Ph.D.

The thesis explores the proposal that macroscopic fluid convection in thin-film processors may be adequately represented by simple linear deterministic models. In addition, it examines the suggestion that the models themselves provide a useful tool in the search for a generalizable 'intrinsic' process heat transfer film coefficient, i.e., one that includes the effects of axial dispersion of the process fluid. Such a parameter would be helpful in the design and scale up of thin-film equipment. The following approach was used to investigate this proposal: first, experimental fluid residence time distributions were obtained t over a range of operating conditions, using an industrial pilot plant thin -film processor. The experimental data were used to select an appropriate linear fluid flow model for the process. The model parameters were evaluated over this range using frequency response techniques. These models were subsequently incorporated into a numerical heat transfer simulation of the thin -film processor. Careful matching of the pilot plant transient temperature responses to those predicted by the simulation yielded the sought after intrinsic (dispersion corrected) heat transfer film coefficients for the processor. / Master of Science

M.S.

The variation of the local coefficients of heat transfer with bed depth, particle density, and particle size were studied using a pyrex pipe four inches in diameter as the fluidizing vessel. The fluidized solids investigated were aerocat cracking catalyst, having a geometric-mean particle diameter of 0.00262 inch and an absolute density of 136.6 pounds per cubic foot; tabular alumina, having a geometric-mean particle diameter of 0.0138 inch and an absolute density of 238.8 pounds per cubic foot; silica gel, having a geometric-mean particle diameter of 0.0187 inch and an absolute density of 135.1 pounds per cubic foot; and superbrite glass beads, having a geometric-mean particle diameter of 0.0138 inch and an absolute particle density of 179.5 pounds per cubic foot. The heating element was a 230-vo1t, 1750-watt, copper-sheathed rod having a diameter of 0.25 inch. Temperature measurements were made with 20 B and S gage, iron-constantan thermocouples insulated with fiber-glass braid over glass wrap. Air varying in temperature from 80 to 85 °F and in humidity from 0.000 to 0.004 pound of water vapor per pound of dry air was used as the fluidizing medium. / Master of Science

Methods for the removal of heat from a sphere, via jet impingement by single and multiple jets was documented experimentally. Average heat transfer rates from a sphere maintained at constant temperature, by means of an internal electronic heater, and subjected to single or multiple jet impingements were obtained and related to the exit conditions of the impinging air jet(s) and to geometric parameters. The heat transfer rate was found to be insensitive to small changes in geometry. The heat transfer rate was found to increase with an increase in mass flow rate. The impingement of two jets was found not to be as efficient as a single jet using the same mass flow rate. Compressibility was found to decrease the heat transfer rate at high values of the Mach number. Attempts to increase the heat transfer rate by increasing the entrainment of the jet by acoustic or mechanical excitation or by the use of an elliptic orifice meet with no success. The decrease in velocity due to the increase in entrainment cancelled any benefit that was gained by increasing the entrainment of the jet. / Master of Science

The behavior of gas-solid reactors known as compact-fixed and moving beds, is analyzed from a theoretical viewpoint. For a compact fixed-bed the solution of the energy balance equations is obtained for the cases of a uniform temperature inside the solid pellets (i.e., the Biot number is zero) and for the case in which there are temperature gradients within the pellets (Bi > 0). For short contact times, beds with Bi > 0 have gas- and solid- temperatures which are greater than the temperatures within beds with Bi = 0. For long times, the situation is reversed. For a compact-moving bed the solution of the mass balance equations is obtained for the cases of a feed-solid with constant concentration and a feed solid with an oscillating concentration. In both cases the steady states obtained are unique, and internal recycling is observed only for a feed-solid with an oscillating concentration. Recycling is that situation when the concentration of the solid falls below that of the gas for a bed in which the feed-solid is greater than the feed-gas. This occurred when the period of oscillation was smaller than the residence time of the solid provided that the residence time of the solid was not very short (i.e., provided that B(,s) > 0.1). For both types of beds there is an equivalence between mass transfer and energy transfer so that the solutions can be interchanged with suitable definitions of dimensionless variables.

A constrained thin film desorption scheme has been experimentally tested to determine the desorption rates for water from an aqueous lithium bromide mixture through a confining membrane. Variable conditions include the inlet concentration, pressure differential across the membrane, and channel height. Desorption takes place in a channel created between two parallel plates with one of the walls being both heated and porous. A hydrophobic porous membrane creates a liquid-vapor interface and allows for vapor removal from the channel. Inlet concentrations of 32 wt%, 40 wt%, and 50 wt% lithium bromide were tested at an inlet sub-atmospheric pressure of 33.5 kPa. Pressure differentials across the membrane of 6 kPa and 12 kPa were imposed along with two channel heights of 170 μm and 745 μm. All cases were run at an inlet mass flow rate of 3.2 g/min, corresponding to Reynolds numbers of approximately 2.5 to 4.5. The membrane surface area for desorption was 16.8 cm². A maximum desorption rate (vapor mass flow rate) of 0.51 g/min was achieved, for the 32 wt%, 12 kPa pressure differential, and 170 μm channel. Increasing the pressure differential across the channel allowed for higher desorption rates at a fixed wall superheat, and delayed the transition to boiling. As the inlet concentration increased the desorber's performance decreased as more energy was required to produce a fixed desorption rate. Results are also presented for the variation in the heat transfer coefficient with the wall superheat temperature. The increase in the channel height had a negative influence on the heat transfer coefficient, requiring larger superheat values to produce a fixed desorption rate. / Graduation date: 2005 / Best scan available for tables and computer code in the appendices. The original is faded.

Thesis (M.S.)--Rochester Institute of Technology, 2007. / Typescript. Includes bibliographical references (leaves 74-76).

Thesis (Ph.D.)--Rochester Institute of Technology, 2005. / Typescript. Includes bibliographical references (leaves 179-189).

Thesis (Ph.D.)--Rochester Institute of Technology, 2006. / Typescript. Includes bibliographical references (leaves 270-275).