Thesis

Experimental assessment, modelling and scale-up of a fixed bed reactor for upgrading the volatile products from the pyrolysis based recycling of fibre reinforced plastic waste

Doctoral student:
Adriana Serras Malillos
Year:
2024
University:
Universidad del País Vasco / Euskal Herriko Unibertsitatea (UPV/EHU)
Director(s):
Alexander Lopez-Urionabarrenechea, Esther Acha
Description:

The work carried out in this doctoral thesis is focused on evaluating the recovery of added-value substances from the treatment of the volatile substances generated in a pyrolysis-based recycling process for fibre reinforced plastic (FRP) waste. This is an innovative objective in the field as current pyrolysis processes do not valorise neither the liquid or gas phase generated during the pyrolysis process. Current FRP waste recycling plants are exclusively focused on the recovery of the fibres.

To this end, pyrolysis experiments have been carried out at laboratory scale. Pyrolysed waste consisted of epoxy and polyester resin based end-of-life (EoL) FRP waste coming from different production sectors. In these pyrolysis experiments, a fixed bed tubular reactor was connected in series to the pyrolysis tank reactor in order to carry out the valorisation of the generated pyrolysis volatiles. Some of these experiments focused on the production of added-value gases. For that, the experimental installation used included a 3 L volume pyrolysis tank-type reactor (operated under slow pyrolysis conditions at 500 °C including, approximately, 100 g of EoL FRP waste sample, and a fixed bed tubular reactor (0.62 m length and 0.02 m diameter) operated at high temperature (900 °C). Results obtained with this experimental configuration showed that it is possible to produce hydrogen rich pyrolysis gases (or pyrolysis gases rich in synthesis gas) almost independently of the type of waste treated. At the same time an aqueous liquid phase is produced, with potential uses or, at least, with easy environmental waste management. The experimental results were consolidated with a preliminary techno-economic analysis. It indicates that profitability of pyrolysis based FRP waste recycling processes could be increased when the valorisation process of the pyrolysis volatiles is integrated. On the other hand, additional experiments were carried out with the focus on the recovery of substances of interest in the liquid phase. Particularly, single aromatic hydrocarbons were targeted. In this latter case, a small tubular pyrolysis reactor (0.36 m length and 0.14 m inner diameter including, approximately, 2.8 g of EoL FRP waste sample) was used to ensure a fast pyrolysis process. Moreover, the pyrolysis and the subsequent volatile treatment were both implemented in the same fixed bed tubular reactor, operated at a lower valorisation process temperature (600 °C) in order to maximise the yield of the generated pyrolysis liquids, and in the presence of different types of zeolites (catalysts that promote the formation of aromatic hydrocarbons). The results obtained in this experimental phase indicate that it is possible to obtain liquids rich in single aromatic hydrocarbons in the presence of H-ZSM5/20Si zeolite, especially when treating polyester-based FRP waste. However, the industrial application of the obtained upgrading pyrolysis liquids would still present significant challenges. From the comparison

XXVIII Summary

of the results obtained in the two mentioned experimental approaches, it is concluded that the most promising solution to be integrated industrially is the valorisation of the generated pyrolysis volatiles towards high added-value pyrolysis gases.

In addition to the described experimental research work, this doctoral thesis was complemented with work related to the industrialisation of the valorisation process of the generated pyrolysis volatiles from the pyrolysis based recycling of FRP waste. Specifically, reactor scale-up calculations based on dimensionless numbers were developed, including the analysis various scenarios combining additional relevant design parameters. Finally, a preliminary kinetic model of the valorisation process of pyrolysis volatiles occurring in the fixed bed reactor was defined and a specialised chemical process software was used to carry out the simulations. Regarding the scale-up, the relevant dimensions of the fixed bed reactor (length, diameter, diameter of the bed particles) were determined in order to be able to treat the higher amount of pyrolysis volatiles that would be generated in a hypothetical pyrolysis plant of higher capacity than the capacity of the laboratory scale reactor used in this thesis work. The target condition to be fulfilled was to ensure the same composition of volatiles at the outlet of the fixed bed reactor as the one obtained at laboratory scale. Particularly, this study was done for the case in which the target by-product was a high added-value pyrolysis gas. The dimensions of the fixed bed reactor were determined by solving the equations related to the relevant dimensionless numbers that include them as variables. Besides, three theoretical correlations were deduced to relate reactor dimensions with the flowrate of pyrolysis volatiles. This correlations serve to determine the dimensions of the reactor depending on the increase of the mass flowrate respect to the flowrate treated at laboratory scale. Regarding the modelling of the valorisation process of the pyrolysis volatiles, the RPlug tubular reactor kinetic model of the specialised Aspen Plus chemical simulator was used. Experimental data obtained in the laboratory was fed in the reactor model, while the kinetic equations of some representative reactions of the valorisation process were extracted from the literature. Kinetic simulations were run at two temperatures (700 °C and 900 °C) and the obtained results were compared with those obtained experimentally in order to determine the validity of the model. It has been concluded that, although the trend of some chemical compounds is equivalent, further work is required on the kinetic model definition, as well as in the quantitative characterisation of the composition of the experimentally measured volatiles. In addition, the valorisation process of the pyrolysis volatiles was also modelled using a Gibbs energy minimisation model reactor. This simulation approach allowed to understand the chemical equilibrium situation of the complex reaction system that takes place in the fixed bed reactor. The comparison of the equilibrium results with both the experimental

Summary XXIX

and kinetic simulation results provided significant understanding on the generation trends of some of the substances present in the pyrolysis gases and liquids. Likewise, their performance in the process in relation to their theoretical maximum performance was observed. Both scale-up calculations and reactor model simulations of the valorisation process of pyrolysis volatiles in a fixed bed reactor for the pyrolysis based recycling of FRP waste was carried out for the first time in this doctoral thesis.

Mention:
International PhD