DESIGN OF AN AUTOMATED FIXED BED REACTOR USED

FOR A CATALYTIC WET OXIDATION PROCESS

A. El Khoury, B. Bejjany

Laboratoire de Chimie Industrielle – Génie des Procédés (EA21), Cnam, 2 rue Conté, Paris III, France

M. Debacq, A. Delacroix

Laboratoire de Chimie Industrielle – Génie des Procédés (EA21), Cnam, 2 rue Conté, Paris III, France

Keywords: Wet Air Oxidation, WAO, Data acquisition, Intelligent sensor module, Regulation, Supervision,

Monitoring.

Abstract: Treatment of polluted industrial wastes is one of the challenging research topics that occupy an important

position in various chemical processes. Wet Air Oxidation (WAO) is one of the emerging processes suited

for the treatment of special aqueous wastes. The system consists of an oxidation in the liquid phase of the

organic matter by molecular oxygen at high temperature (200-325°C) and high pressure (up to 175 bar). It is

an enclosed process with a limited interaction with the environment as opposed to incineration. In this

paper, we will discuss the setup and the design of an automated fixed bed reactor used for wet oxidation of

various types of wastes. The system is controlled by a set of intelligent sensor modules used for data

acquisition. Regulation loops integrated within the sensor modules had been developed in order to control

the gas flow, the reactor temperature and the liquid sampling part. The process supervision and monitoring

had been achieved through the deployment of a SCADA software application. The graphical interface

developed for this purpose monitors the major parts of the process.

1 INTRODUCTION

The identification of highly refractory and non-

biodegradable organic pollutants in wastewater,

especially coming from the chemical and

petrochemical industry, has challenged the

conventional wastewater treatment such as

incineration or biological abatement. There is a clear

need to test and set-up an emerging alternative

technology that can deal with highly concentrated

and/or toxic non-biodegradable organic water

pollutants. However, it seems impossible in the close

future to dispose of one universal method able to

destroy all of the detected pollutants at an acceptable

cost (Masende, 2003). Therefore, Wet Air Oxidation

(WAO) is an efficient process by which organic

pollutants can be transformed by oxidation under

high pressures (50-250 bar) and high temperatures

(200-325°C), into carbon dioxide and water (Mishra,

1995). The process can be performed under milder

conditions (temperatures and pressures) by using a

homogenous or heterogeneous catalyst. Catalytic

Wet Air Oxidation (CWAO) is thus an attractive

process for wastewater treatments of toxic pollutants

such as phenol, pesticides, methyl tert-butyl ether

(MTBE) and their intermediate oxidation

compounds (Pintar, 1992).

Several studies with noble metal catalysts,

mainly Ru and Pt supported on carbon, Al

, TiO

and CeO

have revealed their stability and capacity

to destroy organic pollutants (Imamura, 1988). In

contrast to platinum, ruthenium was found to be an

active metal during the oxidation of acetic acid,

which is very refractory. Comparison of Ru/CeO

and Ru/TiO

showed that titanium oxide was more

stable in acetic and oxidizing medium, but the

loading of Ce on the catalyst significantly changes

the surface properties resulting in a better dispersion

of the noble metals. Thus, Ruthenium and Cerium

metals supported on alumina are considered to be

stable, accurate and cost effective catalysts

(Oliviero, 2000). There are only some tens of

industrial plants in the world and very few

documents are available for the scientific design of

365

El Khoury A., Bejjany B., Debacq M. and Delacroix A. (2007).

DESIGN OF AN AUTOMATED FIXED BED REACTOR USED FOR A CATALYTIC WET OXIDATION PROCESS.

In Proceedings of the Fourth International Conference on Informatics in Control, Automation and Robotics, pages 365-368

DOI: 10.5220/0001622303650368

 SciTePress

such processes due to their complexity and the

delicacy needed for their proper operation

(Debellefontaine, 1999). Therefore, in this work we

will show the essential techniques and equipments

allowing us to control and monitor a pilot scale

reactor designed for the wet oxidation of organic

pollutants.

After presenting the system architecture and the

communication interface, we will attempt to

emphasize on the use of intelligent sensor modules

in order to adequately control the reactor

temperature and pressure as well as the gas flow and

the liquid outlet.

2 SYSTEM ARCHITECTURE

AND COMMUNICATION

The installation (Figure 1) consist of an L-316

tubular fixed-bed reactor (7.6 cm internal diameter

and 70 cm in length), which is placed in the center

of an oven implementing an electrical resistor

controlled by a PID controller. The solution is

introduced to the reactor by a high-pressure pump at

a flow rate ranging from 1 to 10 cm

.min

-1

. The

catalyst is placed between two layers of glass beds in

the reactor. The oxygen is directly fed from a high-

pressure bottle whereas a gas flow indicator and

controller (FIC – Brooks) controls its flow rate. The

effluent of the reactor passes through two

condensers and a gas-liquid separator. The gas phase

is released in the hood after passing through a gas

flow indicator (FI – Brooks) and the liquid phase is

stored in a tank whereas a level indicator (LI –

Bamo) controls a regulation valve (LV - Samson)

prior to liquid evacuation. A backup pressure

regulator (PIC) placed at the gas outlet maintains a

stable pressure inside the system.

Sensors and actuators are plugged into a set of

four intelligent sensor modules (ISM112 – Gantner)

interconnected through an RS485 field bus. An

RS232/RS485 converter enables the supervision

station to communicate with the sensor modules

using the Modbus RTU protocol. The intelligent

sensor module supports measuring methods with 2-,

3-, and 4-wire technique and measuring methods

with 4- and 6-wire bridge connection. Consequently,

the most varying measurement tasks can easily be

solved by means of the different analog inputs and in

combination with the force output, which provides

the local power supply for the transducers. The

module can simultaneously take up and process

sensor signals from several heterogeneous sensors.

As many sensors can be connected as there are

analog and digital signal inputs and outputs

available. With the ISM112 these are 6 sensors at

the most, 4 analog and 2 digital sensors. The RS485

interface permits the simultaneous connection and

operation of a maximum of 32 bus users per

segment. Among analog and digital signal

processing; the intelligent sensor module can handle

a controller variable by which a sensor variable can

be monitored for a definable set value. Deviations of

the sensor variable’s value will be corrected

depending on the set function of the controller (PID-

controller) and will then be assigned to the controller

variable. This corrected value can be assigned to an

analog output and then be used to influence the input

signal by a corresponding connection. Accordingly,

we were able to control and monitor most of the

system parameters in order to boost and optimize the

reaction conditions. A set of five thermocouples

indicates the temperatures at different levels of the

process, especially at the center of the reactor where

the temperature had been adequately controlled and

monitored. Pressure indicators monitor the system

global pressure required for the reactor proper

operation. Possible fluid leakage can be detected

through pressure drops inside the system. Gas flow

is controlled by an algorithm set by the manufacturer

whereas the intelligent sensor modules directly

control the reactor temperature and the regulation

valve through a set of regulation parameters defined

by the Ziegler-Nichols method.

3 TUNING A PID CONTROLLER

The first step in the design strategy is to install and

tune a PID controller (Tan, 2006). The ideal

continuous PID controller returns the controller

output u, as given by equation (1), where K

is the

proportional gain, T

is the integral time, T

the

derivative time, and e the error between the

reference (ref.) and the process output (y).

⎟

⎠

⎞

⎜

⎝

⎛

++=

∫

Tdte

eKu

(1)

We are concerned with small sampling periods

, the equation may be approximated by a discrete

approximation. Replacing the derivative term by a

backward difference and the integral by a sum using

rectangular integration, an approximation may be

given by the equation (2).

ICINCO 2007 - International Conference on Informatics in Control, Automation and Robotics

366

Figure 1: Wet Air Oxidation process diagram. Abbreviations: FI: Flow indicator; FIC: Flow indicator and controller; GC:

Gas Chromatography; LC: Level controller; LI: Level indicator; LV: Regulation valve; PI: Pressure indicator; PIC: Backup

pressure regulator; SCADA: System control and data acquisition; TI: Temperature indicator.

⎟

⎠

⎞

⎜

⎝

⎛

−

++=

−

∑

npn

TTe

eKu

(2)

Index n refers to time instant. By tuning we shall

mean the activity of adjusting the parameters K

, T

and T

. Several tuning aspects may be illustrated by

static considerations. For purely proportional control

= 0 and 1/T

= 0), the control law (2) reduces to

the following equation:

= K

(3)

Considering the feedback loop in Figure 2, where

the controller has the proportional gain K

and the

process has the gain K in steady state, the output x

can be related to the reference (ref.), the load l, and

the measurement noise n by the following equation:

()

nref

+−

(4)

If n and l are zero, then K

should be high in

order to insure that the process output x is close to

the ref. Furthermore, if l is nonzero, a high value

will make the system less sensitive to changes in the

load l. But if n is nonzero, K

should be moderate

otherwise the system will be too sensitive to noise.

Obviously, the setting of K

is a balance between:

stability, noise sensitivity, and load regulation.

Figure 2: Closed loop system identification.

A PID controller may be tuned using the Ziegler-

Nichols frequency response method, according to the

following procedure:

DESIGN OF AN AUTOMATED FIXED BED REACTOR USED FOR A CATALYTIC WET OXIDATION PROCESS

367

(a) Increase the proportional gain until the

system oscillates (Figure 3); that gain is the ultimate

gain K

(b) Read the time between peaks T

at this

setting.

parameters are given in a table.

Figure 3: Ziegler-Nichols frequency response method.

The sample period may be related to the

derivative gain T

. In connection with the Ziegler-

Nichols rules, this implies that T

should

approximately be equal to 1 – 5 percent of the

ultimate period T

. Taking full advantage of this

method; we were able to adequately control the

reactor temperature and the liquid evacuation unit.

When the system reaches the steady state, the

controller allows us to maintain a constant liquid

level in the tank. Consequently, liquid flow can be

continuously evacuated at the system’s outlet.

4 HUMAN MACHINE

INTERFACE (HMI)

An operator’s graphical interface was developed

using the FIX MMI Intellution SCADA software

which combines high performance monitoring and

control with wide range of data acquisition on the

Windows NT/2000/9x operating systems. The FIX

application contains three sets of multithreaded

processes: the user process (HMI), the FIX engine

and the industrial automation device servers. These

processes interact through a client-server

relationship. The user process displays the user

interface and executes blocks of code that are

defined for control algorithms, supervisory control,

analysis and visual presentation. The event-driven

engine maintains a real time database, communicates

with device servers and performs a multitude of

tasks including engineering unit scaling, alarm

processing and historical data collection and

trending. Device servers are the applications that

communicate with Input/Output devices. The FIX

application establishes a communication with the

ISM112 intelligent sensor modules through the

deployment of a Modbus RTU server fully

compliant with the latest Modbus RTU protocol

definitions. Therefore, the ISM112 data registers can

be accessed and modified to the desired values

allowing thus the operator to have full control of the

process variables.

5 CONCLUSION

The aim of this work is to setup an oxidation process

that meets the conditions needed for the aqueous

destruction by oxygen or air of organic pollutants.

The aforementioned techniques and equipments

which in priority are based on regulation and

automation procedures, allowed us to design an

automated fixed bed reactor that fulfills the required

temperatures and pressures conditions (up to 300°C

and 25 bar) usually used for CWAO processes. The

developed monitoring interface allows the operator

to easily manage and control the process parameters.

Chemical runs allowing us to validate the system

efficacy during the oxidation of various types of

aqueous wastes, are in process of completion.

REFERENCES

Debellefontaine, H., Foussard, J.N.,Wet air oxidation for

the treatment of industrial wastes, Waste Manage., 20

(2000) 15.

Imamura, S., Fukuda, I., Ishida, S., Wet oxidation

catalyzed by ruthenium supported on cerium(IV)

oxide, Ind. Eng. Chem. Res., 27 (1988) 718.

Masende, Z.P.G., Kuster, B.F.M., Ptasinski, K.J., Janssen,

F.J.J.G., Katima, J.H.Y., and Scouten, J.C., Platinium

catalysed wet oxidation of phenol in a stirred slurry

reactor: A practical operation window, Appl. Catal. B,

41 (2003) 247.

Mishra, V.S., Mahajani,V.V., Joshi, B., Wet Air

Oxidation, Ind. Eng. Chem. Res., 34 (1995) 2.

Oliviero, L., Barbier Jr, J., Duprez, D., Guerrero-Ruiz, A.,

Bachiller-Baeza, B., Rodriguez-Ramoz, I., Catalytic

wet air oxidation of phenol and acrylic acid over Ru/C

and Ru–CeO

/C catalysts, Appl. Catal. B, 25 (2000)

267.

Pintar, A., Levec, J., Catalytic oxidation of organics in

aqueous solutions : I. Kinetics of phenol oxidation, J.

Catal., 135 (1992) 345.

Tan, W., Liu, J., Chen, T., and Marquez, H.J.,

Comparaison of some well-known PID formulas,

Comput. Chem. Eng., 30 (2006) 1416.

ICINCO 2007 - International Conference on Informatics in Control, Automation and Robotics

368