In this paper, a stochastic approach, namely, a Markov chain with transient and absorbing states, has been used to generate the residence time distribution for any conceivable flow model. All calculations have been carried out with the aid of a computer. The present results compare favorably with those obtained by earlier investigators. However, the present approach is applicable to complex flow systems for which the deterministic mathematical description of them is very complicated, or impossible. The mean and variance of the residence time distribution are readily obtained from this model. Potential applications of this model are numerous, especially for particulate systems.

The effect of cavity size and cavity density on surfaces with artificial nucleation sites, on bubble diameter and trequency of departure in pool nucleate boiling was studied by using high-speed photography and laser fiber probe. A theoretical model has been developed from the analysis of force equilibrium of a bubble on a nucleation site with dueconsideration of the mutual influence of the surrounding bubbles. Preducted results obtained from the proposed model agree satisfactorily withexperimental data.

A comprehensive model for nucleate pool boiling has been developed to predict the effect of cavity size and cavity density on boiling heat transfer. The results of experimental investigations on surfaces with artificial nucleation sites confirmed the theoretical model satisfactorily.

Three methods to determine UNIFAC parameters by chroniatography have been compared. The method of multiple activity coefficients at infinite dilution has been found inferior to the other two methods. According to an analysis on the sensitivities of UNIFAC parameters to γ∞ and (dγ/dx)x=0, it has been discovered that the effect of γ∞ is greater than that of(dγ/dx)x=0. Estimates of parameters and their ranges of values above an appropriate confidence level are discussed.

Dual hot wires were used to separate the simultaneous signals due to temperature fluctuation and velocity fluctuation appearing in high tempera tureturbulence measurements. The coefficients in the sensitivity equation inmeasuring turbulent fluctuation with constant-temperature hot-wire anemometry were deducted, combined with the temperature fluctuation sensitivity equation of constant-current hot-wire anemometry, thus separating the temperature-fluctuation component from the mixed signal, and leading to more accurate determination of the turbulent velocity fluctuation signal and other main turbulence parameters.

A mathematical model for describing the start-up period of a plate distillation column operating under total reflux was derived on the basis of the differential-difference equation expressing the material balance on the trays with holdups in the reboiler, trays and condenser, and an analytical solution was presented. The existence of a constant-composition point during the start-up period was proved mathematically. Experimental work on start-up was carried out in a seven-sieve-tray distillation column of 0.17 meter in diameter and operating with a water-acetic acid system. Good agreement was found between experimental data and calculated results. Experiments also showed the evidence of a constant-composition point in the column during start-up.

The principles of mass and heat transfer for non-isothermal gas absorption with chemical reaction have been applied to the derivation of a mathematical model describing the complexing of carbon monoxide with aluminum cuprous tetrachloride in toluene and the absorption of carbon dioxide with aqueous monoethanolamine, both in packed columns. The fundamental differential equations of the process were treated with the fourth-order Runge-Kutta method, leading to the establishment of the temperature and concentration profiles for both the gas and liquid phases along the packed height. The results conform reasonably well with pilot plant data.

Interfacial emulsification often occurs in the extraction of biological fermentation liquors containing organic acids. As a result of theoretical analysis and experimental work, certain substances present in the fermentation liquor have been identified as the cause leading W/O emulsions. Emulsification caused by surface activity of certain extractants was also examined by using microphotography for both W/O and 0/W emulsions, and a general discussion was then presented on the three phenomena of emulsification, third phase formation and interfacial contamination in extraction processes.

A series of dye-cellulose compounds and their corresponding models were synthesized. The alkali-hydrolysis rate constants of these compounds were measured, and the kinetics and mechanism of the hydrolysis were investigated. The molecular orbital theory was used in analyzing the hydrolytic behavior of twenty compounds including their models. It was found that the hydrolytic behavior of any dye-cellulose might be well approximated by its corresponding model. Fukis superdelocalizability SrN is a good reaction index for correlating the hydrolysis reaction.

A new method is developed which calculated vapor-liquid equilibrium of mixtures directly from the vapor pressure equation of pure components, by using the two-fluid corresponding-state principle at atmospheric or reduced pressures. This method is as rigorous in theory as that based on the equation of state, and as simple in computation as the activity coefficient method. This method may be used satisfactorily in calculating vapor-liquid equilibrium when coordinated with the vdW and Wong mixing rules, but for only homogeneous systems. As an improvement, a simple energy-local-composition mixing rule is suggested. Tested in computing the vapor-liquid equilibria of 37 typical noupolar-nonpolar, polar-polar, and polar-nonpolar binary systems, the proposed method, combined with the suggested rule, has been shown to be more accurate than the one-parameter Wilson equation, but slightly inferior to the two-parameter Wilson equation. The parameter values correlated seem regular against hydrogen-bond change in mixing course of systems.

Generalized power and Metzner constant Ks correlations for both combined helical-ribbon-anchor and helical-ribbon impellers in highly viscous Newtonian and non-Newtonian fluids and operating in the laminar regine are presented on the basis of regressive analysis of experimental data from this work as well as those from the literature. For combined helical-ribbon-anchor impellers the correlation is where for helical-ribbon impellers Moreover, it was confirmed from the results that power consumption for heli cal ribbon impellers is not directly proportional to the length of the impel-ler.

This paper presents two basic equations for computing equilibrium compositions from total pressure for a closed binary vapor-liquid system at constant temperature. Methods for solving these two equations have been explored with an analysis of the error of computation, practical examples of these two equations are then given.

Preparation of ethylene glycol acetate from ethylene using the catalytic system, PdCl2-CuCl2-Cu(OAc)2 in HAc solution, usually involves two steps -acetoxylation and regeneration of catalyst. In this paper, a one-step process has been studied, with oxygen and ethylene introduced simultaneously into a single reactor. It was observed that there must be sufficient partial pressure of oxygen to oxidize Cu+ to Cu++, by keeping oxygen concentration at 8-10% in the raw binary gas mixture of oxygen and ethylene, whereby the activity of the catalytic system could be maintained without formation of CuCl2 residue. By using nitrogen as a diluent, the ratio of O2/C2H4 could be kept in the non-explosive range, and it was found further that 95% of the ethylene in the raw gas could be converted into ethylene glycol monoacetate and diacetate.

An analytical solution for an unsteady state dispersion type tubular reactor model with an immobile zone and recirculation is obtained by a method using the theory of biorthogonal expansion oi operators.