In situ burning (ISB) is a practical method of oil spill cleanup in icy conditions. This study investigates one example of a likely oil spill scenario; burning oil in an ice cavity. In this situation, unique and unexplored physical processes come into play compared with the classical problem of confined pool fires in vessels. The icy walls of the cavity create a significant heat sink causing notable lateral heat losses especially for small cavity sizes (5-10 cm). Melting of ice because of the heat from the flame causes the geometry of cavity to change. Specifically, the diameter of the pool fire increases as the burning advances. This widening causes the fuel to stretch laterally thereby reducing its thickness at a faster rate. The melted ice water causes the oil layer to rise which causes the ullage height to decrease. The decrease in ullage and increase in diameter counteract the reduction in thickness because of widening or stretching of the fuel layer. There thus exists a strong coupling between the burning rate and the geometry change of the pool and cavity. To explore the problem, experiments were performed in circular ice cavities of varying diameters (5 - 25 cm). The change in shape of the ice cavity and the oil layer thickness are recorded using a combination of visual images, mass loss, and temperature data along the centerline and edge of the cavity. The average burning rate of crude oil in a cavity is greater than the corresponding burning rate in a vessel of equal diameter, yet the burning efficiency (% of fuel consumed during combustion) is lower. For example, the average mass loss rate in a 10 cm ice cavity is 50% higher than a steel vessel of similar size. However, the burning efficiency is lower by 50%. Widening of cavity (170%) contributes to the increase in the average mass burning rate. At the same time heat losses through fuel layer increase because of decrease in fuel thickness by widening of the fuel layer. This coupling is analyzed using a mathematical model which can predict burning rate and efficiency of crude oil in an ice cavity for the range of cavity diameters examined. Extension of the model to larger sizes comparable to realistic situations in the Arctic is discussed.

The potential for oil exploration on the Arctic Outer Continental Shelf warrants determination of an efficient method to clean up an oil spill. Traditional spill response equipment may not be practical in an Arctic environment; the presence of ice which may prevent proper deployment of equipment. The remoteness of the areas proposed for oil exploration lack the infrastructure and support networks necessary to stage a response to a large oil spill. These difficulties make it necessary to explore alternative means of oil spill cleanup. In situ burning is one method that may be particularly well-suited for arctic and sub-arctic environments due to the minimal amount of equipment required to achieve an efficient burn, i.e. high mass loss. The Arctic and sub- Arctic environments add an additional level of complexity by introducing a spill medium (ice) that is highly unstable at elevated temperatures. Our experiments sought to calculate the mass loss rate of oil mixtures to determine the efficiency with which they burn within ice channels of varying widths. Since fuel layer thickness is a critical factor in determining the effectiveness of an in situ burn the spread rate of oil along an ice channel was studied. Burning of oil in an ice channel yields low efficiencies (10%) primarily due to the geometric changes of the melting ice channel. The spreading was modeled as a constant flux rectilinear buoyancy-inertia governed flow. The melting causes an increase in the surface area and results in the critical thickness of the oil to be reached sooner. Based on the current bench- scale testing, losses due to ice melting cause the efficiencies of the burning process to be excessively low and not viable to full scale clean up. The results warrant future research to understand how varying other parameters, including starting mass of fuel, influence efficiencies.

The characterization of flammability parameters such as time to ignition, mass loss rate (MLR), and extinction criteria is critical for understanding ignition and burning behavior of timber products. These parameters, often determined with bench scale experiments, have previously been presented in literature. However, standard test methods generally use relatively low trial quantities (e.g., n=3) which can potentially cause large variation in reported values. This study investigates the influence of trial quantity on observed statistical variation in key flammability metrics for timber products (e.g., ignition time, peak MLR, MLR at extinction). Using a conical heater, 100 repeat trials were conducted at incident heat exposures of 20 kW/m2, 40 kW/m2, and 50 kW/m2 on 12.7 mm thick ACX cross laminated plywood samples. Ignition time data was found to exhibit significant positive skew and 20-30 trials were required for the reduction in uncertainty with each additional trial to fall below 0.1s at each heat flux. The normalized uncertainty in ignition time was greatest at 50 kW/m2 and was 20-70% than at 20 kW/m2 and 40 kW/m2. Significant variability was observed in the extinction characteristics of samples exposed to 40 kW/m2 where 39 samples experienced self-extinction while the remainder sustained combustion until burnout. Uncertainty in MLR at extinction for these trials was nearly double that of trials exposed to 20 kW/m2. These results exhibit the significance of large trial quantities when determining flammability characteristics.

"This dissertation documents two interrelated studies that were conducted to more fundamentally understand the scalability of flame heat flux. The first study used an applied heat flux in the bench scale horizontal orientation which simulates a large scale flame heat flux. The second study used enhanced ambient oxygen to actually increase the bench scale flame heat flux itself. Understanding the scalability of flame heat flux more fully will allow better ignition and combustion models to be developed as well as improved test methods. The key aspect of the first study was the use of real scale applied heat flux up to 200 kW/m2. An unexpected non-linear trend is observed in the typical plotting methods currently used in fire protection engineering for ignition and mass loss flux data for several materials tested. This non-linearity is a true material response. This study shows that viewing ignition as an inert material process is inaccurate at predicting the surface temperature at higher heat fluxes and suggests that decomposition kinetics at the surface and possibly even in-depth may need to be included in an analysis of the process of ignition. This study also shows that viewing burning strictly as a surface process where the decomposition kinetics is lumped into the heat of gasification may be inaccurate and the energy balance is too simplified to represent the physics occurring. The key aspect of the second study was direct experimental measurements of flame heat flux back to the burning surface for 20.9 to 40 % ambient oxygen concentrations. The total flame heat flux in enhanced ambient oxygen does not simulate large scale flame heat flux in the horizontal orientation. The vertical orientation shows that enhanced ambient oxygen increases the flame heat flux more significantly and also increases the measured flame spread velocity."

Lors du déclenchement d’un incendie en milieu clos, la quantité d’oxygène du local décroît, entrainant une combustion incomplète. Des gaz chauds imbrûlés peuvent alors s’accumuler dans le local ou dans les gaines de ventilation et un accident thermique peut survenir suite à un apport d’air frais. Ce travail, réalisé pour AREVA, vise à quantifier et d’analyser ce risque, afin de pouvoir le prédire et le prévenir. Tout d’abord, une étude bibliographique a été réalisée afin de définir les paramètres d’auto-inflammation à partir du modèle de Frank-Kamenetskii. Celui-ci permet, après un bilan d’énergie, l’établissement d’un paramètre critique, δC, d’auto-inflammation du mélange. δC réunit la géométrie, la température (et la température ambiante) et la composition du mélange à l’auto-inflammation.Puis, la dégradation thermique du Polyéthylène Haute Densité en fonction de la densité surfacique de flux incident à la surface du matériau et de la sous-ventilation a été caractérisée (cinétique de dégradation, productions gazeuses). Le Cône Calorimètre à Atmosphère Contrôlée a été employé.Ce travail expérimental a permis d’obtenir plusieurs mélanges gazeux suivant les conditions. La dernière partie de l’étude a permis, à partir de δC, de poser le volume de mélange via le rayon comme critère d’auto-inflammabilité des mélanges. En imposant une température, en faisant varier la fraction volumique de chaque gaz combustible entre sa LII et LSI le risque d’accident thermique a été défini. / After the beginning of a fire in a closed room, the oxygen rate in the atmosphere decreases. This implies an incomplete combustion and unburnt gases production. These ones may accumulate in the room or in ventilation pipes, and, after mixing with fresh air, auto-ignite. This could trigger a thermal accident such as backdraft. This present work, conducted for AREVA, aims to analyse this hazard and provide some methods to predict and prevent it. First, a bibliographical research, was carried on to define a mixture’s auto-ignition parameters. This study was based on Frank-Kamenetskii’s model: after establishing the energetics balance between the heat produced by combustion, and the one consumed by conduction, an auto-ignition critical parameter, δC, was defined. It reunites the system’s geometry, temperature (or the room temperature) and composition.Then, the High Density Polythene degradation in a Controlled Atmosphere Cone Calorimeter was studied. The effect on the material’s degradation of under-ventilation and of the energy brought has been tested through the oxygen concentration in the atmosphere and the incident heat flux.During this work many different gas mixtures were analyzed. On the ground of δC formula, the final step was to set the volume, through the radius (characteristic size of the system), as an auto-ignition parameter. Making the concentration of each combustible varying between the LFL and UFL and imposing the temperature allowed to predict this hazard.

An understanding of compartment fire behaviour is important for fire protection engineers. For design purposes, whether to use a prescriptive code or performance based design, life safety and property protection issues are required to be assessed. The use of design fires in computer modelling is the general method to determine fire safety. However these computer models are generally limited to the input of one design fire, with consideration of the complex interaction between fuel packages and the compartment environment being simplified. Of particular interest is the Heat Release Rate, HRR, as this is the commonly prescribed design parameter for fire modelling. If the HRR is not accurate then it can be subsequently argued that the design scenario may be flawed. Therefore the selection of the most appropriate fire design scenario is critical, and an increased level of understanding of compartment behaviour is an invaluable aid to fire engineering assumptions. This thesis details an experimental study to enhance the understanding of the impact and interaction that the size and location of pool fires within an enclosure have upon the compartment fire behaviour. Thirty four experiments were conducted in a reduced scale compartment (½ height) with dimensions of 3.6m long by 2.4m wide by 1.2m high using five typical ventilation geometries (fully open, soffit, door, window and small window). Heptane pool fires were used, located in permutations of three evenly distributed locations within the compartment (rear, centre and front) as well as larger equivalent area pans located only in the centre. This thesis describes the experimental development, setup and results of the experimental study. To assist in the classification of compartment fire behaviour during the experiments, a ‘phi’ meter was developed to measure the time dependent equivalence ratio. The phi meter was developed and configured to measure O₂, CO₂ and CO. The background development, calibration, and experimental results are reported. A review of compartment fire modelling using Fire Dynamics Simulator, has also been completed and the results discussed. The results of this experimental study were found to have significant implications for Fire Safety Engineering in that the size of the fire is not as significant as the location of the fire. The effect of a fire near the vent opening was found to have a significant impact on compartment fire behaviour with the vent located fuel source increasing the total compartment heat release rate by a factor of 1.7 to that of a centrally placed pool fire of the same total fuel area. The assumption that a fire located in the centre of the room provides for the highest heat release rate is not valid for post-flashover compartment fires. The phi meter was found to provide good agreement with the equivalence ratio calculated from total compartment mass loss rates, and the results of FDS modelling indicate that the use of the model in its current form can not be applied to complex pool fire geometries.

[ES] El diseño de puentes, a diferencia de lo que ocurre con el diseño de edificios o con el diseño de túneles ha dejado de lado la consideración de la acción del fuego hasta la fecha. Este vacío normativo, combinado con la gran repercusión económica y social de colapsos de puentes en el pasado como consecuencia de incendios, ha motivado un rápido incremento del número de estudios relativos a la ingeniería frente al fuego en el ámbito de los puentes. Aunque la acción del fuego no resulta del todo desconocida en el ámbito de las estructuras, sí que existen una serie de singularidades que impiden la trasposición directa de recomendaciones o de modelos de fuego simplificados ya desarrollados en otros campos que ya incorporan la acción del fuego en el diseño. En este contexto, el trabajo que a continuación se expone parte de un incendio ocurrido en el estado de Alabama en 2002, cuyas consecuencias fueron la demolición de un puente mixto de 37 metros de vano central, para plantear y validar una metodología que aborda el problema de forma numérica mediante tres modelos acoplados secuencialmente: modelo de incendios, modelo térmico y modelo mecánico Realizada una validación a nivel general se descubre que, aunque la configuración geométrica final obtenida se ajustan bastante a la realidad, la definición del incendio ha supuesto un gran número de hipótesis. Es por ello que se decide, en una segunda parte, realizar una campaña experimental que permita registrar la potencia del fuego, las temperaturas del gas y del acero y las flechas en un puente construido ad-hoc en el campus de la Universitat Politècnica de València. Este puente experimental tenía un vano único de 6 m de luz y fue sometido a cargas de fuego de hasta 1.3 MW. Mediante el empleo de los registros realizados en la campaña experimental se ha validado el modelo de incendio, el modelo térmico y el modelo mecánico. Con todo ello se ha puesto en evidencia la importancia del viento en la acción del fuego, la magnitud de los gradientes térmicos espaciales y la urgencia de desarrollar procedimientos simplificados que permitan la incorporación del fuego como acción en el ámbito de los puentes Las validaciones específicas de cada modelo han permitido además llegar a una serie de conclusiones de gran interés para la realización de futuras campañas experimentales en puentes a mayor escala. / [CAT] El disseny de ponts, a diferència del que passa amb el disseny d'edificis o amb el disseny de túnels ha deixat de banda la consideració de l'acció del foc. Aquest buit normatiu, combinat amb la gran repercussió econòmica i social de col·lapses de ponts com a conseqüència d'incendis, ha motivat un ràpid increment del nombre d'estudis relatius a l'enginyeria del foc del foc en l'àmbit dels ponts. Encara que l'acció del foc no resulta del tot desconeguda en l'àmbit de les estructures, sí que hi ha una sèrie de singularitats que impedeixen la transposició directa de recomanacions o de models de foc simplificats ja desenvolupats en altres camps que ja incorporen l'acció del foc al disseny. En aquest context, el treball que a continuació s'exposa part d'un incendi ocorregut a l'estat d'Alabama en 2002 i que va provocar la demolició d'un pont mixt de 37 metres de va, per plantejar i validar una metodologia que aborda el problema de forma numèrica mitjançant tres models acoplats seqüencialment: model d'incendis, model tèrmic i model mecànic Realitzada una validació a nivell general es descobreix que, encara que la configuració geomètrica final obtinguda s'ajusta en gran mesura a la realitat, la definició de l'incendi ha suposat un gran nombre d'hipòtesis. És per això que es decideix, en una segona part, realitzar una campanya experimental que permeta registrar la potència del foc, les temperatures del gas i de l'acer i les fletxes en un pont construït ad hoc al campus de la Universitat Politècnica de València. Aquest pont experimental presenta un va únic de 6 m de llum i va ser sotmès a càrregues de foc de fins a 1.3 MW. Mitjançant l'ús dels registres realitzats a la campanya experimental s'ha validat el model d'incendi, el model tèrmic i el model mecànic. Amb tot això s'ha posat en evidència la importància del vent en l'acció del foc, la magnitud dels gradients tèrmics espacials i la urgència de desenvolupar procediments simplificats que permetin la incorporació del foc com a acció en l'àmbit dels ponts Les validacions específiques de cada model han permès a més arribar a una sèrie de conclusions de gran interès per a la realització de futures campanyes experimentals en ponts a major escala. / [EN] To date, the fire action has been left aside in the bridge design despite this action has been widely considered in other structures such as building and tunnels. This regulatory vacuum, combined with the great economic and social impact of bridge collapses in recent times as a result of fires, has led to a rapid increase in the number of studies related to fire engineering in the field of bridges. Although the action of fire is not entirely unknown in the field of structures, there are a number of singularities that prevent the direct transposition of recommendations or simplified fire models from such fields. In this context, the study started by using a real fire which occurred in the state of Alabama in 2002 and led to the demolition of the 37-meter main span of a composite concrete and steel bridge to introduce and validate a methodology that numerically addresses the problem by uncoupling the problem in three different models: fire model, thermal model and mechanical model. Once the validation was accomplished at a general level, it was discovered that, although the geometrical data were quite adjusted to reality, the definition of the fire had involved a large number of hypotheses. That is why carrying out an experimental campaign to record the power of the fire, the gas and steel temperatures and the vertical deflections of a bridge built ad-hoc on the campus of the Universitat Politècnica de València became a priority. This 6-meter single span experimental bridge was subjected to fire loads of up to 1.3 MW. Through the use of the information recorded during the experimental campaign, the fire model, the thermal model and the mechanical model were validated. Moreover, the importance of wind in the action of fire, the magnitude of spatial thermal gradients and the urgency of developing simplified procedures which allow the consideration of fire as an action in the field of bridges were also highlighted. Last but not least, the validation of the different models allowed the author to include useful guidelines in order to define future experimental campaigns with more powerful fires and longer span bridges. / Alós Moya, J. (2018). Análisis de la respuesta frente al fuego de puentes mixtos multijácena [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/116625 / TESIS