Philip N. Martin

Active detection of special nuclear material - Recommendations for interrogation source approach for UK prototype active detection system

A WE is developing a prototype active interrogation system to enable robust detection of shielded nuclear material within the context of border choke point detection. This paper describes a study which took place in order to determine the optimum type of radiation and pulse structure to be used with the prototype system. A wide variety of neutron and gamma interrogation sources were considered including the use of sub 10 MeV end-point energy bremsstrahlung or 19F(p,αy)16O characteristic gamma sources and the use of DD, DT, and lower energy beam-target neutron sources such as those produced via. 7Li(p,n)7Be. In each case, where appropriate, both flash systems capable of delivering intense, sub 100 ns pulses of radiation and non-flash repetitively pulsed or continuous wave (CW) sources were considered. Experimental measurements of photo-fission signatures on bare and shielded depleted uranium produced by 100 ns flash sources of 8 MeV bremsstrahlung and 19F(p,αy)16O characteristic gamma sources together with a systematic series of simulations of photofission signatures and associated backgrounds and a review of technological limitations of relevant accelerator technologies were all assessed. An estimate derived from experimental measurements of a minimum number of fissions necessary to detect target quantities of special nuclear material through shielding thickness of interest in the context of border detection is used to determine a charge which must be delivered by the interrogation source and this in turn is used to rank potential interrogation source options. Wider arguments concerned with the ease with which fission signatures may be discriminated above active backgrounds and the utility of different options given operational constraints are then presented and it is concluded that the UK will recommend a flash <;100 ns), 10 MeV end-point energy bremsstrahlung source for the UK active interrogation prototype system.

Pulsed power active interrogation of shielded fissionable material

We report on a collaborative test campaign conducted at the Naval Research Laboratorys Mercury pulsed power facility in December of 2012. The experiment sought to use Mercury in the Intense Pulsed Active Detection (IPAD) [1] mode to interrogate a fissionable material target (depleted uranium, DU) and benchmark the effects of shielding the target with either a low-Z (2% borated high-density polyethylene, BPE) or high-Z (steel) material. A large suite of instrumentation, including 3He, BF3, NaI(Tl), and liquid scintillation detectors were used to measure the delayed γ and neutron signatures from the DU. The test campaign consisted of a series of single IPAD pulses, i.e., “shots,” employing incremental shielding configurations of BPE (up to 50 g/cm2) and steel (up to 150 g/cm2) encapsulating the DU target. We show the results from each detector array, for varying amounts of shielding, in terms of the signal-to-noise vs. time.

Critical Current Operation of the Optimized Self-Magnetic-Pinch Radiographic Diode

The self-magnetic-pinch (SMP) diode is being developed as an intense X-ray source for pulsed power driven radiographic systems. This diode is a critical current limited device described using equations derived from analytic approximations to large aspect ratio (cathode diameter to anode-cathode gap) variant diodes. The radiographic and large-aspect ratio variant of the SMP diode, differ in aspect ratio by around an order of magnitude. Derivation of the critical current limit for the large aspect ratio variant SMP diode requires a number of assumptions; not all of which are valid for the radiographic SMP diode. The applicability of existing critical current approximations to the radiographic SMP diode is discussed. A new critical current description for an optimized radiographic SMP diode configuration is proposed. This new critical current approximation is shown to account well for the radiographic performance (impedance and spot size) of optimized radiographic SMP diodes without the need for an experimentally derived empirical scaling factor.

Development of the Self Magnetic Pinch Diode as a High Brightness Radiographic Source

Summary form only given. The self magnetic pinch (SMP) diode has been developed from a low voltage (<2 MV) to a high voltage (7-8 MV) radiographic source as part of a program to build a hydrodynamics research facility (Hydrus) at AWE Aldermaston. Development of the initial AWE diode design has used the facilities and assistance of a number of UK and US laboratories and companies to carry out both experimental and theoretical investigations into the operation of the diode. Experimental campaigns have been carried out to both investigate the physics operation and to demonstrate performance on the RITS-3 and RITS-6 drivers at SNL, the Mercury and Gamble II drivers at NRL and the Mogul D and EROS drivers at AWE. Modelling of the diode has been carried out mainly using the particle in cell (PiC) code LsP. It has been demonstrated that the diode has achieved the required radiographic performance for Hydros and it will be used as the initial operational diode when the facility is commissioned. Research to develop a long term diode with improved radiographic performance is still continuing.

Active Interrogation of Depleted Uranium Using a Single Pulse, High-Intensity Photon and Mixed Photon-Neutron Source

Active interrogation is a method used to enhance the likelihood of detection of shielded special nuclear material (SNM); an external source of radiation is used to interrogate a target and to stimulate fission within any SNM present. Radiation produced by the fission process can be detected and used to infer the presence of the SNM. The Atomic Weapons Establishment (AWE) and the Naval Research Laboratory (NRL) have carried out a joint experimental study into the use of single pulse, high-intensity sources of bremsstrahlung x-rays and D(\gammab, n)H photoneutrons in an active interrogation system. The source was operated in both x-ray-only and mixed x-ray/photoneutron modes, and was used to irradiate a depleted uranium (DU) target which was enclosed by up to 150 g·cm - 2 of steel shielding. Resulting radiation signatures were measured by a suite of over 80 detectors and the data used to characterise detectable fission signatures as a function of the areal mass of the shielding. This paper describes the work carried out and discusses data collected with 3He proportional counters, NaI(Tl) scintillators and Eljen EJ-309 liquid scintillators. Results with the x-ray-only source demonstrate detection ( > 3\sigmab) of the DU target through a minimum of 113 g·cm - 2 of steel, dropping to 85 g·cm- 2 when using a mixed x-ray/photoneutron source. The 3He proportional counters demonstrate detection ( > 3\sigmab) of the DU target through the maximum 149. 7 g·cm - 2 steel shielding deployed for both photon and mixed x-ray/photoneutron sources.

Neutrons for active detection of special nuclear material: An intense pulsed 7 Li(p,n) 7 Be source

An ongoing program me looking at the active detection of special nuclear material (SNM) is being undertaken by the Atomic Weapons Establishment (A WE) in collaboration with the Naval Research Laboratory (NRL). As part of this programme, pulsed-power driven neutron experiments were conducted at the NRL Mercury accelerator. Mercury was used in a positive polarity mode to produce and accelerate protons into lithium metal foils, generating neutrons via the 7Li(p,n)7Be reaction. 13 shots were carried out at varying machine voltages and over 30 separate neutron and gamma-ray diagnostics were fielded to characterise the angular distribution and energy spectrum of the neutrons generated. Machine performance, neutron, and gamma-ray data are presented and discussed. Neutron yields of up to 1011 neutrons/steradian were recorded, with yields at 60° off axis being approximately 50% of the on axis yield. Previously published analysis [1] of data has been used to validate GEANT4 modelling of the experiments (2). Machine performance data has been used in conjunction with modelled neutron spectra to predict the performance of the Mercury 7Li(p,n)7Be source as a system for detecting SNM.

Prompt sub-MeV neutron production from the 7Li(p,n)7Be reaction on mercury

Summary form only given. Proton-beam-generation experiments have been conducted on the NRL Mercury pulsed-power generator operating in positive polarity with a lithium metal target embedded in the cathode. The accelerating voltage was limited to below 2.7 MV in order to limit the energy of neutrons produced in the 7Li(p,n)7Be reaction (Q = 1.88 MeV) to below 1 MeV. Analyses based on published results1 and calculations presented here are used to predict the angular distribution of neutron yield and spectrum for each shot. Predicted neutron yields are compared to Rh-counter and Al-activation measurements. The results of these comparisons are quite encouraging, showing better than factor-of-2 agreement between the two sets of measurements and the analysis over the voltage range of the shot series. In order to achieve this level of agreement, a series of MCNPX computations has been carried out to determine the spectral contribution of neutrons reflected from the Mercury test-cell environment2 and the associated changes in detector calibrations. The agreement between measurements and modeling provides a check on the voltage calculated using a positive-polarity ion-diode model. For operation at 2.5–2.6 MV, on-axis neutron yields from the p-Li reaction are in the 1011 neutrons/steradian range.

Particle in cell modelling of the quantitative evolution of a pinched electron beam across the a-k gap of a self magnetic pinch diode

Summary form only given. Particle in Cell (PiC) simulations using the Large Scale Plasma (LSP) code of a Self Magnetic Pinch (SMP) diode have been carried out to investigate the electron beam conditions present in the SMP anode-cathode (a-k) gap. The probability distributions of radial charge, macro particle energy and macro particle velocity are described as a function of emission region and position across the a-k gap. The conclusions of these investigations are that the top surface of cathode contributes around 50% of the electron flow while the lower surface accounts for only around 5% with the remaining flow coming from the tapered cathode tip. The energy of the electron beam starts as a low energy peaked distribution at the start of the a-k gap and goes through a broad distribution to one with a high energy bias, retaining a low energy tail largely independent of emission site. The angles which the electrons have across the a-k gap follow distributions which show structure for the emission sites on the top half of the cathode with the quantitative change in angle now being obtainable as a function of position. For emission sites on the lower half of the cathode, and as a position across the a-k gap, the angle distribution data has a broad shape biased to low angle with no clearly observable peaks.

Comparisons of the self pinch diode and the paraxial diode electron distributions at the conversion target using LSP

Simulations of radiographic diodes are routinely used at AWE to investigate the operation of x-ray sources used to radiograph hydrodynamic experiments. The x-ray generation is heavily dependant on the distribution of the electrons striking the Bremsstrahlung conversion target and being able to accurately describe this distribution is key to any given simulation being useful. The time resolved distributions of electrons striking the Bremsstrahlung conversion target for both the self magnetic pinch diode and paraxial diodes are described and compared to both each other and the common fist approximations of a shell of electrons and fully filled cone respectively. It is found that electron distributions in both diodes are not described well by such simple approximations, with the Large Scale Plasma [1] code (LSP) simulations predicting a double peaked distribution going out to around 25 degrees for the paraxial diode and a distributed distribution extending all the way to 90 degrees for the pinch diode.

Towards Optimization of the Paraxial Diode for X-Ray Radiography

The paraxial diode is a gas-filled focusing cell which is routinely used to transport high energy density electron beams for use in pulsed-power driven flash X-ray radiography experiments. Minimization of the radiographic spot size and maximization of the radiation dose is a continuing long-range goal for development of this diode. Suggestions for optimizing the radiographic utility of the paraxial diode are presented within the constraints of the physics of beam focal sweep and emittance growth.