 ~ 1 6 0 ~ + 3 2 ~ 3 3 s + 31p160180f 9 2 s o o 32s'602+ ~31p160~+ 31p160170+ 3 4 ~ 1 6 0 + 2 Table7 Effect of addition of CHF on detection limits in different matrices; detection limits (3 x SD) are given in ng ml-' CHF3 flow rate/ml min-' Analyte and matrix As-1% HNO As-0.02 moll-' NaCl As-0.04 moll-' NaCl As-0.01 moll-' Ca(N03)2 78~e-1% HNO 63Cu-0.02 mol 1-' Na2S0 63Cu-0.02 mol 1-' Na2HP04 64Zn-0.02 moll- Na2HP04 0 0.032 0.23 0.65 0.28 0.88 0.35 0.53 0.42 2 0.014 0.01 0.02 0.04 0.032 0.022 0.089 0.01 1 10-500pgml-' in a 5% HNO matrix.The standard was diluted to yield a 18.93 ng ml-' of As solution in 1% HNO and 760 pg ml-' of NaCl were added to simulate potable water. The results are summarized in Table 8. The table reveals that the sensitivity in the presence of CHF is almost six times higher compared with the absence of CHF and the accuracy and precision are comparably or slightly improved 18.89 & 0.05 uersus 18.51 kO.19 ng ml-' of As (calculated 18.93 ng ml-' of As).It should be pointed out that other potable water samples (laboratory working standards) were also analysed. These samples contain variable amounts of As and other cations (Cu Table 8 Determination of As in NIST/SRM 3 172/B in the presence of 760 ppm of NaCl(1 YO HNO solution) CHF flow rate/ml min-' Parameter 0 2 Blank*/counts s-' 1130 & 30 258 f 1 Mean sensitivityt/counts s- ' per 343 1994 ng ml-' of As As concentration measured/ng ml-l 18.51 f0.19 18.89 f 0.05 Regression coefficient 0.9999 1 1 As concentration calculatedjng ml-' 18.93 ~~ ~ * Quoted errors are & 1 SD.7 Mean sensitivities are for blank subtracted analytical curves. Zn Cd Cr Hg Ni Pb and NH4+) and anions (S04-2 NO3- and P205) in a 760 pg ml-' C1- solution. The sensitivity of the blank subtracted analytical curve without CHF was 330+ 10 counts s-' per ng ml-' of As. Whenever this was sufficient excellent agreement was obtained for As between the present measurements and the mean values (as established by nine independent laboratories). When CHF was used to enhance the sensitivity the present results for As were 10-20% higher than the mean values. This erratic difference is attributed to an interference introduced at m/z 75 by one or more components in solution. Solutions containing each of the matrix components at the same concentrations as in the laboratory standard have been tested separately with and without CHF3 for increments in the blank counts at m/z 75.Increments in the blank signal have been observed only for a phosphate solution (0.02 moll-' Na2HP04) and the inter- ference was attributed to 12C3'P160 2'. The enhancement effect for a 100 ng ml-' As solution with 0.02 moll-' Na2HP04 was studied and the results are also given in Table 2. It is evident that As solutions can also be analysed in the presence of phosphates against an appropriate blank. (b) Solutions of 0.25,0.5 and 1 ng ml-' of As in 0.04 moll-' NaCl solution and 1% HNO produced without CHF a blank of 2683 counts s-' and non-linearly increasing readings between 2900 and 3400 counts s-' whereas with 2 ml min-l CHF the blank was reduced to 411 counts s-l and the blank subtracted analytical curve showed a sensitivity of 1250+38 counts s-' per ng ml-' of As.(c) For 0.5 1 and 5 ng ml-' solutions of Se in 1% HNO without CHF3 at m/z 78 a blank of 5004countss-' and solution readings randomly scattered between 5800 and 6350 counts s-' were observed. With CHF the blank was reduced to 818 counts s-l and the blank subtracted analytical curve yielded a sensitivity of 444 & 39 counts s-' per ng ml-' total of Se. For 82Se without CHF3 the blank was 94 counts s-' and the sensitivity 30+ 15 counts s-' per ng ml-l of total Se whereas with CHF the blank was 88countss-' but the sensitivity increased more than six times to 193 & 23 counts s-' per ng ml-' of total Se. (d) For 0.25-0.5 and 1 ngml-' of Cu in a 0.02moll-' Na2S04 solution and 1% HNO without CHF the blank at 63Cu was 8500 counts s-' and solution readings were randomly scattered between 8200 and 9700 counts s-' whereas with CHF the blank was greatly reduced to 308 countss-' and the blank subtracted analytical curve showed a sensitivity of 1888+25 counts s-l per ng ml-' of total Cu.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 725 Table 9 Calibration of As Se Cu and Zn in interfering matrices with a CHF3 flow rate of 2 ml min-'; quoted errors are f 1 SD Analyte (b) 75As+0.04 mol I-' Nucl- Blank 0.24 ng ml- ' As 0.47 ng ml - ' As 0.97 ng ml-' As (c) "Se- Blank 0.49 ng ml-' Se 0.99 ng ml-' Se 5.04 ng ml-' Se (c) "Se- Blank 0.49 ng ml-' Se 0.99 ng ml- ' Se 5.04 ng ml-' Se ( d ) 63Cu in 0.02 mol 1-' Nu2S04- Blank 0.25 ng ml-' Cu 0.50 ng ml-' Cu 1 ngml-' Cu (e) 64Zn in 0.02 mol I-' Na2HP04- Blank 1.01 ngml-' Zn 4.99 ng ml - ' Zn 10.05 ng ml - ' Zn Signal/ counts s-' 41 1 715 992 1602 Mean sensitivity* Regression coefficient 818 1020 1297 3023 Mean sensitivity Regression coefficient 88 179 304 977 Mean sensitivity Regression coefficient 308 778 1240 2225 Mean sensitivity Regression coefficient 2507 3735 7674 13131 Mean sensitivity Regression coefficient Blank subtracted signal/counts s - ' 0 304 58 1 1191 0 202 479 2505 0 91 216 889 0 470 934 1917 0 1228 5167 10624 Sensitivity/counts s - ' per ppb of element - 1294 1226 123 1 1250fi 38 0.99993 - 409 486 437 444k39 0.99976 - 185 219 176 193 1 2 3 0.99977 - 1880 1686 1917 1888 2 25 0.99981 - 1209 1036 1057 1 100 Ifr 94 0.99955 * Mean sensitivities are for blank subtracted analytical curves.(e) For 1 s and 10 ng ml-' of Zn the blank at 64Zn without CHF was 3700 counts s-' and the sensitivity 414k 38 counts s-'per ng ml-' of total Zn as compared with 2507 counts s-' and 1100+94 counts s-' per ng ml-' of total Zn with CHF respectively. The data with CHF for calibrations (b)-(e) are summarized in Table 9. General Comments For the cases where the net effect of addition of CHF is positive such as As Se Cu and Zn there is a plausible explanation for the decreasing intensity in the blank signal. It is most probable that the corresponding polyatomic interfering species 40Ar35Cl+ 40Ar38Ar+ and 40Ar23Na+ at m/z values of 75 78 63 are produced with a lower reaction rate owing to the competition of species originating from CHF with compo- nents in the matrices.For the blank at m/z 64 in an Na2HP04 solution two interferences were suggested 31P160170f which are considered as non-reactive (see Cu-Na,HPO,) and H31P160,+ which can be eliminated by CHF by abstraction of H thus the blank is significantly reduced. The interpretation of the element count rate increase in the presence of CHF is not straight forward. Allain et aL3 observed a signal enhancement in a 1 moll-' glycerol (and also glucose) solution for Hg (6OO%) Au (325%) Se (250%) As (110%) and Te (190%). For other elements such as Bi Co Eu Ho I In La Nb Ni Pb Pt Sn Sr T1 and U the change in the signal was between 90 and 110% (all relative to signal intensity without glycerol).In their work the effect of glycerol on the blank was not reported. Allain et al. also observed that addition of methane to the argon of up to 6% (v/v) increased the ion signals for As Se and Te. They tried to correlate the ionization enhancement with the IP of the element noting that the effect is observed for elements with IP values of between 9 and 11 eV. Iodine (10.44 eV) was an exception as was also Br (11.30 eV). Their argument was that addition of organic com- pounds modifies the ionization equilibrium over a limited range of energies. The IP for C (11.20 eV) is slightly above this range. In the present experiments no enhancement was observed for Hg (the element most affected in the work by Allain et al.) or for iodide and bromide.An effect on Zn (9.39 eV) and on Be (9.32 eV) which supports their hypothesis was noted but there was also an effect on Cu and Al which have IP values of only 7.73 and 5.99 eV respectively. A further parameter that affects the absolute intensities of the measured elemental and blank currents at the different flow rates of CHF is the relative plasma torch position along the x-axis (the axis of the quadrupole sampling cone and torch). On moving the torch along this axis variable counts for blank and for element signals were observed. This effect was not studied systematically because it was difficult to move the torch without shutting off the discharge. In the present work a 'relative enhancement' has been defined. Generally it was observed that this effect was maintained independent of726 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 the motion of the torch. As this parameter was not optimized it is fairly likely that larger enhancements could be observed by fine tuning the torch position. The authors express their thanks to Professor S. Facchetti Head of the Soil Water and Waste Unit Environment Institute J. R. C. Ispra for his helpful comments and to Dr. H. W. Muntau Head of the Chemistry of the Aquatic Systems Section at the above Unit for supplying the potable water laboratory working standards. References Evans E. H. and Giglio J. J. J. Anal. At. Spectrorn. 1993 8 1. Hill S. J. Ford M. J. and Ebdon L. J. Anal. At. Spectrom. 1992 7 1157. Allain P. Jaunault L. Mauras Y. Mermet M. and Delaporte T.Anal. Chem. 1991 63 1497. Lam J. W. H. and Horlick G. Spectrochim. Acta. Part B 1990 45 1313. Branch S. Corns W. T. Ebdon L. Hill S. and ONeill P. J. Anal. At. Spectrom. 1991 6 155. McLaren J. W. Beauchemin D. and Berman S . S . J. Anal. At. Spectrom. 1987 2 277. Lyon T. D. B. Fell G. S. Hutton R. C. and Eaton A. N. J. Anal. At. Spectrom. 1988 3 265. Lyon T. D. B. Fell G. S. Hutton R. C. and Eaton A. N. J. Anal. At. Spectrom. 1988 3 601. 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Heitkemper D. Creed J. Caruso J. A. and Fricke F. L. J. Anal. At. Spectrom. 1989 4 279. Sheppard B. S. Shen W. Caruso J. A. Heitkemper D. and Fricke F. L. J. Anal. At. Spectrorn. 1990 5 431. Branch S. Ebdon L. Ford M. Foulkes M. and O'Neill P. J. Anal. At. Spectrom. 1991 6 151. Wang J. Evans E. H. and Caruso J. A. J. Anal. At. Spectrom. 1992 7 9.29. Ridout P. S. Jones H. R. and Williams J. G. Analyst 1988 113 1383. Munro S. Ebdon L. and McWeeny D. J. J. Anal. At. Spectrom. 1986 1 211. McLeod C. W. Date A. R. and Cheung Y. Y. Spectrochim. Acta Part B 1986 41 169. Vaughan M. and Horlik G. J. Anal. At. Spectrom. 1989 4 45. Buckley W. T. Budac J. J. Godfrey D. V. and Koenig K. M. Anal. Chem. 1992 64 724. Ebdon L. Foulkes M. E. Parry H. G. M. and Tye C. J. Anal. At. Spectrom. 1988 3 753. Heithmar E. M. Hinners T. A. Rowan J. T. and Riviello J. M. Anal. Chem. 1990 62 857. Lyon T. r). B. and Fell G. S. J. Anal. At. Spectrom. 1990 5 135. Vanhoe H. Vandecasteele C. Versieck J. and Dams R. Anal. Chem. 1989 61 1851. Tan S. H. and Horlick G. Appl. Spectrosc. 1986 40 445. Paper 3 1074431 Received December 20 1993 Accepted March 16 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900719

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 727 Reduction of Polyatomic interferences in inductively Coupled Plasma Mass Spectrometry by Selection of Instrumental Parameters and Using an Argon-Nitrogen Plasma Effect on Multi-element Analyses* Francisco LabordaJ Malcolm J. Baxter Helen M. Crews$ and John Dennis Ministry of Agriculture Fisheries and Food CSL Food Science Laboratory Norwich Research Park Colney Norwich Norfolk UK NR4 7UQ The effect of instrumental parameters and argon-nitrogen plasmas on polyatomic ion formation has been studied in order to reduce their magnitude in routine multi-element analysis without losing detection capability. Special emphasis was placed on the chlorine based polyatomic interferences on V Cr Zn As and Se. A significant reduction in signals from polyatomic ions was attained by using a high aerosol carrier gas flow rate (0.955 I min-') instead of the default flow rate (0.755 I min-') or by adding nitrogen (8%) to the aerosol carrier flow.The ArClf interference produced by 0.05% CI (the maximum concentration expected in digested food stuff samples) was effectively removed by both methods and CIO' and C102+ by addition of nitrogen. Detection limits for elements along the mass range (from Li to U) were on average 2-3 times higher with the mixed gas plasma. This slight degradation of detection limits was not judged to be detrimental to multi- element determinations in five reference materials when the results from using an argon-only plasma (with low and high aerosol carrier flow rates) were compared with the results from the argon-nitrogen plasma.Keywords Inductively coupled plasma mass spectrometry; polyatomic interferences; argon-nitrogen plasma; multi-element analysis Polyatomic ions cause spectroscopic interferences in induc- tively coupled plasma mass spectrometry (ICP-MS) by over- lapping with analytes at the m/z value of interest. These ions are formed by combination of two or more atomic species from precursors in the plasma gas entrained atmospheric gases water added reagents and the sample matrix.' Polyatomic ions from the first three sources cited above are inherent to aqueous ICP-MS systems and the number of significant interfering species is small.2 The most important sources of polyatomic precursors come from the reagents used in the sample preparation mainly acids and the sample matrix itself.Interferences produced by acid used in the digestion or acidification of samples are widely known3 and can be avoided by appropriate selection of the acids. When the precursor is present in the sample matrix its interfering effect has been eliminated in some cases by separating the analyte from the interfering matrix c~mponent,'"~ although more general approaches are based on the control of the ICP-MS system itself. Polyatomic ions containing 0 and/or H can be reduced by reducing the amount of water vapour reaching the plasma using cooled spray chambers- or other desolvation systems.' Optimization of instrumental parameters involved in the con- trol of the plasma discharge (r.f. power and gas flow rate) has been applied to reduce formation of refractory oxides.'0,'' Mixed gas plasmas have been used with different aims in ICP-MS.Oxygen12 has been used mainly to assist in the introduction of organic solvents whilst heliurn,l3 hydrogen14 and methane15 have been used to improve analyte sensitivity. 0xygenl~9~~ and methane25 have been used to reduce polyatomic interferences. Many of these aspects have been reviewed recently.' In the present work two different approaches to overcome polyatomic interferences were studied selection of instrumental parameters in argon-only plasmas and use of argon-nitrogen * Presented at the XXVIII Colloquium Spectroscopicum Internationale (CSI) Post-Symposium 5th Surrey Conference on Plasma Source Mass Spectrometry Durham UK 4-6 July 1993.7 On leave from the Department of Analytical Chemistry University of Zaragoza Spain. .I To whom correspondence should be addressed. plasmas by adding nitrogen to the aerosol carrier gas flow. Special attention was paid to chlorine-based and argon dimer polyatomic interferences. The final aim of this work was to improve the determination of As and Se when part of a routine multi-element analysis procedure for foods and biological samples. Experimental Instrumentation A VG PQ1 inductively coupled plasma mass spectrometer (VG Elemental Winsford Cheshire UK) was used. The sample introduction system consisted of a fixed cross-flow nebulizer and a water cooled double-pass Scott-type spray chamber. Nitrogen was added to the aerosol carrier gas using the second mass-flow controller of the instrument.Standard instrumental and measurement parameters are shown in Table 1. Table 1 Default instrumental ICP-MS conditions Instrumental parameters R.f. power Reflected power Argon gas flow Outer Intermediate Aerosol carrier Sample uptake rate Spray chamber temperature Sample cone Skimmer Measurement parameters Measuring mode Range Number of channels Number of scan sweeps Dwell time Points per peak Collector type 1350 W € 7 w 14 1 min-l 0.5 1 min-l 0.755 1 min-' 0.70 ml min-' 10 "C 1.00 mm Nicone 0.75 mm Nicone Scanning 6.02-239.54 m/z 2048 100 320 ps 5 Pulse728 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 Reagents and Reference Materials Nitric acid (Aristar) and single-element standards (Spectrosol) were obtained from Merck (Poole Dorset UK). Multi-element standards (SPEX ICPMS1 -2 -3 and -4) were obtained from Instruments SA (Stanmore Middlesex UK).Hydrochloric acid (PrimaR) was obtained from FSA Lab. Supplies (Loughborough UK). Purified water (demineralized Milli-Q 18.2 Ma) was used throughout. Four certified materials covering a broad range of chlorine contents were analysed Oyster Tissue Standard Reference Material (SRM) 1566a and Peach Leaves SRM 1547 [National Institute of Standards and Technology (NIST) Gaithersburg MD USA] Lobster Hepatopancreas TORT-1 (National Research Council of Canada Ottawa Ontario Canada) Mussel Tissue MAM2/TM (International Atomic Energy Agency Monaco). Additionally a non-certified reference mate- rial Mixed Diet RM 8431a (NIST) was also analysed.Sample Preparation Digestions of reference materials were performed in stainless- steel pressure decomposition vessels (50 ml capacity). The poly(tetrafluoroethy1ene) (PTFE) liners were cleaned with nitric acid in a microwave oven (CEM 81D) (CEM Microwave Technology Buckingham UK) prior to use. The liners were first rinsed well under running tap water then nitric acid (2.0ml) added and the closed liners subjected to microwaves for 15 min at 25% power. The hot acid was washed out with running tap water and the acid cleaning procedure repeated for a further 20min. Finally the inner surfaces of the liners were rinsed thoroughly with purified water and the liners closed prior to use. For each reference material three samples of approximately 0.500 g were placed in cleaned liners and to one was added an appropriate amount of analyte as spike.Four blanks and one spiked blank were also included in the batch. A similar volume of purified water was added to the unspiked samples and blanks. Nitric acid ( 5 ml) was added to all of the liners. Overnight digestions (6 h) were performed in an air-circulating oven (Gallenkamp Loughborough UK Oven BS) at 150°C after which the vessels were completely cooled at -20 "C (in a temperature controlled cold room) for 30min. The digests were made up to 10.0ml with purified water and stored pending further dilutions. Procedure Experiments to study the effect of instrumental parameters on polyatomic species and analytes were performed using solu- tions containing either 5% (v/v) nitric acid or similar solutions fortified with 0.05% (m/v) chloride as hydrochloric acid.Indium (25 ng ml-l) was added as the internal standard. The selection of the chloride concentration was based on the maximum amount of the element expected in food sample digests. Multi-element solutions of 25 ng ml-I were used except for studies of oxide and doubly charged ions for which 250 ng ml-I solutions were used. Detection limits (LODs) and background equivalent concentrations (BECs) were measured at analyte concentrations of 25 ng ml-I (250 ng ml-' for Se) in 5% (v/v) nitric acid and 5% (v/v) nitric acid plus 0.05% (m/v) chloride matrices. Ion-lens potentials were re-optimized for each of the instrumental conditions studied using ll'In (5% nitric acid). Digested samples were diluted a further ten-fold keeping the nitric acid content at 5%.Internal standardization with In and Bi (50 ng ml-') was used. Analyses were undertaken with three different sets of instrumental conditions. These were (i) default conditions (Table 1); (ii) as ( i ) but with high aerosol carrier gas flow rate; and (iii) as (i) but with nitrogen added to the plasma. Results were corrected in each case for blanks and recoveries. Spikes consisted of the maximum amounts expected of each of the analytes as well as of major components (Na Mg P K and Ca) in the samples. Results and Discussion Effect of Instrumental Parameters on Analyte Intensity The effects of aerosol carrier gas flow rate r.f. power and spray chamber temperature on the "'In signal for an all-argon plasma are shown in Fig.l(a) and (b). For each parameter three different values were tested. These values were chosen to represent the range of values which could be used under normal operating conditions. The optimum value for each parameter was selected by choosing those which after optimiz- ation of the ion-lens potentials gave the maximum response on '151n. A spray chamber temperature of 10°C was selected because although the lowest value tested 1 "C gave a compar- able response it sometimes gave rise to condensation on the cooled surfaces. These optimum conditions corresponded to the default conditions given in Table 1 which are used for the day-to-day running of the instrument in this laboratory. The detection limits using these optimum conditions (ng rnl-l based on 30 of the blank) obtained for "V (O.l) 7 5 A ~ (0.4) 77Se (3) and 78Se (19) compare well with those reported by Hill st al,23 51V (0.96) 75As (3.45) 77Se (15) and 78Se (14) who used simplex optimized conditions for an all-argon plasma.Instrumental conditions which give the maximum elemental response in routine multi-element analyses are a compromise between analyte sensitivity and the possible polyatomic oxide and doubly charged ion interferences." In spite of or because of these compromise conditions several elements remain difficult to analyse by ICP-MS owing to polyatomic inter- ferences. Table 2 shows the isotopes and the interfering poly- atomic species studied in the present work. The larger background signal produced by polyatomic ions generally implies an increase in background noise at the m/z value of interest and hence a decrease in the detection capability.26 ," 0.40 0.50 0.60 0.70 0.80 0.90 1.00 0 10 L I I I I 0.50 0.60 0.70 0.80 0.90 1.00 Flow rate/l min-' Fig.1 Effect of aerosol carrier gas flow rate r.f.power and spray chamber temperature on '151n+ signal (a) spray chamber temperature maintained at 10°C; A 1350; B 1200 and C 1100 W (b) r.f. power maintained at 1350 W; A 1; B 10; and C 25 "CJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 729 Table 2 Polyatomic ions and interfering isotopes of elements studied in the present work Element m/z v 51 Cr 52 53 Mn 55 Polyatomic ion Zn 64 67 68 As 75 Se 74 76 77 78 Furthermore if the polyatomic ions are formed from precursors in the sample matrix for example chloride species a positive systematic error will result unless matrix-matched standards are used or the polyatomic ion signal is reduced. Effect of Instrumental Parameters on Polyatomic Ions In order to reduce polyatomic ion intensities the influence of aerosol carrier gas flow rate r.f.power and spray chamber temperature was studied. The effect of aerosol carrier gas flow rate and r.f. power on the signals of ArAr' ArCl' C10' and ClO,+ are shown in Fig. 2. Chlorine polyatomic species [Fig. 2(b) and (c)] showed a progressive signal decrease with increasing aerosol carrier gas flow rate and decreasing r.f. power. In Fig.2(u) it can be seen that ArAr+ shows a maximum with respect to aerosol carrier gas flow rate which moved to lower flow rates and signal intensity when the r.f.power was reduced although low levels of this polyatomic species were also obtained at the highest flow rate studied. These findings are in agreement with previous work carried out under similar instrumental conditions'' and illustrate that the formation mechanism of ArAr' is different to that of chlorine species. The influence of water loading on the plasma chemistry in ICP-MS has been pointed out by several workers with a reduction in refractory oxide and doubly charged ions being reported when cooled spray chambers were used as a way of reducing the aerosol water content.68 Hutton and Eaton6 have also reported a reduction of ArAr' when reducing the spray chamber temperature this effect being related to three-body collision processes in which oxygen could have a dominant role.The effect of spray chamber temperature on the signals of ArAr' ArCl' and C10+ is shown in Fig. 3. Whereas an increase in the ArAr' signal was observed when the spray chamber temperature was increased the opposite behaviour was observed for the chlorine polyatomic species. The behav- iour of the C10+ signal is difficult to explain because more oxygen would be expected to enter the plasma in the form of water vapour when the spray chamber temperature was increased. However both species containing C1 decreased when the spray chamber temperature increased suggesting that either another C1 species (e.g. containing H which should also increase with increased water loading) which was not moni- tored could have increased or that C1 species are genuinely reduced at increased spray-chamber temperatures.Of the instrumental parameters studied the biggest effect on polyatomic species was observed for aerosol carrier gas flow rate. The most significant reduction of approximately three orders of magnitude was obtained for ArC1+ [Fig. 2(b)] t 20 x lo3 A I t 0 v 1x102 U 10 1 I I I I I I x 1 O 5 I x lo4 1 10 ' 1 I I I 1 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Flow rate/l min-' Fig.2 Effect of aerosol carrier gas flow rate and r.f. power on (a) 36Ar40Ar+ signal (A 1350 W and B 1100 W); (b) 40Ar35C1+ signal (A 1350 W and B 11OOW); chloride concentration 0.05%; and (c) 35C1160+ signal (A 1350W and B 11OOW) and 35Cl'60160+ signal (C 1350 W and D 1100 W); chloride concentration 0.05%.Spray chamber temperature 10 "C throughout whilst for the other polyatomic species studied intensities of around five times lower were obtained. In the case of ArAr' a reduction in signal intensity was also obtained by using spray chamber temperatures close to 0 "C (Fig. 3). Influence of Addition of Nitrogen to the Aerosol Carrier Gas Flow Nitrogen-argon plasmas have been reported to reduce the formation of polyatomic species.'6*18.20 M ontaser and Van HovenZ7 have suggested that the introduction of nitrogen into the aerosol carrier gas flow reduces the temperature of the gas in the axial channel because some of the energy is used to dissociate the diatomic injector gas. Houk et aLZ8 found that the ionization temperature in the central channel is reduced when nitrogen is added to the outer gas.This temperature drop could be responsible for both the reduction of the ionized polyatomic species which require more energy to ionize than atomic species,' as well as for the reported reductions in analyte sensitivities when nitrogen is introduced in the aerosol carrier gas.16 Evans and Ebdoni7 have argued that the reduction of ArAr' ArCl+ and ClCl+ in the presence of nitrogen in the aerosol carrier gas flow can be explained by the competitive formation of ArN' and/or ClN'. The effect of concentration of nitrogen in the aerosol carrier gas is shown in Fig. 4 for different aerosol carrier gas flow rates on the signals of ArAr+ [Fig. 4(u)] ArC1+ [Fig. 4(b)] and ClO' [Fig. 4(c)]. Overall a decrease in polyatomic ion730 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 4 I - 3 v) m C 3 0 4- 9 2 51 X > (0 C (5 ; 7 j 1 I I I I I I 1 0 5 10 15 20 25 30 TemperaturePC Fig.3 Effect of spray chamber temperature on polyatomic ion inten- sities aerosol carrier gas flow rate 0.755 1 min-'; r.f. power 1350 W. A 40A?5Cl+; B 35C1160+; and C 36Ar40Ar+ 1 x lo-' 1 x 1 0 ~ I x 10-2 t- ---4 I 10-3 I l X l O 2 ~ 1 \'" B I 1 x lo-' 1 0 2 4 6 8 1 0 [Nitrogen] (% v/v) Fig. 4 Effect of nitrogen concentration and total aerosol carrier gas flow rate (A 0.655; B 0.755 and C 0.855 1 min-l) on (a) 36Ar40Arf; (b) 40Ar35Cl+; and (c) 35Cl'60+ signals intensities was observed when both the nitrogen concentration and total aerosol carrier gas flow rates were increased. Analyte intensities were similarly reduced.Lower aerosol carrier gas flow rates were not studied because they produced high background signals across the whole m/z range. This increase was probably related t o an increase in the number of photons reaching the detector when nitrogen is introduced into the plasma.18 The background signal was reduced by four orders of magnitude for ArCl' whereas for C10' and ArAr' reductions of around two orders of magnitude were obtained. When nitrogen is introduced into the plasma the formation of NO' is increased," competing with other processes in the plasma involving oxygen. The effect on ArAr+ of adding nitrogen (8% aerosol carrier gas flow rate 0.755 1 min-l) at different spra;y chamber temperatures as a way of varying the amount of oxygen available from water reaching the plasma is shown in Fig.5. In the absence of nitrogen the formation of ArAr+ is influenced by the spray chamber temperature whereas it is independent of this variable in the presence of nitrogen. A similar effect was observed for ArO+ formation [Fig. 5(b)]. These results lend support to the role of oxygen6 in the formation process of ArAr'. Optimization of Signal-to-background and Signal-to-noise Ratios In the experiments reported above aerosol carrier gas flow rate spray chamber temperature and additions of nitrogen showed the most significant effects on the reduction of polya- tomic species. In order to improve the detection limits of the isotopes subjected to interferences a reduction of background signals must give an increase not only in signal-to-background ratios (SBR) but also of signal-to-noise ratios (SNR) since the SNR directly affects the detection limits.The SNR and SBR values for Se and Ar isotopes affected by ArAr+ and ArCl' interferences are shown in Table 3. The reduction of ArAr ' intensity observed when the spray chamber temperature was reduced from 10 to 1 "C produced a parallel improvement in the SBR although no significant improve- ments were observed in the SNR for 76Se and 78Se. The use of a high aerosol carrier gas flow rate produced a significant improvement for both the SBR and the SNR for the four isotopes. With respect to the addition of nitrogen a maximum reduction of the polyatomic signal at a nitrogen concentration of 8% is shown in Fig. 4(4 and (c) with a further reduction of ArAr' and ArCl+ when the aerosol carrier gas flow rate was increased from 0.755 to 0.855 1 min-'.On the other hand 2.5 2.0 "c - 1.5 + L 2 1.0 0.5 0 5.0 4.0 + 5 3.0 + 52 Q 2.0 1 .o I t B ,x-X x-x-x t L I I I I 1 I TemperaturePC 0 5 10 15 20 25 30 Fig.5 Influence of plasma type (A Ar and B Ar-N,) on the effect of spray chamber temperature on (a) ArAr+ and (b) ArO' signalsJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 73 1 Table3 SBR and SNR values for As and Se isotopes affected by polyatomic interferences Aerosol carrier gas flow rate/l min-l 0.755 0.755 0.955 0.755 0.855 Spray chamber temperature/"C 10 1 10 10 10 Aerosol nitrogen (%) ~~~ ~~ ~ Isotope Value 0 0 Matrix 5% HN03:- 76Se SBR* 0.12 0.32 SNR* 5.44 6.90 78Se SBR 1.36 3.55 SNR 39.51 42.28 Matrix 5% H N 0 3 f 0.05% C l - "As SBR 1.07 SNR 44.03 77As SBR 0.16 SNR 6.62 0 0.78 7.55 10.84 56.15 130.82 382.53 24.38 40.90 8 0.33 12.43 5.54 86.57 217.11 816.08 16.48 79.00 8 2.21 11.58 20.74 74.16 389.37 626.88 32.67 65.50 * Analyte concentration 250 ng ml-I.n = 10. it can be seen from Table 3 that when nitrogen was added the best improvement in the SBR was obtained at the highest flow rate while the SNR values were higher at the default flow rate of 0.755 1 min-l. In both cases the data obtained for the SBR and SNR using the nitrogen-argon plasma were better than those obtained using the argon-only plasma at high aerosol carrier gas flow rate. Selection of Alternative Operating Conditions to Reduce Polyatomic Interferences In addition to the default conditions (Table l) two sets of instrumental conditions that favoured the formation of low levels of polyatomic species were selected.These sets were (i) default conditions but with a higher aerosol carrier gas flow rate (0.955 1 min-l); and (ii) default conditions plus the addition of nitrogen (8%) to the aerosol flow. The main drawbacks of using high aerosol carrier gas flow rates reside in the increase in the formation of oxide ions of refractory elements as well as of doubly charged ions of elements of low second ionization The effect of the three sets of instrumental conditions on the formation 1 XI02 S B A + 5 0 2 + I x l Y 10 1 lo-' in-2 * .I 400 500 600 700 800 900 Oxide bond strength/kJ mol-' Fig. 6 Influence of selected instrumental conditions on oxide forma- tion as a function of metal-oxide bond strength A default conditions; B default conditions with high aerosol carrier gas flow rate; and C default conditions with 8% nitrogen in the aerosol flow of the oxides of refractory elements is shown in Fig. 6.At the default aerosol carrier gas flow rate (graph A Fig. 6 ) percent- ages of MO+:M+ ranged from 0.03 to 1.5%. These values increased to more than 20% for the most refractory elements when the high aerosol carrier gas flow rate was used (graph B Fig. 6). The oxide levels were reduced in the presence of nitrogen to a maximum value of 0.6% probably owing to the competitive formation of NO+.18 The influence of same set of instrumental conditions (A default conditions; B default con- ditions with high aerosol carrier gas flow rate; and C default conditions with 8% nitrogen) was also investigated with respect to the formation of doubly charged ions as a function of the Table 4 Detection limits ( 3 4 for selected elements in an argon plasma at standard and high aerosol carrier gas flow rates and nitrogen-argon plasma.Matrix 5% HNOJ Aerosol carrier gas flow rate/l min-' 0.755 0.955 0.755 0.755 0.955 0.755 Aerosol nitrogen (YO) ~~ 0 0 8 0 0 8 Element m/z Solution LOD/ng ml-'* Sample LOD/pg g-'*t Li Be B Mg A1 V Cr Mn Fe c o Ni c u Zn As Se Br Y Mo Cd Sn La Ce Pr Nd Sm Eu Gd Tb DY Ho Er Tm Yb Lu Hg Pb Th U 7 0.8 9 0.1 11 0.6 24 0.3 27 0.2 51 0.1 52 0.3 53 0.3 55 0.07 57 7 59 0.05 60 0.2 63 0.1 65 0.3 64 0.2 66 0.2 68 0.6 75 0.4 74 37 76 120 77 3 78 19 82 2 81 0.3 89 0.02 95 0.1 111 0.2 120 0.2 139 0.02 140 0.04 141 0.01 146 0.1 152 0.06 153 0.02 157 0.09 159 0.01 162 0.05 165 0.01 166 0.03 169 0.01 174 0.02 175 0.01 200 0.3 202 0.3 206 0.1 207 0.07 208 0.03 232 0.02 238 0.01 0.7 0.2 1 0.5 0.4 0.4 0.1 1 0.08 17 0.06 0.1 0.2 0.2 0.4 0.7 1 0.5 86 94 10 17 4 0.4 0.07 0.2 0.6 0.5 0.07 0.09 0.05 0.2 0.1 0.1 0.2 0.03 0.1 0.05 0.2 0.05 0.1 0.04 0.8 0.8 0.2 0.2 0.09 0.07 0.02 20 2 6 0.6 1.1 0.2 0.2 0.5 0.7 23 0.1 3 0.4 0.6 0.5 0.3 5 0.3 49 66 4 9 32 1.5 0.04 0.1 0.2 0.2 0.03 0.04 0.03 0.1 0.1 0.06 0.2 0.02 0.05 0.03 0.05 0.03 0.1 0.01 1 1 0.3 0.5 0.2 0.02 0.02 0.2 0.02 0.1 0.06 0.04 0.02 0.05 0.05 0.01 1 0.01 0.04 0.02 0.04 0.06 0.05 0.1 0.07 7 24 0.7 4 0.5 0.07 0.005 0.03 0.04 0.04 0.005 0.008 0.003 0.02 0.01 0.004 0.02 0.002 0.01 0.002 0.006 0.002 0.005 0.002 0.05 0.06 0.02 0.01 0.007 0.003 0.002 0.2 0.05 0.2 0.1 0.08 0.08 0.02 0.3 0.02 3 0.01 0.02 0.03 0.09 0.05 0.01 0.2 0.1 17 19 2 3 0.8 0.08 0.01 0.03 0.1 0.1 0.01 0.01 0.009 0.03 0.02 0.02 0.03 0.008 0.02 0.01 0.03 0.01 0.02 0.008 0.2 0.2 0.03 0.05 0.02 0.01 0.004 4 0.4 1.2 0.1 0.2 0.04 0.04 0.1 0.1 4 0.03 0.6 0.08 0.1 0.06 0.8 1 0.06 10 13 0.9 2 6 0.3 0.007 0.02 0.03 0.06 0.006 0.008 0.005 0.03 0.02 0.01 0.04 0.004 0.01 0.005 0.01 0.006 0.02 0.004 0.2 0.2 0.05 0.09 0.04 0.004 0.006 * n = 10.f Dilution 1 + 200.732 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 Table 5 BEC in ng ml-' produced by 0.05% chloride (m/v) on selected isotopes affected by polyatomic interferences. Argon plasma at standard and high aerosol carrier gas flow rate and nitrogen- argon plasma Aerosol carrier gas flow rate/l min-' 0.755 0.955 0.755 Aerosol nitrogen (%) Element m/z V 51 52 Cr 53 Zn 67 68 As 75 Se 74 76 0 292.4 4.0 741.9 20.3 1.7 249.3 22.4 360.6 0 47.1 1 .o 128.0 18.3 1.8 3.3 7.0 86.5 8 4.0 0.1 2.5 0.7 0.2 0.2 - 2.8* 10.3 second ionization potential.The mean values for percentages of M2+:M+ for each of the instrumental conditions were for A 0.17 (ranging from 0.004 to 0.9%0); for B 1.04 (ranging from approximately 0.02 up to 4%); and for C 3.13 (ranging from about 0.1 up to 10-20%). The increase in doubly charged ions in the presence of nitrogen can be also explained by the presence of NO+. It has been suggested that NO+ is involved in charge transfer ioniz- ation mechanisms.'* On the other hand the introduction of nitrogen into the plasma implies the formation of significant amounts of nitrogen polyatomic species which can produce additional interference problems.The most important potential interferences from polyatomic species of nitrogen for the elements studied in this work are included in Table 3.3 Trace amounts of Kr (interfering on s2Se) as well as Xe were detected in the nitrogen supply. 77 1548.2 14.6 0.7 Effects of Operating Conditions on the Limits of Detection and 78 28.0 5.4 1.8 Background Equivalent Concentrations * Negative value due to low sensitivity of the isotope and noisy The LODs in 5% nitric acid for elements across the m/z range using the three sets of operating conditions are shown in Table 4. The LODs for solutions were calculated as the concen- signal.Table 6 Determination of selected elements in certified reference material Mussel Tissue MAM2/TM using an argon plasma at standard and high aerosol carrier gas flow rate and a nitrogen-argon plasma. Reference chlorine content 8.71% Element B A1 V Cr Mn Fe c o Ni c u Zn As Se Y Mo Cd Sn La Ce Pr Nd Sm Eu Gd Tb DY Ho Er Tm Yb Lu Hg Pb Th U m/z 11 27 51 52 53 55 57 59 60 65 67 68 75 77 78 82 89 95 111 120 139 140 141 146 152 153 157 159 162 165 166 169 174 175 202 208 232 238 Aerosol carrier gas flow rate/l min-' ~ ~~ 0.755 0.955 0.755 Aerosol nitrogen (%) 0 40.3 f 0.8 155+2 11.0t-0.6 1.62 f 0.03 49.5 + 2.9 64.8 f 0.2 312+7 0.905 k 0.024 1.32 & 0.03 10.0 f 0.1 156k 1 149+2 32.4 + 1.6 172+ 12 30.7 -f 0.4 2.56 f 0.05 0.181 kO.011 0.968 f 0.036 1.44f0.03 0.677 & 0.086 0.145 f 0.001 0.204 f 0.003 0.027 k 0,001 0.107 & 0.002 0.021 f 0.004 0.007 & 0.001 0.022 + 0.004 0.004 f 0.000 0.027 f 0.003 0.005 k 0,001 0.014+0.001 0.002 f o.Oo0 0.007 & 0.003 0.002 f 0.001 0.934 & 0.056 2.16 f 0.02 0.030 f 0.000 0.188+0.002 0 Measured content/pg g- '* 42.7 f 1.6 160+4 2.28 + 0.00 1.38+0.03 4.06k0.11 62.9 f 0.6 326f9 0.873 & 0.027 8.05 k 0.25 166 f 2 165 & 3 14.0 & 0.1 3.80 k 0.02 12.6 k 0.6 2.67 + 0.20 0.171 f0.004 0.592 f 0.066 1.66 f 0.03 0.619f0.038 0.130 & 0.087 0.190 f 0.003 0.033 f 0.OQ3 0.088 f 0.0 14 < LOD < LOD < LOD < LOD 0.035 f 0.007 < LOD < LOD < LOD < LOD < LOD 1.14 f 0.02% 2.09 f 0.01 0.03 1 + 0.0015 0.233 k 0.003 1.27 f 0.03 8 Certified content/pg g-' 40.3 f0.6 165+ 1 1.61 fO.OO 1.35f0.05 1.60f0.01 67.8 kO.1 327 f 0.4 0.898 k 0.005 1.29 f 0.05 7.54 & 0.15 159f 1 154fO 13.1f0.1 2.63 kO.10 3.59 f 0.19 < LOD 0.189 f 0.003 0.698 k 0.01 5 1.40 f 0.03 0.582 f0.014 0.146 k 0.007 0.196 f 0.008 0.03 1 & 0.003 0.093 k0.019 < LOD < LOD < LOD < LOD 0.029 f 0.006 0.005 f 0.001 0.022 + 0.006 < LOD < LOD 0.004 f 0.000 0.862 f0.149 2.13 & 0.08 0.03 5 & 0.001 0.190 -t 0.007 - 1.25 (0.95-1.62) - 67.1 (60.7-75.3) 256.2 (229.2-268.2) 0.88 (0.75-1.07) 7.96 (7.53-8.44) 156.5 (152.8-166.7) 12.8 (11.8-14.4) 2.27 (1.70-2.56) - - - - 1.32 (1.16-1.54) - - 0.95 (0.85-1.06) 1.92 (1.53-2.50)f * Average f standard deviation of duplicates.t Reference value.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 73 3 Table 7 Determination of selected elements in certified reference material Lobster Hepatopancreas TORT-1 using an argon plasma at standard and high aerosol carrier gas flow rate and a nitrogen-argon plasma. Certified chlorine content 5.58% Element B A1 V Cr Mn Fe c o Ni c u Zn As Se Y Mo Cd Sn La Ce Pr Nd Sm Eu Gd Tb DY Ho Er Tm Yb Lu Hg Pb Th U m/z 11 27 51 52 53 55 57 59 60 65 67 68 75 77 78 82 89 95 111 120 139 140 141 146 152 153 157 159 162 165 166 169 174 175 202 208 232 238 Aerosol carrier gas flow rate/l min-l 0.755 0.955 0.755 Aerosol nitrogen (%) 0 5.41 40.15 27.2 4 0.5 7.17 k 1.74 2.81 k 0.25 46.0 f 12.3 20.9 4 0.3 2121f 1 0.461 f 0.002 3.49 1f 0.54 409 4 3 1554 1 149-t-0 38.3 f 1.9 86.0f13.5 22.2 f 12.6 6.96 f 0.02 1.57 f0.03 1.33 f0.06 26.0 4 0.1 0.108 4 0.008 5.15 f 0.01 4.39 4 0.04 0.627 'r_ 0.002 2.54 & 0.03 0.333 4 0.003 0.069 0.005 0.371f0.011 0.041 & 0.002 0.190 f 0.002 0.036 & 0.001 0.086 4 0.001 0.008 'r_ 0.001 0.040 & 0.003 0.005 & 0.000 0.247 k 0.021 9.25 k0.41 0.007 & 0.000 0.098 f 0.006 0 Measured contentlpg g-'* 5.56 f 0.22 27.2 f 0.3 1.74 f 0.15 2.36 f 0.24 4.14f 1.69 20.4 k 0.3 201 k 4 0.45 1 -t 0.008 3.29 & 0.53 399 f 1 169F 1 163f3 27.7 f 0.0 8.14k0.32 11.4 f 0.5 5.73 -t 0.58 1.48 k 0.02 1.20-tO.08 29.7 f 0.2 0.130 f 0.01 6 3.94 f 0.04 3.47 f 0.053 0.5 52 h 0.000 2.13 f 0.03 0.311 fO.010 0.063 & 0.000 0.426 -t 0.022 0.047 rt 0.001 0.208 k 0.021 0.040 3_ 0.001 0.08 1 f 0.002 0.012 f 0.000 0.039 f 0.003 0.010 f 0.001 0.378 & 0.029 8.93 +_ 0.39 < LOD 0.1 14 & 0.009 8 5.82 f 0.06 29.3 & 0.6 1.44 & 0.00 2.66 & 0.35 2.64 & 0.43 21.7 f 0.4 222 f 2 0.480 f 0.014 3.55 f 0.66 409 & 1 160& 1 157+1 27.8 k 0.1 7.40 k 0.04 7.44f0.35 < LOD 1.66 f 0.01 1.29k0.06 26.3 & 0.4 0.1 35 & 0.013 5.11 k0.08 4.32 k 0.01 0.637 fO.010 2.51 kO.01 0.344 & 0.009 0.079 f 0.008 0.423 f 0.016 0.040 & 0.004 0.179 & 0.009 0.030 & 0.002 0.097 k 0.006 0.01 1 f 0.002 0.042 & 0.002 0.005 f 0.001 0.196 & 0.023 9.58 +_ 0.3 1 0.009 f 0.002 0.09 1 f 0.002 Certified content/pg g-' - 1.4f0.3 2.4 f 0.6 23.4 f 1.0 186f11 0.42 f 0.05 2.3 f 0.3 439 & 22 177 f 10 24.6 f 2.2 6.88 & 0.47 - - - 1.5h0.3 26.3 f 2.1 0.139 fO.011 __ - 0.330 f 0.060 10.4 f 2.0 - * Average f standard deviation of duplicates.tration of analyte that produced a signal equal to three times the standard deviation of the blank (ten measurements) and the LODs for the samples were calculated on basis of three standard deviations of the blank and corrected for the sample dilution (1 + 200) assuming a sample mass of 0.5 g. Although some increases in LODs were obtained when the alternative sets of operating conditions were used instead of the default conditions these were not significant except for isotopes affected by either nitrogen polyatomic interferences ("Mn and 68Zn) or by isobaric overlap from nitrogen contami- nants (82Se). Lithium Be B and Ni showed higher detection limits in the presence of nitrogen owing to the increase in the background intensities at their m/z values. In the case of Ni the higher background levels could have been due to the increased cone wear observed when nitrogen was used.Variations in residual Ni have also been observed with respect to spray chamber temperature by Hutton and Eatona6 In this work for argon-only plasmas relative standard deviations (RSDs n=10) of l-6% and 14% were obtained at default and high aerosol carrier gas flow rates respectively. Values between 10 and 20% were obtained for 74Se 76Se and 57Fe in both cases. For argon-nitrogen plasma conditions RSDs of 2-10% were measured for most elements 10-16% for "B 74Se 82Se and 'O0Hg with higher values for 7Li and 57Fe. The presence of chloride in the matrix contributes to the signals for isotopes affected by a chlorine polyatomic inter- ference. The magnitude of these contributions owing to the additional presence of 0.05% (m/v) chloride expressed as BEC of the analyte in the standard matrix (5% nitric acid) is shown in Table 5.An indirect effect of chloride on the argon dimer was also observed on 76Se and 78Se. Proper matrix matching could correct for these interferences. The LODs measured in the presence of chloride increased up to 35 times for isotopes affected by chlorine polyatomic interferences. Significant reductions of BECs were obtained for all these isotopes by using nitrogen. For As and Se an argon-only plasma and default conditions with increased aerosol carrier gas flow rate produced considerable reductions in the BECs. However the addition of nitrogen reduced the BECs even further.Multi-element Analysis of Reference Materials Reference materials covering a broad range of chlorine concen- trations (up to 8.71% for Mussel Tissue) were selected. Multi- element analyses were performed using the three sets of operating conditions. The results obtained are summarized in Tables 6-10. When nitrogen was added to the plasma agree- ment with the certified values was observed for approximately734 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 Table 8 Determination of selected elements in certified reference material Oyster Tissue SRM 1566a using an argon plasma at standard and high aerosol carrier gas flow rate and a nitrogen-argon plasma. Certified chlorine content 0.829% Aerosol carrier gas flow rate/l min-' 0.755 0.955 0.755 Aerosol nitrogen (YO) Certified content/pg 8-l Element B A1 V Cr Mn Fe c o Ni c u Zn As Se Y Mo Cd Sn La Ce Pr Nd Sm Eu Gd Tb DY Ho Er Tm Yb Lu Hg Pb Th U 11 27 51 52 53 55 57 59 60 65 67 68 75 77 78 82 89 95 111 120 139 140 141 146 152 153 157 159 162 165 166 169 174 175 202 208 232 238 0 9.01 k 0.43 119+1 5.51 k0.50 1.39 k0.19 4.46 & 1.53 11.4 k 0.1 477 k 3 0.538 & 0.014 1.85 k 0.02 63.2 f 0.0 798 f 6 760 k 5 15.3f0.5 12.9 f 3.6 9.21 & 11.69 2.49 f 0.30 0.370f0.001 0.411 k0.183 4.15 '.0.03 2.25 f0.12 0.234 & 0.007 0.312+0.002 0.54 k 0.002 0.236 f 0.001 0.045 0.003 0.01 1 f 0.001 0.052 f 0.001 0.009 f 0.001 0.053 f 0.000 0.01 1 -1 0.002 0.033 f 0.003 0.005 5 0.001 0.029 k 0.001 0.006 & 0.002 < LOD 0.350 f 0.004 0.033 & 0.002 0.122f0.004 0 Measured content/pg g-l* 9.13k0.15 122f0 4.26 & 0.05 1.32 k 0.02 1.54k0.22 11.5 f 0.0 462 k 0 0.558 -10.007 1.94f0.00 67.9 f 0.7 935f4 943 k 2 13.3 k0.4 3.07 k 0.52 7.44 & 4.1 1 1.82k0.09 0.348 f 0.012 0.180 f 0.020 4.58 k0.23 2.20 k 0.1 3 0.164 f 0.005 0.259 f 0.005 0.045 k 0.006 0.205 k 0.836 0.052 k 0.1000 <LOT) 0.054 f 0.000 0.01 1 f 0.000 0.041 f 0.003 0.01 1 f 0.001 <LOT) < LOD 0.036 k 0.013 < Lor) < LOD 0.327 & 0.008 0.030 f 0.001 0.137 fO.002 8 9.06 f 0.21 127f2 4.67 k 0.06 1.33 k 0.00 1.40&0.12 11.2 k 0.1 477 k 8 0.547 k 0.000 1.94 kO.10 60.0 0.7 755 f 9 742 k 9 13.3 k0.4 1.84k0.07 2.33 0.60 < LOD 0.392 f 0.001 0.219 f 0.030 4.12k0.15 2.20 0.09 0.223 f 0.013 0.325 f 0.032 0.060 k 0.001 0.224 & 0.02 1 0.044 & 0.001 < LOD 0.069 f 0.003 0.1008 f 0.001 0.063 f 0.01 5 0.012 f 0.003 0.032$0.015 0.006 f 0.001 0.034 + 0.002 0.007 f 0.001 < LOD 0.372 k 0.045 0.035 f 0.002 0.1 14 k 0.006 - 202.5 f 12.5 4.68 f0.15 1.43 k 0.46 12.3 k 1.5 539 & 15 0.57 If 0.1 1 2.25 k 0.44 66.3 k4.3 830 57 14.0f 1.2 2.21 f 0.24 - - - 4.15k0.38 0.37 0.47 - - 0.067 0.017 - 0.0077 - 0.0642 k 0.0067 0.371 f 0.014 0.047 0.132 f0.012 * Average standard deviation of duplicates.7 Reference value. 70% of the isotopes affected by interference from chloride. The remaining 30% are either slightly above or below the certified values with no obvious trend for particular isotopes. The polyatomic interferences from nitrogen listed in Table 2 did not show any effect on the accuracy of the results for the isotopes affected except for 68Zn (Table 9). For the remaining elements good agreement with the certified values was gener- ally obtained for all three sets of operating conditions.Exceptions were a low value for A1 in Oyster Tissue under all three sets of operating conditions which is probably due to the presence of insoluble silicates of this element in the digest and on some occasions poor agreement with the certified values for the isotopes 78Se and "Se (Tables 6-10) 67Zn 95Mo and 208Pb (Table 9) 55Mn (Table 7 and 9) and 57Fe (Tables 6 7 8 and 10). Lower results were observed for some of the rare earth elements when the high aerosol carrier gas flow rate was used as shown for example by the results for Li and Ce from NIST Oyster Tissue and Peach Leaves SRMs.The recoveries of added analyte for rare earth elements ranged from 110 to 140% in spiked blanks and up to 200% in spiked samples when high aerosol carrier gas flow rate conditions were used. This is indicative of a severe matrix effect dependent upon the aerosol carrier gas flow rate and directly proportional to the oxide bond strength of the element. Preliminary results are summarized in Fig. 7 for the effect of increasing the amount of some matrix components (Mg Ca and Na) on the signals of Ce' CeO' and Ce2' when the aerosol carrier gas flow rate was 0.955 1 min-I using "'In as an internal standard (ideally the internal standard should have been an element with a similar oxide bond strength to the analyte). Under these conditions no significant effect was observed with respect to the addition of Mg but for Ca and especially for Na a significant increase of the effect on the signals for Ce' and decrease on CeO' was observed.Levels of doubly charged ions did not change significantly. These results suggest the inhibition of the oxide formation and the enhancement of the analyte signal owing to the competitive formation of the oxides of major cations of the matrix especi- ally Na (oxide bond strength Na > Ca > Mg > K). Similar effects were observed for the remainder of the rare earth elements in the presence of high concentrations of NaJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 73 5 Table 9 Determination of selected elements in certified reference material Peach Leaves SRM 1547 using an argon plasma at standard and high aerosol carrier gas flow rate and a nitrogen-argon plasma.Certified chlorine content 0.036% Aerosol carrier gas flow rate/l min-l ~~ 0.755 0.955 0.755 Aerosol nitrogen (%) Element B A1 v Cr Mn Fe c o Ni c u Zn As Se Y Mo Cd Sn La Ce Pr Nd Sm Eu Gd Tb DY Ho Er Tm Yb Lu Pb Th U Hg mlz 11 27 51 52 53 55 57 59 60 65 67 68 75 77 78 82 89 95 111 120 139 140 141 146 152 153 157 159 162 165 166 169 174 175 202 208 232 238 0 26.4 f 1.3 199 _+ 3 0.425 f 0.060 0.854 f 0.004 1.11 f0.16 89.4 f 1.3 264 f 2 0.082 f 0.003 0.733 & 0.008 3.92 f 0.12 19.6 f 0.1 18.3f0.4 < LOD < LOD < LOD < LOD 2.99 f 0.03 0.141 f0.034 0.032 f 0.014 0.066 f 0.004 9.12 f 0.02 11.6f0.1 1.81 f0.15 6.94 f 0.03 1.15f0.00 0.220 f 0.003 l .l l ~ O . 0 0 0.121 fO.000 0.566 f 0.008 0.087 f 0.002 0.226 2 0.002 0.025 2 0.001 0.133 f 0.001 0.019 f 0.002 < LOD 0.829 f 0.003 0.045 f 0.002 0.010 f 0.000 0 Measured content/pg g-l* 28.0 & 0.8 195 1- 3 0.319&0.014 0.774 -t 0.009 0.766 2 0.020 90.3 -t 1.0 269 & 1 0.086 & 0.006 0.854 & 0.024 3.75 20.10 25.3 f 1.8 23.8 f 2.0 < LOD < LOD <LOD < LOD 2.86 f 0.26 0.075 f 0.014 0.057 f 0.038 0.088 f 0.010 5.29 & 0.03 5.92 f 0.05 1.57 k0.09 4.78 f 0.10 1.14 k 0.01 0.182f0.013 1.18 fO.01 0.125 k 0.002 0.608 f 0.004 0.100f0.004 0.238 f 0.022 0.027 & 0.003 0.147 & 0.002 0.017 2 0.006 < LOD 0.796 & 0.01 1 0.043 f 0.005 0.016 k 0.003 8 28.0 f 0.2 219+1 0.354 f 0.012 0.885 f 0.005 0.865 f 0.025 96.3 f 0.9 280 f 3 0.078 f 0.008 0.707 rt 0.057 3.64 f 0.17 32.6 f 0.6 29.1 f 0.3 < LOD < LOD < LOD < LOD 3.20 f 0.01 0.051 f 0.003 0.020 f 0.005 0.071 f 0.000 10.4 f 0.2 12.7f0.1 1.91 fO.OO 7.83 f0.14 1.11 fO.OO 0.206 f 0.01 1 1.03 f 0.3 0.108 & 0.006 0.493 & 0.023 0.093 f 0.006 0.245 f 0.005 0.028 f 0.002 0.132f0.002 0.020 5 0.001 < LOD 0.855 f 0.026 0.044 & 0.003 0.01 1 f 0.001 Certified content/pg g-' 29f2 249 f 8 0.37 f 0.03 17 98-t-3 2207 - 0.07t 0.69 _+ 0.09 3.7 f 0.4 17.9 & 0.4 0.060 2 0.018 0.120 f 0.009 - - - 0.2t 0.060 f 0.008 0.03 < 0.027 9 t 107 77 I t 0.lt - 0.177 - - - - - - - 0.03 1 f 0.007 0.87 f0.03 0.05-l 0.015t * Average Ifi standard deviation of duplicates.t Reference value. (500pgml-'). The Na concentrations in the digests of the reference materials were not measured.130 110 - m S 0 v1 a m a M .- 2 90 Y - 70 50 m Ce' a CeO+ a Ce2+ 0 10 50 0 100 200 0 200 500 Mg Ca Na Concentration/mg I-' Fig.7 Effect of adding Mg Ca and Na on Ce' CeO' and Ce2+ signals at high aerosol carrier gas flow rate (0.955 1 min-l) Conclusions The results reported were obtained from one batch of the reference materials digested in duplicate in nitric acid using steel pressure decomposition vessels. Under normal default operating conditions when the instrument had been optimized for maximum signal response the addition of 8% nitrogen to the aerosol carrier gas flow improved the measurement of 51V 53Cr 67Zn and 68Zn (except for NIST SRM 1547 Peach Leaves) 75As and 77Se in five reference materials. Of the other 28 elements which were determined in five reference materials those which had certified or reference values were with few exceptions in good agreement with these values when meas- ured under default conditions or under default conditions with either nitrogen added to the aerosol carrier gas flow or with a higher aerosol carrier gas flow rate.The values for some rare earth elements when a high aerosol carrier gas flow rate was used were exceptions. The increase in detection limits when nitrogen was added was not so great as to cause any problems with the multi-element determinations.736 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 Table 10 Determination of selected elements in reference material Mixed Diet SRM 8431a using an argon plasma at standard and high aerosol carrier gas flow rate and a nitrogen-argon plasma Aerosol carrier gas flow rate/l min-' 0.755 0.955 Aerosol nitrogen (Yo) 0.755 Recommended content/ Pg g-' Element B A1 v Cr Mn Fe c o Ni c u Zn As Se Y Mo Cd La Ce Pr Nd Sm Eu Gd Tb DY Ho Er Tm Yb Lu Hg Pb Th U m/z 11 27 51 52 53 55 57 59 60 65 67 68 75 77 78 82 89 95 111 139 140 141 146 152 153 157 159 162 165 166 169 174 175 202 208 232 238 0 3.94 f 0.52 4.82 f 0.05 0.333 k0.192 0.161 fO.041 1.07f0.65 7.86 f 0.02 41.4f 3.2 0.043 f 0.001 0.659 0.060 3.43 f 0.57 15.1 f0.8 14.8 f 0.1 1.85f0.31 4.52 f 1.87 1.22 & 1.56 0.432 & 0.182 0.005 & 0.001 0.305 f 0.028 0.034 f 0.01 8 0.010 f 0.000 0.009 f 0.002 < LOD < LOD < LOD < LOD < LOD < LOD < LOD 0.002 & 0.001 < LOD < LOD < LOD < LOD < LOD 0.136 f 0.044 < LOD 0.006 f 0.001 0 Measured content/pg g-'* 4.01 & 0.28 4.82 & 0.046 0.102+0.028 0.108 k 0.004 0.273 & 0.070 7.86 k 0.02 40.3 & 0.3 0.043 2 0.001 0.659 & 0.060 3.43 40.57 17.6f0.7 16.3k0.2 0.976 4 0.016 0.3 19 _+ 0.358 4.63 f 0.33 0.492 & 0.262 0.003 & 0.002 0.305 f 0.028 0.034 f 0.018 < LOD < LOD < LOD < LOD < LOD < LOD < LOD < LOD < LOD < LOD < LOD < LOD < LOD -= LOD < LOD 0.136 & 0.044 < LOD < LOD 8 4.14 f 0.38 5.18 k 0.09 0.053 f 0.001 0.091 -t 0.002 0.090 f 0,023 7.86 f 0.14 41.8 rt 0.6 0.046 f 0.002 0.716 f0.020 3.23 f 0.02 15.1 f0.2 15.0f0.3 0.992 f 0.028 0.23 1 f 0.042 0.908 f 0.089 < LOD 0.004 f 0.001 0.298 k 0.040 0.041 f 0.004 0.01 3 f 0.002 0.010f0.001 < LOD < LOD < LOD < LOD < LOD < LOD <LOD < LOD < LOD < LOD < LOD < LOD < LOD 0.170 f 0.024 < LOD < LOD - 4.39 f 1.07 0.102 k0.006 8.12 f0.31 37.0 f 2.6 0.038 f 0.008 0.644 f 0.1 51 3.36f0.33 17.0 f 0.6 0.924 & 0.344 0.242 f 0.030 - - - - 0.288 k 0.029 0.042 f 0.01 1 * Average f standard deviation of duplicates.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 References Evans E. H. and Giglio J. J. J. Anal. At. Spectrom. 1993 8 1. Jarvis K. E. Gray A. L. and Houk R. S. Handbook of inductively Coupled Plasma Mass Spectrometry Blackie & Son London 1992. Tan S. H. and Horlick G. Appl. Spectrosc. 1986 40 445. Lyon T. D. B. Fell G. S. Hutton R. C. and Eaton A. N. J. Anal. At. Spectrom. 1988 3 601. Sheppard B. S. Shen W. L. Caruso J. A. Heitkemper D. T. and Fricke F. L. J. Anal. At. Spectrom. 1990 5 431. Hutton R. C. and Eaton A. N. J. Anal. At. Spectrom. 1987,2,595. Browner R. F. and Zhu G. J. Anal. At. Spectrom. 1987 2 543. Zhu G. and Browner R. F. J. Anal. At. Spectrom. 1988 3 781. Lam J. W. and McLaren J. W. J. Anal. At. Spectrom. 1990 5 419. Gray A. L. and Williams J. G. J. Anal. At. Spectrom. 1987,2 81. Gray A. L. and Williams J. G. J. Anal. At. Spectrom. 1987,2,599. Hausler D. Spectrochim. Acta Part B 1987 42 63. Sheppard B. S. Shen W. L. Davidson T. M. and Caruso J. A. J. Anal. At. Spectrom. 1990 5 697. Louie H. and Soo S . Y. J. Anal. At. Spectrom. 1992 7 557. Allain P. Jaunault L. Mauras Y. Mermet J. M. and Delaporte T. Anal. Chem. 1991 63 1497. Evans E. H. and Ebdon L. J. Anal. At. Spectrom. 1989 4 299. Evans E. H. and Ebdon L. J. Anal. At. Spectrom. 1990 5 425. 18 19 20 21 22 23 24 25 26 27 28 Lam J. W. H. and Horlick G. Spectrochim. Acta Part B 1990 45 1313. Branch S. Ebdon L. Ford M. Foulkes M. and O'Neill P. J. Anal. At. Spectrom. 1991 6 151. Beauchemin D. and Craig J. M. Spectrochim. Acta Part B 1991 46 603. Craig J. M. and Beauchemin D. J. Anal. At. Spectrom. 1992 7 937. Ford M. Ebdon L. and Hill S. J. Anal. Proc. 1992 29 104. Hill S. J. Ford M. J. and Ebdon L. J. Anal. At. Spectrom. 1992 7 719. Smith F. G. Wiederin D. R. and Houk R. S. Anal. Chem. 1991 63 1458. Hill S. J. Ford M. J. and Ebdon L. J. Anal. At. Spectrom. 1992 7 1157. Kawaguchi H. Tanaka T. and Mizuike A. Spectrochim. Acta Part B 1988 43 955. Montaser A. and Van Hoven R. L. CRC Crit. Rev. Anal. Chem. 1987 18 45. Houk R. S. Montaser A. and Fassel V. A. Appl. Spectrosc. 1983 5 425. Paper 3105893 J Received September 30 1993 Accepted February 14 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900727

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 737 INTER-LABORATORY NOTE Multi-element Analysis of Archaeological Bronze Objects Using Inductively Coupled Plasma Atomic Emission Spectrometry Aspects of Sample Preparation and Spectral Line Selection I. Segal and A. Kloner lsrael Antiquities Authority P.O. Box 586 Jerusalem 9 1004 Israel 1. B. Brenner* Geochemistry Division Geological Survey of Israel 30 Malkhe lsrael Street Jerusalem 9500 I lsrael Inductively coupled plasma atomic emission spectrometry has been applied to the determination of trace and minor elements in archeological bronzes. Trace element spectral lines were selected on the basis of minimum interference from major element matrix components i.e. Cu Pb Fe and Sn. Using Sc as the internal reference element provided a significant improvement in the relative standard deviation. Accuracy evaluated using standard reference alloys of similar composition was satisfactory.Keywords lnductively coupled plasma atomic emission spectrometry; archaeomefry; bronze; spectral line interference; sample preparation The use of trace and minor elements to fingerprint bronze archaeologic hoards has been reported widely in the litera- ture.14 These studies indicated that the major and trace element composition of the artifacts can reveal additional information on the metallurgical procedures (smelting alloying and casting) used to produce these materials and the source of the raw materials. Most of the analytical data in the literature were obtained using flame atomic absorption spec- trometry.This technique has several disadvantages namely single-element capability inadequate limits of detection and accuracy. Inductively coupled plasma atomic emission spec- trometry (ICP-AES) is now a mature technique for multi- element analysis. However the number of reports describing the use of this technique for the analysis of bronze archaeolog- ical objects is limited. Trampuz-Ore1 et al.' used ICP-AES to determine the provenance of bronze objects from Slovenia. Gluimlia-Mairlo analysed Cu-based objects by ICP-AES and compared artifacts from several archaeological sites although the number of elements determined was limited. Merkel" used ICP-AES for the analysis of Cu ores from Timna located in southern Israel however trace metals were not determined. In the present paper the analytical procedure and perform- ance for the multi-element analysis of archaeological bronzes using ICP-AES is described.Spectral lines for the determi- nation of As Mo Zn Sb Bi Co Cd Ni Au Mn Fe Cr and V are recommended together with their true limits of detection in Cu-Sn-Pb-Zn matrices. The presence of these elements at high concentrations were taken into account in the sample preparation and decomposition procedures. The accuracy of the method was evaluated by analysing standard reference bronzes and related Cu alloys. In the future it is planned to correlate the concentrations of the elements with the provenance and the sources of the archaeological objects using pattern recognition techniques. The aim is to determine the composition and source of the raw materials purity of the metallurgical processes and ulti- mately fingerprint the objects.~~~ ~ * To whom correspondence should be addressed. Experimental Sample Location The bronze objects were collected from two archaeological excavations from Central Israel of the Maresha site mainly Hellenistic in age; and at Bet Guvrin Roman Byzantine and Medieval in age. The archaeology of these sites has been described by K1oner.l' Biblical Maresha consists mainly of three periods of occupation of the Iron Age Late Judean Monarchite Persian Period and Hellenistic. This habitation was terminated by massive destruction in 113 BCE. The archaeo- logical objects studied consisted of bronze and silver coins bronze (occasionally with iron wire bindings) statuettes figur- ines bronze domestic cooking vessels bronze buckles and furniture.The objects analysed from Bet Guvrin were disco- vered in a Roman amphitheatre (2-3 centuries CE) a cave a Crusader church and a medieval fort and consist of a variety of iron bronze and copper objects from different periods. Sample Preparation and Decomposition Owing to the great archaeological value of the objects special attention was given to the pre-preparation and chemical dissolution procedures. Prior to acid dissolution oxidative layers were mechanically removed by scrubbing them with a steel brush. Samples were then cleaned with dilute HCl acetone and de-ionized water. Samples for analysis were removed by micro-drilling in order to minimize damage. The first portions were discarded.In certain cases objects were analysed semi- quantitatively by X-ray fluorescence (XRF) in order to deter- mine major elements and to ascertain the approximate ranges of their concentrations. The dissolution procedure was similar to that described by Hughes et al.l3 Samples weighing 25-50 mg were dissolved in aqua regia (hydrochloric plus nitric acid 3 + 1) in 25-50 ml Pyrex beakers. Samples containing high amounts of Ag and Pb were decomposed in nitric acid. In this procedure Au was not solubilized. In order to determine Au 2ml of aqua regia were added to the sample. The contents of the beakers were heated to 60 "C in order to accelerate dissolution. After cooling 10 ml of de-ionized water were added. Diethylenetriamine was used to avoid precipitation of Ag.14 In most cases complete dissolution was obtained.Samples containing residues were738 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 fused with sodium peroxide and consisted mainly of alumino- silicate impurities. Solutions were made up to volumes of 25-50 ml. Scandium was employed as an internal standard the final concentration amounting to 5 mg 1-I. Table 2 Details of spectrometer instrumentation Monochromator:- Sequential system Grating Dispersion Jobin Yvon JY 38 3600 grooves mm-' range 180-490 nm 0.27 nm mm-' 6 pm in the first order (25 pm slit-widths) Multi-element Calibration The concentrations and range of the elements in the calibration standards were determined on the basis of the data cited in the literature.'-'' These standards and the expected concen- tration ranges are listed in Table 1.It should be noted that although the Cu contents in the sample usually exceeded 90% a standard containing the equivalent of 100% was prepared. The multi-element standards were prepared from concentrated single-element stock solutions (Merck Darmstadt Germany) and were matrix matched with respect to the acid concentration of the samples. The standard for Ag was prepared separately in nitric acid in order to avoid precipitation of AgCl. The elements contained in a particular standard were dic- tated by chemical stability and mutual spectral interference considerations. Thus Ag occurred on its own in order to avoid precipitation. The standard used for the determination of Zn and Bi did not contain Cu owing to mutual spectral line interferences. Po1ychromator:-- System Grating 170-450 nm Dispersion 0.35 nm mm-' R.f. generator Torch Jobin Yvon Ryton de-mountable Jobin Yvon JY 48 1 m 2550 grooves mm-' spectral range Plasma Therm 2.5 kW < 10 W reflected Table 3 ICP operating conditions Pneumatic nebulizer Meinhard TR-C-20 45 psi* 1.2 1 min-l Plasma gas 14 1 min-l Intermediate gas 0.2-0.4 1 min - ' Sheath gas Trassy-Mermet pneumatic nebulizer 0.2 1 min-' Aerosol carrier gas 0.95 ml min-' Washout period Meinhard nebulizer 30 s Integration period Polychromator 10 s monochromator 0.5 s * 1 psi = 6894.76 Pa.Instrumentation and ICP Operating Conditions The major minor and trace elements were determined using a Jobin Yvon JY 48 polychromator and a JY 38 sequential system.The latter instrument was employed for the determi- nation of elements using alternative interference free spectral lines and for the determination using lines that are subject to interference in the polychromator but are suitable for use in the sequential system owing to its superior resolution. Details of the instrumentation and ICP operating conditions are listed in Tables 2 and 3. ference of Zn I 213.856 nm and Cu I 213.851 nm Fe I 226.505 on Cd I1 226.502nm mutual interference of As 1228.812nm on Cd I1 228.802nm Pb 197.272nm on As I 197.297 Sn I 206.858 nm on Sb 206.833 Cu 328.068 nm on Ag 328.068 nm (an elevated linear background was also noted) and Cu I1 223.008 on Bi I 223.061 nm (Figs. 1-7). Consequently inter- ference free Zn I 206.200 As I 193.699 and Ag I 338.289nm were determined using a high-resolution JY 38 sequential system.A spectral line interference correction due to Sn was made for Sb 206.833nm. Tin is a major element in the specimens analysed and the correction was made for concen- trations amounting to 500 mg 1 of Sn (50% in the solid samples). Similarly an inter-element correction for the effect of Fe I 226.505 on Cd I1 226.502nm was made for Fe concentrations amounting to 1000 mg 1-' (up to 100% in the solid samples). In addition a spectral correction was made for the interference of Cu I1 223.008 on Bi I 223.061 nm up to 800 mg 1-' of Cu (80% in the solid sample). It should also be mentioned that in cases of high As content (Lower Bronze Results Selection of Spectral Wavelength The spectral wavelengths and background compensation pos- itions were selected by studying spectral scans in the regions of interest.These are listed in Table 4. Spectral line selection was made on the basis of minimum spectral line interferences and maximum sensitivity. Significant spectral interferences were observed for the following spectral lines mutual inter- Table 1 factors 500-1000 Calibration standards for the analysis of bronzes and copper alloys. Concentrations are given in mg I-' in solution. Dilution Standard Element and line Sn I1 As I Mo I1 Zn I Pb I1 Bi I c o I1 Cd I1 Ni I1 Au I Mn I1 Fe I1 Cr I1 v I1 c u I Ag 1 Sb I s c I1 Wavelength/ nm 189.980 193.699 202.030 206.200 220.353 223.061 228.616 226.502 23 1.604 242.795 257.610 259.940 267.716 3 10.230 324.754 338.289 206.833 361.384 Blank 1 0 20 0 0 0 0 20 0 0 0 0 0 0 0 20 0 0 0 0 - - - - - - - - - - - 800 - - - 5 mg 1-l internal standard 2 3 4 200 - 5 20 - - 20 - 20 - - 400JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 739 7213.856 nm Cu 1213.851 nrn Table 4 Analytical ranges and spectral lines used for the analysis of Cu-Sn-Pb and Ag objects Element and line Sn I1 As I Mo I1 Zn I Pb I1 Bi I c o I1 Cd I1 Ni I1 . Au I Mn I1 Fe I1 Cr I1 v I1 c u I Ag 1 Sb I s c I1 Wavelength/nm 189.980 193.699 202.030 206.200 220.353 223.061 228.6 16 226.502 231.604 242.79 5 257.610 259.940 267.716 310.230 324.754 338.289 206.833 361.384 Background/nm 0.03 1 0.058 0.03 1 0.047 0.03 1 0.05 1 0.05 1 0.03 1 0.03 1 0.03 1 0.03 1 0.03 1 0.05 1 - 0.037 - 0.037 __ -0.037$ - Instrument* SIM SEQ SIM SEQ SIM SIM SIM SIM SIM SIM SIM SIM SIM SIM SIM SEQ SIM SIM Integration time/s 10 0.5 10 0.5 10 10 10 10 10 10 10 10 10 10 10 10 10 0.5 LODt/pg 1-' 349.0 31.0 10.0 2.6 22.0 16.0 5.7 1.7 11.0 3.5 0.7 8.3 2.0 3.5 0.7 1.2 17.0 __ * SIM polychromator JY 48; SEQ sequential JY 38.t LOD limit of detection. $ When Fe is >5% the background for Sb is 0.051 nm. 2 13.806 213.856 213.906 Wavelengthlnm Fig. 1 Spectral scan of Zn I 213.856 nm (corrected for up to 800 mgl-l Cu) age objects) Cd I1 228.802 nm is not suitable owing to the interference of As I 228.812 nm. Additionally when samples contain high Pb contents As 197.297nm is subject to inter- ference from the intense Pb 197.272 nm line. Thus the use of a high-resolution sequential spectrometer was an advantage and allowed alternative spectral lines to be employed.It is evident that the alloy elements of archaeological bronzes cause significant spectral line interferences that must be taken into consideration when selecting spectral lines for the determi- nation of trace elements. Limits of Detection The 2a limits of detection were determined using ultra-pure matrix solutions as blanks. These are listed in Table 4. 226.552 226.452 226.502 Wavelengthlnm Fig. 2 Spectral scan of Cd I 226.502 nm (corrected for up to 1000 mg 1-1 Fe) Precision and Accuracy The precision [relative standard deviation (YO RSD)] of deter- mination without the use of Sc I1 as the internal standard varied from about 0.2 to 6%. The use of Sc I1 as the internal standard resulted in a significant improvement both in the precision (from <0.1 to about 3%) and accuracy (Table 5).Internal standardization compensated for physical interference effects in the sample introduction system. High RSD values observed for Bi could be attributed to the low signal-to- background ratio and the inadequacy of the Sc I1 line to compensate for variations in these intensities. The accuracy of the analytical protocol was determined by740 I Cu1328.068nm I I JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 Cd I I I ! i As1228.812nm 228.712 228.802 228.892 Wavelengthlnm Fig.3 Spectral scan of Cd I 228.802nm. Note As I 228.812 interference t > v) C a C 4- .- 4- - ~~ . . . . . . :,Pb 197.272 nm . . . * . . . . . . @ a . . . .'.*. .... .. . . . : ..'. . . -.-../I.- .-.-. -. . . . I I I 197.247 197.297 197.347 Wavelengthhm Fig. 4 Spectral scan of As I 197.297 nm. Note Pb 197.272 interference analysing certified reference BCS 207 Bronze MBH GM 50 MBH LB 30 and a simulated synthetic standard containing a wide range of trace elements of interest. A comparison of the data obtained with the recommended values for the major minor and several trace elements indicated that the accuracy obtained is satisfactory (Table 5 ) . t 1. v) C a C 4- .- 4- - 1 . .... Sn i 206.858 rimy' Sb 1206.833 nm I I j.783 206.833 206.883 Wave1 engt hln m Fig.5 Spectral scan of Sb I 206.833nm. Note Sn I 206.858nm interference t > v) S a C .id .- 4- - I\ I ' I 'I/ Ag 1328.068 nm I I \ I 1 1 1 I 1 1 1 1 1 1 1 I I I I 328.018 328.1 18 328.068 Wavelengt hln m Fig.6 Spectral scan of Ag I 328.068nm.Note Cu I 328.068 nm interference and significant elevation of the background Conclusions In developing a multi-element method for the determination of trace and minor elements in Cu-Sn-Pb archaeological objects using ICP-AES there are two main concerns selection of the appropriate spectral lines and the dissolution procedure ensuring that all the components (Pb and Sn) are present in the solution and that they remain stable over a reasonable period of time. The addition of diethylenetriamine resulted in prolonged stabilization of Ag bearing solutions. The study indicated that significant spectral line interferences can occur owing to the presence of high concentrations ofJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL.9 Table 5 Analytical data for certified reference bronzes and similar materials with and without internal references 741 Element Sn As Mo Zn Sb Pb Bi c o Cd Cd Ni Mn Fe Cr V c u Ag Element Sn As Mo Zn Sb Pb Bi Co Cd Cd Ni Mn Fe V cu Ag Wavelength/ 189.980 193.699 202.030 206.200 206.8 3 3 220.353 223.061 228.616 226.502 228.802 231.604 257.610 259.943 267.716 310.230 324.754 338.289 nm Instrument SIM SEQ SIM SEQ SIM SIM SIM SIM SIM SEQ SIM SIM SIM SIM SIM SIM SEQ Wavelength/ 189,980 193.699 202.030 206.200 206.833 220.353 223.061 228.616 226.502 228.802 231.604 257.610 259.943 310.230 324.754 338.289 nm In strument SIM SIM SIM SIM SIM SIM SIM SEQ SIM SIM SIM SIM SIM SEQ SEQ SEQ Simulated bronze SRM 207t With IR? Without IR Recommended/ Determined/ RSD mg 1-' mgl-' (%) 100 5 5 5 5 15 5 5 5 5 5 15 5 5 200 - 99.9 4.78 4.82 4.81 4.75 13.8 - 4.86 4.82 4.72 4.87 4.86 4.81 4.92 13.4 205.6 0.47 1.40 0.67 0.3 0.71 0.58 0.55 0.03 0.04 0.50 0.06 0.02 0.10 0.06 0.68 - MBH GM 50 Gun Metal? With IR Recommended Determined RSD (% m/m) (X m/m) (%) 4.73 5.1 0.2 0.05 0.046 2.3 5.08 5.13 1.2 0.08 0.07 2.3 4.12 5.4 0.12 0.05 < 0.01 - - - - - - - 1.45 1.56 0.3 0.05 0.066 0.1 0.23 0.21 0.1 Determined/ RSD Recommended mgl-' (%) (% m/m) 113.2 0.40 5.35 0.58 5.02 0.37 5.14 0.49 3.64 1.80 15.1 1.70 5.10 0.52 5.11 1.6 5.05 0.90 5.05 0.52 5.08 0.70 14.9 0.67 5.05 0.65 5.15 0.56 222.3 0.15 - - - ~- 9.80 0.05 2.50 0.41 <0.078 - - - - - 0.09 0.06 - - - 86.84 0.02 MBH LB 30 Leaded bronze With IR Recommended (% m/m) 10.3 0.02 < 0.01 0.04 9.4 0.025 - - - - 1.52 <0.01 <0.01 - __ - Determined (% m/m) 11.4 < 0.05 <0.1 0.03 10 < 0.01 - - - - 1.66 < 0.02 < 0.05 - .- - With IR Determined (X m/m) 9.70 0.06 2.45 0.38 - - t 0 .1 - - - 0.097 0.06 - - - 85.9 0.02 RSD (%) 0.70 1.50 0.70 0.90 -. - - - __ - 2.20 0.51 - - - 0.29 2.4 Without IR Determined (% m/m) 10.70 0.06 2.48 0.43 0.10 - - - - - 0.097 0.06 - - - 87.3 0.02 RSD (%) 0.25 1.90 0.11 0.26 5.80 - - - - - 0.66 0.51 - - - 0.32 2.4 * IR internal reference. t Bureau of Analysed Samples (Newby Middlesbrough UK) Certificate No. 207 Bronze. 2 MBH Analytical (Barnet Hertfordshire UK) Certificate of Analysis C 33 X No. NGM 50. Q MBH Analytical Certificate of Analysis C 32 X No. LB 30. ~ Cu II 223.01 nm I ' Bi 1223.061 nm r/ I I I I I I I I I I -.-.--.?K... .. _. . . . ..'. .. ... .. .'.:--.z-.- d- I 223.021 223.061 223.101 Wavelengthhm Fig. 7 Spectral scan of Bi I 223.061 nm. Note Cu TI 223.01 nm interference alloying elements such as Cu Pb Sn and Fe. In certain cases unusual concentrations of As in Lower Bronze age implements could interfere with the sensitive line of Cd. The use of a high- resolution sequential spectrometer facilitated the use of these spectral lines and in cases where overlap occurred the selection of alternative lines for analysis. References Ping Y. -D. Recent Advances in the Conservation and Analysis of Artifacts University of London Institute of Archaeology London Craddock P. T. in The Egyptian Mining Temple at Timna ed. Rothenberg B.Institute For Archaeometallurgical Studies London 1988 pp. 169-181. Rothenberg B. The Ancient Metallurgy of Copper Institute For Archaeometallurgical Studies London 1990. Craddock P. T. and Glumlia-Mair A. R. Bronze Working Centers of Western Asia 1000-539 BC ed. Curtis J. Kegan Paul International British Museum London and New York 1988 Shalev S. Goren Y. Levy T. E. and Northover J. P. Archaeometry 1992 34 63. Leese M. N. Craddock P. T. Freestone I. C. and Rothenberg B. Wiener Berichte iiber Naturwissenschaft in der Kunst 198511986 213 90. Key C. A. The Cave of the Treasure ed. Bar Adon P. Israel Exploration Society Jerusalem 1980. Telecote R. F. A History of Metallurgy Metals Society London 1976. Trampuz-Orel N. Milic Z. Hudnik V. and Orel B. in Archaeometry 1990 33 267. 1987 pp. 119-124. pp. 317-326.742 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JUNE 1994 VOL. 9 10 Gliumlia-Mair A. R. Archaeometry 1992,34 107. 14 Dixon K. Russell G. M. Wall G. J. Eddy B. T. Mallett R. C. 11 Merkel J. J. MASCA 1985 3 164. and Royal S. J. The Analysis of Anode Sludges and Their Process 12 Kloner A. Ancient Maresha (In Hebrew) Kadmoniot 1991 Solutions and Beneficiation Products National Institute for p. 9596. Metallurgy Randburg South Africa 1979 Report 201 1. 13 Hughes M. J. Cowell M. R. and Craddock P. T. Archaeometry 1976 18 19. Paper 310691 2E Received November 11 1993 Accepted March 16 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900737

出版商:RSC    

年代:1994

数据来源: RSC