Reactive molecular dynamics simulation of the pyrolysis and combustion of benzene: ultrahigh temperature and oxygen-induced enhancement of initiation pathways and their effect on carbon black generation

Abstract

The pyrolysis and combustion mechanisms of benzene under different chemical environments and temperatures were investigated by a reactive force field based molecular dynamics (ReaxFF MD) simulation using two systems, pure benzene and a mixture of benzene and oxygen gas. The chemical behaviors of this system were investigated under an ultrahigh temperature that can be induced by a high-energy density laser and compared to those at high temperature. According to some experimental data, we assume that an ultrahigh temperature can be used to mimic laser irradiation. The conclusions of this simulation are as follows. First, the ReaxFF MD simulations showed that the decomposition rates of benzene were significantly accelerated by laser irradiation or in the presence of oxygen. Second, additional initiation pathways were opened up by these two factors. The primary initiation pathway involves only the hydrogen atom loss in the pyrolysis of benzene at 3000 K, and the initiation pathways become much more complicated after laser irradiation or the involvement of oxygen. Third, the ReaxFF MD simulations formed a reasonable carbon black (CB) texture of various sizes in the pyrolysis of benzene, and we also focused on the evolution of the texture of CB. The calculation results of the final gaseous products, hydrocarbons, and the formation of CB are in a good agreement with the literature. This study provides a better understanding of the initiation mechanisms of the pyrolysis and combustion of benzene under extreme conditions.

---

1 Introduction

Combustion and pyrolysis are both ubiquitous and the most significant current interest in power source applications, storage, and engineering science. Therefore, significant efforts have been directed to investigate the evolution, reaction mechanism, and kinetics of the pyrolysis and combustion of hydrocarbons, alternative fuels, biofuels, etc.^1–14 Carbon black (CB) is one of the most common products of the combustion and pyrolysis of hydrocarbons. The past four decades have witnessed a rapid development in the study of CB due to its widespread application and distinct properties. The world production of CB is ∼10 Mt per year.¹⁵ CB can be used as antistatic plastics, electrically conductive agents in plastics^16,17 and is extensively used as a catalyst support in proton exchange membrane fuel cells^18–20 due to its excellent electrical conductivity. It also can be used as a reinforcing compound in rubbers^21,22 as a valuable elastomer filler, because of its low thermal expansion coefficient and excellent heat-, chemical-, and weather resistance.²³ It is also a good candidate for thermal interface pastes, because of its excellent thermal conductivity.^24,25

In general, different manufacturing processes of CB share common procedures, i.e., liquid or gaseous hydrocarbons are decomposed at elevated temperatures under oxygen deficient atmosphere, i.e., pyrolysis or partial combustion.²⁶ Nowadays, CB is mainly produced by the furnace back process, where hydrocarbons are partially combusted and immediately quenched with water.^27,28 The reaction conditions including rates of heating, heat treatment temperature, soak time, and ambient gases affect the pyrolysis process and the products.²⁹ For instance, significant differences have been observed in the chemical composition of diesel engine and exhaust soots, because of their differences in the quenching and post quenching environment of soot.³⁰ The sorts of feedstock can also be of crucial importance for the generation of CB. Therefore, in recent years, significant attention has been focused on the oxidation mechanism of ordinary hydrocarbons including alkenes, alkynes, aromatics, and so on. For instance, the pyrolysis and combustion of n-dodecane was studied by Li, et al.;¹² the mechanism and kinetics for the initial steps of pyrolysis and combustion of 1,6-dicyclopropane-2,4-hexyne was investigated by Liu et al.¹³ using reactive molecular dynamic methods; Li et al.³¹ studied the oxidation of toluene at high temperatures, and the simulations by Qian et al.¹⁴ showed the formation of fullerene during the combustion of benzene.

Benzene is an aromatic hydrocarbon, a natural constituent of crude oil and is one of the most elementary petrochemicals. Nevertheless, its presence in some industrial materials such as gasoline is limited in most countries, because of its severe environmental and health effects. Moreover, most nonindustrial applications have been limited by benzene's carcinogenicity.³² The combustion of benzene produces harmful volatile organic compounds (VOCs), namely, the products of benzene combustion in incomplete hydrocarbon oxidation or pyrolysis include the polycyclic aromatic hydrocarbons and eventually the soot or graphitized CB. To decrease the VOCs formation, a detailed understanding of the reaction pathways of benzene under high temperature conditions is essential. The combustion of benzene at flame temperature or catalytic combustion of benzene has attracted significant attention.^33–38

The pyrolysis of benzene can also occur under the laser irradiation, during which some useful nanomaterials can be obtained. For example, carbon nanoparticles were produced by the irradiation of NIR femtosecond (fs) laser pulses to a benzene/water bilayer.³⁹ C60 and C70 structures were obtained during the pyrolysis of benzene-based chemical compound induced by the laser irradiation, where the C/O ratio affects the ratio of production of C60 and C70.⁴⁰ The increase in the C/O ratio increased the productions of C60 and C70, in particularly of C60.

Although significant studies have been performed on the mechanisms and kinetics of the pyrolysis and combustion of benzene at lower temperatures such as <2500 K,^33–38 the exact nature of the chemical process of the pyrolysis and combustion of benzene at ultrahigh temperatures such as under the loading of high-energy density laser is barely reported. The differences between low temperature and high temperature cases and the type and growth of nanomaterials observed during the pyrolysis at high temperatures have not been studied. All these questions are still not fully understood. However, because of ultrahigh temperature and very short reaction time, the pyrolysis and combustion of hydrocarbon (e.g., benzene) by experiment alone is very challenging. Therefore, the reactive molecular dynamics simulation becomes a feasible method to deal with the issue.

In this study, the initiation mechanisms and chemical evolution of the pyrolysis and combustion of benzene were investigated by the ReaxFF based reactive molecular dynamics simulations (RMD). A previous study on the combustion of benzene using ReaxFF MD,¹⁴ where the H atoms were removed during the combustion as soon as they combined into H₂ gas, showed the formation of some compounds such as fullerene, and our results are in agreement with this study. Yet this study did not focus on the initial decomposition of benzene and the chemical evolutions on the existence of H atoms, which is one of our objectives. The combustion of other hydrocarbon such as n-dodecane¹² is also reported. Their study focused on the reaction pathway for the decomposition of benzene and did not mention the carbon black formation. In this study, the pyrolysis and combustion behaviors under ultrahigh temperature (7000 K) which can be induced by high-energy density laser irradiation were compared to those produced under normal high-temperature loading (3000 K). Laser-heated materials such as carbon-based compound can be turned into plasma, and the temperatures can reach to 10 thousands of K.^41,42 Ultrahigh temperature was used to mimic a circumstance under laser irradiation. First, from the ReaxFF MD simulations, the initiation mechanisms of the pyrolysis of benzene were found to be dependent on temperature and chemical environment. Second, the ReaxFF MD simulations led to a reasonable CB texture of various sizes formed in the pyrolysis of benzene, focusing on the CB texture evolution. Third, O₂ was added to provide a benzene/O₂ ratio of 2/3 to investigate the effect of added oxygen. The dissociation rates of benzene in the pure benzene system and benzene in the presence of O₂ system were also analyzed and compared.

The simulation results were compared to the previous experimental literature observations.^1,33–38,43 To some extent, an agreement between the simulation and experimental data was achieved.

2 Methodologies

2.1 Models

Both the pyrolysis of benzene and formation of carbon black have been modeled in a condensed-phase environment at a density corresponding to the density of liquid benzene at room temperature, i.e., 0.879 g cm⁻³. A cell of benzene composed of 100 benzene molecules (1200 atoms) was used to study the pyrolysis of pure benzene. Another cell composed of 100 benzene molecules and 150 O₂ molecules (1500 atoms) was used to study the partial combustion of benzene. The scale effect on the simulation result was investigated. For instance, a calculation was performed on a cell of benzene using a system of 3600 atoms for 1000 ps, and the result in our simulation time was found to be similar to that of a smaller system of 1200 atoms.

2.2 Calculations

A newly developed molecular reactive force field ReaxFF was used to simulate the pyrolysis and combustion of benzene. The validity of ReaxFF to benzene and O₂ was confirmed by the previous studies.^{13,14,31,44–46} After the relaxation, the two cells were subjected to different high-temperature simulations under canonical particle number, volume, and temperature (NVT) conditions with the multiple Nosé–Hoover thermostat method.⁴⁷ A total of four simulations were performed. The information on the temperatures and time of the NVT-MD molecular dynamic simulations is listed in Table 1. The value of 3000 K was used to simulate the initial or subsequent high-temperature pyrolysis or combustion; 7000 K temperature was used to simulate the ultrahigh temperature effect induced by laser irradiation. In Cases 1 and 2, the temperature of 7000 K was only present for 20 ps and after that the temperature was decreased to 3000 K.

Table 1 The information of the temperatures and time of NVT simulations

No. of case	Starting temperature and time/K	The cell	Continuous temperature/K	To simulate
1	7000 (for 20 ps)	Benzene	3000 (for 6 ns)	Laser induced pyrolysis
2		Benzene + O2	3000 (for 2 ns)	Laser induced combustion
3	3000 (for 20 ps)	Benzene	3000 (for 2 ns)	Pyrolysis
4		Benzene + O2		Combustion

---

The ReaxFF is a DFT based reactive force field. The cleavage and formation of chemical bonds can be accurately described during reactions.⁴⁸ In this study, the LAMMPS software package was used^49,50 to perform all the simulations.

2.3 Dynamic trajectory analyses

For each simulation, a time step of 0.1 fs was assigned, and the dynamic trajectories including atomic positions and velocities were recorded every 1 ps and analyzed by a FORTIAN scripts based on the ideas of Strachan et al.⁵¹ to provide information on the chemical reactions. The bond-order cut off values for each pair of elements in ReaxFF can be checked elsewhere.^13,14,31

3 Results and discussion

3.1 Initial steps of the decomposition for oxidation of benzene

In this study, the results of the initial products and the products produced on a longer time scale are presented. The initial products are produced from the earliest decomposition of benzene molecules. The initial products generated very early and are commonly very unstable. They react in a very short time. The products produced on a longer time scale are the results of the evolutions of the initial products. These products are more stable than the initial products and their amounts either kept increasing or were stable during a reasonably long time.

The detailed reaction mechanisms extracted from the dynamic trajectories are presented here. The initial products of the four cases are shown in Table 2. For Case 1, C₆H₅, and H were observed within 1 ps; C₂H₂ and C₄H₄ were observed within 2 ps. Therefore, the pyrolysis pathway of benzene observed in the ReaxFF NVT-MD simulation at T = 7000 K can be represented by Scheme 1 (Table 3), including the hydrogen atom loss and cycloreversion. Case 2 have the same products as C₆H₅, H, C₄H₄, and C₂H₂ on the same steps of reaction as that of Case 1. Besides, the formation of C₆H₆O₂ was observed in 1 ps. C₃H₄O₂, C₃H, HO₂, C₆H₇, and O were observed within 2 ps. Mechanisms of the initial pyrolysis and combustion was complicated by adding oxygen. Therefore, the pathway of benzene and O₂ system observed in the ReaxFF NVT-MD simulation at T = 7000 K can be shown by Scheme 2 including hydrogen atom loss and cycloreversion and from the middle of the ring after adding O₂. For Case 3, the initial products include C₆H₅ and H, which are much simpler. Lower temperature and pure chemical components lead to a simple mechanism of the initial pyrolysis simply involving hydrogen atom loss, as shown in Scheme 3. The products of Case 4 include C₆H₆O₂, C₆H₅, H, and C₂H₂. The addition of oxygen affects the mechanisms of the initial pyrolysis by adhering to the molecule of benzene. Therefore, the pathway of Case 4 can be shown by Scheme 4, including hydrogen atom loss and cycloreversion of the ring into three parts evenly.

Table 2 The initial products and the moments they generated of four NVT simulations

Case 1		Case 2		Case 3		Case 4
Products	Time, ps	Products	Time, ps	Products	Time, ps	Products	Time, ps
C6H5	1.00	C6H5	1.00	C6H5	25.00	C6H6O2	1.00
H	1.00	H	1.00	H	25.00	C6H5	7.00
C2H2	2.00	C6H6O2	1.00			H	7.00
C4H4	2.00	C2H2	2.00			C2H2	12.00
		C4H4	2.00
		C3H4O2	2.00
		C3H	2.00
		HO2	2.00
		C6H7	2.00
		O	2.00

---

Table 3 Initial reaction observed in the pyrolysis or combustion. Times observed corresponding to the frequency of the mechanism observed

---

From the initial products of these cases, at higher temperatures, the mechanism is more complicated. In Case 3, at low temperature and pure chemical component, only one pathway involving the hydrogen atom loss exists for the decomposition of benzene. In Case 1, of the same chemical component, at a much higher temperature, decomposition follows an alternative pathway, i.e., cycloreversion. The comparison of Cases 2 and 4 also support this statement.

Another conclusion is that the mechanism of the initial pyrolysis becomes more complicated by adding oxygen. Cases 1 (two pyrolysis pathways) and 2 (three pyrolysis pathways) are both at 7000 K; however, the initial steps of Case 2 are more complicated, because of the presence of oxygen. The comparison of Cases 3 (one pyrolysis pathway) and 4 (two pyrolysis pathways) also support this statement.

Benzene decomposition pathways at lower temperatures in previous studies^1,33–38 are mainly hydrogen atom loss, forming less atoms rings such as five-membered carbon rings⁴³ and then the decomposition of the rings, as well as a certain branching ratio of benzene cycloreversion. However, the cycloreversions need much longer induction time and proceed in many steps in longer time. At higher temperatures, the cycloreversions become more important and less atoms rings are fewer. The cycloreversions of benzene at higher temperature need much shorter induction time and proceed in fewer steps, i.e., benzene breaks into three C₂H₂ in 1 ps, which in our time resolution for one step.

C₆H₆ → C₆H₅ + H reaction needs to overcome the energy barrier of a single bond, a C–H bond with bond dissociation energy (BDE) of 448.4 kJ mol.⁵² C₆H₆ → C₄H₄ + C₂H₂ reaction needs to overcome the energy barrier of a sp²–sp² C–C bond in benzene with a bond length of 140 pm. The BDE can be calculated by the method of Andreas,⁵³ and was found to be 614.5 kJ mol⁻¹. The later needs higher energy to break than the former. Therefore, at lower temperature C₆H₆ → C₆H₅ + H reaction dominates unimolecular decomposition pathway of benzene. The competing channel to C₄H₄ + C₂H₂ becomes more dominating as the temperature increases.

3.2 Temperature and chemical environment effect on the decomposition rates of benzene

Decay evolutions of benzene in the four cases are shown in Fig. 1. The decompositions of benzene in Cases 1 and 2 basically completed in 30 ps, whereas those in Cases 3 and 4 completed in 2000 ps. The decomposition rates under 7000 K were obviously much higher than those under 3000 K. The decomposition rates of Cases 1 and 2 were essentially the same, and the decomposition rates of Case 4 are more than that of Case 3, indicating that the oxygen gas accelerated the reaction of benzene at lower temperature. For a much higher temperature, the acceleration of oxygen was not obvious because of an originally very high reactivity.

Fig. 1 Decay evolutions of benzene in four simulations.

Both Cases 3 and 4 shows obvious induction time for the decomposition of C₆H₆. The induction time of Case 4 (15 ps) is shorter than that of Case 3 (350 ps), which should be the result of the enhancement of the reaction activity because of the presence of O₂.

The induction time arise from the overcoming of the reaction barrier for the benzene decomposition. The internal energy initially evenly distributed amongst the various degrees of freedom needs to be redistributed so that there is enough energy in the mode that reflects decomposition so that the barrier can be overcome (e.g., a C–H stretch mode for the C₆H₆ → C₆H₅ + H reaction). The induction times in Cases 1 and 2 are much shorter, because much higher temperatures are applied to these systems.

3.3 Evolutions of further chemical reactions and the effect of O₂ on the pyrolysis/combustion of benzene

The evolutions of main chemical products in the early stages of the decomposition under aforementioned temperatures and chemical atmosphere were analyzed and shown in Fig. 2–5.

Fig. 2 (neat benzene, 7000 K for 20 ps and then 3000 K for 6 ns) shows that C₂H₂ (blue line), C₂H₄ (pink line), CH₄ (green line), and H₂ (red line) are created rapidly at the beginning of early 20 ps simulation. At ∼50 ps, the rapid increase in the H₂ stops, and a sharp decrease is observed. For other three products, C₂H₂, C₂H₄, and CH₄, a plateau followed after a slight decrease.

Fig. 2 The evolution of products in the decomposition of benzene during the NVT-MD simulation at T = 3000 K after 20 ps stimulation at T = 7000 K. The initial number of C₆H₆ molecules is 100. At least 50 hydrocarbons (from C1 to C38) were observed, and some of which are in small ratios and not listed in the figures to avoid crowded lines.

Fig. 3 (benzene in O₂ atmosphere, 7000 K for 20 ps and then 3000 K for 2.3 ns) shows that CO (pink line), H₂O (dark cyan line), C₂H₂ (red line), C₂H₄ (yellow line), and H₂ (dark red line) are created rapidly at the beginning of early 20 ps simulation. At ∼100 ps, the rapid increase in H₂ stops, and a sharp decrease is observed. For other four products, CO, H₂O, C₂H₂, and C₂H₄, a plateau is observed after a slight increase. The comparison of Fig. 3 and 4 shows that under the same temperature histories, both C₂H₂ and C₂H₄ are the main products. The presence of O₂ induces the main products of CO and H₂O and also decreases the average size of products.

Fig. 3 Products evolution of benzene in O₂ atmosphere during the NVT-MD simulation at T = 3000 K after 20 ps stimulation at T = 7000 K. The initial number of C₆H₆ molecules is 100. At least 10 hydrocarbons (from C1 to C7) were observed, and some of which are in small ratios and not listed in the figures to avoid crowded lines.

Fig. 4 (neat benzene, 3000 K for 2 ns) shows that the decomposition starts at ∼350 ps and then gradually decreases. C₂H₂ (blue line) is the first and major product, generated in ∼350 ps and reached its largest amount in ∼1000 ps. After that the amount of C₂H₂ was kept constant. C₂H₄ (pink line) and C₂H₃ (dark cyan line) are created at ∼500 ps and increased gradually to high amounts.

Fig. 4 The evolution of products in the decomposition of benzene during the NVT-MD simulation at T = 3000 K. The initial number of C₆H₆ molecules is 100. At least 25 hydrocarbons (from C1 to C17) were observed, and some of which are in small ratios and not listed in the figures to avoid crowded lines.

Fig. 5 shows the decomposition of benzene in the presence of O₂ atmosphere at 3000 K for 2 ns, indicating that the decomposition of benzene and O₂ started very early and then gradually decreased. The levels of O₂ and benzene decreased to very low at ∼500 and 1500 ps, respectively. CO (pink line), H₂O (dark cyan line), CO₂ (yellow line), C₂H₂ (dark pink line, the lower one) are formed rapidly at the beginning of the simulation. After ∼500 ps, plateaus followed after the increase. The products of Case 4 are very similar to those of Case 2. The presence of O₂ also induces the formation CO and H₂O as the main products and also decreased the average size of the products, indicating that the effect of the chemical atmosphere is very significant, probably larger than the effect of the temperature.

Fig. 5 The evolution of products in the decomposition of benzene in the presence of O₂ atmosphere during the NVT-MD simulation at T = 3000 K. The initial number of C₆H₆ molecules is 100. At least 20 hydrocarbons (from C1 to C15) were observed, and some of which were in small ratios and not listed in the figures to avoid crowded lines.

In summary, the analysis of the initial decomposition reactions and the products of benzene shows the dependence on temperature and chemical atmosphere: (1) higher temperatures (7000 K simulating laser ablation) resulted in much higher decomposition rate. In both Cases 1 and 2, benzenes decomposed in 20 ps, and further reactions quickly reached to equilibrium. (2) The presence of O₂ significantly affects the final products.

In the experimental study of the benzene oxidation by Chai et al.,⁴³ acetylene was a major hydrocarbon intermediate, and their result is consistent with our result of all the four cases. For the oxidation of benzene at low pressure (45 mbar) by Detilleux,⁵⁴ fifteen chemical species were detected in benzene flames, including final gaseous products (O₂, CO₂, H₂O, CO, and H₂) and ten hydrocarbons (ranging from C1 to C6). Yang et al.⁵⁵ studied the oxidation of the premixed benzene/oxygen/argon flame with tunable synchrotron photoionization and found that the hydrocarbons include CH₃, C₂H, C₂H₃, C₂H₅, C₃H, C₃H₃, C₃H₅, C₄H, C₄H₃, C₄H₅, C₄H₇, C₅H₃, C₅H₅, C₅H₇, C₆H₅, C₆H₅O, C₇H₇, and C₉H₇ in the flame. CO, CO₂, H₂, and H₂O were also observed in the final gaseous products. Moreover, the isomers of some PAHs were also identified. Most of these free radicals were obtained in our simulations. Some PAHs or CB were also identified in our results. In summary, our calculation results of the final gaseous products and hydrocarbons are in a good agreement with the literature.

3.4 CB texture evolution during the pyrolysis or combustion

The formation of CB was observed during the pyrolysis or combustion of benzene and benzene in the presence of O₂. The CB texture evolutions in Cases 1–4 are shown in Fig. 6–9. The CB texture evolutions during the pyrolysis of pure benzene and benzene in the presence of O₂ after laser irradiation at 7000 K are shown in Fig. 6 and 7, respectively, indicating that in both the cases, benzene gradually decomposed, and then the carbon atoms gathered together to form CB. The differences between the two cases are the size of CB and the time required for its formation. The CB formed in the pure benzene is larger than that formed in benzene in the presence of O₂. Moreover, the generation of the former needs shorter time. After careful investigation, the CBs were found to mainly consist of six-membered ring systems. The carbonaceous enrichment occurs after the carbon atoms escape from the benzene molecules. Nucleus appears when the unstable network becomes aromatic. Moreover, the carbon atoms adjust their positions and approximate forms the six-membered ring systems. During this process, the CB structures of the graphite-like lamellar are built and fly over in the simulation box, absorbing more escaped carbon atoms and a few hydrogen atoms, resulting in further growth.

Fig. 6 Snapshots of benzene during the NVT-MD simulation at T = 3000 K after 20 ps stimulation at T = 7000 K.

Fig. 7 Snapshots of the decomposition of benzene in O₂ atmosphere during the NVT-MD simulation at T = 3000 K after 20 ps stimulation at T = 7000 K.

Fig. 7 (7000 K in the presence of O₂) shows that at 500 ps a polycyclic aromatic hydrocarbon forms (lower right), which then reacts later. It is a critical step in mechanism for rapid growth of aromatic structures. In absence of O₂, the polycyclic aromatic hydrocarbon formed will eventually grow into graphitized CB, similar to that shown in Fig. 6 (7000 K in the absence of O₂). In the presence of O₂, O₂ attacks the polycyclic aromatic hydrocarbons once it forms, and therefore formation of big graphitized CB is more difficult.

The CB texture evolutions during the pyrolysis of the pure benzene and benzene in the presence of O₂ at 3000 K are shown in Fig. 8 and 9, respectively. The pictures also show that in both the cases, the benzene molecules gradually decomposed, and then the carbon atoms gathered together to form CB. The reaction at 3000 K is much slower than that at 7000 K, and the formation of CB needs more time.

Fig. 8 Snapshots of benzene during the NVT-MD simulation at T = 3000 K.

Fig. 9 Snapshots of the decomposition of benzene in the presence of O₂ atmosphere during the NVT-MD simulation at T = 3000 K.

The difference between Cases 3 and 4 are also the size of the CB and the time required for its formation. The size of the CB formed in the pure benzene is larger than that formed in benzene in the presence of O₂ as well as at 3000 K. Moreover, the generation of the former needs shorter time.

The CB growth histories can be represented by the maximal molecular weight (MW) of CBs. The evolution of MW of the CBs is shown in Fig. 10. Under laser irradiation, the MWs of the CBs formed are ∼3443 (black line: 1803 ps) and 1018 (red line: 1689 ps) corresponding to neat benzene and a mixture of benzene and O₂. In the neat benzene and the mixture, the largest CB were observed at 1803 ps and 1689 ps. However, as early as 101 ps, big CB with a MW of 2839 was generated in the neat benzene. Higher C/O ratio leads to earlier large CB formation under laser irradiation. The largest CBs are much bigger at higher C/O ratio.

Fig. 10 Evolution of maximal MW of CBs.

Under a high temperature of 3000 K, the CBs with MWs of ∼1740 (black line 1992 ps) and 856 (red line 1557 ps) were formed corresponding to the neat benzene and a mixture of benzene and O₂. In the neat benzene and mixture, the largest CB was observed at 1992 ps and 1557 ps, respectively. However, as early as 1269 ps, big CB with a MW of 1051 was generated in the neat benzene. Higher C/O ratios led to earlier large CBs formation under high temperatures. The CBs were much bigger at higher C/O ratios.

The summary of the analysis of two points is summarized. First, for a same system, laser irradiation is helpful to produce more or larger CBs in a shorter time, because the laser irradiation results in much higher temperature for complete decomposition of benzene to generate a significant amount of dissociative carbon atoms, supplying sufficient material for the formation of CBs. Moreover, based on this, the laser irradiation of hydrocarbon fuel is considered as a feasible approach to produce some useful nanomaterials such as diamond, fullerene, graphite and nanotube, and all of which are substantially CBs. Second, no matter under laser irradiation or high temperature, higher C/O ratio is beneficial for the generation of more and larger CBs. Higher C/O ratio indicates fewer O₂, and the presence of O₂ will combine more dissociative carbon atoms to form CO₂ and CO, inhibiting the formation of CBs. As a result higher C/O ratio is beneficial for the generation of more and larger CBs.

In all these simulations, the formation of carbon cages or tubes was not observed, probably because of shorter simulating time. When the CB structure of the graphite-like lamellar is built, cages or tubes might be created if the lamellar wraps itself in a long run of simulation. This problem needs to be addressed in the near future.

4 Conclusions

In this study, we investigated the pyrolysis behaviors of benzene in the presence or absence of oxygen, by laser irradiation or simply by high temperature. The pyrolysis and combustion mechanisms of benzene under different chemical environment and temperatures were investigated by ReaxFF MD. Two systems were studied, pure benzene and a mixture of benzene and oxygen gas. The chemical behaviors induced by high-energy density laser compared to those at high temperature were investigated. Ultrahigh temperature was used to mimic laser irradiation. First, the ReaxFF MD simulations showed that the decomposition rates of benzene were significantly accelerated by laser irradiation or in the presence of oxygen. Second, the initiation pathways were also increased by these two factors. The primary initiation pathway only involved a hydrogen atom loss in benzene pyrolysis at 3000 K, and it became more complicated after the laser irradiation or in the presence of oxygen including hydrogen atom loss, cycloreversion of the ring and from the middle of the ring after reacting with an O₂ molecule. Third, the ReaxFF MD simulations formed reasonable carbon black (CB) texture of various sizes in the pyrolysis of benzene, and we also focused on the evolution of the texture of CB. The decay evolution under the aforementioned temperatures shows that both the temperature and C/O ratio mainly affected the decomposition rate of benzene. The formation of CB is very common in the partial combustion of benzene. More CBs were created in the presence of less amount of oxygen.

The calculation results of the final gaseous products, hydrocarbons, and the formation of CB are in a good agreement with the literature. This study provides a better understanding of the initiation mechanisms of the pyrolysis and combustion of benzene under extreme conditions and also obtains a direct observation of the generation of CBs at aromatic level.

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

We greatly appreciate the financial support from the Defense Industrial Technology Development Program (B1520132004), the Science and Technology Innovation Fund of Institute of Chemical Materials, CAEP (no. kjcx-201305), and the National Natural Science Foundations of China (11372291 and 11372292).

References

1. J. M. Simmie, Prog. Energy Combust. Sci., 2003, 29, 599–634.
2. J. A. Miller, R. J. Kee and C. K. Westbrook, Annu. Rev. Phys. Chem., 1990, 41, 345–387.
3. H. Richter and J. B. Howar, Prog. Energy Combust. Sci., 2000, 26, 565–608.
4. J. A. Miller, M. J. Pilling and J. Troe, Proc. Combust. Inst., 2005, 30, 43–88.
5. Z. Qin, V. V. Lissianski, H. Yang, W. C. Gardiner, S. G. Davis and H. Wang, Proc. Combust. Inst., 2000, 28, 1663–1669.
6. G. Black, H. J. Curran, S. Pichon, J. M. Simmie and V. Zhukov, Combust. Flame, 2010, 157, 363–373.
7. C. K. Westbrook, W. J. Pitz, O. Herbinet, H. J. Curran and E. J. Silke, Combust. Flame, 2009, 156, 181–199.
8. H. Wang and M. Frenklac, Combust. Flame, 1997, 110, 173–221.
9. P. Dagaut, Phys. Chem. Chem. Phys., 2002, 4, 2079–2094.
10. F. Battin-Leclerc, Prog. Energy Combust. Sci., 2008, 34, 440–498.
11. S. Pierucci and E. Ranzi, Comput. Chem. Eng., 2008, 32, 805–826.
12. Q. D. Wang, J. B. Wang, J. Q. Li, N. X. Tan and X. Y. Li, Combust. Flame, 2011, 158, 217–226.
13. L. Liu, C. Bai, H. Sun and W. A. Goddard III, J. Phys. Chem. A, 2011, 115, 4941–4950.
14. H. J. Qian, A. C. T. Van Duin, K. Morokuma and S. Irle, J. Chem. Theory Comput., 2011, 7, 2040–2048.
15. R. Adams, Focus Pigm., 2007, 3, 1–2.
16. F. J. B. Calleja, R. K. Bayer and T. A. Ezquerra, J. Mater. Sci., 1988, 23(4), 1411–1415.
17. I. Balberg, Phys. Rev. Lett., 1987, 59(12), 1305–1308.
18. T. Soboleva, X. Zhao, K. Malek, Z. Xie, T. Navessin and S. Holdcroft, ACS Appl. Mater. Interfaces, 2010, 2(2), 375–384.
19. B. Z. Fang, M. S. Kim, J. H. Kim and J. S. Yu, J. Am. Chem. Soc., 2009, 131, 15330–15338.
20. A. Guha, W. J. Lu, T. A. Zawodzinski and D. A. Schiraldi, Carbon, 2007, 45(7), 1506–1517.
21. Z. Rigbi, Reinforcement of rubber by carbon black, Springer, Berlin, Heidelberg, 1980, pp. 21–68.
22. D. J. Kohls and G. Beaucage, Curr. Opin. Solid State Mater. Sci., 2002, 6(3), 183–194.
23. D. Piasta, S. Spange and F. Simon, Langmuir, 2009, 25, 9071–9077.
24. C. K. Leong and D. D. L. Chung, Carbon, 2006, 44, 435–440.
25. Q. Li, L. Dong, Y. Liu, H. A. Xie and C. X. Xiong, Carbon, 2011, 49, 1033–1051.
26. J. B. Donnet, R. C. Bansal and M. J. Wang, Carbon black: science and technology, Marcel Dekker Inc., New York, 2nd edn, 1993.
27. F. C. Lockwood and J. E. Van Niekerk, Combust. Flame, 1995, 103, 76–90.
28. L. Rothbuhr and J. Witte, Process for the production of furnace black, US Pat. 4292291, 1979.
29. S. Ban, K. Malek, C. Huang and Z. Liu, Carbon, 2011, 49, 3362–3370.
30. A. D. H. Clague, J. B. Donnet, T. K. Wang and J. C. M. Peng, Carbon, 1999, 37, 1553–1565.
31. M. C. Xue, D. Q. Wei, J. Q. Li, J. B. Wang and X. Y. Li, J. Phys. Chem. A, 2012, 116(40), 9811–9818.
32. http://en.wikipedia.org/wiki/Benzene, 2014 access.
33. A. Tregrossi, A. Ciajolo and R. Barbella, Combust. Flame, 1999, 117(3), 553–561.
34. G. Vourliotakis, G. Skevis and M. A. Founti, Energy Fuels, 2011, 25(5), 1950–1963.
35. R. Addink and E. R. Altwicker, Environ. Sci. Technol., 2004, 38(19), 5196–5200.
36. T. W. Richards and S. Tamaru, J. Am. Chem. Soc., 1920, 42(8), 1599–1617.
37. L. S. Caretto and K. Nobe, Ind. Eng. Chem. Proc. Des. Dev., 1966, 5(3), 217–222.
38. S. F. Zuo, X. Sun, N. N. Lv and C. Z. Qi, ACS Appl. Mater. Interfaces, 2014, 6(15), 11988–11996.
39. T. Yatsuhashi, N. Uchida and K. Nishikawa, Chem. Lett., 2012, 41(7), 722–724.
40. A. Tregrossi, A. Ciajolo and R. Barbella, Combust. Flame, 1999, 117, 553–561.
41. C. D. David and H. Weichel, J. Appl. Phys., 1969, 40, 3674–3679.
42. F. J. Allen, J. Appl. Phys., 1971, 42, 3145–3149.
43. Y. Chai and L. D. Pfefferle, Fuel, 1998, 77, 313–320.
44. T. Cheng, A. Jaramillo-Botero, W. A. Goddard III and H. Sun, J. Am. Chem. Soc., 2014, 136(38), 13467.
45. C. Kimberly, A. C. T. Van Duin and W. A. Goddard III, J. Phys. Chem. A, 2008, 112(5), 1040–1053.
46. S. Agrawalla and A. C. T. Van Duin, J. Phys. Chem. A, 2011, 115, 960–972.
47. W. Hoover, Phys. Rev. A: At., Mol., Opt. Phys., 1985, 31, 1695–1697.
48. F. Castro-Marcanoa, A. M. Kamatb, M. F. Russo Jr, A. C. T. Van Duinb and J. P. Mathewsa, Combust. Flame, 2012, 159, 1272–1285.
49. S. J. Plimpton, J. Comput. Phys., 1995, 117, 1.
50. H. M. F. Aktulga, J. C. Fogarty, S. A. Pandit and A. Y. Grama, Parallel Comput., 2012, 38, 245.
51. A. Strachan, A. C. T. Van Duin, D. Chakraborty, S. Dasgupta and W. A. Goddard III, Phys. Rev. Lett., 2003, 91, 098301.
52. L. A. Gribov, I. A. Novakov, A. I. Pavlyuchko and O. Yu. Shumovskii, J. Struct. Chem., 2007, 48, 400–406.
53. A. A. Zavitsas, J. Phys. Chem. A, 2003, 107, 897–898.
54. V. Detilleux and J. Vandooren, Combust., Explos. Shock Waves, 2009, 45, 392–403.
55. B. Yang, Y. Y. Li, L. X. Wei, C. Q. Huang, J. Wang, Z. Y. Tian, R. Yang, L. S. Sheng, Y. W. Zhang and F. Qi, Proc. Combust. Inst., 2007, 31, 555–563.

---