ForschungListe aller Forschungsprojekte
Micro-Mechanically Based Modeling of Degradation of Composite Materials with Random Microstructure

Micro-Mechanically Based Modeling of Degradation of Composite Materials with Random Microstructure

Leitung:  Prof. P. Wriggers, Dr. F. Daghia
Team:  Dipl.-Ing. V. Krupennikova
Jahr:  2016
Förderung:  DFG (Graduiertenkolleg 1627)
Bemerkungen:  Prof. P. Wriggers, Dr. F. Daghia

High-performance composite materials, such as carbon-fiber-reinforced polymers, serve as construction materials in various lightweight structures, subjected to high loads. Among these structures are aircraft parts, rotor blades of wind turbines, bridge constructions, etc. [1]. Indeed, the materials must endure the corresponding mechanical loads over their entire service time. Severe environmental conditions, such as moisture, oxidative atmosphere, radiations etc., affect the material microstructure and properties and thus induce material degradation which can lead to premature failure under mechanical loads.

Degradation of the mechanical properties of CFRP due to oxidative environments is of outstanding importance (in Earth atmosphere) and has been extensively investigated in various works. The effect of oxidation on the polymer matrix is proposed as follows [2]. Oxidation of the polymer matrix leads to formation and evolution of volatile molecules and results in the material loss leading, in turn, to the matrix shrinkage (strain). In composites, the oxidation induced strain in the matrix leads to local stress build-up which is determined by heterogeneity in both the composite microstructure and the oxidation state. This mechanism has been validated at the matrix thin film scale [2] and successfully utilized in micro-mechanically based modeling of composite material behavior upon oxidation at the fiber-matrix scale [3].

One of the particular findings presented in the literature [3] is the effect the spacing between the carbon fibers exhibits on the matrix displacement profiles measured on the free surface of oxidized samples: larger matrix displacements correspond to larger distances between the fibers. The strain induced stress releases via fiber-matrix debonding and micro- and meso-crack formation, primarily (as it is observed experimentally) in so called “matrix rich” areas. Thus, a microstructure dependent local decrease of the load transfer capacity of the matrix is expected. The goal of the work in progress is to develop a framework for analysis of the degradation of CFRP upon oxidation which accounts for the random distribution of the phases as a source of variable spacing between the fibers.  

The random microstructure will be taken into account via discretizing the composite material ply by means of a voronoi mesh approach. The largest voronoi cells will be analyzed with respect to fiber-matrix debonding and micro-crack formation due to stress build-up in the matrix, i.e. due to the matrix oxidation. As soon as debonding occurs the stress redistribution within the matrix requires considering a larger domain (the initial voronoi cell and all adjacent voronoi cells) for analysis of the material behavior under oxidation. Micro-crack propagation into adjacent voronoi cells will lead to stress redistribution within the bulk beyond the domain considered, and will require further incorporation of vicinal voronoi cells. In this way the development of meso-cracks can be followed until the predicted level of oxidation is reached.

After certain level of oxidation is reached (depending on oxidation conditions and duration), the domains consisting of damaged voronoi cells will be incorporated into a bulk of a virtual specimen by means of the global-local approach. Finally, the virtual material will be subjected to a virtual tensile test with the load direction perpendicular to the fiber orientation. The effect of micro- and meso-scaled damage is expected to reveal itself at the ply and laminate scale via a decrease of the ultimate stress at failure, in turn caused by a decrease of the matrix load carrying capacity and formation of new sites of stress concentration (tips of cracks). The validity of this approach will be examined via comparison of the virtual results with the results of the real experimental tensile testing.                       

Currently the project is driven on both tracks, the development of the framework for the analysis of CFRP degradation under oxidation and the experimental setup for validation. Implementation of the coupled chemo-mechanical model of matrix material behavior [3] in an AceGen element is the main focus concerning the development of the modeling framework in this stage. The composite material specimens are currently being exposed to the experimental conditions (oxidative environment: air, 150 ˚C); fig. 1 shows the polished surface of a specimen before exposure (a), and after exposure for 15 days (b).

The framework for analysis of CFRP degradation upon oxidation will assist in prediction of the material secure service duration as well as in design of composites with improved properties. The procedure can be further adapted to model the mechanical behavior of other dispersed materials characterized by randomness of microstructure.   



[1] R.F. Gibson, Principles of Composite Material Mechanics. CRC Press Taylor & Francis Group, Boca Raton (FL), 2012.

[2] X. Colin, J. Verdu. Strategy for Studying Thermal Oxidation of Organic Matrix Composites. Composite Science and Technology, 65, 411-419, 2005.

[3] M. Gigliotti, L. Olivier, D.Q. Vu, J.-C. Grandidier, M.C. Lafarie-Frenot. Local Shrinkage and Stress Induced by Thermo-oxidation in Composite Materials at High Temperatures. Journal of the Mechanics and Physics of Solids, 59, 696-712, 2011.