The Selective Laser Melting (SLM) process is employed in high-precision layer-by-layer Additive Manufacturing (AM) on powder bed and aims to fabricate high-quality structural components. To gain a comprehensive understanding of the process and its optimization, both modeling and simulation in conjunction with extensive experimental studies along with laser calibration studies have been attempted. Multiscale and multi-physics-based simulations have the potential to bring out a new level of insight into the complex interaction of laser melting, solidification, and defect formation in the SLM parts. SLM process encompasses various physical phenomena during the formation of metal parts, starting with laser beam incidence and heat generation, heat transfer, melt/fluid flow, phase transition, and microstructure solidification. To effectively model this Multiphysics problem, it is imperative to consider different scales and compatible boundary conditions in the simulations. In this paper, we employ a numerical model for the SLM process, leveraging multi-scale and multi-physics simulation strategies. The model will describe the transition from powder to melt and melt to microstructure solid by applying the appropriate boundary conditions at each stage in the transition process. The model also accounts for temperature-dependent material properties of Ti-6Al-4V alloy, including specific heat capacity, thermal conductivity, viscosity, etc. These effective properties are evaluated under both room temperature and elevated temperature conditions through Molecular Dynamics (MD) simulations. The basic behaviour of melting-related property variation is to be studied and the effect on the melt pool characteristics is simulated. The ultimate aim of the scheme is to plug in temperature-dependent material properties in the model and predict the temporal distribution in the melt pool. The simulation results provide a detailed explanation of the SLM process in all three phases (powder, melt, and microstructure solid).

In this work, we use pseudo-spring-augmented smoothed particle hydrodynamics (SPH) framework to understand how the crack paths differ in edge-cracked plates with inclusions when they are made of functionally graded material (FGM) versus homogeneous plates. Modeling crack propagation in such multi-component structural systems is necessary to uncover the underlying failure mechanisms. While traditionally researchers have used mesh-based techniques like the finite element method to understand crack propagation, these methods have limitations. Consequently, mesh-less techniques such as SPH are gaining popularity. After verifying our framework on a plate made of two materials, we compare and contrast the crack path propagation between FGM plates and homogeneous plates, both having soft and hard inclusions. The crack paths get influenced significantly due to the presence of inclusions. Regardless of the type, in presence of soft inclusions, plates fail due to the failure of the inclusion. On the other hand, cracks tend to deflect away from hard inclusions in both plates. The amount of deflection is governed by the relative stiffness of the plate material. Consequently, the deflection is different in FGM plates compared with homogeneous ones.

We develop a sequential multi-scaling framework for studying the problem of fatigue crack propagation due to liquid droplet impingement. The scope is limited to a hypothetical material and a hypothetical liquid. The multi-scaling is achieved by handshaking the atomistic scale molecular dynamics (MD) simulations with the continuum scale smoothed particle hydrodynamics (SPH). The handshaking, in turn, is performed by evaluating the material, the fracture and the loading properties from MD simulations, and using them as inputs in the continuum scale SPH model. Due to the qualitative agreement of the pressure developed in the fluid and the substrate between the MD simulations and already published results, the liquid droplet impact in SPH is simulated through appropriate surface stresses. Further, we incorporate the pseudo-spring approach within the SPH model to develop a methodology for studying mixed-mode fatigue crack propagation. Our methodology provides good agreement with the existing literature for several cases. Lastly, we calculate the fatigue life of an edge-cracked specimen due to liquid droplet impingement.

The existing smoothed particle hydrodynamics (SPH) approaches for propagating fatigue cracks involve either the deletion of the crack front particle or stopping all its interactions in the total Lagrangian form. Here, we adopt the pseudo-spring-based Eulerian form of SPH to model mode-I fatigue crack propagation. For modeling fatigue crack growth, only the interactions between the crack front particle and its neighbors, which display the largest axial stresses in the connected pseudo-springs, are stopped. We show that our framework can determine accurately the mode-I stress intensity factors (SIFs) and capture both the fatigue crack path and the fatigue life of different specimens.

We study the problem of a projectile impacting on a target at different length scales employing two widely used techniques — molecular dynamics (MD) simulations at atomistic scales and smooth particle hydrodynamics (SPH) at continuum scales. At the atomistic scale, the impact problem is modeled through a short-ranged pair potential whereas, at the continuum scale, the partial differential equations related to the different conservation equations are solved using SPH. SPH is parametrized in a manner that the equation of state and the elastic constants match those of the underlying MD simulations. We evaluate a number of recent improvements to the SPH framework – artificial viscosity, XSPH, Jaumann stress rotation, tensile instability, and mass initialization – by making a systematic comparison between an SPH model with varying degrees of these corrections, and a model based on MD, which acts as the pseudo experiment. We show that if all corrections are incorporated, the SPH results agree well with those from MD simulations. This has been ascertained by comparing the deformation response, the radial distribution function, and the shear strain.

Significant advances in nanoscale research have enabled the continuous miniaturization of devices. With the reduction in the size of the devices, it is important to identify if the continuum scale methods remain applicable to such small-scale systems. Motivated by this, the present work tries to understand the equivalence, or its lack thereof, of the continuum scale Eulerian Smoothed Particle Hydrodynamics (ESPH) and Total Lagrangian Smoothed Particle Hydrodynamics (TLSPH) with the atomistic scale molecular dynamics (MD) simulations. The equivalence is studied using four simple problems — (i) uniaxial tensile testing of a beam, (ii) stress profile in a pre-notched plate under small extension, (iii) head-on collision of two elastic rubber-like rings, and (iv) large deformation of a cantilever beam subjected to an impact at the free end. Using MD simulation data as the pseudo-experimental data, we show that both ESPH and TLSPH provide results that are qualitatively and quantitatively in agreement with the MD simulations if the properties at the continuum scale are obtained directly from the MD simulations, and the same initial conditions are chosen. The comparisons are based on the stress–strain behavior, the distribution of normal and shear stresses, the temporal evolution of the variables such as kinetic energy, etc.

We study the problem of Parkes cantilever beam using a multiscale framework. The problem comprises a cantilever target, which is subjected to a projectile impact at the free end. The two scales involved in our study are—the atomistic scale based molecular dynamics (MD) and the continuum scale based smooth particle applied mechanics (SPAM). Both methods, being particle-based, provide us a unique opportunity to compare (and possibly, improve) the SPAM method using the inputs from the MD simulations. The purpose of this study is to compare the effects of different corrections proposed within the SPAM framework and see which of the results yield a solution closer to MD simulations. For this purpose, three corrections—Artificial viscosity correction, XSPH correction, and Jaumann stress rate correction are studied. The comparison is made by looking at the deformation/fracture pattern, the radial distribution function, the centrosymmetry parameter, and the shear strain.