arXiv subjects

Manoj Settem

Publications and source records attributed to Manoj Settem.

6 recordsLinked to original sources

arXiv2026

Grain boundary evolution in nanoparticles

Grain boundary evolution is a key dynamical process that enables structural rearrangements in nanoparticles and drives them towards low energy configurations. Grain boundaries can also enhance catalytic properties, making it important to understand the elementary processes underlying their evolution for grain boundary engineering in nanoparticles. Compared with bulk materials, nanoparticles have additional rotational and translational degrees of freedom, can accommodate structural changes through shape relaxation, and grain boundaries terminate at free surfaces. However, the atomic-scale mechanisms of GB evolution under these less constrained conditions remain comparatively less understood than in bulk materials. Here, using atomistic simulations, we focus on low energy $\Sigma3$ (coherent twin boundary) and $\Sigma11$ grain boundaries, which are among the persistent compact GB structures that emerge during nanoparticle structural evolution. We identify two fundamental atomic displacements, column shift (C) and screw shift (S), that recur during the evolution of these grain boundaries and their junctions. These displacements occur through different atomic pathways and combine in different ways to generate grain boundary migration, structural transformations, and junction evolution. In particular, the same initial and final grain boundary configurations can be connected through different atomic pathways, and a column shift can occur either as a full shift or through disconnection kinks. C and S remain identifiable even when the grain boundary character changes, for example during a $\Sigma11$ to $\Sigma3$ transformation. An understanding of these elementary GB processes can help identify strategies to control grain boundary evolution and thereby engineer GB structures in nanoparticles.

cond-mat.mtrl-sci↗

arXiv2026

Structural Chart of Copper-Silver Nanoalloys through machine learning

Nanoalloys (or alloy nanoparticles) are an important class of materials that are promising for their functional properties. However, designing synthesis protocols to control their structure and chemical ordering is rather challenging. Part of this difficulty stems from the lack of information on their metastable and stable structures. Here, we develop a general computational framework to construct a structural chart of nanoalloys using 38-atom AgCu nanoalloys as a model system. Initially, the equilibrium structural distribution is sampled using parallel tempering combined with molecular dynamics (PTMD). Using a machine learning (ML) based approach, the vast number of sampled configurations are classified into various structural classes. This ML approach produces a single three-dimensional map in which all structures and compositions can be visualized and discriminated. Finally, a finite-temperature structural chart is constructed which provides information on the dominant structures across the entire range of compositions and temperatures. In addition, the structural chart reveals significant differences in thermal stability between nanoalloys and bulk alloys. The presented framework provides an effective route to compute and map the vast structural and chemical space of multicomponent nanoparticles, paving the way to the rational design of functional nanoalloys.

cond-mat.mtrl-sci↗

arXiv2024

Inherent structural descriptors via machine learning

Finding proper collective variables for complex systems and processes is one of the most challenging tasks in simulations, which limits the interpretation of experimental and simulated data and the application of enhanced sampling techniques. Here, we propose a machine learning approach able to distill few, physically relevant variables by associating instantaneous configurations of the system to their corresponding inherent structures as defined in liquids theory. We apply this approach to the challenging case of structural transitions in nanoclusters, managing to characterize and explore the structural complexity of an experimentally relevant system constituted by 147 gold atoms. Our inherent-structure variables are shown to be effective at computing complex free-energy landscapes, transition rates, and at describing non-equilibrium melting and freezing processes. The effectiveness of this machine learning strategy guided by the generally-applicable concept of inherent structures shows promise to devise collective variables for a vast range of systems, including liquids, glasses, and proteins.

physics.comp-ph↗

arXiv2023

Structural transformations in Cu, Ag, and Au metal nanoclusters

Finite-temperature structures of Cu, Ag, and Au metal nanoclusters are calculated in the entire temperature range from 0 K to melting using a computational methodology that we proposed recently [Settem \emph{et al.}, Nanoscale, 2022, 14, 939]. In this method, Harmonic Superposition Approximation (HSA) and Parallel Tempering Molecular Dynamics (PTMD) are combined in a complementary manner. HSA is accurate at low temperatures and fails at higher temperatures. PTMD, on the other hand, effectively samples the high temperature region and melting. This method is used to study the size- and system-dependent competition between various structural motifs of Cu, Ag, and Au nanoclusters in the size range 1 to 2 nm. Results show that there are mainly three types of structural changes in metal nanoclusters depending on whether a solid-solid transformation occurs. In the first type, global minimum is the dominant motif in the entire temperature range. In contrast, when a solid-solid transformation occurs, the global minimum transforms either completely to a different motif or partially resulting in a co-existence of multiple motifs. Finally, nanocluster structures are analyzed to highlight the system-specific differences across the three metals.

cond-mat.mtrl-sci↗

arXiv2023

Charting nanocluster structures via convolutional neural networks

A general method to obtain a representation of the structural landscape of nanoparticles in terms of a limited number of variables is proposed. The method is applied to a large dataset of parallel tempering molecular dynamics simulations of gold clusters of 90 and 147 atoms, silver clusters of 147 atoms, and copper clusters of 147 atoms, covering a plethora of structures and temperatures. The method leverages convolutional neural networks to learn the radial distribution functions of the nanoclusters and to distill a low-dimensional chart of the structural landscape. This strategy is found to give rise to a physically meaningful and differentiable mapping of the atom positions to a low-dimensional manifold, in which the main structural motifs are clearly discriminated and meaningfully ordered. Furthermore, unsupervised clustering on the low-dimensional data proved effective at further splitting the motifs into structural subfamilies characterized by very fine and physically relevant differences, such as the presence of specific punctual or planar defects or of atoms with particular coordination features. Owing to these peculiarities, the chart also enabled tracking of the complex structural evolution in a reactive trajectory. In addition to visualization and analysis of complex structural landscapes, the presented approach offers a general, low-dimensional set of differentiable variables which has the potential to be used for exploration and enhanced sampling purposes.

cond-mat.mtrl-sci↗

arXiv2022

Tempering of Au nanoclusters: capturing the temperature-dependent competition among structural motifs

A computational approach to determine the equilibrium structures of nanoclusters in the whole temperature range from 0 K to melting is developed. Our approach relies on Parallel Tempering Molecular Dynamics (PTMD) simulations complemented by Harmonic Superposition Approximation (HSA) calculations and global optimization searches, thus combining the accuracy of global optimization and HSA in describing the low-energy part of configuration space, together with the PTDM thorough sampling of high-energy configurations. This combined methodology is shown to be instrumental towards revealing the temperature-dependent structural motifs in Au nanoclusters of sizes 90, 147, and 201 atoms. The reported phenomenology is particularly rich, displaying a size- and temperature-dependent competition between the global energy minimum and other structural motifs. In the case of Au$_{90}$ and Au$_{147}$, the global minimum is also the dominant structure at finite temperatures. In contrast, the Au$_{201}$ cluster undergoes a solid-solid transformation at low temperature (< 200 K). Results indicate that PTMD and HSA very well agree at intermediate temperatures, between 300 and 400 K. For higher temperatures, PTMD gives an accurate description of equilibrium, while HSA fails in describing the melting range. On the other hand, HSA is more efficient in catching low-temperature structural transitions. Finally, we describe the elusive structures close to the melting region which can present complex and defective geometries, that are otherwise difficult to characterize through experimental imaging.

physics.atm-clus↗