Volume 72, Issue 5-6

The effect of TRISO particles' size variations and inter-particle distance on infinite multiplication factor (k inf ) of a pebble bed reactor (PBR) is studied using MCNP code. It is observed that k inf decreases with increasing TRISO particles' size, mainly due to increased total particle surface leading to stronger resonance absorption. Negative temperature reactivity effect is observed at all particle sizes. The k inf increases strongly when particles form in clusters as the clusters behave as a single bigger particle. It is also observed that the white boundary condition is not an appropriate approximation of the unit-cell flux boundary condition in the calculations as it yields significantly higher values of k inf than periodic and reflective boundary conditions.

The existence of boiling phenomena in boiling water-reactors makes these systems susceptible to instabilities. Neutronics coupled with the thermal-hydraulics complicate the problem further. Boiling water reactor instabilities are classified as global or local according to the spatial dependency of the power fluctuations. At present, numerical codes can simulate these instabilities rather well on event basis; but more general physical insight can be gained readily through analytical modeling. One of the earliest models applied to the problem is the parallel channel instability, so called because of the applied thermohydraulic boundary conditions. In this work a new type of instability model is proposed and studied by including the effect of the first axial harmonic of the flux. Thermal-hydraulics is represented with one-dimensional two-phase homogeneous equilibrium flow model equations. Analysis is performed with linearized equations in the frequency domain. Special emphasis is given to the motion of the boiling boundary. The results clearly indicate that the first axial harmonic of the flux affects the stability of the reactor adversely, and must be accounted for to make more realistic predictions.

In the present work, the thermal-hydraulic analysis of reactivity-induced transients in a Low Enriched Uranium (LEU) core of a typical material test research reactor is conducted using the previous program developed Khater et al. The analysis was done for the uncontrolled withdrawal of a control rod under scram-disabled conditions. The initiating event reactivity was considered with and without influence of the reactivity efficiency curve (“S” curve). The results of the transient calculations are analyzed and compared with each other. In the transient without the “S” curve influence, a high primary peak power of 406.18 MW is attained and a clad melt down is occurring after 1.85 s. In the transient with the “S” curve influence, a super prompt highly critical situation is produced (1.762 $ at 0.895 s) with a very high primary peak power of 801.05 MW at 0.912 s. A fast clad melt down is resulting in the hot channel at 1.088 s and a stable film boiling is occurring. This study shows that the influence of the reactivity efficiency curve results in higher peaks in power and temperatures (fuel, clad and coolant) with a fast clad melt down than that of a linear assumption.

Near surface disposal of low and intermediate level radioactive waste is probably the best solution for a long term radioactive waste management today. For safety assurance all near surface disposal aspects and evolution scenarios should be reviewed and analyzed including repository design and site selection, because releases of radionuclides from the repository into the environment cause radiation exposure to the public. Mechanisms of 14 C, 129 I and 137 Cs transportation from the repository into the environment, modelled for the hypothetical radioactive waste disposal facility in Lithuania, are investigated and the results of effective dose calculations for the main exposure pathways are discussed in this paper. The RESRAD-OFFSITE code has been used in this study for modelling purposes. It is shown that the radionuclide transport time into the environment, its radioactive contamination and the radiation exposure to the public mostly depends on the mobility and sorption properties of radionuclides and on the appropriate transport pathways from the repository into the environment.

An extensive analysis of the accuracy in the theoretical reproduction of experimental proton induced reaction cross-sections up to 150 MeV is presented. The predictive capabilities of different models for the description of the nuclear level densities as implemented in the TALYS code and in the modified ALICE code have been assessed by means of a systematic comparison of the theoretical results with experimental EXFOR data relative to target nuclei from 24 Mg to 209 Bi. The results obtained give the possibility to find the best approaches for the cross-section calculation for nuclei from different mass ranges and for different channels.

The finite element-boundary element hybrid method developed previously for reflected systems and restricted to one or two group neutron diffusion theory has been extended to the general multigroup neutron diffusion theory by using the boundary integral equation of multigroup neutron diffusion theory. A linear or bilinear 2-D FEM formulation in the core combined with a 2-D linear BEM formulation in the reflector constitute the basic discretization procedure. Use of the boundary integral equation of multigroup diffusion theory transforms all group-to-group scattering domain integrals into surface integrals in the reflector. Hence the need for a reflector domain mesh is completely eliminated. Via comparisons with pure FEM and BEM solutions of the reflected systems within the context of three and four group diffusion theories, the present formulation is validated and assessed.

The H N method is used to solve radiation transfer problems for plane parallel media for anisotropic scattering. This approach is based on the use of the angular distributions in terms of singular eigenfunctions of the method of elementary solutions. Two analytical expressions are derived for the coefficients of the power series expansions in exit distributions.

The singular eigenfunction approach using C N equations can be extended for high order anisotropic scattering (triplet scattering) to the Albedo problem. Case's eigenfunctions and the orthogonality relations of this scattering can be derived mathematically. Thereby, the infinite medium Green function can be written for an homogeneous medium in terms of singular eigenfunctions. Also using the H N method, the constant source albedo problem can be investigated for this type of scattering.

Using methods of the systems theory of electronic communications and theoretical electrical science, this study describes the transmission of pulse-shaped signals from nuclear radiation detectors from the detector over “electrically long lines” (cables) to the output of the first pulse amplifier. The example of pulses from BF 3 -proportional counters shows, using the Fourier transformation, that pulses from radiation detectors contain a frequency spectrum ranging well above 10 8 Hz. If these pulses are transmitted to the first amplifier over a line length of several meters, the laws of the theory of transmission lines must be taken into account to avoid false signals caused by reflections. In the example, line equations are applied and the influence of the line and the terminating impedance is demonstrated. The influence of the frequency response v(ω) and the phase response δ(ω) of the amplifier is also considered in the sample calculation. The methods presented make it possible to analyze and optimize the transmission and amplification of signals from radiation detectors. Close agreement emerges between empirically observed and calculated pulse shapes.

The preignition probability for a sigmoidal reactivity input in Nuclear Explosive Devices (NEDs) with a strong inherent neutron source strength (i. e. using reactor grade plutonium) is treated comprehensively. The cumulative preignition probability, its probability density as well as the most probable time for preignition during the compression phase are presented in a closed analytical form. This general formalism can be used to calculate the preignition probability for any metallic core composition (either Pu or U or hybrid) possessing an inherently strong or weak background of neutrons.

As part of the analysis of options to generate additional 6 – 8 TWh of electrical energy for Croatian consumers in the next decade, a comparison between natural gas combined cycle plants jointly with wind electricity generators and nuclear power plants has been performed. The criteria for the optimal option are minimal electricity cost and minimal carbon dioxide emissions. Since the comparison has to include the analysis of electricity cost generated during plant life time and since potential long term fuel cost variations (gas, uranium concentrate and uranium enrichment) containing substantial uncertainties, the best solution is to calculate life time levelized electricity costs with probabilistic methods. The results, obtained in form of probabilistic distributions, show that the option with nuclear plants leads to remarkably lower electricity cost and CO 2 emissions than the option with gas and wind plants. The difference is more pronounced when wind generators have low capacity factors.