Modeling of a pilot-scale trickle bed reactor for the catalytic oxidation of phenol
Introduction
To detoxify highly polluted industrial wastewater of low biodegradability, over the past decades, a number of studies have been given to the application of wet air oxidation (WAO) technique for wastewater treatment, as reviewed by Mishra et al. [1]. WAO refers to the process in which liquid pollutants are oxidized by oxygen at elevated temperature (125–320 °C) and pressure (0.5–20 MPa). It is advantageous over conventional biological methods in the respect that high concentration non-biodegradable toxic substances can be degraded with high efficiency. Nevertheless, the operational cost of a typical WAO is tremendously high, owing to the need to escalate system pressure and temperature. In view of this, more recently, researchers have been focusing on the development of a certain catalyst by which the same goal of oxidizing organic pollutants can be achieved at mild pressure and temperature. These attempts give rise to the catalytic wet air oxidation (CWAO) process [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18].
CWAO process can be conducted either in a batch or in a continuous reactor, the latter of which attracts more attention due to its more operational convenience and treatment flexibility. Trickle bed reactor (TBR) is one of the ideal continuous reactor options and the mathematical simulation of a CWAO process taking place in TBR has been widely conducted. Some of the major findings in this regard have been summarized in the following.
On the basis of their studies on the mass transfer coefficients of gas-to-liquid and liquid-to-solid processes in TBR, Goto and Smith [19] established both ‘axial dispersion’ and ‘plug flow’ models to predict the conversion rate of oxidizing formic acid by oxygen at temperatures of 212–240 °C and pressure of 40 atm. The discrepancies between the two models were found negligible and the modeling results were in accordance with experimental data. This work concluded that, in term of their effects on the conversion rate, four mass transfer resistances are listed from the most to the least significance: gas-to-liquid mass transfer, intra-particle diffusion, liquid-to-solid (particle) mass transfer and axial dispersion. Bergault et al. [20] also observed that the hydrogenation of acetophenone in a TBR is much more sensitive to the mass transfer of gas-to-liquid than that of liquid-to-solid. The reviews conducted by Al-Dahhan et al. [21] and Wu et al. [22] further suggested that the incomplete wetting should also be taken into account as one of the factors affecting the TBR performance. Another successful example of the TBR modeling work completed by Avraam and Vasalos [23] also took energy balance into account when establishing the governing equations for the prediction of hydro-processing of oil in a pilot-scale TBR at steady-state. Singh et al. [24] attained promising predictions of phenol conversion on alumina supported CuO.
However, as the CWAO in TBR is a fairly complex process that comprises various steps of mass transfer and chemical reaction, so far more literates have reported on the unsatisfactory prediction of the developed models. In their attempts to predict the hydro-processing of oil, Korsten and Hoffmann [25] had to adjust the wetting efficiency before obtaining good agreement between experimental and simulation results. Valerius et al. [26] encountered the similar problem and later manually modified the catalyst activity of the model so as to fit the experimental data. Pintar et al. [10] acknowledged that the intrinsic reaction rate constant in their simulation model had to be increased by three times to make the model outcome agree well with experimental observations. They attributed such adjustment in model parameters to the possible deactivation in catalytic activity. The adjusted parameter in Bergault et al.’s model [20] was the gas-to-liquid mass transfer coefficient, which was intentionally increased by two to three times to give better simulation results.
A non-steady-state TBR model was first developed by Iliuta and Larachi [27] to describe CWAO of phenol. In spite of this contribution, this model was merely theoretically established without any supporting experimental evidence.
Guo and Al-Dahhan [28] carefully reviewed all the models developed in the past and attributed the discouraging modeling results to the fact that liquid vaporization was predominately ignored by many studies. This finding was agreed by Suwanprasop et al. [29] who assessed the effect of liquid vaporization on the phenol conversion by adjusting gas-to-liquid mass transfer coefficient.
It also has to be stressed that so far no TBR model consisting of dimensionless numbers only has been proposed, leading to the difficulties in applying a very specific model to a more general CWAO process with a wider operational range.
The objective of this study is to develop a TBR model on the basis of dimensionless numbers and parameters that helps to facilitate the scale-up of CWAO process. In doing so, a TBR model was first developed for the oxidation of aqueous phenol by using all the parameters determined from lab-scale studies. The specific TBR model was then normalized and the steady-state dimensionless concentrations of phenol and oxygen were simulated along bed length. Subsequently, a series of continuous CWAO experiments were conducted in a pilot-scale TBR to verify the modeling outcome.
Section snippets
Model assumptions
In developing the TBR model, some general assumptions were made as following:
- (a)
The entire TBR is under isothermal and isobaric conditions;
- (b)
Both the concentrations of phenol (A) and oxygen (B) are time-independent or have reached ‘steady-state’;
- (c)
A is non-volatile compound and there is no existence of A in the gas phase, meaning the possible effect of water vaporization on phenol oxidation is not taken into account in this study;
- (d)
The oxidation takes place between the adsorbed A and B on the catalyst
Materials
Copper supported by activated carbon was chosen as the heterogeneous catalyst for the CWAO of phenol in this work. Cylindrical activated carbon was provided by Norit Co., U.S. and has a mean diameter of 0.8 mm. The copper in elementary form was successfully incorporated into the activated carbon by hydrogen reduction following wet impregnation, resulting in a copper loading of 82.0 mg/g and BET surface area of 919.55 m2/g.
Aqueous phenol (90%) was purchased from Riedel-de Haën AG, Germany. The
Model validation
To verify the precision and accuracy of the TBR model, the effects of the following factors on phenol degradation rate were investigated: gas velocity, liquid velocity, bed temperature and pressure. The simulated phenol outlet concentrations were compared with those in the final samples of each run and depicted in Fig. 2.
Fig. 2 indicates that all the simulation results yield higher outlet phenol concentration or lower conversion rate when compared with the actual situation. Furthermore, out of
Conclusions
In this study a dimensionless TBR model was established that integrates all the mass transfer and the chemical reactions involved in the CWAO of phenol by oxygen when the process has arrived at ‘steady-state’. The major findings of this study are:
- (a)
A dimensionless TBR model is able to scale-up lab observations to pilot-scale applications. Nevertheless, as result of catalyst deactivation and insufficient retention time, the developed dimensionless model achieved satisfactory prediction results for
Acknowledgments
This work was supported by the Research Grants Council (RGC) of Hong Kong Government under the grant No. 613705.
References (32)
- et al.
Catalytic wet air oxidation of carboxylic acids on TiO2-supported ruthenium catalysts
J. Catal.
(1999) - et al.
Glucose hydrogenation on ruthenium catalysts in a trickle bed reactor
J. Catal.
(1998) - et al.
Catalytic wet air oxidation of acetic acid on carbon-supported ruthenium catalyst
J. Catal.
(1997) - et al.
Catalytic oxidation of volatile organic compounds. 1. Oxidation of xylene over a 0.2 wt% Pd/HFAU(17) catalyst
Appl. Catal. B: Environ.
(1999) - et al.
Copper/activated carbon as catalyst for organic wastewater treatment
Carbon
(1999) - et al.
Water pollution abatement by catalytic wet air oxidation in a trickle bed reactor
Catal. Today
(1999) - et al.
Catalytic liquid-phase oxidation of aqueous phenol solutions in a trickle bed reactor
Chem. Eng. Sci.
(1997) - et al.
Catalytic oxidation of organics in aqueous solutions. 1. Kinetics of phenol oxidation
J. Catal.
(1992) - et al.
Cu/Ni/Al layered double hydroxides as precursors of catalysts for the wet air oxidation of phenol aqueous solutions
Appl. Catal. B: Environ.
(2001) - et al.
Characterization of copper catalysts and activity for the oxidation of phenol aqueous solutions
Appl. Catal. B: Environ.
(1998)
Wet air oxidation of phenol using active carbon as catalyst
Appl. Catal. B: Environ.
Modeling and comparison of acetophenone hydrogenation in trickle-bed and slurry airlift reactors
Chem. Eng. Sci.
Evaluation of trickle bed reactor models for a liquid limited reactor
Chem. Eng. Sci.
Catalytic wet oxidation of phenol in a trickle bed reactor2004
Chem. Eng. J.
Modeling of a trickle-bed reactor
Chem. Eng. Process.
Wet air oxidation solid catalysis analysis of fixed and sparged three-phase reactors
Chem. Eng. Process.
Cited by (15)
A comprehensive model of adsorption of resorcinol in rotating packed bed: Theoretical, isotherm, kinetics and thermodynamics
2023, Journal of Environmental Chemical EngineeringEnergy optimization of wet air oxidation reactors with sub-millimeter bubble intensification
2022, Chemical Engineering ScienceCitation Excerpt :Although the WAO reactor dominated by sub-millimeter bubble has been experimentally studied and applied in industry, the relationship between energy consumption and sub-millimeter bubble characteristics for the optimal design and operation of WAO is still unclear. In recent years, several macro-reaction models for the WAO process have been developed, including the nonsteady-state model to describe phenol WAO (Iliuta and Larachi, 2001), the “axial dispersion” and “plug flow” models (Wu et al., 2009), and the scaled-up model of a trickle bed reactor to predict the behavior of phenol WAO (Mohammed et al., 2016). The reaction mechanism, kinetics and CFD modeling of phenol WAO is studied by Makatsa (Makatsa et al., 2021), and the phenol WAO reaction is slightly exothermic.
Performance analysis and modeling of catalytic trickle-bed reactors: a comprehensive review
2021, Journal of Industrial and Engineering ChemistryHierarchical nickel cobalt oxide spinel microspheres catalyze mineralization of humic substances during wet air oxidation at atmospheric pressure
2019, Applied Catalysis B: EnvironmentalOptimal design and operation of an industrial three phase reactor for the oxidation of phenol
2016, Computers and Chemical EngineeringCitation Excerpt :Also the values of the activation energy (EA) and pre-exponential factor (A0) has been estimated by the second approach to be 16,315.735 J/mol and (668,879.2 s−1 (cm3/mol)−1.11), respectively. A wide range of values of activation energy and pre-exponential factor have been reported in the public domain as 53,000 J/mol (Qinglin and Karl, 1998), 55,880 J/mol (Christoskova and Stoyanova, 2000), 61,000 J/mol (Albin et al., 1997) and 75,000 (±3) J/mol (Fortuny et al., 1999), 78,500 J/mol (Wadood and Sama, 2008), while the values of pre-exponential factor have been reported in literatures to be 1.3 × 109(±1 × 108) h−1 bar−1/2 (Fortuny et al., 1999), 3.2 × 1011 L/kgCat h (Wadood and Sama, 2008), which gives a clear indication that the values obtained in this study were within the range reported in the public domain. In order to be sure about precision of the evaluated kinetic parameters, sensitivity analysis for n, m, EA and A0 values is performed.