Techno-Economic Review of Pyrolysis and Gasi�cation Plants for Thermochemical Recovery of Plastic Waste and Economic Viability Assessment of Small-Scale Implementation

Polymers used in the production of consumer products become a part of municipal waste streams after reaching the end of their useful lifespan, but also before even reaching markets, as rejects, scraps, and/or industry by-products as a part of industrial waste streams. Suitability for recovery of industrial wastes differs signi�cantly and needs to be analyzed separately. In this research, a review of the techno-economic parameters of existing recovery plants is done. Dependences between economic and technological parameters, sizes, and types of plants, as well as the composition of input material, are derived. Based on the presented data, a techno-economic analysis of the small-scale implementation of thermochemical recovery plants, for industry-generated residual plastic waste fraction, is conducted. Results show that thermochemical conversion of industrial plastic waste cannot be economically viable on a small scale without a gate-fee. Pyrolysis plants brake even gate-fee is on the level of over 50/86 €/t, while treating only 51%/28% of residual waste, due to strict restrictions regarding feedstock composition. In the case of gasi�cation, it is on the level of 70 €/t, while treating 92% of available waste. Pyrolysis is the only viable solution for treating up to 12 TPD after which gasi�cation also becomes a viable option. Usual capacities are up to 25/100 TPD for pyrolysis/gasi�cation, after which incineration-based technologies need to be considered. The presented results provide decision-makers with a good overview of alternative thermochemical conversion technologies, their technical characteristics, limitations, and possible economic outcomes of their implementation.


INTRODUCTION
Over the last 70 years, the production of plastic materials on a global level increased 240 times (Statista 2019a).Plastic materials replaced numerous other materials and became widely accepted.All of this has resulted in a production of 61.8 million tonnes of plastic material in 2018, on the EU level, but also to generation of 16.8 million tonnes of plastic waste in the same year (Plastics Europe 2020).
Initially, it was assumed that plastic materials are inert and harmless, but over the years it was proved not to be the case.Contrary to other non-hazardous wastes which discarded in nature either degrade with time or stay unchanged, plastic proved to have the worst of both worlds.Due to the long lifetime of polymers, they have a direct impact on the deterioration of the natural environment (Andrady 2003).Also, they can have a negative in uence on the urban environment by causing sewage system blockages (Okunola A et al. 2019) and a reduction in water percolation (Njeru 2006).On the other hand, plastic also slowly degrades, and discarded plastics, due to biotic and abiotic degradation, leads to contamination of soil and water with a wide range of additives (chlorine, heavy metals, retardants, colorants, plasticizers, stabilizers, etc.) which are added to different polymer materials during production to increase their useful properties.
Around 8 million tonnes of plastic products discarded in natural environments nally nd their way into waters contaminating them (Jambeck et al. 2015).Degradation of plastic materials in oceans is much quicker, where, at a glance, can be degraded within a year.It just shreds into tiny microplastics contaminating entire benthic and pelagic ecosystems (Barboza and Gimenez 2015), endangering various marine organisms by entering their food chains (Chiras 2004;Gregory 2009), and in the process releasing toxic chemicals like Bisfenol A (BPA) or polystyrene into water (Okunola A et al. 2019).Also, biodegradable plastics under the in uence of microbes degrade releasing methane and contributing to global warming.
To avoid the negative environmental impact of plastic, it needs to be properly collected, managed, and recovered.Plastic waste comes from post-consumption as well as production stages.While the majority of waste materials enter waste streams after reaching the end of the useful life-span of a product, and a signi cant part of it is disposed of as a part of the municipal solid waste (MSW) stream, manufacturing waste is discarded during the production process, as rejects, damaged products, residues of production raw materials and other production wastes, i.e. before the nal product is even produced.When waste generation is looked upon, it can be seen that manufacturing/industry is responsible for the generation of 10.6% of overall waste generation in the EU, and even 22.2% without major mineral wastes (mineral wastes from construction/demolition and mining activities) (European Commission -Eurostat 2020).Some industrial polymer waste types can be internally recovered (mostly easily recycled clean polymer wastes such as polypropylene (PP) and polyethylene (PE)), some become part of mixed waste streams such as construction and demolition (C&D), and others are handled by plastic recovery facility (PRF) (mostly mixed polymer wastes).In comparison with other waste types, manufacturing waste is usually clean, i.e. without foreign contaminants as it is mainly made of rejects, damaged products, and raw material residues.Thus, despite smaller quantities, it represents a better feedstock for recovery processes.
To encourage the use of different recovery technologies, they need to economically compete with existing solutions.Waste from different sources is mainly sorted and recovered by specialized third-party companies, where materials suitable for material recovery are separated, as they represent the most pro table fraction.On the other hand, residual waste needs to be separately disposed of/treated, mainly by land lling or incineration, which represents an expense.
Thermochemical recovery is one of the solutions that can be applied to many materials that are not suitable for material recovery and can be the most appropriate recovery option for plastic waste.Even though thermochemical recovery of plastic waste yields the best products and/or the most energy compared to other materials, the process can be also sensitive to some constituents of such waste.Here incineration is mainly sensitive to compounds that produce an aggressive atmosphere, like chlorine, thus polyvinyl chloride (PVC) is not suitable.Pyrolysis can be also affected by many contaminants, including chlorine, but the majority of potential problems can be mitigated by the use of different catalysts and/or separation techniques from the gaseous or liquid phase, which increases the cost of the treatment.On the other hand, gasi cation can accept a wide range of input materials with appropriate technology, and some plants are even specialized in treating PVC (Tukker et al. 1999).Thus, all thermochemical conversion technologies require a good feedstock separation process (Business Europe 2019).Thermochemical recycling techniques like pyrolysis and gasi cation can be used for the production of fuels, chemicals, and materials.Depending on the use/transformation of primary products, fossil resources can be replaced with recycled resources obtained from plastic waste and alleviate the problems of dependence on petroleum as raw material, decouple prices of fuels, petrochemical products, and plastics from the oil prices, as well as reduce impacts on the environment, which is all in line with overall EU legislation (Tomić et al. 2022b).Due to this, pyrolysis and gasi cation of solid plastic wastes have been identi ed as one of the most important contributors to solving the problem of solid plastic waste (SPW) disposal within the circular economy (CE) approach (Dogu et al. 2021), and a way to close the loop in entire energy and material recovery mindset (Tomić andSchneider 2018, 2021).
Thermochemical recycling technologies based on pyrolysis and gasi cation due to their robustness and good economics are leading the way, but more research is required to seize the potential of these technologies primarily as a means to increase the e ciency of use of valuable natural resources (Dogu et al. 2021).Thus economic analyses of pyrolysis and gasi cation treatments of plastic waste were carried out in numerous publications.In (Fivga and Dimitriou 2018), Aspen was used for modeling speci c pyrolytic plants with speci c feedstock composition to carry out economic analysis.Similarly is done in (Westerhout et al. 1998) and (Larrain et al. 2020) for the average composition of mixed plastic waste, where in (Westerhout et al. 1998) differences between three different processes are also analyzed.In (Tomić et al. 2022a) economic indices of thermochemical conversion technologies are compared with results for other technologies.In these publications literature review data are used for system and economy modeling.The same approach is used for the analysis of the economy of speci c plastic waste to energy (PWTE) facilities from an Australian perspective (Ghodrat et al. 2019), but no focus was put on feedstock composition at all.Thus, it can be seen that each publication puts focus on speci c waste, speci c situations, and/or speci c technology, whereas only (Westerhout et al. 1998) analyzed more different pyrolysis technologies.A similar approach is used for the analysis of gasi cation-based plastic waste recovery.Thus in (Mazzoni and Janajreh 2017), Aspen is used for modeling municipal waste gasi cation plants and analysis of the in uence of variations in the share of plastic.
Economic analysis of a cogeneration gasi cation plant, based on speci cations from one equipment manufacturer, is conducted in (Arena et al. 2011) and difference in production when treating two different municipal plastic wastes (MPW) is reported.However, in the case of gasi cation, more analyses put focus on bio-waste and municipal waste (Yang et al. 2021).Thus in (Salkuyeh et al. 2018) techno-economic analysis of hydrogen production is assessed, and in (Thunman et al. 2019) biofuel production is analyzed, but in both analyses focus was put on the treatment of the bio-waste component.On the other hand, the municipal solid waste economy is analyzed in (Luz et al. 2015), where also cost of commercialization and the amount of incentive that needs to be paid by the government for waste treatment to achieve pro tability is assessed.It was concluded that government nancial support is essential to reach pro tability as well that larger capacity plants are needed to achieve better economic results.A recent review on co-gasi cation and co-pyrolysis concluded that there is a need for more techno-economic feasibility analyses for the large-scale adoption of such energy technologies (Mariyam et al. 2022).Also, due to renewed focus on hydrogen as an important link in energy transformation, recent publications are focusing on hydrogen production through thermochemical conversion of plastic waste (Wijayasekera et al. 2022;Al-Fatesh et al. 2023).
As in previous research, an economic viability of the use of pyrolysis in the treatment of plastic wastes has been assessed based mainly on speci c, modeled, cases, thus, results cannot be easily used for assessments in other situations, so in this research a techno-economic analysis is conducted from a standpoint of different real plastic waste pyrolysis plants.Regarding gasi cation, a gap in research focused on plastic waste is identi ed, as gasi cation is commonly used for the treatment of waste fractions that are generated/collected in bigger quantities.
In this research, a detailed review of different, existing, pyrolysis and gasi cation plants, with a focus on the treatment of plastic wastes, is done.By this, techno-economic differences between various available technologies from different suppliers are reviewed.A nities towards different polymer materials, but also other materials, trends in production, capacities of plants as well as the differences in investment and operating costs in dependence on used technology, size, and production capacity are reviewed.This way, this analysis takes a step further from previous research where pyrolysis and gasi cationbased treatments are modeled based on speci c equipment producer-obtained data.Thus, this approach to techno-economic analysis, as well as to presenting results, can prove to be a valuable tool for decision-makers and business owners when planning for possible alternative solutions for residual plastic waste treatment options which would reduce land lling and incineration, reduce business entity exposure to market volatility, and/or decrease waste disposal/treatment costs which have been increasing rapidly in recent years.Because the focus is put on individual waste treatment/recovery companies, this research focuses on the small-scale implementation of such technologies which has also been overlooked in previously published research.Such focus on small-scale plants is even more emphasized through the presented case study in which derived results of the techno-economic review are applied to the business case of an industrial plastic waste sorting plant treating waste of very speci c composition.Even though residual waste from this kind of plant favors less demanding technologies like incineration, it is shown that even in these circumstances pyrolysis and gasi cation can be economically pro table when taking into account actual disposal/treatment fees.

REVIEW OF TECHNO-ECONOMIC DATA FOR EXISTING PLANTS
In this research, data from 30 different pyrolysis plants and 22 gasi cation plants are reviewed.The majority of reviewed plants with available data, report suitability for treating MPW fractions which are reported as a part of the The American Society for Testing Materials (ASTM) International Resin Identi cation Coding System (RIC) system -polyethyleneterephthalate (PET), polystyrene (PS), high-density polyethylene (HDPE), low-density polyethylene (LDPE), PP, PVC, and mixed polymer fraction (Other 7).Not all technologies accept all MPW fractions and acceptability of other materials is denoted (mainly in the case of pyrolysis plants) like polyamide (i.e.nylon), acrylonitrile butadiene styrene (ABS), but also rubber, and other materials.

Techno-economic data overview -Pyrolysis
Table 1 presents an overview of feedstock composition and properties requirement data for reviewed pyrolysis plants.It can be seen that all analyzed plants accept HDPE, LDPE, and PP, as pyrolysis of these materials results in higher quality products and high fuel yield.A slightly less common is pyrolysis of PS, which is accepted by about 80% of plants, and Other (7) plastics, which are accepted by almost 60% of the analyzed plants.Reported pyrolysis yields of these individual types of polymers are between 80-90% for PS (The American Chemistry Council 2015) and 60-80% for PE (Arnold 2011).
On the other hand, review data presented in Table 1 show that PET and PVC in many cases are not accepted in input feedstock, while plants that accept these types of plastics usually put restrictions on their share in the input material.One of the reasons for such restrictions is the fuel yield that is only on the level of 30%, which is signi cantly lower when compared to other polymers (Arnold 2011).During pyrolysis, a large share of PET ends up in the form of a carbonized residue, which represents a problem for heat transfer in the pyrolysis chamber as well as for maintenance due to increased deposits on the walls of the exchanger.In addition, pyrolysis of PET can lead to unwanted combustion/oxidation due to the release of oxygen (Arnold 2011) and also leads to the production of polycyclic hydrocarbons and biphenyl derivatives that are harmful to human health and the environment (Park et al. 2020).On the other hand, PVC contains chlorine which is converted into hydrogen chloride gas and dioxins that cause corrosion of the pyrolysis plant itself, but also of the combustion system in which the nal product will be used as a fuel (The American Chemistry Council 2015).Most pyrolysis systems are designed to capture and remove hydrogen chloride and other acids that may be present due to additives in the waste plastic feedstock, but, despite this, the use of materials that contain these compounds is usually avoided.
From other materials that are less commonly mentioned in technical requirements for the plants, there are nylon, i.e. polyamide materials, which are also not wanted in feedstock material in all analyzed facilities, ABS is acceptable in few plants, and some plants restrict its presence to some degree, and rubber is classi ed as part of feedstock material in two facilities.Also, some plants accept some other materials which are separately denoted.Regarding other materials, some datasets emphasized the quality of separation of input material from other contaminants, but that is a common practice for all plants, even if it is not mentioned, and a degree of feedstock quality is required.
Pyrolysis plants require more rigorous pre-sorting to isolate useful feedstock materials, depending on technology parameters, and different plants also have different particle size requirements, thus shredding of sorted waste, where same require particles as little as 6 mm.Analysed technologies have bigger or smaller requirements for maximum moisture content of the raw material which is allowed, where the requirement of less than 1% is not unusual for pyrolysis plants.Also, before the raw material enters a pyrolysis reactor, some technologies require the plastic to be heated and/or partially melted.
Capacities of reviewed pyrolysis plants are shown in Fig. 1.As it can be seen, typical capacities of pyrolysis plants are up to 25 tons per day, although there are also plants with a capacity of over 70 tonnes per day (TPD) (like P11 and P21).
When speci c technology supplier data is reviewed, it can be seen that company Blest offers plants that process even less than 1 tonne per day of plastic waste, Polyfuel and Pyrocrat Systems LLP (P24 and P29) plants that process 3, 6, and 12 TPD (The American Chemistry Council 2015), and Klean Industries (P30) even plants with a capacity of up to 150 tons per day (The American Chemistry Council 2015).
Plants covered by the literature review use different technologies and also differ from each other according to the obtained nal products.Table 2 shows the production of each plant as it is reported, as well as overall production of liquid and gaseous products.As it can be seen, regardless of the pyrolysis technology used, most plants have pyrolytic oil as a nal product (like in the case of plants P1, P2, P8, P9, P15, P16, P22, P23, P26, P28, and P29), which, as such, is put on the market as a semi-product which in majority cases can be used as a crude oil substitute.On the other hand, if there is two (or more) stage condensation in place, lighter and heavier fractions can be obtained separately (like in the case of P3, P6, P7, P17, P18, P25, and P27).Also, through the further distillation process, nal products such as diesel and gasoline can be obtained inhouse (like in the case of P10, P11, P12, P13, P14, P18, P19, P20, P21, and P24), or in a separate facility.In addition to liquid product, which is in most cases the main product of the pyrolysis process, some plants have gaseous and solid fractions (pyrolytic gas and char) as additional products, which are usually used on-site for covering internal energy needs.On the other hand, non-condensing gases can be also used in other technologies/industries, but that is less common (The American Chemistry Council 2015).T thermal pyrolysis From data presented in Table 2, it can be seen that the quantities of liquid and gaseous fractions are of the same order of magnitude regardless of the different plant designs.Maybe the greatest difference is in the use of catalysts, thus, some plants use thermal while others use catalytic pyrolysis to obtain nal products.Nevertheless, there is not much difference in nal production, and average production for thermal pyrolysis plants (per tonne of input feedstock) is on the level of 754 kg of liquid products and 136 kg of gaseous products, while for catalytic pyrolysis plants on the level of 766 kg of liquid and 116 kg of gaseous products.For the calculation of average gaseous production, zero values are not taken into account, as it is assumed that gaseous production is used for covering internal energy needs, and thus, is not always reported.
The amount of catalysts used during the catalytic process is up to 40 kg per ton of waste plastic (Yu et al. 2018).It is important to emphasize that quantity, as well as type of catalyst, depends on the quality and composition of the feedstock.
Also, the use of catalysts in individual plants can be optional, depending on input material composition.From the data presented in Table 1, it can be seen that catalysts are mostly used in facilities that also recover additional, speci c, types of feedstock, like polyamide, rubber, ABS, polyurethane (PUR), ethylene-vinyl acetate (EVA), etc.
Besides pyrolysis oil and non-condensing gases, there are also less wanted solid, carbonized, residues that can be also used to cover part of the energy needs of the facility via incineration, but also can be used in road construction, or as a covering material, or used as an electrode ller (The American Chemistry Council 2015).
Techno-economic data overview -Gasi cation Regarding the acceptance of other materials, polymers are accepted in most cases, while for non-polymer materials not even one data source stated that they are not accepted, but some conditions can be put in.
Thus, when compared to thermal treatment technologies which have very few requirements regarding input material composition (mainly separation of chlorine-containing components and heavy metals (Ramos Casado et al. 2016)), pyrolysis plants require more rigorous pre-sorting to isolate useful feedstock materials, depending on technology parameters, and gasi cation plants usually accept a much wider range of material without major drawbacks.Different plants also require shredding of sorted waste, where in the case of pyrolysis particles as little as 6 mm can be required.Analyzed technologies have bigger or smaller requirements for maximum moisture content of the raw material which is allowed, whereas in some gasi cation plants, this limit is set as low as 0.6%.This is the result of the fact that water in the raw material not only increases the consumption of energy during the process itself but also affects the quality of the nal product as water carries additional unwanted oxygen atoms (Basu 2018).
Capacities of the analysed gasi cation plants are shown in Fig. 2.
As it can be seen from data presented in Fig. 2, typical capacities of gasi cation plants are 20-100 tons per day, although there are also plants with smaller capacities, as small as 7.2 tons per day (G16), as well as plants with a capacity of over 300 (G20 and G21) and up to 750 TPD (G19).
When compared to the capacities of pyrolysis plants, gasi cation plants' capacities are larger, and can reach incinerationbased municipal waste recovery plants which typically process more than 200 TPD, and up to over 2200 TPD (The American Chemistry Council 2015; Tomić et al. 2016).Thus, while pyrolysis can be better used for more speci c cases, to treat smaller quantities of waste with speci c composition, gasi cation plants can be also used for wider-scale waste treatment, especially when smaller sensitivity on input waste composition is taken into account.
Different reviewed plastic waste gasi cation technologies also differ from each other by obtained products.Even though the primary gasi cation product is synthesis gas, which is, for the most part, a mixture of H 2 , CH 4 , CO, CO 2 , some other lighter hydrocarbons, and N 2 (mainly from a gasifying agent), in the majority of gasi cation plants it is converted to other products and/or energy carriers in post-processing, which can be better marketable.As there is no market for synthesis gas per se, in reviewed plants, it is converted into hydrogen, methanol, ethanol, diesel, gasoline, electricity, and heat, depending on the individual plant.Reported production in reviewed plants can be seen in Table 5.As can be seen from data presented in Table 5, the majority of reviewed plants produce electricity, as it is the most marketable product.From reported data on plants that produce electricity as a sole (useful) product, a clear distinction in production between plants that treat (mostly) polymer-based input with low moisture content (G8 -G11) and those that also treat other MSW fractions (G18 -G22) can be seen, as electricity production of the rst group is signi cantly greater.Also, the difference in gasi cation agents cannot be ignored as the rst group uses a steam/air mixture while the second one uses only air.
Air is the most commonly used agent because it is cheaper and simpler to use compared to oxygen.Nitrogen has a diluting effect on synthesis gas and gasi cation by air leads to the production of synthesis gas with a lower calori c value, while gasi cation with pure oxygen yields cleaner gas with a higher calori c value.The use of oxygen-enriched air reduces the dilution effect of nitrogen, and increases the content of other gases such as H 2 , CO, CO 2, and CH 4 and consequently increases the heating effect of synthesis gas (Salaudeen et al. 2019).Steam (water vapor) and carbon dioxide are gasi cation agents used to produce synthesis gas containing higher proportions of H 2 and CO.Steam gasi cation stimulates steam reforming of soot and hydrocarbons, increases the quality of synthesis gas, and increases its calori c value.The use of CO 2 increases the concentration of CO in synthesis gas, which is useful for gas treatment with Fischer-Tropsch synthesis to obtain liquid fuel.
All of these technologies use some additives and catalysts for tar removal and gas cleaning.This is done by the addition of quicklime (calcium oxide), dolomite, olivine, activated carbon, and nickel-based catalysts directly to gasifying bed, and/or through secondary catalytic or thermal cracking, as well as mechanical methods, e.g. using cyclones, electrostatic precipitators, bag lters, ceramic lters or water washers, and high-temperature gasi cation with temperatures over 1000°C (Hayashi et al. 2014).Also, when treating chlorinated raw materials, produced hydrogen chloride (HCl) must be removed from the gasi cation product.HCl can be bound by dolomite addition (reacts with CaO and MgO and forms CaCl 2 and MgCl 2 ) or by the addition of sodium carbonate (Na 2 CO 3 ) (Salaudeen et al. 2019).
Polymer gasi cation only related problems encompass the softening of plastics and ne black powder formation which is deposited on inlet pipe and reactor walls.These problems can be alleviated by the gasi cation of mixed materials, i.e. cogasi cation, of plastics with bio-waste or coal, which can also increase syngas production (Hayashi et al. 2014).

ECONOMY OF PLASTIC WASTE TREATMENT
The economy of the speci c thermochemical conversion technology is determined by the used feedstock, its available quantity (i.e.capacity of the plant), used technology (technical requirements, expenses, and production quantity and quality), and local in uences (tax policy, construction costs and land prices).

Economy of pyrolysis
Feedstock has a direct in uence on the quality of pyrolysis products, their production, and the overall distribution of production.While quality is in the majority of cases addressed through the use of appropriate catalysts, overall production and distribution of products still shows a signi cant dependence on feedstock composition.Thus, four main different groups of industrial pyrolysis plants, by the composition of the processed feedstock, can be identi ed -pyrolysis plants that process entire MPW, plants that process MPW but limit the quantity of PET and PVC, plants that are focused on the treatment of polyole ns, and plants which also treat PS and ABS next to polyole ns.The average production of these three groups of industrial pyrolysis plants is shown in Table 7. From data shown in Table 7, it can be seen that the quantity of liquid products increases with the increased separation of different MPW fractions.This increase is on the level of 6.2% with the limitation of PET and PVC in feedstock material, and 6.5% when only polyole ns, PS and ABS are treated, while liquid production is boosted by 13.9% when accepted fractions are reduced only to polyole ns.
Due to technological differences between plants that produce semi-products like pyrolytic oil, and nal products, in this case, distilled fuels, economic functions for capital and operating costs of these two types of plants are separately collected.The dependence on capital costs of the technology and capacity of the plant is shown graphically in Fig. 3.
As it can be seen from Fig. 3, interpolation of investment cost dependences is done separately for plants that produce pyrolytic oil as a product, and the ones that produce fuel distillates (i.e.diesel, petrol, kerosene, etc.).Additional distillation column cost, for the production of nal products (fuels), increases overall investment costs of plastic pyrolysis plant.On the other hand, in the chart in Fig. 3, data for plants that use catalytic pyrolysis are denoted by darker markers than those that use the thermal pyrolysis process.Thus, it can be seen that catalytic pyrolysis is used mainly for smaller plants.In both cases, linear regression is used for the mathematical modeling of the dependence of capital costs and capacity of the plant as it shows lowest coe cient of determination (R 2 ) value.
On the other hand, the dependence of operating costs, by quantity of produced liquid fuel, and size of the pyrolysis plant is shown in Fig. 4.
Regarding operating costs data shown in Fig. 4, only one data set for the plant with the production of distilled fuels is obtained, but it shows that, like in the case of investment costs, they are higher for the plant with the production of distilled products.The modeled cost function shows the in uence of the economy of size, where speci c cost decreases with the size of a plant, which can be seen from the exponential function which describes the dependence of operating costs and capacity.
Presented functions incorporate all costs of the plant including utilities.Other operating costs that are not incorporated in these curves are costs of feedstock, feedstock pre-treatment as well, and the cost of disposal of residual waste materials.These costs are dependent on the speci c case study and used business model.

Economy of gasi cation
Regarding the in uence of feedstock on gasi cation, it has a direct in uence on the quality/composition and quantity of produced synthesis gas, together with the used gasi cation agent and type of used equipment.As it is the most common way of the use of generated synthesis gas, and due to the lack of data for modeling the economics of production of other nal products, electricity generation is used as the nal product.Also, due to the best results and as the most utilized, this research focuses on steam/air mixture as a gasi cation agent.Thus, it can be seen that average production of electricity in the case of thermal co-gasi cation of plastic waste and biowaste (wood) leads to production of 1230 kWh of electricity per tonne of input material, gasi cation of mainly plastic waste mixture with low moisture content 1540 kWh/t, and gasi cation of sorted PE, as mono-material feedstock, to 1640 kWh/t.From this, it can be concluded that the increase in polymer content and share of polyole ns increases the heating value of synthesis gas and, thus, electricity production.
Due to technological differences between plants that produce different nal products, economic data for capital and operating costs of these two types of plants are separately collected.The dependence on capital costs of the technology and capacity of the gasi cation plants is shown graphically in Fig. 5.In this case, the power function is used for mathematical modeling of dependence of investment costs and capacity of plant which shows little in uence of the economy of size on investment cost.
Regarding the operating cost, for the gasi cation plants with electricity production, it is modeled through the literature data (Thunman et al. 2019;Porcu et al. 2019) as 6.5% of investment cost and incorporates all costs of the plant including utilities.Other operating costs that are not incorporated in these curves are costs of feedstock, feedstock pre-treatment as well, and the cost of disposal of residual waste materials.These costs are dependent on the speci c case study and used business model.
As previous diagrams only show technology cost, the cost of construction of a building suitable for the installation of corresponding technology equipment needs to be analyzed separately.Regarding the speci c cost of construction of industrial buildings varies according to the speci c location, the dependence of price and gross domestic product (GDP) per capita of selected location is identi ed and shown in Table 7. Regression analysis carried over the presented data is conducted in Fig. 6.As can be seen from data presented in Fig. 6, there is a high degree of matching of interpolated curves with gathered data on light-duty factory and heavy-duty factory building space in different places in the world.Thus, for general analyses, the price of construction of industrial buildings can be approximated through the use of these interpolated equations and the GDP per capita of selected locations.

CASE STUDY RESULTS
All plastic waste needs to be sorted and prepared for material recovery, but this process results in signi cant amounts of residual waste that need to be treated or disposed.Usual options for this type of waste are land lling and incineration.Due to limitations that are put on land lling, decreasing plastic waste exports and limited waste incineration capacity cause volatility of these options gate-fees.While building new incineration capacity can be a solution for this problem on a large scale, on a small-scale alternative options need to be found.This limitation is emphasized in the case of local plastic industry waste management where residual waste amounts are much smaller when compared to amounts generated through municipal waste collection system.In this context, alternative energy recovery technologies can prove to be better alternatives to incineration.
To further analyze this, the composition of residual waste from industrial waste sorting facilities is analyzed and the feasibility of investment in small-scale pyrolysis and gasi cation plants that could process of it, thus reducing disposal and land lling, is analyzed.This also enables individual waste sorting facilities to reduce their exposure to market uctuations regarding waste disposal/treatment fees, which are expected to rise as land lling is banned by EU legislation and plastic waste export in third countries heavily reduced.plastic waste is delivered to analyzed industrial plastic waste treatment rms from many different industries in the area, depending on the individually signed contracts.Each delivered load is manually inspected, sorted, and, depending on its main component, undergoes a further mechanical separation process to recover high-value materials for material recycling.At the end of each separation stage, residual waste that does not meet the criteria for material recovery is collected in a warehouse for nal disposal.This residual waste material is sampled as a potential feedstock for the thermochemical recovery.
Industrial residual waste samples were sampled during one workday in November of 2019, from residual material warehouse which has capacity for three months of usual operation.Samples were taken from 12 different spots in the nearly full residual material warehouse, paying attention that each sample is produced in a different timeframe, thus, sampled material represents the median sample of residual industrial waste from industrial waste recovery rm during last three months.After homogenization and quartering of the overall collected residual industrial waste sample, a sample of 1,828.1 grams is obtained and transported for further characterization and analyses.
Taken samples are manually separated on-site using material codes on labels and preliminary examinations through the use of researcher experience (by type and physical properties), ame tests (ease of ignition, burning rate, color of ame, and odors), melting range, or speci c gravity.Based on the obtained results, samples are sorted by constituents.Obtained results are con rmed through the use of Fourier Transform Infrared (FTIR) Spectroscopy analysis (using Perkin Elmer Spectrum One analyzer) by Attenuated Total Re ectance (ATR) technique, due to possible errors of on-site characterization and separation.
The results of the characterization are shown in Fig. 7.
According to information from industrial plastic waste separation facility operators, at full capacity plant generates around 10 TPD of residual waste after separation.From composition results shown in Fig. 7, and polymers found in residual waste after separation, it can be seen that only some of them are commonly used as feedstock in previously reviewed pyrolysis plants (PP, PE, and PET ), while in the case of gasi cation plants also ABS and polycarbonate (PC) are commonly accepted even though not speci cally reported.Other materials are not taken into account as it is stated that they are not accepted, or needed data for the analysis are not found during the literature review.Thus, in this research, the economic feasibility of implementation of small-scale pyrolysis and gasi cation plants, based on common technologies, is investigated in the case study of the residual waste stream from one industrial plastic waste separation facility.As usual ways of disposing of such waste are land lling and incineration, economic analysis will be done in a way that results of the analyses will represent a break-even gate fee of such a plant, i.e. results will show whether can it be economically more pro table to build pyrolysis or gasi cation plant in comparison to paying land ll or incineration plant gate-fee.Even though analysis will be done on an example of a case study, economic functions are derived in a way that they can be used for carrying out pre-feasibility studies anywhere in the world.

PLANT MODELLING BASED ON REVIEW RESULTS AND CASE STUDY NEEDS
Analyzed thermochemical recovery plants are modeled as a way of reducing the cost of disposal for the case study plant.As the vast majority of reviewed pyrolysis plants are processing polyole n wastes or mixed MPW, an analysis of these two cases will be conducted.Thus, Table 8 shows the techno-economic results of a pyrolytic plant that processes only the most pyrolysis-compatible feedstock materials (only polyole ns), as well as when a mixture of all MPW is treated.As it can be seen from Table 8, even if an industrial waste separation plant generates around 10 tonnes of residual waste per day, taking into account the composition of residual waste from the case study plant and the preferred feedstock composition of analyzed plants, treated quantities of waste are much smaller.In the case where only polyole ns are treated, pyrolytic plant with a capacity of just over 3 TPD is needed.In this case, only 27.6% of available residual waste is treated, as that is the share of polyole ns in the residual industrial plastic waste fraction.The use of these materials as feedstock leads to the biggest production of liquid products, but the lower capacity of the plant leads to increased costs.In all analyzed cases, pyrolytic oil is considered to have properties of crude oil and thus can be sold for 67 $/per barrel (Oilprice.com2021).Regarding carbonized residue production, it is calculated so that the balance of the output mass equals the input to the plant.
When other MPW fractions are also treated, taking into account composition of residual waste from the case study plant, 51% of all generated residual plastic waste can be used for fuel production and the capacity of the plant needs to be increased to handle 5.6 TPD of feedstock.Using all MPW fractions, including PET, in input feedstock for a pyrolysis plant decreased speci c production of pyrolytic oil, but also, due to economy of scale, operating costs have decreased.
In the case of modeled gasi cation treatment technologies, the input waste stream consists of a much wider range of materials -polyole n, PET, ABS, and PC.Thus, Table 9 shows the techno-economic results of the gasi cation plant with electricity production.Needed building oor area (equipment and warehouse) 1,000 m2 In the case presented in Table 9, it can be seen that gasi cation plant accepts 91.8% of all generated residual waste from the case study plastic waste separation plant.Thus, gasi cation plant is modeled as a plant with a capacity of 9.2 TPD.The use of this technology leads to greatly increased investment costs, but at the same time reduced operating costs.

Economic results
Table 10 shows the total costs and incomes for two presented cases of pyrolysis plants, as well as calculated NPV in the case where no additional waste treatment cost is charged, i.e. no gate fee.Results are calculated with a discount rate of 9% based on the weighted average cost of capital in the chemical sector and sectors of oil and gas (2021), income tax of 10%, and 10-year amortization period.As it can be seen from Table 10, in both cases these investments are not economically viable without some other income.As an industrial plastic waste sorting facility usually pays for the disposal of its residual waste, investment in a pyrolysis plant can be pro table for such a rm if it can reduce its disposal cost.Thus, the break-even gate fee for both analyzed pyrolysis is calculated under the same conditions.As can be seen, in the rst case, the gate fee which gives zero value of NPV is on level of 89.45 €/t, and in the second case minimum needed gate fee is 50.92 €/t.Thus, the viability of investment mainly depends on the price of alternative disposal of residual plastic waste.
Regarding the case of gasi cation as the nal treatment step for residual waste, economic results for this scenario are shown in Table 11.Presented results in this case are also calculated for cases where a zero gate fee is charged for waste treatment.Regarding the case which economic data were presented in Table 11, even though the capacity of this plant is signi cantly larger, investment is also not economically viable without a waste treatment fee.Thus, the break-even gate fee in this case is on a level of 69.80 €/t.Thus, also in this case, the viability of investment depends on the prices of alternative disposal options.

DEPENDENCE OF DISPOSAL PRICE AND
The price disposal is mainly connected to the local availability of different disposal/treatment solutions.In this case study, the local price for disposal of residual plastic waste reaches up to 80 €/t, while the price for thermal treatment is not publicly available information as it is determined by individual contracts and depends mainly on the characteristics of waste materials that are to be treated.Locally, there is a shortage of waste incineration capacities, thus, incineration prices, per information from local company contact, can reach, and even pass, 100 €/t mark.On the other hand, local companies that try to use the capacity of Scandinavian waste incinerators, which are oversized and therefore do not have enough locally generated waste for operation face the problem of high transport costs.In this case, transportation costs reach 1,800 € per 22-ton truck, which leads to an additional cost of over 80 €/t on top of treatment plant gate fee.
On a broader level, gate fees for the treatment of such waste can cover a signi cant price range.The cost of thermal treatment of refuse-derived fuel (RDF) in cement kilns can range between 40 and 50 €/t in the peak of demand for such treatment (RPS 2014; Government of India 2018), while the price usually ranges from − 20 to + 20 €/t (Schäfer and Moser 2012), where negative price makes sense due to savings enabled by switching from primary fuel (coal) to waste, which saves up to 50 €/t (EcoMondis 2018).Thus, 40 €/t can be taken as a current gate fee, in a situation where the export of plastic waste to Asian countries is heavily reduced.
Thus, three ranges of gate fee acceptability can be identi ed, up to 40 €/t which can be considered for a plant that will be economically viable in a wide range of cases, from 40 to 80 €/t, which can be acceptable in the situations with limited disposal options, and over 80 €/t which can be considered economically viable only in rare situations and in the future if disposal costs increase.Another in uencing factor is the size of the plant which in uence can be seen in the majority of modeled economic functions.Taking these two in uencing factors into account, sensitivity analysis of the pro tability of previously modeled pyrolysis plants is conducted and shown in the following gures.
In the case where only polyole ns are treated, liquid pyrolytic oil production is the highest, thus, for the same gate fee, this kind of can be pro table on a smaller scale.As it can be seen from analysis shown in Fig. 8, in the case of the gate fee of 40 NPV higher than zero for plants with a capacity greater than 5 TPD.On the other hand, in the case of 80 €/t gate fee, it is possible to reduce capacity to around 3.2 TPD and retain pro tability.As can be from the presented data, even in the case where a pyrolysis plant processes only the most suitable polymer wastes, its conditional costeffectiveness can be hard to reach when a small amount of feedstock is available.
Thus, it can be seen why the majority of reviewed pyrolysis plants treat whole MWP with only some restrictions that are dependent on a speci c technology.In that case, higher amounts of feedstock are available, especially when MSW is treated where the MSW collection system collects large quantities of waste which are then recovered in material recovery facility (MRF).This increases the quantity of residual waste fraction available for recovery, which then increases the economic pro tability of pyrolytic recovery facilities.Thus, these kinds of facilities are more suitable for the treatment of MWP of known composition in quantities of over 6 TPD, or in rare situations over 4 TPD.
If there is a need for treatment of other waste plastic materials that are not part of the RIC classi cation, gasi cation represents a better alternative as it is less sensitive to changes in waste composition.When accepted polymers are looked upon, gasi cation plants accept a wide range of materials, which is on a level of incineration-based plants, i.e.only chlorine is in some cases more strictly avoided in the input mix.Even though no gasi cation plant is found with a capacity of under 7.2 TPD, this increases available feedstock for treatment and thus can make gasi cation applicable technology even for smaller-scale applications.Because of that, sensitivity analysis is conducted and pro tability of the electricity-generating gasi cation plant in dependence of gate-fee and capacity, starting from a capacity of 6 TPD, is shown on Fig. 10.
Even though it puts the smallest restrictions on feedstock composition, gasi cation is not so economically sound option on a small scale as its viability is achieved only on higher capacities.In Fig. 10, it can be seen that, in the case of a gate fee of 40 €/t, which is a pro tability limit, NPV is higher than zero for plants of capacity greater than 12 TPD, and if the gate fee is 80 €/t it is possible to reduce capacity to around 8.6 TPD.
Thus, it can be seen that the majority of reviewed gasi cation plants, in the majority of reviewed gasi cation plants, apart from plastic waste, also some other types of bio waste can be treated, as well as different MSW fractions.This increases the available quantities of feedstock, which is the most important when looking at gasi cation as a treatment option.Where existing pyrolysis plants treat up to 80 TPD of polymer waste, with a usual capacity from 1 to 25 TPD, reviewed gasi cation plants have a usual capacity from 7.2 to 100 TPD, and can reach 750 TPD which is in the range of incineration-based waste treatment plants, and also do not have such strict composition requirements.

CONCLUSION
From the presented review of thermochemical conversion plants and techno-economic analysis, it can be concluded that the economic viability of individual thermochemical transformation technologies of plastic waste is much more dependent on the economy of scale than on feedstock composition, but which technology to use is directly connected to the composition of potential feedstock.
A review of technology data for existing plants shows that pyrolysis plants accept only some of the polymers used in production.These are mainly MPW polymers which are labeled as a part of RIC systems, but there are also some restrictions.Most commonly, severe restrictions are put on the share of PVC and PET in feedstock material.If there are other waste materials as a part of feedstock, pyrolysis is not the best choice as review results show that existing plants put some strict limitations regarding their share, or simply do not accept them at all.Thus, plastic waste intended for pyrolysis needs to be of known composition with little contaminants, i.e. it needs to be quality sorted.On the other hand, gasi cation accepts a wide range of polymer materials as well as bio-waste such as wood.The only restrictions that some plants put are on the share of PVC.Thus cation is more suitable as a waste treatment option as there fewer requirements on feedstock quality.Both technologies need some additional As it is reported, feedstock needs to be shredded (for some pyrolysis plants even partially melted), dried, and sorted to reduce inorganic content.
While pyrolysis is a much better option for treating speci c waste fractions, its use is restricted by the quantity of those materials.It is the most suited for the treatment of polyole ns only, but treating mixed MPW fractions reduces liquid production by only 12.2%, which is more than compensated by an increase in the quantity of treated feedstock by 85%.This left the open question of how to treat the rest of the material, which also needs to be treated/disposed of, and represents 72%/49% of overall residual waste.
On the other hand, due to low restrictions on feedstock composition, gasi cation can treat almost all produced residual material, thus boosting available feedstock quantity, once again, by 80%, and recovering 92% of the entire available residual waste.In this case, when treating plastic waste mixture with low moisture content, high synthesis gas yield is achieved.If a gasi cation plant is used for co-gasi cation of plastic with wood waste, its production would decrease by 25%, as can be seen by plant data analysis.
This selectivity of technologies regarding their feedstock composition is mirrored in plant size, where technology's usual capacity follows a decrease in selectivity.Thus, for capacities of up to 10,000 tonnes per year on average pyrolysis imposes itself as a go-to technology, gasi cation if available feedstock quantity is up to around 35,000 tonnes per year, and for higher capacities incineration should be also taken into consideration as a treatment technology.This is also shown in economic results, where with a gate fee of 40 €/t, pyrolysis can achieve pro tability with capacities greater than 5/6 TPD (depending on feedstock composition), whereas gasi cation achieves pro tability after 12 TPD.In the future, with an increase in gate fees of alternative treatment options, the 80 €/t limit decreases needed capacities to achieve pro tability by around 2 TPD for pyrolysis technologies, and 3.5 TPD for gasi cation.
The presented results provide decision-makers with a good overview of alternative thermochemical conversion technologies for residual plastic waste, their technical characteristics, limitations, and possible economic outcomes of their implementation.This can help in steering potential investors, mainly individual plastic waste separation and processing rms, to make the correct decision if they want to decrease their residual waste disposal/treatment cost and/or take volatility in changes in disposal prices out of economic calculations, especially in time when increase in overall gate-fee prices are expected.In future work, this analysis will be expanded on incineration-based technologies to give a holistic view of alternative recovery options, as well as to take into account MPW from MSW streams.Tables Table 1 and 3 are available in the Supplementary Files section.

Figures
Figures

Figure 2 Capacities of reviewed gasi cation plants Figure 3
Figure 2

Figure 9 Economic
Figure 9

Table 2
Production of reviewed pyrolysis plants

Table 3
(Wilk and Hofbauer 2013;Bai et al. 2020sition and properties requirement data for reviewed gasi cation plants.It shows that the majority of analyzed gasi cation plants accept all polymer wastes, while only some of them are specialized for treatment of only particular polymer fractions -plant G8 of only waste polyethylene, G9 waste polyole ns, G16 waste polyole ns, and PVC and G17 only waste PVC.From data presented in Table3, it can be seen that HDPE and LDPE are accepted in all reviewed gasi cation plants except for one that is focused on PVC treatment.Also, PP is accepted in over 90% of gasi cation plants.This is expected due to the reported compositions of synthesis gases.Steam gasi cation of PE generates synthesis gas with over 60% hydrogen (H 2 ) (Erkiaga et al. 2013), while gasi cation of PP leads to the generation of synthesis gas composed of 40% of methane (CH 4 ) and 33% of H 2(Wilk and Hofbauer 2013), which leads to corresponding heating values of 27.5 and 37.9 MJ/kg, respectively.Gasi cation of all other polymers leads to the generation of synthesis gas with a heating value lower than 20 MJ/kg(Wilk and Hofbauer 2013;Bai et al. 2020).The composition of generated synthesis gas in steam gasi cation of most common polymer materials and mixtures is shown in Table4.While gasi cation preference for PE and PP in feedstock is in line with pyrolysis, signi cant differences can be seen when other materials are looked upon.Thus, contrary to pyrolysis, in gasi cation plants, PET is accepted without any limitations in over 80% of cases, while PVC in over 66% of plants without limitations and when cases where allowed share of PVC in input material is limited are included, this number is boosted to 90%.Even though these materials are widely accepted, PET gasi cation leads to generation of synthesis gas with increased CO and CO 2 share due to increased oxygen content(Li et al.

Table 5
Production of reviewed gasi cation plants T thermal gasi cation P plasma gasi cation * kWh/t

Table 7
Construction costs of industrial buildings in the world

Table 9
Techno-economic data for gasi cation plant