Temperature Variations in a 15
Tubular Reactor
151
From Elements of Chemical Reaction Engineering, 2006 Pearson Education, Inc.
Fourth Edition, by H. Scott Fogler. All rights reserved.
152 Radial and Axial Temperature Variations in a Tubular Reactor Chap. 15
15.1.1 Definitions
y
z+z, y+y, z+z
FAzz
Molar Molar
Rate of Rate of
flow rate flow rate
generation accumulation
in x out x+ x
C
zyWAx x zyWAx x x rAxyz x y z A
t
Dividing by DxDyDz and taking the limit as Dx, Dy, and Dz go to zero, we
obtain the molar flux balance in rectangular coordinates
1 WAz C
  ( rWAr )  rA A (153)
r r z t
We will now evaluate the flux terms WA. We have taken the time to derive the
molar flux equations in this form because they are now in a form that is con
sistent with the partial differential equation (PDE) solver COMSOL.
mol mol dm
  
dm2 s dm3 s
The molar average velocity for a binary system is
V yAVA yBVB (157)
The total molar flux of A is given by Equation (154). BA can be
expressed either in terms of the concentration of A, in which case
WA JA CAV (158)
We now need to evaluate the molar flux of A, JA, that is superimposed on the
molar average velocity V.
du
Momentum Transfer t 
dz
The mass transfer flux law is analogous to the laws for heat and momentum
transport, i.e., for constant total concentration
dC
Mass Transfer JAz DAB A (1512)
dz
The general 3dimensional constitutive equation for JA, the diffusional flux of
A resulting from a concentration difference, is related to the mole fraction gra
dient by Ficks first law:
JA cDAB yA (1513)
where c is the total molar concentration (mol/dm3), DAB is the diffusivity of A
in B (dm2/s), and yA is the mole fraction of A. Combining Equations (159)
and (1513), we obtain an expression for the total molar flux of A:
When accounting for diffusional effects, the molar flow rate of species A, FA ,
in a specific direction z, is the product of molar flux in that direction, WAz , and
the crosssectional area normal to the direction of flow, Ac :
FAz AcWAz
In terms of concentration the flux is
dC
WAz DAB A CAUz
dz
The molar flow rate is
dC
FAz WAz Ac DAB A CAUz Ac (1515A)
dz
Similar expressions follow for WAx and WAy . Substituting for the flux WAx,
WAy, and WAz into Equation (152), we obtain
2 2 2
Flow, diffusion, and CA CA CA C C C C
reaction. DAB  
  
 Ux A Uy A Uz A rA A (1516)
This form is x
2
y
2
z
2 x y z t
used in COMSOL
Multiphysics.
2
d CA dC
2
 Uz A rA 0
DAB  (1517)
dz dz
Figure 152 Cylindrical shell of thickness Dr, length Dz, and volume 2prDrDz.
C
Radial Direction Wir De i UrCi (1522)
r
where Ur (m/s) is the average velocity in the radial direction. For now we will
neglect the velocity in the radial direction, i.e., Ur = 0.
Moles of A
WAr Crosssectional area WAr 2
rz
in at r normal to radial flux
Moles of A
WAz Crosssectional area WAz 2
rr
in at z normal to axial flux
Moles of A
Moles of A Moles of A Moles of A
in at r out at ( r r ) in at z out at ( z z )
Moles of A Moles of A
formed Accumulated
WAr2
rz r WAr 2
rz r r WAz 2
rr z WAz 2
rr z z
CA ( 2
rrz )
rA2
rrz 
t
Dividing by 2
rrz and taking the limit as Dr and Dz 0
1 ( rWAr ) WAz C
   rA A
r r z t
Similarly, for any species i and steadystate conditions,
1 ( rWir ) Wiz
   ri 0 (1523)
r r z
Using Equations (1521) and (1522) to substitute for Wiz and Wir in Equation
(1523) and then setting the radial velocity to zero, Ur = 0, we obtain
1 C C
  De i r  De i UzCi ri 0
r r r z z
From Elements of Chemical Reaction Engineering, 2006 Pearson Education, Inc.
Fourth Edition, by H. Scott Fogler. All rights reserved.
Sec. 15.3 Energy Flux 159
For steadystate conditions and assuming Uz does not vary in the axial direction,
C D C C
2 2
C
De 2i e i De 2i Uz i ri 0 (1524)
r r r z z
Q Fi0Hi0 FiHi 0 (1525)
i1 i1
In terms of the molar fluxes and the crosssectional area and (q = Q /Ac )
Ac[ q ( Wi0Hi0 WiHi ) ] 0 (1526)
The q term is the heat added to the system and almost always includes a con
duction component of some form. We now define an energy flux vector, e,
(J/m2 s), to include both the conduction and convection of energy.
e = energy flux (J/sm2) e = Conduction + Convection
e q WiHi (1527)
where the conduction term q (J/m2 s) is given by Fouriers law. For axial (z)
and radial (r) conduction Fouriers laws are
T
qz ke  and qr = ke T

z r
where ke is the thermal conductivity (J/msK). The energy transfer (flow) is
the vector flux times the crosssectional area, Ac, normal to the energy flux
Energy flow = e Ac
Using the energy flux, e, to carry out an energy balance on our annulus (Figure
152) with system volume 2prDrDz, we have
( Energy flow in at r ) er Acr er 2
rz
Accumulation
Energy Flow Energy Flow Energy Flow Energy Flow
of Energy in
in at r out at r r in at z out at z z
Volume ( 2
rrz )
( er2
rz ) r ( er2
rz ) r r ez2
rr z ez2
rr z z 0 (1528)
Dividing by 2prDrDz and taking the limit as Dr and Dz 0,
1 ( rer ) ez
 
  0 (1529)
r r z
The radial and axial energy fluxes are
er qr Wir Hi
ez qz Wiz Hi
1 [ r [ qr Wir Hi ] ] [ qz Wiz Hi ]
   0 (1530)
r r z
and expanding the convective energy fluxes, Wi Hi ,
Neglect
1 1 ( rW ) W H
Radial:   ( rWir Hi )  Hi 
ir
 
ir
i (1531)
r r r r r
( Wiz Hi ) W H
Axial:  Hi iz Wiz i (1532)
z z z
Substituting Equations (1531) and (1532) into Equation (1530), we obtain
upon rearrangement
ri
Recognizing that the term in brackets for steadystate conditions is just the rate
of formation of species i, ri, we have
q H
1  ( rqr ) z Hiri Wiz i 0 (1533)
r r z z
Recalling
T T H T
qr ke  , qz ke  , i CPi  ,
r z z z
From Elements of Chemical Reaction Engineering, 2006 Pearson Education, Inc.
Fourth Edition, by H. Scott Fogler. All rights reserved.
Sec. 15.4 Some Initial Approximations 1511
and
ri i ( rA )
C
Assumption 1. Neglect the axial diffusive term [i.e.,  De i ] wrt the con
z
vective term in the expression involving heat capacities
k rT T
2
T
  ke 2 HRxrA ( UzCP Ci )  0
e  (1535)
r r r z
i z
2
r
Uz 2U0 1  (1536)
R
where U0 is the average velocity inside the reactor and R is the tabular reactor radius.
Energy balance Assumption 2. Assume that the sum CPm CPiCi CA0
iCPi is constant.
with radial and
axial gradients The energy balance now becomes
T k T
2
T
ke 2 e  r  HRxrA UzCPm  0 (1537)
z r r r z
Equation (1536) is the form we will use in the COMSOL problem that fol
lows. In many instances, the term CPm is just the product of the solution den
sity and the heat capacity of the solution (J/kg K).
T
m cCPc a Uht2
R[ T( R,z ) Ta ] (1538)
z
T
m c Ta0
r=0
FA0
T0
mc Ta0 z
This example will highlight the radial effects in a tubular reactor, which up until
now have been neglected to simplify the calculations. Now, the effects of parameters
such as inlet temperature and flow rate will be studied using the software program
COMSOL. Follow the stepbystep procedure in the Web Module and on the CDROM.
From Elements of Chemical Reaction Engineering, 2006 Pearson Education, Inc.
Fourth Edition, by H. Scott Fogler. All rights reserved.
Sec. 15.6 Boundary and Initial Conditions 1513
A B C
E 1 1
k(T) = k0(T0) exp    (E151.3)
R T0 T
with k0 = 1.28 h1 at 300 K. The thermal conductivity ke of the reaction mixture and
the diffusivity De are 0.599 W/m/K and 109 m2/s, respectively, and are assumed to
be constant throughout the reactor. In the case where there is heat exchange between
the reactor and its surroundings, the overall heattransfer coefficient is 1300 W/m2/K
and the temperature of the cooling jacket temperature is assumed to be constant and
is set to 273 K. The other property data are shown in Table E151.1.
TABLE 151.1. PHYSICAL PROPERTY DATA
Solution
Mole Balances: Recalling Equation (1424) and applying it to species A
2 2
CA 1 CA CA C
  De  De 
De   Uz A rA 0 (E151.4)
r
2 r r z
2 z
Rate Law:
E 1 1
rA k( T1 ) exp  
  C (E151.5)
R T1 T A
Stoichiometry: The conversion along a streamline (r) at a distance z from the reac
tor entrance
X(r, z) = 1 CA(r, z)/CA0 (E151.6)
The overall conversion at a given distance z from the reactor entrance is
R
2
CA( r, z )Uzrdr
X ( z ) 1 
0
 (E151.7)
FA0
The mean concentration at any distance z from the reactor entrance is
R
2
CA( r, z )Uzrdr
CA( z ) 
0
2
 (E151.8)
R U0
For plug flow, the velocity profile is
Uz = U0 (E151.9)
The laminar flow velocity profile is
2
r
Uz 2U0 1  (E151.10)
R
T k T
2
T
ke 2 e  r  H Rx
rA UzCPm  = 0 (1537)
z r r r z
Assumptions
1. Ur is zero.
2. Neglect axial diffusion/dispersion flux with regard to convective flux when
summing the products of the heat capacities and their respective fluxes.
3. Steady state.
Cooling jacket
T
mCPC a 2
RUht ( T( R,z ) Ta ) (1538)
z
Boundary conditions
C
At r = 0, then i 0 and T
 0 (E151.11)
r r
C T
At r = R, then i 0 and ke  Uht ( T( R,z ) Ta ) (E151.12)
r r
At z = 0, then Ci = Ci0 and T = T0 (E151.13)
These equations were solved using COMSOL for a number of cases including adi
abatic and nonadiabatic plug flow and laminar flow; they were also solved with and
without axial and radial dispersion. A detailed accounting on how to change the
parameter values in the COMSOL program can be found in the COMSOL Instruc
tions section on the Web in screen shots similar to Figure E151.1. Figure E151.2
gives the data set in SI units used for the COMSOL example.
Define expression
Figure E151.1 COMSOL screen shot of data set for a second order reaction.
Color surfaces are used to show the concentration and temperature profiles, similar
to the black and white figures shown in Figure E151.2. Read through the COMSOL
Web module entitled Radial and Axial Temperature Gradients. One notes in
Figure E152.1 that the conversion is lower near the wall because of the cooler fluid
temperature. These same profiles can be found in color on the Web and CDROM in
the Web Modules.
(a) (b)
Temperature Surface Radial Temperature Profiles
1
350 350
0.9
340 340
0.8
Temperature (K)
0.6 320 320
0.5
310 310
0.4
300 300
0.3
Outlet
0.2 290 290 Half Axial Location
Inlet
0.1
280 280
Average Outlet Conversion=90.3%
0
0.2 0.1 0 0.1 0.2 0 0.02 0.04 0.06 0.08 0.1
Radial Location (m) Radial Location (m)
0.6
Conversion
0.6 0.6
0.5 0.5 0.5
0.4 0.4 0.4
0.3 0.3 0.3
Outlet
0.2 0.2 0.2 Half Axial Location
Inlet
0.1 0.1 0.1
Average Outlet Conversion=90.3%
0 0 0
0.2 0.1 0 0.1 0.2 0 0.02 0.04 0.06 0.08 0.1
Radial Location (m) Radial Location (m)
(c) (d)
Figure E151.2 (a) Temperature surface, (b) temperature surface profiles,
(c) conversion surface, and (d) radial conversion profile.
Analysis: One notes the maximum and minimum in these profiles. Near the wall,
the temperature of the mixture is lower because of the cold wall temperature. Con
sequently, the rate will be lower, and thus the conversion will be lower. However,
right next to the wall, the velocity through the reactor is almost zero so the reactants
spend a long time in the reactor; therefore, a greater conversion is achieved, as noted
by the upturn right next to the wall.
One of the common methods to produce phthalic anhydride is from the partial oxi
dation of oxylene. This simplified reaction for the formation of phthalic anhydride
from oxylene oxidation will be represented by
O
Me
O
+ O2 + 3H2O
Oxylene
Me Phthalic Anhydride O
Construct a model of this reactor and investigate the radial and axial temperature
profiles.
(a) Run your model using values of thermal conductivity from 0.1 to 1000 J/(m
K s).
(b) Compare this model to the plug flow model for mass and heat transfer. Dis
cuss the overall effect of the value of effective thermal conductivity of the bed
on the resulting temperature profile and outlet conversion obtained.
(c) Give a rationale as to why industry does not use tubes of 0.1m in diameter,
but instead uses a smaller tube diameter of 0.0254 m.
Preliminary Calculations
To produce 76 metric ton/y during 350 d/y operation
FC 0.17
F A0 = = = 0.215 mol s
X 0.79
FA 0 79
FT 0= = FA 0 + FB0 + FN2 = FA 0 + FB 0 + FB0
0.012 21
F T 0  FA 0
FB0 = = 0.372 mol s
0.79
1+
0.21
FT = 1.79 mol s
( )
AC = p 0.1m 2 = 0.0314
v0 = 0.0896 m3 s
Solution
In COMSOL we will use a 2D axissymmetric model in the r and z directions. This
reaction takes place in a packedbed reactor, where we will assume that the flow is
plug flow with no axial or radial diffusion.
Assumptions:
1. Ur = 0 (i.e., assume negligible velocity in the radial direction)
2. Uz = constant
3. Neglect axial diffusion/dispersion flux with respect to convective flux
4. Assume constant density and that mean heat capacity of the gas can be
approximated by the values of air
5. Steadystate
Mole Balance:
Starting with Equation (1524)
2Ci De Ci 2C C
De 2 + + De 2i  Uz i + ri = 0 (1524)
r r r z z
Neglecting diffusion in the axial and radial directions gives the plug flow equation:
Ci
Uz + ri = 0 (E152.1)
z
Note that the velocity Uz is constant.
Rate Law:
mol
 rA = k PA PO2 2 (E152.2)
kgcat Pa s
To convert the rate to per reactor volume we multiply by the bulk density of the cat
alyst bed, rB
For the COMSOL model, the partial pressures need to be converted to concentration
using the ideal gas law Ci = Pi /RT
2
 rA = k ( RT ) CACO2 (E152.4)
with
E

RT mol
k = Ae (E152.5)
m3 Pa 2 s
Stoichiometry:
Because of the dilute reactants, e = 0. We also will neglect changes in volumetric
flow rate with temperature and pressure
v = v0
FA FO F
CA = , CO2 = 2 , CC = C (E152.6)
v0 v0 v0
Energy Balance:
Starting with the energy balance from Equation 1537:
2 T ke T T
ke + r + DH Rx rA  Uz CPm =0 (1537)
z 2 r r r z
ke T T
r + DH Rx rA  Uz CPm =0 ( E152.7)
r r r z
The boundary conditions for these PDEs will be:
T
(A)  ke = Uht (T ( R, z )  Ta ) at r = R and 0 z 1 m (E152.8)
r R
Findings:
Temperature (K)
740
1
2
3
4
720
5
700
Temperature (K)
680
660
640
620
600
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
z
Results:
A surface plot of the temperatures in the reactor tube is shown in Figure E152.1
which shows a noticeable difference in temperatures of the fluid at the end of the
reactor starting at about z = 0.7 m. For this value of thermal conductivity, the wall
remains cold at approximately the temperature of the molten salt of 720K. This
can also be seen in the crosssection plot that is shown in Figure E152.2. Each line
on this plot represents a radial position within the reactor. In Figure E152.3 a cross
section plot showing the temperature profile as a function of radial position and axial
distance in the reactor is presented. Each line in this figure represents an axial posi
tion in the reactor. In this plot, the difference between the wall temperature and the
temperature within the reactor is clearly shown. The plug flow condition, in which
the temperature profile has been eliminated in the radial direction, is shown in Fig
ure E152.4. This result was obtained with a thermal conductivity of 1000 J/(s m K).
Temperature (K)
740
1
2
3
4
720
5
700
Temperature (K)
680
660
640
620
600
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
r
Figure E152.3: Temperature Profile in reactor tube for ke=0.78 J/(s m K). Positions: 1)
z=0.0 m, 2) r=0.2m, 3) z=0.4m, 3) z=0.6m, 4) z=0.8m, 5) r=1.0 m This plot shows a large
temperature variation in the radial direction.
Temperature (K)
660
1
2
3
655
4
5
650
645
640
Temperature (K)
635
630
625
620
615
610
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
r
Figure E152.4: Temperature Profile in reactor tube for ke=1000 J/(s m K). Positions: 1)
z=0.0 m, 2) r=0.2m, 3) z=0.4m, 3) z=0.6m, 4) z=0.8m, 5) r=1.0 m This plot shows that
there is no temperature variation with respect to radial position.
From Elements of Chemical Reaction Engineering, 2006 Pearson Education, Inc.
Fourth Edition, by H. Scott Fogler. All rights reserved.
1522 Radial and Axial Temperature Variations in a Tubular Reactor Chap. 15
Table E152.2 gives a summary of the output from the COMSOL program.
Analysis: From Table E152.2 we observe that the removal of heat is effective in
lowering the temperature within the reactor. The advantage of the molten salt cool
ing is that it will help to prevent a runaway reaction in the reactor tubes. As the
radial thermal conductivity is increased, the overall rate of heat transfer from the
fluid to the reactor walls increases. As you can see from the table, the conversion
also decreases.
SUMMARY
To obtain the axial or radial temperature and concentration gradients, the fol
lowing coupled partial differential equations were solved using COMSOL:
2 2
C D C C C
De 2i e i De 2i Uz i ri= 0 (S151)
r r r z z
and
2
T k T T
ke 2 e  r  HRxrA UzCPm  = 0 (S152)
z r r r z
CDROM MATERIAL
Learning Resources
1. Summary Notes
2. Web Module COMSOL Radial and Axial Gradients
External Internal
Skin Boundary Skin Boundary
Partial Pressure Partial Pressure
N2 101 kPa 0
He 0 81 kPa
1 20 m Stratum corneum
2 80 m Epidermis
(c) How would your answers to (a) and (b) change in the dead of winter
when T = 0F?
(d) Critique and extend this problem (e.g., CO2 poisoning).
P154 Instructions: If you have not installed COMSOL Multiphysics, request a trial
version of the software from the Web site http://www.comsol.com/ecre. From
this page you can also download model and documentation files required for
the COMSOL exercises outlined in this chapter.
Load COMSOL Multiphysics and follow the installation instructions.
Doubleclick on the COMSOL Multiphysics icon on your desktop. In the
Model Library, select model denoted 4NonIsothermal Reactor II and click
the Dynamic Help icon on the main toolbar. This opens a Help window in the
COMSOL desktop presenting the detailed documentation of the specific
model. Scroll the side bar to review the model equations. Move further down
the documentation to find the stepbystep instructions that guide you through
the modelbuilding process. The steps also detail how to solve the model and
analyze the results. If you want bypass the model setup process, return to the
Model Library dialog and press OK to open a saved model file.
(a) Why is the concentration of A near the wall lower than the concentration
near the center?
(b) Where in the reactor do you find the maximum and minimum reaction
rates? Why? Instructions: Select the 2D Plot Group 1 > Surface 1 node
to access the surface plot dialog. Type rA (replace cA) in the
Expression edit field to plot the absolute rate of consumption of A
(moles m3 s1).
(c) Increase the activation energy of the reaction by 5%. How do the concen
tration profiles change? Decrease? Instructions: Select the Global Defi
nitions > Parameters node. Multiply the value of E in the constants
list by 1.05 (just type *1.05 behind the existing value to increase or
multiply by 0.95 to decrease). Go to the Study 1 node and select
Compute.
(d) Change the activation energy back to the original value. Instructions:
Select the Global Definitions > Parameters node. Remove the factor
0.95 in the constants list.
(e) Increase the thermal conductivity, ke, by a factor of 10 and explain how
this change affects the temperature profiles. At what radial position do
you find the highest conversion? Instructions: Multiply the value of ke
in the constants lists by 10. Go to the Study 1 node and select Compute.
(f) Increase the coolant flow rate by a factor of 10 and explain how this
change affects the conversion.
(g) In two or three sentences, describe your findings when you varied the
parameters (for all parts).
(h) What would be your recommendation to maximize the average outlet
conversion?
(i) Review Figure E151.2 and explain why the temperature profile goes
through a maximum and why the conversion profile goes through a max
imum and a minimum.
(j) See other problems in the Web Module.
P155 If you have not installed COMSOL Multiphysics, request a trial version of the
software from the Web site http://www.comsol.com/ecre and follow the instal
lation instructions as outlined in P154.
(a) Before running the program, sketch the radial temperature profile down a
PFR for (1) an exothermic reaction for a PFR with a cooling jacket and
(2) an endothermic reaction for a PFR with a heating jacket.
(b) Run COMSOL Multiphysics and compare with your results in (a). Dou
bleclick on the COMSOL Multiphysics icon on your desktop. In the
Model Library, select the model denoted 3NonIsothermal I and press
OK. You can use this model to compare your results in (1) and (2),
above. Click the Dynamic Help icon on the main toolbar to review the
instructions for this model and other models in COMSOL. Change the
velocity profile from laminar parabolic to plug flow. Select the Model 1
> Definitions > Variables 1 node. Change the expression for uz (the
velocity) to u0 (replace the expression 2*u0*(1(r/Ra)2), which
describes the parabolic velocity profile). You can now continue to vary
the input data and change the exothermic reaction to an endothermic one.
(Hint: Select the Global Definitions > Parameters node to access the
list of constants. Do not forget Ta0 the jacket temperature at the end of
the list.) To run a simulation, go to the Study 1 node and select Com
pute. Write a paragraph describing your findings.
(c) The thermal conductivity in the reactor, denoted ke in Figure E151.1,
is the molecular thermal conductivity for the solution. In a plug flow
reactor, the flow is turbulent. In such a reactor, the apparent thermal con
ductivity is substantially larger than the molecular thermal conductivity
of the fluid. Vary the value of the thermal conductivity ke to learn its
influence on the temperature and concentration profile in the reactor.
(d) In turbulent flow, the apparent diffusivity is substantially larger than the
molecular diffusivity. Increase the molecular diffusivity in the PFR to
reflect turbulent conditions and study the influence on the temperature
and concentration profiles. Here you can go to the extremes. Find some
thing interesting to turn in to your instructor. See other problems in the
Web Module.
(e) See other problems in the Web Module.
SUPPLEMENTARY READING
Гораздо больше, чем просто документы.
Откройте для себя все, что может предложить Scribd, включая книги и аудиокниги от крупных издательств.
Отменить можно в любой момент.