Catalytic

Reaction Engineering

Yongdan Li
Professor of Industrial Chemistry
Department of Chemical and Metallurgical Engineering
School of Chemical Technology
Aalto University
Email: [email protected] Nov-Dec, 2018
Kemistintie 1, E404
7. Internal diffusion
 7. Diffusion and Reaction

Catalyst particles

Extrudate D=1-5 mm Fixed bed reactor

L = 3-30 mm

Pellet D=3-10 mm Fixed bed reactor

H=3-10 mm

Granule D=1-20 mm Fixed bed reactor

Bead D=1-5 mm Fixed bed reactor
Sphere D=1-5 mm Fixed bed reactor
Sphere D=20-100 m Fluidized bed &
Slurry bed reactors

3
 7. Diffusion and Reaction

Pore structure of soid catalyst particles

1 μm

10 nm

1 nm

Fig 7.1 Pore structure of the catalyst.

4
 7. Diffusion and Reaction

Diffusivity vs. pore size

Fig 7.2 Diffusivity versus pore size. Classical regions of bulk and knudsen
diffusion and regime of configurational diffusion (adapted from Weisz)
5
 7. Diffusion and Reaction

Pore size of zeolites

Fig 7.3 Pore sizes of zeolites.

6
 7. Diffusion and Reaction

 Zeolites

Fig 7.4 Zeolites.

• Microporous crystalline Al2O3/SiO2 materials with ordered pore structure

(pores 3-10 Å)
• Zeolites act as molecular sieves

7
 7. Diffusion and Reaction

 Shape selectivity: alkylation of toluene with ethene

HCl-AlCl3 ZSM-5 ZSM-5 modified

Composition (%)
Light components 0.2 1.0 0.9
Toluene 48.3 74.4 86.2
p-ethyl toluene 11.9 7.0 11.9
m-ethyl toluene 19.3 14.7 0.4
o-ethyl toluene 3.8 0.3 0.0
heavy aromatics 15.3 1.4 0.1
Distribution of isomers (%)
para 34.0 31.8 96.7
meta 55.1 66.8 3.3
ortho 10.9 1.4 0.0
8
 7.1 Diffusion and Reactions in Homogeneous Systems
For homogeneous systems, the mole balance on species A

WAx WAy WAz C

    rA = A (7-1)
x y z t

For one-dimensional diffusion at steady state:

dWAz
  rA  0 (7-2)
dz
For diffusion through a stagnant film at dilute concentrations

dC A
Fick's law WAz   DAB (7-3) Fig 7.5 The mole balance to species A.
dz • WA Molar flux of species A
(mol/m2·s)
• DAB Binary diffusion
coefficient of A in B (dm2/s)
d 2C A
(7-2) DAB 2
 rA  0 (7-4) • rA Rate of generation of species
A per unit volume (gmol
dz
A/s·dm3)
9
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

Fig 7.6 Mass transfer and reaction steps for a catalyst pellet.

• May affect the overall reaction rate

• Reactant concentrations at the exit of pores higher than inside the pores
(opposite for the products)
• Concentration gradients in the particles can be calculated
10
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

7.2.1 Effective diffusivity

 The effective diffusivity accounts for the fact that:

1. Not all of the area is available

(i.e., the area occupied by solids) for the molecules to diffuse.
2. The paths are tortuous.
3. The pores are of varying cross-sectional areas.

An equation that relates the effective

diffusivity De to either the bulk
diffusivity DAB or the Knudsen
diffusivity DK is

DAB ε p σ c DAB p σ c
De   (7-5)
τ τ
Fig 7.6 Mass transfer and reaction steps for a catalyst pellet.

11
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

Actual distance a molecule travels between two points

τ  tortuosity 
Shortest distance between these two points

Volume of void space

 p (ε p )  pellet porosity 
Total volume (voids and solids)

 max. cross sectional area of pores 

σ c  Constriction factor  f  
 min. cross sectional area of pores 

Fig 7.7 (a) Pore constriction; (b) pore tortuosity

12
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets
7.2.2 Differential equation describing diffusion and reaction in a single catalyst pellet

• Assumptions:
• Steady state
• Spherical particle
• Reaction A  B
• Constant T and P in the particles

Rate of A in at r = WAr  4 r 2 |r (7-6)

Rate of A out at ( r  r )  WAr  4 r 2 |r r (7-7)
Fig 7.8 Shell balance on a catalyst pellet.
 Rate of generation 
of A within a shell    Rate of reaction    Mass catalyst   Volume of shell
   Mass of catalyst   Volume   
of thickness r   

 rA'  ρc  4 rm2 r (7-8)

13
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

The mole balance over the shell thickness △r is

(In at r )  (out at r  r ) + (Generation within r ) = 0 (7-9)
WAr  4 r 2 |r  WAr  4 r 2 |r r  rA' ρ c  4 rm2 r =0

Dividing by (-4π △r) and letting △r→0

d (WAr r 2 ) '
 rA ρ c r 2  0 (7-10)
dr
Molar flux
reaction A  B Equal Molar Counter Diffusion (EMCD)
dC A
WAr   De (7-11)
dr

  dC  
d   De  A  r 2 
(7-10)
  dr   ' (7-12)
 rA ρ c r 2  0
dr
14
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

 Different units for reaction rate can be applied (generation rate for A):
For nth order in the gas-phase concentration of A within the pellet
n-1
Based on per  kmol   m  3
1
unit volume rA  3   knC An kn  k ρ  k ρ S a  
'
n c
''
n c 
 ms  kmol  s
n-1
'  kmol   
3
Based on per m m3
unit mass of catalyst  r    k '
C n
kn  kn S a  
' ''
 kg  s
A n A
 g cat s   kmol 
n-1
Based on per ''  kmol   m3  m
unit surface area
rA  2 
 k '' n
CA kn  
''

 msurface area s   kmol  s
n

rA'  S a (  rA'' )
• Sa=surface area of the catalyst per
rA  ρ c (  r )'
A mass of the catalyst (m2/g cat.)
• c=density of a catalyst particle
rA  ρ c S a (  rA'' )
15
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

  dC   rA'  S a (  rA'' )   dC  
d   De  A  r 2  d   De  A  r 2 
  dr   '   dr   ''
 rA ρ c r 2  0 (7-12)  rA S a ρ c r 2  0 (7-13)
dr dr

  dC A  2 kn
d   De  r  rA''  k n'' C An
  dr  
 kn'' S a ρ c C An r 2  0 (7-14)
dr

2
dCA 2 d CA
2r (  De )  De r 2
 k n Ar  0
C n 2
(7-15)
dr dr

dividing through by –r2De

d 2CA 2 dCA kn n
2
  CA  0 (7-16)
dr r dr De
16
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

This balance equation combines reaction and

diffusion in catalyst particles (spherical, steady
state, isothermal particles)

d 2CA 2 dCA kn n
2
  CA  0 (7-16)
dr r dr De
• Boundary conditions
CA is finite at r = 0
Fig 7.9 Shell balance on a catalyst pellet.

CA = CAs at r = R
Principally same equation is valid also for
other particle geometries （s=a-1）

d 2CA s dCA kn n
2
  CA  0 (7-17)
dr r dr De
17
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

7.2.3 Diffusion with the catalytic reaction, equation in dimensionless form

 Introduce dimensionless variables

CA
ψ (7-18)
CAs
r
λ (7-19)
R

 Change the boundary condition

CA r
CA = CAs at r = R ψ 1 at λ= =1
CAs R
becomes

CA is finite at r = 0 ψ is finite at λ  0

18
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

Rewrite the differential equation in term of the dimensionless variables.

dC A
Start with WAr   De (7-11)
dr

dC A dλ 1
 C As 
dψ dr R

dC A  dC A  dλ dψ  dC A  dλ dC A dψ C As
     (7-20)  (7-21)
dr  dλ  dr dλ  dψ  dr dr dλ R
chain rule

The flux of A in terms of the dimensionless variables

dC A D C  dψ 
WAr   De   e As   (7-22)
dr R  dλ 

19
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

At steady state, the net flow of species A that enters into the pellet at the external pellet
surface reacts completely within the pellet. The overall rate of reaction is therefore equal to
the total molar flow of A into the catalyst pellet.

“a black hole”: All the reactant that diffuses into the pellet is consumed

WA(mol/m2s) Catalyst

Fig 7.10 A black hole.

MA (mol/s) The molar flux into the The external surface

= ×
The overall rate of reaction pellet at the outer surface area of the pellet

dCA dψ
M A  4 R 2WAr |r  R  4 R 2 De |r  R  4 RDeCAs |λ=1 (7-23)
dr dλ

Back 20
7.2 Diffusion and Reactions in Spherical Catalyst Pellets

dC A dψ C As
 (7-21)
dr dλ R

d 2C A d  dC A  d  dψ C As  dλ d ψ  C As 
2
      2 2  (7-24)
dr 2
dr  dr  dλ  dλ R  dr dλ  R 

d 2CA 2 dCA kn n
2
  CA  0 (7-16)
dr r dr De

dividing through by CAs/R2

n 1
d 2 ψ 2 dψ kn R 2C As
  ψ n
0 (7-25)
dλ 2
λ dλ De

21
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets
Mass balance (spherical part.) using dimensionless variables

d 2 ψ 2 dψ
  n ψ 0
2 n (7-26)
dλ 2
λ dλ
n 1
kn R 2C As
 
2
n (7-27)
De

ϕn is called Thiele modulus (n gives the reaction order)

n 1
kn R 2CAs n
kn RCAs rate of surface reaction
 
2
n   (7-28)
De DeC As / R rate of pore diffusion

If Thiele modulus is large = > diffusion may limit the overall rate
If Thiele modulus is small => surface reaction determines the rate

22
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets
Dimensionless balance for a first order reaction
A→B
• Surface reaction rate limited with respect to the adsorption of A and desorption of B
• A and B were weakly adsorbed (i.e., low surface coverage)

rA''  1 k1''CA   rA''  k1''CA

d 2 ψ 2 dψ
  12 ψ  0 (7-29)
dλ 2
λ dλ

And for Thiele modulus: The boundary condition

k1 k1''ρ c S a B.C.1: ψ  1 at λ=1

1  R R n 1 (7-30) B.C.2: ψ is finite at λ  0
De De

23
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

7.2.4 Solution to the differential equation for a first-order reaction

Transformation y  ψλ

dψ 1  dy  y
   2 (7-31)
dλ λ  dλ  λ

d 2 ψ 1  d 2 y  2  dy  2 y (7-32)
  2  2   3
dλ 2
λ  dλ  λ  d λ  λ

d 2 ψ 2 dψ
  1
2
ψ0 (7-29)
dλ 2
λ dλ

d2y
 1
2
y0 (7-33)
dλ 2

24
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

Appendix 1 Second-Order Ordinary Differential Equations

(A-1)

To determine the characteristic roots of

(A-2)

The solution to the differential equation is

(A-3)

where A1 and B1 are arbitrary constants of integration. It can be verified that Equation (A-3)
can be arranged in the form
(A-4)

Equation (A-4) is the more useful form of the solution when it comes to evaluating the
constants A and B because sinh(0) = 0 and cosh(0) = 1.0.

25
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

Appendix 2 Hyperbolic function

26
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets
This differential equation

d2y B.C.1: ψ  1 at λ=1

 1
2
y0 (7-33)
B.C.2: ψ is finite at λ  0
dλ 2

has the following solution

y  A1 cosh 1λ  B1 sinh 1λ (7-34) At λ  0 1  A1 cosh 1  B1 sinh 1

cosh 1λ  1,
Or 0  A1
A1 B 1/ λ   ,
ψ cosh 1λ  1 sinh 1λ (7-35) sinh 1λ  0  B1 
1
λ λ sinh 1

The dimensionless concentration profile is

C A 1 sinh1λ
ψ  (7-36)
C As λ sinh1
27
 7.2 Diffusion and Reactions in Spherical Catalyst Pellets

Equation

C A 1 sinh1λ
ψ  (7-36)
C As λ sinh1

can be used for the simulation of particle

concentration gradients

• Small ϕ1  surface reaction controls =>

concentration of A high in the pores

• Large ϕ1 internal diffusion limits the

rate of reaction.
Fig 7.11 Concentration profile in a spherical catalyst pellet.
 To reduce the value of Thiele modulus
Smaller catalyst particles

28
 7.3 The Internal Effectiveness Factor

The internal effectiveness factor is defined as

Actual overall (observed) rate of reaction

η
Rate of reaction that would result if entire interior surface were (7-37)
exposed to the external pellet surface conditions CAs , Ts
 η is a measure of how far the reactant diffuses into the pellet before reacting.

rA rA' rA''

η  '  '' (7-38)
rAs rAs rAs

To derive the effectiveness factor for a first-order reaction, it is easiest to work in reaction
rates of moles per unit time, MA, rather than in moles per unit time per volume of catalyst.

rA rA  Volume of catalyst particle M A

η   (7-39)
rAs rAs  Volume of catalyst particle M As

29
 7.3 The Internal Effectiveness Factor
If the entire surface were exposed to the concentration at the external surface of the pellet,
CAs, the rate for a first-order reaction would be

Rate at external surface

M As   Volume of catalyst pellet
Volume
4  4 
M As  rAs    R 3   k1C As   R 3  (7-40)
3  3 

The actual rate of reaction is the rate at which the reactant diffuses into the pellet at the
outer surface; that is, all of A that diffuses into the pellet at the outer surface reacts and no A
diffuses back out. (It behaves as a “black hole.”)

dψ
M A  4 RDeCAs |λ=1 (7-23)
dλ

30
 7.3 The Internal Effectiveness Factor

For ψ the equation was obtained already earlier

C A 1 sinh1λ
ψ  (7-36)
C As λ sinh1
Differentiation

dψ  1 cosh λ1 1 sinh 1λ 

  2   1 coth 1  1
(7-41)
λ 1 
dλ  λ sinh 1 λ sinh 1  λ 1

The actual rate of reaction

dψ
M A  4 RDeCAs |λ=1 (7-23)
dλ

M A  4π RDeCAs (1 coth 1  1) (7-42)

31
 7.3 The Internal Effectiveness Factor

4  4 
M As  rAs    R 3   k1C As   R 3  (7-40)
3  3 

M A  4π RDeCAs (1 coth 1  1) (7-42)

M A 4π RDeCAs (1 coth 1  1) 3

η   (1 coth 1  1) (7-43)
M As 4  k R 2
/ D
k1C As   R 3  1 e

3  12

 Internal effectiveness factor

-----for a first-order reaction in a spherical catalyst pellet

3
η (1 coth 1  1) (7-44)
12

32
7.3 The Internal Effectiveness Factor

Fig 7.12(a) Effectiveness factor plot for nth-order kinetics on spherical catalyst particles 33
 7.3 The Internal Effectiveness Factor

For small particles

  small
1
 surface reaction controls
Large particles
  large ( 30)
  << 1
 internal diffusion
limits the reaction rate

Fig 7.12 (b) First-order reaction in different pellet geometrics. 34

 7.3 The Internal Effectiveness Factor
 With large values of  ( 30),   ( 0.1), :

3
η (1 coth 1  1) (7-44)
1
2

3 3 De
η  (7-45)
1 R k1

 Overall reaction rate (1st order reaction, spherical particles, only internal diffusion limits)

 Actual reaction rate 

rA      Reaction rate at CAs 
 Reaction rate at CAs  (7-46)
=  (rAs )
For a first order reaction
rA    (k1C As ) (7-47)

3 De 3
(7-44) rA   (k1C As )  De k1 C As (7-48)
R k1 R
35
 7.3 The Internal Effectiveness Factor
7.3.2 Effectiveness factor with volume change during reaction

When there is volume change，ε ≠ 0 ε Fraction change in volume per

mole of A reacted resulting from the
The correction is obtained from a plot of the change in total number of moles.
ratio of effective factors

 ' Factor in the presence of volume change


 Factor in the absence of volume change

 For example,
A → 2B
if the Thiele modulus were 10 with (ε = 1)
then the effectiveness factor with volume
change would be .
Fig 7.13 Effectiveness factor ratios for first-order kinetics on spherical
  0.8
'
catalyst pellets for various values of the Thiele modulus of a sphere, ϕs, as
a function of volume change. 【From V. W. Weekman and R. L.
Goring,“Influence of Volume Change on Gas-Phase Reactions in Porous
Catalysts.” J.Catal., 4(2),260 (1965).】
36
 7.3 The Internal Effectiveness Factor
7.3.3 Internal diffusion limited reactions other than first order

For reactions of order n

n 1
kn R 2C As
 
2
n
(7-27)
De
For large values of the Thiele modulus, the effectiveness factor is

1/ 2 1/ 2
 2  3  2  3 De (1 n ) / 2
η   (7-49)
n  n  1 
CAs
 n 1 R kn

How to increase the reaction rate for internal diffusion limited reactions？

1. decrease the radius R (make pellets smaller);

2. increase the temperature
3. increase the concentration
4. increase the pore size
37
 7.3 The Internal Effectiveness Factor

7.3.4 Weisz–Prater Criterion for internal diffusion limitations

 The Weisz–Prater Criterion uses measured values of the rate of reaction, -rA(obs), to
determine if internal diffusion is limiting the reaction.

12  3 1 coth 1  1 (7-50)

 The Weisz–Prater parameter

CWP    12
Observed  actual  reaction rate Reaction rate evaluated at C As (7-51)
= 
Reaction rate evaluated at C As A diffusion rate
Actual reaction rate

A diffusion rate

38
 7.3 The Internal Effectiveness Factor

rA'
= ' (7-38)
rAs
k1 k1''ρ c S a rAs'' S a ρ c R 2 rAs' ρ c R 2
1  R R n 1 (7-30)   1
2
 (7-52)
De De DeC As DeC As

 r '
A  obs   rAs' ρ c R 2 
CWP    1 
2
  (7-53)
rAs  DeC As 
'

rA'  obs ρ c R 2
CWP  12  (7-54)
DeC As

 If CWP << 1
No diffusion limitations
No concentration gradient exists within the pellet

 If CWP >> 1
Internal diffusion limits the reaction
39
 7.3 The Internal Effectiveness Factor

Example 7–1 Estimating the Thiele Modulus and Effectiveness Factor

The first-order reaction
AB
was carried out over two different-sized pellets. The pellets were contained in a
spinning basket reactor that was operated at sufficiently high rotation speeds that
external mass transfer resistance was negligible. The results of two experimental
runs made under identical conditions are given in Table 1.

(a) Estimate the Thiele modulus and effectiveness factor for each pellet. R2: η2=0.856
R1: η1=0.182
(b) How small should the pellets be made to virtually eliminate all internal diffusion
resistance, e.g., η = 0.95? R =0.55 mm
3
Table 1 Data from a spinning basket reactor

40
 7.3 The Internal Effectiveness Factor
Solution
(a) 12  3 1 coth 1  1 (7-50)
rA'  obs ρ c R 2
CWP  12  (7-54)
DeC As
rA'  obs ρ c R 2
CWP   12 =3 1 coth 1  1
DeC As

Letting the subscripts 1 and 2 refer to runs 1

and 2
rA2
'
R22 12 coth 12  1
=
rA1 R1 11 coth 11  1
' 2

The terms ρc, De, and CAs cancel because the runs were carried out under identical
conditions. The Thiele modulus is rAs' ρ c
1  R
DeC As
11 R1 R1 0.01m
= 11 = 12 = 12 =1012
12 R2 R2 0.001m
41
7.3 The Internal Effectiveness Factor
 15 10-5   0.001
2
12 coth 12  1
 -5 
=
 3 10   0.01
2
1012 coth 1012   1

12 coth 12  1

0.05=
1012 coth 1012   1

12 =1.65 R 2 =0.001m

11 =1012 =16.5 R1 =0.01m

3 12 coth 12  1 3 1.65coth1.65  1

2 = = =0.856
12
2
1.65 2

3 11 coth 11  1 3 16.5coth16.5  1

1 = = =0.182
11
2
16.5 2

42
 7.3 The Internal Effectiveness Factor
(b) Next we calculate the particle radius needed to virtually eliminate internal
diffusion control (say, η=0.95)
3 13 coth 13  1
0.95=
132
13 =0.9

13  0.9 
=（0.01）  =5.5 10 m
-4
R3 =R1
11  16.5 

A particle size of 0.55 mm is necessary to virtually eliminate diffusion control (i.e.,

η= 0.95)

Analysis: This example is important because it shows us how, with only two
measurements and some assumptions, we can determine internal diffusion
limitations for the two pellet sizes, and predict the pellet size necessary to
completely eliminate internal diffusion.

43
 Summary

 Thiele modulus
n 1
kn R 2CAs large = > diffusion may limit the overall rate
 
2
n
De small => surface reaction determines the rate

 Internal effectiveness factor

For large values of the Thiele modulus, the effectiveness factor is
1/ 2
 2  3
η 
 n 1  n

 Weisz–Prater Criterion
 If CWP << 1
rA'  obs ρ c R 2 No diffusion limitations
CWP   
1
2
No concentration gradient exists within the pellet
DeC As
 If CWP >> 1
Internal diffusion limits the reaction 44
 Summary

 Generation rate of A based on measued bulk concetrations

(no mass transfer limitations =>CAb=CAs=CA)

rA  1 kn CAs
n
 rA  kn CAs
n

 If internal mass transfer limit

(no external limitation CAb=CAs≠CA)

rA  η(  rAs )  ηkn CAs

n

 Effectiveness factor for nth order reaction with large Thiele modulus

3 2
rA  n
kn CAs
n n 1
45
Catalytic
Reaction Engineering

Yongdan Li
Professor of Industrial Chemistry
Department of Chemical and Metallurgical Engineering
School of Chemical Technology
Aalto University
Email: [email protected] Nov-Dec, 2018
Kemistintie 1, E404 46