Key Points
- The joint study achieved a detection limit of ~1.3 picograms of uranium using single-shot laser-induced breakdown spectroscopy (LIBS), through optimizing emission line selection (notably the 682 nm line), light collection efficiency, and hardware. This approach allowed accurate isotopic ratio (²³⁵U/²³⁸U) analysis without calibration standards.
- Unlike traditional methods like ICP-MS, which require sample digestion and are not field-deployable, LIBS provides fast, direct analysis of solid samples in minutes, making it highly suitable for field use and environmental monitoring of trace uranium.
- The technique’s adaptability to other isotopic systems and ongoing instrument miniaturization highlight its promise for broader applications.
A joint study conducted by Applied Spectra (West Sacramento, California) and the Lawrence Berkeley National Laboratory (Berkeley, California) used laser-induced breakdown spectroscopy (LIBS) to measure uranium with a detection limit as low as 1.3 picograms, achieving unprecedented sensitivity by optimizing the uranium emission line, system hardware, and light collection efficiency. The researchers analyzed mock uranium particles on silicon wafers and demonstrated isotopic analysis for 235U:238U ratios using peak fitting without calibration standards. The relative standard deviation ranged from several percent down to about 1% depending on the sample mass and enrichment. All measurements were done with single-shot LIBS, showcasing its potential as a sensitive and robust tool for uranium isotopic analysis.
Richard Russo is the corresponding author for the paper that resulted from this research (1), and will receive, on behalf of the rest of his team, the 2024 Spectrochimica Acta Part B Best Paper Award. The award will be presented at SciX 2025, taking place from October 5 to 10 at the Northern Kentucky Convention Center in Covington, Kentucky. Spectroscopy spoke to Russo about this study and the resulting paper as part of our continuing interview series with SciX award winners.
Can you explain how laser-induced breakdown spectroscopy (LIBS) applies to uranium detection, and why high sensitivity is important?
Although laser-induced breakdown spectroscopy (LIBS) is mainly recognized as an elemental analysis technique, there are increasing initiatives focused on developing this technology for isotopic analysis. The ability to measure isotopic compositions from direct solid samples quickly in field conditions is a key factor driving research and development in LIBS.
Acid digestion or laser ablation with detection using inductively coupled plasma mass spectrometry (ICP-MS) offer high sensitivity for this type of analysis where a limited number of micron-sized particles is present. Although sensitive, these are laboratory-based instruments, not field portable, and generally require more time for analysis than LIBS. Enhanced sensitivity is necessary with LIBS to enable the analysis of minor isotopes that have low relative abundance.
The primary motivation for developing LIBS for this type of analysis is to provide a method for direct analysis using a compact instrument that is ideal for field transport.
How does LIBS compare to other uranium detection methods in terms of sensitivity and speed?
LIBS sensitivity is generally reported in absolute mass, not relative concentration. With ICP-MS, sensitivity is reported relative to the concentration, typically at ppb levels or lower, depending on the sample quantity analyzed and the analysis time. We demonstrated LIBS to detect uranium with a single laser shot from tens of picograms of material, with a limit of detection of approximately 1.3 picograms of uranium.
In terms of speed, LIBS (and laser ablation for ICP-MS) are micro-sampling technologies that enable direct analysis of solid samples. As a result, LIBS measurements require only a few minutes, whereas ICP-MS involves hours for sample digestion procedures, and measurement time. Additionally, when a sufficient sample is available, ppb levels can be achieved from LIBS for some elements by increasing the number of pulses used for sampling and accumulating data. However, with single-particle LIBS analysis, only a few picograms of U are available for analysis, so measuring ppb in such minuscule amounts would require single-atom detection.
What were the main challenges in detecting uranium at picogram levels, and how did your team overcome them?
The two main hurdles are spectral resolution and sensitivity. When considering sensitivity, the key factor is that uranium is not a strong optical emitter in LIBS plasma; more specifically, the emission energy is distributed across hundreds of emission lines. Part of our work involved establishing the optimal emission line for this application. Up until now, the most widely used line has been the 424 nm line due to its relatively large splitting of 24 pm, but it didn’t seem strong enough to help us achieve our goal. A tradeoff on sensitivity versus splitting was required. In this case, since the total amount of mass was very low, we decided to use one of the strongest lines (682 nm) without concern for self-absorption, that has a splitting of about 18 pm. This line is also devoid of spectral interferences. Since we are interested in isotopes, we needed to identify a line with good sensitivity, adequate splitting of the 235U/238U emission lines, and minimal line broadening. The 682nm line fulfilled these conditions.
In terms of spectral resolution, one of the main issues is that traditional off-the-shelf spectrometers for high-resolution applications exhibit low light throughput. Therefore, another significant part of our effort was to develop a specialized spectrometer with a narrow spectral range that provided high light throughput, along with good spectral resolution, particularly tailored to the 682 nm line. Fortunately, we succeeded in this endeavor, which was a significant factor in achieving our goal.
Why is spectral peak height chosen as the analytical parameter, and what advantages does it provide?
Peak height is commonly used for the limit of detection (LOD). The critical parameter is signal-to-background noise, and peak height well defines how net signal can be distinguished from the background fluctuations.
What factors affect the size of particles that LIBS can analyze effectively, and how was this demonstrated in your study?
The total mass of the specific element of interest serves as the primary metric for our analysis. To achieve accurate measurements, we have carefully established a particle calibration scheme that utilizes dried aerosols, helping us effectively determine the particle equivalent masses of uranium. In this instance, we calibrated the instrument by verifying that the total mass was adequately sampled, ensuring our data reflects accurate measurements. It is also important to note that, in terms of sensitivity, the size of the particles plays a significant role, exerting a cubic influence on the total mass. This means that even slight variations in particle size can lead to considerable differences in the measured mass. This understanding is crucial for the accuracy of our measurements and the overall validity of our findings.
What challenges are associated with peak fitting of LIBS spectra for isotopic analysis, and how were they addressed?
Peak fitting is the ideal approach because it eliminates the need for isotopic standards to extract the isotope ratio from the spectra. Peak fitting must rely on a specific line profile, (Gaussian, Lorentzian, Voight) that may not ideally represent the actual LIBS line profile. However, the use of Lorentzian comes very close to fitting the spectra with good accuracy and precision. Peak fitting also relies on the purity of the spectra (spectral overlaps). Again, the 682nm line mitigates most challenges with peak fitting as it is a strong line in a spectrally clean region. Specific for isotopic analysis, the hyperfine structure of the 235U emission line is also important as it changes the line width or even the shape. The availability of the hyperfine structure constants also needs to be considered for optical isotopic analysis.
How could this LIBS technique be adapted for onsite nuclear safeguards and environmental monitoring?
This is our goal–a commercial instrument that can be utilized by trained staff, to simply place a sample in the instrument, press start, and let the software run the experiment. Particles can be located using machine vision, or by raster scanning the laser over a sample substrate. With specific spectrometers, this CONOPS can be utilized for all environmental monitoring applications.
What improvements do you anticipate that could push detection limits into the femtogram
regime?
Improved light collection efficiency, improved ablation efficiency (photons per mass), and further improvements in the spectrometer. Finding a low-noise, gate-able detector as a replacement for the ICCD also would improve results by eliminating the so-called excess noise from the stochastic electron-multiplication process inside the intensifier, regaining the ICCD-induced loss in spectral resolution, and benefiting from a higher quantum efficiency.
Besides uranium, what other elements or isotopes could benefit from this LIBS methodology, and what adaptations would be necessary?
One of the most remarkable features of this approach is that it allows any element in the periodic table with multiple isotopes to be a candidate. We’ve already demonstrated, using LIBS (and LAMIS), the isotopic analysis of H, Sr, Cl, B, O, and others from the optical emission spectra. At this point, we firmly believe that, if a similar approach is adopted for each element or isotope system, involving characterization, selection of the emission line or band of interest, optimization of experimental conditions, and establishment of a data processing strategy, a significant advancement in isotopic analysis by optical emission can be achieved.
How do you envision scaling this method for field deployment in terms of instrument size, robustness, and user expertise?
This is a straightforward engineering approach – once the science is established, we either utilize commercial off-the-shelf (COTS) components or, in some cases, develop specialized components similar to the spectrometer developed for uranium in this work.
What are the main limitations of LIBS for isotopic analysis, and how might future research overcome these?
As mentioned earlier, the main hurdles are sensitivity, spectral resolution, and availability of isotope-specific spectroscopic constants (for example, isotopic shifts, hyperfine structure constants). Generally, most elements do not show large isotopic splitting, so high-resolution spectrometers are needed. Again, depending on the type and total amount of samples available for analysis, the sensitivity required for the application may be covered by a COTS spectrometer that offers the required spectral resolution and light throughput. However, it is essential to note that for field deployment, these COTS spectrometers generally have a large footprint and are somewhat delicate, which limits their portability and suitability for harsh conditions. The size of the spectrometer typically dictates the size of the instrument.
What are your thoughts on this paper receiving the 2024 Spectrochimica Acta Part B Best Paper Award?
The entire ASI-LBNL team is grateful to have achieved this recognition. We hope that such acknowledgment will generate further interest in the use of LIBS for isotopic analysis, both in the laboratory and in the field, as well as for inline applications. Isotopic analysis is currently not considered part of planetary applications, highlighting the need for greater recognition of isotopic analysis using optical spectroscopy, which began before mass spectrometry (MS) but has not received nearly the attention and funding it deserves over the years. I want to thank George Chan and Jhanis Gonzalez for contributing to all the comments.
References
1. Russo, R. E.; González, J. J.; Oropeza, D.; Liu, C.; Chirinos, J.; Chan, G. C.-Y. Chan. Isotopic Analysis of Uranium at the Picogram Level with Single-Shot Laser Induced Breakdown Spectroscopy. Spectrochim. Acta Part B: At. Spectrosc. 2024, 216,106928. DOI: 10.1016/j.sab.2024.106928
