Kinetics of the Reaction between Ozone and p-Hydroxybenzoic Acid

in a Semibatch Reactor. Ya-Wen Ko,*,† Pen-Chi Chiang,† Chia-Line Chuang,† and E. E. Chang‡. Graduate Institute of Environmental Engineering, N...
0 downloads 0 Views 83KB Size
Download PDF
Ind. Eng. Chem. Res. 2000, 39, 635-641

635

Kinetics of the Reaction between Ozone and p-Hydroxybenzoic Acid in a Semibatch Reactor Ya-Wen Ko,*,† Pen-Chi Chiang,† Chia-Line Chuang,† and E. E. Chang‡ Graduate Institute of Environmental Engineering, National Taiwan University, Taipei 106, Taiwan, and Department of Biochemistry, Taipei Medical College, Taipei 105, Taiwan

The kinetic regime in the ozone and p-hydroxybenzoic acid reaction system at pH 3 and 7 can be considered a slow reaction, whereas at pH 10 the rapid reaction mainly occurred in the liquid film and developed in the fast absorption kinetic regime. On the basis of the discussion of kinetic regime and the experimental observations, the kinetic expressions for the p-hydroxybenzoic acid, glyoxal, and dissolved ozone concentration can be developed for both pH 3 and 7 and were verified with the experimental data. The reaction rate constants were also calculated. The agreement between the experimental observations and the model predictions was reasonably good. However, at pH 7, a positive deviation in the decaying glyoxal profile and the overestimation for the p-hydroxybenzoic acid and dissolved ozone concentration were found. The reasons for the discrepancy between the predicted and measured concentrations were also discussed. Introduction Ozonation can oxidize the micropollutants and reduce the disinfection byproduct (DBP) precursors in drinking water. However, many compounds such as low molecular weight aliphatic aldehydes, hydrogen peroxides, organic peroxides, mixed functional and saturated carboxylic acids, and brominated byproducts have been identified as ozonation byproducts.1-4 Aldehydes and chlorinated aldehydes are potential health hazards and are highly biodegradable, which may lead to increasing bacterial populations in distribution systems.2,5,6 Several studies have been conducted to evaluate the factors affecting aldehyde production. It is believed that the aldehyde concentration increases with the ozone dose and ozonation time.3,7-9 Furthermore, the reaction mechanisms relating to the aldehyde formation have been proposed.9-11 On the other hand, because of the complexity of the ozone reaction system, the kinetic study and mathematical simulation for the aldehyde formation are limited. Because ozonation is a gas-liquid process under dynamic conditions, the reaction regime, mass-transfer characteristics, and reaction kinetics are critical to system design and operation. The ascertainment of the kinetic regime and place where the reaction occurs (in the liquid bulk or in the liquid film) will be useful in understanding the reaction system and conducting the kinetic modelization for the ozonation process. According to Beltran et al.,12 four types of kinetic regimes in a gasliquid reaction are visualized: slow, instantaneous, very slow, and fast pseudo-first-order reactions.13,14 The kinetic regime of ozone absorption has been shown to be highly dependent on pH, temperature, ozone partial pressure, gas flow rate, and agitation speed.12,15-20 The objective of this study was to develop the kinetic expression for p-hydroxybenzoic acid, glyoxal, and dis* To whom correspondence should be addressed. Current address: Da-Yeh University, Department of Environmental Engineering, 112, Shan-Jeau Rd., Dah-Tsuen, Chang-Hwa 515, Taiwan. Tel.: +886-4-852-8469, ext 2364. Fax: +8864-853-1157. E-mail: [email protected]. † National Taiwan University. ‡ Taipei Medical College.

solved ozone concentration during the ozonation course. Mass-transfer characteristics and the ascertainment of the kinetic regime at the various pH levels were first evaluated. From the discussion of kinetic regime and the experimental observations, the kinetic models predicting the p-hydroxybenzoic acid, glyoxal, and dissolved ozone concentrations at pH 3 (the direct reaction) and at pH 7 (including both the direct and radical reactions) were proposed and verified with the measured data. p-Hydroxybenzoic acid was chosen as the target compound because it has been identified as a micropollutant in surface waters, a building block in the structure of natural organic matter (NOM), and a potential DBP precursor.11,21,22 The experimental findings at pH 3 have been introduced in previous research.9 Also, the effects of ozone dose and ozonation time on the reduction of total organic carbon (TOC), UV254, and DBPFP (DBP formation potential, including trihalomethane formation potential, THMFP, and haloacetic acid formation potential, HAAFP) were also investigated. Material and Methods A laboratory-scale ozonation system, consisting of a 15-L stainless steel reactor with a built-in diffuser for ozone introduction, an agitator (250 rpm), and internal baffles to facilitate mixing, was used in this investigation. Organic-free and deionized water (Milli-Q SP) containing commercial p-hydroxybenzoic acid (Sigma) at 8 mg/L (TOC ) 5 mg/L) was used as the test sample. Three runs at pH 7 and four runs at pH 10 with different ozone doses (1.19-3.45 mg/(L min)) were conducted at T ) 20 °C. The detailed schematic diagram and operating conditions of this ozonation system is shown elsewhere.9 Dissolved ozone concentrations were analyzed by an ozone monitor (Orbisphere, model 3600) equipped with a membrane-containing cathode. p-Hydroxybenzoic acid was measured quantitatively by HPLC (GBC) using a C18 reverse-phase column (Phenomenex) and a UV detector set at 280 nm. Aldehyde, THM, and HAA were measured using the specific pretreatment procedure and then followed by GC (HP 5890) coupled with ECD

10.1021/ie990111a CCC: $19.00 © 2000 American Chemical Society Published on Web 02/11/2000

636

Ind. Eng. Chem. Res., Vol. 39, No. 3, 2000

(electron capture detector) detection. The detailed analytical methods have been described in previous research.9 Results and Discussion Kinetic Regimes in the Ozonation of p-Hydroxybenzoic Acid. If the kinetic regime of the reaction between ozone and p-hydroxybenzoic acid was known, the kinetic expressions could be properly developed. Thus, the kinetic regimes at pH 3, 7, and 10 were first discussed. In previous work,9 the ascertainment of the reaction regime was achieved on the basis of experimental observation and theoretical criterion. At pH 3, the kinetic absorption of ozone in p-hydroxybenzoic acid is present in the solution instead of the film layer. As the progress of ozonation time continues, the kinetic regime gradually shifts from a slow reaction to a very slow reaction case once p-hydroxybenzoic acid is oxidized. Figure 1 showed the evolution of the p-hydroxybenzoic acid and dissolved ozone concentrations with respect to the progress of reaction time at various pH values. The removal efficiency of p-hydroxybenzoic acid generally increased with the increase of the pH level. At 2.49 mg/ (L min) of applied ozone dose and 5 min of ozonation time, the removal efficiency of p-hydroxybenzoic acid was 28% at pH 3, 37% at pH 7, and 64% at pH 10. As for the concentration profile in Figure 1, the patterns at pH 7 were similar to those at pH 3: once phydroxybenzoic acid was completely oxidized, the dissolved ozone concentration appeared and then accumulated. It is consequently suggested that the reaction regime at pH 7 is probably of the same kind as that at pH 3. On the other hand, at pH 10, the decay of p-hydroxybenzoic acid was apparently more rapid and the dissolved ozone concentration was never detected. Because the predominant pathway at pH 10 is the radical-type reaction, a higher reaction rate is expected. The theoretical criterion suggested by Danckwerts13 and Charpentier14 has been discussed in previous work.9 In short, in this ozonation system when the rate constant is smaller than 7 × 104 M-1 s-1, the criterion for a slow reaction is fulfilled and hence the reaction follows the slow kinetic regime of absorption. Another important condition to clarify the kinetic regime of the ozonation is the enhancement factor (E), defined as the ratio between the actual and maximum physical absorption rates:14

E)

Nt

(1)

KLa[O3]*

At the initial stage, ozone reacted mainly with phydroxybenzoic acid. It was assumed that the competing reaction between ozone and other secondary products was negligible. Therefore, Nt could be expressed as a function of the p-hydroxybenzoic acid’s decomposition rate, accounting for the stoichiometry by

(

Nt ) z -

)

d[p - HBA] dt

t)0

(2)

A polynomial regression of p-hydroxybenzoic acid with respect to time was applied to fit the data. Combining the derivative at t ) 0 and z ) 3 yielded the Nt values under different ozone doses. The average KLa value was 0.083 min-1 from the previous work.9 The [O3]* values

Figure 1. Effect of the applied ozone dose and ozonation time on the p-hydroxybenzoic acid and dissolved ozone concentration at (a) pH 3, (b) pH 7, and (c) pH 10.

were determined from Henry’s constant and the ozone partial pressure at the reactor outlet. The E values calculated from eqs 1 and 2 were around 1.1 at pH 3, 1.3 at pH 7, and 3.0 at pH 10. Therefore, combining the discussion of experimental observation and theoretical examination, the following conclusions can be drawn. At pH 7, the reactions took place in the bulk of the water and developed in the slow absorption kinetic regime, in the same manner as those at pH 3. On the other hand, at pH 10 the rapid reactions mainly occurred in the liquid film and developed in the fast absorption kinetic regime. Modelization of the p-Hydroxybenzoic Acid Degradation, Glyoxal Formation, and Dissolved Ozone Concentration. Because the chemical reactions at pH 3 and 7 are the rate-determining steps (derived from the slow kinetic regime), a kinetic model can be proposed for the prediction of p-hydroxybenzoic acid degradation, glyoxal formation, and dissolved ozone concentration at pH 3 and 7. Figure 2 showed the effects

Ind. Eng. Chem. Res., Vol. 39, No. 3, 2000 637

ates resulting from the reaction of ozone and p-hydroxybenzoic acid, other than glyoxal, and P represents the products formed by the reaction of ozone and glyoxal. It should be noted that eqs 3 and 4 were a short-hand way of describing a whole series of reactions. For example, in eq 3, there are three ozonation reactions, including hydroxyl substitution, ring breaking, and double-bond breaking, taking place in succession to yield glyoxal. Consequently, the value of z ) 3 is determined. Assuming a first-order reaction with respect to the individual reactant, and on the basis of the reaction regime discussed, the reaction rate of p-hydroxybenzoic acid (S) and glyoxal (Gly) in the semibatch reactor can be expressed as

d[S] ) -k1[S][O3] - k2[S][O3] ) -(k1 + k2)[S][O3] dt (6) Figure 2. Effect of the applied ozone dose and ozonation time on the glyoxal formation at pH 7.

of the applied ozone dose and ozonation time on the variation of the intermediate, glyoxal, at pH 7. Similar to the case at pH 3, glyoxal was the major aldehyde species. Glyoxal increased to a maximum, followed by a decrease. The ozonation time at which the peak glyoxal appeared decreased with the increase of ozone dose. However, there were two points different from the situations at pH 3. First, the glyoxal decay with time was more rapid at pH 7 than at pH 3. Second, the peak glyoxal levels at pH 7 reached only 50% of those at pH 3. At pH 3 the direct oxidation is the predominant reaction pathway. Ozone rapidly reacted with p-hydroxybenzoic acid and other intermediates to yield the relative stable products, such as glyoxal, to the peak level. On the other hand, the reaction rate between ozone molecules and the relatively stable glyoxal became slower, which led to a gradual decrease of glyoxal. At pH 7, on the basis of the discussions about reaction regimes, the reactions mainly occurred in the bulk water, similar to the situation at pH 3. However, the generated glyoxal at pH 7 was less than that at pH 3 under the same conditions. This implied that the contribution of the radical reaction at pH 7 was not negligible. Because the applied ozone doses were partly consumed by hydroxide ions, the available ozone with respect to oxidizing p-hydroxybenzoic acid and producing glyoxal became less. Hence, glyoxal formation at pH 7 was less. After glyoxal formed via a direct reaction pathway, the nonselective radical reactions aided in further oxidation and decomposed the glyoxal rapidly, resulting in a fast glyoxal decay. On the basis of the proposed reaction mechanisms,9 experimental observations, and consideration of the direct reaction only (that is, the pH 3 case), the reaction of ozone and p-hydroxybenzoic acid can be expressed by the following simple equations: k1

S + 3O3 98 Gly k2

S + 3O3 98 A k3

Gly + O3 98 P

(3) (4) (5)

where S represents p-hydroxybenzoic acid, Gly represents glyoxal, A designates the products or intermedi-

d[Gly] ) k1[S][O3] - k3[Gly][O3] dt

(7)

Equations 6 and 7 can be integrated under the given initial conditions:

[S]) [S]0,

at t ) 0

(8)

[Gly] ) 0,

at t ) 0

(9)

On the other hand, examining the mass balance of ozone in the semibatch system led to the following expression with the initial conditions

d[O3] ) KLa([O3]* - [O3]) - rO3 ) KLa([O3]* dt [O3]) - 3k1[S][O3] - 3k2[S][O3] - k3[Gly][O3] (10) [O3] ) 0,

at t ) 0

(11)

A Fortran program using the fourth-order Runge-Kutta numerical method was developed to solve eqs 6-11 for the concentrations of p-hydroxybenzoic acid, glyoxal, and ozone as a function of reaction time. The reaction rate constants k1-k3 were found by a trial-and-error procedure and the differences between predicted concentrations and experimental values were used as the criterion for the convergence of iteration. The calculated concentration-time curves during the ozonation course at pH 3 were constructed as lines shown in Figure 3. The resulting average k1, k2, and k3 were 20.0, 66.7, and 28.3 M-1 s-1, respectively. In Figure 3, the predicted trend generally coincided with experimental observations. p-Hydroxybenzoic acid concentrations decreased with reaction time. Glyoxal increased to a maximum and then gradually decreased with the progress of ozonation time. The agreement between the experimental observations and the model predictions for p-hydroxybenzoic acid’s and glyoxal’s concentrations was reasonably well. However, as for the dissolved ozone concentration, the predicted values were significantly higher than the observed data, as shown in Figure 3. It indicated that there are other reactions consuming ozone, except the reactions shown in eqs 3-5, which causes the additional consumption of ozone beyond the model prediction. For example, the intermediates or products A in eq 4, or the products P in eq 5, may react with ozone again, even being completely

638

Ind. Eng. Chem. Res., Vol. 39, No. 3, 2000

Figure 3. Verification of predicted p-hydroxybenzoic acid, glyoxal, and dissolved ozone concentration at pH 3. (Applied ozone dose ) 2.49 mg/(L min).)

Figure 4. Verification of predicted p-hydroxybenzoic acid, glyoxal, and dissolved ozone concentration at pH 7. (Applied ozone dose ) 2.49 mg/(L min).)

oxidized into CO2. Therefore, there was a positive deviation of predicted ozone concentration. Let us now consider the ozonation at pH 7. The radical reaction was also taken into consideration in addition to eqs 3-5:

centration reached the steady state in a short time,

k4

O3 + OH- 98 •O2- + HO2•

k4 ) 70 M-1 s-1 23 (12) k5

S + •OH 98 B

(13)

k6

Gly + •OH 98 P′

(14)

where B represents the products formed by the reaction of p-hydroxybenzoic acid and •OH and P′ represents the products formed by the reaction of glyoxal and •OH. Equation 12 accounts for the initial decomposition of ozone by the hydroxide ions. Because the various radicals or intermediates in the chain propagation reactions were rapidly formed and reacted, a steadystate approximation for these intermediates was used. Consequently, it was assumed that the major reactions consuming ozone were eqs 3-5 and 12. Therefore, according to the reactions listed in eqs 3-5 and 12-14, the reaction rate of p-hydroxybenzoic acid, glyoxal, and ozone concentration at pH 7 can be expressed as

d[S] ) -(k1 + k2)[S][O3] - k5[S][•OH] dt

(15)

d[Gly] ) k1[S][O3] - k3[Gly][O3] - k6[Gly][•OH] dt (16) d[O3] ) KLa([O3]* - [O3]) - rO3 ) KLa([O3]* dt [O3]) - 3k1[S][O3] - 3k2[S][O3] - k3[Gly][O3] k4[O3][OH-] (17) The concentration of the hydroxyl radical is a function of ozone concentration, and it can be regarded as a timeindependent parameter, provided that the radical con-

[•OH] ) R[O3]

(18)

where R is a proportional factor relating [•OH] to [O3]. The same relation has also been derived by Yurteri and Gurol.24 It was shown that R is a function of hydroxide ion, solute concentrations, and radical reaction constants. Hence, with the same initial conditions (eqs 8, 9, and 11), the same approach was applied to find the solution for the prediction model. Figure 4 showed the relationship between the predicted and measured concentration at pH 7. The resulting average R, k5, and k6 were 3.8 × 10-8, 3 × 109, and 2 × 108 M-1 s-1, respectively. As for glyoxal formation, the major discrepancy took place as the glyoxal started decaying. In Figure 2, after the glyoxal concentration reached the maximum value, it decreased rapidly and the concentration profiles were roughly symmetrical. However, the decays in the calculated glyoxal profiles were slower, resulting in positive deviation which appeared after 20 min of ozonation time, shown in Figure 4. The overestimation for the p-hydroxybenzoic acid and dissolved ozone concentration over the entire ozonation time was also observed. Therefore, there must be some other sinks in the system reacting with p-hydroxybenzoic acid, glyoxal, and ozone and reducing their concentrations. As mentioned above, other intermediates or products not being considered by eqs 3-5 and 12-14 may react with p-hydroxybenzoic acid, glyoxal, or ozone again, resulting in the additional consumption. The participation of the superoxide ion radical (•O2-) and organic radicals in the reaction system may be significant. Additionally, at pH 7 the ozone self-decomposition rate is higher than that at pH 3, which also leads to the additional reduction of ozone concentration at pH 7. As for glyoxal, previous investigation25 suggested the OHCCO•, OHC-CO•, and OHC-COO• radicals might be formed during glyoxal ozonation and react with glyoxal to sustain the propagating cycles in the radical chain reactions. Another source of the deviation may come from the rate constants k1 and k2 at pH 7 during the calculation. p-Hydroxybenzoic acid is a dissociating compound with pKa1 ) 4.48 (carboxylic group) and pKa2 ) 9.32 (phenolic group).26 At pH 7 the dominant form

Ind. Eng. Chem. Res., Vol. 39, No. 3, 2000 639 Table 1. Reduction Efficiencies of TOC and UV254 applied ozonation TOC (%) UV254 (%) ozone dose time (mg/(L min)) (min) pH 3 pH 7 pH 10 pH 3 pH 7 pH 10 1.77

2.49

3.45

5 10 15 20 25 30 35 40

0.1 4.5 9.1 12.1 16.3 17.9 29.6 35.3

4.7 12.4 18.8 26.3 32.9 37.1 44.2 45.4

6.9 14.7 18.9 27.5 35.1 41.1 50.2 55.9

20.7 43.9 65.8 83.8 94.9 97.8 97.8 98.7

24.7 47.2 65.4 80.1 92.6 96.8 97.8 98.1

25.7 52.6 75.7 90.4 94.7 96.8 97.5 98.3

5 10 15 20 25 30 35 40

2.0 9.6 15.0 28.2 33.0 38.6 41.1 49.9

6.0 17.1 26.5 36.8 50.3 53.3 57.0 61.3

7.9 15.7 27.5 35.2 44.8 53.2 57.9 62.4

37.3 72.0 94.3 98.5 98.7 98.6 98.9 98.5

34.9 65.1 86.3 97.2 98.3 98.1 98.5 98.7

32.6 66.2 90.7 98.2 98.6 98.5 97.6 98.3

5 10 15 20 25 30 35 40

4.0 17.3 32.4 31.8 43.4 46.7 49.9 55.6

10.2 26.1 37.6 43.3 49.3 50.8 54.2 56.0

12.5 30.4 37.1 44.8 50.4 55.6 57.5 63.4

56.6 93.9 98.6 98.8 99.1 98.7 98.7 98.6

47.6 84.8 97.9 98.3 98.7 98.5 98.5 98.3

37.9 78.0 96.9 98.2 98.3 98.2 98.7 98.7

of the solute reacting with ozone is HO-C6H4-COO-. Previous investigation27,28 has found that the secondorder rate constants increase with pH as does the degree of deprotonation of the dissolved substances. Consequently, applying the rate constant at pH 3 to that at pH 7 will contribute to the discrepancy between the predicted and measured concentrations. Reduction of Organics and DBP Precursors. In addition to aldehydes, the reduction of general organics and DBP precursors was also measured during ozonation of p-hydroxybenzoic acid at pH 3, 7, and 10. Table 1 showed the reduction efficiencies of TOC and UV254 at the various applied ozone doses and ozonation times. The reduction efficiencies of TOC and UV254 generally increased with the increase of applied ozone dose and ozonation time. For these two organic parameters, however, there were some major differences between them. For TOC, the reduction efficiencies were in the order of pH 10 > pH 7 > pH 3. This supported the fact that the radical reaction would promote the degree of complete oxidation. Nevertheless, the reduction efficiencies of TOC at the 40 min of ozonation time and pH 10 were around 56-63%. On the other hand, the high removal (above 95%) for the UV254 substances could be found at each pH level or ozone dose. Also, in a comparison of the reduction efficiencies of UV254 among the three pH levels, no consistent trend was found. As the ozonation time continued, the reduction efficiencies reached the stationary value. Table 2 shows the reduction efficiencies of CHCl3FP and TCAAFP. It was observed that the CHCl3FP removal increased with the increase of ozonation time at pH 3 and 7, whose predominant pathway is the direct reaction. In the radical reaction (pH 10), at 5 min of ozonation time, the CHCl3FP concentration increased first, followed by a rapid decrease. The results indicate that the radicals reacted with organics nonselectively and might enhance the CHCl3 precursor concentration during the ozonation treatment. Additionally, the CHCl3FP produced at 5 min of ozonation time and pH 10 increased with the increase of the applied ozone dose.

Table 2. Reduction Efficiencies of CHCl3FP and TCAAFPa applied ozonation CHCl3FP (%) TCAAFP (%) ozone dose time (mg/(L min)) (min) pH 3 pH 7 pH 10 pH 3 pH 7 pH 10

a

1.77

5 10 20 40

35.0 55.0 79.0 94.8

6.6 -14.0 25.0 26.0 10.0 55.0 59.2 65.5 88.4 91.0 94.5 99.4

25.3 28.4 80.1 97.5

42.4 70.3 93.6 98.1

2.49

5 10 20 40

48.5 65.6 93.8 94.8

18.5 -22.5 35.6 50.7 12.5 74.1 84.6 84.5 98.9 96.5 96.0 99.4

24.8 58.7 96.4 98.7

54.2 75.5 93.9 97.8

3.45

5 10 20 40

53.4 84.2 94.5 95.1

29.9 -32.5 65.4 61.2 36.0 94.7 91.8 90.0 99.2 96.3 94.0 99.5

38.7 88.3 98.1 99.1

52.1 77.8 94.2 98.0

The reduction efficiencies at pH 3 were from previous work.9

The reduction efficiencies of CHCl3FP at 1.77, 2.49, and 3.45 mg/(L min) of applied ozone dose were -14.0%, -22.5%, and -32.5%, respectively. Regarding TCAAFP, ozonation has positive control at three pH values. Nevertheless, at 40 min of ozonation time, the reduction efficiencies for CHCl3FP and TCAAFP at each pH level were almost the same and reached the stationary value. Figure 5 showed the effect of the applied ozone dose on the pseudo-first-order reaction rate constant for UV254, CHCl3FP, and TCAAFP at pH 3, 7, and 10. For UV254, the pseudo-first-order reaction rate constant had a positive correlation with the applied ozone dose at pH 3, 7, and 10. Because there were some connections between UV254 and p-hydroxybenzoic acid, the finding showed in Figure 5a can be explained on the basis of eq 6. However, for CHCl3FP and TCAAFP, the pseudofirst-order reaction rate constant was not associated with the applied ozone dose at pH 7 and 10, whereas the rate constant was a function of the applied ozone dose at pH 3. Therefore, the results suggest that the reaction rate for UV254 was first-order with respect to both UV254 and ozone concentration. On the other hand, at pH 7 and 10, the reaction rates for CHCl3FP and TCAAFP were proportional to their own concentrations only. Conclusions The kinetic regimes in the ozone and p-hydroxybenzoic acid reaction system at the various pH values were evaluated on the basis of the theoretical background and experimental results. At pH 3 and 7, the reactions took place in the bulk of the water and developed in the slow absorption kinetic regime. On the other hand, at pH 10 the rapid reaction mainly occurred in the liquid film and developed in the fast absorption kinetic regime. From the proposed reaction mechanisms, the reaction regime discussed, and experimental observations, the kinetic expressions for the p-hydroxybenzoic acid, glyoxal, and dissolved ozone concentration can be developed for both pH 3 and 7 and verified with the experimental data. The reaction rate constant at pH 3 (considering the direct reaction only) for the reaction producing glyoxal was 20.0 M-1 s-1. The rate constant between the reaction of glyoxal and ozone was 28.3 M-1 s-1. On the other hand, considering both the direct and radical reaction occurring at pH 7, the rate constant between the reaction of p-hydroxybenzoic acid and •OH radical was 3 × 109 M-1 s-1 and that for glyoxal and the •OH radical was 2 × 108 M-1 s-1. The predicted trend

640

Ind. Eng. Chem. Res., Vol. 39, No. 3, 2000

Regarding the reduction of organics and DBP precursors during ozonation, the reduction efficiencies of TOC were in the order of pH 10 > pH 7 > pH 3. This supported the fact that the radical reaction would promote the degree of complete oxidation. As for UV254 and TCAAFP, the reduction efficiencies generally increased with the applied ozone dose and ozonation time. However, the CHCl3FP concentration increased first, followed by a rapid decrease in the radical reaction. Examination of the pseudo-first-order reaction rate constant suggests that the reaction rate for UV254 was first-order with respect to both UV254 and ozone concentration. On the other hand, at pH 7 and 10, the reaction rate for CHCl3FP and TCAAFP was proportional to their own concentrations only. Nomenclature

Figure 5. Effect of the applied ozone dose on the first-order reaction rate constants for (a) UV254, (b) CHCl3FP, and (c) TCAAFP.

generally coincided with experimental observations. p-Hydroxybenzoic acid concentrations decreased with reaction time. Glyoxal increased to a maximum and then gradually decreased with the progress of ozonation time. However, at pH 7, the decays in the calculated glyoxal profiles were slower, resulting in positive deviation which appeared after 20 min of ozonation time. Meanwhile, overestimation for the p-hydroxybenzoic acid and dissolved ozone concentration over the entire ozonation time was also observed. Therefore, other intermediates or products not being considered by the kinetic study may react with p-hydroxybenzoic acid, glyoxal, or ozone again, resulting in additional consumption. The participation of the superoxide ion radical (•O2-) and organic radicals in the reaction system may be significant. On the other hand, the effect of pH value on the rate constant was not considered and that will also contribute to the discrepancy between the predicted and measured concentrations.

A ) the products or intermediates resulting from the reaction of ozone and p-hydroxybenzoic acid, other than glyoxal, defined in eq 4 B ) the products formed by the reaction of p-hydroxybenzoic acid and •OH, defined in eq 13 E ) enhancement factor, defined in eq 1, dimensionless Gly ) glyoxal k1 ) reaction rate constant in eq 3, M-1 s-1 k2 ) reaction rate constant in eq 4, M-1 s-1 k3 ) reaction rate constant in eq 5, M-1 s-1 k4 ) reaction rate constant in eq 12, M-1 s-1 k5 ) reaction rate constant in eq 13, M-1 s-1 k6 ) reaction rate constant in eq 14, M-1 s-1 KLa ) overall mass-transfer coefficient, s-1 Nt ) actual ozone absorption rate, M s-1 P ) the products formed by the reaction of ozone and glyoxal, defined in eq 5 P′ ) the products formed by the reaction of glyoxal and •OH, defined in eq 14 rO3 ) reaction rate of ozone, M s-1 S ) p-hydroxybenzoic acid t ) reaction time, s z ) stoichiometric ratio, ozone mole consumed per phydroxybenzoic acid mole consumed, dimensionless [Gly] ) the glyoxal concentration, M [p-HBA] ) concentration of p-hydroxybenzoic acid dissolved in water, M [O3]* ) equilibrium ozone concentration at the gas-liquid interface (x ) 0), M [S]0 ) the concentration of p-hydroxybenzoic acid at t ) 0, M Greek Letters R ) proportional factor relating [•OH] to [O3], defined by eq 18, dimensionless

Literature Cited (1) Glaze, W. H. Chemistry of Ozone, By-Products, and Their Health Effects. In Ozonation: Recent Advances and Research Needs; American Water Works Associated Research Foundation: Denver, 1986. (2) Weinberg, H. S.; Glaze, W. H. An Overview of Ozonation Disinfection By-Products. In Disinfection By-Products In Water Treatment; CRC Press: Boca Raton, FL, 1996. (3) Gracia, R.; Aragues, J. L.; Ovelleiro, J. L. Study of the Catalytic Ozonation of Humic Substances in Water and Their Ozonation Byproducts. Ozone Sci. Eng. 1996, 18, 195. (4) Weinberg, H. S.; Glaze, W. H. A Unified Approach to the Analysis of Polar Organic By-Products of Oxidation in Aqueous Matrices. Water Res. 1997, 31, 1555. (5) Schechter, D. S.; Singer, P. C. Formation of Aldehydes during Ozonation. Ozone Sci. Eng. 1995, 17, 53.

Ind. Eng. Chem. Res., Vol. 39, No. 3, 2000 641 (6) Standard Methods for the Examination of Water and Wastewater, 19th ed.; APHA, AWWA, WEF: Washington, DC, 1995. (7) Glaze, W. H.; Koga, M.; Cancilla, D. Ozonation Byproducts. 2. Improvement of an Aqueous-Phase Derivatization Method for the Detection of Formaldehyde and Other Carbonyl Compounds Formed by the Ozonation of Drinking Water. Environ. Sci. Technol. 1989, 23, 838. (8) Yamada, H.; Somiya, I. The Determination of Carbonyl Compounds in Ozonated Water by the PFBOA Method. Ozone Sci. Eng. 1989, 11, 53. (9) Ko, Y.-W.; Chiang, P.-C.; Chang, E. E. Ozonation of pHydroxybenzoic Acid Solution. Ozone Sci. Eng. 1998, 20, 343. (10) Gould, J. P.; Weber, W. J., Jr. Oxidation of Phenols by Ozone. J. Water Pollut. Control Fed. 1976, 48, 47. (11) Langlais, B.; Reckhow, D. A.; Brink, D. B. Ozone in Water TreatmentsApplication and Engineering; Lewis Publishers: Chelsea, 1991. (12) Beltran, F. J.; Gomez-Serrano, V.; Duran, A. Degradation Kinetics of p-Nitrophenol Ozonation in Water. Water Res. 1992, 26, 9. (13) Danckwerts, P. V. Gas Liquid Reactions; McGraw-Hill: New York, 1970. (14) Charpentier, J. C. Mass Transfer Rates in Gas-Liquid Absorbers and Reactors. In Advances in Chemical Engineering; Academic Press: New York, 1981; Vol. 11. (15) Gurol, M. D.; Singer, P. C. Dynamics of the Ozonation of Phenol-II Mathematical Simulation. Water Res. 1983, 17, 1173. (16) Beltran, F. J.; Encinar, J. M.; Garcia-Araya, J. F. Ozonation Of o-Cresol in Aqueous Solution. Water Res. 1990, 24, 1309. (17) Sotelo, J. L.; Beltran, F. J.; Gonzalez, M. Ozonation of Aqueous Solutions of Resorcinol and Pholoroglucinol. 1. Stoichiometry and Absorption Kinetic Regime. Ind. Eng. Chem. Res. 1990, 29, 2358. (18) Benitez, F. J.; Beltran-Heredia, J.; Gonzalez, T. Kinetics of the Reaction between Ozone and MCPA. Water Res. 1991, 25, 1345.

(19) Beltran, F. J.; Ovejero, G.; Encinar, J. M.; Rivas, J. Oxidation of Polynuclear Aromatic Hydrocarbons in Water. 1. Ozonation. Ind. Eng. Chem. Res. 1995, 34, 1596. (20) Beltran, F. J.; Encinar, J. M.; Alonso, M. A. Nitroaromatic Hydrocarbon Ozonation in Water. 1. Single Ozonation. Ind. Eng. Chem. Res. 1998, 37, 25. (21) Rice, R. G. The Use of Ozone to Control Trihalomethanes in Drinking Water Treatment. Ozone Sci. Eng. 1980, 2, 75. (22) Yamada, H.; Somiya, I.; Inanami, F. The Effect of Preozonation on the Control of Trihalomethane Formation. Ozone Sci. Eng. 1986, 8, 129. (23) Staehelin, J.; Hoigne, J. Decomposition of Ozone in Water: Rate of Initiation by Hydroxide Ions and Hydrogen Peroxide. Environ. Sci. Technol. 1982, 16, 676. (24) Yurteri, C.; Gurol, M. D. Ozone Consumption in Natural Waters: Effects of Background Organic Matter, pH and Carbonate Species. Ozone Sci. Eng. 1988, 10, 277. (25) Caprio, V.; Insola, A.; Silvestre, A. M. Glyoxal Ozonation Process in Aqueous Solution. Ozone Sci. Eng. 1989, 11, 271. (26) Schwarzenbach, R. P.; Gschwend, P. M.; Imboden, D. M. Environmental Organic Chemistry; John Wiley & Sons: New York, 1993. (27) Hoigne, J.; Bader, H. Rate Constants of Reactions of Ozone with Organic and Inorganic Compounds in WatersI: Nondissociating Organic Compounds. Water Res. 1983, 17, 173. (28) Hoigne, J.; Bader, H. Rate Constants of Reactions of Ozone with Organic and Inorganic Compounds in WatersII: Dissociating Organic Compounds. Water Res. 1983, 17, 185.

Received for review February 16, 1999 Revised manuscript received November 19, 1999 Accepted November 22, 1999 IE990111A