- Email: [email protected]
Thermal hazard analyses for the synthesis of benzoyl peroxide
Journal of Loss Prevention in the Process Industries 43 (2016) 35e41
Contents lists available at ScienceDirect
Journal of Loss Prevention in the Process Industries journal homepage: www.elsevier.com/locate/jlp
Thermal hazard analyses for the synthesis of benzoyl peroxide Yong Zhang, Lei Ni, Juncheng Jiang*, Jun Jiang, Wenxin Zhang, Jiajia Jiang, Mingguang Zhang Jiangsu Key Laboratory of Hazardous Chemicals Safety and Control, College of Safety Science and Engineering, Nanjing Tech University, Nanjing, 210009, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 5 January 2016 Received in revised form 17 April 2016 Accepted 17 April 2016 Available online 21 April 2016
Benzoyl peroxide (BPO), historically, due to its wide applications around the world, has caused many serious fire and explosion accidents. In this paper, in order to prevent such accidents, thermal hazard analyses of synthesis of benzoyl peroxide were studied. Firstly, in order to obtain thermal hazard coefficients, the exothermic processes with different alkaline solutions were studied by RC1e (Reaction Calorimeter). The alkaline solutions are NaOH, NH4HCO3 and Na2CO3, respectively. Secondly, the thermal decomposition of BPO product was studied by PHI-TECⅡto analysis the thermal stability. Finally, the possibility of runaway reactions and thermal risks of synthetic process were evaluated according to the Stoessel criticality diagram. In the first stage, the test results of the reaction heat (DHm), heat release rate (qr) and adiabatic temperature rise (DTad) with different alkaline solutions were NaOH > Na2CO3 > NH4HCO3. In the second stage, according to the analysis of experimental data, the heat release rate of reaction with NH4HCO3 solution was the slowest, while the Maximum Temperature of the Synthesis Reaction (MTSR) and the adiabatic temperature rise (DTad) were lowest when using Na2CO3 solution. The time needed to reach the maximum reaction rate under the adiabatic condition (TMRad) was 0.83 h when using NaOH solution. The temperature was 38.24 C when TMRad is 24 h (TD24). The evaluation results of the process showed that the risks of reactions with NaOH solution or NH4HCO3 solution were not acceptable. Only the risk of the reaction with Na2CO3 solution was acceptable. Therefore, the safety level of synthesis of benzoyl peroxide can be significantly improved by using Na2CO3 solution. Research in this paper can not only improve the safety level of BPO reaction and storage processes, but also provide technical support for stability criterion of BPO decomposition reaction. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Benzoyl peroxide Thermodynamic properties Thermal runaway Reaction kinetics Stability
1. Introduction Benzoyl peroxide (BPO) is a strong oxidizing organic peroxide and easy to decompose under heating. The thermal explosion can be caused by impact, heat, friction, etc (Zhao et al., 2012; Wang et al., 2013; Lee et al., 2014). BPO is widely used as initiator in chemical industries (Zaman et al., 2001). Based on the NFPA standard (National, 2011), the classification level of BPO for flammability and reactivity is 4. In the document of state administration of work safety of China, the synthesis of BPO is a typical high-risk process. In China, the traditional method for the synthesis of BPO is that
* Corresponding author. Mail box 13, No. 200 North Zhongshan Road, Nanjing Tech University, Nanjing 210009, China. E-mail addresses: [email protected] (L. Ni), [email protected] (J. Jiang), [email protected] (M. Zhang). http://dx.doi.org/10.1016/j.jlp.2016.04.007 0950-4230/© 2016 Elsevier Ltd. All rights reserved.
benzoyl chloride and hydrogen peroxide are used as main raw material with alkaline solution. Fan Juan (Fan, 2002) found that the decomposition rate of H2O2 and hydrolysis rate of BPO are rapid. Although low temperature environment can control the above two side reactions effectively, it can also decrease the main reaction rate. Compared with NaOH and Na2CO3, the process of NH4HCO3 dissolved in water is an endothermic process. Therefore, the reaction can keep the process temperature without cooling and the reaction time can be shortened (Duan et al., 2003). Most papers deal with the thermal decomposition of BPO. DaoXing Sun et al.(Sun et al., 2012) used C80 calorimetry and accelerating rate calorimetry (ARC) to study the hazardous characteristics of BPO and obtained the pre-exponential factor (3.61 1019 s1) and activation energy (152.80 kJ mol1). Liu et al. (2013) analyzed the thermal hazards of BPO with incompatibilities (H2SO4, NaOH, and Na2SO3) and found that thermal hazard increased prominently when BPO were mixed with H2SO4, NaOH, and Na2SO3. The thermal
36
Y. Zhang et al. / Journal of Loss Prevention in the Process Industries 43 (2016) 35e41
decomposition of BPP is an autocatalytic reaction. When the BPO was mixed with benzoic acid, benzene or phenol, the hazard significantly increased (Wang et al., 2013; Liu et al., 2015). Acrylic acid and methyl acrylate were identified as catalyst for thermal decomposition of BPO (Huang et al., 2013). In the first step of synthesis of BPO, the thermal hazard of H2O2 mixed with alkaline solutions increased significantly. Chen et al. (2006) used differential scanning calorimetry (DSC) to obtain that the initial decomposition temperature (Tonset) of 31mass% H2O2 was 40.54 C by heating rate of 2 C min1. Wu et al., (2010) found that the Tonset of H2O2 mixed with NaOH was close to room temperature. Many studies have reported the hazards of the decomposition of BPO, but few researches study the hazards of synthesis of BPO mixed with different alkaline solutions. The synthesis of BPO is an exothermic reaction. There is no detailed account of a safety assessment of the synthesis of BPO. In this study, we report the process safety evaluation of the synthesis of BPO mixed with different alkaline solutions. RC1e is used to study the exothermic characteristics for the synthesis of BPO with different alkaline solutions firstly. Secondly, the adiabatic decomposition of raw product is studied by PHI-TECⅡ and the TMRad and TD24 are calculated from the fitting kinetic models. Finally, the risk level of the synthesis of BPO is evaluated based on the Stoessel criticality diagram and the safest process of synthesis of BPO which also has a good yield of BPO can be determined.
2. Experiments 2.1. Reaction path There are two steps for the synthesis of Benzoyl peroxide. Firstly, Hydrogen peroxide (H2O2) is added to alkaline solution to prepare the hydrogen peroxide alkaline solution. Secondly, benzoyl chloride is added to hydrogen peroxide alkaline solution. See Fig. 1.
2.2.2. Reaction calorimeter (RC1e) (1) RC1e is manufactured by METTLER TOLEDO and used to measure the exothermic condition of reaction. RC1e is an automatic synthesis reaction system. A 2 L reactor is used in our experiment. Specific testing principle can be found in the literature (Lee et al., 2014). (2) Experiment conditions: reaction temperature is 10 C. Stirring speed is 100 rmin1. Feed rate is 13.59 g s1 (H2O2) and 13.52 g s1 (benzoyl chloride). Solution concentration is 3.93 mol L1. Alkali solutes are NaOH, NH4HCO3 and Na2CO3.
2.2.3. Nuclear magnetic resonance spectrometer (Bruker AV500 MHz NMR) (1) Bruker AV500 MHz NMR is used to analyze the molecular structure and determinate the material content. (2) Experiment reagent is Sample 2 and solvent is CDCl3.
2.2.4. PHI-TECⅡ (1) PHI-TECⅡ is a quasi-adiabatic calorimeter manufactured by HEL. PHI-TECⅡ is used to obtain the thermal hazard parameters, such as temperature versus time, pressure versus time, self-heating rate and pressure-rising rate. The heat-waitsearch (H-W-S) mode for detecting the self-heating rate is adopted for PHI-TEC Ⅱ (Townsend and Tou, 1980; Sun et al., 2004). The volume of sample ranges from 0.5 to 100 ml. The experimental equipment is shown in Fig. 2. (2) Experiment conditions: the mass of sample 1 is 1.01 g. The volume of test is 10 ml. The temperature range is 60e300 C. Heating step is 5 C. The criterion for no self-heating is 0.02 C min1. 3. The experiment results and analysis of RC1e 3.1. The experiment results of first step
2.2. Experiment reagent and instrument 2.2.1. Experiment reagent and sample (1) Reagent: H2O2 (AR 30 wt%), NaOH (AR), Na2CO3 (AR), NH4HCO3 (AR), benzoyl chloride (AR), sodium dodecyl sulfate (SDS, CP), Sinopharm Chemical Reagent Co., Ltd; deionized water. (2) Sample1 is the product of the synthesis of BPO which is produced by RC1e and then liquid-solid separation. Sample 2 is the product of sample1 being washed and dried.
Because both the adiabatic temperature rise (DTad) and the calculated MTSR are small in first step. Therefore, the worst conditions for the whole process corresponding to either reactions under adiabatic conditions or cooling system failure event are considered in this paper. Tmax ¼ DTad þ Tp. The exothermic
In order to study the adiabatic decomposition condition of BPO effectively, Sample 1 is used as the sample of adiabatic experiment. Sample 2 is used for product analysis.
First steps:
Na2O2 + 2H2O
2NaOH + H2O2 O C Cl + Na2O2
2 O
Second steps:
O
C O O C
Fig. 1. Flow chart of the synthesis process of BPO.
+ 2NaCl
Fig. 2. Schematic diagram of PHI-TECⅡ.
Y. Zhang et al. / Journal of Loss Prevention in the Process Industries 43 (2016) 35e41
37
parameters of different alkaline solution are shown in Table 1. As shown in Table 1 and Fig. 3, all the parameters (DHm, DTad, Tmax and heat release rate) are NaOH > Na2CO3 > NH4HCO3. The reaction with NH4HCO3 is an endothermic reaction and may be an inherent reaction. The reaction with NaOH or with Na2CO3 is exothermic reaction. All the parameters (DHm, DTad, Tmax and heat release rate) in the reaction with NaOH are larger than those of reaction with Na2CO3. The reason may be that NaOH is the strong alkali. Therefore, the reaction with NaOH solution generated the maximum heat. In Fig. 3, the reaction with NH4HCO3 is stable after 6000s and very time-consuming. The reaction times in different alkaline solutions are NaOH > Na2CO3 > NH4HCO3. Although the reaction with NH4HCO3 is an inherent reaction, it is very time-consuming and the production efficiency is very low. Fig. 3. Heat release rate with different alkaline solutions at the first stage.
3.2. The experiment results of second step The second step of the synthesis of BPO is heterogeneous reaction. Therefore, the surfactant (SDS) is added to the reaction and used to avoid caking. But excessive amounts of surfactant can't be used. Because the reactor may generate a lot of foamy and raw material may overflow the reactor. As shown in Fig. 4, the reactions with different alkaline solutions have different heat release rate: NaOH > Na2CO3>NH4HCO3. In order to study the exothermic reaction characteristics under different alkaline solutions, some parameters are considered in the experiment, such as the runaway severity (DTad) (Stoessel, 1993), the maximum temperature of reaction (Tcf) and the maximum temperature of reaction runaway (MTSR). In Table 2, all the parameters values (DHm, DTad, MTSR) are NaOH > NH4HCO3 >Na2CO3. When alkaline solution is NH4HCO3 solution and benzoyl chloride is added to the second step reaction, white gas is generated and reactants may decompose. The reactions with other alkaline solutions don't have the phenomenon. Because OH H2 O2 þ 2C6 C5 COCl then !ðC6 H5 COÞ2 O2 þ 2HCl, NaHCO3 þ HCl/NaCl þ H2 O þ CO2 [. Compared first step with second step, the reaction with the NH4HCO3 solution is an endothermic reaction in first step. But heat release in second step is larger than that of reaction with Na2CO3. Therefore, if we want to compare the risk degree comprehensively for the reactions with two kinds of alkaline solutions, the runaway possibility parameters (TMRad and TD24) should be considered (Marco et al., 2003). In Fig. 5 (a), (b) and (c), the degree of material accumulation (Xac) and Tcf (MTSR) increase to the maximum at the stoichiometric point and then decrease with the reaction time. At the stoichiometric point, the degrees of heat accumulation are NH4HCO3> NaOH > Na2CO3.
Fig. 4. Heat release rate with different alkaline solutions at the second stage.
Table 2 Heat release parameters with different alkaline solutions. Alkaline solutions
Molarity/mol L1
DHm/kJ mol1
DTad/ C
MTSR/ C
NaOH NH4HCO3 Na2CO3
3.93
131.64 74.86 66.98
77.11 51.48 35.76
68.05 54.49 31.07
calculated. See Table 3. In Table 3, the reaction with NH4HCO3 has the highest yield. The reason may be that the reaction spends too much time.
3.3. Product analysis
4. The adiabatic decomposition of sample 1
Sample 2 is obtained by filtering and drying sample 1. Sample 2 dissolves in CDCl3 and is characterized with hydrogen nuclear magnetic spectra. See Fig. 6. There are four peaks, BPO peak, CDCl3 peak, standard peak and water peak. Product yield can be
The test results of sample 1 under adiabatic condition are shown in Table 4 and Fig. 7. The initial decomposition temperature (Tonset) of sample 1 is 67.57 C. BPO decompose and release the gas and heat. Temperature and pressure increase rapidly. Maximum rate of temperature rise is 4598.3 C min1.
Table 1 The parameters of different alkaline solution. Alkaline solutions
DHm/kJ mol1
DTad/K
Tmax/ C
Tmax>25 C
NaOH NH4HCO3 Na2CO3
27.935 10.060 6.720
12.27 4.41 2.74
22.27 5.59 12.74
NO NO NO
4.1. Kinetic models In order to determine the thermokinetic parameters of BPO decomposition, a simple nth order reaction is assumed and the rate equation can be written in the form (Townsend and Tou, 1980):
38
Y. Zhang et al. / Journal of Loss Prevention in the Process Industries 43 (2016) 35e41
Fig. 5. MTSR test results and heat conversion with different alkaline solutions.
Fig. 6. 1HNMR spectrum of BPO in CDCl3.
Y. Zhang et al. / Journal of Loss Prevention in the Process Industries 43 (2016) 35e41
39
Number
Alkaline solutions
Productivity/%
(R2). Then, Ea ¼ 99.69 kJ mol1, A ¼ 6.12 1011 s1. In order to further evaluate the risk of reaction runaway, the time to reach the maximum reaction rate (q) should be calculated.
1 2 3
NaOH NH4HCO3 Na2CO3
0.8381 0.8478 0.8238
4.2. Time to reach the maximum reaction rate (q)
Table 3 Productivity of different alkaline solutions.
The equation of the time to reach the maximum reaction rate is shown as follow (Chen et al., 2012):
Table 4 Test data of sample 1 under adiabatic condition. Parameter
Sample1
F
4.78 67.57 199.24 131.67 3.967 4598.3
Decomposition temperature of self-heating/ C Maximum temperature/ C Adiabatic temperature rise/ C Maximum pressure/Mpa Maximum rate of temperature rise/ C min1
ln q ¼
Ea 1 InA R T
(4)
Submitting the Ea and A from above simulation to the Equation (4), the temperature TD24 (TD8) can be calculated whenq ¼ 24 h or 8 h. See Table 6. When the reaction temperature is higher than the MTSR, it may have the second reaction. The TMRad can also be obtained and shown in Table 7. In Table 6, the DTad of sample 1 is 629.38 C. Based on the Zurich hazard analysis (Zogg and Zürich, 1987), the reaction severity is high. The temperature in summer is easy to reach the TD24 (38.24 C). In Table 7, the TMRad of the reaction with Na2CO3 is higher than those of the reactions with other alkaline solutions. Base on the assessment criteria, the probability of a runaway reaction is almost impossible. 5. Safety assessment of the synthesis of benzoyl peroxide
Fig. 7. Temperature (pressure) histories of sample 1 under adiabatic decomposition condition.
Tf T n n1 T T n n1 dT Ea ¼ kDTad cA0 ¼ A exp DTad f cA0 dt DTad RT DTad (1) Then,
dT 1 DTad k¼ DT dt ad Tf T
!n c1n A0
(2)
Based on the experiment or calculated values (T, Tf, DTad and dT/ dt), if suitable reaction order is selected, lnk represents linear correlation to1/T. See Equation (3).
ln k ¼ ln A
Ea RT
(3)
where, dT/dt is temperature rising rate, C s1. Tf is the highest decomposition temperature, C. k is reaction rate constant. cA0 is the initial concentration of component A, mol L1. XA is conversion of component A. n is reaction order. A is pre-exponential factor. Ea is activation energy, J mol1. Based on Equations (2) And (3), experiment and fitting curves of Ink vs 1/T are shown in Fig. 8. n increases from 0.5 to 2.5 which the step is 0.5. To get more accurate results, n increases from 1.9 to 2.5 which the step is 0.1. The results are shown in Table 5. When n ¼ 2.2, the fitting results have the highest correlation coefficient
Base on the criticality classes of thermal runaway scenario (Nanchen et al., 2009), the critical temperatures are summarized in Table 8. In the reactions with NaOH or NH4HCO3 alkaline solutions, Tr < TD24 < MTSR < MTT. The risk level is 5 and the risk is not acceptable. After loss of control of the synthesis reaction, the decomposition reaction will be triggered (MTSR > TD24) and the technical limit will be reached during the runaway of the secondary reaction. The decomposition of BPO could lead to critical pressure increase. There is no safety barrier between the main and secondary reaction. Only quenching or dumping can be used (Stoessel, 2008). In the reactions with Na2CO3, Tr < MTSR < TD24 < MTT. The risk level is 2 and the risk is low. But, if the reaction mass is maintained for a longer time under heat accumulation conditions, the decomposition reaction could be triggered and reach the MTT. The BPO decomposition releases heat rapidly. When the reaction temperature reaches the boiling point, the system pressure will rise quickly due to the generation of a large amount of steam. The reactants may spill from the reactors and even explode. Therefore, the heat accumulation needs to be avoided and no special safety measures are required. From the evaluation result, although the reaction with NH4HCO3 is endothermic reaction in the first step, the severity and probability of runaway reaction in second step are higher than that of reaction with Na2CO3. Therefore, reaction with Na2CO3 has the lowest risk level among the reactions with three alkaline solutions. 6. Conclusions In this paper, a detailed risk assessment — encompassing the determination of the severity and the probability of the synthesis of BPO with different alkaline solutions — has been studied. The main findings resulted in the following: (1) In the first step of the synthesis of BPO, DHm, DTad and heat release rate are NaOH > Na2CO3 > NH4HCO3. The reaction
40
Y. Zhang et al. / Journal of Loss Prevention in the Process Industries 43 (2016) 35e41
4
15 Experiment(n =2.2) Simulation
2 2
R =0.96851 0
5
-2
lnK
lnK
n=0.5 n=1 n=1.5 n=2 n=2.5
10
0
-4 -5
-6
-10
-8 -10
0.0022
0.0024
0.0026
0.0028
0.0030
-15 0.0020
0.0022
0.0024
1/T
0.0026
0.0028
0.0030
1/T Fig. 8. Calculated and fitting curves of Ink vs 1/T of sample 1.
Table 5 Kinetic parameters of sample 1 with different reaction order. n
1.9
2
2.1
2.2
2.3
2.4
2.5
R2 Ea A
0.96775 92.57 5.37 1010
0.96818 94.95 1.21 1011
0.96842 97.32 2.72 1011
0.96851 99.69 6.12 1011
0.96845 102.06 1.38 1012
0.96827 104.43 3.10 1012
0.96799 106.80 6.97 1012
Table 6 Thermal decomposition characteristic data of samples modified by thermal inertia factor. Sample 1
DTad,sample/ C
629.38 38.24 47.38
TD24/ C TD8/ C
(5) Base on the Stoessel criticality diagram, only the risk level of reaction with Na2CO3 is level 2. Therefore, among the three kinds of alkaline solutions, the thermal hazard for the synthesis of BPO with Na2CO3 alkaline solution is lowest. Further, vent-sizing experiments can be performed for the synthesis of BPO with alkaline solution to prevent over pressurization because of the produced gas.
Table 7 TMRad of different alkaline solutions.
Acknowledgments
Alkaline solutions
MTSR/ C
TMRad/h
Extended assessment criteria
NaOH NH4HCO3 Na2CO3
68.05 54.49 31.07
0.83 3.55 59.45
frequent probable remote
Table 8 Risk classification with different alkaline solutions. Alkaline solutions
Tr/ C
MTSR/ C
TD24/ C
MTT/ C
The risk level
NaOH NH4HCO3 Na2CO3
10
68.05 54.49 31.07
38.24
100
5 5 2
with NH4HCO3 is an endothermic reaction and may be an inherent reaction. (2) In the second step of the synthesis of BPO, DHm, DTad and heat release rate are NaOH > NH4HCO3> Na2CO3. The reaction with NH4HCO3 has highest heat accumulation. (3) Based on the experiment values of PHI-TECⅡ, Ea of the sample 1 is 99.69 kJ mol1, A is 6.12 1011 s1 and TD24 of BPO is 38.24 C. (4) TMRad of different alkaline solutions is NaOH < NH4HCO3 < Na2CO3. The reaction with NaOH has the minimum TMRad and the maximum runaway possibility.
The authors are grateful for the support given by key project of National Natural Science Foundation of China under Grant No. 21436006, National Natural Science Foundation of China under Grant No. 51176070, Jiangsu Natural Science Foundation of China (BK20141457, SBK2015043673), Project of the Natural Science Foundation of Jiangsu Province, China “Analyzing model and optimizing technology for safety capacity of a chemical industrial park” (BK 2012824), Jiangsu Province “333 high-level talents project” (BRA2015313). Nomenclature A cA0 Ea dT/dt k MTSR MTT n TMRad Tcf TD8 TD24 Tf
pre-exponential factor the initial concentration of component A, mol L1 activation energy, J mol1 temperature rising rate, C s1 rate constant the Maximum Temperature of the Synthesis Reaction, C Maximum temperature for technical reasons, C reaction order the time needed to reach the maximum reaction rate under the adiabatic condition, s the maximum temperature of reaction runaway, C the temperature when TMRad is 8 h, C the temperature when TMRad is 24 h, C the highest decomposition temperature, C
Y. Zhang et al. / Journal of Loss Prevention in the Process Industries 43 (2016) 35e41
the maximum temperature of reaction, C normal process temperature, C the conversion rate of component A F thermal inertia factor q the time to reach the maximum reaction rate DHm molar reaction heat, kJ mol1 DTad adiabatic temperature rise, C DTad,sample the revised adiabatic temperature rise, C
Tmax Tr XA
References Chen, K.Y., Lin, C.M., Shu, C.M., Kao, C.S., 2006. An evaluation on thermokinetic parameters for hydrogen peroxide at various concentrations by DSC. J. Therm. Anal. Calorim. 85, 87e89. Chen, Li-Ping, Liu, Ting-Ting, Yang, Qian, Chen, Wang-Hua, 2012. Thermal hazard evaluation for iso-octanol nitration with mixed acid. J. Loss Prev. Process Ind. 25, 631e635. Duan, Haibao, Cui, Chunwei, Cai, Guocheng, Zhang, Rui, Peng, Qijun, 2003. New method for synthesis of benzoyl peroxide. Chem. Ind. Eng. Prog. (China) 22 (3), 293e294. Fan, Juan, 2002. Discussion about preparation of benzoyl peroxide. Appl. Chem. Ind. (China) 31 (6), 1e2. Huang, Yi-Hao, I, Yet-Pole, Chen, Nan-Chi, Wu, Sheng-Hung, Horng, Jao-Jia, Wu, Yung-Tin, Wen, I-Jyh, 2013. Thermal runaway reaction evaluation of benzoyl peroxide using calorimetric approaches. J. Therm. Anal. Calorim. 113, 595e598. Lee, Ming-Hsun, Chen, Jiann-Rong, Shiue, Gong-Yih, Lin, Yan-Fu, Shu, Chi-Min, 2014. Simulation approach to benzoyl peroxide decomposition kinetics by thermal calorimetric technique. J. Taiwan Inst. Chem. Eng. 45, 115e120. Liu, S.H., Hou, H.Y., Chen, J.W., Weng, S.Y., Lin, Y.C., Shu, C.M., 2013. Effects of thermal runaway hazard for three organic peroxides conducted by acids and alkalines with DSC, VSP2, and TAM III. Thermochim. Acta 566, 226e232. Liu, Shang-Hao, Hou, Hung-Yi, Shu, Chi-Min, 2015. Thermal hazard evaluation of the
41
autocatalytic reaction of benzoyl peroxide using DSC and TAM III. Thermochim. Acta 605, 68e76. Marco, Eissen, Andreas, Zogg, Konrad, Hungerbuhler, 2003. The runaway scenario in the assessment of thermal safety: simple experimental access by means of the catalytic decomposition of H2O2. J. Loss Prev. Process Ind. 16, 289e296. Nanchen, A., Steinkrauss, M., Stoessel, F., 2009. Utilisation of the criticality classes within TRAS410. Forsch. Im. Ingenieurwes. 73, 3e10. National, F.P.A., 2011. NFPA 704, Standard System for the Identification of the Hazards of Materials for Emergency Response. National Fire Protection Association. Stoessel, Francis, 1993. What is your thermal risk? Chem. Eng. Prog. 10, 68e75. Stoessel, Francis, 2008. Thermal Safety of Chemical Processes: Risk Assessment and Process Design. Wiley. Sun, J.H., Li, X.R., Hasegawa, K., Liao, G.X., 2004. Thermal hazard evaluation of complex reactive substance using calorimeters and Dewar Vessel. J. Therm. Anal. Calorim. 76 (3), 883e893. Sun, Dao-Xing, Miao, Xiao, Xie, Chuan-Xin, Gu, Jing, Li, Rong, 2012. Study on thermal properties and kinetics of benzoyl peroxide by ARC and C80 methods. J. Therm. Anal. Calorim. 107, 943e948. Townsend, D.I., Tou, J.C., 1980. Thermal hazard evaluation by an accelerating rate calorimeter. Thermochim. Acta 37, 1e30. Wang, Tien-Szu, Liu, Shang-Hao, Qian, Xin-Ming, You, Mei-Li, Chou, Wei-Lung, Shu, Chi-Min, 2013. Isothermal hazards evaluation of benzoyl peroxide mixed with benzoic acid via TAM III test. J. Therm. Anal. Calorim. 113, 1625e1631. Wu, Sheng-Huang, Chi, Jen-Hao, Huang, Chun-Chin, Lin, Nung-Kai, Peng, Jiou-Jhu, Shu, Chi-Min, 2010. Thermal hazard analyses and incompatible reaction evaluation of hydrogen peroxide by DSC. J. Therm. Anal. Calorim. 102, 563e568. Zaman, F., Beezer, A.E., Mitchell, J.C., et al., 2001. The stability of benzoyl peroxide formulations determined from isothermal microcalorimetric studies. Int. J. Pharm. 225, 135e143. Zhao, Bao-Dong, Wan, Ping-Yu, Liu, Ji-Qiang, Wang, Jian-Guo, Wang, Tao, 2012. Investigation of thermal instability of benzoyl peroxide in the presence of carbazole and its derivatives. Thermochim. Acta 543, 232e238. Zogg, H.A., Zürich, V., 1987. “Zurich” Hazard Analysis: a Brief Introduction to the “Zurich” Method of Hazard Analysis. Zurich Insurance Group, Risk engineering.
Contents lists available at ScienceDirect
Journal of Loss Prevention in the Process Industries journal homepage: www.elsevier.com/locate/jlp
Thermal hazard analyses for the synthesis of benzoyl peroxide Yong Zhang, Lei Ni, Juncheng Jiang*, Jun Jiang, Wenxin Zhang, Jiajia Jiang, Mingguang Zhang Jiangsu Key Laboratory of Hazardous Chemicals Safety and Control, College of Safety Science and Engineering, Nanjing Tech University, Nanjing, 210009, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 5 January 2016 Received in revised form 17 April 2016 Accepted 17 April 2016 Available online 21 April 2016
Benzoyl peroxide (BPO), historically, due to its wide applications around the world, has caused many serious fire and explosion accidents. In this paper, in order to prevent such accidents, thermal hazard analyses of synthesis of benzoyl peroxide were studied. Firstly, in order to obtain thermal hazard coefficients, the exothermic processes with different alkaline solutions were studied by RC1e (Reaction Calorimeter). The alkaline solutions are NaOH, NH4HCO3 and Na2CO3, respectively. Secondly, the thermal decomposition of BPO product was studied by PHI-TECⅡto analysis the thermal stability. Finally, the possibility of runaway reactions and thermal risks of synthetic process were evaluated according to the Stoessel criticality diagram. In the first stage, the test results of the reaction heat (DHm), heat release rate (qr) and adiabatic temperature rise (DTad) with different alkaline solutions were NaOH > Na2CO3 > NH4HCO3. In the second stage, according to the analysis of experimental data, the heat release rate of reaction with NH4HCO3 solution was the slowest, while the Maximum Temperature of the Synthesis Reaction (MTSR) and the adiabatic temperature rise (DTad) were lowest when using Na2CO3 solution. The time needed to reach the maximum reaction rate under the adiabatic condition (TMRad) was 0.83 h when using NaOH solution. The temperature was 38.24 C when TMRad is 24 h (TD24). The evaluation results of the process showed that the risks of reactions with NaOH solution or NH4HCO3 solution were not acceptable. Only the risk of the reaction with Na2CO3 solution was acceptable. Therefore, the safety level of synthesis of benzoyl peroxide can be significantly improved by using Na2CO3 solution. Research in this paper can not only improve the safety level of BPO reaction and storage processes, but also provide technical support for stability criterion of BPO decomposition reaction. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Benzoyl peroxide Thermodynamic properties Thermal runaway Reaction kinetics Stability
1. Introduction Benzoyl peroxide (BPO) is a strong oxidizing organic peroxide and easy to decompose under heating. The thermal explosion can be caused by impact, heat, friction, etc (Zhao et al., 2012; Wang et al., 2013; Lee et al., 2014). BPO is widely used as initiator in chemical industries (Zaman et al., 2001). Based on the NFPA standard (National, 2011), the classification level of BPO for flammability and reactivity is 4. In the document of state administration of work safety of China, the synthesis of BPO is a typical high-risk process. In China, the traditional method for the synthesis of BPO is that
* Corresponding author. Mail box 13, No. 200 North Zhongshan Road, Nanjing Tech University, Nanjing 210009, China. E-mail addresses: [email protected] (L. Ni), [email protected] (J. Jiang), [email protected] (M. Zhang). http://dx.doi.org/10.1016/j.jlp.2016.04.007 0950-4230/© 2016 Elsevier Ltd. All rights reserved.
benzoyl chloride and hydrogen peroxide are used as main raw material with alkaline solution. Fan Juan (Fan, 2002) found that the decomposition rate of H2O2 and hydrolysis rate of BPO are rapid. Although low temperature environment can control the above two side reactions effectively, it can also decrease the main reaction rate. Compared with NaOH and Na2CO3, the process of NH4HCO3 dissolved in water is an endothermic process. Therefore, the reaction can keep the process temperature without cooling and the reaction time can be shortened (Duan et al., 2003). Most papers deal with the thermal decomposition of BPO. DaoXing Sun et al.(Sun et al., 2012) used C80 calorimetry and accelerating rate calorimetry (ARC) to study the hazardous characteristics of BPO and obtained the pre-exponential factor (3.61 1019 s1) and activation energy (152.80 kJ mol1). Liu et al. (2013) analyzed the thermal hazards of BPO with incompatibilities (H2SO4, NaOH, and Na2SO3) and found that thermal hazard increased prominently when BPO were mixed with H2SO4, NaOH, and Na2SO3. The thermal
36
Y. Zhang et al. / Journal of Loss Prevention in the Process Industries 43 (2016) 35e41
decomposition of BPP is an autocatalytic reaction. When the BPO was mixed with benzoic acid, benzene or phenol, the hazard significantly increased (Wang et al., 2013; Liu et al., 2015). Acrylic acid and methyl acrylate were identified as catalyst for thermal decomposition of BPO (Huang et al., 2013). In the first step of synthesis of BPO, the thermal hazard of H2O2 mixed with alkaline solutions increased significantly. Chen et al. (2006) used differential scanning calorimetry (DSC) to obtain that the initial decomposition temperature (Tonset) of 31mass% H2O2 was 40.54 C by heating rate of 2 C min1. Wu et al., (2010) found that the Tonset of H2O2 mixed with NaOH was close to room temperature. Many studies have reported the hazards of the decomposition of BPO, but few researches study the hazards of synthesis of BPO mixed with different alkaline solutions. The synthesis of BPO is an exothermic reaction. There is no detailed account of a safety assessment of the synthesis of BPO. In this study, we report the process safety evaluation of the synthesis of BPO mixed with different alkaline solutions. RC1e is used to study the exothermic characteristics for the synthesis of BPO with different alkaline solutions firstly. Secondly, the adiabatic decomposition of raw product is studied by PHI-TECⅡ and the TMRad and TD24 are calculated from the fitting kinetic models. Finally, the risk level of the synthesis of BPO is evaluated based on the Stoessel criticality diagram and the safest process of synthesis of BPO which also has a good yield of BPO can be determined.
2. Experiments 2.1. Reaction path There are two steps for the synthesis of Benzoyl peroxide. Firstly, Hydrogen peroxide (H2O2) is added to alkaline solution to prepare the hydrogen peroxide alkaline solution. Secondly, benzoyl chloride is added to hydrogen peroxide alkaline solution. See Fig. 1.
2.2.2. Reaction calorimeter (RC1e) (1) RC1e is manufactured by METTLER TOLEDO and used to measure the exothermic condition of reaction. RC1e is an automatic synthesis reaction system. A 2 L reactor is used in our experiment. Specific testing principle can be found in the literature (Lee et al., 2014). (2) Experiment conditions: reaction temperature is 10 C. Stirring speed is 100 rmin1. Feed rate is 13.59 g s1 (H2O2) and 13.52 g s1 (benzoyl chloride). Solution concentration is 3.93 mol L1. Alkali solutes are NaOH, NH4HCO3 and Na2CO3.
2.2.3. Nuclear magnetic resonance spectrometer (Bruker AV500 MHz NMR) (1) Bruker AV500 MHz NMR is used to analyze the molecular structure and determinate the material content. (2) Experiment reagent is Sample 2 and solvent is CDCl3.
2.2.4. PHI-TECⅡ (1) PHI-TECⅡ is a quasi-adiabatic calorimeter manufactured by HEL. PHI-TECⅡ is used to obtain the thermal hazard parameters, such as temperature versus time, pressure versus time, self-heating rate and pressure-rising rate. The heat-waitsearch (H-W-S) mode for detecting the self-heating rate is adopted for PHI-TEC Ⅱ (Townsend and Tou, 1980; Sun et al., 2004). The volume of sample ranges from 0.5 to 100 ml. The experimental equipment is shown in Fig. 2. (2) Experiment conditions: the mass of sample 1 is 1.01 g. The volume of test is 10 ml. The temperature range is 60e300 C. Heating step is 5 C. The criterion for no self-heating is 0.02 C min1. 3. The experiment results and analysis of RC1e 3.1. The experiment results of first step
2.2. Experiment reagent and instrument 2.2.1. Experiment reagent and sample (1) Reagent: H2O2 (AR 30 wt%), NaOH (AR), Na2CO3 (AR), NH4HCO3 (AR), benzoyl chloride (AR), sodium dodecyl sulfate (SDS, CP), Sinopharm Chemical Reagent Co., Ltd; deionized water. (2) Sample1 is the product of the synthesis of BPO which is produced by RC1e and then liquid-solid separation. Sample 2 is the product of sample1 being washed and dried.
Because both the adiabatic temperature rise (DTad) and the calculated MTSR are small in first step. Therefore, the worst conditions for the whole process corresponding to either reactions under adiabatic conditions or cooling system failure event are considered in this paper. Tmax ¼ DTad þ Tp. The exothermic
In order to study the adiabatic decomposition condition of BPO effectively, Sample 1 is used as the sample of adiabatic experiment. Sample 2 is used for product analysis.
First steps:
Na2O2 + 2H2O
2NaOH + H2O2 O C Cl + Na2O2
2 O
Second steps:
O
C O O C
Fig. 1. Flow chart of the synthesis process of BPO.
+ 2NaCl
Fig. 2. Schematic diagram of PHI-TECⅡ.
Y. Zhang et al. / Journal of Loss Prevention in the Process Industries 43 (2016) 35e41
37
parameters of different alkaline solution are shown in Table 1. As shown in Table 1 and Fig. 3, all the parameters (DHm, DTad, Tmax and heat release rate) are NaOH > Na2CO3 > NH4HCO3. The reaction with NH4HCO3 is an endothermic reaction and may be an inherent reaction. The reaction with NaOH or with Na2CO3 is exothermic reaction. All the parameters (DHm, DTad, Tmax and heat release rate) in the reaction with NaOH are larger than those of reaction with Na2CO3. The reason may be that NaOH is the strong alkali. Therefore, the reaction with NaOH solution generated the maximum heat. In Fig. 3, the reaction with NH4HCO3 is stable after 6000s and very time-consuming. The reaction times in different alkaline solutions are NaOH > Na2CO3 > NH4HCO3. Although the reaction with NH4HCO3 is an inherent reaction, it is very time-consuming and the production efficiency is very low. Fig. 3. Heat release rate with different alkaline solutions at the first stage.
3.2. The experiment results of second step The second step of the synthesis of BPO is heterogeneous reaction. Therefore, the surfactant (SDS) is added to the reaction and used to avoid caking. But excessive amounts of surfactant can't be used. Because the reactor may generate a lot of foamy and raw material may overflow the reactor. As shown in Fig. 4, the reactions with different alkaline solutions have different heat release rate: NaOH > Na2CO3>NH4HCO3. In order to study the exothermic reaction characteristics under different alkaline solutions, some parameters are considered in the experiment, such as the runaway severity (DTad) (Stoessel, 1993), the maximum temperature of reaction (Tcf) and the maximum temperature of reaction runaway (MTSR). In Table 2, all the parameters values (DHm, DTad, MTSR) are NaOH > NH4HCO3 >Na2CO3. When alkaline solution is NH4HCO3 solution and benzoyl chloride is added to the second step reaction, white gas is generated and reactants may decompose. The reactions with other alkaline solutions don't have the phenomenon. Because OH H2 O2 þ 2C6 C5 COCl then !ðC6 H5 COÞ2 O2 þ 2HCl, NaHCO3 þ HCl/NaCl þ H2 O þ CO2 [. Compared first step with second step, the reaction with the NH4HCO3 solution is an endothermic reaction in first step. But heat release in second step is larger than that of reaction with Na2CO3. Therefore, if we want to compare the risk degree comprehensively for the reactions with two kinds of alkaline solutions, the runaway possibility parameters (TMRad and TD24) should be considered (Marco et al., 2003). In Fig. 5 (a), (b) and (c), the degree of material accumulation (Xac) and Tcf (MTSR) increase to the maximum at the stoichiometric point and then decrease with the reaction time. At the stoichiometric point, the degrees of heat accumulation are NH4HCO3> NaOH > Na2CO3.
Fig. 4. Heat release rate with different alkaline solutions at the second stage.
Table 2 Heat release parameters with different alkaline solutions. Alkaline solutions
Molarity/mol L1
DHm/kJ mol1
DTad/ C
MTSR/ C
NaOH NH4HCO3 Na2CO3
3.93
131.64 74.86 66.98
77.11 51.48 35.76
68.05 54.49 31.07
calculated. See Table 3. In Table 3, the reaction with NH4HCO3 has the highest yield. The reason may be that the reaction spends too much time.
3.3. Product analysis
4. The adiabatic decomposition of sample 1
Sample 2 is obtained by filtering and drying sample 1. Sample 2 dissolves in CDCl3 and is characterized with hydrogen nuclear magnetic spectra. See Fig. 6. There are four peaks, BPO peak, CDCl3 peak, standard peak and water peak. Product yield can be
The test results of sample 1 under adiabatic condition are shown in Table 4 and Fig. 7. The initial decomposition temperature (Tonset) of sample 1 is 67.57 C. BPO decompose and release the gas and heat. Temperature and pressure increase rapidly. Maximum rate of temperature rise is 4598.3 C min1.
Table 1 The parameters of different alkaline solution. Alkaline solutions
DHm/kJ mol1
DTad/K
Tmax/ C
Tmax>25 C
NaOH NH4HCO3 Na2CO3
27.935 10.060 6.720
12.27 4.41 2.74
22.27 5.59 12.74
NO NO NO
4.1. Kinetic models In order to determine the thermokinetic parameters of BPO decomposition, a simple nth order reaction is assumed and the rate equation can be written in the form (Townsend and Tou, 1980):
38
Y. Zhang et al. / Journal of Loss Prevention in the Process Industries 43 (2016) 35e41
Fig. 5. MTSR test results and heat conversion with different alkaline solutions.
Fig. 6. 1HNMR spectrum of BPO in CDCl3.
Y. Zhang et al. / Journal of Loss Prevention in the Process Industries 43 (2016) 35e41
39
Number
Alkaline solutions
Productivity/%
(R2). Then, Ea ¼ 99.69 kJ mol1, A ¼ 6.12 1011 s1. In order to further evaluate the risk of reaction runaway, the time to reach the maximum reaction rate (q) should be calculated.
1 2 3
NaOH NH4HCO3 Na2CO3
0.8381 0.8478 0.8238
4.2. Time to reach the maximum reaction rate (q)
Table 3 Productivity of different alkaline solutions.
The equation of the time to reach the maximum reaction rate is shown as follow (Chen et al., 2012):
Table 4 Test data of sample 1 under adiabatic condition. Parameter
Sample1
F
4.78 67.57 199.24 131.67 3.967 4598.3
Decomposition temperature of self-heating/ C Maximum temperature/ C Adiabatic temperature rise/ C Maximum pressure/Mpa Maximum rate of temperature rise/ C min1
ln q ¼
Ea 1 InA R T
(4)
Submitting the Ea and A from above simulation to the Equation (4), the temperature TD24 (TD8) can be calculated whenq ¼ 24 h or 8 h. See Table 6. When the reaction temperature is higher than the MTSR, it may have the second reaction. The TMRad can also be obtained and shown in Table 7. In Table 6, the DTad of sample 1 is 629.38 C. Based on the Zurich hazard analysis (Zogg and Zürich, 1987), the reaction severity is high. The temperature in summer is easy to reach the TD24 (38.24 C). In Table 7, the TMRad of the reaction with Na2CO3 is higher than those of the reactions with other alkaline solutions. Base on the assessment criteria, the probability of a runaway reaction is almost impossible. 5. Safety assessment of the synthesis of benzoyl peroxide
Fig. 7. Temperature (pressure) histories of sample 1 under adiabatic decomposition condition.
Tf T n n1 T T n n1 dT Ea ¼ kDTad cA0 ¼ A exp DTad f cA0 dt DTad RT DTad (1) Then,
dT 1 DTad k¼ DT dt ad Tf T
!n c1n A0
(2)
Based on the experiment or calculated values (T, Tf, DTad and dT/ dt), if suitable reaction order is selected, lnk represents linear correlation to1/T. See Equation (3).
ln k ¼ ln A
Ea RT
(3)
where, dT/dt is temperature rising rate, C s1. Tf is the highest decomposition temperature, C. k is reaction rate constant. cA0 is the initial concentration of component A, mol L1. XA is conversion of component A. n is reaction order. A is pre-exponential factor. Ea is activation energy, J mol1. Based on Equations (2) And (3), experiment and fitting curves of Ink vs 1/T are shown in Fig. 8. n increases from 0.5 to 2.5 which the step is 0.5. To get more accurate results, n increases from 1.9 to 2.5 which the step is 0.1. The results are shown in Table 5. When n ¼ 2.2, the fitting results have the highest correlation coefficient
Base on the criticality classes of thermal runaway scenario (Nanchen et al., 2009), the critical temperatures are summarized in Table 8. In the reactions with NaOH or NH4HCO3 alkaline solutions, Tr < TD24 < MTSR < MTT. The risk level is 5 and the risk is not acceptable. After loss of control of the synthesis reaction, the decomposition reaction will be triggered (MTSR > TD24) and the technical limit will be reached during the runaway of the secondary reaction. The decomposition of BPO could lead to critical pressure increase. There is no safety barrier between the main and secondary reaction. Only quenching or dumping can be used (Stoessel, 2008). In the reactions with Na2CO3, Tr < MTSR < TD24 < MTT. The risk level is 2 and the risk is low. But, if the reaction mass is maintained for a longer time under heat accumulation conditions, the decomposition reaction could be triggered and reach the MTT. The BPO decomposition releases heat rapidly. When the reaction temperature reaches the boiling point, the system pressure will rise quickly due to the generation of a large amount of steam. The reactants may spill from the reactors and even explode. Therefore, the heat accumulation needs to be avoided and no special safety measures are required. From the evaluation result, although the reaction with NH4HCO3 is endothermic reaction in the first step, the severity and probability of runaway reaction in second step are higher than that of reaction with Na2CO3. Therefore, reaction with Na2CO3 has the lowest risk level among the reactions with three alkaline solutions. 6. Conclusions In this paper, a detailed risk assessment — encompassing the determination of the severity and the probability of the synthesis of BPO with different alkaline solutions — has been studied. The main findings resulted in the following: (1) In the first step of the synthesis of BPO, DHm, DTad and heat release rate are NaOH > Na2CO3 > NH4HCO3. The reaction
40
Y. Zhang et al. / Journal of Loss Prevention in the Process Industries 43 (2016) 35e41
4
15 Experiment(n =2.2) Simulation
2 2
R =0.96851 0
5
-2
lnK
lnK
n=0.5 n=1 n=1.5 n=2 n=2.5
10
0
-4 -5
-6
-10
-8 -10
0.0022
0.0024
0.0026
0.0028
0.0030
-15 0.0020
0.0022
0.0024
1/T
0.0026
0.0028
0.0030
1/T Fig. 8. Calculated and fitting curves of Ink vs 1/T of sample 1.
Table 5 Kinetic parameters of sample 1 with different reaction order. n
1.9
2
2.1
2.2
2.3
2.4
2.5
R2 Ea A
0.96775 92.57 5.37 1010
0.96818 94.95 1.21 1011
0.96842 97.32 2.72 1011
0.96851 99.69 6.12 1011
0.96845 102.06 1.38 1012
0.96827 104.43 3.10 1012
0.96799 106.80 6.97 1012
Table 6 Thermal decomposition characteristic data of samples modified by thermal inertia factor. Sample 1
DTad,sample/ C
629.38 38.24 47.38
TD24/ C TD8/ C
(5) Base on the Stoessel criticality diagram, only the risk level of reaction with Na2CO3 is level 2. Therefore, among the three kinds of alkaline solutions, the thermal hazard for the synthesis of BPO with Na2CO3 alkaline solution is lowest. Further, vent-sizing experiments can be performed for the synthesis of BPO with alkaline solution to prevent over pressurization because of the produced gas.
Table 7 TMRad of different alkaline solutions.
Acknowledgments
Alkaline solutions
MTSR/ C
TMRad/h
Extended assessment criteria
NaOH NH4HCO3 Na2CO3
68.05 54.49 31.07
0.83 3.55 59.45
frequent probable remote
Table 8 Risk classification with different alkaline solutions. Alkaline solutions
Tr/ C
MTSR/ C
TD24/ C
MTT/ C
The risk level
NaOH NH4HCO3 Na2CO3
10
68.05 54.49 31.07
38.24
100
5 5 2
with NH4HCO3 is an endothermic reaction and may be an inherent reaction. (2) In the second step of the synthesis of BPO, DHm, DTad and heat release rate are NaOH > NH4HCO3> Na2CO3. The reaction with NH4HCO3 has highest heat accumulation. (3) Based on the experiment values of PHI-TECⅡ, Ea of the sample 1 is 99.69 kJ mol1, A is 6.12 1011 s1 and TD24 of BPO is 38.24 C. (4) TMRad of different alkaline solutions is NaOH < NH4HCO3 < Na2CO3. The reaction with NaOH has the minimum TMRad and the maximum runaway possibility.
The authors are grateful for the support given by key project of National Natural Science Foundation of China under Grant No. 21436006, National Natural Science Foundation of China under Grant No. 51176070, Jiangsu Natural Science Foundation of China (BK20141457, SBK2015043673), Project of the Natural Science Foundation of Jiangsu Province, China “Analyzing model and optimizing technology for safety capacity of a chemical industrial park” (BK 2012824), Jiangsu Province “333 high-level talents project” (BRA2015313). Nomenclature A cA0 Ea dT/dt k MTSR MTT n TMRad Tcf TD8 TD24 Tf
pre-exponential factor the initial concentration of component A, mol L1 activation energy, J mol1 temperature rising rate, C s1 rate constant the Maximum Temperature of the Synthesis Reaction, C Maximum temperature for technical reasons, C reaction order the time needed to reach the maximum reaction rate under the adiabatic condition, s the maximum temperature of reaction runaway, C the temperature when TMRad is 8 h, C the temperature when TMRad is 24 h, C the highest decomposition temperature, C
Y. Zhang et al. / Journal of Loss Prevention in the Process Industries 43 (2016) 35e41
the maximum temperature of reaction, C normal process temperature, C the conversion rate of component A F thermal inertia factor q the time to reach the maximum reaction rate DHm molar reaction heat, kJ mol1 DTad adiabatic temperature rise, C DTad,sample the revised adiabatic temperature rise, C
Tmax Tr XA
References Chen, K.Y., Lin, C.M., Shu, C.M., Kao, C.S., 2006. An evaluation on thermokinetic parameters for hydrogen peroxide at various concentrations by DSC. J. Therm. Anal. Calorim. 85, 87e89. Chen, Li-Ping, Liu, Ting-Ting, Yang, Qian, Chen, Wang-Hua, 2012. Thermal hazard evaluation for iso-octanol nitration with mixed acid. J. Loss Prev. Process Ind. 25, 631e635. Duan, Haibao, Cui, Chunwei, Cai, Guocheng, Zhang, Rui, Peng, Qijun, 2003. New method for synthesis of benzoyl peroxide. Chem. Ind. Eng. Prog. (China) 22 (3), 293e294. Fan, Juan, 2002. Discussion about preparation of benzoyl peroxide. Appl. Chem. Ind. (China) 31 (6), 1e2. Huang, Yi-Hao, I, Yet-Pole, Chen, Nan-Chi, Wu, Sheng-Hung, Horng, Jao-Jia, Wu, Yung-Tin, Wen, I-Jyh, 2013. Thermal runaway reaction evaluation of benzoyl peroxide using calorimetric approaches. J. Therm. Anal. Calorim. 113, 595e598. Lee, Ming-Hsun, Chen, Jiann-Rong, Shiue, Gong-Yih, Lin, Yan-Fu, Shu, Chi-Min, 2014. Simulation approach to benzoyl peroxide decomposition kinetics by thermal calorimetric technique. J. Taiwan Inst. Chem. Eng. 45, 115e120. Liu, S.H., Hou, H.Y., Chen, J.W., Weng, S.Y., Lin, Y.C., Shu, C.M., 2013. Effects of thermal runaway hazard for three organic peroxides conducted by acids and alkalines with DSC, VSP2, and TAM III. Thermochim. Acta 566, 226e232. Liu, Shang-Hao, Hou, Hung-Yi, Shu, Chi-Min, 2015. Thermal hazard evaluation of the
41
autocatalytic reaction of benzoyl peroxide using DSC and TAM III. Thermochim. Acta 605, 68e76. Marco, Eissen, Andreas, Zogg, Konrad, Hungerbuhler, 2003. The runaway scenario in the assessment of thermal safety: simple experimental access by means of the catalytic decomposition of H2O2. J. Loss Prev. Process Ind. 16, 289e296. Nanchen, A., Steinkrauss, M., Stoessel, F., 2009. Utilisation of the criticality classes within TRAS410. Forsch. Im. Ingenieurwes. 73, 3e10. National, F.P.A., 2011. NFPA 704, Standard System for the Identification of the Hazards of Materials for Emergency Response. National Fire Protection Association. Stoessel, Francis, 1993. What is your thermal risk? Chem. Eng. Prog. 10, 68e75. Stoessel, Francis, 2008. Thermal Safety of Chemical Processes: Risk Assessment and Process Design. Wiley. Sun, J.H., Li, X.R., Hasegawa, K., Liao, G.X., 2004. Thermal hazard evaluation of complex reactive substance using calorimeters and Dewar Vessel. J. Therm. Anal. Calorim. 76 (3), 883e893. Sun, Dao-Xing, Miao, Xiao, Xie, Chuan-Xin, Gu, Jing, Li, Rong, 2012. Study on thermal properties and kinetics of benzoyl peroxide by ARC and C80 methods. J. Therm. Anal. Calorim. 107, 943e948. Townsend, D.I., Tou, J.C., 1980. Thermal hazard evaluation by an accelerating rate calorimeter. Thermochim. Acta 37, 1e30. Wang, Tien-Szu, Liu, Shang-Hao, Qian, Xin-Ming, You, Mei-Li, Chou, Wei-Lung, Shu, Chi-Min, 2013. Isothermal hazards evaluation of benzoyl peroxide mixed with benzoic acid via TAM III test. J. Therm. Anal. Calorim. 113, 1625e1631. Wu, Sheng-Huang, Chi, Jen-Hao, Huang, Chun-Chin, Lin, Nung-Kai, Peng, Jiou-Jhu, Shu, Chi-Min, 2010. Thermal hazard analyses and incompatible reaction evaluation of hydrogen peroxide by DSC. J. Therm. Anal. Calorim. 102, 563e568. Zaman, F., Beezer, A.E., Mitchell, J.C., et al., 2001. The stability of benzoyl peroxide formulations determined from isothermal microcalorimetric studies. Int. J. Pharm. 225, 135e143. Zhao, Bao-Dong, Wan, Ping-Yu, Liu, Ji-Qiang, Wang, Jian-Guo, Wang, Tao, 2012. Investigation of thermal instability of benzoyl peroxide in the presence of carbazole and its derivatives. Thermochim. Acta 543, 232e238. Zogg, H.A., Zürich, V., 1987. “Zurich” Hazard Analysis: a Brief Introduction to the “Zurich” Method of Hazard Analysis. Zurich Insurance Group, Risk engineering.