Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Part 1. Fatigue of Textile Composites . . . . . . . . . . . . . . . . . 1
Chapter 1. Fatigue Behavior and Damage
Evolution of 2D and 3D Textile-
Reinforced Composites . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2. Experimental methodologies . . . . . . . . . . . . . . . . . . . . . 5
1.3. Fatigue behavior and damage evolution in
2D E-glass plain weave textile-reinforced epoxy composite . . . . .. . . . . . . 9
1.3.1. Quasi-static tensile behavior and
damage observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3.2. Fatigue life and damage metrics . . . . . . . . . . . . . . . . . 15
1.3.3. Fatigue damage observation and evolution . . . . . . . . . . 18
1.3.4. Postfatigue mechanical properties and
damage observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.4. Fatigue behavior and damage evolution
in single-ply non-crimp 3D orthogonal weave
E-glass reinforced epoxy composite . . . . . . . . . . . . . . . . . . . 24
1.4.1. Quasi-static tensile behavior and
damage observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
1.4.2. Fatigue life and damage metrics . . . . . . . . . . . . . . . . . 34
1.4.3. Fatigue damage observation and evolution . . . . . . . . . . 40
1.4.4. Postfatigue mechanical properties and
damage observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
1.5. Fatigue behavior and damage evolution in 3D rotary
braided carbon reinforced epoxy composite . . . . . . . . . . . . . . . 49
1.5.1. Quasi-static tensile behavior and damage
observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
1.5.2. Fatigue life and damage metrics . . . . . . . . . . . . . . . . . 55
1.5.3. Fatigue damage observation and evolution . . . . . . . . . . . 58
1.5.4. Postfatigue mechanical properties . . . . . . . . . . . . . . . . 60
1.6. Fatigue behavior and damage evolution in
non-crimp stitched and unstitched carbon
reinforced epoxy composite . . . . . . . . . . . . . . . . . . . . . . . . . 63
1.6.1. Quasi-static tensile behavior . . . . . . . . . . . . . . . . . . . 64
1.6.2. Fatigue life and damage metrics . . . . . . . . . . . . . . . . . 67
1.6.3. Fatigue damage observation and evolution . . . . . . . . . . . 71
1.6.4. Postfatigue mechanical properties . . . . . . . . . . . . . . . . 73
1.7. Remarks and perspectives . . . . . . . . . . . . . . . . . . . . . . . 78
1.8. Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Chapter 2. Fatigue Limit: A Link to
Quasi-Static Damage? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
2.1. Fatigue limit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
2.2. Damage development stages and load
thresholds for quasi-static tension . . . . . . . . . . . . . . . . . . . . . 90
2.3. Damage development in quasi-static tension
and in the progression of fatigue loading . . . . . . . . . . . . . . . . . 93
2.4. Experimental data on the fatigue limit and the
quasi-static damage thresholds for textile composites . . . . . . . . . . 96
2.4.1. Fatigue limit for glass fiber reinforced
composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
2.4.2. Fatigue limit for carbon fiber reinforced
composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
2.5. Summary and conclusion on the fatigue life limit . . . . . . . . . 102
2.6. Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
Part 2. Fatigue of Short Fiber Reinforced
Composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Chapter 3. Experimental Observations of
Fatigue of Short Fiber Reinforced Composites . . . . . . . . . . . 109
3.1. Injection molded SFRC . . . . . . . . . . . . . . . . . . . . . . . . . 110
3.2. SN curve behavior of SFRC . . . . . . . . . . . . . . . . . . . . . . 113
3.2.1. Fiber-based parameters . . . . . . . . . . . . . . . . . . . . . . 115
3.2.2. Loading-based parameters . . . . . . . . . . . . . . . . . . . . 121
3.2.3. Environmental effects . . . . . . . . . . . . . . . . . . . . . . . 123
3.2.4. Specimen configurations . . . . . . . . . . . . . . . . . . . . . 125
3.3. Loss of stiffness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
3.3.1. Collection of loss of stiffness data . . . . . . . . . . . . . . . 131
3.3.2. Comparison of the loss of stiffness curves . . . . . . . . . . . 133
3.4. Future outlook and modeling strategy . . . . . . . . . . . . . . . . 136
3.5. Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Chapter 4. Fatigue Modeling of SFRC:
A Master SN Curve Approach . . . . . . . . . . . . . . . . . . . . . . 145
4.1. Overall framework and modeling strategy . . . . . . . . . . . . . 145
4.2. Choice of a mean field homogenization method . . . . . . . . . . 151
4.2.1. Benchmarking of schemes with full FE solution . . . . . . . 153
4.3. Damage modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
4.3.1. Fiber–matrix debonding: equivalent bonded inclusion approach . . . . . 159
4.3.2. Matrix damage . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
4.3.3. Validation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
4.4. MSNC approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
4.4.1. Scaling of SN curves using the endurance limit . . . . . . . 166
4.4.2. MSNC approach . . . . . . . . . . . . . . . . . . . . . . . . . . 168
4.4.3. Validation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
4.4.4. Comparison of the MSNC scheme with other schemes . . . . . . . . . 177
4.5. Component-level simulations . . . . . . . . . . . . . . . . . . . . . 181
4.6. Conclusions and future outlook . . . . . . . . . . . . . . . . . . . 184
4.7. Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
Preface
Safety, durability and reliability considerations of composite material components necessitate an in-depth knowledge of different aspects of the materials behavior. In particular, composite components exposed to cyclic loading throughout their life-time could and usually do suffer from mechanical performance degradation, namely ‘fatigue’. In the last three or four decades, intense and continuous investigation has been primarily devoted to the fatigue behavior of unidirectional laminated composites, as manufacturing of these long fiber reinforced plastics was suitable for the earlier applications in the aeronautic and naval industry. Nowadays, development of new resins, improvement of the impregnation techniques, and availability of a broad range of materials and architecture for reinforcements allow the application of composites for mass scale production of lightweight components. Therefore, there is high demand from various industries, such as the automotive, aeronautical, marine, and energy production industries, among others, for knowledge of the fatigue mechanisms and design of suitable fatigue resistant structural, as well as non-structural, composite elements subjected to long-term fluctuating loads.
In the broad family of fiber reinforced composite materials, textile reinforced and injection molded short fiber reinforced composites, in the Authors’ opinion, have some mechanical and manufacturing peculiarities suitable for several applications, also covering long-term variable loadings.
The automated, computer controlled machines for industrial manufacturing of textile reinforcements and the increasing machine operation speed makes such preforms competitive from the productivity and affordability viewpoints. A huge variety of interlacement architectures for composite reinforcements, ranging from two- to three-dimensional fabrics, are now available. These possess excellent drapability and versatility, which is extremely important for complex double-curvature shape components.
Ease of manufacturing, low cost and reasonable specific mechanical properties make short fiber reinforced composites (SFRC) an attractive material for a large number of industrial applications, particularly in the automotive industry. The injection molding manufacturing process allows rapid and inexpensive fabrication of composite components with complex features like notches, curved surfaces and irregular thickness. This attribute, combined with reasonable specific properties, makes them very attractive for a wide range of semi-structural applications.
This book presents an overview of the fatigue behavior of these two different reinforced composites, namely textile composites (Part 1) and short fiber reinforced composites (Part 2).
Part 1, dedicated to textile composites, gathers several understandings on the fatigue imparted damage evolution. Chapter 1 presents experimental investigations of the mechanical behavior and the fatigue damage development in 2D and 3D textile reinforced composites. For each material, a complete description of the damage evolution is detailed covering the prefatigue quasi-static, the fatigue and the postfatigue quasi-static tensile behavior. The prefatigue quasistatic tensile loading provides the relevant mechanical properties and understanding of the initiation and development of the damage mechanisms. The tensile-tensile fatigue loading, accompanied by the fatigue life diagram, gives in-depth understanding on the damage modes sequence related to the reinforcement architecture. Moreover, analogies of the static and cyclic damage mechanisms are highlighted to obtain an initial connection of the damage thresholds and the “fatigue limit”, extensively discussed in Chapter 2. The postfatigue quasi-static behavior points out the effect of the fatigue on the mechanical properties and on the modification of the initiation and development of the damage modes.
The aim of Chapter 2 is to extend the notion of “fatigue limit” (namely, the cyclic load level below which the fatigue failure at a given large number of cycles does not happen) to textile composites. In particular, the relationship, if it exists, of the damage thresholds and the fatigue limit. The considered damage thresholds (i.e. the load levels distinguishing the different stages of the damage evolution) were identified, as in Chapter 1, using acoustic emission (AE) registration during the quasi-static tensile loading. The extensive dataset of various glass and carbon fiber reinforced thermoset textile composites confirms the link between the quasi-static damage thresholds and the fatigue life limit. However, the complex fatigue behavior of textile composites does not indicate a unique correlation of the fatigue limit and the quasi-static damage thresholds. This correlation is shown to be very different for glass and carbon reinforced textile composites.
Part 2 collects different perspectives on the fatigue response of short fiber reinforced composites. Chapter 3 compares a wide set of experimental measurements to highlight the dependency of the fatigue behavior on a large number of factors, including: fiber and matrix materials, fiber distribution, environmental and loading factors. Fatigue behavior of composites manifests itself in two easily observable trends. First is the SN curve; it represents the number of cycles to failure if a certain load is applied. Second is the loss of stiffness; SFRC materials suffer from continuous loss of stiffness during cyclic loading. Both these aspects are described in detail with emphasis on statistical comparisons to underline dependence on various factors. This chapter reinforces the different challenges one must overcome during the fatigue simulation of SFRC, some of which are dealt with in Chapter 4.
In Chapter 4, a component level simulation method for SFRC is detailed by introducing a new multiscale hybrid modelling concept for fatigue simulation. This proposed approach combines test results and simulation results (hybrid approach) on different scales (microscopic simulation, macroscopic fatigue behavior), overcoming the problems of depending on too many expensive tests (pure test-based macroscopic approach) and a lack of efficiency (pure microscopic simulation approach). Four different steps are identified for successful component level simulation. First, the appropriate homogenization scheme is recognized. Next, the micro-mechanics damage modelling of SFRC is built within the framework of the chosen homogenization scheme. The damage at the micro-scale is then related to the macroscale fatigue properties by means of a Master SN curve approach. Finally, all of the elements are integrated with manufacturing simulation software and fatigue solvers to achieve component level simulation. Each of the four steps are described in detail with validation of the models and examination of the assumptions at every step of model development.
Fatigue of fiber reinforced composites, like a number of other engineering problems, offer two contrasting approaches. The first is that of a material scientist who would like to understand reality with the highest possible accuracy, the second is that of an engineer who desires to develop tools to safely design parts which might be subject to complex conditions. Though not mutually exclusive, every investigation of fatigue of composite materials has in general a certain bias towards one of the two approaches. In this book, we have tried to give a flavor of both approaches: Chapter 1 can be viewed as a scientists’ approach towards the understanding of fatigue of textile composites; whereas Chapter 4 presents a predominantly engineers’ approach to fatigue of composite materials. In both cases, however, the perspective of the other (scientist or engineer) has also been respected and dwelt upon at different points.
The results and understandings collected in the book are achievements of several years of fruitful collaborations with many colleagues and friends. We particularly acknowledge: Giulia Gramellini, Juan Pazmino and Vanni Neri Tomaselli (former masters students at Politecnico di Milano and Erasmus Visiting Scholars in KU Leuven); Yasmine Abdin, Jose Manuel Beas, Katleen Vallons and Ignaas Verpoest (KU Leuven); Alexander E. Bogdanovich and Dmitri D. Mungalov (former members of R&D department at 3Tex Inc.).