Abstract:Spin coating is a wide-spread, quick, and inexpensive method to create nanometer-thick thin films of various polymers, such as polystyrene, on top of solid substrates. Since the film thickness determines the mechanical, optical, and degradation properties of the coated film, it is essential to develop a simple method to predict thickness based on other manipulatable factors. In this study, a three-dimensional manifold relating initial solution concentration, thin film coverage thickness, and monodisperse bulk molecular weight is developed utilizing curve-fit machine learning. The model is able to receive polystyrene bulk molecular weight and desired thin film thickness as input and output an accurate prediction of initial solution concentration required to generate a coating of a desired thickness.

What problem does this paper attempt to address?

This paper discusses how to utilize machine learning to establish a three-dimensional correlation model between the thickness, solution concentration, and molecular weight of polystyrene (PS) films. In this study, a machine learning method of curve fitting was used to analyze a dataset of PS samples prepared by spin-coating, in order to predict the unknown third factor (film thickness) given two factors (molecular weight and concentration). The objective of the research is to develop a model that can predict based on the relationship between initial solution concentration, film thickness, and molecular weight of high-molecular-weight polymers. The thickness of the film is crucial for its mechanical, optical, and degradation properties, thus requiring a simple method for controlling and predicting its thickness. Polystyrene is widely used in various fields, such as semiconductors, optics, and biomedical engineering, due to its mechanical strength, transparency, and processability. In this paper, the researchers proposed, for the first time, a Python-based curve fitting machine learning model that accurately predicts the film thickness based on input molecular weight and concentration. This allows researchers to adjust the concentration based on any available molecular weight to generate a film with the desired thickness, or determine the molecular weight of unknown polystyrene by measuring the film thickness spun-coated at a known concentration. The paper also discusses how molecular weight and concentration affect film thickness, and how mathematical models in previous studies explain these relationships through theoretical physics and experimental data. By revealing the complex connections between polystyrene film thickness, molecular weight, and solution concentration, this research not only deepens our understanding of film physics but also provides a platform for material design and introduces new capabilities for predicting film engineering. In summary, this paper presents a new method for predicting the thickness of spin-coated polystyrene films, which contributes to optimizing film performance and advancing the development of precision-engineered film coatings in various industries.

Utilizing Machine Learning to Model Interdependency of Bulk Molecular Weight, Solution Concentration, and Thickness of Spin Coated Polystyrene Thin Films

Utilizing machine learning to model interdependency of bulk molecular weight, solution concentration, and thickness of spin coated polystyrene thin films

Radial Basis Function Neural Network and Overlay Sampling Uniform Design Toward Polylactic Acid Molecular Weight Prediction

A Machine Learning Approach Capturing Hidden Parameters in Autonomous Thin-Film Deposition

A Machine Learning Approach to Model the Electrospinning Process of Biocompatible Materials

The Effect of Machine Learning Algorithms on the Prediction of Layer-by-Layer Coating Properties

A Machine Learning Model for the Prediction of Water Contact Angles on Solid Polymers

Machine Learning-enhanced Efficient Spectroscopic Ellipsometry Modeling

Enhancing Copolymer Property Prediction through the Weighted-Chained-SMILES Machine Learning Framework

Machine Learning-Assisted Identification of Copolymer Microstructures Based on Microscopic Images

Analyzing and Predicting the Viscosity of Polymer Nanocomposites in the Conditions of Temperature, Shear Rate, and Nanoparticle Loading with Molecular Dynamics Simulations and Machine Learning

Multi‐source data‐driven approach for prediction of melt density during polymer compounding

Numerical simulation, clustering and prediction of multi-component polymer precipitation

Deep learning-assisted morphological segmentation for effective particle area estimation and prediction of interfacial properties in polymer composites

Monolayer and Multilayer Compact Film Formations on Different Substrates by Physical Deposition

Predicting Adhesive Free Energies of Polymer--Surface Interactions with Machine Learning

Predicting and Accelerating Nanomaterials Synthesis Using Machine Learning Featurization

Transfer Learning of Full Molecular Weight Distributions via High-Throughput Computer-Controlled Polymerization

Deep-Learning-Enhanced Modeling of Electrosprayed Particle Assembly on Non-spherical Droplet Surfaces

Extrapolative Machine Learning Models for Copolymers

Representing Polymers as Periodic Graphs with Learned Descriptors for Accurate Polymer Property Predictions