Xylidine-polluted groundwater purification. Ozonation and catalytic wet oxidation/Ksulidiiniga reostatud pohjavee puhastamine. Osoneerimine ja kataluutiline marghapendus.
Xylidine-polluted groundwater can be found in the sites of former Soviet military missile bases due to spills of a xylidine-based missile fuel--samine. The fuel consists of triethylamine (50 wt%) and xylidines or dimethylanilines (45 wt%) . The precise composition of the rocket fuel can be found in the literature . Similar contamination of groundwater occurs at former military bases in Barta and Tasi, West Latvia . Triethylamine is an easily volatile component and therefore we focused on the treatment of xylidines, the main unvolatile toxic components of samine.
Xylidines are toxic and mutagenic compounds. Typically, xylidines are a mixture of isomers in which the 2,4-, 2,5-, and 2,6-isomers are dominant. Molecular structures of isomers are presented in Fig. 1. They are relatively soluble, mobile, and persistent in groundwater . Based on the effects seen in animals, it is likely that xylidine can be absorbed through skin in sufficient amounts to cause dangerous methemoglobinemia . The maximally allowable concentration of aromatic amines (including xylidines) in drinking groundwater is 0.1 [micro]g/L .
Several investigations have been carried out on the purification of water contaminated with aromatic amines, including xylidines. Some of them are presented in our previous study .
Among physical wasterwater treatment methods, adsorption on granulated activated carbon (GAC) has been the most common treatment because of its simplicity, effectiveness, and relatively low price. The method is largely used in the purification of groundwater [8, 9] from organic pollutants. Despite the fact that adsorption gives a good purification effect, the pollution is not totally eliminated but only transferred from one medium to another. Rising costs, particularly those associated with the disposal of used GAC, are raising the competitiveness of chemical oxidation methods.
The main target of this study was to investigate the feasibility of the treatment of xylidine-polluted groundwater by means of chemical methods and to develop a theoretical basis for the design of the treatment processes. To achieve the target, two processes were experimentally studied and mathematically modelled: ozonation and catalytic wet oxidation (CWO) using GAC as a catalyst.
[FIGURE 1 OMITTED]
The reaction rate coefficients of 2,4-xylidine ozonation are not available in the literature. In order to design an ozone reactor for 2,4-xylidine destruction on a rational basis, these kinetic parameters must be evaluated from the laboratory experiments.
Another target was to evaluate the reaction kinetics of CWO in the presence of GAC. The CWO process could be used as a regeneration unit in the integrated adsorption process. The pollutant is concentrated on the surface of GAC and then degraded in the wet oxidation unit.
Ozonation is a well-established technology capable of destroying a wide range of organic molecules, including toxic organics. Ozone is a very reactive and powerful oxidizing agent ([E.sup.0] = 2.07 V) compared to other oxidizing agents, such as chlorine (1.36 V) and [H.sub.2][O.sub.2] (1.78 V) . A considerable disadvantage of ozone can be high production costs. Moreover, high capital cost of the ozonation unit makes ozonation technologies best suited to large-scale applications .
The reaction between ozone and aromatics includes the insertion of oxygen into the benzene ring, which results in the breaking of the double bonds. The opening of the benzene ring usually requires 1-2 moles of ozone per mole of aromatics .
Aromatic amines decompose to several by-products during ozonation (e.g., nitrobenzene and azobenzene, acetic and formic acid) [13-15] before mineralization. Most of the reaction products are biodegradable and less toxic than the reactants.
The kinetic parameters for reactions between xylidines and ozone are not available in the literature. In order to design a larger-scale ozone contactor (e.g., a bubble column, stirred tank reactor, etc.) for the treatment of polluted groundwater, the reaction rate coefficients should be known. In the present work, a wetted-wall reactor was used for evaluating the kinetic parameters of the reaction. The wetted-wall reactor provides an advantage in reaction kinetics research because of its precisely known mass transfer area and absence of foaming. However, a wetted-wall reactor can be used for fast reactions only; otherwise the contact time is too short for sufficient concentration changes to be observed. The coefficients of the ozonation reaction rate for aromatic amines show adequate values [16, 17] in order for the above-mentioned reactor system to be used.
A steady-state mathematical model that takes into account the mass transfer between phases, the reaction between ozone and xylidine, xylidine vaporization, the enhancement of the mass transfer due to the reaction between ozone and xylidine, and the stoichiometric ratio of the consumed ozone to the reacted xylidine was developed for the wetted-wall reactor.
The developed model and experimental set-up can be used in the future for the estimation of the kinetic parameters of reactions involving other similar compounds. The estimated rate coefficients could be used for the design of bubble ozonation columns. The values of the volumetric ozone mass-transfer coefficient can be easily obtained from the oxygen mass transfer data .
Catalytic wet oxidation
Among the treatments of nitrogen-containing organic compounds in water, wet air oxidation (WAO) appears as one of the most promising . The idea of the wet oxidation process is to enhance contact between molecular oxygen and the organic matter to be oxidized . Nevertheless, in order to obtain high conversions for pollutants, the non-catalytic techniques need severe operating conditions in terms of temperature and pressure. The CWO process with molecular oxygen using GAC at mild conditions (temperature below 200[degrees]C, pressure below 15 atm) reduces significantly the severity of the wet oxidation process and constitutes an economically attractive pretreatment step for non-biodegradable aqueous pollution due to its low capital and operating costs .
The CWO process can be carried out in a tubular packed-bed reactor operating in the trickle flow regime using air oxygen as an oxidant. In these conditions, chlorinated and nitroaromatic compounds (e.g. phenol) are oxidized over a solid catalyst in a three-phase system under mild conditions of oxygen partial pressure (6-12 atm) and temperature (120-160[degrees]C) . The activated carbon has been demonstrated to have catalytic effect on the wet oxidation of phenols. The oxidation can be carried out even in mild conditions of air pressure (4.7 atm) and temperature (40[degrees]C) with 48% of phenol conversion .
The different steps of the reaction mechanism of the wet oxidation of nitroaromatics involved in the process can be summarized as shown in Fig. 2 .
It clearly appears that acetic acid and ammonia are the most refractory compounds. Both oxides and supported noble metals can be used in the CWO of organic compounds. However, catalysts active in steps I and II could be different from those that are active and selective to N2 in step III. The end products are very resistant to total oxidation even under severe conditions and their oxidation to carbon dioxide is usually the rate-limiting step . A reaction mechanism similar to the nitroaromatics oxidation scheme (Fig. 2) is supposed to take place in the 2,4-xylidine degradation process; the first step for xylidine oxidation is probably just the formation of nitroaromatics.
[FIGURE 2 OMITTED]
Chemical oxidation methods, such as ozonation and catalytic wet oxidation in the presence of GAC, were studied experimentally. The following chapter gives a brief overview of the experimental part of the study.
All the experiments were carried out on a model solution (2,4-xylidine and distilled water). The 98% pure 2,4-xylidine solution was obtained from MERCK-Schuchardt (Germany).
The ozonation kinetics and mass transfer experiments were carried out on a wetted-wall reactor (Fig. 3). The height of the wetted-wall ozonation column (a glass tube) was 1 m and its inner diameter was 0.015 m. The column was carefully set in a vertical position and was equipped with special sections prior to the gas and liquid inlets and outlets in order to maintain a stable annular flow. The use of the glass tube made it possible to observe the complete wetting of the inner wall.
Ozonation experiments were carried out at room temperature (20[degrees]C). The operation mode of the reactor was countercurrent. To achieve a complete film distribution, the reactor was washed with a NaOH solution and rinsed with distilled water before runs. Polluted water was fed into the top of the reactor with the help of a peristaltic pump. Ozone was generated from pressured air by a laboratory ozone generator (Trailigaz "Labo 10"). The ozone-air mixture was fed into the system through a fluoroplast pipe from the bottom of the reactor. Sampling was carried out while the system was at a steady state. The liquid phase concentration of the 2,4-xylidine was measured inside the feed tank and at the outlet of the liquid. The ozone concentration was measured at the inlet and outlet of the gas phase. In mass transfer experiments (no reaction), the ozone concentration was measured additionally in a liquid effluent. The volumetric flow rate of the gas feed was measured by a rotameter. The pH was measured for all the samples before and after ozonation. The main values of the experimental parameters as well as experimental results are presented in . The experimentally observed parameters varied in a range presented in Table 1.
It was observed that the pH of the solution decreased during the ozonation (by about 1 unit), indicating that in the degradation process acids are formed. The colour of the liquid changed from colourless to slightly yellowish at the outlet. The HPLC analyses showed the appearance and decay of 18 peaks. These peaks present the ring-cleavage and final acidic products, which was confirmed by the biodegradability growth in tests. In addition to oxalic acid, the analysis showed the formation of propionic, formic, and acetic acids during the ozonation.
[FIGURE 3 OMITTED]
Catalytic wet oxidation
In the study of CWO the influence of the reaction conditions, viz. temperature and oxygen partial pressure, were evaluated. The CWO experiments were carried out in 0.3 L Parr Series 4561 Mini Reactor equipped with Series 4840 Temperature Controller (made in the USA). A scheme of the reactor system is presented in Fig. 4. The conditions of the CWO experiments are presented in Table 2.
In the CWO experiments, the commercial Granulated Acicated Carbon GAC NORIT RB 1 (made in The Netherlands) was used in adsorption and as the catalyst. High purity compressed oxygen was used as the oxidant (AGA-Finland).
Before the experiments, the reactor and sampling tube were cleaned with distilled water. The cold reactor was then charged with 100 g of 2,4-xylidine saturated wet GAC and 100 mL of residual liquid solution from adsorption experiments. The adsorption isoterms, xylidine diffusion coefficient in GAC pores, and breakthrough curves are presented in . The reactor was then sealed and heated to the appropriate test temperature and agitated at 900 rpm stirrer speed. Impeller rotation was set to eliminate any external mass transfer resistance. The thermal equilibrium was achieved in about 1 h after the desired temperature was entered into the temperature controller and the zero sample (5 mL) was taken by opening the liquid sampling valve. Then the oxygen inlet valve was opened and the oxygen pressure was adjusted to a desired level. The stopwatch was immediately started. Samples were taken throughout the run to assess the dependence of the reaction rate on 2,4-xylidine concentration after 10, 30, 60, and 120 min by closing the the [O.sub.2]-inlet valve and opening the liquid sampling valve. After sampling, the sample valve was closed and the [O.sub.2]-valve was re-opened, which flushed the sample tube at the same time.
[FIGURE 4 OMITTED]
The 2,4-xylidine concentration in the liquid phase was determined by a High Performance Liquid Chromatorgaph (Hewlett-Packard HP1100 series with UV and RI detectors) analyser. Analyses used MERCK LiChrosolv[R] HPLC grade methanol, analytical grade (N[H.sub.4])[H.sub.2]P[O.sub.4] and [(N[H.sub.4]).sub.2]HP[O.sub.4], and a 15 cm YMC-Pack Pro C18 column and method from YMC Co, Ltd (Japan). The ozone concentration in the gas phase was measured using a spectrophotometer (Unicam Helios Beta) and iodometrically in the liquid phase . The pH of the inlet and outlet liquids was measured using a Vernier PH-BTA pH sensor. In addition, the COD was measured with the Hach standard method [28a], using a Portable Datalogging Spectrophotometer Hach DR/2010 at wavelength 605 nm. [BOD.sub.7] was measured with a standard method [28b], using a Marvet Junior dissolved oxygen meter.
In CWO the total organic carbon (TOC) content was analysed by Shimadzu Total Organic Carbon Analyzer TOC-5050A.
MODELLING: RESULTS AND DISCUSSION
A steady-state wetted wall reactor model was developed taking into account the mass transfer between the phases (N), the reaction between ozone and 2,4-xylidine (R), the 2,4-xylidine vaporization, as well as the stoichiometric ratio of the consumed ozone to the reacted 2,4-xylidine. The enhancement of the ozone mass transfer, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], due to the reaction between ozone and 2,4-xylidine was calculated in the model using an equation based on the second-order reaction kinetics . The model was based on countercurrently connected and perfectly mixed mass transfer stages (Fig. 5).
[FIGURE 5 OMITTED]
The mass balances for the ith (i = 1 for ozone and i = 2 for 2,4-xylidine) component in the jth stage are:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)
for the liquid phase and
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2)
for the gas phase, where F is the cross-sectional area of the reactor, l is the height of the stage (l = L/n = 0.1 m, where L is the total length of the reactor and n is the number of stages), [H.sup.i] is Henry's dimensionless coefficient calculated for ozone according to [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] at 20[degrees]C) and for 2,4-xylidine obtained from the literature ([H.sup.xyl] = 0.27 ). The stiochiometric coefficient and enhancement factor for 2,4-xylidine are equal to one 1 ([m.sup.xyl] = 1 and [[phi].sup.xyl] = 1) The mass-transfer coefficient for 2,4-xylidine, [k.sup.xyl.sub.L], was calculated using the equation based on the penetration mass transfer theory in dilute solutions  and for ozone was estimated experimentally (Eq. 6) from the ozone mass-transfer experiments (no reaction).
The liquid hold-up, [h.sub.L] is calculated from the following equation:
[h.sub.L] = [pi]d [delta]/F, (3)
where d is the column's inner diameter, F is the cross-sectional area, and the liquid film thickness for the laminar flow, [delta], a is obtained from the following equation :
[delta] = [cube root of ([mu][LAMBDA]/g[[rho].sup.2])], (4)
where [LAMBDA] is the liquid mass flow rate per wetted perimeter, [mu] is the liquid dynamic viscosity, [rho] is the liquid density, and g is the gravitational acceleration.
The developed mathematical model was coupled with the parameter estimation program . First, the ozone mass-transfer coefficient, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] for the wetted wall column was estimated from the pure water-ozone mass-transfer experiments (self-decomposition of ozone was ignored). The objective function was used to minimize the differences between the observed and predicted ozone concentrations as follows:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (5)
where N is the number of observations.
The dependence of [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] [m/s] on the liquid flow rate [LAMBDA] in the laminar-wavy film flow range (0.01-0.09 kg/m s) in the wetted-wall reactor resulted in the following relation with the estimated parameters e and p:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (6)
The reaction rate and the stoichiometric coefficient for ozone [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] were estimated from the experimentally measured concentrations of [c.sup.xyl.sub.Lout] and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], obtained from the experiments. The sum of the residual squares
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (7)
was minimized in the estimation. The quality of the data fitting was checked with the degree of explanation, [R.sup.2] defined as follows:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (8)
The estimated parameters, 95% confidence limits, and the degree of explanation, [R.sup.2], are presented in Table 3.
The elaborated steady-state wetted-wall ozonation reactor model was able to describe the evolution of the pollutant and ozone concentrations in the liquid and gas phases along the reactor length coordinate in various conditions, as presented in Fig. 6.
The model for the wetted-wall reactor showed a good agreement when validated against experimental data (see Fig. 7).
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
Kinetic models for catalytic wet oxidation
Several completely different sets of kinetic models have been reported for CWO . In this study the reaction was modelled in terms of the Langmuir-Hinshelwood kinetic model and with the power law kinetics (second-order kinetic model). It can be noted that both models predict satisfactory 2,4-xylidine degradation in the CWO process. HPLC analysis showed the formation and decay of several intermediate compounds. The exact concentrations of these species were not measured and the CWO models were developed without intermediates.
The Langmuir-Hinshelwood kinetic model for CWO reaction is theoretically most justified. It assumes an irreversible surface reaction between the absorbed reactants (2,4-xylidine and oxygen). The reaction rate of xylidine is a function of the concentrations of the reacting species on the surface of the GAC. The surface reaction of chemisorbed 2,4-xylidine and oxygen is rate controlling as follows :
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (9)
where k is the reaction rate coefficient, and [K.sup.xyl] and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] are the adsorption coefficients of 2,4-xylidine and oxygen, respectively. The temperature dependence of the adsorption coefficient on higher temperatures is presented by the van't Hoff equation as follows:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (10)
where [K.sub.0] is the pre-exponential factor and [DELTA]H is the heat of adsorption.
The temperature dependence of the reaction rate coefficient, k, is determined by the Arrhenius equation:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (11)
where [k.sub.0] is the pre-exponential factor and [E.sub.a] is the activation energy. The oxygen concentration in the liquid phase was calculated using the following oxygen solubility equation :
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (12)
where [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the oxygen partial pressure.
In the optimization routine, the oxygen adsorption coefficient parameters, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], as well as the pre-exponential factor, [k.sub.0], and activation energy, [E.sub.a], for the Langmuir-Hinshelwood kinetic model (Eq. 9) were estimated. The values of parameters in the 2,4-xylidine adsorption coefficient (Eq. 10) in the temperature range of 415-450 K were separately estimated from the mass balances at the end of the pre-heating period before oxygen was added.
Power-rate law model
The simplest rate equation that would adequately fit the experimental data is the power-rate law model. The reaction kinetics is presented formally through a direct oxidation reaction in the liquid phase, including oxygen and 2,4-xylidine concentrations in the liquid phase. The rate of reaction is given by :
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (13)
where k is the reaction rate constant, [c.sup.xyl] and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] are the concentrations of 2,4-xylidine and dissolved oxygen in the liquid phase, and superscripts x and y are reaction orders.
This model is entirely empirical, and it does not take into account the adsorption on the catalyst surface in the heterogeneous CWO, which are included in Eq. 9. The equation is correct for the wet oxidation without catalyst. In the parameter estimation routine, the pre-exponential factor, [k.sub.0], activation energy, [E.sub.a], and reaction orders x and y were estimated.
The two models were implemented so that it was easy to link the model into the optimization algorithm of the Flowbat program framework . The Davidson optimization routine was used for parameter estimation. All the parameters except [K.sup.kyl.sub.0] and [DELTA][H.sup.xyl] were evaluated simultaneously.
On the basis of the 2,4-xylidine decay, the apparent activation energies and pre-exponential constants were determined for the reactions of the CWO and the WAO, with and without catalyst, using a parameter estimation program. The objective function was
[PHI] = [N.summation over (l=1)] [([c.sup.xyl,calc.sub.l] - [c.sup.xyl,exp.sub.l]).sup.2], (14)
where N is the total number of observations.
The results of the parameter estimation of the power-rate law and Langmuir-Hinshelwood kinetic model kinetics are presented in Tables 4 and 5.
As can be seen from Table 4, the reaction between 2,4-xylidine and oxygen is formally a typical second-order reaction, first-order in respect to both reactants. In WAO experiments, the oxygen partial pressure was kept constant and the oxidation reaction was first-order in respect to 2,4-xylidine.
A satisfactory fit was achieved with both the CWO models when validated against experimental data. The effect of temperature on the 2,4-xylidine degradation under constant oxygen pressure, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], is presented in Fig. 8 and the effect of oxygen partial pressure under constant temperature, T = 430 K, is presented in Fig. 9. Predictions for the WAO without catalyst in Fig. 8 were calculated from the power-law kinetic model (Eq. 13) and for the CWO, from the Langmuir-Hinshelwood kinetic model (Eq. 9). Predictions for the CWO by the power-law are presented in .
Catalytic wet oxidation under operating conditions completely oxidized 2,4-xylidine. The TOC degradation was significant but not 100% complete. The TOC reduction was incomplete because the 2,4-xylidine degradation pathway involves side reactions that lead to the formation of stable acids. The HPLC analysis indicated that over 20 intermediates were formed. Some of the organic compounds, such as propionic and formic acids, were determined throughout the analysis.
[FIGURE 8 OMITTED]
[FIGURE 9 OMITTED]
COMPARISON OF TREATMENT METHODS
Ozonation gives a good effect in destructing xylidines but complete mineralization is not feasible. The same concerns the CWO process because organic acids, which are formed throughout the oxidation process, are resistant to wet oxidation. In addition the ozonation process is feasible for large-scale applications whilst adsorption onto GAC with subsequent CWO can be economically used for smallscale purification plants.
Before CWO, the pollutant is adsorbed on the surface of GAC as the first step (see breaktrough curves in ). In the second step of CWO, the 2,4-xylidine is oxidized in the presence of dissolved oxygen and the GAC acts as a catalyst. In our experiments, the 2,4-xylidine was almost completely oxidized in the CWO reactor within an hour at relatively mild temperatures (140-175[degrees]C) and oxygen partial pressures (4-10 atm). Non-catalytic wet oxidation was not so effective for 2,4-xylidine removal.
It must still be born in mind that in the study a model solution was used in laboratory experiments and the kinetic parameters were estimated for the xylidines-distilled water system and therefore before real groundwater purification additional laboratory experiments must be carried out with in situ water, taking into account the deviations due to the water quality (additional organics, pH, etc.).
Xylidines are carcinogenic compounds originating from missile fuel and are dominating in polluted groundwater in the areas of abandoned Soviet missile bases in Eastern Europe. Two chemical methods for xylidine-polluted water purification were studied--ozonation and catalytic wet oxidation.
Ozonation of 2,4-xylidine in aqueous solution was carried out on a wetted-wall column. A laboratory-scale ozone contactor was designed and a steady-state wetted-wall reactor model was developed. The model was based on counter-currently connected and perfectly mixed mass transfer stages. It was possible to describe the changes of the pollutant and ozone concentrations along the reactor length coordinate in various conditions. The mathematical model elaborated for the wetted-wall reactor was observed to be in good agreement with the experimental data--the degree of explanation exceeded 96% for both the ozone mass transfer and the 2,4-xylidine ozonation data.
The developed wetted-wall reactor model is suitable for the investigation of ozonation kinetics for other organic compounds with similar nature. It is easy to include the intermediates into the reaction scheme, as the equation system for each stage of the ozonation column is solved numerically.
The catalytic wet oxidation with GAC as a catalyst at mild conditions (temperature below 200[degrees]C, total pressure below 15 bar) was experimentally studied. The 2,4-xylidine was adsorbed on the surface of the GAC and then degraded in the wet oxidation unit. The GAC type catalyst is active enough for 2,4-xylidine oxidation. The 2,4-xylidine was almost completely oxidized in the CWO reactor within an hour, while the intermediates such as ring cleavage byproducts and organic acids, which are formed throughout the oxidation process, are resistant to degradation. The non-catalytic wet oxidation did not degrade 2,4-xylidine at mild conditions and needs severer conditions.
Two kinetic models for predicting 2,4-xylidine degradation in the catalytic wet oxidation were studied. The Langmuir-Hinshelwood and the power-rate law kinetic models were both based on 2,4-xylidine decay. A satisfactory fit was achieved with both when validated against experimental data. The Langmuir-Hinshelwood kinetic model should be used in the reactor design as it has solid theoretical foundation.
In order to obtain more exact results, for the modelling of 2,4-xylidine degradation advanced CWO kinetic models that take into account the formation of intermediate products and their adsorption and desorption should be used. These intermediate products should be identified and quantitatively measured during the experiment. In this study only some of these intermediates were analytically identified (formic, acidic, and propionic acids).
Despite the formation of refractory compounds, the oxidation processes increase the biodegradability of the aqueous solution, since these intermediate products and organic acids are more biodegradable than 2,4-xylidine. A combination of chemical oxidation (ozonation, CWO) and biological treatment could be successfully used.
This study gives a mathematical tool and rate coefficients for designing two chemical methods for xylidine-polluted groundwater purification: ozonation in a bubble column and adsorption onto GAC NORIT RB1 in a packed bed reactor with a subsequent catalytic wet oxidation in the presence of GAC. The results and models developed can be used to design a purification process when only one contaminant is present. The same approach can also be used for real water containing many compounds if they are presented through one lumped parameter, such as COD, BOD, TOC, etc.
Received 3 February 2004, in revised form 5 March 2004
[1.] Raukas, A. (ed.) Endise Noukogude Liidu sojavae jaakreostus ja selle likvideerimine. EV Keskkonnaministeerium, Tallinn, 1999.
[2.] Rozkov, A., Vassiljeva, I., Kurvet, M., Kahru, A., Preis, S., Kharchenko, A., Krichevskaya, M., Liiv, M., Kaard, A. & Vilu, R. Laboratory study of bioremediation of rocket fuel-polluted groundwater. Water Res., 1999, 33, 1303-1313.
[3.] Avontis, J., Kokars, V., Malers, I., Valtere, S., Whittaker, H., Cooper, D. & Whittaker, R. Contamination and environmental remediation of former missile sites near Liepaja. Crit. Rev. Anal. Chem., 1998, 28, 97-100.
[4.] Pereira, W. E., Rostad, C. E. & Garbarino, J. R. Groundwater contamination by organic bases derived from coal tar wastes. Environ. Toxicol. Chem., 1983, 2, 283-294.
[5.] Gosselin, R. E., Smith, R. P. & Hodge, H. C. Clinical Toxicology of Commercial Products. 5th Edn. 1984, Williams & Wilkins, Baltimore, II-208.
[6.] EU Council Directive 98/83/EC of 3 November 1998 on the Quality of Water Intended for Human Consumption.
[7.] Reinik, J, Viiroja, A. & Kallas, J. Xylidine-polluted groundwater purification. Adsorption experiments and breakthrough calculations. Proc. Estonian Acad. Sci. Chem., 2001, 50, 205-216.
[8.] Shih, T. C., Wangpaichitr, M. & Suffet, M. Evaluation of granular activated carbon technology for the removal of methyl tertiary butyl ether (MTBE) from drinking water. Wat. Res., 2003, 37, 375-385.
[9.] Namasivayam, C. & Kavitha, D. Adsorptive removal of 2-chlorophenol by low-cost coir pith carbon. J. Haz. Mat., 2003, 98, 257-274.
[10.] LaGrega, M. D., Buckingham, P. L. & Evans, J. C. Hazardous Waste Management. McGraw-Hill Book Co., Singapore, 1994, 506-521.
[11.] Rodgers, J. D. & Bunce, N. J. Treatment methods for the remediation of nitroaromatic explosives. Water Res., 2001, 35, 2101-2111.
[12.] Hautaniemi, M., Kallas, J., Munter, R. & Trapido, M. Modelling of chlorophenol treatment in aqueous solutions. 1. Ozonation and ozonation combined with UV radiation under acidic conditions. Ozone Sci. Eng., 1998, 20, 259-282.
[13.] Sarasa, J., Cortes, S., Ormad, P., Gracia, R. & Ovelleiro, J. L. Study of the aromatic byproducts formed from ozonation of anilines in aqueous solution. Water Res., 2002, 36, 3035-3044.
[14.] Sauleda, R. & Brillas, E. Mineralization of aniline and 4-chlorophenol in acidic solution by ozonation catalyzed with Fe2+ and UVA light. Appl. Cat. B: Env., 2001, 29, 135-145.
[15.] Beltran-Heredia, J., Torregrosa, J., Dominguez, J. R. & Peres, J. A. Kinetics of the reaction between ozone and phenolic acids present in agro-industrial wastewaters. Water Res., 2001, 35, 1077-1085.
[16.] Hoigne, J. & Bader, H. Rate constants of reactions of ozone with organic and inorganic compounds in water. I. Non-dissociating organic compounds. Water Res., 1983, 17, 173-184.
[17.] Hoigne, J. & Bader, H. Rate constants of reactions of ozone with organic and inorganic compounds in water II. Dissociating organic compounds. Water Res., 1983, 17, 185-194.
[18.] Gottschalk, C., Libra, J. A. & Saupe, A. Ozonation of Water and Waste Water. Wiely-VCH, Weinheim, Germany, 2000, 104-106.
[19.] Zimmermann, F. J. & Diddams, D. G. The Zimmermann process and its application in the pulp and paper industry. TAPPI, 1960, 43, 710-711.
[20.] Deiber, G., Foussard, J. N. & Debellefontaine, H. Removal of nitrogeneous compounds by catalytic wet air oxidation. Kinetic study. Environ. Pollut., 1997, 96, 311-319.
[21.] Harf, J., Hug, A., Vogel, F. & von Rohr, P. R. Scale-up of Catalytic Wet Oxidation under moderate conditions. Environ. Progr., 1999, 18, 14-20.
[22.] Miro, C., Alejandre, A., Fortuny, A., Bengoa, C., Font, J. & Fabregat, A. Aqueous phase catalytic oxidation of phenol in a trickle bed reactor: effect of the pH. Wat. Res., 1999, 33, 1005-1013.
[23.] Fortuny, A., Font, J. & Fabregat, A. Wet air oxidation using active carbon as catalyst. Appl. Cat., 1998, 19, 165-173.
[24.] Oliviero, L., Barbier, J. Jr., & Duprez, D. Wet air oxidation of nitrogen-containing organic compounds and ammonia in aqueous media. Appl. Cat. B: Envir., 2003, 40, 163-184.
[25.] Mantazavinos, D., Livingston, A. G., Hellenbrand, R. & Metcalfe, I. S. Wet air oxidation of polyethylene glycols; mechanisms, intermediates and implications for integrated chemical-biological wastewater treartment. Chem. Eng. Sci., 1996, 51, 4219-4235.
[26.] Reinik, J., Jakobsson, K. & Kallas, J. 2,4-Xylidine degradation with ozonation: mass transfer and reaction kinetics. Ozone Sci. Eng. (in press).
[27.] Guidance Manual, Alternative Disinfectants and Oxidants, Ch. 3. EPA, 1999.
[28.] Standard Methods for the Examination of Water and Wastewater. American Public Health Association, Wachington, DC, 1995, (a) no. 5220, (b) no. 5210.
[29.] Dankwerts, P. V. Gas-Liquid Reactions. McGraw-Hill Book Company, 1970, 115-118.
[30.] Masschelein, W. J. Fundamental properties of ozone in relation to water sanitation and environmental applications. In Proc. IOA Specialised Symposium on Fundamental and Engineering Concepts for Ozone Reactor Design. Toulouse, France, 2000, 1-30.
[31.] Standards of Performance for New Stationary Sources: Volatile Organic Compound Emissions from the Synthetic Organic Chemical Manufacturing Industry Wastewater, Federal Register, Vol. 63, No. 236. EPA, 1998.
[32.] Coulson, J. M. & Richardson, J. F. Chemical Engineering, Vol. 1. Pergamon Press, Oxford, 1990, 468-488.
[33.] Treybal, R. E. Mass Transfer Operations. McGraw-Hill Book, Inc, New York, 1980, 50-54.
[34.] Keskinen, K. & Aitamaa, J. FLOWBAT User's Instruction Manual Version: 5.6. NESTE, Engineering Process Technology, Finland, 1998.
[35.] Hoeevar, S., Krasovec, O. U., Orel, B., Arico, A. S. & Kim, H. CWO of phenol on two differently prepared CuO-Ce[O.sub.2] catalysts. Appl. Cat., 2000, 28, 113-125.
[36.] Satterfield, C. N. Heterogenous Catalysis in Practice. McGraw-Hill, New York, 1980, 45-60.
[37.] Tromans, D. Temperature and pressure dependent solubility of oxygen in water: a thermodynamic analysis. Hydrometallurgy, 1998, 48, 327-342.
[38.] Lixiong, L., Chen, P. & Gloyna, E. F. Generalized kinetic model for wet oxidation of organic compounds. AICHE J., 1991, 37, 1687-1697.
[39.] Kallas, J. & Reinik, J. Xylidine-polluted groundwater purification. Wet oxidation--wet air oxidation using granulated active carbon. Sci. Proc. Riga Technical Univ. Ser.: Material Sci. Appl. Chem., 2002, 5, 208-215.
Janek Reinik (a,b) * and Juha Kallas (c)
(a) Department of Chemical Engineering, Tallinn University of Technology, Ehitajate tee 5, 19086 Tallinn, Estonia
(b) National Institute of Chemical Physics and Biophysics, Akadeemia tee 23, 12618 Tallinn, Estonia
(c) Department of Chemical Technology, Lappeenranta University of Technology, P.O. Box 20, FIN53851 Lappeenranta, Finland
* Corresponding author, firstname.lastname@example.org
Table 1. Experimentally observed variables Variable Notation Unit Range Solution pH pH -- 3.34-7.52 Liquid residence time t s 4.6-12.6 Liquid flow rate [Q.sub.L] L/min 0.04-0.18 Gas flow rate [Q.sub.G] L/min 1.2-2.1 2,4-Xylidine [c.sup.xyl.sub.Lin], mg/L 54-200, 45-182 concentration [c.sup.xyl.sub.Lout] Ozone concentration [MATHEMATICAL mg/L 13.2-22.7, EXPRESSION NO 9.6-20.3 REPRODUCIBLE IN ASCII] Biodegradability in [BOD.sub.7]/ -- 0-0.01 [COD.sub.in] Biodegradability out [BOD.sub.7]/ -- 0.05-0.26 [COD.sub.out] Table 2. Experimental conditions of the CWO experiments Variable Notation Unit Range Temperature T [degrees]C 140-170 Total pressure [P.sup.TOT] atm 10-15 Oxygen partial pressure [P.sup.O2] atm 4-10 2,4-Xylidine concentration [c.sup.Xyl] g/L 8-11 GAC concentration [c.sup.GAC] [g.sub.dry]/L 265 Table 3. Estimated parameters of 2,4-xylidine ozonation Parameter Unit Value 95% confidence limits Lower e 4.43 x [10.sup.-4] 1.61 x [10.sup.-5] p 0.411 0.287 [MATHEMATICAL L/(mol s) 0.316 x [10.sup.5] 1.67 x [10.sup.3] EXPRESSION NOT REPRODUCIBLE IN ASCII] [MATHEMATICAL mol/mol 2.45 1.74 EXPRESSION NOT REPRODUCIBLE IN ASCII] Parameter 95% confidence [R.sup.2] limits Upper e 7.25 x [10.sup.-5] 0.994 p 0.429 [MATHEMATICAL 4.65 x [10.sup.3] 0.961 EXPRESSION NOT REPRODUCIBLE IN ASCII] [MATHEMATICAL 3.16 EXPRESSION NOT REPRODUCIBLE IN ASCII] Table 4. Estimated parameters for catalytic wet oxidation Parameter Unit Value Model Langmuir- Hinshelwood, Eq. 9 [k.sub.0] mol/min 5.72 x [10.sup.3] Eq. 11 [E.sub.a] kJ/mol 31.3 Eq. 11 [MATHEMATICAL [m.sup.3]/mol 8.8 x [10.sup.-3] Eq. 10 EXPRESSION NOT REPRODUCIBLE IN ASCII] [MATHEMATICAL kJ/mol -13.2 Eq. 10 EXPRESSION NOT REPRODUCIBLE IN ASCII] [K.sub.0. [m.sup.3]/mol 3.45 x [10.sup.-3] Eq. 10 sup.xyl] [DELTA] kJ/mol -18.1 Eq. 10 [H.sup.xyl] Power-rate Model, Eq. 13 [k.sub.0] mol/min 80.3 Eq. 11 [E.sub.a] kJ/mol 5.59 Eq. 11 x -- 1.01 Eq. 13 y -- 1.00 Eq. 13 Table 5. Estimated parameters for wet air oxidation (experiments without catalyst) Parameter Unit Value Model Power-rate model, Eq. 13 [k.sub.0] 1/min 1.18 Eq. 11 [E.sub.a] kJ/mol 20.3 Eq. 11
|Printer friendly Cite/link Email Feedback|
|Author:||Reinik, Janek; Juha, Kallas|
|Publication:||Estonian Academy of Sciences: Chemistry|
|Date:||Sep 1, 2004|
|Previous Article:||Advances in petrol additives research/Edusammud bensiini lisandite uurimisel.|
|Next Article:||Kinetics of [[sup.3]H]WAY100635 binding to [5-HT.sub.1A] receptors in rat hippocampal membranes/[[sup.3]H]WAY100635 sidumise kineetika serotoniin 1A...|