[en] Effects of multigroup neutron cross section generation procedures on core physics parameters for compact fast spectrum reactors have been examined. Homogeneous and space-dependent multigroup cross section sets were generated in 11 and 27 groups for a representative fast reactor core. These cross sections were used to compute various reactor physics parameters for the reference core. Coarse group structure and neglect of space-dependence in the generation procedure resulted in inaccurate computations of reactor flux and power distributions and in significant errors regarding estimates of core reactivity and control system worth. Delayed neutron fraction was insensitive to cross section treatment, and computed reactivity coefficients were only slightly sensitive. However, neutron lifetime was found to be very sensitive to cross section treatment. Deficiencies in multigroup cross sections are reflected in core nuclear design and, consequently, in system mechanical design

[en] Neutron streaming and heterogeneity effects are studied in detail in a lattice of fuel pins with voided coolant channels and in a lattice of bubbles in a pool of boiling fuel. A multi-energy Monte Carlo transport calculation of streaming effects in a lattice of bubbles is presented. Streaming is found to cause a change in k, the reactor eigenvalue or multiplication factor, of 0.01 in the lattice studied. Benoist diffusion coefficients are computed and used in a multigroup anisotropic diffusion calculation of streaming effects in this lattice. The multi-energy Monte Carlo and multigroup anisotropic diffusion estimates of streaming effects are found to agree very well. A multi-energy Monte Carlo transport estimate of streaming effects is presented for a cell similar to that planned for the Clinch River Breeder Reactor (CRBR). Effective anisotropic diffusion coefficients are computed for the true CRBR lattice cell. Three sets of Benoist diffusion coefficients are computed for corresponding cylindricized CRBR cells. All four sets of anisotropic diffusion coefficients are used in multigroup anisotropic diffusion calculations of streaming effects in the CRBR lattice. The multi-energy Monte Carlo and anisotropic diffusion estimates of streaming are found to be in reasonable agreement, but minor discrepancies are noted in the streaming reactivities computed with the four sets of anisotropic diffusion coefficients

[en] Over a period of 30 years, more than a hundred Zero Power Reactor (ZPR) critical assemblies were constructed at Argonne National Laboratory. The ZPR facilities, ZPR-3, ZPR-6, ZPR-9 and ZPPR, were all fast critical assembly facilities. The ZPR critical assemblies were constructed to support fast reactor development, but data from some of these assemblies are also well suited for nuclear data validation and to form the basis for criticality safety benchmarks. A number of the Argonne ZPR/ZPPR critical assemblies have been evaluated as ICSBEP and IRPhEP benchmarks. Of the three classes of ZPR assemblies, engineering mockups, engineering benchmarks and physics benchmarks, the last group tends to be most useful for criticality safety. Because physics benchmarks were designed to test fast reactor physics data and methods, they were as simple as possible in geometry and composition. The principal fissile species was ²³⁵U or ²³⁹Pu. Fuel enrichments ranged from 9% to 95%. Often there were only one or two main core diluent materials, such as aluminum, graphite, iron, sodium or stainless steel. The cores were reflected (and insulated from room return effects) by one or two layers of materials such as depleted uranium, lead or stainless steel. Despite their more complex nature, a small number of assemblies from the other two classes would make useful criticality safety benchmarks because they have features related to criticality safety issues, such as reflection by soil-like material. ZPR-3 Assembly 6 consisted of six phases, A through F. In each phase a critical configuration was constructed to simulate a very simple shape such as a slab, cylinder or sphere that could be analyzed with the limited analytical tools available in the 1950s. In each case the configuration consisted of a core region of metal plates surrounded by a thick depleted uranium metal reflector. The average compositions of the core configurations were essentially identical in phases A - F. ZPR-3 Assembly 6F (ZPR-3/6F), the final phase of the Assembly 6 program, simulated a spherical core with a thick depleted uranium reflector. ZPR-3/6F was designed as a fast reactor physics benchmark experiment with an average core ²³⁵U enrichment of approximately 47 at.%. Approximately 81.4% of the total fissions in this assembly occur above 100 keV, approximately 18.6% occur below 100 keV, and essentially none below 0.625 eV - thus the classification as a 'fast' assembly. This assembly is Fast Reactor Benchmark No. 7 in the Cross Section Evaluation Working Group (CSEWG) Benchmark Specifications and has historically been used as a data validation benchmark assembly. Loading of ZPR-3/6F began in late December 1956, and the experimental measurements were performed in January 1957. The core consisted of highly enriched uranium (HEU) plates, depleted uranium plates, perforated aluminum plates and stainless steel plates loaded into aluminum drawers, which were inserted into the central square stainless steel tubes of a 31 x 31 matrix on a split table machine. The core unit cell consisted of three columns of 0.125 in.-wide (3.175 mm) HEU plates, three columns of 0.125 in.-wide depleted uranium plates, nine columns of 0.125 in.-wide perforated aluminum plates and one column of stainless steel plates. The maximum length of each column of core material in a drawer was 9 in. (228.6 mm). Because of the goal to produce an approximately spherical core, core fuel and diluent column lengths generally varied between adjacent drawers and frequently within an individual drawer. The axial reflector consisted of depleted uranium plates and blocks loaded in the available space in the front (core) drawers, with the remainder loaded into back drawers behind the front drawers. The radial reflector consisted of blocks of depleted uranium loaded directly into the matrix tubes. The assembly geometry approximated a reflected sphere as closely as the square matrix tubes, the drawers and the shapes of fuel and diluent plates allowed. According to the logbook and loading records for ZPR-3/6F, the reference critical configuration was loading 5 which was critical on January 4, 1957. The subsequent loadings were very similar but were less clean for criticality because there were modifications made to accommodate reactor physics measurements other than criticality. Accordingly, ZPR-3/6F loading 5 was selected as the only configuration for this benchmark. As documented below, it was determined to be acceptable as a criticality safety benchmark experiment. A very accurate transformation to a simplified model is needed to make any ZPR assembly a practical criticality-safety benchmark. There is simply too much geometric detail in an exact (as-built) model of a ZPR assembly. This is especially true of ZPR-3/6F because of the complex core loading required to approximate a sphere with rectangular plates in a square matrix.

[en] A major safety concern for nuclear reactors in space power applications is the effect of accidental submersion of the reactors in water. Such a situation might be postulated, for example, as a consequence of a launch pad accident. The classes of reactors proposed most frequently for use in space are fast spectrum reactors, for which submersion results in a softened core neutron spectrum caused by the displacement of the liquid metal coolant by the water. The softened spectrum alters the neutron balance in the core - neutron capture and fission are increased while leakage from the core is reduced. Water outside the submerged core introduces an increased number of reflected thermalized neutrons into the core. The net effect is a function of the specific features of the reactor design (composition, size, etc.) and can be positive or negative depending upon the contributions of the individual effects. Analysis of the magnitude of the effect requires an accurate evaluation of the individual components. At present a designer must rely on detailed calculations performed after key design parameters are settled to determine the effects of submersion. The purpose of our work is to develop generic features of the submersion phenomenon to provide designers a means to an a priori knowledge of the impact of potential design choices on submersion reactivity

[en] Generic features of the interaction between core nuclear and mechanical designs and reactor control system design for compact fast spectrum space power reactors have been examined. Reactivity worths of various control concepts were evaluated for representative fast spectrum cores. In addition, special characteristics of each control concept that significantly affect core nuclear and mechanical design were considered. Ex-core control methods based on reflector control and in-core control devices such as control rods lead to divergent core designs and to different types of design problems. Total control worth of ex-core control devices is limited and is strongly dependent on core size. Reflector control also results in unfavorable radial power shifts, but ex-core control does avoid unnecessary reactor vessel penetrations. Control rods have characteristics essentially opposite to those of ex-core devices. Design demands on the primary control system are shown to be reduced by including a slow-acting secondary system based on in-core dispersed poison

[en] The TREAT Upgrade (TU) reactor design is presently nearing completion. The reactor will be used to test LMFBR fuel under simulated accident conditions. The physics of the TU core is complicated by a number of factors related to the planned application of the facility. In this paper the design approach used to produce the core fissile loading spatial distribution needed to satisfy the requirement to test a number of different test clusters in various test loops in a transient operating mode is described. The TU core consists of a driver comprised of the pre Upgrade Zircaloy-clad TREAT fuel assemblies and a central converter region of new Inconel-clad fuel assemblies surrounding the test loop location. Separating the two concentric zones is a buffer zone that is made up of new Inconel clad assemblies operating at lower temperatures than the converter region

No abstract available

[en] Over a period of 30 years, more than a hundred Zero Power Reactor (ZPR) critical assemblies were constructed at Argonne National Laboratory. The ZPR facilities, ZPR-3, ZPR-6, ZPR-9 and ZPPR, were all fast critical assembly facilities. The ZPR critical assemblies were constructed to support fast reactor development, but data from some of these assemblies are also well suited for nuclear data validation and to form the basis for criticality safety benchmarks. A number of the Argonne ZPR/ZPPR critical assemblies have been evaluated as ICSBEP and IRPhEP benchmarks. Of the three classes of ZPR assemblies, engineering mockups, engineering benchmarks and physics benchmarks, the last group tends to be most useful for criticality safety. Because physics benchmarks were designed to test fast reactor physics data and methods, they were as simple as possible in geometry and composition. The principal fissile species was ²³⁵U or ²³⁹Pu. Fuel enrichments ranged from 9% to 95%. Often there were only one or two main core diluent materials, such as aluminum, graphite, iron, sodium or stainless steel. The cores were reflected (and insulated from room return effects) by one or two layers of materials such as depleted uranium, lead or stainless steel. Despite their more complex nature, a small number of assemblies from the other two classes would make useful criticality safety benchmarks because they have features related to criticality safety issues, such as reflection by soil-like material. ZPR-3 Assembly 11 (ZPR-3/11) was designed as a fast reactor physics benchmark experiment with an average core ²³⁵U enrichment of approximately 12 at.% and a depleted uranium reflector. Approximately 79.7% of the total fissions in this assembly occur above 100 keV, approximately 20.3% occur below 100 keV, and essentially none below 0.625 eV - thus the classification as a 'fast' assembly. This assembly is Fast Reactor Benchmark No. 8 in the Cross Section Evaluation Working Group (CSEWG) Benchmark Specificationsa and has historically been used as a data validation benchmark assembly. Loading of ZPR-3 Assembly 11 began in early January 1958, and the Assembly 11 program ended in late January 1958. The core consisted of highly enriched uranium (HEU) plates and depleted uranium plates loaded into stainless steel drawers, which were inserted into the central square stainless steel tubes of a 31 x 31 matrix on a split table machine. The core unit cell consisted of two columns of 0.125 in.-wide (3.175 mm) HEU plates, six columns of 0.125 in.-wide (3.175 mm) depleted uranium plates and one column of 1.0 in.-wide (25.4 mm) depleted uranium plates. The length of each column was 10 in. (254.0 mm) in each half of the core. The axial blanket consisted of 12 in. (304.8 mm) of depleted uranium behind the core. The thickness of the depleted uranium radial blanket was approximately 14 in. (355.6 mm), and the length of the radial blanket in each half of the matrix was 22 in. (558.8 mm). The assembly geometry approximated a right circular cylinder as closely as the square matrix tubes allowed. According to the logbook and loading records for ZPR-3/11, the reference critical configuration was loading 10 which was critical on January 21, 1958. Subsequent loadings were very similar but less clean for criticality because there were modifications made to accommodate reactor physics measurements other than criticality. Accordingly, ZPR-3/11 loading 10 was selected as the only configuration for this benchmark. As documented below, it was determined to be acceptable as a criticality safety benchmark experiment. A very accurate transformation to a simplified model is needed to make any ZPR assembly a practical criticality-safety benchmark. There is simply too much geometric detail in an exact (as-built) model of a ZPR assembly, even a clean core such as ZPR-3/11 loading 10. The transformation must reduce the detail to a practical level without masking any of the important features of the critical experiment. And it must do this without increasing the total uncertainty far beyond that of the original experiment. Such a transformation is described in Section 3. It was obtained using a pair of continuous-energy Monte Carlo calculations. First, the critical configuration was modeled in full detail - every plate, drawer, matrix tube, and air gap was modeled explicitly. Then the regionwise compositions and volumes from the detailed as-built model were used to construct a homogeneous, two-dimensional (RZ) model of ZPR-3/11 that conserved the mass of each nuclide and volume of each region. The simple cylindrical model is the criticality-safety benchmark model. The difference in the calculated k_eff values between the as-built three-dimensional model and the homogeneous two-dimensional benchmark model was used to adjust the measured excess reactivity of ZPR-3/11 loading 10 to obtain the k_eff for the benchmark model.

[en] The integral Fast Reactor (IFR) being developed by Argonne National Laboratory (ANL) combines the advantages of metal-fueled, liquid-metal cooled reactors and a closed-loop fuel cycle. Some of the primary advantages are passive safety for the reactor and resistance to diversion for the heavy metal in the fuel cycle. in addition, the IFR pyroprocess recycles all the long-lived actinide activation products for casting into new fuel pins so that they may be burned in the reactor. A key component in the Fuel Cycle Facility (FCF) recycling process is the electrorefiner (ER) in which the actinides are separated from the fission products. In the process, the metal fuel is electrochemically dissolved into a high-temperature molten salt, and electrorefined uranium or uranium/plutonium products are deposited at cathodes. This report addresses the new and innovative aspects of the criticality analysis ensuing from processing metallic fuel, rather than metal oxide fuel, and from processing the spent fuel in batch operations. in particular, the criticality analysis employed a mechanistic approach as opposed to a probabilistic one. A probabilistic approach was unsuitable because of a lack of operational experience with some of the processes, rendering the estimation of accident event risk factors difficult. The criticality analysis also incorporated the uncertainties in heavy metal content attending the process items by defining normal operations envelopes (NOES) for key process parameters. The goal was to show that reasonable process uncertainties would be demonstrably safe toward criticality for continuous batch operations provided the key process parameters stayed within their NOES. Consequently the NOEs became the point of departure for accident events in the criticality analysis

[en] The continuous-energy Monte Carlo neutron transport code VIM and its auxiliaries are briefly described. The ENDF/B cross section data processing procedure is summarized and its benchmarking against MC²-2 is reviewed. Several representative applications at ANL are described, including fast critical assembly benchmark calculations and STF and TREAT Upgrade benchmark calculations. 2 figures