Polymer matrix composites do not exist naturally, so they must be manufactured. Behind this simple statement hides a complex reality with strong consequences for these materials. In fact, due to the great diversity of resins, reinforcements and additives and the great freedom of shape and size offered, these materials prove to be extremely versatile since they can be shaped by adjusting their properties to the specific requirements of an application. This flexibility nevertheless introduces a counterpart. More than any other material, composites require close integration of knowledge of the constituent materials, the study of manufacturing processes and the performance of the parts obtained. Although the properties of composites are mainly determined by those of their constituents, experience proves that their method of development has a significant influence on their level of performance in at least two ways:
-
by the generation of molding imperfections more commonly referred to as defects which alter the performance of the part (porosity, fiber misalignment, under-polymerization of resins, internal stresses, etc.);
-
by modifying the parameters defined during the design phase of the composite part (orientation and distribution of fibers).
Manufacturing technology is therefore an essential link in the design-manufacturing chain of the composite part. It thus plays a large part in the growth of composites due to the costs it generates, its ability to transform technical parts of varying sizes, its flexibility of use, the speeds it allows and its reproducibility. The choice of a manufacturing process is also guided by its environmental impact and by the adequacy with the technical need.
A key factor in the development of technologies, and therefore of composites, is the ability to develop processes that meet the requirements of the target market. Over the past three decades many technologies have been developed to serve the mass market and high performance markets. Although empiricism has often accompanied the first steps of these technologies, the most important progress has come from the effort to understand and model the physical phenomena associated with these processes. More or less sophisticated models have been developed on the basis of knowledge that is still partial but nevertheless sufficient to help engineers and technicians to implement robust technologies. They help to understand how process parameters and constituent characteristics affect the final properties of composite parts. Historically, design or calculation offices were the first to use business software to compare manufacturing strategies and choice of materials and thus provide support to production teams. The current challenge is that tools, methods, models, rules are also used directly in manufacturing workshops to:
-
anticipate real practical difficulties;
-
respond in a very interactive way to production contingencies;
-
optimize composite molding techniques;
-
define the best process monitoring strategy (measurement of pressure, temperature, etc.);
-
reduce process development times.
Thus, this type of approach makes it possible to gradually replace the very strongly anchored empiricism in the world of manufacturing workshops with predictive quantitative rules.
The purpose of this article is to present the main phenomena involved in the molding processes of structural composites with a polymer matrix and to propose a simplified mathematical description which derives directly from the major physical principles adapted to heterogeneous and anisotropic fibrous media. Structural composites refer to thermoplastic or thermosetting polymers reinforced with continuous fibers. Molding covers technologies where polymer flows play an important role in the manufacture of the part.