Technical developments for accurate determination of amount of samples used for TOF measurements

Activity determination of 241,243Am samples has been performed with two separate methods of calorimetry and gamma-ray spectroscopy. Decay heat measurements of the samples were carried out by using a calorimeter, and activities of the samples were accurately determined with uncertainties less than 0.45%. The primary source of uncertainty in the calorimetric method is the accuracy of available half-life data. Gamma-ray detection efficiencies of a HPGe detector were determined with uncertainties of 1.5% by combining measured efficiencies and Monte Carlo simulation. Activities of the samples were determined with uncertainties less than 2.0% by gamma-ray spectroscopy and were concordant with those of the calorimetry.


Introduction
Accurate neutron capture cross sections of Minor Actinides (MAs) produced in fission reactors are required in order to promote the R&D of the nuclear waste transumutation system for MAs.However, there are gaps between required accuracies and current accuracies on relevant neutron capture cross sections [1].Current accuracies are 6% for the 237 Np capture cross section in the neutron energy region from 0.5 to 500 keV, 8% for 241 Am, and 10% for 243 Am from 0.5 to 1350 keV.On the other hand, required accuracies are 3% for 237 Np, 2% for 241 Am, and 2% for 243 Am.Additionally, discrepancies exist between measured capture cross section data deviating from their uncertainty ranges.To meet the requirement, "Research and development for Accuracy Improvement of neutron nuclear data on Minor ACtinides (AIMAC)" has been started to improve the reliability of the neutron cross section data of MAs by a combination of neutron capture and neutron total cross section measurements.The purpose of the project is to reduce the uncertainties of the capture cross sections of 237 Np and 241,243 Am to half of the present values.
It has been pointed out that uncertainty due to the sample mass is one of the factors contributing to the discrepancies of measured cross section data [2][3][4].For neutron capture and neutron total cross section measurements, the uncertainty due to the sample mass affects normalizations and accuracies for measured cross section data.However, the sample mass is not identified in some cases, resulting in unrecognized systematic uncertainties.Therefore, as a part of the AIMAC project, we have developed the techniques accurately determining the sample mass by two different methods: calorimetric method, and gamma-ray spectroscopic method.As for a e-mail: terada.kazushi@jaea.go.jp the calorimetric method, decay heat of the samples was measured accurately by using a calorimeter.Since Q-values associated with radioactive decays of MAs are well known, activities of the samples can be obtained accurately.In the spectroscopic method, gamma-ray emission probabilities of 241,243 Am, 239 Np have been determined with high precision for accurately quantifying the amounts of the samples, by direct detection of emitted gamma-rays from the samples, with a planar type High-Purity Germanium (HPGe) detector [5].An efficiency curve of the HPGe detector was obtained by combining measured efficiencies and Monte Carlo simulation.Activity measurements of the samples were performed.

Calorimetric method
Americium-241 undergoes alpha-decay with a half-life of 432 y, and 237 Np (half-life of 2.144×10 6 y) is produced.Americium-243 undergoes alpha-decay with a half-life of 7370 y and produces 239 Np, which is in radioactive equilibrium with 243 Am due to a short half-life of 2.35 d [6].Two 241 Am samples with nominal activities of 480 and 950 MBq and three 243 Am samples with nominal activities of 60, 120 and 240 MBq were supplied from Khlopin Radium Institute for neutron capture and neutron total cross section measurements at the Japan Proton Accelerator Research Complex (J-PARC).These Am samples were prepared by compression molding of mixtures of AmO 2 powder and Y 2 O 3 powder into the form of pellets with a diameter of 10 mm and a thickness of 0.5 mm, then the pellets were sealed in Al containers with a thickness of 0.1 mm.Isotopic purities of these 241,243 Am samples were analyzed by thermal ionization mass spectrometry and alpha-ray spectroscopy.The isotopic purities of 241 Am in the 241 Am samples were more than 99.9%, and isotopic contaminations of 239 Pu were 0.09%.On the other hand, isotopic compositions of 243 Am, 241 Am, 244 Cm and 242 Cm in the 243 Am samples were 97.36%, 2.39%, 0.26% and 1.00 × 10 −4 %, respectively.The impact of these impurities is described below.
Decay heat measurements of the 241,243 Am samples have been performed with the TAM-IV calorimeter manufactured by TA Instruments.The precision of the calorimeter was ±100 nW, and baseline stability was ±200 nW/24 h in the certification sheet.For absolute calibration of the calorimeter, joule heat from a resistor of 10 k with an uncertainty of 0.01% was measured while changing a current with a multimeter (DMM7510) having an uncertainty of 1 ppm.As the result, we have confirmed the uncertainty of the heat measurements of 200 nW.Gamma-ray energies emitted from the beta minus decay of 239 Np are relatively high at around 300 keV by comparing with those of 241 Am (60 keV).The 241,243 Am samples were held in the measuring container while sandwiched between two tungsten shielding disks with diameters of 20 mm and thicknesses of 12 mm in order to reduce the leakage of decay gamma-rays from the 241,243 Am samples to the outside of the calorimeter.The heat leakage due to the gamma-rays from the 243 Am samples were less than 0.1% according to calculations by using GEANT4 [7].On the other hand, the gamma-ray energies emitted from the 241 Am samples are low (60 keV), and heat leakage from the 241 Am sample was negligibly.To reduce the influence caused by air compression when the sample is inserted into the calorimeter, measurements of the 241,243 Am samples were performed until the measured heat reached equilibrium.The measuring time for each sample was 12 days at least.Figures 1 and 2 show the heating values from the 241,243 Am samples.Uncertainties of the measured heating values for the 241,243 Am samples were less than 0.26% taking into account fluctuations in measurement values and the baseline stability.
Activity (A cal ) of the sample is defined as follows: where H the is measured heating value, Q is Q-value of radioactive decay for target nuclide, k is correction factor for the impurities in the 241,243 Am samples.Qvalues for alpha-decay of 241 Am and 243 Am are 5637.81± 0.12 keV and 5438.1 ± 0.9 keV, respectively [6].For the 241 Am samples, contributions due to the impurities of 2. Decay heat of the 243 Am samples with nominal activities of 60, 120 and 240 MBq, respectively.
239 Pu were 0.08%.The heat leakage caused by the gammarays of the 241 Am samples was negligibly.The decay heat from the daughter nuclide of 237 Np does not affect due to a long half-life (2.144×10 6 y).On the other hand, impurities of 241 Am, 244 Cm and 242 Cm in the 243 Am samples have a great influence on obtained activities because their half-lives are significantly shorter than that of 243 Am.Correction factor caused by these impurities were 0.4118 as for the 243 Am samples.Finally, activities of the 241,243 Am samples were determined with uncertainties less than 0.45%.The following points were taken into consideration for the uncertainties of the activities of the 241,243 Am samples: (1) half-lives of the target nuclides ( 241 Am: 0.15%, 243 Am: 0.30%); (2) the uncertainties caused by decay heat measurements ( 241 Am: < 0.04%, 243 Am: < 0.26%); (3) the uncertainties due to the Q-values (< 0.02%).The uncertainties of the half-lives of 241,243 Am were the most dominant uncertainty component.

Gamma-ray spectroscopic method
Gamma-rays emitted from the 241,243 Am samples were measured with a planar type low energy photon HPGe detector (GLP-36360/13P4) having dimensions of 36 mm × 13 mm manufactured by ORTEC.The detector was p-type Ge crystal with a Be window with a thickness of 0.254 mm, where the cap-to-crystal distance was 7 mm.The Ge detector was cooled with the X-COOLER III made by ORTEC.The energy resolution at 122 keV ( 57 Co gamma-ray) was 0.6 keV.
Measurements of the 241,243 Am samples were performed with a source-detector distance of 150 cm.Fullenergy peak efficiencies of the detector were determined by a combination of measurements for standard gammaray sources and Monte Carlo simulations using the PHITS code [8].Two individual standard gamma-ray sources ( 60 Co and 137 Cs) supplied by JRIA having activities of 1.0 MBq with uncertainties of 1.5% (1σ ) were used.Measurements of the standard sources were carried out for a long period of time to reduce statistical uncertainties in peak areas.The dead time was less than 0.2%.Peak areas were obtained by simultaneously fitting a Gaussian, a skewed Gaussian and a smoothed step function.The skewed Gaussian shows contribution of incomplete charge collection.The smoothed step function indicates contributions from incomplete charge collection effects and Compton scattering of gammarays in the detector.X-rays from 137 Cs were used to confirm the reliability of the efficiencies at low energies.The efficiencies were determined with the obtained peak areas, the absolute activities of the sources and gammaray emission probabilities of the 60 Co and 137 Cs nuclides.For the standard sources, evaluated data of half-lives and gamma-ray emission probabilities were taken from the International Atomic Energy Agency (IAEA) [9].Contribution from cascade summing was sufficiently small for the measurements made at 150 cm.The efficiency curve was interpolated by PHITS calculations taking into account physical parameters of Ge detector.The details of the PHITS calculations were described in [5], so it is briefly described in the present paper.We used Monte Carlo calculations to determine the parameters of the HPGe detector (radius, length and dead layer thickness ) to match the measured efficiencies, and precise interpolation of the efficiencies was achieved.The PHITS calculations were in good agreement with the measured efficiencies with an uncertainty of 1.5% ranging from 60 to 300 keV.
Measurements of the 241,243 Am samples were performed as well as the standard sources.Dead time of the measurements were less than 6.9%.Areas of 59.54 keV 241 Am peak and 277.60 keV 239 Np peak in the measured gamma-ray spectra were determined by the Gaussian fitting.Statistical uncertainties of the peak areas were less than 0.1%.Activity of the samples is defined as follows by gamma-ray measurements; where Y is the peak area of gamma-rays, is gamma-ray detection efficiency, I γ is gamma-ray emission probability, T is the measuring time.We have preciously measured the gamma-ray emission probabilities of 241,243 Am and 239 Np by combination of gamma-and alpha-ray spectroscopic methods with uncertainties less than 1.2% [5], and the measured gamma-ray emission probabilities of 59.54 keV on 241 Am 35.64 ± 0.46 and 277.60 keV on 239 Np 14.34 ± 0.16 were used for activity determination.Corrections of gamma-ray attenuation caused by the 241,243 Am samples and the air between the samples and the detectors were made.The correction factors were ranging from 1.179 to 1.195 for the 241 Am samples and 1.037 to 1.048 for the 243 Am samples.Finally, activities of the 241,243 Am samples were derived with uncertainties less than 2.0% by taking into consideration the following components: (1) the statistical uncertainties of peak areas for the 241,243 Am measurements (< 0.1%); (2) the uncertainties of gamma-ray detection efficiencies (< 1.5%); (3) the uncertainties of gamma-ray emission probabilities of 241 Am and 239 Np.

Results
The activities of the 241,243 Am samples were determined by two individual methods of calorimetry and gamma-ray spectroscopy.Figures 3 and 4 indicate the obtained results.
It is noted that excellent activity determination became available by using the calorimeter with the uncertainties less than 0.45%.Their results were in good agreements.

Conclusions
We have developed the techniques for accurately determining the sample mass by: a calorimetric method, and a gamma-ray spectroscopic method.Decay heat of the 241,243 Am samples was measured accurately by using the calorimeter with a precision of 200 nW, and thus activities of the samples were obtained accurately with uncertainties less than 0.45%.In gamma-ray spectroscopic method, the efficiency curve of the HPGe detector was derived by combining measured efficiencies and Monte Carlo simulation.Activities of the 241,243 Am samples were obtained with uncertainties less than 2%.The activities of the 241,243 Am samples obtained by the two separate methods were in good agreement, and highly accurate quantification of the 241 Am samples with an uncertainty approaching 0.15%, limited by available half-life data, was enabled by calorimetry.
The author would like to thank the accelerator and technical staff at J-PARC for operation of the accelerator and the neutron production target and for the other experimental supports.
Present study includes the result of "Research and Development for accuracy improvement of neutron nuclear data on minor actinides" entrusted to the Japan Atomic Energy Agency by the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT).

Figure 1 .
Figure 1.Decay heat of the 241 Am samples with nominal activities of 480 and 950 MBq.

Figure 3 .
Figure 3. Activities of the 241 Am samples determined by calorimetry and gamma-ray spectroscopy.

Figure 4 .
Figure 4. Activities of the 243 Am samples determined by calorimetry and gamma-ray spectroscopy.