Search

V. S. O. P. ('Very Superior Old Programs) is a system of codes lurked together for the simulationof reactor life histories and temporary in-depth research. In comprises neutron cross sectionlibraries and processing routines, repeated neutron spectrum evaluation, 2-D diffusion calculationwith depletion and shut-down features, in-core and out-of--pile fuel management, fuel cyclecost analysis, and thermal hydraulics (at present restricted to 's). Various techniques havebeen employed to accelerate the iterative processes and to optimize the internal data transfer.The storage requirement is confined to 17 M-Bytes.The code system has extensively been used for comparison studies of reactors, their fuel cycles,simulation of safety features, developmental research, and reactor assessments .

In the next generation of fusion experiments, which are planned to reach thermonuclear ignition conditions, a deuterium-tritium mixture will be used as fuel. If significant D-T burning is achieved in these devices a large number of 14.1 MeV neutrons will be produced which will activate the components used in the experiments. The resulting activation radiation will create additional problems of access to the machine for maintenance and repair, which should be taken into account in the design and construction phase of these experiments. To give an idea of the dose rates to be expected, some activation calculations for a typical large Tokamak ignition experiment are presented. It is assumed that the machine is operated in a pulsed mode at an average of 100 plasma discharges within one week and that 10$^{20}$ neutrons per pulse are produced. The investigation shows that 30 days after shutdown of the machine the total absorbed dose rate in the vicinity of the device is about 0.2 rem/h which limits the working time in this area to about 10 h per quarter. For repairs inside the vacuum vessel or dismantling of the machine it is assumed that the machine can be divided into eight identical segments. The dose rates from an isolated segment and from the remaining part of the machine after removal of one segment are calculated. The results for both cases are of the same order of magnitude and are in the range of 2 to 4 rem/h for decay times between 5 and 30 days. Hence, all repairs inside the vacuum vessel or dismantling of the machine must be done by remote control.

This report presents some model designs of fusion reactor blankets with low lithium and tritium inventories. The lithium inventory is minimized by using a thin breeding zone which is operated in a soft neutron spectrum. Enriched lithium (28 % up to 90 % $^{6}$Li) is used as breeding material and beryllium is employed for neutron multiplication and moderation. Due to the low lithium inventory, the equilibrium tritium inventory can also be kept to a minimum provided the tritium extraction system is sufficiently efficient. A parametric study has been performed to investigate the relevant neutronic and photonic characteristics of this blanket : concept. Different blanket models are compared with respect to their breeding potential, the nuclear heating,and the lithium, beryllium, and tritium inventories involved. The advantages of low lithium and tritium inventories from the safety point of view are weighed against the drawbacks, like the need of enriched lithium, the large beryllium inventory, and the contamination of the coolant with tritium produced in beryllium.

In many problems of reactor physics anisotropic scattering of neutrons in the high energy range has an important influence on the neutron transport through matter. To take this fact into account in neutronic calculations computer codes based on the Monte Carlo method and on the method of diserete ordinates are used. The different techniques for the handling of anisotropic scattering in these transport codes are described and discussed. For a comparison of the different methods the neutron flux distribution in a simple assembly consisting of natural lithium is investigated by three computer codes : a modified version of the Monte Carlo code ESP, the discrete ordinates transport code ANISN, and the multigroup Monte Carlo code MORSE which uses the same group cross sections as ANISN. In general, the agreement between the different codes is good ; even better than one would expect from the approximations made in the codes ANISN and MORSE. Some discrepancies occur at the inner and outer boundary of the assembly ; but it cannot be concluded that these are due to an incorrect handling of anisotropic scattering in some of the codes.