Actinium-225 has important medical applications as an agent for targeted alpha therapy, but existing supplies of this isotope are limited. The 230Th(p,2n)229Pa reaction has been considered as a route to 225Ac, however, there is no available cross section data for this reaction in the literature. The purpose of this work is to measure the 230Th(p,2n)229Pa reaction cross section in the energy range where this reaction has been calculated to peak to determine the feasibility of using this reaction for 225Ac production.

As there is not a commercial source of 230Th, for this work, thorium, naturally enriched in 230Th, was separated from uranium ore. A novel procedure was developed for this separation consisting of leaching, liquid-liquid extraction, precipitations and ion exchange chromatography. The thorium was fabricated into accelerator targets using electrodeposition. The final targets were deposited onto a thin (10 μm) titanium backing and were highly uniform and stable. Target thicknesses ranged from ~900 to 1900 μg Th/cm2, which was sufficient to make the relevant cross section measurements.

Chemical processing of the irradiated targets was necessary to determine the activity of the protactinium activation products. The target processing was optimized for rapid, high yield and high radiopurity chemical separations. Column chromatography was used for the separation and the final chemical processing procedure is based on a Dowex 1x8 column in HCl media. The chemical studies showed that titanium was an ideal target backing material due to its chemical behavior and dissolution properties. CL resin (TrisKem International) was considered for the target processing, and, while ultimately not used, it was characterized further and batch studies are presented for radium, actinium and thorium in HCl and HNO3 as well as radium, thorium and protactinium in HF. Six column separation studies were also performed with thorium, radium, actinium and protactinium in a variety of solution conditions. Further extraction studies were done with radium and actinium using Pb resin (Eichrom) and Rose Bengal. Both radium and actinium show a strong affinity (k’ >10,000) for the resin at intermediate pHs from solutions with Rose Bengal with the pH range and magnitude of uptake significantly increased by the presence of Rose Bengal compared to basic solutions without this large counter ion.

Thorium and protactinium tracer isotopes were necessary for all of the chemical development studies and were produced with isotope generators. A 237Np/233Pa isotope generator was made and extensively characterized. It was in continual use for over a year with an average yield of ~75% before breakthrough occurred. A 227Ac/227Th isotope generator was also made based on an existing procedure described in the literature.

Nuclear data measurements were done for several relevant protactinium isotopes, including half-life measurements for 228Pa and 229Pa, and modern measurements of the major gamma-ray emissions from the decay of 231Pa. For many of these measurements, the error envelopes have been significantly decreased compared to previous measurements, while remaining in good agreement with the existing literature values.

The 230Th enriched targets were irradiated at the Center for Accelerator Mass Spectrometry at Lawrence Livermore National Laboratory with proton energies ranging from 14.1 to 16.9 MeV. Excitation functions in this energy range are reported for the first time in the literature for the 230Th(p,2n)229Pa and 230Th(p,3n)228Pa reactions. The peak measured value of the 230Th(p,2n)229Pa reaction was 182 土 12 mb at 14.4 土 0.1 MeV. Based on the measured 230Th(p,2n)229Pa reaction cross section, this reaction could reasonably be used for 225Ac production, although significant amounts of relatively isotopically pure 230Th would be needed for significant production.

Antineutrino detectors may be able to fill an important role in the global nuclear nonproliferation regime by discovering an undeclared nuclear reactor or monitoring a known reactor tens of kilometers away. However, such detectors are large capital investments, and a substantial portion of their cost is the photomultiplier tubes that collect the light from antineutrino interactions in the detector volume and convert it into a usable electronic signal. As such, maximizing the light collection efficiency, and thus performance, of each photomultiplier tube is paramount. Wavelength-shifting plates may be able to aid in this goal. Wavelength-shifting plates increase the amount of light collected by each photomultiplier tube, which translates into a greater energy resolution of the detector. At the same time, however, the wavelength-shifting plates smear out the collection of this light in time, as light captured by the plates is delayed for different amounts of time before reaching a photomultiplier tube. This reduces the ability of post-processing algorithms to successfully reconstruct the location at which an event occurred in the detector. This is expected to have a countervailing effect to the improved energy resolution provided by the wavelength-shifting plates. This work explores the trade-off between these two effects, utilizing Monte Carlo simulations validated against experiments in air and in a water-Cherenkov detector to predict the effects that wavelength-shifting plates will have on large-volume water-Cherenkov detector performance. It was found that wavelength shifting plates improve the overall signal-to-background ratio of a 16 meter diameter, 16 meter height cylindrical detector by 13%, allowing a 3000 MWth undeclared nuclear reactor to be discovered 22% faster, in 16 days rather than 20.4 days when a detector without wavelength-shifting plates is utilized. This improvement is significant as it demonstrates that the improvement wavelength-shifting plates provide to light collection outweighs their effects on timing, so that wavelength-shifting plates may be a cost-effective way to improve the overall performance of large-volume water-Cherenkov antineutrino detectors intended for nuclear nonproliferation missions.

This dissertation addresses the need for new non-destructive assay instruments capable of quantifying the fissile isotopic composition of spent nuclear fuel and of independently verifying the declared amounts of special nuclear materials at various stages of the nuclear fuel cycle. High-energy delayed gamma-ray spectroscopy can provide the ability to directly assay fissile and fertile isotopes in the highly radioactive environment of the spent fuel assemblies and to achieve the safeguards goal of measuring nuclear material inventories for spent fuel handling, interim storage, reprocessing facilities, and final disposal and repository sites.

The delayed gamma-ray assay concept is investigated within this context with the objective of assessing whether the delayed gamma-ray assay instrument can provide sufficient sensitivity, isotope specificity and accuracy as required in nuclear material safeguards applications. Preliminary system design analysis indicates that the delayed gamma-ray response is affected by multiple parameters: type and intensity of the interrogating source, the configuration of the interrogation setup, the time pattern of the interrogation, and the resolution and count rate limit of the gamma-ray detection system.

In order to handle the variety of factors associated with the delayed gamma-ray assay of spent nuclear fuel, a high-fidelity response modeling technique is introduced. The new algorithm seamlessly combines transport calculations with analytical decay/depletion, and discrete gamma-ray source reconstruction codes. Its performance was benchmarked in the dedicated experimental campaign involving accelerator-driven photo-neutron sources and samples containing fissile and fertile isotopes.

Analytical estimations of the intensity of the delayed gamma-ray response and the passive background rate are utilized to develop a concept of the non-destructive instrument for the assay of spent nuclear fuel. The modeling technique is then applied to more detailed parametric study. These simulations included extensive spent fuel inventories, and accounted for realistic assay configurations and instrumentation. The results of this preliminary analysis indicate that the delayed gamma-ray assay of spent nuclear fuel assemblies can be performed with available neutron generator and detection technology.

The sensitivity of the delayed gamma-ray spectra to the actinide content of the spent nuclear fuel is investigated. The simplest analysis of the delayed gamma-ray response is based on the analysis of integrated count rates and peak ratios. More powerful analytical and numerical methods are likely needed for determining the relative concentrations of fissile and fertile isotopes in samples with complex compositions.

Neutron-induced reaction cross sections on short-lived neutron-rich nuclei, such as fission fragments play a crucial role for a wide range of nuclear physics applications, from nuclear energy and astrophysics to U.S. stockpile stewardship and other national security missions. However, our ability to model these cross sections are limited by theoretical uncertainties that range over orders of magnitude. Indirect methods are utilized to determine cross sections when direct measurements are not possible. These indirect methods can provide an experimental constraint on nuclear properties used in determining the (n,g) cross section using Hauser-Feshbach reaction modeling including the nuclear level density (NLD), gamma-ray strength function (gSF), and parameterizations of the optical model potential. One example of this is the fission fragment 92Sr. The direct measurement of 92Sr neutron-capture cross section is not possible, as the half-life of 2.66 hours is too short to allow the manufacturing of a target and subsequent irradiation of the target using neutrons. While it may not be possible to measure neutron capture on 92Sr, beta decay of 93Rb (Q=7.466(9) MeV) can be used to probe average nuclear properties of the 93Sr compound nucleus. In this dissertation we present the results of such a study where the emitted beta-delayed gamma rays were measured using a total absorption spectrometer (TAS) known as the Summing NaI(Tl) detector (SuN) to simultaneously determine the gamma-ray energies and excitation energies. The experiment utilized a combination of the SuN detector together with a plastic scintillator to measure coincidence events with the emitted beta particle, and a Tape Station for Active Nuclei (SuNTAN) to remove gamma-ray background from the decay daughter activity. The measured 93Sr gamma-ray energies as a function of excitation energy were analyzed using the beta-Oslo Method to extract statistical nuclear properties that were implemented in the reaction code TALYS. The resulting calculated 92Sr neutron capture cross section is constrained by the experimental uncertainty and systematic uncertainties of the beta-Oslo Method.

Gnowee/COEUS v1.0 is a metaheuristic software package that has been developed at the University of California, Berkeley (UC Berkeley), in collaboration with Lawrence Livermore National Laboratory (LLNL), to design optimized Energy Tuning Assemblies (ETA). ETAs are designed in order to modify the neutron source spectrum and produce an objective spectrum given a large set of constraints and changeable variables. The software package is based on a general-purpose metaheuristic optimization algorithm, Gnowee, which uses COEUS to couple the algorithm to the the Monte Carlo N-Particle (MCNP) radiation transport package. The initial successful application, of the software package, was the design of a conical ETA to spectrally shape a simplified monoenergetic 14.1 MeV neutron point source to a thermonuclear and prompt fission neutron spectrum for technical nuclear forensics (TNF) purposes. Like TNF, many other applications in the nuclear engineering field require neutron energy distributions that cannot be obtained with currently available neutron sources. With an increased demand in neutron beam applications, numerous facilities, including the NIF, are interested in employing an accurate and efficient optimization design methodology, such as Gnowee/COEUS, to tailor available neutron spectra and intensity for specific requirements.

The dissertation research included the following steps: (a) development of a fast and efficient optimization methodology and software package for tailoring neutron energy, (b) identification of the neutron transport simulation code package to be coupled with the optimization part, (c) experimental validation of optimization software package (d) applying the optimization package for specific ETA designs, and (e) performing simple experiments on specific ETA designs.

This dissertation describes further efforts made in developing and efforts that have been made in generating a generalized, problem independent Gnowee/COEUS v2.0. The new software package includes the ability to design an ETA within a high fidelity model of a neutron producing facility, with realistic source configurations, a wider range of possible optimization functions, and a larger set of possible variables and constraints. The first part of the dissertation describes the improved COEUS v2.0 and the possibilities of applying the code to design ETAs within LLNL's National Ignition Facility (NIF) Target Chamber (TC). The improved modeling of the NIF TC environment by a set of Monte Carlo and deterministic codes is described, including the modeling and characterization of its system components and instrumentation, including Diagnostic Instrument Manipulator (DIM) 90-78, Target and Diagnostic Manipulator (TANDM) 90-348, SNOUT and large Target Option Activation Device (HTOADs) used for validation and ETA experiments. In addition, a detailed description of the activation foils used to cover a large range of neutron energy spectra is included, as well as the analysis and the unfolding techniques of the obtained experimental results. The second part of the dissertation focuses on the improvements and experimental validation of the full 3-D Monte Carlo model of the NIF Target Chamber. A discussion about various system errors and material uncertainties that could explain certain differences in modeling and experimental results is included. This thesis shows the ability of the newly developed software package to shape the NIF's high neutron flux output of a mainly monoenergetic 14.1 MeV neutron source peak, to an energy range including 8 to 10 MeV for integral benchmarking applications, or to highly peaked 8 or 10 keV spectra for a Boron Neutron Capture Therapy (BNCT) application. The final results provide a powerful demonstration of the tailoring capabilities of Gnowee/COEUS v2.0. This could allow various neutron producing facilities, such as the NIF, to expand their user base in various nuclear science and engineering applications, including detector characterization and calibration, study of radiation damage to various materials, cross section measurements for neutron activation studies, or medical applications such as BNCT or isotope separation.

Graphics processing units, or GPUs, have gradually increased in computational power from the small, job-specific boards of the early 1990s to the programmable powerhouses of today. Compared to more common central processing units, or CPUs, GPUs have a higher aggregate memory bandwidth, much higher floating-point operations per second (FLOPS), and lower energy consumption per FLOP. Because one of the main obstacles in exascale computing is power consumption, many new supercomputing platforms are gaining much of their computational capacity by incorporating GPUs into their compute nodes. Since CPU-optimized parallel algorithms are not directly portable to GPU architectures (or at least not without losing substantial performance), transport codes need to be rewritten to execute efficiently on GPUs. Unless this is done, reactor simulations cannot take full advantage of these new supercomputers.

WARP, which can stand for ``Weaving All the Random Particles,'' is a three-dimensional (3D) continuous energy Monte Carlo neutron transport code developed in this work as to efficiently implement a continuous energy Monte Carlo neutron transport algorithm on a GPU. WARP accelerates Monte Carlo simulations while preserving the benefits of using the Monte Carlo Method, namely, very few physical and geometrical simplifications. WARP is able to calculate multiplication factors, flux tallies, and fission source distributions for time-independent problems, and can run in both criticality or fixed source modes. WARP can transport neutrons in unrestricted arrangements of parallelepipeds, hexagonal prisms, cylinders, and spheres.

WARP uses an event-based algorithm, but with some important differences. Moving data is expensive, so WARP uses a remapping vector of pointer/index pairs to direct GPU threads to the data they need to access. The remapping vector is sorted by reaction type after every transport iteration using a high-efficiency parallel radix sort, which serves to keep the reaction types as contiguous as possible and removes completed histories from the transport cycle. The sort reduces the amount of divergence in GPU ``thread blocks,'' keeps the SIMD units as full as possible, and eliminates using memory bandwidth to check if a neutron in the batch has been terminated or not. Using a remapping vector means the data access pattern is irregular, but this is mitigated by using large batch sizes where the GPU can effectively eliminate the high cost of irregular global memory access.

WARP modifies the standard unionized energy grid implementation to reduce memory traffic. Instead of storing a matrix of pointers indexed by reaction type and energy, WARP stores three matrices. The first contains cross section values, the second contains pointers to angular distributions, and a third contains pointers to energy distributions. This linked list type of layout increases memory usage, but lowers the number of data loads that are needed to determine a reaction by eliminating a pointer load to find a cross section value.

Optimized, high-performance GPU code libraries are also used by WARP wherever possible. The CUDA performance primitives (CUDPP) library is used to perform the parallel reductions, sorts and sums, the CURAND library is used to seed the linear congruential random number generators, and the OptiX ray tracing framework is used for geometry representation. OptiX is a highly-optimized library developed by NVIDIA that automatically builds hierarchical acceleration structures around user-input geometry so only surfaces along a ray line need to be queried in ray tracing. WARP also performs material and cell number queries with OptiX by using a point-in-polygon like algorithm.

In the initial testing where 10^6 source neutrons per criticality batch are used, WARP is capable of delivering results that are anywhere from 4 to 800 pcm away from MCNP 6.1 and Serpent 2.1.18, but with run times that are 11-82 times lower, depending on problem geometry and materials. On average, WARP's performance on a NIVIDIA K20 is equivalent to approximately 45 AMD Opteron 6172 CPU cores. Larger batches are typically perform better on the GPU, but memory limitations of the K20 card restricted batch size to $10^6$ source neutrons.

WARP has shown that GPUs are an effective platform for performing Monte Carlo neutron transport with continuous energy cross sections. Currently, WARP is the most detailed and feature-rich program in existence for performing continuous energy Monte Carlo neutron transport in general 3D geometries on GPUs, but compared to production codes like Serpent and MCNP, WARP has limited capabilities. Despite WARP's lack of features, its novel algorithm implementations show that high performance can be achieved on a GPU despite the inherently divergent program flow and sparse data access patterns. WARP is not ready for everyday nuclear reactor calculations, but is a good platform for further development of GPU-accelerated Monte Carlo neutron transport. In it's current state, it may be a useful tool for multiplication factor searches, i.e. determining reactivity coefficients by perturbing material densities or temperatures, since these types of calculations typically do not require many flux tallies.

Organic scintillators have been used in conjunction with photomultiplier tubes to detect fast

neutrons since the early 1950s. The utility of these detectors is dependent on an understanding

of the characteristics of their response to incident neutrons. Since the detected light in

organic scintillators in a fast neutron radiation field comes primarily from neutron-proton

elastic scattering, the relationship between the light generated in an organic scintillator and

the energy of a recoiling proton is of paramount importance for spectroscopy and kinematic

imaging. This relationship between proton energy deposited and light production is known

as proton light yield.

Several categories of measurement methods for proton light yield exist. These include

direct methods, indirect methods, and edge characterization techniques. In general, measurements

for similar or identical materials in the literature show a large degree of variance

among the results. This thesis outlines the development of a new type of indirect method

that exploits a double neutron time of flight technique. This new method is demonstrated

using a pulsed broad spectrum neutron source at the 88-Inch Cyclotron at Lawrence Berkeley

National Laboratory.

The double time of flight method for proton light yield measurements was established

using two commercially available materials from Eljen Technology. The first is EJ-301, a

liquid scintillator with a long history of use. Equivalent materials offered by other manufacturers

include NE-213 from Nuclear Enterprise and BC-501A from Saint-Gobain Crystals.

The second material tested in this work is EJ-309, a liquid scintillator with a proprietary

formulation recently introduced by Eljen Technology with no commercial equivalents. The

proton light yield measurements were conducted in concert with several system characterization

measurements to provide a result to the community that is representative of the material

itself. Additionally, the errors on the measurement were characterized with respect to systematic

uncertainties, including an evaluation of the covariance of data points produced and

the covariance of fit parameters associated with a semi-empirical model.

This work demonstrates the viability of the double time of flight technique for continuous

measurement of proton light yield over a broad range of energies without changes to the

system configuration. The results of the light yield measurements on EJ-301 and EJ-309

suggest answers to two open questions in the literature. The first is that the size of the

scintillation detector used to measure the proton light yield should not effect the result if

the spatial distributions of Compton electrons and proton recoils are equivalent. Second, the

shape of the scintillation detector should not effect the light yield with the same constraint

on the spatial distributions.

A characterized hardware and software framework has been developed, capable of producing

proton light yield measurements on additional materials of interest. The acquisition,

post processing, error analysis, and simulation software were developed to permit characterization

of double time of flight measurements for a generic system, allowing it to be utilized to

acquire and analyze data for an array of scintillation detectors regardless of detector size or

geometric configuration. This framework establishes an extensible capability for performing

proton light yield measurements to support basic and applied scientific inquiry and advanced

neutron detection using organic scintillators.

The level density and $\gamma$-ray strength function of $^{243}$Pu have been determined using the Oslo method.\

A 12~MeV deuteron beam from the University of Oslo cyclotron was used to populate excited states in the quasi-continuum of $^{243}$Pu using the $^{242}$Pu$(d,p)$ reaction. The distribution of primary $\gamma$-rays as function of the excitation energy has been extracted from particle-$\gamma$ coincidence data. Based on the Brink-Axel asumption that the primary $\gamma$-ray spectrum is proportional to the product of the $\gamma$-ray transmission coefficient (which only depends on the transition energy) times the level density at the excitation energy of the final state. The $\gamma$-ray strength function is calculated from the $\gamma$-ray transmission coefficient assuming pure dipole radiation. Both, the level density and $\gamma$-ray strength function, are normalized using available experimental data from libraries. \

The level density of $^{243}$Pu follows closely the constant-temperature level density formula. An enhancement of the $\gamma$-ray strength function is seen at low energies that is similar to that previously measured in other actinides. This structure is interpreted as the M1 scissors resonance. Its centroid $\omega=2.42(5)\:\mathrm{MeV}$ and its total strength $B=10.1(15)\:\mu_N^2$ are in excellent agreement with sum-rule estimates.

The measured level density and $\gamma$-ray strength function were then used to calculate the $^{242}$Pu$(n,\gamma)$ cross section within the Hauser-Feshbach formalism.