Subscriber access provided by University of South Dakota
Article
Simulation of a non-isothermal industrial hydrotreating reactor using Simulink Ali Fooladi Toosi, Mohammad Sadegh Samie, Ali Dashti, and Mahmud Atarian Shandiz Energy Fuels, Just Accepted Manuscript • Publication Date (Web): 13 Jun 2014 Downloaded from http://pubs.acs.org on June 17, 2014
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
Simulation of a non-isothermal industrial hydrotreating reactor using Simulink Ali Fooladi Toosi, Mohammad Sadegh Samie, Ali Dashti, * Mahmud Atarian shandiz
Chemical Engineering Department, Faculty of Engineering, Ferdowsi University of Mashhad, Mashhad, Iran
Keywords: Modeling, Simulation, Hydrotreating, Trickle – bed reactor, Simulink.
Abstract
A steady-state three-phase heterogeneous model was applied to simulation of a VGO hydrotreating reactor in both pilot and industrial-plant under non-isothermal conditions. Three main reactions including HDS, HDA and HDN were considered and various kinetic models evaluated to justify the commercial reactor predictions. Influence of important operating variables such as feed API gravity, temperature, pressure and LHSV were studied. Evaluation of simulated results showed that components concentration profiles are very close in liquid and solid phase and the model can be renewed to pseudo two-phase model. Thus, the differentialalgebraic equations (DAEs) system were converted to particular stiff ordinary differential equations (ODEs) and solved simultaneously with Simulink toolbox in MATLAB. The
ACS Paragon Plus Environment
1
Energy & Fuels
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 25
simulation was validated by pilot and industrial-plant hydrotreater data with good agreement. By implementing the hydrotreater simulation in the Simulink well, the dynamic behavior studies are more simply and easily for any commercial hydrotreating reactors.
1- Introduction Nowadays, crude oil is still the most important world energy source for clean fuels supplement. Available crude oils becoming more and more heavier makes refining more difficult than ever. Furthermore, the increasing demand for high valuable products such as gasoline and middle distillates and more strict environmental rules dictates refiners to maximize the product quality and reduce more impurity contents. Catalytic hydrotreating (HDT) process is widely applied in the petroleum refinery industry to upgrade heavy oils and remove impurities such as sulfur, nitrogen, oxygen, metal-containing compounds as well as polynuclear aromatics. To maximize the product quality yield, study the effect of process conditions on hydrotreating is necessary. 1-5 Depending on the feed type and the desired product quality, the name of process will change. In the case of naphtha where sulfur is the main undesirable heteroatom, the process called hydrodesulfurization
(HDS).
Hydrotreating
is
used
for
straight–run
gas
oil
and
hydrodemetallization (HDM) process for heavy oils. A hydrocracking process is used when a change in the molecular weight of the feed is purposed. Selection of process type depends directly on the amount of impurity content and on the levels of conversion required.6 In ages, various residue hydroconversion processes developed and commercially employed using fixed bed, moving bed, ebullated bed, slurry bed or a combination. In literature, the typical operating conditions of these reactors presented. Deactivation rate of catalyst is the main aspect
ACS Paragon Plus Environment
2
Page 3 of 25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
for selecting the required process type. Now most of the hydrotreating reactors are multiphase catalytic fixed-bed classified to co-current and counter current gas-liquid flow trickle-bed.7-10 Korsten and Hoffmann developed a well-known plug flow reactor model for the HDS which their model has been base for the next studies.11 Matos and Guirardello have presented another model to describing hydrocracking process.12 Possibly, Chowdhury et al.13 accounted HDA reaction in hydrotreating modeling and simulation of diesel for the first time. Reaction kinetics dependence of hydrocracking process with catalyst type was investigated by Marafi et al.14 Rodriguez and Ancheyta developed the model for hydrodenitrogenation (HDN), hydrogenation of aromatics (HDA) and HDM reactions.6 Mederos et al. described a dynamic model for hydrotreating and studied time effect on operating conditions.15 Murali et al.9 carried out experimental HDT process in micro reactor using CoMo type catalyst and developed kinetic model with HDS and HDA reactions to evaluate commercial reactor. They used different order of reaction in proposed kinetic rates of HDS and HDA. However, the type of catalyst used in our simulation is similar to Murali et al.9 ones, but the predicted results of industrial reactor not confirmed well by those kinetic models. Alvarez et al. developed a model for hydroprocessing of heavy oil and study the quench effect on exit product.16,17 Chen et al. 18 applied HDS and HDA reactions for modeling and simulation of a commercial HDT reactor based on kinetic models of Korsten and Hoffmann
11
and Chowdhury
et al.13 They considered vapor-liquid equilibrium (VLE) effects and indicated the significance of VLE in simulation results. However, the feed gravity is determinant in accounting VLE effect. More heavier feed, need higher severity as well as higher temperature and pressure, thus less vapor phase content can be present in the system to observe VLE effects. Jarullah et al. have developed a model for the HDS of heavy crude oil.19 Recently, Alvarez and Ancheyta have
ACS Paragon Plus Environment
3
Energy & Fuels
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 25
studied on the start of run problems and represented optimum conditions for dynamic state of hydroprocessing.20 To solve the governing differential equations of the modeling, the different softwares were used. Rodriguez et al21 and Jimenez et al22 used MATLAB software whereas Mederos et al23,24 used FORTRAN language and Jarullah et al25,26 used gPROMS for solving the model differential equations. In this study, we have used MATLAB Simulink toolbox for modeling and simulation of the nonisothermal trickle-bed HDT reactor in both pilot and industrial scale. The flexibility, ability for solving differential equations with different mathematics methods in the shortest possible time are some advantageous of this toolbox. Model validation has been carried out in comparison with pilot data taken from literature11 and industrial data27. As stated by Rodrıguez and Ancheyta6, “Because kinetic information was taken from different sources, in which the operating conditions, the type of catalyst and feed, and the experimental setup, among other factors, are also different, it is almost impossible to have an exact representation of experimental data generated in other reaction systems, and recalculation of kinetic parameter values is mandatory.” in presented model, kinetic parameters were reevaluated because of different catalyst type, feed, and operating conditions. In most HDT simulation process, only the HDS reaction is considered. There are a few researches
6,15
considering the three important reactions including HDS, HDN
and HDA together. In our study, these three reactions are considered. The predicted results showed good agreement with mentioned experimental data. In this simulation, the linear increasing temperature gradient obtained along the reactor-bed emphasizes the necessity of using quench streams.
2- Modeling and simulation of the HDT reactor
ACS Paragon Plus Environment
4
Page 5 of 25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
First, the well-known three-phase plug flow reactor model was applied for the HDT modeling.11 However, the obtained results of our simulation showed that the liquid and solid phase concentration profiles are very close on the catalyst surface. Thus, by neglecting mass transfer resistances between liquid and solid phases, the initial trickle-bed model converted to pseudo two-phase plug flow reactor model. The model considers mass-transfer at the gas-liquid and liquid-solid interfaces, and involves correlations to predict mass-transfer coefficients, gas solubilities and specification of hydrocarbon feedstock shown in Table 1. The model verification is well done by pilot-scale data reported by Korsten and Hoffmann.11 Then, the validated model is used to simulate an industrial-plant data as presented in Table 2.27 Simulation results showed very good agreement with industrial data. The industrial hydrotreater is dimensioned as: the internal diameter of reactor is 2.9 m and a capacity of catalyst for loading up is 72.63m3. Table 2 shows the physical and chemical properties of the feedstock. A commercial Co-Mo supported on alumina was used. The feed stream is treated under following operating conditions: pressure of 5.17 MPa, temperature of 345°C, liquid hourly space velocity (LHSV) of 2.3 h-1and H2/HC ratio of 163 NM3/M3.27 The main assumptions for the modeling simplification are: The steady-state operation with no pressure drop is considered. No catalyst deactivation is happened. Mass transfer resistance in the gas side of the gas-liquid interface is negligible. Mass transfer resistance in the liquid side of the liquid-solid interface is negligible.
ACS Paragon Plus Environment
5
Energy & Fuels
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 25
Pseudo two-phase instead of three-phase model was applied. So, the differential-algebraic equations (DAEs) were converted to particular stiff ordinary differential equations (ODEs). Vapor pressure for hydrocarbons is negligible. There are no concentration gradients in solid phase because of slowness of the reaction. Gas and liquid velocities are constant through the reactor. Hydrogen pressure and components concentration can only change in axial direction. The reactor model considers main reactions such as HDS, HDN and HDA. There is no reaction in gas phase and all the reactions take place only on the catalyst surface. Thus, the following equation expressed changing in the molar gas flow rate of gaseous compound in the gas phase:
= k . a .
.
∗ C − k . a .
.
.
∗ [P ]
(1)
Where i = H2, H2S. The mass balance equation of gaseous compounds in the liquid phase is:
=
.
∗ ! − C " +
$% .&'
∗ r)*
(2)
Where i = H2, H2S. The mass balance equation of organic compounds in the liquid phase can be written:
=
$% .&+
∗ r,
(3)
The energy balance equation results following equation to predict the temperature profile along the reactor bed:
ACS Paragon Plus Environment
6
Page 7 of 25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
-. -/
=
$0
123 4 ∗54 ∗$4 6237 ∗57 ∗$7 8
∑
