2 Prediction of Diffusivities PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 26

PREDICTION OF DIFFUSIVITIES

1) Diffusion in gases
Diffusivities are best determined by experimental
measurements.
The diffusivity of some common gases diffusing in
air at 0C and 1 atm (can be obtained from the text
book (McCabe, Smith and Harriott, Appendix 19).
The value of diffusivities can also be predicted using
various empirical equations.

EKC 217: Mass Transfer - Prediction of Diffusivities

The equations developed by Chen and Othmer is:


0.5

1
1

0.01498 T

MA MB

P(TcATcB ) 0.1405(VcA0.4 VcB0.4 ) 2


1.81

DAB

------ (1.50)

where:

DAB = Diffusivity (cm2/s)


T
= Temperature (K)
MA, Mb
= Molecular weights of component A & B
TcA, TcB
VcA, VcB

respectively
= Critical temperatures of A & B respectively (K)
= Critical molar volumes of A & B respectively
(cm3/gmol)
= Pressure (atm)

EKC 217: Mass Transfer - Prediction of Diffusivities

Another rigorous equation called Chapman-Enskog equation is


used to predict diffusivities:

DAB

0.001858 T

2
P AB
D

1.5

where:

DAB
T
MA, MB
P
AB
D
k

MA MB

M AM B

0.5

------ (1.51)

= Diffusivity (cm2/s)
= Temperature (K)
= Molecular weights of component A & respectively
= Pressure (atm)
= average collision diameter
= collision integral based on the Lennard-Jones potential
= f (kT/AB )
= Boltzmanns constant
= Lennard-Jones force constant for common gases

AB A B
EKC 217: Mass Transfer - Prediction of Diffusivities

The collision integral, D decreases with increasing T, that makes


DAB increase with more than the 1.5 power of the absolute T.
For diffusion in air at T = 300 1000 K, DAB varies with about T 1.71.8, and T 1.75 can be used to extrapolate from room temperature
data.
Therefore, we can say that:

DAB T 1.75 x 1/P

Hence, the diffusivity at T2 and P2 relative to the standard


temperature and pressure (STP) can be calculated as follows:
1.75

DAB , 2

PSTP T2

DAB , STP
P2 TSTP

EKC 217: Mass Transfer - Prediction of Diffusivities

------ (1.52)

Eq. (1.51) is relatively complicated to use, and often some of the


constants such as AB are not available of difficult to estimate.
Hence, semi empirical method by Fuller et al. is often preferred:

DAB

1.00 x 10 7 T 1.75 (1 / M A 1 / M B ) 0.5

P A

1/ 3

1/ 3 2

------ (1.53)

where: A = sum of structural volume increments


Eq. (1.53) was obtained by correlating many recent data and uses
atomic values from Table 6.2-2. This method can be used for
mixtures of nonpolar gases or for a polar-nonpolar mixture.

EKC 217: Mass Transfer - Prediction of Diffusivities

EKC 217: Mass Transfer - Prediction of Diffusivities

Example 5:
Predict the volumetric diffusivity for benzene in air at 100C and 2
atm by using the rigorous equation (1.51) and by extrapolating from
the published value for 0C and 1 atm.
Solution:
From Appendix 19 (McCabe, Smith and Harriott), the force
constants are as follows:

/k

Benzene

412.3

5.349

78.1

Air

78.6

3.711

29

EKC 217: Mass Transfer - Prediction of Diffusivities

Thus,

AB

5.349 3.711

4.53
2

AB / k 412.3 x 78.6 180


kT

AB

373

2.072
180

From Appendix 19, D = 1.062.


Substitute values into Eq. (1.51) gives:

DAB

78.1 29

0.001858 x 3731.5
78.1 x 29

2 x 4.532 x 1.062

EKC 217: Mass Transfer - Prediction of Diffusivities

0.5

0.0668 cm 2 /s

From Appendix 18 at standard temperature and pressure,


DAB 0.299 ft 2 /h 0.0772 cm 2 /s

At 373 K and 2 atm,


1.75

1 373
DAB 0.0772

2 273

0.0666 cm 2 /s

Therefore, agreement with the value calculated from Eq. (1.51)


is very good.

EKC 217: Mass Transfer - Prediction of Diffusivities

Example 6:
In an oxygen-nitrogen gas mixture at 1 std atm, 25C, the
concentration of oxygen at two planes 2 mm apart are 10 and
20 vol %, respectively. Calculate the flux of diffusion of the
oxygen for the case where:

a) The nitrogen is non-diffusing


b) There is equimolecular counterdiffusion of the two gases.

DO2-N2 at 0C and standard atmospheric pressure


= 1.81 x 10-5 m2/s

EKC 217: Mass Transfer - Prediction of Diffusivities

10

Solution:

Given:

PT = 1 atm = 1.0133 x 105 N/m2


T = 25C = 298 K
DO2-N2 = 1.81 x 10-5 m2/s (0 C , 1 atm)
Diffusivity of O2-N2 mixture at 25C , 1
atm:

A+B

A
NB = 0

pA1
= 0.2PT

pA2
= 0.1PT

2 mm

1.5

298
DO 2 N 2, 298K 1.81 x 10 x

273
2.06 x 10-5 m 2 /s
5

EKC 217: Mass Transfer - Prediction of Diffusivities

Component A = O2
Component B = N2

11

pA1 = 0.2 PT
pA2 = 0.1 PT
Then, pB1 = PT 0.2PT = 0.8PT
pB2 = PT 0.1PT = 0.9PT
a) For the case where nitrogen is non-diffusing, Eq. (1.37) is used:

DAB PT
NA
( p A1 p A2 )
RT ( z2 z1 ) pBM

and

pBM

pB 2 pB1

ln( pB 2 / pB1 )

Calculate pBM :

pBM

(0.9 0.8) PT

0.849 PT
ln(0.9 PT / 0.8PT )

EKC 217: Mass Transfer - Prediction of Diffusivities

12

PT
PT

1.178
pBM 0.849 PT

Substitute the known values into Eq. (1.37):

N O2

DAB PT

( p A1 p A2 )
RT ( z2 z1 ) pBM
2
m
2.06 x 10-5
1.178
5 N
s

0.2 0.1 x 1.0133 x 10


2
Nm
m
(8314
)(298 K)(0.002m)
kmole K
-5 kmole
4.95 x 10
m2 s

EKC 217: Mass Transfer - Prediction of Diffusivities

13

b) When there is equimolecular counter diffusion of the two gases,


the following equations are applied:

DAB p A1 p A2
JA
RT ( z2 z1 )

DAB pB1 pB 2
JB
RT ( z2 z1 )

Substitute the known values:


2
m
5
5 N
0.2 0.1 x 1.0133 x 10 2
2.06 x 10
s
m
J O2
Nm
8314
(298 K )(2 x 10 3 m)
kmole K
-5 kmole O 2
4.21 x 10
m2 s
EKC 217: Mass Transfer - Prediction of Diffusivities

14

For nitrogen:
2

m
N
5
2.06 x 10
0.8 0.9 x 1.0133 x 105
s
m

J N2
Nm
8314
(298 K)(2 x 10 3 m)
kmole K
- 4.21 x 10-5 kmole N 2 /m 2 s

EKC 217: Mass Transfer - Prediction of Diffusivities

15

Exercise: Estimation of Diffusivity of a Gas Mixture


Normal butanol (A) is diffusing through air (B) at 1 atm abs. Using
the Fuller et al. method, estimate the diffusivity DAB for the
following temperatures and compare with the experimental data:
a) For 0C
b) For 25.9C
c) For 0C and 2 atm abs.

Ans:
a) Deviation of +10% from the experimental value (in Table 6.2-1).
b) Deviation of +4% from the experimental value.
c) DAB = 3.865 x 10-6 m2/s

EKC 217: Mass Transfer - Prediction of Diffusivities

16

2) Diffusion in Small Pores


The diffusion process is called Knudsen diffusion.
Occurs when the pore size is much smaller than the normal mean free
path during gas diffusion in very small pores of a solid, for processes
like adsorption, drying of porous solids or membrane separation.
The diffusivity for a cylindrical pore is:
DK 9700r

where:

T
M

------ (1.54)

DK = Knudsen diffusivity, cm2/s


T = temperature, K
M = molecular weight
r = pore radius, cm

Diffusivity in the pore:


EKC 217: Mass Transfer - Prediction of Diffusivities

1
1
1

D pore DAB DK
17

------ (1.55)

3) Diffusion in Liquids
A widely used correlation for liquid diffusivity of small molecules,
called Wilke-Chang equation:

DAB 1.173 x 10
where:

DAB
T
B
VA

MB

0.5
(

M
)
T
16
B

------ (1.56)

B V A

0.6

= diffusivity, m2/s
= absolute temperature, K
= viscosity of B in Pa.s
= molar volume of solute as liquid at its normal
boiling point, m3/kgmole (taken from Table 6.3-2)
= association parameter of the solvent
= molecular weight of solvent B (kg/kgmole)

EKC 217: Mass Transfer - Prediction of Diffusivities

18

Recommended values for :

Water

2.6

Methanol
Ethanol

1.9
1.5

Benzene, heptane and other unassociated solvents

1.0

Eq. (1.56) is valid only at low solute concentrations and does not
apply when the solution has been thickened by addition of highmolecular-weight polymers.

EKC 217: Mass Transfer - Prediction of Diffusivities

19

EKC 217: Mass Transfer - Prediction of Diffusivities

20

For water as solute, VA = 0.0756 m3/kgmole.


If other compound is the solute, e.g.: acetone (CH3COCH3) in
water, the value of VA can be calculated by considering the atomic
volumes as given in Table 6.3.2 Atomic and Molar Volumes at the
Normal Boiling Point.
Acetone (CH3COCH3) has 3 carbons, 6 hydrogens and 1 oxygen.
Therefore,

VA = 3(14.8 x 10-3) + 6(3.7 x 10-3) + 1(7.4 x 10-3)


= 0.0740 m3/kgmole
EKC 217: Mass Transfer - Prediction of Diffusivities

21

EKC 217: Mass Transfer - Prediction of Diffusivities

22

Example 7:
Predict the diffusion coefficient of acetone (CH3COCH3) in water at
25C and 50C using the Wilke-Chang equation. The experimental
value is 1.28 x 10-9 m2/s at 25C (298 K).
Solution:
From Appendix 9 (McCabe and Thiele textbook), the viscosity of
water at 25C is 0.90x 10-3 Pa.s and at 50C, 0.55 x 10-3 Pa.s. From
Table 6.3-2, for CH3COCH3 with 3 carbons + 6 hydrogens + 1
oxygen,

VA = 3(0.0148) + 6(0.0037) + 1(0.0074) = 0.074 m3/kgmole


EKC 217: Mass Transfer - Prediction of Diffusivities

23

For water, the association parameter () is 2.6 and MB = 18.02 kg


mass/kgmole.
For 25C, T = 298 K, substituting known values into Eq. (1.56):

DAB 1.173 x 10

0.5
(

M
)
T
16
B

B V A

0.6

(1.173 x 10-16 )(2.6 x 18.02)1/2 (298)

(0.9 x 10-3 )(0.0740) 0.6


1.277 x 10-9 m 2 /s

EKC 217: Mass Transfer - Prediction of Diffusivities

24

For 50C, T = 323 K:

DAB 1.173 x 10

0.5
(

M
)
T
16
B

B V A

0.6

(1.173 x 10-16 )(2.6 x 18.02)1/2 (323)

(0.55 x 10-3 )(0.0740)0.6


2.25 x 10-9 m 2 /s
Therefore, the predicted diffusivity is very close to the experimental
value.

EKC 217: Mass Transfer - Prediction of Diffusivities

25

Schmidt number
26

The ratio of the kinematic viscosity to the molecular diffusivity is


known as Schmidt number, Sc.
Sc

------ (1.57)

Similar to Prandtl number:

Pr

cp

------ (1.58)

Sc for gases in air at 0C and 1 atm is given in Appendix 18, range


from 0.5 2.0.
Sc for liquids range from 102 to 105 for typical mixtures.
EKC 217: Mass Transfer - Prediction of Diffusivities

You might also like