Return to Document Menu
ICP OPTICAL EMISSION
application note
The Analysis of Fe/Nd Alloys
Michelle Uchiyama and Dennis Yates
PerkinElmer Instruments, 761 Main Avenue, Norwalk, CT 06859 USA
Introduction Experimental
A variety of iron/neodymium alloys Instrumentation
are used in the computer industry for All emission data were acquired with a PerkinElmer Optima 3000 ICP
recording data in high density on spectrometer. Instrumental operating conditions and parameters are listed in
magnetic media. The performance of Table 1.
the alloy in this application is critically
Table 1: Instrumental Operating Conditions and Parameters
dependent on the major and minor
elemental composition. Analysis of these ICP RF Power (watts) 1100 Viewing Height (mm) 15
materials is difficult since the wavelength Plasma Argon Flow (L/min) 15 Max. Integration Time (s) 200
richness of the emission spectra from Neb. Argon Flow (L/min) 1.00 Min. Integration Time (s) 40
the major components (Fe and Nd) Aux. Argon Flow (L/min) 1.0 Slit Type Normal
can interfere with most of the minor
components. The typical approach to
dealing with the analytical interfer-
Analytical wavelengths and background correction points (BCP) are included
ences is through matrix matching of
in Table 2.
standards and the use of interelement
correction techniques. Commonly, a
single analytical wavelength is utilized Table 2:Analytical Wavelengths (nm)
for each major and minor component. In Wavelength Lower Upper Wavelength Lower Upper
this study the PerkinElmer Optima Element (nm) BCP BCP (nm) BCP BCP
3000™ ICP-OES spectrometer was Al 167.02 0.000 0.015 309.27 0.019 0.076
employed. The instrumentation B 182.53 0.017 0.017
includes an echelle optical system with
Ca 315.88 0.019 0.019 393.36 0.024 0.072
a segmented-array charge-coupled
Ce 413.38 0.076 0.025 413.77 0.038 0.051
device (CCD) detector in the focal
plane. The detector is an integrating 418.66 0.025 0.013 456.23 0.028 0.028
device and permits measurements in Co 228.61 0.021 0.021 230.78 0.014 0.028
excess of 5000 emission lines covering Cu 324.75 0.069 0.020
the range of 167 nm to 782 nm. Thus Dy 353.17 0.043 0.033 394.47 0.132 0.048
selection of multiple wavelengths per
Fe 234.34 0.029 0.014 259.94 0.024 0.024
element was possible.
The Fe/Nd alloys were analyzed for 273.95 0.033 0.025
Al, B, Ca, Ce, Co, Cu, Dy, Fe, Gd, La, Gd 342.25 0.032 0.021
Mg, Mn, Nd, Ni, Pr, Sm, Si, V, and Y La 379.48 0.023 0.035 408.67 0.012 0.050
using 34 wavelengths. Multiple Mg 279.55 0.034 0.043
wavelengths were employed for the
Mn 257.61 0.047 0.039
determination of certain elements
Nd 401.22 0.025 0.025 406.11 0.025 0.037
when levels were low or questions of
spectral interferences arose. Lack of 430.36 0.026 0.040
agreement may indicate an unidenti- Ni 231.60 0.021 0.021
fied interference. Elemental concentra- Pr 390.84 0.000 0.036
tions in the prepared samples ranged Si 251.61 0.046 0.015
from 0.2 ppm (Ce) to 600 ppm (Fe).
V 270.09 0.033 0.033 292.40 0.027 0.036
We would like to thank Mr. Robert
Haire and the Howmet Corporation for Y 371.03 0.034 0.000
their kind assistance in providing
prepared samples and standards.
Calibration information is given in Table 3. Standards were used to Results and Discussion
bracket the analysis range. Data were acquired using peak height for Three Fe-Nd alloy standards
single pixel windows and peak area for double pixel windows. containing major and minor
component concentrations within
the calibration range were analyzed
using the Optima ICP-OES. Analyti-
cal results are compared with
"known" values for these solutions
Table 3: Calibration Data (%) obtained in a round robin study
conducted by the alloy manufactur-
Element Standard 1 Standard 2 Element Standard 1 Standard 2
er. Results with associated precision
Al 0.0000 0.6010 La 0.0000 0.2001 information are summarized in
B 0.0000 1.5840 Mg 0.000 0.2020 Table 4. Also included in Table 4 is
Ca 0.2009 0.0000 Mn 0.2019 0.0000 a summary of results and precision
Ce 0.0000 0.2001 Nd 36.0300 25.9240 obtained for standards 1 and 2 after
Co 5.6110 3.2900 Ni 0.0000 0.2024 a two-hour period when they are
Cu 0.4955 0.0000 Pr 0.0000 0.9857 read as samples. Data collection for
Dy 3.8320 0.0000 Si 0.1920 0.0000 three replicates requires approxi-
Fe 51.4680 63.6020 V 1.7360 3.0020 mately seven minutes and, as can
Gd 0.2015 0.0000 Y 0.0000 0.2070 be seen in Table 4, short-term
precision is excellent. Read-back of
the standards indicates drift over
the two-hour period to be minimal,
generally less than 1%.
Table 4: Summary of Results and Precision
Element Sample 1 Sample 2 Sample 3 STD 1 STD 2
Wavelength "True"Found (%RSD) "True" Found (%RSD) "True"Found (%RSD) "True" Found(%RSD) "True"Found (%RSD)
(nm)
Al 167.02 0.154 0.156 11.61 0.303 0.328 3.99 0.443 0.480 4.27 0.601 0.641 2.12 0.000 0.009 72.47
Al 309.27 0.154 0.167 2.48 0.303 0.306 0.00 0.443 0.444 0.74 0.601 0.605 0.58 0.000 0.006 74.55
B 182.52 0.207 0.221 1.50 0.417 0.433 0.54 0.829 0.848 0.89 1.584 1.599 0.36 0.000 0.011 30.01
Ca 315.88 0.096 0.100 0.88 0.047 0.051 0.88 0.017 0.020 0.45 0.000 0.000 117.05 0.201 0.202 0.63
Ca 393.36 0.096 0.097 0.39 0.047 0.049 0.07 0.017 0.019 0.24 0.000 0.000 25.03 0.201 0.202 0.29
Ce 413.38 0.019 0.024 1.55 0.483 0.054 0.00 0.101 0.106 1.37 0.200 0.201 0.79 0.000 0.002 39.38
Ce 413.77 0.019 0.020 6.25 0.483 0.050 2.68 0.101 0.102 0.64 0.200 0.202 0.52 0.000 0.001 58.20
Co 228.61 5.004 5.143 0.63 4.581 4.680 0.77 4.012 4.105 0.44 3.290 3.332 0.89 5.611 5.651 0.64
Co 230.78 5.004 5.153 0.67 4.581 4.684 0.81 4.012 4.110 0.41 3.290 3.332 0.83 5.611 5.653 0.70
Cu 324.75 0.248 0.243 0.60 0.125 0.124 0.71 0.060 0.060 0.39 0.000 0.000 91.35 0.495 0.497 0.74
Dy 353.17 2.480 2.606 0.39 1.220 1.292 0.10 0.618 0.612 0.45 0.000 0.002 106.43 3.832 3.858 0.75
Dy 394.47 2.480 2.518 0.40 1.220 1.249 0.17 0.618 0.631 0.27 0.000 0.003 104.97 3.832 3.809 0.92
Fe 234.34 55.361 57.034 0.27 59.382 60.324 0.08 61.909 62.439 0.16 63.602 63.490 0.20 51.468 51.736 0.62
Fe 259.94 55.361 55.282 0.37 59.382 59.591 0.09 61.909 62.381 0.20 63.602 63.648 0.24 51.468 51.673 0.21
Fe 273.95 55.361 56.531 0.31 59.382 60.092 0.09 61.909 62.406 0.16 63.602 63.522 0.21 51.468 51.750 0.62
Gd 342.24 0.101 0.107 0.66 0.050 0.054 0.58 0.020 0.022 0.62 0.000 0.000 60.23 0.201 0.203 0.58
La 379.47 0.020 0.016 2.17 0.049 0.042 0.65 0.104 0.097 0.52 0.200 0.201 0.37 0.000 0.000 nd
La 408.67 0.020 0.023 4.53 0.049 0.048 1.19 0.104 0.101 0.86 0.200 0.201 0.39 0.000 0.003 32.84
Mg 279.55 0.017 0.017 0.85 0.050 0.048 0.73 0.104 0.104 0.20 0.202 0.203 0.30 0.000 0.000 97.11
Mn 257.61 0.101 0.104 0.63 0.049 0.051 0.68 0.019 0.020 0.43 0.000 0.000 87.15 0.202 0.202 0.63
Nd 401.22 33.5 33.592 0.36 30.492 30.889 0.11 28.410 28.522 0.16 25.924 26.081 0.22 36.030 36.110 0.23
Nd 406.11 33.5 33.660 0.36 30.492 30.946 0.12 28.410 28.570 0.17 25.924 26.111 0.21 36.030 36.099 0.23
Ni 231.6 0.019 0.018 1.06 0.050 0.048 0.41 0.103 0.104 0.15 0.202 0.205 0.79 0.000 0.000 339.13
Pr 390.84 0.102 0.104 1.18 0.253 0.256 0.26 0.514 0.518 0.22 0.986 0.985 0.29 0.000 -0.001 51.68
Si 251.61 0.099 0.105 0.91 0.049 0.048 0.53 0.019 0.020 1.80 0.000 0.000 99.04 0.192 0.192 0.64
V 270.09 1.989 2.032 0.64 2.286 2.314 0.72 2.593 2.624 0.37 3.002 3.027 0.75 1.736 1.742 0.68
V 292.40 1.989 2.018 0.33 2.286 2.313 0.07 2.593 2.616 0.15 3.002 3.007 0.23 1.736 1.749 0.67
Multiple wavelengths were chosen
for each major component (> 2.0
wt%), while all minor components
were analyzed at the most sensitive
wavelength free of spectral interfer-
ences. Agreement between "true" and
found values was excellent for Al, B,
Ca, Co, Cu, Dy, Fe, Gd, Mg, Mn, Nd,
Ni, Pr, Si, V, and Y. In almost all cases,
differences were no more than 5% of
the known value. Additionally, the
ability to use more than one wave-
length for each element adds confi-
dence to the analytical results and
provides the flexibility to detect
interferences in unusual samples.
Analytical lines for La and Ce
exhibited spectral interferences from
both Fe and Nd. Examples of the
spectra are shown in Figures 1
and 2. Although four analytical
lines were used for cerium and two
lines for lanthanum, none provided
Figure 1. Scans in the 379.487-nm region – Lanthanum.
acceptable results, particularly at
lower levels.
Figure 2: Scans in the 418.660-nm region – Cerium.
To correct for these interferences, calculated. A summary of the data for values is excellent. Since processing
a multivariant calibration technique La nd Ce, as well as Sm, are shown in of the analytical data occurs as it is
was utilized. The multi-component Table 5. The top part of the table collected, this technique is not
spectral fitting (MSF) technique contains the MSF results. For subject to differences in plasma
requires collection of pertinent comparison, results acquired using conditions, which can exist when
spectral data for standards and conventional data processing are interelement correction factors are
samples as well as a spectral scan of listed in the bottom half of the table. measured, thereby affecting their
a pure solution of each interfering Comparing results in the sections of accuracy. Secondly, more than one
element at a concentration which the table directly gives an excellent interferent can be accurately
approximates (± 1 order of magni- indication of the improvement corrected for.
tude) that in the unknown. This possible using MSF. The improve-
spectral information is then utilized ment can be substantial.
to solve a series of simultaneous Using the MSF technique, agree-
equations and the analytical result ment between "expected" and found
Table 5: MSF Data
Sample 1 Sample 2 Sample 3
Wavelength Expected Found %RSD Expected Found %RSD Expected Found %RSD
(nm) (WT%) (WT%) (WT%)
Samarium
442.43 0.0999 0.1016 0.05 0.0497 0.0506 0.56 0.0193 0.0196 1.4
Lanthanum
379.47 0.0198 0.0192 1.71 0.0492 0.0493 0.77 0.1038 0.1046 0.45
Cerium
418.66 0.0187 0.200 1.93 0.0483 0.0501 0.46 0.1007 0.104 1.02
Results shown below were taken on the same samples using normal data processing:
Samarium
442.43 0.0999 0.1116 0.83 0.0497 0.0588 0.66 0.0193 0.0243 2.13
Lanthanum
379.47 0.0198 0.018 2.08 0.0492 0.0479 0.96 0.1038 0.1034 1.08
Cerium
418.66 0.0187 0.0332 1.96 0.0483 0.0716 0.72 0.1007 0.120 1.24
Neodymium and iron interference modeled except for samarium where only neodymium was modeled. NON-MSF data
acquired using peak height and manual background correction.
Visit our website at [Link]
PerkinElmer Instruments PerkinElmer, Optima and Optima 3000 are trademarks of Pe r k i n E l m e r, Inc.
Quattro Pro is a registered trademark of Corel Corporation.
761 Main Avenue
Norwalk, CT 06859-0010 USA
Tel: 800-762-4000 or
(+1) 203-762-4000 D-6249
Fax: (+1) 203-762-4228 © 1999 PerkinElmer, Inc. Printed in U.S.A. ICPA559
KG1 2 9 9 0 1
