We address the problem of stochastic simulation of soil particle-size curves (PSCs) in heterogeneous aquifer systems. Unlike traditional approaches that focus solely on a few selected features of PSCs (e.g., selected quantiles), our approach considers the entire particle-size curves and can optionally include conditioning on available data. We rely on our prior work to model PSCs as cumulative distribution functions and interpret their density functions as functional compositions. We thus approximate the latter through an expansion over an appropriate basis of functions. This enables us to (a) effectively deal with the data dimensionality and constraints and (b) to develop a simulation method for PSCs based upon a suitable and well defined projection procedure. The new theoretical framework allows representing and reproducing the complete information content embedded in PSC data. As a first field application, we demonstrate the quality of unconditional and conditional simulations obtained with our methodology by considering a set of particle-size curves collected within a shallow alluvial aquifer in the Neckar river valley, Germany.

Surface-bound particulates containing radionuclides in the environment can become airborne through the process of resuspension. Once airborne, these radionuclides can be inhaled or ingested to deliver an internal dose of ionizing radiation. To that end, the resuspension factor method is a powerful tool for predicting a person's exposure to airborne particles from surface contaminations, and therefore is used to determine protective and intervening measures. The resuspension factor is calculated as the ratio measured airborne to surface mass concentration and has been found to generally decrease exponentially with time. Current models of the resuspension factor are empirical and have failed to predict recent measurement, motivating a stronger basis and physical model for the system. Additionally, federal guidances conservatively suggest an unphysical model of particulate radioactivity impact wherein the entirety of the radiation is absorped. For this dissertation, two- and three-compartment catenary models were derived which build on measured resuspension rate constants under various influences. These models were fit to a set of historic observations of resuspension factors using an instrumental uncertainty-weighting to resolve the large variances early in time which otherwise inflate calculations. When compared to previous resuspension models, our physical models better fit the data achieving reduced-chi-squared closer to 1. An experiment was undertaken to validate our basic environment resuspension models in an urban environment without wind. A resuspension chamber is constructed by placing an acrylic tube atop a poured concrete surface and lowering a low-volume air sampler head from above. Europium oxide powder was dispersed upon the surface or from above the air sampling height to emulate ideal compartmentalized release scenarios, and air is sampled on an hourly, daily, or weekly basis. Sampler filters then were evaluated for Europium content using neutron activation and gamma spectroscopy. Hourly measurements following airborne release are within an order of magnitude of early-timeframe historic resuspension factors (~10^−6 m^−1), whereas daily and weekly measurements from surface release demonstrate a gradual decrease in resuspension factor (∼10^−8 m^−1). These results support a need to critically assess the resuspension factor definition and its relationship to "initial suspension" and the indoor background, non-anthropogenic resuspension. Finally, a simulated model was generated to demonstrate loss of alpha radiation from relevant transuranic radioparticles. This was accomplished using the Geant4 Monte Carlo particle transport code. This basic model demonstrated a clear loss of average intensity and energy of exiting particles which are both directly related to the absorped dose. The data shows a loss from 10 to 90% of intensity to occur at particle sizes approaching the range of alphas within them, and a loss of roughly half the initial alpha energy at around the same particle sizes. The results establish a first-order baseline for a particulate self-absorption model which complement existing dosimetry models for inhaled radionuclides.

A laboratory process exists for the extraction of kafirin protein from sorghum grain in order to form kafirin encapsulating microparticles. This laboratory process extracts approximately 2 g of protein and takes in excess of 60 hours from start to finish. A scaled-up extraction process based on the current laboratory process, consisting of a 100 L extraction vessel, was established in order to extract large volumes of kafirin protein from sorghum grain. Approximately 2.5 kg of kafirin protein, which contained approximately 80 % protein after defatting, was extracted from red sorghum grain. This blended kafirin protein, which was the product of combining 9 batches done on the up-scaled process, was needed in order to obtain a consistent base raw material for further experimentation. The blended kafirin was used to investigate the formation of kafirin encapsulating microparticles. This was achieved by means of the solvent phase separation technique with acetic acid as the solvent phase. A series of experiments, selected from a partial factorial design, were used to screen how the formation of microparticles was affected by various parameters. The parameters investigated were solvent to protein ratio, stirring speed, water addition rate and number of water droplets. The morphology of the various microparticles produced was analysed by means of light microscopy, FTIR and particle size analysis, and the different formed microparticles characterised. From the screening partial factorial experimental design, it was determined that the acetic acid concentration was crucial for the formation of microparticles. Microparticles did not form at a low mass ratio (2.3) of glacial acetic acid solvent to protein. Water addition rate and stirring rate also affected microparticle formation while the number of water droplets was insignificant. Therefore, using a high solvent to protein mass ratio (6.8), additional refined partial factorial experiments were conducted. These experiments focused on the effect of water addition rate and stirring speed on the final kafirin microparticle size. Ultimately, a polynomial model was developed to predict the final kafirin microparticle size using only the water addition rate and stirring speed as inputs. The model had an R2 value of 0.986 and was found to relatively accurate during validation. The model also identified that three distinct regions existed within the workspace: _ A region containing large particles due to protein mass agglomeration and crosslinking, which occurs at low stirring speeds (< 400 rpm) and high water addition rates (> 5 mL/min) _ A region where only small individual microparticles exist, which occurs at high stirring speeds (< 800 rpm) and low water addition rates (> 2 mL/min) _ A region where moderate particles existed as uniform agglomerates of the microparticles, which occurs at moderate stirring speeds (+- 600 rpm) and moderate water addition rates (+- 3.5 mL/min) Ultimately these kafirin microparticles, prepared from protein extracted in an up scaled process, were used to form qualitative microparticle films. The microparticle films were made without plasticiser and without dewatering the microparticles. Furthermore these films were made from microparticles in the regions identified in the model. This qualitative film formation showed that agglomerated microparticles can form films. This could be beneficial for the feasibility of a commercialised process for kafirin microparticle films since the production time would be shorter and less energy intensive. / Dissertation (MEng)--University of Pretoria, 2016. / Chemical Engineering / MEng / Unrestricted

The specific surface area, determined by the BET method, is a parameter often used to scale results of mineral studies of surface reactions in terms of rate and capacity to the field scale. Such extrapolations of results from small-scale laboratory experiments to the field-scale are important within many environmental applications. An example of this is for the prediction of radionuclide retention in the bedrock surrounding a deep repository for radioactive waste, following failure of the engineered barriers, where radionuclides may sorb onto minerals constituting the geological environment. As a first step, the approach used in this work is to study the relationship between specific surface area and the particle size (0.075-8 mm) and to approach the field scale via measurements on large, centimetre-sized pieces, for seven natural minerals commonly found in granite: apatite, biotite, chlorite, K-feldspar, hornblende, labradorite and magnetite. The underlying assumption is that sorption of radionuclides can be related to specific surface area of a particular mineral in a continuation of this project.The results show that the phyllosilicates biotite and chlorite have a specific surface area that is about 10 times larger than the other minerals. Over the range of particle size fractions studied, the specific surface area varies between 0.1 and 1.2 m2g-1 for biotite and chlorite. The other studied minerals have specific surface areas varying between 0.01 m2g-1 for the largest fraction and up to 0.06 - 0.12 m2g-1 for the smallest. Results show linear relationships between the specific surface area and the inverse of the particle size for all studied minerals for small particle sizes, as expected. For some minerals, however, the data seemingly can be divided in two linear trends, where a change in internal surface area, surface roughness and/or particle geometry as the particle size decreases may explain this behaviour. Interestingly, for larger particles, there is a deviation from the linearity observed for small particles. Tentatively, this behaviour is attributed to a disturbed zone, created by the mechanical treatment of the material during particle size reduction, extending throughout small particles, but not altering an undisturbed core of the larger particles. In agreement with this, measurements on large pieces show a surface area 5 to 150 times lower than expected from the linear trends observed for the crushed material, implying an overestimation of the surface area and possibly also of the sorption capacities of the rock material from simple extrapolations of experimental results employing finely crushed material to the field situation. / QC 20110929 / Äspö Radionuclide sorption

Laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) has risen to among the top tier techniques for the direct analysis of solid samples. However, significant problems remain that must be solved to achieve the full analytical potential of LA-ICP-MS. Inefficient conversion of aerosol to ions within the ICP or transmission through the MS interface may decrease precision, sensitivity, and/or accuracy. Although fundamental mechanisms that govern ion production and transmission have been studied extensively in solution-nebulization (SN) ICP-MS instruments, significant gaps in our understanding remain. Furthermore, it is unclear to what extent differences between the aerosols generated during SN and LA influence either ion production or transmission. In this work, I initially investigated differences in the spatial distributions of Ca, Ba, and Sc ions generated by LA and SN using high-resolution LIF imaging. Ions formed from aerosol generated by LA at low fluence were distributed over much greater axial and narrower radial distances than SN aerosol. Additionally, I investigated the effects of solvent, laser fluence, and ablation atmosphere (He vs Ar) on ion distributions in the ICP. Unlike solvent, changing laser fluence and ablation atmosphere produced considerable changes in the ion signal intensity and spatial distribution during LA. At greater laser fluence, the radial distance over which ions were distributed dramatically increased. Surprisingly, when helium was mixed with argon as carrier gas, ion signals decreased. Many of these effects were assumed to be related to changes in the number and size of particles generated during LA. In a follow-up study, relative contributions to ion densities in the ICP from particles of different sizes were investigated. LIF images were recorded while filtering particles above a threshold size on-line. Micron-sized particles contributed the majority of ions formed in the ICP. For Ba, Ca, and Sc, differences in the axial position where nanometer- and micron-sized particles vaporized were 2, 1, and less than 1 mm, respectively. I also performed experiments to identify changes in the ion signal related to changing ablation conditions vs. changing ICP conditions associated with helium additions to the carrier gas. LIF images were recorded during different combinations of He/Ar added upstream and/or downstream of the ablation cell. Changes in the ion signal during ablation in helium vs argon did not always match expectations based on changes in particle numbers and sizes measured with SEM. The results force re-examination of some of the fundamental assumptions about the effect of carrier gas composition on the performance of LA-ICP-MS. The research described in this dissertation provides valuable insight into fundamental aspects of key ICP processes related to LA generated aerosol.

京都大学 / 新制・課程博士 / 博士(理学) / 甲第24184号 / 理博第4875号 / 新制||理||1698(附属図書館) / 京都大学大学院理学研究科生物科学専攻 / (主査)准教授 半谷 吾郎, 准教授 Huffman Michael Alan, 教授 今井 啓雄 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DGAM

No description available.

No description available.

No description available.

No description available.