AbstractThis chapter presents a description of most of the instruments that are currently in use for the measurement of plutonium and uranium using passive methods (without an external source). This includes the acquisition electronics as well as Singles counting methods, coincidence counting methods and multiplicity counting methods. The Singles counting applications include the measurement of waste and curium bearing materials. The coincidence counting applications include bulk plutonium, bulk uranium, waste and holdup measurements and fresh fuel assemblies. The multiplicity application description includes advantages and disadvantages and multiplicity detector design. There is also a description of some non-3He systems. The chapter concludes with a description of additional concepts: neutron imagers, list-mode data analysis, distributed source term analysis, unattended monitoring and MCNP modeling for detector design.

Results from a preliminary accelerator-based experiment conducted at the Harwell Tandem Van de Graaff generator to estimate the thick target integrated over angle yield of 6.129 MeV gamma-rays from alpha-particles in the MeV range stopping in natural carbon are briefly outlined. The results should be useful in planning future detailed experiments to generate data for applications and also to benchmark thin target cross section evaluations. The estimated yields referred to a natural C-13 abundance of 1.09 atom % at 5.597, 5.801, and 5.999 MeV in units of 6.129 MeV gamma-rays per 10(8) alpha-particles are: 1.04, 1.62 and 3.31, respectively. An indicative overall relative uncertainty at the 68 % confidence level is estimated to be about 6%, although we caution that, being undertaken only as a feasibility study, the checks and balances we would usually conduct were not performed.

Results from a preliminary accelerator-based experiment conducted at the Harwell Tandem Van de Graaff generator to estimate the thick target integrated over angle yield of 6.129 MeV γ-rays from α-particles in the MeV range stopping in natural carbon are briefly outlined. The results should be useful in planning future detailed experiments to generate data for applications and also to benchmark thin target cross section evaluations. The estimated yields referred to a natural 13C abundance of 1.09 atom % at 5.597, 5.801, and 5.999 MeV in units of 6.129 MeV γ-rays per 108 α-particles are: 1.04, 1.62 and 3.31, respectively. An indicative overall relative uncertainty at the 68 % confidence level is estimated to be about 6%, although we caution that, being undertaken only as a feasibility study, the checks and balances we would usually conduct were not performed.

The capabilities of current safeguards neutron instrumentation can be expanded by acquiring and analyzing data in new ways. A new List Mode Collar (LMCL) was created by retrofitting a traditional safeguards neutron coincidence counting system, the Neutron Coincidence Collar, with newly developed preamplifiers on each of its 18 He-3 tubes in active interrogation mode. By applying list mode data acquisition (LMDA) and analysis, a signal from each He-3 tube can now be recorded and analyzed to determine the neutron count rate measured in each individual(3)He tube. Further, the count rate distribution across all He-3 tube locations in the system can be determined. This is the first time that a full list mode method has been implemented using a traditional neutron coincidence collar and spatial response data has been obtained. The objective of this research was to develop and evaluate a laboratory demonstration list mode neutron coincidence collar (LMCL) that can extract a greater number of useful signatures than are currently generated using the current collar. A new analysis method called the "List Mode Response Matrix" has also been developed for use in combination with these measurements to improve the detection of missing fuel pins (partial defects) during the nondestructive assay of fresh nuclear fuel assemblies, although analysis algorithms will be reported separately from this experimental evaluation. In this paper, 18 channel spatial response data are presented for a(252)Cf source moving within the LMCL detector cavity. LMDA, combined with the addition of multiple preamplifiers, facilitates this capability by increasing the number of signals that can be measured simultaneously; allowing for in-depth analysis of neutron coincidence events to determine the location and distribution of nuclear material within the measurement geometry. This approach enables a spatial response measurement to form a "pattern" or item signature unique to the fissioning source's distribution; in the case of the LMCL, the distribution of fresh low-enriched uranium fuel within the assembly. A major benefit provided by LMDA is that this can all be achieved from a single measurement pulse train in offline analysis. To experimentally demonstrate this concept, a Mirion Technologies (Canberra) model JCC-71 Neutron Coincidence Collar has been retrofitted with new electronics designed at Oak Ridge National Laboratory. This paper presents the results from the experimental evaluation of the laboratory demonstration system. The results indicate equivalent performance of the LMCL to the original collar in addition to improved spatial response capabilities, while maintaining the original system footprint.

Double pulsing in neutron coincidence counters is the result of a pulse processing chain-3He proportional counter timing incompatibility known to nondestructive assay system designers. It is not currently widely acknowledged, or accounted for, in the user community. However, it is gaining attention as list mode data acquisition and analysis becomes more commonly used for system diagnostics, revealing features that have been overlooked by historic timing gate selection using traditional shift register data acquisition methods. Double pulsing increases the apparent number of measured neutron events from a source. Therefore, it may contribute to a falsely increased count rate if present at the operational high voltage used when assaying samples, even with set predelays. The authors have previously studied the effects of double pulsing, using list mode data acquisition and analysis on neutron pulse trains in three common neutron coincidence counting systems used in routine measurements for international safeguards: a Canberra Industries JCC-71 Neutron Coincidence Collar, a variant on the Canberra Industries JCC-51 Active Well Neutron Coincidence Counter (the Large Volume Active Well Neutron Coincidence Counter), and an AnTech Inc. N2071 Neutron Coincidence Collar. This non-ideal behavior was isolated to the Amptek A111 Charge Sensitive Preamplifier & Discriminator chip used in both the Canberra Industries and Antech systems. The double pulsing fraction was calculated in post-analysis for various high voltage settings in these A111-based systems, assuming no predelay settings. In this work, the authors expand upon this identification and analysis to make it more translatable between list mode data acquisition and analysis and shift register-based analysis. By investigating the double pulsing fractions in a JCC-71 Uranium Neutron Collar as a function of predelay settings, as well as describing and performing alternative tests and analyses to diagnose and quantify the double pulsing using shift register logic, this work hopes to complete the picture of double pulsing identification and analysis.

Neutron coincidence counting is a well-established technique used for the nondestructive quantification of special nuclear material during international safeguards inspections. The neutron counters are commonly designed with an annular body, centered about an inner well or cavity into which a measurement item is placed, and the moderating annulus is populated with He-3 tubes connected to a varying number of preamplifiers. The Canberra Industries JAB-01 preamplifier/amplifier/discriminator board is employed within the company's neutron coincidence counters, built for use by the International Atomic Energy Agency. Non-ideal behavior of these boards was identified during a detector characterization, using list mode data acquisition, of a Canberra Industries JCC-71 Neutron Coincidence Collar implementing four JAB-01 boards. List mode data acquisition and analysis reveals features that have commonly been overlooked by historic timing gate selection while using shift register data acquisition methods, which are routinely adopted in international safeguards. It has been shown that double pulsing effects are not fully captured within the predelay setting; therefore, they may influence the response of the system within the standard operating regime. We set out to identify and correct for double pulsing in our post analysis of neutron pulse trains, while isolating this behavior to the relevant system. To understand and potentially address these concerns, the responses of two different JAB-01 board systems - the JCC-71 Neutron Coincidence Collar and a modified JCC-51 Active Well Neutron Coincidence Counter - are compared with the responses of an AnTech Inc. N2071 Neutron Coincidence Collar that also uses an amplifier built on the Amptek A111 Charge Sensitive Preamplifier & Discriminator chip, and a JCC-71 that employs custom preamplifiers designed at Oak Ridge National Laboratory.

Recent IAEA circulars and policy papers have sought to implement safeguards when any purified aqueous uranium solution or uranium oxides suitable for isotopic enrichment or fuel fabrication exists. Under the revised policy, IAEA Policy Paper 18, the starting point for nuclear material under safeguards was reinterpreted, suggesting that purified uranium compounds should be subject to safeguards procedures no later than the first point in the conversion process. In response to this technical need, a combination of simulation models and experimental measurements were employed to develop and validate concepts of nondestructive assay monitoring systems in a natural uranium conversion plant (NUCP). In particular, uranyl nitrate (UO2(NO3)2) solution exiting solvent extraction was identified as a key measurement point (KMP), where gamma-ray spectroscopy was selected as the process monitoring tool. The Uranyl Nitrate Calibration Loop Equipment (UNCLE) facility at Oak Ridge National Laboratory was employed to simulate the full-scale operating conditions of a purified uranium-bearing aqueous stream exiting the solvent extraction process in an NUCP. Nondestructive assay techniques using gamma-ray spectroscopy were evaluated to determine their viability as a technical means for drawing safeguards conclusions at NUCPs, and if the IAEA detection requirements of 1 significant quantity (SQ) can be met in a timely way. This work investigated gamma-ray signatures of uranyl nitrate circulating in the UNCLE facility and evaluated various gamma-ray detector sensitivities to uranyl nitrate. These detector validation activities include assessing detector responses to the uranyl nitrate gamma-ray signatures for spectrometers based on sodium iodide, lanthanum bromide, and high-purity germanium detectors. The results of measurements under static and dynamic operating conditions at concentrations ranging from 10–90gU/L of natural uranyl nitrate are presented. A range of gamma-ray lines is examined, including attenuation for transmission measurement of density and concentration. It was determined that transmission-corrected gamma-ray spectra provide a reliable way to monitor the 235U concentration of uranyl nitrate solution in transfer pipes in NUCPs. Furthermore, existing predictive and analysis methods are adequate to design and realize practical designs. The 137Cs transmission source employed in this work is viable but not optimal for 235U densitometry determination. Validated simulations assessed the viability of 133Ba and 57Co as alternative densitometry sources. All three gamma-ray detectors are viable for monitoring natural uranium feed; although high-purity germanium is easiest to interpret, it is, however, the least attractive as an installation instrument. Overall, for monitoring throughput in a facility such as UNCLE, emulating the uranium concentration and pump speeds of the Springfields conversion facility in the United Kingdom, an uncertainty of less than 0.17% is required in order to detect the diversion of 1 SQ of uranyl nitrate through changes in uranium concentration over an accountancy period of one year with a detection probability of 50%. Although calibrated gamma-ray detection systems are capable of determining the concentration of uranium content in NUCPs, it is only in combination with verifiable operator declarations and supporting data, such as flow rate and enrichment, that safeguards conclusions can be drawn.

In this paper we develop and investigate several criteria for assessing how well a proposed spectral form fits observed spectra. We consider the classical improved figure of merit (FOM) along with several modifications, as well as criteria motivated by Poisson regression from the statistical literature. We also develop a new FOM that is based on the statistical idea of the bootstrap. A spectral simulator has been developed to assess the performance of these different criteria under multiple data configurations.

At many points in the safe and transparent handling of plutonium materials the relative isotopic composition of the principle isotopes needs to be known. Sometimes this information may be of primary interest - such as in the verification of safeguard declarations or in the confirmation of the reactivity of mixed oxide fuel. At other times, e.g., for radioactive waste characterization, the isotopic composition may be needed to calculate specific thermal power or specific spontaneous fission rates for the item under study, which can subsequently be combined with calorimetric and correlated neutron counting measurements, respectively, in order to make quantitative assessments of the mass of Pu and associated nuclides that are present in an item.The Multi-Group Analysis code MGA is a highly regarded and widely used computer code for the analysis of high resolution gamma ray spectra in order to extract the relative isotopic composition of plutonium for a diversity of items with minimal prior information. It has been honed over many years to give reliable results for a broad range of measurement scenarios commonly encountered in the fuel cycle. The nuclear industry is not dormant however and the demands on such codes continue to shift as a combination of technology and necessity open up new application areas. For example, while MGA had its origins in the analysis of clean spectra on product material principally for nuclear safeguards applications taken with germanium detectors having good low-energy resolution, it is now widely applied to the characterization of drummed waste forms and the complex spectra from such items acquired with much larger volume and poorer resolution detectors often used in such applications for the dual use of quantitative assay of the many gamma-emitters.This new domain of operational experience resulted in the need to enhance MGA to deal with spectra of poor statistical quality and also to cope with some of the complications that arise in the analysis of unusual spectra. Together with some additional changes made to incorporate feedback since the release of version 9.63 (which had minor revisions denoted by the letters A through H) of the code this has resulted in the creation of MGA v10.In this paper we shall outline the main changes to the code explaining why they were conceived and implemented. We illustrate what kinds of measurement problems can now be addressed over and above the previous capabilities which have been preserved and verified by the same set of regression tests that have been applied to previous generation of the code.

A new extended-range multi-modality tomographic gamma scanner (TGS) system is presented. New features for this system include a dual mode transmission beam that extends the dynamic range of TGS assays to drum densities that were previously unfeasible. The details of this system are described with particular emphasis placed on the unique elements of the system. Representative data highlighting the performance of the system with high-density drums are presented and discussed. These results demonstrate that extending the range of applicability for the TGS methodology is achieved.

The failure of the point model for highly multiplying Pu metal items is largely a result of the fact that at larger masses the leakage multiplication is not constant throughout the item. This represents a violation of the underlying assumptions of the point model. One way to tackle this dilemma is to introduce finite extent parameters into the model by integrating the multiplication terms present in the basic point model equations over the spatial dimension(s). Here we consider simple cases of weapons grade Pu spheres and squat cylinders in both a non-reflective geometry and surrounded by a basic packing material reflector. The reduced spatial moments, gn = /, as a function of Pu mass are evaluated and the prospect of using these functions as weighting factors for terms in the point model is considered. INTRODUCTION It is well established that application of multiplicity counting to the assay of Pu metal in the kg range results in an assay bias [1,2]. This is partially attributed to a violation of the spatial invariance assumption in the point model expressions used for the detector response function during the analysis step. A weighted point model approach has been developed by Krick and colleagues at Los Alamos National Laboratory [3] in which some of the parameters in the point model are replaced by more complex forms chosen semi-empirically to match the results of analog Monte Carlo simulations of the multiplicity counting experiment. In this way the point model equations are treated as physically based guides to the shape of the calibration curve with empirically determined, problem dependent, coefficients. In applying the point model to a spatially extended problem, several of the assumptions upon which it is built may be challenged – for example the assertion that the neutron detection efficiency is single valued. For the assay of compact metallic items perhaps the largest influence comes from the spatial variation in the multiplication throughout the object. This is the only aspect of the problem we shall consider. Our aim here is to delineate this effect in isolation to better understand the physics at work.

The application of quantitative high-resolution gamma-ray spectrometry for the non-destructive assay of plutonium bearing items, such as waste drums, is complicated by self-attenuation if the plutonium is present as lumps. By definition, lumps are small compared to the bulk matrix and so are not accounted for in the gross matrix correction yet can exert a significant influence on the assay result. Compared to a calibration using dilute standards, self-attenuation results in an under-reporting of the mass of plutonium present. The availability of representative standards is unrealistic for diverse waste streams and so a means to detect and compensate for the presence of lumps is needed. An experimental approach that can in principle generate an item specific correction factor is to exploit the differential attenuation between a set of gamma-lines of known relative emission intensity. In the case of routine measurements of drummed Pu wastes the choice of lines is often limited, the most appropriate often being those at 129 keV and 414 keV from 239Pu. This paper discusses the problems and potential of exploiting this pairing in a simple dual energy approach to the long standing and challenging problem of self-attenuation.

A general purpose passive neutron box counter has been designed, constructed and factory calibrated. The instrument is intended to sort and assay Transuranic Uranium (TRU) waste according to the Waste Isolation Pilot Plant (WIPP) criteria in containers ranging from drums to large boxes and crates. A multi-position Cf Add-A-Source (AAS) capability has been built into the system to determine gross matrix correction factors. The Cf source capsule runs in a U- shaped guide tube beneath the powered roller conveyor used to move the containers into the assay cavity from the loading station. The factory calibration involved measuring a wide range of matrix materials and densities in 208-liter (55 US-gal.) barrels, Standard Waste Box (SWB), Standard Large Box (SLB-2), and Ten Drum Overpack (TDOP) containers. For each container a Volume Weighted Average (VWA) rate and an AAS perturbation factor was determined and the relationship between them was established for the full range of conditions expected to be encountered operationally. In this paper we describe the calibration procedure which used Cf-252 as a surrogate for Pu-240 to map out the spatial responses for the various container-matrix combinations. A particular challenge was the scale of the measurement campaign which was directly related to the large volume of some of the containers. The reduction of the data was also challenging because for the larger items with high concentrations of hydrogen steep spatial gradients were observed in the response. For this reason simple volume-element averaging of the data to derive VWA quantities was inadequate and numerical-integration approaches of the 3-dimensional maps were explored. The response maps were also used to create point-source contributions to the Total Measurement Uncertainty (TMU), but this work is not the subject of this paper. The wide range in container size also required varying numbers of AAS interrogation position. For the drums a single AAS position was used while for the SWB and TDOP six positions were used to achieve better coverage of the cross-section. The SLB-2 has a larger cross section still and ten positions of the AAS were used for this container. The AAS calibration parameterizations for the various containers will be described along with the difficulties associated with a calibration of large containers.

When Pu is present in waste items as lumps, the Reals neutron coincidence signal may be bolstered by virtue of self induced fission events. This gives rise to a positive bias to the assay result and leads to a one sided contribution to the total measurement uncertainty. In the general it is not feasible to determine to magnitude of this effect experimentally and allowance for it must therefore be estimated using separate ad hoc rules. In this paper we to develop a simple model for the self-multiplication enhancement in small lumps allowing the importance of the effect in waste assay to be quantified. In addition, an approach is suggested for how to propagate an uncertainty contribution in to the final result.

Neutron multiplicity analysis has been a valuable technique for safeguards measurements of plutonium oxide and mixed oxides. Historically, most of these measurements have been performed using shift register based electronics. The shift register data acquisition lacks certain flexibility because the basic coincidence parameters (e.g., pre-delay, gate width, and long delay) must be fixed prior to the start of the measurement and the values may potentially, therefore, be sub optimal. List mode or time stamped data acquisition records the arrival time of each pulse thereby preserving the history of the pulse stream and allowing analysis and reanalysis using software analogs to the shift register circuit with adjustable parameters. Until recently, the data rates encountered in the assay of modest amounts of plutonium using efficient multiplicity counters were beyond the capacity of readily available personal computers. The calibration of the large epithermal neutron multiplicity counter (LEMC) for assay of plutonium scrap materials is used as a vehicle to compare the performance of the multiplicity shift register and a commercially available list mode acquisition module.

Passive Neutron Coincidence Counting (PNCC) and Passive Neutron Multiplicity Counting (PNMC) based on (Multiplicity) Shift Register (M)(SR) pulse train correlation analyzers is a long established and important non destructive assay method used in the quantification of plutonium and other spontaneously fissile materials across the fuel cycle. Very high efficiency neutron chambers (>60%) are now available and are being applied to ever more demanding items including impure materials with a high (α, n) rate and articles with a high self-leakage multiplication. This trend means that high instantaneous count rates are commonly encountered such that the multiplicity histogram extends to high order; in other words the number of events detected in a single coincidence gate can be large. This poses a problem in that the likelihood of accidental (chance) coincidences due to random events and overlapping (super) fission histories increases and precision is lost in correcting for them. The epithermal design is one attempt to reduce the capture time distribution to minimize the accidentals coincidence rate but the field of application is so broad that high instantaneous rates are still encountered. This inevitably results in the need to apply a correction to the observed Singles, Doubles and Triples rate for dead time losses. When the instantaneous counting rate is high the uncertainties in the applied corrections can be the accuracy limiting factor in the derived counting rate. Controlling and compensating for dead time losses so that target accuracy is achieved is a crucial aspect of a successful design and implementation process. Dead time losses can be reduced substantially on new systems intended for special use by using dedicated preamplifier-discriminators for each 3He-filled proportional counter together with de-randomizer circuitry and fast encoding electronics. These adaptations are costly, however, and may be difficult to retrofit to existing systems. In this work we therefore take a fresh look at the way in which corrections for dead time losses are applied to the recorded MSR data. We note several interesting empirical correlations observed in experimental data which allow dead time parameters to be extracted. We also comment on the self consistency constraints which exist and can be exploited between PNCC and PNMC results and also between expressions for the

We present analytical expressions for the gate utilization factors (GUFs), up to fourth order, for both signal triggered and random triggered histograms based on a three-component capture time profile. These are useful for refined design performance calculations of passive neutron multiplicity counters using shift register correlation analysis. To our knowledge, these expressions are new.