In the Laboratory
Rare Gas Viscosities: A Learning Tool
Arthur M. Halpern
Department of Chemistry, Indiana State University, Terre Haute, IN 47809; a-halpern@indstate.edu
Viscosity is one of the main topics that students en-
counter in the treatment of transport processes in physical
chemistry. Most texts treat this topic beginning with Newtons
law of viscosity, which is the basis of the laminar flow model
involving the transport of momentum across a lateral velocity
gradient. The application of gas kinetic theory to Newtons
law allows one to express the viscosity, , of a hard-sphere
(i.e., semiperfect) gas as
= k1 c (1)
where is the density, c is the mean speed, is the mean free
path, and k1 is a constant. For rigid spheres, k1 is assigned
the value of 0.499 (1). (Levine presents a value of 5/32, i.e.,
0.4909 [2, p 462].) If we express using the perfect gas law,
PM/RT, and replace c and by their respective gas kinetic
expressions (2, pp 432, 443), namely
1/2
c = 8RT
M time-dependent pressure profile of a gas being pumped from a
and
= 1 RT
2
1/2 2 PN A
we get the following rather simple equation for the viscosity
of the semiperfect, elastic-sphere gas
MT 1/2
= k2
2
where M is the molar mass, T is the absolute temperature,
and the (spherical) molecular diameter. The cgs value of k2
is 2.715 1021 (and thus is obtained in poise, i.e., g cm1 s1).
Students are alerted to two presumably surprising predictions
that stem from eq 4, namely, that because the P dependence
of and cancel each other, the gas viscosity should be
pressure independent, and it should increase with increasing
temperature (as T 1/2). Most students find the first prediction
counterintuitive and the second contrary to their experience
with most liquids. The experimentally observed near-P-
capillary tube
to
vacuum
pump
gas
ballast inlet
bulb
Figure 1. Diagram of the apparatus. In our apparatus the ballast
bulb is 1 L; the capillary tube has a radius of 0.06 cm and a length
of 40 cm.
214 Journal of Chemical Education Vol. 79 No. 2 February 2002 JChemEd.chem.wisc.edu
independence of and its T dependence provided early and
strong support for the validity of Maxwells kinetic theory.
Another interesting consequence of the approximate P-indepen-
dence of is that the free-fall velocity of an object in a gas,
for example air, should be pressure invariant. An observation
related to this phenomenon, the independence of the resistance
of a swinging pendulum to the air density, was reported
by Boyle in 1660about 200 years before development of
kinetic theory.
Another interesting aspect of eq 4 is that it allows a
microscopic molecular quantity, , to be determined from the
measurement of a bulk macroscopic property, gas viscosity.
Indeed, many texts present problems in which the student
finds from the viscosity of a particular gas.
In our physical chemistry laboratory, students determine
the viscosities of a series of gases (viz. the rare gases He, Ne, Ar,
Kr, and Xe, as well as SF6, an honorary rare gas) using the
(2) evacuation or transpiration method, in which they obtain the
fixed volume through a capillary tube at constant temperature
(3). A diagram of the apparatus is shown in Figure 1.
This method, a direct application of Poiseuilles law, in
(3)
which laminar flow is assumed, is capable of providing gas
viscosities with impressively high precision (e.g. relative
standard deviation / 103). The time dependence of the
pressure of a gas (having a viscosity ) as it is being evacuated
from a system of fixed volume V and temperature T through
a capillary tube of radius r and length L is predicted (3) to be
(4)
P0
P= (5)
1 + k P0 t
where P0 is the initial pressure and k is a constant with
r4
k= = k
16VL
(6)
Students determine k by calibrating the system1 using (dry)
air of known viscosity, for example 182.8 P at 20 C (4). A
plot of P versus t for Ar with the regression fit to eq 5 is
shown in Figure 2.
It is important to perform these experiments under con-
ditions for each gas in which laminar flow is expected to take
place and hence eq 1 to be valid. At high pressures nonlaminar
flow occurs, and at low pressures molecular flow (non-bulk-
transport) proceeds. The onset of the high-pressure regime
is given by (3)
1/2
P high > 8000RTL (7)
M r3
and the low pressure regime by
10 1/2
P low < r RT (8)
M
 In the Laboratory

180
Table 1. Pressures at Which Nonlaminar (Phigh) and
160
Molecular (Plow) Flow Become Significant
Gas M Phigh/torr Plow/torr
140
He 4.003 388 1.9
P / torr

120
Ne 20.179 276 1.4
100 Ar 39.948 140 0.7

80 Kr 83.80 108 0.5

Xe 131.29 79 0.4
60
SF6 146.054 48 0.2
40
0 20 40 60 80 100 120 140 160 180

t/s
Figure 2. Plot of P versus t for evacuation of argon from the appa- Table 2. Experimental, Calculated, and Literature Values
ratus at 21.0 C. () experimental data; (- - -) fit to eq 5. for Viscosity and Collision Diameter
exptl / lit
Gas /Pa visc /b calcd /c
P Value/P Ref
350
He 182 194.1 4 0.1 2.172 2.674
196.4 5
Ne 298 311.1 4 0.1 2.571 3.074
300 313.9 5
Viscosity / P

Ar 223 221.7 4 0.2 3.612 3.974

222.8 5
250 Kr 247 250.3 4 0.2 4.092 4.249
249.7 5
Xe 229 226.0 4 0.3 4.817 4.662
200 226.5 5
SF6 156 148.3 6 0.4 6.107 5.770
NOTE: Experimental viscosities were obtained in the physical chemistry
150
1 2 3 4 5 laboratory experiment and corrected to 20.0 C assuming T 1/2-dependence
(eq 4).
Row Number aStandard deviations of viscosities reported here.

Figure 3. Plot of versus row number for the rare gases: () data bValues from ref 4 and eq 4.

from physical chemistry lab; () data from ref 4. cValues from Gaussian 98W and eq 9.

Note that Phigh and Plow depend on the viscosity and molar
mass of the gas. To make sure one is obtaining P(t) profiles
in the laminar flow regime, one must know the gas viscosity
beforehand. Without knowing this quantity one can perform a
preliminary determination of and use that value to determine
the appropriate pressure range for a more accurate measurement.
Table 1 contains Phigh and Plow values for the gases studied in
this experiment. Notice the limited pressure ranges for the
heavier gases, Xe and SF6.
A comparison of the gas viscosities, obtained using this
technique, with literature values is contained in Table 2. When
students examine their viscosity values of the rare gases they
intuitively expect to observe a trend, yet a plot of versus
row number does not depict smooth, monotonic behavior
(Fig. 3). After being reminded to consider eq 4, students
understand that is expected to vary as M 1/2/ 2.They then
realize that since both M and increase with increasing row
number (but at different rates), might not necessarily be
expected toand indeed, does notchange monotonically
with row number (or atomic mass).
Students obtain values of the collision diameters for the
gases using their viscosity values and eq 4 and then see that these
quantities do, in fact, increase monotonically with increasing row
number, as expected. This exercise is an opportune time to
remind students that atomic size is not a definitive quantity:
its value depends on the method used to obtain it (e.g., from
gas viscosity, thermal conductivity, liquid density, crystal
structure, bond lengths in the case of diatomic molecules, etc.).
The question then arises of how reasonable the rare
gas atomic diameters they obtain from viscosity measurements
arethat is, how they compare with other determinations
of atomic size.
One interesting basis for comparison is easily provided by
computational methods, which are now accessible to student
practitioners. Thus, they can obtain the values of the rare
gases and SF6 by using the VOLUME keyword in Gaussian
98W (7). This utility obtains the volume from a Monte Carlo
integration of the electron density within a contour cutoff
of 0.001 electrons/bohr3 (0.007 electrons/3). Because this is a
statistical calculation, we chose to obtain higher precision by
increasing the number of points sampled from the density
matrix (i.e., we employed 5 105 points instead of the default
value of 20 by adding iop(6/45=500000) to the Route Section
in the Gaussian 98W job). In these calculations, we used the
MP2 method (to account for some degree of electron corre-
lation) and the standard SDD (6D, 10F) basis set (employing
electrostatic core potentials) because we span the rare gases down
to Xe. This calculation gives the molar volume, Vm. The

JChemEd.chem.wisc.edu Vol. 79 No. 2 February 2002 Journal of Chemical Education 215

 In the Laboratory

respective diameter, calcd, was obtained from the relation 5

1/3
calcd = 6V
(9)
N A
4

The results of these calculations are summarized in Table

/
2 and shown in Figure 4, in which we plot the diameters
obtained from the viscosities, visc, versus row number. The 3
agreement between these values for the five rare gases and
SF6 is reasonably good, considering their disparate origins and
the different approximations that underlie each method.
2

Note
1. In our apparatus, r 0.06 cm, L 40 cm, and V 1.4 L.
Figure 4. Plot of versus row number: () from ref 4 and eq 4; ()
from Gaussian 98W and eq 9.
Literature Cited
1. Present, R. D. Kinetic Theory of Gases; McGraw-Hill: New
York, 1958; pp 221222.
2. Levine, I. N. Physical Chemistry, 4th ed.; McGraw-Hill: New
York, 1995.
3. Halpern, A. M. Experimental Physical Chemistry. A Laboratory
Textbook, 2nd ed.; Prentice Hall: Upper Saddle River, NJ,
1997; pp 331342.
4. Handbook of Chemistry and Physics, 1st student ed.; CRC Press:
Boca Raton, FL, 1988; pp F-24F-26.
5. Kestin, J.; Khalifa, H. E.; Wakeham, W. A. J. Chem. Phys.
1977, 67, 42544259.
6. Kiyokatsu, U.; Kogoshi, K. J. Inorg. Nucl Chem. 1974, 989992.
7. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Mont-
1 2 3 4 5
Row Number
gomery, J. A. Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.;
Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.;
Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.;
Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.;
Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.;
Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.;
Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.;
Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.;
Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98,
Revision A.9; Gaussian, Inc.: Pittsburgh PA, 1998.

216 Journal of Chemical Education Vol. 79 No. 2 February 2002 JChemEd.chem.wisc.edu