Calibrating Passive Neutron Multiplicity and Waste Counters Using Calibrated ²⁵²Cf Sources: Estimating a_Pu-Values Without Pu

ABSTRACT

²⁵²Cf sources are especially interesting for calibrating passive multiplicity counters without the need for Pu because:

• they are genuinely point-like and being lightly encapsulated offer a nearly isotropic unperturbed fission spectrum of neutrons
• for all practical purposes they can be considered as a pure source of spontaneous fission neutrons which considerably simplifies interpolation. That is the leakage self multiplication, M L, can be taken as unity and the ratio a of ( á, n)-to-(SF, n) neutrons can be taken as zero.
• it is a well studied multiplicity system with the factorial moments íi of the distribution P(í) being well established.
• the energy spectrum is similar to that of the spontaneous spectra of the Pu isotopes
• it is readily available with well known outputs , Y, determined absolutely by reference to Mn-baths operated by National Standards Laboratories

In this work we exploit these features in the framework of the point-model interpretational equations discussed in ASTM C 1500 ‘Standard Test Method for Nondestructive Assay of Plutonium by Passive Neutron Multiplicity Counting’. In particular we: extend the equations to include the delayed neutron contribution showing how to extract the gate utilization factors (GUFs) accurately from the data available; extend the solutions given in the Appendices to cases where the Triples GUF is not equal to the square of the Doubles GUF; and, illustrate how to project ²⁴⁰Pu _eff performance from measured ²⁵²Cf.

INTRODUCTION

The field of international nuclear safeguards demands the highest accuracy from non destructive radiometric assay in order to minimize the amount of unaccounted special nuclear material. The reason is that over many assays random uncertainties average out while systematic errors do not. Therefore, when the flow of materials is high even a small unrecognized systematic bias can result in a significant quantity of material being unaccounted for over a short period of time. Increasingly Monte Carlo calculations are being applied to develop the calibration of waste and safeguards passive neutron assay systems. Modern Monte Carlo particle interaction and transport codes, for example MCNPX [1], are very powerful and within the limits of our knowledge of basic nuclear data can faithfully model all of the important physical processes taking place in the item and detector. They provide a means to interpolate or extend the calibration when representative reference materials are unavailable or impractical to apply. Although in principle they may be used to estimate the response absolutely, given that the details of construction are adequately specified in practice, this is rarely done. It is considered better practice to normalize the model calculations to a carefully performed and highly controlled experiment(s). In this way the overall accuracy can often be improved because certain sources of systematic error, largely common to all calculations (e.g. moderator density and as built dimensions, effective volume of ³He proportional counter gas etc.) cancel out when one deals in ratios. ²⁵²Cf spontaneous fission sources provide a readily available and convenient surrogate for Pu and other neutron emitters in many cases. The emission spectrum is reasonable well known and they are available with very small dimensions lightly encapsulated. For many practical applications they can therefore be regarded as point-like. This considerably simplifies the interpretation of experimental results and often satisfies the assumptions of simple analytical models so that a means exists to gain added confidence with the Monte Carlo results. Based on the results of an intercomparison exercise [2] we conclude that the current state of the practice of several National Laboratories around the world for the absolute determination of the total neutron output of commercially available Cf sources is 0.3 – 0.4 % relative at the 68% confidence level. Pu items containing masses known to greater accuracy than this are possible but for benchmarking purposes remain limiting because of large relative uncertainties in spontaneous fission half-lives and multiplicity yield per fission. Given that Cf is readily available and can be neutron calibrated to a fraction of a percent we are therefore motivated to develop methods that take full advantage of this fact. Also we recognized that the ²⁴⁰Pu-effective worth of an item could easily be re-expressed in terms of a ²⁵²Cf-effective mass with the conversion being amenable to simple cross calibration. With the application of suitable detector specific response characteristics then the relationship would be a basic nuclear parameter.

It is worth noting explicitly that Cf reference sources of known emission rate are commonplace whereas Cf sources whose mass is accurately known are rare. In contrast Pu reference materials are traditionally available as oxides of well known weight, well known Pu composition and well known Pu:oxide weight fraction. This strongly influences our thinking. In principle however there is no technical reason why an isotopically pure, non-multiplying metallic ²⁴⁰Pu sample with negligible ( á, n) contribution cannot be manufactured with a mass known to comparable accuracy (certainly<0.1%, say) of the current best reference materials. The availability of such materials for basic research would be highly desirable in our endeavors to reduce the uncertainties in basic nuclear data parameters of interest in neutron counting applied to safeguards and waste assay.

In this paper we consider the application of point model neutron multiplicity equations (see [3], [4] and references given there for details) and in particular how ²⁵2Cf of known emission rate can be used to characterize a passive neutron counter operating with shift register neutron correlation analyzer electronics. In particular we revisit the approximate form of the point model equations present in Appendix X1 (“Other Multiplicity Solutions”) of ASTM C1500-02 [5].