Search

Dynamic processes controlling explosive volcanic eruptions are poorly constrained because of lack of direct observations. To investigate driving mechanisms, physical processes in eruptive conduits, and behavior of eruptive products at the Earth’s surface, I studied heterogenous bubble nucleation in rhyolite magmas, syn-sintering microlite growth in pyroclastic obsidian, and thermal evolution of pyroclastic density currents (PDCs). Nucleation and growth rates were constrained using microlite and bubble textural measurements and high temperature experiments. Thermal evolution of PDCs was investigated using paleomagnetism of pumice and lithic clasts in PDC deposits. Heterogenous bubble nucleation experiments were performed employing decompression experiments using rhyolitic glass with two size populations of magnetite microlites. Variations in bubble number densities (BNDs) with ΔP and experimental dwell time indicates that BND is controlled by both ΔP and magnetite microlite number densities (MNDs), as well as activation energies/irregularities of different nucleation sites on crystal faces. Results from this study have implications for classical nucleation theory. MNDs in static growth experiments suggest that obsidian pyroclasts in the Mono Craters, CA eruption sintered in ~≤7 hours. Varieties of microlite morphologies and orientations suggests repeated in-conduit fragmentation and sintering at multiple depths prior to being erupted. During the North Mono eruption, volatile concentrations and microlite textures indicate that pyroclasts were extracted from many depths in the conduit. In the final phase, however, higher MNDs indicate longer stalling while lower volatile concentrations record equilibration at shallow depths. Pumice and lithic samples from Crater Lake, OR PDC deposits were thermally demagnetized to 650–700 ⁰C and via alternating field (AF) to 100 mT. Thermoremanent magnetization (TRM) vectors are randomly oriented in pumice and lithic clasts. This dictates that pumice and wall-rock lithics cooled prior to final deposition, and accessory lithic clasts were heated and cooled prior to final deposition. This requires that final PDC deposits are collections of material deposited at relatively cool temperatures. I posit that early fall and PDC deposits cooled and were scoured and entrained by later highly erosive PDCs generated by tall plinian column collapse. Voluminous proximal lithic breccia are deposits of these scouring PDCs.

Today nucleate boiling is widely used in numerous industrial applications such as cooling processes because of the high achieved heat transfer rates for low temperature differences. It remains a possible cooling solution for the next generation of central processing units (CPU), which dissipate heat fluxes exceeding the capabilities of today’s conventional forced air cooling. However, nucleate boiling is a very complex and elusive process involving many mechanisms which are not fully understood yet and a comprehensive model is still missing. For this study a new experimental setup was designed, constructed and commissioned to investigate bubble nucleation, growth, departure and interaction during nucleate pool boiling from a silicon device fully immersed in fluorinert FC-72. The location of bubble nucleation is controlled by artificial cavities etched into the silicon substrate. Boiling is initiated with a heater integrated on the back and micro-sensors indicate the wall temperature at the bubble nucleation site. During this work three different silicon test section designs were fabricated and boiling experiments on these substrates successfully conducted. Bubble growth, bubble departure frequencies and bubble departure diameters for different dimensioned artificial cavities, varied pressure and increasing wall temperature were measured from high-speed imaging sequences. Bubble interactions like vertical and horizontal coalescence were visualised and their impact on the boiling heat transfer investigated. The influence of spacing between two neighbouring artificial cavities on bubble nucleation and departure frequencies, vertical coalescence frequencies and departure diameters was analysed. The acquired data are used as input for a numerical code developed by our collaborators (Brunel University, UK and Los Alamos National Laboratories, USA) and are a first step to validate the code. The code studies the interactions between bubble nucleation sites on solid surfaces as a network. The simulations will help design boiling substrates utilised for chip cooling applications with optimal artificial cavity distribution to maximise the cooling heat transfer.

The research concerns the flash boiling of fuels injected into combustion engines. This is the phenomenon where an injected fuel jet flash vaporizes due to a sudden reduction in the injection pressure in the combustion environment. The cavitation bubbles that form, grow and can expand the jet resulting in interference of jets for multi-port injectors. The problem is controlled by bubble nucleation and growth phenomenon in the fuel jet. The initial conditions that trigger the process, namely superheating of the jets for a range of fuel systems including single component systems, and surrogates for petroleum-based fuels were investigated. The work is primarily experimental and focuses on the bubble nucleation phenomenon for miscible mixture systems. The experimental apparatus uses a bubble column to simulate a droplet going through a rising temperature gradient beyond its boiling point until nucleation is observed. Although primarily experimental, computations were also performed to compare the experimental results to what theory would dictate. Results from the experiments were then processed and the superheat limits were compared to theoretical results obtained using both the thermodynamic and kinetic limit of superheat and general agreement was observed.

A heater designed to monitor surface temperature fluctuations during pool boiling experiments while the bubbles were simultaneously being observed has been fabricated and tested. The heat source was a transparent indium tin oxide (ITO) layer commercially deposited on a fused quartz substrate. Four copper-nickel thin film thermocouples (TFTCs) on the heater surface measured the surface temperature, while a thin layer of sapphire or fused silica provided electrical insulation between the TFTCs and the ITO. The TFTCs were micro-fabricated using the liftoff process to deposit the nickel and copper metal films. The TFTC elements were 50 microns wide and overlapped to form a 25 micron by 25 micron junction. TFTC voltages were recorded by a DAQ at a sampling rate of 50 kHz. A high-speed CCD camera recorded bubble images from below the heater at 2000 frames/second. A trigger sent to the camera by the DAQ synchronized the bubble images and the surface temperature data. As the bubbles and their contact rings grew over the TFTC junction, correlations between bubble behavior and surface temperature changes were demonstrated. On the heaters with fused silica insulation layers, 1-2 C temperature drops on the order of 1 ms occurred as the contact ring moved over the TFTC junction during bubble growth and as the contact ring moved back over the TFTC junction during bubble departure. These temperature drops during bubble growth and departure were due to microlayer evaporation and liquid rewetting the heated surface, respectively. Microlayer evaporation was not distinguished as the primary method of heat removal from the surface. Heaters with sapphire insulation layers did not display the measurable temperature drops observed with the fused silica heaters. The large thermal diffusivity of the sapphire compared to the fused silica was determined as the reason for the absence of these temperature drops. These findings were confirmed by a comparison of temperature drops in a 2-D simulation of a bubble growing over the TFTC junction on both the sapphire and fused silica heater surfaces. When the fused silica heater produced a temperature drop of 1.4 C, the sapphire heater produced a drop of only 0.04 C under the same conditions. These results verified that the lack of temperature drops present in the sapphire data was due to the thermal properties of the sapphire layer. By observing the bubble departure frequency and site density on the heater, as well as the bubble departure diameter, the contribution of nucleate boiling to the overall heat removal from the surface could be calculated. These results showed that bubble vapor generation contributed to approximately 10% at 1 W/cm^2, 23% at 1.75 W/cm^2, and 35% at 2.9 W/cm^2 of the heat removed from a fused silica heater. Bubble growth and contact ring growth were observed and measured from images obtained with the high-speed camera. Bubble data recorded on a fused silica heater at 3 W/cm^2, 4 W/cm^2, and 5 W/cm^2 showed that bubble departure diameter and lifetime were negligibly affected by the increase in heat flux. Bubble and contact ring growth rates demonstrated significant differences when compared on the fused silica and sapphire heaters at 3 W/cm^2. The bubble departure diameters were smaller, the bubble lifetimes were longer, and the bubble departure frequency was larger on the sapphire heater, while microlayer evaporation was faster on the fused silica heater. Additional considerations revealed that these differences may be due to surface conditions as well as differing thermal properties. Nucleate boiling curves were recorded on the fused silica and sapphire heaters by adjusting the heat flux input and monitoring the local surface temperature with the TFTCs. The resulting curves showed a temperature drop at the onset of nucleate boiling due to the increase in heat transfer coefficient associated with bubble nucleation. One of the TFTC locations on the sapphire heater frequently experienced a second temperature drop at a higher heat flux. When the heat flux was started from 1 W/cm^2 instead of zero or returned to zero only momentarily, the temperature overshoot did not occur. In these cases sufficient vapor remained in the cavities to initiate boiling at a lower superheat.

A molecular level understanding of nucleation occurring in condensed phases is important in a number of industrial applications like polymer foaming, sonolysis, manufacturing of drugs, polymorphic transformations, and metallic alloy production. The recently developed statistical mechanical approaches based on density functional theory and Monte Carlo (MC) simulation can provide valuable insight into these nucleation processes. In this work, we applied density functional theory to study bubble nucleation in a micellar solution. Our results showed that the presence of surfactant molecules lowers the free energy barrier of bubble nucleation. We also found that under moderate negative pressures, the stable micelle may evolve to form the critical bubble and the resulting free energy barrier is lower than that in the absence of this mechanism. A kinetic consideration revealed that the above mechanism is less effective at lower pressures closer to spinodal. Further, the mechanism of bubble nucleation from the stable micelle correlated well with the liquid-liquid miscibility. In another study, we aimed at understanding crystal nucleation in binary Lennard-Jones (LJ) mixtures. An important factor dictating the crystal nucleation behavior is the underlying phase diagram. Thus, we developed thermodynamic integration (TDI) methods that predict accurately the coexistence points on the phase diagram. In particular, we introduced a method to directly calculate Gibbs free energy difference between the solid and the liquid phases by connecting these phases by a reversible path. Further, we showed that this technique extends naturally to a binary mixture, which allows us to predict its melting temperature. We also developed a new technique to calculate sublimation temperature of single component and binary systems by TDI method. We used these methods to predict eutectic and spindle-shaped phase diagram for the LJ binary mixtures. We then studied the free energy surface for crystal nucleation using MC simulation aided by umbrella sampling technique at a few representative points on these phase diagrams. The emphasis in this work was to identify proper order parameters that correctly describe the crystal nucleation. Our work provides a useful starting point to study how the phase diagram affects the crystal nucleation behavior.

The exsolution of volatiles in magma initiates the formation of gas bubbles as a result of the development of a state of supersaturation from either magma ascending in the conduit, decrease in temperature, or the crystallization of anhydrous solid phases. In this analysis rhyolitic glasses were isothermally decompressed in a series of controlled nucleation experiments to investigate the kinetics of bubble nucleation and growth. A two-fold approach was used to synthetically saturate the high silica melt. The first method uses blocks of solid obsidian, while the second technique uses powdered obsidian. The second method evenly saturates the melt by minimizing the volume to surface area ratio, hence the diffusion distance. In all experiments glass charges were saturated with H 2 O at 100 MPa and 850°C, and then isothermally decompressed to final confining pressures of 30, 50, and 70 MPa. Samples were held at the final pressure for periods ranging from 5 to 300 seconds to allow for nucleation to take place. Bubble-bearing glass charges were imaged under SEM and petrographic microscopes, and digitally analyzed. All glass charges produced by the block method contained either no bubbles or only a few bubbles. The only indication of bubble nucleation was limited to a zone on the outer periphery of the glass charges, implying that in the block method, the time given for hydration was insufficient for homogeneous saturation of the melt. We directed our attention to the bubble-bearing glasses generated by the powder experiments. Vesicle Size Distribution was used to determine the kinetic behavior of bubbles at the onset of nucleation and their development in the melt within the known residence time. The more uniform bubble distribution in the melts using the powder method gave rise to nuclei density populations ranging from 10 9 - 10 11 bubbles/cm 4 . The mean vesicle size for all experiments was 0.009 mm in diameter. Growth rates were calculated to be approximately 3.8 x 10 -2 cm/s. Nucleation rates of 10 7 - 10 9 bubbles cm -3 s -1 illustrate how minor fluctuation in pressure can drastically increase the onset of bubble formation in viscous silicate melts. The data generated by VSD was then applied to a conceptual model used to describe explosive volcanic eruptions.