JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 199 I VOL. 6 205 Application of a High Resolution Inductively Coupled Plasma Mass Spectrometer to the Measurement of Long-lived Radionuclides Chang-Kyu Kim,* Riki Seki and Shigemitsu Moritat Department of Chemistry University of Tsukuba Tsukuba 305 Japan Shin-ichi Yamasaki and Akito Tsumura Division of Water Quality Science National Institute of Agro- Environmental Sciences Tsukuba lbaraki 305 Japan Yuichi Takaku Marubun 8- 1 Odenma-cho Nihonbashi Chuo-Ku Tokyo 1 70 Japan Yasu hito IgarashiS Division of Radioecology National Institute of Radiological Sciences 3609 Isozaki Nakaminato lbaraki 3 7 7 - 72 Japan Masayoshi Yamamoto Low Level Radioactivity Laboratory Kanazawa University Tatsunokuchi lshikawa 923- 12 Japan Some long-lived radionuclides such as '"c 226Ra 232Th 237Np 238U 239Pu and 240Pu were measured using high resolution inductively coupled plasma mass spectrbmetry (HR-ICP-MS).By using HR-ICP-MS with an ultrasonic nebulizer the detection limits of these nuclides were 0.0024.02 pg ml-l and the sensitivities were 10 times better than those obtained using HR-ICP-MS without the ultrasonic nebulizer. More accurate isotopic data were also obtained using HR-ICP-MS than with quadrupole ICP-MS at lower concentrations of the analyte because of the improvement in counting statistics that can be obtained with HR-ICP-MS due to the greater efficiency of ion trans- mission. A comparison of the measurement of the 240Pu to 239Pu ratio is shown. Keywords High resolution inductively coupled plasma mass spectrometry; long-lived radionuclide detection limit It is important in the 'post-Chemobyl era' to measure long- lived radionuclides in our environment correctly and precisely.Information on the behaviour of these nuclides in the environ- ment is indispensable in estimating potential radiation effects on humans. Conventional radiometric methods are however time consuming and tedious owing to the low concentrations and low specific activity of the nuclides. For instance when measuring 237Np at the global fallout level in the soil by ct- spectrometry it takes a week or more to obtain several counts of ten resulting in extremely low sample throughput. This is typical of analyses of long-lived radionuclides in the general environment therefore more sensitive and accurate methodo- logy is needed.Inductively coupled plasma mass spectrometry with a quad- rupole mass analyser (ICP-MS) has been applied to the deter- mination of many long-lived radionuclides. These studies although successful possess some weaknesses. For example there are a number of background molecular species which exist in the mass range below 80 and thus analyte peaks in this area may overlap because the resolution of the convention- al ICP-MS device is limited to unit mass only. The detection limits are not sufficient in some instances. In order to measure long-lived radionuclides in environmental samples directly without chemical separation new analytical techniques involv- ing higher sensitivity and lower detection limits than can be achieved with ICP-MS are still required.Recently high resolution ICP-MS (HR-ICP-MS) using a double focusing mass analyser to enable mass measurement * Present address Korea Institute of Nuclear Safety P.O. Box 16. t Present address Power Reactor and Nuclear Fuel Development Cor- Daeduk-danji Daejon 305-353 Korea. poration Tokai lbaraki 3 19- 1 I Japan. To whom correspondence should be addressed. at high resolution whilst also achieving high sensitivity was developed for the determination of ultra-trace element^.^.' In the present paper an HR-ICP-MS instrument was used for the Table 1 Operating conditions for HR-ICP-MS ICP cotidtiom- R.f. powerfkW Coolant gas flow-rate/] min-' Auxiliary gas flow-rate/l min-l Solution upt ake-rate/ml m in-' Carrier gas flow-ratel1 min-' Without ultrasonic nebulizer With ultrasonic nebulizer Ultrzrsonic nebulizer- conditions- Solution uptake-rate/ml min-' Heater temperature/'C Condenser temperature/'C Load coil to aperturelmm Aperture di ame t er/mm High resolution magnet supply/A Accelerating voltagefkV Interface ~mditiom- Double focusing mass analyser.conditions- 1.2 0.5 1 . 1 0.5 2.0 2 120 1 7 I 2.4 4.0 13 Element Data acquisition MCA* Dwell time/ Channel for Number of mass range/u channels ms one peak sweeps search Tc 98.8-99.3 50 640 24 5 Ra 225.4-226.6 50 640 24 5 Th 231.4-232.6 50 640 24 3 U 237.4-238.8 50 640 24 3 Pu 238.4-240.6 SO 640 24 3 Np 236.4-237.6 50 640 24 5 * MCA multi-channel analyser.206 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1991 VOL. 6 Collector )Magnet( slit I Collector Fig.1 A schematic diagram of the HR-ICP-MS arrangement measurement of long-lived radionuclides as a preliminary ex- periment on the direct measurement of radionuclides in envi- ronmental samples. Experimental Instrumentation The HR-ICP-MS assembly (PlasmaTrace VG Elemental Winsford Cheshire UK) consisted of an ICP a sampling in- terface an electrostatic analyser (ESA) a magnetic sector and detectors. A schematic diagram is shown in Fig. 1. Ions pass through a narrow adjustable slit and are injected into the field of the ESA which is an energy focusing device. The ESA in combination with a magnetic sector focuses ions with different energies onto the same point on the collector. The spectrometer has the capability of adjustable resolution by changing the width of the source slit and the collector slit.An ultrasonic nebulizer (USN) (Applied Research Labora- tories Ecublens Switzerland) was used in order to increase the amount of sample reaching the plasma. This nebulizer was of a continuous feed type. The sample solution was delivered to the nebulizer by means of a Gilson Minipuls peristaltic pump (Gilson Villiers-Le-Bel France). Aerosols generated at the transducer surface were transported by Ar gas to the heating tube and the condenser for desolvation and then to the ICP. The operating conditions of the HR-ICP-MS instrument are given in Table 1. The measurements were performed not in the high-resolution mode but in the relatively low resolution mode (MIAM = 400; the slits were opened) in order to obtain a greater ion transmission.Spectra were obtained by scanning the ESA. Reagents The nitric and hydrochloric acids used for preparing the solu- tions were of super analytical reagent grade (Tama Pure AA- 100 Tama Chemical Kawasaki Japan). Standard solutions of T c (5.5 pg ml-I; 3.5 mBq ml-I) 226Ra (1.36 pg ml-I; 50.3 mBq ml-I) 232Th (1 .O pg ml-I; 4.0 nBq ml-I) 237Np (2.0 pg ml-I; 0.052 mBq ml-I) 238U (1.0 pg ml-I; 12.4 nBq ml-I) and 239Pu and 240Pu (total concentration 20.8 pg ml-I; 72.5 mBq ml-I) were prepared by successive dilution of the stock solution with 1 mol dm-3 HNO,. De-ionized water (Milli-Q Japan Milipore Kita-Shinagawa Tokyo Japan) with a resistivity >14 MR cm was used to prepare the solutions. Results and Discussion Mass Spectra Figs. 2 4 show the comparative mass spectra for 'Vc 226Ra 232Th ?37Np ?W 239Pu and 240Pu with both concentric nebuliz- ers and USNs.The spectrum for each nuclide was clear but the peaks did not have a typical flat-topped shape. With a sector type mass spectrometer working in the low-resolution mode a flat topped mass peak is usually obtained when a con- stant ion beam is achieved. It is difficult however to obtain a constant ion beam using ICP-MS because of the intrinsic short-term fluctuations of the ICP.* At a low counting rate the effects of these fluctuations may be greater resulting in the degradation of the peak shape. At higher counting rates a flat topped mass peak is obtained. Calibration however can usually be carried out not only using peak height but also with peak area so that the peak flatness is of no concern.C 5 2 I 1 1 0 I I I - 98.9 99.0 93.1 99.2 225.8 226.0 226.2 226.4 m/z Fig. 2 Ultrasonic nebulizer was used ( c ) and ( d ) ultrasonic nebulizer was not used Mass spectra for (u) and (c) WTc ( 1.76 pg ml-'; 1.12 mBq ml-') and (h) and ( d ) "'Ra (1.36 pg m1-I; 50.3 mBq m1-I) by HR-ICP-MS. (a) and ( h )JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 199 I VOL. 6 207 100 60 c 30 e 3 .- E O 0 ~ v) 231.8 232.0 232.2 232.4 236.8 237.0 237.2 237.4 Q c -5 100 OJ 10 iij - (D 50 5 0 0 I I 237.8 238.0 238.2 238.4 236.8 237.0 237.2 237.4 mlz Fig. 3 HR-ICP-MS. (a) (h) and (c) ultrasonic nebulizer was used (6) ultrasonic nebulizer was not used Mass spectra for (a) *j2Th ( I .O pg ml; 4.0 nBq ml-') (c) 23xU ( 1 .O pg rn1-I; 26 pBq ml-I) and ( h ) and ( d ) '"Np (2.0 pg ml-I; 0.052 mBq ml-') by .- m Q c .E 40 - m 0 2 iij 20 0 238.8 239.0 239.2 239.4 239.8 240.0 240.2 240.4 Fig.4 Mass spectra for 23'Pu and 240Pu (20.8 pg ml-'; 72.5 mBq mi-') by HR-ICP-MS. (a) Ultrasonic nebulizer was used ( h ) ultrasonic nebulizer was not used Detection Limits and Sensitivities Fig. 5 shows the background readings and sensitivities for some nuclides using ICP-MS and HR-ICP-MS indicating that lower background and greater sensitivity is achieved using HR-ICP-MS compared with conventional ICP-MS. In ICP-MS the relatively high background is mainly due to photons from the plasma reaching the photon sensitive detector by way of multiple reflections inside the instrument. A weak discharge in the lens system at the elevated ion-lens voltage and the pres- ence of charged species may also contribute to the back- ground.' However it is possible that this weak discharge only occurs when the wrong operating conditions are used.") In HR- ICP-MS fewer photons and ions might be scattered inside the device and because of the longer more complicated ion flight path and the use of narrow slits and lenses fewer reach the de- tectors.Therefore the continuum background of the HR-ICP- MS instrument was generally lower than that of the ICP-MS instrument. In HR-ICP-MS increased transmission of the ions can also be expected due to the acceleration of the ions by a high ion acceleration potential of 4 kV. Therefore higher sen- sitivity could be achieved by using HR-ICP-MS even with a conventional nebulizer than by using ICP-MS.The efficiency of the sample introduction into an ICP can be improved by substituting a USN which gives a rich spray of small droplets (4 pm). Hence the amount of sample reaching the plasma from the nebulizer was increased from 1% (usual with concentric type nebulizers) to about 10% with a USN."." As shown in Fig. 5 and Table 2 the sensitivity using HR- ICP-MS with a USN was increased about 10 times. In Table 2 the data are mean values of triplicate measurements for one sample. Detection limits of the order of 2-8 fg ml-' were achieved for the nuclides tested with the exception of and YJ. The relatively high detection limits for 'j2Th and 23xU were attributable to the 'memory effects' in the spectrometer or the 'high blank value' of the nuclides present in the reagents208 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 199 1 VOL.6 Concentration HR-ICP-MS HR-ICF-MS of standard - without USN with CSN Nuclide Half-life/years solution/pg ml-' Counts Detect ion limit */ counts Detect ion 1 imi t/ pg ml-' pg ml-' "Tc 2 . 1 4 ~ 10' 5.5 'I 95 0.23 456 0.008 11' 0.14 yBq' 8' 5 pBq' ??6Ra 1.6oX1O3 1.36 I17 0.03 2683 0.006 12' 1 mBq' 27' 0.2 mBq' 23r-l-h 1.4 Ix 10"' 1 .o - - 2435 0.02 379' 0.08 nBq' 2 . 1 4 ~ 1 0 ~ 1.98 94 0.04 1814 0.002 7' 1 yBq' 6' 0.05 pBq' '3"P ?MU 4 . 4 7 ~ 10' 1 .o - - 1919 0.02 302' 0.2 nBq' 23Ypu 2.4 1 x I 04 16.76 815 0.04 1246 1 0.005 14' 10 pBq' 6' 0.1 mBq' ?40pu 6 . 5 7 ~ 1 O3 4.04 :!05 0.05 3027 0.004 3' 0.03 mBq' 13' 3 pBq' * Detection limits calculated by using the equation in references 3 and 16.t Counts of the blank solution ( 1 mol dm-j HNO,). Table 3 Comparison measurement of the 240Pu to 23'Pu ratio by ICP-MS and HR-ICP-MS using a concentration of 20.8 pg ml-' for 23'Pu and 24"Pu which equals 72.5 mBq ml-' in activity Parameter ICP-MS HR-ICP-MS HR-ICP-MS without USN with USN Operating conditions- Mass range 238-245 238.8-240.4 MCA channels 5 12 50 Channel for peak search - 24 Dwell time 80 ms 640 ms Sweeps 3200 5 2u)Pu to z7yPu ratio* 0.256 0.252 0.243 Accuracy 6.2% 4.6% 0.8% Precision t 25.4% 7.8% 2.0% * The value certified by the Oak Ridge National Laboratory was 0.241. t Relative standard deviation of triplicate measurements for ICP-MS and relative Poissoin standard deviation for HR-ICP-MS. and water used.These detection limits however show an im- portant potential for HR-ICP-MS in the analytical study of en- v ironmen tal radioactiv ities. In the current environment the radioactivity concentrations of the nuclides discussed in this paper are of the order of 1 mBq g-I or less. For a given activity of 1 mBq using an activ- ity counting technique it takes 1 x los s (about I d) to obtain 100 counts assuming a unit efficiency. A 100% efficiency in the actual measurement cannot be expected. For instance the efficiency of a-spectrometry the 2n gas flow counting method and liquid scintillation counting are at most about 20 40 and 90% respectively. Gamma-spectrometry exhibits only several per cent. of the measurement efficiency. Because the detection limits are functions of the background and the sensi- tivity of the method and low background measurements are achieved with radiometric measurement (especially in a- spectrometry using a Si detector) detection limits as low as less than one to several mBq (total activity) are usually ob- tainable.By using the measurement conditions of HR-ICP-MS de- scribed in this paper low detection limits were obtained and total solution consumption was less than 10 ml; i.e. the amount of 226Ra needed for detection was only 2 mBq (6 pCi) showing two advantageous features of ICP-MS over conven- tional activity measurements. Radium-226 has the shortest half-life among the nuclides tested. Isotopic Information Although ICP-MS gives isotopic information for the analyte at a low concentration (pg ml-I level) the accuracy and precision are not as good because of the poor counting statistics.In HR- ICP-MS which has a higher sensitivity better isotopic data can be achieved. Plutonium has accumulated in the environment from several sources and the isotopic composition of Pu in environmental samples is characteristic of its origin. Table 3 shows a com- parison of the results for the measurement of the 240Pu to 23yPu ratio by ICP-MS HR-ICP-MS without a USN and HR-ICP- MS with a USN. The deviation of the 240Pu to 239Pu ratio from the certified value provided by the Oak Ridge National Labora- tory was 0.8% by using HR-ICP-MS with a USN under the op- erating conditions described. It should be stated however that the method of counting is not the only factor affecting the accuracy and the precision of the isotopic information.Previous measurementsI3 of Pb isotope ratios at a higher concentration (50 ng ml-I) than the present Pu using HR-ICP-MS gave a minimum relative stan- dard deviation (RSD) of 2% while with ICP-MS a 0.2% RSD was obtained. The measurement time per unit mass was 30 s in both instances. This is explained by the fact that the scanning speed in HR-ICP-MS is not as fast as that in ICP-MS so that fluctuations in the ICP' may have a greater influence on the result. In determining isotope ratios all the factors affecting the measurement should be taken into consideration. In conclusion these preliminary results suggest the feasibi-JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 199 1. VOL. 6 209 a) c - 237Np 239Pu ICP-MS HR-ICP-MS without ultrasonic nebulizer HR-ICP-MS with ultrasonic nebulizer Fig.5 Comparison of ( a ) background and ( h ) sensitivity in ICP-MS and HR-ICP-MS. wTc 1.8 pg ml-I; '37Np 2.0 pg m1-I; and 239Pu 3.5 pg ml-' lity of using HR-ICP-MS for the direct measurement of long- lived radionuclides in environmental samples. Further applica- tions in the field are expected. The authors thank Drs. R. C. Hutton and N. Bradshaw VG Elemental and T. Shimamura Marubun for their helpful dis- cussion; N. Bradshaw for providing Fig. 1; and M. Yoshida for drawing the figures. Part of this work was covered by a Grant in aid for Scientific Research from the Ministry of Edu- cation Science and Culture Japan under contract No. 63303015. Appendix Because the analytes referred to in this paper are all radio- active precautions are necessary in handling the solutions.However the amount of activity used was at the mBq level i.e. environmental level and even if the total amount was taken into the body almost no radiation hazard would be anticipated. The annual limits of intake (ALI) through inges- tion recommended by the International Commission on Ra- diological Protection (ICRP) International Atomic Energy Agency for w o r k e r ~ ~ ~ . ~ ~ are 1 x lox 7 x 104 3 x 104 3 x lo3 5 x los and 2 x 10' Bq for T c 226Ra 232Th 237N ' PY 23xU 239Pu and 240Pu respectively. Evidently the amounts used in the method described in this paper were far less than the ALI (1Os-1O'* times less) suggesting that any dis- charge was permissible.However laws regulations guide- lines etc. set by governments or recommendations by ICRP etc. should be strictly observed in handling radioactive materials. 1 2 3 4 5 6 7 8 9 10 1 1 12 13 14 15 16 References Kim C. K. Otsuji M. Takaku Y. Kawamura H. Shiraishi K. Iga- rashi Y . Igarashi S. and Ikeda N. Radioisotopes 1989,38. 15 1. Kim C. K.. Takaku Y. Yamamoto M. Kawamura H. Shiraishi K. Igarashi Y. Igarashi S. Takayama H. and Ikeda N. J . Radioonol. Nucl. Chem. 1989 132 13 1. Kim C. K. Oura Y. Takaku Y. Nitta H. Igarashi Y. and Ikeda N. J. Radioanal. Nucl. Chem. 1989 136 353. Igarashi Y. Kim C. K. Takaku Y. Shiraishi K. Yamamoto K.. and Ikeda N. Anal. Sci. 1990,6. 157. Morita S. Kim C. K.. Takaku Y. Seki R.. and Ikeda N. Int. J. Appl. Radial. Res. in the press. Bradshaw N. Hall E. F. H. and Sanderson N. E. J. Anal. At. Spec- from. 1989.4 80 1. Morita M. Ito H. Uehiro T. and Otsuka K. Anal. Sci. 1989,5609. Winge R. K. Eckels D. E. Dekalb E. L. and Fassel V. A. J. Anal. At. Speiwom. 1988,3. 849. Kawaguchi H. Tanaka T. and Mizuike A. Spe,Pc.trochim. Acta. Part B 1988,43,955. Hutton R. C.. personal communication. Fassel V. A. and Bear B. R. Spectim*hirn. Acta Part B. 1986 41 1089. Olson K. W.. Haas W. J. and Fassel V. A. Anal. Chem.. 1977. 49 632. Takaku U. unpublished results. International Commission on Radiological Protection (ICRP) Publica- tion 30 Part 1 (Annals of the ICRP Vol. 2 No. 3-4). Pergamon Press Oxford 1978. International Commission on Radiological Protection (ICRP) Publica- tion 30 Part 2 (Annals of the ICRP Vol. 4 No. 3/4) Pergamon Press Oxford 1980. Curie L. A. And. Chem.. 1968.40. 586. Paper 01039731 Recei\,ed September 3rd 1990 Accepted Decemher- 1 I th I990