Evaluation of the delayed neutron data for thermal fission of ²³⁵U is attempted through an indirect method based on the comparison between measured and calculated integral quantities such as kinetic parameters and reactivity worths of fuel rods and burnable poison rods for the critical assembly Semi-Homogeneous Experiment (SHE). The most probable values for both of the effective delayed neutron fraction β_eff f and the decay constants λ's of delayed neutron precursors are determined from the condition that the sum of the squared deviations of the ratios of calculated to measured values from unity is minimized for the integral quantities, The most probable value of β_eff obtained in the present work is significantly large compared with that obtained with the Keepin's delayed neutron data set, and is closer to the value obtained with the ENDF/B-V data file than that with the ENDF/B-IV. On the other hand, the most probable λ values obtained in the present work are not away from those for thermal fission filed in the Keepin's delayed neutron data set.

View full abstract

Measurements on the neutron and γ-ray attenuation in multi-legged air filled ducts have been carried out using a Cockroft-Walton type neutron generator. The measured spectra were obtained with an NE-213 liquid scintillator using pulse shape discrimination to resolve neutron and γ-ray pulse height data and using a spectral unfolding code to convert these data to energy spectra. Results are given for two rectangular duct geometries. A series of integral measurements have also been carried out with a long counter for fast neutrons and a radiation survey meter to measure the γ-ray dose. The biological dose arising due to the streaming of neutrons was measured with a rem counter. The results show the relative decrement of the integral counts along the axis of the duct. The spectral measurements reveal the presence of resonant windows of oxygen present in the concrete medium. Also the 14 MeV component is found to be dominant in the first leg of the duct whereas degraded neutrons dominate in the second leg. The results are compared with MORSE calculations.

View full abstract

The Large Scale Test Facility (LSTF) of the Rig-of-Safety Assessment No. 4 (ROSA-N) Program is a volumetrically scaled (1/48) pressurized water reactor (PWR) system with an electrically heated core used for integral simulation of small break loss-of-coolant accidents (SBLOCAs) and operational transients. Three 2.5% cold-leg SBLOCA experiments were conducted at LSTF. In the experiments, the break was oriented at the side, bottom and top of the horizontal cold leg, respectively. The loop seal clearing in the bottom break case was later than in the side break case since a larger amount of liquid had to be discharged until the loop seal clearing. The loop seal clearing in the top break case was later than in the side break case because of the smaller discharge flow rate. The core liquid level drop due to boiloff after the loop seal clearing in the bottom break case was earliest among three cases because of the largest mass loss before the loop seal clearing and in the top break case latest because of the latest occurrence of the loop seal clearing. However, the effect on the system transients such as the pressure and core liquid level transients was small since the transient time was quite long. Analyses to the experimental results were performed with the RELAP5/MOD2 code. Shortcomings in the RELAP5 code calculation results were resolved by reducing the interfacial drag in the hot leg and the core. The inclusion of Shrock's model for the side, bottom and top break of a large horizontal pipe was also tested in the analyses. However, consistent results with data were not obtained since the void fraction in the broken cold leg was not calculated properly.

View full abstract

Experiments of H₂-HT isotope separation were carried out with a 0.3 mmφ hot wire column of 30 mm in diameter and 1, 500 mm in length. The column was operated at 0.1 MPa, and the temperature of Mo hot wire was regulated by the density of an electric current to be 773 or 988 K. An activity of tritium, whose value lay in the order of nCi/cm³, was measured by 200 cm³ flow-through ionization chambers, which were directly inserted to the flow lines of the feed, top and bottom draw-offs, respectively.
The experimental results of dependence of flow rate and cut on separation factors were in good agreement qualitatively with those from theory based on a Newton iterative numerical solution of the two-dimensional convection-diffusion equation. A magnitude of separative power was, however, several times smaller than that from the analysis, when the ordinary diffusion coefficient and the thermal diffusion factor were obtained by the Chapman-Cowling formula and the Lennard-Jones (12-6) potential. Such a discrepancy in magnitude could not be reduced by rational variations of values of intermolecular potential parameters.

View full abstract

Two factors, which are important for selecting the extractor type suitable for liquid metal-molten salt system were studied, i.e. the formation and the coalescence processes of liquid metal drop. The drop formation process for liquid metal dispersion in the continuous phase is predictable from semi-empirical correlation reported on aqueous solution-organic solvent systems. The height of droplet bed being accumulated on drop-settling portion is predictable from the coalescence time of single drop on a flat metal interface. The coalescence of metal drop on a clean interface was very fast. The extractor type of liquid metal dispersion in molten salt is considered to be suitable for the pyrochemical extraction process.

View full abstract

In order to calculate airflow around a nuclear site, a new weighted interpolation method for sparse wind data at various terrain heights is developed. This method employs weighting functions for vertical distance and the topographic barrier between the station and a grid point, in addition to a weighting function for horizontal distance which is employed in usual methods. The new weighting functions are developed from the analysis of field measurements taken in complex terrain. This method can represent the wind-fields from the bottom to the upper boundary better than usual methods which use only the horizontal weighting factor, even when the upper wind data are not available.

View full abstract

Effective combination of measured data with theoretical analysis has permitted deriving a method for more accurately estimating the power distribution in BWRs. Use is made of least squares method for the combination between relationship of the power distribution with measured values and the model used in FLARE or in the three-dimensional two-group diffusion code. Trial application of the new method to estimating the power distribution in JPDR-1 has proved the method to provide reliable results.

View full abstract

Transient CHF (critical heat flux) tests of a 4 × 4 rod bundle were analyzed by the subchannel analysis program MENUETT. MENUETT is based on a non-equilibrium, five equation, two-phase flow model and is available both for steady state and transient analyses. Turbulent mixing and void drift effects are taken into account to calculate cross flows in fuel rod bundles. The tendency of calculated subchannel mass fluxes and qualities agreed with experimental data. By using a critical quality correlation obtained from steady state CHF data, the position of the earliest boiling transition could be predicted regardless of non-uniform axial heat flux distributions. This transition occurrence time was predicted within a difference of 0.10.3 s from the experimental time. MENUETT applicability was confirmed for transient calculations predicting thermal-hydraulic behavior in bundles.

View full abstract

In most studies, the Monte Carlo criticality code KENO-IV has been compared with other Monte Carlo codes, but evaluation of its performance with different box descriptions has not been done so far. In Monte Carlo computations, any fractional savings of computing time is highly desirable. Variation in computation time with box description in KENO for two different fast reactor fuel subassemblies of FBTR and PFBR is studied. The K_eff of an infinite array of fuel subassemblies is calculated by modelling the subassemblies in two different ways (i) multi-region, (ii) multi-box.
In addition to these two cases, excess reactivity calculations of FBTR are also performed in two ways to study this effect in a complex geometry.
It is observed that the K_eff values calculated by multi-region and multi-box models agree very well. However the increase in computation time from the multi-box to the multi-region is considerable, while the difference in computer storage requirements for the two models is negligible. This variation in computing time arises from the way the neutron is tracked in the two cases.

View full abstract

Methyl nitrate (CH₃NO₃) and ethyl nitrate (C₂H₅NO₃) generate in the radioiodine treatment system of reprocessing, when organic iodides are reacted with silver-impregnated silica gel (an adsorbent, e. g. AC6120) or with hyperazeotropic nitric acid in Iodox process⁽¹⁾. Since such alkyl nitrates are potentially explosive, their accumulation in the system should be avoided by an appropriate device ; however, the related literature is little. The present work offers a method of decomposing them through their reactions with nitrogen dioxide (NO₂) on zeolites.

View full abstract