- Email: [email protected]
Line selection and interference correction for the analysis of tungsten alloy by inductively coupled plasma atomic emission spectrometry
Talanta ELSEVIER
Talanta 44 (1997) 47-51
Line selection and interference correction for the analysis of tungsten alloy by inductively coupled plasma atomic emission spectrometry Z u n U n g Bae a'*, S a n g H a k Lee a, S u n g H o Lee b ~Department of Chemistry, Kyungpook National University, Taegu 702-701, South Korea bCentral Research Institute, Korea Tungsten Co., Dalsung, Kyungpook 711-860, South Korea
Received 17 October 1995; revised 20 May 1996; accepted 30 May 1996
Abstract A method for the analysis of tungsten alloy to determine selected elements using inductively coupled plasma atomic emission spectroscopy is described, with emphasis on line selection and spectral interference. The spectral interference coefficients were calculated for the spectral lines of selected major and trace elements. These values were used to select analytical lines and to calibrate concentrtions of the analytes. The detection limits of the elements for this method were determined and compared with those obtained by flame atomic absorption spectrometry and direct current carbon arc emission spectrometry. The results indicated that the detection limits for all of the elements determined by the proposed method are significantly better than those obtained by other techniques. In this study, the analytical reliability of the proposed method was estimated by comparison of the analytical data for the two types of tungsten alloys produced by the Korean Tungsten Company with those obtained by the matrix matching method and the results indicated that the accuracy of multi-element analysis is satisfactory. Keywords: Inductively coupled plasma atomic emission spectrometry; Interference correction; Line selection; Tungsten alloy
I. Introduction
Tungsten alloys are of great importance in various branches of industry. They are mainly employed for radiation shields, balance weights and inertial navigation systems [1]. A detailed knowledge of the levels of the major and trace elements in metal alloys is essential since they may have * Corresponding author. Fax: (82) 53-950-6330.
either a deleterious or beneficial effect upon the mechanical and physical properties of alloys [25]. Trace element analysis of metal alloys can be a problem owing to the difficulty of dissolution and the instability of the sample solutions. Inductively coupled plasma atomic emission spectroscopy (ICP-AES) is a popular tool in the determination of trace elements [6-22], but a current limitation is the spectral interferences observed with elements that emit line-rich spectra. In order to
0039-9140/97/$15.00 © 1997 Elsevier Science B.V. All rights reserved PII S0039-9140(96)02007-3
48
z.u. Bae et al. / Talanta 44 (1997) 47 51
overcome this problem, a number of studies have been carried out to develop methods for isolating elements from complex matrices, including coprecipitation [9-11], complex formation [12,13] hydride conversion [14], adsorption [15], ion exchange [16] and chromatography [17-19]. However, these methods are concomitant with the risk of contamination from the chemicals and tedius time-consuming procedures. In this study, the interfering spectral lines for nearly all of the elements in tungsten alloy that can be determined by ICP-AES were investigated. Spectral interferences, which arise from the incomplete isolation of the net signal at the analysis line, seriously affect the analytical results, particularly when high concentrations of matrix constituents are present. For the analytical lines of elements of interest we calculated the spectral interference correction coefficient (K0.), which is defined by the ratio of the spurious concentration observed for element i and the actual concentration of interference j to evaluate quantitatively the amount of spectral interference on the individual lines. The corrected concentrations of the elements for the real samples were obtained using the K o. values and the results were compared with those obtained by other analytical methods such as flame atomic absorption spectrometry (FAAS) and direct current carbon arc atomic emission spectrometry (d.c. arc AES). The detection limits for the analytes were also evaluated for the procedure suggested in this paper.
2. Experimental 2.1. Instrumentation
For ICP-AES, a JY 38 Plus ICP system (JobinYvon, Longjumeau, France) was employed. Argon was used to purge the optical path and the monochromator for wavelengths in the range 180-195 nm. Since solutions containing hydrofluoric acid were employed in the present study, a hydrofluoric acid-resistant sample introduction system was used. This system consists of a platinum-iridium concentric nebulizer (JobinYvon; Cat. No. 20925080), a PTFE nozzle, a
Scott-type coaxial plastic spray chamber (JobinYvon; Cat. No. 11285268), a plastic sheathing tube and a sapphire sample introduction tube. A force-fed Miniplus II peristaltic pump (JobinYvon; Cat. No. 21357000) was used to deliver sample solution at a rate of 1.5 ml rain -~. The other operating conditions for ICP-AES are listed in Table 1. For FAAS a Model 3030B atomic absorption spectrometer (Perkin-EImer, Norwalk, CT, USA) was used, and for d.c. arc AES a Model 70-000 system (Thermo Jarrell Ash, Franklin, MA, USA) with a 3.4 m focal length monochromator with a ruled grating was employed. 2.2. Samples and reagents
Two types of tungsten alloys (Grades P and T) used for real sample analysis were supplied by the Korea Tungsten Company (Kyungpook, Korea). Water was purified with a Milli-Q system (Millipore, Bedford, MA, USA). Concentrated hydrofluoric acid and nitric acid were of analytical-reagent grade. Metal standard solutions for ICP measurements were prepared by diluting 1000 ppm stock solutions obtained from Spex Industries, (Edison, N J, USA). Internal standard and buffer materials used for d.c. arc AES were silver chloride (Zeebac; reagent grade) and Specpure graphite powder (Bay Carbon; spectrographic grade), respectively. 2.3. Sample preparation 2.3.1. I C P - A E S and A A S
Tungsten and related samples were dissolved in a mixture of high-purity hydrofluoric and nitric Table 1 Operating conditionsfor ICP-AES Sample delivery Ar gas flows: Cooling Coating Nebulizing Observation height Slit widths Spectral range Generator frequency
1.5 ml rain- ~usingperistalticpump 13 15 lmin-I 0.2 0.4 1min 0.3-0.5 1min15 mm above upper coil Entrance 20 /~m, exit 20 /~m 170 800 nm 40.68 MHz
49
Z.U. Bae et al. / Talanta 44 (1997) 47 51 Table 2 ICP-AES interference correction coefficients (Ko) for tungsten alloy analysis" Analyte line (nm)
K0 Co
AI Co Cr Fe Nb Ni Ta Ti V W
396.152 238.892 284.325 238.204 313.079 231.604 238.706 323.451 290.882 207.911
Nb
0.004 0.109 0.076 0.085 0.520 0.418 0.002 0.066 0.005
1.247 0.005 0.204 0.038 0.005 69.46 0.074 8.308 0.006
Ta
Ti
W
0.719 0.083 0.204 0.260 0.655 7.652
0.528 0.005 0.102 0.003 84.16 0.158 0.009
0.018 0.222 0.020 0.027 0.025 0.107 0.264 0.005 0.028
0.185 0.253 0.008
0.004 0.005
'~Only the K,/values for major elements are listed. The analyte wavelengths with the lowest EK~ values are listed.
acids, adding 3 ml of each acid slowly to 1 g of sample. The mixture was then heated in a microwave digestion system (CEM; Model MDS-81D) until dissolution was achieved. The system was operated at 90% power (576 W) for 3 min, 0% for 3 min and 50% (320 W) for 3 min in sequence. The resulting solution was diluted with deionized water to give a final volume of 100 ml. In order to determine minor components in samples using the matrix matching method, standard solutions were prepared using aqueous solutions containing the same percentage of major constituents as those in the real samples. 2.3.2. D.c. arc A E S
For d.c. arc AES, the sample was mixed with buffer (2 parts sample + 3 parts buffer) with a Wig-L-Bug shaker for 60 s. The composition of buffer used was 1 part silver chloride (Zeebac; reagent grade) and 2 parts Specpure graphite powder (Bay Carbon; spectrographic grade). The mixture of 75 mg was then loaded into graphite cups (Bay Carbon; Model S-12). Trace element analysis was carried out with a current of 10 A and an arcing period of 40 s. The signal at the 244.793 nm line of Ag was used as an internal standard.
3. Results and discussion 3. I. Line selection
Line selection was performed for the major and trace elements using a solution containing 10 g lof each of the major components. Spectral interferences of spectral lines of the elements to be determined were studied by scanning the spectra of the interfering elements using the profiling mode of the analysis program. For the simultaneous determination of elements of interest in tungsten alloy, the most sensitive line for each element cannot always be used owing to the possible interferences of spectral lines of major components. In order to evaluate quantitatively the amount of spectral interferences, the interference ocrrection coefficients (K¢) for all of the lines of the elements were calculated. K 0 is defined by Eq. (1) and the values are listed in Table 2 [23-25]. spurious concentration observed for element i (ng ml- ~) K,jactual concentration of interferent j (/t g ml - l)
(1)
A K~ value of 0.01 means that a spurious concentration of 100 ppb was observed for the element i
50
Z.U. Bae et al. / Talanta 44 (1997) 47-51
Table 3 Detection limits (ppm) for elemental analysis of tungsten alloy by ICP-AES
Elements
Wavelength (nm)
K0 method
FAAS
D.c. Arc AES
A1 Co Cr Fe Nb Ni Ta Ti V W
396.152 238.892 284.325 238.204 313.079 231.604 238.706 323.451 290.882 207.911
0.010 0.022 0.010 0.015 0.025 0.050 0.480 0.005 0.013 0.030
0.091 0.080 0.055 0.040
0.07 0.06 0.05 0.05 0.52 0.05 0.53 0.04 0.09 0.50
in the matrix solution of 1.00% (w/v). Therefore, it is recommended to select the wavelength at which the sum of K o. values of the interferent elements are the lowest. The corrected concentrations using the K~j values were calculated from the equation
0.070 0.051 1.0
C~ = < - E K~ Cj
(2)
where C~, Ci and Cj are the corrected, measured and matrix concentration, respectively. Thus the corrected concentrations should provide more reliable
Table 4 Analytical results for multi-element analysis by ICP-AES using Ko and matrix matching methods Sample
Grade P
Grade T
a
Element
Wavelength (nm)
Co Nb Ta Ti W Fe Ni A1 Cr V
238.892 313.079 238.706 323.451 207.911 238.204 231.604 396.152 284.325 290.882
Co Nb Ta Ti W Fe Ni AI Cr V
238.892 313.079 238.706 323.451 207.911 238.204 231.604 396.152 284.325 290.882
Certified
Ku methoda
value
5.0+0.1 2.0 _+0.1 2.0 + 0.1 85.0 + 0.8
10.0+0.5 2.00 _+0.1 7.0 _+0.2 8.00 _+0.2 60.0 -+ 0.8
Errors are standard deviations for five measurements.
5.05_+0.10 <0.03 1.90 _+0.10 2.05 _+0.08 85.40 _+0.61 95 4- 10 140 + 18 < 0.01 < 0.01 < 0.02 10.20_+0.15 1.95 _+0.10 7.20 -+ 0.12 7.93 -+ 0.13 59.3 _+0.75 115 _+ 15 120 -+ 10 < 0.01 < 0.02 <0.02
Matrix matching
Concentration
method a
unit
5.06_+0.09 1.89 + 0.15 2.08 + 0.09 85.45 + 0.50 90 + 10 145 4- 20
10.15_+0.10 1.94 +_0.09 7.25 -+ 0.15 7.90 + 0.12 59.8 _+0.80 120 _+ 13 115 + 15
% % % % % ppm ppm ppm ppm ppm % % % % % ppm ppm ppm ppm ppm
Z.U. Bae et al. / Talanta 44 (1997) 47 51
data in the determination of trace elements in a tungsten matrix,
3.2. Detection limit The detection limits for 10 elements are listed in Table 3. These detection limits are the concentrations required to give a signal three times greater than the standard deviation of the background fluctuation. The relative standard deviation of the background was determined for each selected analytical line for 10 replicates measured with a 0.5 s integration time. The average value of the standard deviation was found to be about 1% for the matrix solution. This value does not significantly differ from that found for pure waer. On the other hand, the signal-tobackground ratios were substantially decreased in the presence of matrix, which explains the difference between the detection limits measured in water and with matrix. In comparison with the detection limits in aqueous solution, the ICP-AES detection limits in the matrix solution were decreased by factors varying from 1 to 20. The values of the detection limits for the elements determined by the present technique are also compared with those obtained by FAAS and d.c. arc AES in Table 3. The results indicate that the detection limits for all the elements except V, Ta and Ni determined by the proposed method are better than those obtained by the other techniques.
3.3. Analysis of tungsten alloys Determinations of major and trace elements in two types of tungsten alloys obtained from the Korea Tungsten Company were performed by the Ko correction method. The results are presented in Table 4. The certified values listed in Table 4 were obtained by collaborative trials. The concentrations of major elements obtained by the K,j correction method are in good agreement with the certified values and those obtained by the matrix matching technique. The analytical data for the trace elements were also
51
found to agree with those obtained by the matrix matching method.
References [1] M.B. Bever, Encyclopedia of Materials and Engineering, MIT Press, Boston, MA, 1986, p. 4159. [2] H. Yamaguchi, K. Yamada, O. Kujirai and R. Hasegawa, Bunseki Kagaku, 42 (1993) 145. [3] D.Herbert, A. Bbdolvahab, L. Benno, K. Erich, F. Gernot and G. Manfred, Mikrochim. Acta, 108 (1992) 163. [4] H. Danninger, W. Pisan and B. Lux, Int.J. Refract. Hard Met., 5 (1986) 144. [5] E. Kny and H.M. Otner, Int. J. Refract. Hard Met., 4 (1985) 77. [6] G. Henrion, J. Gelbrecht and H. Lippert, Z. Chem., 20 (1980)108. [7] F. Mogi, K. Itoh, N. Okamoto, M. Narita and M. Fujine, Denki Seiko, 59 (1988) 263. [8] X. Xu, F. Zhang and J. Zhang, Youkuangye, 8 (1989) 45. [9] K.C. Ng and J.A. Caruso, Anal. Chem., 55 (1983) 1513. [10] O. Kujirai, M. Kohri, K. Yamada and H. Okochi, Anal. Sci., 6 (1990) 379. [ll] O. Kujirai, K. Yamada, M. Kohri and H. Okochi, Fresenius' J. Anal. Chem., 339 (1991) 133. [12] K.C. Ng, M. Zerezghi and J.A. Caruso, Anal. Chem., 56 (1984) 417. [13] J.M.E. Almendro, C.B. Ojeda, A.G. de Torres and J.M.C. Pavon, Analyst, 117 (1992) 1749. [14] E.D. Salin and MM. Habib, Anal. Chem. 56 (1984) 1186. [15] Z. Guangyao, S. Zhixing, L. Xingyin and C. Xijun, Anal. Lett., 25 (1992) 561. [16] L.L.W. Somasiri, A. Birnie and A.C. Edwards, Analyst, l l6 (1991) 601. [17] E.D. Salin and G. Horlick, Anal. Chem., 51 (1979) 2284. [18] K.A. Forbes, J.F. Vecchiarelli, P.C. Uden and R. Barnes, Anal. Chem., 62 (1990) 2033. [19] N.S. Safronova, S. Matveeva, Y.I. Favelinsky and V.A. Ryabukhin, Analyst, 120 (1995) 1427. [20] R. Ullmann and H. Ringer, Fresenius' J. Anal. Chem., 323 (1986) 139. [21] I.B. Brenner, S. Erlich, G. Vial, J. MacCormack, P. Grosdaillon and Aric L.E. Asher, J. Anal. At. Spectrom., 2 (1987) 637. [22] M. Carre, O. Diaz de Rodriguez, J.M. Mermet, M. Bridenne and Y. Marot, J. Anal. At. Spectrom., 6 (1991) 49. [23] P.W.J.M. Boumans, in P.W.J.M. Boumans (Ed.), Inductively Coupled Plasma Emission Spectroscopy, Part 1, Wiley, New York, 1987, p. 421. [24] R.I. Botto, in R.M. Barnes (Ed.), Developments in Atomic Plasma Spectrochemical Analysis, Heyden, Philadelphia, 1981, p. 141. [25] R.I. Botto, Anal. Chem., 54 (1982) 1654.
Talanta 44 (1997) 47-51
Line selection and interference correction for the analysis of tungsten alloy by inductively coupled plasma atomic emission spectrometry Z u n U n g Bae a'*, S a n g H a k Lee a, S u n g H o Lee b ~Department of Chemistry, Kyungpook National University, Taegu 702-701, South Korea bCentral Research Institute, Korea Tungsten Co., Dalsung, Kyungpook 711-860, South Korea
Received 17 October 1995; revised 20 May 1996; accepted 30 May 1996
Abstract A method for the analysis of tungsten alloy to determine selected elements using inductively coupled plasma atomic emission spectroscopy is described, with emphasis on line selection and spectral interference. The spectral interference coefficients were calculated for the spectral lines of selected major and trace elements. These values were used to select analytical lines and to calibrate concentrtions of the analytes. The detection limits of the elements for this method were determined and compared with those obtained by flame atomic absorption spectrometry and direct current carbon arc emission spectrometry. The results indicated that the detection limits for all of the elements determined by the proposed method are significantly better than those obtained by other techniques. In this study, the analytical reliability of the proposed method was estimated by comparison of the analytical data for the two types of tungsten alloys produced by the Korean Tungsten Company with those obtained by the matrix matching method and the results indicated that the accuracy of multi-element analysis is satisfactory. Keywords: Inductively coupled plasma atomic emission spectrometry; Interference correction; Line selection; Tungsten alloy
I. Introduction
Tungsten alloys are of great importance in various branches of industry. They are mainly employed for radiation shields, balance weights and inertial navigation systems [1]. A detailed knowledge of the levels of the major and trace elements in metal alloys is essential since they may have * Corresponding author. Fax: (82) 53-950-6330.
either a deleterious or beneficial effect upon the mechanical and physical properties of alloys [25]. Trace element analysis of metal alloys can be a problem owing to the difficulty of dissolution and the instability of the sample solutions. Inductively coupled plasma atomic emission spectroscopy (ICP-AES) is a popular tool in the determination of trace elements [6-22], but a current limitation is the spectral interferences observed with elements that emit line-rich spectra. In order to
0039-9140/97/$15.00 © 1997 Elsevier Science B.V. All rights reserved PII S0039-9140(96)02007-3
48
z.u. Bae et al. / Talanta 44 (1997) 47 51
overcome this problem, a number of studies have been carried out to develop methods for isolating elements from complex matrices, including coprecipitation [9-11], complex formation [12,13] hydride conversion [14], adsorption [15], ion exchange [16] and chromatography [17-19]. However, these methods are concomitant with the risk of contamination from the chemicals and tedius time-consuming procedures. In this study, the interfering spectral lines for nearly all of the elements in tungsten alloy that can be determined by ICP-AES were investigated. Spectral interferences, which arise from the incomplete isolation of the net signal at the analysis line, seriously affect the analytical results, particularly when high concentrations of matrix constituents are present. For the analytical lines of elements of interest we calculated the spectral interference correction coefficient (K0.), which is defined by the ratio of the spurious concentration observed for element i and the actual concentration of interference j to evaluate quantitatively the amount of spectral interference on the individual lines. The corrected concentrations of the elements for the real samples were obtained using the K o. values and the results were compared with those obtained by other analytical methods such as flame atomic absorption spectrometry (FAAS) and direct current carbon arc atomic emission spectrometry (d.c. arc AES). The detection limits for the analytes were also evaluated for the procedure suggested in this paper.
2. Experimental 2.1. Instrumentation
For ICP-AES, a JY 38 Plus ICP system (JobinYvon, Longjumeau, France) was employed. Argon was used to purge the optical path and the monochromator for wavelengths in the range 180-195 nm. Since solutions containing hydrofluoric acid were employed in the present study, a hydrofluoric acid-resistant sample introduction system was used. This system consists of a platinum-iridium concentric nebulizer (JobinYvon; Cat. No. 20925080), a PTFE nozzle, a
Scott-type coaxial plastic spray chamber (JobinYvon; Cat. No. 11285268), a plastic sheathing tube and a sapphire sample introduction tube. A force-fed Miniplus II peristaltic pump (JobinYvon; Cat. No. 21357000) was used to deliver sample solution at a rate of 1.5 ml rain -~. The other operating conditions for ICP-AES are listed in Table 1. For FAAS a Model 3030B atomic absorption spectrometer (Perkin-EImer, Norwalk, CT, USA) was used, and for d.c. arc AES a Model 70-000 system (Thermo Jarrell Ash, Franklin, MA, USA) with a 3.4 m focal length monochromator with a ruled grating was employed. 2.2. Samples and reagents
Two types of tungsten alloys (Grades P and T) used for real sample analysis were supplied by the Korea Tungsten Company (Kyungpook, Korea). Water was purified with a Milli-Q system (Millipore, Bedford, MA, USA). Concentrated hydrofluoric acid and nitric acid were of analytical-reagent grade. Metal standard solutions for ICP measurements were prepared by diluting 1000 ppm stock solutions obtained from Spex Industries, (Edison, N J, USA). Internal standard and buffer materials used for d.c. arc AES were silver chloride (Zeebac; reagent grade) and Specpure graphite powder (Bay Carbon; spectrographic grade), respectively. 2.3. Sample preparation 2.3.1. I C P - A E S and A A S
Tungsten and related samples were dissolved in a mixture of high-purity hydrofluoric and nitric Table 1 Operating conditionsfor ICP-AES Sample delivery Ar gas flows: Cooling Coating Nebulizing Observation height Slit widths Spectral range Generator frequency
1.5 ml rain- ~usingperistalticpump 13 15 lmin-I 0.2 0.4 1min 0.3-0.5 1min15 mm above upper coil Entrance 20 /~m, exit 20 /~m 170 800 nm 40.68 MHz
49
Z.U. Bae et al. / Talanta 44 (1997) 47 51 Table 2 ICP-AES interference correction coefficients (Ko) for tungsten alloy analysis" Analyte line (nm)
K0 Co
AI Co Cr Fe Nb Ni Ta Ti V W
396.152 238.892 284.325 238.204 313.079 231.604 238.706 323.451 290.882 207.911
Nb
0.004 0.109 0.076 0.085 0.520 0.418 0.002 0.066 0.005
1.247 0.005 0.204 0.038 0.005 69.46 0.074 8.308 0.006
Ta
Ti
W
0.719 0.083 0.204 0.260 0.655 7.652
0.528 0.005 0.102 0.003 84.16 0.158 0.009
0.018 0.222 0.020 0.027 0.025 0.107 0.264 0.005 0.028
0.185 0.253 0.008
0.004 0.005
'~Only the K,/values for major elements are listed. The analyte wavelengths with the lowest EK~ values are listed.
acids, adding 3 ml of each acid slowly to 1 g of sample. The mixture was then heated in a microwave digestion system (CEM; Model MDS-81D) until dissolution was achieved. The system was operated at 90% power (576 W) for 3 min, 0% for 3 min and 50% (320 W) for 3 min in sequence. The resulting solution was diluted with deionized water to give a final volume of 100 ml. In order to determine minor components in samples using the matrix matching method, standard solutions were prepared using aqueous solutions containing the same percentage of major constituents as those in the real samples. 2.3.2. D.c. arc A E S
For d.c. arc AES, the sample was mixed with buffer (2 parts sample + 3 parts buffer) with a Wig-L-Bug shaker for 60 s. The composition of buffer used was 1 part silver chloride (Zeebac; reagent grade) and 2 parts Specpure graphite powder (Bay Carbon; spectrographic grade). The mixture of 75 mg was then loaded into graphite cups (Bay Carbon; Model S-12). Trace element analysis was carried out with a current of 10 A and an arcing period of 40 s. The signal at the 244.793 nm line of Ag was used as an internal standard.
3. Results and discussion 3. I. Line selection
Line selection was performed for the major and trace elements using a solution containing 10 g lof each of the major components. Spectral interferences of spectral lines of the elements to be determined were studied by scanning the spectra of the interfering elements using the profiling mode of the analysis program. For the simultaneous determination of elements of interest in tungsten alloy, the most sensitive line for each element cannot always be used owing to the possible interferences of spectral lines of major components. In order to evaluate quantitatively the amount of spectral interferences, the interference ocrrection coefficients (K¢) for all of the lines of the elements were calculated. K 0 is defined by Eq. (1) and the values are listed in Table 2 [23-25]. spurious concentration observed for element i (ng ml- ~) K,jactual concentration of interferent j (/t g ml - l)
(1)
A K~ value of 0.01 means that a spurious concentration of 100 ppb was observed for the element i
50
Z.U. Bae et al. / Talanta 44 (1997) 47-51
Table 3 Detection limits (ppm) for elemental analysis of tungsten alloy by ICP-AES
Elements
Wavelength (nm)
K0 method
FAAS
D.c. Arc AES
A1 Co Cr Fe Nb Ni Ta Ti V W
396.152 238.892 284.325 238.204 313.079 231.604 238.706 323.451 290.882 207.911
0.010 0.022 0.010 0.015 0.025 0.050 0.480 0.005 0.013 0.030
0.091 0.080 0.055 0.040
0.07 0.06 0.05 0.05 0.52 0.05 0.53 0.04 0.09 0.50
in the matrix solution of 1.00% (w/v). Therefore, it is recommended to select the wavelength at which the sum of K o. values of the interferent elements are the lowest. The corrected concentrations using the K~j values were calculated from the equation
0.070 0.051 1.0
C~ = < - E K~ Cj
(2)
where C~, Ci and Cj are the corrected, measured and matrix concentration, respectively. Thus the corrected concentrations should provide more reliable
Table 4 Analytical results for multi-element analysis by ICP-AES using Ko and matrix matching methods Sample
Grade P
Grade T
a
Element
Wavelength (nm)
Co Nb Ta Ti W Fe Ni A1 Cr V
238.892 313.079 238.706 323.451 207.911 238.204 231.604 396.152 284.325 290.882
Co Nb Ta Ti W Fe Ni AI Cr V
238.892 313.079 238.706 323.451 207.911 238.204 231.604 396.152 284.325 290.882
Certified
Ku methoda
value
5.0+0.1 2.0 _+0.1 2.0 + 0.1 85.0 + 0.8
10.0+0.5 2.00 _+0.1 7.0 _+0.2 8.00 _+0.2 60.0 -+ 0.8
Errors are standard deviations for five measurements.
5.05_+0.10 <0.03 1.90 _+0.10 2.05 _+0.08 85.40 _+0.61 95 4- 10 140 + 18 < 0.01 < 0.01 < 0.02 10.20_+0.15 1.95 _+0.10 7.20 -+ 0.12 7.93 -+ 0.13 59.3 _+0.75 115 _+ 15 120 -+ 10 < 0.01 < 0.02 <0.02
Matrix matching
Concentration
method a
unit
5.06_+0.09 1.89 + 0.15 2.08 + 0.09 85.45 + 0.50 90 + 10 145 4- 20
10.15_+0.10 1.94 +_0.09 7.25 -+ 0.15 7.90 + 0.12 59.8 _+0.80 120 _+ 13 115 + 15
% % % % % ppm ppm ppm ppm ppm % % % % % ppm ppm ppm ppm ppm
Z.U. Bae et al. / Talanta 44 (1997) 47 51
data in the determination of trace elements in a tungsten matrix,
3.2. Detection limit The detection limits for 10 elements are listed in Table 3. These detection limits are the concentrations required to give a signal three times greater than the standard deviation of the background fluctuation. The relative standard deviation of the background was determined for each selected analytical line for 10 replicates measured with a 0.5 s integration time. The average value of the standard deviation was found to be about 1% for the matrix solution. This value does not significantly differ from that found for pure waer. On the other hand, the signal-tobackground ratios were substantially decreased in the presence of matrix, which explains the difference between the detection limits measured in water and with matrix. In comparison with the detection limits in aqueous solution, the ICP-AES detection limits in the matrix solution were decreased by factors varying from 1 to 20. The values of the detection limits for the elements determined by the present technique are also compared with those obtained by FAAS and d.c. arc AES in Table 3. The results indicate that the detection limits for all the elements except V, Ta and Ni determined by the proposed method are better than those obtained by the other techniques.
3.3. Analysis of tungsten alloys Determinations of major and trace elements in two types of tungsten alloys obtained from the Korea Tungsten Company were performed by the Ko correction method. The results are presented in Table 4. The certified values listed in Table 4 were obtained by collaborative trials. The concentrations of major elements obtained by the K,j correction method are in good agreement with the certified values and those obtained by the matrix matching technique. The analytical data for the trace elements were also
51
found to agree with those obtained by the matrix matching method.
References [1] M.B. Bever, Encyclopedia of Materials and Engineering, MIT Press, Boston, MA, 1986, p. 4159. [2] H. Yamaguchi, K. Yamada, O. Kujirai and R. Hasegawa, Bunseki Kagaku, 42 (1993) 145. [3] D.Herbert, A. Bbdolvahab, L. Benno, K. Erich, F. Gernot and G. Manfred, Mikrochim. Acta, 108 (1992) 163. [4] H. Danninger, W. Pisan and B. Lux, Int.J. Refract. Hard Met., 5 (1986) 144. [5] E. Kny and H.M. Otner, Int. J. Refract. Hard Met., 4 (1985) 77. [6] G. Henrion, J. Gelbrecht and H. Lippert, Z. Chem., 20 (1980)108. [7] F. Mogi, K. Itoh, N. Okamoto, M. Narita and M. Fujine, Denki Seiko, 59 (1988) 263. [8] X. Xu, F. Zhang and J. Zhang, Youkuangye, 8 (1989) 45. [9] K.C. Ng and J.A. Caruso, Anal. Chem., 55 (1983) 1513. [10] O. Kujirai, M. Kohri, K. Yamada and H. Okochi, Anal. Sci., 6 (1990) 379. [ll] O. Kujirai, K. Yamada, M. Kohri and H. Okochi, Fresenius' J. Anal. Chem., 339 (1991) 133. [12] K.C. Ng, M. Zerezghi and J.A. Caruso, Anal. Chem., 56 (1984) 417. [13] J.M.E. Almendro, C.B. Ojeda, A.G. de Torres and J.M.C. Pavon, Analyst, 117 (1992) 1749. [14] E.D. Salin and MM. Habib, Anal. Chem. 56 (1984) 1186. [15] Z. Guangyao, S. Zhixing, L. Xingyin and C. Xijun, Anal. Lett., 25 (1992) 561. [16] L.L.W. Somasiri, A. Birnie and A.C. Edwards, Analyst, l l6 (1991) 601. [17] E.D. Salin and G. Horlick, Anal. Chem., 51 (1979) 2284. [18] K.A. Forbes, J.F. Vecchiarelli, P.C. Uden and R. Barnes, Anal. Chem., 62 (1990) 2033. [19] N.S. Safronova, S. Matveeva, Y.I. Favelinsky and V.A. Ryabukhin, Analyst, 120 (1995) 1427. [20] R. Ullmann and H. Ringer, Fresenius' J. Anal. Chem., 323 (1986) 139. [21] I.B. Brenner, S. Erlich, G. Vial, J. MacCormack, P. Grosdaillon and Aric L.E. Asher, J. Anal. At. Spectrom., 2 (1987) 637. [22] M. Carre, O. Diaz de Rodriguez, J.M. Mermet, M. Bridenne and Y. Marot, J. Anal. At. Spectrom., 6 (1991) 49. [23] P.W.J.M. Boumans, in P.W.J.M. Boumans (Ed.), Inductively Coupled Plasma Emission Spectroscopy, Part 1, Wiley, New York, 1987, p. 421. [24] R.I. Botto, in R.M. Barnes (Ed.), Developments in Atomic Plasma Spectrochemical Analysis, Heyden, Philadelphia, 1981, p. 141. [25] R.I. Botto, Anal. Chem., 54 (1982) 1654.