Water Research 37 (2003) 319328

Catalytic wet oxidation of phenol in a trickle bed reactor over a Pt/TiO2 catalyst
Clayton B. Maugans, Aydin Akgerman*
Chemical Engineering Department, Texas A&M University, College Station, TX 77843-3122, USA Received 1 January 2002; received in revised form 1 June 2002; accepted 14 June 2002

Abstract Catalytic wet oxidation of phenol was studied in a batch and a trickle bed reactor using 4.45% Pt/TiO2 catalyst in the temperature range 1502051C. Kinetic data were obtained from batch reactor studies and used to model the reaction kinetics for phenol disappearance and for total organic carbon disappearance. Trickle bed experiments were then performed to generate data from a heterogeneous ow reactor. Catalyst deactivation was observed in the trickle bed reactor, although the exact cause was not determined. Deactivation was observed to linearly increase with the cumulative amount of phenol that had passed over the catalyst bed. Trickle bed reactor modeling was performed using a three-phase heterogeneous model. Model parameters were determined from literature correlations, batch derived kinetic data, and trickle bed derived catalyst deactivation data. The model equations were solved using orthogonal collocations on nite elements. Trickle bed performance was successfully predicted using the batch derived kinetic model and the three-phase reactor model. Thus, using the kinetics determined from limited data in the batch mode, it is possible to predict continuous ow multiphase reactor performance. r 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Phenol; Trickle bed reactors; Wet oxidation; Catalytic wet oxidation; Multiphase reactors

1. Introduction Organic compounds are common pollutants in industrial waste streams and waste sites. To remove the pollutants there are many remediation technologies, of which only reactive destruction results in the mineralization of the waste. Bioremediation and incineration are two typical examples of reactive destruction technologies. However, both these remediation technologies have their limitations and are not always the optimal solution. Bioremediation is ideally suited for low concentrations, with maximum concentrations as low as 50 ppm in some examples [1]. Incineration is ideal for highly concentrated liquid streams, typically around 350,000 ppm or higher chemical oxygen demand (COD)
*Corresponding author. Tel.: +1-979-845-3375; fax: +1979-845-6446. E-mail address: a-akgerman@tamu.edu (A. Akgerman).

content before the energy requirements become selfsustaining [2]. Neither process is optimal when the toxic organic compound concentration falls between the extremes. Some alternative destruction technologies have also been studied in the eld including wet air oxidation (WAO), and supercritical water oxidation (SCWO). WAO involves treating the aqueous organics with air at elevated temperature and pressures [312] and SCWO is the same, but occurring above the critical point of water [1320]. A natural extension to WAO is catalytic wet oxidation (CWO), which is similar to WAO but with the addition of a catalyst. The catalyst allows milder operating conditions than WAO while yielding similar if not superior kinetic performance. CWO research has been conducted on a variety of organic compounds using numerous catalysts with varying results [2139]. CWO is in use today as a treatment technology in certain applications such as coal gas efuent oxidation in Japan

0043-1354/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 3 - 1 3 5 4 ( 0 2 ) 0 0 2 8 9 - 0

320

C.B. Maugans, A. Akgerman / Water Research 37 (2003) 319328

Nomenclature Ar Ci Deff Ea hG hL H k kL a kLS K P r R rate RTP t T vt Arrhenius preexponential factor concentration of i (mol/L or ppm) effective diffusivity (m2/s) activation energy (J/mol) gas holdup liquid holdup Henrys constant (atm/(mol solute/mol solvent)) reaction rate constant (units vary) gasliquid interfacial mass transfer coefcient (1/min) liquidsolid interfacial mass transfer coefcient (L/min) reaction adsorption constant (L/mol) pressure (atm) radius (m) catalyst radius (m) reaction rate (mol/min/g-cat) room temperature and pressure time (min) temperature (K) total volumetric owrate

catalyst mass (g-cat)

Subscripts A reactive TOC organic compounds B unreactive TOC organic compounds bed bed G gas L liquid Oxy oxygen Ph phenol S solid TOC total organic carbon Superscripts 0 initial 0 dimensioness variable * catalyst site Greek Letters e porosity r density (g/L) x catalyst deactivation r2 second-order gradient

and Britain, as well as for cyanide wastewater oxidation in Japan [27], and for destroying organic undesirables in an ammonium sulfate crystal production facility in Italy [9]. In this research heterogeneous CWO was studied as an organic wastewater treatment technology using phenol as a model compound. Phenol was selected because it is a common industrial pollutant, is toxic, and is difcult to biodegrade at high-concentration levels. This model compound was studied in two different reactor congurations; rst a batch reactor was used to determine technology viability as well as determination of the reaction kinetics, then a ow reactor system was developed primarily for model conrmation and scaleup, but also to evaluate this technology in a industrially more practical ow reactor system. One of the objectives of this study was to determine whether trickle bed reactor operation can be predicted from kinetics in terms of lumped parameters, such as total organic carbon (TOC), determined in a batch reactor by a limited number of experiments.

Maugans and Akgerman [30]. A reevaluation of those data in this paper suggests that Eq. (1) is a more suitable kinetic model. The rate expression was rederived using the following mechanism based on reaction between adsorbed phenol and disassociatively adsorbed oxygen. I: Ph S "PhS ; II: O2 2S "2OS ; III: IV: PhS OS -AS S ; AS "A S :

If it is assumed that the surface reaction, reaction III is the rate determining step, all other reactions are in dynamic equilibrium, and rate IV is very fast, then the following rate expression is obtained: qCPh ratePh;intrinsic   W qt VL

p COxy p 2 : 1 KPh CPh KOxy COxy K1 CPh

2. Reaction kinetics from batch experiments: phenol destruction Batch experiments originally performed using powdered catalyst yielded the kinetic expression reported in

Parameters were determined from the original batch reactor data reported by Maugans and Akgerman [3031] and are given in Table 1. Fig. 1 shows prediction of experimental phenol conversion in the batch reactor with Eq. (1).

C.B. Maugans, A. Akgerman / Water Research 37 (2003) 319328

321

3. Reaction kinetics from batch experiments: TOC destruction The overall TOC reduction mechanism was described in terms of a parallel reaction scheme, which reduces to the reactions below [30]. A - B krateX CO2 : In this reaction scheme, A denotes phenol and all other intermediates produced from the oxidation of phenol that further oxidize into refractory components B (certain organic acids) or completely mineralize into CO2. The initial value of A is known, since it is the same as the initial TOC value. The initial value of B is assumed to be 0 at time=0. The TOC reaction scheme
rateY

reduces to Eqs. (2) and (3) as shown by Maugans [31]. qCTOC rateTOC;intrinsic   W qt VL

! p KX p 2 CA COxy ; 2 1 KOxy COxy

where CA concentration of phenol and all other intermediates that further oxidize, is determined by rateA;intrinsic  qCA  W qt VL

! p KX p 2 KY CA COxy 1 KOxy COxy 3

0 CA

Table 1 Apparent Arrhenius constants for Eqs. (2)(4) Ari K1 KPh KOxy KX KY 8.52 10 (L g-cat/min/mol ) (73.0 1012) 5.76 1012 (L/mol) (74.2 1011) 3.09 108 (L/mol) (71.5 108) 2.94 1014 (L1.5 g-cat/min/mol0.5) (71.4 1014) 6.42 107 (L1.5 g-cat/min/mol0.5) (73.0 107)
12 1.5 0.5

and the initial conditions (initial phenol concentration) and CTOC [0] are known (and equal). Kinetic parameters for Eqs. (1)(3) are also given in Table 1 and Fig. 2 shows the prediction of the measured TOC proles in the batch reactor with Eqs. (2) and (3).

Eai (J/mol) 110,000 (722,000) 100,000 (726,000) 67,000 (716,000) 125,000 (72300) 79,000 (72400)

4. Trickle bed reactor studies 4.1. Materials Phenol solution (90 wt% phenol 10 wt% water) was obtained from Fisher Scientic. The catalyst was powdered 4.45% Pt/TiO2 with a maximum diameter of less than 105 mm, obtained from Engelhard. The binder used with the catalyst, as described below, was a 40% SiO2 solution of Ludoxt Binder. The oxygen source in

1 0.9 0.8 0.7 150 C 165 C 175 C 185 C 200 C

1-Conversion

0.6 0.5 0.4 0.3 0.2 0.1 0 0 20 40 60

80

100

120

Time (min)

Fig. 1. Batch reactor data, phenol conversion predictions with the kinetic model at different temperatures (P 48 atm, W 2 g-cat/L, 0 Cph 1150 ppm).

322

C.B. Maugans, A. Akgerman / Water Research 37 (2003) 319328

1 0.9 0.8 0.7 150 C 165 C 175 C 185 C 200 C

1-Conversion

0.6 0.5 0.4 0.3 0.2 0.1 0 0 20 40 60

80

100

120

Time (min)

Fig. 2. Batch reactor data, TOC conversion predictions with the kinetic model at different temperatures (P 48 atm, W 2 g-cat/L. 0 Cph 1150 ppm).

these experiments was compressed air provided by Brazos Valley Welding. All water used in experimentation and cleaning was distilled and deionized water. TOC analysis reagents included J.T. Baker sodium persulfate and EM 98% phosphoric acid. HPLC analyses employed EM HPLC grade methanol and a 15 cm Supelcosil LC-8 column from Supelco. All glass beads used in the reactor were Pyrext. 4.2. Equipment Experiments were conducted in a trickle bed ow reactor schematically represented in Fig. 3. All tubing and ttings used were stainless steel. The gas ow rate was adjusted and controlled by a mass ow controller. The main pump (for water) was a variable speed, reciprocating pump while the metering pump (for phenol solution) was a smaller variable speed, reciprocating pump (Minipump). Heating was provided by electric heaters located external to the reactor, which heated a bed of aluminum beads in which the reactor was immersed. Reactor temperature was monitored by a sliding thermocouple in a thermowell. Two diaphragmtype backpressure regulators (BPR) were used in series to control reactor pressure. System pressure was monitored with a Bordon tube pressure gauge. The reactor efuent was cooled before expansion through the BPRs using tap water and a double pipe heat exchanger. 4.3. Experimental procedure Deionized water was pumped at a constant ow rate via the main pump. Into this ow stream, a stream of

concentrated phenol was added through the MiniPump. Both pumps were reciprocating piston pumps, so to dampen out pulses the feed stream was sent through a 1/ 1600 tubing coil. The feed was then preheated in a 1/800 preheater coil immersed in the hot aluminum bed surrounding the reactor. Air was used as the oxygen source and owed from the pressurized air tank, through the mass ow controller, through the preheater, and into the top of the reactor where it was mixed with the phenol solution feed before entering the reactor. The reactor conguration was such that the feed passed over a bed of inert glass beads before entering the catalyst zone. After exiting the reactor the efuent was cooled and depressurized. The reactor was loaded with 2.29 g of pelletized catalyst and run continuously until the completion of the ow reactor studies. During extended periods when no data were taken, water was run through the reactor to avoid disruptive cooling/heating and reactor shutdown. Liquid owrate was maintained constant at 12 g/min and the air inlet owrate was maintained at 171 mL/min (measured at room temperature and pressure). The owrates were kept constant throughout the trickle bed studies in order to maintain consistent hydrodynamics in the reactor to eliminate the effect of reactor hydrodynamics on conversion. Space time was controlled by adjusting the concentration of the phenol feed. Pressure was maintained at approximately 4176 atm. The temperature range studied was from 1501C to 2051C and the feed concentration range varied from 200 to 1200 ppm phenol. Because it was found that the temperature prole along the catalyst bed of the reactor was constant (711C), the thermocouple was left in place at the bottom of the catalyst bed for all experiments.

C.B. Maugans, A. Akgerman / Water Research 37 (2003) 319328

323

Fig. 3. Trickle bed reactor system.

4.4. Catalyst preparation The research catalyst was supplied in a ne powder form and unsuitable for xed bed reactor experiments. To pelletize the powder, 5.5 g of catalyst powder was mixed with 6 mL of Ludoxt binder. The resulting paste was then extruded, dried, crushed, sieved, and calcined overnight at 3201C. This relatively low calcination temperature was chosen since signicant catalyst deactivation was observed when calcining at temperatures near 5001C. The pellets were then tested in high temperature water for physical stability and structural integrity. Below the boiling point of water the catalyst was observed to be unaffected by the hot water. At boiling, the catalyst quickly disintegrated into a powdery form, due to the formation and ashing of water vapor inside the pellet pores. Table 2 gives the physical properties of the catalyst. In addition to Table 2, the following physical properties were measured: median catalyst particle size of 450 mm, bed density of 0.79 g/mL, bed porosity of 0.68, and solid material density of 2.4 g/(mL solid mater). The pelletized catalyst contained 0.64 g of 4.45% Pt/TiO2 powdered catalyst per gram of pellets (the remainder being binder). 4.5. Homogeneous reactions In the ow reactor system the reactants were in a high temperature reaction zone for a brief period of time

Table 2 Catalyst surface areas Stock powder Single point surface area P/P0 (m2/g) BET surface area (m2/g) BJH adsorption (m2/g) MJH desorption (m2/g) Cumulative pore V (m3/g) Avg. pore diameter ( (A) LUDOX bound pellets 1101C dried 52.7 51.7 56.1 63.9 71.0 0.19 116 0.18 118 Calcined 50.4 53.7 62.2

before and after the catalyst bed. This had the potential of introducing error to the kinetic study as there was the possibility of signicant homogeneous wet oxidation occurring outside the catalyst bed. Experiments were therefore performed in which the catalyst bed was replaced with similar sized inert glass beads. The reactor was operated at 1851C, 41 atm, 12 g feed/min, 172 mL (RTP)/min air, and a feed concentration of 417 ppm phenol corresponding to 319 ppm TOC. Although there was about 10% phenol conversion to certain

324
350

C.B. Maugans, A. Akgerman / Water Research 37 (2003) 319328

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0
0 50 100 150 200

TOC Conc. (ppm)

300 250 200 150 100 50 0

Exit Inlet

Conversion

Phenol TOC Linear (TOC)

50

100 g Phenol Passed

150

Time on Stream (min)

Fig. 4. Extent of homogeneous (non-catalytic) reaction in trickle bed reactor, measured in terms of TOC reduction.

Fig. 5. Catalyst deactivation at base condition (T 1751C, 0 =460 ppm). P 44 atm, vt;L 12 g/min, Cph

intermediates, there was no reduction in TOC concentration (Fig. 4) indicating that homogeneous oxidation was not signicant at the reaction conditions. The slight reduction from the theoretical value of 319 ppm TOC to measured value of 300 ppm TOC in Fig. 4 is probably due phenol conversion into intermediates, which have less carbon mass fraction compared to phenol.

6. Mathematical modeling: phenol equations A trickle bed reactor is a three-phase reactor, and it was assumed that the reaction was only occurring on the solid phase surface. It was necessary for the reactants to transfer into the solid catalytic phase for reaction to occur. To model this transfer a mass balance for each of these phases was performed. Derivations can be found in [31]. Gas phase:
0 hG rbed vt;G P0 Oxy dPOxy

5. Results: catalyst deactivation Catalyst deactivation was observed in this study, although the exact cause is uncertain. Fig. 5 shows reactor performance at the reference condition throughout the course of a continuous reactor study. The reference condition was 1751C, 44 atm pressure and 460 ppm initial phenol condition, the reaction was repeated at this condition periodically. Fig. 5 shows a linear decrease in conversion with respect to the cumulative amount of phenol passed over the catalyst bed. A linear regression of the above deactivation data was performed and used in conjunction with the rate expressions in the trickle bed model to account for catalyst deactivation as shown in Eq. (4). ratei 0:641 xratei;intrinsic ; 4

R1 T Wtotal kL a

dW 0
0 P0 Oxy POxy

!
0 0 COxy ;L COxy

HOxy rmolar;water

0;

liquid phase:
0 hL vt;L rbed C0 Oxy dCOxy Wtotal dW 0 ! 0 POxy P0Oxy 0 0 kL a COxy;L COxy HOxy rmolar;water  0 0  3fDeff ;Oxy 1 eS COxy dCOxy  0; 2 0 R dr rR 0 0 hL vt;L rbed CPh dCPh Wtotal dW 0  0 0  3fDeff ;Ph 1 eS CPh dCPh  0; R2 dr0 rR

where the intrinsic rate expressions are those given by Eqs. (1)(3). x was found to be 0.0055 cumulative grams of phenol passed through the reactor. The factor 0.64 was included to account for the catalyst dilution due to the binder. Therefore: ratePh 0:64 1 x ratePh;intrinsic ; rateOxy 0:64 1 x rateoxy;intrinsic 6ratePh : 5 6

solid phase:
0 Deff CPh 0 r2 Cph rS ratePh 0; R 0 Deff COxy

10

Eq. (6) was based on the assumption that 6 mol of oxygen were consumed per mole of phenol. This was based on TOC reduction rather than complete mineralization.

0 rS rateOxy 0: r2 COxy

11

Experimentally no temperature gradient along the length of the reactor was observed. An internal energy balance on the pellets showed no signicant temperature

C.B. Maugans, A. Akgerman / Water Research 37 (2003) 319328

325

increase due to the heat of reaction on the pellet scale, therefore the system was considered isothermal. Eqs. (7)(11) were solved simultaneously, using Eqs. (1), (5) and (6), and by the application of orthogonal collocation on nite elements (OCFE). Five hundred nite elements were used with 4 collocation points used for the gas and liquid phase and 6 collocation points used for the solid phase [31,40].

using Eqs. (2)(5) and 6, and by the application of OCFE.

8. Parameters The equations solved for modeling trickle bed performance involved a large number of parameters, many of which were unknown for this reactor system and were therefore determined from published correlations. Molecular diffusivity was calculated using the SiddiqiLucas [41] method. Liquid and gas holdup were determined by calculating the liquid fraction of the void space using a correlation developed by Meng and Chung [42] and the measured bed porosity. The wetting efciency was determined based on a correlation published by Herskowitz and Smith [43]. Analysis of the wetting efciency showed it to be greater then 0.97 at all times, allowing the assumption of complete surface wetting to be made. The reactor gasliquid mass transfer coefcient, kL a was determined by averaging the kL as as calculated by the correlations proposed by Wild [44] and Midoux [45]. The Ellman correlation [46] was used to calculate the pressure drop as a parameter necessary to the Midox correlation.

7. Mathematical modeling: TOC equations The mass balance equations for the TOC system were: liquid phase:
0 hL vt;L rbed CTOC ;L planar Aplanar CTOC Wtotal 0 3fDeff ;TOC 1 eS CTOC ;L R2  spherical  Aspherical CTOC  0; rR 0 hL vt;L rbed CA ;L planar Aplanar CA Wtotal 0  3fDeff ;A 1 eS CA ;L spherical  A C  0; spherical A R2 rR 0 hL vt;L rbed COxy ;L planar Aplanar COxy kL a Wtotal ! 0 P0 Oxy POxy 0 0 COxy;L COxy HOxy rmolar;water

12

13

9. Comparison of experimental data and model predictions Using a Fortran program, Eqs. (7)(11) were solved simultaneously along the bed of the reactor with the exit proles of each nite element used as the inlet proles for each successive element in the reactor. A total of 500 nite elements along the bed of the reactor were used for each nal efuent data point simulation. According to the model the catalyst pellets frequently became oxygen decient towards the center of the pellets, resulting in no reaction occurring in the oxygen decient zones. Hydrolysis reactions, although unlikely, may occur under such conditions; however, it was assumed in the model that no reactions occur in the absence of oxygen. Final calculated efuent conversions are compared to the experimental values for phenol oxidation in the trickle bed reactor and can be seen in Figs. 6 and 7. Similarly overall TOC destruction values for the same experiments were modeled by Eqs. (12)(18) and are shown in Figs. 8 and 9. As can be seen in Figs. 69, the conversion was a function of temperature as well as space time, where space time was calculated as g-catalyst/(g-phenol/min). By using this space time denition it was possible to vary space time without adjusting the mass owrate of the aqueous feed, but by adjusting the concentration of the feed. On each chart, as space time increases the

0 3fDeff ;Oxy 1 eS COxy ;L

R2

 spherical  Aspherical COxy 

rR

0; 14

gas phase: hG rbed vt;G P0 Oxy Aplanar Pplanar Oxy R1 T Wtotal ! 0 0 POxy POxy 0 0 kL a COxy;L COxy 0: HOxy rmolar;water

15

solid phase:
0 Deff ;TOC CTOC spherical Bspherical CTOC rS rateTOC 0; R 0 Deff ;A CA spherical Bspherical CA rS rateA 0; R 0 Deff ;Oxy COxy

16 17

spherical rS rateOxy 0; 18 Bspherical COxy R where rateTOC and rateA are represented analogous to Eq. (5). For the TOC calculations rateOxy is also represented by rateTOC since oxygen reaction was assumed stoichiometrically one to one with TOC. Similarly Eq. (12)(18) were solved simultaneously,

326
1 0.9 0.8 0.7 Model Actual

C.B. Maugans, A. Akgerman / Water Research 37 (2003) 319328

1 0.9 0.8 0.7
Model Actual

Conversion

Conversion
170 190 210 230 250 270 290

0.6 0.5 0.4 0.3

0.6 0.5 0.4 0.3

0.2 0.1 0 150

0.2 0.1 0 0 100 200 300 400 500 600

Space Time (g-cat/(g-phenol/min))

Space Time (g-cat/(g-phenol/min))

Fig. 6. Phenol conversion in a trickle bed reactor at 2032071C. Fig. 9. TOC conversion in a trickle bed reactor at 1931971C.
1 0.9 0.8 0.7
Model Actual

0.6 0.5 0.4 0.3 0.2 0.1 0 50 100 150 200 250 300 350 400

the predicted data points from the model would precisely overlap each other for the same condition, but in this case they do not. The reason for this is due to the catalyst bed deactivating with time and this deactivation, while not indicated on the two-dimensional chart, is accounted for in the reactor model. This result also applies to Figs. 79.

Conversion

10. Summary and conclusions Based on batch reactor experiments a kinetic model was developed to model the catalytic reaction for phenol oxidation, as well as another model for overall TOC destruction. A trickle bed reactor was operated with pelletized catalyst in order to generate performance data from a three-phase ow reactor. The technology was able to achieve up to 80% TOC destruction in the tested conguration. Predicted results are then compared with the experimental data. The complex three-phase reactor was successfully modeled without the use of data generated in the ow reactor (with the exception of catalyst deactivation). Predictions for phenol oxidation using batch derived kinetics were successfully applied to a complex threephase trickle bed reactor over a temperature range from 1502051C and a space velocity range from 150 to 650 gcat/(g-phenol/min). Additionally a lumped TOC model was developed that successfully predicted batch as well as ow reactor performance. It can be concluded that the three-phase trickle bed reactor can be modeled based on batch derived kinetic data and established phase behavior correlations.

Space time (g-cat/(g-phenol/min))

Fig. 7. Phenol conversion in a trickle bed reactor at 1931971C.

1 0.9 0.8 0.7 Model Actual

Conversion

0.6 0.5 0.4 0.3 0.2 0.1 0 150 200 250 300 350 400 450

Space Time (g-cat/(g-phenol/min))

Fig. 8. TOC conversion in a trickle bed reactor at 2032071C.

conversion of phenol and TOC also increases. As expected, increasing temperature (by moving from Figs. 69) the reactant conversion also increases. In Fig. 6, for example, it can be seen that two of the conditions tested were repeat experiments and two different experimental data points are noted for each condition. As expected, the two experimental ndings for reaction conversions are in close agreement, but do not exactly overlap. It is typically expected however that

Acknowledgements This project has been funded by Grants 104TAM2358 and 026TAM3358 in part with Federal Funds as part of

C.B. Maugans, A. Akgerman / Water Research 37 (2003) 319328

327

the program of the Gulf Coast Hazardous Substance Research Center which is supported under cooperative agreement R815197 with the United States Environmental Protection Agency and in part with funds from the State of Texas as part of the program of the Texas Hazardous Waste Research Center. The contents do not necessarily reect the views and policies of the US EPA or the State of Texas nor does the mention of trade names or commercial product constitute endorsement or recommendation for use.

References
[1] Autenrieth RL, Bonner JS, Akgerman A, Okaygun M, McCreary EM. Biodegradation of phenolic wastes. J Hazardous Mater 1991;28:29. [2] Imamura S. Catalytic and noncatalytic wet oxidation. Ind Eng Chem Res 1999;38:1743. [3] Carlos TMS, Maugans CB. Wet air oxidation of renery spent caustic: a renery case study. Proceedings of the NPRA Environmental Conference, San Antonio, TX, 2000. [4] Copa WM, Momont JA. In: H. M. Freeman, editor. Section 8.6, wet oxidation, standard handbook of hazardous waste treatment and disposal. New York, NY: McGraw-Hill, 1986. p. 8.7993. [5] Copa WM, Randall TL, Welhelmi AR. Wet air oxidation of hazardous wastes. In: Cheremisinoff PN, editor. Encyclopedia of environmental control technology. Houston, TX: Gulf Publishing Company, 1989. p. 31432 [Chapter 11]. [6] Dietrich MJ, Randall TL, Canney PJ. Wet air oxidation of hazardous organics in wastewater. Environ Prog 1985;4:171. [7] Jackman AP, Powell RL, Hazardous waste treatment technologies. Park Ridge, NJ: Noyes Publications, 1991. p. 90134. [8] Gitchel WB, Meidl JA, Burant W. Carbon regeneration by wet air oxidation. Chem Eng Prog 1975;71:90. [9] Giudici D, Maugans CB, Integration of wet air oxidation (WAO) into the ammonium sulfate crystallization process for increased recovery and elimination of waste disposal: a case study. Proceedings of the Third European Conference of Chemical Engineering, Nurnberg, Germany, 2001. [10] Teletzke GH. Wet air oxidation. Chem Eng Prog 1964;60:33. [11] Mishra VS, Mahajani VV, Joshi JB. Wet air oxidation. Ind Eng Chem Res 1995;34:248. [12] Zimmermann FJ. New waste disposal process. Chem Eng 1958;25:117. [13] Abeln, J, et al. Supercritical water oxidation (SCWO): a process for the treatment of industrial waste efuents. Proceedings of the 19th IT3 Conference, Portland, OR, 1999. [14] Cocero MJ, Alonso E, Fdz-Polanco F. Supercritical water oxidation of wastewater and sludges. Proceedings of the Third European Conference of Chemical Engineering, Nurnberg, Germany, 2001.

[15] Crooker PJ, et al. , Operating results from supercritical water oxidation plants. Ind Eng Chem Res 2000;39: 4865. [16] Gopalan S, Savage PE. Reaction mechanism for phenol oxidation in supercritical water. J Phys Chem 1994; 98:12646. [17] Martino CJ, Savage PE. Total organic carbon disappearance kinetics for the supercritical water oxidation of monosubstituted phenols. Environ Sci Tech 1999;33:1911. [18] Savage PE. Organic chemical reactions in supercritical water. Chem Rev 1999;99:603. [19] Schmieder H, Abeln J. Supercritical water oxidation: state of the art. Proceedings of the 19th IT3 Conference, Portland, OR, 1999. [20] Shaw RW, et al. Supercritical water: a medium for chemistry. Chem Eng News 1991;69(51):26. [21] Akyurtlu JF, Akyurtlu A, Kovnklioglu S. Catalytic oxidation of phenol in aqueous solutions. Catal Today 1998;40:343. [22] Fortuny A, et al. Water pollution abatement by catalytic wet air oxidation in a trickle bed reactor. Catal Today 1999;53:107. [23] Hamoudi S, Belkacemi K, Larachi F. Catalytic oxidaiton of aqueous phenolic solutions catalyst deactivation and kinetics. Chem Eng Sci 1999;54:3569. [24] Imamura S. Catalytic and noncatalytic wet oxidation. Ind Eng Chem Res 1999;38:1743. [25] Ishii T, et al. Method for purication of waste water. US Patent 5,158,689, 1992. [26] Levec J, Smith JM. Oxidation of acetic acid solutions in a trickle-bed reactor. AIChE J 1976;22:159. [27] Luck F. A review of industrial catalytic wet air oxidation processes. Catal Today 1996;27:195. [28] Luck F, Djafer M., Rose J-P, Cretenot D. Athoss: a novel process for sludge disposal. Proceedings of the AWWA 18th Federal Convention, 1999. [29] Matatov-Meytal YI, Sheintuch M. Catalytic abatement of water pollutants. Ind Eng Chem Res 1998;37:309. [30] Maugans CB, Akgerman A. Catalytic wet oxidation of phenol over a Pt/TiO2 catalyst. Water Res 1997;31:3116. [31] Maugans CB, Catalytic wet oxidation of phenol in batch and trickle bed reactors over a Pt/TiO2 catalyst. Dissertation, Texas, A & M University, December 1999. [32] Taguchi J, Okuhara T. Selective oxidative decomposition of ammonia in neutral water to nitrogen over titaniasupported platinum or palladium catalyst. Appl Catal A 2000;194:89. [33] Pintar A, Besson M, Gellezot P. Catalytic wet air oxidation of Kraft bleaching plant efuents in the presence of titania and zirconia supported ruthenium. Appl Catal B 2000;30:123. [34] Pintar AM, Besson P. Gellezot catalytic wet air oxidation of Kraft bleaching plant efuents in a trickle-bed reactor. Proceedings of the Third European Conference of Chemical Engineering, Nurnberg, Germany, 2001. [35] Pintar A, Levec J. Catalytic liquid-phase oxidation of refractory organics in waste water. Chem Eng Sci 1992; 47:2395. [36] Pintar A, Levec J. Catalytic liquid phase oxidation of phenol aqueous solution. A kinetic investigation. Ind Eng Chem Res 1994;33:3070.

328

C.B. Maugans, A. Akgerman / Water Research 37 (2003) 319328 [43] Herskowitz M, Smith JM. Trickle bed reactors: a review. AIChE J 1983;29:1. [44] Wild G, Larachi F, Charpentier JC. Heat and mass transfer in gasliquid-sold xed bed reactors. In: Quintard M, Todorovic M, editors. Heat and mass transfer in porous media. Elsevier: Amsterdam, The Netherlands, 1992. p. 616. [45] Midoux N, Morsi BI, Purwasasmita M, Laurent A, Charpentier JC. Interfacial area and liquid side mass transfer coefcient in trickle bed reactors operating with organic liquids. Chem Eng Sci 1984;39:781. [46] Ellman MJ, Midoux N, Laurent A, Charpentier JC. A new, improved pressure drop correlation for trickle-bed reactors. Chem Eng Sci 1988;43:2201.

[37] Sadana A, Katzer J. Involvement of free radicals in the aqueous-phase catalytic oxidation of phenol over copper oxide. J Catal 1975;35:140. [38] Santos A, et al. Catalytic wet oxidation of phenol: kinetics of the mineralization rate. Ind Eng Chem Res 2001;40: 2773. [39] Tukac V, Hanika J. Catalytic wet oxidation of substituted phenols in the trickle bed reactor. 1998; 71: 262. [40] Finlayson B. Nonlinear analysis in chemical engineering. New York: McGraw-Hill, 1980. [41] Siddiqi MA, Lucas K. Correlations for prediction of diffusion in liquid. Canad J Chem Eng 1986;64:839. [42] Meng SF, Chung T. Liquid holdup and axial dispersion in trickle-bed reactors. Chem Eng Sci 1996;51:5357.