A preliminary study of molten salt power reactors was made. The most promising fuel carrier salts were the fluorides and chlorides of the alkali metals, zirconium, and beryllium. The chlorides were found to have lower melting points but were less stable and more corrosive than the fluorides. A Li/sup 7/ F- - BeF/sub 2/ mixture with ThF/sub 4/ and UF/sub 4/appeared to perform best. Of the numerous alloys tested as container material, Inconel and a nickel-- molybdenum alloy INOR-8 appeared to be the most resistant to corrosion. To study the performance, safety, economics, and construction costs of a typical molten salt reactor, a reactor of specific type and size was chosen for study. The reference design reactor was a two-region homogeneous converter with a core salt of 70 mole% Li/sup 7/F and 30% BeF/sub 2. ThF/sub 4/ and enough VF/sub 4/ for criticality were added. Study indicated that a molten salt reactor would produce economical power, but the problem of developing a salt core and a container metal which would last for many years of operation needed further study. (M.C.G.)

Second and third tests were conducted on an expansion bellows and Zircaloy-stainless steel transition joint assembly, duplicating that used in the core-pressure vessel assembly of the HRT. The assembly was exposed for 1140 hours in the second test and and 167 hours in the third to a solution containing 0.04m UO2SO4, 0.02m H2SO4, and 0.005m CuSO4 in a 100A dynamic corrosion loop under conditions simulating reactor startup, shutdown, and steady-state operation. Results indicated that the transition joint assembly as tested was mechanically sound and leak-tight. Pit-type corrosion continued on the expansion bellows where the upper retaining tangent contacts the bellows. Results are included of specimens exposed to the circulation solution during transition joint and bellows test.

A third test of the mockup of the Zircaloy - stainless steel transition joint as used in the HRT reactor vessel was completed. No additional mechanical deflections of the bellows were made, but 12 additional thermal cycles of the unit were carried out. . The joint and bellows functioned properly and were leak-tight. Corrosion of stainless steel in 0.04 m UO2SO4 containing equimolar beryllium sulfate was not greater at 10-15 fps and much less at 15-75 fps, than in HRT solution at 200 and 225 C. Increasing the beryllium sulfate concentration to 0.08 m increased corrosion at 225 C and resulted in an unstable solution at 300 C.

A general program has been written to plot curves on the Oracle curve plotter. A description of the code and complete instructions for preparation of input tapes and operation of the code are given. The code tape is available from the Mathematics Panel or from the author.

A general program has been written to plot curves on the Oracle curve plotter. The supplement includes changes to slow down some of the loops and minimize the possibility of read-around errors and changes to handle special cases.

Increasing temperature in an aqueous liquid system results in a decrease in liquid density and an increase in vapor density. When level is measured by the differential pressure method, temperature increase results in a decrease in the effective span f the instrument and a shift in zero. When the instrument reference leg is uncompensated, both zero and span shifts occur with the effects being most pronounced at the high levels. When the reference leg is compensated, zero shifts are eliminated (instrument zero is at 100% level for this type installation). For control purposes at levels above 50% of full level, the compensated reference leg installation will give the best results. For measurement of levels below 50%, the uncompensated installation is best. As errors in level indications will be produced, in either type installation, by temperature, pressure and steam withdrawal (power) effects, the placement of taps should be such as to place the normal or control level in the region of 50% indicated level.

A qualitative theory for cake formation in ThO2 slurries is presented. The sphere formation which occurs in "Standard" slurry can be explained on the same basis. The irregular and unpredictable yield strength is easily fitted into the theory. On the basis of this theory, the writer has been led to the conclusion that only colloidally stable slurries, or slurries with crystallites or comminution-resistant particles which are large enough to overcome colloidal effects (probably larger than 0.1u) can be used with assurance in the ThO2 reactor system. This conclusion holds for 300 gm/kg slurries as well as for more concentrated mixtures.

A study of the degree and effects of radiation damage to the motor winding of the HRT 400-! canned motor circulating pumps was made. With the gamma flux estimate and the available radiation damage data, the effective life of the stator windings was estimated to be about one year, for normal operation of the pump and motor.

Data are presented for all toroid runs which circulated aqueous thorium oxide slurries between Aug, 1054, and October, 1956. In addition, a tabulation of the properties of numerous thoria preparations is presented.

Run SM-2 was run to determine whether a sulfated ThO2 slurry could be handled in a hydrodynamic system similar to the HRT test blanket. It was found that the ThO2 concentration circulating in the pipes was consistently 50% or less of the expected concentration based on the quantity of the oxide changed. The run lasted 1730 hours. It was terminated because of a slurry leak a few days before a shutdown had been scheduled. Severe erosion was found in the pump impeller and flow nozzle . Chloride concentration high enough to cause concern over possible stress corrosion cracking occurred on several occasions in the pressurizer.

This technical report presents an overview of nuclear reactor maintenance to be used in planning a nuclear reactor. There are certain basic maintenance fundamentals that are common to all types of reactors that may be' incorporated in a power producing facility. Basically, there are only two types of maintenance procedures. The direct type, which is common to conventional steam plants, may be used in some areas where the radioactivity is low enough. In most parts of the plant, maintenance will of necessity be remote due to the high level of radioactivity. For simplicity of description in this report all reactor types are divided into two general classes: solid fuel types and circulating fuel types. The report lists nine types of reactor power plant types with advantages and disadvantages maintenance-wise of each.

In analyzing stresses, deflections, and forces in piping and certain structural systems, it is useful to have a set of equations giving the deflections at any point of a curved beam in terms of the various loads acting on the beam. This technical report presents the deflection equations for a curved circular-arc beam of variable length. The cases treated include in-plane and out-of-plane bending due to forces and moments applied to the end of the beam in three principal directions, and uniformly distributed loads applied along the beam in these three directions. Deflections were calculated by the strain energy method using Castigliano's theorem.

A falling ball viscometer under development at ORNL, employing a flow system and an electromagnetically operated dash-pot pump, was evaluated for possible use with aqueous ThO2 slurry systems under reactor irradiation. Interchangeable check valve inserts were designed and fabricated to investigate several pump designs. Magnetic flux concentrations which originally prevented circulation of the ferritic stainless steel viscometer ball through the electromagnetic pump were eliminated by substitution of iron-magnetic stainless steel inserts. Viscosity was correlated through a logarithmic plot of the dimensionless Reynolds number vs. Froude number with the ratio of ball diameter to tube diameter as a parameter. The relation is linear in the laminar flow region.

The distillation method (Argonne Sampler) for the analysis of sodium oxide in NaK was used to calibrate a plug indicator. The description of the equipment, procedure, and experiences with the Argonne Samplers are presented in great detail to aid future users of this method. Although this method is not very precise, it has been thoroughly checked out and is recommended as a standard means for sampling and analysis for oxide in liquid metal systems. (auth)

It is concluded that an excess of O2 must be added to the HRT fuel circulating stream in amounts sufficient to recombine the D2 in dead-end lines and serve as corrosion protection.

A plot of the pressure-drop data shows that the pressure drop through a single HRT off-gas adsorption unit is equal to 2.66 times the O2 flow (measured in liters/min at 70 F, 14.7 psia).

The assembly of a second thorium oxide slurry corrosion test facility, loop BS, has been completed and 2010 hr of operation on slurry have been logged. This second test loop has proved satisfactory from an .operational standpoint. Corrosion data and operational observations are given for six thorium oxide slurry runs made at 300 C in 100A pump loops BS and CS. A new development model of the rotator for an in-pile slurry toroid is described.

This technical report described the formulation of a set of two-group neutron diffusion equations and the solution for the critical fuel cross section in a one-dimensional, multi-region spherical reactor. A subsequent report will describe the ORACLE code developed for survey calculations using this method.

General considerations involving the transportation of irradiated fuel and radioactive wastes are reviewed. It is assumed that many reactors will supply feed to a few large multipurpose chemical plants which ultimately send radioactive waste to a few disposal sites. General economic considerations of irradiated fuel reprocessing, economic aspects of the nuclear economy complex, growth predictions of the nuclear power economy in the U.S., general requirements for the shipment of fuel and waste, regulations applicable to fuel shipment, and permissible radiation levels are discussed.

The fact that material transport does occur under conditions of finite temperature difference in a flowing molten metal system was established. The rate mass transfer was thought to be either diffusion limited or solution rate limited. It is believed that the mass transfer of structural materials in Na or NaK systems is solution rate limited. The limiting process has not been qualitatively or quantitatively confirmed for the Inconel-Na or Inconel-NaK system. Increasing the maximum system wall temperature increases the amount of mass transfer, at least above 1300 deg F. The effect of the total temperature difference across the system on the amount of mass transfer was determined.

The general Wigner-Eisenbud theory is used to develop a method of analysis for the cross sections of fissionable nuclei. The method is employed in giving a reasonable description of the low energy cross sections in U/sup 235/. The single level fit for U/sup 235/ is known to be unreasonable. Many level expressions for the cross sections are derived--the only approximation to the general theory being the neglect of all but a small group of resonances. It is shown that regardless of the number or definition of the fission channels the many-level expressions require few level parameters: the E/sub lambda /, GAMMA / sub lambda n/, GAMMA /sub lambda gamma / and GAMMA /sub lambda F/ of the single level theory for each resonance and a few additional parameters pertinent to the interference between levels. The interference terms are described and shown to be important. The shape and size of the U/sup 235/ cross sections below 2 ev are fitied to within one per cent using (a) only one negative energy resonance of smaller size than in the single level fits (b) no additional levels to fit the shape other than the observed levels at positive energies (c) three interference parameters …

The Eyewash program, written by James H. Alexander and Nancy D. Given, provides solutions of reactor criticality problems in spherical geometry by means of the group diffusion method. It employs thirty lethargy groups (plus one thermal group) in nine regions. The input consists principally of specifying the geometrical scaling factor, boundaries and compositions of the various regions, and temperature level. The output includes the value of vc that would render the system critical, the relative fission density distribution, fissions, absorptions, and leakages in each lethargy group in each region, and, if desired, an edit of the flux at each space point, each lethargy, and an edit of the macroscopic cross sections for each lethargy, each region. OCUSOL-A is a minor modification and extension of Eyewash. It provides for the computation and editing, on the supervisory control typewriter, of the total absorptions in selected nuclides in the various regions. This information is useful in the computation of breeding ratios and the preparation of detailed neutron balances, and in the estimation of flux-averaged cross sections for use in estimating the rate of change of concentration of the various nuclides with burn-up. The program also provides for saving and transferring the final fission …

A series of two-group calculations has been made on the Oracle for the purpose of obtaining critical-mass and flux distribution data for various ORR core configurations. The 3G3R code of Bate, Einstein, and Kinney was used, together with the RSP code developed by Nelson. This made it possible to obtain results for the three-dimensional case. The results, which are presented graphically, are intended to serve as a guide for the design of experiments until such time as actual measurements are available. The calculations were performed for the "clean cold" case, and it should be realized that the presence in the core of experiments and of fission products built up during operation will materially alter the flux patterns found. It is believed that the critical-mass data are accurate to within 10%. Within the fuel region it is believed that the thermal-flux patterns are the also accurate to this degree. Comparison of the results with MTR critical experiments, however, indicates that the thermal flux in the reflector in the vicinity of the fuel-reflector interface may have been underestimated by a factor of as much as 1.3. It should also be recalled that in a two-group calculation the "fast flux" is often a …

Preliminary results of tests wit the Westinghouse 100A pump indicate that Al2)3 and tungsten carbide coated bearing-journal assemblies prepared by the Linde process are not promising as substitutes for the graphitar-stellite combination. The front Al2O3 assembly failed at start-up with water and both front and rear tungsten carbide assemblies failed after 121 hours with water at 245 C.