2021 Volume 62 Issue 2 Pages 278-283
Thin-walled metal tubes, because of their low cost, high stiffness and strength combined with a relatively low density, have been extensively applied in transportation vehicle industry as energy absorbers. The traditional methods applied to the metal tube fabrication, however, are time-consuming and materially-inefficient processes. In the present study, taken AlSi10Mg alloy as the model material, the hexagon thin-walled tubes were prepared by Selective laser melting (SLM). The effect of post heat treatment on the energy absorption characteristics of hexagon thin-walled tubes is systematically studied. The results revealed that the subsequent heat treatment significantly improved the energy absorption properties of the hexagonal thin-walled tube, especially the solution heat treatment. Moreover, the energy absorption properties of the AlSi10Mg hexagonal tube obtained in the present work is higher than most of recently reported thin-walled AlSi10Mg tubes fabricated by SLM.
Fig. 4 (a) Quasi-static compressive force-displacement curves of triangular, square and hexagon thin-walled tubes simulated by the finite element simulation; (b) Comparison of quasi-static compression experiment and simulation results of hexagonal thin-walled tubes without post heat treatment and under solution treatment.
An astonishing increase in traffic accidents has been witnessed accompanying the rapid development of the transportation vehicle industry.1) According to Global Status Report 2015, more than 1.2 million people die annually in traffic accidents worldwide, and 50 million suffer non-fatal injuries. Moreover, it is estimated that the traffic injuries could be the seventh leading causes of deaths by the year 2030, if no protective measures are taken. Apart from severe injuries to human beings, traffic accidents also lead to catastrophic damages to the structures of the vehicles and surrounding environment. A most effective approach to minimize the loss of life and property in traffic accidents is to install energy absorption structures to the vehicles, which are designed to effectively and stably dissipate most kinetic impact energies to maximize passenger safety when accidental collisions occur.2,3) Thus, numerous kinds of structures with excellent energy-absorbing capacity have been developed and investigated in the past decades, including ring and ring systems, thin-walled tubes, honeycombs, cellular materials, etc.4–7) Among these energy-absorbing structures, thin-walled metal tubes are extensively applied, because of their low cost, high stiffness and strength combined with a relatively low density.8,9) During impact events, thin-walled metal tubes can effectively absorb the kinetic energy of the structures by performing plastic deformation. As energy absorbers, the thin-walled metal tubes can also be used in many different ways, such as lateral compression, axial crushing, splitting and curling, expanding and so on.5,10)
As a unique thin-walled structure, thin-walled metal tube is generally with a thickness of 0.2 to 4.0 mm. The traditional manufacturing method for the thin-walled metal tube is to produce a stamping and then machine it to the required thickness. However, the machining process is a time-consuming and materially-inefficient process. Therefore, new methods to manufacture thin-walled metal tube with smaller thicknesses and higher performances have become a new research direction. Laser additive manufacturing (LAM) technology, capable of producing any precise metal products, offers a valuable opportunity for fabricating the thin-walled metal tubes by controlling the widths of the deposited tracks and their hatch spacing.11–14) For instance, the ultra-thin plate can be prepared by the LAM when the deposited tracks are only superposed along the building direction. Thus, the LAM technology can dramatically simplify the production process of thin-walled metal tubes compared with the traditional manufacturing methods.
To search for optimum designs for the thin-walled metal tubes, the structures of the tubes are usually designed and optimized by means of numerical simulations, with the aid of parallel computing using commercial finite element software such as LS-DYNA, PAM-CRASH, ABAQUS, and RADIOSS.15,16) However, due to the complexity of the non-equilibrium LAM process, such problems as residual stresses, directional grain growth, micro-segregation, and the creation of non-equilibrium phases, the actual performance of the fabricated structures by the LAM is always lower than the performance predicted by the numerical simulations.17,18) One of the best ways to achieve LAMed structures with desired mechanical properties is the post heat treatment, which can efficiently release the residual stress by reducing the yield strength and creep of the LAMed structures to improve the structural stability.19,20) Moreover, the heat treatment can also tailor the microstructures of the LAMed samples, such as particle size, chemical composition, phases, crystallite size, morphology, and chemical disorder, to optimize the performances of the LAMed structures.
In the present study, the effect of post heat treatment on the energy absorption characteristics of hexagon thin-walled AlSi10Mg tubes prepared by SLM is systematically studied. The results revealed that the subsequent heat treatment significantly improved the energy absorption properties of the hexagonal thin-walled tube, especially the solution heat treatment. Moreover, the energy absorption properties of the AlSi10Mg hexagonal tube obtained in the present work is higher than most of recently reported thin-walled AlSi10Mg tubes fabricated by SLM.
The AlSi10Mg alloy samples were fabricated in an SLM machine, which utilized the ytterbium fiber laser with a wavelength of 1070 nm. The diameter of the laser beam spot was 0.1 mm. SLM experiments were conducted inside a working chamber, which is filled with argon gas to prevent oxidization of deposited samples. The diameters of the pre-alloy AlSi10Mg powders were normally distributed between 20 and 45 µm. Owing to the rapid solidication conditions, the majority of the particles produced by gas atomization in an argon atmosphere are spherical or near-spherical in shape. Substrates were pre-heated to 100°C before SLM experiments to reduce the thermal stress. Before SLM process, the substrate surface was carefully polished with grit papers so as to clear away the oxide thin films and then thoroughly cleaned in ethanol. The laser power was 300 W and the scan speed was 750 mm/s. The layer thickness in z direction during multi-layer deposition was set as 30 µm. After each layer was finished, the build platform was lowered by a distance equal to the layer thickness. A schematic of the SLM process is shown in Fig. 1. The 3D model used in the present study is obtained by CAD and Solidworks, and the model is converted into STL format file. The STL file is converted into G code. The control structure of 3D printing machine controls the laser path to scan and melt the powder spreading area, and then continue to spreading powder and scanning is repeated till to complete the printing of the entire structure. Final sample is obtained after post-processing.
Schematic of the SLM process.
The microstructural features of as-fabricated AlSi10Mg specimens were characterized by scanning electron microscopy (SEM). The samples were polished following the standard polishing procedures before SEM observation. To evaluate the mechanical properties at room temperature, the tensile specimens were machined from as-fabricated samples with the tensile axis parallel to the laser scanning direction. The gauge length of the specimens was 10 mm and the width was 2 mm. The average tensile strength was measured from 3 specimens by using a universal testing machine at a strain rate of 5 × 10−4 s−1.
2.2 SimulationsThe finite element model for the collapse simulation of tubes was numerically analyzed using the nonlinear code PAM-CRASH, which is widely used in multiple fields such as aerospace, ship, and railway industry. It is of excellent performance in explicit analysis of various physical processes with complex nonlinear characteristics of material, geometry and boundary. The height of tube studied in this work is 75 mm and the thickness is 1 mm. The Belytschko-Tsay four-node shell element with five integral points along the thickness of element was used to modeling the tubes. The finite element mesh size is selected as 2 mm for a balance of solution efficiency and accuracy. The material used for tube is AlSi10Mg with solution treatment and annealing. The 105# constitutive model, elastic-plastic-shell in PAM-CRASH, is adopted to describe the material behavior of tubes subjected to axial compression. The strain-rate effect is not considered in constitutive model for the insensitivity of aluminum alloy. The top plate is set to compress the tube axially at a constant velocity of 2 m/s, and the axial crushing displacement is set at 40 mm, while the displacement and rotation of bottom plate is restrained. The 36# contact model of SELF-IMPACTING NODE-TO-SEGMENT WITN EDGE TREATMENT was utilized to simulate the contact behavior of tube during structural collapse, and that between two plates and the thin-wall tube was prescribed using 33# contact model of SYMETRIC NODE-TO-SEGMENT WITN EDGE TREATMENT. The contact thickness is 0.5 mm, and the constant friction coefficient is set to 0.15.
An absorbing structure is generally expected to absorb more energy with higher energy absorption efficiency as possible in a controllable crushing pattern. There are multiple indicators proposed to evaluate the crash worthiness of energy absorbing structure, for instance, energy absorption (EA), specific energy absorption (SEA), peak crash force (PCF), mean crash force (MCF), and crash force efficiency (CFE). The peak force should be reduced as possible to prevent the passengers from overloading impact. Some important indicators are given in eqs. (1)–(4).
The energy absorption can be mathematically formed as,
| \begin{equation} \mathit{EA} = \int_{0}^{\delta}F(x)dx \end{equation} | (1) |
| \begin{equation} \mathit{SEA} = \mathit{EA}/M \end{equation} | (2) |
| \begin{equation} \mathit{MCF} = \frac{\mathit{EA}}{\delta} \end{equation} | (3) |
| \begin{equation} \mathit{CFE} = \frac{\mathit{MCF}}{\mathit{PCF}} \times 100\% \end{equation} | (4) |
Tensile test, which is the most direct way to evaluate the impact of microstructures and flaws on the mechanical properties of deposited samples, is carried out for the AlSi10Mg fabricated by the LAM. Figure 2 displays the room-temperature tensile engineering stress-strain curves of the as-fabricated AlSi10Mg alloy (labeled as SLM). Clearly, the samples fabricated by the SLM exhibits a relatively high tensile and yield strengths of about 300 MPa and 220 MPa, respectively, but shows a low ductility of about 4%. Such poor plasticity restricts their applications as excellent energy absorption structures. To improve the plasticity of the as-deposited AlSi10Mg alloy, the fabricated samples are further processed by the post heat treatment to tailor their microstructures and mechanical properties. As shown in Fig. 2, the heat treatment has a dramatic influence on the mechanical properties of the as-deposited samples. As the sample is annealed at 300°C for 2 h (labeled as SLM+Annealing), there is an obviously decrease in the tensile and yield strengths, while a large improvement in the ductility (about 6%). As the sample is solution heat treated at 530°C for 2 h (labeled as SLM+Solution), there is a slight decrease in the tensile and yield strengths, but the ductility dramatically increases to about 11%. Figure 3 depicts the SEM micrographs of the microstructure of the as-built and heat-treated SLM AlSi10Mg samples. For the as-built SLM AlSi10Mg sample, as shown in Fig. 3(a), the grey features are primary α-Al matrix decorated with white fibrous Si network. The fine dispersion of fibrous Si networks in the Al matrix has a positive effect on the mechanical properties of the as-built SLM AlSi10Mg samples. The effect of post heat treatment on the microstructures of the as-built samples is shown in Figs. 3(b) and (c). Clearly, the microstructure becomes coarser after the as-built sample annealed at 300°C for 2 h, as shown in Fig. 3(b). As the sample is solution heat treated at 530°C for 2 h, the microstructure is further coarsened as illustrated in Fig. 3(c). Therefore, the solid solution treatment at 530°C for 2 h is chosen as the best post-treatment process for the SLM deposited AlSi10Mg alloy.
Room-temperature tensile engineering stress-strain curves of the AlSi10Mg alloys fabricated by SLM (labled as SLM), under a post annealed (450°C for 2 h, labled as SLM+Annealing), and a solution heat treated (450°C for 2 h, labled as SLM+Solution) treatment.
Micrographs of the as-built and heat-treated SLM AlSi10Mg specimen microstructures. (a) as-built; (b) annealed at 300°C for 2 h; (c) solution heat treated at 530°C for 2 h.
In addition to the mechanical properties of the material itself, the configuration of the energy-absorbing structure also has a great influence on the energy absorption characteristics of thin-walled tube. Therefore, it is necessary to determine the structural configuration for the deposited specimen. To screen the best thin-walled tube configuration, the compressive processes of three structures, with the cross-sectional shapes of triangle, quadrilateral and hexagon, were simulated by the finite element simulation to predict the deformation behavior. As shown in Fig. 4, the crushing force of hexagonal tubes is significantly higher than that of quadrilateral and triangular tubes, especially after the crush displacement is 15 mm. While in the early stage of crushing, the comparative advantage is not obvious, mainly because the complete fold has not been formed in the initial stage of crushing. The energy absorption index for three kinds of thin-walled tube are listed in Table 1, where three tubes with same mass were considered, the EA of three tubes is 202.8 kJ, 261.3 kJ and 315.8 kJ, respectively. The density of the as-built specimens is 2.56 g/mm3, measured by Archimedes method. Corresponding SEA calculated by eq. (2) is 10.61 kJ/kg, 13.67 kJ/kg and 16.52 kJ/kg, respectively. The SEA of hexagonal tubes is higher than that of triangular and square tubes by 55.7% and 20.8%, respectively. The peak crush force of three tubes is close to 18.5 kN, the MCF and CFE of hexagon tube is higher than that of triangle and square tube. It is indicated that hexagonal tube is advantaged on energy absorption resulting in selection as the optimal shape for tube of the experimental validation.
(a) Quasi-static compressive force-displacement curves of triangular, square and hexagon thin-walled tubes simulated by the finite element simulation; (b) Comparison of quasi-static compression experiment and simulation results of hexagonal thin-walled tubes without post heat treatment and under solution treatment.
After the geometrical section of tube was determined, two hexagonal tubal specimens were manufactured using the SLM additive manufacturing technology. One of them was prepared without any treatment and the other with solution treatment. An actual hexagonal thin-walled tube, fabricated by the SLM, is shown in Fig. 5(a). Then the quasi-static compression experiments of the hexagonal thin-walled tubes were carried out. Figures 5(b) and (c) show the whole crushing processes of the hexagonal thin-walled tubes after annealed and solution treatments. The corresponding compression force-displacement curves of these two processes are presented in Fig. 4(b). For comparison, the simulation results of the solution-treated structure are also shown in this figure. Clearly, the initial peak force of the untreated tube reaches 28.5 kN at a crushing stroke of 2 mm. For solution-treated tube, the initial peak force is decreased significantly to half of that of the untreated tube. In addition, the deformation displacement of tube is dramatically increased to 38 mm after solution treatment, which is obviously higher than that of the untreated one (5 mm). As listed in Table 2, the EA of three tubes is 93.0 kJ, 232.5 kJ and 315.8 kJ, respectively. Corresponding SEA calculated by eq. (2) is 4.87 kJ/kg, 12.16 kJ/kg and 16.52 kJ/kg, respectively. With the solution treatment, the SEA, EA of tube is enhanced by about 148.8%. The simulational and experimental value of PCF is in good agreement with a relative error of 5.7%, while that of SEA, MCF, and EA is about 35.8%, which is mainly by the structural asymmetry of experimental specimen.
(a) An actual hexagonal thin-walled tube fabricated by the SLM additive manufacturing technology, (b) Snapshots of the experimental deformation process for thin-walled tube without post heat treatment. (c) Snapshots of the experimental deformation process for thin-walled tube under post solution heat treatment. (d) Snapshots of the simulational deformation process for thin-walled tube under post solution heat treatment.
The deformation process of the thin-walled tubes without heat treatment and under solution treatment are respectively shown in Figs. 5(b) and (c). Apparently, for the thin-walled tube without heat treatment, obvious cracks appeared in the plastic folding area when the compression deformation reached 5 mm. While for the tube under solution treatment, a stable non-extensive mode of deformation is generated during the crushing process and no obvious cracks are found until it’s compressed to 38 mm. It can be seen from Fig. 5(c) that the thin-walled tube has a horizontal fold bending phenomenon at the compression displacement of 21 mm, coupled with inward deformation during the folding process. Whereas this folding hinge dose not generate in the tube without heat treatment (Fig. 5(b)). This dramatic enhancement in the deformation displacement is mainly due to the redistribution of material microstructure after the solution treatment, which brings the improvement of material plasticity. By comparing the experimental and simulational curves of the solution-treated structure, it can be found that the average crushing force by the simulational result is in good agreement with experimental result. The peak crush force obtained by the experiment is 17.5 kN, which is also close to the simulational result of 18 kN. However, the force-displacement curve of the tube with solution treatment is relatively smooth, while there presents fluctuation (wave crest) characteristics on the simulational curve. Actually, every wave crest of force-displacement curve corresponds to a beginning of the generation of folding hinge, and the subsequent load decline is related to the collapse of the folding hinge. As the structure defect of actual specimen is difficult to be considered in the ideal numerical model, which has a big impact on the generation of folding hinge. This can be confirmed by comparing the snapshots of the deformation processes for the experiments and simulations shown in Figs. 5(c) and (d). For the ideal simulation model, the thin-walled tube specimen produced a symmetrically extensional deformation mode during the crushing process. However, the SLM fabricated tube is not absolutely axisymmetric, which actually introduces certain structural defect, resulting in the formation of non-extensible deformation modes. Nevertheless, the present thin-walled tube manufactured with SLM and under solution heat treatment greatly simplify the fabrication process. More importantly, the SEA of AlSi10Mg hexagonal tube obtained in the present work is 12.16 kJ/kg, which is higher than that of the most of circular AlSi10Mg tubes21) and sinusoidally corrugated AlSi10Mg tubes22) fabricated with SLM.
The effect of post heat treatment on the energy absorption characteristics of hexagon thin-walled AlSi10Mg tubes prepared by SLM is systematically studied. It is found that the subsequent heat treatment, especially the solution heat treatment (530°C for 2 h), significantly improved the energy absorption properties of the hexagonal thin-walled tube, and could increase its SEA from the untreated 4.87 kJ/kg to solution-treated 12.16 kJ/kg. Moreover, the SEA of AlSi10Mg hexagonal tube obtained in the present work is higher than most of recently reported thin-walled AlSi10Mg tubes fabricated by SLM.
The research presented in this work was supported by National Key R&D Program of China (No. 2016YFB1200504), Liaoning Provincial Innovation Team Program for Higher Education (LT2016010), Dalian Fund Plan of Science and Technology Innovation (2019J11CY017).
