Accurately predicting copolymer properties plays a pivotal role in the field of polymer informatics. This endeavor necessitates a comprehensive understanding of polymer structures, adept feature engineering, and proficient application of machine learning algorithms. In traditional methodologies, features for each monomer structure were generated independently, thus, segregating features from individual monomers. This approach results in a less informative representation, with limited applicability. To address these challenges, we introduce an innovative machine learning framework, named weighted-chained-SMILES. By constructing a representative SMILES notation, more intricate information can be encapsulated within the generated features. Our experimental results to predict the thermal properties demonstrate that our approach not only delivers competitive predictive performance but also exhibits enhanced adaptability across a diverse range of molecular representations. The versatility showcased by our model suggests promising potential for tackling more complex copolymer systems and extending its predictive capabilities to various other polymer properties.

Chelating agents are commonly employed in microelectronic processes to prevent metal ion contamination. The ligand fragments of a chelating agent largely determine its binding strength to metal ions. Identification of ligands with suitable characteristics will facilitate the design of chelating agents to enhance the capture and removal of metal ions from the substrate in microelectronic processes. This study employed quantum chemical calculations to simulate the binding process between eleven ligands and the hydrated forms of Ni²⁺, Cu²⁺, Al³⁺, and Fe³⁺ ions. The binding strength between the metal ions and ligands was quantified using binding energy and binding enthalpy. Additionally, we explored the binding interaction mechanisms and explained the differences in binding abilities of the eleven ligands using frontier molecular orbitals, nucleophilic indexes, electrostatic potentials, and energy decomposition calculations based on molecular force fields. Based on our computational results, promising chelating agent structures are proposed, aiming to guide the design of new chelating agents to address metal ion contamination issues in integrated circuit processes.

Balancing the piezoelectric stress coefficient (e₃₃) and the elastic constant (C₃₃) of AlN-based materials is challenging but crucial for its application in bulk acoustic wave resonators. In this work, via first-principles calculations, we demonstrate that e₃₃ and C₃₃ can be simultaneously enhanced by adding boron (B) in the Sc_xAl_1−xN system. Compared with Sc_xAl_1−xN, the large C₃₃ in B_0.125Sc_x−0.125Al_1−xN is caused by the shorter and stiffer B–N bonds, and the enhancement of e₃₃ mainly comes from a larger sensitivity of atomic coordinates with respect to the strain $(|du/d\epsilon |)$ . Further investigation shows that the introduction of B in Sc_xAl_1−xN significantly decreases the ratio of lattice parameter c over a due to the planar coordination preference of bulk BN, which is accompanied by an increased average $|du/d\epsilon |$ and e₃₃. Such a negative correlation can be attributed to the fact that $|du/d\epsilon |$ of N is proportional to the asymmetry of transversal bond strengths but inversely proportional to the overall average bond strengths around N atoms. Such a bonding character observed in this work could be beneficial to the design of wurtzite materials with large piezoelectric coefficients and high elastic constants.

Polyamines have become important chemical components used in several integrated circuit manufacturing processes, such as etching, chemical mechanical polishing (CMP), and cleaning. Recently, researchers pointed out that polyamines can be excellent enhancers in promoting the material removal rate (MRR) of Si CMP, but the interaction mechanism between the polyamines and the silicon surface has not been clarified. Here, the micro-interaction mechanisms of polyamines, including ethylenediamine (EDA), diethylenetriamine (DETA), triethylenetetramine (TETA), tetraethylenepentamine (TEPA), and pentaethylenehexamine (PEHA), with the Si(1, 0, 0) surface were investigated through molecular dynamics (MD) simulations using the ReaxFF reactive force field. Polyamines can adsorb onto the Si(1, 0, 0) surface, and the adsorption rate first accelerates and then tends to stabilize with the increase in the quantity of -CH₂CH₂NH-. The close connection between the adsorption properties of polyamines and the polishing rate has been confirmed by CMP experiments on silicon wafers. A comprehensive bond analysis indicates that the adsorption of polyamines can stretch surface Si–Si bonds, which facilitates subsequent material removal by abrasive mechanical wear. This work reveals the adsorption mechanism of polyamines onto the silicon substrate and the understanding of the MRR enhancement in silicon CMP, which provides guidance for the design of CMP slurry.

Optimizing the retaining ring structure can improve the quality of Chemical Mechanical Polishing (CMP). This study establishes a two-dimensional Computational Fluid Dynamics-Discrete Element Method (CFD-DEM) model, while the model is validated by experiments. The results graphically demonstrate the influence of the retaining ring groove design on the motion of the slurry abrasive particles. The size of the retaining ring groove appears to have a threshold value, above which the abrasives start to have significant distribution in the wafer region. As the groove size continues to increase, the number of abrasives entering the ring increases abruptly and oscillates at specific nodes. The abrasive transfer rate increases with the number of grooves in the early stage but reaches a limit at a certain number of grooves. Meanwhile, the retaining ring position affects the transfer of the abrasives. This study provides a base for optimizing the design of retaining rings.

The piezoelectric and elastic properties are critical for the performance of AlN-based 5G RF filters. The improvement of the piezoelectric response in AlN is often accompanied by lattice softening, which compromises the elastic modulus and sound velocities. Optimizing both the piezoelectric and elastic properties simultaneously is both challenging and practically desirable. In this work, 117 X_0.125Y_0.125Al_0.75N compounds were studied with the high-throughput first-principles calculation. B_0.125Er_0.125Al_0.75N, Mg_0.125Ti_0.125Al_0.75N, and Be_0.125Ce_0.125Al_0.75N were found to have both high C₃₃ (>249.592 GPa) and high e₃₃ (>1.869 C/m²). The COMSOL Multiphysics simulation showed that most of the quality factor (Q_r) values and the effective coupling coefficient (K_eff²) of the resonators made with these three materials were higher than those with Sc_0.25AlN with the exception of the K_eff² of Be_0.125Ce_0.125AlN, which was lower due to the higher permittivity. This result demonstrates that double-element doping of AlN is an effective strategy to enhance the piezoelectric strain constant without softening the lattice. A large e₃₃ can be achieved with doping elements having d-/f- electrons and large internal atomic coordinate changes of du/dε. The doping elements–nitrogen bond with a smaller electronegativity difference (ΔEd) leads to a larger elastic constant C₃₃.