JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 245 Analysis of Some Low Silicon Content Alloys by Inductively Coupled Plasma Atomic Emission Spectrometry* I. HlavaCek and 1. HlavaCkova Analytika Company Limited U Elektty 650 194 05 Prague 9 Czech Republic Inductively coupled plasma atomic emission spectrometry procedures have been developed for the analysis of some low silicon content alloys such as low-alloy steels nickel metal FeMo high-alloy steel cobalt alloy FeNb and FeV. In most cases the special emphasis has been placed on the determination of silicon together with the other analyte elements. Keywords lnductively coupled plasma atomic emission spectrometry; multi-element analysis; silicon; alloys; sample decomposition; phosphoric acid For the determination of low silicon contents of up to about 0.3% Si (sample mass of 0.500 g) it is possible to replace molecular absorption spectrometry (MAS) originally used i.e.UV/VIS spectroscopy with inductively coupled plasma atomic emission spectrometry (ICP-AES) or flame atomic absorption spectrometry (FAAS) for some metallic materials. The sample decomposition procedure (sample dissolution with hydro- chloric acid nitric acid or their mixture) is sufficient for the dissolution of silicon and some other elements. The dissolution conditions i.e. heating at a temperature of about 60°C does not result in the exclusion of silicon from the sample solution. In this way it is possible to dissolve metallic samples with matrices containing largely iron nickel chromium or manga- nese. The silicon determination in FeMo however results in incorrect values of silicon content because molybdenum sili- cides such as MoSi MoSi and possibly Mo,Si are not decomposed with hydrochloric and nitric acids.It was found that sufficient decomposition of molybdenum silicides is ensured by dissolving them with a mixture of sulfuric phos- phoric and if necessary nitric acids. The sample solution must be evaporated until white fumes of sulfur trioxide appear. Therefore ferroalloys containing low silicon content ( FeV and FeNb) Co alloys (Real) and steel Poldi AKRB samples were also decomposed with a mixture of sulfuric and phosphoric acids in a study of the hardly soluble components. Experimental Instrumentation An ARL 33000 LA sequential atomic emission spectrometer with an inductively coupled argon plasma was used for the determination of analyte elements under compromise instru- mental conditions i.e.at an observation height of 18 mm and an argon carrier gas flow rate of about 1.2 1 min-l. A Perkin- Elmer 503 flame atomic absorption spectrometer was used for comparative analysis by FAAS. Sample Preparation The average chemical composition of some alloys containing low and medium silicon contents is presented in Table 1. Table 2 summarizes the sample dissolution procedures. Essentially the alloy samples in chip (alloys and steels) or the fine powder form (ferroalloys especially the FeNb particle sizes should be less than 0.05mm without separating the various size fractions) are dissolved with phosphoric and sulfuric acids (except for low-alloy steels and nickel metal) in polytetrafluoroethylene (PTFE) vessels at a temperature of between 150 and 200°C without the addition of hydrofluoric acid.The decomposition time has to be prolonged for about 10-15 min after the sample dissolution. In some instances it is appropriate to start the sample decomposition with hydro- chloric and nitric acids. The low-alloy steel and nickel metal samples are dissolved only with hydrochloric and nitric acids in glass beakers. In this way the alloy samples are quantitatively converted into a soluble form. If the FeNb samples are analysed the sample solutions have to be stabilized with the addition of oxalic acid. In the analysis of FeV it was found that the aluminium content values obtained are lower.Therefore the residual aluminium content was determined by FAAS after fusing insoluble residues with potassium disulfate see Table2. The sample decomposition of some alloys was also performed in a microwave oven using practically the same procedures as in PTFE beakers under atmospheric pressure. The decomposition time was reduced from 2-3 h to about 10-15 min for all the samples analysed. Blank and synthetic samples were used for calibration and were prepared by the same procedures as the real alloy samples. The matrix element composition was simulated using high- purity metals. The synthetic calibration samples also contained known additions of the analyte elements in the concentration range to be considered. For practical purposes the matrix composition of the calibration samples can also be simulated using suitable certified or internal reference materials with known additions of some analyte elements.Table 1 Chemical composition of some low and medium silicon content alloys (YO) Alloy Low-alloy steel Ni metal FeMo High-alloy steel AKRB Cobalt alloy Real FeNb FeV t Some cobalt alloy Real types also contain Mo and Ni. * Presented at the XXVIII Colloquium Spectroscopium Internationale (CSI) York UK June 29-July 4 1993.246 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 Table 2 Sample preparation Sample Evaporation Final Alloy mass/g Dissolution to white fames volume/ml Low-alloy 1 .ooo 10 ml HN03 conc. - 100 1 .ooo 10 ml HNO conc. - 100 steel Ni metal FeMo 0.500 10 ml HNO conc. 10 ml H2S04 (1 + 1) 250 High-all0 y 0.500 10ml HN03 ( l + l ) 10 ml H,SO (1 + 1) 250 Co-alloy 0.500 20 ml HCl (1 + 1) 10 ml HNO conc. 250 10 ml HCl (1 + 1) 10 ml HCl (1 + 1) 10mlHC1(1+1) 25 ml H,PO conc.steel AKRB 10ml HC1 (1+1) 25 ml H,PO conc. Real 10 ml HzS04 (1 + 1) 25 ml H3P0 conc. 25 ml H3P0 conc. 25 ml H3P04 conc. 2 g NH,OII*HCl FeNb* 0.200 - 10 ml H,SO (1 + 1) 250* FeV7 0.500 10 ml HC1 conc. 10 ml H2S0 (1 + 1) 2507 5ml HNO (l-kl) * FeNb sample solution is stabilized with addition of 2.5 g of oxalic acid (COOH),*2H20. -f Aluminium contained in FeV sample was not quantitatively converted into a solution by the described procedure. Insoluble residues were analysed after filtration of a sample solution washed with warm HC1 (1 -t 9) and water ashed and fused with 2 g of potassium disulfate in a quartz crucible.After dissolution of the melt in water and dilution to 100 ml with water in a calibrated flask the determination of aluminium was performed by flame atomic absorption spectrometry using the spectral line 309.3 nm in the fuel rich N,0-C2H2 flame. The total aluminium content in FeV is given by the sum of aluminium contents found by ICP-AES and FAAS. Results and Discussion The analyte elements were determined using the spectral lines recommended by the manufacturer see Table 3 except for cobalt tantalum titanium and tungsten. The original Co I1 238.892 nm line was used only for the analysis of nickel metal where a low iron content was found because the line was severely affected by the nearby Fe I1 238.863 nm line. In the Table 3 Analyte elements and analytical spectral lines for ICP-AES Element and line A1 I B I Cd I1 c o I1 Cr I c u I Fe I1 Mn I1 Mo I Ni I1 Si I Ta Ti I v I1 W I Zn I1 Wavelength/nm 394.401 249.678 226.502 238.892 360.533 324.754 259.940 257.610 317.035 23 1.604 251.611 265.327 363.546 311.071 400.875 202.548 Secondary slit- widt h/pm 75 50 75 75 75 75 75 75 75 75 75 75 75 75 50 75 Table 4 Interferences from added elements at a concentration of 1 % for different Ni lines (nm) Interfering element A1 c o Cr c u Fe Mn Mo Si Ti v W Background equivalent concentration (YO Ni) Ni TI 221.647 nm Ni I 232.003 nm 0.0018 0.005 < 0.005 0.0004 0.047 < 0.0005 - - < 0.0005 0.00007 0.0009 o.Ooo1 0.00016 0.001 1 0.0002 0.0004 0.007 0.0002 0.044 o.Ooo1 0.0002 - - 0.0002 0.0085 0.0013 0.0004 Ni I1 231.604 nm < 0.00005 - - - 0.00005 - Table 5 Interferences from added elements at a concentration of 1% for different tungsten lines (nm) Background equivalent concentration (YO W) Interfering element A1 c o Cr cu Fe Mn Mo Ni Ti v w I1 207.91 1 0.0017 0.00018 0.00027 0.0012 0.00012 0.00014 0.00075 0.00012 0.0003 0.00 12 w I1 w I1 w I1 W I 224.875 218.936 239.709 400.875 - < 0.0002 - 0.0002 - 0.0001 - < 0.001 - - - - - - - - 0.0037 0.0012 0.0042 0.00025 - - - 0.0003 - 0.001 - 0.00005 - - - - - 0.09 - - - - - 0.0004 Table 6 Interferences from added elements at a concentration of YO Interfering element c o Cr c u Fe Mo Nb Ni Ti V W Background equivalent concentration (%) < 0.005 Ni co 0.0001 3 0.008 Zn 0.01 Zn 0.0005 Cd 0.03 c o 0.005 Cr < 0.0005 Si 0.00 15 Ta 0.0004 Zn 0.0004 c o 0.0075 Si < O.oO15 V 0.0007 A1 0.0055 Cu 0.0005 Si 0.0005 Ta 0.007 Ti 0.001 A1 0.0005 co 0.001 Ti 0.01 1 v 0.0003 Cr 0.00014 co 0.0016 Si Note Co I 350.228 nm Co I1 238.892 nm Ta I1 240.063 nm Co I 350.228 nm Ta I1 240.063 nm Ti I 363.546 nm Co I 350.228 nm Ti I1 323.452 nm Co I 350.228 nmJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL.9 247 other instances the determination of cobalt was performed using the Co I 350.228 nm line at an observation height of 26 mm. Similarly the determination of tantalum in FeNb was performed using the Ta I1 240.063 nm line at an observation height of 18 mm instead of the Ta 265.327 nm line of low intensity which is severely influenced by the presence of chromium. For the determination of titanium in steel Poldi AKRB the Ti I1 323.452 nm line at an observation height of 18 mm was selected instead of the Ti I 363.546 nm line which is affected by the presence of molybdenum.For the determi- nation of nickel and tungsten in the cobalt alloy Real more suitable spectral lines were investigated. Besides the original Ni I1 231.604 nm line the Ni I1 221.647 and Ni I 232.003 nm lines were examined. For comparison the spectral interferences found at the optimal observation height of 18 mm are presented in Table 4. Finally the determination of nickel was performed using the Ni I1 231.604nm line because it was the most significant spectral line. The determination of tungsten was originally performed using the W I 400.875 nm line of very low intensity.Therefore the W I1 207.911 W I1 224.875 W I1 218.936 and W I1 239.709 nm spectral lines were studied. The Table 7 Comparison of results for silicon content (YO) in real and certified reference materials (CRMs) low-alloy steels Sample 871 -CRM BCS-CRM 149 BCS-CRM 270 R 42616 R 42716 R 42816 R 42916 R 43016 R 43116 Certificate value MAS* 0.0095 0.013 0.0025 < 0.002 0.050S 0.05 5 0.052 - 0.043 0.053 - 0.057 - 0.045 - 0.046 - - ICP-AES 0.0105 0.001 5 0.056 0.056 0.051 0.057 0.064 0.05 1 0.053 * MAS Molecular absorption spectrometry. t Correlation coefficient for determination of silicon (MAS and ICP- AES) was found to be 0.993. Table 8 Comparison of results (%) for CRMs of nickel metal W I1 207.911 nm line appeared to be the most advantageous spectral line compared with the original line at an observation height of 18mm.The most important interferences from accompanying elements found for tungsten are shown in Table 5. The spectral interferences of the matrix and the other accompanying elements of the spectral line intensities of the analyte elements are reported in Table 6. The contingent interferences were corrected for the increase or decrease in the spectral background due to differences in the sample matrix composition if necessary. The proposed procedures were verified by means of Czechoslovakian British and German certified reference materials. For comparison and verification of accuracy the samples were also analysed by other analytical methods such as gravimetry titrimetry MAS and FAAS. The following tables present results obtained for some metallic materials analysed by ICP-AES.Comparison of the ICP-AES and MAS results for the low silicon contents in low-alloy steels is given in Table 7. Table 8 presents the analytical results of ICP-AES and FAAS and certified values for nickel metal. Similarly the analytical results obtained by ICP-AES FAAS and other analytical methods are shown in Tables 9-13 for ferromolyb- denum high-alloy steel Poldi AKRB cobalt alloy Real FeNb and FeV respectively. The sample decomposition of FeNb with sulfuric and phos- phoric acids was also used for the determination of nitrogen by titrimetry after distillation. The results for the nitrogen contents in FeNb are compared with those found by vacuum extraction (analyser Balzers EAN 202) are given in Table 14.The precision of the determination of analyte elements characterized by means of the basic statistical data is presented in Tables 15-17 for FeMo FeNb and FeV respectively. The limits of determination defined as ten times the standard deviation of the background noise were mostly within the range 0.002-0.02% for a sample mass of 0.500 g in 250 ml of solution i.e. within the range 40-400 ng ml-I for the analyte elements. The limit of determination achieved also depends on the sample matrix composition the other accompanying elements and their mutual spectral influence. Sample Method Cd c o Cr cu Fe Mn Si Zn &AN-CRM Certificate 0.0003 0.078 - 0.01 1 0.0 16 0.105 0.035 0.0057 5- 15-01 5 FAAS 0.0004 0.064 < 0.005 0.01 1 0.0 15 0.106 - 0.005 ICP-AES 0.001 0.064 < 0.005 0.012 0.014 0.109 0.035 0.006 &AN-CRM Certificate 0.002 0.21 0.079 0.101 0.0138 0.163 0.0 19 5-51-016 FAAS 0.0017 0.193 < 0.005 0.083 0.108 0.014 - - - ICP-AES 0.00 15 0.195 0.005 0.084 0.107 0.014 0.155 0.0195 Table 9 Comparison of results (YO) for real and CRM samples of ferromolybdenum Sample 02 03 41 42 43 90 & AN-CRM 4-4-02 21 1J16A Method FAAS others FAAS others FAAS others FAAS others FAAS others FAAS others Certificate FAAS Certificate ICP-AES ICP- AES ICP-AES ICP-AES ICP-AES ICP-AES ICP-AES ICP- AES A1 - < 0.02 < 0.02 < 0.02 < 0.02 < 0.02 < 0.02 - - - - - - - < 0.02 < 0.02 - Cr < 0.02 < 0.02 < 0.02 < 0.02 < 0.02 < 0.02 < 0.02 0.02 < 0.02 0.02 0.07 0.04 0.02 < 0.02 < 0.02 - - c u 0.24 0.26 0.49 0.53 0.18 0.185 0.15 0.15 0.20 0.205 0.36 0.37 0.52 0.495 0.50 0.192 0.02 Mn 0.01 < 0.01 0.01 <0.01 0.01 < 0.01 0.01 <0.01 0.01 < 0.01 0.08 0.08 < 0.01 < 0.01 < 0.01 - 0.004 Ni 0.04 0.05 0.04 0.05 0.07 0.06 0.05 0.05 0.05 0.06 0.05 0.05 0.045 0.04 0.02 - - Si 0.03* 0.25 0.04* 0.03 0.03* 0.03 0.02* 0.035 0.04* 0.03 0.60* 0.63 0.091 0.10 0.095 0.226 0.22 v < 0.01 < 0.01 < 0.01 < 0.01 < 0.01 <0.01 <0.01 < 0.01 < 0.01 < 0.01 < 0.01 < 0.01 0.01 - - * Gravimetry.248 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL.9 Table 10 Comparison of results (%) for real samples of high-alloy steel AKRB. Correlation coefficients for determination of manganese molybdenum silicon titanium and tungsten (chemical methods and ICP-ALES) were found to be 0.999 0.999 0.999 0.998 and 0.999 respectively c o Cr c u Mn Mo Si Ti v W Sample Method A1 B* B7 1608 P1 Chemical - - - - 12.691 - 0.271 0.5155 0.109 0.5259 - 1.9755 - - 0.52 0.003 - FAAS 2.01 - - 0.00 1 0.039 0.235 0.55 ICP-AES 2.00 0.065 0.063 0.002 12.53 0.037 0.245 0.52 0.08 0.535 0.002 1.955 1608 P2 Chemical - FAAS 2.35 ICP-AES 2.31 0.05 0.55 0.002 13.04 0.087 0.26 0.76 0.24 0.735 0.003 2.22 1608 P3 Chemical - - - - 13.741 - 0.441 1.0135 0.41 0.899 - 2.619 0.005 - ICP-AES 2.61 0.085 0.090 0.0025 13.66 0.037 0.42 1.04 0.385 0.90 0.007 2.62 1608 P4 Chemical - - - - 14.825 - 0.581 1.X$ 0.5% 1.0% - 2.999 - - 0.006 - FAAS ICP-AES 2.87 0.10 0.095 0.0025 14.47 0.038 0.56 1.295 0.46 1.09 0.005 3.03 1608 P5 Chemical - - 0.006 - FAAS ICP-AES 3.18 0.12 0.113 0.005 15.22 0.041 0.66 1.56 0.61 1.195 0.006 3.41 3102 X1 Chemical - FAAS - - - 13.295 - 0.301 0.755 0.269 0.735 - 2.21* - 0.090 0.255 0.79 0.72 0.004 - - 0.0015 - - - - - FAAS - - 0.0015 - 0.040 0.405 - - - 0.002 - 0.042 0.54 - - - - - 15.841 - 0.68j 1.561 0.64 1.174 - 3.329 - - - 14.401 - 0.311 1.19$ 0,199 0.8oEj - 2.879 - - I 0.002 - 0.043 0.64 - - - 0.031 0.28 - - - 0.003 0.002 - - - - ICP-AES 2.71 0.02 0.018 0.002 14.10 0.030 0.29 1.22 0.18 0.805 0.003 2.95 3102 X Chemical - - - - 15.121 - 0.325 122$ 0.229 0.995 - 2.8% - 0.004 - ICP-AES 3.16 0.09 0.095 0.003 14.76 0.030 0.29 1.20 0.21 1.04 0.004 2.88 - - - - 0.0015 - 0.032 0.28 - FAAS * Boron content determined directly without separation.7 Boron content determined after hydroxide separation. $ Titrimetry. 9 MAS. Table 11 Comparison of results (YO) for real samples of cobalt alloy Reail ~~ Method FAAS and others FAAS and others FAAS and others FAAS and others FAAS and others FAAS and others FAAS and others FAAS and others FAAS and others ICP-AESl I C P - A E S ICP-AES ICP-AES ICP-AES ICP-AES ICP-AES ICP-AES ICP-AES A1 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 c u 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 Fe 2.22* 2.13 3.04* 2.85 2.18* 2.00 2.22* 2.07 2.10* 1.98 0.33* 0.32 0.38* 0.37 0.15* 0.14 0.18* 0.18 Mn 0.207 0.175 0.20t 0.17 0.21t 0.175 0.197 0.15 0.207 0.16 0.497 0.50 0.507 0.50 0.527 0.54 0.50t 0.485 Mo Ni Si 1.939 1.925 1.94 1.85 1.924 1.925 1.899 1.87 1.9% 1.91 0.97 1.169 1.17 1.279 1.26 1.125 1.16 1.025 v 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 w 4.85* 4.90 4.72* 4.77 4.60* 4.71 4.79* 4.88 4.68* 4.81 4.35" 4.33 4.78* 4.65 4.35* 4.35 4.54* 4.63 Sample 1638 P 1640 P 1641 P 1642 P 1643 P - 0.015* 0.02 0.70* 0.70 0.025 0.74" 0.73 - 1734 P1 1734 P - 0.05 2.19 2.20 0.075 1.89 1.93 - 1735 P1 1735 P * MAS. j.Titrimetry. 8 Gravimetry. j Sample decomposition for ICP-AES was performed by means of a microwave oven. Table 12 Comparison of results (%) for real and CRM samples of ferroniobium Sample 5 S Method FAAS and others FAAS and others FAAS and others Certificate value ICP-AES ICP-AES ICP-AES A1 Cr 0.04 0.04 0.01 < 0.02 0.14 0.14 c u 0.02 0.02 0.02 0.03 0.0 1 < 0.02 Mn 1.71 1.67 5.46 5.20 0.27 0.34 Mo 0.03 0.03 0.02 < 0.02 0.02 0.02 Ni 0.06 0.04 0.02 < 0.02 0.02 < 0.02 Si Ta Ti 4.98 0.23 5.02 0.25 0.40 0.63 0.32 0.63 1.047 4.3 1.03 4.76 4.9 0.47 v 0.05 0.04 0.03 0.025 0.06 0.085 - W - 0.96 1.59 - - 0.285 0.97 10.77* 10.6 0.72 - - 0.24 < 0.02 NHKG 1 .oo 1.70 < 0.02 0.24 BCS-CRM 362 FAAS and others Certificate value Certificate value ICP-AES ICP-AES ICP-AES 0.195 0.195 0.01 < 0.02 1.88 1.81 0.01 < 0.02 0.01 < 0.02 - 0.045 0.055 - 0.73" 0.73 1.79 1.86 1.03 1.055 4.757 0.44 4.76 0.465 0.306 1.32 0.30 1.12 3.85 0.567 3.29 0.515 0.04 0.045 0.075 0.355 - - - 1.605 2.53 2.42 1.86 1.765 0.24 - CRM 576-1 0.20 0.19 - 0.055 0.055 - 0.275 1.45 - CRM 579-1 * Titrimetry (after separation and hydrolysis of K,SiF,).t MAS (with Malachite Green).249 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 Table 13 Comparison of results (YO) for real and CRM samples of FeV Sample 58 D 4087 D 4088 D 4089 D 4130 D 4131 D 4132 D 4133 BCS-CRM 20512 BAM-CRM 531-1 Method FAAS and others FAAS and others FAAS and others FAAS and others FAAS and others FAAS and others FAAS and others FAAS and others Certificate value FAAS Certificate value ICP-AES ICP-AES ICP-AES ICP-AES ICP-AES ICP-AES ICP-AES ICP-AES ICP-AES ICP-AES A1 total 1.80 1.77 1.76 1.70 1.88 1.85 1.66 1.58 1.75 1.74 1.80 1.70 1.10 0.97 1.60 1.69 2.0* 1.95 1.92 1.59 1.61 Cr 0.15 0.11 0.08 0.08 0.135 0.14 0.10 0.10 0.12 0.12 0.10 0.11 0.15 0.14 0.11 0.11 0.19 0.205 0.365 c u 0.01 0.01 0.025 0.025 0.025 0.025 0.025 0.025 0.03 0.03 0.05 0.05 0.04 0.04 0.12 0.12 0.247 0.258 0.019 0.023 Mn 0.18 0.17 0.18 0.175 0.17 0.17 0.16 0.16 0.16 0.15 0.16 0.16 0.20 0.19 0.17 0.17 0.25 0.255 0.15 0.14 Ni 0.06 0.05 0.065 0.075 0.10 0.105 0.07 0.075 0.05 0.06 0.05 0.05 0.04 0.05 0.11 0.1 1 < 0.02 0.045 - 0.02 Si 0.70t 0.66 - 0.71 - 0.76 - 0.73 0.72t 0.72 0.72t 0.74 0.83t 0.82 0.70T 0.62 1.02* 1.03 1.005 0.69 0.685 * Information value not certified. t Gravimetry.Table 14 Comparison of results for nitrogen content (%) in real samples of FeNb Vacuum extraction Sample (Balzers Ni-bath) Distillation* S 0.010 0.006 0.01 1 45 0.163 0.168 0.162 59 0.112 0.106 0.112 72 0.123 0.123 0.118 * Parallel determinations. Table 15 Precision of determination of analyte elements (%) in FeMo eSAN-CRM 4-4-02 samples Parameter A1 Cr c u Mn Ni Si v n* 10 8 15 13 13 16 8 Average < 0.02 < 0.02 0.50 < 0.01 0.04 0.095 < 0.01 SDt - - 0.012 - 0.004 0.012 - - - 2.4 - 10.0 12.6 - RSDS * n =Number of analyses.t SD = Standard deviation. $ RSD = Relative standard deviation. Table 16 Precision of determination of analyte elements (YO) in FeNb BCS-CRM 362 samples Parameter A1 Cr cu Mn Mo Ni Si Ta Ti v W n 5 5 5 4 4 4 6 6 6 4 4 Average 1.605 0.195 ~ 0 . 0 2 1.81 <0.02 <0.02 0.73 4.76 0.465 0.045 0.24 SD 0.026 0.021 RSD 1.6 10.8 - 4.4 - - 4.1 1.9 5.3 24.4 9.2 - 0.080 - - 0.030 0.089 0.025 0.011 0.022250 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 Table 17 Precision of determination of analyte elements (%) in real and CRM samples of FeV Sample Parameter A1 total Cr c u Ian Ni Si FeV 80 n Average SD RSD BAM-CRM n 531-1 Average SD RSD 33 2.715 0.069 2.5 1.61 0.038 2.4 25 33 0.19 0.012 6.3 0.365 0.013 3.6 24 34 0.0065 0.001 1 16.9 25 0.023 0.0024 10.4 3 4. 0.21 0.005 2.2 0.14 0.004 2.7 2 5 33 0.075 0.005 6.5 0.02 0.003 25 13.7 33 0.68 5 0.017 2.5 0.685 0.019 2.8 24 Conclusion It was found that the described ICP-AES procedures can be used successfully in metallurgical laboratories for the multi- element analysis of some low silicon content alloys. Silicon can be determined together with the other analyte elements without the application of hydrofluoric acid. In some instances the microwave oven was used to reduce the digestion time. The described ICP-AES procedures have been used for the multi-element analysis of high silicon content ferroalloys in everyday laboratory practice for the past years. Paper 3/03933A Received July 7 1993 Accepted October 12 1993