Étude d'un pilote simulant un réseau et permettant de tester la qualité d'un matériau en contact avec l'eau

Study of a network simulating pilot to assess the quality of materials in contact with water

Centre de Recherche et de Contrôle des Eaux de la Ville de Paris 144-156, avenue P.-Vaillant-Couturier, 75014 Paris

Résumé

Un pilote a été conçu pour apprécier le comportement bactériologique, microtoxicologique et physicochimique des matériaux en contact avec l'eau (tuyaux, joints, revêtements). Ce pilote qui reproduit les conditions réelles d'exploitation des réseaux de distribution est un système dynamique fermé.

Dans un premier temps, une étude de l'évolution dans l'eau d'une population bactérienne bien définie a été entreprise sur quatre souches usuelles de l'eau. Elle a montré qu'après une période de croissance les bactéries atteignent un état quasi-stationnaire, très stable, qui permet d'observer les modifications dues aux tuyaux. Une étude systématique de tous les facteurs de variabilité des résultats (effet mécanique, force ionique, qualité de l'eau, nutriments) a abouti à la totale maîtrise du système.

Tous les types de matériaux peuvent ainsi être testés dans tous les types d'eau et dans une large gamme de températures, aussi bien du point de vue microbiologique que microtoxicologique et physicochimique. Des résultats pratiques sont montrés.

Abstract

A dynamic testing device was defined to evaluate the bacteriological, microtoxicological and physicochemical behaviour of materials (tubings, gaskets, coatings) in contact with drinking water. This closed loop circuit reproduces the real conditions of the domestic distribution networks.

In the first step, the evolution of a well defined bacterial population was studied with four strains isolated from water. We determined that bacteria grow until they reach a stable quasi-stationary phase.

In a first time, the irreproducibility of the kinetics meant that some factors of variability were out of control. To gain a total control of the system, a systematic study of the interfering parameters was performed. It showed that the stationary phase was highly influenced by the variable quality of the network water, such as the periodical presence of micropollutants in distribution water, which are toxic towards bacteria. Moreover, it showed that traditional washing of bacteriological glassware was a decisive factor of irreproducibility, since by burning the glassware at 550°C the reproducible growth of bacteria was obtained. The total organic carbon (TOC) level also was shown an essential parameter, which is of particular importance for the Pseudomonas strain. Bacteria growth was also influenced by other parameters such as the quality of the peristaltic pump tubing and the water ionic strength and dissolved oxygen.

The control of these parameters gave reproducible and stable quasi-stationary phases. Constant quality springwater was used to avoid the variable toxicity level of tap water. Finally, when the system was under control the variation of the tubing quality was easily detected.

As a conclusion, this study demonstrated that in a well controlled environment, every type of materials can be tested in every type of water, in a wide range of temperatures, at the microbiological, microtoxicological and physicochemical levels. Practical results are shown.

As a conclusion, it is clear that the use of a well defined and stable water and the use of a well defined inoculum are mandatory for valuable interlaboratories assays.