Burning Behavior of Fuel on Water Under the Influence of Waves.

Sauer, Nathaniel

Etd

Burning Behavior of Fuel on Water Under the Influence of Waves.

Public Deposited

As the world’s energy production continues to rely on fossil fuels, environmental spills will pose a significant hazard to marine and shoreline ecosystems. A quick and effective response to ocean oil spills is critical to minimizing the subsequent environmental damage. In-situ burning (ISB) is a method of cleanup and containment of this spilled fuel by collecting and burning the fuel in-place on the ocean surface. ISB is particularly well regarded for its speed, effectiveness, and low cost, and as such has become the focus of much research. Limited research has been completed on the impact of waves on hydrocarbon pool burning. This study attempts to investigate the influence of surface water waves on the burning characteristics of hydrocarbon fuels floating on water at three distinct scales. For each scale, a separate wavetank was used, two of which were specifically designed and constructed for this work. The primary guideline for informing in-situ burning operations is the Beaufort scale, which correlates wind speed with wave height. A common threshold for in-situ burning is a Beaufort sea state of 3 or less for conducting in-situ burn operations. However, this is only in reference to boom performance and the ability to collect and ignite the oil and makes no reference to potential burn performance. Using typical ocean wave distributions for corresponding wavelengths and periods, the corresponding approximate wave steepness for each Beaufort sea state is calculated. The primary goal of this study is to quantify the potential effect on burning that each sea state could produce. Wave steepness is the dimensionless parameter that will be used throughout this work, and is defined as wave height divided by wavelength. It is hypothesized that two factors are most greatly affecting the heat loss induced by the wave: the distance of water motion across the bottom of the fuel, and the frequency of this water motion. For linear water waves of regular period and wavelength and with a known water depth; the wavelength is traditionally calculated with a form of the dispersion equation. Where the wavelength is an iterative solution depending only on the wave period and the tank depth. In this way the wave steepness captures both the distance and frequency of the wave motions. The distance of the motion is captured in the wave height, which for deep water waves the horizontal motions are approximately equal to the vertical motions. Additionally, the frequency of oscillations is captured in the wavelength, as the wavelength is a function of the wave period. This work experimentally examines the interaction of waves with burning pools at three distinct sizes: 10 cm, 80 cm, and 2 m. Three wave tanks were used to investigate the burning of floating hydrocarbon fuels. Two of these wave tanks were custom-built at Worcester Polytechnic Institute for this study. The third and largest tank used in this study is located at the US Army Cold Regions Research Facility in Hanover, NH. Current data is available for kerosene fuel at 10 cm and 80 cm, and crude oil at 2 m. Small-scale experiments were conducted in a wavetank 2.4 m long, 0.24 m wide, and 0.28 m tall and filled with fresh water. A custom-made Arduino-based wavemaker generated wave profiles based on the height and period of the experimenters’ choosing. Thin layers (0.6 cm) of kerosene were burned at a 10 cm diameter, and examined against 9 different wave profiles, with wave steepness (S) ranging from (S = 0.003 − 0.080). The no wave cases experienced a burn rate on average of approximately 0.5 mm/min, with the worst-case wave burning at approximately 0.35 mm/min a decrease of 30%. A study on subsurface water velocities and wave motion displacement was also conducted in the small-scale wavetank, using Particle Image Velocimetry (PIV) and particle tracking. Medium-scale experiments were conducted in a modular wavetank built exclusively for examining the burning rate of a fuel floating on water. This wavetank was 3.66 m (12 ft) long, 1.83 m (6 ft) wide, and 0.91 m (3 ft) tall, and filled with fresh water to a static depth of 0.61 m (2 ft). Waves generated in this tank had a more than 0.86 R2 agreement with their ideal sinusoidal profiles. Kerosene pools 80 cm in diameter were examined at 1, 2, and 3 cm thicknesses and with 4 different wave profiles with steepness (S) ranging from (S = 0.008 − 0.0174). At this scale, fuel layer thickness was maintained at steady state using a pump and flowmeter. Surface heat flux was also measured and a burning model incorporating the measured regression rate and heat flux allowed for the calculation of heat loss to the water sublayer. Flame heights and centerline heat flux experienced reductions with increasing wave steepness. Experimentally determined regression rates were also observed to decrease with increasing waves. The regression rate for 1 cm thick fuel layers was reduced by a maximum of 27%, for 2 cm layers a maximum of 19%, and for 3 cm layers a maximum of 12%. The calculated heat loss to the water sublayer also increased with increasing wave steepness. Large-scale experiments were conducted at the U.S. Army Cold Regions Research Facility in Hanover, NH. The above-ground wavetank used measured 13.7 m long, 2.4 m wide, and 2.25 m tall and during testing was filled with 30 parts per thousand saltwater. Crude oil in 8 cm thick layers was examined for a baseline and two wave cases, with wave steepness (S) values of (S = 0.008 & S = 0.009). The burn area for these experiments was a 1.9 m x 1.7 m pool with an approximated equivalent diameter of 2.03 m. Thermocouple data from these experiments was used to evaluate the rate at which a hot thermal wave progressed through the burning fuel layer. Flame heights and global burn rates were observed to decrease with increasing wave steepness. Calculated thermal penetration rates were also seen to decrease with increasing wave steepness. Burning rate models for each scale are presented and are based primarily on experimentally captured temperature data. Results across the three scales show a decrease in fuel temperatures, a decrease in flame height, and a decrease in burning rate with the addition of waves. Burning rate models show good agreement between calculated and experimentally determined values, especially for the medium scale where fuel layer thickness was maintained, and regression rate was measured. Thicker fuels tend to experience less of a reduction in burning behavior with waves. Changes in boilover occurrence and behavior were also observed with the addition of waves. Although it is hard to determine the precise impact of waves at ISB scales, this study presents reasonable ranges for sea-state equivalent waves.

Creator