Prediction of terephthalic acid (TPA) yield in aqueous hydrolysis of polyethylene terephthalate (PET)
This work provides a tool for material scientists to optimize PET recycling conditions, saving time and resources, but it is incremental as it applies existing ML methods to a new dataset in a specific domain.
The study tackled predicting terephthalic acid (TPA) yield in PET hydrolysis by developing machine learning models using 381 experimental data points, with the ANFIS model achieving the best performance as measured by R-squared and RMSE metrics.
Aqueous hydrolysis is used to chemically recycle polyethylene terephthalate (PET) due to the production of high-quality terephthalic acid (TPA), the PET monomer. PET hydrolysis depends on various reaction conditions including PET size, catalyst concentration, reaction temperature, etc. So, modeling PET hydrolysis by considering the effective factors can provide useful information for material scientists to specify how to design and run these reactions. It will save time, energy, and materials by optimizing the hydrolysis conditions. Machine learning algorithms enable to design models to predict output results. For the first time, 381 experimental data were gathered to model the aqueous hydrolysis of PET. Effective reaction conditions on PET hydrolysis were connected to TPA yield. The logistic regression was applied to rank the reaction conditions. Two algorithms were proposed, artificial neural network multilayer perceptron (ANN-MLP) and adaptive network-based fuzzy inference system (ANFIS). The dataset was divided into training and testing sets to train and test the models, respectively. The models predicted TPA yield sufficiently where the ANFIS model outperformed. R-squared (R2) and Root Mean Square Error (RMSE) loss functions were employed to measure the efficiency of the models and evaluate their performance.