- Email: [email protected]
Determination of Cr, Ni, Cu, Zn and Cd in niobium by radiochemical proton-activation analysis
Talantu. Vol. 29. pp. 285 to 290. 1982 Printed in Great Brilsin.
DETERMINATION OF Cr, Ni, Cu, Zn AND Cd IN NIOBIUM BY RADIOCHEMICAL PROTON-ACTIVATION ANALYSIS W. G. Sektion Analytik und Hiichstreinigung,
FAIX
and V. KRIVAN
Universitlt
Ulm. Oberer Eselsberg N26. D-7900 Ulm, F.R.G.
(Receioed 3 Seprember 1981. Accepted 26 September 1981)
Summary-A
rapid radiochemical proton-activation technique based on the utilization of short-lived indicator radionuclides (fij2 = ‘7- 100 min) for the determination of Cr, Ni, Cu. Zn, and Cd in niobium is described. It involves the irradiation of the samples with 13-MeV protons, the post-irradiation decontamination of the sample surface, a rapid sample dissolution, a separation procedure based on anionexchange from HF medium, and counting the eluate with a high-resolution gamma-spectrometer. In addition, for the determination of Ni. a specific separation procedure is proposed. For a 20-min irradiation with a beam intensity of 10 pA and a delay time of 20 min. the limits of detection are 4 t&g for Cr, 0.5 rig/g for Ni, 60 rig/g for Cu. 0.5 pg/g for Zn and 0.2 pg,/g for Cd. For Cr. Ni, and Cu. the results obtained by this technique are compared with data obtained by radiochemical neutron-activation analysis and atomic-absorption spectrometry.
Several powerful activation-analysis techniques have been developed in recent years for trace analytical
characterization of niobium, enabling determination of about 30 elements,’ but none of them is capable of detecting nickel in niobium of good or high-purity grade. Radiochemical neutron-activation analysis (NAA) involving irradiation with relatively high thermal neutron flux (8 x 1013 n.cm-2.sec-‘) does not allow detection of nickel contents below 0.1 &g,’ mainly because of the low isotopic abundance of the target nuclide 64Ni (0.95%) in the natural element. Limits of detection of Zpg/g have been reported for direct optical emission spectrography” and 0.3 pg/g for a spectrophotometric technique involving a specific separation of nickeL4 These limits of detection are not low enough for apphcation of the techniques to the determination of nickel in high-purity niobium. In our previous work,5 it was shown that protonactivation analysis by use of (p,n)-reactions and shortlived indicator radionuclides (tl,2 = 20-40 min) is a very sensitive technique for the determination of nickel, and in addition of chromium and copper. However, the analysis cannot be performed instrumentally, because of strong activation of the matrix through the reaction 93Nb(p,n)g3mNb (tllZ = 6.95 hr). radiochemical separation becomes Consequently, necessary if this sensitive determination technique is to be used. The purpose of this work was to develop a radiochemical proton-activation technique for the determination of Cr, Ni, Cu, Zn, and Cd in niobium, involving a rapid decomposition of the sample and a rapid removal of the radionuclides produced from the matrix. Although, in contrast to Ni, the elements Cr, Cu. Zn, and Cd can be determined very sensitively by radiochemical NAA,2.6.7 the present technique can be
of interest as a second complementary ing high sensitivity.
method offer-
EXPERIMENTAL
Chemicals and apparatus
All reagents used prior to irradiation were of “suprapur” grade. They were of “pro analysi” grade for post-irradiation procedures. The separations were carried out on Dowex 1X8 (200400 mesh) strongly basic anion-exchange resin in the F--form. The original concentrations of hydrofluoric acid and nitric acid were 40% and 650/ respectively. The separation procedure was developed by using the radioisotopes 54Mn, 6*Cu, 6sZn, “Ga, and 115m~“sCd as tracers. In the tracer experiments, a single-channel analyser with a well-type 3 x 3 in. NaRTI) detector was used for counting. For y-ray spectrometry, a GefLi) detector having an energy resolution of 1.9 keV FWHM for the 1.332-MeV y-ray of “Co and an efficiency of 200/” relative to a 3 x 3 in. NaI(Tl) detector and a peak-to-Compton ratio of 40: 1 was used. The detector was connected to a Canberra 8180 multichannel analyser. Samples and standards
Analyses were performed samples:
on the following
niobium
(a) Nb-P, Plansee, Reutte, Austria (b) Nb-Es, Heraeus, Hanau, F.R.G. (c) Nb-WCT, Teledyn Wah Chang, Albany, Oregon, U.S.A. (d) Nb-R-l, Max-Planck-Institut fur Metallforschung, Stuttgart, F.R.G. (e) Nb-R-2, Max-Planck-Institut fur Metallforschung, Stuttgart, F.R.G. From the niobium samples, targets with a thickness of 0.61 mm were cut with a diamond saw. In order to remove possible surface contamination, the samples were etched for 10 set in a 9: 1 v/v mixture of 40% hydrofluoric acid and 65% nitric acid. Thick metal targets of pure Cr, Ni, Cu. Zn, and Cd were used for standardization. A monitor foil of pure niobium was placed on the side of the target exposed to the irradiation. 285
W. G. FAIXand V. KRIVAN
286
After the bombardment, the monitor foils were counted with a y-ray spectrometer, and the relative radioactivity of 93mMo produced in the irradiation of samples and standards was taken as the basis for the quantitative evaluation. The results obtained were corrected for the difference in the depth of penetration of the proton? in the samples and standards.
generation of gas (0.2-0.5 ml). Then the solution was diluted with water Jo about 12 ml to give about 5M hydrofluoric acid concentration. The dissolution was carried out in a 20-ml polyethylene syringe which was connected by a polyethylene tube, through a tube pump, to the ionexchange column. After dilution, the pump was started and the sample solution was brought to the column at about 6-7 ml/min. Elution was done with 2M hydrofluoric acid. The first 25 ml of eluate were collected in a polyethylene bottle for y-spectrometric measurements. The polyethylene column was 10 cm in height and 0.6 cm in internal diameter and filled with the resin Dowex 1X8 (200-400 mesh), pretreated sequentially with 20ml of 10M hydrochloric acid, 20 ml of water and 40 ml of 2M hydrofluoric acid. On the basis of recent studies on the extractability of dithizonates from hydrofluoric acid media,” the indicator radionuclide for the determination of Ni, 6oCu, was directly separated from the hydrofluoric acid solution after the ion-exchange by extraction with two lo-ml portions of 5 x IOe3M dithizone in chloroform, and then the hydrofluoric acid solution was washed with 5 ml of chloroform. The organic phase (25 ml) was used for counting with the y-spectrometer. Standards were measured in the solid stated for 1-5 min and the count-rates were corrected for the geometry of the 25 ml of solution.
and post-irradialion etching Irradiations were performed in the cyclotron of the Nuclear Research Centre at Karlsruhe. Niobium samples were irradiated in water-cooled target-holders with 13-MeV protons at a beam current of 10 PA for 10-30 min. Standards were irradiated with a beam current of 1OOnA for 3Osec. In order to remove possible surface contamination after irradiation the samples were etched for 30 set in a 9: 1:5 v/v mixture of 40% hydrofluoric acid, 65% nitric acid and water, then for 10sec in 10M hydrochloric acid, and finally washed with water. Irradiation
Radiochemical
separation
procedure
(Fig.
I)
The systematic studies on the anion-exchange characteristics of the elements in hydrofluoric acid medium by Faris’ and in hydrofluoric acid and hydrofluoric/nitric acid medium by IIS” was the basis for the development of the necessary radiochemical separation procedure. After surface decontamination, the samples were dissolved in 3 ml of 7:3 v/v 40% hydrofluoric acid/6576 nitric acid mixture heated initially to 80”. The acid mixture contained 1OOpg of each of the inactive carriers of Mn, Cu. Zn. Ga, and In. If there was incomplete decomposition (large samples), further dropwise addition of nitric acid followed. After the dissolution, the unconsumed nitric acid was removed by dropwise addition of formic acid until there was no further
RESULTS AND DISCUSSION The method Table 1 reports the relevant reactions and data for the activation of the matrix element, niobium. After a 20-min irradiation with 13-MeV protons at a beam
Irradiated niobium E, = 13 MN;
Surtacr
Ip = lOpA; t,= IO--2Omin)
drcontaminatlan
AddAlan of 0.2-0.5 ml of HCOOH Dilution with \O to l2ml
Column: PE, h = IO cm, + = 0.6 Resin: DOWOX IX6 l200-400 mesh) Pretmatmrnt 20ml 2Oml 40ml
,.,,._...
+6ml/min)
..,...... .,,.....
IOM HCl Hz0 PMHF
Elution with 2M HF
I
2Sml
Eluato
I
High rrsolutiong - spectrometer counting
Fig. 1. Scheme of the post-irradiation
radiochemical procedure.
287
Cr, Ni, Cu, Zn and Cd in niobium
Table 1. Reactions induced in niobium by 13-MeV protons
Reaction
MeV
Q3Nb(p,n)93mMo
- 1.2
Q3Nb(p.pn)Q2mNb
-8.8
93Nb(p,an)89mZr Q3Nb(p,an)8QeZr
-5.6 -5.6
Protein energy at f~,,,+
Major y-rays.
Intensity,
MeV
%
6.95 hr
0.2632 0.6846 1.4772
61.2 91.0 99.4
12.5
10.16d 4.16 min 78.4 hr
0.9345 0.5878
99.0 89.5
23.0 -
0.5110 0.9092
47.0 99.87
Q-value, Cl12
current of 10 PA, the samples give a dose-rate of 5-10 rem/hr at a distance of 10 cm, which is caused mainly by 93mM~ produced from the matrix. Consequently, the samples must be processed behind appropriate lead shielding. Data on the analytical nuclear reactions and the properties of the ihdicator radionuclides produced are listed in Table 2 along with primary interference reactions.r2-I4 In the determination of Cr, Ni, and Cu, the possible interferences caused by (p,pn) and (p,an) reactions have been experimentally verified at 15 MeV,S and found negligible even when the content of the interfering elements is 10 times that of the element to be determined. In the determination of Zn, interference is possible from Ga and Ge, but through the reactions 96Ga(p,n)69Ge and 72Ge(p,n)72As it was proved that it could be neglected because of the extremely low content of Ga and Ge in niobium. Similarly, Sn and In can’interfere in the determination of Cd, but neutron-activation analysis failed to detect these elements in the niobium samples used and the NAA detection
MeV
-
limit is much lower’ than that at which these elements interfere in the cadmium determination. In trace analysis at very low levels, reliable surface decontamination can be an important factor in the accuracy. Therefore, in addition to the pre-irradiation etching and packing of the samples in clean aluminium foils for irradiation, a post-irradiation etching was performed in hydrofluoric/nitric acid mixture followed by washing in dilute hydrochloric acid. The weight loss in the post-irradiation etching is cu. 0.05% and is negligible. The separation procedure described allows decontamination factors > 10’ to be obtained for the nuclides 89m*gZr,92mNb and 93mM~ produced from the matrix. The chemical yields obtained for the complete post-irradiation procedure in ‘the presence of 1oOpg of carrier for each radionuclide and about 1OOmg of niobium are given in Table 3. They rep
IO0 t MN
Fig. 2. y-Ray spectrum of the eluate of a niobium sample (Nb-P) obtained by using the following experimental conditions: proton energy = 13 MeV, beam intensity = 10@, irradiation time = 20 min. cooling time = 18 min, counting time = 8.3 min.
0
-
60Ni(p,n)60Cu
Ni
69.09
18.6
12.4
12.8 24.0
12.3 7.6
68Zn(p,n)68Ga
“°Cd(p,n)“oIn
“‘Cd(p,n)“‘mln “*Cd(p,n)“‘In
‘13Cd(p,n)“3mIn 1’6Cd(p,n)“6”In
Zn
Cd
3.71
26.23
83.76
Isotopic abundance, %
cu
“Ni(p,‘#“Cu
52Cr(p,n)52mMn
Cr
Element
Principal reaction
-0.5 -1.3
-1.9 -3.4
-4.7
-3.7
-4.1
-4.7
-6.9
-5.5
MeV
Q-value,
0.5110 0.8260 1.3325 1.7920 0.5110
23.4
0.5110 1.0774 0.5110 0.6577 0.5363 0.5110 0.6064 0.6182 0.3917 0.4170 0.8188 1.0971 1.2934 2.1120
68.3 69.1
99.48 54.0
7.6 14.4
0.5110 0.66% 0.9619
38.4
9.76
0.5110 1.4343
MeV
Major y-ray,
21.3
rw min
121.0 97.9 87.0 43.0 1.2 5.3 64.1 30.0 17.0 53.0 80.0 16.0
176.0 3.2
185.0 8.5 6.7
186.0 19.2 87.3 44.9 196.0
193.0 100.0
%
Intensity,
1’7Sn(p,an)“3mIn “51n(n,y)“6mIn 1’gSn(p,an)“6mIn
“sSn(p,an)“‘mIn “31n(p,pn)“21n ’ ‘6Sn(o,an)’ “In ..
’ “Sn(p,an)’ “In
“Ga(p,pr$sGa 72Ge(p,an)68Ga
Wu(p,pn)Wu 66Zn(p,an)62Cu
64Zn(p,an)60Cu
56Fe(p,an)52mMn
Interfering reaction
Table 2. Data on production and properties of the indicator radionuclides and possible primary interference reactions
0.35 4.3 14.4 7.6 8.6
-4.3 -3.7
0.66
27.5
60.0
69.1 27.8 48.9
48.9
91.7
0, /”
Isotopic abundance.
-5.1 -9.4 -6.7
- 10.3 -8.7 -7.4
- 10.8 -9.3 -11.9
- 10.9
-13.1
MeV
Q-value,
3 P ?1 r. x g
289
Cr. Ni. Cu, Zn and Cd in niobium Table 4. Limits of detection*
Table 3. Recovery of the indicator radionuclides with regard to the decomposition and separation in the presence of 100 pg of carrier for each element and 100 mg of Nb
Element Yield, s:,
Element Mn cu Zn Ga In
97.1 99.1 96.8 98.3 97.8 98.1
f + + f + +
15 min (including
4 ndg 20 rig/g 0.5 ng/gt 60 rig/g 0.5 @/g 0.2 lQ3lg
Cr Ni
0.7 0.3* 1.4t 0.8 1.0 1.1
cu
* For group separation by ion-exchange. t For specific separation including both exchange and the extraction separation steps.
about
Limit of detection
5 1 min transport
the
ion-
time). The
specific separation of Cu requires an additional 10 min. In Table 4, the limits of detection are given along with the conditions under which they were obtained. The detection limit for Ni (0.5 rig/g)) was obtained by counting the 1.3325-MeV y-line of “OCu with a
* Experimental conditions assumed; proton energy = 13 MeV; beam intensity = 10 PA; irradiation time = 20 min; cooling time = 20min; counting time = 1Omin; sample processed = Nb-R-2. t After specific separation of Cu and using a scintillation counter; cooling time = 30 min.
3 x 3 in. NaI(TI) detector after specific separation. This limit of detection can be improved by a factor of about 2 if the OSll-MeV y-line of 6oCu is used for the counting. However, in counting the 0.511-MeV-line of 6oCu, it must be considered that the radionuclides 6’Cu (tr,a = 3.3 hr) and 62Cu (ttll = 9.76min) also contribute to formation of this peak and correct time
Table 5. Contents of Cr. Ni and Cu determined in niobium of different grades of purity Sample Nb-P Nb-ES Nb-WCT Nb-R-1 Nb-R-2
Cr 9.6 29 7 0.43
k 1.7 &g f 5 nglg + 2 nglg f 0.08 pg/g <4ng/g
Contents determined Ni
cu
53 * 4&g 0.11** 0.03 pg/g 0.11 + 0.04 pg/g 0.14 f 0.02 jig/g 25+7nglg
0.6 + 0.2 &g 0.38 f 0.04 )cg/g 73 f 21 rig/g 1.3 + 0.3 /tg/g < 60 rig/g
Table 6. Comparisons of results obtained by different determination techniques for Cr. Ni and Cu Sample
Element
Content determined This technique Other techniques
Nb-P
Cr
9.6 f 1.7j&g
Nb-ES
Ni Cr
53 *4&g 29+5ng/g
cu
Nb-WCT
Cu
0.38 f 0.04 &g
73 * 21 rig/g
10.9 * 66 f 23 f 35 f
0.9 /.lgjg’ 6 &g’ 6 rig/g’’
0.35 f 0.43 f 0.56 f 0.46 f 65 f
0.12 j@/g’
’ Radiochemical neutron-activation analysis6 b Radiochemical proton-activation analysis using radionuclidet5 ’ Radiochemical .proton-activation analysis using radionuclide” d Flameless atomic-absorption spectrometry16 ’ Radiochemical neutron-activated analysis using radionuclide2 f Radiochemical neutron-activation analysis using radionuclide*O
12ng/gb 0.08 cBl’Bd 0.14 pg/g’ 0.03 j&g’
12ng,/gr 60*6ng/g “2Mn as indicator 65Zn as indicator 64Cu as indicator’ 66Cu as indicator
290
W. G. FAIXand V. KRIVAN
normalization is possible only on the basis of the complex decay curve.
also thank Dr. R. Caletka and Dr. C. S. Sastri. This project was financially supported by Bundesministerium fir Forschung und Technologie, Bonn.
Analysis of niobium
The technique developed has been applied to the analysis of niobium samples of different grades of purity. The results are given in Table 5 as means of at least 3 determinations, and the corresponding average deviations. Zinc and cadmium could not be detected in any of the samples analysed, so only a limiting concentration can be given, for instance # 0.5 @g for Zn and $0.2 PgJg for Cd in niobium Nb-R-2. As an example, the y-ray spectrum of the separated fraction after ion-exchange of the niobium sample Nb-P is shown in Fig. 2. For the determination of Cr, Ni and Cu, the accuracy of the method was checked by comparing the results with those obtained by other analytical techniques, viz. radiochemical neutron-activation analysis and atomic-absorption spectrometry. As can be seen from Table 6, which summarizes the results for niobium samples of different grades of purity, on the whole a satisfactory degree of accuracy could be achieved. Acknowled~ements-Grateful acknowledgement is made to Kernforschungszentrum Karlsruhe for making available the irradiation facilities free of charge. The authors thank Dr. K. Schulze, Max-Planck-lnstitut ftir Metallforschung, Stuttgart, for supplying the precious high-purity niobium samples and for valuable discussions. For discussions they
REFERENCES I. V. Krivan, Pure Appl. Chem., in the press. 2. W. G. Faix, R. Caletka and V. Krivan. Z. Anul. Chem.. 1981,307.409. 3. L. S. Brooks, Spectrochim. Acta, 1965, 21, 1023. 4. 0. Grossman and H. Grosse-Ruyken. Z. Anal. Chem.. 1968, 233, 14. 5. V. Krivan, J. Radioanal. Chem., 1975, 26, 151. 6. W. G. Faix and V. Krivan. Z. Anal. Chem.. 1980. 302. 269. 7. R. Caletka, W. G. Faix and V. Krivan, J. Rudioanal. Chem., in the press. 8. C. F. Williamson, J.-P. Boujot and J. Picard, Tub/es of Range and Stopping Power of Chemical Elements of Charged Particiei of Energy 0.35 to 500 MeV, Rappoit CEA-R 3042. Saclav. 1966. 9. J. P. Faris, Anal. &em., 1964, 36, 1157. IO. W. G. Faix, R. Caletka and V. Krivan, ibid.. 1981. 53, 1719. Il. R. Caletka and V. Krivan, Z. Anal. Chem., in the press. 12. W. Seelmann-Eggebert, G. Pfennig and H. Miinzel, Charr of rhe Nuclides. 4th Ed., Gersbach Verlag, Miinchen, 1974. 13. K. A. Keller, H. Miinzel and H. Lange, Q-Vulues for Nuclear Reactions, Landolt-Bbrnstein. New Series, Group 1, Vol. 5. Springer, Berlin, 1973. 14. G. Er&mann and W. Soyka. The Gamma Rays of rhe Radionuclides. Verlag Chemie. Weinheim, 1979. 15. W. G. Faix, J. W. Mitchell and V. Krivan, J. Radioanal. Chem.. 1979. 53, 97. 16. W. G. Faix and V. Krivan. unpublished results.
DETERMINATION OF Cr, Ni, Cu, Zn AND Cd IN NIOBIUM BY RADIOCHEMICAL PROTON-ACTIVATION ANALYSIS W. G. Sektion Analytik und Hiichstreinigung,
FAIX
and V. KRIVAN
Universitlt
Ulm. Oberer Eselsberg N26. D-7900 Ulm, F.R.G.
(Receioed 3 Seprember 1981. Accepted 26 September 1981)
Summary-A
rapid radiochemical proton-activation technique based on the utilization of short-lived indicator radionuclides (fij2 = ‘7- 100 min) for the determination of Cr, Ni, Cu. Zn, and Cd in niobium is described. It involves the irradiation of the samples with 13-MeV protons, the post-irradiation decontamination of the sample surface, a rapid sample dissolution, a separation procedure based on anionexchange from HF medium, and counting the eluate with a high-resolution gamma-spectrometer. In addition, for the determination of Ni. a specific separation procedure is proposed. For a 20-min irradiation with a beam intensity of 10 pA and a delay time of 20 min. the limits of detection are 4 t&g for Cr, 0.5 rig/g for Ni, 60 rig/g for Cu. 0.5 pg/g for Zn and 0.2 pg,/g for Cd. For Cr. Ni, and Cu. the results obtained by this technique are compared with data obtained by radiochemical neutron-activation analysis and atomic-absorption spectrometry.
Several powerful activation-analysis techniques have been developed in recent years for trace analytical
characterization of niobium, enabling determination of about 30 elements,’ but none of them is capable of detecting nickel in niobium of good or high-purity grade. Radiochemical neutron-activation analysis (NAA) involving irradiation with relatively high thermal neutron flux (8 x 1013 n.cm-2.sec-‘) does not allow detection of nickel contents below 0.1 &g,’ mainly because of the low isotopic abundance of the target nuclide 64Ni (0.95%) in the natural element. Limits of detection of Zpg/g have been reported for direct optical emission spectrography” and 0.3 pg/g for a spectrophotometric technique involving a specific separation of nickeL4 These limits of detection are not low enough for apphcation of the techniques to the determination of nickel in high-purity niobium. In our previous work,5 it was shown that protonactivation analysis by use of (p,n)-reactions and shortlived indicator radionuclides (tl,2 = 20-40 min) is a very sensitive technique for the determination of nickel, and in addition of chromium and copper. However, the analysis cannot be performed instrumentally, because of strong activation of the matrix through the reaction 93Nb(p,n)g3mNb (tllZ = 6.95 hr). radiochemical separation becomes Consequently, necessary if this sensitive determination technique is to be used. The purpose of this work was to develop a radiochemical proton-activation technique for the determination of Cr, Ni, Cu, Zn, and Cd in niobium, involving a rapid decomposition of the sample and a rapid removal of the radionuclides produced from the matrix. Although, in contrast to Ni, the elements Cr, Cu. Zn, and Cd can be determined very sensitively by radiochemical NAA,2.6.7 the present technique can be
of interest as a second complementary ing high sensitivity.
method offer-
EXPERIMENTAL
Chemicals and apparatus
All reagents used prior to irradiation were of “suprapur” grade. They were of “pro analysi” grade for post-irradiation procedures. The separations were carried out on Dowex 1X8 (200400 mesh) strongly basic anion-exchange resin in the F--form. The original concentrations of hydrofluoric acid and nitric acid were 40% and 650/ respectively. The separation procedure was developed by using the radioisotopes 54Mn, 6*Cu, 6sZn, “Ga, and 115m~“sCd as tracers. In the tracer experiments, a single-channel analyser with a well-type 3 x 3 in. NaRTI) detector was used for counting. For y-ray spectrometry, a GefLi) detector having an energy resolution of 1.9 keV FWHM for the 1.332-MeV y-ray of “Co and an efficiency of 200/” relative to a 3 x 3 in. NaI(Tl) detector and a peak-to-Compton ratio of 40: 1 was used. The detector was connected to a Canberra 8180 multichannel analyser. Samples and standards
Analyses were performed samples:
on the following
niobium
(a) Nb-P, Plansee, Reutte, Austria (b) Nb-Es, Heraeus, Hanau, F.R.G. (c) Nb-WCT, Teledyn Wah Chang, Albany, Oregon, U.S.A. (d) Nb-R-l, Max-Planck-Institut fur Metallforschung, Stuttgart, F.R.G. (e) Nb-R-2, Max-Planck-Institut fur Metallforschung, Stuttgart, F.R.G. From the niobium samples, targets with a thickness of 0.61 mm were cut with a diamond saw. In order to remove possible surface contamination, the samples were etched for 10 set in a 9: 1 v/v mixture of 40% hydrofluoric acid and 65% nitric acid. Thick metal targets of pure Cr, Ni, Cu. Zn, and Cd were used for standardization. A monitor foil of pure niobium was placed on the side of the target exposed to the irradiation. 285
W. G. FAIXand V. KRIVAN
286
After the bombardment, the monitor foils were counted with a y-ray spectrometer, and the relative radioactivity of 93mMo produced in the irradiation of samples and standards was taken as the basis for the quantitative evaluation. The results obtained were corrected for the difference in the depth of penetration of the proton? in the samples and standards.
generation of gas (0.2-0.5 ml). Then the solution was diluted with water Jo about 12 ml to give about 5M hydrofluoric acid concentration. The dissolution was carried out in a 20-ml polyethylene syringe which was connected by a polyethylene tube, through a tube pump, to the ionexchange column. After dilution, the pump was started and the sample solution was brought to the column at about 6-7 ml/min. Elution was done with 2M hydrofluoric acid. The first 25 ml of eluate were collected in a polyethylene bottle for y-spectrometric measurements. The polyethylene column was 10 cm in height and 0.6 cm in internal diameter and filled with the resin Dowex 1X8 (200-400 mesh), pretreated sequentially with 20ml of 10M hydrochloric acid, 20 ml of water and 40 ml of 2M hydrofluoric acid. On the basis of recent studies on the extractability of dithizonates from hydrofluoric acid media,” the indicator radionuclide for the determination of Ni, 6oCu, was directly separated from the hydrofluoric acid solution after the ion-exchange by extraction with two lo-ml portions of 5 x IOe3M dithizone in chloroform, and then the hydrofluoric acid solution was washed with 5 ml of chloroform. The organic phase (25 ml) was used for counting with the y-spectrometer. Standards were measured in the solid stated for 1-5 min and the count-rates were corrected for the geometry of the 25 ml of solution.
and post-irradialion etching Irradiations were performed in the cyclotron of the Nuclear Research Centre at Karlsruhe. Niobium samples were irradiated in water-cooled target-holders with 13-MeV protons at a beam current of 10 PA for 10-30 min. Standards were irradiated with a beam current of 1OOnA for 3Osec. In order to remove possible surface contamination after irradiation the samples were etched for 30 set in a 9: 1:5 v/v mixture of 40% hydrofluoric acid, 65% nitric acid and water, then for 10sec in 10M hydrochloric acid, and finally washed with water. Irradiation
Radiochemical
separation
procedure
(Fig.
I)
The systematic studies on the anion-exchange characteristics of the elements in hydrofluoric acid medium by Faris’ and in hydrofluoric acid and hydrofluoric/nitric acid medium by IIS” was the basis for the development of the necessary radiochemical separation procedure. After surface decontamination, the samples were dissolved in 3 ml of 7:3 v/v 40% hydrofluoric acid/6576 nitric acid mixture heated initially to 80”. The acid mixture contained 1OOpg of each of the inactive carriers of Mn, Cu. Zn. Ga, and In. If there was incomplete decomposition (large samples), further dropwise addition of nitric acid followed. After the dissolution, the unconsumed nitric acid was removed by dropwise addition of formic acid until there was no further
RESULTS AND DISCUSSION The method Table 1 reports the relevant reactions and data for the activation of the matrix element, niobium. After a 20-min irradiation with 13-MeV protons at a beam
Irradiated niobium E, = 13 MN;
Surtacr
Ip = lOpA; t,= IO--2Omin)
drcontaminatlan
AddAlan of 0.2-0.5 ml of HCOOH Dilution with \O to l2ml
Column: PE, h = IO cm, + = 0.6 Resin: DOWOX IX6 l200-400 mesh) Pretmatmrnt 20ml 2Oml 40ml
,.,,._...
+6ml/min)
..,...... .,,.....
IOM HCl Hz0 PMHF
Elution with 2M HF
I
2Sml
Eluato
I
High rrsolutiong - spectrometer counting
Fig. 1. Scheme of the post-irradiation
radiochemical procedure.
287
Cr, Ni, Cu, Zn and Cd in niobium
Table 1. Reactions induced in niobium by 13-MeV protons
Reaction
MeV
Q3Nb(p,n)93mMo
- 1.2
Q3Nb(p.pn)Q2mNb
-8.8
93Nb(p,an)89mZr Q3Nb(p,an)8QeZr
-5.6 -5.6
Protein energy at f~,,,+
Major y-rays.
Intensity,
MeV
%
6.95 hr
0.2632 0.6846 1.4772
61.2 91.0 99.4
12.5
10.16d 4.16 min 78.4 hr
0.9345 0.5878
99.0 89.5
23.0 -
0.5110 0.9092
47.0 99.87
Q-value, Cl12
current of 10 PA, the samples give a dose-rate of 5-10 rem/hr at a distance of 10 cm, which is caused mainly by 93mM~ produced from the matrix. Consequently, the samples must be processed behind appropriate lead shielding. Data on the analytical nuclear reactions and the properties of the ihdicator radionuclides produced are listed in Table 2 along with primary interference reactions.r2-I4 In the determination of Cr, Ni, and Cu, the possible interferences caused by (p,pn) and (p,an) reactions have been experimentally verified at 15 MeV,S and found negligible even when the content of the interfering elements is 10 times that of the element to be determined. In the determination of Zn, interference is possible from Ga and Ge, but through the reactions 96Ga(p,n)69Ge and 72Ge(p,n)72As it was proved that it could be neglected because of the extremely low content of Ga and Ge in niobium. Similarly, Sn and In can’interfere in the determination of Cd, but neutron-activation analysis failed to detect these elements in the niobium samples used and the NAA detection
MeV
-
limit is much lower’ than that at which these elements interfere in the cadmium determination. In trace analysis at very low levels, reliable surface decontamination can be an important factor in the accuracy. Therefore, in addition to the pre-irradiation etching and packing of the samples in clean aluminium foils for irradiation, a post-irradiation etching was performed in hydrofluoric/nitric acid mixture followed by washing in dilute hydrochloric acid. The weight loss in the post-irradiation etching is cu. 0.05% and is negligible. The separation procedure described allows decontamination factors > 10’ to be obtained for the nuclides 89m*gZr,92mNb and 93mM~ produced from the matrix. The chemical yields obtained for the complete post-irradiation procedure in ‘the presence of 1oOpg of carrier for each radionuclide and about 1OOmg of niobium are given in Table 3. They rep
IO0 t MN
Fig. 2. y-Ray spectrum of the eluate of a niobium sample (Nb-P) obtained by using the following experimental conditions: proton energy = 13 MeV, beam intensity = 10@, irradiation time = 20 min. cooling time = 18 min, counting time = 8.3 min.
0
-
60Ni(p,n)60Cu
Ni
69.09
18.6
12.4
12.8 24.0
12.3 7.6
68Zn(p,n)68Ga
“°Cd(p,n)“oIn
“‘Cd(p,n)“‘mln “*Cd(p,n)“‘In
‘13Cd(p,n)“3mIn 1’6Cd(p,n)“6”In
Zn
Cd
3.71
26.23
83.76
Isotopic abundance, %
cu
“Ni(p,‘#“Cu
52Cr(p,n)52mMn
Cr
Element
Principal reaction
-0.5 -1.3
-1.9 -3.4
-4.7
-3.7
-4.1
-4.7
-6.9
-5.5
MeV
Q-value,
0.5110 0.8260 1.3325 1.7920 0.5110
23.4
0.5110 1.0774 0.5110 0.6577 0.5363 0.5110 0.6064 0.6182 0.3917 0.4170 0.8188 1.0971 1.2934 2.1120
68.3 69.1
99.48 54.0
7.6 14.4
0.5110 0.66% 0.9619
38.4
9.76
0.5110 1.4343
MeV
Major y-ray,
21.3
rw min
121.0 97.9 87.0 43.0 1.2 5.3 64.1 30.0 17.0 53.0 80.0 16.0
176.0 3.2
185.0 8.5 6.7
186.0 19.2 87.3 44.9 196.0
193.0 100.0
%
Intensity,
1’7Sn(p,an)“3mIn “51n(n,y)“6mIn 1’gSn(p,an)“6mIn
“sSn(p,an)“‘mIn “31n(p,pn)“21n ’ ‘6Sn(o,an)’ “In ..
’ “Sn(p,an)’ “In
“Ga(p,pr$sGa 72Ge(p,an)68Ga
Wu(p,pn)Wu 66Zn(p,an)62Cu
64Zn(p,an)60Cu
56Fe(p,an)52mMn
Interfering reaction
Table 2. Data on production and properties of the indicator radionuclides and possible primary interference reactions
0.35 4.3 14.4 7.6 8.6
-4.3 -3.7
0.66
27.5
60.0
69.1 27.8 48.9
48.9
91.7
0, /”
Isotopic abundance.
-5.1 -9.4 -6.7
- 10.3 -8.7 -7.4
- 10.8 -9.3 -11.9
- 10.9
-13.1
MeV
Q-value,
3 P ?1 r. x g
289
Cr. Ni. Cu, Zn and Cd in niobium Table 4. Limits of detection*
Table 3. Recovery of the indicator radionuclides with regard to the decomposition and separation in the presence of 100 pg of carrier for each element and 100 mg of Nb
Element Yield, s:,
Element Mn cu Zn Ga In
97.1 99.1 96.8 98.3 97.8 98.1
f + + f + +
15 min (including
4 ndg 20 rig/g 0.5 ng/gt 60 rig/g 0.5 @/g 0.2 lQ3lg
Cr Ni
0.7 0.3* 1.4t 0.8 1.0 1.1
cu
* For group separation by ion-exchange. t For specific separation including both exchange and the extraction separation steps.
about
Limit of detection
5 1 min transport
the
ion-
time). The
specific separation of Cu requires an additional 10 min. In Table 4, the limits of detection are given along with the conditions under which they were obtained. The detection limit for Ni (0.5 rig/g)) was obtained by counting the 1.3325-MeV y-line of “OCu with a
* Experimental conditions assumed; proton energy = 13 MeV; beam intensity = 10 PA; irradiation time = 20 min; cooling time = 20min; counting time = 1Omin; sample processed = Nb-R-2. t After specific separation of Cu and using a scintillation counter; cooling time = 30 min.
3 x 3 in. NaI(TI) detector after specific separation. This limit of detection can be improved by a factor of about 2 if the OSll-MeV y-line of 6oCu is used for the counting. However, in counting the 0.511-MeV-line of 6oCu, it must be considered that the radionuclides 6’Cu (tr,a = 3.3 hr) and 62Cu (ttll = 9.76min) also contribute to formation of this peak and correct time
Table 5. Contents of Cr. Ni and Cu determined in niobium of different grades of purity Sample Nb-P Nb-ES Nb-WCT Nb-R-1 Nb-R-2
Cr 9.6 29 7 0.43
k 1.7 &g f 5 nglg + 2 nglg f 0.08 pg/g <4ng/g
Contents determined Ni
cu
53 * 4&g 0.11** 0.03 pg/g 0.11 + 0.04 pg/g 0.14 f 0.02 jig/g 25+7nglg
0.6 + 0.2 &g 0.38 f 0.04 )cg/g 73 f 21 rig/g 1.3 + 0.3 /tg/g < 60 rig/g
Table 6. Comparisons of results obtained by different determination techniques for Cr. Ni and Cu Sample
Element
Content determined This technique Other techniques
Nb-P
Cr
9.6 f 1.7j&g
Nb-ES
Ni Cr
53 *4&g 29+5ng/g
cu
Nb-WCT
Cu
0.38 f 0.04 &g
73 * 21 rig/g
10.9 * 66 f 23 f 35 f
0.9 /.lgjg’ 6 &g’ 6 rig/g’’
0.35 f 0.43 f 0.56 f 0.46 f 65 f
0.12 j@/g’
’ Radiochemical neutron-activation analysis6 b Radiochemical proton-activation analysis using radionuclidet5 ’ Radiochemical .proton-activation analysis using radionuclide” d Flameless atomic-absorption spectrometry16 ’ Radiochemical neutron-activated analysis using radionuclide2 f Radiochemical neutron-activation analysis using radionuclide*O
12ng/gb 0.08 cBl’Bd 0.14 pg/g’ 0.03 j&g’
12ng,/gr 60*6ng/g “2Mn as indicator 65Zn as indicator 64Cu as indicator’ 66Cu as indicator
290
W. G. FAIXand V. KRIVAN
normalization is possible only on the basis of the complex decay curve.
also thank Dr. R. Caletka and Dr. C. S. Sastri. This project was financially supported by Bundesministerium fir Forschung und Technologie, Bonn.
Analysis of niobium
The technique developed has been applied to the analysis of niobium samples of different grades of purity. The results are given in Table 5 as means of at least 3 determinations, and the corresponding average deviations. Zinc and cadmium could not be detected in any of the samples analysed, so only a limiting concentration can be given, for instance # 0.5 @g for Zn and $0.2 PgJg for Cd in niobium Nb-R-2. As an example, the y-ray spectrum of the separated fraction after ion-exchange of the niobium sample Nb-P is shown in Fig. 2. For the determination of Cr, Ni and Cu, the accuracy of the method was checked by comparing the results with those obtained by other analytical techniques, viz. radiochemical neutron-activation analysis and atomic-absorption spectrometry. As can be seen from Table 6, which summarizes the results for niobium samples of different grades of purity, on the whole a satisfactory degree of accuracy could be achieved. Acknowled~ements-Grateful acknowledgement is made to Kernforschungszentrum Karlsruhe for making available the irradiation facilities free of charge. The authors thank Dr. K. Schulze, Max-Planck-lnstitut ftir Metallforschung, Stuttgart, for supplying the precious high-purity niobium samples and for valuable discussions. For discussions they
REFERENCES I. V. Krivan, Pure Appl. Chem., in the press. 2. W. G. Faix, R. Caletka and V. Krivan. Z. Anul. Chem.. 1981,307.409. 3. L. S. Brooks, Spectrochim. Acta, 1965, 21, 1023. 4. 0. Grossman and H. Grosse-Ruyken. Z. Anal. Chem.. 1968, 233, 14. 5. V. Krivan, J. Radioanal. Chem., 1975, 26, 151. 6. W. G. Faix and V. Krivan. Z. Anal. Chem.. 1980. 302. 269. 7. R. Caletka, W. G. Faix and V. Krivan, J. Rudioanal. Chem., in the press. 8. C. F. Williamson, J.-P. Boujot and J. Picard, Tub/es of Range and Stopping Power of Chemical Elements of Charged Particiei of Energy 0.35 to 500 MeV, Rappoit CEA-R 3042. Saclav. 1966. 9. J. P. Faris, Anal. &em., 1964, 36, 1157. IO. W. G. Faix, R. Caletka and V. Krivan, ibid.. 1981. 53, 1719. Il. R. Caletka and V. Krivan, Z. Anal. Chem., in the press. 12. W. Seelmann-Eggebert, G. Pfennig and H. Miinzel, Charr of rhe Nuclides. 4th Ed., Gersbach Verlag, Miinchen, 1974. 13. K. A. Keller, H. Miinzel and H. Lange, Q-Vulues for Nuclear Reactions, Landolt-Bbrnstein. New Series, Group 1, Vol. 5. Springer, Berlin, 1973. 14. G. Er&mann and W. Soyka. The Gamma Rays of rhe Radionuclides. Verlag Chemie. Weinheim, 1979. 15. W. G. Faix, J. W. Mitchell and V. Krivan, J. Radioanal. Chem.. 1979. 53, 97. 16. W. G. Faix and V. Krivan. unpublished results.