No abstract

A computer model of the simultaneous heat, mass, and momentum transfer processes occurring throughout an entire cooling tower is described in this paper. The model includes the flexibility to analyze the several configurations, fill arrangements, and flow distributions commonly used by the power industry. The fundamental governing equations are solved using a finite-integral technique to provide a quasi-two-dimensional description of the flow and cooling process within the tower. The model has been successfully compared with field data from cooling towers at three TVA power plants as well as data from other utilities. Each of these towers was significantly different in design, thereby demonstrating the versatility of the model for correctly predicting the cooling performance of mechanical and natural draft towers, as well as crossflow and counterflow orientations, for a range of meteorological and plant operating conditions.

This paper addresses an investigation of flue gas injection on natural draft cooling tower performance through numerical simulation. The control volume finite difference method was used for discretizing the governing equations in axisymmetric form on a boundary-fitted grid. The five independent variables addressed in this study are flue gas flow rate, flue gas temperature, radial injection location, injection orientation, and liquid entrainment in the flue gas. The flue gas temperature was found to have the most significant effect on tower performance (cold water temperature), because it strongly affects the buoyancy within the tower. The total air flow through a tower is driven by buoyancy forces, and the cooling performance is a strong function of the airflow rate.

Rate constants for the formation (kf) and dissociation (kd) of complexes of Mnxl with the ligands 1 ,I O-phenanthroline (phen), 2.2'-bipyridine (bipy), and 2,2',2"-terpyridine (terpy) have been measured, a t low temperatures in anhydrous methanol, by the stopped-flow method. A t 298.1 K and / = 0.20 mot I-1 (NaCIO,) : log (k$I mol-l s-l) (extrapolated) = 4.95, 4.24, and 3.72; AH$ = 10.6 f 0.4, 11.7 f 0.1, and 12.8 * 1 .O kcal mol-; AS,$ = -0.4 f 1.6, 0.1 f 0.1. and 1.3 f 3-9 cal K -l mol-l; log (kd/s-l) = 1 .I 5, 1.57, and -1.33; AH,$ = 13.5 f 0.3, 13.4 f 0.5, and 13.7 f 0.6 kcal mol-l; and AS,$ = -8.2 f 0.9. -6.4 f 1.9, and -18 & 2 cat K -l moI-l for phen, bipy, and terpy respectively. The kinetically determined stability constants in methanol for the complexes (log K, = 3.8,2-7, and 5.0 for phen, bipy, and terpy respectively) are very similar to values determined in aqueous solution. Values of AH,$ are significantly larger than the activation enthalpy associated with MnIImethanol solvent exchange and possible reasons for this unexpected behaviour are discussed.THERE is widespread agreement l-.l that rates of formation of metal complexes in aqueous solution are controlled largely by the rate of exchange of water molecules between the inner sphere of a metal ion (MS6n2f, where S is a solvent molecule) and the bulk solvent. An interchange mechanism (1) has been proposed in mostcases (L = ligand) in which rapid outer-sphere ion-pair or ion-dipole association (equilibrium constant KO) is followed by rate-determining solvent exchange (rate constant kex). If the concentration of metal ion is in a large excess the mono-complex is predominantly formed and the rate law is as in equation (2). In the simplestcases where K,[MS6n+] < 1, the second-order formation rate constant (kf) is given by equation (3). Although this equation has been extended with some success to (3) rates of formation reactions in anhydrous methanolJ5y6 further studies in a variety of non-aqueous solvents

No abstract