Absorption kinetics of ozone in water with ultrasonic radiation

Ultrasonics Sonochemistry 14 (2007) 552–556 www.elsevier.com/locate/ultsonch

Absorption kinetics of ozone in water with ultrasonic radiation Hui Zhang a

a,*

, Lijie Duan b, Daobin Zhang

a

Department of Environmental Engineering, Wuhan University, P.O. Box C319, Luoyu Road 129#, Wuhan 430079, China b Department of Environmental Science, Wuhan University, P.O. Box C319, Luoyu Road 129#, Wuhan 430079, China Received 7 April 2006; received in revised form 1 September 2006; accepted 8 September 2006 Available online 3 November 2006

Abstract A mathematical model was proposed to depict classical unsteady state method that was used to determine volumetric mass transfer coefficient of ozone from gaseous phase to aqueous phase during sonolysis. The rate constant of ozone self-decomposition with ultrasonic radiation, which was one of the parameters in the model, was determined with separate experiments. The results showed that self-decomposition rate constants of ozone were enhanced by ultrasound. The self-decomposition rate constant of ozone is linearly dependent on ultrasonic power, but the increase of the decomposition rate could not enhance ozone mass transfer coefficient. The volumetric mass transfer coefficients of ozone were also enhanced by ultrasonic radiation, while ultrasonic power had little effect on volumetric mass transfer coefficient of ozone. The degassing effect of ozone due to ultrasonic radiation was insignificant in the sparged system when ozone was bubbled during sonolysis.  2006 Elsevier B.V. All rights reserved. Keywords: Ultrasound; Ozone; Self-decomposition; Absorption; Kinetics

1. Introduction Since Dahi reported disinfection of water by means of ultrasound and ozone (US/O3) process in 1976 [1], it has been used for the destruction of aqueous pollutants such as trinitrotoluene (TNT) and cyclotrimethylene-trinitramine (RDX) [2], humic substances [3,4], 4-nitrophenol [5], carbon tetrachloride [6], methyl tert-butyl ether [7,8], cyclohexene [9], nitrobenzene and 4-chlorophenol [10], pentachlorophenol [11], textile dyes [12–18], and tetraphenyl porphine tetrasulfonic acid [19]. In the combined US/O3 system, ozone is decomposed thermolytically in the vapor phase of a cavitation bubble by sonolysis [20]: O3 ðgÞþÞÞÞ ! O2 ðgÞ þ Oð3 PÞðgÞ 3



Oð PÞðgÞ þ H2 OðgÞ ! 2 OHðgÞ

ð1Þ ð2Þ

These decomposition reactions occur in the gas phase. The reaction products migrate to the interfacial sheath of *

Corresponding author. Tel.: +86 27 68775837; fax: +86 27 68778893. E-mail address: [email protected] (H. Zhang).

1350-4177/$ - see front matter  2006 Elsevier B.V. All rights reserved. doi:10.1016/j.ultsonch.2006.09.005

the bubble where they subsequently react in the aqueous phase [11]. In addition, ultrasonic radiation has been demonstrated to increase the mass transfer of ozone to solution by means of increasing volumetric mass transfer coefficient [1,4,9,14], which is a critical factor in determining the overall efficacy of the process in the design of large-scale reactors for coupled gas–liquid reaction such as US/O3 system [21]. However, quantification of the mass transfer rates – the all-important information in the efficient design of coupled gas–liquid sonochemical systems – is still lacking [21]. Although a number of techniques have been developed to measure the volumetric gas–liquid mass transfer coefficient, the dynamic methods are preferably used, as they are fast, experimentally simple, and applicable to various systems [22]. When ozone mass transfer coefficient was measure by means of this classical unsteady state method with the absorption of ozone, the kinetic data of sonolytic decomposition of ozone were needed when establishing the mass balance equation of the absorption. Therefore the ozone decomposition experiments were required to determine ozone decomposition rate constants with ultrasonic radiation. In this case, degassing may be a factor in

H. Zhang et al. / Ultrasonics Sonochemistry 14 (2007) 552–556

sonolytic ozonation due to the high Henrys law constant of ozone, especially at low frequency such as 20 kHz [9]. However, little attention has been paid when ozone mass transfer coefficient was measured. Only Weavers and Hoffmann took this factor into account, and by investigating the decomposition of ozone in a continuously stirred tank reactor mode under conditions closed to the atmosphere, open to the atmosphere, and open to the atmosphere with ozone gas bubbling, mass transfer coefficients and selfdecomposition rate constants of ozone were determined [9]. In this study, we proposed another simpler method to determine mass transfer coefficient of ozone in a semi-batch reactor, and the effect of ultrasonic power on mass transfer enhancement was investigated. 2. Model development In a semi-batch reactor where ozone is continuously bubbled into the reactor containing a given volume of liquid, dissolved ozone mass balance in the reactor can be written as: d½O3  ¼ k L aA ð½O3   CÞ  kC  k L aD ð½O3   ½O3 0 Þ dt

ð3Þ

where kLaA is the volumetric mass transfer coefficient of gaseous ozone absorbed into the water (min1), kLaD is the volumetric mass transfer coefficient of dissolved ozone desorbed into the degassing bubbles or air over the water level in the semi-batch reactor (min1), [O3] is the dissolved ozone concentration at any time t (mM), k is the first-order reaction rate constant of sonolytic decomposition of ozone (min1) (the first-order reaction would be verified by the following experimental results), [O3]* is the ozone saturation concentration equilibrium with inlet gaseous ozone concentration [O3]g which would be determined from Roth and Sullivan [23] (mM), [O3]* 0 is the ozone saturation concentration equilibrium with gaseous ozone concentration in the degassing bubbles or at the top of the reactor (mM), and t is the time (min). Assuming [O3]* 0 is negligible [9], Eq. (3) can be rewritten as: d½O3   ¼ k L aA ½O3   ðk þ k L aA þ k L aD Þ½O3  dt

ð4Þ

Solving above equation with initial condition: t = 0, [O3] = 0, we obtain ½O3  ¼

k L aA ½O3  f1  exp½ðk þ k L aA þ k L aD Þtg k þ k L aA þ k L aD ð5Þ

Defining U = kLaA[O3]* and W = k + kLaA + kLaD, Eq. (5) could be simplified as: ½O3  ¼

U ½1  expðWtÞ W

ð6Þ

553

The parameters of U and W can be determined based on the data of dissolving ozone concentration [O3] versus time t using Matlab. Then the dissolved ozone concentration at steady state [O3]ss could be calculated as U/W and kLaA is obtained as follows: k L aA ¼

U ½O3 

ð7Þ

If decomposition rate constant k is available, kLaD can be calculated as k L aD ¼ W  k  k L aA

ð8Þ

To determine the decomposition rate constant k, ozone decomposition experiments would be conducted where quiescent water containing a given concentration of dissolved ozone is irradiated by ultrasound. In this case, we have: d½O3  ¼ k½O3   k L a0D ð½O3   ½O3 0 Þ dt

ð9Þ

where k L a0D is the volumetric mass transfer coefficient of dissolved ozone desorbed into the degassing bubbles or air over the water level in the batch reactor when no gas is bubbled (min1). Similarly, [O3]* 0 is negligible [9], and Eq. (9) can be integrated as ln

½O3  ¼ ðk þ k L a0D Þt ½O3 0

ð10Þ

where [O3]0 is the dissolved ozone concentration at time t = 0 (mM). A graph of ln([O3]/[O3]0) versus time t would give a straight line with a slope as ðk þ k L a0D Þ. Decomposition rate constant k can be obtained if k L a0D is available. Since k L a0D is difficult to measure, k L a0Doxygen , the volumetric mass transfer coefficient of dissolved oxygen desorbed into the degassing bubbles or air over the water level in the batch reactor is measured. k L a0D is then calculated according to Higbie’s penetration theory [24] sffiffiffiffiffiffiffiffiffiffiffiffiffiffi Dozone 0 0 ð11Þ k L aD ¼ k L aDoxygen Doxygen where Dozone and Doxygen are molecular diffusion coefficients of ozone and oxygen in water. They are determined to be 1.89 · 109 m2/s and 2.41 · 109 m2/s from Johnson and Davis [25], and Wilke and Chang [26], respectively. To determine k L a0Doxygen , degassing experiments would be conducted where quiescent water containing a given concentration of dissolved oxygen is irradiated by ultrasound. In this case, we have d½O2   ¼ k L a0Doxygen ð½O2   ½O2  Þ dt

ð12Þ

where [O2] is the dissolved oxygen concentration (mM), and [O2]* is the oxygen saturation concentration. Integrating above equation with initial condition: t = 0, [O2] = [O2]0, we obtain:

554

ln

½O2   ½O2  0  ¼ k L aDoxygen t ½O2 0  ½O2 

H. Zhang et al. / Ultrasonics Sonochemistry 14 (2007) 552–556

ð13Þ

A graph of ln(([O2][O2]*)/([O2]0 [O2]*)) versus time t would give a straight line with a slope as k L a0Doxygen .

oxygen concentration was periodically measured with DO probe (Orien 810A) and the oxygen saturation concentration was measured to be 0.25 ± 0.1 mM. 4. Results and discussion

3. Materials and methods It can be seen from Fig. 2 that ln(([O2][O2]*)/ ([O2]0[O2]*)) is linearly dependent on time with correlation coefficients range from 0.993 to 0.997. The oxygen mass transfer coefficients for degassing k L a0Doxygen would be determined by Eq. (13). In the absence of ultrasonic radiation, oxygen is actually desorbed into the air over the liquid level, and the mass transfer coefficient is as low as 0.0079 min1. The presence of ultrasonic radiation greatly enhanced mass transfer rate. Using Eq. (11), ozone mass transfer coefficients for degassing k L a0D could be calculated. Table 1 showed degassing effect due to ultrasonic radiation

0.0 P=0 W

ln{([O2]-[O2]*)/([O2] 0-[O2]*)}

In ozone absorption experiments, distilled water (pH 5.6–5.8) was fed into a 180 ml glass reactor. Sonication was performed with a KS-250 ultrasonic generator (250 W, 20 kHz, Ninbo Kesheng Instrument Co., China) equipped with a titanium probe transducer. The tip of the probe was 1 cm in diameter and was placed 3 cm into the liquid layer. The sonication was administered in pulses with a 50% duty cycle. The reactor was immersed into a water bath to keep the temperature around 25 ± 2 C (see Fig. 1). A magnetic stirrer provided complete mixing of the solution in the rector. Ozone was bubbled into the water using an ozone generator (XFZ-5QI, China). The gas flow rate was determined with a bubble flow meter. It was fixed at 100 ml min1 and the superficial velocity of gas was 0.12 cm s1 accordingly. The gaseous ozone concentration was monitored by the iodometric method with potassium iodide solution [27], and the dissolved ozone concentration was measured by the indigo method [28]. In ozone decomposition experiments, ozone was bubbled into the distilled water till a steady state ozone concentration in water was reached. Then the gas stream was stopped and ultrasonic generator was turned on. The dissolved ozone concentration was measured periodically by the indigo method [28]. In oxygen degassing experiments, oxygen was bubbled into the distilled water till a steady state oxygen concentration in water was reached. Then the gas stream was stopped and ultrasonic generator was turned on. The atmosphere over the liquid layer was purged with air to achieve the same oxygen content as that in the air. The dissolved

P=100W

-0.5 P=175W P=200W

-1.0

-1.5

-2.0 0

10

20

30

40

n

time (5 min) Fig. 2. Estimation of mass transfer coefficient k L a0Doxygen from data of dissolved oxygen concentration and time (n = 1: without ultrasonic radiation; n = 0: with ultrasonic radiation).

Fig. 1. The experimental setup.

H. Zhang et al. / Ultrasonics Sonochemistry 14 (2007) 552–556 Table 1 Volumetric mass transfer coefficients for degassing

555

0.30

Power (W)

k L a0D ðmin1 Þ

kLaD (min1)

0 100 175 200

0.0070 0.054 0.12 0.15

0.022 0.069 0.044 0.057

0.25

k (min-1)

0.20 0.15 0.10

0

0.05 -1 -2

0.00 0

P=100W

50

100

150

200

250

power (W)

P=175W

-3

Fig. 4. The effect of ultrasonic power on ozone decomposition rate constant.

P=200W

-4 -5 -6 0

5

10

15 20 time (min)

25

30

Fig. 3. The effect of ultrasonic power on ozone decomposition ([O3]0 = 0.32–0.38 mM).

is more pronounced with the increasing ultrasonic power. A 6.7–20.4-fold higher volumetric mass transfer coefficient for degassing k L a0D was observed than in the absence of ultrasonic radiation. Fig. 3 illustrated ozone decomposition under different ultrasonic power conditions. The results showed that ozone disappearance rate increased significantly with ultrasonic radiation. The correlation coefficients ranged from 0.987 to 0.996, indicating first-order decomposition hypothesis was in agreement with the experimental results. Based on Eq. (10), the slope determined from Fig. 3 is the sum of k þ k L a0D . Since k L a0D has been obtained in Table 1, ozone decomposition rate constants can be calculated. Fig. 4 indicated that ozone decomposition rate constant increased with the increase of ultrasonic power. A linear dependence of ozone decomposition rate constant k with a variation of ultrasonic power P can be quantified as follows [8]: k ¼ c0 þ cP

rates. However, the optimum power level is dependent on the operating frequency, and no maximum was observed when operating at 20 kHz [30]. Therefore under the frequency condition of 20 kHz in this study, ozone decomposition rate constant would increase with the increase of ultrasonic power. Fig. 5 illustrated ozone absorption under different ultrasonic power conditions. The steady state ozone concentration is more quickly reached with ultrasonic radiation than that in the absence of ultrasonic radiation. The calculated volumetric mass transfer coefficients kLaA in the presence of ultrasonic radiation were in the range of 0.31– 0.35 min1 compared with 0.26 min1 in the absence of ultrasonic radiation, indicating a 19.1–34.6% higher mass transfer coefficient was achieved in the presence of ultrasonic radiation. To investigate whether mass transfer enhancement was due to the increase of ozone decomposition rate in the presence of ultrasonic radiation, Hatta numbers were calculated as [24]

1.0 0.8

ð14Þ

where c0 is the ozone decomposition rate constant in the absence of ultrasonic radiation, c is the slope of the k versus P line and P is the ultrasonic power. The correlation coefficient is obtained as high as 0.9997. The primary mechanism for the US enhanced ozone decomposition is regarded as the thermal (pyrolysis) process via reactions (1) and (2) [4,8]. The increase in acoustic intensity will result in greater sonochemical effects in the collapsing bubble [29]. Kang et al. [8] thought there is an optimum power density which can be applied during sonochemical irradiation in order to obtain maximum reaction

[O3 ]/[O3 ]ss

ln([O3]/[O3] 0)

P=0W

0.6

P=0W

0.4

P=100W

0.2

P=175W

0.0

P=200W

0

3

6 9 time (min)

12

15

Fig. 5. The effect of ultrasonic power on ozone absorption ([O3]g = 1.01–1.54 mM, line: simulated).

556

Ha ¼

H. Zhang et al. / Ultrasonics Sonochemistry 14 (2007) 552–556

pffiffiffiffiffiffiffiffiffiffiffiffiffiffi Dozone k kL

ð15Þ

where kL is liquid side mass transfer coefficient and obtained from Calderbank and Moo-Young [31]. The very low Hatta numbers (5.28 · 103–7.19 · 103) indicated that mass transfer of ozone was not enhanced by ozone decomposition reaction, i.e., the enhancement factor E is equal to one. Therefore the calculated mass transfer coefficient was physical mass transfer coefficient. When gas bubbles containing ozone enter the ultrasonic reactor, the greater mixing due to ultrasonic radiation leads to the turbulence [4], which reduces the liquid film thickness [14]. According to the two-film theory, liquid side mass transfer coefficient kL is inverse proportional to the liquid film thickness. The decrease of the liquid film thickness would result in the increase of kL. In addition, one of the mechanical effects of ultrasound is the break up of gas bubbles containing ozone [4], which lead to the larger specific surface area aA. Therefore volumetric mass transfer coefficient kLaA would increase in the presence of ultrasonic radiation. However, the ultrasonic power had little effect on ozone absorption. The calculated volumetric mass transfer coefficients kLaA were nearly the same (0.31–0.35 min1) for all three ultrasonic powers. The volumetric mass transfer coefficient for degassing kLaD could be negligible compared with volumetric mass transfer coefficient kLaA, indicating that degassing effect was not pronounced in the sparged system when ozone was bubbled during sonolysis. This is in agreement with the results by Weavers and Hoffmann [9]. 5. Conclusion The classical unsteady state method was employed to determine volumetric mass transfer coefficient of ozone from gaseous phase to aqueous phase during sonolysis. The separate ozone decomposition experiments showed that rate constant of ozone self-decomposition with ultrasonic radiation was enhanced by ultrasonic radiation and increased linearly with the increase of ultrasonic power. The volumetric mass transfer coefficient of ozone and degassing coefficient were then determined based on ozone absorption and decomposition experiments. Ultrasound power has little effect on volumetric mass transfer coefficient though the presence of ultrasound enhanced ozone mass transfer rate. The degassing effect of ozone due to

ultrasonic radiation was insignificant in the sparged system when ozone was bubbled during sonolysis. Acknowledgements This study was supported by Wuhan Municipal Science and Technology Bureau, China through ‘‘The Chengguang Project’’ (Grant No. 20015005061). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31]

E. Dahi, Water Res. 10 (1976) 677. R.A. Sierka, Ozone Sci. Eng. 6 (1984) 275. R.A. Sierka, G.L. Amy, Ozone Sci. Eng. 7 (1985) 47. T.M. Olson, P.F. Barbier, Water Res. 28 (1994) 1383. P.F. Barbier, C. Petrier, J. Adv. Oxid. Technol. 1 (1996) 154. I. Hua, M.R. Hoffmann, Environ. Sci. Technol. 30 (1996) 864. J.W. Kang, M.R. Hoffmann, Environ. Sci. Technol. 32 (1998) 3194. J.W. Kang, H.M. Hung, A. Lin, M.R. Hoffmann, Environ. Sci. Technol. 33 (1999) 3199. L.K. Weavers, M.R. Hoffmann, Environ. Sci. Technol. 32 (1998) 3941. L.K. Weavers, F.H. Ling, M.R. Hoffmann, Environ. Sci. Technol. 32 (1998) 2727. L.K. Weavers, N. Malmstadt, M.R. Hoffmann, Environ. Sci. Technol. 34 (2000) 1280. H. Destaillats, A.J. Colussi, J.M. Joseph, M.R. Hoffmann, J. Phys. Chem. A104 (2000) 8930. N.H. Ince, G. Tezcanlı´, Dyes Pigment 49 (2001) 145. R. Lall, R. Mutharasan, Y.T. Shah, P. Dhurjati, Water Environ. Res. 75 (2003) 171. G. Tezcanli-Gu¨yer, N.H. Ince, Ultrasonics 42 (2004) 603. A.D. Martins, V.M. Canalli, C.M.N. Azevedo, M. Pires, Dyes Pigment 68 (2006) 227. I. Gu¨ltekin, N.H. Ince, Ultrason. Sonochem. 13 (2006) 208. H. Zhang, L.J. Duan, D.B. Zhang, J. Hazard. Mater. 138 (2006) 53. Y. Bando, S. Yamaguchi, Y. Toyoda, K. Doi, M. Nakamura, K. Yasuda, A. Oda, Y. Kawase, J. Chem. Eng. Jpn. 37 (2004) 1070. E.J. Hart, A. Henglein, J. Phys. Chem. 90 (1986) 3061. Y.G. Adewuyi, Environ. Sci. Technol. 39 (2005) 3409. A. Kumar, P.R. Gogate, A.B. Pandit, H. Delmas, A.M. Wilhelm, Ind. Eng. Chem. Res. 43 (2004) 1812. J.A. Roth, D.E. Sullivan, Ind. Eng. Chem. Fundam. 20 (1981) 137. P.V. Danckwerts (Ed.), Gas–Liquid Reactions, McGraw-Hill, New York, 1970. P.N. Johnson, R.A. Davis, J. Chem. Eng. Data 41 (1996) 1485. C.R. Wilke, P. Chang, AICHE J. 1 (1955) 264. D.L. Flamm, Environ. Sci. Technol. 11 (1977) 978. H. Bader, J. Hoigne´, Water Res. 15 (1981) 449. Y.G. Adewuyi, Ind. Eng. Chem. Res. 40 (2001) 4681. L.H. Thompson, L.K. Doraiswamy, Ind. Eng. Chem. Res. 38 (1999) 1215. P.H. Calderbank, M.B. Moo-Young, Chem. Eng. Sci. 16 (1961) 39.