Hardening increments in aluminum alloys [5].
1. Introduction
Due to its relative low density and high strength, the 2
This chapter provides information about mechanical behavior of welded joints of aluminum alloys in terms of properties, determined by tensile, indention and fatigue tests, as well as, the fatigue crack growth conditions in different zones of the welded joints.
1.1. Precipitation and mechanical properties of aluminum alloys
The precipitation hardening process requires that the second component in the aluminum alloy, is sufficiently soluble to allow extensive dissolution at an elevated temperature (solubilization treatment temperature) and that the solubility is considerably reduced at lower temperatures, such is the case of Al-Cu alloys (Figure 1) [1].
According to Figure 1, to induce precipitation hardening, the alloy is heated at a temperature higher than the solvus temperature to produce a homogeneous solid solution α, allowing dissolution of the second phase θ and eliminates the segregation of the alloy. Once, the solubilization temperature is reached, the alloy is cooling at high rate in order to limit the diffusion process of the atoms toward potential sites of nucleation. Finally, the supersaturated solid solution α
Considering the equilibrium forces between the line tension
As
The actual strength of the particle under this circumstance becomes irrelevant, as the bypassing operation becomes dependent only upon the interparticle spacing. If, however, the strength of the particle is such that the maximum resistance force is attained before sin θ=1, then particles will be sheared and the dislocation will pass through the particle (Figure 4).
Therefore, it follows that, for a given interparticle spacing (given volume fraction and particle size), hard particles will give the maximum precipitation hardening, and this condition defines the maximum degree of hardening attainable. Soft particles give a lesser degree of hardening. Consideration of the relationship between the applied stress and the dislocation bowing, following Orowan [2], leads to the Orowan equation:
where Δ
Ashby [3] further developed his equation to take into account the interparticle spacing, and the effects of statistically distributed particles. The Ashby-Orowan relationship is given as:
Application of the Taylor factor for polycrystalline materials, expressing the microstructural parameters in terms of the volume fraction and real diameter and converting shear stress to tensile stress, yields [4]:
where Δ
|
|||
|
|
|
|
2014 | 97 | 417 | 320 |
6061 | 55 | 276 | 221 |
7075 | 103 | 503 | 400 |
1.2. Welding in age hardened condition
The 2
Figure 6, shows the weld thermal cycles for gas metal arc welding (GMAW) process in a 6061-T6 (Al-Si-Mg) alloy at different preheat conditions and their correlation between the C precipitation curve. In this sense, Myhr et al. [9] studied the microstructural evolution during the cooling weld thermal cycle in Al-Si-Mg alloys (Figure 7). They determined that when the peak temperature
In addition to the microstructural transformation, after welding porosity and liquation cracking could also exist, which affect directly the mechanical behavior of the welded joints.
Porosity in welds of aluminum alloys is very complicated to control, because of the high hydrogen diffusion in liquid aluminum (Figure 8), the environmental interaction and the high rate of solidification.
When a fusion welding process in aluminum is performed, the hydrogen diffusion in melted metal could be as high as 1.00 cm3/g, this fact produces the formation of gas bubbles. If we consider the bubble formation in liquid metal as schematically represented in Figure 9, the bubble begins to ascend when the surface tension is overcome by the buoyant force, which tends to push the bubble to the free surface.
The critical radius
where κ is the detaching angle of the bubble, σ* the interfacial energy between liquid and gas (~ 1 Jm-2 for the majority of the gas-metal systems),
Liquation in welds of aluminum alloys can occur in the partially melted zone (PMZ). The PMZ is the region outside the fusion zone where grain boundary liquation occurs during welding. Figure 12 shows a portion of the PMZ in GMAW of 6061-T6 aluminum made with high silicon content filler metal (ER4043).
The liquation phenomenon occurs along grain boundary, although it can be presented in the grain interior. When liquation is presented, cracking can occur along grain boundary because of the tensile strains generated during welding. The weld metal composition is determined by base metal and filler metal compositions, as well as, the dilution ratio. The dilution ration is related with the amount of filler metal diluted with the base metal to form the weld metal. Metzger [12] observed liquation cracking in full penetration, gas tungsten arc welding (GTAW) on a 6061 alloy made with Al-Mg filler metal at high dilution ratios, but not in similar welds made with Al-Si filler metals at any dilution ratio. Huang et al. [13] have conducted studies on liquation cracking in the PMZ of full penetration welds of a 6061 alloy. They found that liquation cracking occurred in GMAW welds when an ER5356 (Al-Mg) filler metal is employed but not with an ER4043 (Al-Si).
2. Mechanical behavior of aluminum alloys welds
2.1. Indentation
The hardness of a material represents the plastic deformation resistance by indentation. The hardness number
For usual indentation, the Vickers hardness number
where, sometimes,
Concerning instrumented indentation tests (IIT), which allow to plot a load-depth curve, the calculation of a hardness number can use the maximum distance (maximum indentation dept
Classical indentation has been used to determine the hardness evolution in precipitation hardening aluminum alloys welds [8, 15, 16]. Ambriz et al. [8] determined Vickers microhardness profiles and mapping representation in welds of 6061-T6 aluminum alloy (Figure 13).
A significant difference for the hardness of weld metal, and HAZ with respect to base material was observed. This indicates that mechanical properties after welding will be different. It should be noted a soft zone formation in both sides of the welded joint, the hardness decrease in the soft zone is around 43% with respect to base material. This characteristic is the result of the thermodynamic instability of the β’’ needle-shaped precipitates (hard and fine precipitates) promoted by the high temperatures reached during the welding process. Indeed the temperatures reached during the welding process are favorable to transform the β’ phase, rod-shaped according to the transformation diagram for the 6061 alloy (Figure 6).
Considering the hardness heterogeneity of the welded joints, instrumented indentation tests (IIT) was performed in base metal, weld metal and HAZ (soft zone). Figure 14, shows the evolution of the applied load as a function of indentation depth for 6061-T6 and 7075-T651 aluminum alloys welds.
Moreover, from the instrumented indentation tests it is possible to calculate the elastic modulus which is deduced by the inverse of the unloading curve (1/
where
To calculate
wher
Additionally, the yield strength
where
Instrumented indentation test allows to determine the properties given in Table II. Yield strength values determined by instrumented indentation are similar to those obtained by micro-traction test (Figure 16). It means that it is possible to characterize a local zone in a welded material by instrumented indentation where it is not possible by global test (tensile test).
|
|
|
|
|
Base metal | 98.6 | 266 | 0.46 | 0.015 |
Weld metal | 78.0 | 133 | 0.48 | 0.022 |
HAZ | 87.4 | 106 | 0.46 | 0.016 |
On the other hand, the elastic modulus obtained by instrumented indentation does not correspond with those reported in tension or compression tests for aluminum alloys (68-72 GPa) and it is not possible to establish a clear difference between weld metal and HAZ. In fact in a recent study, Chicot et al. [19] established that the elastic modulus obtained by instrumented indentation corresponds to a bulk modulus. This was explained by the fact that for the case of indentation, the elastic and plastic strain is triaxial, whereas in tensile test, the strain used to determine the elastic modulus is uniaxial.
2.2. Tensile properties
As possible to deduce from indentation test (Figure 13), the tensile mechanical properties of aluminum alloys welds hardened by precipitation, are not homogeneous along the welded joint. Tensile properties in welds obtained by several welding processes of this alloys have been studied. For instance, V. Malin [15] studied the relation between weld thermal cycles and the microstructural transformation with tensile properties of a 6061-T6 alloy welded by GMAW. In their research, tensile tests samples were taken from the welded joint as shown in Figure 15. It is to say, a global structure effect was considered during tensile test.
The failure zone after tensile test was localized in the HAZ (soft zone) where the hardness of the welded joint is minimal, this result is in agreement with Ambriz et al. [20].
Considering the hardness profile evolution, the true stress-strain curves in weld metal and HAZ (soft zone) of a 6061-T6 alloy welds were determined by means of micro-traction test [18]. The individual behavior is presented in Figure 16, as well as, its respective comparison with base metal.
The HAZ presents a reduction of the tensile strength with respect to base metal and weld metal of around 41 and 19%, respectively. This aspect was related to the over-aging phenomenon and it is explained in terms of the microstructural transformation (Figure 5), and precipitation sequence. Although, weld metal shows higher tensile strength than HAZ, a lower ductility is observed for weld metal. This characteristic was attribute to the porosity formation during the solidification and the high silicon content of the filler metal (ER4043) which, when mixed with the melted base metal, leads to a microstructure of eutectic silicon, which is a brittle phase that adversely affects the tensile mechanical properties of the welded joint. A summary of tensile mechanical properties for 6061-T6 and 7075-T651 aluminum alloys welds are presented in Table III.
|
|
|
|
|
|
|
6061-T6 | 68 | 279 | 310 | 14.0 | 408 | 0.08 |
6061-T6 (weld metal, ER4043) | 68 | 151 | 226 | 4.00 | 464 | 0.20 |
6061-T6 (HAZ, soft zone) | 68 | 120 | 183 | 13.0 | 300 | 0.16 |
7075-T651 | 72 | 530 | 568 | 8.00 | 794 | 0.08 |
7075-T651 (weld joint, ER5356) | 68 | 165 | 260 | 2.80 | 677 | 0.25 |
2.3. Fatigue of aluminum alloys welds
Fatigue or fatigue damage is the consecutive modification of the materials properties with respect to the application of a cyclic stress, which can conduct to the fracture. Under uniaxial cyclic loading conditions it is possible to distinguish a stress ratio
As a function of
Some practical applications involve cyclic loading at a constant amplitude, but irregular loads as a function of time are commonly encountered. In this case, we will discuss some results in terms of a constant amplitude loading. The simplest fatigue test consists of subjecting a specimen to a cycling loading (different levels of stress amplitude
In Figure 18, it is possible to identify different domains: (i) Low cycle fatigue. In this case a high stress level is applied on the sample (normally over the yield strength of the material). Because of the high deformation during the test, the number of cycles to failure tends to be lower (102 to 104). (ii) High cycle fatigue. This is related with an elastic behavior on a macro scale of the sample, i.e. the stress level is not higher than the yield strength of the material. The failure is expected for a large number of cycles, for instance, more than 105. In fact the boundary between low and high cycle fatigue is not well defined by a specific number of cycles. The most important difference is that low cyclic fatigue is associated with macro-plastic deformation on each loading cycle. When the stress level in high cycle fatigue is applied, a fatigue limit or endurance limited is presented, which is represented by an asymptote in the Wöhler curve. In some metallic materials it could be obtained when the number of cycles is in the order of 106 to 107. (iii) Fatigue gigacycle. This domain corresponding with a very high number of cycles and it has been observed that fatigue limit tends to decrease when the number of cycles increases.
It is well known that fatigue damage is a surface phenomenon as indicated by Forsyth [22], who determined the presence of reliefs linked to the formation of localization deformation bands named persistent bands. The surface topography is traduced by the formation of intrusions and extrusions as shown schematically in Figure 19.
For a uniaxial tensile test, these bands resulting in the formation of micro-cracks (state I in Figure 19), which are orientated at 45 degrees with respect the traction axe. Only certain grains are affected by the formation of those bands. The persistent bands orientation and the formation of cracks on the state I, are important in the case of uniaxial and multiaxial loading. Brown and Miller [23, 24] introduced a useful notation in multiaxal loading for facets A and B, which are schematized in Figure 20.
The type B facets provides a shearing vector which enters into the material, and they are more dangerous than type A facets, from which the shearing vector is tangent to the free surface of the sample. The intrusions and extrusions formation associated to the slip persistent bands, as well as, the micro-propagation of cracks in the state I are of interest at a distance of the grain size (small fatigue cracks). Thus, considering that micro-cracks are related with the crystallography aspect, once the crack encounters the first grain boundary it begins to bifurcate according to state II and the propagation at a perpendicular direction of the principal stress is obtained.
Additionally to the fatigue damage mechanism mentioned previously, in the case of welding, the stress concentration factor due to the geometry of the welding bead has a special importance. In this sense, Ambriz et al. [20] has been quantified the effect of the welding profile generated by modified indirect electric arc (MIEA) technique on the fatigue life of a 6061-T6 aluminum alloy. In order to determine the stress concentration factor
Subsequently, uniaxial fatigue test a cyclic loading with a sinusoidal wave form at a frequency of 35 Hz and load ration
Figure 22 shows the maximum stress
where
Regarding the geometry of the welding profile, comparison of the fatigue performance exhibited by MIEA welds with the results reported in the literature [25] for a single V joint configuration for the same aluminum alloy shows a significant improvement in fatigue life for the MIEA welded samples. The
Considering that, if the loading condition promotes the formation of a principal crack (Figure 19), it will grow according to the power law region as shown schematically in Figure 24.
The fatigue crack growth rate d
where
Considering the stable crack growth propagation region shown in Figure 24, the experimental results of
where
|
|
|
|
6061-T6 (rolling) | 5.00×10-7 | 2.426 | 0.89 |
6061-T6 (transverse) | 3.97×10-8 | 3.320 | 0.97 |
Weld metal | 2.63×10-14 | 8.550 | 0.92 |
HAZ | 3.77×10-9 | 6.650 | 0.89 |
Figure 25, shows the fatigue crack growth for base metal (6061-T6) in the rolling and transverse to rolling direction.
This graph shows that the microstructural characteristics (anisotropy) does not have an important influence in terms of fatigue crack growth as could be expected, taking into account that yield strength in rolling direction is higher than transverse direction. However, this is not the case for weld metal and HAZ (Figures 25b and c), in which the crack tends to propagate faster than that in base metal. In the case of weld metal (Figure 25 b), the faster crack growth rate in comparison with base metal is related to the low toughness due to the high silicon (~ 5.5 wt. percent) content provided by the filler metal during welding. Similarly, for the HAZ it is possible to observe that the crack growth is faster than base metal, aspect which is attribute to the microstructural transformation of fine needle shape precipitates β’’ into coarse bar shape precipitates β’ produced by the thermal effect.
3. Conclusion
Mechanical behavior in welds of precipitation hardened aluminum alloys are still under development and the softening phenomena in the heat affected zone should be better understood. Valuable information could be obtained by the precise understanding of the weld thermal cycles in conjunction with the C transformation curve and its microstructural effect in mechanical properties. In this sense, our research group is conducting experiments to control the weld thermal cycle by means of localized chillers and heaters in the fusion zone and heat affected zone to observe the mechanical properties evolution of the welded joints.
References
- 1.
Gladman T. Precipitation Hardening in Metals. Materials Science and Technology 1999;15 30-36. - 2.
Orowan E. Internal Stress in Metals and Alloys; 1948. - 3.
Ashby MF. Oxide Dispersion Strengthening. Gordon and Breach; 1958. - 4.
Gladman T. The Physical Metallurgy of Microalloyed Steels. The Institute of Materials; 1997. - 5.
ASM. Properties and Selection: Non Ferrous Alloys and Special Purpose Materials. ASM International; 1992. - 6.
Askeland D., Fulay P and Wright W. The Science and Engineering of Materials; 2010. - 7.
Edwards G., Stiller K., Dunlop GL. and Couper MJ. The precipitation sequence in Al-Mg-Si alloys. Acta Materialia 1998;46 (11) 3893-3904. - 8.
Ambriz RR., Barrera G., García R. and López VH. Effect of the Weld Thermal Cycles of the Modified Indirect Electric Arc on the Mechanical Properties of the AA6061-T6 Alloy. Welding International 2010;24 (4) 42-51. - 9.
Myhr OR., Grong O., Fjaer HG. and Marioara CD. Modelling of the Microstructure and Strenght Evolution in Al-Mg-Si Alloys During Multistage Thermal Processing. Acta Materialia 2004;52 4997-5008. - 10.
Matters G. The Welding of Aluminum and its Alloys. CRC; 2002. - 11.
Grong O. Metallurgical Modelling of Welding. The Institute of Materials; 1997. - 12.
Metzger GE. Some Mechanical Properties of Welds in 6061 Aluminum Alloy Sheet. Welding Journal 1967;46 (10) 457-469. - 13.
Huang C. and Kou S. Liquation Cracking in Full Penetration Al-Mg-Si Welds. Welding Journal 2004;4 111-122. - 14.
Oliver WC. and Pharr GM. An Improved Technique for Determining Hardness and Elastic Modulus Using Load and Displacement Sensing Indentation Experiments. Journal of Materials Research 1992;7 (6) 1564-1583. - 15.
Malin V. Study of Metallurgical Phenomena in the HAZ of 6061-T6 Aluminum Welded Joints. Welding Journal 1995;9 305-318. - 16.
Liu G., Murr LE., Niou CS., McClure JC. and Vega FR. Microstructural Aspects of the Friction Stir Welding of 6061-T6 Aluminum. Scripta Materialia 1997;37 (3) 355-361. - 17.
Hay JC., Bolshakov A. and Pharr GM. Critical Examination of the Fundamental Relations Used in the Analysis of Nano-Indentation Data. Journal of Materials Research 1999;14 (6) 2296-2305. - 18.
Ambriz RR., Chicot D., Benseddiq N., Mesmacque G. and de la Torre SD. Local Mechanical Properties of the 6061-T6 Aluminum Weld Using Micro-Traction and Instrumented Indentation. European Journal of Mechanics A/Solids 2011;30 307-315. - 19.
Chicot D., Roudet F., Zaoui A., Louis G. and Lepingle V. Influence of Visco-Elastic-Plastic Properties of Magnetite on the Elastic Modulus: Multicyclic Indentation and Theoretical Studies. Materials Chemistry and Physics 2010;119 75-81. - 20.
Ambriz RR., Mesmacque G., Ruiz A., Amrouche A. and López VH. Effect of the Welding Profile Generated by the Modified Indirect Electric Arc Technique on the Fatigue Behavior of 6061-T6 Aluminum Alloy. Materials Science and Engineering A 2010;527 2057-2064. - 21.
Pineau A. and Bathias C. Fatigue des Matériaux et des Structures 1. Lavoisier; 2008. - 22.
Forsyth PJE. Some Metallographic Observations on the Fatigue of Metals. Journal of the Institute of Metals 1951;80 181. - 23.
Brown MW. and Miller KJ. A Theory for Fatigue Failure Under Multiaxial Stress-Strain Conditions. Proceedings of the Institution of Mechanical Engineerings 1973;187 745-755. - 24.
Miller KJ. Metal Fatigue-Past, Current and Future. Proceedings of the Institution of Mechanical Engineerings 1991;205 1-14. - 25.
ASM. Fatigue and Fracture. ASM Iternational; 1996. - 26.
Moreira PMGP., de Jesus AMP., Ribeiro AS. and de Castro PMST. Fatigue Crack Growth in Friction Stir Welds of 6082-T6 and 6061-T6 Aluminium Alloys: A Comparison. Theoretical and Applied Fracture Mechanics 2008;50 81-91. - 27.
Ambriz RR., Mesmacque G., Ruiz A., Amrouche A., López VH. and Benseddiq N. Fatigue Crack Growth Under a Constant Amplitud Loading of Al-6061-T6 Welds Obtained by Modified Indirect Electric Arc Technique. Science and Technology of Welding and Joining 2010;15 (6) 514-521.