Explosion characteristics of mixtures containing

Process Safety and Environmental Protection 119 (2018) 218–222

Contents lists available at ScienceDirect

Process Safety and Environmental Protection journal homepage: www.elsevier.com/locate/psep

Explosion characteristics of mixtures containing hydrogen peroxide and working solution in the anthraquinone route to hydrogen peroxide Xuewu Jia a , Feng Sun a,b , Yi Fei a , Manping Jin a , Fan Zhang a,∗ , Wei Xu a , Ning Shi a , Zhiguo Lv b a b

State Key Laboratory of Safety and Control for Chemicals, SINOPEC Research Institute of Safety Engineering, Qingdao 266071, China College of Chemical Engineering, Qingdao University of Science and Technology, Qingdao 266042, China

a r t i c l e

i n f o

Article history: Received 28 February 2018 Received in revised form 9 August 2018 Accepted 9 August 2018 Available online 11 August 2018 Keywords: Hydrogen peroxide Anthraquinone process Explosive property

a b s t r a c t The explosive properties of mixtures of aqueous hydrogen peroxide (H2 O2 ) and working solution (WS), and the main components of the working solution (1, 3, 5-trimethyl benzene(TMB), trioctylphosphate (TOP)) were investigated by the drop weight test. The explosion range was interpreted by thermal calculation, and the calculated results agreed well with the experimental test. The explosion mechanism of the TMB/H2 O2 mixture is the partial oxidation of TMB by H2 O2 , which is qualitatively obtained by analyzing the gaseous and liquid products of the TMB/ H2 O2 mixture after explosion. Finally, a proposed explosion mechanism was suggested. © 2018 Published by Elsevier B.V. on behalf of The Institution of Chemical Engineers.

1. Introduction Hydrogen peroxide (H2 O2 ) is a green inorganic chemical product, the decomposition product of which is water. It is widely used in various areas of industry, such as chemical synthesis, bleaching of pulp, and environmental protection (Hage and Lienke, ´ 2005; Bilinska et al., 2017; Xiang et al., 2010; Corma et al., 2001; Malakootian and Moridi, 2017; Khan et al., 2015). H2 O2 is used in the paper pulping and water treatment industries where it is being used as an environmentally friendly alternative to chlorine, the use of which as a bleaching agent/disinfectant has been prohibited by law in some parts of the world. With the increasing awareness of environmental protection, H2 O2 gradually replaces some of the chemicals that pollute the environment. As such the demand of H2 O2 is increasing rapidly. Specially, H2 O2 with high concentration (50–70 wt.%) is used more widely because of its high oxidation capacity (Lu et al., 2018). The main industrial process currently used for the production of H2 O2 is the anthraquinone process. 2-Ethylanthraquinone (EAQ) is dissolved in a solvent mixture of C9–C10 aromatics (the main component is 1, 3, 5-trimethyl benzene (TMB)) and trioctylphosphate (TOP) to form a working solution (WS) (Chen, 2008).The WS is always used because the anthraquinone (AQ) and anthrahydro-

∗ Corresponding author.

quinone (AHQ) have different solubility. AQ dissolves in nonpolar, aromatic solvents, and AHQ dissolves well in polar solvents (TOP). The primary reactions are represented in Fig. 1, where the AQ is hydrogenated into AHQ, and AHQ is then oxidized into H2 O2 and AQ. The produced H2 O2 is usually extracted from the WS with water. The WS separated from H2 O2 is recycled to the reduction process. Thus, the anthraquinone process efficiently produces H2 O2 from H2 and O2 in very high yield. The concentration of H2 O2 extracted from the oxidized working solution, using demineralized water in liquid-liquid sieve tray columns, is usually 27.5 wt.% to 35 wt.%. It is more energy efficient to raise the concentration of H2 O2 to 50∼70 wt.% during extraction. However, the higher the concentration, the more H2 O2 becomes susceptible for forming explosive mixtures (Goor et al., 2007; Othmer, 2018). It is well documented that there is a H2 O2 -organic-water explosion zone upon extraction of H2 O2 from the organic working solution using water. The concentration of H2 O2 should not exceed a maximum of 44 wt.% (Othmer, 2018). The limit of this explosion zone depends on the nature of the organic material (in this case, the components of the WS) and the operating conditions. With the progress of the technology, the composition of the WS has changed and different factories use different WS. So it is necessary to develop a method to test the explosion range of the liquid system and investigate the mechanism of the explosion to ensure the safety of the process.

https://doi.org/10.1016/j.psep.2018.08.007 0957-5820/© 2018 Published by Elsevier B.V. on behalf of The Institution of Chemical Engineers.

X. Jia et al. / Process Safety and Environmental Protection 119 (2018) 218–222

219

Fig. 1. The scheme of the anthraquinone process.

In open literature, the explosion ranges of the H2 O2 /organic mixtures are mainly restricted to the organics that are soluble to H2 O2 (Bretschger and Shanley, 1947; Shanley and Greenspan, 1957; Shanley and Perrin, 1958; Monger et al., 1961; Schreck et al., 2004; Chi et al., 2012). Schreck et al. (2004) investigated the thermal explosion hazard of mixtures of 2-propanol (2-PropOH), 2-methyl-2-propanol (TBA), 2-methyl-2-butanol (TAA) and 2methyl-2-pentanol (THA) with aqueous H2 O2 in various ratios by heating the mixtures under confine mentor by exposing them to a shock wave. They compared the range of the explosion limits of different organics using different experimental methods. Chi et al. (2012) investigated the influence of acetone on the thermal stability of H2 O2 by Differential Scanning Calorimetry (DSC) and Vent Size Package 2(VSP2). Details of the explosive properties of mixtures of aqueous H2 O2 and insoluble organics are hardly known. To determine the explosive properties of such types of dangerous mixtures, various methods are described. Frequently used test methods are the drop weight test (Schreck et al., 2004), the BAM 50/60 test (Koenen et al., 1961), the TNO 50/70 steel tube test (Groothuizen et al., 1974), and the USA GAP test (Mason and Aiken, 1972; Conner, 1974). In this work, the explosion range of the WS/H2 O2 , TMB/H2 O2 and TOP/H2 O2 mixtures was determined by the drop weight test (Recommendations on the transport of dangerous goods, 2009). To investigate the explosion mechanism of the TMB/H2 O2 mixtures, the explosion experiment was conducted in a closed bomb test. A reaction mechanism was proposed based on the gaseous and liquid products formed during this test.

2. Test procedure

Fig. 2. The drop weight test device.

2.1. Drop weight test

2.2. Closed bomb test

In the industrial process, the organic phase and the water phase are mixed well in an extraction column, but in the sample cell in the drop weight test, the organic phase and the water phase are layered. In order to simulate the situation in the working conditions, a surfactant (Sodium dodecylbenzenesulphonate, SDBS) was added to the WS/H2 O2 mixtures to enable a good contact between the two phases, and the results were compared with the system without surfactant. The amount of surfactant was so little (0.1 wt.%) that the reaction of the surfactant with H2 O2 can be ignored. The drop weight test was carried out with a falling weight machine (UN Test A.1) as shown in Fig. 2 (Recommendations on the transport of dangerous goods, 2009).40 mm3 of the sample was confined in a steel cylinder beneath a piston. The freely falling 5 kg weight was dropped from a height of one meter directly on the piston. The effect was judged by the sound, which was a very loud noise upon explosion produced by the sample. On a certain given composition, take the first test. If the result is positive, finish the experiment, and the result is taken as positive. If the result is negative, the test is continued until a positive result arises, and the result is taken as positive. If the result of the fourth test is still negative, finish the experiment, and the result is taken as negative.

The sample amount of the drop weight test is so small (40 mm3 ) that the amounts of gaseous and liquid products are not sufficient for analysis. In order to collect and analyze the gas and liquid products, the closed bomb test is (Recommendations on the transport of dangerous goods, 2009) carried out. The closed bomb test device was shown in Fig. 3. The apparatus consisted of a cylindrical steel pressure vessel of 89 mm in length, 60 mm in external diameter and 50 mm in internal diameter. Two flats are machined on opposite sides, and the firing plug was fitted on the bottom plug while the bursting disc was fitted on the above plug. The mixtures were ignited under confinement. For liquid samples, the ignition system consisted of a 25 cm long Ni/Cr wire with a diameter 0.6 mm and a resistance of 3.85/m. The wire was wound, using a 5 mm diameter rod, in the shape of a coil and was attached to the firing plug electrodes. The distance between the bottom of the vessel and the underside of the ignition coil should be 20 mm. The wire was heated by a constant power supply able to deliver 10 A. The TMB, TOP, EAQ and SDBS used in the experiment had a purity of 99.7% and were purchased from the Aladdin Industrial Corporation (Shanghai, China), and high-purity water with an electrical

220


Fig. 3. The closed bomb test device. Fig. 5. Explosion range of TMB/H2 O2 mixture (black dotted line: the stoichiometric line of combustion of H2O2 and the WS; blue dotted line: estimated explosion range) (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

Fig. 4. Explosion range of the WS/H2 O2 mixtures. (䊉: with surfactant, : without surfactant;dotted line: the stoichiometric line of combustion of H2 O2 and WS).

conductivity of 0.267 ␮S cm−1 was supplied by the ultrapure water machine (Item No. ULPQX-I, ULUPURE Ultrapure Technology Co. LTD, Sichuan, China).

Fig. 6. Explosion range of TOP/H2 O2 mixture (dotted line: the stoichiometric line of combustion of H2 O2 and the WS).

3. Experimental results The explosion ranges of the WS/H2 O2 , TMB/H2 O2 and TOP/H2 O2 mixture investigated by the drop weight test were shown in Figs. 4–6, respectively. The shape of the explosion range obtained from the explosion tests was as expected. All authors (Bretschger and Shanley, 1947; Shanley and Greenspan, 1957; Shanley and Perrin, 1958; Monger et al., 1961; Schreck et al., 2004) published the point of intersection (POI) of the upper and lower explosion limits on or nearby the stoichiometric line of combustion of H2 O2 and the organic material in question. Using the drop weight test, a POI was found at 42.7 wt.% H2 O2 , 7.3 wt.% WS and 50 wt.% water (without surfactant), at 37.2 wt.% H2 O2 , 6.9 wt.% WS and 56.9 wt.% water (with surfactant), at 40.0 wt.% H2 O2 , 5.2 wt.% TMB and 54.8 wt.% water and at 39.4 wt.% H2 O2 , 5 wt.% TOP and 55.6 wt.% water. In Fig. 4, the explosion range of the WS/H2 O2 mixture is the shape of a triangle, and the upper and lower explosion limits meet at the stoichiometric line of complete combustion, which is similar to that of the mixtures of H2 O2 with soluble organics (Schreck et al., 2004).The explosion range of the WS/H2 O2 mixture with surfactant was substantially larger than that without surfactant. The addition of surfactant made the contact area of the mixtures much larger, which made the reaction more violent. In the experiment without surfactant, the water phase and the organic phase are lay-

ered. The minimum content of H2 O2 in the explosion range of the WS/H2 O2 mixture was 42.7 wt.%. During the extraction of H2 O2 in the anthraquinone process, it is possible to reach 45 wt.%, so the mixtures at the bottom of the extraction column may enter the explosion range of the WS/H2 O2 mixture. This limit should be considered carefully when the concentration of H2 O2 would be increased during extraction. To investigate the influence of the major components of the WS on the explosion characteristics of the mixture, the explosion range of TMB/H2 O2 and TOP/H2 O2 mixtures was investigated, see Figs. 5 and 6. The shape of the explosion ranges was similar with each other. But the detailed explosion ranges were not same. There is a strong dependence of the explosive properties of the investigated mixtures on the chemical structure of the organics. The POI for these mixtures were at 40.0 wt.% H2 O2 , 5.2 wt.% TMB and 54.8 wt.% water and at 39.4 wt.% H2 O2 , 5.0 wt.% TOP and 55.6 wt.% water. The explosion range of the TOP/H2 O2 mixture was larger than that of the TMB/H2 O2 mixture.

4. Thermal calculation The mixture of TMB/H2 O2 was chosen as the example to investigate the relationship between the heat of reaction and the explosion


221

Table 1 The gas phase products. No.

Component (H2 O2 :H2 O:TMB)

O2/%

CO/ppm

combustible gas /%LEL

1 2 3

35%:35%:30% 45%:45%:10% 63%:27%:10%

24 30 33

900 1100 650

3 6 4

range. The reaction heat was calculated based upon the following scheme H2 O2 → H2 O(l) + 0.5O2 (g)H = −98.4 kJ/mol(ShanleyandPerrin, 1958)

(1)

C9 H12 +12O2 → 9CO2 +6H2 O(l)H = −5256 kJ/mol(Pedleyet al., 1986)

(2)

This scheme is thermally identical with direct reaction between organics and H2 O2 . The stoichiometric line of combustion of H2 O2 and the organic means that the oxidant and the organic react complete exactly right. For the points below the stoichiometric line, it is an oxidant-rich mixture and the remaining H2 O2 was assumed to decompose completely. For the points above the stoichiometric line, it is an organic-rich mixture and the remaining organic was assumed to remain unchanged (Shanley and Perrin, 1958). Based on the scheme, the heat of decomposition and heat of reaction, respectively, were calculated within the triangular chart representing mixtures of H2 O2 , water and TMB in Fig. 5. The points with the same energy were drawn in the chart. An enthalpy value was selected by the least square method between the isoenthalpy line and the experimental explosion range. From Fig. 5, we can see that the isoenthalpy envelope and the experimentally determined explosion range agreed well. For the TMB/ H2 O2 mixture tested, this enthalpy value was 3.86 kJ/g (the blue dotted line on Fig. 5), which is in the range of the literature (Shanley and Perrin, 1958) (3.35∼5.02 kJ /g). The envelope had a single discontinuity, located at a composition of zero oxygen balance to CO2 and H2 O. This is in agreement with the shape of the sensitivity area as determined experimentally.

Fig. 7. The total ion chromatogram (TIC) of main liquid phase products.

Table 2 The main liquid products analyzed by GC–MS. Sequence of the peak

Time/min

Substance

Chemical name

1

15.106

CH3 COOH

Acetic acid

2

15.794

3

16.088

4

16.997

3- Methyl benzaldehyde

5

18.222

3,5-Dimethylbenzaldehyde

6

18.336

Phenol

7

21.732

3,5-Dimethylphenol

8

24.884

m-Cresole

Benzaldehyde CH3 CH2 COOH

Propionic acid

5. Reaction mechanism To investigate the explosion mechanism of the TMB/H2 O2 mixture, the gas and liquid products were analyzed using a gas detector (New-Cosmos, XP-302M-A, Japan) and a Gas Chromatography-Mass Spectrometer (GC–MS).The TMB/H2 O2 mixture (the ratio of TMB: H2 O2 : H2 O = 10: 45: 45) was investigated in the closed bomb. This ratio is in the explosion zone in Fig. 5. The bursting disc burst after ignition. The gas products were detected, and the results were shown in Table 1. The presence of O2 -enriched gas (24%) means that H2 O2 decomposed. The production of CO and combustible gas (which can be oxidized on the surface of platinum wire) means that H2 O2 reacted with TMB. The liquid products after the reaction of TMB and H2 O2 wasanalyzed by GC–MS and a total ion chromatogram (TIC) is shown in Fig. 7. The main liquid products are mentioned in Table 2. Based on the experimental results of the gaseous and liquid products, a proposed mechanism for the oxidation of TMB and H2 O2 is shown in Fig. 8. The liquid phase products were complicated which could be seen in Fig. 7. Based on the gaseous and liquid results, the explosion mechanism of the TMB/H2 O2 system was supposed to be a partial oxidation reaction of H2 O2 and TMB. The partial oxidation took place in a short time, and a large amount of energy released instantaneously which caused a chemical explosion. To investigate the partial oxidation reaction mechanism of the TMB/H2 O2 system, we focused on the products with benzene ring. First, 2,4Dimethylbenzaldehyde was found in the liquid products, which meant that the methyl of TMB can be oxidized by H2 O2 (path 1); in the meantime, 2,4-dimethylphenol was found, which meant that

Fig. 8. A proposed mechanism for the reaction of TMB and H2 O2.

222


the C C bond between the benzene ring and methyl broke in the high energy of explosion which has been proved theoretically (path 2) (Chenoweth et al., 2008). Then, the free radical can react with the hydroxyl radicals to terminate the reaction or react with H2 O2 or H2 O to be further oxidized (path 3). Many homologues of benzaldehyde and phenol containing one methyl or none methyl were found in the liquid products. The formation path of the products could be assumed as follows: the free radical was formed by broking the C C bond between the benzene ring and the other methyl, and the free radical abstracted the hydrogen atom of H2 O2 or H2 O to form the homologues. Large amount of energy was released in the oxidation reaction, which induced the explosion. The results indicate that H2 O2 and the free radicals generated in the decomposition of H2 O2 are the oxidant in some steps of proposed mechanism above. Thus, the explosion of the TMB/H2 O2 system is the oxidation of the TMB by H2 O2 , and this is consistent with the experimental result (Table 2). 6. Conclusions An experimental method has been described for investigating the explosion range of the systems containing H2 O2 , water and the WS or the main components of the WS (1, 3, 5-trimethyl benzene (TMB), trioctylphosphate (TOP)). The results obtained in this investigation indicate that explosion may occur in the aqueous H2 O2 and insoluble organics systems when the concentration of H2 O2 is above a certain value. It has been shown that the isoenthalpy line of 3.86 kJ/g of mixture is almost identical with the experimental limit of sensitivity to drop weight test in the TMB/H2 O2 system. Moreover, based on the gaseous and liquid results, the explosion mechanism of the TMB/ H2 O2 system was supposed to be a partial oxidation reaction of H2 O2 and TMB. Acknowledgment The author acknowledge the national key research and development program of China for supporting this research under Grant 2016YFB0301701. References ´ Bilinska, L., Gmurek, M., Ledakowicz, S., 2017. Textile wastewater treatment by AOPs for brine reuse. Process Safety Environ. Prot. 109, 420–428. Bretschger, M.E., Shanley, E.S., 1947. Concentrated Hydrogen Peroxide, in.

Chen, Q., 2008. Development of an anthraquinone process for the production of hydrogen peroxide in a trickle bed reactor—from bench scale to industrial scale. Chem. Eng. Process. Process. Intensif. 47, 787–792. Chenoweth, K., van Duin, A.C.T., Goddard III, W.A., 2008. ReaxFF reactive force field for molecular dynamics simulations of HydrocarbonOxidation. J. Phys. Chem. A 112, 1040–1053. Chi, J.H., Wu, S.H., Charpentier, J.C., Yet-Pole, I., Shu, C.M., 2012. Thermal hazard accident investigation of hydrogen peroxide mixing with propanone employing calorimetric approaches. J. Loss Prev. Process Ind. 25, 142–147. Conner, J., 1974. Explosion risks of unstable substances, test methods employed by EM 2 (home office) RARDE. Proceedings First International Loss Prevention Symposium. Corma, A., Nemeth, L.T., Renz, M., Valencia, S., 2001. Sn-zeolite beta as a heterogeneous chemoselective catalyst for Baeyer-Villiger oxidations. Nature 412, 423–425. Goor, G., Glenneberg, J., Jacobi, S., 2007. Hydrogen peroxide. In: Ullmann’s Encyclopedia of Industrial Chemistry. Groothuizen, T.M., Hartgerink, J.W., Pasman, H.J., 1974. Phenomenology, Test Methods, and Case Histories of Explosions in Liquids and Solids, Loss Prev Safety Prom. Process Ind. 239. Hage, R., Lienke, A., 2005. Applications of transition-metal catalysts to textile and wood-pulp bleaching. Angew. Chemie Int. Ed. 45, 206–222. Khan, W., Najeeb, I., Tuiyebayeva, M., Makhtayeva, Z., 2015. Refinery wastewater degradation with titanium dioxide, zinc oxide, and hydrogen peroxide in a photocatalytic reactor. Process. Saf. Environ. Prot. 94, 479–486. Koenen, H., Ide, K.H., Swart, K.H., 1961. Sicherheitstechnische Kenndaten Explosionsfahigen Stoffe (Safety technical characteristic properties of explosive substances). Explosivestoffe 9, 30. Lu, X., Xu, H., Yan, J., Zhou, W.J., Liebens, A., Wu, P., 2018. One-pot synthesis of ethylene glycol by oxidative hydration of ethylene with hydrogen peroxide over titanosilicate catalysts. J. Catal. 358, 89–99. Malakootian, M., Moridi, A., 2017. Efficiency of electro-Fenton process in removing Acid Red 18 dye from aqueous solutions. Process. Saf. Environ. Prot. 111, 138–147. Mason, M.C., Aiken, E.G., 1972. Methods of evaluating explosives and hazardous materials. U.S. Bureau Mines JC 8541. Monger, J.M., Sello, Harry, Lehwalder, D.C., 1961. Explosion Limits of Liquid Systems Containing Hydrogen Peroxide, Water, and Oxygenated Organic Compounds, vol. 6., pp. 23–27. Othmer, K., 2018. Hydrogen peroxide. In: Encyclopedia of Chemical Technology. John Wiley & Sons, Inc. Pedley, J.B., Naylor, R.D., Kirby, S.P., 1986. Thermochemical Data of Organic Compounds[M]. Chapman and Hall. Recommendations on the transport of dangerous goods, 2009. Manual of Tests and Criteria, United Nations. Schreck, A., Knorr, A., Wehrstedt, K.D., Wandrey, P.A., Gmeinwieser, T., Steinbach, J., 2004. Investigation of the explosive hazard of mixtures containing hydrogen peroxide and different alcohols. J. Hazard. Mater. 108, 1–7. Shanley, E.S., Greenspan, F.P., 1957. Highly concentrated hydrogen peroxide physical and chemical properties. Ind. Eng. Chem. 39, 1536–1543. Shanley, E.S., Perrin, J.R., 1958. Prediction of the explosive behavior of mixtures containing hydrogen peroxide. Jet Propul., 382–385. Xiang, Y., Zhang, W., Hao, Z., 2010. Synergetic decolorization and deodorization of sludge protein foaming solution by 60Co ␥-ray irradiation/H2O2 oxidation. Process. Saf. Environ. Prot. 88, 285–291.