Finite Element Analysis research report

University of Alberta
Selective Catalytic Reduction in a Fixed-Bed Reactor
by
Anirban Chakraboty (1199390)
Ryan Lee Robles (1100884)
A preliminary report submitted to Dr. Joseph Mmbaga
in partial fulfillment of the requirements for the completion of
CHE 610
Department of Chemical and Materials Engineering
December 10, 2010
Edmonton, Alberta

i
Abstract
The practicality of using COMSOL modelling to represent an experimentally validated
process has been investigated. The selective catalytic reduction of nitric oxide by ammonia in a
simple, cylindrical fixed-bed catalytic reactor has been modelled using simple two-dimensional
geometry. Equations of mass and momentum transport have been coupled along with the
Brinkman equation to account for porosity in the system. Available kinetic data was
implemented and the overall conversion percentage was calculated from concentration results
obtained in the program.
The results obtained for the provided set of conditions were within 5% error of the
experimental data. It was concluded the model was fairly representative of the system and a
parametric study was conducted to examine the effect of: reactant concentrations, water and
porosity.
It was found a nominal temperature range and reactant ratio of unity will provide optimal
conversion percentages. As well, the effect of water on the conversion is significant only below
a certain temperature. Finally, there were no apparent effects of porosity and implicitly,
permeability and porosity on the reaction.
A post-discussion was also included to comment on related works and future directions with
COMSOL modelling.
Ultimately, COMSOL was found to be a relatively accurate and valid initial point for
investigations concerning selective catalytic reactions.

ii
Table of Contents
Abstract i
Table of Contents ii
List of Figures and Tables iii
I. INTRODUCTION 1
II. PHYSICAL MODEL 2
Geometry Definition 2
Chemical Equations 3
III. GOVERNING EQUATIONS AND BOUNDARY CONDITIONS 4
Mass Transport 4
Momentum Transport 4
Brinkman Equation 5
IV. PRECALCULATIONS 5
Properties of Air 5
Space Velocity 6
Permeability and Pressure Drop 7
V. RESULTS 8
COMSOL Multiphysics 8
Calculation of Conversion Percentage 9
Validation 10
VI. PARAMETRIC STUDY 11
Effect Reactant Ratio 11
Effect of Water 12
Effect of Porosity 14
CONCLUSION 16
VII. SUPPLEMENTARY DISCUSSION 17
NOMENCLATURE 27
REFERENCES 28
APPENDIX A: Sample Calculations A1
APPENDIX B: MATLAB Codes B1
APPENDIX C: Sample Tutorial of COMSOL Multiphysics C1

iii
List of Figures
Figure 1. Three-dimensional representation of the catalytic reactor 2
Figure 2. Two-dimensional representation of the catalytic reactor 2
Figure 3. Air density and dynamic viscosity as temperature changes 6
Figure 4. One-dimensional concentration profile of reactants 8
Figure 5. Two-dimensional concentration profile of reactants 9
Figure 6. Comparison of experimental and simulation results for space
velocity of 100 000 h-1
10
Figure 7. Conversion of NO at several NH3/NO ratios 11
Figure 8. Conversion of NO under dry and wet conditions 12
Figure 9. Conversion of NO under different porosities 14
Figure 6.1. Reduction of NO under different NH3 Conditions 18
Figure 6.2. Reduction of NO with Ag/Alumina Catalysts 19
Figure 6.3. Inner wall temperature change with the fluidity enthalpy at
different pressures in the supercritical regions
25
Figure 6.4. Temperature Distribution inside the boiler 26
List of Tables
Table 1. Kinetic parameters for a wet, packed bed reactor 3
Table 2. Corresponding permeability and pressure drops for porosity, 0.5 7
Table 3. Percent improvement of NO conversion depending on NH3: NO 12
Table 4. Kinetic parameters for a dry, packed bed reactor 12
Table 5. Corresponding permeability and pressure drops for porosity, 0.5 15
Table 6.1 Various conversion parameters (extracted from Fino, 2009) 22

1
I. INTRODUCTION
Combustion and its methods of control are major factors influencing the livelihood of
society. According to a study by the U.S. Department of Energy (1996), about 85% of the
energy used in the country originated from combustion sources. In addition, the transportation
system relies almost entirely on combustion- aircrafts are wholly powered by on-board fuel
burning and most public transit systems are diesel-engine motorized. Commonplace appliances
such as lawn mowers and chain saws are also found to be gasoline-powered. The cement
manufacturing sector is also heavily dependent of the heat energy produced by combustion.
Rotary kilns, commonly used to produce cement clinkers are highly inefficient devices
employing this process (Trubasev, 2005). On a larger scale, the metals-refining industry relies
heavily on combustion, employing furnaces for production of the raw product and heat-treating
and annealing ovens downstream in converting to the finished good.
However, there is a downside associated with combustion- environmental pollution. The
major pollutant emissions produced are unburned and partially burned hydrocarbons, such as
aldehydes; nitrogen oxides, NOx; carbon monoxide; sulfur oxides; and greenhouse gases, in
particular CO2. The drive to become environmentally conscious has led to a continually
changing atmosphere in terms of more stringent emission regulations and standards. For
instance, amendments to the Clean Air Act in 1963 on several occasions have imposed stricter
standards in response to the realization the photochemical smog in the Los Angeles basin was
primarily a result of automobile emissions of unburned hydrocarbons and NOx (Haagen-Smit,
1952). Engines and power plants have commonly incorporated selective catalytic reduction
since the 1970s for the reduction of NOx (Fang, 2003). Depending on the specific requirement,
several other SCRs have also been utilized: plasma type catalyst (Okubo et al, 2007), lean NOx
trap (LNT) (MECA, 2000), and urea-SCR (Tronconi et al, 2007).
For the present study, the SCR of NO by NH3 presented by Chae et al (2000) is
examined. Chae‟s paper investigates SCR using the widely known Vanadium catalysts
(V2O5/TiO2). The significance of the project is in the simulative validation of the understandings
of the behaviour and the conversion of the NOx (namely NO) in the reactors with respect to the
NH3 which acts as a highly efficient reducing agent. The fundamentals of the reactor outlined in
the paper reveals the conditions used in the experiment and the conversion profile obtained at
different temperatures.

2
The main purpose of this study is to verify the feasibility of using COMSOL modelling
by validating results against experimental data. Depending on the accuracy of the simulation, a
parametric study will be conducted to both determine the effects of several operating conditions
on the reaction and confirm other experimental conjectures.
II. PHYSICAL MODEL
Geometry Definition
Figure 1 shows the three-dimensional cylindrical representation of a tubular fixed-bed
catalytic reactor.
Figure 1. Three-dimensional representation of the catalytic reactor
An axial two-dimensional model of the fixed-bed reactor wherein the catalyst bed
behaves as the porous media was assumed. In other for this model to be valid, a reasonable
assumption of plug flow was made. A vertical cross-section of the reactor is shown in Figure 2.
Figure 2. Two-dimensional representation of the catalytic reactor
H =7.2 cm
INLET
OUTLET
W =10 cm
7.2 cm
10 cm

3
Chemical Equations
Several reactions have been proposed to represent the SCR of NO by NH3 (Devadas, 2006)
(1)
(2)
(3)
Equations (1), (2) and (3) are referred to as the standard, fast and slow SCRs, respectively.
The above reaction represents the maximum conversion of NO using a V2O5-WO3/TiO2
catalyst and incorporates NH3 oxidation. The reaction kinetics can be described the Eley-Rideal
mechanism (Frans et al, 1987) with values of constants as provided in Table 1.
(4)
(5)
The reaction rate constants and adsorption equilibrium constant are calculated using the
Arrhenius Equation
(6)
(7)
(8)
Table 1. Kinetic parameters for a wet, packed bed reactor
Kinetic Parameter Value
Ea,NO 12.7
Ea,NH3 [kcal/mol] 57.6
HNH3 [kcal/mol] 22.2
3.04 x 106
9.98 x 108
69.1
Temperature Range, T [C] 200-500

4
III. GOVERNING EQUATIONS AND BOUNDARY CONDITIONS
Coupled with the reaction mechanics within the reactor, the two main equations
governing the selective catalytic reactor system are the mass and momentum transport equations
(Comsol, 2010). The heat of reaction is not considered due to the small initial concentrations of
the two reactants.
Mass Transport
The mass transfer in the reactor domain is given by
(9)
Since the mass equation is assumed at steady state, the stationary simplification is
(10)
Where denotes the diffusion coefficient and is the reaction term.
For the boundary conditions, the initial concentrations of NO and NH3 are specified
(11)
At the outlet, the mass flow through the boundary is dominated by convection, thus assuming
any mass flux due to diffusion across the boundary is negligible
(12)
Therefore,
(13)
Analogous to the momentum transport, at the walls of the reactor, assume an insulated
boundary condition. (I.e. no mass is transported across the boundaries)
(14)
Momentum Transport
The incompressible Navier-Stokes equation is used to describe the momentum transport.
Equation (15) governs the flow of a Newtonian fluid in the laminar flow regime.
(15)
(16)

5
In this particular model, steady-state is assumed which will cancel the first term in both
equations above. If the assumptions of no convection and negligible internal force (F) are
applied, the equation can further be simplified as
(17)
Where denotes the dynamic viscosity, u the velocity, and the density of the fluid.
At the inlet, a velocity vector normal to the boundary is specified
(18)
At the outlet boundary, a pressure is specified
(19)
A no-slip boundary condition is applied on the reactor walls
(20)
Brinkman Equation
The momentum equation (17) is slightly altered to account for the porosity of the reactor.
(21)
Where denotes the permeability and the porosity.
IV. PRECALCULATIONS
Properties of Air
Exhaust gases from both refineries and automobiles can be appropriately represented by
air. However, the intrinsic properties of air of interest in this case are largely temperature-
dependent. While the material browser available in COMSOL will allow for these properties to
be automatically calculated, the permeability and pressure drop will be calculated based on the
values of density and dynamic viscosity. Thus, an approximation of these properties is obtained
by fitting the density vs. temperature and dynamic viscosity vs. temperature data available in
Perry‟s Handbook (2008) to a power series. Figure 3 is a plot of the two properties with respect
to temperature. The fitted equations are respectively indicated on the plot.

6
Figure 3. Air density and dynamic viscosity as temperature changes
Space Velocity
Chae (2000) provides data for a specific space velocity. Since the simulated model is
two-dimensional, a conversion from three-dimensions must be derived. Appendix A contains a
sample calculation of converting the space velocity
The space velocity of the reactor is given by
(22)
Where denotes the reactor volume.
The value converted into its two-dimensional equivalent is
(23)
Where denotes the surface velocity at the inlet of the reactor.
0.00E+00
1.00E-05
2.00E-05
3.00E-05
4.00E-05
5.00E-05
6.00E-05
7.00E-05
8.00E-05
9.00E-05
1.00E-04
0
0.5
1
1.5
2
2.5
3
3.5
4
0 200 400 600 800 1000
DynamicViscosity,Ns/m2
Density,kg/m3
Temperature, K
Density
Dynamic Viscosity
r = 377.5T-1.014 m = 3.671-10T1.853

7
Permeability and Pressure Drop
A significant piece of information missing in the paper pertains to the permeability,
which in combination with porosity, will affect the pressure drop in the system. As a valid
approximation, Incropera et al (2007) provides a method for estimating these parameters.
The Ergun equation is used to calculate the pressure drop per unit height and is given as
(24)
Where denotes the particle diameter.
The pressure drop per unit height calculated from equation (24) is then substituted into
the Darcy equation to obtain the permeability.
(25)
Where denotes the cross-sectional area and is the volumetric flow rate.
MATLAB was used to calculate the parameter values and the codes are available in
Appendix B. Table 2 summarizes the pressure drop and permeability values used in the initial
simulation.
Table 2. Corresponding permeability and pressure drops for porosity, 0.5
Temperature
C
Porosity = 0.5
P, Pa K, m2
200 28884 3.31 x 10-10
250 33011 3.49 x 10-10
300 37638 3.63 x 10-10
350 42736 3.73 x 10-10
400 48282 3.81 x 10-10
450 54260 3.87 x 10-10
500 60656 3.92 x 10-10

8
V. RESULTS
COMSOL Multiphysics
By defining all the equations and parameters outlined in the report, a successful 2D
representation of the catalytic reactor was built using the COMSOL Multiphysics software. The
readily available Transport of Diluted Species and Brinkman Equations were coupled and both
one- and two-dimensional profiles of the NO concentration were obtained. Appendix C contains
a sample tutorial for building the simulation model. Plots were generated using MATLAB,
EXCEL and the export feature available in COMSOL.
Figure 4 is the concentration profile of both reactants throughout the reactor under wet,
packed bed conditions for  . Figure 5 is the two-dimensional concentration profile of
NO under the same conditions. The top of the reactor is considered reference (origin) point.
Figure 4. One-dimensional concentration profile of reactants
0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
0 0.02 0.04 0.06 0.08
Concentration,mol/m3
Reactor Height, m
NH3
NO

9
Figure 5. Two-dimensional concentration profile of reactants
Figure 5 shows the conversion of NO occurs immediately at the reactor inlet and a
majority of the initial reactants are converted well within the first 0.01m.
As seen in Figure 4, the consumption of NO in the reactor is slightly higher than that of
the NH3. This will result in unconverted ammonia (NH3 slip). The amount of unconverted
ammonia unconverted must be monitored carefully and with high reliability in order to meet both
process and environmental regulations. The NH3 slip will be highly dependent on process
conditions and will be investigated.
Calculation of Conversion Percentage
(26)

10
Validation
The results obtained by provided Chae (2000) for a space velocity of 100 000 h-1
were
used as the validation basis. The data was extracted using the WinDig software. Appendix A
contains a sample calculation of the percent conversion of NO. Appendix B contains the
MATLAB code for the overall calculations. Figure 6 displays a comparison of the results
obtained using COMSOL with those provided in the paper.
Figure 6. Comparison of experimental and simulation results for space velocity of 100 000 h-1
As seen in the figure, the overall trend in conversion percentage is captured adequately
with slight discrepancies occurring in the intermediate dynamics. Ultimately, the COMSOL
model is fairly representative of the experimental data, consistently within 5% agreement
throughout the complete temperature range. Any differences may be attributed to parameter
estimations, as was the case with the density and dynamic viscosity or general misinformation as
the initial paper (Chae 2000) did not include detailed information about the physical model used.
Considering the effectiveness of the COMSOL model, a parametric study was conducted with
confidence to investigate the effects of different operating conditions.
0
10
20
30
40
50
60
70
80
90
100
200 250 300 350 400 450 500
PercentConversionofNO
Temperature, K
Chae (2000) Model
COMSOL Multiphysics

11
VI. PARAMETRIC STUDY
The effects of altering several parameters on the NO conversion were investigated to
verify experimental results and intuition. For instance, increasing the initial NH3 concentration
should logically result in higher conversion of NO per pass through the reactor, but also results
in more unused NH3. The COMSOL model was used to verify this conjecture, as well as
examine the effects of changing conditions and altering the porosity.
Effect of Reactant Ratio
As seen in Figure 4 the reducing agent is consumed in tandem with the nitric oxide, but at a
slightly slower rate. This unused amount of NH3, known as NH3 slip, must be minimized for
several reasons. However, the amount of reducing agent must be adequate enough to allow for
full conversion of the nitrogen oxides. In order to investigate this property, several reactant
ratios (NH3: NO) were simulated. Figure 7 displays the conversion of NO for a reactant ratio of
0.85, 1 and 1.15 with respect to operation temperatures.
Figure 7. Conversion of NO at several NH3/NO ratios
0
10
20
30
40
50
60
70
80
90
100
200 250 300 350 400 450 500
PercentConversionNO
Temperature, K
NH3:NO = 1
NH3:NO = 0.85
NH3:NO = 1.15

12
It is apparent from the Figure 7 the change in ratio directly affects the NO conversion.
Notice, however, there is no discernible difference in conversion for temperatures up to 300C.
At temperatures greater than 300C, discrepancies in conversion percentages become more
apparent. Using the smaller ratio as the base case, Table 3 summarizes the conversion
improvement with increased ammonia concentrations for temperatures (350 - 500C).
Table 3. Percent improvement of NO conversion depending on NH3: NO
Temperature NH3: NO = 0.85  1 NH3: NO = 1  1.15
350 14.2 1.89
400 14.5 3.01
450 14.0 5.32
500 13.4 8.35
According to Ciardelli (2007) for SCR, the optimum working temperature range is from
300- 400C with a NH3/NO ratio of unity. This is prominently supported by the figure as
noticeable differences begin to manifest at the lower bound temperature. At these temperature,
increasing the ratio from 0.85 to unity results in an average improvement of about 14%. While
increasing the ratio to above par, only results in an average of increase of about 4.65%.
Therefore, within this range, to meet practical conversion specifications while minimizing the
NH3 slip, it is obvious the NH3/NO ratio must be unity. However, for temperatures lower than
300C, it may be more economical and practical to operate at NH3/NO ratio should be less than
unity.
Effect of Water
The exhaust gases either from automobiles or plants constituting combustion processes
will almost always contain moisture. As a result, it may be more practical to simulate selective
catalytic reduction under wet conditions.
Topsoe (1996) suggests water has a significant effect on NH3 SCR due to the reactive
competition between H2O and NH3 for adsorption on the catalytic sites. In order to investigate
this phenomenon, the conversion percentages of NO simulated using dry conditions. Table 3

13
indicates the alternate conditions of a dry packed-bed reactor as provided by Chae (2000).
Figure 8 plots the results obtained from COMSOL using the dry and wet conditions.
Table 4. Kinetic parameters for a dry, packed bed reactor
Kinetic Parameters Packed-Bed Reactor Dry Conditions
Ea,NO 11.5
Ea,NH3 [kcal/mol] 42.8
HNH3 [kcal/mol] 21.5
2.79 x 106
6.38 x 105
59.6
Figure 8. Conversion of NO under dry and wet conditions
According to Topsoe (1996), the effect of water is considerable at temperatures below
400C. This is apparent in Figure 8, wherein an average decrease in conversion of 26% is
observed for the temperature range (200-300C). At the conversion intersect occurring at 350C
the wet conditions become more efficient, registering an average improvement of 6.2%. The
0
10
20
30
40
50
60
70
80
90
100
200 250 300 350 400 450 500
PercentConversionofNO
Temperature, K
Dry Conditions
Wet Conditions

14
hindrance introduced by water weakens at higher temperatures as the conditions now enable the
system to overcome the higher activation energies. Thus, at higher operating temperatures, it
may be beneficial to operate under wet conditions.
Effect of Porosity
Altering the porosity of the reactor will result in change in both permeability and pressure
drop across the system. To investigate the effects, conversions from using a porosity of 0.7 were
compared with the default results. Table 4 contains the permeability and pressure drops
calculated using the Darcy and Ergun equations for the new porosity. Figure 9 summarizes the
conversions achieved using different reactor porosities.
Figure 9. Conversion of NO under different porosities
0
10
20
30
40
50
60
70
80
90
100
200 250 300 350 400 450 500
PercentConversionNO
Temperature, K
Porosity = 0.5
Porosity = 0.7

15
Table 5. Corresponding permeability and pressure drops for porosity, 0.5
Temperature
C
Porosity = 0.5
P, Pa K, m2
200 4307 2.22 x 10-9
250 47987 2.40 x 10-9
300 5364 2.55 x 10-9
350 5998 2.66 x 10-9
400 6696 2.75 x 10-9
450 7455 2.82 x 10-9
500 8272 2.87 x 10-9
Regardless of the varied changes in both parameters, there was no apparent effect on the
NO conversion. As a result, it was concluded the effect of porosity and thus, permeability and
pressure drop on the system appear to be negligible.

16
CONCLUSION
The impact of exhaust emissions on the environment has always garnered international
attention. Consistent amendments to the Clean Air Act and other environmental restrictions are
constantly becoming more stringent in an effort to curb degradation of the planet by pollution.
Combustion processes, for instance, result in harmful emissions such as carbon dioxide (CO2),
sulphur dioxide (SO2), nitric oxides (NOx) and dust (Siemens, 2007). Thus, many measures are
implemented in automobiles and power plants to meet emission standards.
For many de-nitrification processes, such as flue gas treatment, selective catalytic
reduction with ammonia is used in conjunction with front-end primary measures to convert
nitrogen oxides to nitrogen and water at high temperatures.
A COMSOL model was built to simulate a fixed-bed reactor employing this process
using operating conditions provided by Chae et al (2000). The simulation model will be
examined for its applicability to practical utilization. The simulation results were validated and a
parametric study was conducted to investigate other experimental results.
The results from COMSOL matched within 5% error with those presented in the primary
paper suggesting COMSOL modelling may be a viable preliminary standard for results.
From the parametric study, it was concluded:
1. For NH3/NO ratios of (0.85, 1, 1.15), there are no differences in NO conversion for
temperatures below 300C.
2. The operating temperature range of 300-400C and reactant ratio of unity will allow
for optimal conversion of NO with minimal NH3 slip as determined experimentally.
3. The hindrance effect of moisture on the conversion is more significant at
temperatures below 400C. However, more realistic results may be obtained
considering wet conditions.
4. There is no apparent effect of porosity on the conversion percentage. Consequently,
permeability and pressure drop play a minor role in SCR.
Ultimately, the COMSOL model can be effectively applied as a preliminary study into
the selective catalytic reduction of nitric oxides by ammonia; and potentially a viable gateway
into other SCR processes.

17
VII. SUPPLEMENTARY DISCUSSION
The discussion has taken into account several key papers in the domain of SCR and has
performed its relevant analysis on the matter which is documented in the following verses.
Intensive research and analysis has been carried out in the related areas of technology of
selective catalytic reduction. A report on the “Kinetics of the biofuels-assisted SCR of NOx over
Ag/alumina-coated Microchannels” (Murzin, 2009) has claimed when hexadecane was used at a
temperature of 350C and above the reduction of NO with hexadecane was almost four times
higher than when octane was used. It was also claimed that the geometry of the reactant
molecules have an effect on the reactivity respectively. Two mains reactions have been studied
(6.1)
(6.2)
Oxidation Effect
Earlier studies done in the area have reported the presence of Ag/Alumina (Al2O3)
catalyst have been more active in the presence of oxygen (Miyadera et al, 1993). At
temperatures below 300C, the reduction of NO as reported gained a certain cardinal point
(Murzin et al, 2009). Also in the case of higher temperatures, the point of inflexion has been
found at 6% more than the initial concentration of oxygen. Comprehensively, it is observed the
change of temperature gives a different experimental result in the studies as performed. The
following graphical representation seen in Figure 10 gives a vivid description of the results as
mentioned. Figure 11 is the COMSOL model in which the difference between the two figures is
exhibited. The SCR the concentration of the NO and the NH3 was transformed to 1.0 and 2.5
respectively.

18
Figure 6.1. Reduction of NO under different NH3 Conditions

19
Figure 6.2. Reduction of NO with Ag/Alumina Catalysts

20
The second figure has been obtained as the result of the changes in the concentration of
the both the NO and the NH3 (reductant), where visualization is of the magnitude of the rate of
change in the concentration profile and its overall distribution. The intrinsic reason prompting
the showcase of the COMSOL version of the altered concentration of the NO and the NH3 as a
reducing agent is the known initial concentration of the two gases cannot be same at the
beginning and the end of the reaction. By means of usage of different catalyst, in compatibility
with the substrates there should be different kinetics of the reaction and the changing
concentration of the component gases in the phase.
Streamlined Comprehension
This project and its subsequent report have undergone an amalgamated study of the other
papers in terms of its conclusions and have strived to reach or predict its future scopes. On close
analysis it is found the approaches of the paper (Murzin et al, 2009) on SCR has laid multiple
emphasis on the temperature and the relative size of the hexadecane molecule. Creating a
synergy between the presence of the catalyst (bio fuel) and the Ag/Alumina, some papers such as
in initial discussion have been able to build up a multilateral way to approach and investigate the
individual contributions of each of the parameters. However, within the limited scope and
magnitude of this report, more extensive parameter studies have not been included, but are
suggested for future work. Positive outcome being, it can be comprehended logically COMSOL
4.0a with its latest GUI and additional features possesses the potential of successful evaluation of
the multitude of details incorporated into any SCR research.
Another paper (Ribeiro et al, 2006) investigating the role of ammonia as the principle
component in the reaction phase of SCR, has proposed ammonia as the hydrogen carrier. It has
been claimed in this model NO and NO2 are first released in the gas phase and correspondingly
reduced by the platinum surface. Eventually, this indirectly proposes the investigation of the
catalytic coefficients of the Pt and setting them in the COMSOL modeling to observe the
outcome. With having a moderate number of models within the elements of the Pt group as for
example palladium, a pattern for the usages of the catalyst in the same family and their
respective contributions can be observed. In the due course of time, it will help to save time and
cost of doing more elaborate experiments and researchers can benefit from this pattern created,

21
which can help in the predicting the reaction kinetics and outcome of experiments in SCR with
Pt group catalyst, without having done real life experiments.
In the report, it can be found it establishes itself or has the potential to form a platform for
SCR using Vanadium based catalysts. Proceeding in the same way as mentioned, it can take
leverage creating patterns that can help in the prediction of further experiments related to SCR.
Overall, it squeezed the potential of the COMSOL software to create a model in the 2D platform
with a distinct profile distribution.
In reference to the project, it can be observed in the pre-calculation part of the report
impetus has been given on the incorporation of the material properties of the physical quantities
used in the reaction, from the material browser. The discussion as the analytical part of the
report argues and advocates the more extensive application of the material properties of the
reactants and put them into play and studies the respective outcomes. In conjunction with other
reports, researchers strive to depict the utilization and role of lanthanum ferrite crystals for the
reduction of NO by means of hydrogen (Fino et al, 2009); it has tried to eliminate the role of
ammonia in the reaction phase.
The most interesting and best part of this study is the researchers have stressed the sole
usage of H2 on the catalyst contribution of perovskite (Fino et al, 2009) catalysts. It also states in
accordance to the report by Topsoe (1996) and Ciardelli (2007) the optimal working condition
will be between 300 and 400C, which in correspondence agrees with the studies by Murzin
(2009). Crucially, one very pivotal part of the studies by Fino et al was an effort to derive out a
correlation between the catalyst structure and the activity.
It suffices to mention, not all the papers in this domain have carried out an intensive
investigation of all the parameters; there is still a caveat in the exact role of the catalyst bricks
which have not been completely searched (Fino et al, 2009). However, the following table can
deliver some important data about the half conversion, peak temperatures and maximum
conversion rates at different operating conditions.

22
Table 6.1 Various conversion parameters (extracted from Fino, 2009)
While discussing the porosity of the reactor and its effect on the total reaction phase, it
has been depicted alteration of the reactor porosity will lead to the change in the permeability
and pressure drop. Also, the graphical representation shows the percentage conversion of the
NO goes up from 30% to a peak of over 90%; a downturn occurs at a temperature of 350C.
However, the studies by Murzin et al (2009) claim in their substantiation at the temperature
range above 350C the hexadecane molecule is better oxidized and the effect was impressive on
the overall SCR. Interestingly, results from COMSOL perceived a 5% error in comparison with
the primary literature; the paper by Murzin obtained high standard errors. Although the
researchers claim their data has been in sync with the model created, including the standard
errors, it is conclusive from our COMSOL model with further incorporation of data and
operating condition alterations, there is reasonable optimism with COMSOL and its model
predictions. Furthermore, the report strongly advocates a sensitivity analysis of the coefficients
of the partial differential equations in an aim to create better synergy between the actual results
and the models. A sensitivity analysis can help eliminate the errors as much as possible and can
optimally help to reduce the gap between the error and the optimal result to be obtained.

23
Data and interpretation about Heat transfers in Tubes
Some interesting data prompted inclusion and investigation into a paper on the heat
transfer and role of the frictional features in a rifled tube. The rifled tube is the tube with inner
ridges. The paper discusses a project that investigates and then discuses the flow and distribution
of the heat in the walls of the tube, having its attributes to the rifling (vertically upward) in the
supercritical and the subcritical temperature zones. The pressure range of the machine was from
12kPa to 30MPa. There was as perceived a wide range of pressure which helped in gauging the
exact relevance of the change in the unit of the temperature and the pressure in the flowing liquid
inside which shows the pattern of the heat flow and the distribution of the pattern that can be
compared with a simulated model. The results of the experiments showed the heat flux was
ranging from 130 to 720kW/ m2. The pivots that were measured were the temperature
distribution and the pressure drops. Normal pressure drops, enhanced and the inner wall heat
flux on the heat transfer characteristics were collected and analysed.
The experimental data when analysed suggested the researchers Chan et al designed and
developed a high pressure water experimental system in China to characterise the heat transfer
coefficients and the resistance characteristics in the smooth and the rifle; both in the supercritical
and the subcritical temperature, respectively.
What is important about this study?
As in accordance to the primary paper as for example, Swenson et al, studied and
investigated the effects of the nucleate boiling against the boiling. Results have exhibited the
nucleate boiling can be substantiated to elevated vapour quantities in the rifled tube, rather than
the smooth tube. As reported another group by Watson et al also studied about the rotational
flow in the tubes and they found out that rotational flow in the ribbed tube can significantly
increase the CHF and the critical vapour quality respectively.
Interestingly, it was found the rifled tube or the ribbed tube having different geometries can
affect the „heat transfer enhancements‟ properties. Numerically and logically it depicts the
shape, or rather the geometries of the tubes, harbouring the fluid flow is proportional to the heat
transfer mechanism.
Results: The inner wall heat flux of the heat exchanger was 380kW/m2
and the mass flux
was 520kg/m2
s. Moderate vapour quantity was at 335C and the saturation of the fluid at 324C.

24
The temperature difference as recorded, between the wall and the bulk fluid was 360C and the
fluid saturation temperature at 347.4C. The temperature difference between the wall and the
temperature of the core was 12C. In the interpretation of the temperature difference, the rifled
or the ribbed tube worked quite significantly as a heat exchanger. The diagram which has been
given over here gains special attention because of the fact the diagram depicts the change or the
cardinal point of the graph. It can be seen there was an arrest of the heat transfer mechanism
with the increase in the pressure subsequently.
How was the Ht in the near critical region?
The occurrence of the DNB (Departure from Nucleated Boiling) was avoided and can be
termed as a great achievement in case of the heat transfer mechanism in the total machine design.
The calibration of the COMSOL model with the help of the heat transfer module both in case of
the Ht and the heat transfer in the solids and the liquids also would usher in a histogram
depiction of the total pattern of the heat flow distribution in the total flow surface area. It would
actually help to see and observe the trends and the pattern of the heat distribution in the ribbed
which is actually the ridged tubes. To capture and then later on while in doing some other
studies remains the aim and the utmost focus of the project in COMSOL now and also in the due
course of time will be to predict the results of different Heat Transfer experiments in solids and
in fluids without having a large amount of data input and subsequent validations. The aim of the
future projects involving the contribution of the COMSOL project model should optimally be
finding the correlations also between the different parameters as the conversion ratio, the
temperature, the porous structure of the media and the porosity at which the media is operating.
As for example the project uses two separate porosities of the media. This can be aptly called a
multiphysics problem along with a sense of hetero catalytic activity as seen in the presentation
on Monolith Reactors. By the aid of the COMSOL, it can be found in different graphical
representations, in the industrial sectors and mainly in the engines of the aviation sector many
potential dangers can be avoided. The paper (D yang et al, 2010) cites some interesting facts
about the fluid temperature at the supercritical pressure and the effect of this on the
thermodynamic properties of the fluid respectively. When included in COMSOL, the results
obtained it can have immense significance on the studies concerning transport phenomenon.

25
Figure 6.3. Inner wall temperature change with the fluidity enthalpy at different pressures in the
supercritical regions (source: D yang et al, 2010)
Another paper by researchers Alsairafi et al has studied the relation between the
conversion and the temperature in the reactor, which is a boiler in this case. It also reports the
oxygen concentration and the effect of the oxygen on the conversion rate of the NOx. The paper
takes into consideration Equations 15 and 16, modelled into the FLUENT software.

26
Figure 6.4. Temperature Distribution inside the boiler (source: Alsairafi et al, 2010)
However, as the model has not been found in chromatic dimensions, it is certain that this
is going to be an issue at least in this report regarding the observation and the analysis of the
model. The observation and the comprehensive analysis of the observational data depicted that
the maximal Oxygen concentration was at the HRSG (Heat recovery steam generator) inlet
section and the minimal concentration was at the zone near the burner. The reasons cited for
these two events were the concentration of the oxygen was believed to be maximal at the region
of the inlet of the reactor and the minimal was found at the burner where most of the oxygen has
been used up. The distribution of the maximum NO formation was also reported at the burner
sites where the temperature was highest conclusively spearheading towards the fact the NO
formation was dependent on the temperature. This seems a very well-defined point which can be
of extreme relevance to the current project if there is any chance of integrating our model with
different temperature variations.
Finally the report comprehends itself having presented the utmost discussion and analysis
within its eligible scope and limit and trusts itself to have delivered intrinsic ideas about future
horizons in terms of studying transport phenomenon.

27
NOMENCLATURE
A cross-sectional area, m2
dp particle diameter, m
D diameter, m
 heat of adsorption, kcal/mol
Ea Enthalpy, kcal/mol
H height of the reactor, m
k reaction rate constant, 1/s, mol/cm3
/s
K adsorption equilibrium constant, cm3
/mol
P pressure, Pa
Q volumetric flow rate, m3
/s
R equilibrium constant, 1.9872cal/mol/K
SCR selective catalytic reduction
T temperature, K
u velocity, m/s
V reactor volume, m3
W width of the reactor
SV space velocity, m-1
Greek
permeability
porosity
r density, kg/m3
 dynamic viscosity, Ns/m2
Superscripts and subscripts
 standard
NH3 ammonia
NO nitric oxide

28
References
Alsairafi, A.A., A. Khoshhal and M Rahimi, “The CFD Modeling of NOx Emission, HiTAC and
Heat Transfer in an Industrial Boiler,” Numerical Heat Transfer, Part A: Applications, 58,
pp. 295-312 (2005).
Chae, H.J., S.T. Choo, H. Choi, and I. Nam, “Direct Use of Kinetic Parameters for Modeling and
Simulation of a Selective Catalytic Reduction Process,” Ind. Eng. Chem. Res., 39, pp.
1159-1170 (2000).
Ciardelli, C., I. Nova, E. Tronconi, D. Chatterjee, B. Bandl-Konrad, M. Weibel, and B. Krutzsch,
“Reactivity of NO/NO2-NH3 SCR system for diesel exhaust aftertreatment:
Identification of the reaction network as a function of temperature and NO2 feed
content,” Applied Catalytic B: Environmental, 70, pp. 80-90 (2007).
Cumaranatunge, L., S.S. Mulla, A. Yezerets, N.W. Currier, W.N. Delgass and F.H. Ribiero,
“Ammonia is a hydrogen carrier in the regeneration of Pt/BaO/Al2O3 NOx traps with
H2,” Journal of Catalysis, 246, pp. 29-34 (2007).
Comsol Multiphysics 4.0, “Thermal Decomposition” Tutorial, COMSOL AB, (2010).
Devadas, M., O. Kröcher, M. Elsener, A. Wokaun, N. Söger, M. Pfeifer, Y. Demel, and
L. Mussmann, “Influence of NO2 on the selective catalytic reduction of NO with
ammonia over Fe‐ZSM5”, Applied Catalysis B: Environmental, 67, pp. 187–196 (2006).
Fang, H. L., and H.F.M. DaCosta, “Urea thermolysis and NOx reduction with and
without SCR catalysts,” Applied Catalytic B: Environmental, 46, pp. 17-34 (2003).
Frans, J., J.G. Janssen, F.M.G. Van den Kerkhof, H. Bosch, and J.R.H. Ross, “Mechanism of the
Reaction of Nitric Oxide and Oxygen Over Vanadia Catalysts”, J. Phys. Chem, 91, pp.
6633-6638 (1987).
Fino, D., S. Furfori, S. Bensaid, N. Russo, “Towards practical application of lanthanum ferrite
catalysts for NO reduction with H2,” Chemical Engineering Journal, 154, pp. 348-354
(2009).
Haagen-Smit, A.J., “Chemistry and Physiology of Los Angeles Smog,” Industrial Engineering
Chemistry, 44, pp. 1342-1346 (1952).
He, Y., D.B. Brown, S. Lu, M.J. Paratore et al, “Opportunities and Challenges for Blended 2-
Way SCR/DPF Aftertreatment Technologies,” SAE Technical Paper, 2009-01-0274,
(2009).
Murzin, D.Y., J.R. Hernández Carucci, A. Kurman, H. Karhu, K. Arve, K. Eränen, J. Wärnå, and
T. Salmi, “Kinetics of the biofuels-assisted SCR of NOx over Ag/alumina-coated
Microchannels,” Chemical Engineering Journal, 154, pp.34-44 (2009).

29
Incropera, F.P., and D.P. DeWitt, “Fundamentals of Heat and Mass Transfer,” 6th
ed., John
Wiley & Sons, Hoboken, NJ (2007).
Lee, P., “Effects of B20 Fuel and Catalyst Entrance Section Length on the Performance of
UREA SCR in a Light-Duty Diesel Engine.” Unpublished dissertation. Wayne State
University (2009).
Manufacturers of Emission Controls Association (2000). “Catalyst-Based Diesel Particulate
Filters and NOx Adsorbers: A Summary of the Technologies and the Effects of Fuel
Sulfur,” http://www.meca.org/galleries/default-file/cbdpf-noxadwp.pdf
Miyadara, T., “Alumina-supported silver catalysts for the selective reduction of nitric oxide with
propene and oxygen-containing organic compounds,” Applied Catalysis B:
Environmental, 2, pp. 199-205 (1993)
Okubo, M., T. Kuroki, K. Yoshida, and T. Yamamoto, “Simultaneous Reduction of Diesel
Particulate and NOx Using Oxygen-Poor Nonthermal Plasma Application,” Industry
Applications Conference. 42nd IAS Annual Meeting., IEEE, New Orleans, LA,
pp. 1864-1870 (2007).
Perry, R.H., and D.W. Green, “Perry‟s Chemical Engineers‟ Handbook,” 8th
ed., McGraw-Hill,
NY, pp. 2-211-2-216 (2008).
Siemens AG (2007). 76181 Karlsruhe, Germany, “Minimizing Ammonia Consumption in
DeNOx Plants In situ Laser Gas Analyzer LDS 6 Monitors NH3 Slip in Real-time,”
http://www.siemens.com/processanalytics
Stepanek, J., P. Koci, F. Frantisek, M. Marek and M. Kubicek, “Investigation of combined DOC
and NSRC diesel car exhaust catalysts,” Computer and Chemical Engineering, 34,
pp. 744-752 (2009)
Topsoe, N.Y., H. Topsoe, and J.A. Dumesic, “Vanadia/Titania Catalysts for Selective Catalytic
Reduction of Nitric Oxide by Ammonia,” Journal of Catalysis, 163, pp. 409-417 (1996).
Trubasev, P., and P. Besedin, “Criteria for the Thermodynamic Efficiency of Cement Clinker
Production from Natural Raw Material,” Theoretical Foundations of Chemical
Engineering, 39, pp. 628-634 (2005).
Tronconi, E., I. Nova, C. Ciardelli, D. Chatterjee, and M. Weibel, “Redox features in the
catalytic mechanism of the “standard” and “fast” NH3-SCR of NOx over a V-based
catalyst investigated by dynamic methods,” Journal of Catalysis, 245, pp. 1-10 (2007).
U.S. Department of Energy (1996), “1996 Annual Energy Review,” DOE/EIA-0384(96).
http://www.eia.doe.gov/bookshelf.html

Finite Element Analysis research report

More Related Content

What's hot (20)

Viewers also liked (8)

Similar to Finite Element Analysis research report (20)

Finite Element Analysis research report