BACKGROUNDMetallic components are the cornerstone of modern industries such as aviation, aerospace, automobile manufacturing, and energy production. The stringent requirements for high-performance metallic components impede the optimization of materials selection and manufacturing. Laser-based additive manufacturing (AM) is a key strategic technology for technological innovation and industrial sustainability. As the number of applications increases, so do the scientific and technological challenges. Because laser AM has domain-by-domain (e.g., point-by-point, line-by-line, and layer-by-layer) localized forming characteristics, the requisite for printing process and performance control encompasses more than six orders of magnitude, from the microstructure (nanometer- to micrometer-scale) to macroscale structure and performance of components (millimeter- to meter-scale). The traditional route of laser-metal AM follows a typical “series mode” from design to build, resulting in a cumbersome trial-and-error methodology that creates challenges for obtaining high-performance goals.ADVANCESWe propose a holistic concept of material-structure-performance integrated additive manufacturing (MSPI-AM) to cope with the extensive challenges of AM. We define MSPI-AM as a one-step AM production of an integral metallic component by integrating multimaterial layout and innovative structures, with an aim to proactively achieve the designed high performance and multifunctionality. Driven by the performance or function to be realized, the MSPI-AM methodology enables the design of multiple materials, new structures, and corresponding printing processes in parallel and emphasizes their mutual compatibility, providing a systematic solution to the existing challenges for laser-metal AM. MSPI-AM is defined by two methodological ideas: “the right materials printed in the right positions” and “unique structures printed for unique functions.” The increasingly creative methods for engineering both micro- and macrostructures within single printed components have led to the use of AM to produce more complicated structures with multimaterials. It is now feasible to design and print multimaterial components with spatially varying microstructures and properties (e.g., nanocomposites, in situ composites, and gradient materials), further enabling the integration of functional structures with electronics within the volume of a laser-printed monolithic part. These complicated structures (e.g., integral topology optimization structures, biomimetic structures learned from nature, and multiscale hierarchical lattice or cellular structures) have led to breakthroughs in both mechanical performance and physical/chemical functionality. Proactive realization of high performance and multifunctionality requires cross-scale coordination mechanisms (i.e., from the nano/microscale to the macroscale).OUTLOOKOur MSPI-AM continues to develop into a practical methodology that contributes to the high performance and multifunctionality goals of AM. Many opportunities exist to enhance MSPI-AM. MSPI-AM relies on a more digitized material and structure development and printing, which could be accomplished by considering different paradigms for AM materials discovery with the Materials Genome Initiative, standardization of formats for digitizing materials and structures to accelerate data aggregation, and a systematic printability database to enhance autonomous decision-making of printers. MSPI-oriented AM becomes more intelligent in processes and production, with the integration of intelligent detection, sensing and monitoring, big-data statistics and analytics, machine learning, and digital twins. MSPI-AM further calls for more hybrid approaches to yield the final high-performance/multifunctional achievements, with more versatile materials selection and more comprehensive integration of virtual manufacturing and real production to navigate more complex printing. We hope that MSPI-AM can become a key strategy for the sustainable development of AM technologies. Download high-res image Open in new tab Download PowerpointMaterial-structure-performance integrated additive manufacturing (MSPI-AM).Versatile designed materials and innovative structures are simultaneously printed within an integral metallic component to yield high performance and multifunctionality, integrating in parallel the core elements of material, structure, process, and performance and a large number of related coupling elements and future potential elements to enhance the multifunctionality of printed components and the maturity and sustainability of laser AM technologies.

Material-structure-performance integrated laser-metal additive manufacturing.

Laser powder bed fusion is a dominant metal 3D printing technology. However, porosity defects remain a challenge for fatigue-sensitive applications. Some porosity is associated with deep and narrow vapor depressions called keyholes, which occur under high-power, low-scan speed laser melting conditions. High-speed x-ray imaging enables operando observation of the detailed formation process of pores in Ti-6Al-4V caused by a critical instability at the keyhole tip. We found that the boundary of the keyhole porosity regime in power-velocity space is sharp and smooth, varying only slightly between the bare plate and powder bed. The critical keyhole instability generates acoustic waves in the melt pool that provide additional yet vital driving force for the pores near the keyhole tip to move away from the keyhole and become trapped as defects.

https://science.sciencemag.org/content/sci/370/6520/1080.full.pdf

Critical instability at moving keyhole tip generates porosity in laser melting.

Additive manufacturing (AM), also known as three-dimensional (3D) printing, has boomed over the last 30 years, and its use has accelerated during the last 5 years AM is a materials-oriented manufacturing technology, and printing resolution versus printing scalability/speed trade-off exists among various types of materials, including polymers, metals, ceramics, glasses, and composite materials Four-dimensional (4D) printing, together with versatile transformation systems, drives researchers to achieve and utilize high dimensional AM Multiple perspectives of the AM of structural materials have been raised and illustrated in this review, including multi-material AM (MMa-AM), multi-modulus AM (MMo-AM), multi-scale AM (MSc-AM), multi-system AM (MSy-AM), multi-dimensional AM (MD-AM), and multi-function AM (MF-AM) The rapid and tremendous development of AM materials and methods offers great potential for structural applications, such as in the aerospace field, the biomedical field, electronic devices, nuclear industry, flexible and wearable devices, soft sensors, actuators, and robotics, jewelry and art decorations, land transportation, underwater devices, and porous structures

Additive manufacturing of structural materials

The recent explosion of machine learning (ML) and artificial intelligence (AI) shows great potential in the breakthrough of metal additive manufacturing (AM) process modeling, which is an indispensable step to derive the process-structure-property relationship. However, the success of conventional machine learning tools in data science is primarily attributed to the unprecedented large amount of labeled data-sets (big data), which can be either obtained by experiments or first-principle simulations. Unfortunately, these labeled data-sets are expensive to obtain in AM due to the high expense of the AM experiments and prohibitive computational cost of high-fidelity simulations, hindering the direct applications of big-data based ML tools to metal AM problems. To fully exploit the power of machine learning for metal AM while alleviating the dependence on “big data”, we put forth a physics-informed neural network (PINN) framework that fuses both data and first physical principles, including conservation laws of momentum, mass, and energy, into the neural network to inform the learning processes. To the best knowledge of the authors, this is the first application of physics-informed deep learning to three dimensional AM processes modeling. Besides, we propose a hard-type approach for Dirichlet boundary conditions (BCs) based on a Heaviside function, which can not only exactly enforce the BCs but also accelerate the learning process. The PINN framework is applied to two representative metal manufacturing problems, including the 2018 NIST AM-Benchmark test series. We carefully assess the performance of the PINN model by comparing the predictions with available experimental data and high-fidelity simulation results, using finite element based variational multi-scale formulation method. The investigations show that the PINN, owed to the additional physical knowledge, can accurately predict the temperature and melt pool dynamics during metal AM processes with only a moderate amount of labeled data-sets. The foray of PINN to metal AM shows the great potential of physics-informed deep learning for broader applications to advanced manufacturing. All the data-sets and the PINN code will be made open-sourced in https://yan.cee.illinois.edu/
 once the paper is published.

https://link.springer.com/content/pdf/10.1007/s00466-020-01952-9.pdf

Machine learning for metal additive manufacturing: predicting temperature and melt pool fluid dynamics using physics-informed neural networks

• Common defects and anomalies in powder bed fusion metal additive manufacturing. • Formation mechanism and practical mitigation strategies are discussed. • Defects/anomalies are classified as powder-related, processing-related, and post-processing related issues. • Properties such as mechanical behavior and corrosion resistance of defective parts are discussed. • Current challenges, gaps, and future trends are discussed. Metal additive manufacturing is a disruptive technology that is revolutionizing the manufacturing industry. Despite its unrivaled capability for directly fabricating metal parts with complex geometries, the wide realization of the technology is currently limited by microstructural defects and anomalies, which could significantly degrade the structural integrity and service performance of the product. Accurate detection, characterization, and prediction of these defects and anomalies have an important and immediate impact in manufacturing fully-dense and defect-free builds. This review seeks to elucidate common defects/anomalies and their formation mechanisms in powder bed fusion additive manufacturing processes. They could arise from raw materials, processing conditions, and post-processing. While defects/anomalies in laser welding have been studied extensively, their formation and evolution remain unclear. Additionally, the existence of powder in powder bed fusion techniques may generate new types of defects, e.g., porosity transferring from powder to builds. Practical strategies to mitigate defects are also addressed through fundamental understanding of their formation. Such explorations enable the validation and calibration of models and ease the process qualification without costly trial-and-error experimentation. 

Defects and anomalies in powder bed fusion metal additive manufacturing

State-of-the-art metal 3D printers promise to revolutionize manufacturing, yet they have not reached optimal operational reliability. The challenge is to control complex laser-powder-melt pool interdependency (dependent upon each other) dynamics. We used high-fidelity simulations, coupled with synchrotron experiments, to capture fast multitransient dynamics at the meso-nanosecond scale and discovered new spatter-induced defect formation mechanisms that depend on the scan strategy and a competition between laser shadowing and expulsion. We derived criteria to stabilize the melt pool dynamics and minimize defects. This will help improve build reliability.

Controlling interdependent meso-nanosecond dynamics and defect generation in metal 3D printing

The Air Force Research Laboratory Additive Manufacturing Modeling Series was executed to create calibration and validation data sets relevant to models of laser powder bed fusion-processed metallic materials. This article describes the data generated for the 2nd of 4 challenge questions which was specifically focused on microscale process-to-structure modeling needs. This work describes the experimental methods, and the resulting characterization data collected from a series of single-track and multi-track deposits built with an EOS M280 from the nickel-based alloy IN625. In general, track dimensions followed common scaling behaviors as a function of processing parameters in quasi-steady-state regions, but significant systematic track geometry variations were quantified in transient regions with more dynamic energy input processes.

Michael G. Chapman

Papers

Controlling interdependent meso-nanosecond dynamics and defect generation in metal 3D printing

AFRL Additive Manufacturing Modeling Series: Challenge 2, Microscale Process-to-Structure Data Description

AFRL AM Modeling Challenge Series: Challenge 3 Data Package

AFRL AM Modeling Challenge Series: Challenge 2 Data Package