Catalytic Pyrolysis Process to Produce Styrene from Waste Expanded Polystyrene Using a Semi-Batch Rotary Reactor

Abstract

Thermal and catalytic pyrolysis of waste expanded polystyrene (WEPS) was studied to obtain mainly styrene monomer, which can be recycled in the polystyrene industry. Initially, preliminary experiments were carried out in a static semi-batch glass reactor with basic catalysts and without catalysts, using toluene as solvent at 250 °C, determining their styrene yields to select the best catalyst. MgO turned out to be the best catalyst due to its stability and cost. This catalyst was characterized by XRD, BET area, SEM-EDS, Raman spectroscopy, UV–VIS, and TGA. The kinetic equation for WEPS pyrolysis in the glass reactor was determined as a first-order reaction. The heat of reaction, the Gibbs free energy change, and the entropy change were calculated. Finally, WEPS pyrolysis experiments were carried out using a rotating semi-batch steel reactor, at higher temperatures and without using solvents, evaluating the styrene yield and its performance for its possible industrial application. In this reaction, the activity remained almost constant after four catalyst regenerations. The best styrene yield was 94 wt%, which could be one of the highest reported in the literature. This result may be associated with the back-mixing obtained in the rotary reactor, in contrast to the performance observed in the static glass reactor.

1. Introduction

Waste expanded polystyrene (WEPS) is abundant in plastic waste floating in the oceans. Like most synthetic polymers, WEPS degrades very slowly by the UV radiation effect from the sun, producing smaller and smaller particles, up to micro- or nano-sized polymer fragments, which can be easily ingested by marine fauna, such as mussels, oysters, fish, seabirds, and whales. Ingestion of plastic microparticles can harm animals or microorganisms [1,2,3,4,5,6]. Microscopic plastic particles in fish and wildlife are massive and well documented.

Lebreton et al. [7] conducted a study of the so-called “plastic island” or “great Pacific garbage patch” (GPGP), located between California and Hawaii in the North Pacific Ocean. In this investigation, the authors estimated that the patch area comprises 1.6 million km² with a concentration of plastics between 10 and 100 kg/km², as well as predicting that it contains a total of 1.8 × 10¹² plastic parts weighing 78,909 tons, composed of fractions categorized into four size classes: microplastics (0.05–0.5 cm), mesoplastics (0.5–5 cm), macroplastics (5–50 cm), and megaplastics (>50 cm). For microplastics, they estimated 6380 tons (8.08 wt%); for mesoplastics, 9971 tons (12.64 wt%); for macroplastics, 20,195 tons (25.6 wt%); and finally, for megaplastics, 42,362 tons (53.68 wt%). In addition, in this study, it was estimated that the plastic material in the form of foam, mainly WEPS, is constituted as follows: a microplastic fraction of 2 tons, a mesoplastic fraction of 8 tons, and a macroplastic fraction of 24.8 tons, making a total of 34.8 tons of total weight. The average mass concentration of plastics measured within the patch limits showed an exponential increase in recent decades, from an average of 0.4 kg/km² in the 1970s to 1.23 kg/km² in 2015.

In the Solid Waste Inventory 2020 prepared by the Secretariat of Environment (SEDEMA) of the Government of Mexico City [8], it was reported that during this year, 12,306 tons of municipal solid waste were generated daily. On the other hand, in the Basic Diagnosis for the Integral Management of Solid Waste 2020 published by the Ministry of Environment and Natural Resources (SEMARNAT) of the Mexican Federal Government [9], it is specified that the percentage by weight average of WEPS is 1.55% of the total solid waste generated in Mexico. Therefore, it is estimated that around 190.74 tons of WEPS originated daily in Mexico City in 2020, of which only 0.09 tons/day were recycled, as reported in the 2020 Solid Waste Inventory mentioned above.

Chaukura et al. [10] conducted a review of the WEPS potential uses, and they mention that it can be reprocessed for recycling; it can also be reused to obtain polymer-clay nanocomposites; high-value-added products such as adhesives, paints, superabsorbents, polystyrene nanofibers, flocculants, and ion exchangers; and the styrene monomer generation through a pyrolysis process. The present work proposes this last method to take advantage of WEPS by producing its styrene monomer through a pyrolysis process and thus avoid its final disposal in landfills, open-air dumps, or the oceans.

Several studies have been carried out related to the WEPS pyrolysis, using different catalysts or without a catalyst; in some of these investigations, the pyrolysis was carried out directly inside a furnace, using a few milligrams of polystyrene [11,12,13]; in other studies, different reactors were used, such as a simple flask [14], semi-continuous reactors, or either glass or stainless steel [15,16,17,18,19,20,21,22,23,24,25,26,27], or more complex continuous reactors, such as the autothermal fixed bed reactor [28] or the conical spouted bed reactor [29]. From the review of these studies, it was highlighted that when using acid solids as catalysts, high amounts of benzene, non-condensable gases, and polycyclic aromatic hydrocarbons are obtained; the latter can be carcinogenic [15]. On the contrary, using basic solids as catalysts results in higher styrene yield and fewer non-condensable gases. Therefore, the present study aims to obtain styrene preferably because it has a higher value-added benefit than benzene, since, for styrene production, benzene and ethylene are used as raw materials. For this reason, basic solids were used as catalysts.

Regarding the kinetics of WEPS pyrolysis, several studies have been carried out [30,31,32,33,34].

Preliminary experiments were carried out to select the best catalyst, and with this catalyst, experiments were carried out in a rotary reactor to evaluate the feasibility of its application on an industrial scale, for which the catalyst regeneration was also studied to be reused in the WEPS pyrolysis process. The rotary reactor performance in the WEPS pyrolysis has not been reported previously. In this reactor, it was crucial to avoid leaks, which were controlled with a low rotation speed (1.6 rpm) and with the installation of a mechanical seal and an expansion joint at the outlet of the pyrolysis vapors.

2. Materials and Methods

For this investigation, WEPS was collected from used disposable cups and plates that were washed, as well as from discarded packaging material—this type of WEPS is usually clean, so it was not necessary to wash it. Subsequently, a size reduction of the WEPS collected was carried out to obtain fragments between 0.5 and 2 cm in a commercial crusher (Lasser Mills, manufactured in Mexico City, Mexico). In this way, WEPS was prepared to make it easier to dissolve in the solvent.

2.1. WEPS Solubility Experiments

Due to the WEPS’s low density (0.012 g/mL), it was necessary to reduce its volume, dissolving it in a suitable solvent. Therefore, WEPS was dissolved in toluene, orange essential oil, and a 50% toluene/50% orange oil mixture. For this purpose, previously weighed pieces of WEPS were added to 100 mL of solvent until no more material was dissolved, and by weight difference between what was disbanded and what was not dissolved, the amount of WEPS that was dissolved was determined.

The criteria to select the best solvent were as follows: that WEPS had good solubility in the solvent, that it was not toxic (or carcinogenic), and that it had low cost and availability in the market.

By dissolving WEPS in a solvent while reducing its volume, the trapped air inside WEPS is also eliminated, which is necessary for the pyrolysis process since it must be performed entirely without oxygen.

2.2. Determination of the WEPS Average Molecular Weight

To carry out the WEPS characterization, its average molecular weight was estimated by measuring the viscosity of the WEPS solutions in toluene at different concentrations at 25 °C with the help of a Brookfield viscometer (relative error: ±1%). As is known [35], the viscosity data can be used to calculate the WEPS average molecular weight using the Mark–Houwink equation:

[η] = KM^α

(1)

where [η] = intrinsic viscosity, M = average molecular weight, and K and α are constants for a particular polymer–solvent system. According to Gowariker et al. [35], the constants K and α for the polystyrene-toluene system have the following values: K = 11 × 10⁻³ mL/g and α = 0.725. The intrinsic viscosity is given by the equation:

[ƞ] = (ln ƞ_r/C)_C→0

(2)

where C is the concentration in g/mL and η_r is the relative viscosity which is given by the equation:

η_r = µ/µ₀

(3)

where μ is the viscosity of the WEPS solutions in toluene, as determined above, and µ₀ is the viscosity of the pure solvent. If ln (η_r)/C is plotted against concentration C (in g/mL), the intrinsic viscosity [η] will be given by the y-intercept of this graph.

2.3. Pyrolysis Experiments in the Glass Reactor without Stirring

Preliminary WEPS pyrolysis experiments were performed without catalysts, and with MgO and calcined dolomite as catalysts; they were named G0, G1, and G2, respectively. These variables were kept constant: the external temperature at 400 °C, the internal temperature at 250 °C [14], WEPS: catalyst ratio 10:1 by weight [20], and N₂ flow at 0.1 L/min as carrier gas. In all the experiments, 25 g of WEPS dissolved in 100 mL of toluene at atmospheric pressure was used in the semi-batch experimental equipment, as shown in Figure 1.

Number 1 corresponded to the entry of N₂ as the carrier gas; its flow was measured with the rotameter marked with number 2 (Cole Parmer, Vernon Hills, IL, USA, from 0.1 to 0.5 L/min). Number 3 corresponded to the glass reactor with a capacity of 1 L. The reactor was heated with an electrical resistance (number 4). Number 5 corresponded to the reactor lid, which had four mouths. A neoprene and Teflon gasket was placed between the glass reactor and the cover to prevent leaks. A manometer (Metron, Reston, VA, USA) was placed in one of the lid mouths with a scale from 0 to 2 kg/cm². Another of the mouths of the lid was the carrier gas inlet to the reactor (red). A thermowell was placed in another of the mouths of the lid to indicate the temperature inside the reactor. The last mouth allowed the exit of the pyrolysis vapors.

A thermocouple was connected from the electrical resistance to the temperature controls (number 6) to maintain the heating temperature at the established value, which consisted of a rheostat, a digital electronic pyrometer, and a relay that constituted an automatic system.

The pyrolysis vapors were condensed with three glass refrigerants (numbers 7). The liquid products were recovered in the glass separator–accumulator (separation funnel, number 8), where it was possible to take fractions of liquid products at a given time using the valve at the bottom.

The cooling system (number 9) consisted of a vessel containing a mixture of water and methanol (50%), to which dry ice was added, and, in this way, the temperature of the cooling fluid was reduced to −5 °C. A submersible pump was installed to re-circulate the cooling fluid. The non-condensable gases exited the system along with the carrier gas through a hose towards the atmosphere (number 10, green).

The purpose of carrying out these preliminary experiments was to select the catalyst on the basis of the one that provided the highest styrene yield at the lowest cost.

The liquid product of WEPS pyrolysis, using MgO as a catalyst, was analyzed with an infrared spectrometer (Perkin Elmer, Frontier model, Waltham, MA, USA). The composition of the liquid products in each experiment was determined using a Varian CP-3380 gas chromatograph, with a flame ionization detector and a 30 m × 0.25 mm ID × 0.25 μm capillary column of polyethylene glycol as stationary phase (relative error: ±4.5%). These are the analysis conditions recommended by the standard method ASTM D5135-16.

An experiment was carried out at a temperature of 400 °C with the catalyst that provided the highest styrene yield, called G3.

2.4. WEPS Pyrolysis Kinetics

The kinetics of the WEPS pyrolysis reaction was determined by evaluating the weight of the liquid products as a function of time in the setup of Figure 1. The WEPS weight loss was determined by subtracting the weight of the liquid product obtained in a measured time from the weight of WEPS initially charged to the reactor. WEPS dissolved in toluene was loaded into the reactor (number 3). The pyrolysis reaction started when approximately 200 °C was reached, the temperature at which the solvent had already evaporated (the toluene boiling point is 111 °C). It was observed that at 117 °C, most of the toluene added as solvent was eliminated, and thus this moment was considered as the start of the reaction (t = 0 min) for kinetic determination purposes. All the distilled solvent collected in the container (number 8) was extracted.

After this step, the pyrolysis reaction time (t) began, and vapors from thermal or catalytic pyrolysis began to form and condense in the gas–liquid accumulator–separator (number 8). Then, sampling was performed every 10 min, removing the liquid product accumulated at each time in the separation funnel number 8, which was analyzed by gas chromatography and weighed (m) on an analytical balance (Sartorius Basic).

The WEPS pyrolysis kinetics was determined on a weight basis to express the kinetic equation as reported in the literature. The kinetics equation for a heterogeneous system can be stated as [19]:

−dm/dt = kmⁿ

(4)

where m = WEPS mass in a given reaction time (g), t = reaction time (min), k = specific reaction constant, and n = reaction order. In various studies of the WEPS pyrolysis [19,29,30,31,32,33,34], it has already been reported that the WEPS pyrolysis kinetics is of the first order; if n = 1, Equation (4) can be integrated as follows:

$$-{{\displaystyle \int }}_{{\mathrm{m}}_0}^{\mathrm{m}}\mathrm{dm}/\mathrm{m}=\mathrm{k}{{\displaystyle \int }}_0^{\mathrm{t}}\mathrm{dt}$$

(5)

Solving this integral, Equation (6) was obtained, which was used to obtain k with the experimental data:

ln (m₀/m) = kt

(6)

2.5. Thermodynamic Analysis of the WEPS Pyrolysis Reaction

The change in enthalpy of reaction for the WEPS pyrolysis is given by the equation:

$$\Delta {\mathrm{H}}_{\mathrm{r}}=\sum \left\{{\mathrm{n}}_{\mathrm{P}}\right[{\mathrm{H}}_{\mathrm{f}}°+{{\displaystyle \int }}_{298}^{\mathrm{T}}\mathrm{Cp}\mathrm{dT}]{\}}_{\mathrm{P}}-{\mathrm{n}}_{\mathrm{PS}}[{\mathrm{H}}_{\mathrm{f}}°+{{\displaystyle \int }}_{298}^{\mathrm{T}}\mathrm{Cp}\mathrm{dT}{]}_{\mathrm{PS}}$$

(7)

where n_P = number of moles of product P, n_PS = number of moles of WEPS, H_f° = enthalpy of formation, and C_p = heat capacity.

The change in Gibbs free energy of the WEPS pyrolysis reaction is given by the equation [33]:

ΔG_r = Ea + RT ln ((K_B T)/(h A))

(8)

where K_B = Boltzmann’s constant (1.38066 × 10⁻²³ J/°K), h = Planck’s constant (6.62608 × 10⁻³⁴ J s), Ea = activation energy (J/mol), and A = pre-exponential factor (s⁻¹).

The change in entropy of the WEPS pyrolysis reaction can be calculated from the Gibbs free energy definition:

ΔS_r = (ΔH_r − ΔG_r)/T

(9)

The necessary data to realize the calculation of these parameters were obtained from the bibliography [36,37,38].

2.6. Pyrolysis Experiments in the Semi-Batch Rotary Reactor

To carry out experiments, a pilot plant-type semi-batch rotary reactor was built with a length of 24.5 cm long by 16 cm in internal diameter, with a capacity of 4.9 L; it was made of carbon steel, with a mechanical seal at the outlet of the gaseous products, as shown in Figure 2, and can be seen in the images of Figure A1 and Figure A2. A thermowell with a sliding thermocouple was built to measure the temperature in the center of the reactor axially, which was connected to an electronic temperature indicator. The cylindrical reactor was placed on eight bearings that allowed its free rotation. A 1/4 HP motor and a speed reducer, with a speed of 1.6 rpm, were installed to provide rotation of the reactor through a chain. At the reactor outlet, a stainless-steel condenser was connected to recover the liquid products. Between the mechanical seal and the condenser, a pressure gauge was installed to monitor the system pressure and an expansion joint to prevent leaks due to the thermal expansion of the connections. The most critical aspect of the operation of this reactor was leak control. It was observed that to control leaks, it was better to use a low rotation speed and install a mechanical seal and an expansion joint in the gas outlet pipe.

A liquid–gas separator tank was installed at the condenser outlet to recover the liquid phase. The liquid products accumulated at the bottom, making it possible to collect samples at different times of the pyrolysis reaction by the valve placed at the bottom of the liquid–gas separator tank.

The reactor was heated with direct fire, for which two burners were installed, one for LP gas and the other for the non-condensable gaseous pyrolysis products. A volumetric gas meter was placed to quantify the production of non-condensable gases from pyrolysis; at the outlet of this meter, the gases were sent to one of the two burners. Likewise, another volumetric gas meter was installed to quantify the consumption of LP gas.

A glass fiber insulation with a thickness of 4 cm was placed around the reactor to prevent heat loss. The cooling system consisted of a 1/2 HP pump that sucked the water contained in a vessel and recirculated it to the condenser.

To prevent product leaks, in all experiments, a Teflon gasket was placed on the reactor lid and sealed with red silicone, which is resistant to high temperatures. Unfortunately, a fraction of the Teflon gasket burned and degraded, so changing the Teflon gasket in each experiment was necessary.

All WEPS pyrolysis experiments were performed with a WEPS: catalyst ratio of 10:1 by weight [20]. In each experiment, 460 g of WEPS were loaded into the rotating reactor, previously melted to reduce its volume and eliminate the trapped air. Neither solvent nor carrier gas (N₂) was used to make the process more cost-effective and sustainable.

The experimentation started with the first WEPS pyrolysis experiment without a catalyst—it was called R-1 (

Table 1. Summary of experiments carried out in the semi-batch rotary reactor.

Experiment	Features
R-1	No catalyst
R-2	Fresh catalyst
R-3	Fresh catalyst
R-4	Fresh catalyst
R-5	Fresh catalyst, improved insulation
R-6	Regenerated catalyst, improved insulation
R-7	Two times regenerated catalyst, improved insulation
R-8	Three times regenerated catalyst, improved insulation
R-9	Four times regenerated catalyst, improved insulation
R-10	No catalyst, improved insulation

). Subsequently, experiments R-2, R-3, and R-4 were carried out using a fresh MgO catalyst. In all experiments, the reaction time, including heating from 25 °C, was 3 h. The samples of liquid products obtained in each experiment, taken every 10 min, were analyzed with the Varian CP-3380 gas chromatograph mentioned above.

Coatings were placed both at the front and rear of the rotating reactor to avoid lateral heat loss, for which insulation made of steel mesh, fiberglass, and cement was used (Figure A2). An 8.5 cm thick ceramic fiber layer was also placed around the reactor, in addition to the 4 cm thick fiberglass insulation. With these improvements, experiment R-5 and the following experiments were carried out.

In order to investigate whether good results were obtained using the same regenerated MgO catalyst, the solid residues from the previous experiments were calcined at 600 °C for 4 h in order to remove the carbon coating from the MgO catalyst that remained at the end of the pyrolysis, and the catalyst thus regenerated was used in experiment R-6.

The spent catalyst from experiment R-6, in turn, was calcined, and the regenerated catalyst was used in experiment R-7. Then, the solid residue from run R-7 was calcined for the third time, and the regenerated catalyst was applied in experiment R-8. Finally, the solid residue from run R-8 was calcined for the fourth time, and the regenerated catalyst was used in experiment R-9. In this last experiment, the non-condensable gases generated in the pyrolysis were stored in a balloon for their corresponding analysis. A Gow Mac gas chromatograph (series 750) with a 1/8″ packed column of dimethylpolysiloxane as a stationary phase was used because light hydrocarbons were obtained as non-condensable gases. A second experiment without a catalyst was carried out (R-10). All these experiments with the rotary reactor are summarized in Table 1.

2.7. Catalyst Characterization

2.7.1. N₂ Physisorption (BET Area)

The N₂ adsorption–desorption isotherm of the fresh MgO catalyst was determined with NOVA Quantachrome Instruments version 11.0 equipment, (Anton Paar group, Boynton Beach, FL, USA) using the physisorption of N₂ at 77 °K (relative error: ± 0.1%). Before adsorption, the sample received a degassing pretreatment at 300 °C for 5 h under a vacuum of 1 × 10⁻⁴ torr. The sample weight was 0.2482 g, and the sample volume was 0.06708 cm³. This analysis determined the BET-specific area, pore volume, and pore diameter distribution by applying the standard BET model [39]. The T-plot method was used to quantify the pore volume (Vp) of the samples [40]. The pore diameter distribution was obtained from the BJH model [41] using isothermal desorption and assuming the geometry of the pores as cylindrical.

2.7.2. X-ray Diffraction

XRD diffractograms of MgO catalyst, before and after the pyrolysis reaction using the glass reactor, were obtained with a BRUKER diffractometer, model D8 Focus, made in Karlsruhe, Germany, equipped with a Cu tube anode (30 kV, 20 mA), using CuKα radiation. The scan was performed in the range of 2 θ~10–100°, with a rate of 2°/min (relative error: ± 4.5%).

2.7.3. Scanning Electron Microscopy (SEM-EDS)

Images of the MgO catalyst microstructure were taken employing scanning electron microscopy (SEM), before and after the pyrolysis reaction using the glass reactor, with field emission and high resolution in a JEOL microscope (model JFM-6701-F, Tokyo, Japan), using secondary electrons (relative error: ±0.5%).

2.7.4. Raman Spectroscopy

Samples of the spent catalyst from the pyrolysis experiments carried out with the rotary reactor, as well as the regenerated catalyst, were analyzed by Raman spectroscopy, using Renishaw In Via Raman equipment, (New Mills, Gloucestershire, UK) a lamp with a wavelength of λ = 532 nm, and a power of 100%; the equipment was calibrated with a Si wafer at 520 cm⁻¹.

2.7.5. UV–VIS Spectroscopy

Samples of the solid residues from the pyrolysis experiments produced with the rotary reactor, as well as their respective regenerated catalysts, were also analyzed by UV–VIS spectroscopy with a Varian Cary 100 spectrometer, manufactured in Mulgrave, Australia; the equipment was calibrated with a polytetrafluoroethylene (PTFE) disc.

2.7.6. Thermogravimetric Analysis (TGA)

Samples of the spent catalyst from the pyrolysis experiments carried out with the rotary reactor were analyzed by TGA using SDT Q600 TA Instruments equipment (Artisan Technology Group, Champaign, IL, USA) with a sample weight of between 5 and 8 mg and an airflow of 10 mL/min (relative error: ±0.5%).

2.7.7. Calcination of Samples

The MgO catalyst samples, before and after the pyrolysis reaction, were calcined to determine the carbon content to evaluate the relevance of the catalyst regeneration. Calcination of the fresh MgO catalyst and the solid residues from experiments was carried out at 600 °C for 4 h; the sample weight was 1 g.

3. Results and Discussion

3.1. WEPS Solubility Experiments

The results of the experiments on the solubility of WEPS in toluene, orange oil, and a mixture of 50% orange oil/50% toluene were 0.6018, 0.4163, and 0.5633 g/mL of solvent, respectively. The result of the solubility of WEPS in toluene was similar to that obtained by García et al. [42]; they reported 0.6 g/mL.

Toluene was found to be the most suitable solvent to dissolve WEPS since it offered the highest solubility among the solvents tested; the safety sheet for toluene mentions that there is no evidence that it is carcinogenic, and that toluene has relatively low cost and good availability in the market.

3.2. Determination of the WEPS Average Molecular Weight

As can be seen in Figure 3, the viscosity increased exponentially with the concentration of WEPS in toluene, which is in agreement with the results reported by Kol et al. [43].

Using the data on the viscosities of the WEPS solutions in toluene at different concentrations, the graph in Figure 4 was obtained, ln (η_r)/C against WEPS concentration in toluene in g/mL. The intrinsic viscosity was the y-intercept, so [η] = 39.285 mL/g.

Applying the Mark–Houwink Equation (1) and using the constants reported by Gowariker et al. [35], K = 11 × 10⁻³ mL/g and α = 0.725, resulted in a WEPS average molecular weight of M = 79,517.15 g/mol.

3.3. Pyrolysis Experiments in the Glass Reactor without Stirring

The infrared (IR) spectrum obtained from the WEPS pyrolysis liquid products of experiment G1, carried out in the glass reactor, using MgO as a catalyst, is shown in Figure 5. The functional group characterization of this spectrum was as follows: (I) Bands between 3030–3080 cm⁻¹ and 910–990 cm⁻¹ corresponded to the vinyl group, characteristic of styrene. (II) The band in 1631 cm⁻¹ suggests the presence of the C=C bond. (III) Bands between 730–770 cm⁻¹ and small bands between 1700–1900 cm⁻¹ indicate the presence of monosubstituted benzene. (IV) Bands between 1450 and 1500 cm⁻¹ correspond to the presence of aromatic groups. (V) Bands between 2930 and 2980 cm⁻¹ characterize the C-H groups. (VI) The band in 1580 cm⁻¹ is attributed to substitution on the benzene ring. The presence of the vinyl group and monosubstituted groups in the benzene ring account for the presence of styrene and other aromatics such as toluene and ethylbenzene.

The styrene yield obtained without a catalyst, with MgO and calcined dolomite as catalysts at 250 °C, and the experiment G3 at a temperature of 400 °C with the selected catalyst are detailed in

Table 2. Styrene yields obtained from the WEPS pyrolysis experiments in the static glass reactor.

Experiment	Catalyst	Temperature (°C)	Styrene Yield (wt%)
G0	Without catalyst	250	23.27
G1	MgO	250	56.29
G2	Calcined dolomite	250	50.69
G3	MgO	400	66.42

.

In the preliminary experiments G0 without catalysts and G1 with an MgO catalyst carried out with the glass reactor at 250 °C (Table 2), a higher styrene yield was obtained with the MgO catalyst, of 56.29 wt% against 23.27 wt% without a catalyst. By comparison, Zhang et al. [20] reported a styrene yield of 56.07 wt% without a catalyst and a slightly higher styrene yield of 62.96 wt% with an MgO catalyst at 350 °C. Furthermore, as can be seen in Figure 6, the styrene yield as a function of time was much higher in the preliminary experiment G1, carried out with an MgO catalyst, compared to experiment G0, carried out without a catalyst.

Zhang et al. [20] used BaO as a catalyst, resulting in a higher styrene yield; they obtained 71.35 wt% at 350 °C. However, BaO costs about five times more than MgO (Aldrich). Therefore, it was decided to use MgO as the most appropriate catalyst for WEPS pyrolysis.

In the catalytic degradation of polystyrene using solid acids, considerable amounts of benzene and ethylbenzene are formed [20]. These products are partially attributed to the subsequent degradation and hydrogenation of the styrene produced, which results in a decrease in the styrene fraction in oils obtained. The production of benzene and indane derivatives is also one of the characteristics of the oils produced by solid acid catalysts, which were not detected in oils obtained with solid base catalysts. In a mechanism model proposed by Zhang et al. [20], the formation of benzene and indane derivatives with solid acid catalysts begins with a proton attack on the branched phenyl group to produce a π-complex cation, which is converted to a σ-bonded complex cation that is released as benzene, as shown in Scheme 1:

On the other hand, with basic catalysts, the fraction of styrene in the produced oils increased to about 75 wt% and the fraction of styrene containing the monomer and dimer to about 90 wt%. Since benzene and indane derivatives were not detected in oils produced with solid bases nor in those obtained with simple thermal degradation of polystyrene, and since a considerable amount of styrene dimer was observed in these oils, it is considered that the depolymerization of polystyrene using solid bases proceeds in a similar way to simple thermal pyrolysis. Thermal degradation of polystyrene begins with the random formation of polymeric radicals, while catalytic degradation with solid bases can begin with the formation of carbocations by removal of a hydrogen atom from polystyrene adsorbed on basic sites, according to the reaction mechanism model proposed by Zhang et al. [20], which can be seen in Scheme 2:

3.4. WEPS Pyrolysis Kinetics

WEPS weight loss over time in the experiment without a catalyst (G0), performed in the glass reactor at a temperature of 250 °C, is shown in Figure 7a. If it is graphed ln (m₀/m) against time t, according to Equation (6), and Figure 7b results, where it can be seen that it is very close to a straight line. Fitting to a straight line using the least squares method resulted in a slope of k = 0.0146 min⁻¹, with a correlation coefficient of R² = 0.9965.

In the WEPS catalytic pyrolysis experiment G1, in which MgO was used as a catalyst with the glass reactor at a temperature of 250 °C (Figure 8a), a similar decrease as a function of time was observed as in the case of the reaction without a catalyst. By applying Equation (6), the straight line of Figure 8b was obtained; using the least squares method, the slope gave k = 0.0156 min⁻¹, with a correlation coefficient of R² = 0.9927. Therefore, it was corroborated that the WEPS pyrolysis reaction follows first-order kinetics, using toluene as a solvent.

3.5. Thermodynamic Analysis of WEPS Pyrolysis Reaction

Equations (7)–(9) were used to calculate the enthalpy change (ΔH_r), Gibbs free energy change (ΔG_r), and the entropy change (ΔS_r) of the WEPS pyrolysis reaction at different temperatures that are shown in

Table 3. Calculated thermodynamic parameters of WEPS pyrolysis.

Temperature (°K)	ΔHr (KJ/mol)	ΔGr (KJ/mol)	ΔSr (KJ/mol °K)
573	9340.07	175.41	15.99
673	16,745.95	202.31	24.58
773	25,765.10	229.34	33.03
873	36,122.29	256.47	41.08
973	47,572.83	283.70	48.60

.

The change in enthalpy of WEPS pyrolysis ΔH_r was positive, confirming that the reaction was endothermic. Furthermore, the increase in the entropy of the reaction ΔS_r with temperature suggests that the WEPS pyrolysis was favored with higher temperatures (>673 °K), and a better styrene yield was obtained at a higher temperature (>673 °K). However, the increase in Gibbs free energy ΔG_r with increasing temperature indicated that the reaction was unfavorable, moving away from the equilibrium at very high temperatures (>823 °K). This is in agreement with the results obtained by Mo et al. [33], who observed the same behavior of these thermodynamic parameters in the WEPS pyrolysis.

3.6. Pyrolysis Experiments in the Semi-Batch Rotary Reactor

The WEPS weight loss as a function of time and the weight evolution over time of the liquid products obtained in the R-8 pyrolysis experiment carried out with the rotary reactor, with a WEPS/MgO catalyst ratio of 10:1, using no solvent nor carrier gas, is shown in Figure 9. The graphs corresponding to all the pyrolysis experiments that were carried out using the rotary reactor are shown in Figure A3. The same WEPS weight loss as a function of time and the weight evolution over time of the liquid fraction trend was observed in all the experiments, even when the regenerated MgO catalyst was used, or no catalyst was used. At the beginning of the pyrolysis, a large volume of liquid products with a very high styrene concentration was obtained. As the reaction progressed, a decrease in the concentration of styrene was observed, and the concentration of toluene, ethylbenzene, and heavier products (methyl styrene, propenyl benzene, butenyl benzene, and the dimer) increased, but a considerable reduction in the volume of liquid products (oil) was observed.

Figure 10 shows the average of the internal temperature as a function of time of all the experiments carried out in the rotating reactor, which is represented by the blue curve, in which two inflection points were observed. In the first case, a slight drop in temperature was perceived between 50 and 80 min, which can be attributed to the fact that at that time, the greatest cracking of WEPS took place, which required a large amount of energy since the WEPS pyrolysis is highly endothermic. However, this depression in temperature was not observed in the temperature test that was conducted without WEPS load, which corresponded to the orange curve in Figure 10. In the second case, no temperature increase was observed in the first 10 min of heating in the blue graph because the WEPS charged to the rotating reactor melted in those first minutes of heating, which was not observed in the WEPS no-load temperature test (orange plot).

The yields of the pyrolysis reaction of the experiments carried out with the rotary reactor of the products in gas, liquid, and solid form are shown in

Table 4. Yields of products of the gaseous, liquid, and solid fraction of the experiments carried out with the rotary reactor.

Experiment	Gas Yield (wt%)	Liquid Yield (wt%)	Solid Yield (wt%)
R-1	18.39	81.54	0.07
R-2	9.59	89.36	1.05
R-3	21.69	77.15	1.16
R-4	13.24	85.85	0.91
R-5	12.63	86.60	0.77
R-6	1.67	97.49	0.84
R-7	5.20	93.54	1.26
R-8	0.19	99.23	0.58
R-9	7.17	91.80	1.03
R-10	10.81	89.08	0.11

. It was observed that in the experiments R-1 and R-10, carried out without a catalyst, a lesser amount of solid residue was obtained. Carbon was deposited on a surface, and when MgO was used as a catalyst, it had a large surface area (the MgO BET area was 45.63 m²/g). On the other hand, in the experiments in which no catalyst was used, the carbon was only deposited on the inner wall of the rotating reactor.

The final concentrations in weight % of the liquid products of all the experiments carried out in the rotary reactor are detailed in

Table 5. Final concentrations of the liquid products in the oil of the experiments carried out with the rotary reactor.

Experiment	Styrene (wt%)	Toluene (wt%)	Ethylbenzene (wt%)	Heavier Products (wt%)
R-1	92.76	2.36	0.84	4.04
R-2	87.26	6.35	4.75	1.64
R-3	89.80	6.15	2.61	1.44
R-4	91.08	6.37	1.70	0.85
R-5	92.05	5.52	1.84	0.59
R-6	92.87	6.58	0.36	0.19
R-7	94.37	4.55	0.60	0.48
R-8	95.08	4.22	0.41	0.29
R-9	92.70	6.25	0.77	0.28
R-10	96.88	2.10	0.47	0.55

. In the case of experiment R-10, a higher final concentration of styrene was obtained because it reached a temperature of 507.9 °C (carried out with the improved insulation). In comparison, in experiment R-1, a lower final concentration of styrene was obtained because the maximum temperature reached was lower, at 438.4 °C; both experiments were performed without a catalyst. It was observed that the by-products toluene, ethylbenzene, and heavier liquid products were obtained in the absence of a catalyst. It is also shown in Table 5 that those final concentrations of toluene double, and in some experiments triple, were obtained in the presence of the MgO catalyst than in its absence (experiments R-1 and R-10). Therefore, the MgO catalyst affected the production of this by-product.

A relevant aspect of the results is knowing the styrene yield evolution as a function of time, as shown in Figure 11a–c. The evolution of the styrene yield over time of the experiments carried out using the rotating reactor without catalysts R-1 and R-10 is shown in Figure 11a. In the case of experiment R-1, where there was only simple insulation (glass fiber with a thickness of 4 cm), the reactor was heated to 438 °C, so a smaller slope of temperature against time was observed. In the case of experiment R-10, composite insulation was built (glass fiber plus ceramic fiber). The reactor was heated to 508 °C, so a greater temperature slope against time was observed.

In the case of the experiments carried out with the rotary reactor using fresh MgO catalyst (Figure 11b), a variation in the styrene yield similar to experiment R-1 was observed. Experiments R-2, R-3, and R-4 were carried out using the 4 cm thick fiberglass covering as insulation. In the case of experiment R-5, where the fiberglass coating plus the ceramic fiber layer was used as insulation, it was observed that the beginning and the end of the yield curve diverged over time concerning the other experiments (R-2 to R-4).

Figure 11c shows the styrene yield as a function of the time of the experiments carried out with the rotary reactor using regenerated MgO catalyst (R-6 to R-9). With the improved isolation, which consisted of the side coatings, the 8.5 cm thick ceramic fiber layer around the reactor, and the 4 cm thick fiberglass insulation, mentioned above, greater homogeneity in the styrene yield as a function of time was perceived. In the case of experiment R-6, the catalyst was regenerated for the first time, and no deactivation process was observed. The same MgO catalyst used and regenerated for the second time was used again in experiment R-7, and no decrease in performance was noted. Likewise, the same catalyst was regenerated for the third time and evaluated in experiment R-8; no deactivation was observed. Finally, in the experiment, R-9 was reused the four-time-regenerated MgO catalyst, and no deactivation was observed.

In all experiments, an evolution of the styrene yield over time similar to the development of temperature over time was observed (Figure 10), suggesting that the styrene yield had a directly proportional dependence on the heat flux or the temperature increase.

The results of the final styrene yield, LP gas consumption, and the maximum average temperature reached in the experiments carried out with the rotary reactor without using a catalyst are shown in

Table 6. Final styrene yield, LP gas consumption, and the maximum average temperature reached in the experiments made without a catalyst.

Experiment	Maximum Average Temperature (°C)	Styrene Yield (wt%)	LP Gas Consumption (L)
R-1	438.4	75.63	391
R-10	507.9	86.31	395

. It was observed that in experiment R-1, the styrene yield was lower than in experiment R-10 because the maximum temperature inside the reactor was lower due to the poor insulation, although the LP gas consumption was similar. Therefore, with the same LP gas consumption, a higher styrene yield can be obtained by improving the insulation; in this way, the thermal efficiency was increased by reducing heat losses.

Table 7. Final styrene yield, LP gas consumption, and the maximum average temperature reached in the experiments using the fresh MgO catalyst.

Experiment	Maximum Average Temperature (°C)	Styrene Yield (wt%)	LP Gas Consumption (L)
R-2	484	77.98	374
R-3	445	69.29	368
R-4	450.2	78.20	363
R-5	479.6	79.72	364

shows the results of the experiments carried out with the rotary reactor using fresh MgO catalyst, without any thermal pretreatment. It was observed that lower styrene yields were obtained in these experiments because the experiments R-2 to R-4 were carried out before improving the insulation, and thus the temperatures reached inside the reactor were lower than in the experiments where the two layers of thermal insulation were placed.

It was observed that the styrene yields in experiments R-2, R-4, and R-5 were similar; the average was 78.5%, while in experiment R-3, it deviated 11% less from the average. Furthermore, it was noted that the higher the reactor temperature, the higher the yield of styrene obtained.

The results of the final styrene yield, LP gas consumption, and the maximum average temperature reached in the experiments carried out with the rotary reactor using the same MgO catalyst regenerated are shown in

Table 8. Final styrene yield, LP gas consumption, and the maximum average temperature reached in the experiments carried out with regenerated MgO catalyst.

Experiment	Maximum Average Temperature (°C)	Styrene Yield (wt%)	LP Gas Consumption (L)
R-6	504.1	90.54	351
R-7	519.5	88.28	343
R-8	528	94.35	345
R-9	513.2	85.09	338

. It was observed that the highest styrene yield achieved was 94.35 wt% in experiment R-8, which coincided with the maximum average temperature reached inside the reactor, which was 528 °C, being the largest of all experiments.

Table 6, Table 7 and Table 8 show that when using the rotary reactor, high styrene yields were obtained, higher than those obtained in other studies in the literature [19,29], even with the regenerated MgO catalyst and without a catalyst. Therefore, it is technically feasible to implement it for industrial use by applying the regeneration of MgO catalyst or without a catalyst. Furthermore, the type of reactor used in the present study probably greatly favors the internal mass transfer due to the rotation of the reactive bed, that is, a good back-mixing of WEPS is achieved.

The lower temperatures in experiments R-1 to R-4 were because they were performed before the improvements in insulation were carried out. However, these temperature differences made it possible to determine the divergences in styrene yield with temperature.

Experiment R-10, carried out without a catalyst and with a maximum average temperature reached 507.9 °C, can be compared with experiment R-6, where a maximum average temperature of 504.1 °C was achieved, in which 4% more styrene yield was obtained. In experiments R-9 and R-10, a similar styrene yield was obtained, although an average maximum temperature of 513.2 °C was reached in experiment R-9. When pyrolysis was carried out at these temperatures, the behavior was different from that at low temperatures, remembering that in the experiments at 250 °C, previously carried out with the glass reactor, it was observed that the behavior was exactly the opposite (Figure 6). For the above, this process could also be carried out without a catalyst to obtain good results. By operating this process using the rotary reactor and without a catalyst, only carbon was obtained as solid residue, which could be valuable. When MgO was used as a catalyst, at the end of pyrolysis, the carbon remained attached to the catalyst’s outer surface, and, in this way, the carbon was not valuable.

Performing the analysis of the non-condensable gases of the R-9 experiment, a total distribution of solid, liquid, and gaseous products (by weight) resulted as follows: butane 0.6%, butylene 6.57%, toluene 5.74%, ethylbenzene 0.71%, styrene 85.09%, heavier liquids 0.26%, and carbon 1.03%.

Table 9. Comparison of present study results with some relevant investigations.

Reference	Temperature (°C) *	Oil Yield (wt%)	Styrene Content in Oil (wt%)	Styrene Yield (wt%)	Reactor Type	Catalyst
Park et al. [14]	450	95.1	76.31	72.57	Flask with paddle stirrer (semi-batch)	No catalyst
Miandad et al. [16]	450	80.8	48	38.78	Stainless steel static reactor (semi-batch)	No catalyst
Imani Moqadam et al. [19]	410	97.3	80.98	78.79	Fluidized bed reactor (semi-batch)	Silica–alumina
Zhang et al. [20]	350	93.4	76.4	71.35	Stainless steel static reactor (semi-batch)	BaO
Kim et al. [25]	390	95.9	71.63	68.69	Stainless steel reactor with a mechanical agitator (semi-batch)	No catalyst
Monroy-Alonso et al. [27]	390	88.6	75.2	66.62	Stainless steel static reactor (semi-batch)	No catalyst
Artexte et al. [29]	500	--	--	70.57	Conical spouted bed reactor (continuous)	No catalyst
Carniti et al. [30]	400	--	13.5	--	Glass tubes sealed under vacuum (batch)	No catalyst
This study	528	99.23	95.08	94.35	Steel rotary reactor (semi-batch)	MgO

* Temperature at maximum styrene yield.

shows the highest oil yields, styrene content in oil, and styrene yield obtained in some relevant investigations. When making a comparison with the best result of the present work, it was observed that a higher styrene yield (94.35 wt%) was achieved in this study, derived from the fact that a high yield of oil (99.23 wt%) was also obtained at a high temperature (528 °C). The highest yield of styrene reported in each study in Table 9 was carried out in static semi-batch reactors, in some with a mechanical stirrer, and only one was continuous in a spouted bed regime. The difference between the present study and the others is that the reactor used was rotatory with basic powdered catalyst, which allowed for back-mixing in the reactive bed so that higher styrene yields were achieved in this study.

The liquid products obtained in this pyrolysis process were subjected to vacuum distillation to purify the styrene produced, thus lowering the boiling point in the distillation, preventing spontaneous polymerization of styrene at high temperatures.

An average of the amount of styrene produced per liter of LP gas consumed in the experiments carried out with the rotary reactor was made, which resulted in 1.144 g of styrene/L of LP gas (in the gas phase), which is equivalent to 0.535 kg of styrene/kg of LP gas. These data were used to evaluate the amount of LP gas required for styrene production in this process. Currently in Mexico City, the sale price of 0.535 kg of styrene monomer is USD 2.39, the cost of LP gas is USD 1.03 per kg, the cost of electricity is USD 0.09, and the cost of WEPS is estimated at USD 0.29, which makes an operating cost to produce 0.535 kg of styrene USD 1.41 in this pilot plant. Therefore, it can be said in a preliminary way and from an overall view that this process could be profitable on an industrial scale, although the investment cost of the equipment, as well as the cost of styrene purification, labor, maintenance, and other costs must also be taken into account.

3.7. Catalyst Characterization

3.7.1. N₂ Physisorption (BET Area)

Figure 12 shows the N₂ adsorption and desorption isotherm of the fresh MgO catalyst at a temperature of 77 °K, wherein it was observed that the isotherm was type IV with a mesoporous pore diameter, according to the IUPAC classification [44].

Using the DH cumulative desorption method, a surface of 45.63 m²/g of fresh MgO catalyst was obtained. Utilizing the BJH cumulative desorption method, a pore volume of 0.1016 cm³/g was obtained. With the BJH method by desorption, an average pore diameter of 3.645 nm was obtained, as seen in the pore diameter distribution in Figure 13.

3.7.2. X-ray Diffraction

The diffractogram of MgO catalyst before pyrolysis (Figure 14) showed a remarkable coincidence with the MgO diffractogram (93.36%), which was called periclase (red color in Figure 14 and Figure 15).

Figure 15 shows the diffractogram of the MgO catalyst after pyrolysis of the experiment carried out with the glass reactor (G1), in which the presence of coke was observed. This analysis shows that the MgO catalyst was not modified after pyrolysis since the periclase reflections remained constant.

3.7.3. Scanning Electron Microscopy (SEM-EDS)

Figure 16a shows the image obtained by scanning electron microscopy of the MgO catalyst before pyrolysis. Figure 16b corresponds to the mapping of Mg by EDS of the catalyst, Figure 16c corresponds to the mapping of Si, Figure 16d corresponds to the Ca mapping, Figure 16e corresponds to the O mapping, and the chemical compositions are detailed in

Table 10. MgO catalyst chemical composition before and after pyrolysis.

Element	Before Pyrolysis (wt%)	After Pyrolysis (wt%)
O	79.94	34.82
C	0.00	48.95
Mg	19.3	12.95
Si	0.38	2.33
Ca	0.38	0.71
S	0.00	0.24

.

Figure 17 shows the images of the MgO catalyst after pyrolysis of the experiment carried out with the glass reactor (G1) obtained with the scanning electron microscope. Its chemical composition is reported in Table 10. This analysis performed with scanning electron microscopy and EDS shows that, after the pyrolysis reaction, carbon was formed on the MgO catalyst, and the presence of traces of sulfur that did not exist before the WEPS pyrolysis was found.

Table 11. Chemical compositions of the solid residues from the rotary reactor experiments by EDS.

Element	Weight %
	R-3	R-4	R-5	R-6	R-7
C	8.75	32.77	11.55	12.38	19.05
O	45.52	39.85	35.59	32.99	22.63
F	5.33	2.90	1.59	13.95	23.60
Mg	17.86	8.42	48.61	13.49	21.75
Al	6.05	4.42	0.15	0.56	1.18
Si	14.19	11.12	1.42	12.34	7.37
S	0.15	0.16	0.32	0.35	0.39
Ca	1.18	0.35	0.77	13.04	2.27
Fe	0.96	0.00	0.00	0.89	1.75

details the chemical compositions obtained by EDS from the solid residues generated with the rotary reactor after the pyrolysis reaction, from experiments R-3 to R-5 carried out with fresh MgO catalyst, R-6 carried out with MgO catalyst regenerated once, and R-7 carried out with MgO catalyst regenerated twice.

The presence of F can be explained by the degradation of the Teflon gasket that was placed between the reactor lid and the reactor body; it was observed that in the experiments R-6 and R-7 that corresponded to the regenerated catalyst, the F not only was not lost over the catalyst, but it increased with each regeneration. The presence of Si and other elements came from the remains of the red silicone sealant. Finally, the presence of Fe can be explained by the tiny fragments of the inner wall of the rotary reactor that were detached when the catalyst was recovered after pyrolysis.

3.7.4. Raman Spectroscopy

Figure 18 shows the Raman spectra obtained from the solid residues of the experiments carried out with the rotary reactor. According to Johnson and Thomas [45], the bands at 1337 and 1595 cm⁻¹ corresponded to a highly disordered carbon. Therefore, most likely, it was not of the graphitic type. For this reason, the total elimination of this carbon can be easily carried out at a temperature of 600 °C, at which the catalyst was freed from its carbon cover. This way, its regeneration was carried out. Consequently, these bands no longer appeared in the spectrum of the calcined residue (Figure 18).

3.7.5. UV–VIS Spectroscopy

Figure 19 shows the UV–VIS spectra by diffuse reflectance obtained from the solid residues of the experiments carried out with the rotary reactor and their corresponding calcined residues. The change in the spectra of the residues of the experiments carried out with the rotary reactor concerning their respective calcined residues was observed. According to Zhang et al. [46], the calcined residues correspond to the UV–VIS spectrum of MgO contaminated with a little Si, which agrees with Section 3.7.3. (Table 11) regarding the presence of Si.

3.7.6. Thermogravimetric Analysis (TGA)

Figure 20 shows the TGA analysis of the solid residues from experiments R-3 to R-7 carried out with the rotary reactor; it was observed that most of the weight loss took place between 500 and 600 °C.

3.7.7. Calcination of Samples

The calcination results were the following: the weight loss of the MgO catalyst without pyrolysis reaction was 10.58 wt%, which is attributed to humidity. On the other hand, the weight loss in the calcination of the solid residue from the G1 experiment carried out with the glass reactor was 46.79 wt %, and this result of the carbon content of the G1 run solid residue approximately coincided with the result (48.95 wt %) shown in Section 3.7.3. (Table 10).

Table 12. Weight loss during the spent catalyst calcination (1 g) and the TGA analysis results (5–8 mg) of the experiments carried out with the rotary reactor.

Experiment	Carbon Weight % with Calcination	Carbon Weight % with TGA
R-3	14.15	13.7376
R-4	10.75	13.1138
R-5	8.69	9.3167
R-6	8.74	9.0359
R-7	12.76	11.9905

shows the calcination results of the solid residues generated in the experiments R-3 to R-7 carried out with the rotary reactor; they were compared with the results of the thermogravimetric analysis (TGA) discussed in Section 3.7.6. It was observed that the carbon content results were similar.

Experiment G1 was carried out with the static glass reactor, and experiments R-3 to R-7 were carried out with the rotating reactor, so it was shown that the movement of the reactor helped the WEPS back-mixing and the deposit of carbon on the catalyst was lower. However, it should also be considered that the experiments with the rotary reactor were carried out at a higher temperature.

4. Conclusions

Preliminary experiments were carried out in a static glass reactor at 250 °C to determine which catalyst would be used in the WEPS pyrolysis. MgO was selected as the best catalyst due to its stability, low cost, and availability in the market. Later, catalytic and thermal WEPS pyrolysis experiments were carried out with a rotating steel reactor at higher temperatures, in which good results were obtained. The catalyst regeneration was studied for its reuse in pyrolysis. The highest styrene yield obtained was 94.35 wt%, with the catalyst regenerated three times, which coincided with the highest average temperature reached inside the rotary reactor, which was 528 °C. The best styrene yield was directly related to a higher temperature, with the temperature being the predominant variable in WEPS pyrolysis. The effect of the MgO catalyst was more significant at low temperatures than at higher temperatures. However, the movement in the rotating reactor also helped to obtain a higher styrene yield than when the reactor was static because the reactant bed back-mixing took place, and heat transfer was improved.

Given that good styrene yields were obtained with the rotary reactor, even with the MgO catalyst regenerated several times and without a catalyst, its industrial application is feasible and technically recommendable for styrene production from WEPS. It is crucial to have a good insulation system to minimize heat losses, operate the rotary reactor with a low rotation speed (1.6 rpm), and install mechanical seals and expansion joints for better control of leaks. This process implementation at an industrial level could contribute to the circular economy.

MgO catalyst is mesoporous and does not undergo any modification after pyrolysis. Thus, it can be reused with good results, employing its regeneration at 600 °C before reuse, with which the pyrolysis remaining carbon cover can be easily removed, since this carbon is of the highly disordered type, that is, not graphitic, which was verified with Raman spectroscopy. Furthermore, the presence of F, Si, and Fe impurities, among others generated at the pyrolysis end, do not affect the catalytic activity of the MgO regenerated with the pyrolysis solid residue calcination.

It is confirmed that the WEPS pyrolysis follows a first-order kinetics, according to reports in the literature. From the thermodynamic analysis, the increase in the reaction entropy ΔS_r with the temperature rise suggests that the WEPS pyrolysis is favored with higher temperature since a higher styrene yield is obtained at a higher temperature. ΔH_r and ΔG_r were calculated to characterize the WEPS pyrolysis thermodynamically.