The production of detrimental carbonaceous deposits in jet aircraft fuel systems results from the involvement of trace heteroatomic species in the autoxidation chain that occurs upon fuel heating. Although it has been known for many years that these sulfur-, nitrogen-, and oxygen-containing species contribute to the tendency of a fuel to form deposits, simple correlations have been unable to predict the oxidation rates or the deposit forming tendencies over a range of fuel samples. In the present work, a chemical kinetic mechanism developed previously is refined to include the roles of key fuel species classes, such as phenols, reactive sulfur species, dissolved metals, and hydroperoxides. The concentrations of these fuel species classes in the unreacted fuel samples are measured experimentally and used as an input to the mechanism. The resulting model is used to simulate autoxidation behavior observed over a range of fuel samples. The model includes simulation of the consumption of dissolved oxygen, as well as the formation and consumption of hydroperoxide species during thermal exposure. In addition, the chemical kinetic mechanism is employed with a global deposition submechanism in computational fluid dynamics (CFD) simulations of deposit formation occurring in nearisothermal as well as non-isothermal flowing environments. Experimental measurements of oxygen consumption, hydroperoxide formation, and deposition are performed for a set of seven fuels. Comparison with experimental measurements indicates that the methodology offers the ability to predict both oxidation and deposition rates in complex flow environments, such as aircraft fuel systems, using only measured chemical species class concentrations for the fuel of interest.

No abstract

No abstract

No abstract

A comprehensive numerical and experimental investigation on micrometer-sized water droplet impact dynamics and evaporation on an unheated, flat, dry surface is conducted from the standpoint of spray-cooling technology. The axisymmetric time-dependent governing equations of continuity, momentum, energy, and species are solved. Surface tension, wall adhesion effect, gravitational body force, contact line dynamics, and evaporation are accounted for in the governing equations. The explicit volume of fluid (VOF) model with dynamic meshing and variable-time stepping in serial and parallel processors is used to capture the time-dependent liquid-gas interface motion throughout the computational domain. The numerical model includes temperature- and species-dependent thermodynamic and transport properties. The contact line dynamics and the evaporation rate are predicted using Blake's and Schrage's molecular kinetic models, respectively. An extensive grid independence study was conducted. Droplet impingement and evaporation data are acquired with a standard dispensing/imaging system and high-speed photography. The numerical results are compared with measurements reported in the literature for millimeter-size droplets and with current microdroplet experiments in terms of instantaneous droplet shape and temporal spread (R/D(0) or R/R(E)), flatness ratio (H/D(0)), and height (H/H(E)) profiles, as well as temporal volume (inverted A) profile. The Weber numbers (We) for impinging droplets vary from 1.4 to 35.2 at nearly constant Ohnesorge number (Oh) of approximately 0.025-0.029. Both numerical and experimental results show that there is air bubble entrapment due to impingement. Numerical results indicate that Blake's formulation provides better results than the static (SCA) and dynamic contact angle (DCA) approach in terms of temporal evolution of R/D(0) and H/D(0) (especially at the initial stages of spreading) and equilibrium flatness ratio (H(E)/D(0)). Blake's contact line dynamics is dependent on the wetting parameter (K(W)). Both numerical and experimental results suggest that at 4.5 < We < 11.0 the short-time dynamics of microdroplet impingement corresponds to a transition regime between two different spreading regimes (i.e., for We < or = 4.5, impingement is followed by spreading, then contact line pinning and then inertial oscillations, and for We > or = 11.0, impingement is followed by spreading, then recoiling, then contact line pinning and then inertial oscillations). Droplet evaporation can be satisfactorily modeled using the Schrage model, since it predicts both well-defined transient and quasi-steady evaporation stages. The model compares well with measurements in terms of flatness ratio (H/H(E)) before depinning occurs. Toroidal vortices are formed on the droplet surface in the gaseous phase due to buoyancy-induced Rayleigh-Taylor instability that enhances convection.

Flow experiments using heated Jet-A fuel and additives were performed to study the effects of treated surfaces on surface deposition. The experimental apparatus was designed to view deposition due to both thermal oxidative and pyrolytic degradation of the fuel. Carbon burnoff and scanning electron microscopy were used to examine the deposits. To understand the effect of fuel temperature on surface deposition, computational fluid dynamics was used to calculate the two-dimensional temperature profile within the tube. Three kinds of experiments were performed. In the first kind, the dissolved O2 consumption of heated fuel is measured on different surface types over a range of temperatures. It is found that use of treated tubes significantly delays oxidation of the fuel. In the second kind, the treated length of tubing is progressively increased which varies the characteristics of the thermal-oxidative deposits formed. In the third type of experiment, pyrolytic surface deposition in either fully treated or untreated tubes is studied. It is found that the treated surface significantly reduced the formation of surface deposits for both thermal oxidative and pyrolytic degradation mechanisms. Moreover, it was found that the chemical reactions resulting in pyrolytic deposition on the untreated surface are more sensitive to pressure level than those causing pyrolytic deposition on the treated surface.

No abstract

No abstract