CFD Simulation of Ethanol Steam Reforming System For Hydrogen Production

chemengineering
Article
CFD Simulation of Ethanol Steam Reforming System
for Hydrogen Production
Alon Davidy ID
Israel Military Industries Ltd., Ramat Hasharon 47100, Israel; alon.davidy@gmail.com; Tel.: +972-03-904-9118

Received: 26 May 2018; Accepted: 30 July 2018; Published: 2 August 2018
Abstract: Hydrogen could be a promising source fuel, and is often considered as a clean energy
carrier as it can be produced by ethanol. The use of ethanol presents several advantages, because it is
a renewable feedstock, easy to transport, biodegradable, has low toxicity, contains high hydrogen
content, and easy to store and handle. Reforming ethanol steam occurs at relatively lower
temperatures, compared with other hydrocarbon fuels, and has been widely studied due to the high
yield provided for the formation of hydrogen. A new computational fluid dynamics (CFD) simulation
model of the ethanol steam reforming (ESR) has been developed in this work. The reforming
system model is composed from an ethanol burner and a catalytic bed reactor. The liquid ethanol is
burned inside the firebox, then the radiative heat flux from burner is transferred to the catalytic bed
reactor for transforming the ethanol steam mixture to hydrogen and carbon dioxide. The proposed
computational model is composed of two phases—Simulation of ethanol burner by using Fire
Dynamics Simulator software (FDS) (version 5.0) and a multi-physics simulation of the steam
reforming process occurring inside the reformer. COMSOL multi-physics software (version 4.3b)
has been applied in this work. It solves simultaneously the fluid flow, heat transfer, diffusion with
chemical reaction kinetics equations, and structural analysis. It is shown that the heat release rate
produced by the ethanol burner, can provide the necessary heat flux required for maintaining the
reforming process. It has been found out that the mass fractions of the hydrogen and carbon dioxide
mass fraction are increased along the reformer axis. The hydrogen mass fraction increases with
enhancing the radiation heat flux. It was shown that Von Mises stresses increases with heat fluxes.
Safety issues concerning the structural integrity of the steel jacket are also addressed. This work
clearly shows that by using ethanol which has low temperature conversion, the decrease in structural
strength of the steel tube is low. The numerical results clearly indicate that under normal conditions
of the ethanol reforming (The temperature of the steel is about 600 ◦ C or 1112 ◦ F), the rupture time of
the HK-40 steel alloy increases considerably. For this case the rupture time is greater than 100,000 h
(more than 11.4 years).
Keywords: ethanol burner; ethanol steam reformer; catalytic bed reactor; hydrogen production;
fuel cell; CFD; Fire Dynamic Simulation software (FDS); Large Eddy Simulation (LES); radiative heat
flux; convective heat flux; Multi-physics Simulation; Darcy Flow Equation; chemical kinetics;
arrhenius coefficients; diffusion equation; heat transfer; thermal expansion; structural analysis;
steel tube; HK-40 alloy; ultimate tensile strength; rupture time
1. Introduction
In recent decades, there has been a continuous effort to reduce global environmental pollution
and fossil oil consumption [1]. According to Akande et al. [2] the demand for hydrogen has
increased recently due to progress in fuel cell technologies. Fuel cells are electrochemical devices
described as continuously operating batteries and are considered as a clean source of electric
ChemEngineering 2018, 2, 34; doi:10.3390/chemengineering2030034 www.mdpi.com/journal/chemengineering

ChemEngineering 2018, 2, 34 2 of 24
energy, containing high energy efficiency, and its resulting emission is just water [3]. It can be
produced from different kinds of renewable feedstocks, such as ethanol. The use of ethanol as a
raw material presents several advantages because it is easy to transport, biodegradable, has low
toxicity, contains high hydrogen content, and easy to store and handle [4]. Furthermore, ethanol is
economically, environmentally and strategically attractive as an energy source. Ethanol can be a
hydrogen source for countries that lack fossil fuel resources, but have significant agricultural economy.
This is feasible, because virtually any biomass can now be converted into ethanol as a result of recent
advances in biotechnology [5]. These attributes have made H2 obtained from ethanol reforming a
very good energy vector, especially in fuel cells applications. Hydrogen production from ethanol has
advantages compared to other H2 production techniques, including steam reforming of methanol and
hydrocarbons. Unlike hydrocarbons, ethanol is easier to reform and is also free of sulfur, which is a
catalyst poison in the reforming of hydrocarbons [6]. Also, unlike methanol, which is produced from
hydrocarbons and has a relatively high toxicity, ethanol is completely biomass-based and has low
toxicity and as such it provides less risk to the population [7]. The fact that methanol is derived from
fossil fuel resources also renders it an unreliable energy source in the long run. Conclusively, amongst
the various processes and primary fuels that have been proposed for hydrogen production in fuel cell
applications, steam reforming of ethanol is very attractive. Ethanol steam reforming (ESR) proceeds
at temperatures in the range of 300–600 ◦ C, which is significantly lower than those required for CH4
or gasoline reforming. This is an important consideration for the improved heat integration of fuel
cell vehicles.
The main objective of the computational fluid dynamics (CFD) modeling in this work is to analyze
the structural integrity and performance of an industrial ESR system with considerable modeling
accuracy that can help evaluate various modeling assumptions usually employed.
Extensive work has been performed on the mathematical modeling the development of
first-principles reformer models. With the dramatic increase of computing power, CFD modeling has
become an increasingly important platform for reformer modeling and design, combining physical
and chemical models with detailed representation of the reformer geometry. When compared
with first-principles modeling, CFD is a modeling technique with powerful visualization and
computational capabilities to deal with various geometry characteristics, transport equations and
boundary conditions.
A 3D CFD simulation pioneering study of the ESR in microreactors with square channels has
been carried out by Uriz et al. [8]. A phenomenological kinetic model based on simple power-law
rate equations and considering the following reactions—ethanol dehydrogenation to acetaldehyde,
acetaldehyde steam reforming to H2 and CO2 , ethanol decomposition to CO, CH4 and H2, and the
water-gas shift—describes satisfactorily the ESR over a Co3 O4 -ZnO catalyst. The CFD computational
study shows that high reforming temperatures (above 625 ◦ C) should be avoided. This is because the
decomposition of ethanol competes effectively with the dehydrogenation of the alcohol to acetaldehyde,
which is the key intermediate of the ESR process, and results in a reduced hydrogen yield and an
increased content of CO in the reformate stream. Their work shows that micro reactors can help to
overcome these difficulties by increasing the surface area-to-volume ratio and the catalyst loading.
By using these micro reactors, the operating temperature can be reduced while increasing the selectivity
and maintaining the level of ethanol conversion [8].
Uriz et al. [9] showed that CFD is a very useful tool for advancing in the development and
optimization of these technologies. The interest of CFD for assisting in the understanding of the
behavior of hydrogen technologies is also important. Nevertheless, there are also limitations regarding
the description of chemical transformations and the physics of some complex phenomena such as in
multi-physics systems.
Lao et al. [10] developed a CFD model of an industrial scale steam methane reforming reactor
(reforming tube) used to produce hydrogen. The CFD model of an industrial-scale reforming tube has
been developed in ANSYS Fluent (version 15.0) with realistic geometry characteristics to simulate the
transport and chemical reaction phenomena with approximate representation of the catalyst packing.
In this work a CFD simulation study of the ESR system has been performed. The proposed
computational work is composed of two phases—simulation of ethanol burner by using fire dynamics
simulator (FDS) software and a multiphysics simulation of steam reforming process occurring inside
the reformer. COMSOL multi-physics software has been applied in this work. It solved simultaneously
the fluid flow, heat transfer, diffusion with chemical reaction kinetics equations and structural analysis.
It is probably the first time that FDS software has been applied in order to simulate the combustion
processes of the burner inside the gas heated heat exchange reformer (GHHR). The FDS software is
described in detail in Sections 2.1 and 2.2. The multiphysics reformer model is described in Section 2.3.
As far as I know, this work is the first coupled CFD and structural analysis study on the ESR system.
The structural analysis of the reformer steel tube is essential in order to verify that the structural
integrity of the tube under the operating conditions of the reformer is maintained. Steel tube rupture
and cracks may cause release of the hydrogen gas to the reformer facility. Hydrogen has wide
flammability limits and very low ignition energy [11]. Therefore, hydrogen present safety concerns at
limited ventilation conditions because of the danger of explosive mixture formation that may cause
severe damage [12–14]. Diéguez et al. [12] showed examples of application of CFD to safety issues
such as hydrogen leakages, hydrogen flames, detonation and application of the simulation results to
evaluate possible physiological injuries. They performed a CFD for simulating the Hydrogen leakage.
1.1. Mechanism of Steam Ethanol Reforming Process

The main reactions that occur during the ESR process on Ni catalyst are [15]:
• Ethanol steam reforming:
C2 H5 OH + 3H2 O → 2CO2 + 6H2
• Ethanol decomposition to methane:
C2 H5 OH → CO + H2 + CH4
• Ethanol dehydration:
C2 H5 OH → C2 H4 + H2 O
• Ethanol dehydrogenation:
C2 H5 OH → CH3 CHO + H2
• Ethanol decomposition to acetone:
2C2 H5 OH → CH3 COCH3 + CO + 3H2
• Water-gas shift reaction:

CO + H2 O → CO2 + H2
1.2. Ethanol Burner Steam Reformer

The reforming system described in this work is composed from ethanol burner and a catalytic bed
reactor. The liquid ethanol is burned inside the firebox. The heat from the flue gases is transferred to the
catalytic bed reactor (the theoretical model of the reformer is described in Section 2.3) for transforming
the ethanol steam mixture to hydrogen and carbon dioxide. The catalyst is made from Ni/Al2 O3 .
The schematic of the system is described in [16].
Burner Heat Transfer
The heat produced from ethanol burning from the combustion products is transferred to catalytic
bed walls in by three modes (see Figure 1):
(1) Convection
ChemEngineering 2018, 2, x FOR PEER REVIEW 4 of 24
(2) Radiation
(3) (2) Radiation
Conduction
(3) Conduction
Figure
Figure 1. 1. Heattransfer
Heat transfermechanisms
mechanisms in
insteam
steamreformer [17].
reformer [17].
1.2.1.1.2.2 Convection
Convection
Heat is transferred through fluids in motion and between a fluid and solid surface in relative
Heat is transferred through fluids in motion and between a fluid and solid surface in relative
motion. When the motion is produced by forces other than gravity, the term forced convection is
motion. When the motion is produced by forces other than gravity, the term forced convection is
used. In engines the fluid motions are turbulent. Heat is transferred by forced convection between
used.the
In in-cylinder
engines the fluidand
gases motions are turbulent.
the cylinder Heatcylinder
head, valves, is transferred by piston
walls, and forcedduring
convection between
induction,
the in-cylinder gases and the cylinder head, valves,
compression, expansion and exhaust processes [18]. cylinder walls, and piston during induction,
compression, expansion and exhaust processes [18].
1.2.3 Radiative Heat Transfer
1.2.2. Radiative Heat
There are Transfer
two sources of radiative heat transfer within the burner: The high temperature
burned gases and the soot particles in the ethanol flames. In the burner, most if the fuel burns in
There are two sources of radiative heat transfer within the burner: The high temperature burned
turbulent diffusion flame as fuel and air mix together. The flame is highly luminous, and soot
gasesparticles
and theare
soot particles
formed at an in the ethanol
intermediate flames.
step In the burner,
in the combustion most
process if the fuel burns in turbulent
[18].
diffusion flame as fuel and air mix together. The flame is highly luminous, and soot particles are
1.2.4 Conduction Heat Transfer
formed at an intermediate step in the combustion process [18].
Heat is transferred by molecular motion, through solids and through fluids at rest, due to
1.2.3.temperature
Conductiongradient.
Heat Transfer
It is transferred by conduction through the reformer walls [18].
Heat is transferred
2. Materials by molecular motion, through solids and through fluids at rest, due to
and Methods
temperature gradient. It is transferred by conduction through the reformer walls [18].
2.1. Fire Dynamic Simulation Modeling of the Burner
2. Materials and Methods
The FDS has been developed at the Building and Fire Research Laboratory (BFRL) at the
National Institutes
2.1. Fire Dynamic of Standards
Simulation ModelingandofTechnology
the Burner (NIST), e.g., McGrattan et al. [19,20]. This software
calculates simultaneously the temperature, density, pressure, velocity, and chemical composition
The FDS
within hasnumerical
each been developed
grid cellatatthe Building
each discreteand
timeFire Research
step. Laboratory
It also calculates the(BFRL) at the heat
temperature, National
Institutes of Standards
flux, and mass loss and Technology
rate of the enclosed (NIST), e.g., McGrattan
solid surfaces. The latteretisal.used
[19,20].
in theThis
casesoftware
where thecalculates
fire
heat release rate is unknown. The major components of this software are:
simultaneously the temperature, density, pressure, velocity, and chemical composition within each
numericalHydrodynamic Model—FDS
grid cell at each code step.
discrete time is formulated based on the
It also calculates CFDtemperature,
of fire-drivenheat
fluidflux,
flow.and
Themass
FDS numerical solution can be carried out using either a Direct Numerical Simulation (DNS)
loss rate of the enclosed solid surfaces. The latter is used in the case where the fire heat release rate is method
or Large Eddy Simulation (LES). The latter is relatively low Reynolds numbers and is not severely
unknown. The major components of this software are:
Hydrodynamic Model—FDS code is formulated based on CFD of fire-driven fluid flow. The FDS
numerical solution can be carried out using either a Direct Numerical Simulation (DNS) method or
Large Eddy Simulation (LES). The latter is relatively low Reynolds numbers and is not severely limited
in grid size and time step as the DNS method. In addition to the classical conservation equations
considered in FDS, including mass species momentum and energy, thermodynamics-based state
equation of a perfect gas is adopted along with chemical combustion reaction for a library of different
fuel sources.
Combustion Model—For most applications, FDS uses a mixture fraction combustion model.
The mixture fraction is a conserved scalar quantity that is defined as the fraction of gas at a given point
in the flow field that originated as fuel. The model assumes that combustion is mixing controlled,
and that the reaction of fuel and oxygen is infinitely fast. The mass fractions of all of the major
reactants and products can be derived from the mixture fraction by means of “state relations”,
empirical expressions arrived at by a combination of simplified analysis and measurement [21].
Radiation Transport—Radiative heat transfer is included in the model via the solution of the
radiation transport equation for a non-scattering gray gas. In a limited number of cases, a wide
band model can be used in place of the gray gas model. The radiation equation is solved using a
technique similar to a finite volume method for convective transport, thus the name given to it is the
Finite Volume Method (FVM) [19]. FDS also has a visual post-processing image simulation program
named “smoke-view”.
2.2. Governing Equations of FDS Software

This section introduces the basic conservation equations for mass, momentum and energy for
a Newtonian fluid. These are the same equations that can be found in almost any textbook on fluid
dynamics or CFD.
2.2.1. Mass and Species Transport

Mass conservation can be expressed either in terms of the density, ρ [19]:
∂ρ .
+ ∇·ρu = m000 b (1)
∂t
where ρ is the density (kg/m3 ), and u is the velocity field (m/s). In terms of the individual gaseous
specie, Yα :
∂ . .
(ρYα ) + ∇·ρYα u = ∇·ρDα ∇Yα + m000 α + m000 b,α (2)
∂t
Dα is the diffusion coefficient of α component of the mixture (m2 /s).
2.2.2. Momentum Transport

The momentum equation in conservative form is written [19]:
∂
(ρu) + ∇·ρuu + ∇ p = ρg + fb + ∇·τij (3)
∂t
where fb is the force term (Pa/m). The stress tensor, τij (Pa), is defined [19]:
( !
2 1 i = j 1 ∂ui ∂u j
τij = µ 2Sij − δij (∇·u) ; δij = ; Sij = + i, j = 1, 2, 3 (4)
3 0 i 6= j 2 ∂x j ∂xi
The term Sij is the symmetric rate-of-strain tensor, written using conventional tensor notation.
The symbol µ is the dynamic viscosity of the fluid. The overall computation can either be treated as
a DNS, in which the dissipative terms are computed directly, or as a LES, in which the large-scale
eddies are computed directly and the subgrid-scale dissipative processes are modeled. The numerical
algorithm is designed so that LES becomes DNS as the grid is refined. Most applications of FDS
are LES. For example, in simulating the flow of smoke through a large, multi-room enclosure, it is
not possible to resolve the combustion and transport processes directly. However, for small-scale
combustion experiments, it is possible to compute the transport and combustion processes directly.
2.2.3. LES
The most distinguishing feature of any CFD model is its treatment of turbulence. Of the three main
techniques of simulating turbulence, FDS contains only LES and DNS. There is no Reynolds-averaged
Navier-Stokes (RANS) capability in FDS. LES is a technique used to model the dissipative processes
(viscosity, thermal conductivity, material diffusivity) that occur at length scales smaller than those
that are explicitly resolved on the numerical grid. This means that the parameters µ, k and D in the
equations above cannot be used directly in most practical simulations. They must be replaced by
surrogate expressions that “model” their impact on the approximate form of the governing equations.
This section contains a simple explanation of how these terms are modeled in FDS. There is a small
term in the energy equation known as the dissipation rate, ε (Pa/s) the rate at which kinetic energy is
converted to thermal energy by viscosity [19]:

2
2 2 2
2
ε = τij ·∇u = µ 2Sij Sij − 3 (∇· u ) = µ 2 ∂u
∂x + 2 ∂v
∂y + 2 ∂w
∂z +
2 2 2 2 (5)
2
∂v
∂x + ∂u
∂y + ∂w
∂y + ∂v
∂z + ∂u
∂z + ∂w
∂x − 3
∂u
∂x + ∂v
∂y + ∂w
∂z
Following the analysis of Smagorinsky, the viscosity µ is modeled:

1
2 2 2
µ LES = ρ(Cs ∆) 2Sij : Sij − (∇·u)2 (6)
3
where Cs is an empirical constant and ∆ is a length on the order of the size of a grid cell. The bar above
the various quantities denotes that these are the resolved values, meaning that they are computed from
the numerical solution sampled on a coarse grid (relative to DNS). The other diffusive parameters, the
thermal conductivity and material diffusivity, are related to the turbulent viscosity by [16]:
µ LES c p µ
k LES = ; (ρD )t,LES = LES (7)
Prt Sct
The turbulent Prandtl number Prt and the turbulent Schmidt number Sct are assumed to be
constant for a given scenario. The model for the viscosity, µ LES , serves two roles: Firstly, it provides a
stabilizing effect in the numerical algorithm, damping out numerical instabilities as they arise in the
flow field, especially where vorticity is generated. Second, it has the appropriate mathematical form to
describe the dissipation of kinetic energy from the flow.
2.2.4. Energy Transport

The energy conservation equation is written in terms of the sensible enthalpy, hs (J/kg) [19]:
∂ Dp . . .
(ρhs ) + ∇·(ρhs u) = + q000 − q000 b − ∇·q00 + ε (8)
∂t Dt
The sensible enthalpy (J/kg) is a function of the temperature [19]:
hs = ∑ Yα hs,α (9)
α

where sensible heat of each component in the mixture is calculated by:
where sensible heat of each component in the mixture is calculated by:
ZT
T
T 0  dT 0

T(T) =c p, cTp,α  dT
hs,h s,α (10)
(10)
T0 T0
.
wherewhere q  HRR
q000 is the is theper
HRRunitper unit volume
volume from a chemical
from a chemical reactionreaction
(W/m3 ),
.
(W/m qb energy
q000 b ),is the
3 is the energy
transferred
.
to thetransferred
evaporating to the evaporating
liquid 3
(W/m ) liquid 00
and q (W/m 3 ) and qthe
represents  represents
conductivetheand
conductive
radiative and radiative
heat (W/m2 ):
fluxes heat
fluxes (W/m2):
.
q  = kT -  h∑
q00 = k∇ T − s,a a h ρD ∇Y + q00 r
a (11)
a  Da Ya + qr
s,a (11)
a
2.2.5. Equation of State

2.2.5. Equation of State
The pressure is calculated by using the ideal gas equation of state (the burning occurs at
Thepressure
atmospheric pressure which
is calculated
is muchbyless
using thethe
than ideal gas pressure
critical equation of stateair):
of the (the burning occurs at
atmospheric pressure which is much less than the critical pressure of the air):
ρRT
p =RT (12)
p= W (12)
W
where p is the pressure in (Pa), T is the temperature in (K), R is the universal gas constant and W is the
where p is the pressure in (Pa), T is the temperature in (K), R is the universal gas constant and W
molecular mass of the gaseous mixture in (J/mol).
is the molecular mass of the gaseous mixture in (J/mol).
2.2.6. FDS Modelling of the Ethanol Burner
2.2.6. FDS Modelling of the Ethanol Burner
The geometry of the
The geometry ethanol
of the burner
ethanol burnermodel
modelis
is shown inFigure
shown in Figure2.2.
Figure
Figure 2.2.The
Thegeometry
geometry of
of the
theethanol
ethanolburner.
burner.
At bottom
At the the bottom of the
of the burner,
burner, ethanolisisinjected
ethanol injected and
and ignited.
ignited.The
Theethanol burning
ethanol in still
burning air isair is
in still
controlled by buoyancy (see Figure 3) [22].
controlled by buoyancy (see Figure 3) [22].
Figure 3. The physics of liquid fuel burning [22].

Figure 3. The physics of liquid fuel burning [22].
It was assumed that the heat of combustion of ethanol is 26,780 (kJ/kg) [23]. The thermal
It was assumed that the heat of combustion of ethanol is 26,780 (kJ/kg) [23]. The thermal
conductivity of the liquid ethanol is 0.17 (W/(m·K)). The specific heat of the liquid ethanol is 2.45
conductivity of the liquid ethanol is 0.17 (W/(m·K)). The specific heat of the liquid ethanol is
(kJ/(kg·K)) and its density is 787 (kg/m3) [24]. There are three reasons that lead to justify this
2.45 (kJ/(kg · 3
assumption: and its density is 787 (kg/m ) [24]. There are three reasons that lead to justify
K))
this assumption:
(1) The thermal diffusivity of the ethanol is low (the ratio between the thermal conductivity and the
(1) The heat capacity
thermal multiplied
diffusivity by the
of the density).
ethanol Therefore,
is low thebetween
(the ratio heat frontthe
penetrates
thermalwithin a short and
conductivity
distance
the heat insidemultiplied
capacity the ethanol by
pool.
the density). Therefore, the heat front penetrates within a short
(2) The radiative
distance heat
inside the flux is pool.
ethanol attenuated inside the liquid phase.
(3) The internal flow inside the liquid can be neglected (natural convection). Therefore, the
(2) The radiative heat flux is attenuated inside the liquid phase.
convective heat transfer inside the ethanol liquid pool can be neglected.
(3) The internal flow inside the liquid can be neglected (natural convection). Therefore, the convective
The FDS simulation is based on the lower heating values of the ethanol. This model is flexible.
heat transfer inside the ethanol liquid pool can be neglected.
There is an option to increase the heat release rate (HRR) by using fuel with higher heating value
(suchFDS
The as gasoline or diesel).
simulation is based on the lower heating values of the ethanol. This model is flexible.
There is an option to increase the heat release rate (HRR) by using fuel with higher heating value (such
2.3. Multiphysics Analysis of the Reformer Model
as gasoline or diesel).
The second part deals with numerical analysis of the reformer. Figure 4 shows the geometry of
2.3. Multiphysics
this system. Analysis of the Reformer Model
The second part deals with numerical analysis of the reformer. Figure 4 shows the geometry of
this system.
Steel Tube
Catalytic bed–
H2O + C2H5OH
Figure
Figure 4. 4.of3D
3D plot theplot of the reformer.
reformer.
TheThe reformer
reformer is made
is made of catalyst
of catalyst bed bed material
material and steel
and steel tube. tube.
It hasItbeen
has been assumed
assumed that the radius
that the
radius of the reformer is 0.2 m. The height of the reformer is 0.5 m. The thickness of the steel (HK-40
of the reformer is 0.2 m. The height of the reformer is 0.5 m. The thickness of the steel (HK-40 alloy)
alloy)
tubetube is 0.01
is 0.01 m (10
m (10 mm).mm). The
The reformationchemical
reformation chemical reactions
reactions occur
occurininaaporous
porouscatalytic bed
catalytic bed where the
where the heat is supplied through the ethanol burner to drive the endothermal reaction system. The
heat is supplied through the ethanol burner to drive the endothermal reaction system. The reactor is
reactor is enclosed with steel tube. Ethanol and steam are mixed in stoichiometric amounts and enter
enclosed with steel tube. Ethanol and steam are mixed in stoichiometric amounts and enter through
through the inlet of the catalytic reactor. COMSOL multi-physics software (version 4.3b) has been
the inlet of the catalytic reactor. COMSOL multi-physics software (version 4.3b) has been applied
applied in this work. It solved simultaneously the fluid flow, heat transfer, diffusion with chemical
in thiskinetics
reaction work.equations
It solvedandsimultaneously the fluid flow, heat transfer, diffusion with chemical reaction
structural analysis.
kinetics equations and structural analysis.
2.3.1. Model KineticsReformer Bed
2.3.1. Model Kinetics—Reformer Bed
Inside the reformer, steam and ethanol react to form hydrogen and carbon dioxide:
Inside the reformer, steam and ethanol react to form hydrogen and carbon dioxide:
C2 H5OH + 3H 2O  2CO2 + 6H 2 (13)
C H OH + 3H O → 2CO + 6H
The rate constant of the ESR reaction2 is 5temperature 2dependent according
2 2 to [8]: (13)
 isEtemperature
The rate constant of the ESR reaction  dependent according to [8]:
k =Aexp    (14)
 Rg T  E
k = A exp − (14)
Rg T
where A (frequency factor) is 7 × 10 (1/s) and E (activation energy) is 65.8 (kJ/mol) [15].
5
where A (frequency factor) is 7 × 105 (1/s) and E (activation energy) is 65.8 (kJ/mol) [15].
2.3.2. Fluid FlowReformer Bed
The flow
2.3.2. FluidofFlow—Reformer
gaseous species through
Bed the reformer bed is described by Darcy’s Law:
The flow of gaseous species through 
  the reformer bed is described by Darcy’s Law:
      psr   =0 (15)
     κ
∇· ρ − ∇ psr = 0 (15)
η
Here, ρ denotes the gas density (kg/m3 ), η is the viscosity (Pa·s) and κ is the permeability of the
porous medium (m2 ) and psr is the pressure inside the reformer bed (Pa). All other boundaries are
impervious, corresponding to the condition:
k
− ∇ psr ·n = 0 (16)
η
2.3.3. Energy Transport—Reformer Bed

The energy transport equation is used to describe the average temperature distribution in the
porous bed:
∂Tsr
ρC p + ∇·(−k sr ∇ Tsr ) + ρC p u·∇ Tsr = Q (17)
∂t
The volumetric heat capacity of the bed is given by:

ρC p t
= ε ρC p f
+ (1 − ε) ρC p s
(18)
In the above equations, the indices “f” and “s” denote fluid and solid phases, respectively, and ε
is the volume fraction of the fluid phase. Furthermore, Tsr is the temperature (K) and k sr is the thermal
conductivity of the reformer bed w/(m·K). Q represents a thermal source (w/m3 ), and u is the fluid
velocity field (m/s). Assuming that the porous medium is homogeneous and isotropic, the steady-state
equation becomes:

∇·(−k sr ∇ Tsr ) + ρC p u·∇ Tsr = Q (19)
The volumetric heat source due to the reaction is calculated:
Q = ∆Hr ·r (20)
where r represents the reaction rate. The steam reformation of ethanol is endothermic, with an enthalpy
of reaction of ∆Hr = 1.74 × 105 J/mol. Equation (19) also accounts for the conductive heat transfer in
the steel tube. As no reactions occur in this domain, the description reduces to:
∇·(−k steel ∇ Tsr ) = 0 (21)
where k steel is the thermal conductivity of the steel. The temperature of the gas is 700 K at the inlet.
At the outlet, it is assumed that convective heat transport is dominant:
n·(−k sr ∇ Tsr ) = 0 (22)
The thermal properties of the reformer are presented at Table 1 [9].
Table 1. Thermal properties of the reformer [10].
Material Property Value

rho 3960 (kg/m3 )
Cp 880 (J/(kg·◦ C))
k 33 (w/(m·◦ C))
2.3.4. Mass Transport—Reformer Bed

The mass-balance equations for the model are the Maxwell-Stefan diffusion and convection
equations at steady state:
!
n ∇p ∇T

∇· ρωi u − ρωi ∑ Dij ∇xj + xj − ωj
p
− DiT
T
= Ri (23)
j=1
Here ρ denotes the density (kg/m3 ), ωi is the mass fraction of species i, x j is the molar fraction of
species j, Dij is ij component of the multicomponent Fick diffusivity (m2 /s), DiT denotes the generalized
thermal diffusion coefficient (kg/(m·s)), T is the temperature (K), and Ri is the reaction rate (kg/(m3 ·s)).
2.3.5. Calculation of the Binary Diffusion Coefficients—Reformer Bed

The diffusion coefficients of the components in the mixture, Dij are obtained from the
Chapman-Enskog theory [25]:
r
1 1
T3 Mi + Mj
Dij = 5.9543·10−4 (24)
pσij2 Ωij
In which σij and Ωij are the Lennard-Jones collision diameter and collision integral between
one molecule of i and one molecule of j respectively. The collision diameter is obtained from
σij = 1/2 σi + σj and Ωij is taken from [26]:

1.06036 0.19300 1.03587 1.76474

Ωij = 2
+ + + (25)
τij exp 0.47635τij exp 1.52996τij exp 3.89411τij
In which τij = kT/ε ij is the dimensionless temperature, where k is Boltzmann constant and
√
ε ij = ε i ε j is the maximum attractive energy between one molecule of i and one molecule of k.
The values of σi and ε i for the substances used in this work are given in Table 2 [27].
Table 2. Properties of the gaseous species [27].
Lennard-Jones Collision Diameter σi (Angstroms)

Ethanol 4.455
H2 O 2.655
CH4 3.882
CO 3.590
CO2 3.996
H2 2.968
Maximum Attractive Energy between Two molecules σi /k B (K)
Ethanol 391.0
H2 O 363.0
CH4 136.5
CO 110.3
CO2 190.0
H2 33.3
The diffusion coefficients as a function of the temperature are shown in Figure 5.

The diffusion coefficients as a function of the temperature are shown in Figure 5.
Figure 5. The binary diffusion coefficients of the main species as a function of the temperature.
Figure 5. The binary diffusion coefficients of the main species as a function of the temperature.
This figure clearly shows that the spreading (diffusion) gaseous species inside the reformer
This figure clearly shows that the spreading (diffusion) gaseous species inside the reformer
increases with the temperature.
increases with the temperature.
2.3.6. Structural Analysis—Reformer Bed and Steel Tube
2.3.6. Structural Analysis—Reformer Bed and Steel Tube
It was assumed that the tube is made of Steel HK-40. HK alloy, known as HK 40, is an austenitic
It was assumed
Fe-Cr-Ni that
alloy that hasthe tube
been is made heat
a standard of Steel HK-40.
resistant HK alloy,
material. Withknown as HK
moderately 40,temperature
high is an austenitic
Fe-Cr-Ni
strength, oxidation resistance, the alloy is used in a wide variety of industrial applications temperature
alloy that has been a standard heat resistant material. With moderately high such as:
strength, oxidation
Ammonia, resistance,
methanol the alloyreformers;
and hydrogen is used in a widepyrolysis
ethylene variety of industrial
coils applications
and fittings; such as:
steam super
heater methanol
Ammonia, tubes and fittings [28]. Thereformers;
and hydrogen thermo-physical and pyrolysis
ethylene thermomechanical
coils andproperties of the steel
fittings; steam superare
heater
tubeslisted in Table[28].
and fittings 3 [10,29].
The thermo-physical and thermomechanical properties of the steel are listed in
Table 3 [10,29].
Table 3. Thermo-physical and thermomechanical properties of steel HK-40 [10,27].
Table 3. Thermo-physical and thermomechanical

Material Property properties of steel HK-40 [10,27].
Value
E 186 × 109 (Pa)
Material Property
nu Value
0.3
Erho 7720
186 ×(kg/m
3)
109 (Pa)
alpha
nu 17.64 × 10 −6 (1/°C)
0.3
Cp
rho 502 (kg/m3 )
(J/(kg·°C))
7720
k
alpha 29.58×
17.64 10−6 (1/◦ C)
(w/(m·°C))
Cp 502 (J/(kg·◦ C))
3. Results k 29.58 (w/(m·◦ C))
This section divided into two parts. In Section 3.1 the thermal results of Fire Dynamics
3. Results
Simulation (FDS) software are shown. In Section 3.2 the COMSOL multi-physics results for the
reformer are presented.
This section divided into two parts. In Section 3.1 the thermal results of Fire Dynamics Simulation
(FDS)3.1.
software are shown.
Fire Dynamics Simulator In Section
Software 3.2 the COMSOL multi-physics results for the reformer
Results
are presented.
One advantage of FDS simulation is that it can provide much detailed information on the
ethanol burner,
3.1. Fire Dynamics including
Simulator the Results
Software local and transient gas velocity, gas temperature, species
One advantage of FDS simulation is that it can provide much detailed information on the ethanol
burner, including the local and transient gas velocity, gas temperature, species concentration, solid wall
temperature, fuel burning rate, radiative heat flux, convective heat flux and HRR. The temperature
field at t = 77.5 s is shown in Figure 6.
concentration,
ChemEngineering 2018, solid
2, 34 wall temperature, fuel burning rate, radiative heat flux, convective heat flux and13 of 24
concentration, solid wall temperature, fuel burning rate, radiative heat flux, convective heat flux and
HRR. The temperature field at t = 77.5 s is shown in Figure 6.
HRR. The temperature field at t = 77.5 s is shown in Figure 6.
Figure Temperature
6. 6.
Figure Temperaturefield ◦ C) inside
field ((°C) insidethe
theburner
burnerat tat= t77.5
= 77.5
s. s.
Figure 6. Temperature field (°C) inside the burner at t = 77.5 s.
The maximal temperature at t = 77.5 ss approaches

approaches to 815 ◦ The
°C. velocity field of flue gases is
The The
maximal
maximal temperatureatatt t==77.5
temperature 77.5 s approaches toto815
815°C. C.
TheThe velocity
velocity fieldfield of gases
of flue flue gases
is is
shown
shown in Figure
in Figure 7.
7. 7.
shown in Figure
Figure 7. Velocity field (m/s) of the flue gases at t = 77.5 s.

Figure
Figure 7. 7.Velocity
Velocityfield
field(m/s)
(m/s) of
ofthe
theflue gases
flue at tat= t77.5
gases s. s.
= 77.5
Figure 8 shows the radiation heat flux emitted by the burner.
Figure
Figure 8 shows
8 shows thethe radiationheat
radiation heatflux
flux emitted
emitted by
bythe
theburner.
burner.
Figure 8. Radiation heat flux produced by the ethanol burner.

Figure 8. Radiation heat flux produced by the ethanol burner.
Figure 8 shows that the maximal radiation heat flux is 100,000 (W/m2). It should be noted that
Figure 8 shows that Figure 8. Radiation
the maximal heat fluxheat
radiation produced
flux by
is the ethanol
100,000 burner.2 ). It should be noted that
(W/m
the FDS simulation is based on the lower heating values of the ethanol. This model is flexible. There
the FDS
is ansimulation is basedthe
option 8toshows
increase onHRR
the by
lower heating
using values
fuel with of heating
higher the ethanol.
value This
(such model
2). It as
is flexible.
gasoline There is
or diesel).
Figure that the maximal radiation heat flux is 100,000 (W/m should be noted that
an option
As
the we
to simulation
increase
FDSshall
the
see later, HRR
is the HHR
based
bythe
using
on produced
fuel
lower by
with
the
heating
higherofheating
ethanol
values burner, canvalue
provide
the ethanol.
(such asisgasoline
the necessary
This model
orflux
diesel).
heatThere
flexible.
As weis an option to increase the HRR by using fuel with higher heating value (such as gasoline or diesel). flux
shall
required see
for later, the
maintaining HHRthe produced
reforming by the
process. ethanol burner, can provide the necessary heat
required
As we forshall
maintaining theHHR
see later, the reforming
producedprocess.
by the ethanol burner, can provide the necessary heat flux
3.2. Multi-Physics
required Analysisthe
for maintaining of the Operation
reforming of the Ethanol Steam Reformer
process.
3.2. Multi-Physics Analysis of the Operation of the Ethanol Steam Reformer
A parametric study has been performed in order to analyze the influence of the heat flux on the
3.2. Multi-Physics Analysis of the Operation of the Ethanol Steam Reformer
reformer performance andbeen
its structural integrity.
A parametric study has performed in orderThe numerical
to analyze theresults for heat
influence flux
of the of flux
heat 95,300on the
(W/m 2) are presented in Section 3.2.1. The numerical results for heat flux of 200,000 (W/m2) are
reformer performance and its structural integrity. The numerical results for heat flux of 95,300 (W/m2 )
presented in Section 3.2.2.
reformer performance and The
its structural
are presented in Section 3.2.1. numericalintegrity. Theheat
results for numerical
flux of results
200,000for heat 2flux
(W/m ) areofpresented
95,300 in
(W/m
Section
2) are presented in Section 3.2.1. The numerical results for heat flux of 200,000 (W/m2) are
3.2.2.
3.2.1. Multiphysics Results for Low Heat Flux Input
presented in Section 3.2.2.
Figure 9 shows
3.2.1. Multiphysics the 3D
Results fortemperature field Input
Low Heat Flux inside the reformer.
3.2.1. Multiphysics Results for Low Heat Flux Input
Figure 9 shows the 3D temperature field inside the reformer.
Figure 9 shows the 3D temperature field inside the reformer.
Figure 9. 3D plot of the reformer temperature field.

As can be seen
ChemEngineering 2018,from
2,from Figure
x FOR 9, the temperature at the bottom section of reformer is lower
PEER REVIEW 24 than
15 of than
As can be seen Figure 9, the temperature at the bottom section of reformer is lower
the temperature at the upper side. The temperature of the steel is much higher than the temperature
the temperature at the upper side. The temperature of the steel is much higher than the temperature
of of
thethe As can be seen
catalyst. from Figure 9, thereasons:
temperature at thethe
bottom section of reactions
reformer is lower than
catalyst.This
This is because
is because ofoftwo
two Firstly,
reasons: Firstly, endothermic
the endothermic reactions absorbabsorb theand
the heat, heat,
and the temperature at the upper side. The temperature of the steel is much higher than the temperature
secondly, the catalyst is much thicker than the steel tube. Figure 10 shows the mass fractions of thethe
secondly, the catalyst is much thicker than the steel tube. Figure 10 shows the mass fractions of
of the catalyst. This is because of two reasons: Firstly, the endothermic reactions absorb the ◦ heat, and
species (ethanol,
species (ethanol,CO CO2 and
2 and H
H22O)O) along
along the reformer axis
the reformer axisfor
forinlet
inlettemperature
temperatureofof600 600
°C.C.
secondly, the catalyst is much thicker than the steel tube. Figure 10 shows the mass fractions of the
species (ethanol, CO2 and H2O) along the reformer axis for inlet temperature of 600 °C.
Figure
10.10. Mass
Mass fractions
fractions ofofof
the
the species
thespecies (ethanol,
species(ethanol, CO and
andHHH
CO222and 2O) along the reformer axis.
Figure
Figure 10. Mass fractions (ethanol, CO 2 O)
2O) along
along thethe reformer
reformer axis.axis.
TheTheThe
massmass
mass fractions
fractions
fractions ofofethanol
of the the
the ethanol
ethanol
and steamand
and steam
steam decay
decay decay
along the along
along the reformer
the
reformer reformer axis.
axis.
axis. The TheTheethanol
ethanol ethanol
conversion
conversion
conversion is 80.3%.
is 80.3%.The
The ethanol
ethanol and
and the
the steam
steam decays
decays at
at the
the same
same slope,
slope,
is 80.3%. The ethanol and the steam decays at the same slope, and according to Equation (13) the and
and according
according to to Equation
Equation
(13)(13)
amounts thethe
ofamounts
amounts
ethanol ofofethanol
and ethanol
the steamand
andthe steam
steam are
theproportional
are are proportional
proportional to each
to
to each other. each other.
other.
Similar Similar
Similar
values values
values
have been have
have been in
been
reported
reported
reported in in
Reference
Reference[4].[4].From
Fromthisthis figure
figure itit can
can be seen
seen that
that thetheHRR
HRRproduced
producedbybythethe ethanol
ethanol
Reference [4]. From this figure it can be seen that the HRR produced by the ethanol burner, can provide
burner,
burner,cancanprovide
providethetherequired
requiredheat heatflux
flux for
for maintaining
maintaining the the reforming
reformingprocess.
process.Figure
Figure 1111 shows
shows
the required heat flux for maintaining the reforming process. Figure 11 shows the mass fractions of the
thethe mass
mass fractions
fractions ofofthe
theethanol
ethanolalong
alongthe thereformer
reformer axis for
for inlet
inlet temperatures
temperatures of 400
of 400°C,
°C,500 °C
500 and
°C and
ethanol along the reformer axis for inlet temperatures of 400 ◦ C, 500 ◦ C and 600 ◦ C.
600600
°C.°C.
Figure 11. Mass fractions of the ethanol along the reformer axis for various inlet temperatures.

FromFrom this figure it can be seen that for inlet temperature of 500
this figure it can be seen that for inlet temperature of 500 ◦ C,°C, thethe ethanol conversion is
ethanol conversion is 70.0%.
70.0%. For inlet temperature◦ of 400 °C, the ethanol conversion is 43.8%. Thus, the ethanol conversion
From this figure
For inlet temperature it can
of 400 C, be
theseen that conversion
ethanol for inlet temperature
is 43.8%. of 500 the
Thus, °C, the ethanol
ethanol conversion
conversion is
increases
increases with the inlet temperature. Figure 12 shows the mass fractions of hydrogen along the
70.0%.
with the For inlet temperature of 400 °C, the ethanol conversion is 43.8%. Thus, the ethanol
inlet temperature. Figure 12 shows the mass fractions of hydrogen along the reformer axis. conversion
reformer axis.
increases with the inlet temperature. Figure 12 shows the mass fractions of hydrogen along the
reformer axis.
Figure 12. Mass fractions of the H2 along reformer axis.

This figure clearly shows a considerable increasing of hydrogen mass fraction along the
This figure
reformer clearly
axis. shows a in
The increase considerable
Hydrogen massincreasing
fractionofishydrogen
89.4%. The mass
sumfraction along the
of hydrogen, reformer
ethanol,
This
axis. carbon figure
The increase clearly
inand
Hydrogen shows
massa considerable
fraction increasing
is 89.4%. The of hydrogen
sum is mass
of1hydrogen, fraction along the
dioxide, steam mass fraction at the reformer output as expectedethanol,
accordingcarbon dioxide,
the mass
reformer axis. The increase in Hydrogen mass fraction is 89.4%. The sum of hydrogen, ethanol,
and steam mass fraction
conservation law. Theat 3Dthe reformerstress
Von-Mises output is 1 as expected
distribution according
of reformer is shownthe mass conservation
in Figure 13. law.
carbon dioxide, and steam mass fraction at the reformer output is 1 as expected according the mass
The 3D Von-Mises stress distribution of reformer is shown in Figure 13.
conservation law. The 3D Von-Mises stress distribution of reformer is shown in Figure 13.
Figure 13. 3D plot of the reformer Von-Mises stress field.

As expected, the Figure
maximal 13.stress is located
3D plot at the reformer
of the reformer head
Von-Mises corner.
stress The Von-Mises thermal
field.
stresses which is developed inside the steel are much higher than the Von-Mises stresses which are
As expected, the maximal stress is located at the reformer head corner. The Von-Mises thermal
developed inside the catalyst material. This is because the temperatures inside the catalyst material
stresses
As which the
expected, is developed
maximalinside
stressthe steel areat
is located much higher than
the reformer the Von-Mises
head corner. The stresses which thermal
Von-Mises are
developed
stresses which inside the catalyst
is developed material.
inside Thisare
the steel is because the temperatures
much higher inside the catalyst
than the Von-Mises material
stresses which are
developed inside the catalyst material. This is because the temperatures inside the catalyst material
are much lower than the temperature in the steel tube (Since the reforming reaction is endothermic, it
are more
absorbs much heat
lowerandthancools
the temperature in thematerial).
of the reformer steel tube (Since the reforming
According reaction
to this figure, is maximal
the endothermic,
stress in
it absorbs
the steel more heattoand
tube reached cools ofvalue
maximal the reformer material). According to this figure, the maximal stress
of 1.3 GPa.
in the steel tube reached to maximal value of 1.3 GPa.
are much lower than the temperature in the steel tube (Since the reforming reaction is endothermic,
3.2.2.itMultiphysics Results
absorbs more heat and for High
cools Heat
of the Flux material).
reformer Input According to this figure, the maximal stress
3.2.2. Multiphysics Results for High Heat Flux Input
in thenumerical
The steel tube reached
results to
formaximal value
heat flux of of 1.3 GPa.
200,000 (W/m2 ) are presented in this section. Figure 14
The numerical results for heat flux of 200,000 (W/m2) are presented in this section. Figure 14
shows the 3D temperature field inside the reformer.
3.2.2.
showsMultiphysics Results field
the 3D temperature for High Heat
inside theFlux Input
reformer.
The numerical results for heat flux of 200,000 (W/m2) are presented in this section. Figure 14
shows the 3D temperature field inside the reformer.

The maximal temperature obtained in the steel jacket for the case of 200,000 (W/m2) is about 652
°C. Figure
The 15 shows
maximal the mass fractions
temperature obtained ofin
thethe
species
steel(ethanol, COthe
jacket for 2 and H2O)
case ofalong (W/m2 )axis.
the reformer
200,000 is about
652 ◦ C. Figure 15 shows
The maximal the mass
temperature fractions
obtained of the
in the steel species
jacket for the(ethanol, CO2(W/m
case of 200,000 and2)H O) along
is2about 652 the
reformer axis. 15 shows the mass fractions of the species (ethanol, CO2 and H2O) along the reformer axis.
°C. Figure
Figure 15. Mass fractions of the species (ethanol, CO2 and H2O) along the reformer axis.
It can be seen from Figures 10 and 15 that the mass fractions of ethanol and steam decay with
Figure 15. Mass fractions of the species (ethanol, CO2 and H2O) along the reformer axis.
increasing heat
Figure 15.flux.
MassThe ethanol
fractions of conversion
the species is about 85.4%.
(ethanol, Figure
CO2 and H2 O)16along
shows
thethe mass fractions
reformer axis. of
hydrogen along the reformer axis.
It can be seen from Figures 10 and 15 that the mass fractions of ethanol and steam decay with
It can be seen
increasing heat from Figures
flux. The 10 conversion
ethanol and 15 that is the mass
about fractions
85.4%. Figure of
16 ethanol
shows the and steam
mass decay
fractions of with
increasing
hydrogenheat flux.the
along The ethanol
reformer conversion is about 85.4%. Figure 16 shows the mass fractions of
axis.
hydrogen along the reformer axis.
Figure 16. Mass fractions of the H2 along the reformer axis.

Figure 16 indicates that the increase in hydrogen mass fraction is 89.9%. The hydrogen mass
Figure 16 indicates
Figure that
16 indicates thetheincrease
that increasein hydrogen massfraction
fractionis is 89.9%. The hydrogen
massmass
fraction is increased. Figure 17 shows the in hydrogen distribution
temperature mass along 89.9%.
the z The
axis hydrogen
of the reformer
fraction is increased.
fraction Figure
is increased. temperature distribution along the z
Figure 17 shows the temperature distribution along the z axis of the reformer for
17 shows the axis of the reformer
for the two cases.
the two cases.
for the two cases.
Figure 17. Temperature distribution along the z axis of the reformer.

It can be seen from Figure 17 that the temperature increases with the increasing heat flux. As a
resultIt can bethe
seen from Figure 17fraction
that theincreases
temperature increases with the increasing
The heat flux. As a
It can ofbethat
seen hydrogen
from Figuremass
17 that the temperature(the reforming
increasesrate
withincreases). 3D Von-Mises
the increasing heat flux. As a
result of that the hydrogen mass fraction increases (the reforming rate increases). The 3D Von-Mises
resultstress distribution
of that of reformer
the hydrogen massisfraction
shown in Figure 18.
increases (the reforming rate increases). The 3D Von-Mises
stress distribution of reformer is shown in Figure 18.
stress distribution of reformer is shown in Figure 18.

Figure 18.18.3D
Figure 3Dplot
plotof
of the
the reformer Von-Misesstress
reformer Von-Mises stress field.
field.
Figure 19 shows the Von-Mises distribution along the z axis of the steel tube for the two cases.
Figure 19 shows
Figure thethe
19 shows Von-Mises distribution
Von-Mises distributionalong the zzaxis
along the axisofofthe
thesteel
steel tube
tube forfor
the the
twotwo cases.
cases.
Figure 19. Von-Mises distribution along the z axis of the steel tube.
Figure 19 indicates that Von-Mises stresses increases with the heat flux. The effect of increasing
Figure 19 indicates that Von-Mises stresses increases withand
the heat
Figure fluxes
19 increases
indicates thatthe temperature
Von-Mises inside the
stresses increases withthe
steel tube theheat flux.the
enhances The
heat flux. effect of
thermal
The increasing
stresses.
effect The
of increasing
the heat
thermal fluxes
stressesincreases
which the
are temperature
developed inside
inside the
the steel
steel tube
tube areand enhances
much higherthe thermal
than those stresses.
which The
were
the heat fluxes increases the temperature inside the steel tube and enhances the thermal stresses.
thermal
obtainedstresses
in the which are developed
previous section. Theinside the steel
increase tubeheat
of the are much higher than
flux enhances a those
little which were
the ethanol
The thermal
obtained
stresses
in but
which are
the also
previous
developed
section.
inside the
The increase
steel tube are much higher than those which
conversion, increases the thermal stressesofdeveloped
the heat inside
flux enhances a little
the steel tube. Thistheeffect
ethanol
can
wereconversion,
obtained in butthe previous
also increasessection. Thestresses
the thermal increase of the heat
developed flux
inside theenhances
steel tube. aThis
little the can
effect ethanol
conversion, but also increases the thermal stresses developed inside the steel tube. This effect can cause
to mechanical failure in the steel tube. In order to maintain the structural integrity of the tube, HK-40
alloy has been applied in this work. HK alloy, known as HK 40, is an austenitic Fe-Cr-Ni alloy that
cause to mechanical failure in the steel tube. In order to maintain the structural integrity of the tube,
HK-40 alloy has been applied in this work. HK alloy, known as HK 40, is an austenitic Fe-Cr-Ni alloy
has
thatbeen
has abeen
standard heat resistant
a standard materialmaterial
heat resistant with moderately high temperature
with moderately strength, and
high temperature oxidation
strength, and
resistance. Figure 20 shows the effect of temperature increasing on the
oxidation resistance. Figure 20 shows the effect of temperature increasing on the tensile ultimate tensile ultimate strength of HK
40 alloy and
strength of HK31040 stainless
alloy and steels
310[30].
stainless steels [30].
According
AccordingtotoFigure Figure20, 20,the
the maximal
maximal stress in the
stress steel
in the tubetube
steel reached to maximal
reached valuevalue
to maximal of 482ofMPa 482
(70 ksi). From Figures 19 and 20 it can be seen that the thermal stresses developed
MPa (70 ksi). From Figures 19 and 20 it can be seen that the thermal stresses developed for low heat for low heat flux are
much lower
flux are much thanlowerthethan
ultimate tensile strength
the ultimate of the HK-40
tensile strength of the (According to Figure
HK-40 (According 20, the 20,
to Figure Vonthe Mises
Von
stresses developed in the middle section of the tube are less
Mises stresses developed in the middle section of the tube are less than 450 MPa). than 450 MPa).
From
From Figure
Figure 20, 20, itit can
can be be also
also seen
seen that
that the
the decrease
decrease in inthe
theultimate
ultimate tensile
tensile strength
strength isis less
less than
than
10% ◦ C or 1112 ◦ F). Since ethanol steam reforming occurs at
10%(The
(Thetemperature
temperatureofofthe thesteel
steelisisabout
about600600 °C or 1112 °F). Since ethanol steam reforming occurs
relatively lower
at relatively temperatures
lower temperatures compared with other
compared withhydrocarbon fuels, thefuels,
other hydrocarbon decrease
the indecrease
the strength
in theof
the Steel tube
strength of the is low.
SteelTherefore,
tube is low. theTherefore,
structural integrity of the integrity
the structural HK-40 steel tubeHK-40
of the is kept.steel
Figure
tube21isshows
kept.
the effect
Figure 21of temperature
shows the effect stress rupture of HK
of temperature 40-2
stress steel alloy
rupture of HK [30].
40-2 steel alloy [30].
Figure
Figure2121clearly clearlyindicates
indicates that under normalnormal
that under conditions of the ethanol
conditions of thereforming
ethanol(the temperature
reforming (the
of the steel tube is about 600 is ◦ C about
or 1112 ◦
temperature of the steel tube 600F),
°Cthe
or rupture
1112 °F),time increases
the rupture considerably.
time For this case the
increases considerably. For
rupture time is greater than 100,000 h (more than 11.4
this case the rupture time is greater than 100,000 h (more than 11.4 years). years).
Figure20.
Figure 20.The
Theeffects
effectsofofthe
thetemperature
temperatureon
onthe
theultimate
ultimate tensile
tensile strength
strength of
of HK
HK 40
40 and
and 310
310 steels
steels [28].
[28].
Figure 21. The effects of the temperature on the rupture time of HK 40-2 alloy [30].
Figure 21. The effects of the temperature on the rupture time of HK 40-2 alloy [30].
4. Discussion
4. Discussion
Hydrogen could be a promising fuel, and if often considered as a clean energy carrier as it only
Hydrogen could be a promising fuel, and if often considered as a clean energy carrier as it only
emits water during combustion. It can be produced from different kinds of renewable feedstocks,
emitssuchwater during combustion.
as ethanol. The use of ethanol It canasberaw produced
materialfrom different
presents severalkinds of renewable
advantages because feedstocks,
it is a
such renewable
as ethanol. The use
feedstock, of to
easy ethanol as raw
transport, material presents
biodegradable, several contains
has low toxicity, advantages highbecause
hydrogenit is a
renewable
content,feedstock,
and easyeasy to to transport,
store biodegradable,
and handle. Ethanol steam has low toxicity, contains
reformation occurs at high hydrogen
relatively content,
lower
and easy to storecompared
temperatures and handle. with Ethanol steam reformation
other hydrocarbon fuels and has occurs at relatively
been widely studiedlower due totemperatures
the high
yield provided
compared with other forhydrocarbon
the formation fuels of hydrogen.
and hasInbeen this reformation
widely studied reaction,
due water
to theand high ethanol
yield react
provided
for theover a catalystof
formation bed to produce
hydrogen. Inathis
mixture of hydrogen
reformation rich gas.
reaction, A new
water andtool has been
ethanol reactdeveloped in
over a catalyst
this work in order to analyze the reformer operation. The
bed to produce a mixture of hydrogen rich gas. A new tool has been developed in this work proposed computational work is in
composed of two phasessimulation of ethanol burner by using FDS software and multiphysics
order to analyze the reformer operation. The proposed computational work is composed of two
simulation of steam reforming process occurring inside the reformer. This software calculates the
phases—simulation of ethanol burner by using FDS software and multiphysics simulation of steam
temperature, density, pressure, velocity, and chemical composition within each numerical grid cell
reforming
at eachprocess
discrete occurring
time step. Itinside
computes the the
reformer. This heat
temperature, software calculates
flux, and mass loss the temperature,
rate. There are three density,
pressure,
major components of the model: Hydrodynamic, combustion, and radiation models. FDS code is time
velocity, and chemical composition within each numerical grid cell at each discrete
step. formulated
It computes the on
based temperature, heat flux,
CFD of fire-driven fluidand mass
flow. ThelossFDSrate. Theresolution
numerical are three canmajor components
be carried out
of theusing
model: Hydrodynamic,
either a DNS methodcombustion,LES. FDS uses anda radiation models.
mixture fraction FDS codemodel.
combustion is formulated
The mixturebased on
CFD fraction is a conserved
of fire-driven scalarThe
fluid flow. quantity
FDS that is defined
numerical as the fraction
solution can be of gas at out
carried a given
using point in the
either a DNS
flow field that originated as fuel. The model assumes that combustion
method LES. FDS uses a mixture fraction combustion model. The mixture fraction is a conserved is mixing controlled, and that
scalarthe reactionthat
quantity of fuel and oxygen
is defined is infinitely
as the fractionfast. Radiative
of gas heat transfer
at a given point inistheincluded in thethat
flow field model via
originated
the solution of the radiation transport equation for a non-scattering gray gas. The radiation equation
as fuel. The model assumes that combustion is mixing controlled, and that the reaction of fuel and
is solved using a technique similar to a finite volume method for convective transport, thus the name
oxygen is infinitely fast. Radiative heat transfer is included in the model via the solution of the
given to it is the FVM. One advantage of FDS simulation is that it can provide much detailed
radiation transport
information equation
on the for a non-scattering
fire, including the local and gray gas.gas
transient Thevelocity,
radiation gasequation
temperature, is solved
speciesusing
a technique similarsolid
concentration, to awall
finite volume method
temperature, compositefor convective
burning rate,transport, thustransfer,
radiation heat the name given to it is
convection
the FVM. One advantage
heat transfer and HRR. of FDSbeen
It has simulation
found out is that
thatthe
it can provide
maximal much detailed
temperature at t = 77.5 information
s approaches on the
fire, including
to 815 °C. Thethe magnitude
local and transient gas velocity,
of the radiation heat fluxgas temperature,
is 95,300 (w/m2). species concentration, solid wall
temperature, The composite
second partburningdeals with rate,numerical
radiation analysis of theconvection
heat transfer, reformer. The heatreformation
transfer and chemical
HRR. It has
been found out that the maximal temperature at t = 77.5 s approaches to 815 C. The magnitudetoof the
reactions occur in a porous catalytic bed where the heat is supplied through ◦
the Ethanol burner
driveheat
radiation the endothermal
flux is 95,300reaction
(w/m2 ).system. The reactor is enclosed with a HK-40 alloy steel tube.
Ethanol and steam are mixed in stoichiometric amounts and enter through the inlet of the catalytic
The second part deals with numerical analysis of the reformer. The reformation chemical reactions
reactor. The COMSOL multi-physics software has been applied in this work. It solved
occur in a porous catalytic bed where the heat is supplied through the Ethanol burner to drive the
simultaneously the fluid flow, Heat transfer, diffusion with chemical reaction kinetics equations and
endothermal reaction system. The reactor is enclosed with a HK-40 alloy steel tube. Ethanol and
steam are mixed in stoichiometric amounts and enter through the inlet of the catalytic reactor.
The COMSOL multi-physics software has been applied in this work. It solved simultaneously the
fluid flow, Heat transfer, diffusion with chemical reaction kinetics equations and structural mechanics.
The diffusion equations values are obtained from the Chapman-Enskog theory. As far as I know,
this work is the first CFD structural simulation study of ESR system. It is probably the first time that
FDS software has been applied to simulate the ethanol burner.
reformer performance and its structural integrity. The numerical results were obtained for heat fluxes
of 95,300 (W/m2 ) and 200,000 (W/m2 ). It is shown that the HRR produced by the ethanol burner, can
provide the necessary heat flux required for maintaining the reforming process.
It has been found out that the mass fractions of the ethanol and steam decays along the
reformer axis. The hydrogen and carbon dioxide mass fraction are increased along the reformer
axis. The hydrogen mass fraction increases with enhancing the radiation heat flux. For the first
case (heat flux magnitude of 95,300 (W/m2 )) it was found out that the ethanol conversion is 80.3%.
The ethanol and the steam are decaying at the same rate. Similar values have been reported in the
literature. A sensitivity test has been carried out in order to analyze the influence of inlet temperature
on ethanol conversion. The results show that for inlet temperature of 500 ◦ C, the ethanol conversion
is 70.0%. For inlet temperature of 400 ◦ C, the ethanol conversion is 43.8%. Therefore, the ethanol
conversion increases with the inlet temperature. There is a considerable increasing of Hydrogen mass
fraction along the reformer axis. The increase in hydrogen mass fraction is 89.4%. The sum of hydrogen,
ethanol, carbon dioxide and steam mass fraction at the reformer output is 1, as expected according the
mass conservation law.
Similar physical behavior has been observed in the second case. The ethanol conversion is about
85.4%. The increase in hydrogen mass fraction along the reformer is 89.4%. The thermal stresses which
are developed inside the steel tube are higher for the case of heat flux of 200,000 (W/m2 ). The increase
of the heat flux enhances a little the ethanol conversion, but also increases the thermal stresses of
the steel tube. This effect can cause to mechanical failure in the steel tube. In order to maintain the
structural integrity of the tube, HK-40 alloy has been applied in this work. HK alloy, known as HK
40, is an austenitic Fe-Cr-Ni alloy that has been a standard heat resistant material with moderately
high temperature strength, and oxidation resistance. This alloy is used in a wide variety of industrial
applications such as: Ammonia, methanol, and hydrogen reformers; ethylene pyrolysis coils and
fittings; and steam super heater tubes and fittings.
It has been found out that by increasing the heat flux of the reformer, the decrease in the tensile
yield strength of HK-40 alloy is less than 10% (the temperature of the steel is about 600 ◦ C or 1112 ◦ F).
The effect of increasing the heat fluxes increases the temperature inside the steel tube and enhances
the thermal stresses (produced by the thermal expansion). Since ethanol steam reforming occurs at
relatively lower temperatures compared with other hydrocarbon fuels, the decrease in the strength of
the steel tube is low. Thus, the structural integrity of the HK-40 steel tube is kept.
The numerical results for clearly indicates that under normal conditions of the ethanol reforming
(the temperature of the steel tube is about 600 ◦ C or 1112 ◦ F), the rupture time increases considerably.
For this case the rupture time is greater than 100,000 h (more than 11.4 years). The described algorithm
described in this work may be applied other reformer system (diesel, methanol, or methane).
5. Conclusions
A new tool has been developed in this work in order to analyze the burner and steam reformer
operation. This model may be implemented for simulating other reforming systems (such as methanol,
and diesel). Future work will focus on performing coupled heat transfer, chemical reactions, and creep
analyses on the reformer.
Funding: This research received no external funding.

Conflicts of Interest: The author declares no conflict of interest. No funding sponsors had any role in the
numerical analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish
the results.
References
1. Poran, A.; Tartakovsky, L. Performance and emissions of a direct injection internal combustion engine devised
for joint operation with a high-pressure thermochemical recuperation system. Energy 2017, 124, 214–226.
[CrossRef]
2. Akande, A.; Aboudheir, A.; Idem, R.; Dalai, A. Kinetic modeling of hydrogen production by the catalytic
reforming of crude ethanol over a co-precipitated Ni-Al2 O3 catalyst in a packed bed tubular reactor. Int. J.
Hydrogen Energy 2006, 31, 1707–1715. [CrossRef]
3. Marino, F.; Boveri, M.; Baronetti, G.; Laborde, M. Hydrogen production from steam reforming of bioethanol
using Cu/Ni/K/γ-Al2 O3 catalysts. Effect of Ni. Int. J. Hydrogen Energy 2001, 26, 665–668. [CrossRef]
4. Bineli, A.R.; Tasić, M.B. Catalytic Steam Reforming of Ethanol for Hydrogen Production: Brief Status.
Chem. Ind. Chem. Eng. Q. 2016, 22, 327–332. [CrossRef]
5. Akande, A.J. Production of Hydrogen by Reforming of Crude Ethanol. Master’s Thesis, University of
Saskatchewan, Saskatoon, SK, Canada, 2005.
6. Athanasios, N.F.; Kondaridesm, D.I. Production of hydrogen for fuel cells by reformation of biomass-derived
ethanol. Catal. Today 2002, 75, 145–155.
7. Klouz, V.; Fierro, V.; Denton, P.; Katz, H.; Lisse, J.P.; Bouvot-mauduit, S.; Mirodatos, C. Ethanol reforming for
hydrogen production in a hybrid electric vehicle: Process optimization. J. Power Sources 2002, 105, 26–34.
[CrossRef]
8. Uriz, I.; Arzamendi, G.; López, E.; Llorca, J.; Gandía, L.M. Computational fluid dynamics simulation of
ethanol steam reforming in catalytic wall micro channels. Chem. Eng. J. 2011, 167, 603–609. [CrossRef]
9. Uriz, I.; Arzamendi, G.; Diéguez, P.M.; Gandía, L.M. Computational fluid dynamics as a tool for designing
hydrogen energy technologies. In Renewable Hydrogen Technologies: Production, Purification, Storage,
Applications and Safety; Elsevier: Oxford, UK, 2013; Chapter 17; pp. 401–435.
10. Lao, L.; Aguirre, A.; Tran, A.; Wu, Z.; Durand, H.; Christofides, P.D. CFD modeling and control of a steam
methane reforming reactor. Chem. Eng. Sci. 2016, 148, 78–92. [CrossRef]
11. Lewis, B.; Von Elbe, G. Combustion Flames and Explosion of Gases, 2nd ed.; Academic Press Inc.: New York, NY,
USA; London, UK, 1961.
12. Diéguez, P.M.; López-San, M.J.; Idareta, I.; Uriz, I.; Arzamendi, G.G.; Luis, M. Hydrogen Hazards and Risks
Analysis through CFD Simulations. In Renewable Hydrogen Technologies: Production, Purification, Storage,
Applications and Safety; Elsevier: Oxford, UK, 2013; Chapter 18; pp. 437–452.
13. Tartakosky, L.; Sheintoch, M. Fuel Reforming in Internal Combustion Engines. Prog. Energy Combust. Sci.
2018, 67, 88–114. [CrossRef]
14. European Industrial Gases Association AISBL. Combustion Safety for Steam Reformer Operation; IGC Doc
172/12/E; European Industrial Gases Association AISBL: Brussels, Belgium, 2012.
15. Olafadehan, O.A.; Ayoola, A.A.; Akintunde, O.O.; Adeniyi, V.O. Mechanistic Kinetic Models for Steam
Reforming of Concentrated Crude Ethanol on Ni/Al2O3 Catalyst. J. Eng. Sci. Technol. 2015, 10, 633–653.
[CrossRef]
16. Loock, S.; Ernst, W.S.; Thomsen, S.G.; Jensen, M.F. Improving Carbon Efficiency in an Auto-Thermal Methane
Reforming Plant with Gas Heated Heat Exchange Reforming technology. In Proceedings of the 7th World
Congress of Chemical Engineering, Glasgow, UK, 10–14 July 2005.
17. Carlsson, M. Carbon Formation in Steam Reforming and Effect of Potassium Promotion. Johnson Matthey
Technol. Rev. 2015, 59, 313–318. [CrossRef]
18. Heywood, J.B. Internal Combustion Engine Fundamentals; McGraw-Hill Book Company: New York, NY, USA, 1988.
19. McGrattan, K. Fire Dynamics Simulator (Version 5)—Technical Reference Guide Volume 1: Mathematical Model;
NIST Special Publication 1018; National Institute of Standards and Technology U.S. Department of
Commerce: Gaithersburg, MD, USA, 2010.
20. McGrattan, K.; Forney, G.P. Fire Dynamics Simulator (Version 5)—User’s Guide; NIST Special Publication 1019;
National Institute of Standards and Technology U.S. Department of Commerce: Gaithersburg, MD, USA, 2010.
21. McGrattan, K. Numerical Simulation of the Caldecott Tunnel Fire, April 1982; NISTIR 7231; National Institute of
Standards and Technology U.S. Department of Commerce: Gaithersburg, MD, USA, 2005.
22. Hu, L. A review of physics and correlations of pool fire behavior in wind and future challenges. Fire Saf. J.
2017, 91, 41–55. [CrossRef]
23. Hamins, A.; Kashiwagi, T.; Bruch, R.R.; Toten, G.E. Fire Resistance of Industrial Fluids, ASTM STP 1284;
Toten, G.E., Reichel, J.E., Eds.; American Society for Testing and Materials: Philadelphia, PA, USA, 1995.
24. Thermo-Physical Properties of Ethanol. Available online: https://www.thermalfluidscentral.org/
encyclopedia/index.php/Thermophysical_Properties:_Ethanol (accessed on 8 June 2018).
25. De-Souza, Z.G.M.; Moraes, F.F. Parametric Study of Hydrogen Production from Ethanol Steam Reforming in
Membrane Micro-reactor. Braz. J. Chem. Eng. 2013, 30, 355–367. [CrossRef]
26. Neufeld, P.D.; Jansen, A.R.; Aziz, R.A. Empirical equations to calculate 16 of the transport collision integrals
Ω(l,s)* for the Lennard-Jones (1216–) potential. J. Chem. Phys. 1972, 57, 1100–1102. [CrossRef]
27. Perry, R.H.; Green, D.W.; Maloney, J.O. Perry’s Chemical Engineers’ Handbook, 7th ed.; McGraw Hill: New York,
NY, USA, 1997.
28. Kubota Metal Corporation. Fahramet Division, Alloy Data Sheet, Heat Resistant Alloy HK40; Kubota Metal
Corporation: Orillia, ON, Canada, April 1991.
29. Steel Founders’ Society of America. Steel Casting Handbook; Supplement 9, High Alloy Data Sheets Heat
Series; Steel Founders’ Society of America: Crystal Lake, IL, USA, 2004.
30. Van Echo, J.A.; Roach, D.B.; Hall, A.M. Short Time Tensile and Long-Time Creep Rupture Properties of
the HK-40 Alloy and Type 310 Stainless Steel at Temperature to 2000 ◦ C. ASME J. Basic Eng. Trans. 1967,
89, 465–478. [CrossRef]
© 2018 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

CFD Simulation of Ethanol Steam Reforming System For Hydrogen Production

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

CFD Simulation of Ethanol Steam Reforming System For Hydrogen Production

Transféré par

Droits d'auteur :

Formats disponibles

chemengineering

ChemEngineering 2018, 2, 34; doi:10.3390/chemengineering2030034 www.mdpi.com/journal/chemengineering

1.1. Mechanism of Steam Ethanol Reforming Process

• Ethanol decomposition to methane:

• Ethanol decomposition to acetone:

2C2 H5 OH → CH3 COCH3 + CO + 3H2

• Water-gas shift reaction:

1.2. Ethanol Burner Steam Reformer

2.2. Governing Equations of FDS Software

2.2.1. Mass and Species Transport

2.2.2. Momentum Transport

Following the analysis of Smagorinsky, the viscosity µ is modeled:

2.2.4. Energy Transport

ChemEngineering 2018, 2, x FOR PEER REVIEW 7 of 24

2.2.5. Equation of State

Figure 3. The physics of liquid fuel burning [22].

ChemEngineering 2018, 2, x FOR PEER REVIEW 9 of 24

2.3.3. Energy Transport—Reformer Bed

The volumetric heat source due to the reaction is calculated:

∇·(−k steel ∇ Tsr ) = 0 (21)

n·(−k sr ∇ Tsr ) = 0 (22)

The thermal properties of the reformer are presented at Table 1 [9].

Table 1. Thermal properties of the reformer [10].

Material Property Value

2.3.4. Mass Transport—Reformer Bed

2.3.5. Calculation of the Binary Diffusion Coefficients—Reformer Bed

1.06036 0.19300 1.03587 1.76474

Table 2. Properties of the gaseous species [27].

Lennard-Jones Collision Diameter σi (Angstroms)

The diffusion coefficients as a function of the temperature are shown in Figure 5.

Table 3. Thermo-physical and thermomechanical

The maximal temperature at t = 77.5 ss approaches

Figure 7. Velocity field (m/s) of the flue gases at t = 77.5 s.

ChemEngineering 2018, 2, x FOR PEER REVIEW 14 of 24

Figure 8. Radiation heat flux produced by the ethanol burner.

Figure 9. 3D plot of the reformer temperature field.

Figure 9. 3D plot of the reformer temperature field.

ChemEngineering 2018, 2, x FOR PEER REVIEW 15 of 24

ChemEngineering 2018, 2, x FOR PEER REVIEW 16 of 24

Figure 12. Mass fractions of the H2 along reformer axis.

Figure 13. 3D plot of the reformer Von-Mises stress field.

Figure 13. 3D plot of the reformer Von-Mises stress field.

Figure 14. 3D plot of the reformer temperature field.

Figure 16. Mass fractions of the H2 along the reformer axis.

Figure 17. Temperature distribution along the z axis of the reformer.

Figure 18. 3D plot of the reformer Von-Mises stress field.

Funding: This research received no external funding.

Vous aimerez peut-être aussi