Soutenance mémoire
Architectured materials
Introduction of architectured materials
Architectured materials, which inherits from the concept of composite and hybrid materials, refers to heterogenous materials in which the topology of different phases is engineered in order to yield structural or functional effects [1,2]. It is an emerging concept in recent years. The motivation comes from material development in order to meet more and more exigent requirements for industrial applications. Taking Figure 1-1 (a) for example, it shows a property map between Young’s modulus and density of different material families [1]. While most space in the middle of the map is occupied, some blank areas appear on the chart. One hole is the upper zone on the left, which presents the combination of a low density and a high Young’s modulus. This can be very interesting to reduce the weight but keep the high strength of some structures at the same time, such as lightening automotive structures to reduce CO2 emission. However, this combination of properties is hard to access by classical homogeneous materials, architectured materials are born to meet the need of pushing the limit of existing property maps [3].
Metallic architectured materials
In this section, metallic architectured material is introduced individually because it is more closely related to the present work. The constitution of metallic architectured materials is usually related to composition gradient or localized deformation. Metallic compositionally graded materials are the most intensely researched because the alloying element contents play a key part of their properties. The graded composition is often realized by diffusion process of interstitial atoms like C and N, as they effect mechanical properties prominently in alloys.
Different studies are carried out on the fabrication of metallic architectured materials in Fe-C system because of numerous phases in the ferrous system. Chehab et al. [10] proposed a process for compositionally graded steel with carburization or decarburization technique to overcome contradiction of the elongation and the resistance of steels. This study also proposed a method to estimate the global necking strain of the architectured materials using Considère’s criterion.
Possible architectures are shown in Figure 1-7. In this study, for the architectures that do not cover the whole surface, i.e., Figure 1-7 (c)(d)(e)(h)(i), two kinds of masks were made to selectively carburize or decarburize austenite steel according to different architectures and different base steel.
Corrugation geometry
In this part, corrugation, the interested form in this study, will be discussed. Studies have been carried out on isolated corrugations and corrugated sandwich structures both in simulation works and experiments because of the special mechanical response when they are loaded. Corrugated forms are easy to fabricate, and corrugated sandwich panels have a smaller density compared to full-filled ones. Corrugated sandwich panel presented a higher tensile strength and bending stiffness along the longitude direction in comparison to simple monolithic panels of the same amount materials [31].
Extended variations are applicated on corrugation-filled sandwich panels to explore enhanced performances. Adding foam infill in the sandwich voids can reduce the deflection of the panel with the cost of a slight weight increasing [32]. Sandwich cores with graded thickness can also reduce the panel deflection and absorb more plastic energy [33]. Preliminary periodic corrugated embossment on plain steel sheets can modify the Young’s modulus and rigidity [34].
Results of these studies also show that geometric parameters of corrugations have an important influence on the mechanical behavior of corrugated sandwich panels. Therefore, corrugated form is very promising and worth further exploration.
For the development of metallic materials, especially for steels which are one of the most widely used structural materials, there is always a contradiction between strength and ductility.
In order to bring a solution to this problem, Fraser [6] studied systematically the behavior of isolated corrugation and corrugated reinforcement composite in Figure 1-10 by changing geometrical parameters (P, h, t in Fig.1-5 (a)) and also material properties (Young’s modulus, yield strength, work hardening coefficient and exponent).
This study carried out by Finite Element Method (FEM) simulation proves that different parameters of the corrugated core could influence the composite global tensile behavior on a system of 4130 steel reinforcement embedded in a copper matrix. Compared with straight reinforcement of the same volume fraction, corrugated ones can postpone the necking with the cost of a loss in ultimate tensile strength (UTS), as shown in Figure 1-11 (a). For corrugations with a larger amplitude, it is even possible to attain a more important necking stain than that of the copper matrix, which allows access to the blank region of the material property map. This delay came from the boosts of work hardening rate during tensile test shown in Figure 1-11 (b).
Laser treatment on metallic materials
Laser processing has been developed very fast since laser was in 1960s. It refers to the use of thermal effect between laser beam and material surface to complete the processing, such as laser cutting, laser drilling, laser welding, laser surface treatment, laser shock peening, etc. It can be applied on various, including metals, polymers, ceramics, or the combinations of them.
Laser processing has many advantages. It has a great flexibility thanks to adjustable energy density, spot size and moving speed. As there is no direct physical contact between laser and material surface, no wear takes place even with materials with high hardness. Its impact is limited within the treated zone, the heat affected zone (HAZ) is small. Besides, in the past ten years, laser processing has been combined with the fast developed additive manufacturing technology to become the trending LAM.
It can be used as an external heat source to modify the properties of metallic materials for its flexibility and rapidity. Compared to traditional heat treatment in furnace which results in a homogenous microstructure of the whole part, laser heat treatment can realize a localized heat treatment without affecting too much the remaining of the specimen thanks to its precision. By changing treatment parameters, such as input power, moving velocity, focus distance, treatment duration, etc., the thermal cycle can be changed, and different results can be achieved. This part is focused on laser processing on ferrous materials. Laser hardening is one of the most developed laser process and is a common surface strengthening method nowadays. Besides, laser treatment can bring other modifications on metallic materials, for example, laser annealing can reduce the YS and UTS but increase ductility [36], laser carburizing or nitriding can bring precipitation on the treated surface to improve its strength.
These kinds of treatments will be introduced in the following paragraphs. To better understand laser processing, it begins with a presentation of phase transformation and their features in Fe-C system, followed by different laser treatments on ferrous materials, including laser annealing, laser hardening and laser alloying. And laser processing on other metallicmaterials is also reviewed.
Introduction of phase transformation in Fe-C system
As steels are the most widely used metallic material, laser heat treatment on steels is also one of the most well-studied and extensively practiced subjects among laser applications. For its short interaction time and small penetration depth, most of the treatments aim to modify the surface property to adapt to its function, such as increase the wear resistance. The metallurgy of Fe-C system has much to exploit because microstructural and mechanical properties can be well controlled by the chemical component and heat treatment cycle. Each phase in the Fe-C phase diagram (Figure 1-13) has diverse features and their combination brings different effects to yield stress, work hardening rate and fracture strength of steels. Basic features of principal phases are summarized as follows.
Laser carburization
As introduced above, laser hardening is usually used as a surface strengthening method to increase surface layer hardness by forming martensite. This process can be further developed by inducing other alloying elements to reinforce hardening effect. This laser alloying process is usually about small interstitial atoms like carbon or nitrogen. It enables the introduction of carbon or nitrogen atoms into the metallic matrix in order to form solid solution or carbides/nitrides to increase the surface hardness and wear resistance [55]. It is extensively studied on ferrous materials and other alloy systems. This section will present principles and recent studies of laser carburization on steels.
Traditionally surface carburization is obtained by liquid salt bath, solid carbonaceous material, gas treatment in furnace or plasma technique, especially by gas or plasma which are well-researched and widely used in industry. At high temperature, carbon atoms are decomposed from alkanes or CO2 gas atmospheres and transferred into the matrix. The carbonrich layer is hardened by quenching. Figure 1-21 is an example of 0.19% carbon AISI 8620 steel carburized in air/propane atmosphere with constant CO2 for more than 10h at 925 °C [56].
A fully martensitic layer of nearly 1 mm is formed and the hardness is improved to 750 Hv compared to 500 Hv in the core. Conventional gas carburization can increase the carbon content to around 0.8%, and super-carburized layer containing 2-4% carbon can be realized by repeated carburizing treatment. At this stage, not only martensite is generated but also numerous carbides are precipitated and spheroidized, which improves the wear resistance still further [57].
Conclusions and the present work positioning
This literature review presents the concept of architectured materials and its development in recent years. It refers to heterogeneous materials where the geometric topology of each component is well designed to yield certain synergistic effect between them when it is loaded [1,2]. Some examples are given for better understanding, especially concerning metallic architectured materials and corrugated geometry. It is a popular trend to develop heterogeneous materials to achieve a trade-off between contradictory properties of homogeneous materials, like increasing strength and ductility at the same time [4]. Most existing metallic architectured materials are fabricated by ‘post assembly’ of different components. This usually results in a poor bonding and cohesion at the interface, which is not strong enough and leads to cracks and an early-stage facture. Therefore, the quality of interface and connectivity is the key point to the mechanical performance of architectured materials. Instead of ‘post assembly’, the interface decohesion problem can be resolved by making use of different existing phases out of one initial material. A well-arranged spatial architecture of different phases, such as ferrite and martensite, may have a synergistic effect.
Microstructural characterization and mechanical properties
Microstructural observations are realized on the section of as-received materials and after laser treatment for two studied steel sheets AISI 430 and annealed martensitic steel using both optical microscope (OM) and scanning electron microscope (SEM) equipped with energy dispersive X-ray spectrometry (EDS) and electron backscatter diffraction (EBSD) detector. The rolling direction and transverse direction are not specified on the as-received AISI 430 sheets and microstructural examination do not reveal evident difference along different directions. And annealed martensitic steel is fully heat treated above Ac3 before laser treatment, following tests show that there is not any obvious difference in neither microstructure nor mechanical properties in neither initial rolling direction and transverse direction. Therefore, in this work, rolling direction and transverse direction of the sheets will not be distinguished in characterization nor laser treatment.
Prior to the microstructural observations, the samples are prepared by grinding with silicon carbide (SiC) abrasive papers with grit sizes of 240, 400, 800, 1000, followed by cloth polishing using 9, 3, and 1 µm diamond suspension solutions, each step for 3 minutes. AISI 430 steel samples are finished by 0.04 µm colloidal silica suspension (OP-S suspension produced by Struers) for 5 minutes and are etched in Kalling’s No.1 solution (1.5 g CuCl2, 33 ml HCl, 33 ml ethanol, 33 ml distilled water) for a few seconds to reveal the microstructure. The etching solution outlines the grain boundaries and carbides, and colors martensite in dark brown in OM. Annealed martensitic steel samples are etched by 3% nital solution for around 10 seconds. Ferrite appears light, cementite is dark in OM, and martensite is etched. Figure 2-1 shows the microstructure of as-received AISI 430 section (a) by OM (b) by SEM. It presents a complete equiaxed ferritic microstructure with evenly distributed very fine chromium carbide precipitation. The average grain size is 11.3 μm.
Tensile test and DIC equipment
Fully and localized laser treatments are realized on steel sheets. According to predefined dimensions, tensile specimens are cut from steel sheets. Even though laser treatment introduces internal stress into steel sheets due to phase transformation and distort them, it is better to keep the cutting step after laser treatment. Because certain laser parameters heat steel sheets to very high temperature, even above fusion point, if this is applied on already cut specimen, the heat dissipation will be affected by specimen borders and localized overheat will take place. Laser treated zone risks to be less homogeneous according to geometries and this ununiformity will cause stress concentration and early fracture of specimens. That is the reason why tensile specimens are cut from the sheets are cut after laser treatment despite of some disadvantages like distortion or slight centering defects. Figure 2-6 shows several examples of single straight a corrugation reinforced architectured samples.
Glow Discharge Optical Emission Spectrometry analysis
In this study, the chemical composition measurement is carried out by Glow Discharge Optical Emission Spectrometry (GD-OES) analysis. GD-OES is a surface analytical technique, an atomic emission spectrometer system employing a non-thermal glow discharge source for atomic excitation [92]. In a glow discharge, cathodic sputtering is used to remove material layer by layer from the sample surface. The atoms removed migrate into the plasma where they are excited through collisions with electrons or metastable carrier gas atoms. The characteristic spectrum emitted by these excited atoms is measured by the optical spectrometer [93]. The distribution of elements is determined qualitatively by depth profiling using a Horiba JY 10000 RF spectrometer equipped with a 4 mm diameter anode. The power to the plasma is supplied by a radio-frequency generator at a frequency of 13.56 MHz. A moderate power of 40 W is used for all analyses. High-purity argon is employed as the discharge gas, at a constant pressure of 750 Pa. The following atomic emission lines (O: 130 nm, C: 156 nm, Fe: 372 nm, Cr: 425 nm, Ni: 225 nm, Mn: 403 nm, Si: 288 nm, P: 178 nm, S: 181 nm) are selected. Compared with EDS detector, which is not sensible enough for quantitative measurement of light elements like carbon or oxygen because of low dispersive energy, GD-OES analysis detects wavelength instead, so it allows precise measurement for all elements. The sputter erosion of glow discharge creates a crater of 4 mm diameter as shown in Figure 2-10. The sputtered atoms are deposited on the border of the crater. The crater shape with the chosen parameters is relatively clean: the crater bottom is sufficiently flat and, the bottom and the surface roughness similar. This method is advantageous because with progressive erosion, evolution of chemical composition along the depth can be detected layer by layer. In this study, two kinds of GD-OES analysis are interested. For AISI 430 direct laser hardening, the oxygen signal is especially followed to study the oxidation layer formed on the surface due to the treatment with and without protective shielding argon gas, which is only a qualitative measurement. On the contrary, exact carbon contents through the whole thickness of AISI 430 sheets for laser carburizing treatment are analyzed quantitatively.
Characterization of laser hardened AISI 430 sheet
Microstructure
Figure 3-6 (a)-(h) is the macrograph of the single-track laser-treated area at different powers with Ar shielding gas. In (a) and (e), there is a slight growth of the grain size on the surface layer with 10 kJ/m. In (b) and (c), the grains in the middle of the treated zone are almost equiaxed and homogenous through the thickness by symmetry laser spots. In (f) and (g), grains near the front surface, where the laser is applied, are columnar in the normal direction of the arc-shaped treated zone. In (d) and (h), columnar grains are produced along the thickness in the center line and parallel to the adjacent surface. Columnar grains are formed along the solidification direction in a melt pool because the high power (25 kJ/m) results in a total fusion in the treated zone. The columnar grains in (h) are more elongated than (d) due to the higher cooling rate. Grain size decreases progressively with the distance from the center, except for (a), where the laser powers on both sides are relatively low. No evident porosity is observed even for 25 kJ/m, where total fusion is achieved through the section.
Microhardness
The microhardness test shows that the base material presents a Vickers hardness of around 160 Hv. Figure 3-10 reveals the hardness along the middle line of the section with a singletrack (a) and multi-track (b) laser treatment at 100 W from the two sides in both air and Ar.
Track centers are marked by dashed lines. With a single-track laser, a maximum hardness of around 310 Hv is produced in the treated zone, which is 90% higher than the base material. The hardened zone in Ar is narrower than in air, and the maximum value obtained is also lower. The hardened zone in Ar presents a wave-like profile with the peak values almost unchanged in the middle of each track, although the hardened zone in air presents a hardness that repeatedly dropped to around 220 Hv, even for the laser track center.
For other conditions, except at 50 W from each side where the highest temperature in the adjacent layer of the sheets is below Ac1, microhardness also increases at the different levels. This softening with the multi-track laser treatment exists for samples with a high input power in both air and Ar, although in Ar the required power is higher (25 kJ/m)
Tensile test results of architectured materials and discussion
Figure 3-19 present experimental and numerical stress-strain behavior of specimens with one to four reinforcements introduced by localized laser hardening, (a) with straight reinforcements and (b) with corrugated reinforcements of h=1 mm. Both straight and corrugated reinforcements are efficient to increase the UTS compared to base material and the enhancement is more important with the increasing reinforced volume. But the loss in necking strain is larger with the increasing number of reinforcements. Unfortunately, there is no evident differences between straight and corrugation reinforced specimens with the same number of reinforcements, neither in experiments nor in FEM simulations. Unlike what is pointed out in the study of Fraser [6], corrugated reinforcements do not enhance the necking strain compared to straight ones.
Specimens reinforced with one to four corrugations of h=2 mm present similar trend, UTS is improved but the necking strain decreases with the increasing reinforced volume. But comparing straight reinforcements and corrugations, the necking strain is not improved, it is even worsened by corrugated reinforcements with h=2 mm.
Future improvements
The first reason of the negative effect of corrugated reinforcements is the large local strain of matrix. The strain between adjacent reinforcements is the most important, and sometimes it can be twice as much of the average strain when the corrugation height is larger. Therefore, the first trial in the following step is to utilize only one reinforcement. But as the available lasers have relatively small spot size (~ 1 mm) and the architecturation effect may be more visible with a larger volume reinforced, the gauge part width should be reduced. That is the reason why the tensile specimens are 5-mm-wide in the gauge length in the following part instead on 10 mm in this chapter.
The second reason of the architecturation effect is not satisfying is that the corrugations are not unbent enough to generate the boost in work hardening rate. The unbending process needs to be advanced. From Equation 3-9, it can be deduced that a smaller corrugation ratio H/P results in a smaller strain when the corrugation is fully unbent. In addition, another reason to reduce corrugation height H is that the robot arm which moves according to preprogrammed
trajectory utilizes an approximate linear velocity. When the architecturation geometries become more complicated and it needs to change much the moving direction, for example, when H=3 mm, the corrugation hardened by laser becomes less coherent. Also, as the reinforcement thickness should be adjusted to keep the same strengthened volume, the laser spot size needs to be more reduced with higher corrugations. This may lead to an untreated region in the middle of the sheet thickness and the reinforcement may not be enough strengthened. Thus, in the following step, the ratio H/P will be reduced, smaller H and larger P will be adopted. A more strengthened reinforcement is also expected. A higher yield stress of reinforcement can avoid the corrugation to be stretched and foster the unbending when it is under traction.
The boost in work hardening effect may be more evident and it will postpone the necking strain [112]. In the following step, a more effective hardening method needs to be investigated to introduce a more strengthened reinforcement in the ductile matrix in order to enhance the necking strain by corrugations.
Conclusions
In this chapter microstructural and mechanical property changes of thin AISI 430 sheets were studied here under different treatment conditions relating to the linear energy density, inert shielding gas, and treatment sides. Several conclusions emerge from microstructural and mechanical characterization.
(1) Despite the poor carbon content and grain coarsening, the laser heat treatment was shown to enhance the mechanical behaviors of 18% Cr FSS by the formation of martensite and chromium carbides during rapid cooling. Local microhardness can be increased by 90% at most using a single-track laser treatment. Fully treated samples by multi-track laser treatment are characterized for the first time, and the yield stress is increased by 60% and UTS by 45.8%.
(2) Thick oxide layers formed at high temperatures on the weld-side surfaces, especially above Tm, fragilize the laser-treated material and make the results less repeatable. Argon shielding gas is efficient to mitigate these disadvantages.
(3) Double-sided laser treatment enables a more homogeneous treatment through the thickness, hence allowing for an optimal trade-off between strengthening and loss of ductility.
(4) With the flexibility of the laser treatment trajectory and the small spot size, localized heat treatment can be achieved, thus enabling the architecturation of laser-based materials.
In the second part of this chapter, the laser parameter which leads to the most hardening effect was adopted to fabricate architectured materials with straight and corrugated reinforcements.
Different geometries and reinforcing volumes were carried out and the mechanical properties of architectured materials are evaluated. Some conclusions are issued from this study.
(1) Straight and corrugated reinforcements can increase YS and UTS of specimens compared to untreated material. Larger reinforced volume leads to higher reinforcing effect.
(2) While straight reinforced specimens present a homogeneous strain field during uniaxial tensile test, corrugation reinforced specimens have a heterogeneous strain field. A larger corrugation inclination ration H/P results in a more heterogeneous strain map.
(3) For a given reinforcement geometry, increasing the reinforced volume can increase the UTS
but reduce the ductility. But for the same reinforced volume, different geometries do not change much the stress-strain behavior. Corrugated reinforcements do not enhance the necking strain compared to straight ones in this material system.
(4) The lack of enhancement is due to the excessive local strain in the matrix and delayed corrugation unbending, the following step is to optimize the geometry. A more effectivehardening treatment may also benefit the enhancement in necking strain.
|
Table des matières
Introduction
Chapter 1 Literature review
1.1 Architectured materials
1.1.1 Introduction of architectured materials
1.1.2 Examples of architectured materials
1.1.3 Metallic architectured materials
1.1.4 Corrugation geometry
1.2 Laser treatment on metallic materials
1.2.1 Introduction of phase transformation in Fe-C system
1.2.2 Laser hardening
1.2.3 Laser carburization
1.2.4 Laser softening
1.2.5 Laser treatment on other metallic materials
1.3 Laser treatment applications on architectured materials
1.4 Conclusions and the present work positioning
Chapter 2 Materials and experimental procedure
2.1 Materials
2.1.1 Introduction and properties
2.1.2 Microstructural characterization and mechanical properties
2.2 Experimental procedure
2.2.1 Laser treatment platform and temperature measurement
2.2.2 Tensile test and DIC equipment
2.2.3 Finite Element Method simulation of uniaxial tensile test
2.2.4 Micro-hardness test
2.2.5 Microstructural observation and phase identification
2.2.6 X-ray diffraction analysis
2.2.7 Glow Discharge Optical Emission Spectrometry analysis
Chapter 3 Laser hardening of 430 ferritic stainless steel for enhanced mechanical properties and corrugation reinforced architectured materials
3.1 Introduction
3.2 Laser beam characterization
3.3 Temperature evolution
3.4 Characterization of laser hardened AISI 430 sheet
3.4.1 Microstructure
3.4.2 Microhardness
3.4.3 Surface state
3.4.4 Tensile test
3.4.5 Discussion of laser hardening effect on microstructural and mechanical properties
3.5 Corrugation reinforced architectured materials by localized laser hardening
3.5.1 Identification of matrix and reinforcement mechanical properties
3.5.2 Prediction of isolated straight and corrugated reinforcement
3.5.3 Tensile test results of architectured materials and discussion
3.5.4 Future improvements
3.6 Conclusions
Chapter 4 Laser carburization strengthening effect of AISI 430 ferritic stainless steel and corrugation reinforced architectured materials
4.1 Introduction
4.2 Temperature evolution
4.3 Characterization of laser assisted carburized AISI 430 FSS
4.3.1 Microstructural analysis and phase identification
4.3.2 Carburization efficiency
4.3.3 Microhardness
4.3.4 Tensile test
4.3.5 Fractography
4.3.6 Discussion
4.4 Corrugation reinforced materials by localized laser carburization
4.4.1 Identification of matrix and reinforcement mechanical properties
4.4.2 Prediction of isolated straight and corrugated reinforcement
4.4.3 Tensile test results of architectured materials and discussion
4.4.4 Future improvements
4.5 Conclusions
Chapter 5 Corrugation reinforced architectured materials by direct laser hardening on annealed carbon steel
5.1 Introduction
5.2 Results
5.2.1 Mechanical and microstructural characterization of the matrix and reinforcement
5.2.2 Identification of matrix and reinforcement mechanical properties for FEM simulation
5.2.3 Prediction of isolated straight and corrugated reinforcement mechanical properties
5.2.4 FEM simulation results of straight and corrugation reinforced specimens
5.2.5 Tensile test of architectured samples
5.2.6 Fractography
5.3 Discussion
5.3.1 Effect of corrugated reinforcement on work hardening
5.3.2 Perspective of the corrugation effect on necking strain
5.4 Conclusions
Conclusions and Perspectives
References
