T. W. F. Russell

Bubble breakage in pipeline flow

Abstract The rate of bubble breakage in turbulent liquid flow was examined using a population balance containing a bubble breakage model to analyze experimental bubble breakage rate data. The bubble breakage model was based on high-speed motion photography observations of the breakage process in turbulent liquid pipeline flow. The bubble breakage model predicts the number of bubbles formed from a breakage, the size of the bubbles formed and the rate of bubble breakage. Bubble breakage was determined to be binary; bubble breakage size was described by a breakage size function in which unequal bubble sizes had a higher probability of being formed compared to equal bubble sizes; and the breakage rate was assumed to be first order with respect to the number of bubbles of a given size. The value of the breakage rate constant was found to be approximately equal to the frequency of the second mode of oscillation of the maximum stable bubble size in a given turbulent flow.

Molecular beam distributions from high rate sources

A mathematical model that describes the evaporated flux from high rate sources like those used in the deposition of semiconductor films were developed from mass balances with suitable constitutive equations describing the molecular beam distribution. The model was experimentally verified in laboratory scale equipment. Good agreement between the measured distributions and model results were obtained using a single fitting parameter. The fitting parameter is material specific and was correlated to the Knudsen number at a fixed location in the nozzle. A generalized procedure to evaluate this parameter and estimate the incident flux from a molecular beam source is outlined.

Economics of processing thin‐film solar cells

The commercial‐scale processing potential for photovoltaic devices based on polycrystalline and amorphous silicon thin films is evaluated by analyzing in detail the semiconductor deposition process for the following material layers: (CdZn)S/CuInSe2, CdS/CdTe, (CdZn)S/Cu2S, and amorphous silicon. The current status of both laboratory‐scale and unit‐operations‐scale experiments is tabulated and the goals required for reasonable commercial‐scale production detailed. The costs of manufacture are estimated and it is shown that all thin‐film materials which have achieved 10% efficiency in the laboratory can be manufactured at a cost well below one dollar per watt, provided the manufacturing facilities can be designed to produce between 100 000 and 1 000 000 square meters per year. The critical thin‐film research areas are identified for each of the materials listed above.

Reaction analysis of the formation of CuInSe2 films in a physical vapor deposition reactor

The reaction kinetics for the formation of CuInSe2 films by reacting Cu/In layers with elemental selenium are compared with those for H2Se. The species mole fractions as a function of time in a single Se-source physical vapor deposition (PVD) reactor are found to be essentially the same as those obtained in a chemical vapor deposition (CVD) reactor with flowing H2Se, indicating that the same chemical equation representation can be used in both cases. The chemical engineering reaction analysis model developed previously by us is shown to predict adequately the experimental data in both reactors. The model is employed to predict three-source behavior. The effects of rate of species delivery and substrate temperature on the time to make CuInSe2 is presented quantitatively.

Effect of temperature on copper indium selenization

The reaction kinetics of copper indium diselenide formation is studied by measuring species composition as a function of time at 250°C and 325°C in a tubular chemical vapor deposition reactor. This extends our previous modeling and experimental study at 400°C. The initial copper–indium alloy is analyzed at the reaction temperatures using high-temperature X-ray diffraction measurements. This modifies the understanding of the chemistry of the copper indium growth kinetics and a new set of model equations is presented. The specific reaction rate constants and activation energies for the chemical reactions are obtained, enabling one to calculate the holding time for the reaction.

Horizontal bubble flow

Abstract An experimental investigation of cocurrent bubble flow in 0.0254 m and 0.0508 m diameter horizontal pipelines has been performed. Gas and liquid mass velocities ranged from 0.00955 to 0.675 and 2720 to 6040 kg/m 2 sec, respectively, and gas-phase holdups or void fractions ranged from 0.13 to 7.59%. High speed motion pictures revealed that the gas, introduced into the liquid with a concentric nozzle, emerged in the form of a rough jet which was ultimately sheared into 1 times; 10 minus;3 to 3 times; 10 minus;3 m diameter bubbles. Approximately 4 meters downstream from the nozzle, a well developed bubble flow was observed where bubble number density and axial velocity were constant with respect to axial position in the pipeline. Bubble velocities ranged from 0.001 to 0.57 m/sec greater than the average liquid velocities. Bubble radial and circumferential spatial distributions were found to be a strong function of the degree of turbulence in the liquid phase. Because of these turbulent flow conditions, bubble shapes were much different than those of equivalent diameter bubbles rising in stagnant liquids. A sphere-ellipsoid of revolution model was developed for characterization of bubble shape and computation of gas-liquid interfacial area and two-phase pressure drop.

Chemical Vapor Deposition of Hydrogenated Amorphous Silicon from Disilane

The authors describe hydrogenated amorphous silicon (a-Si:H) thin films deposited at growth rates of 1 to 30 A/s by chemical vapor deposition (CVD) from disilane source gas at 24 torr total pressure in a tubular reactor. The effects of substrate temperature and gas holding time (flow rate) on film growth rate and effluent gas composition were measured at temperatures ranging from 360{sup 0} to 485{sup 0}C and gas holding times from 3 to 62s. Effluent gases determined by gas chromatography included silane, disilane and other higher order silanes. A chemical reaction engineering model, based on a silylene (SiH/sub 2/) insertion gas phase reaction network and film growth from both SiH/sub 2/ and high molecular weight silicon species, Si/sub n/H/sub 2n/, was developed. The model predictions were in good agreement with experimentally determined growth rates and effluent gas compositions.

Technology development versus new ideas development by universities

A logical approach to technology development which stresses the planning and interpretation of experiments useful for the design of commercial scale equipment is presented. The utility of the approach is illustrated by considering the design of a reactor to make copper indium diselenide.

Heat Transfer in Tubular Fluid-Fluid Systems

Publisher Summary This chapter provides a critical review of the current state of the art in the field of two-phase flow with heat transfer and discusses procedures that can be used for the design of tubular fluid–fluid systems. Both heat transfers without phase change and with phase change are discussed in detail. In each case, the analysis is based on an understanding of the flow patterns and the hydrodynamics of the system. In heat transfer with phase change, it is necessary to understand the phase-change phenomenon on the molecular level to model effectively the mass- and heat-transfer processes. The assumption of a saturated vapor phase greatly simplifies the calculation without a significant loss in accuracy. The design of two-phase contactors with heat transfer requires a firm understanding of two-phase hydrodynamics to model effectively the heat- and mass-transfer processes. The chapter also identifies the areas where further theoretical and experimental research is essential.

Continuous Deposition Of Photovoltaic Grade CdS Sheet at the Unit Operations Scale

Uniform photovoltaic grade CdS sheet has been reproducibly deposited on a continuously moving flexible substrate in a reel to reel vacuum coater. Materials characterization by scanning electron microscopy, photoluminescence, and resistivity revealed that continuously deposited CdS is essentially equivalent to material that was deposited for making high efficiency Cu2S/CdS cells in the laboratory scale process. Cells made using continuously deposited CdS sheet had efficiencies as high as 7.85%.