Oxidation reaction and thermal stability of 1,3-butadiene under oxygen and initiator

2.1 Materials

1,3-Butadiene (mass purity > 99.90 %, molecular weight 54.09 g·mol⁻¹) was obtained from Guangdong Walter Gas Co., ltd., China. 2,2′-Azobisisobutyronitrile (AIBN) (98 % Shanghai Macklin Biochemical Co., ltd, China), cumyl hydroperoxide (CHP) (≥80 %, Aladdin Industrial Corporation, China), N₂, O₂ (99.99 %, Nanning Zhong Yi Chuang Gas Co., ltd, China), KI (99.50 %, Aladdin Industrial Corporation, China), and Na₂S₂O₃ (99.95 %, Aladdin Industrial Corporation, China) were used in the experiments.

2.2 Thermal oxidation of 1,3-butadiene by MCPVT

Based on the pressure vessel test (PVT, an experimental method used by the United Nations Specialized Committee on Dangerous Goods to assess the intensity of decomposition of chemical hazards) (Okada et al., 2014), a mini closed pressure vessel test (MCPVT, vessel volume: 25 mL) system was designed to measure the change of pressure (P) and temperature (T) during the decomposition of chemicals hazards. The experimental device comprises four parts: a mini closed pressure container, a temperature control system, a pressure/temperature sensor, and a pressure/temperature recorder, as shown in Fig. 1. The thermal oxidation reaction of 1,3-butadiene was evaluated by monitoring the T and P behaviors of the container. The amount of substance was calculated with the ideal gas equation (PV = nRT) to monitor the decomposition reaction. An internal glass bottle (3 mL) was settled into the stainless-steel vessel as a sampling container. The sample mass of 1,3-butadiene is 0.33 g, and the initial pressure of oxygen delivered from the gas cylinder to the container through the intake valve was set to 0.6 MPa. The autoxidation reaction in pure 1,3-butadiene without an initiator was performed at first. Then, about a 2 % mass fraction of the initiators, CHP and AIBN, were used to initiate the oxidation of 1,3-butadiene for the other two types of oxidation models. Isothermal experiments were conducted at 323–363 K for 8 h in an oxygen atmosphere. When the reactor was heated, the inner temperature and pressure were changed and recorded by the recorder and signal processor. After the reaction, the reactor was cooled to room temperature in a cold-water bath. The products were collected and analyzed.

Close

Fig. 1

The reaction device of 1,3-butadiene oxidation.

Export to PPT

2.3 Analysis of 1,3-butadiene peroxide by iodimetry

Peroxide was both impact-sensitive and thermally unstable. It is of paramount importance to determine the peroxide content of 1,3-butadiene by iodimetry at different temperatures. Several oxidation products were dissolved in potassium iodide-starch solution. The generated peroxides oxidized the potassium iodide to an equivalent amount of iodine (Eq. (1)). The iodine molecules were linked with starch to form a complex, which turned the solution blue. Then, the standardized sodium thiosulfate solution was used to titrate the blue solution for the end of the color disappeared (Eq. (2)). The results were quantified as milligrams per kilogram (mmol·kg⁻¹) of peroxide.

2.4 Differential scanning calorimetry (DSC)

Dynamic scanning experiments were performed on a Mettler TA Q2000 system. It was used for performing experiments for withstanding relatively high pressure to approximately 10.0 MPa. STAR^e software was applied to obtain thermal curves for further analysis (Mettler, 2004; You et al., 2010; Liu et al., 2011). Thermal decomposition characteristics of oxidation products of 1,3-butadiene were analyzed by differential scanning calorimetry (DSC). To measure the total heat, heat being evaluated during the complete cure, dynamic DSC analysis was performed from 303 to 573 K at a heating rate of 10 K⋅min⁻¹, and the flow of dynamic N₂ was 35.0 mL·min⁻¹. The detection sensitivity is 0.04 μW. Different masses of the three oxidation products were placed into the testing crucible. The test cell was sealed manually by a special tool equipped with Mettler’s DSC.

2.5 GC-MS and HS-GC-MS analysis

The oxidation processes of 1,3-butadiene produced solid, liquid, and gaseous products. The main gas and liquid products were qualitatively analyzed by gas chromatography-mass spectrometry (GC–MS, GC/MS-QP2010 Ultra, SHIMADZU, Japan), which was performed on a GC-2010 plus gas chromatography coupled with an electron impact ionization detector (EID, 70 eV) and a Rxi-5SilMS fused silica capillary column (30 m × 0.25 mm × 0.25 μm, J&W Scientific Inc.). A sample autoinjector was employed for all liquid samples, and the gas sample was injected manually. The analytical procedures were as follows: the temperature was maintained at 333 K for 1 min, then increased to 373 K at a rate of 3 K·min⁻¹, and kept for 3 min. The carrier gas was ultrahigh pure helium at a constant flow rate of 4.0 mL·min⁻¹. The injection temperature and volume were maintained at 373 K and 1.0 μL; the split ratio was 30:1, while the interface and ion source temperatures were set at 473 and 493 K, respectively. A quadrupole mass filter was used with an m/z range of 18–300 in full-scan mode. The volatile components in the solid products were analyzed by headspace sampler-gas chromatography-mass spectrometry (HS-GC–MS) coupled with an Agilent Technologies G1888 Network Headspace Sampler. The zone temperatures were as follows: oven temperature 393 K, loop temperature 393 K, transmission line temperature 403 K. The event times were set as follows: GC cycle time 10 min, vial equilibrating time 15 min, loop fill time 0.2 min, inject time 1 min, pressurized time 0.2 min, and equilibrating loop time 0.05 min. Qualitative analysis of the products was based on the cracking patterns and retention times observed in the mass spectrometry and gas chromatography analyses, respectively.

3.1 Pressure behavior of 1,3-butadiene oxidation in MCPVT

The oxidation of 1,3-butadiene with oxygen has been the subject of numerous publications to date, with the first report by Robey in 1944 (Robey et al., 1944). Because of its particular conjugated structure, 1,3-butadiene is readily oxidized by oxygen to generate peroxide or polyperoxide, ending in a —O—O— bond. The peroxide is unstable and decomposes into free radicals that have a negative effect on the polymerization process. Therefore, it is vital to understand the oxidation behavior concerning oxygen and initiators in the synthesis process of styrene-butadiene rubber because their use is promoted based on their polymerization properties. The oxidation process was conducted using a self-designed MCPVT device under isothermal conditions. The MCPVT experiment was carried out under an oxygen atmosphere for three oxidation conditions: autoxidation, CHP-initiated, and AIBN-initiated. The pressure and inner temperature were monitored when the vessel was heated at 343 K for 8 h.

3.1.1

3.1.1 Autoxidation of 1,3-butadiene initiated by oxygen

The results of the autoxidation process and the contrast experiment in a nitrogen atmosphere are shown in Fig. 2. The temperature vs time (T-t) and pressure vs time (P-t) curves in Fig. 2(b) have no noticeable change, while the curves of Fig. 2(a) exhibit a significant decline. The P-t curve gradually declines from 0.9990 MPa to 0.9510 MPa (Table 1). It was assumed that 1,3-butadiene and oxygen are ideal gases and the two substances react in equal amounts during the reaction. Then, the amount of 1,3-butadiene remaining after the reaction was equal to the amount of oxygen, and the oxygen consumption was half of the total pressure (Table 1). The change of pressure with temperature was consistent with the ideal gas equation. Since V and R are constant values, the relationship of the amount of the total gas with time was plotted as shown in Fig. 3. Curve 1 is an approximately straight line; it indicates that a small amount of obvious chemical reactions occurred in the nitrogen atmosphere. Moreover, curve 2 for an oxygen atmosphere indicates that the reaction processes of 1,3-butadiene have occurred, with a low 1,3-butadiene conversion rate of 3.94 %.

Close

Fig. 2

The P-t and T-t curves of 1,3-butadiene: (a) in an oxygen atmosphere at 343 K; (b) in a nitrogen atmosphere at 383 K.

Export to PPT

Table 1 The oxygen consumption of 1,3-butadiene oxidation.

	Autoxidation	CHP-initiated oxidation	AIBN-initiated oxidation
Initial Pressure (MPa)	0.9987	1.0445	0.9822
Final Pressure (MPa)	0.9510	0.7838	0.8824
ΔPressure (MPa)	0.0477	0.2607	0.0998
Oxygen consumption (MPa)	0.02385	0.1304	0.0499
Oxygen consumption (mmol)*	0.2406	1.3156	0.5035

* Calculated according to the ideal gas equation.

Close

Fig. 3

The n-t curves of 1,3-butadiene: (1) in a nitrogen atmosphere; (2) in an oxygen atmosphere.

Export to PPT

3.1.2

3.1.2 Organic peroxide CHP initiated oxidation

Organic peroxide cumene hydroperoxide (CHP) has been broadly employed in the chemical industry. It is predominantly used in producing phenol and acetone by catalytic cleavage and as an initiator in the acrylonitrile–butadiene–styrene (ABS) copolymer polymerization process (Chen et al., 2008; Hsu et al., 2012; Wu et al., 2012a). CHP is easily recognized as a typical thermally unstable and hazardous substance. Therefore, this study first simulated the thermal hazard characteristics of CHP via MCPVT to predict the temperature of decomposition and the runaway behaviors. Fig. 4 demonstrates P-t and T-t curves for the decomposition of initiator CHP under the heating rate of 1.76 K·min⁻¹. The T-t curve exhibits a sharp exothermic peak. The temperature was rapidly elevated to about 408.9 K by self-decomposition when the reaction temperature reached ca. 394.65 K, accompanied by a zooming rise in pressure. It indicated that the rapid decomposition of CHP released a large amount of heat and produced two free radicals (Liu et al., 2019). The chemical entity oxidized 1,3-butadiene using the CHP as an initiator. The pressure behavior of the 1,3-butadiene oxidation process induced by 2 masses% CHP is shown in Fig. 5. The P-t curve displays a sharp drop and decreases from 1.0648 MPa to 0.7838 MPa (Table 1), indicating the rapid reaction of 1,3-butadiene, with a 1,3-butadiene conversion rate of 21.56 %. Also, there is a gradual visible color change of the product from colorless to yellow over time. This process is slightly different from the autoxidation approach (Table 1).

Close

Fig. 4

The P-t and T-t curves of the CHP initiator.

Export to PPT

Close

Fig. 5

The P-t and T-t curves of 1,3-butadiene initiated by CHP.

Export to PPT

3.1.3

3.1.3 Azo compounds AIBN initiated oxidation

Azo compounds AIBN is a typical self-reactive material widely used for blowing agents and initiators (Guilun et al., 2013; Yamamoto and Takahashi, 2016). The pressure behavior of AIBN is shown in Fig. 6. When the oxidation temperature was raised to 342.28 K, it could be seen that there was a distinct exothermic peak on the T-t curves of solid AIBN during the linear heating process, causing an increase in the system temperature from 342.28 K to 381.40 K with the rate of 0.72 K·s⁻¹. Sequentially, the system pressure increases from 0.5255 MPa to 3.6625 MPa rapidly. In that respect, the chemical nature of the AIBN initiator is the free radical formed following the thermal initiation reaction.

Close

Export to PPT

Close

Fig. 6

The P-t and T-t curves of the AIBN initiator.

Export to PPT

The oxidation of 1,3-butadiene initiated by AIBN was also investigated. According to Fig. 7, time course studies of MCPVT analysis have shown a quick decline of the pressure, with the value decreasing from 0.9899 MPa to 0.8824 MPa (Table 1), which indicates a chemical reaction occurred between reactants. Similarly, this process was accompanied by a gradual visible color change of the product from colorless to brown over time. These oxidized products were identified based on GC–MS and HS-GC–MS analyses in this work below.

Close

Fig. 7

The P-t and T-t curves of 1,3-butadiene initiated by AIBN.

Export to PPT

3.2 Oxidation kinetic

The oxidation reaction of 1,3-butadiene was affected by the presence of oxygen and initiator and further affected its thermal stability. In order to control the oxidation reaction process, the influence of reaction conditions (temperature and initiator) on the oxidation reaction rate of 1,3-butadiene was studied. The initiators and the peroxides products would decompose into many free radicals during its deep radical oxidation process. Those free radicals would lead to the reaction of 1,3-butadiene being complicated. Thus, this paper only discussed the kinetics of the initial oxidation reaction of 1,3-butadiene. To determine the initial oxidation rate of 1,3-butadiene under oxygen and initiator, two kinetic models, viz. autoxidation and initiated oxidation, were applied. In an autoxidation model, 1,3-butadiene reacted with oxygen and gave rise to polyperoxide, which was unstable and decomposed into a mass of free radicals. It would initiate the dimerization of 1,3-butadiene. Dimerization of 1,3-butadiene through the Diels-Alder reaction is second-order in the gas phase (Aldeeb et al., 2004). Based on that, the autoxidation model was expressed as the following equation:

To simplify the equation, the C₄H₆:O₂ ratio in the experiment was near unity. The alternating (1 to 1) copolymer of 1,3-butadiene and oxygen was mainly formed in the oxygen atmosphere (Hendry et al., 1968a,b). Under this condition, the same amount of 1,3-butadiene and oxygen was consumed in the oxidation reaction. That was, [B] = [O₂]. Since the two reactions occurred almost simultaneously, these could be treated as parallel reactions (Ren et al., 2015) Then, the total rate equation for Eqs. (3) and (4) is:

In addition, the oxidation and dimerization reactions mainly form liquid or solid products, and the vapor pressure of gas products can be negligible. It was assumed that both 1,3-butadiene and oxygen are ideal gases; an ideal gas state equation was employed to quantify the number of reactants.

Meanwhile, the mechanism for the oxidation of 1,3-butadiene under the initiator was shown as follows (Mayo et al., 1957; Mayo, 1958; Hendry et al., 1968a,b). Where B is a 1,3-butadiene molecule, B·, and BO₂· are radicals of 1,3-butadiene and oxygen units, and In· is a radical from the initiator.

Initiation reactions

Propagation reactions

Termination reactions

For long kinetic chains at high oxygen pressures (where reactions (13), 15, and 16 can be neglected), the conventional steady-state assumption gives

Then

Assuming that 1,3-butadiene is an ideal gas, the concentration of 1,3-butadiene can be approximated by the ideal gas equation ( $\mathit{PV}=nRT$ ). Eq. (18) can be rearranged as

Then.

The activation energy E_a and frequency factor A were determined by plotting the experimental rate constants obtained at different temperatures from the linear form of the Arrhenius equation (Eq. (21)).

From Eq. (21), the value of activation energy can be determined based on the slope, and the intercept can obtain the frequency factor.

For the experiment, the initial temperatures of the 1,3-butadiene oxidation process were 323 K, 333 K, 343 K, 353 K, and 363 K. The kinetic calculations began with the decrease of pressure, which was recorded as 0 h. Fig. 8 shows the plots of 1/P vs time t at different temperatures for the autoxidation model. Fig. 9 displays the curves of ln P vs time t at different temperatures for two initiated oxidation models. The kinetic parameters are displayed in Table 2. It indicated the reaction rates had different degrees of increasing with an increasing temperature. The initial reaction rate of autoxidation increased lightly with the increasing temperature, while the reaction rates of CHP and AIBN-initiated oxidation were remarkably improved. Adding initiators could be accelerated the oxidation to form more peroxides, which would be attributed to the formation of free radicals through decomposition (Wang et al., 2018). Furthermore, the reaction rate initiated by CHP was greater than that of AIBN-initiated. Such a trend suggests that organic peroxide compound has more effect than azo compound in enhancing the reactivity of 1,3-butadiene toward oxygen.

Close

Fig. 8

Plot of 1/P vs t for 1,3-butadiene autoxidation model.

Export to PPT

Close

Fig. 9

Curves for initiated oxidation models of 1,3-butadiene; (a) CHP-initiated oxidation; (b) AIBN-initiated oxidation.

Export to PPT

The activation energies were calculated by the Arrhenius equation (Eq. (21)), and plots of ln k vs T⁻¹ consisted of straight lines (Fig. 10). The slope to calculate the activation energy is displayed in Table 3. Autoxidation's apparent activation energy (E_a) was calculated to be 20.85 kJ·mol⁻¹. In two initiated oxidation models, the apparent activation energies E_a were 33.30 kJ·mol⁻¹ and 56.27 kJ·mol⁻¹, respectively, which were more extensive than autoxidation. There were two reasons for these differences. First, the autoxidation of 1,3-butadiene followed second-order reaction kinetics, whereas the initiated oxidation models were fitted to first-order reactions. Those corresponded to different reaction mechanisms, autoxidation, and free radical initiated mechanisms. Second, the initial onset temperature T₀ of 80 mass% CHP and pure AIBN were calculated to be 373 K and 383 K, respectively (Wu et al., 2012b; Zhang et al., 2015). Activation energy is the minimum energy required to initiate a reaction, which implies that a reaction with higher activation energy either needs a higher reaction temperature to gain adequate energy to initiate the reaction (Vuppaladadiyam et al., 2019). Therefore, the E_a of CHP-initiated oxidation was lower than that of AIBN-initiated oxidation. It indicated that the oxidation rates of different reaction systems and activation energy varied significantly. Hendry et al. (1968a,b) studied that the rate of the liquid-phase oxidation of 1,3-butadiene initiated by AIBN was 0.036 Mole·h⁻¹ at 760 Torr oxygen at 50 ℃ for 7.57 h. Aldeeb et al (Aldeeb et al., 2004) reported the reaction of 1,3-butadiene with oxygen fit with second-order, and the activation energy was 20.1 ± 2.1 kcal·mol⁻¹. Adusei et al (Adusei and Fontijn, 1993) researched the rate coefficients for consumption of ground-state oxygen atoms by reaction with 1,3-butadiene in the temperature range of 324 K to 363 K have been measured. The results revealed that the average rate coefficient and the activation energy were 1.74 × 10⁻¹¹ cm³·mol⁻¹·s⁻¹ and 27.42 kJ·mol⁻¹, respectively. Liu et al (Liu et al., 1988) investigated that the absolute rate constants for the gas-phase reaction of the OH radical with 1,3-butadiene in an argon atmosphere were measured at 1 atm over the temperature range 305–1173 K. At temperatures below 600 K, k(OH + 1,3-butadiene) = (1.4 ± 0.1) × 10⁻¹¹ exp[(440 ± 40)/T]cm³·mol⁻¹·s⁻¹. The thermal polymerization and oxidation characteristics of 1,3-butadiene were investigated using an ARC by Liu et al (Liu et al., 2020). The result showed that the activation energy of 1,3-butadiene oxidation was 112.6 kJ·mol⁻¹.

Close

Fig. 10

Plots of ln k vs T⁻¹.

Export to PPT

3.3 Peroxide value of 1,3-butadiene peroxide

In the presence of oxygen, 1,3-butadiene reacts with oxygen and possibly results in the formation of peroxide. In order to further confirm this result, the peroxide values of three kinds of oxidation processes were analyzed by iodimetry. The iodimetry results showed that 1,3-butadiene would produce different peroxide content under different oxidation processes. The effect of the main reaction conditions (reaction time, temperature, the molar ratio of 1,3-butadiene-to-oxygen, and initiators) on the peroxide value of the oxidation reaction was investigated, with the results shown in Fig. 11. The influence of reaction time on peroxide value is shown in Fig. 11(a). The relationship between the peroxide value and the oxidation reaction time is as follows: when the reaction time was from 0 h to 20 h, the peroxide value increased slowly from 0 to 2.37 mmol·kg⁻¹; when the reaction time was from 20 h to 26 h, the peroxide value increased rapidly from 2.37 mmol·kg⁻¹ to 27.76 mmol·kg⁻¹; at t = 26 h, the maximum peroxide value is obtained; when the reaction time was from 26 h to 30 h, the peroxide value decreased quickly from 27.76 mmol·kg⁻¹ to 8.5 mmol·kg⁻¹.

Close

Fig. 11

Effects of the reaction conditions on the peroxide value of autoxidation reaction. (a) Reaction time; (b) Reaction temperature; (c) The C₄H₆: O₂ molar ratio; (d) Initiated oxidation.

Export to PPT

The initial reaction of 1,3-butadiene with oxygen is to produce peroxide, but peroxide is unstable and prone to thermal decomposition reaction. That is, the oxidation reaction of 1,3-butadiene with oxygen is:

From the above reactions, it can be seen that in the initial reaction (22) stage, the reaction is dominated by the formation of peroxide, and the peroxide concentration increases with time. Due to the interaction of Eq. (22) and (23), the peroxide concentration will reach the maximum of 27.76 mmol·kg⁻¹ (at 26 h and 338 K). Free radicals are very active and can trigger many reactions, including autocatalytic reactions. At a specific temperature (e.g. 338 K), the peroxide concentration cannot maintain a stable maximum value because Eq. (24) decreases with time. In order to reduce the thermal decomposition of peroxide, it is possible to reduce the reaction temperature.

The effect of temperature on peroxide value is shown in Fig. 11(b). The concentration first increased from 3 mmol·kg⁻¹ to 27.85 mmol·kg⁻¹ with increasing temperature from 318 K to 338 K at 26 h, indicating more and more peroxides were produced. Then it decreased with increasing temperature after 338 K. Peroxides are temperature-dependent and decompose at high temperatures. Handy and Rothrock have reported that the primary product from the oxidation of 1,3-butadiene at 353 K is a nonvolatile copolymer of 1,3-butadiene and oxygen, which contained 1,2- and 1,4-butadiene units as polybutadiene, but some or all of these units are separated by peroxide (—O—O—) links (Handy and Rothrock, 1958). The unstable —O—O— bond of peroxide was prone to decompose under high temperatures and would give rise to many free radicals that promote the deep oxidation of 1,3-butadiene with oxygen. The free radical was conducted in combination with those MCPVT observation results.

The effect of the molar ratio of 1,3-butadiene-to-oxygen on peroxide value was assessed. Fig. 11(c) shows that the peroxide value increased with an increasing molar ratio, indicating that 1,3-butadiene reacted completely with oxygen and produced more peroxides. The effect of initiators on peroxide value was discussed in Fig. 11(d). As proposed in Section 3.2, the reaction rate of the 1,3-butadiene oxidation process increased in the initiators' presence. So, the reaction time of 1,3-butadiene oxidation initiated by initiators was reduced to 8 h to detect the peroxide value. The results showed that the CHP-initiated oxidation would form a high level of peroxide during its oxidation process at 343 K, reaching the highest value of 12.92 mmol·kg⁻¹, then decreased with the reaction temperature increased. The peroxide value of AIBN-initiated reached the highest value of 12.09 mmol·kg⁻¹, which was less than that of CHP-initiated. Thus, it is notable that 1,3-butadiene had a vital ability for peroxide production in contact with oxygen and tended to proceed with the decomposition of peroxide to form oxidation products.

3.4 Thermal decomposition characteristics of oxidation products

DSC is a standard instrument used to measure organic matter's oxidation stability and organic peroxide's decomposition heat (Chen et al., 2008; Zhang et al., 2015). Li et al. reported that peroxide was an important energetic material with potential hazards because its decomposition releases heat (Li et al., 2019). In order to investigate the relationship between peroxide value and heat release, peroxides with different peroxide values were measured by DSC. The exothermic onset temperature (T₀) and the heat of decomposition (Q_DSC) of peroxide were measured by the thermal curves of the DSC by dynamic thermal scanning under a heating rate of 10 K·min⁻¹.

Different peroxide values were measured to analyze exothermic reactions, as shown in Fig. 12. It can be seen that there are distinct exothermic peaks on the DSC curves of 1,3-butadiene oxidation products. The results show that the peroxide value increased from 15.3 mmol·kg⁻¹ to 33.6 mmol·kg⁻¹, corresponding to the decomposition heat Q_DSC increased from 897.2 J·g⁻¹ to 1887 J·g⁻¹ (Table 4). According to the United Nations Publication Recommendations on the Transport of Dangerous Goods (Persson et al., 2018), the substance could be classified as the fifth type of hazardous material because its Q_DSC exceeds 250.0 J·g⁻¹. It indicated that the oxidation products were hazardous. Besides, the T₀ was determined to be advanced and decreased with the increase of peroxide value. It indicated that a higher starting temperature would inevitably cause significant runaway behavior to appear earlier, resulting in the safety of production, transportation, and usage being threatened.

Close

Fig. 12

Heat flow vs temperature for autoxidation products.

Export to PPT

DSC also measured the oxidation products of 1,3-butadiene initiated by initiators, and the results are shown in Fig. 13. The heat flow curves of CHP and AIBN-initiated products performed strongly exothermic peaks at the temperature range of 334.62–453.84 K, and its decomposition heat was 826.6 J·g⁻¹ and 1152.8 J·g⁻¹, respectively. However, these peak patterns showed a significant difference from autoxidation, which residual initiators in the oxidation products may cause. In the previous work, to prevent the thermal explosion of 80 mass% CHP for transportation, manufacturing process, or use, Wu et al. analyze thermal hazards and safe storage of 80 mass% CHP (Wu, 2012). The T₀ of 80 mass% CHP by DSC was calculated to be 373 K, and the Q_DSC was determined to be 1200 J g⁻¹ with a heating rate of 10 K·min⁻¹. In addition, the thermal decomposition characteristic of AIBN was investigated by Zhang et al (Zhang et al., 2015). The T₀ of AIBN by DSC was calculated to be 383 K, and the Q_DSC was determined to be 1024 J g⁻¹ with a heating rate of 10 K·min⁻¹. Therefore, the exothermic peak was widened, and the T₀ was advanced in the presence of the initiator. The thermal parameters of three kinds of products are shown in Table 4. The decomposition of peroxide can cause potential thermal hazards and release radicals to form many secondary products.

Close

Fig. 13

Heat flow vs temperature for initiated oxidation products. (a) CHP-initiated oxidation; (b) AIBN-initiated oxidation.

Export to PPT

3.5 Products of 1,3-butadiene oxidation

To further characterize the complicated 1,3-butadiene oxidized products, the oxidized products were collected and monitored by GC-MS and HS-GC–MS. The relative quantities of each identified component were estimated based on the peak area. The gas products are shown in Table 5. Similar oxidized products were produced from three oxidation models for autoxidation, CHP-initiated oxidation, and AIBN-initiated oxidation of 1,3-butadiene. Furan is the main gas product, with a low 1,3-butadiene conversion rate of 0.19 %, 0.87 %, and 0.21 %, respectively. In the case of CHP-initiated oxidation, additional products of 2-propenal and formic acid were formed with a relative content of 0.30 % and 1.65 %, respectively.

The liquid products are summarized in Table 6. In the case of autoxidation, the six main products were 1,2-divinlcyclobutane, 3-butene-1,2-diol, 2(5H)-furanone, 4-vinylcyclohexene, allyl alcohol and 2,6-cyclooctadien-1-one with relative contents of 0.75 %, 0.64 %, 3.63 %, 25.38 % 1.33 %, and 0.58 %, respectively. The main dimerization products were 1,2-divinlcyclobutane and 4-vinylcyclohexene, whereas the other four were oxidation products. It can be inferred that autoxidation of 1,3-butadiene was undergone with a low conversion rate. As a comparison, the products of CHP and AIBN-initiated oxidation were also determined. Many aldehydes, acids, and alcohols were produced with a high 1,3-butadiene conversion rate. In the case of AIBN-initiated oxidation, 4-vinylcyclohexene (28.67 %) was still the major product, but the yield of oxidized products had increased to 23.30 %. It indicated AIBN initiator promoted the oxidation of 1,3-butadiene. Besides, the quantities of the main products resulting from CHP-initiated oxidation were 2-propenal, allyl alcohol, and 3-cyclohexene-1-carboxylic acid, with relative content of 10.08 %, 15.96 %, and 8.22 %, respectively. Compared to autoxidation, the yields of similar oxidized products for 3-butene-1,2-diol, 2(5H)-furanone, allyl alcohol, and 2,6-cyclooctadien-1-one were increased. It indicated that the CHP initiator significantly affects the oxidation of 1,3-butadiene. Table 7 displays the volatile components in solid products of 1,3-butadiene oxidation. The main products were 2(5H)-furanone and 4-vinylcyclohexene. Therefore, results from the GC-MS and HS-GC-MS indicated that each oxidation mode produced a more or less similar set of oxidized products. For two initiated oxidation models, CHP and AIBN are initiators of free radical chain reactions, but the initial decomposition temperatures were different. The oxidation of 1,3-butadiene was initiated by the free radicals generated from the initiators. At the same temperature and time, the degrees of 1,3-butadiene oxidation initiated by various initiators were different, which led to different results for the measured product distribution. For the autoxidation model, there are two steps: the reaction of 1,3-butadiene with oxygen was dominated by the formation of peroxides, and peroxides would further decompose to initiate the complex reaction. The initial oxidation reaction could also lead to the different distribution of 1,3-butadiene reaction products.

There are two reactions of 1,3-butadiene in an oxygen atmosphere: (1) oxidation reaction; (2) polymerization. Under a nitrogen atmosphere, the reaction products of 1,3-butadiene are not detected even if the temperature is 373 K. When the temperature is 383 K, a small amount of chemical reaction occurs. The main products are 1,3-butadiene dimers, including 4-vinylcyclohexane and 1,2-divinylcyclobutane. However, the 1,3-butadiene oxidation reaction can quickly occur in an oxygen atmosphere. When the temperature is 343 K, there are many reaction products, including oxidation products (oxygenated compounds) and polymers (such as dimers). The results show that the autoxidation reaction can significantly reduce the reaction temperature of 1,3-butadiene.

A possible scheme of 1,3-butadiene oxidation is proposed, as shown in Scheme 1. There are two possible pathways. The first pathway is the oxidation process of 1,3-butadiene. As a result of O-atom addition to the π bond in 1,3-butadiene, two possible biradical isomers could be produced after the initiator and O₂ addition. The isomerization of 1,4-buteneperoxy radical (isomer A) would lead to the formation of 3,6-dihydro-1,2-dioxin (DHD), a highly reactive peroxide that may initiate the polymerization of the 1,3-butadiene (Aldeeb et al., 2004). This peroxide was instability and formed 2,5-dihydrofuran. It is known that 2,5-dihydrofuran undergoes H₂ elimination to produce furan (Assa et al., 1986; Wellington and Walters, 2012). The other isomer, 1,2-buteneperoxy radical (isomer B), was favorable to isomerizing vinyloxirane, which underwent rapid ring expansion to 2,3-dihydrofuran at relatively low temperatures (Crawford et al., 1976) The reactions of 2,3-dihydrofuran were reported (Assa and Menashe, 1989) as unimolecular isomerization to crotonaldehyde and cyclopropanecarboxaldeldehyde. The formation of crotonaldehyde can be inferred based on previous work (Lifshitz and Laskin, 1994) on the thermal reactions of 2-methyl-4,5-dihydrofuran, which showed that 2,3-dihydrofuran could isomerize either directly to a noncyclic aldehyde or via the cyclopropane to form the noncyclic aldehyde. Besides, at one-atmosphere pressure, collisions would be instrumental in the biradical isomer B undergoing a 1,2-hydrogen shift to form 3-butenal (Brezinsky et al., 1985). It was expected to react in a way characteristic of aldehydes (Colket et al., 1977). In the presence of abundant OH, H, and O, the easily abstractable aldehyde H is removed and is followed by the loss of CO. In this case, the resonantly-stabilized long-lived allyl radical is formed and is postulated to react with either O or HO₂ in a radical–radical reaction to form acrolein (Brezinsky et al., 1985). The two biradical isomers could form 2-butene-1,4-diol and 3-butene-1,2-diol via H-addition (Handy and Rothrock, 1958; Zhang et al., 2019). The deep radical oxidation of 1,3-butadiene is a remarkably complex reaction. The reaction of 1,3-butadiene with oxygen could give rise to an unstable peroxide. The spontaneous homolysis of the —O—O— bond in the initial autoxidation product will produce highly reactive free radicals. Then, the generated free radicals can react with peroxides or 1,3-butadiene to form complex products. Rapid oxidation reactions must be avoided because they can lead to runaway reactions or explosion hazards.

Close

Scheme 1

The possible scheme of 1,3-butadiene oxidation.

Export to PPT

The other pathway is the dimerization of 1,3-butadiene. Dimerization of 1,3-butadiene through Diels-Alder type cycloaddition was suggested based on previous work (Doering et al., 1972; Berson and Dervan, 1973) to predict a reaction mechanism. Two isomers of 1,3-butadiene, cis-1,3-butadiene and trans-1,3-butadiene, would dimerize into 4-vinylcyclohexane, 1,5-cyclooctadiene, trans-1,2-divinylcyclobutane, and cis-1,2-divinylcyclobutane dimers. These reactions would take place through the formation of the intermediate octa-1,7-diene-3,6-diyl diradicals (trans, trans), (cis, trans), and (cis, cis) (Aldeeb et al., 2004). In the presence of abundant O₂, 4-vinylcyclohexane and 1,5-cyclooctadiene would oxidize into 3-cyclohexene-1-carboxaldehyde, 3-cyclohexene-1-carboxylic acid, 5,10-dioxatricyclo[7.1.0.0^4,6]decane, 9-oxabicyclo[6.1.0]non-4-ene, 2,6-cyclooctadien-1-ol, and 2,6-cyclooctadien-1-one.