Thermo-Chemical conversion of polyolefin-based facemask using bench-scale pyrolysis system

The generation of personal protective equipment (PPE) waste due to the impact of COVID has increased multi-fold globally. In this study, pyrolysis of polyolefin-based PPEs was carried out using a bench-scale reactor of 2 kg per batch capacity. Thermogravimetric (TGA) analysis of face masks was carried out to identify the optimal parameters for the pyrolysis process. Different combinations of catalysts (ZSM-5 and montmorillonite), catalyst to feed ratio (2.5% and 5%), experiment duration (2 h and 3 h), and process temperature (450 degrees C and 510 degrees C) were tested to determine the maximum yield of the pyrolysis oil. The oil and char obtained from the pyrolysis of PPEs were analyzed for its gross calorific value (GCV), elemental analysis (CHNS), and chemical composition. Based on the experiments conducted, the optimum pyrolysis temperature, catalyst, catalyst to feed ratio, and batch time for maximum oil yield (55.9% w/w) were determined to be 510 degrees C, ZSM-5, 5%, and 2 hours, respectively. Oil was free of sulfur and had a calorific value of 43.7 MJ/kg, which is comparable to commercial diesel fuel and makes it a suitable alternative fuel for ships, boilers, and furnaces.

Thermal degradation model of used surgical masks based on machine learning methodology.

Background: The COVID-19 pandemic has leveraged facial masks to be one of the most effective measures to prevent the spread of the virus, which thereby has exponentially increased the usage of facial masks that lead to medical waste mismanagements which pose a serious threat to life. Thermal degradation or pyrolysis is an effective treatment method for the used facial mask wastes and this study aims to investigate the thermal degradation of the same. Methods: Predicted the TGA experimental curves of the mask components using a Machine Learning model known as Artificial Neural Network (ANN). Significant findings: Three different parts of the mask namely- ribbon, body, and corner were separated and used for the analysis. The thermal degradation behavior is studied using Thermogravimetric Analysis (TGA) and this is crucial for determining the reactivity of the individual mask components as they are subjected to a range of temperatures. Using the curves obtained from TGA, kinetic parameters such as Activation energy (E) and Pre-exponential factor (A) were estimated using the Coats-Redfern model-fitting method. Using the determined kinetic parameters, thermodynamic quantities such as a change in Enthalpy (ΔH), Entropy (ΔS), and Gibbs-Free energy (ΔG) were also calculated. Since TGA is a costly and time-consuming process, this study attempted to predict the TGA experimental curves of the mask components using a Machine Learning model known as Artificial Neural Network (ANN). The dataset obtained at a heating rate of 10°C/min was used to train the 3 different neural networks corresponding to the mask components and it showed an excellent agreement with experimental data (R2 > 0.99). Through this study, a complex chemical process such as thermal degradation was modelled using Machine Learning based on available experimental parameters without delving into the intricacies and complexities of the process.