Abstract

The main aim of this experiment was to find the diffusivity of acetone in air at 50°C and 1atm. In
order to find the diffusivity, firstly, the rate of decrease in the level of acetone in the capillary
tube for every 30 minutes was noted. An air pump was used to supply air to make sure that the
partial pressure of acetone is zero at the mouth of the capillary tube. The readings were helpful in
plotting a graph and the slope of the graph was substituted into an equation derived from Fick’s
first law. The equation was under the Stefan Winkelmann’s method for calculating the
diffusivity. Finally, using this method, the experimental diffusivity was calculated to be

−6 m2
3.889 ×10 . This value is around 68% deviant from the actual value for diffusivity that was
s
calculated theoretically using the Fuller correlation. Another experiment was done, but with 20-

m2
minute intervals, and the diffusivity was found to be 2.077 ×10−5 . This value is 70% deviant
s
from the true value.

Introduction

If a fluid, with two or more components, has a concentration gradient, then there is a net
movement of mass of the fluid in the direction which reduces the concentration gradient. This is
termed as mass transfer and it occurs in gases, liquids or in both phases at the same time. If a
liquid is evaporated in a narrow and vertical tube at constant temperature, there is a transfer of
mass from the surface of the fluid, only by molecular diffusion. The diffusivity of a vapour arose
from its liquid can be found by using the Stefan-Winkelmann method. It states that, by
controlling the rate of fall of liquid, the evaporation rate, and with the concentration gradient, we
can calculate the value for diffusivity (Coca, Bueno, & Alvarez, 1980).

In this experiment, the value for the diffusivity was determined using a volatile liquid, acetone,
as the evaporation source. The acetone was evaporated in a vertical glass tube and the top of the
tube had an opening through which a vapour free gas was passed. To make sure there were no
eddy currents in the tube and a constant temperature of 50°C, a water bath was included below
the narrow tube. In addition, eddy currents will affect the diffusion rate and this results errors in

1
the calculated diffusivity and the mass transfer would no longer be just by molecular diffusion.
The evaporation rate can be found from the rate of fall of liquid. A Vernier callipers was used to
take the readings of the liquid level for every 30 minutes. Using the concentration gradient value,
and the readings of the liquid level, we calculated the diffusivity of acetone vapour using the
Stefan-Winkelmann method.

The rate of mass transfer can be calculated by the below equation

'
N A =D ( CL )( CC )
A T

Bm
(Eq. 0)

By considering the evaporation of acetone, we have,

N 'A = ( Mρ ) dLdt
L
(Eq. 0)

Combining Eq. 1 and Eq. 2,

CA CT
( Mρ ) dLdt =D ( L )( C ) Bm
(Eq. 0)

After integrating, with boundary conditions L = L0 at t = 0, and rearranging we get,

ρL
t
=
L+ L 0 2 MD ( )( CC C ) ( L−L )
A
Bm

T
0 (Eq. 0)

A graph of t /(L+ L0 ) against ( L−L0 ) will give the gradient as s,

ρL C Bm
s= ( 2 MD )( CA CT ) (Eq. 0)

Diffusivity D, can then be calculated from Eq. 5

2
 ρ L C Bm
D= (Eq. 0)
2 M C A CT s

Objectives

1. To calculate the vapor diffusion coefficient of acetone in air at 50 °C using the Stefan-
Winkelmann method.
2. To compare the diffusivity values obtained with different time intervals.

Procedures

Start-up procedures

Firstly, a water bath was filled with clean and filtered water to nearly 20mm from the top.
Secondly, the main power on the panel was switched on. Next, the set-value on the digital
temperature controller was set to 50°C. Lastly, the heater was turned on and the water was heated
to 50°C.

Priming procedure: Capillary tube

A pipette was used to fill in the capillary tube with acetone until the level of acetone in the tube
falls just below the surface of water in the water bath. The tube was filled with acetone after
cleaning it with a weak solution of detergent for the first time, and with acetone for the second
and third times.

Shut-down procedures

Firstly, the heater was switched off along with the air pump. Then, the set-point value on the
temperature controller was adjusted to around 22°C. The water was allowed to cool down until a
point where it was safe to touch.

Experimental procedures

1. The start-up and the priming procedures were performed.

3
2. The capillary tube was filled with acetone to a depth of around 25mm. To make sure it was
the acetone level was below that of water, it was first immersed in the water bath and seen
through a telescope as a trial.
3. The capillary tube was inserted from the top of the fitting and was tightened.
4. The initial level of the acetone was observed through the telescope.
5. The air pump was connected to one end of the capillary tube and the other end was left open
to atmosphere.
6. After half-an-hour, the air pump was switched, and the flexible tubing was disconnected. The
capillary tube ends were closed with the caps.
7. The level of acetone was observed using a telescope and Vernier callipers was used to find
the level of acetone.
8. Steps 5-8 were repeated for every 30 minutes.

Results

Experiment 1: Data collected for every 30 minutes

Table 1: Experimental data for evaporation rate of acetone taken for every 30 minutes

Time, t Level of Acetone, L+L0, (mm) t/L+L0 (ks/mm)

(ks) L (mm)
0.0 10.9 21.8 0
1.8 12.1 23 0.0783
3.6 13.5 25.6 0.1406
5.4 15.4 28.9 0.1869
7.2 17.4 32.8 0.2195
9.0 19.1 36.5 0.2466

4
 Plot of t/(L+L0) against (L-L0)

0.3

0.25 f(x) = 0.03 x + 0.04

T/(L+L0) (in ks/mm)

0.2

0.15

0.1

0.05

0
0 1 2 3 4 5 6 7 8 9
(L-L0) (in mm)

Figure 1: Graph of t/(L+L0) against (L-L0)

Sample Calculations

The slope was the graph (Figure 1) was found to be 0.0283 ks/mm2.

This value is converted to s/m2 using the below conversion

ks 1000 s 10 6 m m 2 6 s
s=0.0283 2
× × 2
=28.3 ×10 2
mm 1 ks 1m m

Using the Antoine equation, along with the constants given, we can get vapor pressure of acetone

A 1161
B− 7.02447−
¿ C +T 224+50
P =10 =10 =612.694 mmHg

Using Raoult’s law, vapor mole fraction of acetone can be found

P¿ 612.694
y A= = =0.806
P 760

Assuming an ideal gas law to be valid, we have,

n P 101325 mol
C T= = = =37.714 3
V RT 8.314 ×(50+ 273.15) m

5
Log mean molecular concentration is calculated below.

CB1 = CT

Patm −P¿ 101325−81685.82 mol

C B 2=C T
( P atm )
=37.714 × (
101.325
=7.3098 3
m )
(C B 1 −C B 2) 37.714−7.3098 mol
C Bm= = =18.53 3
C 37.714 m
ln B 1 ( )
CB 2
ln
7.3098( )
Concentration at the junction

P¿ 81685.82 mol
C A=
( ) C =37.714 ×
P atm T 101325 (
=¿30.404 3
m )
Given, MA and ρ A , and with all the above calculated variables, we can calculate diffusivity using
Eq. 6

ρL C Bm 790 ×18.53
D AB= =
2 M C A CT s 2 ×0.058 ×30.404 × 37.714 ×28.3 ×106

−6 m2
D AB=3.889 ×10
s

Experiment 2: Data collected for every 20 minutes

Table 2: Experimental data for evaporation rate of acetone taken for every 20 minutes

Time, t Level of Acetone, L+L0, (mm) t/L+L0 (ks/mm)

(ks) L (mm)
0.0 16.35 32.7 0
1.2 16.75 33.1 0.036
2.4 18.00 34.35 0.070
3.6 19.00 35.35 0.102
4.8 20.20 36.65 0.131

6
 6.0 21.00 37.35 0.161

Plot of t/(L+L0) against (L-L0)

0.18
0.16
0.14
T/(L+L0) (in ks/mm)

0.12
0.1
0.08 f(x) = 0.01 x + 0.08
0.06
0.04
0.02
0
0 0.5 1 1.5 2 2.5 3 3.5
(L-L0) (in mm)

Figure 2: Graph of t/(L+L0) against (L-L0)

The calculations in Experiment 1 were repeated for Experiment 2. Using the same calculation
methods, the calculated DAB for acetone in air, for every 20-minute intervals, is shown below.

s
s=5.3× 106
m2

−5 m2
D AB=2.077 × 10
s

Discussion

Using the fuller correlation (Welty, Rorrer, & Foster, 2019), the diffusivity, at 50 °C, of acetone

m2
in air was calculated to be (1.219 ± 0.1)×10−5 . The table below (table 3) shows the
s
percentage errors for both the experiments.

Table 3: Percentage errors in the experimental diffusivity.

Experiment 1 (30-minute intervals) Experiment 2 (20-minute intervals)

7
 (3.889 ×10−6)−(1.219 ×10−5) (2.077 × 10−5 )−(1.219 ×10−5)
| (1.219× 10−5 ) |
×100 %
| (1.219 ×10−5 ) |
×100 %

Percentage error is 68.1% Percentage error is 70.39%

The percentage errors for both the experiments are significantly high and unexpected. The
parallax error might be involved which is associated with the error when taking the reading of
the liquid acetone level. Moreover, parallax error occurs when the liquid acetone meniscus level
in the capillary tube is not parallel with the reference line of the telescope. The readings might be
further affected with the errors from the Vernier callipers. The value from the Vernier scale
could be taken wrongly which can result in 0.05mm-0.5mm difference from the actual value. A
possible alternative to taking manual readings would be to use a data logging system and let the
computer take the readings. This reduces the human error. Moreover, at the tip of the capillary
tube, the mole fraction of acetone is assumed to be zero (or the partial pressure of acetone is
assumed to be zero) which results in some error as well because the vapours of acetone might not
be fully blown away by the air that is pumped into the capillary tube.

For Experiment 2, it is worth be noted that, although the diffusivity value is 70% deviant, none
of the points are near to the trend line of the graph while for Experiment 1, the most points are
either near to or touch the trend line of the graph. This tells that taking readings for longer times
gives closer and better points which eventually lead to a more accurate experimental diffusivity.
Another possible reason for a large error could be convective eddy currents in the capillary tube
and hence the diffusivity is affected.

Not following proper priming procedures, like not cleaning the capillary tube and the pipette
from previous use, might result in a error in the diffusion coefficient of acetone because the
presence of other liquids in the capillary tube will affect the evaporation rate of the acetone in the
capillary tube. Moreover, small air bubbles might be trapped at the mouth of the capillary tube
before the experiment was started. This would again affect the diffusion coefficient. To minimize
this, a porous surface can be used at the top of the capillary tube for a free flow of solute and
solvent.

8
An important thing to note is that even the theoretical value is not 100% accurate and it does
involve an error of around 10-20%. The fuller correlation might give a different diffusivity value
compared to the Hirshfelder equation.

Conclusion

To sum up, in this experiment, the diffusion coefficient of acetone in air at 50°C was found using
the Stefan-Winkelmann method. The experimental diffusivity was calculated to be

−6 m2
3.889 ×10 for 30-minute intervals. This value has 68% error for diffusivity compared to the
s
theoretically calculated diffusivity using the Fuller correlation. Another experiment was done,

m2
but with 20-minute intervals, and the diffusivity was found to be 2.077 ×10−5 . This value is
s
70% deviant from the true value. It is important to note as to what led to a high percentage
errors. The different reasons involve the human errors when taking readings of the evaporation
rates, convective eddy currents and the parallax errors. Overall, the experiment did give helpful
results to get a diffusivity value that is less than 100% deviant from the true value.

References

1. Coca, J., Bueno, J. L., & Alvarez, R. (1980). Gaseous Diffusion Coefficients by the Stefan-
Winkelmann Method Using a Polymer-Solvent Mixture as Evaporation Source. Industrial &
Engineering Chemistry Fundamentals, 19(2), 219–221.
2. Welty, J. R., Rorrer, G. L., & Foster, D. G. (2019). Fundamentals of momentum, heat, and
mass transfer. Hoboken, NJ: Wiley.

9
Appendix

10
11