Anal. Chem. 1997, 69, 2988-2993
Analysis of Fission Products Using Capillary Electrophoresis with On-Line Radioactivity Detection Gregory L. Klunder,* John E. Andrews, Jr., Patrick M. Grant, and Brian D. Andresen
Forensic Science Center, Lawrence Livermore National Laboratory, P.O. Box 808, L-371, Livermore, California 94550 Richard E. Russo
Lawrence Berkeley National Laboratory, M/S 70-193A Berkeley, California 94720
Capillary electrophoresis has been used to separate metal ions characteristically associated with nuclear fission. Indirect UV absorbance and on-line radioactivity detection were used simultaneously to monitor the analytes. The radioactivity detector consists of conical plastic scintillating material with the capillary passing through the center to provide a 4π detection geometry. The wide end of the cone is optically coupled to a photomultiplier tube. Transient isotachophoretic techniques were employed to stack large volumes of samples which had low specific activities. Radioactivity detection of 152Eu and 137Cs was achieved at the nanocurie level for 80-100 nL injections. The detector is approximately 80% efficient, enabling samples resident in the detector window for 0.1 min to be reliably assayed. The separation of 137Cs and 137mBa isotopes, which are in secular equilibrium, was modeled to demonstrate the effects of the rapid decay of 137mBa. Separation of metal cations by capillary electrophoresis has been demonstrated by numerous investigators and can be considered routine for several applications.1-5 Due to the similarities in ionic mobilities, complexing agents are added to the buffer to enhance the separation of the metal ion species. Numerous complexing ligands have been studied for different target ions and various applications. Although the lanthanides can be the most difficult to separate due to their size and charge similarities, R-hydroxyisobutyric acid (HIBA) incorporated in the background electrolyte has proven to be very effective.5 Since most metal ions do not exhibit significant optical absorption in solution, UVabsorbing electrolytes are used in the buffer for indirect detection. Detection limits of approximately 100 ppb and separation of mixtures consisting of 19 ions have been reported.4 This success in metal ion analysis indicated that capillary electrophoresis could be used for the rapid separation and detection of nuclear species of diverse interest. In these instances, it is necessary to determine whether an analyte is radioactive, in addition to measuring the presence and concentration of stable carrier ions. (1) Timerbaev, A. R. J. Cap. Electrophor. 1995, 2, 165-174. (2) Shi, Y.; Fritz, J. S. J. Chromatogr. 1993, 640, 473-479. (3) Chen, M.; Cassidy, R. M. J. Chromatogr. 1993, 640, 425-431. (4) Weston, A.; Brown, P. R.; Jandik, P.; Jones, W. R.; Heckenberg, A. L. J. Chromatogr. 1992, 593, 289-295. (5) Vogt, C.; Conradi, S. Anal. Chim. Acta 1994, 294, 145-153.
2988 Analytical Chemistry, Vol. 69, No. 15, August 1, 1997
On-line radioactivity detection for capillary electrophoresis has undergone limited development, as is summarized in Table 1.6-15 Most of the focus has been directed toward isotopes used for radiopharmaceuticals and radiolabeled biochemical analyses, e.g., 99mTc, 32P, and 35S. Reactor coolant water for regulatory compliance, high-level wastes, and environmental characterization are areas that have been addressed by other techniques, such as ion chromatography (IC), and could benefit from the advantages of CE.16-18 Nuclear fission products are principal analytes in these areas of investigation. Rapid separations with high resolution would enable CE to analyze active species that have short halflives or are difficult to separate by other techniques. Additionally, reduced sample volumes would minimize worker exposure, which can be especially hazardous when handling high-level radioactive wastes (such as can be found at the Hanford site).19 However, the combination of very small sample size and short analysis time can sometimes be a disadvantage for the optimum detection of radioactive species. Because some fission products of interest have short half-lives, increasing the separation time for these species is not practical, while injecting larger samples can result in band broadening and poor peak resolution. To (6) Pentoney, S. L., Jr.; Quint, J. F.; Zare, R. N. Anal. Chem. 1989, 61, 16421647. (7) Pentoney, S. L., Jr.; Zare, R. N.; Quint, J. F. In Analytical Biotechnology: Capillary Electrophoresis and Chromatography; Horvath, C., Nikelly, J. G., Eds.; ACS Symposium Series 434; American Chemical Society: Washington, DC, 1990; pp 60-89. (8) Gordon, J. S.; Vasile, S.; Hazlett, T.; Squillante, M. IEEE Trans. Nucl. Sci. 1993, 40, 1162-1164. (9) Altria, K. D.; Simpson, C. F.; Bharij, A. K.; Theobald, A. E. Electrophoresis 1990, 11, 732-734. (10) Kaniansky, D.; Rajec, P.; Svec, A.; Havasi, P.; Macasek, F. J. Chromatogr. 1983, 258, 238-243. (11) Kaniansky, D.; Rajec, P.; Svec, A.; Marak, J.; Koval, M.; Lucka, M.; Franko, S.; Sabanos, G. J. Radioanal. Nucl. Chem. 1989, 129, 305-325. (12) Tracht, S.; Toma, V.; Sweedler, J. V. Anal. Chem. 1994, 66, 2382-2389. (13) Tracht, S.; Cruz, L.; Stobba-Wiley, C. M.; Sweedler, J. V. Anal. Chem. 1996, 68, 3922-3927. (14) Westerberg, G.; Lundqvist, H.; Kilar, F.; Langstrom, B. J. Chromatogr. 1993, 645, 319-325. (15) Poirier, M. J.; Glajch, J. L.; Barry, E. F. Wintergreen Conference on Chromatography; 17th International Symposium on Capillary Chromatography and Electrophoresis, Wintergreen, VA, May 7-11, 1995; pp 58-59. (16) Gaur, S. J. Chromatogr. A 1996, 733, 57-71. (17) Fjeld, R. A.; Guha, S.; Devol, T. A.; Leyba, J. D. J. Radioanal. Nucl. Chem. 1995, 194, 51-59. (18) Smith, M. R.; Farmer, O. T., III; Reeves, J. H.; Koppenaal, D. W. J. Radioanal. Nucl. Chem. 1995, 194, 7-13. (19) Grate, J. W.; Strebin, R.; Janata, J.; Egorov, O.; Ruzicka, J. Anal. Chem. 1996, 68, 333-340. S0003-2700(97)00042-5 CCC: $14.00
© 1997 American Chemical Society
Table 1. Radioactivity Detection in Capillary Electrophoresis isotopes (decay mode, energy, half life) 32P
(β-, 1.71 MeV, 14.28 d)
32P
(β-, 1.71 MeV, 14.28 d) (IT, 140 keV, 6 h) 14C (β-, 156 keV, 5730 y), 32P (β-, 1.71 MeV, 14.28 d), 35S (β-, 167 keV, 87.2 d) 3H (β-, 19 keV, 12.3 y), 35S (β-, 167 keV, 87.2 d) 35S (β-, 167 keV, 87.2 d), 32P (β-, 1.71 MeV, 14.28 d) 11C (β+, 511 keV, 20.3 min), 18F (β+, 511 keV, 1.835 h) 99mTc (IT, 140 keV, 6 h) 137Cs (β-, 514 keV, 30.07 y), 137mBa (γ, 662 keV, 2.55 min), 152Eu (EC + β-, 122-1408 keV, 13.5 y) 99mTc
detector CdTe semiconductor, plastic scintillator/PMT (cooled) CdTe semiconductor NaI(Tl) scintillator/PMT Geiger-Mueller counter, scintillator/PMT
refs 6, 7 8 9 10, 11
off-column method, peptide-binding membrane imaged onto a CCD camera off-column method, phosphor imaging plate (BaFBr:Eu2+)/PMT plastic scintillator (poly(vinyltoluene))/PMT
14
BGO (bismuth germanate) crystal/PMT plastic scintillator (poly(vinyltoluene))/PMT
15 this work
address these difficulties, we have incorporated transient isotachophoresis using hydrodynamic injections, which combines isotachophoresis and zone electrophoresis to achieve on-column stacking. Conventional isotachophoresis (ITP) with conductivity detection has been used as a method of preconcentrating larger sample volumes, although the analyte zones do not physically separate. However, transient isotachophoresis (TITP) allows the stacking of large injection volumes, followed by separation and detection by optical techniques.20 In this work, we developed an on-line nuclear detector, similar to others reported in the literature, for the measurement of radioactive species separated by CE.6,7,13 Although room-temperature, semiconductor cadmium-telluride (CdTe) detectors can provide high-energy resolution, they have been determined to be less efficient in this particular application.6 Plastic scintillators are easily machinable and provide good efficiency, but they provide poorer energy resolution. For CE analyses, on-column radiation detection is limited to β and γ particles of sufficient energy to escape through the capillary. The minimum detectable energy for β particles is calculated to be approximately 150 keV for a 100 µm i.d. column with 125 µm fused-silica walls.21 Since CE is a rapid separation technique, any given analyte will be resident in the detection window for only a short time. In order to optimize the detection efficiency and count time, the largest scintillator that could be implemented without severe degradation of peak resolution was designed and fabricated. This paper reports the development of separation chemistry and the on-line radioactivity detector for the analysis of nuclear fission products. EXPERIMENTAL SECTION Instrumentation. A HP3DCE capillary electrophoresis system (Hewlett-Packard, Waldbronn, Germany) with uncoated fusedsilica capillaries was used for all separations reported in this work. The capillaries were either 100 µm i.d. (Polymicro Technologies, Phoenix, AZ) or 75 µm with a bubble window (200 µm) for optical detection (Hewlett-Packard). Total capillary length was 50 cm, and a constant voltage of 15 kV (300 V/cm) was used. The optical (20) Mazereeuw, M.; Tjaden, U. R.; Reinhoud, N. J. J. Chromatogr. Sci. 1995, 33, 686-697. (21) Knoll, G. Radiation Detection and Measurement; J. Wiley & Sons: New York, 1979.
12 13
Figure 1. Diagram of the on-line CE radioactivity detector. Scintillator material is dye-doped poly(vinyltoluene) (Bicron BC-400).
detector and the radioactivity detector were located 8.4 and 12.8 cm, respectively, before the outlet end of the capillary. The optical detection system employed a deuterium light source and a diode array detector for indirect UV absorption. Absorption was measured at 220 nm, referenced to 300 nm, with a 16 nm bandwidth and a 20 Hz sampling rate. An optical interface with a slit 50 µm wide and 620 µm long maintained the alignment of the capillary with the diode array detector. Sample injections were primarily hydrodynamic and are reproducible to 1-2% RSD according to the manufacturer specifications. Electrokinetic injections were also investigated and will be discussed. Radioactivity Detector. Plastic scintillator materials were machined into cones of diameter 9 mm on the wide end, tapering to 3 mm, with a total length of 10 mm (Figure 1). The plastic was poly(vinyltoluene) doped with a scintillating dye (BC-400, Bicron, Newbury, OH) with a refractive index (RI) of 1.581 and λmax of 423 nm. The manufacturer reported the light output of the BC-400 plastic scintillator as 65% relative to that of anthracene, which has a light output of 40-50% relative to that of NaI(Tl). The CE capillary passed through a 400 µm hole drilled through the middle of the cone, thereby providing 4π geometry for a 6 mm length of the capillary. Detection volumes were 26.5 and 47.1 nL for the 75 and 100 µm i.d. capillaries, respectively. The wide surface of the scintillating cone was optically coupled to a PMT through optical grease (RI ) 1.465, BC-630, Bicron). All other surfaces of the scintillator were coated with white, diffuse reflective material (BC-620, Bicron or Teflon tape) and surrounded with a lead shield. The wide surface of the scintillator filled the window of a small PMT (Model H5773, Hamamatsu, San Jose, CA), which operated on a +12V input with a voltage gain of 0-1 V, controlled by a 10 kΩ potentiometer. Detection was achieved with a gated Analytical Chemistry, Vol. 69, No. 15, August 1, 1997
2989
photon counter (Model SR400, Stanford Research Systems, Inc., Palo Alto, CA) using 0.1 s integration time and -6.0 mV discrimination level. Black electrical tape was wrapped around the entire assembly to reduce background light. Resultant overall dimensions of the detector package were 2 cm × 2 cm × 6 cm long. The analog signal from the photon counter was processed through an A/D converter (HP35900, Hewlett-Packard), and the data were displayed and stored in a separate channel of the CE data acquisition system. Reagents. Buffers used for the separations were creatinine (30 mM) and R-hydroxyisobutyric acid (HIBA, 8 mM) at a pH of 4.5. Standards of stable analytes were prepared at concentrations of 20 ppm from 1000 ppm stock solutions (SPEX Industries, Inc., or NIST spectrometric standards). Reliable radioactive analytes were available from the Lawrence Livermore National Laboratory Nuclear Chemistry Division: 137Cs at 1.401 × 105 dpm/100 µL and 152Eu at 1.093 × 107 dpm/100 µL (which also contained 154Eu at 1.073 × 105 dpm/100 µL). Although 152Eu is not a fission product, its chemical behavior is identical to that of Eu nuclides formed by fission. Radioactive species can be supplied with or without a carrier substance (a stable isotope of the same element). If produced by the no-carrier-added (NCA) method, the radionuclide will have a high specific activity (activity/mass). Conversely, if carrier is present, the specific activity will be lower. Prior to present analyses, the specific activities of the two radionuclides were unknown. Radioactivity assays were accomplished with a high-resolution Ge(Li) detector. The individual quantity of activity injected for each CE experiment was determined by postrun analysis of an outlet vial with a 2 in. × 2 in. NaI(Tl) well detector. The reagent used for injection volume stacking was high-purity 0.1 M hydrochloric acid (Seastar Inc.). CAUTION: Handling of radionuclides with high specific activity and penetrating radiation emission can present radiological hazards. Proper safety procedures for the handling, storage, and disposal of radioisotopes should be observed.
Figure 2. Separation of a mixture of lanthanide ions (80 nL hydrodynamic injection of 2.5 ppm of each ion) with indirect UV absorption detection. Buffer was 30 mM creatinine/8 mM HIBA at pH ) 4.5.
RESULTS AND DISCUSSION Sample Injection and Matrix Effects. Sample injection volumes (lengths) in capillary electrophoresis should be kept small to achieve sharp, well-defined peaks with a large number of theoretical plates.22 However, when analyte concentrations are very low, it is necessary to increase the injection volume. The electropherogram in Figure 2 shows the separation of several fission product species (the lanthanides) with an injection size of approximately 80 nL, i.e., 1.8 cm of capillary length (3.8% of total column length of the 75 µm i.d. capillary). A creatinine/HIBA buffer was used for indirect UV absorbance detection with a bubble capillary, and each cation was present at a concentration of 2.5 ppm in dilute acid solution (pH ) 2.0). Because of their similar mobilities, the lanthanides can be difficult to separate, especially with larger injection volumes. The peaks were baseline resolved but exhibited some tailing that became more evident when higher concentrations were analyzed. The number of theoretical plates measured for La3+ was approximately 100 000 but decreased to about 20 000 for Tm3+ for a 50 cm column. The data demonstrated that adequate separation could be achieved with large amounts of injected analyte. This capability often may be necessary for radioactive samples with low specific activities.
An alternative to increasing the injection volume is preconcentration of the analyte on the column by injecting the sample electrokinetically. Although this technique can produce improved detection limits, there is a preference for faster moving ions to undergo greater preconcentration than slower moving ions.23 In addition, matrix effects can produce complications when using electrokinetic injections, as shown in Figure 3. The two electropherograms in Figure 3a demonstrate that a dilute acid matrix has only a small effect on electrokinetic injections of Ce3+. For identical injection conditions (10 kV, 2.0 s), the sample prepared in dilute acid (pH ) 2.0) resulted in a peak of only slightly larger area than one prepared in a more neutral solution (pH ) 4.5). In both cases, peak tailing resulted from the large injection size. However, another fission product of interest, Cs+, was much more susceptible to matrix effects upon electrokinetic injection, as can be seen in Figure 3b. The possibility of a significant matrix influence on the resultant analysis suggests that electrokinetic injections should be used only after prior verification of the method for particular analytes of interest. Matrix effects can also cause problems when hydrodynamic injections are employed. Large-volume injections of lanthanides in dilute acid (pH ) 2.0) did not significantly affect the separation, as shown in Figure 2. However, matrix effects can seriously perturb large injections of faster moving ions in a more acidic medium (see Figure 4). A 98 nL hydrodynamic injection, 1.3 cm long (2.6% of the total length of a 100 µm i.d. column), of a 10 ppm Cs+/Ba2+ standard at pH ) 2.0 resulted in severe distortion of Cs+ and Ba2+ peaks (Figure 4a). Sample overloading and the acid medium caused the analyte peaks to split into two portions, which were not baseline resolved. However, Figure 4b demonstrates how these substandard peaks could be sharpened to welldefined signals by means of transient isotachophoresis. Isotachophoretic conditions allow ions to be stacked into distinct analytic zones based on their mobilities. The sample is bracketed between a fast leading electrolyte and a slower tailing electrolyte, inducing the ions to stack into concatenated zones. The concentration of ions in the zones is regulated by the mobility and concentration of the leading electrolyte, described by the
(22) Huang, X.; Coleman, W. F.; Zare, R. N. J. Chromatogr. 1989, 480, 95-110.
(23) Huang, X.; Gordon, M. J.; Zare, R. N. Anal. Chem. 1988, 60, 375-377.
2990 Analytical Chemistry, Vol. 69, No. 15, August 1, 1997
Figure 3. Electrokinetic injections of (a) Ce3+ (10 kV, 2.0 s) and (b) Cs+ (10 kV, 3.0 s). The samples were an aqueous solution of the chloride salt (pH ) 4.5) and an acidic solution in dilute HNO3 (pH ) 2.0); 20 ppm concentrations.
Kohlrausch regulating function.24 Detection of the analyte zones in isotachophoresis (ITP) is normally achieved by conductivity measurements. Transient isotachophoresis (TITP) is a combination of isotachophoresis and capillary zone electrophoresis. The capillary is filled with the buffer, and the leading electrolyte is injected before the sample. The analyte ions are initially stacked into concentrated zones under the established ITP conditions. As the leading electrolyte diffuses into the buffer, ITP conditions no longer persist, and the analyte ions separate in a zone electrophoretic mode.20,25-27 Transient isotachophoresis enables oncolumn preconcentration of cations and also allows detection by indirect UV absorption. In Figure 4b, TITP conditions were established by injecting a small volume of 0.1 M HCl as the leading electrolyte prior to injection of the sample, followed by creatinine buffer as the tailing electrolyte. For Cs+ and Ba2+, stacking by TITP resulted in well-defined peaks with reduced peak widths compared to those from separations under conventional CE conditions. However, some peak fronting was still evident due to effects of electromigration dispersion. (24) Bullock, J.; Strasters, J.; Snider, J. Anal. Chem. 1995, 67, 3246-3252. (25) Reinhoud, N. J.; Tjaden, U. R.; van der Greef, J. J. Chromatogr. A 1993, 653, 303-312. (26) Gebaur, P.; Thormann, W.; Bocek, P. J. Chromatogr. 1992, 608, 47-57. (27) Krivankova, L.; Gebauer, P.; Thormann, W.; Mosher, R. A.; Bocek, P. J. Chromatogr. 1993, 638, 119-135.
Figure 4. Electropherograms of a Cs+/Ba2+ standard without (a) and with (b) a leading injection of HCl (pH ) 1) for stacking. Injection size was 98 nL, and analyte concentrations were 10 ppm each.
Radioactivity Detection. The radioactivity detector (Figure 1) was located 37.2 cm from the inlet of the capillary and 4.4 cm before the optical detector. An analyte which passed the radioactivity detector at 6 min would have a velocity of 6.2 cm/min and a minimum resultant peak width of 0.1 min arising from the size of the detector. Thus, the minimum temporal separation of any radioactive peaks on this system must be approximately 0.1 min in order to achieve analyte separation. The separation effectiveness is determined by peak full width at half maximum (fwhm), which must also incorporate the width of the sample. Background counts for the scintillation detector were on the order of 150 counts/min (cpm), with no cooling of the PMT. However, at the beginning and end of each experiment (as can be seen in Figure 6a, although not shown at the end), a large transient in the scintillator detector output was observed. Pentoney et al. also reported similar behavior, which was hypothesized to arise from the application of high voltage to the capillary.6 In the present work, this effect did not interfere with any separations data, since detector response returned to stable background levels in less than 1 min. Initial evaluation of the detector design was performed with the separation of 152Eu, a fission product which β-decays with both electron and positron emission with a 13.5 year half-life. Figure 5 displays data from a separation of 152Eu monitored with both the radioactivity and optical detectors. The radioelectropherogram in Figure 5a shows a single large peak arising from the decay of Analytical Chemistry, Vol. 69, No. 15, August 1, 1997
2991
Figure 5. Electropherograms with (a) radioactivity detection and (b) indirect UV absorption. The sample of 152Eu injected was approximately 4 nCi in a volume of 40 nL. A 17-point Savitzy-Golay smoothing function was applied to the radioactivity data.
Figure 6. Electropherograms with (a) radioactivity detection and (b) indirect UV absorption. The sample of 137Cs injected was approximately 15 nCi in a volume of 98 nL. A 17-point SavitzyGolay smoothing was applied to the radioactivity data.
152Eu
daughter nuclides were individually detected with the on-line radioactivity detector. Figure 6 shows a radioelectropherogram (Figure 6a) and an indirect UV electropherogram (Figure 6b) for separation of a 137Cs sample. Stacking by TITP, as previously described, was employed for this separation and resulted in well-defined peaks. Figure 6b depicts a relatively large peak for stable Ba2+, indicating an aged sample that had decayed undisturbed for some time. The peaks which appear after the Ba2+ peak are attributed to the matrix but have not been identified. In Figure 6a, the first peak, 137Cs, contains ∼65% of the total measured activity, while the second peak, 137mBa, contains ∼35%. For the purposes of this experiment, the activity of the 137Cs could be considered constant; however, by the time the 137mBa peak reached the radiation detector (2.7 min), its activity had been reduced by a factor of 2. During the time of the separation, 137Cs was constantly generating 137mBa in the attempt to regain equilibrium. This real-time decay resulted in the region between the two peaks, where the signal did not return to baseline. Real-time decay of 137mBa contributes to the shape of the peak, which has a width of 0.5 min (0.2 half-lives). Consideration of a semiquantitative model will elucidate the separation of the two isotopes from secular equilibrium. Under the initial conditions, the 137Cs activity equals the 137mBa activity, and an instantaneous separation of the two species would result in two distinct peaks of equal intensity. However, after separation,
during separation; net photopeak counts were 768, with a sample residence time of 0.14 min, as determined from the peak’s fwhm. The amount of activity injected was determined by assaying the outlet vial after allowing adequate time for the entire sample to migrate off the capillary. In the sample, the total activity was measured to be 9125 dpm (4.1 nCi, 152 Bq), corresponding to an on-line detection efficiency of ∼60%. The electropherogram in Figure 5b was obtained by indirect UV absorption detection and a bubble capillary, and the concentration of Eu3+ was estimated to be 10 ppm from previous calibration data. A more complex fission product is 137Cs, which decays by the following mechanism: 137
Cs f 137mBa + β- (514 keV) f 137Ba + γ (662 keV)
Radioactive 137Cs has a 30.0 year half-life and β-decays to a metastable 137mBa, which has a 2.55 min half-life and decays in turn to stable 137Ba with emission of a 662 keV γ-ray. Due to the large difference in half-lives, the two unstable nuclides are in radioactive secular equilibrium and remain so unless disturbed by chemical separation. Even if separated, after approximately 25 min (∼10 half-lives of 137mBa), the equilibrium will reestablish such that the two isotopes will have equal activities. A common tactic for nuclear detection is to report 137Cs as a simple γ-decay of 662 keV with a 30 year half-life. However, since CE is a relatively fast separation technique, both the parent and the 2992 Analytical Chemistry, Vol. 69, No. 15, August 1, 1997
decay is effectively negligible. The model calculations fit the observed data (Figure 6a) very well. At a separation time of 2.50 min, the relative amount of activity in each of the two peaks was calculated to be 66.4% for 137Cs and 33.6% for 137mBa, compared to the observed values of 65.3% and 34.7%, respectively. The efficiency for the detection of 137Cs was determined to be ∼80%, somewhat higher than that for 152Eu. CONCLUSIONS
Figure 7. Model demonstrating the separation of 137Cs from 137mBa in secular equilibrium. The solid bars represent the level of Cs activity and the open bars the level of Ba activity, as a function of time after the chemical separation.
the 137Cs will continuously regenerate 137mBa, which will also undergo separation in situ and decay to 137Ba. Figure 7 depicts the results of a model which incorporates the decay and production of 137mBa over the course of a 137Cs separation. Activity is plotted as a profile of speciated radioactivity vs separation time. At time zero, the two isotopes are present at identical activity levels (137Cs, solid bars; 137mBa, open bars). As the two are separated, the amount of initial 137mBa activity remaining is determined by the decay rate law:
A(t) ) A0(1 - e-0.693t/t1/2) where A(t) is the amount of activity at time t, A0 is the initial amount of activity, and t1/2 is the half-life of 137mBa. The rate of isotope production from parent decay is described by
A(t) ) A0(e-0.693t/t1/2) During the time required for CE separation, the amount of 137Cs
On-line radioactivity detection of species produced by nuclear fission was achieved at low nanocurie detection levels. The small volumes required for analysis by CE minimize worker exposure and generated waste. Improvements in detection configuration, such as coincidence counting and cooling the PMT, would reduce the background and improve the efficiency. Voltage programming could also be used to increase the counting time and lower the detection limits, as demonstrated by Pentoney et al.6,7 Additional experimental tactics, such as larger injection volumes, radiation detection under stopped-flow conditions, or postseparation analyte trapping with conventional counting, could also be implemented to lower overall limits of detection. ACKNOWLEDGMENT This work was funded by the Department of Energy Nonproliferation and National Security Program, Office of Research and Development (NN-20). The work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract No. W-7405-ENG-48.
Received for review January 14, 1997. Accepted May 9, 1997.X AC970042E X
Abstract published in Advance ACS Abstracts, July 1, 1997.
Analytical Chemistry, Vol. 69, No. 15, August 1, 1997
2993
