In the past decades, much concern has been expressed regarding organic micropollutants generating endocrine disruption. In particular, estrogenic endocrine disruptors, compounds that interfere with the estrogenic hormonal mechanisms, are in the center of environmental scien- tists attention. Numerous endocrine disrupting effects have been observed at concentrations corresponding to those found in aquatic environments, such as feminization of fauna, infertility, reproductive disturbance, cancer, or developmental disruption. Wastewater treatment plants ef- fluents have been identified as the main source of estrogenic endocrine disruptors in the aquatic environment, due to inappropriate treatment. Promising alternative treatment systems based on the use of ligninolytic enzymes (e.g. laccases) have recently emerged. This work falls within the framework of these new techniques. Althoughno environmentally safe threshold can clearly be set, focusing on the removal of global estrogenicity in water instead of concentrations of targeted estrogenic compounds seems a relevant research. To our knowledge, the use of these recently emerged enzymatic processes at an industrial scale (such as for the treatment of urban wastewater), oriented towards water overall estrogenicity, has not been implemented yet.The general objective of this work was to develop ans study a process of removal of estro- genicity by laccases from white-rot fungi, in laboratory, with the purpose of design and scale-up for industrial implementation. This work consisted of the conception, characterization, testing, study and modeling of this process.First, in order to enable the study and the scaling-up of a process of estrogenicity removal, op- timizing the technologies of quantification of estrogenicity in water was a real necessity. Therefore, a study of the methods of estrogenicity monitoring, within a treatment process framework, was conducted. Based on a wide literature review, existing methods were gathered and assessed with the aim of their use as monitoring and design tools. Fromthat review, four methods were selected and tested according to numerous criteria and their compatibility with our process: three bioassays (Yeast Estrogen Screen assay; Lyticase-assisted Yeast Estrogen Screen assay; MCF-7 cell based reporter gene assay) and one analytical method (High Performance Liquid Chromatography with UV detection, HPLC-UV). Concentration-response curves towards the reference 17β-estradiol, in several solvents, were acquired. A fitting model was developed for further expression of all measurements in Estradiol Equivalents. The methods were used to evaluate the estrogenicity of bisphenol A, triclosan and nonylphenol, along with estrogenicity of mixtures of bisphenol A and 17β-estradiol. The testing of these four methods enabled the assessment of their sensitivity, re- producibility, and implementation. Based on that experimental assessment, the Lyticase-assisted Yeast Estrogen Screen (LYES) assay was selected and systematized to be further applied, in combination with an adapted protocol of HPLC-UV analysis, to the monitoring of estrogenicity removal in two lab-scaled reactors. The LYES assay demonstrated a real methodological potential for thescale-up of an estrogenicity removal process and could be used as a design tool. For both reactors, results have indicated that HPLC-UV and LYES assay methods are completely inter- changeable in the case of bisphenol A monitoring (in the conditions used in this work). This work also highlighted the peculiar behavior of mixtures of bisphenol A and 17β-estradiol in terms of estrogenicity, and the parallel observation of an enhancement of bisphenol A estrogenicity removalin presence of 17β-estradiol.In the second part of this work, a batch reactor with laccases in solubilized form was developed and estrogenicity removal was assessed. Kinetics data for the degradation of estrogenic endocrinedisruptors were acquired with the LYES assay (estrogenicity) and the HPLC-UV method (concen- trations). Degradations were performed from several solutions of bisphenol A, 17β-estradiol, and mixtures. In the case of 17β-estradiol and mixtures, conversions reached minimum 90% within 1 hour of degradation at our conditions, with no dependency on pollutant initial concentrations. In the case of bisphenol A, conversions varied from 0 to 100% after 6 hours of degradation and were shown tobe strongly dependent on BPA initial concentrations, indicating the laccases deac- tivation by substrate. Bisphenol A byproducts of degradation were also analyzed, which indicated their absence of estrogenicity and their potential linear evolution with BPA degradation. Acquired kinetics data were exploited for the modeling of the batch degradations kinetics. At this stage of the work, no clear kinetics conclusions could be made in this part.From an industrial point of view, the use of immobilized enzymes is more cost-effective, due to the improvement of enzymes recovery and stability. Therefore, in the third part of this work, a continuous column-shaped packed-bed reactor composed of laccases immobilized on a silica support was developed. The packed-bed reactor was built and tested in laboratory. Residence time distributions, pressure drop calculations, and tracer degradations were performed on the re- actor for its characterization.Immobilization and activity measurements protocols were applied. Similar monitoring than in the batch reactor (LYES assay and HPLC-UV) were performed during continuous degradations of bisphenol A, 17β-estradiol, and mixtures in the packed-bed reactor. Acquired kinetics data were exploited to study and model the kinetics occurring in the packed-bed reactor. Mass transfer phenomena and laccases deactivation by substrate in the reactors were investigated and modeled, revealing, depending on the pollutant, the presence of slight external mass transfer limitation and the strong dependence of the kinetics on laccases deactivation. A design tool, in the form of a mathematical model for the design of a packed-bed reactor with immobilized laccases for the degradation of bisphenol A and related estrogenicity, was developed. The model was validated (simple validation) on experimental data. The model was then used, as a comparison, for the design of a reactor with similar conditions than a documented technique of bisphenol A degradation by ozone. The design resulted in a reactor twice smaller than for degradation by ozone, for the same conversion.In the current context of environmental crisis, this work proposed a first version of a promising practical solution for the treatment of environmentally problematic e-EDCs, oriented towards water overall estrogenicity monitoring and removal, and using natural biocatalysts. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished