A Global Optimizer for Nanoclusters

Abstract

We have developed an algorithm to automatically build the global minimum and other low-energy minima of nanoclusters. This method is implemented in PyAR (https://github.com/anooplab/pyar) program. The global optimization in PyAR involves two parts, generation of several trial geometries and gradient-based local optimization of the trial geometries. While generating the trial geometries, a Tabu list is used for storing the information of the already used trial geometries to avoid using the similar trial geometries. In this recursive algorithm, an n-sized cluster is built from the geometries of n−1 clusters. The overall procedure automatically generates many unique minimum energy geometries of clusters with size from 2 up to n using this evolutionary growth strategy. We have used our strategy on some of the well-studied clusters such as Pd, Pt, Au, and Al homometallic clusters, Ru-Pt and Au-Pt binary clusters, and Ag-Au-Pt ternary cluster. We have analyzed some of the popular parameters to characterize the clusters, such as relative energy, singlet-triplet energy difference, binding energy, second-order energy difference, and mixing energy, and compared with the reported properties.

Idarubicin–Gold Complex: From Crystal Growth to Gold Nanoparticles

“…The trigonal plane Au 6 -clustered structure is directly inspired from previous publications. 23 − 25 Please note that only relevant results are mentioned in the main text, whereas most of the computed data are reported in the Supporting Information . In addition, the electrostatic potential (ESP) and corresponding contour maps (MESP) have been considered for structures sI to sIII.…”

Role of atomicity in the oxygen reduction reaction activity of platinum sub nanometer clusters: A global optimization study

“…The global optimization of nanoclusters is carried out using automated structure prediction strategy employed in Python for aggregation and reaction (PyAR). The details of the implementation of the GO algorithm can be found in references Nandi et al 28 and Khatun et al 29 In a nutshell, the structural prediction of clusters is initiated by constructing clusters of size n from the (n − 1) th cluster (termed as seed) by the addition of atoms (termed as monomers) from random directions. The generated cluster geometries are optimized using the semiempirical XTB 6.2.2 program to get refined geometries 30,31 .…”

“…The vibrational frequencies are calculated to confirm the nature of optimized structures. For checking similarity of the resulting cluster geometries, we have considered two parameters; (i) energy difference between two clusters, (ii) fingerprint distance defined as the Euclidean distance between the eigenvalues obtained from the diagonalization of Coulomb matrix constructed between two cluster geometries 41,42,29 . A threshold of 10 −4 eV for energy difference and 1 Å for fingerprint deviation are applied for removing similar geometries.…”

BAl4Mg−/0/+: Global Minima with a Planar Tetracoordinate or Hypercoordinate Boron Atom

“…Several trial geometries of BAl 4 Mg −/0/+ are generated by chemical intuition and the cluster building procedure implemented in the Python program for aggregation and reaction (PyAR) [83,84]. Modeling by intuition was conducted, targeting for ptB and ppB based on similar reported molecules.…”

“…Modeling by intuition was conducted, targeting for ptB and ppB based on similar reported molecules. The automated cluster building was done as follows: first, a diatomic molecule was generated from two randomly chosen atoms from B, Al, and Mg. To achieve the optimized geometry of these diatomic molecules, another randomly chosen atom was added following the procedure described in [84] to generate several (N) estimated geometries. All these geometries were optimized, unique minima were chosen, and the further addition of random atoms was continued until the target chemical formula was reached.…”