Measurement of niobium reaction rate for material surveillance tests in fast reactors
2021-01-28ChikaraItoShigetakaMaedaToshihikoInoueHidekiTomitaTetsuoIguchi
Chikara Ito,Shigetaka Maeda,Toshihiko Inoue,Hideki Tomita,Tetsuo Iguchi
(1.Fuels and Materials Department,Japan Atomic Energy Agency,Oarai,Ibaraki 311-1393,Japan;2.Experimental Fast Reactor Department,Japan Atomic Energy Agency,Oarai,Ibaraki 311-1393,Japan;3.Department of Energy Engineering,Nagoya University,Nagoya 464-8603,Japan)
Abstract:A highly accurate and precise technique for measurement of the 93Nb(n,n’)93mNb reaction rate was established for the material surveillance tests,etc.in fast reactors.The self-absorption effect on the measurement of the characteristic X-rays emitted by 93mNb was decreased by the dissolution and evaporation to dryness of niobium dosimeter.A highly precise count of the number of 93Nb atoms was obtained by measuring the niobium solution concentration using inductively coupled plasma mass spectrometry.X-rays of 93mNb were measured accurately by means of comparing the X-ray intensity of irradiated niobium solution with that of the solution in which stable 93Nb was added.The difference between both intensities indicates the effect of 182Ta,which is generated from an impurity tantalum,and the intensity of X-rays from 93mNb was evaluated.Measurement error of the 93Nb(n,n’)93mNb reaction rate was reduced to be less than 4%,which was equivalent to the other reaction rate errors of dosimeters used for Joyo dosimetry.In addition,an advanced technique using Resonance Ionization Mass Spectrometry was proposed for the precise measurement of 93mNb yield,and 93mNb will be resonance-ionized selectively by discriminating the hyperfine splitting of the atomic energy levels between 93Nb and 93mNb at high resolution.
Key words:Isord-10;niobium;reaction rate;material surveillance test;fast reactor;characteristic x-rays;experimental fast reactor joyo;resonance ionization mass spectrometry;hyperfine structure
1 Introduction
Because neutron fluence and spectrum are key parameters in various irradiation tests and material surveillance tests,they must be evaluated accurately.The reactor dosimetry test was performed by the multiple foil activation method in the experimental fast reactor Joyo of the Japan Atomic Energy Agency[1].A niobium dosimeter[2]which is an activation foil for Joyo dosimetry has been applied to fast neutron dosimetry.
The inelastic scattering reaction of93Nb has a sensitive neutron energy range from 105to 106eV,and the energy distribution of the reaction cross section[3]is similar to that of the displacement cross section[4]of iron.Therefore,a niobium dosimeter is suitable for evaluation of the fast neutron fluence and the displacement per atom for iron.Moreover,a niobium dosimeter has an advantage for long-term irradiation tests because93 mNb,which is produced by the inelastic scattering reaction,has a long half-life (16.4 y).
This study established a highly accurate technique for measurement of the niobium reaction rate.The self-absorption effect was decreased by the dissolution and evaporation to dryness of the niobium dosimeter.Dosimeter weight was measured precisely by means of mass spectrometry after irradiation.Regarding interfering fluorescence due to182Ta produced by impurities in the niobium dosimeter,the fluorescence effect of182Ta was estimated quantitatively by means of an experiment in which the X-ray intensity of the solution was compared with that of the solution in which stable93Nb was added.
This technique was applied to the Joyo dosimetry.The93Nb(n,n’)93 mNb reaction rates were measured at the neutron field in the Joyo irradiation core and compared to the reaction rates calculated from the neutron flux and spectra evaluated by the multiple foil activation method.
2 Experimental method
2.1 Measurement of 93mNb activity in niobium dosimeter
93mNb emits low energy photons and characteristic X-rays of 16.6 keV (Kα) and 18.6 keV (Kβ).The measurement of these low energy X-rays is not accurate because of the self-absorption effect.The intensities of X-rays from a niobium foil of 2.5 μm thickness,as used for the Joyo dosimetry,were reduced by 20% to 30%.
The pretreatment procedure of X-ray measurement is illustrated in Figure 1.An irradiated niobium dosimeter of less than 1 mg was dissolved in a mixed solution of hydrofluoric and nitric acids.After diluting the niobium solution with purified water to a concentration of approximately 10 parts-per-million (ppm),the solution of 100 μg was dropped onto a plastic film and was dried using an electric heater.
Fig.1 X-ray source preparation procedure for 93mNb X-ray measurement
The self-absorption effect was decreased by the above procedure.The intensities of Kαand KβX-rays emitted by93mNb were measured by means of X-ray spectroscopy using a high purity germanium semi-conductor detector (ORTEC,Lo-AX 51370/20-P),as shown in Figure 2.The93mNb activity was determined from the counted X-rays intensity,the K X-ray emission rate and the detection efficiency.
Fig.2 Pulse height spectrum of X-rays and γ-rays emitted by a prepared source
2.2 Measuring the number of 93Nb atoms in niobium dosimeter
The number of93Nb atoms is not normally determined by measuring the dosimeter’s weight because of embrittlement of thin niobium foil under neutron irradiation[5].Instead,the concentration of93Nb in the solution was measured by means of mass spectrometry using an inductively coupled plasma mass spectrometer (PerkinElmer,ELAN 6100),which can measure the m/z values of trace elements in liquids with high precision.A curve was obtained by measuring the concentration of the solution with calibrated93Nb using the internal standard method by adding calibrated zirconium to the niobium dosimeter solution.The above procedure is applicable to the measurement of niobium impurity in reactor component structure materials.Therefore,retrospective estimation of the fast neutron fluence of reactor structure components is possible by measuring the93mNb concentration.
2.3 Quantitative evaluation of 182Ta induced fluorescence effect
A niobium dosimeter is made of high purity niobium wire or foil containing tantalum at 102ppm because niobium coexists with tantalum in the Columbite-Tantalite ((Fe,Mn)(Ta,Nb)2O6),from which the niobium dosimeter is made.The γ-rays emitted by182Ta (half-life:114 d),originating from the tantalum impurity,excite93Nb,and the excited93Nb emits characteristic Kαand KβX-rays in a manner similar to93mNb.
To evaluate the effect of182Ta induced fluorescence,the intensities of X-ray peaks from the dissolved irradiated niobium dosimeter were compared to those obtained after addition of stable93Nb to that solution.
The resulting X-ray spectra are shown in Figure 3.The difference in intensities (B-A,in Figure 3) indicates182Ta impurity’s effect on added93Nb.Therefore,93mNb X-rays were measured accurately after the difference in intensities became negligible.
Fig.3 Pulse height spectra of X-rays and γ-rays emitted by a prepared source before and after addition of 93Nb
Moreover,93mNb activities can be evaluated without waiting until the end of irradiation.Figure 4 shows the relation between the mass of93Nb and KαX-ray intensity.The intercept on the vertical axis (mass of93Nb=0 in Figure 4) of the extrapolation line between two arbitrary points represents only the X-ray intensity emitted by93mNb generated by neutron irradiation.
Fig.4 Estimation of 93mNb activity using the relation of 93Nb mass and Kα X-ray intensity
2.4 Calculating the reaction rate of 93Nb(n,n’)93mNb
93mNb activities were calculated from measured X-ray intensities,X-ray emission probabilities and the detection efficiency.The reaction rate of93Nb(n,n’)93mNb was calculated using93mNb activities,the number of93Nb atoms in the dosimeter and the reactor operation data.Tables 1 and 2 show,respectively,the evaluation of experimental errors in the measured93mNb activity and the number of93Nb atoms.Both93mNb activity and the number of93Nb atoms were measured with an uncertainty of less than 3%.Consequently,the experimental error of the reaction rate is within 4%.
Tab.1 Experimental error in the measurement of 93mNb activity
Tab.2 Experimental error in the measurement of the number of 93Nb atoms
3 Measuring reaction rate of neutron irradiation field in fast reactor
3.1 Irradiation condition
In the tests,irradiation test subassemblies with niobium foils of 2.5 μm in thickness were loaded and irradiated in the fuel region (core center),the reflector region (9th row) and the in-vessel storage rack of the Joyo irradiation core (i.e.,the Mark-II core)[6].The cross-sectional view of Joyo core and the positions where niobium dosimeters were loaded are illustrated in Figure 5.
Fig.5 Niobium dosimeter irradiation positions in Joyo neutron field
3.2 Neutron flux and spectra measurements by foil activation method
Determination of the neutron flux and spectra at the irradiation positions of niobium dosimeters was based on the foil activation method[1],which used a dosimeter set consisting of iron,nickel,copper,titanium,cobalt,tantalum,scandium,neptunium (237Np) and uranium (235U and238U) to cover the neutron spectrum over the energy range from 0.1 eV to 20 MeV.Measured reaction rates of the dosimeters and the neutron reaction cross sections of the JENDL/D-99 library[3]were used for adjustment of the initial guess neutron spectrum calculated from the two-dimensional transport code DORT.The adjusted neutron spectra by the foil activation method are shown in Figure 6.The evaluated fast neutron fluence (greater than 0.1 MeV) was 1.55×1024n/m2~2.29×1026n/m2.
The93Nb(n,n’)93mNb reaction rate was calculated using the cross-section JENDL/D-99 library and neutron flux obtained by the above multiple foil activation method.
Fig.6 Neutron spectra at niobium dosimeter irradiation positions in Joyo neutron fields
3.3 93Nb(n,n’)93mNb reaction rate measurements
The results for the93Nb(n,n’)93mNb reaction rate measured by niobium dosimeters were compared with those obtained by the multiple foil activation method.The reaction rates by both methods are shown in Table 3.The ratios of reaction rates measured from niobium dosimeters to those obtained by the multiple foil activation method range from 0.97 to 1.03 and were in good agreement within the experimental uncertainty of approximately 4%.
By means of the technique noted above,the measurement error of the93Nb(n,n’)93mNb reaction rate was reduced to less than 4%,which was equivalent to the other reaction rate errors of dosimeters used for the Joyo dosimetry.
Tab.3 93Nb(n,n’)93mNb reaction rate measurement results
4 Resonance Ionization Mass Spectrometry proposal
Decay of182Ta is required to measure the characteristic X-rays of93mNb.An advanced technique by Resonance Ionization Mass Spectrometry (RIMS) was proposed for the precise measurement of93mNb yield.In the method,93mNb is resonance-ionized selectively by discriminating the hyperfine splitting of the atomic energy levels between93Nb and93mNb at high resolution.
The high resolution resonance ionization spectroscopy method by high repetition rate Ti:Sapphire laser system was developed for the precise measurement of93mNb yield.High resolution resonance ionization spectroscopy of stable niobium was demonstrated by means of the present laser system,in which hyperfine splitting was resolved clearly[7].Isotope analysis by high resolution RIMS might be applied to the transmutation detectors used for neutron fluence measurement[8].
The efficiency of the measurement will be verified experimentally with niobium dosimeters irradiated in Joyo.
5 Conclusion
This study established a high precision measurement technique for the93Nb(n,n’)93mNb reaction rate.The self-absorption effect was decreased by the dissolution and evaporation to dryness of the niobium dosimeter.It was confirmed that the93Nb(n,n’)93mNb reaction rate could be measured with an uncertainty of within 4% by application of the established method to fast neutron dosimetry in the Joyo irradiation field.
Acknowledgements:The authors would like to thank Mr.K.Sukegawa,T.Masui and T.Saikawa of Inspection Development Co.,Ltd.for their technical assistance during the experiments.
A part of this work was supported by JSPS KAKENHI Grant Number 26420868.