Research Article
Discrepancy in the k0-values of 134Cs, 152Sm, 75Se and Experimental Implementations in k0 Standardization Techniques
Department of Physics, Federal University Dutse, Jigawa State, Nigeria
LiveDNA: 237.25022
The Nigeria Research Reactor-1 (NIRR-1) is a Miniature Neutron Source Reactor (MNSR) and a low power, which was designed to have one central control rod (CCR) that performs safety and regulatory functions and to serve as a neutron source1. The CCR are strong neutron absorbers that can be inserted or withdrawn from the reactor core. They are used to compensate for the excess reactivity necessary for long-term core operation and to adjust the power level of the reactor in order to bring the core to power, follow load demands and shutdown the reactor2. Jonah et al.2,3 have standardized NIRR-1, via the relative and the k0-standardization method for the use in Neutron Activation Analysis (NAA).
The implementation of the k0-standardization concept in NAA involved the co-irradiation of the sample and a neutron fluence rate monitor to determine a composite nuclear constant4,5, which eliminates the use of standards. When the composite nuclear constant is obtain accurately, these are used to evaluate the concentration of nuclides6. In the implementation of the k0-standardization concept, the k-values, epithermal neutron flux shape factor (α) and sub-cadmium-to-epithermal neutron flux ratio (ƒ) are important parameters used7,8. These parameters are dependent on each irradiation facility and must be determined for standardization and application purposes. In order to extend its utilization to include the k0-standardization method, the following neutron spectrum parameters in inner and outer irradiation channels were determined by the Cd-ratio for multi-monitor method and published elsewhere3. The results were compared with the neutron spectrum parameters of other reactor facilities with similar core configuration such as the Slowpoke and Miniature Neutron Source Reactor facilities available in the literature of Kennedy et al.4.
In the efforts to implement the k0-standardization concept in NAA techniques in many laboratories worldwide, the problem of how accurate the k0-values available in literature were in order to produce accurate results9. De Corte and Simonits10 observed that some of the recommended k0-values available in the Atomic and Nuclear Data Tables have discrepancies and Moens et al.5, Simonits et al.11 and De Corte and Simonits10 revealed that the k0-values in the Atlas of Neutron Resonances are inconsistent with the recommended data. In line with these observation, this study seek to determine the k0-values of the following nuclides; 134Cs,152Sm and75Se reported to be discrepant. The Synthetic Multi-element Standard (SMELS) recommended by the k0-users for the validation of k0-NAA method and the standard reference materials, NIST-1633b (Coal Fly Ash) were used12,13.
The k0 values of the discrepant nuclides; 134Cs, 152Sm and 75Se have been obtained experimentally in this study using the NIRR-1 facilities. The experimental k0 values results obtained in this study were in good agreement with the recommended values, except for 152Sm whose deviation was 28%. Furthermore, there was a large deviation in the range of 18.29-39.53% from the results of this study compare to the empirical calculated values for all the nuclides and their respective peak energies. Hence, this study revealed the unreliability of the nuclear data and observed that the experimental determination of the k0 values is necessary in order to use the data in k0-standardization method.
The Synthetic Multi-elemental Standards (SMELS) Type I and III was weighed and an amount was measured based on a phenol formaldehyde resin (Bakelite). The measured samples were spiked with 30 different materials as well as the NIST 1633b (Coal Fly Ash). The samples were wrapped in polyethylene films that have been cleaned appropriately and prepared for irradiation. The irradiation were performed using NIRR-1 operating at half-full power of about 15 kW with neutron flux 5×1011n/cm2 sec was performed in 201712,2.
The samples were irradiated for 7 h in the inner irradiation channel B-2. For the determination of 152Sm, the samples were left for two days after irradiation before counting. Each of the samples were counted for 30 min using GEM-30195 HPGe Coaxial, vertical dipstick detector (ORTEC). This similar procedures is applied in the determination of the following elements; Na, K, As, Zn, Sb, Dy Br, Mo, La, Au, W, Ho, U, Ga, Lu, Ba, Yb.
For the determination of 134Cs and 75Se, the irradiated samples were kept for 15 days before measurement using the GEM-30195 HPGe. Each of the samples were counted for 60 min. This procedure is applied also in the determination of the elements; Sc, Ce, Co, Cr, Eu, Gd, Lu, Ba, Mo, Nd, Rb, Sb, Ta, Tb, Th, Yb, Zn, Cd, Fe, Sr, Ag, Hf, Ir, Hg, Zr, Te, Os.
Based on the regime of irradiation, all the irradiated induced activities were measured using the GEM-30195 HPGe Coaxial, vertical dipstick detector (ORTEC). The relative efficiency of the detector is 30% and the resolution is 1.95 Kev at 1.33 MeV for 60Co. The gamma-ray acquisition system employed in this study consists of the MAESTRO emulation software compatible with Multi-channel Analyzer (MCA). This software was used for the peak identifications and evaluations14,15. The efficiency of the GEM-30195 HPGe Coaxial detector as a function of source-detector geometries was determined using standard gamma ray sources16, in order to determine the k0-values. The full energy peak efficiency at 1 and 5 cm geometry of the HPGe detector was determined using gamma ray sources and an efficiency curve at the respective geometries were fitted to obtain the efficiencies of the discrepant nuclides.
The specific activity (Asp) for the activation product of the element of interest was obtained using Eq. 1:
(1) |
= | Saturated factor with tirr = Irradiation time | |
D = e-λtd | = | Decay factor with td decay time |
C | = | 1-e-λtc Counting time correction with tc = counting time |
= | Decay constant | |
Np | = | Net peak area |
W | = | Weight of the measured samples |
The equation for concentration calculation based on the k0-standardization method by the use of Høgdahl convention for the so-called 1/v nuclides needs the neutron spectrum parameters17, which are the shape factor of the epithermal neutron flux (α), approximated by a 1/E1+α distribution and the thermal-to-epithermal neutron flux ratio (ƒ). Several author including De Corte et al.13,18, Jonah et al.2 and Rossbach and Blaauw7 had enumerated the means of obtaining these neutron spectrum parameters based on the experimental methods which were required in both the Høgdahl convention and the Wescott formalism. However, with NIRR-1 and other similar reactors with stable neutron flux characteristics, it had been recommended that the Cd-ratio for multi-monitor method be performed for neutron spectrum monitoring. Moreover, this method had been observed as the most accurate for α-monitoring3,19.
In this investigation, the neutron spectrum parameters (i.e., f and α) in the inner irradiation channel (B2) of NIRR-1 for NAA applications had been determined using the k0-standardized method using Cu foil as the monitor and results published3.
The experimental k0-values was determined by imputing the evaluated specific activity using Eq. 2 as written in the Hogdahl formalism:
(2) |
is the ratio of resonance integral to thermal cross-section corrected for non-ideal epithermal neutron flux:
is the thermal to epithermal flux ratio and ε is the full energy peak detector efficiency. The determination of the resonance integral I0(α,) for the non-ideal epithermal neutron flux have been done elsewhere2.
The values of the thermal and epithermal fluxes were obtained using Eq. 3 and 4:
(3) |
(4) |
θ | = | Isotopic abundance |
c | = | Concentration of foil monitor (Cu) |
Iλ | = | Gamma abundance |
w | = | Weight of foil monitor |
= | Average cross section for fast neutron | |
M | = | Atomic weight |
NA | = | Avogadro’s number |
Np | = | Net peak area of the foil monitor |
The empirical calculated k0-values of each of the nuclides of interest were determined using the Eq. 5:
(5) |
σth | = | Thermal absorption cross section for neutron (velocity 2200 m sec1) |
γ | = | Gamma-ray intensity |
The values of these parameters were obtained from literature Glascock20 and Blaauw21.
The k0-values of the following discrepant nuclides; 134Cs, 152Sm and 75Se have been re-measured (experimentaly) using Certified Reference Materials; Synthetic Multi-element Standard (SMELS) I and III, NIST 1633b Coal fly Ash. The experimental deviation (%) and empirical deviation (%) from the recommended values of De Corte and Simonits10 were presented in Table 1. As shown, the sources of the uncertainties in the results of both the recommended and experimental results may be due to peak analysis or from determination of neutron flux parameters f and α which are used in the evaluation process.
From the results in Table 1, the k0 value of 152Sm obtained in this study experimentally showed the highest percentage deviation of 28.14% compared with the recommended values of De Cort and Simonits10. The 134Cs k0 value at the peak energy of 802 keV showed the second highest deviation of 4.38% compared with recommended values. The other peak lines of 134Cs (562, 569 and 602 keV) showed slight deviation of 3.38, 0.27 and 1.89%, respectively.
The percentage deviation of the nuclides as a function of their respective peak lines for each of the studied nuclides was shown in Table 1. Table 1 showed how the experimental values and the empirical calculated values deviated (%) from the recommended values. As indicated in Table 1, the deviation of the empirical calculated from the recommended were as follows; 152Sm at 103 keV (39.52%), 75Se at 136 and 401.7 keV were 24.55 and 18.29% and for 134Cs at 562, 569, 602 and 802 keV were 18.34, 21.16, 19.19 and 22.16%, respectively.
The distribution of the k0 values as obtained experimentally in this study9 and including the empirical calculated and the recommended k0 values10 were shown in Fig. 1. As observed, the deviation between the recommended, experimental and the calculated k0 values was large for all the peak lines of 134Cs and 152Sm. Moens et al.5 also reported this similar observation of large deviations that appeared in the comparison of the calculated k0-values with recommended k0-values. According to Moens et al.5, it revealed the unreliability of the nuclear data. Moens et al.5 observed that the experimental determination of the k0 values should be done in order for the data to be use in k0- standardization computations.
Table 1: | Empirical calculation and measured k0-values compared with the recommended values and the measured k0-values |
Source: De Corte and Simonits10 and St-Pierre and Kennedy9 |
Fig. 1: | k0 value distributions |
Source: De Corte and Simonits10 and St-Pierre and Kennedy9 |
This was in agreement with this study as it was cleared that the discrepancy between the calculated k0 values and the recommended values were very high.
The distribution of the k0 values between the recommended and the experimental k0 values have been plotted as shown in Fig. 2 to show how many of the data fully agreed with each other.
As revealed in Fig. 2, the results of this work compared well with the recommended data for most of the nuclides except for 152Sm whose recommended k0-value of 2.31×101. The experimental value obtained was 1.66×101, which represent a deviation of 28% as seen in Table 1. The reason may be due to the discrepant nuclear data used in the calculation of k0-value for this nuclide.
The relationship between the experimental and the empirical calculated k0 values and how far it deviated (%) from the recommended values was illustrated in Fig. 3. As seen in Fig. 3, the deviation from the empirical calculated values were high for all the nuclides in this studies. This showed that much effort should be emphasized on the experimental determination of the k0 values before using the data in the k0 standardization method. It also revealed that using the empirical calculation will lead to a large discrepancy in the determination of elemental concentrations.
As revealed in this study, with the exception of 152Sm all the studied nuclides had good agreement with the recommended values. In comparison between this study and that measured k0 values of 134Cs for all the peak lines were in good agreement. It was observed that the uncertainties of the k0 values of 134Cs for all the peak lines were higher compared to this study except for gamma line of 602 and 802 keV (134Cs). The reason for the difference may be coming from the type of detector used in each of the cases. Hence, the results of this work showed very good agreement with the recommended data and the measured value. This showed the accuracy of the adopted irradiation and analytical protocols employed in the laboratory for the determination of elemental concentrations and facilities used for this work.
Fig. 2: | A plot of recommended k0 values and the experimental values |
Fig. 3: | Deviation (%) distribution of the experimental values from the empirical calculated k0-values of each of the nuclides of interest |
It is concluded that experimental determination should be performed rather than empirical calculations for these discrepant nuclides before applying the data in the k0 standardization technique. The data obtained in this study will aid in elemental determination of concentrations and the established protocols will be applied in the determination of other nuclides; Sc, Ce, Co, Cr, Eu, Gd, Lu, Ba, Mo, Nd, Rb, Sb, Ta, Tb, Th, Yb, Zn, Cd, Fe, Sr, Ag, Hf, Ir, Hg, Zr, Te, Os in author’s laboratory.
This study revealed the discrepancy observed in the nuclear data libraries used in determination of elemental concentration using k0 standardization methods. This study uncover critical areas and shows high discrepancy using empirical calculation of the k0 values employed in this field. Thus, this new development emphasis the implementation of experimental determination of the k0 values for the standardization methods.