The following document provides measurements, equations and figures describing the minimum power coil geometries for D.C. magnetic mirrors.

Abstract: "A scheme is discussed in which D+ ions would be accelerated to about 100 kev in a conventional accelerator, then be sent through a gas target frmo which about half the ions would emerge neutral with very little scattering or energy loss; the neutral beam would then cross into a magnetic field and part of the atoms would be ionized and trapped by colliding with the already trapped ions. The smallest mirror machine with a central field of 30-40 kilogauss, 2;1 mirror ratio and β = 01 to 0.5 that would trap most of the neutral beam in the steady state, would be about a meter in diameter and require the order of hundreds of amperes of D+ ions into the neutralizer. The smallest stellerator with a field of 30 kilogauss and β = 0.1 that would trap most of the neutral beam would be about a meter long and require about 10 amperes of D+ ions. A preliminary mirror machine experiment that might be done using the 1/2-ampere, 100-kev D+ beam from the UCRL spare MTA injector is discussed. A method of starting these machine is discussed."

A knowledge of the mechanical and physical properties of uranium is important in understanding and explaining its pile behavior. The effects of irradiation on the properties of uranium will be dealt with in another chapter. However, it is important to know and understand the pre-irradiation properties of uranium prior to investigating the effects of irradiation.

This twenty-second quarterly report of the research and development activities of the Radiological Sciences Department, Hanford Atomic Products Operations, as before, includes some items charged to control but included for general interest. Such are identified as "not charged to research." The previous report of this series was numbered: HW-34408 January 10, 1955

A study was performed to determine the economic incentives associated with in-pile boiling, with the explicit assumption that this method of operation was technically feasible. A non-boiling reactor plant was used as a comparison, and the net electric generation was held constant at 400 Mw. The economic effect of variations in specific power from 47 to 32 kw/ft., and steam qualities of 10 to 60% were studied. The results showed that, under the conditions used, the unit power cost during single purpose operation would be in the range 5.9 to 6.1 mils/kw-hr at the generator bus-bar, compared to an estimated 6.6 mils/ kw-hr. for the non-boiling plant. These gains are realized under conditions of relatively high specific powers, greater than 32 kw/ft., relatively modest bulk outlet steam quality (greater than 20%), and high-pressure steam (1000 psia or higher) used directly in the turbines. During the dual market period, the power rate is about the same. An increase of 0.2 mils/kw-hr. is indicated when the specific power is lowered- Capital investment savings of perhaps four to five million dollars may be possible, but savings of this magnitude have little effect on the over-all power cost. The power cost savings stated above …

The conservative design criteria applied to process tubing may seriously hinder the advancement of high temperature reactor development. Factors of safety of 3 to 4 applied to an arbitrarily defined yield strength are used to calculate allowable internal working pressures. The effect of biaxial stressing on the tubes has been shown (2) to cause yielding at a lower internal pressure. However, tests conducted on biaxially stressed short lengths of tubing indicated that the ultimate tensile strength of the material was not affected.

Construction of the HRT reactor shield tank was completed, and the inside surfaces were painted. The roof structure for the tank is being assembled in preparation for an acceptance pressure test. Service piping and instrument lines are being installed in the central room area by ORNL craft forces. This work is approximately 50% complete. Fabrication of all temperature system components, except the blanket outer storage tanks, has been completed.

A preliminary chemical flowsheet is presented of a fluoride volatility process for recovering and decontaminating uranium from heterogeneous reactor fuels after dissolution in a fused salt. In laboratory work, a gross β decontamination factor of > 10 4 was obtained in the fluorination of a UF4-NaF-ZrF4 melt by passing the product UF6 through NaF at 650°C. The solubility of UF6 in molten NaF-ZrF4 was shown in kinetic studies to cause a lag in the evolution of UF6 from the fluorinator. Corrosion of nickel in the fluorination step appeared to be 2-4 mils/hr during the time that uranium was present. The average corrosion rate over the process as a whole was less than O.4 mil/hr. Earlier studies were reported in ORNL-1709 and 1877.

It was suggested to the writer by K.A. Eschbach that liquid magnesium might offer certain advantages as a reactor coolant. As a result of this suggestion, a preliminary investigation of the possibilities of this material was made. Definite advantages for a restricted class of applications were found, but a detailed evaluation would seem to require further basic experimental research on the heat transfer, corrosion and flux-mechanical properties of the substance.

Pressure drops are reported for forced circulation flow of steam-water mixtures in a 23.5 foot long, 1.43 inch ID, 0.1 inch thick, horizontal annulus. The inner surface of the annulus was uniformly heated over a range from 97,000 to 233,000 Btu/hr-ft², exit pressures extended from 100 to 500 psig, and exit steam qualities varied from 0 to 60% by weight. Liquid water entered the annulus and boiling lengths up to 15 feet were investigated. Moreover, the Woods and the Martinelli and Nelson methods of calculating two-phase pressure drop were applied to the experimental conditions, and the deviations between the analytical and the test results are presented.

A nuclear safety indicator has been described. This is an instrument which will make it possible to determine the safety of performing contact maintenance work on certain long columns used in the continuous flow processing of plutonium.

Fourier analyses are made on two functions. The first is a step function forward from periodic samples of a sinusoid. If the frequency of the sinusoid is less than one-half of the sampling frequency, it is shown that the step function has no frequency components less than one-half of the sampling frequency other than that of phase, and duration with respect to the interval of the analysis. It is shown that the insertion of a blank space in the period of analysis reduces greatly the uncertainty of the amplitude of the initial sinusoid as estimated from the results of the Fourier analysis. The results of the analyses are useful in the design and evaluation of certain analog data-analyzing systems.