 |
| About Bioline | All Journals | Testimonials | Membership | News |
|
||||||
|
||||||
International Journal of Environmental Research, Vol. 4, No. 3, July-September, 2010, pp. 507-512 Article Determination of the apparent reaction rate constants for ozone degradation of substituted phenols and QSPR/QSAR analysis Liu, H. 1 , Tan, J. 1 , Yu, H. X. 2 , Liu, H. X. 1 , Wang, L.S. 2 , and Wang, Z.Y. 1,2* 1 School of Biological and Chemical Engineering, Jiaxing University, Zhejiang Jiaxing, 314001, People’s Republic of China Date of Submission: 03-Sep-2009 Code Number: er10056 Abstract Although extensive experimental work has been carried out during the last several years, experimental reaction rate constants are available only for hundreds of compounds. Therefore, it is useful to develop a theoretical prediction method, which can be used to obtain estimates of the necessary kinetic parameters. One of the most successful approaches to predict chemical properties starting only from molecu-lar structural information is quantitative structure-activity/property relationships modeling (QSAR/QSPR). The purpose of this paper is to study the relationships between concentrations of 26 substituted phenols and reaction times during the ozonation process and determine the reaction orders and apparent reaction rate constants (-lgk΄). Then, optimized geometries of the substituted phenols were carried out at the B3LYP/6-311GFNx08 level using the Gaussian 03 software package. The structural and thermodynamic parameters ob-tained were taken as theoretical descriptors to establish a novel QSPR/QSAR model for -lgk΄ of the substi-tuted phenols, with a regression coefficient R = 0.909 and standard deviation SD = 0.141. Finally, the stability of the model for -lgk΄ predictions was checked by the t-test, showing satisfactory results. Results obtained reveal the reliability of QSPR/QSAR model for the prediction of ozone degradations rate constant of organic compounds.Keywords: Ozonation, Degradation rate, Density functional theory (DFT), Multiple linear regressions, Structural and thermodynamic parameters, E LUMO Introduction With the development of the chemical industry, substituted phenols have been increasingly used for the synthesis of drugs and other chemicals. Substi-tuted phenolic compounds, most of which are persis-tent and/or toxic organic pollutants, can be catego-rized as carcinogens, or malformation and mutation caus-ing substances. Among these compounds, chlorophe-nol has been considered as one of 129 controlled prior-ity pollutants by the U.S. Environmental Protection Agency (Keith, et al., 1979). In recent years, increasing attention has been paid to phenols in wastewater (Kuscu, et al., 2005; Fang, et al., 2006; Subramanyam, et al., 2007; Dalal, et al., 2007; Xie, et al., 2008). Be-cause the molecular structure of substituted phenols has a benzene ring with high chemical stability, it is difficult to completely degrade them by conventional biochemical or physical chemistry means. Ozonation is one of the most efficient technolo-gies for treating wastewaters. Due to intensive oxida-tion, ozone can degrade most organic compounds into CO 2 and H 2 O. A number of novel advanced ozonation processes have been developed to improve oxidation efficiency. Shen et al. (2008) studied the kinetics and mechanisms of degradation of p-chloronitrobenzene in water by ozonation and concluded that the phenols can undergo ring-opening reactions to produce low molecular carboxylic acids and finally CO 2 . The main intermediate products were phenol, p-chlorophenol, p-nitrophenol, 2-chloro-5-nitrophenol, 5-chloro-2-nitrophenol, 5-nitro-catechol, para-benzoquinone, 5-nitro-1, 2, 3-trihydroxy phenol, trihydroxy semiquinone and glycolic acid. Sanchez-Polo et al. (2007) compared the efficiency of UV photodegradation in combina-tion with various advanced oxidation processes (O 3 , UV/H 2 O 2 , O 3 /activated carbon) for the degradation of naphthalenesulfonic acids in aqueous solution and investigated the kinetics and the mechanisms involved in these processes. Chu et al. (2007) investigated the ozonation of synthetic wastewater containing an azo dye, CI Reactive Black 5, using a microbubble generator and a conventional bubble contactor. Harrison et al. (2007) explored the degradation mechanisms of cit-ronellal reactions with ozone and OH radicals. Gramatica et al. (1999) studied the tropospheric degra-dation of organic compounds by OH, NO 3 radicals and ozone and developed statistical models for predicting the oxidation rate constants of OH and NO 3 for many heterogeneous compounds by the quantitative struc-ture-activity relationship (QSAR) and quantitative structure-property relationship (QSPR) method. In ad-dition, QSPR/QSAR models were developed to predict degradation rate constants of tropospheric ozone and to study the degradation reactivity mechanism of 116 diverse compounds (Ren, et al., 2007). The aim of the present study was to analyze the ozonation efficiency and the relationship between degradation rate and the structure of substituted phenols. The apparent reaction rate constants (-lgk΄) for the ozone degradation of 26 common substituted phenols were measured for the first time in this study. In addition, optimized geometries of substituted phenols were carried out at the B3LYP/6-311GFNx08 level using the Gaussian 03 program. Finally, correlation of the model between the apparent reaction rate constants and calculated parameters was estab-lished by the QSPR/QSAR method. Materials and Methods Ozone was generated by an ozone generator (Jinghua Jianqiao Environmental Protection Science and Technique Co., Ltd., DJ-Q2020A, China). The ex-periments were conducted in a 250 mL three-necked flask. The initial concentration of substituted phenol was 5.00Χ10 -4 mol/L and the volume of solution was 100 mL. During experiments, ozone was continuously introduced into the reactor and maintained at a con-stant concentration (0.00118 mol/L). Excess ozone in the outlet gas was absorbed by 10% sodium thiosul-fate solution. All experiments were conducted at 298.15 K. During the reaction process, the concentrations of substituted phenols were detected after different reac-tion time periods by UV spectrophotometer at their maximum absorption wavelengths (Spectrumlab 752s, LengGuang. Tech., China). The reaction order and ap-parent reaction rate constants were obtained from the chemical reaction rate equation. At the same time, blank experiments with 4-nitrophenol and 2,3-dichlorophenol was carried by replacing ozone with the continuous introduction of air into the reactor at the same rate. After an equal time to the ozone degradation, we found that the concentrations of these two compounds were virtually unchanged. The results indicate that disap-pearance of substituted phenols was due to the reac-tion with ozone alone. All calculations for the 26 substituted phenols were carried out with the Gaussian 03 program. The geometries of all the substituted phenols were opti-mized at the B3LYP/6-311GFNx08 level and frequency cal-culations were performed to ensure they were at the potential energy surface minima. The structural and thermodynamic parameters were calculated. Structural parameters in this study included molecular volume (V i ) , molecular average polarizability (α), dipole mo-ment (μ), energy of the highest occupied molecular orbital(EHOMO ), energy of the lowest unoccupied mo-lecular orbital(E LUMO), the most negative atomic par-tial charge in molecule (q - ) and the most positive par-tial charge on a hydrogen atom (qH + ). Thermodynamic parameters calculated were as follows: standard en-thalpies (Hθ), standard Gibbs energies (Gθ), standard entropy (Sθ), standard heat capacities at constant vol-ume Cθv, and thermal correction to energy Eθth. To determine the optimum number of components for the correlation model, the leave-one-out (LOO) cross-validation procedure was used to validate the derived QSPR/QSAR model by the SPSS for Windows (version 12.0) software program. The quality of the derived QSPR/QSAR model was evaluated in terms of the LOO cross-validation correlation coefficient (q), the squared regression coefficient (R), the standard deviation (SD) and the t-test. Results and Discussion The reaction equation for ozone degradation of substituted phenols can be expressed as follows. Substituted phenols + O 3 -> Ps Based on Equation 1, the ozone degradation rate equa-tion can be presented as follows. -dC t /dt =KCtmCO3 n Equation 2 where C t (mol/L) and C O3 (mol/L) are the concen-trations of substituted phenols and ozone in aqueous solution, respectively, at reaction time t; k is the reac-tion rate constant, and m and n are the reaction orders of substituted phenols and ozone, respectively. Because the concentration of ozone was always saturated in the ozonation process, it can be regarded as a constant, assuming that C O3 has no in-fluence on the ozone diffusion rate under stirring in aqueous solution. Thus Equation 2 can be simplified as Equation 3. -dC t /dt = k ΄Ct m Equation 3 where k2 is an apparent reaction rate constant and m is then the total reaction order. When total reaction order (m) is zero, the reaction equation can be shown as Equation 4. C t = C0 - k΄t Equation 4 where C0 is the initial concentration of substituted phenol in the reaction system.If the total reaction or-der (m) is unity, the reaction equation can be shown as Equation 5. lg(C t /[C]) = lg(C0 /[C]) - k΄t Equation 5 where [C] is the unit concentration. The concentrations versus reaction time of ozone degradation for 26 substituted phenols were investi-gated during the ozonation processes and the experi-mental results for the four substituted phenols (1,4-dihydroxybenzene, 2-naphthol, 3-chlorophenol and 2-nitrophenol) are shown in [Figure - 1]. It can be seen that the concentration of substituted phenols decreased linearly with reaction time. Among these four com-pounds, the ozone degradation of 1,4-dihydroxybenzene was the fastest, while 2-naphthol was the slowest. Thus, it can be concluded that the aqueous ozone degradation reaction rate equation is in agreement with Equation 4. Therefore, the reaction order is zero and the apparent reaction rate constant (-k΄) is equal to the linear slope in [Figure - 1]. All substituted phenols and their calculated struc-tural parameters at the B3LYP/6-311GFNx08 level are listed in [Table - 1] and their calculated thermodynamic param-eters and apparent reaction rate constants (-lgk΄) are listed in [Table - 2]. Using the resulting structural and ther-modynamic parameters as variables, correlation equa-tions for the apparent rate constants were developed by multiple linear regressions with SPPS 12.0, in which the apparent rate constants were the independent vari-ables listed in [Table - 3]. The regression coefficients (R), standard deviations (SD), the regression coefficients of LOO cross-validation (q) and the Root Mean Square of Prediction (RMSEP) are also listed in [Table - 3]. The optimum equation was determined by com-paring the regression coefficients (R and q). As shown in [Table - 3], the values of R increased with the number of variables. Thus the four-variable Equation.9 was selected as the optimum equation with R = 0.909, SD = 0.141 and q = 0.856.The optimum equation contains four variables E LUMO , α, q - and S . Inspection of Equa-tion 9 may lead to the following interpretations: (1) - lgk΄ decreases with E LUMO . This is because reaction activity increases with the ELUMO value so it is easier for substituted phenols with larger E LUMO values to degrade than those with smaller E LUMO values. (2) The smaller the q - (the value is negative) value is, the smaller -lgk΄ will be. This is because if the charge is more nega-tive, the electron acting with ozone is more easily lost and therefore -lgk΄ decreases. (3) Furthermore,Sθ ex-presses the degree of disorder: the larger the degree of disorder, the larger the degradation ability. (4) In addi-tion, if a increases, -lgk΄ increases, i.e. the apparent rate constant decreases. The volume of the molecule increases with increasing a, the molecule is thus more stable and -lgk΄ increases.The predicted -lgk΄ of all the substituted phenols and the differences between them and experimental values are listed together in [Table - 2]. From [Table - 2], we can see the experimental values of -lgk΄ were close to the values predicted by Equation 9. The maximum deviation between the val-ues predicted by Equation 9 and the experimental val-ues is -0.290 for the compound 2,6-dinitrophenol and the second large difference is -0.262 for 3--methoxyphenol.The standard regression coefficients and t-values of the independent variables in Equation 9 are listed in [Table - 4]. The order of the standard re-gression coefficients is as follows: α > Sθ > E LUMO > q - .Thus, it can be concluded that α effects -lgk most strongly. Moreover, it can be seen that all t-values are larger than the standard t-value, indicating that all four variables are significant. Therefore, it can be concluded that the optimum equation (Equation 9) obtained in this study is robust. Furthermore, in order to check the reliability of the pre-dictive model developed in this study, the 26 substi-tuted phenols in [Table - 1] were divided into two groups: the first three compounds of every four in order were included in the first group (training sets) and the re-maining compounds were all included in the second group (external test sets). Using the same regression method as mentioned above, validation models that fitted the -lgk΄ values with the parameters of the 20 compounds in first group were created as shown in Equation 10. Equation 10 n = 20, R = 0.918, SD = 0.149, q = 0.833, RMSEP= 0.130 The RMSEP value of Equation 10 for the external test set (the second group) is 0.130. Moreover, the SD value of Equation 10 (0.149) is smaller and the values of R and q are similar to those obtained from Equation 9. t-values of the independent variables for Equation 10 are listed in [Table - 4]. indicating that all the values are larger than the standard t-value. These results confirm that the model obtained is reliable and has good pre-dictive ability. Conclusion Ozonation is one of the most efficient technolo-gies for treating wastewaters. However, the experimental determination of such reaction rate constants is diffi-cult, costly and time-consuming, and there are many uncertainties in chamber conditions. Therefore, reli-able theoretical models to estimate rate constants of the degradability of chemicals are strongly required. Among them, quantitative structure-activity/property relationships (QSAR/QSPR) study is a useful and ef-fective alternative approach to predict rate constants of this process. In this study, the ozone degradations of 26 substituted phenols in aqueous solutions were investigated at 298.15 K. The results show that the ozonation reaction order is zero and the apparent reac-tion rate constants of all substituted phenols were obtained from the chemical reaction rate equation. Based on the optimized geometries of substituted phenols, using the Gaussian 03 program, a novel QSPR/ QSAR model for apparent reaction rate constants (-lgk2) was developed by a multiple linear regression method. The optimum model (Equation 9) obtained in this work contains four variables E LUMO , q-, α and S·, for which the regression coefficient R = 0.909 and the standard deviation SD = 0.141. Furthermore, the opti-mum equation shows that -lgk2 increases with increas-ing q - and a and decreases with increasing E LUMO and S·. The results of the t-test indicate that the model exhibits optimum stability. Acknowledgements Financial support by the provincial Natural Sci-ence Foundation of Zhejiang (Y507280) is gratefully acknowledged.[12] References
Copyright 2010 - International Journal of Environmental Research The following images related to this document are available:Photo images[er10056t3.jpg] [er10056t2.jpg] [er10056t4.jpg] [er10056f1.jpg] [er10056t1.jpg] |
| |||||||||
