Determination of impurities in highly pure platinum by inductively coupled plasma-atomic emission spectrometry

Talanta ELSEVIER

Talanta 42 (1995) 1959-1963

Determination of impurities in highly pure platinum by inductively coupled plasma-atomic emission spectrometry Xinhua Zhang*, Huifen Li, Yufang Yang Institute of Precious Metals, Kunming, 650221, People's Republic of China

Received 7 November 1995; revised 9 June 1995; accepted 19 June 1995

Abstract

In this work, a cyclone spray chamber system is used in conjunction with an inductively coupled plasma-atomic emission spectrometer instead of the conventional Scott-type chamber system to reduce the lower limit of detection achieved by the instrument, and an internal standard element (Y) is introduced to eliminate the effects caused by the drift in the plasma background level. An ICP-AES method for the determination of 13 impurity elements in a highly pure platinum sample has been developed. In this method, it is not necessary either to add a platinum matrix to the calibration standard or to separate and concentrate the elements to be determined in the samples. The effect of the platinum matrix on the elements to be analyzed is corrected for by a background equivalent concentration subtraction method. The determination ranges of the method are as follows: 0.00010-0.0050% for Mg, Mn, Cu, Ag, Fe and Zn; 0.00030-0.015% for Au, Ir, Ni and Pb; 0.00050-0.025% for Rh and A1; and 0.00080-0.040% for Pd. The method is simple, rapid and accurate, and can be applied to the analysis of 99.9-99.995% pure platinum. Keywords: Cyclone spray chamber; ICP-AES; Impurity elements; Multielement analysis; Platinum

1. Introduction

In a previous paper [l], we reported the determination of 13 impurities in 99.99% pure platinum with the ICPQ-1015 direct spectrometer. When there is a demand for the analysis of samples of this metal having a higher purity, the instrument cannot meet the required lower limit of detection of impurity elements. As Browner and B o o m [2] pointed out, the sample introduction system does not always share the excellent properties of the optical and electronic components found in most modern atomic spectrometers. F o r this reason we intro* Corresponding author.

SSDI 0039-9140(95)01675-9

duced a cyclone spray chamber system (CSCS) [3] into the original instrument, and compared the analytical characteristics of the CSCS with those of the conventional Scott-type chamber system (STCS). Under the same operating conditions, the detection limits of the elements to be determined were reduced two- to threefold when the CSCS was used. This means that the lower limit of detection and the amount of sample used can be reduced by use of the CSCS technique. An internal standard element (Y) was introduced to eliminate the effects caused by the drift in the plasma background level [4]. Efforts were made to develop a method for the determination of 13 impurities in 99.9-99.995% pure platinum.

1960

X. Zhang et al. Talanta 42 (1995) 1959-1963 To ICP

Table 2 Wavelength of analytical lines Element Wavelength(nm) Element Wavelength(rim) Pd Rh Ir Au Ag Fe Ni

! Carrier g u

c

.

363.469 343.489 224.268 267.594 328.068 259.940 231.604

Cu AI Mg Mn Zn Pb Y

324.754 396.152 279.553 257.610 202.548 220.353 371.030

2.3. Preparation o f solutions waste tank

Fig. 1. Diagram of the CSCS. 2. Experimental

2.1. Apparatus The determinations were carried out with an ICPQ-1015 direct spectrometer, a CTM-50 monochromator, and an automatic background correction unit. A diagram of the CSCS is shown in Fig. 1. The working conditions given in Table 1 were used throughout the investigation.

2.2. Reagents Hydrochloric acid and nitric acid (suprapure grade), platinum metal (purity, 99.99938%), Y203 (purity, 99.99%), impurity elements (purity more than 99.9%), and doubly distilled water were used.

Calibration standard solutions According to the determination ranges, a set of at least three calibration standard solutions were prepared. These solutions contained 10% (v/v) hydrochloric acid, 2 lag m l - i o f the internal standard element (Y), and different concentrations o f the impurity elements to be determined, but they did not contain a platinum matrix. Matrix solution A matrix solution containing 20 mg ml-~ o f platinum was prepared; the solution also contained the same amount of hydrochloric acid and internal standard element as the calibration standard solutions. Sample solutions The platinum sample was dissolved in aqua regia and was converted to its chloro complex by repeated evaporation with hydrochloric acid. The residue was dissolved in 2 ml of 1:1 hydrochloric acid and the solution was diluted to 10 ml.

Table 1 Working conditions Component/parameter

Description/value

Torch Nebulizer Forward r.f. power (kW) Reflected power (W) Argon coolant flow rate (I rain-i) Argon plasma flow rate (I rain -I) Argon cartier gas flow rate (I rain-~) Height of observation above load coil (ram) Integration time (s)

A set of three concentric quartz tubes Concentric 1.0 <5 12 1.0 0.85 15 20

1961

X. Zhang et al. / Talanta 42 (1995) 1959-1963 Table 3 Comparison of the characteristics of the two systems Element

Pt Pd Rh Ir Ag Au Cu Ni Zn Pb Fe Y

STCS

CSCS

lb

la

BEC (~gml - I )

DL (~tg ml - I )

Ib

la

BEC (~gml -I )

DL (~gml - t )

0.97 3.22 4.78 4.72 1.49 0.69 1.14 2.23 1.36 5.41 0.90 0.26

5.68 15.20 25.44 74.80 66.04 16.54 56.34 51.78 96.73 27.97 72.21 61.33

2.01 2.70 2.39 0.63 0.23 0.43 0.22 0.45 0.14 2.40 0.13 0.043

0.051 0.064 0.056 0.019 0.0068 0.012 0.0061 0.013 0.0042 0.058 0.0037 0.0013

0.72 1.94 3.50 1.53 0.48 0.51 0.39 1.15 0.43 4.36 0.32 0.084

10.75 29.19 51.63 59.54 66.52 35.94 60.14 60.83 54.39 45.01 59.89 60.44

0.72 0.71 0.73 0.26 0.077 0.14 0.066 0.19 0.079 1.07 0.054 0.014

0.020 0.020 0.020 0.0077 0.0022 0.0042 0.0019 0.0057 0.0024 0.029 0.0016 0.00042

2.4. Analytical procedures The wavelengths of the analytical lines used are shown in Table 2. Firstly, the optical system of the instrument was calibrated by using a solution of any single element of a suitable concentration, and the gain (voltage) of the spectrometer photomultipliers were adjusted by using a standard solution of high concentration. Secondly, the spectral intensities of the analytes in the calibration standards were determined. Calibration curves were prepared from the spectral intensities of these analytes, and the concentrations of analytes in the sample were determined. Finally, the background equivalent concentration (BEC) at the wavelength of the analytical line for the analytes were determined with the platinum matrix solution. The above operations were controlled by a computer program, and printed results were provided automatically.

3. Results and discussion

3.1. Main characteristics of CSCS The analytical characteristics of the CSCS and the STCS were determined and compared by using a solution containing 19~tgml -I of each analyte and a blank solution, under the same conditions. The results are summarized in Table 3. In this table, I~ represents the intensity of the analytical line for analyte A in a solution with concentration Ca, Ib represents the intensity of the blank at the same wavelength as for

analyte A, BEC (equal to K/b), and DL (equal to O.03C,/Ia/Ib) is the detection limit of the analyte A. The ICPQ-1015 direct spectrometer is equipped with photomultipliers the gain (voltage) of which can be adjusted automatically. When the concentrations of the analyte and the operating conditions are kept constant, the gain with the CSCS is reduced by 80- 100 V relative to that with the conventional STCS. This means that the CSCS works better in the target ranges of the instrument than the STCS. In other words, there are wider dynamic ranges with the CSCS than with the STCS. More aerosols, and smaller and more evenly distributed droplets are introduced from the central tube at the bottom of the chamber into the torch after the solutions are aspirated into the chamber and moved along a spiral path, which provides more opportunities for collision with the wall of the chamber, as shown in Fig. 1. Thus, the noise which originates from the fluctuation of the aerosols is also reduced, and a more stable signal is obtained. From the results obtained in Table 3, it is obvious that, when the CSCS is used, Ia for a specified element is increased while the corresponding lb and BEC are decreased, and hence I Jib is also increased. The detection limits obtained with the CSCS are reduced two-to-threefold compared with those obtained with the STCS. In other words, the sensitivity obtained with the CSCS is increased two-to-threefold compared with those obtained with the STCS. This means that the lower limits of detection of the instrument and the amount of sample used can be reduced by

X. Zhang et al. / Talanta 42 (1995) 1959-1963

1962

Table 4 Influence o f platinum matrix on the intensity of analytical lines Element

Platinum concentration (mg m l - ~) 0

Mg Mn Cu Ag Fe Ni AI Zn Pd Rh lr Pb Au Au ~

10

15

20

I,,

Zno

/~n

lno

I.°

1.o

l,°

l.o

66.2 69.5 39.7 44.0 24.6 17.8 32.2 57.5 29.8 37.4 34.8 22.2 8.1 2.5

100 100 100 100 100 100 100 ! 00 100 100 100 100 100 100

68.2 72.0 40.6 44.6 28.6 19.2 31.7 72.2 30.1 37.9 40.2 24.3 71.5 2.6

103 104 102 101 116 108 99 126 101 101 116 110 879 104

69.1 73.2 41.0 45.0 30.7 20.0 31.8 79.4 30.3 38.1 43.0 25.3 101 2.7

104 105 103 102 125 112 99 138 102 102 124 114 1243 108

71.1 75.7 41.8 45.8 33.0 21 .I 32.2 88.6 30.9 38.8 46.4 26.8 _b 2.8

107 109 105 104 134 118 100 154 104 104 133 121 _ 112

len, Intensity m e a s u r e d ; / n o , intensity normalized to 100%. a Determined by use o f the CTM-50 monochromator. b The intensity measured is beyond the range of determination.

use of the CSCS technique. On the basis of the results obtained, a method has been developed for the determination of 13 impurities in 99.999.9955% pure platinum with the use of 20 mg m l - ~ sample solutions.

3.2. Effect of platinum matrix and correction of spectral interferences Spectral interference is one of the most important interferences in ICP-AES. The background caused by the intense emission of the platinum matrix is the main source of interference and must be corrected for before any element can be determined. The influence of the platinum matrix on the intensity of the analytical lines is shown in Table 4. The investigation was carried out with a solution containing 1 lag ml-~ of each of the analytes, and 0, 10, 15, or 2 0 m g m l -~ of platinum. The interference of the matrix platinum on the analyte determination was also observed by using the automatic background correction unit, with solutions containing 10 lag ml -~ of analyte and a solution containing 20 mg m l of platinum. From the experimental results, no spectral interferences are found among the platinum group elements. This is in agreement with the results obtained by Kibright and Tinsley

[5]. Because the analytical line from the gold channel, Au 242.795 nm, suffers serious interference due to the platinum spectral line (Pt 242.804 nm), gold was determined with use of another line, Au 267.595 nm, by using the CTM-50 Monchromator. The analytical line, Ir 224.268 nm, suffers a slight interference from the spectral line of copper (Cu 224.261 nm), but can be neglected in the present work. The analytical lines of Zn, Fe, Pb and Ni suffer slight interferences from the spectral wings of the platinum lines. Because the values of these spectral wings are small and constant at a fixed concentration of platinum, they are also treated as a background. Almost no spectral lines of other analytes are influenced by the platinum matrix. It was found previously [6] that the Pb 220.350 nm spectral line suffers interference from the Pd 220.349 nm spectral line when a large amount of palladium is present, but no such interference was found with a trace amount of palladium. The analytical line, Pd 363.470 nm, suffers interference from the Ar 363.446 nm spectral line, and thus the sensitivity of palladium is reduced. The effect of platinum on the analytes was corrected for by a BEC subtraction method. It can be formulated briefly as follows: Ctr = C a p - - C e q , where C,p, V~q, and Ctr are the apparent concentration, background equivalent concentration and true concentrations of the

X. Zhang et al./ Talanta 42 (1995) 1959-1963

1963

Table 5 Results of the synthetic samples CAL.-B

Element CAL.-A

C,da Ctr Recovery RSD Cap C~da Ctr Recovery RSD C,p Ceq (%) (lagml -I) (lag ml - j ) (lag ml -t) (%) (%) (lag ml -I) (lag ml -t) (lag ml -I) (lag ml -I) (%) Mg Mn Cu Ag Fe Zn Au Ir Ni Pb Rh AI Pd

0.034 0.014 0.029 0.034 0.514 0.513 0.539 0.659 0.188 0.702 0.177 0.104 0.0

0.148 0.114 0.127 0.128 0.611 0.611 0.924 1.502 0.575 1.088 0.774 0.715 0.947

0.10 0.10 0.10 0.10 0.10 0.10 0.40 0.40 0.40 0.40 0.60 0.60 1.0

0.11 0.10 0.10 0.09 0.10 0.10 0.38 0.39 0.39 0.39 0.60 0.61 0.95

110 100 100 90 100 100 95 92 98 98 100 102 95

0.2 1.6 0.2 0.2 0.9 0.6 2.1 1.2 0.7 1.4 0.7 3.1 1.9

0.436 0.433 0.419 0.414 0.909 0.899 2.093 2.218 1.737 2.240 2.594 2.531 3.909

0.40 0.40 0.40 0.40 0.40 0.40 1.60 1.60 1.60 1.60 2.40 2.40 4.00

0.40 0.42 0.39 0.38 0.40 0.39 i .55 1.56 1.55 1.54 2.42 2.43 3.91

100 105 98 95 100 98 97 98 97 96 96 101 98

0.8 0.6 0.7 0.5 2.3 1.8 1.4 4.4 3.0 3.0 2.9 0.6 5.2

a Cad, concentration added. analyte, respectively. Here, C ~ p r a c t i c a l l y includes the b a c k g r o u n d , the spectral wing a n d even the very w e a k spectral line o f the p l a t i n u m m a t r i x . T h e p u r i t y o f the p l a t i n u m m a t r i x , f r o m which the B E C values o f the a n a l y t e are determined, is directly related to the reliability o f the correction. T h e p l a t i n u m m a t r i x used in o u r research is a special p r o d u c t o f o u r Institute, the p u r i t y o f which is 99.99938%. T h e B E C o f the m a t r i x is a l m o s t c o n s t a n t ; it is n o t necessary to d e t e r m i n e the B E C value each time o f a n a l y sis with the s a m e b a t c h o f m a t r i x if the a n a l y t ical c o n d i t i o n s are c o n t r o l l e d strictly.

3.3. Determination ranges and results f o r synthetic samples T h e d e t e r m i n a t i o n ranges are as follows: 0 . 0 0 0 1 0 - 0 . 0 0 5 0 % for M g , M n , Cu, Ag, F e a n d Zn; 0 . 0 0 0 3 0 - 0 . 0 1 5 % for A u , Ir, N i a n d Pb; 0 . 0 0 0 5 0 - 0 . 0 2 5 % for R h a n d AI; a n d 0 . 0 0 0 8 0 0.040% for Pd. T h e results for two synthetic s a m p l e s are s u m m a r i z e d in T a b l e 5, in which the R S D

(relative s t a n d a r d d e v i a t i o n ) values for each s a m p l e were o b t a i n e d statistically f r o m 12 replicate d e t e r m i n a t i o n s in f o u r e x p e r i m e n t s d u r i n g a period of 2 months.

Acknowledgment T h e a u t h o r s t h a n k P r o f e s s o r Y i b i n Q u for her help in p r e p a r i n g this m a n u s c r i p t .

References [1] X. Zhang, H. Li and Y. Yan, Fenxi Huaxue (Analytical chemistry), 19(10) (1991) i199-1201. [2] R.F. Browner and A.W. Boom, Anal. Chem., 56 (1984) 786A. [3] Z. He, D. Wang and H. Ye, Guangpuxue Guangpufenxi (Spectrosc. Spectr. Anal.), 70) (1987) 43-49. [4] S.A. Myers and D.H. Tracy, Spectrochim. Acta, Part B, 38(9) (1983) 1227-1253. [5] G.F. Kilbright and H.M. Tinsley, Talanta, 26(1) (1979) 41-45. [6] X. Zhang, H. Li and Y. Yang, Fenxi Shiyanshi (Anal. Lab.), 10(3) (1991) 70-71.