1. NAME AND TITLE OF DATA LIBRARY

TRANSMIT: Experimental Neutron Transmission Data Used to Test Total Cross Sections.

2. NAME AND TITLE OF DATA RETRIEVAL PROGRAMS

XTRSIC: A Program to Print Experimental Transmission Data by Element.

3. CONTRIBUTOR

Oak Ridge National Laboratory, Oak Ridge, Tennessee.

4. HISTORICAL BACKGROUND AND INFORMATION

Since evaluated cross section data are ultimately used to predict neutron transmission through shields comparatively thicker than the samples normally used to make cross section measurements, the Defense Nuclear Agency sponsored experiments to test the data with samples of realistic thicknesses. These transmission checks are necessary because the total cross-section evaluations are based on measurements from various sources using different techniques covering different energy regions and often with inadequate energy resolution to resolve all of the structure in the cross sections. This situation results in two major sources of errors in the file affecting the cross-section magnitude and its energy scale. The energy scale differences in the data may be solved, for a given element, at the time the evaluation is made when there is sufficient energy overlap between the various data sets. However, for practical shielding applications involving several elements, it is difficult to insure that the energy scale of the different evaluations is consistent unless a high degree of absolute accuracy is achieved in each evaluation. The finite energy resolution of the transmission data, upon which the total cross section's evaluation is based, may cause very serious errors when all of the structure is not resolved. This is so because the average transmission obtained at a given energy for a specific sample thickness cannot be converted to an average cross section which would be valid for other energy resolutions and/or sample thickness. Above 0.5 MeV neutron energy, total cross sections are usually obtained from transmission measurements using time-of-flight with a "white neutron source" and recoil proton detectors for sample thicknesses between 0.2 and 0.3 atoms per barn. Our measurements which were obtained with similar techniques will, therefore, test the adequacy of these data with sample thicknesses 3 to 4 times larger. From 0.2 to 0.5 MeV the data situation varies very much from nuclei to nuclei and the evaluations are often based on very fragmentary data from diverse sources. It is very difficult to estimate, a priori, the adequacy of the evaluations in this energy region but, typically, we may expect errors to be as large as 30 to 40%.

5. APPLICATION OF THE DATA

The data may be compared with the uncollided flux transmission probabilities based on any newly evaluated or newly acquired total cross section file.

6. SOURCE AND SCOPE OF DATA

Neutron transmission measurements, from 0.2 to 20.2 MeV, have been made for the shielding materials carbon, oxygen, aluminum, silicon, calcium, iron, and the compound silicon dioxide. The measurements were performed at the ORELA Shield Test Station with a resolution of about 0.12 nsec/meter on sample thicknesses varying from 0.65 to 0.9 atoms/barn. The transmission measurements were compared with the predictions obtained from the Defense Nuclear Agency evaluated cross-section library as of October, 1972. Since the total cross-section files for these elements are also the ones present in the ENDF/B-III library, we also checked its total cross-section files for all of these elements with the exception of oxygen. There are serious discrepancies between our data and the predictions based on the evaluated files. These discrepancies are often large in the energy region from 0.2 to 0.6 MeV.

The experimental technique, the data reduction method, the experimental results and the comparison with the predictions of the evaluations are described in the packaged documentation.

7. DISCUSSION OF THE DATA RETRIEVAL PROGRAM

The transmission data is formatted to provide a list of the transmission for each energy and estimated error. The retrieval program simply locates the desired element(s) and prints the data.

8. DATA FORMAT COMPUTER

BCD/EBCDIC card images; IBM 360/370.

9. TYPICAL RUNNING TIME

A complete printout required 28 seconds of CPU time on the IBM 360/75 computer.

10. REFERENCE

F. G. Perey, T. A. Love, and W. E. Kinney, "A Test of Neutron Total Cross-Section Evaluations from 0.2 to 20 MeV for C, O, Al, Si, Ca, Fe, and SiO₂," ORNL-4823 (ENDF-178) (December 1972).

11. CONTENTS OF LIBRARY

Included are the referenced document and one (1.2MB) DOS diskette which contains the transmission data, the input and output for the sample problem.

12. DATE OF ABSTRACT

October 1972; reviewed May 1984.