CONVERSION OF POLYPROPLENE- RICH DISPOSABLE FACE MASK THROUGH CATALYTIC PYROLYSIS TO PRODUCE HIGH-VALUE HYDROCARBON-OIL

Abstract

The exponential rise in the global energy demand along with the steep depletion of conventional fossil fuels necessitate an alternative and sustainable energy source. Moreover, the profusion of waste plastic generation (projected to be around 155 to 265 metric tons per year by 2060) without proper disposal methods has led to a global concern that needs a potent solution (Lebreton and Andrady, 2019). Likewise, the upsurge in the waste generation of disposable face masks during the COVID-19 pandemic as well as its persistence in the environment over the years without a proper disposal method has led to a global concern that needs a potent solution. Their non biodegradable nature and prolonged presence in the environment leads to their progressive breakdown into micro and nano-size plastic fragments that can be easily ingested by animals and humans, thus posing a severe threat to the society (X. Chen et al., 2021). A thermochemical conversion approach provides a promising solution for converting waste to energy, meeting the demands of both, the depleting fossil fuel crisis as well as the requirement for proper disposal methods. Different thermochemical processes like pyrolysis, gasification, combustion, hydrothermal liquefaction, etc., are available to convert various feedstocks into high-value products. Amongst these, pyrolysis is preferred because of the cost-efficient, easy-to-use, and environment-friendly nature of the process (Czajczyńska et al., 2017; Praveenkumar et al., 2024). Moreover, the presence of high amount of carbon, hydrogen, and volatile matter in the disposable face mask illustrated by its ultimate analysis (carbon: 77.77%, hydrogen: 13.4%, nitrogen: 0.05%, oxygen: 8.78%) and proximate analysis (volatile matter: 83.35%, ash: 11.95%, fixed carbon: 4.70%) makes it a potential source of energy that can be harvested via pyrolysis technique. Thus, present study explores the feasibility of converting disposable face mask to energy using a low-cost waste (spent aluminium hydroxide/oxide nanoparticle adsorbent) derived catalyst. Subsequently, analysis of the textural and morphological characteristics of the catalyst using different analytical techniques such as XRD, FTIR, BET, FESEM and NH3-TPD demonstrates a mesoporous structure, with a surface area of 161.20 m2 g-1 and total pore volume of 0.25 cm3 g-1. Furthermore, the presence of all three acidic sites denotes to the availability of large number of binding sites, that can enhance the reactivity of the process. Thermogravimetric analysis of the non-catalytic and catalytic pyrolysis (using spent AHNP derived !-Al2O3) of disposable face mask was conducted at varied heating rates of 10oC/min, 20oC/min, 30oC/min, 40oC/min, and 50oC/min, respectively. Iso-conversional methods, Kissinger Akahira Sunose (KAS) and Ozawa Flynn Wall (OFW) were used for the kinetic study. Additionally, the thermodynamic parameters of the process namely Gibb’s free energy (∆G, kJmol-1), enthalpy (∆H, kJmol-1), and entropy (∆S, kJmol-1K-1) were also determined. Results found that addition of catalyst to the process benefits the overall efficacy of the process by reducing the activation energy (Ea) (without catalyst; OFW-Ea: 188.7 kJ/mol, KAS-Ea: 186.2 kJ/mol; with bare alumina (!-Al2O3); OFW-Ea: 183.2 kJ/mol, KAS-Ea: 180.4 kJ/mol) as well as lowering the disordered state of the process. The results illustrated the potential efficacy of utilizing spent aluminium hydroxide/oxide adsorbent based catalyst in place of high-cost commercial catalysts. However, conventional analysis techniques do not provide insights into the influence of characteristics of feedstock on the process kinetics. Therefore, a study was conducted to exemplify the efficacy of using machine learning for the predictive modeling of the pyrolysis process kinetics in order to understand the complexities of the interrelations of predictor variables and their influence on activation energy. The activation energy for pyrolysis of waste plastics was evaluated using machine learning models namely Random Forest, XGBoost, CatBoost, and AdaBoost regression models. Characteristics of the feedstock, types of plastic, conversion values and iso-conversional methods (OFW, KAS, and Friedman (FR)) used for kinetic evaluation, were taken as predictor variables. A total 674 data points were collected from literature as well as the present investigation. Feature selection based on the multicollinearity of data and hyperparameter tuning of the models utilizing RandomizedSearchCV was conducted. Random forest model outperformed the other models with R2 value of 0.941. Shapely additive explanations projected fixed carbon content, ash content, conversion value, and carbon content as significant parameters of the model. Moreover, on comparing the predicted results of activation energy with the results estimated using TGA data through iso-conversional method, an absolute error range of 0.1 to 1.8% (for OFW) and 0.2 to 1.2% (for KAS) in the conversion range of 0.1 to 0.9 was observed. These results further substantiate the reliability and robustness of the model developed using machine learning approach. Subsequently, the modelling and optimization of the catalytic pyrolysis of face masks incorporating a waste-derived catalyst to analyze the effect of process parameters (temperature, feed-to-catalyst ratio, and inert gas flow rate) on the oil yield of the process was conducted. The experimental study was carried out in a semi-batch reactor using spent AHNP derived !-Al2O3 ii as a catalyst. Response surface methodology and machine learning were used for the optimization and modeling of the process. Both response surface methodology (rotational central composite design, R2-0.95) and machine learning (decision trees regression, R2-0.83) demonstrated a higher prediction accuracy and lower error margins. Both the modeling processes provide results with high efficacy. However, the employment of machine learning adds on to the knowledge of the interrelations of the predictor variable with the output of the model by employing the explainable artificial intelligence (XAI). Pyrolysis temperature was spot lighted to be the predominant parameter followed by feed-to-catalyst ratio. Experimental oil yield (13.5%) obtained at optimized parameters (516 oC temperature, 3:1 feed-to-catalyst ratio, and 163 mL/min inert gas flow rate) was compared with those predicted through response surface methodology (13.7%) and decision trees regression (13.12%), showcasing an absolute error range of 0.2-0.4 wt%. Thus, the results highlight the successful modeling and optimization of the catalytic pyrolysis of face masks. Further, at optimized conditions, the effects of metal doped spent AHNP derived !-Al2O3 catalysts on the pyrolysis product distribution and composition were investigated. Monometallic (Ni, Co, Fe impregnated) and bimetallic (Ni-Co, Ni-Fe, Co-Fe impregnated) waste derived alumina catalysts were explored for their catalytic performance to produce commercial range fuels from disposable face mask and further compared with the bare alumina (!-Al2O3) based catalytic as well as non-catalytic processes. Catalysts were used in an ex-situ mode. Characterization of the metal doped catalysts highlighted bi-modal mesoporous structure with surface area in the range of 140.42 to 161.20 m2 g-1 that helps to enhance the stability and activity of the catalysts. Ni-Co/Al and Co-Fe/Al exhibited higher oil production of 14.9 wt% while Fe/Al demonstrated increased non-condensable gas production of 78.8 wt%. Ni/Al showed a higher selectivity for naphthene and paraffins likewise, Fe/Al exhibited a higher selectivity for olefins in the oil. Likewise, different catalysts exhibit distinctive advantages in terms of product distribution and composition due to their individual characteristics. However, in terms of the calorific value of oil, Ni/Al presented the best results with HHV of 44.64 MJ/kg. The present study demonstrates the feasibility of valorizing the face mask into energy-dense oil comparable to commercial range fuel using a spent adsorbent based catalyst. The potential environmental effects associated with the non-catalytic and catalytic pyrolysis of disposable face mask was estimated using life cycle analysis (LCA) and the economic assessment iii of the processes was conducted in terms of operational cost analysis. Mono-/bimetal doped spent adsorbent (aluminium oxide/hydroxide nanoparticles) derived catalysts were utilized in the process and assessed for their impact on the environment. The impact of individual process units on the distinct environmental indices of CML 2001-August 2016 methodology of ‘Sphera’s LCA for experts’ software was assessed. The scenario comprising of recycling the energy produced via combustion of non-condensable gases along with utilizing the excess energy as well as char for energy and coal credits demonstrates a significant reduction in the LCA impact values of the overall process. The sub-processes significantly effecting the environment factors were evaluated to be the pyrolysis process, combustion of hydrocarbon-oil, catalyst, and shredding of the face mask. The lack of bias, robustness, and reliability of the LCA results were further established by the Monte Carlo uncertainty analysis showcasing a standard deviation of less than 10%. Based on the operational cost assessment, the pyrolysis process can be arranged in a sequence as Al < non-catalytic < Fe/Al < Ni-Fe/Al < Ni/Al < Co-Fe/Al < Ni-Co/Al < Co/Al. Additionally, the operational cost of the process presented in the study is significantly lower than the studies utilizing the commercial based catalyst. Thus, the present study highlights the eco-friendly nature of the non-catalytic and catalytic pyrolysis of the disposable face mask utilizing a spent adsorbent derived catalyst and also provides a comprehensive insight for the scale-up of the process following a circular economy approach.