Metal-free carbocatalysts for the selective production of ethylbenzene and cumene from mixed plastic waste

Abstract

Conventional production of ethylbenzene and cumene relies on fossil fuels and results in significant CO₂ emissions. As a sustainable alternative, plastic waste can be used to produce high-value aromatics via catalytic fast pyrolysis (CFP). However, the existing CFP approaches for converting plastics to aromatics offer low selectivity of these two products or require high-pressure hydrogen that is detrimental to the process cost-competitiveness. In this study, we demonstrate a one-pot CFP approach based on the use of mixed plastic waste, mild conditions (i.e., 550 °C, atmospheric pressure and no H₂) and most importantly, a metal-free carbocatalyst derived from tire char, another otherwise polymer waste causing significant environmental concern. This approach has been proven to achieve two high-purity products, ethylbenzene and cumene, while reducing CO₂ emissions by 12–80% compared to conventional processes, based on values reported across multiple references. It could also be highly cost-competitive as opposed to the virgin products from naphtha. Intensive characterization and density functional theory calculations further reveal that the defect-rich amorphous carbon in the tire-derived carbocatalyst created by chemical etching plays a dual role in abstracting and transferring hydrogen, enhancing the in situ interactive synergy between individual plastics that simultaneously promotes both the hydrogenation of monomers/radicals from polystyrene (PS) and the aromatization of short-chain aliphatics from low-density polyethylene (LDPE) and polypropylene (PP).

---

Green foundation

1. Catalytic fast pyrolysis of mixed plastics into high-value aromatics offers a circular and scalable pathway for plastic valorization.

2. The demonstrated record-high yields and selectivity toward ethylbenzene and cumene from mixed plastics are able to significantly avoid the use of fossil fuel and costly hydrogen, with reduced CO₂ emissions by up to 80% compared to the conventional naphtha-based production route.

3. For a greener approach in the future, electrification of the pyrolysis reactor via the use of Joule heating reactors or others is crucial, which can eliminate the combustion sector that is the largest emission source in this process.

1 Introduction

The rapid rise in the global population has driven a steady increase in the production of plastics, with an annual growth rate of 8.4% since 1950. By 2019, the global production of plastics had amounted to 9.2 billion tonnes, whereas only 10% had been recycled.^1–3 Meanwhile, on a weight basis, approximately 60% of plastic waste is made up of hydrocarbons, including polystyrene (PS), polyethylene (PE), and polypropylene (PP).^4,5 The traditional recycling methods, relying on physical collection, sorting, and mechanical processing, often degrade the quality of these plastics and have thus shown very limited success in the long-term management of plastic waste. Similarly, the chemical recycling methods, which are most often based on the depolymerization of plastic waste back to monomers for the re-manufacturing of virgin plastics, are energy-intensive and also inefficient for mixed plastics. Breakthrough approaches are still urgently needed to accommodate the mixed plastics and convert them to high-value products that should also have a comparable market size to the quantity of plastic waste produced in the same year.

The aromatic compounds, including benzene (B), ethylbenzene (E), toluene (T) and alkyl benzene such as cumene (C), are extensively used as solvents, fuel additives, and raw materials in the chemical industry, playing a crucial role in the production of a variety of end-use products.^6–8 The market size of ethylbenzene grows steadily, with an estimated volume of 36 million tonnes in 2025 and projected to reach 42 million tonnes by 2030.⁹ Cumene (C₉H₁₂), in particular, is one of the top five large-scale produced chemical reagents, primarily due to its extensive application in the production of phenol and acetone.¹⁰ Its global demand is projected to reach US$17.63 billion by 2025.^11,12 Regarding the conventional production of ethylbenzene, as illustrated in Fig. 1A, it involves an initial dehydrogenation of ethane (typically derived from naphtha cracking) to produce ethylene, which then reacts with benzene to form ethylbenzene.¹³ The process is energy-intensive due to the use of harsh conditions and even costly due to the use of noble metal catalysts.^14,15 On the other hand, cumene is produced via the Friedel–Crafts alkylation of benzene with propylene (Fig. 1B) upon the use of Lewis acid catalysts such as AlCl₃–HCl, which is highly corrosive. The process also requires a large excess of benzene that, in turn, results in the formation of undesirable polyalkylated by-products, such as diisopropylbenzenes (C₁₂H₁₈).^16,17 Finally, with respect to the production of aromatics from plastic waste, one approach is catalytic pyrolysis (Fig. 1C) based on the use of a variety of different catalysts, especially zeolite or metal-doped zeolites with moderated acidity.^18–22 However, the selectivity of benzene, toluene, ethylbenzene, and xylene (BTEX) is usually less than 50% among all the produced mono-aromatic hydrocarbons (MAHs), along with a broad range of polycyclic aromatic hydrocarbons (PAHs), which are the precursors for coke deposits that deactivate catalysts rapidly. Alternatively, as shown in Fig. 1D, a two-stage hydrogenation process involving initial hydrolysis of PS and subsequent hydrocracking of styrene monomers into ethylbenzene was proposed.^23,24 However, it is only capable of dealing with a single plastic, specifically PS, making it unsuitable for mixed plastic waste. The large consumption of high-pressure hydrogen is also concerning, as it is certainly detrimental to the economic viability of the process.

Fig. 1 Comparison of previous studies on BTEC synthesis with the present work. Panel A: the conventional ethylbenzene production route based on naphtha cracking. Panel B: the conventional production of cumene based on the Friedel–Crafts catalyst. Panel C: zeolite-catalyzed pyrolysis of mixed plastics. Panel D: a two-stage hydrogenation of PS. Panel E: the present work, focusing on the catalytic fast pyrolysis of mixed plastics consisting of LDPE, PP and PS using tire waste-derived carbocatalysts. Steps (1) and (2) in Panel E refer to the pyrolysis of the tire chip and subsequent demineralization and chemical activation of pyrolytic char, respectively.

To date, metal-free carbocatalysts have been explored as a promising substitute to replace the traditional noble metals in a variety of applications,^25,26 such as the hydrogenation of alkene hydrocarbons by the use of graphene oxide with frustrated Lewis acid–base pairs (FLPs) as active sites,^27,28 hydrogenation of aromatic nitro compounds by fullerene²⁹ and catalytic pyrolysis of plastic waste to liquid fuels by activated carbon.^30–33 However, to the best of our knowledge, no study has explored the use of tire waste for the synthesis of functional carbocatalysts for chemical recycling of plastics into valuable aromatics. Yet, it is still unknown whether the unique properties, such as the inherent sulfur- or oxygen-bearing functional group and carbon defects in the activated tire, are catalytically active.

In this work, we investigate the co-pyrolysis of PS with polyolefins (LDPE and PP) for the selective production of ethylbenzene and cumene using a carbon-based catalyst derived from scrap tire. More specifically, as illustrated in Fig. 1E, the process is facile, employing mild conditions including an optimum operating temperature of 550 °C, atmospheric pressure and no use of hydrogen. The effects of BET surface area, acidity, defect sites, and ash content of the carbocatalysts were examined to evaluate their influence on product selectivity. Meanwhile, variations in plastic ratios and pyrolysis temperatures were made to investigate the synergistic interactions among different polymers. In addition, density functional theory (DFT) calculations were performed to elucidate the role of the active sites and confirm the proposed reaction mechanism. The combined experimental and theoretical insights provide a solid foundation for developing efficient, metal-free catalytic systems for the sustainable valorization of mixed plastic waste.

2 Materials and methods

2.1 Materials

Plastic reagents, including PS, PP, and LDPE with 99.9% purity, were purchased from Nanochemazone and sieved to a particle size of 200–500 µm prior to use. Tire char, obtained through pyrolysis of mixed tire chips at 800 °C, was sourced from the industry. HCl (32 vol%), H₂SO₄ (98 wt%), and KOH were purchased from Thermo Fisher, Sigma-Aldrich, and Merck, respectively. Analytical-grade chemicals, including ethylbenzene, toluene, styrene, and cumene, were sourced from Sigma-Aldrich. Commercial activated carbon (AC) was obtained from Norit RX3 Extra Pty, Ltd. LDPE waste from various plastic packaging was provided by Urban Mining Industries Pty Ltd, while PS and PP waste were collected from local rubbish bins. Prior to use, PP and LDPE were washed with distilled water, dried, crushed using liquid nitrogen and a coffee grinder, and sieved to a particle size below 800 µm. PS waste was first dissolved in acetone, then the solvent was evaporated at approximately 80 °C. The residue was dried at room temperature overnight, crushed, and sieved to a particle size below 800 µm. To identify contaminants in the plastic waste, FTIR and CHNS analyses were conducted, and the results were compared with those of plastic reagents, as shown in Fig. S1 and Table S1.

2.2 Methods

2.2.1 Synthesis of the catalyst. As shown in Fig. 1E, the synthesis of the metal-free carbocatalyst began with the pyrolysis of scrap tires sourced from a commercial collection center. In the industrial process, scrap tires were shredded, de-wired, and pyrolyzed in a vertical retort at approximately 800 °C for 5 h to produce fresh tire char. The resulting char was subsequently subjected to acid and base washing at 80 °C, followed by KOH activation at 800 °C and thorough rinsing to remove impurities and enhance the surface area. Detailed synthesis conditions are listed in Table S2. For the initial demineralization step and post-washing stage, various acid/base/water systems were tested. For the sample treated by a sequence of HCl washing + KOH activation + final H₂SO₄ washing, it was found to be the most active for the pyrolysis of plastics and was hereafter referred to as activated tire char (ATC). The other samples were numbered from 1 to 6 in Table S2. The activation temperature and mass ratio of KOH to FTC were selected based on the previous studies.^34,35 The carbon yield of the obtained ATC is 16.8 wt% based on the scrap tire feedstock.

2.2.2 Catalyst characterization. CHNS analysis of the catalyst composition was conducted using a Thermo Scientific Flash Smart. The instrument calibration was performed using 1–2 mg of BBOT (2,5-Bis(5-tert-butyl-benzoxazol-2-yl) thiophene). For each analysis, 1–2 mg of finely ground catalyst was used. The thermal stability and ash content of the ATC catalyst were assessed using a Shimadzu DTG-60H thermogravimetric analyzer (TGA). The surface area and pore structure of catalysts were analyzed using a Micromeritics 3Flex surface characterization analyzer. Prior to the measurements, the samples were subjected to degassing at 90 °C for one hour and were subsequently heated at 150 °C for nine hours under vacuum. The Brunauer–Emmett–Teller (BET) method was employed to determine the surface area, while the Barrett–Joyner–Halenda (BJH) method was used to assess the pore structure. Raman spectroscopy was performed to investigate the carbon characteristics of the samples using a Renishaw Raman microscope equipped with a 514 nm argon laser. The analysis was carried out with a total of four accumulations to ensure accuracy.

XPS analysis was conducted using a Thermo Scientific Nexsa Surface Analysis Platform, equipped with a monochromated Al Kα source. All peaks in the spectrum were calibrated using the carbon (C1s) peak at 284.8 eV as a reference. Data processing was performed using Avantage software, with the analysis referencing the binding energies of sulfur-bearing compounds as previously reported in ref. 36 and 37. SEM imaging and elemental distribution analysis of various ATC samples were carried out using a JEOL JSM-7001F scanning electron microscope, coupled with energy-dispersive X-ray spectroscopy (EDS). High-resolution transmission electron microscopy (HRTEM) observation was performed using a Tecnai G2 T20 electron microscope at an accelerating voltage of 200 kV.³⁸

Temperature-programmed desorption (TPD) was employed to evaluate hydrogen uptake (H₂-TPD), acidity (NH₃-TPD), and basicity (CO₂-TPD) using 100 mg of each sample. For these analyses, the samples underwent pre-treatment at 800 °C for 20 minutes under an argon (Ar) flow for H₂-TPD and under helium (He) for NH₃-TPD and CO₂-TPD, and then were cooled to 50 °C. The absorption process involved introducing 10% H₂/Ar for H₂-TPD, 15% NH₃/He for NH₃-TPD, and CO₂/He for CO₂-TPD, each at 50 °C for 1 hour. The infrared spectra of the catalysts were recorded using a Thermo Nicolet 380 spectrometer. Sample pellets were made with 5 wt% catalyst in KBr powder and pressed at 4–8 MPa. After treating the pellets at 500 °C for 1 hour and evacuating at 10⁻³ Pa for 2 hours, a background spectrum was recorded at room temperature and automatically subtracted. To determine the acidity, the catalysts were exposed to pyridine at room temperature for 30 minutes until saturation, then evacuated at 40 °C for 1 hour. The spectra were recorded in the 400–4000 cm⁻² range.

2.2.3 Catalytic activity test.
2.2.3.1 Pyroprobe system. Fast pyrolysis for catalyst screening was performed using the CDS Pyroprobe 5200, with detailed procedures described in our previous publication.³⁹ Pyrolysis experiments were conducted on both individual and mixed plastics, with and without the catalyst, using different mass ratios as specified in Table S3. For each run, feedstocks were mixed with and without the ATC catalyst and loaded into a quartz micro-reactor. The mixture was rapidly heated to 550 °C (∼140 °C s⁻¹) and held for 25 s, and the temperature was selected based on prior studies showing full decomposition of LDPE, PP, and PS between 500 and 600 °C.⁴⁰ In addition to pyrolysis under helium (He), PS pyrolysis in 10 vol% H₂/He was conducted at 60 psi (≈4 atm). This involved purging the H₂/He mixed gas into the interface for 15 s to increase the pressure before reverting to the low-pressure mode. A blank run using only ATC was always performed after each run to purge out the residues in the sample-transferring line. Each condition was tested at least twice to determine the standard error. The area-based selectivity of each compound was determined from GC-FID peak areas and calculated using the normalized area percentage (area%). This approach assumes a linear correlation between FID response and mass concentration.⁴¹ The selectivity, S(i), was calculated as:

where A(i) is the peak area measured for species i and A_total refers to the areas for all the species measured.

2.2.3.2 Fixed-bed reactor system. To validate the Pyroprobe results, pyrolysis was also conducted in a fixed-bed reactor according to the conditions listed in Table S3. Most experiments were carried out at 550 °C in a preheated furnace, consistent with the Pyroprobe conditions. The furnace was protected by nitrogen during the heating process. Once the temperature was reached, a sample-laden holder was rapidly inserted into the furnace center (a heating rate of ∼300 °C min⁻¹), and the resulting vapors were passed through cold traps cooled with dry ice and isopropanol. The mass and corresponding yields of wax, char, and liquid were directly measured. The liquid fraction was collected by rinsing the trap with 5 mL of acetone and analyzed via GC-MS, ¹³C-NMR, and ¹H-NMR. GC-MS was conducted in an Agilent GCMS 5973 with a DB-5MS column. For NMR analysis, the pyrolytic oil was diluted in d-chloroform and analyzed according to the procedure shown in previous work.⁴⁰ Gas products were collected in gas bags, analyzed by GC-TCD, and quantified using standard methods. The yields of the target products in the fixed-bed reactor experiment were determined using an external calibration method. Ethylbenzene, toluene, cumene, and styrene were used as calibration standards. For each compound, volumes of 0.5 mL, 1.0 mL, and 1.5 mL were individually dissolved in 5 mL of acetone, corresponding to the same solvent volume used to wash the impinger during liquid product collection. The mass of each standard was calculated based on its density, and calibration curves were constructed by correlating the calculated mass with the corresponding GC/MS peak area. The resulting calibration plots are presented in Fig. S2. Product yields of key compounds (styrene, toluene, ethylbenzene, and cumene) were calculated based on external calibration curves (eqn (1)).where: A(i)_{in experiment} is the peak area of compound i in the experimental sample. A(i)_{in calibration} is the peak area of compound i obtained from the calibration curve. Mass(i)_{in calibration} is the calculated mass of compound i used in the calibration. Mass plastic is the mass of the plastic feedstock used in each experiment.

2.2.4 DFT calculations. Spin-unrestricted density functional theory calculations were conducted using the projector-augmented-wave method as implemented in the Vienna Ab initio Simulation Package (VASP).⁴² The exchange–correlation functional takes the Perdew–Burke–Ernzerhof form.⁴³ Since the catalyst supercell is large, the plane wave basis energy cut-off of 400 eV, and a k-mesh of 1 × 1 × 1 were adopted during optimizations. The energy and force convergence criteria were set to 1 × 10⁻⁵ eV and 0.02 eV Å⁻¹, respectively. To calculate the Gibbs free energy change of the desired reaction process at the working temperature, we express the Gibbs free energy (G) of each reaction species as follows:

G = E + G_corr (2)

where E is the total energy obtained from DFT calculations, E_ZPE is the zero-point energy, and S is the entropy. T is the working temperature, and is the inner energy increment due to the increase in the temperature. The G_corr term can be obtained by adsorbent frequency analysis, followed by post-processing using the VASPKIT code.⁴⁴ During the adsorbent frequency analysis, only the adsorbent is allowed to move. An amorphous carbon network was adopted to represent the activated carbon char catalyst. The network model was constructed based on the “GAP_annealed_taC_512at_10_5.xyz” amorphous carbon model as revealed in Deringer et al.'s work.⁴⁵ After adding a vacant layer of 20 Å, creating a defect, and removing extra carbon atoms, the amorphous carbon network was fully optimized.

2.2.5 Techno-economic analysis and CO₂ emission assessment. Finally, to evaluate the economic viability and carbon footprint of our process, a techno-economic analysis was performed for a hypothetical plant with a plastic treatment capacity of 240 tonnes per day (i.e., 10 tonnes per h), a size that is reasonable for the efficiency of cost and heat integration.^46–48 All assumptions and detailed methodologies are described in the SI (Methods, Tables S4–S6). The system boundaries for this study and the conventional processes used for comparison are presented in Fig. S3 and S4, with detailed explanations provided in the SI. To improve process efficiency, heat integration was implemented, and the corresponding details are illustrated in Fig. S5 in the SI. Two scenarios were examined to assess the impact of energy efficiency and by-product credit on the minimum selling price (MSP). The first scenario consumed the entire gas products and a portion of the char for the combustion section, leaving most of the char as a by-product. In contrast, the second scenario used solid char for combustion, with gas products left as by-products. Both scenarios assume the spent ATC catalyst to be a valuable by-product or a fuel. For each scenario, it starts from plastic pre-treatment to pyrolysis of mixed plastics in a fixed-bed reactor, product separation, combustion, and finally flue gas cleanup. Aspen Plus Process Simulator (V14 aspenONE) was used to develop the flowsheet for each scenario.

The CO₂ emission calculations in this study are based on two scopes. Scope 1 emissions refer to direct emissions released as a result of activities within the facility, including emissions from process reactions and fuel combustion used to power the facility. Scope 2 emissions arise from indirect sources, primarily associated with electricity supplied to the facility. Scope 1 emissions were calculated from the CO₂ generated during combustion, obtained directly from process simulations, as well as from other process-related emissions, for which CO₂ emission factors were adopted from the literature. Scope 2 emissions were calculated based on electricity consumption, with the associated CO₂ emissions derived from the energy usage of each process unit. The detailed calculations are provided in the SI.

3 Results and discussion

3.1 Catalyst characterization

As indicated in Fig. 1E, the synthesis of the metal-free carbocatalyst was initiated from the pyrolysis of tire chips. The initial pyrolysis steps result in the production of fresh tire char (FTC), which contains 13.8 wt% ash primarily dominated by sulfur, silica (SiO₂) and zinc oxide (ZnO) in Table S7. These elements are used as the vulcanization additives during the manufacturing of tires.^49,50 The properties of a typical ATC sample, as obtained through initial acid washing with HCl and final rinsing with H₂SO₄, are detailed in Table S8, where its FTC precursor and a commercial activated carbon (AC) sample are included for comparison. The combination of HCl leaching and H₂SO₄ rinsing reduced the overall ash content down to only 1.9 wt%, which is mostly acid-insoluble silica. Furthermore, as indicated by the surface scan results from XPS in Fig. S6, the ATC catalyst is metal-free on its surface, as opposed to the abundance of Si, Zn, and S on the FTC surface.

Additionally, the microstructures of the FTC precursor and ATC are shown in Fig. 2a. Compared to the FTC precursor with a small BET surface area of 33 m² g⁻¹, ATC exhibits a large surface area of 1057 m² g⁻¹ that is highly porous. Its surface area is also comparable to the commercial AC with a BET surface area of 1517 m² g⁻¹. Such a high surface can be explained by the chemical activation role of the KOH reagent^51–53 used for the chemical activation. TEM images in Fig. 2b-1 and b-2 show that FTC and ATC are highly amorphous. In particular, ATC displays abundant structural defects, highlighted by white circles. A magnified view in Fig. 2b-3 further reveals that ATC consists of highly curled small flakes, characteristic of amorphous carbon,⁵⁴ while the white lines indicate vacancy sites corresponding to defect regions.^55,56 In addition, Raman spectra shown in Fig. 2c-1 reveal structural defects and imperfections, which are quantified by the I_D1/I_G ratio, where a higher ratio indicates a greater defect. The ATC displays an intensity ratio of 4.15 between the D₁-band (∼1250 cm⁻¹) and the G-band (∼1550 cm⁻¹) (also see Table S8). This ratio is higher than that of FTC (3.19) and AC (2.74). These findings support the development of carbon active sites and amorphous structures according to the activation process, which is further supported by the largest ratio of I_D3/(I_D2 + I_D3 + I_G) for ATC.

Fig. 2 Properties of carbocatalysts derived from waste tires. Panel a: summary of carbocatalyst synthesis and SEM images of FTC, AW-TC (acid-washed tire char), and ATC (white scale bar is 10 µm). Panel b: TEM images of (1) FTC, (2) ATC, where the white circles highlight defect sites, and (3) a magnified view of the defect regions. Panel c(1): Raman spectra of ATC compared with FTC and commercial activated carbon (AC), including peak deconvolution. Panel c(2): total acidity of ATC, FTC, and AC determined by NH₃-TPD and pyridine-FTIR.

Regarding the total acidity shown in Fig. 2c-2 and Fig. S7, it is also evident that the ATC is the most acidic for both the acidity measured by NH₃-TPD and pyridine-FTIR. It is also a Lewis acid catalyst with a lean Brønsted acid site. The Lewis acidity can be attributed to the abundant carbon defects, which create electron vacancies in the unsaturated carbon structure, and the Brønsted acidity can be from S-/O-bearing functional groups.

3.2 Catalyst screening using a Pyroprobe micro-reactor

For the catalyst screening purpose, a micro-reactor, namely Pyroprobe coupled with GC–TCD–FID–MS⁵⁷ was used first. As detailed in the Methods section, catalysts were physically mixed with plastics (reagent-grade) at varying ratios and pre-loaded into the micro-reactor prior to isothermal pyrolysis. The three individual plastics were first processed at 550 °C, without the use of any catalysts. As depicted in Fig. 3a and b, pyrolysis of PS alone led to the primary formation of its monomer styrene purely derived from the thermal breakdown, while PP and LDPE predominantly decomposed into aliphatic compounds with a broad range of retention time from C4 (butene) to C20 (eicosene), aligning with the observations from previous studies.^58,59 Given that PS primarily produces aromatic compounds from its own pyrolysis, it was then mixed with the ATC carbocatalyst at a large mass ratio of 10 and subjected to pyrolysis. As demonstrated in Fig. 3c (GC/MS spectra in Fig. S8), the ATC carbocatalyst is the most active for the highest BTEC selectivity. The catalyst amount is critical, achieving a BTEC selectivity of 78% (area-based) for an ATC-to-PS mass ratio of 10, as shown in Fig. 3d.

Fig. 3 Selectivity of BTEC from the pyrolysis of reagent-grade plastics (LDPE, PP, and PS) at 550 °C in Pyroprobe. (a) Product distribution from non-catalytic pyrolysis of single plastics, with “others” including PAHs and oxygenated compounds from plastic impurities. (b) Respective GC/MS spectra for the pyrolytic oil samples in (a). (c) Pyrolytic oil of PS upon the use of different catalysts at a catalyst to PS mass ratio of 10. (d) BTEC selectivity as a function of the mass ratio of the ATC carbocatalyst to PS. (e) Pyrolytic oil composition as a function of PP addition percentage in the PS–PP binary mixture. (f) Pyrolytic oil composition as a function of LDPE addition percentage in the PS–LDPE binary mixture. (g) Pyrolytic oil composition from the ternary mixture of PS–PP–LDPE. (h) Respective GC/MS spectra for the pyrolytic oil samples in (g). In panels e–h, the mass ratio of ATC to plastic feedstock was fixed at 10.

Next, by fixing the ATC mass ratio at 10, the binary mixtures of PS–PP and PS–LDPE were tested. As shown in Fig. 3e and f, with the increase in the mass percentage of PP or LDPE, the selectivity of BTEC decreases parabolically rather than linearly, caused by the weighted averaging of the individual results. These findings indicate synergistic interactions between the two aliphatic plastics and PS, which enhance the selectivity toward BTEC. Finally, the ternary blends of PS–LDPE–PP with a fixed mass ratio of 1 : 1 for PP to LDPE were tested. All the blends were also mixed with ATC at a mass ratio of 10. Clearly, as demonstrated in Fig. 3g and h, the introduction of PS was in favor of the formation of BTEC, as expected. More intriguingly, a blending ratio of 6 : 1 : 1 for the three plastics (for the presence of 75 wt% PS in the blend) reached the highest BTEC selectivity of 80%, which is also comparable to the 78% selectivity achieved with PS alone. Detailed information regarding the composition of the compounds produced during pyrolysis, as well as GC/MS spectra, can be found in Fig. S8 and Tables S9–S16.

3.3 Scale-up experiment in a fixed-bed reactor

Based on the above-mentioned screening results, a fixed-bed reactor with a gram-scale capacity, which is a hundred times larger than the Pyroprobe, was used to validate the high performance of the ATC carbocatalyst. Based on the use of PS alone, the pyrolysis temperature and the catalyst to PS mass ratio were further fine-tuned in the fixed bed. As demonstrated in Fig. S9, a reaction temperature of 550 °C is still the best in providing the most intense peaks for BTEC. Interestingly, the catalyst to PS mass ratio of 3, instead of 10 for Pyroprobe tests, was found to be optimum in achieving the most intense peaks for our target products, whilst PAHs are invisible. This could be due to better mixing of the catalyst with plastics in a relatively large facility. Accordingly, these two conditions are fixed, whilst different plastics, single, binary and ternary, were tested. All the products including solid residue/char, oil and gas were collected. Based on the product yields shown in Table S17, a satisfactory mass balance of 89–99 wt% was confirmed for the reactor system.

The reagent-grade PS and its binary mixtures with PP or LDPE were examined. Fig. 4a-1 for PS alone confirmed a pronounced color change and composition of the liquid product caused by the addition of the ATC catalyst. In the absence of the catalyst, the liquid is rich in styrene monomers, along with PAHs, which account for the yellow color of the liquid product. In contrast, upon the addition of the catalyst, instead of styrene, the BTEC species, especially ethylbenzene is much more dominant. The liquid color is also transparent for the complete disappearance of the PAH species. The area-based selectivity analysis in Table S18 confirmed a selectivity of 86% for the BTEC group, which aligns closely with the Pyroprobe result. Furthermore, regarding the PAHs formed from PS alone, Table S17 indicates a preferred transformation of these species into solid char, along with the increased formation of gaseous species especially H₂ (Table S19). Clearly, the cracking and even dehydrogenation of PAHs was effectively catalyzed by the ATC carbocatalyst. With respect to the binary blends of PS with aliphatic PP or LDPE (25 wt% each), as demonstrated in Fig. 4a-2 and 3 (compositions in Tables S20 & S21), in the absence of the catalyst, the predominance of styrene is not changed, whilst the liquid product is turbid due to the presence of waxy species derived from PP or LDPE. However, the introduction of the ATC catalyst is clearly striking, resulting in the disappearance of wax and a full transparency of the liquid color. Compared to the 50 wt% BTEC yield achieved from PS alone, the BTEC yield from binary PS–PP and PS–LDPE was boosted to 55 wt% and 62 wt%, respectively (Table S17). These results support a positive synergy between PS and LDPE/PP for the high yield of BTEC. For comparison, the binary blend of PP–LDPE at an equal mass ratio was also examined. As shown in Fig. 4a-4 and Table S22, in the absence of the catalyst, abundant wax was formed, which echoes the turbid appearance of the liquid product and the large carbon number species with a long retention time in the GC–MS spectra. By contrast, the addition of ATC is effective at reducing the turbidity and cleaving the long-chain wax into short-chain compounds up to C₁₆.

Fig. 4 Upscaled production of BTEC from (a) reagent-grade plastics and (b–c) real-world plastic waste. Panels a(1)–(4): GC/MS spectra and collected oil from the pyrolysis of PS, PS (75 wt%)–PP (25 wt%), PS (75 wt%)–LDPE (25 wt%), and PP (50 wt%)–LDPE (50 wt%), with and without ATC, respectively. Panel b (left panel): pulverized plastic waste. Right top panel (#1–4): photos of the liquid samples collected, and bottom panel: the respective GC/MS and ¹³C-NMR spectra. Panel c(1): product yield and BTEC yield in samples #1–4. Panel c(2): comparison of this study with other studies using pyrolysis at a temperature range of 400–650 °C. (Note: The BTEX or BTEC yield obtained from other studies was calculated as the product of selectivity and oil yield.) References: [A], ref. 62, [B] ref. 60, [C] ref. 61, [D] ref. 63, [E] ref. 64, [F] ref. 65, [G] ref. 66, [H] ref. 67, [I] ref. 68, and [J] ref. 60. Panel c(3): mass percentages of individual BTEC from the ATC-assisted pyrolysis of PS only; and panel c(4): mass percentages of individual BTEC from the ATC-assisted pyrolysis of the PS + PP + LDPE blend (6 : 1 : 1).

Finally, the real-world plastic waste samples were tested, which were sourced from food packaging soft plastics for LDPE, sample vials for PP, and styrofoam for PS. The former two samples were pulverized directly (Fig. 4b), while the latter one was dissolved in acetone first. Subsequently, acetone was evaporated and the solid residue was densified and pulverized. The pulverized samples, blended with and without the catalyst, were subjected to pyrolysis under the same conditions as those used for the reagent-grade plastics. As demonstrated in Fig. 4b, upon the inclusion of the catalyst and the increase in the mass fraction of PS in the ternary blend, the liquid turned transparent with invisible deposits. The product yield distribution in Table S23 confirmed the absence of wax, while the yield of the overall liquid remains as high as 74–80% across all cases. The respective GC–MS spectra of liquid (Fig. 4b and compositions in Tables S24 and 25) confirmed the predominance of BTEC in all three samples with the use of the ATC catalyst. These results are in close agreement with those obtained for reagent-grade plastics, except for an increased char yield observed at the PS : PP : LDPE ratio of 6 : 1 : 1 (Table S23). As shown in Fig. S1, the FTIR spectra exhibit similar characteristic peaks for both plastic waste and reagent-grade plastics. However, CHNS analysis (Table S1) detected small amounts of heteroatoms, specifically sulphur and nitrogen, in the plastic waste. These impurities could be derived from the remnants of the labels which have a different structure than the bare bottles. To further validate this hypothesis, thermogravimetric analysis (TGA) was performed on spent ATC catalysts used in the pyrolysis of both waste plastics and reagent-grade plastics, with the results shown in Fig. S10. The differences in weight loss profiles confirm variations in the amount of volatile (heavy) compounds retained within the ATC. Nevertheless, the liquid products remain the same between the reagent-grade and real plastics. To further support this and examine the liquid product from the real plastics, C¹³-NMR analysis was conducted. A standard solution of BTEC with known concentrations of individual species was also analyzed and used to quantify the mass yield of BTEC based on an external standard calibration procedure outlined in the Methods section. Evidently, the spectra shown in Fig. 4b support the abundance of aromatic rings from BTEC in the last three samples.

The quantified mass yield of BTEC, along with the yields of the overall liquid and wax is summarized in Fig. 4c-1. Fig. 4c-2 further benchmarks our BTEC yields against the published data. Clearly, in matching the maximum 35 wt% yield for BTEX achieved from the use of zeolite catalyst,^60,61 a blending ratio of 2 : 1 : 1 for PS : PP : LDPE is good enough. Nevertheless, for another study⁶² on pyrolysis of polyethylene using ZSM-5, they only managed to achieve this yield at 700 °C. Similarly, for the study using zeolite,⁶¹ the ternary blend of PS : PP : LDPE with the same mass ratio as ours required a reaction duration as long as 75 minutes. The overall liquid yield is only 60 wt%, as opposed to 75 wt% achieved in this work. Most importantly, for our best mass ratio of 6 : 1 : 1 for PS : PP : LDPE, the selectivity and yield of BTEC reach 90% (Table S25) and 62 wt% (Table S17), respectively. Particularly, among the BTEC group, ethylbenzene is the most abundant product for a mass percentage of 62–67% (Fig. 4c-3 & 4). It is followed by cumene with a mass percentage of 22–28.2%. Such high selectivity for both products has yet to be reported.

3.4 Reusability of ATC

Cycle tests were performed to evaluate the durability of the ATC carbocatalyst. As shown in Fig. 5, repeated use of ATC resulted in a gradual yellowing of the liquid product and a reduction in BTEC selectivity from 90% initially to 63% after the fifth cycle. This decline can be attributed to the accumulation of heavy PAHs and solid coke on the catalyst surface. By mild regeneration through combustion in air at 400 °C, the catalyst substantially regained its performance, producing a transparent liquid product with BTEC selectivity restored to ∼80%. The carbon loss during regeneration was approximately 10–20 wt%, which is mostly ascribed to the trapped heavy compounds as well as partial oxidation of the carbon matrix inherent to the carbocatalyst.

Fig. 5 Reusability of ATC using a fixed-bed reactor for the pyrolysis of PS at 550 °C with PS to ATC ratios of 1 to 3. (a) Collected pyrolytic oil, (b) GC/MS of pyrolytic oil, and (c) BTEC selectivity. T, E, S, and C denote toluene, ethylbenzene, styrene, and cumene, respectively.

To assess whether the regeneration treatment affected defect density and acidity, fresh ATC and spent–regenerated ATC were compared using FTIR–pyridine adsorption and temperature-programmed oxidation (TPO) analyses. As shown in Fig. S11, the regenerated ATC exhibits a slightly lower total acidity than the fresh ATC, indicating partial loss of defect and/or functional sites during regeneration. In addition, the TPO profiles reveal that, up to 550 °C (i.e., the upper temperature relevant to the process), the regenerated ATC displays higher oxidation resistance than fresh ATC. This behavior is likely due to the preferential removal of more reactive and disordered carbon fractions and/or surface functional groups during the regeneration process.

3.5 Property–activity correlation for the ATC catalyst

As demonstrated, ATC exhibited superior catalytic activity over FTC and AC. We therefore investigated the relationship between catalyst properties and their selectivity toward BTEC. Several carbocatalysts were prepared by varying the pre- and post-treatments applied after KOH activation, as detailed in Table S2. As a result, the prepared samples differ in the BET surface area, defect density, ash content, and acidity. The correlation of each of these properties with the selectivity of BTEC from the pyrolysis of PS alone is summarized in Fig. 6a–d. Clearly, all these properties are influential in determining the final BTEC selectivity. As shown in Fig. 6a, the BET surface area is the most influential, showing a linear trend with a sharp gradient for all the datum points but the commercial AC sample. The low activity of AC, with the largest BET surface area, can be explained by its high graphitization extent and the shortage of defect carbon sites (I_D1/I_G in Fig. 6b), as well as its lowest acidity in Fig. 6c. Finally, regarding the ash-forming metal species, as evidenced in Fig. 6d, they are detrimental, showing a reverse trend for their general correlation with BTEC selectivity. This is strong evidence of the catalytic role of carbon defect sites and even the associated sulfur/oxygen-bearing sites. Inferably, these active sites are mostly created by the catalyst synthesis process including demineralization and chemical activation. To prove this, we subjected the commercial AC to the same process and found that its activity was increased markedly. As shown in Fig. 6e and f, the overall BTEC selectivity was improved from 79% for original AC to around 90% post the treatment of the AC, along with a significant improvement in the selectivity of toluene and ethylbenzene.

Fig. 6 Correlation of BTEC selectivity from Pyroprobe-based pyrolysis of PS at 550 °C with (a) BET surface area, (b) degree of carbon defects (I_D1/I_G), (c) acidity, and (d) ash/impurity content of the carbocatalysts. (e) GC–MS chromatograms and (f) BTEC selectivity of liquid products collected from fixed-bed reactor pyrolysis of PS at 550 °C using AC before and after KOH activation, following a method similar to that employed for ATC synthesis. T, E, S, and C denote toluene, ethylbenzene, styrene, and cumene, respectively.

3.6 Proposed reaction mechanism

All the experimental results confirm the importance of PS as a principal source for the formation of BTEC. More specifically, alkyl benzene radicals could be initially derived from the thermal decomposition of PS alone, whereas their subsequent conversion into BTEC could be catalyzed by the ATC carbocatalyst. The non-metal active sites on the ATC should be highly active in facilitating hydrogen abstraction and transfer, resulting in the formation of ethylbenzene (C₈H₁₀), toluene (C₇H₈), and cumene (C₉H₁₂) as the major products, with only trace amounts of benzene (C₆H₆). To prove this hypothesis, we tested the catalytic pyrolysis of pure styrene (C₈H₈) under nitrogen and hydrogen atmospheres at 550 °C in the Pyroprobe. As demonstrated in Fig. 7a-1, styrene remained stable under a nitrogen atmosphere, whereas it successfully converted to BTEC with a selectivity of 70% in 10 vol% H₂/He at a total pressure of 60 psi. The resultant GC–MS spectrum is also highly analogous to the case of PS mixed with ATC. Operating at a low H₂ pressure of 60 psi was even beneficial for maintaining a relatively stable performance of the catalyst during the cycle test, as demonstrated in Fig. 7a-2. The H₂-TPD results (Table S8 and Fig. S12) confirmed the hydrogen adsorption capability of ATC. Regarding the active site for the abstraction of hydrogen, the structural defects and functional groups within the carbonaceous matrix of ATC likely play a critical role in hydrogen uptake and dissociation. On the one hand, the fresh tire char (FTC) rich in ash-forming minerals such as zinc sulfide (ZnS, a Lewis acid too)⁶⁹ was inactive (Fig. 3c). The property–activity correlation results in Fig. 6c also indirectly support this statement. On the other hand, as confirmed in other studies, the graphene-like structure and delocalized charge distribution in carbon materials can give rise to frustrated Lewis acid–base pairs (FLPs), which facilitate the activation of H₂ and enhance the hydrogenation of C=C bonds.^27,70 This is apparently the case for the ATC carbocatalyst, as it is rich in both acid and base sites as opposed to its other two counterparts (Fig. S7b and c). Additionally, based on the product distribution that has been experimentally confirmed, Fig. 7a-3 illustrates the possible reaction routes for BTEC from the feedstock of PS alone. In principle, in the presence of defective carbon sites acting as a Lewis acid, the loss of a hydride anion from the alkyl position of PS can initiate cleavage of the long PS chain.⁷¹ As a result, a tertiary benzylic radical is formed,^71,72 which can subsequently react with hydrogen via intermolecular hydrogen transfer and undergo β-scission, leading to the formation of a benzyl terminus. This terminus can undergo further cleavage, forming a styryl radical, which is subsequently converted into ethylbenzene through reacting with the hydrogen radical. Additionally, toluene is formed when the benzyl terminus is protonated at the benzylic ring and then undergoes β-scission. After PS cleavage into a benzyl terminus, the remaining part of the PS chain forms an α-methyl benzyl terminus, which undergoes β-scission to generate a cumyl radical. This radical then reacts with hydrogen to form cumene. In contrast, on the Brønsted acid site, which is scarce within the ATC catalyst, PS degradation primarily leads to the formation of carbocations, which further undergo fragmentation to produce benzene.^72,73 The required hydrogen can be sourced directly from the functional oxygen groups bound to hydrogen, the dehydrogenation of PAHs that ultimately lead to coke/char (Table S17), hydrogen radicals derived from alkyl groups, or even hydrogen gas released during catalytic cracking. The gas product analysis in Table S19 confirmed a strong capability of ATC on the formation of hydrogen and other light hydrocarbons from the catalytic pyrolysis of PS alone. The elemental analysis of the spent ATC in Table S26 even confirmed a considerable increase in the hydrogen content in the spent ATC, which is mostly due to the chemically adsorbed hydrogen that could not be removed by acetone washing.

Fig. 7 Proposed reaction mechanism. (a) Pyrolysis of PS or its styrene monomer with the addition of ATC in the Pyroprobe at 550 °C and a catalyst-to-plastic mass ratio of 10: (1) GC/MS spectra collected under He and 10% H₂ in He, (2) cycling performance of ATC during PS pyrolysis under He and 10% H₂ in He, and (3) the proposed mechanism for the conversion of PS to TEC on the ATC surface. (b) Pyrolysis of aliphatic plastics (PP or LDPE) alone: (1) yields of gas products with and without ATC, (2) selectivity towards benzene (B), toluene (T), ethylbenzene (E), and benzene derivatives, and (3) the proposed reaction mechanism. (c) Catalytic pyrolysis of PS–PP–LDPE ternary blends: (1) selectivity of ethylbenzene versus the weighted average based on single-plastic results, (2) selectivity of cumene versus the weighted average based on single-plastic results, and (3) the proposed mechanism.

The catalytic performance of ATC on the pyrolysis of aliphatic PP and LDPE is also noteworthy. For the liquid produced derived from the binary LDPE–PP (1 : 1 for mass ratio) in Fig. 4a-4, it is evident that the ATC carbocatalyst is active in promoting the cleavage of the long-chain aliphatics. Simultaneously, the yields of light gases were also enhanced considerably, including H₂, CH₄, C₂H₄/C₂H₆ and C₃H₆/C₃H₈, as demonstrated in Fig. 7b-1. The molar yields of these gas products also differ greatly between the two plastics. The LDPE alone favors the production of lighter ethylene (C₂H₄), methane (CH₄) and H₂, in contrast to the dominance of propylene (C₃H₆) and H₂ from PP. Evidently, after mixing with ATC, the dehydrogenation of LDPE or PP was also catalyzed by ATC, which should occur on the Lewis acid sites too. Additionally, as shown in Fig. 7b-2, the selectivity of mono-aromatics including benzene (B), toluene (T), ethylbenzene (E), and benzene derivatives (detailed compositions in Tables S27 and S28), was enhanced remarkably from the catalytic pyrolysis of two plastics alone. In particular, for the pyrolysis of PP with the addition of ATC, the total selectivity for its benzene derivatives was increased from negligible (without catalyst) to around 35% based on the GC–FID peak areas. It may be due to the methyl-branched structure of PP, which can be easily cracked into lighter gases (e.g., C₂–C₃, Fig. 7b-1 and Table S19) that subsequently undergo aromatization to form aromatic compounds.^74,75 In addition, the abundant mesopores within the ATC could exert a sieving effect, promoting the cyclization of alkene (C_nH_2n).⁷⁶ Upon the catalytic role of Lewis acid sites, the abstraction of hydrogen could further take place from the alkene intermediate, leading to the formation of MAHs including BTEC, as illustrated in Fig. 7b-3.

Finally, as hydrogen donors are essential for the conversion of PS to BTEC while hydrogen abstraction can occur from the pyrolysis of LDPE and PP, we speculated that for the catalytic pyrolysis of the PS–PP–LDPE ternary blend, a strong synergy occurs between the three plastics in promoting the intermolecular hydrogen transfer on the ATC carbocatalyst surface. This is rectified by Fig. 7c-1 where the experimentally observed selectivity of ethylbenzene (E) from the ternary blends is obviously larger than the respective average calculated based on the results from the individual plastics. Likewise, Fig. 7c-2 for the selectivity of cumene (C) is also supportive of a strong synergy between individual plastics. More specifically, as shown in Fig. 7c-3, it is inferred that the H-abstraction could occur preferentially between hydrocarbon radicals derived from PS (e.g., styryl, benzyl, or cumyl radical) and the long hydrocarbon chain (e.g., wax) in PP or LDPE, resulting in both the formation of BTEC and shorter aliphatic compounds. Subsequently, the short chain radicals from LDPE or PP could undergo cyclization and aromatization, converting into monocyclic aromatics including BTEC.

3.7 DFT calculation on the active sites

DFT calculations were performed to rationalize the active sites on the ATC carbocatalyst and their roles in the reaction pathways illustrated in Fig. 7a–c. The key mechanism involves the abstraction of hydrogen from the reactant by the ATC, followed by the passage of hydrogen from the ATC to radicals or light compounds, including monomers derived from plastics. To verify this mechanism and identify the active sites responsible for facilitating hydrogen migration, several possible functional groups within the ATC were examined. The graphitic form was first chosen as the matrix to incorporate potential active sites, including oxygen-bearing groups, sulfur-bearing groups, and unsaturated carbon sites near defects. Although dimethyl heptene was the most abundant aliphatic species detected by GC–MS, heptane (C₇H₁₆) was selected as a simplified model compound to represent aliphatics produced from the catalytic cracking of LDPE or PP (Fig. 4a-4 and Table S22). The energy change for the reaction C₇H₁₆ + * → C₇H₁₅ + *H, where * denotes the active site, was evaluated across various sites. As shown in Fig. 8, only defect carbon sites (−0.61 eV) and carbon defects adjacent to oxygen functional groups (−0.56 eV, −0.80 eV, and −1.46 eV) exhibited negative energy changes for hydrogen abstraction from heptane. In contrast, other functional groups showed positive energy changes (from +1.57 eV to +3.42 eV), indicating that hydrogen abstraction is energetically unfavorable at those sites.

Fig. 8 DFT calculations of H-abstraction energy of heptane initiated by oxygen functional groups, sulphur functional groups, and carbon defects. (a): Configurations of defect carbon sites, (b) and (c): configurations of sulphur functional groups, (d)–(f): configurations of oxygen and sulfur functional groups, and (g)–(i): configurations of oxygen functional groups and carbon defects, and (j): configurations of oxygen functional groups.

3.8 DFT calculation of the reaction mechanism

We further proceeded to investigate the defective carbon sites within the amorphous region to better represent the actual structure of ATC (Fig. 9a). Various types of unsaturated carbon sites around hole defects were evaluated for their ability to initiate hydrogen abstraction from heptane, as shown in Fig. S13. The energy changes for these reactions varied widely, ranging from +1.79 eV to −0.83 eV across different carbon active sites. Based on these results, we selected two different active sites that exhibited moderate H-abstraction energies. In particular, we focused on two carbon sites: a secondary carbon(I) in Fig. 9b that can be referred to as active site I and a tertiary carbon(II) in Fig. 9c that can be referred to as active site II. The energy change for the reaction (C₇H₁₆ + * → C₇H₁₅ + *H) reaches −0.65 eV and 0.88 eV for the two sites, respectively. Subsequently, we investigated several radicals and a monomer (styrene) as potential recipients for the abstracted hydrogen. The results indicate that strong hydrogen adsorption can increase the energy required for the subsequent desorption and transfer of hydrogen, potentially deteriorating the performance of active site I. This concern is supported by the variation in energy changes for the migration of adsorbed hydrogen from both active sites to various radicals and styrene monomers to form toluene, ethylbenzene, and cumene, as shown in Fig. S14 and S15. Although active site II exhibits higher energy for H abstraction, it shows lower energy for hydrogenation of various radicals and styrene monomers. For example, the energy change for the formation of ethylbenzene from a styryl radical is −1.07 eV on active site II (Fig. S15), compared to +0.51 eV on active site I (Fig. S14). Nevertheless, for both active sites, the formation of ethylbenzene from a styryl radical and the formation of cumene from a cumyl radical exhibit a similar low energy change relative to the hydrogenation of other radicals. This finding explains the predominance of ethylbenzene and cumene in our liquid products. It also suggests that defective carbon sites exhibiting moderately negative energy changes for hydrogen abstraction serve as effective active sites for initiating the reaction, while sites showing moderately positive values are more suitable for supplying hydrogen during subsequent reaction steps.

Fig. 9 DFT calculation results: (a) configuration of a defect in amorphous carbon (the blue area shows the defect), (b) active site I, representing a defect similar to that in graphitic carbon (the red circle shows H abstracted from the defect site), (c) active site II, representing a defect similar to that in amorphous carbon (the red circle shows H abstracted from the defect site), (d) stepwise reaction mechanism corresponding to the Gibbs free energy profile shown in (e) for active site I and (f) for active site II.

Finally, as illustrated in Fig. 7c-3, we propose a strong synergistic effect between the three types of plastics in promoting the formation of toluene, ethylbenzene, and cumene. Specifically, for the model aliphatic compound heptane (C₇H₁₆) derived from LDPE and/or PP, hydrogen abstraction can be initiated by the styryl radical generated from PS, leading to the formation of C₇H₁₀, which can subsequently undergo a Diels–Alder cyclization reaction and eventually be converted into toluene (C₇H₈) via a total of 18 steps shown in Fig. 9d. Simultaneously, the hydrogenation of the styryl radical takes place. The respective Gibbs free energy change profile is outlined in Fig. 9e for active site I and in Fig. 9f for active site II, with all the free energies calibrated at 550 °C. Notably, both pathways exhibit negative ΔG values, strongly supporting the thermodynamic feasibility of the proposed aromatization mechanism. These results also confirm that active sites with moderate hydrogen abstraction energies are effective at facilitating the reaction. For comparison, the coupling of ethylbenzene formation from the styrene monomer and the aromatization of heptane to toluene were also considered, with detailed breakdowns and Gibbs free energy profiles provided in Fig. S16 and S17 for the two active sites, respectively. The results show that the hydrogenation of styrene to ethylbenzene on active site II is thermodynamically favorable, with a negative Gibbs free energy of −1.24 eV. These findings suggest that ATC with defect carbon sites can effectively initiate hydrogen abstraction from aliphatics, followed by aromatization, and that styryl radicals serve as more favorable hydrogen acceptors than styrene monomers. Overall, the results confirm that the formation of BTEC proceeds predominantly through radical-mediated, hydrogen-involving reaction pathways in the presence of ATC. Specifically, the defect sites generated through the activation approach are crucial for these reactions. Indeed, by subjecting the commercial activated carbon (AC) to the same activation treatment, we confirmed an 11% increase in the BTEC selectivity, as demonstrated in Fig. 6e. These results indicate that the synthesis of the carbocatalyst under optimal conditions can be extended to other carbon materials, such as commercial activated carbon, to achieve high catalytic activity toward aromatic production from mixed plastic waste.

3.9 Techno-economic analysis: conventional process vs. this study

Finally, to evaluate the economic viability and carbon footprint of our process, techno-economic analysis was performed for the hypothetical plant of 240 tonnes per day. Two scenarios were examined to assess the impact of energy efficiency and by-products on the minimum selling price (MSP). Scenario 1 (Fig. 10a) does not treat light gas as a by-product; instead, it is combined with the spent catalyst and is used as a combustion fuel. In contrast, Scenario 2 (Fig. S18) considers light gas as a by-product and uses only the spent catalyst as the combustion fuel. The optimal scenario was determined based on the lower MSP, which in this case is Scenario 1. As illustrated in Fig. 10a, the overall process flowsheet of Scenario 1 can be divided into five main sections: plastic pre-treatment, pyrolysis, product separation, combustion, and flue gas clean-up. Fig. 10b depicts the final mass flow, which was calculated according to the experimental observation for the pyrolysis of mixed plastics, the performance of the separation section, and the heat balance within the whole process. Clearly, our process can produce two major high-purity products including ethylbenzene (4.9 tonnes per h at 92 wt% purity, boiling point = 136 °C) and cumene (2.2 tonnes per h at 87 wt% purity, boiling point = 154.2 °C) along with the by-product of a liquid mixture of toluene and ethylbenzene (0.48 tonnes per h at 76 wt% purity) and 1.5 tonnes per h solid char (Table S29).

Fig. 10 Techno-economic analysis of fast pyrolysis of plastic waste to produce high-purity ethylbenzene and cumene. (a) Process flow diagram for Scenario 1. (b) Mass balance of inputs and outputs in the process. (c) Comparison of CO₂ emissions between this study and other processes producing the same chemicals. Symbols [1], [2], and [3] refer to ref. 77, 78 and 79, respectively, and ** refers to ref. 47. (d) Cost breakdown and MSP for Scenario 1. (e) Sensitivity analysis of MSP variation in Scenario 1.

Additionally, Fig. 10c and Table S30 confirmed a CO₂ emission rate of only 1.14 kg-CO₂/kg-products from our process, which is mostly from the combustion section, followed by the emission from the transportation of plastics from the materials recovery facility (MRF). Nevertheless, such a rate is comparable to and even slightly lower than the emission from virgin BTX production (1.2 kg-CO₂/kg-products),⁴⁷ and is significantly lower than the traditional naphtha cracking process for the production of ethylbenzene (1.3–5.2 kg-CO₂/kg-products), as estimated across multiple references 77–79. This is partially attributed to the moderate temperature required for fast pyrolysis and the simplified process design, including heat integration here. Furthermore, using plastic waste as feedstock, rather than conventional chemical materials such as naphtha, helps reduce CO₂ emissions, as plastic waste is generally not considered a carbon-intensive input. Finally, Fig. 10d shows a minimum selling price (MSP) of US$1.28 per kg for the most abundant product ethylbenzene, which falls in the ethylbenzene market price range of approximately US$1.00–1.40 per kg over the past five years.^9,80 This price is also predominated by the feedstock cost, accounting for nearly 50% of the total cost, followed by capital depreciation. Fig. 10e for the sensitivity analysis also confirmed the top three influential variables including feedstock cost, internal rate of return and plant size. By doubling the plant size, setting an internal rate of return of 15%, or reducing the feedstock price to US$0.32 per kg, the MSP of ethylbenzene could drop by a minimum of 20%. Accordingly, the MSP of ethylbenzene could fall below the lowest market price of the virgin product, making the whole process highly profitable. Detailed cost breakdowns for both scenarios can be found in Tables S31–S34.

In summary, this study demonstrates a facile process for the catalytic pyrolysis of mixed plastics into high-value aromatics, especially high-purity ethylbenzene and cumene, that are traditionally synthesized via multi-steps, the use of fossil fuel, and harsh conditions that are unsustainable. More intriguingly, upon a prior demineralization and chemical activation, the waste tire pyrolysis-derived char was demonstrated to be superior in delivering an unprecedentedly high yield and selectivity for aromatics. This is primarily attributed to the abundance of defects on its carbonaceous matrix, which favors the synergistic interactions between different plastics. The hydrogen abstraction and intermolecular transfer from aliphatic LDPE and PP were promoted on the Lewis acid sites, providing sufficient hydrogen for the hydrogenation of styrene and other radicals derived from PS. Additionally, the aromatization of the light aliphatic hydrocarbon derived from LDPE and PP was also catalyzed by the tire-derived carbocatalyst. Finally, through rigorous heat integration, the overall process demonstrated significantly low carbon emissions compared to conventional methods for ethylbenzene synthesis, while also exhibiting comparable economic viability. This work offers a promising pathway for transforming plastic and tire waste into valuable chemical feedstocks, laying the groundwork for modifying existing petrochemical operations and supporting the emergence of new industries focused on sustainable, circular waste-to-value technologies.

3.10 Future perspective

Overall, the results obtained to date are highly promising for the future scale-up of tire char valorization, an otherwise low-value by-product of waste tires. More importantly, these results establish a solid foundation for converting plastics into high-value chemicals such as ethylbenzene and cumene, whose production from fossil fuels remains carbon-intensive.

Moving forward, efforts will focus on deploying this technology through larger-scale and continuous-feeding tests. More specifically, to further explore and maximize the activity of tire char, additional studies are underway to compare carbocatalysts derived from different types of tires and from different rubber components within a tire (i.e., natural and synthetic rubbers), as well as to investigate reactor design and process integration.

In addition, although the optimal plastic waste composition identified in this study contains a high proportion of polystyrene (PS), future work will examine feedstocks with higher fractions of widely used plastics, such as polyolefins (PE). For this purpose, the properties of the carbocatalyst, including acidity, together with the reaction conditions, may need to be further fine-tuned to enhance the cyclization and aromatization reactions shown in Fig. 7(b).

Finally, the life cycle assessment (LCA) conducted in this work was limited to CO₂ emission calculations. A more comprehensive LCA addressing the full environmental impact, including tire char production, transportation, and final disposal, should be undertaken in future studies.

4 Conclusions

Based on a comprehensive study of the catalytic pyrolysis of three different plastics (PP, LDPE, and PS) using a carbocatalyst derived from tire char (ATC), the following conclusions were drawn:

1. Catalytic pyrolysis of PS at 550 °C using ATC increased BTEC (benzene, toluene, ethylbenzene, and cumene) selectivity from 31.2 area% (non-catalytic) to 87 area%. For mixed PP and LDPE, the use of ATC boosted monocyclic aromatic selectivity to over 30 area% at a catalyst-to-plastic ratio of 3 : 1. The catalyst is also relatively stable. In particular, post-regeneration by air treatment at 400 °C restored the BTEC selectivity to approximately 80 area% in the sixth cycle.

2. Co-pyrolysis of PS with PP and/or LDPE exhibited a strong synergistic effect in enhancing BTEC selectivity. PP or LDPE acts as a hydrogen donor, while aromatic radicals derived from PS serve as acceptors. This intermolecular hydrogen transfer takes place on the surface of the ATC carbocatalyst. The optimal feedstock ratio of PS : PP : LDPE = 6 : 1 : 1 achieved a BTEC selectivity of 90 area% with a product yield of 62 wt%.

3. Density functional theory (DFT) calculations confirm that the defect sites within the amorphous carbon structure of ATC play dual roles, facilitating both hydrogen abstraction and hydrogen transfer, which collectively enhance BTEC selectivity. The calculations further show that the formation of ethylbenzene and cumene via styryl and cumyl radicals is more favorable at these defect sites than through direct conversion from the styrene monomer. Moreover, these defect sites, which possess moderate hydrogen abstraction energies, favor the aromatization reaction, as evidenced by the more negative Gibbs free energies for the overall reaction pathway.

4. Techno-economic analysis (TEA) indicated that the minimum selling price (MSP) of ethylbenzene in the best-case scenario was estimated at US$1.28 per kg, making it competitive with its petroleum-derived counterparts. Sensitivity analysis showed that doubling the plant capacity, setting the internal rate of return to 15%, or lowering the feedstock cost to US$0.32 per kg could reduce the MSP by at least 20%, demonstrating strong economic viability. Moreover, the process could reduce CO₂ emissions by approximately 12–80% compared to conventional ethylbenzene production methods, based on values reported across multiple references.

Conflicts of interest

There are no conflicts of interest to declare.

Data availability

All the data supporting the conclusions of this study are available upon request.

Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d5gc06756a.

Acknowledgements

This project was supported by Tire Stewardship Australia (TSA) and the Australian Research Council Industrial Transformation Research Hub (IH230100011). The first author gratefully acknowledges Monash University for awarding the Indonesian Women in STEM Scholarship and the Faculty of Engineering for the tuition fee scholarship. The authors would like to thank Dr Anthony De Girolamo from the Department of Chemical and Biological Engineering and Mrs Made Ganesh Darmayanti from the School of Chemistry, Monash University, for their assistance with CHNS and NMR measurements, respectively. The use of the Monash X-ray Platform and the support provided by Ms Yvonne Hora are also gratefully acknowledged. In addition, the authors thank Dr Xi-Ya Fang and the Monash Centre for Electron Microscopy for their valuable technical support and contribution to the characterization work.

References

1. S. Huang, H. Wang, W. Ahmad, A. Ahmad, N. Ivanovich Vatin and A. M. Mohamed, et al., Plastic Waste Management Strategies and Their Environmental Aspects: A Scientometric Analysis and Comprehensive Review, Int. J. Environ. Res. Public Health, 2022, 19(8), 4556.
2. B. C. Gibb, Plastics are forever, Nat. Chem., 2019, 11(5), 394–395.
3. R. Hossain, M. T. Islam, A. Ghose and V. Sahajwalla, Full circle: Challenges and prospects for plastic waste management in Australia to achieve circular economy, J. Cleaner Prod., 2022, 368, 133127.
4. R. J. Conk, J. F. Stahler, J. X. Shi, J. Yang, N. G. Lefton and J. N. Brunn, et al., Polyolefin waste to light olefins with ethylene and base-metal heterogeneous catalysts, Science, 2024, 385(6715), 1322–1327.
5. G. W. Coates and Y. D. Y. L. Getzler, Chemical recycling to monomer for an ideal, circular polymer economy, Nat. Rev. Mater., 2020, 5(7), 501–516.
6. C. Zhang, G. Kwak, Y.-J. Lee, K.-W. Jun, R. Gao and H.-G. Park, et al., Light hydrocarbons to BTEX aromatics over Zn-modified hierarchical ZSM-5 combined with enhanced catalytic activity and stability, Microporous Mesoporous Mater., 2019, 284, 316–326.
7. J. Wang, J. Jiang, X. Wang, S. Pang, Y. Sun and X. Meng, et al., Enhanced BTEX formation via catalytic fast pyrolysis of styrene-butadiene rubber: Comparison of different catalysts, Fuel, 2020, 278, 118322.
8. P. Gaurh and H. Pramanik, Production of benzene/toluene/ethyl benzene/xylene (BTEX) via multiphase catalytic pyrolysis of hazardous waste polyethylene using low cost fly ash synthesized natural catalyst, Waste Manage., 2018, 77, 114–130.
9. Ethylbenzene Market Size and Share Analysis - Growth Trends and Forecasts (2025–2030). Mordor Intelligence, 2024; https://www.mordorintelligence.com/industry-reports/ethylbenzene-market (last accessed November 2025).
10. A. Samad, H. Saghir, I. Ahmad, F. Ahmad and H. Caliskan, Thermodynamic analysis of cumene production plant for identification of energy recovery potentials, Energy, 2023, 270, 126840.
11. A. Ayub, M. Zulkefal and H. Sethi, Artificial-Intelligence-Assisted Investigation of Quality and Yield of Cumene Production, Mater. Proc., 2024, 17(1), 10.
12. A. Samad, H. Saghir, I. Ahmad, F. Ahmad and H. Caliskan, Thermodynamic analysis of cumene production plant for identification of energy recovery potentials, Energy, 2023, 270, 126840.
13. M. R. Rahimpour, M. Jafari and D. Iranshahi, Progress in catalytic naphtha reforming process: A review, Appl. Energy, 2013, 109, 79–93.
14. F. Zhang, M. Zeng, R. D. Yappert, J. Sun, Y.-H. Lee and A. M. LaPointe, et al., Polyethylene upcycling to long-chain alkylaromatics by tandem hydrogenolysis/aromatization, Science, 2020, 370(6515), 437–441.
15. M. Dou, Z. Chen, S. Zhao, K. Ma, D. Yang and X. Ma, et al., One-Step Production of Ethylbenzene from Ethane and Benzene on the PtZnPr/ZSM-5 Catalyst, Ind. Eng. Chem. Res., 2024, 63(24), 10596–10607.
16. G. Buelna and T. M. Nenoff, A one-step catalytic separation process for the production of cumene, Catal. Lett., 2005, 102(3), 285–288.
17. S. Misaki, M. Takabatake, S. Hasegawa, Y. Manaka, W.-J. Chun and K. Motokura, Effect of zeolites on the alkylation of aromatics with alkanes using a Pd nanoparticle/solid acid cooperative catalytic system, Green Carbon, 2024, 2(3), 282–290.
18. G. Gałko, M. Rejdak, D. Tercki, M. Bogacka and M. Sajdak, Evaluation of the applicability of polymeric materials to BTEX and fine product transformation by catalytic and non-catalytic pyrolysis as a part of the closed loop material economy, J. Anal. Appl. Pyrolysis, 2021, 154, 105017.
19. X. Tian, Z. Zeng, Z. Liu, L. Dai, J. Xu and X. Yang, et al., Conversion of low-density polyethylene into monocyclic aromatic hydrocarbons by catalytic pyrolysis: Comparison of HZSM-5, Hβ, HY and MCM-41, J. Cleaner Prod., 2022, 358, 131989.
20. H. Xiao, J. Harding, S. Lei, W. Chen, S. Xia and N. Cai, et al., Hydrogen and aromatics recovery through plasma-catalytic pyrolysis of waste polypropylene, J. Cleaner Prod., 2022, 350, 131467.
21. Y. Han, J. Yu, T. Chen, X. Liu and L. Sun, Study on catalytic pyrolysis mechanism of natural rubber (NR) over Zn-modified ZSM5 catalysts, J. Energy Inst., 2021, 94, 210–221.
22. M. N. Akhtar, S. Riaz, N. Ahmad and E. A. Jaseer, Pioneering Aromatic Generation from Plastic Waste via Catalytic Thermolysis: A Minireview, Energy Fuels, 2024, 38(13), 11363–11390.
23. Z. Zhang, D. Li, J. Wang and J. Jiang, Cascade upcycling polystyrene waste into ethylbenzene over Fe2N @ C, Appl. Catal., B, 2023, 323, 122164.
24. R. Li, Z. Zhang, X. Liang, J. Shen, J. Wang and W. Sun, et al., Polystyrene Waste Thermochemical Hydrogenation to Ethylbenzene by a N-Bridged Co, Ni Dual-Atom Catalyst, J. Am. Chem. Soc., 2023, 145(29), 16218–16227.
25. J. Zhang, Z. Zhao, Z. Xia and L. Dai, A metal-free bifunctional electrocatalyst for oxygen reduction and oxygen evolution reactions, Nat. Nanotechnol., 2015, 10(5), 444–452.
26. X. Liu and L. Dai, Carbon-based metal-free catalysts, Nat. Rev. Mater., 2016, 1(11), 16064.
27. A. Primo, F. Neatu, M. Florea, V. Parvulescu and H. Garcia, Graphenes in the absence of metals as carbocatalysts for selective acetylene hydrogenation and alkene hydrogenation, Nat. Commun., 2014, 5(1), 5291.
28. D. R. Dreyer, H.-P. Jia and C. W. Bielawski, Graphene Oxide: A Convenient Carbocatalyst for Facilitating Oxidation and Hydration Reactions, Angew. Chem., Int. Ed., 2010, 49(38), 6813–6816.
29. B. Li and Z. Xu, A Nonmetal Catalyst for Molecular Hydrogen Activation with Comparable Catalytic Hydrogenation Capability to Noble Metal Catalyst, J. Am. Chem. Soc., 2009, 131(45), 16380–16382.
30. Y. Zhang, D. Duan, H. Lei, E. Villota and R. Ruan, Jet fuel production from waste plastics via catalytic pyrolysis with activated carbons, Appl. Energy, 2019, 251, 113337.
31. E. Huo, C. Liu, L. Xin, Y. Zhang, Y. Zhao and M. Qian, et al., Jet fuel range hydrocarbon production by co-pyrolysis of low density polyethylene and wheat straw over an activated carbon catalyst, Sustainable Energy Fuels, 2021, 5(23), 6145–6156.
32. W. Mateo, H. Lei, E. Villota, M. Qian, Y. Zhao and E. Huo, et al., Synthesis and characterization of sulfonated activated carbon as a catalyst for bio-jet fuel production from biomass and waste plastics, Bioresour. Technol., 2020, 297, 122411.
33. X. Lin, H. Lei, E. Huo, M. Qian, W. Mateo and Q. Zhang, et al., Enhancing jet fuel range hydrocarbons production from catalytic co-pyrolysis of Douglas fir and low-density polyethylene over bifunctional activated carbon catalysts, Energy Convers. Manage., 2020, 211, 112757.
34. S. Doja, L. K. Pillari and L. Bichler, Processing and activation of tire-derived char: A review, Renewable Sustainable Energy Rev., 2022, 155, 111860.
35. A. S. Al-Rahbi and P. T. Williams, Production of activated carbons from waste tyres for low temperature NOx control, Waste Manage., 2016, 49, 188–195.
36. H. Hu, Y. Fang, H. Liu, R. Yu, G. Luo and W. Liu, et al., The fate of sulfur during rapid pyrolysis of scrap tires, Chemosphere, 2014, 97, 102–107.
37. L. Qiu and G. Xu, Peak overlaps and corresponding solutions in the X-ray photoelectron spectroscopic study of hydrodesulfurization catalysts, Appl. Surf. Sci., 2010, 256(11), 3413–3417.
38. F. Shahid, M. Marshall and A. L. Chaffee, Active carbon monoliths from soft brown coal: A systematic study of their preparation, pore structure modification and characterization, J. Ind. Eng. Chem., 2024, 132, 424–436.
39. S. Yang, L. Shi, Q. Zhou, B. Qian, A. De Girolamo and L. Zhang, Elucidating the synergistic interaction and reaction pathway between the individual lignocellulosic components during flash pyrolysis, Chem. Eng. J., 2022, 432, 134372.
40. W. N. Dewi, Q. Zhou, M. Mollah, S. Yang, I. M. S. K. Ilankoon and A. Chaffee, et al., Synergistic interaction between scrap tyre and plastics for the production of sulphur-free, light oil from fast co-pyrolysis, Waste Manage., 2024, 179, 99–109.
41. Q. Zhou, J. Gu, J. Wang, A. De Girolamo, S. Yang and L. Zhang, High production of furfural by flash pyrolysis of C6 sugars and lignocellulose by Pd-PdO/ZnSO4 catalyst, Nat. Commun., 2023, 14(1), 1563.
42. G. Kresse and J. Furthmüller, Efficiency of ab initio total energy calculations for metals and semiconductors using a plane-wave basis set, Comput. Mater. Sci., 1996, 6(1), 15–50.
43. J. P. Perdew, K. Burke and M. Ernzerhof, Generalized Gradient Approximation Made Simple [Phys. Rev. Lett. 77, 3865 (1996)], Phys. Rev. Lett., 1997, 78(7), 1396.
44. V. Wang, N. Xu, J.-C. Liu, G. Tang and W.-T. Geng, VASPKIT: A user-friendly interface facilitating high-throughput computing and analysis using VASP code, Comput. Phys. Commun., 2021, 267, 108033.
45. V. L. Deringer, M. A. Caro, R. Jana, A. Aarva, S. R. Elliott and T. Laurila, et al., Computational Surface Chemistry of Tetrahedral Amorphous Carbon by Combining Machine Learning and Density Functional Theory, Chem. Mater., 2018, 30(21), 7438–7445.
46. M. Larrain, S. Van Passel, G. Thomassen, U. Kresovic, N. Alderweireldt and E. Moerman, et al., Economic performance of pyrolysis of mixed plastic waste: Open-loop versus closed-loop recycling, J. Cleaner Prod., 2020, 270, 122442.
47. G. Yadav, A. Singh, A. Dutta, T. Uekert, J. S. DesVeaux and S. R. Nicholson, et al., Techno-economic analysis and life cycle assessment for catalytic fast pyrolysis of mixed plastic waste, Energy Environ. Sci., 2023, 16(9), 3638–3653.
48. F. Riedewald, Y. Patel, E. Wilson, S. Santos and M. Sousa-Gallagher, Economic assessment of a 40,000 t/y mixed plastic waste pyrolysis plant using direct heat treatment with molten metal: A case study of a plant located in Belgium, Waste Manage., 2021, 120, 698–707.
49. J. Neethirajan, A. R. Parathodika, G.-H. Hu and K. Naskar, Functional rubber composites based on silica-silane reinforcement for green tire application: the state of the art, Functional Composite Materials, 2022, 3(1), 7.
50. S. Li, T. Q. Tran, Q. Li, B. Ji, A. S. Brand and W. Zhang, Zn leaching recovery and mechanisms from end-of-life tire rubber, Resour., Conserv. Recycl., 2023, 194, 107004.
51. Y. Zhang, Y.-P. Zhao, L.-L. Qiu, J. Xiao, F.-P. Wu and J.-P. Cao, et al., Insights into the KOH activation parameters in the preparation of corncob-based microporous carbon for high-performance supercapacitors, Diamond Relat. Mater., 2022, 129, 109331.
52. O. Oginni, K. Singh, G. Oporto, B. Dawson-Andoh, L. McDonald and E. Sabolsky, Influence of one-step and two-step KOH activation on activated carbon characteristics, Bioresour. Technol. Rep., 2019, 7, 100266.
53. J. Wang and S. Kaskel, KOH activation of carbon-based materials for energy storage, J. Mater. Chem., 2012, 22(45), 23710–23725.
54. J. Y. Huang, HRTEM and EELS studies of defects structure and amorphous-like graphite induced by ball-milling, Acta Mater., 1999, 47(6), 1801–1808.
55. Z. Lin, G. Shao, W. Liu, Y. Wang, H. Wang and H. Wang, et al., In situ TEM observations of the structural stability in carbon nanotubes, nanodiamonds and carbon nano-onions under electron irradiation, Carbon, 2022, 192, 356–365.
56. Y. Wang, Q. Zhang, L. Li, F. Wu, D. Geng and W. Hu, Turning Defects Into Advantages: Structures, Synthesis, and Applications of 2D Amorphous Carbon, Adv. Funct. Mater., 2025, 35, e09481.
57. W. N. Dewi, Q. Zhou, M. Mollah, S. Yang, I. M. S. K. Ilankoon and A. Chaffee, et al., Synergistic interaction between scrap tyre and plastics for the production of sulphur-free, light oil from fast co-pyrolysis, Waste Manage., 2024, 179, 99–109.
58. R. Miandad, M. A. Barakat, A. S. Aburiazaiza, M. Rehan and A. S. Nizami, Catalytic pyrolysis of plastic waste: A review, Process Saf. Environ. Prot., 2016, 102, 822–838.
59. S. Budsaereechai, A. J. Hunt and Y. Ngernyen, Catalytic pyrolysis of plastic waste for the production of liquid fuels for engines, RSC Adv., 2019, 9(10), 5844–5857.
60. R. Miandad, M. A. Barakat, M. Rehan, A. S. Aburiazaiza, I. M. I. Ismail and A. S. Nizami, Plastic waste to liquid oil through catalytic pyrolysis using natural and synthetic zeolite catalysts, Waste Manage., 2017, 69, 66–78.
61. R. Bagri and P. T. Williams, Catalytic pyrolysis of polyethylene, J. Anal. Appl. Pyrolysis, 2002, 63(1), 29–41.
62. P. Gaurh and H. Pramanik, A novel approach of solid waste management via aromatization using multiphase catalytic pyrolysis of waste polyethylene, Waste Manage., 2018, 71, 86–96.
63. U.-e.-S. Amjad, A. Tajjamal, A. Ul-Hamid, A. Faisal, S. A. H. Zaidi and L. Sherin, et al., Catalytic cracking of polystyrene pyrolysis oil: Effect of Nb2O5 and NiO/Nb2O5 catalyst on the liquid product composition, Waste Manage., 2022, 141, 240–250.
64. H. Li, X. Chen, P. Fu, B. Tang, X. Zhuansun and Z. Sun, et al., Synthesis of metal and nitrogen co-doped activated carbon catalysts for the co-production of monocyclic aromatics and hydrogen-rich gas from the pyrolysis of biomass and plastic, Energy, 2025, 316, 134652.
65. B. Valizadeh, J. Lee and Y.-K. Park, Enhanced generation of light olefins and aromatic-enriched oil via catalytic pyrolysis of instant waste plastic rice containers over activated biochar, J. Anal. Appl. Pyrolysis, 2025, 186, 106905.
66. H. Lin, L. Zhu, Y. Liu, V. S. Vladimirovich, B. Oboirien and Y. Zhou, Plastic upgrading via catalytic pyrolysis with combined metal-modified gallium-based HZSM-5 and MCM-41, Funct. Compos. Mater., 2024, 18(11), 125.
67. L. Kong, J. Wang, K. Dong, Z. Sun, B. Tang and N. Zhao, et al., Catalytic pyrolysis characteristics of polystyrene by biomass char-supported nanocatalysts, J. Anal. Appl. Pyrolysis, 2024, 179, 106511.
68. H. Shafaghat, S. Gulshan, A.-C. Johansson, P. Evangelopoulos and W. Yang, Selective recycling of BTX hydrocarbons from electronic plastic wastes using catalytic fast pyrolysis, Appl. Surf. Sci., 2022, 605, 154734.
69. Q. Zhou, S. Yang, H. Wang, Z. Liu and L. Zhang, Selective deoxygenation of biomass volatiles into light oxygenates catalysed by S-doped, nanosized zinc-rich scrap tyre char with in situ formed multiple acidic sites, Appl. Catal., B, 2021, 282, 119603.
70. B. Wang, G. Liu, X. Deng, Z. Deng, W. Lin and Z. Li, Replacement of Pd nanoparticles: Hydrogenation promoted by frustrated Lewis acid-base pairs in carbon quantum dots, J. Catal., 2020, 383, 304–310.
71. B. Pukánszky, J. P. Kennedy and T. Kelen, Tüdős F. Cationic reactions in the melt, Polym. Bull., 1981, 5(8), 469–476.
72. C. Marquez, C. Martin, N. Linares and D. De Vos, Catalytic routes towards polystyrene recycling, Mater. Horiz., 2023, 10(5), 1625–1640.
73. M. Marczewski, E. Kamińska, H. Marczewska, M. Godek, G. Rokicki and J. Sokołowski, Catalytic decomposition of polystyrene. The role of acid and basic active centers, Appl. Catal., B, 2013, 129, 236–246.
74. O. D. Bozkurt and H. E. Toraman, Conversion of Polypropylene into Light Hydrocarbons and Aromatics by Metal Exchanged Zeolite Catalysts, Langmuir, 2024, 40(18), 9636–9650.
75. X. Huang, H. Li, W.-D. Xiao and D. Chen, Insight into the side reactions in methanol-to-olefin process over HZSM-5: A kinetic study, Chem. Eng. J., 2016, 299, 263–275.
76. G. Giannetto, A. Montes, N. S. Gnep, A. Florentino, P. Cartraud and M. Guisnet, Conversion of Light Alkanes into Aromatic Hydrocarbons.VII. Aromatization of Propane on Gallosilicates: Effect of Calcination in Dry Air, J. Catal., 1994, 145(1), 86–95.
77. S. R. Nicholson, N. A. Rorrer, T. Uekert, G. Avery, A. C. Carpenter and G. T. Beckham, Manufacturing Energy and Greenhouse Gas Emissions Associated with United States Consumption of Organic Petrochemicals, ACS Sustainable Chem. Eng., 2023, 11(6), 2198–2208.
78. . Emission factor (ethyl benzene production). https://www.climatiq.io/data/emission-factor/1ccd1b8d-7eaa-4122-8d34-301b1328180c, 2023, last accessed February 2026.
79. P. Brian and J. W. M. Flannery, Greenhouse Gas Index for Products in 39 Industrial Sectors: Petrochemicals, https://www.rff.org/publications/working-papers/the-greenhouse-gas-index-for-products-in-39-industrial-sectors/, 2022, last accessed November 2025.
80. . Ethylbenzene Price Trend and Forecast. https://www.procurementresource.com/resource-center/ethylbenzene-price-trends, 2024, last accessed November 2025.

---