Short Communication on “Direct compositional quantification of (U-Th)O2 - MOX nuclear fuel using ns-UV-LIBS and chemometric regression models”
Introduction
Majority of world's nuclear reactors use uranium (U) based fuels (Natural to ∼10% 235U). However, in India, due to scarcity of high quality U ore, there has been a long interest in using thorium (Th) as a fertile element to produce 233U fissile isotope, which has the best neutronic property in the thermal neutron energy region and is a probable replacement of natural U in the Indian nuclear reactors. One-third of world's Th reserves is in India [1], [2]. Based on that, the Indian nuclear program is planned in three stages. The third stage is planned upon the 233U-Th fuel after generating 233U from Th placed around the nuclear reactor core as a blanket to capture escaped fission neutrons in Fast Breeder Reactors which use (U-Pu)O2 fuel in the 2nd stage. The 1st stage is the production of nuclear power from natural uranium and generation of 239Pu, which is used in the 2nd stage as fuel [3]. While the initiation of the third stage will take place in the future, for the timely development of thorium-based technologies for the entire thorium fuel cycle, an Advanced Heavy Water Reactor (AHWR) is being developed which will use (Th-233U)O2 fuel of different composition varying from 3 to 3.75% for AHWR and 18–22.5% in AHWR300-LEU (low enriched uranium). In a AHWR reactor the composition of fuel (U wt.%) varied depending on the position of fuel rod in reactor, i.e., whether it is a centre circle fuel or peripheral fuel rod.
Accurate elemental composition analysis in nuclear materials is an important step for quality assurance at different stages of fuel fabrication from the point of view of the neutron economy inside a reactor. Presently, chemical and electrochemical methods are used for elemental composition analysis of (U-Th)O2 fuel samples. Both these mentioned method have two common drawbacks – dissolution of sample and generation of nuclear waste. Thoria being a refractory material, the dissolution of sintered (U-Th)O2 pellets is difficult and time consuming. In this respect, LIBS have huge potential for U determination in (U-Th)O2 fuel pellets, especially in view of hazardous environment and remote application required in the nuclear industry. LIBS has many advantages over other conventional direct sample analysis techniques such as on-site analysis, pseudo non-destructive, noncontact, minimal to no sample preparation, minimum sample requirements, and capability to measure all elements of the periodic table. ICP-OES/MS methods may be used for this study, but required huge dilution after sample dissolution and hence instead of major matrix elements ICP-OES/MS method is used for trace elemental study after bulk separation. Sarkar et al. demonstrated the applicability of laser induced breakdown spectroscopy (LIBS) for U-Th MOX samples using uni-variative calibration with ∼10% accuracy and precision [4]. One drawback of the proposed method was among the emission lines used in the study, only U(II) 263.553 nm can be used for analyzing a maximum of 32% U composition in MOX sample, for the remaining lines the limit is 18–20% U. These limits make the reported method borderline for analyzing AHWR300-LEU fuel samples. Moreover, for a technique to be utilized by the nuclear industry for quality assurance, 0.5–1% is required for acceptance as a routine technique and <5% when “go/no go” type answers are required. Among the different methods reported in the literature for improving analytical performance of LIBS, the use of UV-wavelengths for ablation [5], [6], [7] and multivariate regression calibration like partial least square regression (PLSR) or principle component regression (PCR) has shown significant potential [8], [9], [10], [11], [12], [13].
In the present work, we measured LIBS spectra at UV (266 nm), Vis (532 nm) and IR (1064 nm) laser along with two different multivariate regression models (PLSR and ASD-PLSR) to obtain U quantification in MOX sample with ≤3% precision which will fulfill the “go/no go” type answer in quality assurance laboratory.
Section snippets
System, samples and analysis
A common LIBS configuration described elsewhere was used for present study [14]. Briefly, a Nd:YAG laser with 1064 nm, 532 nm and 266 nm wavelength was focused using a 1 inch diameter plano-convex lens (f = 10 cm) to produce a laser induced plasma. The spot size of laser ablation was 150 μm. The emission light was collected through a collimator, (CC52, Andor, UK), placed at 6 cm from the plasma and fed to an echelle spectrometer (Mechelle, ME5000, Andor, UK) through an optical fiber of 200 μm
Experimental conditions optimization
The maximum signal to noise (SNR) ratio was used as the figure of merit to optimize the laser energy (EL) and acquisition time delay (td). SNR was calculated by varying the td at particular EL values using the CS sample with 35 wt% U composition. U-Th MOX sample's emission spectra being very complex; the large number of lines makes the spectra appear like a band of emission lines. It is difficult to select pure emission lines (no spectral interferences) for U. To identify and confirm the
Conclusion
We demonstrated the capability of LIBS for rapid determination of U in MOX fuel samples with analytical performance acceptable to a nuclear quality control lab. The effect of different pre-treatment procedures was studied on raw data before PLSR. Maximum intensity normalization was found to give better analytical results for U determination using PLSR. Under optimum analytical condition, PLSR was utilized to study the effect of laser wavelength for U determination in MOX sample. Although good
Acknowledgements
The authors (MS and AS) are thankful to Dr. S. Kannan, Head, Fuel Chemistry Division, Prof. B.S. Tomar, Director, Radiochemistry and Isotope Group, B.A.R.C. for their support and encouragement of the LIBS work. The authors (RER and XM) acknowledge support from the Office of Basic Energy Sciences, Chemical Sciences Division of the U.S. Department of Energy under contract number DE-AC02-05CH11231 at the Lawrence Berkeley National Laboratory.
References (18)
- et al.
Laser-induced breakdown spectroscopy for determination of uranium in thorium-uranium mixed oxide fuel materials
Talanta
(2009) - et al.
Wavelength dependence on the elemental analysis of glass by laser induced breakdown spectroscopy
Spectrochim. Acta Part B At. Spectrosc.
(2008) - et al.
Quantitative micro-analysis by laser-induced breakdown spectroscopy: a review of the experimental approaches
Spectrochim. Acta Part B At. Spectrosc.
(2002) - et al.
Wavelength dependence in the analysis of carbon content in coal by nanosecond 266 nm and 1064 nm laser induced breakdown spectroscopy
Plasma Sci. Technol.
(2015) - et al.
Multiple emission line analysis for improved isotopic determination of uranium — a computer simulation study
Spectrochim. Acta Part B At. Spectrosc.
(2013) - et al.
Advancing the analytical capabilities of laser ablation molecular isotopic spectrometry for boron isotopic analysis
Spectrochim. Acta Part B At. Spectrosc.
(2014) - et al.
Laser ablation molecular isotopic spectrometry of water for 1D2/1H1 ratio analysis
Spectrochim. Acta Part B At. Spectrosc.
(2013) - et al.
Exploring laser-induced breakdown spectroscopy for nuclear materials analysis and in-situ applications
Spectrochim. Acta Part B At. Spectrosc.
(2012) - et al.
Evaluation of the prediction precision capability of partial least squares regression approach for analysis of high alloy steel by laser induced breakdown spectroscopy
Spectrochim. Acta Part B At. Spectrosc.
(2015)
Cited by (13)
Raw signal improvement using beam shaping plasma modulation for uranium detection in ore using laser-induced breakdown spectroscopy
2022, Analytica Chimica ActaCitation Excerpt :LIBS has been successfully applied to industrial analysis [10–12], space exploration [13], and nuclear engineering [14,15], and its feasibility in uranium detection was first demonstrated by Wachter and Cremers in 1987 [16]. In recent years, many studies focus on the evolution of uranium plasma [17–19], isotope analysis [20–22] and the limit of detection (LOD) [23–35] or quantitative performance [36–38] of uranium in different matrixes. Harilal et al. [17] investigated the spatial-temporal evolution of uranium plasma under different ambient pressure conditions, revealing that the lifetime of uranium species would grow with the increase of ambient pressure.
Determination of the uranium elemental concentration in molten salt fuel using laser-induced breakdown spectroscopy with partial least squares-artificial neural network hybrid models
2022, Spectrochimica Acta - Part B Atomic SpectroscopyCitation Excerpt :Several applications of LIBS in nuclear fuel analysis are documented in the literature. Our group has shown the application of LIBS in elemental composition analysis of (U,Th)O2 samples using both the traditional calibration-based approach [13] and with chemometrics approach [14]. The Partial Least Square (PLS) chemometrics method produces results with <5% precision against <10% in calibration-based approach.
Uranium corrosion characterization by handheld laser-induced breakdown spectroscopy
2021, Spectrochimica Acta - Part B Atomic SpectroscopyCitation Excerpt :Fiber optic based LIBS systems could potentially be adapted for glovebox use [13], however our preference when working with radioactive samples is to deploy an instrument wholly contained within the glovebox. LIBS has been used for the direct analysis of uranium samples including uranium ores [14,15], uranium oxides [16,17], uranyl fluoride [18], uranium in surrogate nuclear debris [19], and mixed actinide fuel containing uranium and plutonium [20]. Portable and handheld LIBS systems have been demonstrated [21] and commercialized within the last 5 years.
Laser ablation spectrometry for studies of uranium plasmas, reactor monitoring, and spent fuel safety
2021, Spectrochimica Acta - Part B Atomic SpectroscopyCitation Excerpt :It should be noted that the plasmas of high-Z elements such as U, Pu, or Th will, in most scenarios, result in highly congested emission spectrum and difficulties in its interpretation. Chemometrics, based on multivariate statistical analysis, offers an alternative approach that can significantly improve the analytical performance of LIBS in nuclear applications [81]. The recent focus in the literature on improvement of the measured intensity and selectivity of emission lines of interest is well reflected in a thorough analysis by Chan et al. in [71], which showed that the largest improvement in the effectiveness of isotopic analysis using laser ablation spectroscopy based techniques is through enhancement of the measured spectral emission intensity and SBR rather than searching for elusive emission lines or bands with very well resolvable isotopic splitting.
Laser-induced breakdown spectroscopy of uranium in the vacuum ultraviolet range
2020, Spectrochimica Acta - Part B Atomic SpectroscopyCitation Excerpt :LIBS is thoroughly described in other papers [7–9] and features some unique perks that make up for its relatively low sensitivity: thanks to its rapidity, compactness, and ability to be used remotely and to detect all elements in one measurement, LIBS can be especially relevant for in situ or online measurements of nuclear materials. In fact, it has been used by several teams as a mean of achieving quality control, both for the production of nuclear fuel [10,11] and for its recycling [12], for studies on uranium corrosion [13], and is being considered by some authors as a promising method for nuclear safeguards and forensics [5,6,14–16]. Although the typical detection system in LIBS operates in the UV–visible range, LIBS was also developed in the VUV range to analyze metalloids and non-metals, which have intense emission lines below 200 nm, in order to improve their detection limit.