No abstract

No abstract

No abstract

This paper presents a mathematical model developed for the operating simulation of a trickle-bed reactor. Furthermore, the model has been fixed to emulate the operation of the three reactors used at the gas oil hydrotreatment reactors available at Amuay refinery (PDVSA). The differential mass balances of the chemical species present in the process system are defined based on the mass conservation equation at the gas and the liquid phase. Moreover, mass transfer resistance on the inner catalyst particle is evaluated in terms of the catalyst effectiveness factor. Also, mass transfer coefficients, gas solubilities and transport properties of the chemical species available in the gas and liquid phase are evaluated with the help of mathematical correlations obtained from gas oil desulfurization experiments adjusted at the operating conditions available in the Amuay refinery gas oil hydrotreatment units. In addition, the model includes mathematical expressions for the rate of reaction (HDS and HDN) described with the help of kinetic equations of the Langmuir-Hinshelwood type. Furthermore, the kinetic parameters involved in the chemical reactions considered in the reactor model were estimated using a multidimensional simplex algorithm under the commercial reactor process conditions. Introduction The study of trickle-bed reactors has been one of the most important and fascinated topics faced in the Chemical Engineering field due to the fact that these reactors have proved their versatility in several applications involved in the oil refining industry, petrochemical complexes and biochemical processes1. Basically, a trickle-bed reactor is provided with a column that in some applications could be considerably high (over 20-meter height) loaded with a fixed packed bed of catalyst particles, throughout this bed gas and liquid flow on a downward direction. However, several design schemes show the possibility of upward co-current flow. Moreover, there are processes where the gas stream is fed following an upflow path in counter-current flow. In terms of co-current flow design, the liquid represents the disperse phase and it flows through the catalyst particle bed mainly as thin films that cover the outer catalyst surface. On the other hand, the gas is considered the continuous phase. Studies reported from the Faculty of Chemical Engineering at the University of Liege2–4 have given special attention to the modeling of liquid flow patterns in trickle-bed reactors. There are several advantages in the use of trickle-bed reactors such as:The liquid flow approaches the theoretical plug flow, which represents a warranty for higher catalytic conversion inside the reactor.Less catalyst lost per operation run.Lack of movable parts inside the reactor.Safe operation at high pressures and temperatures.The liquid flow can be fixed in terms of catalyst wetting and mass and heat transfer resistance. The hydrotreatment of vacuum gas oils requires the use of trickle-bed reactors suited to the mode of co-current descendent flow of gas and liquid in order to achieve the removal of heteroatoms such as sulfur (HDS) and nitrogen compounds (HDN) present in the liquid feedstock. The vacuum gas oils hydrotreatment facilities at the refinery consist of two units provided with three reactors. These vessels are placed in a parallel flow distribution that allows the processing of 85 KBPD of cracked and virgin gas oils achieving a 90 wt% level of desulfurization on a 2 years operation time base. In order to fulfill quality restrictions concerning the production of desulfurized vacuum gas oils is necessary the accurate knowledge of the entire chemical factors related with the hydrodesulfurization phenomena. This tutoring process could be accomplish by following two different approaches:Pilot plant test runs.

A model for a non-isothermal, semi-batch (stagnant slurry and flowing gas), laboratory scale Fischer-Tropsch (FT) slurry reactor is developed. The model assumes the existence of FT and water-gas shift (WGS) reactions. The reactor configuration is assumed to be the same as one used by Bhattacharjee et 0/. (1986). Gas-slurry mass transfer coefficients, solubility parameters and other physical transport and kinetic parameters used in the model are obtained from the reported studies of Lieb and Kuo (1984), Bhattacharjee et 0/. (1986), Deckwer et 0/. (1982, 1986) and Karandikar eI al. (1987) for the IT slurry system. The model is used to evaluate the relevant kinetic constants and the heat generation parameters forthe IT reaction from the experimental data of Bhattacharjee er 0/. (1986). The nature of the heat generation curves indicates that multiple steady states are likely to occur in a non-isothermal IT slurry reactor. The ignition temperatures are calculated as functions of gas hourly space velocity, activation energy for the Ff reaction, reactor pressure, and coolant temperature and flow rate. In general, these temperatures are in good agreement with those reported by Bhattacharjee er at. (1986). The exact values of the ignition temperature are strongly affected by the magnitudes of the activation energy and the heat of IT reaction. Once the reactor is ignited, the catalyst changes its character leading to the multiple branches of heat generation and product distribution curves. The extinction temperature was, therefore, not observed in Bhattacharjee's experiments.