Evaluation de la birefringence en ligne et modelisation du procede de soufflage de gaine multicouche (French text).

摘要

The recent development of multilayer film from the combination of a variety of different polymer resins allows the elaboration of technical films with enhanced optical and barrier properties. Our objective is to develop tools that will allow us to improve our knowledge of the multilayer film blowing process.; In order to gain general knowledge of the process an experimental campaign was realized to identify the effect of the main processing conditions (take up ratio, blow up ratio, temperature of the cooling air flow, etc.) on the radius and temperature profile of monolayer and multilayer bubbles. We were able to observe that an increase of the blow up ratio and a decrease of the take up ratio resulted in a diminution of the solidification position of the polymer melt.; The second step we took was to improve the in-line birefringence measurement system in order to be able to complete a scan of the bubble length in less than one minute. From these data we were able to evaluate the average axial and transverse stress generated inside the monolayer and multilayer bubble during their formation. This allowed us to relate the process conditions to the axial and transverse stress profile.; A Newtonian thermo mechanical model that describes the multilayer film blowing process was developed. In order to obtain the appropriate set of equations a novel approach was used instead of the classical development (for the monolayer film blowing process) which is based on the thin membrane theory and its use of a mobile reference framework tangential to the skin surface of the bubble. The mechanicals and thermal equations were developed in a global fixed reference framework. This caused the appearance of a small parameter, the ratio of the die gap over the die radius, in all the equations. This allowed us to realize an expansion development on the variables which make it possible to analyze the equations at various orders of magnitude of the parameter. With this approach it is possible to demonstrate results which were previously assumed. For example, we are able to demonstrate that the temperature profile in the radial direction is at least one order of magnitude lower than the temperature profile in the axial direction. This validates, in a formal way, hypotheses which were previously made. It is also possible to develop the heat transfer equation at a higher order of magnitude in order to be able to derive the radial temperature profile. (Abstract shortened by UMI.)