(47a) Characterization and Impact of Oxygenates in Plastic Waste Pyrolysis Oils on Steam Cracking Process Efficiency

Names and affiliations: M. Kusenberg^1,*, S. De Langhe^1,*, B. Parvizi¹, A. Jamil Abdulrahman¹, R.J. Varghese¹, S. Ukkandath Aravindakshan², A Kurkijärvi², A. Munoz Gandarillas², J. Jamieson², S. De Meester³, K.M. Van Geem¹

¹ Laboratory for Chemical Technology (LCT), Ghent University, B-9052 Zwijnaarde, Belgium;

² Neste Corporation, Technology Centre, Kilpilahti, POB 310, FI-06101 Porvoo, Finland;

³ Laboratory for Circular Process Engineering (LCPE), Ghent University, B-8500 Kortrijk, Belgium

^*Authors M. Kusenberg and S. De Langhe are shared first authors

Mechanical recycling of mixed packaging waste has limitations due to the required separation into pure polyolefin streams which is often technically impossible and hence leads to low quality recyclates [1]. In this context, chemical recycling via pyrolysis, upgrading and subsequent steam cracking of pyrolysis oils towards valuable base chemicals such as ethylene and propylene is a promising recycling route that ultimately leads to new virgin-quality polymers [2-8].

In this joint study, Neste and the Laboratory for Chemical Technology (LCT), Ghent University demonstrated the steam cracking of distilled fractions from pyrolysis oils derived from real packaging waste using a continuous steam cracking unit. Neste, as a leading company in processing pyrolysis oils into high-quality feedstock for steam cracking, and LCT, as pioneers in academic research on steam cracking, collaborated to explore the differences in pyrolysis oils derived from the two most commonly used polymers in plastic packaging: polyethylene (PE) and polypropylene (PP). The study focused on the naphtha and diesel-range distilled fractions of pyrolysis oils rich in these polymers. Neste supplied the plastic waste-derived pyrolysis oil and shared its expertise in producing drop-in circular products for the petrochemical industry. LCT distilled the pyrolysis oil into the desired fractions, conducted feedstock analysis, and performed bench-scale steam cracking experiments. This collaboration aimed to highlight the unique characteristics of PE and PP pyrolysis oils, importance of upgrading, as well as the opportunities and challenges in their potential applications as sustainable petrochemical feedstocks.

Prior to cracking, pyrolysis oils were characterized via comprehensive two-dimensional gas chromatography (GC × GC) including normal and reversed-phase column configurations coupled to various detectors with a special focus on oxygenated components.

The results of our research show that all analyzed samples contain substantial amounts of oxygenates (1-10 wt.%) (see Table 1) and it was found that ketones are the most prevailing oxygenate group.

Table 1. Oxygenate analysis via reversed-phase GC × GC-FID.

The high concentration of ketones likely originates from secondary reactions occurring during the pyrolysis process. It can be assumed that CO, present in the pyrolysis gasses as a direct consequence of oxygen-containing impurities in the respective waste fractions, reacts with radicals present in the reaction mixture [9, 10]. In that way, a carbonyl group is introduced into the structure, with a radical carbonyl carbon. Subsequent intramolecular hydrogen transfer or recombination with another radical then results in the formation of a ketone. Due to the increased prevalence of secondary and tertiary radicals during PP decomposition, ketones derived from PP consequently exhibit a high degree of branching [11-14].

Steam cracking experiments with 20/80 weight-based blends of pyrolysis oil with conventional fossil naphtha showed slightly increased ethylene yields of the blends compared to pure fossil naphtha (see Figure 1). The PP-derived feedstocks show an overall lower ethylene yield compared to the PE-derived pyrolysis oil-naphtha blends as well as the pure fossil naphtha. This is in agreement with the findings presented in our previous studies [15, 16] as well as those reported by Hajekova et al. [17]. It can therefore be established that addition of PP-based pyrolysis oils always leads to reduced ethylene yields. In terms of the oxygenate composition of the feedstocks it can be seen that it translated into considerably higher CO yields in the steam cracking effluent. This shows that the pyrolysis oils are not feasible as steam cracker feedstocks based on the CO yields, as these would lead to off-spec products in industrial units and potential catalyst poisoning downstream of the crackers. Thus, it is shown that blending alone is likely not sufficient to alleviate the contaminant effects and hence removal of oxygen using hydrotreatment is highly recommended prior to industrial application [18-21]. Furthermore, there is a considerable increase in polyaromatic products for the PE-naphtha and PE-diesel blends, where the PE-naphtha/fossil naphtha blend yields ~3 times and the PE-diesel/fossil naphtha blend ~6 times more polyaromatic products as compared to pure fossil naphtha. When cracked at industrial scale, such an increase in heavy products and especially polyaromatics would be a reducing factor for the run-length of the cracker which confirms that there is still a quality gap between pyrolysis oil quality and efficient industrial-scale steam cracking requirements. The most effective solution to reduce heavy product formation is hydrotreatment leading to a reduced concentration of unsaturated compounds in the feedstocks.

Figure 1. Yields in wt.% of the most important products at a coil outlet temperature of 880 °C.

This work shows that distilled pyrolysis oils with conventional naphtha are promising steam cracking feedstocks if intermediate processing steps are performed to remove the oxygenates and other impurities using, for instance, hydrotreatment. With the findings of this work, such upgrading processes can be designed in a more effective way.

References

[1] Plastics Europe - Association of Plastics Manufacturers, "Plastics - The Facts 2019," 2019.

[2] M. Kusenberg et al., Waste Manage, 2022, 138, 83-115.

[3] K. Ragaert et al., Waste Manage, 2017, 69, 1, 24-58.

[4] M. Solis and S. Silveira, Waste Manage, 2020, 105, 128-138.

[5] O. Dogu et al., Prog Energy Combust Sci, 2021, 84, 100901.

[6] M. Larrain et al., J Clean Prod, 2020, 270, 122442.

[7] Z. A. Hussein et al., Int J Environ Anal Chem, 2021, 1-18.

[8] M. Kusenberg et al., Fuel Process Technol, 2022, 238, 107474.

[9] A. López et al., Chem Eng J, 2011, 173, 1, 62-71.

[10] J.-B. Peng et al., Chem, 2019, 5, 3, 526-552.

[11] Y. Zhang et al., ACS Sustainable Chemistry & Engineering, 2022, 10, 1, 91-103.

[12] M. Kusenberg et al., Fuel Process Technol, 2022, 227, 107090.

[13] H. Bockhorn et al., J Anal Appl Pyrolysis, 1999, 48, 2, 93-109.

[14] S. Ray and R. P. Cooney, "Chapter 9 - Thermal Degradation of Polymer and Polymer Composites," in Handbook of Environmental Degradation of Materials (Third Edition), M. Kutz Ed.: William Andrew Publishing, 2018, pp. 185-206.

[15] M. Kusenberg et al., Waste Manage, 2022, 141, 104-114.

[16] M. Kusenberg et al., Chem Eng J, 2023, 473, 145259.

[17] E. Hájeková et al., J Anal Appl Pyrolysis, 2007, 79, 1, 196-204.

[18] R. De Bruycker et al., J Anal Appl Pyrolysis, 2016, 122, 468-478.

[19] K. M. Sundaram and B. Stancato, "How much is too much? Feed contaminants and their consequences," presented at the AIChE 2018 Spring National Meeting, Orlando, Florida, 22-26 April 2018, 32d.

[20] A. H. Zacher et al., Green Chem, 2014, 16, 2, 491-515.

[21] R. L. Ware et al., Energy Fuels, 2020, 34, 12, 16181-16186.