JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY DECEMBER 1991 VOL. 6 673 Effect of the Matrix on the Determination of Some Impurities in Europium(ii1) Oxide by Flame and Electrothermal Atomic Absorption Spectrometry Vera Spevackova Karel Kratzer and Ma’ja Cejchanova Department of Nuclear Chemistry Faculty of Nuclear Science and Engineering Czech Technical University 115 19 Prague 1 Brehova 7 Czechoslovakia The possibility of the direct determination of Co Cr Cu Fe Mn Ni Pb and Zn in europium oxide by flame and electrothermal atomic absorption spectrometry was studied. The purification of the europium solution used for the experiments was necessary in order to eliminate possible contaminants. Copper was measured at 327.4 nm because of spectral interference from europium at 324.75 nm.The results show that the direct determination of the elements under study is possible. Keywords Europium; impurity; flame atomic absorption spectrometry; electrothermal atomic absorption spectrometry; spectral interference The determination of Co Cr Cu Fe Mn Ni Pb and Zn in pure europium oxide is necessary in the electronics indus- try where the highest content of these elements has to be lower than 0.005%. Atomic absorption spectrometry (AAS) flame and electrothermal atomization was chosen for the analysis because of its selectivity and sensitivity. The verification of the method and the study of the possible matrix effect on the determination of each element under study was the aim of this work. In preliminary experiments on synthetic solutions an increase in the absorbance signal for Pb Zn and especially Cu after the addition of a europium solution was observed.This might be caused by impurities of Pb Zn and Cu being present in the europium solution used and/or by a matrix effect. In order to eliminate the possible contaminants purification of the europium solution was carried out. Copper Pb and Zn can be extracted from europium solution using diethyldithiocarbamate (DDC). From the published values of extraction constants1** for all the elements under study the optimum experimental condi- tions can be established. The extraction order is Cu>Pb>Zn; the pH value of the extraction should be greater than 4.’ Experimental Apparatus A Varian Techtron 875 spectrometer equipped with a burner or a GTA-95 graphite furnace were used.Measure- ments of pH were made with a Radiometer PHM 62 pH meter. Radioactivity measurements were made with a TESLA NZQ 7 17T and a NaI(T1) wall type crystal. Reagents Standard metal solutions (1 mg ml-l in 2% HN03) were obtained from Aldrich and DDC (0.02 mol dm-3) in butyl acetate from Merck. The following reagents were also used hydrochloric acid (pro analysi); acetate buffer solution (pH 5); europium oxide (99.9%) final solution 100 mg ml-l of Eu in 0.1 mol dm-3 HNO,; and carrier-free radionuclides lszEu and 6sZn. Purification of Europium Oxide In order to check the proposed extraction procedure labelling of the solution by radioisotopes was performed. Owing to the higher values of extraction constants of Cu and Pb compared with Zn all three elements were quantita- tively extracted under the same experimental conditions as for the quantitative extraction of Zn.The following proce- dure was used the pH of the solution under study was established at pH 5 with acetate buffer and 5 ml aliquots containing 0.5 pg ml-l of Zn and 10 mg ml-l of Eu labelled with 6sZn and lS2Eu were shaken with 5 ml of 0.02 mol dm-3 DDC for 3 min. After separation of the phases the measurements of the activities were performed. The extraction yield of Zn was 99.9% and the extraction of Eu was lower than 0.5%. Influence of the Europium Matrix The following model solutions corresponding to the highest limit of impurities in europium oxide (0.005% for each element) were prepared in approximately 0.1 mol dm- HN03 solution A element under study 0.5 pg ml-l; Table 1 Conditions of measurement by flame AAS Wavelength/ Slit-width/ Element nm nm c o Cr c u Fe Mn Ni Pb Zn 240.7 357.9 324.8t 248.3 279.5 232.0 217.0 213.9 0.2 0.2 0.5 0.2 0.2 0.2 1 .o 1 .o Flame Air-C2H2 Air-C2H2 Air-C2H2 Air-C2H2 Air-C2H2 Air-C2H2 Air-C2H2 N20-CZH2 Background corrected Yes No No Yes Yes Yes Yes Yes Detection limit*/ pg ml-* 0.3 0.2 0.1 0.2 0.07 0.3 0.3 0.02 *Detection limit = 30 standard deviations of the blank.tPreliminary experiment.674 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY DECEMBER 199 1 VOL. 6 Table 2 Conditions of measurement by electrothermal atomiza- tion Ash Atomization Detection Element temperature/"C temperature/"C limiting co 900 2300 40 Cr 1 000 2600 25 c u 900 2300 20 Fe 800 2300 60 Mn 800 2300 20 Ni 900 2400 50 Pb 350 2000 40 Zn 3 50 1900 5 Table 3 Results of the flame AAS measurement (concentration of each element 0.5 pg ml-l) n=3 FouncUpg ml-l Element Solution A Solution B Solution C Absorbance c o Cr Cut Fe Mn Ni Pb Zn 0.49 f 0.02 0.49 k 0.02 0.5 1 + 0.02 0.49 f 0.02 0.51 f 0.02 0.49 f 0.02 0.50 f 0.02 0.49 k 0.02 0.48 k 0.01 0.49 f 0.02 0.48 f 0.02 0.49 f 0.03 0.5 1 f 0.03 0.5 1 k 0.03 0.5OkO.01 -# ND* ND ND ND ND ND§ NDll -# 0.017 0.014 0.052 0.035 0.060 0.023 0.0 14 0.085 *ND= not detectable.tMeasured at 324.8 nm in preliminary experiments. #Value much higher than expected. §Without purification of Eu solution a value of 0.04 pg ml-l of VWithout purification of Eu solution a value of 0.13 pg ml-l of Pb was obtained. Zn was obtained. Table 4 Influence of Eu on the determination of Cu at other wavelengths Wavelength/ Solution A/ Solution B/ Solution C/ nm pg ml-' pg ml-l pg ml-l 327.4 0.5 1 f 0.02 0.48 k 0.02 0.02 f 0.02 2 17.9 0.52 f 0.03 0.47 f 0.03 0.05 f 0.02 ties of 0.005%.For lower concentrations a furnace tech- nique is necessary. When measuring Cu a spectral interfer- ence of Eu appeared (&,=324.754 nm &,=324.753 nm).6 This interference is not mentioned in frequently used manuals because the ratio of intensities (ICU:IEu= 500 1) is favourable for Cu which in the majority of samples is present in higher concentrations than Eu and therefore in such samples is not of great importance. For this reason other resonance lines for Cu were studied and the line at 327.4 nm was found to be unaffected by Eu and therefore could be used for direct determination.The results are given in Table 4. solution B 0.5 pg ml-l of the element under study + 10 mg ml-1 of Eu; and solution C 10 mg ml-l of Eu. All solutions were measured by flame and electrothermal AAS at the appropriate The measurement conditions are given in Tables 1 and 2. Results and Discussion The results of the measurements using flame AAS are given in the Table 3. From this table it follows that with the exception of Cu all the elements under study could be measured directly in a Eu matrix. The detection limit in flame AAS is sufficient for concentration limits of impuri- Conclusion The possibility of the direct determination of Co Cr Cu Fe Mn Ni Pb and Zn was verified. Preliminary observa- tions of increases in the absorption signals of Pb and Zn were caused by impurities in the synthetic Eu solution and after its purification they disappeared. For Cu there is a spectral interference of Eu at the wavelength 324.8 nm and therefore the line at 327.4 nm was used. References Stary J. and Freiser H. IUPAC-Chelating Agents Part IV Pergamon Press Oxford 1978. Kratzer K. Ph.D. Thesis University of Prague 1967. Koch 0. G. and Dedic G. A. Handbuch der Spurenanalyse Springer Berlin 1974. Manual of Flame Techniques Varian Techtron Springfield IL 1979. Manual of Flameless Techniques Varian Techtron Springfield IL 1982. Welz B. Atomabsorptionspektrometrie Verlag Chemie Wein- heim 1983 p. 312. Paper I I002 74K Received January 21st 1991 Accepted July 3rd 1991