Experimental and Kinetic Simulation Study of the High-Temperature Pyrolysis of 1,2,4-Trimethylbenzene, 1,3,5-Trimethylbenzene and n-Propylbenzene

Abstract

This paper reports a comparative study on the high temperature pyrolysis characteristics of three C₉H₁₂ isomers, including n-propylbenzene (PBZ), 1,3,5-trimethylbenzene (T135MBZ), and 1,2,4-trimethylbenzene (T124MBZ), via single-pulse shock tube (SPST) experiments and kinetic simulations. The SPST experiments were conducted in the temperature range of 1100–1700 K, at pressures of 10 bar and 15 bar, with a fixed fuel concentration of 200 ppm. The reaction time was approximately 1.8 ms for all of the experiments. The distributions of the pyrolysis products were quantitatively analyzed as functions of pressure and temperature. A detailed kinetic mechanism was used to simulate the experimental results, and it is demonstrated that the mechanism can capture the pyrolysis characteristics reasonably well. Both experimental and simulation results reveal that PBZ exhibits higher fuel reactivity than T124MBZ and T135MBZ under the studied conditions. Pyrolysis of all three C₉H₁₂ isomers generates key soot precursors, including acetylene and benzene. Sensitivity and rate-of-production (ROP) analyses indicate similar primary pyrolysis pathways. The benzyl radical is first formed through the dehydrogenation reaction and then it undergoes a series of decomposition reactions leading to the detected small hydrocarbon species. This study not only provides an in-depth understanding of the high temperature pyrolysis characteristics of the three C₉H₁₂ isomers, but also provides essential validation data for the development and optimization of chemical kinetic mechanisms for alkyl aromatic hydrocarbons.

1. Introduction

Fossil fuels remain the dominant global energy source, but their combustion contributes to climate change and air pollution. Alkyl aromatic hydrocarbon compounds are important components in real fuels, i.e., gasoline, diesel, and kerosene [1,2,3,4], and previous studies have shown that the co-pyrolysis of solid biomass and plastic waste can increase the formation of aromatic hydrocarbons [5]. Among the various aromatic hydrocarbons, the isomers of C₉H₁₂ fuels including 1,2,4-trimethylbenzene (T124MBZ), 1,3,5-trimethylbenzene (T135MBZ), and n-propylbenzene (PBZ) are widely used to represent the aromatic hydrocarbons in real fuels to develop surrogate models [6,7,8]. In addition, aromatic hydrocarbon compounds are also considered to be an important source of soot formation because they can easily transform to polycyclic aromatic hydrocarbons (PAHs) during combustion processes [9]. The formation of soot will not only reduce the combustion efficiency, but also pose a huge threat to humans and the environment. Hence, in order to improve the combustion efficiency and to reduce emissions, a better understanding of the combustion and pyrolysis properties of aromatic hydrocarbon compounds are necessary.

While the oxidation properties of C₉H₁₂ (e.g., ignition delay times, flame speeds, and species profiles) have been extensively studied [10,11,12,13,14,15,16,17,18,19,20,21], their pyrolysis characteristics—particularly in a comparative framework—remain poorly understood. Zhao et al. [22] studied the pyrolysis of PBZ in a flow reactor at temperatures from 950 to 1450 °C. Gudiyella et al. [23] studied the pyrolysis characteristics of PBZ in an SPST within the temperature range of 850–1150 K and pressure at 50 atm, and established a detailed chemical kinetic mechanism, which provided a good simulation of fuel consumption and the distribution of major aliphatic, monoaromatic, and polycyclic aromatic hydrocarbons. Yuan et al. [11] investigated the pyrolysis of PBZ in a flow reactor at 0.04, 0.2, and 1 atm pressures by synchrotron vacuum ultraviolet photoionization mass spectrometry. Sun et al. [24] investigated the high temperature pyrolysis characteristics of PBZ at a pressure of 20 bar and temperature of 950–1700 K using an SPST. Compared with PBZ, studies on the pyrolysis characteristics of T135MBZ are scarce [25].

For the C₉H₁₂ isomer, there have also been few systematic studies. Ji et al. [26] experimentally studied the laminar flame speeds and extinction strain rates of benzene, PBZ, toluene, o-xylene, m-xylene, p-xylene, T124MBZ, and T135MBZ under atmospheric pressure and an initial temperature of 353 K. Hui et al. [20] investigated the laminar flame speeds and premixed extinction limits of PBZ, T124MBZ, T135MBZ, and toluene to evaluate the effect of different alkyl substitutions to the benzene ring on flame propagation and extinction. Liu et al. [21] studied the oxidation reactions of PBZ, iso-propylbenzene, T135MBZ, and T124MBZ in a jet-stirred reactor to find the most suitable alternative transportation fuel. Liang et al. [15] studied the ignition delay times (IDT) of PBZ, T135MBZ, and T124MBZ in a high-pressure shock tube (HPST) and simulated the experimental results using an updated detailed kinetic mechanism.

Based on the above considerations, although extensive studies have been performed on the three C₉H₁₂ isomers, comparative studies on the pyrolysis properties are rather scarce. Thus, this work intends to perform a comparative study on the high temperature pyrolysis characteristics of PBZ, T135MBZ, and T124MBZ via a novel SPST experimental system. To uncover the underlying chemical kinetics of pyrolysis processes of the three fuels, a combustion chemical kinetic mechanism is constructed to simulate the experimental results, and a sensitivity analysis and ROP analysis are further performed.

2. Experimental Methods

The experiments were carried out via an SPST at the North University of China. The detailed experimental procedures have been described in previous studies [27] and only a brief introduction is provided here. The total length of the facility is 4.55 m and the inner diameter is 44 mm. The diaphragm divides the shock tube into a 3.05 m driven section and a 1.5 m driver section. The schematic diagram of the system is shown in Figure 1. An optical window was included in the experimental system, but the optical window was not used for any measurements throughout the experiment. In order to ensure that the experiment was carried out only under single-pulse shock wave heating conditions, the driven section of this shock tube was connected to the Dump Tank via a baffle valve with an internal diameter of 44 mm to absorb the reflected shock waves generated during the experiment. The incident shock-wave velocity was measured by four piezoelectric sensors of the PCB series mounted on the side wall and one Kistler 6125B piezoelectric pressure transducer mounted on the end cap of the driven chamber. The distances of the four pressure sensors from the end caps of the low-pressure section were 204.5 cm, 154.5 cm, 104.5 cm, and 54.5 cm. The reaction time τ [28] in this text is defined as the time interval between the initial point of the reflected shock wave peak surface and the point at which the pressure of the reflected shock-wave peak surface drops to 80%. A time record diagram of the pressure history after the reflected shock wave is shown in Figure 2. The GasEq program [27] was used to calculate the temperature and pressure following the shock wave. To prevent condensation of the liquid fuel in the driven section, the entire experimental setup was maintained at 100 °C, while the sampling section was heated to 210 °C. The collected samples were analyzed using an Agilent 7820 A gas chromatograph (GC) equipped with a flame ionization detector (FID) and a thermal conductivity detector (TCD), as well as an Agilent 7890B-5975B gas chromatography-mass spectrometer (GC-MS). Sampling was triggered by a fast-response solenoid valve, starting approximately 3 ms after the reflected shock wave reached the endwall and lasting for 100 ms. Pyrolysis product concentrations were quantified using calibrated reference standards. For species without direct calibration, the effective carbon number method was applied to estimate their concentrations. Additional products not covered by the 13 GC standards were measured through multi-point calibration using custom-prepared mixtures of varying concentrations. The in-house mixture was a mixture of different concentrations of gas diluted by argon after evaporation of a standard liquid sample with a purity of 99.9%. With the exception of 13 standard gases, the rest of the products were measured with the gas chromatography-mass spectrometer (GC-MS).

The overall uncertainty of the product concentration, reaction time, and temperature in the experiments is generally consistent with other similar facilities and related studies [29,30,31]. The measured temperature variations behind the reflected shock wave, due to uncertainties in shock velocity measurements, are within ±2% [32]. For product concentrations, the uncertainty of calibration species calculated using repeated sampling of standard gases is approximately ±10% [33]. The effective reaction time was measured by He et al. [34], and the uncertainty was ±5%. Therefore, this experiment has high accuracy and can provide valuable data for the validation of detailed kinetic mechanisms.

The three fuels used in the experiment were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd., Shang Hai, China and the detailed experimental conditions are shown in

Table 1. Experimental conditions of C₉H₁₂ fuels.

Fuel	XFuel	T5 (K)	P5 (bar)	XKr	XAr	Reaction Time (ms)
PBZ	0.02%	1100–1500	10.0, 15.0	0.2%	99.78%	1.8
T124MBZ		1200–1700
T135MBZ		1200–1700

. In Table 1, T₅ represents the temperature behind the reflected shock and P₅ represents the pressure behind the reflected shock. Helium (He) was used as the driving gas, argon (Ar) was the dilution gas, and Krypton (Kr) was the internal standard gas. The system was calibrated with 13 gas chromatography standards from Jinghan Gas Co., Ltd. Wu Han, China and Shanghai Maclean’s Biochemical Technology Co., Ltd., Shang Hai, China and the concentration of pyrolysis products was calculated using the calibrated standards. The standard gas was a mixture of hydrogen, methane, ethane, ethylene, propane, propylene, acetylene, propadiene, propylene, 1-butene, 2-butene, and 1,3-butadiene diluted with nitrogen, all at a concentration of 150 ppm. All fuels, He and Ar had purities of ≥99.9% (fuels) or ≥99.999% (gases). In this experiment, the initial fuel concentration was 200 ppm diluted by Ar gas. The low fuel concentration was used to avoid the clogging of the analytical system by the soot particles formed during the pyrolysis process entering the gas chromatography line. The Ar-diluted fuel mixture was prepared by the Dalton partial pressure method in a vacuum gas distribution tank at 100 °C, and the pressure was monitored by two membrane vacuum gauges with a maximum capacity of 10 torr and 1000 torr (the specific model was Inficon CDG100) and a digital pressure gauge (the specific model was the KY2010 digital precision pressure gauge produced by Beijing Kaihang Weiye Co., Ltd., Beijing, China), and the pressure was allowed to stand for more than 12 h after the pre-distribution was completed. In addition, the inner surface of the driven section was cleaned with silk stained with absolute ethanol after each group of experiments to remove carbon deposits and ensure the accuracy of the experiment. Before each set of experiments, test experiments were performed with pure argon to verify that the contaminant levels were within the acceptable range of experimental error.

In addition, due to the limitations of the analytical equipment, it is hard to calculate the carbon balance. The formation of soot and large polycyclic aromatic hydrocarbons (PAHs) under high-temperature conditions—which fall outside of the detection range of this study—further complicate carbon balance determination. Nevertheless, based on prior pyrolysis research involving fuels with well-defined compositions, the uncertainty in carbon balance has been maintained within 85–115% across all experimental conditions. The accuracy of carbon balance primarily depends on factors such as the absorption in the mixture tank, shock tube, and sampling line, as well as the GC and GC-MS analytical methods. Previous studies on other fuels have confirmed the reliability of this experimental system in assessing the carbon balance. While this study does not quantify large hydrocarbon or soot-related compounds, the measured distributions of small C1–C6 species remain crucial for developing detailed reaction mechanisms for aromatic hydrocarbons. These smaller molecules play a fundamental role in determining the high-temperature combustion behavior of aromatic fuels.

3. Kinetic Modeling

Separate detailed chemical kinetics have been developed for each of the three studied isomers, but these mechanisms are primarily designed for specific fuels. For example, the JetSurF 2.0 mechanism [35] was developed mainly for benzene, toluene, ethylbenzene, and PBZ, but it does not include branched alkyl aromatics. The Liu mechanism [21] showed acceptable performance in the prediction of jet-stirred reactor experimental results for four C₉H₁₂ isomeric transportation fuels. The detailed kinetic mechanism developed by Kukkadapu et al. [36] for C₇–C₁₁ methylated aromatics included sub-mechanisms for T124MBZ and T135MBZ. In addition, it is worth noting that the Kukkadapu mechanism was developed based on the reaction mechanism of alkylbenzene that Mehl et al. [14] constructed, and the reaction of C₀–C₄ is taken from AramcoMech 2.0 [20]. The NUIGMECH1.1 mechanism has been developed by re-evaluating the kinetics and thermochemistry of C₀–C₄ base chemistry based on existing studies and experimental diagnostics [27,37,38,39,40]. To ensure consistency in the underlying mechanism development, a new kinetic mechanism was constructed by coupling the NUIGMech1.1 core mechanism [31,41,42] with the sub-mechanisms for T124MBZ [37], T135MBZ [37], and PBZ [35]. The detailed mechanism includes 1105 species and 6219 reactions. The current mechanism has been validated against the experimental data of IDTs [15]. Thus, the current mechanism was employed to systematically analyze the pyrolysis chemistry of three C₉H₁₂ isomers. Based on the SPST experimental system, the pyrolysis process of the fuel was simulated by a closed-type homogeneous reactor with constrained volume and solved using the energy equation module implemented in the Chemkin II and SENKIN software [43,44]. In addition, the sensitivity and rate-of-production analyses performed in this study were simulated with ChemkinII software. The simulation run time was defined based on the experimental results. The averaged reaction time during the shock tube experiment was 1.8 ms, and this was defined as the simulation time. At the end point of the simulation process, the species concentrations were derived to compare with experimental results. For comparisons, the other mechanisms related to the studied fuels were also used for simulations. Specifically, the mechanism developed by Liu et al. [21] was used for modeling the pyrolysis of the three fuels, and the other three mechanisms corresponding to PBZ [22], T124MBZ [45], and T135MBZ [46] were also employed to simulate the experimental results. The results are provided in the Supplementary Materials. These mechanisms generally show very similar prediction results.

4. Results and Discussions

4.1. Experimental and Simulation Results

In this section, experimental and simulation results of the species concentrations as a function of temperature are shown to clarify the high-temperature properties of the three fuels. Figure 3 shows the concentration distribution curves of the three fuels as a function of pressure and temperature. For each fuel, the measured fuel concentration profiles at pressures of 10 and 15 bar are almost identical. The pyrolysis rate of PBZ decreases gradually with the increase in temperature, and reaches the minimum rate at 1400 K, and the pyrolysis degree of PBZ is about 90% at this time. The experimental results demonstrate that reactant fuel concentrations undergo rapid depletion with increasing temperature, approaching near-zero values. However, a fuel concentration of zero was not achieved at the studied conditions, primarily due to the constrained reaction time of 1.8 ms and studied temperature range, which limits the extent of fuel decomposition. Compared to PBZ, the two trimethylbenzene fuels begin pyrolysis at a relatively high temperature, which is about 100 K higher than that of PBZ. The consumption rates of two trimethylbenzene fuels are very similar. Furthermore, the simulated curve for PBZ demonstrates a larger deviation from the experimental results compared with the other two studied isomers. Generally, it can be seen that all of the simulated curves of the three fuels show a steep changing trend as the fuel pyrolysis begins. The current model simulation results reveal that PBZ exhibits accelerated decomposition kinetics at temperatures exceeding 1150 K, corresponding to residual fuel concentrations below 50 ppm—a trend consistent with observations for both T124MBZ and T135MBZ. This kinetic discrepancy may stem from incomplete representation of polycyclic aromatic hydrocarbon growth mechanisms in current models, potentially leading to over-predicted reactivity. To address this limitation, further enhanced PAH speciation measurements as well as PAH chemistry are needed to improve the kinetic model’s predictive capabilities regarding soot precursor formation. As is shown in Figure 3, measured fuel concentration profiles at pressures of 10 and 15 bar are very close and within the uncertainties. The simulation results at both pressures are almost identical, and this is because the rate constants of the direct decomposition reaction are very close under the studied pressure (10 and 15 bar) and temperature conditions. Only when the pressure shows large differences (i.e., from 5 to 100 bar) does the rate constant change become slightly greater under high-temperature conditions (usually >1500 K). Thus, the kinetic modeling results for all the three fuels demonstrate the minimal effect of pressure.

The temperature-concentration curve of the pyrolysis products of PBZ is shown in Figure 4. The results of species measured at two pressures are very close. Ethene and toluene are the main pyrolysis products, and their concentrations change with temperature, as shown in Figure 4b,f. The concentrations of ethene and toluene increase gradually with the change in temperature, and the peak temperature points are reached at 1400 K and 1350 K, respectively. The corresponding peak concentrations are 150 ppm and 110 ppm. Then, the concentrations gradually decrease with the increase in temperature. The concentrations of methane and acetylene increase with the increase in temperature. The concentration profiles of ethylbenzene and styrene with temperature are shown in Figure 4g,h. The variations in the concentrations of ethylbenzene and styrene as a function of temperature are similar, both of which gradually increase with the increase in temperature. After reaching the peak temperature point of the concentration, concentration decreases with the increase in pyrolysis temperature. Ethylbenzene reaches the peak concentration point at about 1200 K, and the maximum concentration is about 20 ppm. Styrene reaches its peak concentration around 1250 K, with a maximum concentration of about 24 ppm. The concentration profiles of benzene with temperature are shown in Figure 4e. It can be seen that the concentration gradually increases with the increase in temperature, and does not show a decreasing trend. The pyrolysis products of PBZ contain many soot precursors, among which ethene and toluene are the main pyrolysis products with concentrations of more than 100 ppm, and they can be generated stably at high temperatures. To sum up, at the studied conditions, pressure has minor effects on the pyrolysis species, while the species concentrations vary slightly with pressures changing from 10 to 15 bar. The trends in the pyrolysis products over the temperature range can be captured by the current model.

The temperature-concentration curves of the fuel and pyrolysis products of T124MBZ are shown in Figure 5. The concentrations of methane and acetylene as a function of temperature are shown in Figure 5b,c, and it can be seen that their concentrations both increase gradually with increasing temperature. The concentrations of toluene, p-xylene, and styrene with temperature are shown in Figure 5h–j. The concentrations of these three products increase gradually with the increase in temperature. And, after reaching the peak temperature point of concentration, their concentrations decrease with the increase in pyrolysis temperature. The peak concentration values are reached around 1500 K, with the peak concentrations of 12 ppm, 15 ppm, and 15 ppm. The concentrations of ethene and benzene as a function of temperature are shown in Figure 5f,g. The concentrations of ethene and benzene increase with the increase in temperature, and stabilize gradually after reaching their peak concentrations. Small but detectable quantities of propene, propyne, and naphthalene were also observed, the concentrations of which vary with temperature, as shown in Figure 5e,k,l. As can be seen from Figure 5, the prediction results of the mechanism on the residual amounts of fuel are in good agreement with the experimental results. Meanwhile, major species like methane and benzene are also well captured by the current model. The current model underpredicts the concentrations of most products, while the trends are captured. Specifically, from Figure 5, the predicted species concentrations of ethane, allene, ethene, toluene, styrene, propyne, and naphthalene from kinetic modeling are lower compared with the experimental results. Some of these species present low concentrations during the pyrolysis of T124MBZ and T135MBZ, as shown from the experimental results. This indicates that some reaction pathways leading to the formation of these species are missing in the current model.

The temperature-concentration curves of the fuel and pyrolysis products of T135MBZ are shown in Figure 6. The concentrations of methane and ethene as a function of temperature are shown in Figure 6b,e, and it can be seen that the concentrations of methane and ethene gradually increase with increasing temperature. The concentration profiles of ethane, p-xylene, and styrene are shown in Figure 6f,h,I. The concentrations of these three products gradually increase with the increase in temperature. And, after reaching the peak temperature point of the concentration, concentration decreases with the increase in the pyrolysis temperature. The peak temperature points are reached at around 1450 K, with peak concentrations of 6 ppm, 12 ppm, and 8 ppm. The concentrations of ethene and benzene changing with temperature are shown in Figure 6e,f. The concentrations of ethylene and benzene increase with the increase in temperature, and stabilize gradually after reaching their peaks. There are also small amounts of allene and naphthalene, the concentrations of which vary with temperature, as shown in Figure 6i,g. As can be seen from Figure 6, the prediction results of the mechanism for the residual amounts of reactants, methane, and p-xylene are in good agreement with the experimental results. And, the prediction results for the concentrations of other products are also basically the same. Overall, the products of the three C₉H₁₂ fuels are very similar, and the general trend in the same product is also the same. Compared with PBZ, the products of the two trimethylbenzenes have several more aromatic compounds and lower reactivity, and the molecular structure affects their reaction pathways and reactivity.

4.2. Sensitivity Analysis

To elucidate the chemical kinetics governing the pyrolysis of the three C₉H₁₂ fuels (PBZ, T135MBZ, and T124MBZ), sensitivity analyses were conducted for both the parent fuels and their major pyrolysis products. For the sensitivity analysis, the pre-exponential factor (A-factor) was systematically perturbed by a factor of 2.0 (increase) and 0.5 (decrease) from its nominal value. The sensitivity coefficient defined as ln(x+/x−)/ln(k+/k−) was computed for every reaction. Here, k is the original pre-exponential factor of the target reaction, k− is the k factor divided by 2.0, k+ is the k factor multiplied by 2.0, and x+ and x− are the corresponding species concentrations obtained with k+ and k−, respectively. For a consistent comparison of the pyrolysis mechanisms across these isomers, the analysis was performed at identical fuel conversion rates. Thus, according to the experimental and modeling results shown in Figure 3, the sensitivity analysis of PBZ pyrolysis was performed at 1100 K, 1300 K, and 1500 K under pressures of 10 and 15 bar. Figure 7 shows the sensitivity analysis results for PBZ pyrolysis at 10.0 bar. The definitions of the relevant abbreviations in the sensitivity analysis are shown in

Table 2. The definitions of the relevant abbreviations in the sensitivity analysis.

Abbreviation	P-XYL	O-XYL	IND	MEXYLYLE	MEBZCYBUT
Chemical structural formula
Abbreviation	T135MBJ	FULVENE	PBZJA	PBZJB	PBZJC
Chemical structural formula

.

For PBZ pyrolysis, at 1100 K, PBZ is mainly decomposed to form benzyl and ethyl radicals via PBZ(+M)=C₆H₅CH₂+C₂H₅(+M), which is the dominant reaction that plays a major role in promoting the formation of ethene and toluene. And, benzyl is an important intermediate reactant in the pyrolysis process of PBZ. It has high reactivity and promotes the further decomposition of PBZ. At 1300 K, the primary reaction driving PBZ decomposition is C₆H₅CH₃(+M)=C₆H₅CH₂+H(+M), where benzyl radicals react with H atoms to form stable toluene. This reaction also significantly enhances the production of ethane and toluene. Notably, it emerges as the dominant pathway for both product formation and PBZ consumption at 1100 K and 1300 K. At the same time, the reaction of PBZ dehydrogenation has been inhibiting product formation throughout the pyrolysis process.

Similarly, based on experimental and simulation results, sensitivity analysis of T124MBZ and T135MBZ pyrolysis is conducted at 1300 K, 1450 K, and 1700 K with pressures of 10 and 15 bar. Figure 8 and Figure 9 show the sensitivity analysis results of T124MBZ at 10.0 bar.

For T124MBZ, the main reaction that plays a significant role in promoting the formation of methane from the pyrolysis product is T124MBZ+H=P-XYL+CH₃. T124MBZ undergoes a dehydrogenation reaction with hydrogen radicals to form p-xylene and methane. There are two main inhibition reactions for the generation of methane: T124MBZ+H=P-XYLCH₂+H₂ and T124MBZ+H=O-XYLCH₂+H₂, both of which are the main pathways of T124MBZ consumption. Further, it is also noted that one of the two reactions, namely T124MBZ+H=P-XYLCH₂+H₂, inhibits the formation of benzene.

Similar to the analysis of T124MBZ, the dehydrogenation reaction of T135MBZ greatly contributes to its consumption and is the initial reaction of the T135MBZ pyrolysis process, which promotes the formation of both methane and benzene. Different from T124MBZ, due to the different positions of the methyl group on the benzene, the important intermediate reactants of T135MBZ in the pyrolysis process are mainly p-xylene and T135MBJ, while o-xylene and p-xylene would appear in the T124MBZ pyrolysis process.

The observed sensitivities in fuel consumption stem from the distinct decomposition pathways favored by each fuel’s molecular structure and the stability of the resulting radicals. For PBZ, the most sensitive reaction is PBZ(+M)=C₆H₅CH₂+C₂H₅(+M), as the weak C–C bond adjacent to the aromatic ring readily cleaves to form a resonance-stabilized benzyl radical and an ethyl radical, both of which are kinetically favorable under combustion conditions. In contrast, T124MBZ consumption is most sensitive to two competing pathways—T124MBZ(+M)=O-XYLCH₂+H(+M) and T124MBZ(+M)=P-XYLCH₂+H(+M). Meanwhile, T135MBZ primarily decomposes via T135MBZ(+M)=T135MBJ+H(+M).

4.3. ROP Analysis

In order to understand the formation pathways of the main products of the three C₉H₁₂ fuels, rate-of-production (ROP) analysis was performed on the three fuels at 1300K. For PBZ, C₂H₄ and C₆H₅CH₃ products were selected for ROP analysis based on the experimental results, and the results are shown in Figure 10. As can be seen from the figure, C₆H₅CH₃ is mainly formed by the reaction of benzyl groups with hydrogen radicals, and this reaction contributes a high amount to the yield of C₆H₅CH₃ in a very short time. Combined with the reaction path analysis provided in Supplementary Materials, it is found that a large number of ethyl radicals and benzyl groups will be generated in the first step of pyrolysis. Then, these radicals undergo a dehydrogenation reaction to produce a large number of hydrogen radicals, improving the yield of C₆H₅CH₃. In addition, the reaction C₆H₅C₃H₅-2+C₆H₅CH₂=C₆H₅C₃H₄+C₆H₅CH₃ has a certain impact on the yield of C₆H₅CH₃ throughout the entire reaction time. For C₂H₄, it is mainly generated by the dehydrogenation of ethyl radicals.

Similarly, CH₄ and C₆H₆ products were selected for ROP analysis for the two trimethylbenzenes, and the results are shown in Figure 11 and Figure 12. As can be seen from Figure 10 and Figure 11, methane is continuously and steadily formed during the reaction time. Most of the first steps of the pyrolysis reaction of T135MBZ and T124MBZ are dehydrogenation and part of it is to remove a methyl radical. The hydrogen and methyl radicals produced by this step can promote the formation of CH₄. For C₆H₆, it is mainly produced by the reaction of toluene with hydrogen radicals, and the yield increases continuously throughout the reaction time. Due to the different main decomposition pathways of PBZ, T124MBZ, and T135MBZ in the first step of pyrolysis, the main products of the three fuels are different.

5. Conclusions

In this study, the pyrolysis product distributions and their concentration dependences on the pyrolysis temperature and pressure of three C₉H₁₂ fuels, namely n-propylbenzene (PBZ), 1,3,5-trimethylbenzene (T135MBZ), and 1,2,4-trimethylbenzene (T124MBZ), were studied by using an SPST experiment and kinetic simulation. The main conclusions are summarized as follows.

(1) PBZ begins pyrolysis at about 1100 K, while T124MBZ and T135MBZ begin pyrolysis at about 1200K, indicating that the reactivity of PBZ is higher than that of T124MBZ and T135MBZ under the study conditions of this experiment.

(2) The kinetic simulation of the pyrolysis process of the three C₉H₁₂ fuels by using a detailed kinetic mechanism are in reasonably good agreement with the experimental results. The pyrolysis products of the three fuels are very similar, but there are some differences. Specifically, PBZ produces a lower amount of the most important soot precursors, i.e., acetylene, than T124MBZ and T135MBZ.

(3) Sensitivity analysis and ROP analysis reveal that the main consumption path of PBZ is to generate benzyl group through a decomposition reaction and it then undergoes secondary reactions, while T135MBZ and T124MBZ are mainly consumed through dehydrogenation reactions.

(4) Generally, as is shown in both this paper and the Supplementary Materials, the contemporary detailed kinetic mechanisms can capture the reactivity trends in the pyrolysis products over the temperature range, and they can also show some quantitative results for some species. However, for many species, especially for species with minor quantities, notable discrepancies between the predicted model and experimentally measured concentrations exist. From the sensitivity and ROP analyses, it can be seen that fuel chemistry related to the aromatics is required via experimental or theoretical chemistry approaches to obtain more reliable reaction rate constants. However, further methods to minimize experimental uncertainty are also required because discrete points still exist from the SPST facility. Thus, reliable mechanism analysis or optimization methods, i.e., curve-matching methods could be a possible solution to resolve this issue.

(5) In summary, the high temperature pyrolysis characteristics of the three C₉H₁₂ fuels were systematically studied via experimental and simulation methods. The results not only build a foundation for a deep understanding of the pyrolysis process of the three C₉H₁₂ fuels, but they also provide reliable basic experimental data for the subsequent development of chemical kinetic mechanisms. However, it should be mentioned that due to the lack of a more accurate analytical instrument, large molecules such as polycyclic aromatic hydrocarbons at low concentrations were not detected and quantitated in this study. However, further mechanism optimization is still needed to better predict the experimental results, which represents an important problem to be solved in future studies.