S. Elmaleh

Theorie de l'epuration par filtration biologique immergee

Resume Submerged biological filtration results from a combination of deep bed filtering and biological treatment. It is a valuable process when used after a physico-chemical treatment or in denitrification for drinking water. A mathematical model for dissolved pollution abatement is proposed, based on a limiting factor which is oxygen concentration in the biolayer. The results of this phenomenon are (i) apparition of a limiting biolayer thickness beyond which per cent removal is no longer a function of biolayer thickness but a function of biolayer area. (ii) apparition of soluble and insoluble metabolic products which reduce the apparent effectiveness of biological filter. This model has four parameters and the results are discussed and compared to experimental results.

Influence sur la cinetique biochimique de la concentration en carbone organique a l'entree d'un reacteur developpant une polyculture microbienne en melange parfait

Resume Des experiences ont ete entreprises dans le but de determiner la cinetique de croissance microbienne dune polyculture alimentee en melange parfait par une solution dun substrat complexe. Il apparait que la concentration de leffluent ne depend pas seulement du temps de passage impose mais aussi de la concentration en substrat applique a lentree du reacteur. Une cinetique a pu etre determinee a laide du carbone organique total comme indice de pollution: k = k o ( S / S + αS o )

Biological filtration through a packed-column

Abstract A model of the biological removal of dissolved pollutants by filtration through a packed-column has been developed. The theoretical predictions concerning porosity change, oxygen depletion and purification efficiency have been tested with a dissolved substrate: the model allows, after laboratory trials, the design and optimization of a filter.

Etude experimentale de la filtration biologique immergee

Abstract The results of submerged biological filtration results in deep beds have been examined. By analysing the different operating variables and by the observation of biomass, experimental results enable one to verify how representative is the model established by Grasmick, Elmaleh & Ben Aim, published in Water Research , (1979). A simple experimental method of characterizing the degradability of an effluent, is proposed. This enables one to determine the optimum dimensions of the reactor by using the model equations.

Epuration hydrocarbonne en couche fluidisee triphasique etude experimentale et modelisation

Abstract Biological treatment by immobilized cells on a submerged support allows important volume reduction of units owing to the high micro-organism concentration in the process. The efficiency of classical reactors, i.e. preoxygenated fixed bed, is, nevertheless, limited by oxygen supply to the reactive medium and the unit must be worked discontinuously because of the bed clogging. A 9.4 cm i.d. three-phase fluidized reactor has been tested in an attempt to solve these problems; the turbulence induced by air injection should remove the excess biomass. Step injections of synthetic sewage have been made for different values of the working parameters. Interesting efficiency results have been observed: elimination efficiency of 10 kg BOD 5 m −3 day −1 without recycle, is independent of space-time over a wide range. Limitations of a three-phase fluidized reactor have also been found: insufficient oxygen transfer to the liquid phase due to rapid bubble coalescence is a consequence of fluidizing small particles and inefficient sloughing which allows the reactor to be segregated depending on biofilm thickness and, sometimes, clogged in the upper part of the column. A model based on previous work in fixed bed has been proposed. Interphase mass transfer and zero order reaction within the biofilm are taken into account and the liquid phase is assumed to be perfectly mixed. The two parameters of the model are a combination of the numbers of transfer units within the biofilm and through the boundary layer and of the number of reaction units in the biofilm. The limiting substrate may be oxygen, organic carbon or any other nutrient. Experimental efficiencies can be predicted by the model with 15% accuracy. Particles size distribution must be corrected with regard to segregation. Physicobiochemical assumptions can be used in the modelling of a reactor which would not be perfectly macromixed and could be basically used in a unified theory.

Transient state microbial growth in perfect mixing

Abstract Most experimental studies about bacterial growth in transient state have shown that the specific growth rate changes cannot be predicted by classical kinetic equations. The models generally proposed to fit with experimental results are highly complex and sophisticated. However, simple considerations about mass transfer through a standard cell allow to derive a transfer function between the substrate concentration and the specific growth rate. This modelization may take also into account the influence of an inlet multicomponent substrate concentration on the equilibrium in a chemostat. The theoretical equations have been used in the prediction, with good precision, of the response of a polyculture in perfect mixing grown with beef extract to a step disturbance on the inlet concentration.

Representation de l'ecoulement macroscopique dans un bassin d'aeration par une cascade de deux melangeurs parfaits inegaux

Resume Le melangeage dans un bassin daeration au cours de lepuration par boue activee a ete relativement peu etudie. On admet generalement un regime piston ou dagitation parfaite. On a determine la DTS dans un bassin de 12 m3 aere par un systeme a moyennes bulles. Le melangeage est independant du debit liquide dans la gamme des essais mais varie en fonction croissante du debit gazeux. Les courbes de DTS ont ete interpretees a laide dune cascade de deux reacteurs parfaitement agites inegaux. On ajuste la courbe experimentale a laide de labscisse de son maximum. On a pu ainsi relier le parametre du modile au debit daeration. La concordance entre les courbes experimentales et calculees est bonne.

Influence du temps de passage et des conditions d'agitation sur la taille du floc developpe dans une unite a boues activees

Influence of residence time and mixing intensity on pollution abatement, respirometric activity and floc size has been studied in a mechanically stirred activated sludge laboratory plant. Preliminary results confirm a tight dependance between floc size and shape and efficiency, bacterial activity and effluent clarification.

Détermination du macromélange du Liquide lt des solides dans une unitéà Boues activées en fonctionnement normal avec recyclage

Resume Les ecoulements des phases liquide et solide dans les unites a boues activees ont ete relativement peu etudies. On a determine les distributions des temps de sejour (DTS) des deux phases en laboratoire sans perturber le fonctionnement de linstallation. Les courbes de DTS ont ete obtenues par deconvolution des signaux dentree et de sortie pour la phase liquide et par extrapolation du signal de sortie pour la phase solide. Lexploitation des DTS des deux phases a permis de mettre en evidence une diference entre les moyennes des temps de sejour pour les deux phases. Cette differences est expliquee par la distribution de porosite dans le bassin daeration. Ces resultats montrent linteret de la distribution des temps de contact dans le calcul des unites.