²³⁹Pu/²⁴⁰Pu Isotope Ratios Can Be Determined Using LIBS

Laser Induced Breakdown Spectroscopy (LIBS) is also known as laser ablation/optical emission spectroscopy. This type of high-resolution emission spectroscopy has received considerable attention as a versatile analytical technique. The technique has several advantages over other forms of analysis for nuclear applications; those of greatest importance at the Los Alamos Plutonium Facility include reduction of sample size, direct analysis in inhomogeneous matrices, reduced turnaround time between sample submittal and results, and in-situ analysis capability. The Plutonium Facility currently employs standard inductively coupled plasma mass spectrometry/atomic emission spectroscopy for most chemical analyses, with dissolution of a fairly large sample (~ 0.5 g) normally required. Isotope ratios are determined by thermal ionization mass spectrometry (TIMS) or gamma spectrometry. LIBS has been previously applied to actinide analysis, including recent work involving U isotope ratio determination.

Figure 1. A LIBS plasma plume produced from a lump of electrorefined Pu metal. The laser impinges on the sample vertically from the top, with emission collected horizontally by the 40 mm diameter, f/3 lens at far left. The sample chamber viewport is approx. 2.5" in diameter.

Observation of isotope shifts through the use of optical emission spectroscopy is not a common application for LIBS, mainly because of the very high resolution needed. However, a French research group at CEA Saclay has recently reported a LIBS isotope ratio determination for uranium using the U(II) line at 424.437 nm with a ²³⁸U/²³⁵U isotope shift of 1.39 cm^-1 (0.025 nm). Their observed linewidth of 0.67 cm^-1 (0.012 nm) was approximately twice as broad as the instrument limit of 0.305 cm^-1 (0.0055 nm). This linewidth was produced using experimental conditions of 308-nm laser wavelength, 0.46-GW/cm² focused laser energy density, a delay time of less than 500 ns, and 3 Pa (~20 mtorr) of air as the buffer gas.

One of the largest plutonium isotope shifts occurs in the Pu(I) emission line at 594.52202 nm, which has a ²³⁹Pu/²⁴⁰Pu isotope shift of 0.355 cm^-1, approximately one-fourth of the isotope shift of the U(II) transition. The smaller value for this isotope shift is partially due to the 1 atomic mass unit (amu) difference between common Pu isotopes as compared to the 3-amu difference between the two most abundant natural isotopes of uranium. This small isotope shift makes an isotope ratio determination more challenging. Under the experimental conditions reported for the observation of the uranium isotope ratio, where the linewidth is 0.67 cm^-1, the emission from the plutonium isotopes would not be resolved, and an isotopic ratio measurement of ²³⁹Pu/²⁴⁰Pu using LIBS would not be possible. The linewidth reported in the French work is attributed to Doppler and Stark broadening in the very hot plasma of the French experiment. A higher buffer-gas pressure and longer delay times should result in a cooler plasma, in which smaller isotope shifts may be observed. We show that this approach allows the LIBS measurement of the ²³⁹Pu/²⁴⁰Pu isotope ratio.

The LIBS system employs the following major components: a pulsed laser, sample chamber, emission spectrometer, detector, and computer. Work within the Plutonium Facility requires that radioactive samples be contained inside a glove box. Only the sample chamber and a minimum of optics reside within the glove box (see Figure 1); all other optical components are mounted on external optical tables. A pulsed and Q-switched Nd:YAG laser was used, operating at its fundamental wavelength of 1064 nm and generating pulse widths of ~5 ns. The laser energy is typically attenuated to ~25 mJ and focused into a less than 100-mm spot on the sample surface, producing an energy density of ~1011 W/cm². This high energy density is sufficient to both vaporize the sample surface and to generate a high-temperature plasma consisting of both neutral atoms and ions. The plasma species are electronically excited and emit light characteristic of all elements present in the ablation plume. The sample is enclosed in a vacuum chamber that is backfilled with helium at 100 torr pressure. This atmosphere allows some translational and electronic cooling of the plasma without excessive quenching of the emission. The emission is collected and collimated before exiting the glove box through a second viewport. The light is then focused onto the entrance slit of a high-resolution spectrometer, where it is wavelength-dispersed and collected on an intensified charge-coupled device (ICCD) camera. A broad continuum emission background is avoided by electronically delaying the ICCD by 1 ms following the laser pulse collecting the emission over a gate width of 5 ms.

A Pu transition was selected for analysis based on its large isotope shift, high intensity, and minimal spectral interference from neighboring lines. Atomic emission dominates the plasma at longer ICCD delay times, and as the plasma cools, expansion slows, thus allowing more electron-ion recombinations to occur. As a result, atomic emission is favored because of a decrease in both Doppler and Stark line broadening. Thus, the Pu atomic line at 594.52202 nm (16815.576 cm^-1) was chosen for this work, with a previously observed ²³⁹Pu/²⁴⁰Pu isotope shift of 0.355 cm^-1.

Two samples of greatly differing isotopic composition were chosen for characterization. The first sample consisted of Pu metal of a nominal 239/240 isotopic ratio of 93/6. The second sample was PuO₂ with a nominal ratio of 49/51. The oxide sample was analyzed as a pressed pellet.

The ²³⁹Pu/²⁴⁰Pu isotope shift of 0.355 cm^-1 from the plutonium atomic line at 594.52202 nm is clearly resolved in our plasma conditions (see Figure 2). Curve-fitting yields integrated peak-area ratios that match isotopic ratios as measured by TIMS or gamma spectrometry within ±0.5%. The curve fit typically produces a peak separation of ~6.7 pixels, with full-width-at-half-maximum at ~3.2 pixels. The observed plutonium linewidths are calculated to be 0.19 cm^-1 (0.0067 nm) based on the known dispersion of the spectrometer. These linewidths are within the experimental error of the ideal instrument-limited linewidth, which is calculated to be 0.15 cm^-1 (0.0052 nm) based upon the known modulation transfer function for the ICCD system.

Figure 2. LIBS produced a one-dimensional spectrum of the Pu emission line from ²³⁹Pu/²⁴⁰Pu isotopes in PuO₂ at 594.52202 nm, clearly showing resolution of the ²³⁹Pu/²⁴⁰Pu isotope shift. The LIBS setup at Los Alamos produces conditions better suited to such fine discriminations in actinides than a similar French experiment does (recently conducted at CEA Saclay).

In the uranium work by the French group, the uranium plasma is dominated by ionic lines that exhibit a larger degree of broadening. They study the ²³⁸U/²³⁵U isotope shift using an ionic line at 424.437 nm with an isotope shift of 0.025 nm, or 1.39 cm^-1, and report line broadening exceeding the instrument limit by a factor of 2 (observed linewidth ~0.012 nm, instrument limit 0.0055 nm). A Doppler width of 0.0106 nm was calculated for the U(II) line, with Stark width ~0.002 nm at a delay time of 350 ns and ~20 mtorr buffer-gas pressure.

Although the experimental apparatus used by the French group closely resembles that of our work, there are major differences in the experimental conditions that greatly affect intrinsic linewidth. In particular, our conditions of increased buffer gas pressure and longer delay time allow the plasma to achieve much lower translational temperatures, in spite of much higher laser energy densities. The use of a longer wavelength emission line for analysis provides greater peak separation. These effects result in smaller linewidths and adequate resolution that allow the plutonium isotopic ratio to be determined. The linewidths observed under our conditions for plutonium should be applicable to all of the light actinides (Th-Cm) under the same experimental conditions because the first ionization potentials are similar, ranging from 5.97 eV for Am to 6.27 eV for Np, and changes in the relative masses are small. Since the reported largest isotope shifts for Pu and U are in the range of 0.3­0.5 cm^-1/amu, the isotopic shifts for all of the light actinides should be similar, and LIBS should be applicable for isotopic ratio determinations across the light actinide series.

We have shown that LIBS may be applied to plutonium isotopic analysis. The technique is sensitive, essentially nondestructive, and can produce accurate results with reasonable precision. The accuracy of the technique critically depends on adequate spectral resolution of isotopic emission. We find that our conditions are well suited to resolution of 1-amu isotope shifts, and we expect that analysis of all actinides are possible under our conditions, of which the most important are long delay times and high buffer-gas pressures.

This article was contributed by Coleman A. Smith (NMT-11). Other researchers on the project include Max A. Martinez and D. Kirk Veirs (NMT-11 and David A. Cremers (C-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