Reduction of bias in neutron multiplicity assay using a weighted point model
Abstract
Accurate assay of most common plutonium samples was the development goal for the nondestructive assay technique of neutron multiplicity counting. Over the past 20 years the technique has been proven for relatively pure oxides and small metal items. Unfortunately, the technique results in large biases when assaying large metal items. Limiting assumptions, such as unifoh multiplication, in the point model used to derive the multiplicity equations causes these biases for large dense items. A weighted point model has been developed to overcome some of the limitations in the standard point model. Weighting factors are detemiined from Monte Carlo calculations using the MCNPX code. Monte Carlo calculations give the dependence of the weighting factors on sample mass and geometry, and simulated assays using Monte Carlo give the theoretical accuracy of the weightedpointmodel assay. Measured multiplicity data evaluated with both the standard and weighted point models are compared to reference values to give the experimental accuracy of the assay. Initial results show significant promise for the weighted point model in reducing or eliminating biases in the neutron multiplicity assay of metal items. The negative biases observed in the assay of plutonium metal samples are caused by variations in the neutron multiplication formore »
 Authors:

 William H.
 Merlyn S.
 Douglas R.
 Publication Date:
 Research Org.:
 Los Alamos National Laboratory
 Sponsoring Org.:
 USDOE
 OSTI Identifier:
 977484
 Report Number(s):
 LAUR041149
TRN: US1002931
 Resource Type:
 Conference
 Resource Relation:
 Conference: Submitted to: 7th international conference on facility operationssafeguards interface, Charleston, SC, February 29March 5, 2004
 Country of Publication:
 United States
 Language:
 English
 Subject:
 72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS; ACCURACY; ENERGY SPECTRA; GEOMETRY; MULTIPLICITY; NEUTRONS; OXIDES; PLUTONIUM; SHAPE
Citation Formats
Geist, W H, Krick, M S, and Mayo, D R. Reduction of bias in neutron multiplicity assay using a weighted point model. United States: N. p., 2004.
Web.
Geist, W H, Krick, M S, & Mayo, D R. Reduction of bias in neutron multiplicity assay using a weighted point model. United States.
Geist, W H, Krick, M S, and Mayo, D R. Thu .
"Reduction of bias in neutron multiplicity assay using a weighted point model". United States. https://www.osti.gov/servlets/purl/977484.
@article{osti_977484,
title = {Reduction of bias in neutron multiplicity assay using a weighted point model},
author = {Geist, W H and Krick, M S and Mayo, D R},
abstractNote = {Accurate assay of most common plutonium samples was the development goal for the nondestructive assay technique of neutron multiplicity counting. Over the past 20 years the technique has been proven for relatively pure oxides and small metal items. Unfortunately, the technique results in large biases when assaying large metal items. Limiting assumptions, such as unifoh multiplication, in the point model used to derive the multiplicity equations causes these biases for large dense items. A weighted point model has been developed to overcome some of the limitations in the standard point model. Weighting factors are detemiined from Monte Carlo calculations using the MCNPX code. Monte Carlo calculations give the dependence of the weighting factors on sample mass and geometry, and simulated assays using Monte Carlo give the theoretical accuracy of the weightedpointmodel assay. Measured multiplicity data evaluated with both the standard and weighted point models are compared to reference values to give the experimental accuracy of the assay. Initial results show significant promise for the weighted point model in reducing or eliminating biases in the neutron multiplicity assay of metal items. The negative biases observed in the assay of plutonium metal samples are caused by variations in the neutron multiplication for neutrons originating in various locations in the sample. The bias depends on the mass and shape of the sample and depends on the amount and energy distribution of the ({alpha},n) neutrons in the sample. When the standard point model is used, this variablemultiplication bias overestimates the multiplication and alpha values of the sample, and underestimates the plutonium mass. The weighted point model potentially can provide assay accuracy of {approx}2% (1 {sigma}) for cylindrical plutonium metal samples < 4 kg with {alpha} < 1 without knowing the exact shape of the samples, provided that the ({alpha},n) source is uniformly distributed throughout the sample and has an average neutron energy close to the O({alpha},n) average neutron energy. Better assay results can be obtained if there is some knowledge of the plutonium geometry, because weighting factor curves can be calculated for any specified geometry.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2004},
month = {1}
}