We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 15
VOLUME 94,
NUMBER 3 MAY 1, 1954
A Model for Collision Processes in Gases. I. Small Amplitude Processes in Charged
and Neutral One-Component Systems*
P. L. Buamuaoans E. P. Gross,t ax M. Keoox
Astrowmy Department, Hareord Unisersty, Combidge, Massachusds ond Laboratory for Insulation Research,
‘Massachusetts Intute of Fcinolgy, Cambridge, Massechuats
(Received November 27,1953)
‘A kinetic theory approach to cllsion processes in ionized and
‘neural gases is presented. This approach is adequate for the uni
fed treatment of the dynamic properties of gases over a continuous
range of pressures from the Knudsen Init to the high-pressure
Tint where the aerodynamic equations are valid. Itisalso possible
to satisfy the correct. microscopic boundary” conditions. The
‘method consists in altering the eolison terms in the Boltzmann
equation. The modified colison terms are constructed. so that
‘ac calisin conserves partiele number, moment, and eneTgy
bother characteristics such as persistence of veloities and angular
Gependence may be included. ‘The present article lustrates the
technique for a simple model involving the seeumption of =
Callison time independent of velocity this model i applied to
the study of small amplitude ocilations of one-component ionized
and neutral gases. The initial value problem fo unbounded space
{s solved by performing a Fourier transformation on the spuce
4, INTRODUCTION
pi nea sats oak
Sten at qa eaeanas
sre fe tae
tribution function f(v,r,!) is defined by the condition
Snr eia tet tarn i
Pade iinet he
Taal gi a teal
sniaat tea I fae
ay coins get
atte he eat
ze let vey
Se EAS CS ane,
SHRSElaprnta tnd G. Coming, The Molomai! Thay of
owe Gases (Cbg Utara Pre Leon, 82),
Bd en Comber
variables and a Laplace transformation on the time vaviable.
For uncharged gases there results the correct adiabatic limiting
law for sound- ave propagation at high pressures and in adlton,
fone obtains a theory of absorption and dispersion of sound for
arbitrary pressures. For ionized gases the dference in the nature
ofthe organization in the low-pressure plasma oscillations and in
high-pressure sound-type osllations is studied. Two important
‘asta are dlstingwlahed. Ifthe wavelengths ofthe onilatons are
Jong compared to either the Debye length or the mean fee path,
ll change in frequency is obtained asthe ealson frequency
varies from aero to infinity. The accompanying absorption is
tlt reaches its maximm value wen the callisfon frequency
fsquals the plasma frequency. The second case refers to waves
Shorter than both the Debye length and the mean fee path; these
waves are characterized by a very heavy absorption,
which the macroscopic theory does not give a correct
description, During recent years interest in such prob-
Jems has increased considerably. An extreme example
of the breakdown of the aerodynamic theory, vi. in
the Knudsen region, has been known for some time;
here the mean free path of a molecule is large compared
to some linear dimension of the apparatus, and behavior
at the boundaries becomes important. Other examples
are provided by high-frequency sound waves ina
rarefied gas and by plasma oscillations.
‘The solution of the Boltzmann equation is, in general,
‘4 matter of considerable difficulty even in cases corre-
sponding to the physically simplest situations. Signifi
‘ant progress has been confined practically to the study
of two limiting cases in which two different approxima-
tion procedures can be applied. A criterion for the
range of validity of the approximate methods is pro-
vided by the comparison of some characteristic time
1 (or characteristic length 7.) for the relevant process
with the average time 7. (or mean free path L,) be-
tween molecular collisions.
For high density (7>re or 12>L.) the Enskog-
‘Chapman (E-C) theory may be used. The frst approxi-
mation of the theory consists in assuming local thermo-
dynamic equilibrium and a common drift velocity for
all molecular species. The next approximation corrects
the distribution function by terms proportional to the
first derivatives of temperature, velocity, and density;
this corresponds to the aerodynamic equations with
coeficients of heat conduction, viscosity, and diffusion,
‘The high-density region (72>r.) isin fact the range in
which the aerodynamic equations provide an adequate
description. Higher approximations of the E-C theory
lead to correction terms proportional to higher deriva-
sit512
tives of 7, q, p. The successive approximations of the
E-C theory correspond to an expansion of the distribu-
tion function in powers of the mean free path L.. If we
consider sound waves with wavelength Z>>L the first
and second approximations are already sufficient 10
sive all significant features of the process, However,
‘when L becomes comparable with Z,, it is necessary to
go to the third and even higher approximations to
obtain adequate results; as discussed by Herzfeld?
important dynamic contributions appear only in ap-
proximations higher than the second. The third approxi-
mation already involves formidable labor and has been
used to solve only the simplest problems. Consideration
of higher approximations is, in any case, of doubtful
value as the entire procedure breaks down in just the
range where the contributions from these higher-order
terms become important. In addition, the boundary
conditions in many problems cannot be specified
properly within the scope of the E-C formalism. A
different approach, using expansions in terms of
Hermite polynomials in velocity space, has been given
by Grad. He uses some moments of low order in addi
tion to the usual ones representing p,q, and 7. The
procedure involves a gain in simplicity over the E-C
theory but is still quite complicated. Tt is basically a
high-density theory and is capable of dealing only with
boundary conditions which can be specified in terms of
the moments appearing in the theory. The limitations
of any theory based on the use of a finite number of
‘moments will be brought out in the discussions of this
series of papers.
‘The opposite limiting case (rr, or L