Researchers Evaluate Isotope Ratio Techniques for Actinide Analysis

Isotope ratio and isotope dilution analyses play a key role in the characterization of nuclear materials for both defense and energy applications. In many cases we use thermal ionization mass spectrometry (TIMS), which offers optimum accuracy and precision: less than 0.001% relative standard deviation (RSD). However, in cases where less-precise and less-accurate data (~0.5% to 3.0% RSD) will meet customer needs, or when very rapid results are required, isotope ratio analysis can be undertaken by other means. Quantitative elemental and isotopic data can be obtained directly from the sample surface through the use of cathodic sputtering (glow discharge mass spectrometry-GDMS), laser ablation (laser ablation inductively coupled plasma mass spectrometry-LA-ICPMS), or by ion beam bombardment (secondary ion mass spectrometry-SIMS). In this article, we describe the capabilities of two plasma-based isotope ratio analysis techniques: GDMS and ICPMS.

Figure 1. Scanning electron microscope image of a sputter crater in glass after 5 hrs. of analysis using a secondary cathode. The bright areas around the perimeter of the crater are coated with Ta sputtered from the secondary cathode.

ICPMS is the method of choice for most trace element determinations in radioactive materials at LANL. The ICP ion source is an extremely efficient ion generator that operates at near-atmospheric (~760 torr) pressures. Nearly complete analytical coverage of the periodic table is available from ICPMS. Samples may be introduced as solutions, slurries, or suspensions of very fine particles generated by laser ablation. Samples are nebulized, ignited in an rf-generated argon plasma at >7000°C, and sampled into the mass spectrometer through a series of cones. The sampling cones, along with a powerful differential pumping system, permit ions to make the transition from the high-pressure ion source region to the low-pressure (~10-7 to 10-8 torr) mass analysis region.

In a glow discharge (GD) ion source, direct atomization and ionization are achieved via cathodic sputtering of the sample surface in a low-pressure (0.1 to 10 torr) argon atmosphere. Unlike other direct sampling techniques (e.g., SIMS or LA-ICPMS), the GD ion source exhibits relatively uniform elemental response and can rapidly produce quantitative trace analyses (parts per billion) without the need for strictly matrix-matched standards. The GD ion source is highly efficient and is capable of measuring ~99% of the elements in the periodic table. Its chief limitation is that samples must conduct electricity. For ceramics, soils, and other insulating materials, bulk analyses are performed by mixing a powdered sample with a conducting metal binder, or by using a tantalum (Ta) secondary cathode. A radiofrequency-powered GD ion source also permits direct sputter atomization of insulating materials.

Mass bias in mass spectrometry arises when physical phenomena present during ionization, sampling, or analysis result in the preferential loss of either light or heavy ions from an ion population. Thus, the measured isotope ratio is different from the actual value. Although mass bias in ICPMS and GDMS is a predictable function of element mass and can be corrected, it is generally far more severe than in TIMS. For this reason, isotope ratio measurements made using plasma-source mass spectrometry are inherently less accurate than TIMS measurements.

Extensive research has been done on isotope ratio analysis using both double-focusing single collector and "multicollector" ICPMS. The latter are equipped with up to nine independent detectors that measure multiple ion beams simultaneously and can attain sensitivity and precision that rivals that of multicollector TIMS instruments. Even so, single-collector, double-focusing ICPMS (and GDMS) instruments are far less expensive, easier to maintain, and offer exceptional analytical versatility. For "real-world" samples, a double-focusing, single-collector ICPMS is capable of accuracy and precision of 0.5% to 2.0% RSD, depending on the actual concentration of the isotopes in the unknown. Isotope dilution analysis, preconcentrat-on, and on-line separations for key elements are also possible using solution-based ICPMS.

In contrast, only a few investigators have used GDMS to measure isotope ratios. NMT-1 researchers have measured isotope ratios of several elements (B, W, Tl, Pb, U) in a variety of metal and glass standards using a double-focusing, single-collector GDMS instrument. The mass ranges can be selected and scanned by changing the magnetic field intensity. Total elapsed time required for each analysis (i.e., 8 to 10 sets of 8 to 15 scans) was 30 to 40 minutes. All uncertainties are reported at the 2-s level.

First, we performed isotope ratio analyses of tungsten (W, 0.102 wt. %) and lead (Pb, 240 ppm) in a steel standard (NIST SRM 1264a). The precision of W isotope ratios measured using GDMS was ~0.4% to 1.5% RSD, and the accuracy was less than1% RSD. Mass bias in Pb isotopes was estimated using the departure of the 186W/183W ratio from the known value, and assuming an exponential mass fractionation law. The precision of Pb isotope ratios in SRM 1264a, measured by GDMS, was ~1.5% to 3.6% RSD.

For the next set of analyses, we analyzed a Na-Ca-Al silicate-based glass-ceramic standard (NIST SRM 611) using a Ta secondary cathode (Figure 1). The standard glass contains trace amounts of Pb (426 ppm), Tl (62 ppm), and U (461.5 ppm). Furthermore, the isotope ratios of Pb (²⁰⁸Pb/²⁰⁶Pb and ²⁰⁷Pb/²⁰⁶Pb) in this material have been characterized using TIMS. Isobaric interference from (TaNa)⁺ ion clusters (mass = 203.9378 amu) prevented the collection of ²⁰⁴Pb at a mass resolution of 1500, although the remaining Pb isotopes (²⁰⁶Pb, ²⁰⁷Pb, ²⁰⁸Pb) were interference-free. Mass bias, measured using ²⁰⁵Tl/²⁰³Tl, varied from ­0.7% to +0.9% over 9 runs, much lower than typical values for ICPMS (> 1.5%). These data also support our theoretical calculations, which predict that mass bias in the GD ion source is less severe than in the ICP ion source. The GDMS Pb isotope determinations were typically accurate to within 1% of TIMS Pb isotope measurements and precise to < 2% RSD.

GDMS permits the collection of multi-element isotope ratio data in a single run; thus, a uranium isotope ratio (²³⁵U/²³⁸U) was also measured in the glass standard, along with Pb and Tl. The extremely low abundance of ²³⁵U (ca. 1 ppm) resulted in low signal intensities, and the ²³⁵U/²³⁸U ratios obtained by GDMS were less precise (2.3% to 9.3% RSD) and less accurate (0.4% to -7.8% RSD) than the Pb and Tl measurements. However, the uranium isotope ratio data are still useful. The GDMS analyses indicated that the U in the glass standard was depleted in ²³⁵U (²³⁵U/²³⁸U = 2.2x10^-3 to 2.5x10^-3). This was confirmed by NIST personnel and in published reports. Note that the characterization of depleted U in the glass standard took less than 1 hour by GDMS.

Conclusions
Most commercial, magnetic-sector ICPMS and GDMS instruments are single-collector, double-focusing mass spectrometers designed for rapid determinations of trace and major elements. However, sector GDMS and ICPMS may also be used to acquire reasonably precise and accurate (0.5% to 2.0% RSD) isotope ratios on trace and major elements. Both techniques are very rapid and require minimal sample preparation but with decreased sensitivity, accuracy, and precision relative to TIMS. Moderate-precision data may be useful in some applications; however, isotope ratios acquired using GDMS and ICPMS may be most useful as screening techniques before TIMS data are obtained from selected samples.

This article was submitted by David M. Wayne (NMT-1.) Others contributing to this work are Wei Hang and Vahid Majid (CST-9) and E. Larry Callis, Debbie Figg, Diane McDaniel, and Cris Lewis (NMT-1).

NMT | LANL | DOE
Phone Book | Search | Help/Info

L O S  A L A M O S  N A T I O N A L   L A B O R A T O R Y
Operated by the University of California for the US Department of Energy