Production of Aviation Fuel-Range Hydrocarbons Through Catalytic Co-Pyrolysis of Polystyrene and Southern Pine

Abstract

Sustainable aviation fuels (SAFs), produced from waste and renewable sources, are a promising means for reducing net greenhouse gas emissions from air travel while still maintaining the quality of air transportation expected. In this work, the catalytic co-pyrolysis of polystyrene and pine with red mud (bauxite residue) and ZSM-5 catalysts at temperatures of 450 °C, 500 °C, and 550 °C was investigated as a method for producing aromatic hydrocarbons with carbon numbers ranging from 7 to 17 for use as additives to blend with SAF produced through other methods to add the required quantity of aromatic molecules to these blends. The maximum yield of kerosene-range aromatic hydrocarbons was 620 mg per gram of feedstock (62% of feedstock was converted to kerosene-range hydrocarbons) obtained at 550 °C in the presence of ZSM-5. Additionally, it was noted that a positive synergy exists between pine and polystyrene feedstocks during co-pyrolysis that cracks solid and liquid products into gaseous products similarly to that of a catalyst. The co-pyrolysis of pine and polystyrene without a catalyst produced on average 17% or 36.3 mg more kerosene-range hydrocarbons than predicted, with a maximum yield of 266 mg of C7–C17 aromatic hydrocarbons per gram of feedstock (26.6% conversion of initial feedstock) obtained at 550 °C.

1. Introduction

One of society’s greatest concerns in the 21st century is global warming and climate change [1]. Although an increase in global temperatures of 1.5 °C compared to pre-industrial times is seen by scientists as the best possible scenario, this will lead to severe detrimental impacts on biodiversity and human populations [2]. Transportation accounts for approximately 23% of global greenhouse gas (GHG) emissions and is likely to rise as global populations continue to increase [3]. Technological and engineering developments are decreasing emissions in the transportation sector, but there is a need for additional improvements to meet either 1.5 °C or 2 °C climate change goals [3]. Electrification of the transportation sector through technologies such as electric vehicles, mass transit rail systems, electric or hybrid buses, and onshore power for boats have all been implemented with the effect of decreasing carbon dioxide emissions [4,5,6,7]. However, the electrification of commercial aviation aircraft is a futile effort. Current battery technology limits battery energy density to values ranging from 150 to 250 Wh/kg [8]. For comparison, commercially available jet fuel contains a minimum of 11900 Wh/kg [9].

The potential for electrifying commercial air travel is extremely limited with the current gap in battery energy density. Instead, an alternative solution to reduce net emissions from commercial aviation is sustainable aviation fuels (SAFs) [10]. SAFs are defined by the International Civil Aviation Organization (ICAO) as aviation fuels that promote conservation and responsible economic development by reducing net GHG emissions [11]. Jet fuel contains appreciable concentrations of paraffins, olefins, naphthenes, and aromatics [10,12]. Paraffins, olefins, and naphthenes are dense in energy and produce low emissions while maintaining critical fuel properties such as freezing point, flash point, and thermal stability [10]. Aromatics, while not as clean burning or energy-dense as other hydrocarbons are an essential component of jet fuel for their role in expanding seals within fuel lines at high elevations to prevent fuel leakage and improper combustion [12]. The American Society of Testing and Materials (ASTM) has approved several methods of converting renewable materials such as waste cooking oils or energy-rich fuel crops through enzymatic or thermochemical means into kerosene-range, carbon 7 to carbon 17 (C7–C17), hydrocarbons [13,14,15,16]. Because these methods use feedstocks primarily composed of linear alkanes and alkenes such as fatty acids, the fuels produced from these technologies are rich in paraffins, olefins, and naphthenes, but they either lack entirely or have meager quantities of necessary aromatic compounds [12].

This limitation in current SAF production technologies has frequently relegated SAF to an additive role where it is only used when blended in small proportions with petroleum-based aviation fuels instead of being a drop-in fuel [17]. This work proposes a novel solution for producing drop-in SAF. If a renewable aviation fuel additive, rich in aromatics is blended with SAF synthesized using traditional methods, the resulting fuel blend would contain both the necessary fraction of aromatics and linear hydrocarbons. Since many of the traditional feedstocks used for producing SAF do not contain aromatic rings, alternative feedstocks, and SAF production mechanisms must be implemented.

This work suggests the co-pyrolysis of pine and polystyrene (PS) as an effective means of producing an aromatic-rich SAF additive. Pine and other woody biomasses are primarily composed of lignin and cellulose, which produce large quantities of aromatic molecules when undergoing pyrolysis [18,19]. Pine was specifically selected for its ready availability in the southern United States and its high lignin content, 25–35%, compared to hardwood trees, 15–25% [20]. Polystyrene is a plastic polymer made of repeating aromatic monomers that similarly degrade into aromatic molecules when exposed to heat [21,22,23]. As a pure hydrocarbon polymer, it is a rich source of hydrogen that can assist in upgrading biomass-derived oil during the co-pyrolysis process [24]. Prior research has indicated that the co-pyrolysis of biomass and waste plastics can lead to increased yields of aromatic molecules through Diels–Alder and alkylation reactions that occur during the co-pyrolysis reaction [19,25,26].

Catalytic fast pyrolysis of polystyrene and catalytic co-pyrolysis of polystyrene and pine blends were investigated to determine the most effective means for converting these waste products into aromatic additives for use in SAF. Bauxite residue, or red mud (RM), and Zeolite Socony Mobil-5 (ZSM-5) were selected as the catalysts for this process. Red mud contains high concentrations of Fe₂O₃, or hematite, which provide active oxygen species sites to promote catalytic deoxygenation and hydrogenation processes that promote higher yields of liquid and gaseous products [27,28]. ZSM-5 is a common cracking catalyst in the petroleum industry and is particularly attuned to facilitating acid-catalyzed oligomerization and decarboxylation reactions [19,22]. Furthermore, its high specific surface area, typically greater than 250 m²/g, makes it more active than red mud [29]. Reaction temperatures ranging from 450 °C to 550 °C were investigated as identified through a prior work by the author, which indicated that the maximum thermal degradation of pine and polystyrene occurred at temperatures lower than 550 °C [30].

In recent years, many works have focused on increasing the selectivity of monoaromatic hydrocarbons (MAHs) while decreasing polyaromatic hydrocarbon (PAH) yields from the co-pyrolysis of plastic and biomass [22,23,31,32]. Performing catalytic pyrolysis of polystyrene with ZSM-5, researchers reported that catalytic pyrolysis decreased the content of PAH in the effluent gas stream [31]. Using Py-GC/MS, Guo et al. [22] performed co-pyrolysis of poplar and polystyrene in the presence of HZSM-5 and iron-doped Fe/HZSM-5. Their results indicated that catalytic co-pyrolysis increased the selectivity of MAH such as ethylbenzene and toluene [22]. Another team of researchers reported a similar increase in MAHs using metal-doped biomass char-supported catalysts when pyrolyzing polystyrene [23]. When performing non-catalyzed co-pyrolysis of polystyrene and pine, Rajput et al. [32] also achieved a lower yield of PAH and postulated that co-pyrolysis reduced PAH synthesis by promoting ring-opening reactions. These results indicate that both co-pyrolysis with biomass and the use of catalysts can diminish PAH yields from the pyrolysis of polystyrene. As PAH is undesirable for aviation fuels, this current work will similarly seek to decrease the yield of PAH and determine the extent of the effects of catalysis and co-pyrolysis on their decrease.

Other works have investigated the co-pyrolysis of polystyrene with nonwoody biomasses and have reported increased bio-oil yields compared to biomass-only pyrolysis [24,25]. Chaturvedi et al. [25] refrained from using catalysts when performing co-pyrolysis of rice husks and polystyrene because many catalysts that increase bio-oil yields and quality are expensive and can quickly deactivate. Another opinion from Van Nguyen et al. [24] highlighted that many works use a high plastic-to-biomass ratio, which, while giving high yields, will not be sustainable because plastics are inherently derived from non-renewable sources. Taking these works into consideration, this current research compares the effect of in situ catalysis with the synergistic effects originating from the co-pyrolysis of biomass and polystyrene to determine if the use of catalysts is justifiable. Furthermore, it contrasts the observed yields of biomass and plastic co-pyrolysis to theoretically predicted values assuming separately performed pyrolysis of pine and polystyrene to determine if the co-pyrolysis mechanism increases the oil yield from biomass or if higher yields that have been previously reported are the result of plastic volatiles only.

This research is intended to fill a gap in the field’s present understanding of what conditions will optimize both the liquid yield and selectivity of C7–C17 aromatic hydrocarbons from the catalytic co-pyrolysis of pine and polystyrene. Additionally, this research expands upon prior works’ knowledge of the synergistic effects originating from the co-pyrolysis of polystyrene and pine and the effect of both catalysis and pyrolysis temperature on these synergies. The concentrations of C7–C17 hydrocarbons in the resultant oils were determined using Gas Chromatography/Mass Spectrometry (GC/MS) calibrated with known representative compounds, a feat that has not been previously reported for the co-pyrolysis of pine and polystyrene. The results of this work will also provide a solution to synthesizing the required aromatics for drop-in SAF blends.

2. Results and Discussion

2.1. Feedstock Characteristics

The polystyrene had high proportions of volatile matter and fixed carbon (FC) and negligible moisture content (MC) and ash content (

Table 1. Proximate and ultimate analysis of polystyrene and pine feedstocks.

	Proximate Analysis				Ultimate Analysis
	MC (% Wet Basis)	Volatiles (% Dry Basis)	Ash (% Dry Basis)	FC * (% Dry Basis)	C (%)	H (%)	N (%)	O * (%)	S (%)
PS	0.33	82.87	0.03	17.10	91.69	8.29	<0.01	<0.01	0.02
Pine	8.86	81.17	0.31	18.52	46.41	6.87	0.01	46.71	<0.01

* By difference.

), indicating that the polystyrene used in this study had few filler and binding materials, which would have appeared as ash. The average values reported for the pine (Table 1) are like those reported in the literature [33,34,35].

2.2. Pyrolysis Products

The results from the two trials performed for each of the 27 experimental conditions were averaged together, and two-way ANOVA was performed for each pyrolysis product to compare the effects of reaction temperature and catalysis. Theoretically predicted values for the yield of each pyrolysis product of the blended feedstock were also determined. Statistically significant results are reported throughout the following sections, and the degree of freedom (d.f.), F statistic (F), and p-value for each statistically significant result are given. A summary of all the ANOVA statistics and post hoc testing is provided in the Supplementary Materials (Tables S2 and S3).

2.2.1. Effect of Temperature on Pyrolysis Yields

As the temperature increased, there was a consistent and statistically significant decrease in the solid char yield (Figure 1a,b and Table S1) for both polystyrene pyrolysis (d.f. = 2, F = 7.9, p = 0.010) and pine pyrolysis (d.f. = 2, F = 31.6, p = 0.000). The higher temperatures provided more energy for depolymerizing large macromolecules that could not be broken down at lower temperatures [23]. Zhang et al. [34], using pine sawdust and polyethylene, and Kong et al. [23], with polystyrene and biochar, reported similar trends.

For the polystyrene feedstock (Figure 1a and Table S1), higher temperatures also led to statistically significant reductions in the yield of gaseous products (d.f. = 2, F = 8.0, p = 0.010) and statistically significant gains in the oil yield (d.f. = 2, F = 15.8, p = 0.001). Liu et al. [36] reported a similar increase in the oil yield for polystyrene pyrolysis up to 600 °C before the oil yield began to lessen and the gas yield began to rise at even higher temperatures. In another work, the gas yield of polystyrene pyrolysis was reported to decrease, and the oil yield increased until 450 °C at which point it began the liquid yield fell and the gas yield expanded as liquid products were further cracked into non-condensable gases [23]. The variations in the optimal reaction temperature for maximum oil yield can be attributed to the design of the pyrolysis reactor and the size of the particles being pyrolyzed with fluidized-bed processes, which required higher temperatures than fixed-bed systems with finer particles which require lower temperatures than larger ones [35,37,38]. For the temperature ranges tested, higher temperatures provided more energy for splitting the polystyrene feedstock and led to an increased oil yield. At temperatures greater than those tested in this work, the additional energy provided further degrades the liquid oil into gas. While this result is desirable for hydrogen syngas production, it is not conducive for producing aviation fuels and the pyrolysis temperature should not exceed 600 °C if liquid products are desired.

For the pine feedstock (Figure 1b and Table S1), the gas yield (d.f. = 2, F = 27.6, p = 0.000) and the oil yield (d.f. = 2, F = 10.3, p = 0.005) had a statistically significant dependence on the reaction temperature. The gas yield continuously increased with increasing temperatures as volatile products were further decomposed, whereas the oil yield peaked at 500 °C before dropping at higher temperatures. Both of these trends are well documented in the literature and exhibit no irregularities [39,40].

2.2.2. Effect of Catalyst on Pyrolysis Yields

For the pine feedstock (Figure 1b and Table S1), catalysis only significantly affected the oil yield (d.f. = 2, F = 7.4, p = 0.013). While adding red mud and ZSM-5 marginally improved the gas yield, these differences were not statistically significant from the no catalyst baseline. However, there was a statistically significant decrease in the oil yield because of the catalysts. Both the red mud and the ZSM-5 catalysts led to lower oil yields, with the effect of ZSM-5 being more pronounced. The catalysts facilitated cracking reactions that split would-be oil products further into non-condensable gases. The observed difference between the two catalysts is due to ZSM-5 being a stronger cracking catalyst because of its higher specific surface area, leading to a more significant conversion of liquid products into gases [22,29]. It is generally accepted that catalytic activity shifts biomass pyrolysis products towards smaller non-condensable products [22,41,42].

The catalysts had a more pronounced effect on the yields of polystyrene pyrolysis (Figure 1a and Table S1) with statistically significant variations in gas yields (d.f. = 2, F = 56.3, p = 0.000) and oil yields (d.f. = 2, F = 44.4, p = 0.000) present. Both the red mud and the ZSM-5 enlarged gas yields and diminished oil yields. Kong et al. [23], working with metal-doped bio-char catalysts, reported that oil yields decreased for catalytic polystyrene pyrolysis by 5–7%, values comparable to those observed using ZSM-5 in this work. Other authors using various catalysts and combinations of plastic and biomass feedstocks have reported the same trend [24,29]. Both catalysts increased β-scission and C-C cracking reactions that converted liquid products to non-condensable gases; however, the ZSM-5 catalyst had a much more pronounced effect due to both its stronger acidic sites and higher specific surface area [18,21,22].

2.2.3. Pyrolysis Yields of Blended Feedstock

Increasing the reaction temperature for the co-pyrolysis of polystyrene and pine (Figure 1c and Table S1) resulted in a statistically significant reduction in char yield (d.f. = 2, F = 26.0, p = 0.000) and a corresponding increase in the oil yield (d.f. = 2, F = 6.7, p = 0.017). Generally, higher temperatures favored higher gas yields, although this trend was not statistically significant. Rajput et al. [32] reported decreasing char yields at higher temperatures for the co-pyrolysis of sawdust and polystyrene, as higher temperatures enhanced the synergistic effect between the two feedstocks. The same work reported increasing gas yields and decreasing oil yields at temperatures of 550 °C and above [32]. The added energy at higher temperatures facilitated complete degradation of the feedstock into volatiles, which led to higher liquid yields as a result of secondary Diels–Alder and alkylation reactions, as well as generally larger gas yields [22,24,33].

The catalysts also had a statistically significant effect on the yields of char (d.f. = 2, F = 5.7, p = 0.025), as well as the yields of gas (d.f. = 2, F = 8.2, p = 0.010) for the co-pyrolysis of polystyrene and pine (Figure 1c and Table S1). The red mud catalyst led to lower char yields and higher gas yields. Interestingly, the gas yields decreased with increasing temperature, unlike the no catalyst and ZSM-5 catalyst scenarios. This result is likely due to the conversion of hematite, Fe₂O₃, into magnetite, Fe₃O₄, which occurs around 462 °C in red mud and results in the catalyst being less reactive towards oxygenated species in the biomass, which resulted in more aqueous phase products [28]. The ZSM-5 catalyst led to a statistically significant (Table S3, p = 0.000) increase in gas yield, an expected result as the numerous acidic sites on the catalyst facilitated the degradation of both the feedstock and the volatiles produced into smaller non-condensable gases [18,21,22]. Zhang et al. [34] reported that increasing reaction temperature led to decreasing char yields and increasing gas yields for the co-pyrolysis of pine and polyethylene with a spent fluidized catalytic cracking catalyst, confirming the results obtained here.

The use of catalysts, however, did not have a significant effect (d.f. = 2, F = 2.8, p = 0.115) on the oil yield of the co-pyrolysis process (Figure 1c and Table S1). During the individual pyrolysis of polystyrene and pine, the catalysts, particularly the ZSM-5, decreased the oil yield substantially, a result not repeated during co-pyrolysis. The synergistic effects originating from the interaction between pine volatiles and polystyrene volatiles and polystyrene volatiles and pine char led to increased cracking of the initial feedstock, as seen by lower char yields than theoretically predicted, and of the volatile feedstock, as seen by greater gas yields and lesser oil yields than theoretically predicted [33]. The addition of either the red mud or ZSM-5 did not lead to significant further cracking beyond that which occurred because of co-pyrolysis. This indicates that the cracking effects of the catalysts were effectively replaced by the synergistic interactions between the blended feedstocks which had an effect like that of the red mud or ZSM-5 [25]. This is a key result because the use of catalysts increases the cost and complexity of a pyrolysis process, and if co-pyrolysis can be used to increase feedstock cracking without the use of catalysts then that is preferable. With regard to SAF production from polystyrene, co-pyrolysis can be used to increase the cracking of the feedstock and replace ZSM-5 in that role.

2.3. Analysis of Oil Products

The percentage of carbon, hydrogen, oxygen, and sulfur (ultimate analysis) found in the oil produced from each feedstock was determined (

Table 2. Ultimate analysis and water content of oil and predicted values for blended feedstock.

		Ultimate Analysis					Water Content (% at Given Temperature)
		C (%)	H (%)	N (%)	O * (%)	S (%)	450 °C	500 °C	550 °C
PS	No Catalyst	91.9	7.99	0.01	0.02	0.08	0.00	0.00	0.00
	Red Mud	90.9	7.93	0.01	1.15	0.01	0.00	0.00	0.00
	ZSM-5	91.7	7.96	0.01	0.26	0.07	0.00	0.00	0.00
Pine	No Catalyst	68.5	7.85	0.01	23.6	0.08	15.15	12.65	11.14
	Red Mud	65.9	7.60	0.01	26.5	0.04	15.81	13.74	12.17
	ZSM-5	71.2	7.72	0.01	20.9	0.16	11.74	10.8	9.32
50/50 Blend	No Catalyst	86.8	7.95	0.01	5.23	0.01	0.00	0.00	0.00
	Red Mud	88.2	8.52	0.01	3.26	0.01	0.00	0.00	0.00
	ZSM-5	87.6	8.23	0.01	4.06	0.1	0.00	0.00	0.00
Theoretical	No Catalyst	80.2	7.92	0.01	11.8	0.08	7.58	6.33	5.57
	Red Mud	78.4	7.77	0.01	13.8	0.02	7.91	6.87	6.09
	ZSM-5	81.5	7.84	0.01	10.6	0.12	5.87	5.4	4.66

* By difference.

). There were no significant variations as a result of pyrolysis temperature and catalyst usage, so the results from all trials were averaged to provide the given values. The theoretically expected values for the ultimate analysis of the blended feedstock are also reported for comparison to the experimentally observed values (Table 2).

The polystyrene oil largely reflected the original composition of the polystyrene feedstock, with small impurities of sulfur and oxygen likely originating from either the catalysts or traces of pine oil left in the reactor setup. The pine oil had significantly higher carbon content and lower oxygen content than the original pine feedstock. This is a result of oxygen being removed through deoxygenation or dehydration reactions and is well documented in the literature [23,43,44].

Further evidence of the existence of synergistic effects occurring during the co-pyrolysis of polystyrene and pine is demonstrated by the oil produced from the blended feedstocks having a higher carbon content and lower oxygen content than theoretically predicted from an equal blend of pine and polystyrene oils. A portion of this effect is likely due to the difference in the water content between the pine oil and the co-pyrolysis oil, as explained in the next paragraph. However, the simultaneous increase in hydrogen content above that which was theoretically predicted indicates that there are synergistic effects that eliminated oxygen when converting the feedstocks into oil. These synergistic effects are due to intermolecular hydrogen transfer with the polystyrene acting as a hydrogen donor, which accelerated the rates of dehydration, deoxygenation, and Diels–Alder reactions within the pine volatiles [19,22,24,25].

The water content of all the oils produced from the polystyrene-containing feedstocks was 0.00%. During the co-pyrolysis of pine and polystyrene, the exclusively hydrophobic oils produced from the polystyrene feedstock extracted the organic oil fraction from the liquid products produced from the pine feedstock. Then, when the liquid product was centrifuged, the hydrophobic oil, containing compounds originating from both pine and polystyrene, was separated from the aqueous phase consisting of water and other hydrophilic products originating from the pine feedstock [25]. The water content of the pine pyrolysis oils decreased because of increasing temperature (d.f. = 2, F = 21.2, p = 0.000), and when the ZSM-5 catalyst (d.f. = 2, F = 21.5, p = 0.000) was used (

Table 3. The concentration of kerosene-cut hydrocarbons, styrene, and C7–C17 hydrocarbons containing oxygen, and the yield of C7–C17 hydrocarbons per gram of feedstock for each experimental condition.

			Conc. of All C7–C17 Compounds *	C7–C17 Containing Oxygen *	Conc. of Styrene Only	Oil Yield of Pyrolysis Reaction	Yield of C7–C17 Hydrocarbons
			(mg/g Oil)	(mg/g Oil)	(mg/g Oil)	(g Oil/g Feed)	(mg/g Feed)
Polystyrene	None	450 °C	451.83	0.00	423.44	0.8850	399.87
		500 °C	472.90	0.00	444.44	0.8900	420.88
		550 °C	467.32	0.00	428.90	0.9050	422.92
	Red Mud	450 °C	530.25	0.00	458.60	0.8625	457.34
		500 °C	525.76	0.00	477.05	0.8825	463.98
		550 °C	559.42	0.00	456.13	0.9000	503.48
	ZSM-5	450 °C	672.60	0.00	277.14	0.8250	554.90
		500 °C	725.89	0.00	251.85	0.8400	609.74
		550 °C	720.14	0.00	349.20	0.8625	621.12
Pine	None	450 °C	191.33	181.85	N/A	0.0506	9.68
		500 °C	241.12	230.52	N/A	0.0604	14.56
		550 °C	227.21	192.29	N/A	0.0503	11.43
	Red Mud	450 °C	261.10	260.49	N/A	0.0506	12.87
		500 °C	274.14	271.72	N/A	0.0532	14.23
		550 °C	276.24	274.99	N/A	0.0519	13.98
	ZSM-5	450 °C	357.42	322.03	N/A	0.0428	15.29
		500 °C	332.14	305.74	N/A	0.0506	16.80
		550 °C	376.52	296.35	N/A	0.0508	19.14
50/50 Blend	None	450 °C	575.22	13.68	373.41	0.4325	248.77
		500 °C	556.26	13.86	341.68	0.4207	234.01
		550 °C	590.11	17.34	368.99	0.4504	265.78
	Red Mud	450 °C	514.81	20.89	323.29	0.4121	212.15
		500 °C	513.02	20.56	359.77	0.4258	218.44
		550 °C	534.43	12.31	388.98	0.4422	236.32
	ZSM-5	450 °C	531.78	32.58	300.26	0.3896	207.19
		500 °C	577.83	26.47	335.24	0.4281	247.38
		550 °C	543.87	21.96	337.59	0.4294	233.51
Theoretical 50/50 Blend	None	450 °C	321.58	90.92	211.72	0.4678	204.77
		500 °C	357.01	115.26	222.22	0.4752	217.72
		550 °C	347.27	96.15	214.45	0.4777	217.18
	Red Mud	450 °C	395.68	130.25	229.30	0.4559	235.11
		500 °C	399.95	135.86	238.53	0.4672	239.10
		550 °C	417.83	137.50	228.07	0.4753	258.73
	ZSM-5	450 °C	515.01	161.02	138.57	0.4339	285.09
		500 °C	529.01	152.87	125.93	0.4453	313.27
		550 °C	548.33	148.18	174.60	0.4567	320.13

* A list of compounds identified by GC/MS, those containing oxygen, and those excluded from concentration calculations due to insufficient match quality are provided in Table S7.

; two-way ANOVA statistics reported in Table S5). Both higher temperatures and increased catalytic cracking induced more dehydration and deoxygenation reactions which led to the oil being less polar and water being less miscible within it. The red mud had the opposite effect and slightly increased the water content of the pyrolysis oil, which is likely a result of the redox reactions catalyzed by the iron in the red mud, which preserved oxygen in the oil and increased the miscibility of water in the final product [22].

2.4. Chemical Composition of Oil Products

2.4.1. Effect of Catalysts on Oil Compound Distribution

The distribution of compounds in each oil varied as a result of the feedstock composition, the pyrolysis temperature, and the usage of catalysts (Figure 2, Figure 3 and Figure 4). The use of red mud only slightly altered the composition of the polystyrene oil by marginally increasing the amount of polyaromatic hydrocarbons (PAHs) and shifting the distribution of compounds towards slightly lower carbon numbers (Figure 2b). These trends are likely the result of the red mud weakly catalyzing C-C cracking reactions and alkylation reactions, although they are limited in their effect because red mud is not a strong catalyst [19,28,29].

Conversely, ZSM-5, with many strong acidic sites, led to a significant variation in the oil composition [19,22]. The oil produced in the presence of ZSM-5 had significantly less styrene, more toluene and ethylbenzene, and more PAH while also decreasing the occurrence of large aromatic macromolecules (Figure 2c). The ZSM-5 catalyst promoted alkylation reactions, which converted styrene into indene, naphthalene, phenanthrene, and their derivatives [22]. A further reduction in styrene was achieved as ZSM-5 catalyzed secondary C-C cracking reactions, which converted the products of β-scission into monoaromatic hydrocarbons (MAHs) such as ethylbenzene [23].

These characteristics make the oil produced using the ZSM-5 catalyst more desirable to produce SAF because a greater proportion of the produced oil is within the kerosene cut and because there is a greater variety of compounds within this range. This variety is more conducive for producing aviation fuels that meet ASTM requirements because the overall properties of the fuel, such as flash point, cloud point, heating value, and conductivity, will not be dominated by a single compound and will instead be influenced by each of the compounds which tends to temper the properties and reduce the variability of key properties within the fuel [12]. One drawback of the ZSM-5 catalyzed pyrolysis is the increase in three-ring PAHs, which are not suitable for aviation fuels.

The pine pyrolysis oil had a greater degree of variability and additional classifications of molecules, including cycloolefins (cycloalkenes), fatty acids, and naphthenes (cycloalkanes), with the vast majority of molecules containing oxygen (Figure 2d–f). When pyrolysis was performed with red mud catalyst, the area percentage of aromatics and PAH decreased, and the area percentage of fatty acids and cycloolefins rose (Figure 2e). These shifts in product distributions are the result of the red mud catalyzing dehydration reactions of levoglucosan and other cellulosic compounds, as well as facilitating redox reactions that lead to the formation of more alcohols and esters [19,22]. Pyrolysis in the presence of ZSM-5 removed the occurrence of fatty acids and naphthenes and increased the area percentages of aromatics and cycloolefins compared to the no-catalyst oil (Figure 2f). Presumably, the ZSM-5 catalyst cracked large alkane and alkene chains found in the fatty acids and naphthenes and simultaneously catalyzed the oligomerization of short olefin molecules into aromatic compounds through Diels–Alder reactions [18,24]. A shared feature of both catalysts was to shift the distribution towards lower carbon numbers through catalytic cracking of large molecules [22].

2.4.2. Effect of Reaction Temperature on Oil Compound Distribution

The effects of temperature on the distribution of compounds in oils produced from pine varied depending on the catalyst used, but in general, higher temperatures resulted in a smaller proportion of low-molecular-weight molecules (C7 and below) and an increased proportion of high carbon number compounds (C15 and above) and PAH (Figure 3a–d and Figure S2). These trends are a result of small carbon number molecules being further cracked under more energetic conditions into non-condensable gases, while resonance-stabilized aromatic and PAH molecules did not crack as readily [45]. Furthermore, when increasing temperatures in the presence of ZSM-5, the olefins produced by cracking reactions were incorporated into PAH through Diels–Alder and alkylation reactions [18,22,26].

Pyrolysis temperature had little effect on the distribution of compounds in the oil produced from polystyrene regardless of catalyst. Universally, across all three catalytic conditions, higher temperatures led to slightly higher percentages of styrene and slightly lower percentages of larger aromatic macromolecules originating from incomplete depolymerization of the feedstock. This shift is a result of higher energy conditions increasing the rate of β-scission and C-C cracking reactions [23].

2.4.3. Compound Distribution of Co-Pyrolysis Oil

The compound distribution of the oil produced from the co-pyrolysis of polystyrene and pine (Figure 4) largely resembles that of the polystyrene oil (Figure 2 and Figure S1). The polystyrene volatiles dominate the oil blend as much of the pine volatiles are lost as non-condensable gases or remain in the aqueous phase [25]. The red mud and ZSM-5 catalysts only had the effect of increasing the amount of PAH present in the oil and did little to alter the weight distribution of compounds. Higher reaction temperatures also tended to favor lower-weight molecules for the same reasons described previously.

As discussed before, the addition of catalysts only minimally altered the composition of the oil beyond what was accomplished through the synergistic effects of the co-pyrolysis process. Compared to the oil from polystyrene pyrolysis (Figure 2a), the co-pyrolysis oil (Figure 4a) had fewer high molecular weight compounds and a greater variety of compounds in the desired C7–C17 range. While the co-pyrolysis process did not decrease the occurrence of heavy aromatic macromolecules (C24 peak) as effectively as the polystyrene oil produced with the ZSM-5 catalyst (Figure 2c), the co-pyrolysis oil had only a fraction of the amount of PAHs present, including much lower quantities of undesirable 3-ring PAH. The synergistic interactions between the pine volatiles and char and the polystyrene volatiles increased the rate of Diels–Alder reactions that led to higher proportions of MAH [22,25]. As the reactants of this reaction are generally the same as those of the alkylation reactions that produce PAH, this competition explains the decreased quantities of PAH observed [25,32].

2.5. Reaction Mechanisms of Pyrolysis, Catalytic Pyrolysis, and Co-Pyrolysis

The following proposed reaction pathways (Figure 5) explain the observed distributions of compounds and are validated as fitting pathways by the GC/MS analysis of the pyrolysis oils produced. The primary mechanisms of thermal depolymerization of polystyrene are β-scission, C-C bond cleavage, molecular elimination, and free radical reactions [22,23,24,46]. β-scission results in the formation of styrene monomers and their dimers and trimers, whereas elimination and free radical reactions lead to the formation of bibenzyl and other related aromatic compounds [22,24,46]. All pyrolysis oils produced from polystyrene feedstocks contained bibenzyl and related compounds such as trans-stilbene, with the occurrence of these compounds decreasing during ZSM-5 catalyzed-pyrolysis and co-pyrolysis. Non-catalyzed C-C bond cleavage results in larger polystyrene chain fragments, which then form aromatic macromolecules through recombination and oligomerization reactions [23,46]. This was observed as pyrolysis oils produced without a catalyst or with the weaker red mud catalyst had large proportions of high carbon number aromatic compounds. Catalyzed C-C cracking, whether through co-pyrolysis or with the aid of ZSM-5, tends to produce monoaromatic hydrocarbons and olefin fragments [22,23]. These products can react with themselves and the products of β-scission through Diels–Alder reactions to form diphenylcyclopropane and related compounds or through alkylation reactions to form indene and its derivatives or PAH [19,21,47]. Consistent throughout all reaction pathways is intermolecular and intramolecular hydrogen transfer, which typically originates from β-scission or dehydrogenation reactions and aids in producing and stabilizing the various secondary products such as indene, diphenylcyclopropane, and PAH [19,21,22,26,47]. In this study, ZSM-5 catalyzed polystyrene pyrolysis oils at all temperatures contained larger proportions of monoaromatic hydrocarbons such as toluene and ethylbenzene as well as indenes and PAHs than non-catalyzed or red mud catalyzed polystyrene pyrolysis oils.

The thermal degradation of biomass has been well studied and operates through dehydration, decarboxylation, and decarbonylating reactions [18,19,24,47]. Cellulose and hemicellulose produce levoglucosan as an intermediary, which ultimately presents itself as furans and other oxygenated 5-carbon compounds [18,19,22,47]. Lignin depolymerizes to form mostly phenolics and ethers [24]. This was observed as the pine pyrolysis oils produced with and without catalysts contained furans, furfurals, cyclopentones, and phenolics, as predicted. During co-pyrolysis with polystyrene, hydrogen transfer from the polystyrene enables dehydration reactions, which convert these phenolics into unoxygenated monoaromatic hydrocarbons [19,24]. Furans are similarly converted during co-pyrolysis to a mixture of benzene aromatics through decarboxylation reactions and Diels–Alder reactions, which are made possible through hydrogen transfer and reactions with olefins produced from polystyrene decomposition [21,24,26]. The oils produced via co-pyrolysis contained greater percentages of monoaromatic hydrocarbons than polystyrene pyrolysis oils and significantly fewer oxygen-containing compounds than those obtained from pine pyrolysis. A small portion of these aromatics underwent alkylation reactions to form PAH, although this reaction was less prevalent during co-pyrolysis compared to ZSM-5-catalyzed depolymerization of polystyrene [19,21,26].

2.6. Concentration of Products in Pyrolysis Oils

2.6.1. Effect of Pyrolysis Temperature and Catalyst on Concentration

The pyrolysis reaction temperature did not have a significant effect on the concentration of kerosene-cut hydrocarbons or the concentration of styrene for pyrolysis of polystyrene without a catalyst or in the presence of red mud (Table 3). For the oil produced in the presence of ZSM-5, however, increasing temperatures led to increasing concentrations of C7–C17 hydrocarbons and a higher concentration of styrene. At higher temperatures, the catalytic activity of the ZSM-5 increased, and more of the initial feedstock was depolymerized through β-scission while a higher rate of catalytic C-C bond cleavage produced more products in the C7–C17 range [22]. The increased concentration of styrene is likely because it is resonance stabilized and a more thermodynamically favored product [48].

By far, the largest indicator of both the concentration of kerosene-cut hydrocarbons and the concentration of styrene in the polystyrene oils was the type of catalyst used. Red mud increased the concentration of kerosene-cut hydrocarbons while decreasing the percentage of styrene in that fraction from 93.2% for the no-catalyst case to 86.2% (Table 3). The ZSM-5 induced an even larger change, increasing the concentration of C7–C17 compounds by more than 50% compared to the un-catalyzed process and decreasing the percentage of styrene in the oil to 41.5% (Table 3). The use of catalysts increased the yield of kerosene-range compounds by facilitating β-scission and C-C bond cracking reactions that resulted in lighter-weight aromatics [22,23]. In particular, ZSM-5, with a high specific surface area and strong acidic sites, catalyzed C-C bond cracking reactions that led to the production of toluene, ethylbenzene, and α-methylstyrene instead of styrene monomers [22,23,29]. ZSM-5 also catalyzed alkylation reactions that produced PAH, a set of reactions not observed in the no-catalyst or red mud-catalyzed processes [19,21,26]. It should be noted that these results also indicate that polystyrene pyrolysis, either with no catalyst or red mud, can be an effective means of recycling polystyrene into styrene monomers to produce new polystyrene polymer products [49,50].

The oil produced from pine pyrolysis had much lower concentrations of C7–C17 hydrocarbons and was largely composed of oxygenated compounds (Table 3). At higher temperatures and in the presence of ZSM-5, deoxygenation and dehydration reactions were more prevalent [18,21]. Conversely, the iron in red mud catalyzed redox reactions, increasing the proportion of oxygen-containing compounds, particularly alcohols and organic acids [22,42].

For the oils produced through the co-pyrolysis of polystyrene and pine, the use of catalysts decreased the concentration of kerosene-cut compounds and had only minimal effects on the proportions of oxygen-containing compounds and styrene monomers in the oil (Table 3). As stated previously, the catalysts had a diminished impact during the co-pyrolysis process because the synergistic effects originating from pine and polystyrene volatile interactions induced many of the same reactions catalyzed by the red mud and ZSM-5 [22,25]. The slightly decreased concentration of C7–C17 hydrocarbons is a result of the combined cracking effects of both the catalyst and the synergistic effects originating from co-pyrolysis [19].

2.6.2. Comparison of Co-Pyrolysis Oil to Theoretical Predictions

Compared to the theoretical values, the actual co-pyrolysis oils had higher concentrations of C7–C17 molecules, lower concentrations of oxygen-containing molecules within the kerosene-cut range, and higher concentrations of styrene (Table 3). While these results are expected as the organic phase collected from the co-pyrolysis process is predominately composed of polystyrene volatiles, they still clearly indicate that positive synergistic effects occurred during the co-pyrolysis process [25]. During co-pyrolysis, decarboxylation and Diels–Alder reactions convert oxygen-containing compounds into aromatics, while hydrogen transfer and dehydration reactions change phenolics into nonoxygenated aromatics [19,21,24,26]. The result of these reactions is a significant decrease in oxygen-containing compounds compared to what was theoretically predicted and an increase in kerosene-cut compounds as lighter olefins that would be lost as non-condensable gases are incorporated into aromatics through the Diels–Alder reactions [21,24].

2.7. Overall Kerosene-Cut Yield

The total yield of C7–C17 hydrocarbons, in milligrams, produced per gram of initial feedstock was calculated by multiplying the concentration of C7–C17 compounds per gram of pyrolysis oil with the oil yield per gram of feedstock (Equation (S3)). Due to both the high yield of oil and the high percentage of C7–C17 hydrocarbons within that oil, as a feedstock, polystyrene produced the most kerosene-cut hydrocarbons per gram of feedstock. Higher temperatures and the use of catalysts further increased the yield by more completely cracking the initial feedstock into kerosene-cut compounds via β-scission and C-C cracking reactions as described previously [21,22]. Conversely, the pine feedstock had very low yields due to a small oil yield that only marginally increased using catalysts.

The uncatalyzed co-pyrolysis of polystyrene and pine led to yields of kerosene-cut hydrocarbons that were 22% higher than theoretically predicted. The maximum yield of 266 mg/g at 550 °C means that 26.6% of the initial mass of feedstock was recovered as kerosene-cut fuel that could be blended with SAF to produce aviation fuel with the necessary concentrations of aromatic compounds. This high yield can be attributed to the occurrence of Diels–Alder and hydrogen transfer reactions, which occur during co-pyrolysis and recover a higher proportion of pine volatiles as organic-phase oil than during pine pyrolysis individually [22,25].

Both the co-pyrolysis of polystyrene and pine in the presence of red mud and the co-pyrolysis of polystyrene and pine in the presence of ZSM-5 led to yields of kerosene-cut hydrocarbons lower than what was theoretically predicted. While the combined effects of the co-pyrolysis process and the catalysts further cracked some desired compounds into non-condensable gases, the largest reason for the less-than-predicted yield was the increased cracking of large aromatic molecules, a result that did not occur when the catalysts were used during polystyrene pyrolysis [19,33]. Guo et al. [22] reported a similar trend in their work with poplar, polystyrene, and HZSM-5, although no justification for this observation was provided. A likely explanation for this occurrence is that pine volatiles, which are produced at temperatures from 225 to 500 °C, occupied the active sites of the catalysts before that of the polystyrene volatiles, which have a maximum weight loss at 420–440 °C [35]. The pine volatiles then prevented the secondary cracking of the polystyrene volatiles, which occurred readily during polystyrene-only pyrolysis. Another possibility is that some of these heavy aromatic molecules were produced through the Diels–Alder reactions that occur synergetically during the co-pyrolysis process [22,25].

3. Materials and Methods

3.1. Materials

The feedstocks used in this study were polystyrene and southern loblolly pine (Pinus taeda). The polystyrene plastic was obtained as disposable clear plastic cutlery bought from www.Amazon.com (accessed on 11 January 2023) (Fuling brand, ASIN: B08MF4T6C3, Chongqing, China). The plastic utensils were crushed into a uniform particle size using a 5.5 horsepower granulator with a 4 mm screen (SG-2042NCH, Shini, Willoughby, OH, USA). The pine biomass was obtained from Auburn University forestry tracts, manually stripped of its bark using a chainsaw with a log peeler attachment, sun-dried under natural conditions, and chipped using a Morbark Eager Beaver M12R (Morbark LLC, Winn, MI, USA). After further solar drying the pine biomass was again size reduced using a hammer mill with a 4 mm screen (MKHM158B, USA Pellet Mill, Opa-Locka, FL, USA).

Two catalysts were used in the experiment: Zeolite Socony Mobil-5 (ZSM-5) and bauxite residue (red mud). The ZSM-5 catalyst was purchased from ACS Material (Pasadena, CA, USA) and received in a calcined and pelletized form with 30% binder by weight. The ZSM-5 catalyst has a silica-to-alumina ratio (SiO₂/Al₂O₃) of 38, a minimum specific surface area of 250 m²/g, and an average pore size of 5 Å [51]. The catalyst was ground to a fine powder using a horizontal planetary ball mill (MSK-SFM-1S, MTI Corporation, Richmond, CA, USA). The powdered ZSM-5 was passed through a No. 140 mesh sieve to ensure no large particles remained. The red mud (bauxite residue) was obtained as a waste material from Almatis Burnside, Inc. (Gonzales, LA, USA). The catalyst has a specific surface area of 21.92 m²/g and an average pore size of 6.15 nm [52]. The red mud was not acid-washed or calcined and was first dried and then crushed using a mortar and pestle. The pulverized catalyst was passed through a No. 140 mesh sieve to guarantee a uniform particle size.

3.2. Feedstock Characterization

Proximate analysis of the polystyrene and pine feedstocks, including determining the moisture, ash, volatiles, and fixed carbon contents, was conducted. The moisture content was determined according to ASTM standard E871-82 [53] using an automatic moisture analyzer (MJ33, Mettler Toledo, Greifensee, Switzerland) and reported on a wet basis. The ash content was measured per ASTM standard E1755-01 [54] specifications using a programmable muffle furnace (LT 9/13, Nabertherm, Lilienthal, Germany). Using the same muffle furnace, the volatile matter was determined according to ASTM standard E872-82 [55]. The fixed carbon content was determined by difference. The ash, volatiles, and fixed carbon contents were reported on a dry basis. Ultimate analysis of the feedstocks was performed using an Elementar elemental analyzer (vario MICRO select, Elementar, Langenselbold, Germany), which measured the carbon, nitrogen, hydrogen, and sulfur content directly, with the oxygen content being determined by difference.

3.3. Experimental Setup

A bench-scale fixed-bed pyrolysis reactor was used in the investigation, and the schematic of the reactor is shown in Figure 6. The reactor consisted of a 19-inch long, 1-inch outer diameter (0.875-inch inner diameter with walls approximately 0.0625 inches thick) 304 stainless steel tube. The pipe was fitted with Swagelok^® (Swagelok, Solon, OH, USA) ferrule fittings on either end, with one end connecting to a pressurized nitrogen gas line and the other connecting to the condenser train assembly. The reaction tube was placed horizontally inside a ceramic tube furnace with a 12-inch heating zone (TF55035 A-1, Lindberg/MPH, Riverside, MI, USA). The condenser train assembly consisted of one custom-made 304 stainless steel Dreschel Bottle and three glass Dreschel Bottles (5516-08, Ace Glass Incorporated, Vineland, NJ, USA) connected with vinyl tubing. The stainless-steel bottle was used for its high heat tolerance and cooled by natural convection in air and the three glass bottles were placed in an ice bath to aid in condensation of the pyrolysis vapors. Following the condenser train assembly, the pyrolysis gases passed through an electrostatic precipitator (ESP) made from a 2-inch schedule 40 PVC pipe (2.049-inch average inner diameter with walls a minimum of 0.154 inches thick) with a 1/16” inch diameter mild steel TIG rod functioning as the anode. The connections and fittings for the ESP were 3D printed, and the ESP was fitted with a stainless-steel collection bottle at the bottom to capture any pyrolysis oil condensed in the ESP. The ESP was charged at 20 kV and grounded from the exterior of the PVC pipe shell. After passing through the ESP, the gas stream was exhausted to the atmosphere.

Glass wool was inserted into the end of the reactor tube nearest the nitrogen gas inlet line before loading the reactor with feedstock to ensure that the feedstock remained within the portion of the tube that would be located within the heating zone of the furnace. For the pure polystyrene and pure pine feedstocks, 40 g of feedstock was massed and poured directly into the reactor tube. For the 1:1 blend of polystyrene and pine, 20 g of each type of feedstock was combined and stirred by hand to form a homogenous mixture before being poured into the reactor tube. For trials with catalysts added to the feedstock, 40 g of pure feedstock or 1:1 blended feedstock was mixed by hand with 4.5 g of the respective catalyst to form a 90% feedstock and 10% catalyst mixture by weight. The mixed feedstock/catalyst mixture was then loaded into the reactor tube.

During each experimental trial, ultra-high purity (UHP) grade nitrogen gas (Airgas, Opelika, AL, USA) at 20 PSIG was used as a carrier purge gas and flowed through the reactor at a volumetric flow rate of 0.75 L per minute. The ceramic heater was set with a fixed ramp rate of 50 °C per minute up to the set reactor temperature and the reactor tube was held at the fixed temperature for three hours. This extended reaction time does not reflect the required time for the reaction to occur but was instead used to prevent erroneous results originating from incomplete pyrolysis at the edges of the heating region. The three glass condensers were held in an ice bath at approximately 0 °C, and the ESP was charged at a voltage of 20 kV.

3.4. Experimental Design

There were three variables altered during the investigation: the pyrolysis temperature, the feedstock composition, and the type of catalyst used, if any. Each variable had a total of three levels. The pyrolysis temperature varied between 450 °C, 500 °C, and 550 °C. The feedstock composition was either 100% pine, a 1:1 blend of pine and polystyrene, or 100% polystyrene, and there was either no catalyst, 4.5 g of red mud, or 4.5 g of ZSM-5 added to the feedstock blend. Each unique combination of the variable levels was tested for a total of 27 unique experiments. Each experiment was duplicated to corroborate the results and if there was a statistically significant variation between the two trials, it was assumed that an error occurred, and the given experiment was repeated for a third time. The most common error leading to statistically significant outliers was incomplete pyrolysis of the entire feedstock, and as such, these results were deemed invalid and removed from consideration.

3.5. Mass Balance

Before loading the pyrolysis reactor and beginning the experiment, the reaction tube and condenser train, including the stainless-steel collection bottle from the ESP, were each massed. After loading 40 g of feedstock for trials without catalysts or 44.5 g of blended feedstock and catalyst for the remaining trials, the reaction tube was again massed to ensure that the proper amount of feedstock was loaded. Upon completion of the three-hour pyrolysis experiment, the furnace was switched off, and the nitrogen purge rate was increased to 2 L per minute for 10 min to force any pyrolysis vapors or condensed oils from the reaction tube into the condenser train. After the reaction tube cooled to room temperature, the end of the reactor tube was opened, and any pooled oil was scrapped out of the reactor with a spatula and deposited in the steel Dreschel Bottle from the condenser train. The reaction tube was massed to determine the amount of solid residue (char or coke) remaining in the tube by the difference with the initial mass of the tube. By a similar method, the mass of liquid products collected was determined from the final mass of the condenser train. The mass of the gas phase was obtained by difference, as shown in Equation (1), and the percent mass fraction of each phase was calculated using Equation (2).

$$m_g=m_f-m_s-m_{cl} \left(g\right)$$

(1)

$${\%m}_{phase}={\displaystyle \frac{m_{phase}}{m_f}}\times 100$$

(2)

where $m_g$ is the mass of the gas phase, $m_f$ is the mass of the original feedstock loaded (weight of catalyst excluded), $m_s$ is the mass of the char (4.5 g were subtracted from the total observed mass of solids for each trial using a catalyst to account for the mass of the remaining catalyst), $m_{cl}$ is the total mass of the combined liquid phase, ${\%m}_{phase}$ is the percent mass fraction of a given phase, and $m_{phase}$ is the mass of a given phase.

The liquid phase was centrifuged at 2000 rpm (Allegra 6 Centrifuge, Beckman Coulter, Brea, CA, USA) for 10 min to separate the organic phase from the aqueous phase. After centrifugation, the hydrophobic organic phase was extracted from the centrifuge vial using a syringe. The mass of both the extracted organic oil phase and the remaining aqueous phase was measured. The percent mass fraction of the combined liquid phase was further separated into the organic oil and aqueous phases and calculated using Equation (2), where $m_{phase}$ was replaced with either the mass of the oil or the aqueous phase.

3.6. Product Analysis

The total organic carbon (TOC) and total organic nitrogen (TON) of the aqueous phase were determined using a Shimadzu TOC/TON analyzer (TOC-L CPN/TNM-L, Shimadzu, Kyoto, Japan) with the samples diluted in a 200:1 ratio.

Characterization of the pyrolysis oil included ultimate analysis, determining the moisture content and higher heating value of the oil, and chemical analysis by GC/MS. Ultimate analysis was performed using an Elementar elemental analyzer, as documented previously. The moisture content was determined using an automated volumetric Karl-Fischer titrator (V20 Volumetric KF Titrator, Mettler Toledo, Greifensee, Switzerland) with one component reagent for volumetric Karl Fischer titration for aldehydes and ketones (CombiTitrant 5 Keto, Millipore Sigma, Burlington, MA, USA) and its associated solvent (CombiSolvent Keto, Millipore Sigma, Burlington, MA, USA) being used for the titration. The higher heating value (HHV) was measured using an IKA C200 bomb calorimeter with 500 mg of sample loaded (C200, IKA, Wilmington, NC, USA).

The GC/MS samples were prepared by dissolving 50 mg of oil in 2 mL of dichloromethane (DCM) (BDH1113-4LG, VWR, Radnor, PA, USA). The samples were run in Agilent Technologies GC/MS (GC, 8890, Agilent Technologies, Santa Clara, CA, USA; MS, 5977C, Agilent Technologies, Santa Clara, CA, USA). The integration parameters were set such that the total number of identified and integrated peaks for each sample was between 30 and 50. The NIST20 library (National Institute of Standards and Technology, Gaithersburg, MD, USA) was used to identify each of the integrated peaks. Further details on the GC/MS analysis are provided in the supplemental information.

The chemical compounds of greatest interest when considering the production of aviation fuels include paraffins, naphthenes, aromatics, olefins, and oxygen-containing compounds with carbon numbers from C7 to C17. From the GC/MS data, the percent fraction of a given class of compounds was determined by summing the percent area peak of each qualifying compound of a given class together to obtain the overall percentage. A similar process was performed when grouping compounds by carbon number (Figure 2, Figure 3, Figure 4 and Figures S1–S3).

GC/MS Calibration

From the compounds observed in the collected oils, thirteen representative compounds were selected to prepare a calibration solution. The compounds selected included aromatics of varying carbon numbers, a two-ring polyaromatic hydrocarbon (PAH), a three-ring PAH, and various oxygenated aromatic compounds (Table S6). A detailed explanation of the calibration process is provided in the Supplemental Information.

3.7. Statistical Analysis

A two-way factorial ANOVA test was used to determine the significance of the variations in the yield of the pyrolysis products. A separate test was performed for each product (char, gas, oil, and aqueous) from each feedstock (polystyrene, pine, and blended feedstock) for a total of 12 tests. The two factors were catalysis (with three independent levels: no catalyst, red mud, and ZSM-5) and temperature (with three independent levels: 450 °C, 500 °C, and 550 °C). A critical value, α, of 0.05 was used. F statistics were calculated for the catalysis factor, the temperature factor, and the interaction of the two factors. The F statistics were also converted to p-values (representing the probability of a given result occurring assuming that a given factor did not affect the product yields) and reported as such.

Post hoc testing was performed to determine which combinations of levels were statistically significant. Tukey’s Honestly-Significant-Difference (Tukey HSD) was used to determine this again, utilizing a 95% confidence level, α = 0.05. The upper and lower limits of the 95% confidence interval and the p-value of the difference were reported. All statistical testing was performed using the R (version 4.4.0) software package [56].

3.8. Theoretical vs. Ovserved Results

The presence or absence of synergistic effects occurring during the co-pyrolysis of pine and polystyrene was determined according to the methods presented in earlier works by Chen et al. [57] and Zhuang et al. [58]. Applied to this work, this method involved comparing the observed values of liquid yield and concentration of kerosene cut hydrocarbons from the 1:1 blend of pine and polystyrene with the theoretical values predicted by averaging the observed values from pyrolysis of pure pine feedstock and pyrolysis of pure polystyrene feedstock individually. The observed values for both the blended feedstock and the pure feedstocks were determined using the methods presented earlier. The theoretical values, assuming no synergistic effects, were calculated using Equations (3) and (4).

$${\%}_{theoretical oil yield }=0.5\left({\%}_{pine oil yield}\right)+0.5\left({\%}_{polystyrene oil yield}\right)$$

(3)

$${\left[kerosene\right]}_{theoretical}=0.5\left({\left[kerosene\right]}_{pine}\right)+0.5\left({\left[kerosene\right]}_{polystyrene}\right)$$

(4)

where $\left[kerosene\right]$ is the concentration of the kerosene cut in the pyrolysis oil.

4. Conclusions

This work investigated the potential for producing kerosene-cut aromatic aviation fuels through the pyrolysis of polystyrene and the co-pyrolysis of polystyrene and pine at various reaction temperatures ranging from 450 °C to 550 °C and in the presence of different catalysts, including red mud and ZSM-5. Additionally, calibrated GC/MS was performed to quantify the exact concentration of compounds in the pyrolysis oils, a first-of-its-kind feat for the co-pyrolysis of polystyrene and pine.

The addition of ZSM-5 and red mud increased cracking reactions in the polystyrene feedstock, which increased the gas yield and decreased the liquid yield. Catalytic pyrolysis produced 5.5% and 1.5% more gas and 5% and 1% less oil for ZSM-5 and red mud, respectively. The synergistic effects originating from the co-pyrolysis of polystyrene and pine exhibited similar catalytic effects to that of the ZSM-5 catalyst and induced increased cracking of solid and liquid products into gases. The co-pyrolysis process also diminished the effects of the catalysts so that they did not have a significant additional effect on the co-pyrolysis yields. Averaged across all temperatures, the blended feedstocks produced 8.25% more gas and 3.6% fewer liquid products than theoretically predicted, providing evidence that increased cracking reactions occurred. The effect of temperature on the yield of pyrolysis products varied from feedstock to feedstock, but in general, increasing temperatures tended to lead to increased liquid yields and decreased solid yields.

The products of polystyrene pyrolysis were mainly styrene, ethylbenzene, naphthalene, phenanthrene, and other related monoaromatic hydrocarbons (MAHs) and polyaromatic hydrocarbons (PAHs). The use of a ZSM-5 catalyst increased the yield of C7–C17 hydrocarbons by 43.6% compared to polystyrene pyrolysis without a catalyst. Although not suitable for sustainable aviation fuel (SAF) production, the pyrolysis of polystyrene with red mud produced 41% styrene monomers by weight, indicating that this is a viable process for recycling polystyrene into new polymers.

The oil products from co-pyrolysis exhibited characteristics more closely related to the pure polystyrene oil with higher carbon content and lower oxygen content than theoretically predicted. However, compared to the polystyrene oil, the co-pyrolysis oil had 85% less PAH than the ZSM-5 catalyzed polystyrene oil, a desirable characteristic for SAF. During co-pyrolysis, pine volatiles were converted into nonoxygenated aromatics through Diels–Alder, and dehydration reactions were made possible through hydrogen transfer from the polystyrene feedstock. This resulted in yields of kerosene-cut hydrocarbons that were 22.5% higher than theoretically predicted, with a maximum yield of 266 mg of C7–C17 hydrocarbons produced per gram of feedstock through co-pyrolysis at 550 °C. The addition of catalysts only decreased the yield and offered little to no additional benefits.

In conclusion, pyrolysis of polystyrene at 550 °C with ZSM-5 catalyst or co-pyrolysis of pine and polystyrene at 550 °C without a catalyst is recommended as the optimal conditions for producing kerosene-range aromatic hydrocarbons for their use as additives to SAF. These aromatic-rich fuels can be blended with traditional SAF to produce aviation fuels with the proper concentration of aromatics to qualify as turbine-ready fuels. Future continuations of this work will involve blending the aromatic-rich oils produced in this study with SAF produced through conventional methods and testing those blends for compliance with ASTM standards.