Options
Process development and digitalization of laser-directed energy deposition processing high-strength 7XXX series aluminum alloy
Citation Link: https://doi.org/10.15480/882.13078
Publikationstyp
Doctoral Thesis
Date Issued
2024
Sprache
English
Author(s)
Advisor
Referee
Title Granting Institution
Technische Universität Hamburg
Place of Title Granting Institution
Hamburg
Examination Date
2024-06-11
Institute
TORE-DOI
Citation
Technische Universität Hamburg (2024)
Additive manufacturing has gained considerable attention from both industry and academia in recent years, primarily due to its numerous advantages over conventional subtractive manufacturing technologies. The layer-by-layer deposition process in additive manufacturing offers substantial design freedom, which proves advantageous for weight reduction, improving manufacturability for structures with intricate geometries, and cost-saving. In fusion-based additive manufacturing, materials undergo a process of melting and solidification. The energy sources used for melting materials include laser, arc, and electron beam. Laser-based additive manufacturing has gained widespread popularity due to its ability to operate without a vacuum atmosphere, unlike electron-beam melting, and its superior geometrical accuracy compared with arc-based additive manufacturing.
Lightweight materials and designs are in high demand in the automobile and aerospace industries because of their potential to reduce carbon emissions and minimize environmental impacts. Aluminum alloys with a high density-strength ratio are commonly utilized as lightweight materials. However, the range of applicable Al alloys in laser-based additive manufacturing is limited due to various processing challenges. These challenges include issues such as oxidation, high reflectivity, porosity formation, and cracking susceptibility. Therefore, it is crucial to address these challenges and establish an optimal process window for Al alloys in laser-based additive manufacturing. Al-Zn-Mg-Cu alloys, also known as 7XXX series alloys, are precipitation-strengthening alloys that demonstrate superior strength characteristics when subjected to appropriate ageing treatment. The rapid heating and cooling involved in laser-based additive manufacturing can be advantageous for processing 7XXX series alloys. The non-equilibrium solidification resulting from the rapid cooling promotes grain refinement and increases the solubility of solutes in the α-Al matrix, thereby enhancing the precipitation-strengthening effects during ageing treatment. Consequently, it is valuable to explore the processability of 7XXX series alloys in laser-based additive manufacturing.
The primary objective of this thesis is to develop an appropriate process window for laser-directed energy deposition (L-DED) in order to process high-strength 7XXX series alloys. Furthermore, the aim is to investigate the relationship between process-microstructure-property. Additionally, considerable attention has been given to understanding the mechanism behind defect formation, specifically porosity and cracks.
7XXX series alloys are processed using wire- and powder-based L-DED, respectively. When it comes to structural integrity, specifically in terms of porosity levels and cracking, wire-based feedstock demonstrates superior processability compared with powder-based feedstock. The high porosity levels observed in structures produced by powder-based L-DED are attributed to two main factors: the presence of intrinsic pores inside powder materials and the evaporation of volatile alloying elements during processing. The installation of a self-designed shielding gas nozzle in wire-based L-DED has been found to greatly improve surface roughness, reduce porosity levels, and enhance processing stability. A delayed hot cracking is noticed in wire-based L-DED processing high-strength Al alloys. The delayed hot cracking is correlated to the deposition length of structures and the number of deposited layers. After conducting a comprehensive analysis of the solidification conditions captured by an infrared thermal camera, microstructural characterization, and residual stress analyzed using high-energy synchrotron X-ray diffraction, it has been determined that the initiation mechanism behind delayed hot cracking in wire-based L-DED is attributed to inadequate backfilling to shrinkage during solidification. This insufficient backfilling results in solidification cracking. Additionally, the competitive growth pattern observed between grains with a preferential growth direction and other misaligned grains relative to the heat flow direction contributes to the occurrence of delayed cracking. Regarding the mechanical properties, as-built structures produced using wire-based L-DED exhibit superior strength while maintaining a satisfactory level of ductility. The superior mechanical properties observed in wire-based L-DED structures can be attributed to two main factors. Firstly, the cyclic heating effects that occur during successive deposition to previously solidified layers act as an ageing treatment leading to the precipitation of strengthening phases. Secondly, the presence of large columnar grains with an epitaxial growth pattern dominates the microstructure, further enhancing the overall strength of the material.
Process monitoring and quality control are two prominent areas of research in additive manufacturing, particularly for achieving consistent properties in structures intended for serial production and industrialization. Existing methodologies rely on machine learning algorithms for post-process quality control, which can aid in ensuring quality and optimizing resource utilization. Nevertheless, these methods cannot prevent material waste since they are implemented after the structures have already been produced.
Another objective of this thesis is to leverage machine learning algorithms for in-situ adjustment of process parameters in laser-based additive manufacturing, to ensure processing stability.
The high-speed camera captures processing images that contain various characteristics, including the melt pool, plume, and spatter. These characteristics exhibit distinct patterns under different processing states. To effectively analyze and classify these images, a convolutional neural network is employed. The convolutional neural network is trained to learn and recognize the specific characteristics of the process images, enabling it to accurately identify the corresponding processing states. The successful categorization of processing states through the identification of processing images using a convolutional neural network has been validated. This enables the in-situ adjustment of process parameters based on the current processing states, thereby ensuring the maintenance of processing stability. As a result, a consistent quantity and distribution of porosity level can be achieved across various deposition layers.
It can be concluded through the comprehensive analysis conducted in this thesis regarding the L-DED processing high-strength 7XXX series Al alloys as follows: It has been demonstrated that using wire as the feedstock in L-DED shows better processability than using powder. Crack-free thin-wall structures with low porosity levels can be built in wire-based L-DED while achieving a good combination of strength and ductility in these structures simultaneously. Additionally, a methodology involving process monitoring and machine learning has been proposed and proven to be effective in maintaining processing stability and consistency.
Lightweight materials and designs are in high demand in the automobile and aerospace industries because of their potential to reduce carbon emissions and minimize environmental impacts. Aluminum alloys with a high density-strength ratio are commonly utilized as lightweight materials. However, the range of applicable Al alloys in laser-based additive manufacturing is limited due to various processing challenges. These challenges include issues such as oxidation, high reflectivity, porosity formation, and cracking susceptibility. Therefore, it is crucial to address these challenges and establish an optimal process window for Al alloys in laser-based additive manufacturing. Al-Zn-Mg-Cu alloys, also known as 7XXX series alloys, are precipitation-strengthening alloys that demonstrate superior strength characteristics when subjected to appropriate ageing treatment. The rapid heating and cooling involved in laser-based additive manufacturing can be advantageous for processing 7XXX series alloys. The non-equilibrium solidification resulting from the rapid cooling promotes grain refinement and increases the solubility of solutes in the α-Al matrix, thereby enhancing the precipitation-strengthening effects during ageing treatment. Consequently, it is valuable to explore the processability of 7XXX series alloys in laser-based additive manufacturing.
The primary objective of this thesis is to develop an appropriate process window for laser-directed energy deposition (L-DED) in order to process high-strength 7XXX series alloys. Furthermore, the aim is to investigate the relationship between process-microstructure-property. Additionally, considerable attention has been given to understanding the mechanism behind defect formation, specifically porosity and cracks.
7XXX series alloys are processed using wire- and powder-based L-DED, respectively. When it comes to structural integrity, specifically in terms of porosity levels and cracking, wire-based feedstock demonstrates superior processability compared with powder-based feedstock. The high porosity levels observed in structures produced by powder-based L-DED are attributed to two main factors: the presence of intrinsic pores inside powder materials and the evaporation of volatile alloying elements during processing. The installation of a self-designed shielding gas nozzle in wire-based L-DED has been found to greatly improve surface roughness, reduce porosity levels, and enhance processing stability. A delayed hot cracking is noticed in wire-based L-DED processing high-strength Al alloys. The delayed hot cracking is correlated to the deposition length of structures and the number of deposited layers. After conducting a comprehensive analysis of the solidification conditions captured by an infrared thermal camera, microstructural characterization, and residual stress analyzed using high-energy synchrotron X-ray diffraction, it has been determined that the initiation mechanism behind delayed hot cracking in wire-based L-DED is attributed to inadequate backfilling to shrinkage during solidification. This insufficient backfilling results in solidification cracking. Additionally, the competitive growth pattern observed between grains with a preferential growth direction and other misaligned grains relative to the heat flow direction contributes to the occurrence of delayed cracking. Regarding the mechanical properties, as-built structures produced using wire-based L-DED exhibit superior strength while maintaining a satisfactory level of ductility. The superior mechanical properties observed in wire-based L-DED structures can be attributed to two main factors. Firstly, the cyclic heating effects that occur during successive deposition to previously solidified layers act as an ageing treatment leading to the precipitation of strengthening phases. Secondly, the presence of large columnar grains with an epitaxial growth pattern dominates the microstructure, further enhancing the overall strength of the material.
Process monitoring and quality control are two prominent areas of research in additive manufacturing, particularly for achieving consistent properties in structures intended for serial production and industrialization. Existing methodologies rely on machine learning algorithms for post-process quality control, which can aid in ensuring quality and optimizing resource utilization. Nevertheless, these methods cannot prevent material waste since they are implemented after the structures have already been produced.
Another objective of this thesis is to leverage machine learning algorithms for in-situ adjustment of process parameters in laser-based additive manufacturing, to ensure processing stability.
The high-speed camera captures processing images that contain various characteristics, including the melt pool, plume, and spatter. These characteristics exhibit distinct patterns under different processing states. To effectively analyze and classify these images, a convolutional neural network is employed. The convolutional neural network is trained to learn and recognize the specific characteristics of the process images, enabling it to accurately identify the corresponding processing states. The successful categorization of processing states through the identification of processing images using a convolutional neural network has been validated. This enables the in-situ adjustment of process parameters based on the current processing states, thereby ensuring the maintenance of processing stability. As a result, a consistent quantity and distribution of porosity level can be achieved across various deposition layers.
It can be concluded through the comprehensive analysis conducted in this thesis regarding the L-DED processing high-strength 7XXX series Al alloys as follows: It has been demonstrated that using wire as the feedstock in L-DED shows better processability than using powder. Crack-free thin-wall structures with low porosity levels can be built in wire-based L-DED while achieving a good combination of strength and ductility in these structures simultaneously. Additionally, a methodology involving process monitoring and machine learning has been proposed and proven to be effective in maintaining processing stability and consistency.
Subjects
high-strength 7XXX aluminum alloy
Laser-directed energy deposition
microstructure and mechanical properties
porosity and hot cracking
process development
processing monitoring and controlling
DDC Class
620.1: Engineering Mechanics and Materials Science
621: Applied Physics
Loading...
Name
Wang_Mengjie_Process development and digitalization of laser-directed energy deposition processing high-strength 7XXX series aluminum alloy.pdf
Size
156.9 MB
Format
Adobe PDF