- Email: [email protected]
Determination of lanthanum, cerium, praseodymium and neodymium in alloy steels by inductively coupled plasma atomic-emission spectrometry
Talanra, Vol. 37, No. 8, pp. 815-818, 1990 Printed in Great Britain. All rights reserved
0039.9140/9B $3.00+ 0.00 Copyright 0 1990Pergamon Press plc
DETERMINATION OF LANTHANUM, CURIUM, PRASEODYMIUM AND N~ODYMI~ IN ALLOY STEELS BY INDUCTIVELY COUPLED PLASMA ATOMIC-EMISSION SPECTROMETRY ANDRZEI M. GROSSMAN and JBBZY CIBA Institute of Analytical and Genera1 Chemistry, Silesian Technical University, 44-101 Gliwice, Poland JERZY JURCZYK and WALDEMARSPIBWOK Institute of Ferrous Metallurgy, 44-101 Gliwice, Poland {Receiyed 30 May 1989. Revised 33 October 1989. Accepted 8 December 1989) ~~-Inductively coupled plasma spectrometry has been applied to the det~ation of La, Ce, Pr and Nd in alloy steels. Spectral interterence by other alloying elements as well as by the lantbanides themselves was studied. The influence of other lanthanides on the Pr and Nd lines could be dealt with by correction equations. It was found that within the range of concentrations corresponding to mild alloy steel, at least one of the lines selected for determining the l~~a~d~ was free from interferences. The detection limits for La, Ce, Pr and Nd in steel were 5 x lo-$, 1.5 x IO-‘, 1 x 1O-4 and 2.4 x 10~‘% respectively. The procedure was tested on standard samples and by the standard-addition method.
Addition of small amounts of rare-earth elements to steel si~ificantly modifies its properties. I$2 Calorimetric methods allow the determination of either the total lanthanide content” or only the cerium.‘-* The individual rare-earth elements can be determined by XRF spectrometry but prior separation from the matrix is required ‘O-12.Inductively coupled plasma atomic-emission spectrometry (ICP-AES) has been used for the determination of Ce and Y in steels and Nimonic alloys, but Fe, Cr and Zr interfered with the Ce measurement at 418.66 nm.13 The technique has also been used for dete~ination of a number of lanthanides in geological samples after precipitation or chromatographic separation.‘4-‘q A wide range of prominent lines for use in ICP-AES determination of several elements, including La, Ce, Pr and Nd, has been given by Winge et ~1.~’ and Bouman has studied the mutual spectra1 interferences of rare earth elements at chosen lines.21 On the basis of the information above we decided to investigate the use of ICP-AES for the direct determination of La, Ce, Pr and Nd in steels. Reagents Standard solutions of the lanthanides (1 mg/ml) were prepared by dissolving the
“Specpure” oxides (Johnson and Matthey) in con~ntrat~ nitric acid. Solutions of matrix components for studying interference e&cts were made from the metals or suitable salts. All reagents were of analytical grade. Procedure The sample of steel turnings (0.5 g) was heated with 10 ml of a 3: 1 v/v mixture of concentrated hydrochloric and nitric acids in a covered beaker for 30 min. The solution was evaporated almost to dryness, 10 ml of concentrated perchloric acid were added, and the mixture was heated until white fumes were evolved, The solution was cooled, 15 ml of water were added and the mixture was heated again. The residue was collected on a medium fast filter paper, and washed with 25 ml of 0.1&f perchloric acid in small portions, the filtrate and washings being collected in a SO-ml standard flask and diluted to the mark with water. The samples used in the standard-additions method were prepared by the same procedure, with suitable amounts of lanthanide solutions added. Standard samples for calibration curves were prepared by dissolving 0.5 g of “Armco” iron as above, with addition of standard lanthanide solutions to cover the range O-40 pg/mI. Individual series of standards were made for each rare-earth element.
815
ANDRZEJM. GROSSMAN et al.
816
The solvent blank was a 10 g/l. solution of “Armco” iron prepared as above. RESULTS AND CONCLUSIONS
The choice of line for analysis was based on spectral scans over a range of f0.5 nm either side of each of the lines recommended by Winge et al.” (18 lines for La, 13 for Ce, 10 for Pr and 19 for Nd), for solutions of the pure analytes and interferents. This was done to identify the lines which would suffer least from mutual spectral interferences as well as interference from iron (the main constituent of the matrix). On the basis of the results and literature data concerning the influence of other components of the steel matrix,22g23two lines were selected for each lanthanide (Table 1). Calibration graphs were prepared by use of the standards made by addition of the individual elements to the “Armco” iron solution. High correlation coefficients (>0.99997) were obtained for regression analysis of the results, over the range O-40 pgg/ml. The dectection limits C,_were calculated as the concentration equivalent to three times the standard deviation of five measurements of the background signal. As the relative standard deviations of the background (RSDB) were less than 1%, detection limits (C,,,) were calculated for RSDB values assumed to be equal to 1% of the background signal, as recommended by Winge et a1.,20for comparison of
Table 1. Apparatus and operating conditions ICP-ARL 3520B sequential spectrometer Incident power Reflected power Coolant argon Plasma argon Nebulizer argon Liquid uptake rate
1.18 kW 10 w 1I l./min 0.87 l./min 0.9 l./min 2.2 ml/min
Entrance slit Exit slit Integration time Scanning mode Analysis mode
17 pm 20 pm 1 set 5se.c Analysis lines
Line La11 La11 CeII CeII PrII PrII NdII NdII
Identification number 4 5 1 4 2 3 3 4
Wavelength, nm
333.149 492.179 413.765 456.236 390.844 440.882 406.109 415.608
Spectral order 2
I 1 1 1 1 1 1
High voltage (arbitrary units) 5 5 9 9 9 9 I I
Table 2. Coefficients of analytical equations I = A + BC in the concentration range C = &40 mg/l., and values of detection limits Line
A
B
Correlation coefficient
C, mgll.
C,,, mgk
La 4 La 5 Ce 1 Ce 4 Pr 2 Pr 3 Nd 3 Nd 4
18.5 50.6 188.3 196.6 83.4 178.5 104.3 106.1
47.2 26.9 54.7 48.1 35.6 51.2 32.1 25.3
0.999987 0.999976 0.99997 1 0.9p9971 0.999984 0.99998 1 0.999980 0.999989
0.005 0.018 0.021 0.015 0.010 0.023 0.024 0.036
0.013 0.057 0.10 0.12 0.07 0.11 0.10 0.09
these results with literature data based on this assumption. The values of C, and C,,, are gathered in Table 2. The lines selected for La and Ce determination are free from spectral interference by other lanthanides whereas the lines chosen for Pr and Nd are not, and correction equations must be applied, as shown in Table 3. The coefficients needed are calculated from the concentrationdependence of the interferent signal measured at the wavelength used for determination of the analyte. The corrections were assumed to be additive, and their use resulted in good linearity of calibration graphs prepared from the results for mixtures of the analyte with other lanthanides, even for the 390.844 line for PrII, where the interferences are greatest (Fig. 1). The influence of other alloying elements on chosen lines of the lanthanides was examined by means of spectral scans of solutions of the interferents (of known concentrations) as well as by measurements of the signals for these elements at the wavelengths used for the lanthanides. The concentrations of these solutions of the alloying elements are given in Table 4, and are those that would be obtained by applying the sample preparation procedure to an iron or steel containing the element at the level also stated in Table 4. These levels are generally the
Table 3. Correction equations for praseodymium and neodymium Line Pr2 Pr 3 Nd 3 Nd 4
Bouation C=C,,,-O.l7lC,-0.00121CNd c = c, - 0.0373& C = C, - O.O217Cc, C = C, - O.O0956Cc,- O.OllSC,
C = true concentration (mg/l.). C, = concentration calculated from analytical curve (mg/l.). Cc, = concentration of interfering element (mg/l.).
Determination
of La. Ce, Pr and Nd in steels
mg/l. Fig. 1. The effects of correction of Pr signal at 390.844 nm: calibration curve for pure standard solutions of Pr; A measured values for samples with different amounts of other lanthanides; 0 corrected values.
highest used in alloy steels. We assumed that the interference of additives could be considered as negligible (within the range of concentrations examined) if the difference between the signals for the interferent solution and water was lower than that for CL,%. In other cases the critical concentration ratio“’ (CCR) was calculated according to
CC&s Ai/Li
where X, and Xi are the net signals for the lanthanide and interferent respectively, C, is the concentration of lanthanide and Ci the concentration of interferent quoted in Table 4. The results are given in Table 4 and show that the greatest interference is caused by V, Zr and Ti, but within the range of concentrations corresponding to low alloy steel and in many cases to
Table 4. The critical concentration Lanthanide and line, nm
Fe
La 333.749 La 492.179 Ce 413.765 Ce 456.236 Pr 390.844 Pr 440.882 Nd 406.109 Nd 415.608
133,000 88,600 17,400 23,300 48,900 11,100 15,100
Cr
Ni
54,800 92,000 7300 11,800 20,200 13,100 31,100 26,100 41,400 31,200 -
-Signifies
10.0 100
5.2 52
5.9 59
high-alloy steel, at least one line of each lanthanide is free from interference. It is advisable to prepare the lanthanide standards in a solution made from “Armco” iron as the differences caused by the changes in iron content in steel samples with different levels of alloying agents can then be neglected. The data in Table 4 are based on measurements taken with only the highest interferent concentration tested, and corrections for lower concentrations can be calculated by simple proportion from that information, but it must be taken into consideration that any background shift may form part of the measured signal. The method was tested on standard samples with certified cerium contents. Because of lack of certified samples containing all four lanthanides examined, the standardadditions method was applied. Standard samples used in the Polish metallurgical industry were used as the matrix; their composition is given in Table 5. The method used for decomposing the samples resulted in the amount of elements such as W and Nb in solution being considerably lower than that expected from the certified value, the bulk being separated with the silica. The results obtained are presented in Table 6. Those for Ce are in good agreement with the certified value. The ratios of La, Pr and Nd to Ce found for the standard samples correspond to their ratios in the mischmetall used in production of the standards. Good results were also obtained by the standard-additions method. On the basis of these results ICP spectrometry may be recommended for rapid precise determination of La, Ce, Pr and Nd in alloy steels.
ratios (CCR) for individual interferents and lanthanides
Mn 21,200 28,900 21,000 21,200 53,500
co
5.5 55
the difference in signal is less than C,,,.
w
v
19,000 20,300 11,100 2600 2940 160 2600 6600 1500 1820 410 97 1100 1310 510
Concentrations Solution, g/l. Level in steel, %
817
1.0 10
Zr 3780 1000 460 1540 1660 1350 3
Ti Nb 3900 350 7300 530 3410 1300 6100 11,200 3800 _ 1420 5600 3600
MO
Cu
Al
1000 -
-
-
of the interferents 0.5 5
1.1 11
1.0 10
0.5 5
1.0 10
0.2 2
1.0 10
1.0 10
ANDRW M. GROSSMAN ef al.
818
Table 5. IMZ standard samples of steel Concentration Sample
C
Mn
Si
P
1.714. 1.19/l 1.25/l
0.017 0.13 (0.005) 0.013 0.20 1.00 0.30 0.020 1.48 0.26 0.28 0.018
1.3411 1.26/l 1.82
0.11 0.31 0.10
0.84 0.48 1.11
0.31 0.29 0.27
0.014 0.021 0.017
S 0.032 0.017 0.025
Cr
Ni
of alloying agent, % Cu
MO
V
W
0.004 0.012 0.016 0.0017
(0.030) 0.009 0.73 1.87 0.020 0.21
0.11 0.65 4.05 (0.05) 0.14 0.39 0.085 0.069 0.060
Ti
zr
(0.-&)5) ‘1Ip 0.101 -
1 -
Ce -
1 I (0.0015) 0.02
*“Armco” iron. Table 6. Determination of lanthanides (%) in standard samples &95% confidence interval (n = 3); quantities in parentheses are amounts added (extent for Ce in 1.82, which is the cer&d value) Sample 1.19/l
1.25/l
1.26/l
1.34/l
1.82
La 333.749 La 492.179
0.198 + 0.002 0.197 f 0.003 (0.200)
0.049 f 0.002 0.049 * 0.002 (0.050)
0.098 f 0.002 0.098 f 0.002 (0.100)
0.012 f 0.001 0.012 f 0.001 -
La 413.765 Ce 456.236
0.001 f 0.002 0.001 f 0.002 (0.000) 0.050 f 0.002 0.049 f 0.002 (0.050)
0.102 + 0.003 0.098 f 0.003 (0.100)
0.000 f 0.001 0.000 f 0.001 (0.000) 0.051 f 0.002 0.050 f 0.002 (0.050)
0.196 f 0.004 0.195 f 0.003 (0.200)
0.020 f 0.001 0.020 f 0.001 (0.020 f 0.002)
0.196 f 0.003 0.199 & 0.003 (0.200)
0.099 f 0.002 0.099 * 0.002 (0.100)
0.0002 f 0.001* 0.003 + 0.001’ -
0.098 + 0.002 0.098 & 0.002 (0.100)
-0.001 + 0.001 0.000 f 0.001 (0.000)
0.198 + 0.003 0.198 f 0.003 (0.200)
0.000 f 0.001 0.001 f 0.001 (0.000) 0.048 f 0.002 0.048 f 0.002 (0.050)
Line, nm
Pr 390.844 Pr 440.882 Nd 406.109 Nd 415.608
0.007 f 0.002 0.006 f 0.002 -
*With correction for vanadium.
REFERENCES 1. M. Kepka, Hutn. Listy, 1977, 32, 171. 2. P. F. Wandby, Intern. Metals Rev., 1978, 23, 74. 3. P. K. Spitsyn and I. G. Surin, Zh. Analit. Khim., 1975,
30, 284. 4. T. I. Romantseva and L. K. Kharitonova, Zavodsk. Lab., 1979, 45, 495. 5. I. G. Surin and P. K. Spitsyn, ibid., 1980, 46, 886. 6. Z. Zhou and Y. Chen. Fenxi Huaxue, 1985, 13, 289. 7. L. I. Kharlamova, T. A. Borcheva and W. T. Solomatin, Zavodsk. Lab., 1974, 40, 1169. 8. T. Capalla, J. Jurczyk and K. Szeja, Pr. Inst. Mefal. Zelaza, 1979, 29, 59.
9. J. Pietrosz and J. Czyz, Hum. Lisry, 1978, 33, 587. 10. A. T. Kashuba and C. R. Hines, Anal. Chem., 1971,43, 1758. 11. J. Jurczyk, I. Sheybal and W. Smolec, Pr. Inst. Metal. Zelaza, 1981, 33, 93.
12. T. Sofilic, Metalurgija (Sisak), 1985, 24, 131. 13. V. Rett and I. HlavaEek, Hum. Lisfy, 1979, 34, 428.
14. J. A. C. Broekaert, F. Leis and K. Laqua, Spectrochim. Acta, 1979, 34B, 73.
15. A. Bolton, J. Hwang, A. Vander Voet, ibid., 1983,3BB, 165. 16. I. B. Brenner, E. A. Jones, A. E. Watson and T. W. Steele, Chem. Geol., 1984, 45, 135. 17. R. Aulis, A. Bolton, W. Doherty, A. Vander Voet and P. Wong. Spectrochim. Acta, 1985, 40B, 377. 18. S. J. Buchanan and L. S. Dale, ibid., 1986, 41B, 237. 19. H. Iwasaki and H. Haraguchi. Anal. Chim. Ada, 1988, 208, 163.
20. R. K. Winge, V. J. Peterson and V. A. Fassel, Appl. Spectrosc.,
1979, 33, 206.
21. P. W. J. M. Boumans, J. A. Tielrooy and F. J. M. J. Maessen, Spectrochim. Acta, 1988, 43B, 173. 22. A. N. Zaidel, V. K. Prokofev, S. M. Raiskii, V. A. Slavnyi and E. Y. Shreider, Tables of Spectral Lines, Plenum Press, New York, 1970. 23. H. L. Parson, A. Forster and D. Anderson, An Aflas of Spectral Interference in ICP Spectoscopy. Plenum Press, New York, 1980. 24. P. W. J. M. Boumans, Spectrochim. Acra, 1980,3!IB, 57.
0039.9140/9B $3.00+ 0.00 Copyright 0 1990Pergamon Press plc
DETERMINATION OF LANTHANUM, CURIUM, PRASEODYMIUM AND N~ODYMI~ IN ALLOY STEELS BY INDUCTIVELY COUPLED PLASMA ATOMIC-EMISSION SPECTROMETRY ANDRZEI M. GROSSMAN and JBBZY CIBA Institute of Analytical and Genera1 Chemistry, Silesian Technical University, 44-101 Gliwice, Poland JERZY JURCZYK and WALDEMARSPIBWOK Institute of Ferrous Metallurgy, 44-101 Gliwice, Poland {Receiyed 30 May 1989. Revised 33 October 1989. Accepted 8 December 1989) ~~-Inductively coupled plasma spectrometry has been applied to the det~ation of La, Ce, Pr and Nd in alloy steels. Spectral interterence by other alloying elements as well as by the lantbanides themselves was studied. The influence of other lanthanides on the Pr and Nd lines could be dealt with by correction equations. It was found that within the range of concentrations corresponding to mild alloy steel, at least one of the lines selected for determining the l~~a~d~ was free from interferences. The detection limits for La, Ce, Pr and Nd in steel were 5 x lo-$, 1.5 x IO-‘, 1 x 1O-4 and 2.4 x 10~‘% respectively. The procedure was tested on standard samples and by the standard-addition method.
Addition of small amounts of rare-earth elements to steel si~ificantly modifies its properties. I$2 Calorimetric methods allow the determination of either the total lanthanide content” or only the cerium.‘-* The individual rare-earth elements can be determined by XRF spectrometry but prior separation from the matrix is required ‘O-12.Inductively coupled plasma atomic-emission spectrometry (ICP-AES) has been used for the determination of Ce and Y in steels and Nimonic alloys, but Fe, Cr and Zr interfered with the Ce measurement at 418.66 nm.13 The technique has also been used for dete~ination of a number of lanthanides in geological samples after precipitation or chromatographic separation.‘4-‘q A wide range of prominent lines for use in ICP-AES determination of several elements, including La, Ce, Pr and Nd, has been given by Winge et ~1.~’ and Bouman has studied the mutual spectra1 interferences of rare earth elements at chosen lines.21 On the basis of the information above we decided to investigate the use of ICP-AES for the direct determination of La, Ce, Pr and Nd in steels. Reagents Standard solutions of the lanthanides (1 mg/ml) were prepared by dissolving the
“Specpure” oxides (Johnson and Matthey) in con~ntrat~ nitric acid. Solutions of matrix components for studying interference e&cts were made from the metals or suitable salts. All reagents were of analytical grade. Procedure The sample of steel turnings (0.5 g) was heated with 10 ml of a 3: 1 v/v mixture of concentrated hydrochloric and nitric acids in a covered beaker for 30 min. The solution was evaporated almost to dryness, 10 ml of concentrated perchloric acid were added, and the mixture was heated until white fumes were evolved, The solution was cooled, 15 ml of water were added and the mixture was heated again. The residue was collected on a medium fast filter paper, and washed with 25 ml of 0.1&f perchloric acid in small portions, the filtrate and washings being collected in a SO-ml standard flask and diluted to the mark with water. The samples used in the standard-additions method were prepared by the same procedure, with suitable amounts of lanthanide solutions added. Standard samples for calibration curves were prepared by dissolving 0.5 g of “Armco” iron as above, with addition of standard lanthanide solutions to cover the range O-40 pg/mI. Individual series of standards were made for each rare-earth element.
815
ANDRZEJM. GROSSMAN et al.
816
The solvent blank was a 10 g/l. solution of “Armco” iron prepared as above. RESULTS AND CONCLUSIONS
The choice of line for analysis was based on spectral scans over a range of f0.5 nm either side of each of the lines recommended by Winge et al.” (18 lines for La, 13 for Ce, 10 for Pr and 19 for Nd), for solutions of the pure analytes and interferents. This was done to identify the lines which would suffer least from mutual spectral interferences as well as interference from iron (the main constituent of the matrix). On the basis of the results and literature data concerning the influence of other components of the steel matrix,22g23two lines were selected for each lanthanide (Table 1). Calibration graphs were prepared by use of the standards made by addition of the individual elements to the “Armco” iron solution. High correlation coefficients (>0.99997) were obtained for regression analysis of the results, over the range O-40 pgg/ml. The dectection limits C,_were calculated as the concentration equivalent to three times the standard deviation of five measurements of the background signal. As the relative standard deviations of the background (RSDB) were less than 1%, detection limits (C,,,) were calculated for RSDB values assumed to be equal to 1% of the background signal, as recommended by Winge et a1.,20for comparison of
Table 1. Apparatus and operating conditions ICP-ARL 3520B sequential spectrometer Incident power Reflected power Coolant argon Plasma argon Nebulizer argon Liquid uptake rate
1.18 kW 10 w 1I l./min 0.87 l./min 0.9 l./min 2.2 ml/min
Entrance slit Exit slit Integration time Scanning mode Analysis mode
17 pm 20 pm 1 set 5se.c Analysis lines
Line La11 La11 CeII CeII PrII PrII NdII NdII
Identification number 4 5 1 4 2 3 3 4
Wavelength, nm
333.149 492.179 413.765 456.236 390.844 440.882 406.109 415.608
Spectral order 2
I 1 1 1 1 1 1
High voltage (arbitrary units) 5 5 9 9 9 9 I I
Table 2. Coefficients of analytical equations I = A + BC in the concentration range C = &40 mg/l., and values of detection limits Line
A
B
Correlation coefficient
C, mgll.
C,,, mgk
La 4 La 5 Ce 1 Ce 4 Pr 2 Pr 3 Nd 3 Nd 4
18.5 50.6 188.3 196.6 83.4 178.5 104.3 106.1
47.2 26.9 54.7 48.1 35.6 51.2 32.1 25.3
0.999987 0.999976 0.99997 1 0.9p9971 0.999984 0.99998 1 0.999980 0.999989
0.005 0.018 0.021 0.015 0.010 0.023 0.024 0.036
0.013 0.057 0.10 0.12 0.07 0.11 0.10 0.09
these results with literature data based on this assumption. The values of C, and C,,, are gathered in Table 2. The lines selected for La and Ce determination are free from spectral interference by other lanthanides whereas the lines chosen for Pr and Nd are not, and correction equations must be applied, as shown in Table 3. The coefficients needed are calculated from the concentrationdependence of the interferent signal measured at the wavelength used for determination of the analyte. The corrections were assumed to be additive, and their use resulted in good linearity of calibration graphs prepared from the results for mixtures of the analyte with other lanthanides, even for the 390.844 line for PrII, where the interferences are greatest (Fig. 1). The influence of other alloying elements on chosen lines of the lanthanides was examined by means of spectral scans of solutions of the interferents (of known concentrations) as well as by measurements of the signals for these elements at the wavelengths used for the lanthanides. The concentrations of these solutions of the alloying elements are given in Table 4, and are those that would be obtained by applying the sample preparation procedure to an iron or steel containing the element at the level also stated in Table 4. These levels are generally the
Table 3. Correction equations for praseodymium and neodymium Line Pr2 Pr 3 Nd 3 Nd 4
Bouation C=C,,,-O.l7lC,-0.00121CNd c = c, - 0.0373& C = C, - O.O217Cc, C = C, - O.O0956Cc,- O.OllSC,
C = true concentration (mg/l.). C, = concentration calculated from analytical curve (mg/l.). Cc, = concentration of interfering element (mg/l.).
Determination
of La. Ce, Pr and Nd in steels
mg/l. Fig. 1. The effects of correction of Pr signal at 390.844 nm: calibration curve for pure standard solutions of Pr; A measured values for samples with different amounts of other lanthanides; 0 corrected values.
highest used in alloy steels. We assumed that the interference of additives could be considered as negligible (within the range of concentrations examined) if the difference between the signals for the interferent solution and water was lower than that for CL,%. In other cases the critical concentration ratio“’ (CCR) was calculated according to
CC&s Ai/Li
where X, and Xi are the net signals for the lanthanide and interferent respectively, C, is the concentration of lanthanide and Ci the concentration of interferent quoted in Table 4. The results are given in Table 4 and show that the greatest interference is caused by V, Zr and Ti, but within the range of concentrations corresponding to low alloy steel and in many cases to
Table 4. The critical concentration Lanthanide and line, nm
Fe
La 333.749 La 492.179 Ce 413.765 Ce 456.236 Pr 390.844 Pr 440.882 Nd 406.109 Nd 415.608
133,000 88,600 17,400 23,300 48,900 11,100 15,100
Cr
Ni
54,800 92,000 7300 11,800 20,200 13,100 31,100 26,100 41,400 31,200 -
-Signifies
10.0 100
5.2 52
5.9 59
high-alloy steel, at least one line of each lanthanide is free from interference. It is advisable to prepare the lanthanide standards in a solution made from “Armco” iron as the differences caused by the changes in iron content in steel samples with different levels of alloying agents can then be neglected. The data in Table 4 are based on measurements taken with only the highest interferent concentration tested, and corrections for lower concentrations can be calculated by simple proportion from that information, but it must be taken into consideration that any background shift may form part of the measured signal. The method was tested on standard samples with certified cerium contents. Because of lack of certified samples containing all four lanthanides examined, the standardadditions method was applied. Standard samples used in the Polish metallurgical industry were used as the matrix; their composition is given in Table 5. The method used for decomposing the samples resulted in the amount of elements such as W and Nb in solution being considerably lower than that expected from the certified value, the bulk being separated with the silica. The results obtained are presented in Table 6. Those for Ce are in good agreement with the certified value. The ratios of La, Pr and Nd to Ce found for the standard samples correspond to their ratios in the mischmetall used in production of the standards. Good results were also obtained by the standard-additions method. On the basis of these results ICP spectrometry may be recommended for rapid precise determination of La, Ce, Pr and Nd in alloy steels.
ratios (CCR) for individual interferents and lanthanides
Mn 21,200 28,900 21,000 21,200 53,500
co
5.5 55
the difference in signal is less than C,,,.
w
v
19,000 20,300 11,100 2600 2940 160 2600 6600 1500 1820 410 97 1100 1310 510
Concentrations Solution, g/l. Level in steel, %
817
1.0 10
Zr 3780 1000 460 1540 1660 1350 3
Ti Nb 3900 350 7300 530 3410 1300 6100 11,200 3800 _ 1420 5600 3600
MO
Cu
Al
1000 -
-
-
of the interferents 0.5 5
1.1 11
1.0 10
0.5 5
1.0 10
0.2 2
1.0 10
1.0 10
ANDRW M. GROSSMAN ef al.
818
Table 5. IMZ standard samples of steel Concentration Sample
C
Mn
Si
P
1.714. 1.19/l 1.25/l
0.017 0.13 (0.005) 0.013 0.20 1.00 0.30 0.020 1.48 0.26 0.28 0.018
1.3411 1.26/l 1.82
0.11 0.31 0.10
0.84 0.48 1.11
0.31 0.29 0.27
0.014 0.021 0.017
S 0.032 0.017 0.025
Cr
Ni
of alloying agent, % Cu
MO
V
W
0.004 0.012 0.016 0.0017
(0.030) 0.009 0.73 1.87 0.020 0.21
0.11 0.65 4.05 (0.05) 0.14 0.39 0.085 0.069 0.060
Ti
zr
(0.-&)5) ‘1Ip 0.101 -
1 -
Ce -
1 I (0.0015) 0.02
*“Armco” iron. Table 6. Determination of lanthanides (%) in standard samples &95% confidence interval (n = 3); quantities in parentheses are amounts added (extent for Ce in 1.82, which is the cer&d value) Sample 1.19/l
1.25/l
1.26/l
1.34/l
1.82
La 333.749 La 492.179
0.198 + 0.002 0.197 f 0.003 (0.200)
0.049 f 0.002 0.049 * 0.002 (0.050)
0.098 f 0.002 0.098 f 0.002 (0.100)
0.012 f 0.001 0.012 f 0.001 -
La 413.765 Ce 456.236
0.001 f 0.002 0.001 f 0.002 (0.000) 0.050 f 0.002 0.049 f 0.002 (0.050)
0.102 + 0.003 0.098 f 0.003 (0.100)
0.000 f 0.001 0.000 f 0.001 (0.000) 0.051 f 0.002 0.050 f 0.002 (0.050)
0.196 f 0.004 0.195 f 0.003 (0.200)
0.020 f 0.001 0.020 f 0.001 (0.020 f 0.002)
0.196 f 0.003 0.199 & 0.003 (0.200)
0.099 f 0.002 0.099 * 0.002 (0.100)
0.0002 f 0.001* 0.003 + 0.001’ -
0.098 + 0.002 0.098 & 0.002 (0.100)
-0.001 + 0.001 0.000 f 0.001 (0.000)
0.198 + 0.003 0.198 f 0.003 (0.200)
0.000 f 0.001 0.001 f 0.001 (0.000) 0.048 f 0.002 0.048 f 0.002 (0.050)
Line, nm
Pr 390.844 Pr 440.882 Nd 406.109 Nd 415.608
0.007 f 0.002 0.006 f 0.002 -
*With correction for vanadium.
REFERENCES 1. M. Kepka, Hutn. Listy, 1977, 32, 171. 2. P. F. Wandby, Intern. Metals Rev., 1978, 23, 74. 3. P. K. Spitsyn and I. G. Surin, Zh. Analit. Khim., 1975,
30, 284. 4. T. I. Romantseva and L. K. Kharitonova, Zavodsk. Lab., 1979, 45, 495. 5. I. G. Surin and P. K. Spitsyn, ibid., 1980, 46, 886. 6. Z. Zhou and Y. Chen. Fenxi Huaxue, 1985, 13, 289. 7. L. I. Kharlamova, T. A. Borcheva and W. T. Solomatin, Zavodsk. Lab., 1974, 40, 1169. 8. T. Capalla, J. Jurczyk and K. Szeja, Pr. Inst. Mefal. Zelaza, 1979, 29, 59.
9. J. Pietrosz and J. Czyz, Hum. Lisry, 1978, 33, 587. 10. A. T. Kashuba and C. R. Hines, Anal. Chem., 1971,43, 1758. 11. J. Jurczyk, I. Sheybal and W. Smolec, Pr. Inst. Metal. Zelaza, 1981, 33, 93.
12. T. Sofilic, Metalurgija (Sisak), 1985, 24, 131. 13. V. Rett and I. HlavaEek, Hum. Lisfy, 1979, 34, 428.
14. J. A. C. Broekaert, F. Leis and K. Laqua, Spectrochim. Acta, 1979, 34B, 73.
15. A. Bolton, J. Hwang, A. Vander Voet, ibid., 1983,3BB, 165. 16. I. B. Brenner, E. A. Jones, A. E. Watson and T. W. Steele, Chem. Geol., 1984, 45, 135. 17. R. Aulis, A. Bolton, W. Doherty, A. Vander Voet and P. Wong. Spectrochim. Acta, 1985, 40B, 377. 18. S. J. Buchanan and L. S. Dale, ibid., 1986, 41B, 237. 19. H. Iwasaki and H. Haraguchi. Anal. Chim. Ada, 1988, 208, 163.
20. R. K. Winge, V. J. Peterson and V. A. Fassel, Appl. Spectrosc.,
1979, 33, 206.
21. P. W. J. M. Boumans, J. A. Tielrooy and F. J. M. J. Maessen, Spectrochim. Acta, 1988, 43B, 173. 22. A. N. Zaidel, V. K. Prokofev, S. M. Raiskii, V. A. Slavnyi and E. Y. Shreider, Tables of Spectral Lines, Plenum Press, New York, 1970. 23. H. L. Parson, A. Forster and D. Anderson, An Aflas of Spectral Interference in ICP Spectoscopy. Plenum Press, New York, 1980. 24. P. W. J. M. Boumans, Spectrochim. Acra, 1980,3!IB, 57.