No abstract

No abstract

It has been reported previously (1) that the reaction of N, N'-(n's-(3-chlorophenyl)formamidine with ethyl malonate may be controlled so that the predominant product is the substituted acrylate (II) rather than the anilide (IV) obtained by Dains (2). Compounds of type II are of interest because of the ease with which they may be cyclized to 4-hydroxyquinoline derivatives which are intermediates in the preparation of certain important antimalarial drugs (SN-7618 or Chloroquine, etc.). The present work was undertaken with the purpose of determining the general applicability of the reaction for the preparation of substituted 4-hydroxyquinoline derivatives. Substituted diphenylformamidines are easily obtained in good yield and we have shown that several of these react with ethyl malonate in the same way as the m-chloro derivative studied originally. We have encountered an interesting effect of an orf/io-substituent on the rate of the reaction.In order to facilitate the investigation of the reaction of ethyl malonate with the formamidines, the acrylates (II) were not isolated, but after removing unchanged formamidine,3 the crude reaction mixture was subjected to cyclizing conditions, followed by saponification of the quinoline esters (VI) produced; the acids (VII) were precipitated by the addition of mineral acid and their weights taken as a measure of the extent of the first reaction. This is advantageous because it is difficult to separate quantitatively the low-melting acrylates (II) and anilides (IV). Both are cyclized by heating in an inert solvent, but the quinoline anilides (V) are not hydrolyzed by aqueous alkali and separate fairly completely from the alkaline solutions from which they may be removed before acidifying to precipitate the acids (VII). The yield of acrylate from the reaction of a formamidine with ethyl malonate is of course somewhat higher than the value implied by the weight of acid produced, since this figure includes the losses in the cyclization and saponification steps. That these losses are quite low and consistent, however, may be demonstrated by subjecting pure acrylates prepared from ethoxymethylene malonic ester (3) to cyclization followed by saponification under the same conditions employed with the crude acrylates. It is probable that the presence of anilide in the latter lowers the yield in the cyclization step

The purpose of the present study is to improve the current prediction capabilities of the entrainment fraction in horizontal gas-liquid flow. Since it is recognized that waves at the gas-liquid interface are the main source of entrainment, an experimental and theoretical work has been carried out to characterize the waves at the gas-liquid interface and to develop a model for entrainment calculations based on such characteristics. The model consists of three sub-models, namely, onset of entrainment, maximum entrainment and entrainment values in between. The onset of entrainment model determines the conditions at which the gas starts shearing the wave crests through a force balance between drag and surface tension forces. The maximum entrainment model provides the maximum fraction of liquid that can be entrained at high gas velocities by integration of the turbulent velocity profile to a determined dimensionless film thickness within the buffer sub layer. The entrainment fraction in between onset and maximum boundaries is calculated from an equilibrium between atomization and deposition rates. The atomization rate is calculated by first determining the wave mass flux in the liquid film and second by calculating the fraction of a single wave that is sheared by the gas through a force balance. The deposition rate is calculated as a linear function of the droplet concentration in the gas. Closure relationships have been developed from data for wave celerity, frequency, amplitude and width which are used in the entrainment model. A review of the most used correlations for calculating the entrainment fraction is presented and their performance evaluated. The present model shows better prediction than available models when compared to the acquired experimental data and the available experimental data in the literature.

No abstract