Multi-scale modeling of fixed-bed Fischer Tropsch reactor

Minhaj Ghouri, Shaik Afzal, Rehan Hussain, Jan Blank, Dragomir B. Bukur, Nimir O. Elbashir

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

A multi-scale pseudo-homogeneous one-dimensional model for the Fischer Tropsch fixed-bed reactor has been developed in this study using a detailed mechanistic kinetic scheme proposed in an earlier study (Todic et al., 2013). The developed model is capable of predicting the concentration and temperature profiles at the micro-(catalyst pores) level as well as the macro-(reactor bed) level for a cobalt-based catalyst (15 wt% Co/Al2O3). The uniqueness of this model is that it tries to combine different levels of complexity (product distribution modeling, particle diffusion and reactor bed modeling) into one single model. The predictability of this model has been validated experimentally using an advanced high-pressure FTS reactor unit over a wide range of testing conditions. It quite accurately predicts experimentally measured CH4 selectivity at different gas hourly space velocities but less accurately predicted CO conversion. On the other hand, a hydrocarbon product distribution has been predicted using a MATLAB® code that was written to estimate the FTS kinetic model's parameters (Todic et al., 2013) based on the experimental data collected using bench scale FTS fixed-bed reactor. The optimization of this model was done using a Genetic Algorithm (GA). The findings showed excellent predictability of the experimentally measured hydrocarbon product distribution profile of the catalyst, specifically paraffin formation rates which are the main products of the cobalt-based catalyst. This comprehensive model also involved modified thermodynamic equation of state and currently is upgraded to enable a direct comparison of the gas-phase and supercritical solvent-assisted FTS reactions under a variety of conditions using the experimental data reported in a recent study by our team (Kasht et al., 2015).

Original languageEnglish
Pages (from-to)38-48
Number of pages11
JournalComputers and Chemical Engineering
Volume91
DOIs
Publication statusPublished - 12 Sep 2016

Fingerprint

Catalysts
Hydrocarbons
Cobalt
Gases
Kinetics
Carbon Monoxide
Equations of state
Paraffin
Paraffins
MATLAB
Macros
Genetic algorithms
Thermodynamics
Testing
Temperature

ASJC Scopus subject areas

  • Chemical Engineering(all)
  • Computer Science Applications

Cite this

Multi-scale modeling of fixed-bed Fischer Tropsch reactor. / Ghouri, Minhaj; Afzal, Shaik; Hussain, Rehan; Blank, Jan; Bukur, Dragomir B.; Elbashir, Nimir O.

In: Computers and Chemical Engineering, Vol. 91, 12.09.2016, p. 38-48.

Research output: Contribution to journalArticle

Ghouri, Minhaj ; Afzal, Shaik ; Hussain, Rehan ; Blank, Jan ; Bukur, Dragomir B. ; Elbashir, Nimir O. / Multi-scale modeling of fixed-bed Fischer Tropsch reactor. In: Computers and Chemical Engineering. 2016 ; Vol. 91. pp. 38-48.
@article{a6cca052dfc242aca6b7442acd27dc0f,
title = "Multi-scale modeling of fixed-bed Fischer Tropsch reactor",
abstract = "A multi-scale pseudo-homogeneous one-dimensional model for the Fischer Tropsch fixed-bed reactor has been developed in this study using a detailed mechanistic kinetic scheme proposed in an earlier study (Todic et al., 2013). The developed model is capable of predicting the concentration and temperature profiles at the micro-(catalyst pores) level as well as the macro-(reactor bed) level for a cobalt-based catalyst (15 wt{\%} Co/Al2O3). The uniqueness of this model is that it tries to combine different levels of complexity (product distribution modeling, particle diffusion and reactor bed modeling) into one single model. The predictability of this model has been validated experimentally using an advanced high-pressure FTS reactor unit over a wide range of testing conditions. It quite accurately predicts experimentally measured CH4 selectivity at different gas hourly space velocities but less accurately predicted CO conversion. On the other hand, a hydrocarbon product distribution has been predicted using a MATLAB{\circledR} code that was written to estimate the FTS kinetic model's parameters (Todic et al., 2013) based on the experimental data collected using bench scale FTS fixed-bed reactor. The optimization of this model was done using a Genetic Algorithm (GA). The findings showed excellent predictability of the experimentally measured hydrocarbon product distribution profile of the catalyst, specifically paraffin formation rates which are the main products of the cobalt-based catalyst. This comprehensive model also involved modified thermodynamic equation of state and currently is upgraded to enable a direct comparison of the gas-phase and supercritical solvent-assisted FTS reactions under a variety of conditions using the experimental data reported in a recent study by our team (Kasht et al., 2015).",
author = "Minhaj Ghouri and Shaik Afzal and Rehan Hussain and Jan Blank and Bukur, {Dragomir B.} and Elbashir, {Nimir O.}",
year = "2016",
month = "9",
day = "12",
doi = "10.1016/j.compchemeng.2016.03.035",
language = "English",
volume = "91",
pages = "38--48",
journal = "Computers and Chemical Engineering",
issn = "0098-1354",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Multi-scale modeling of fixed-bed Fischer Tropsch reactor

AU - Ghouri, Minhaj

AU - Afzal, Shaik

AU - Hussain, Rehan

AU - Blank, Jan

AU - Bukur, Dragomir B.

AU - Elbashir, Nimir O.

PY - 2016/9/12

Y1 - 2016/9/12

N2 - A multi-scale pseudo-homogeneous one-dimensional model for the Fischer Tropsch fixed-bed reactor has been developed in this study using a detailed mechanistic kinetic scheme proposed in an earlier study (Todic et al., 2013). The developed model is capable of predicting the concentration and temperature profiles at the micro-(catalyst pores) level as well as the macro-(reactor bed) level for a cobalt-based catalyst (15 wt% Co/Al2O3). The uniqueness of this model is that it tries to combine different levels of complexity (product distribution modeling, particle diffusion and reactor bed modeling) into one single model. The predictability of this model has been validated experimentally using an advanced high-pressure FTS reactor unit over a wide range of testing conditions. It quite accurately predicts experimentally measured CH4 selectivity at different gas hourly space velocities but less accurately predicted CO conversion. On the other hand, a hydrocarbon product distribution has been predicted using a MATLAB® code that was written to estimate the FTS kinetic model's parameters (Todic et al., 2013) based on the experimental data collected using bench scale FTS fixed-bed reactor. The optimization of this model was done using a Genetic Algorithm (GA). The findings showed excellent predictability of the experimentally measured hydrocarbon product distribution profile of the catalyst, specifically paraffin formation rates which are the main products of the cobalt-based catalyst. This comprehensive model also involved modified thermodynamic equation of state and currently is upgraded to enable a direct comparison of the gas-phase and supercritical solvent-assisted FTS reactions under a variety of conditions using the experimental data reported in a recent study by our team (Kasht et al., 2015).

AB - A multi-scale pseudo-homogeneous one-dimensional model for the Fischer Tropsch fixed-bed reactor has been developed in this study using a detailed mechanistic kinetic scheme proposed in an earlier study (Todic et al., 2013). The developed model is capable of predicting the concentration and temperature profiles at the micro-(catalyst pores) level as well as the macro-(reactor bed) level for a cobalt-based catalyst (15 wt% Co/Al2O3). The uniqueness of this model is that it tries to combine different levels of complexity (product distribution modeling, particle diffusion and reactor bed modeling) into one single model. The predictability of this model has been validated experimentally using an advanced high-pressure FTS reactor unit over a wide range of testing conditions. It quite accurately predicts experimentally measured CH4 selectivity at different gas hourly space velocities but less accurately predicted CO conversion. On the other hand, a hydrocarbon product distribution has been predicted using a MATLAB® code that was written to estimate the FTS kinetic model's parameters (Todic et al., 2013) based on the experimental data collected using bench scale FTS fixed-bed reactor. The optimization of this model was done using a Genetic Algorithm (GA). The findings showed excellent predictability of the experimentally measured hydrocarbon product distribution profile of the catalyst, specifically paraffin formation rates which are the main products of the cobalt-based catalyst. This comprehensive model also involved modified thermodynamic equation of state and currently is upgraded to enable a direct comparison of the gas-phase and supercritical solvent-assisted FTS reactions under a variety of conditions using the experimental data reported in a recent study by our team (Kasht et al., 2015).

UR - http://www.scopus.com/inward/record.url?scp=84992303768&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84992303768&partnerID=8YFLogxK

U2 - 10.1016/j.compchemeng.2016.03.035

DO - 10.1016/j.compchemeng.2016.03.035

M3 - Article

VL - 91

SP - 38

EP - 48

JO - Computers and Chemical Engineering

JF - Computers and Chemical Engineering

SN - 0098-1354

ER -