Anda di halaman 1dari 13

On the Maxwell-Stefan approach to

multicomponent diusion
Dieter Bothe
Center of Smart Interfaces
Technical University of Darmstadt
Dedicated to Herbert Amann on the occasion of his 70
|
anniversary
Abstract
We consider the system of Maxwell-Stefan equations which describe mul-
ticomponent diusive uxes in non-dilute solutions or gas mixtures. We
apply the Perron-Frobenius theorem to the irreducible and quasi-positive
matrix which governs the ux-force relations and are able to show normal
ellipticity of the associated multicomponent diusion operator. This pro-
vides local-in-time wellposedness of the Maxwell-Stefan multicomponent
diusion system in the isobaric, isothermal case.
Keywords: Multicomponent Diusion, Cross-Diusion, Parabolic Systems
1 Introduction
On the macroscopic level of continuum mechanical modeling, uxes of chemical
components (species) are due to convection and molecular uxes, where the
latter essentially refers to diusive transport. The almost exclusively employed
constitutive law to model diusive uxes within continuum mechanical models
is Ficks law, stating that the ux of a chemical component is proportional to the
gradient of the concentration of this species, directed against the gradient. There
is no inuence of the other components, i.e. cross-eects are ignored although
well-known to appear in reality. Actually, such cross-eects can completely
divert the diusive uxes, leading to so-called reverse diusion (up-hill diusion
in direction of the gradient) or osmotic diusion (diusion without a gradient).
This has been proven in several experiments, e.g. in a classical setting by Duncan
and Toor; see [5].
To account for such important phenomena, a multicomponent diusion ap-
proach is required for realistic models. The standard approach within the the-
ory of Irreversible Thermodynamics replaces Fickian uxes by linear combi-
nations of the gradients of all involved concentrations, respectively chemical
potentials. This requires the knowledge of a full matrix of binary diusion coef-
cients and this diusivity matrix has to fulll certain requirements like positive
semi-deniteness in order to be consistent with the fundamental laws from ther-
modynamics. The Maxwell-Stefan approach to multicomponent diusion leads
1
to a concrete form of the diusivity matrix and is based on molecular force
balances to relate all individual species velocities. While the Maxwell-Stefan
equations are successfully used in engineering applications, they seem much less
known in the mathematical literature. In fact we are not aware of a rigorous
mathematical analysis of the Maxwell-Stefan approach to multicomponent diu-
sion, except for [6] which mainly addresses questions of modeling and numerical
computations, but also contains some analytical results which are closely related
to the present considerations.
2 Continuum Mechanical Modeling
of Multicomponent Fluids
We consider a multicomponent uid composed of n chemical components
.
.
Starting point of the Maxwell-Stefan equations are the individual mass balances,
i.e.

|
j
.
+ div (j
.
u
.
) = 1
tot
.
, (1)
where j
.
= j
.
(t, y) denotes the mass density and u
.
= u
.
(t, y) the individual
velocity of species
.
. Note that the spatial variable is denoted as y, while the
usual symbol x will refer to the composition of the mixture. The right-hand side
is the total rate of change of species mass due to all chemical transformations.
We assume conservation of the total mass, i.e. the production terms satisfy

n
.=1
1
tot
.
= 0. Let j denote the total mass density and u be the barycentric
(i.e., mass averaged) velocity, determined by
j :=
n

.=1
j
.
, j u :=
n

.=1
j
.
u
.
.
Summation of the individual mass balances (1) then yields

|
j + div (ju) = 0, (2)
i.e. the usual continuity equation.
In principle, a full set of n individual momentum balances should now be
added to the model; cf. [9]. But in almost all engineering models, a single set of
Navier-Stokes equations is used to describe the evolution of the velocity eld,
usually without accounting for individual contributions to the stress tensor. One
main reason is a lack of information about appropriate constitutive equations
for the stress in multicomponent mixtures; but cf. [14]. For the multicomponent,
single momentum model the barycentric velocity u is assumed to be determined
by the Navier-Stokes equations. Introducing the mass diusion uxes
j
.
:= j
.
(u
.
u) (3)
and the mass fractions Y
.
:= j
.
,j, the mass balances (1) can be rewritten as
j
|
Y
.
+ju Y
.
+ div j
.
= 1
tot
.
. (4)
In the present paper, main emphasis is on the aspect of multicomponent dif-
fusion, including the cross-diusion eects. Therefore, we focus on the special
2
case of isobaric, isothermal diusion. The (thermodynamic) pressure j is the
sum of partial pressures j
.
and the latter correspond to c
.
1T in the general
case with c
.
denoting the molar concentration, 1 the universal gas constant and
T the absolute temperature; here c
.
= j
.
,`
.
with `
.
the molar mass of species

.
. Hence isobaric conditions correspond to the case of constant total molar
concentration c
tot
, where c
tot
:=

n
.=1
c
.
. Still, species diusion can lead to
transport of momentum because the `
.
are dierent. Instead of u we therefore
employ the molar averaged velocity dened by
c
tot
v :=
n

.=1
c
.
u
.
. (5)
Note that other velocities are used as well; only the diusive uxes have to be
adapted; see, e.g., [18]. With the molar averaged velocity, the species equations
(1) become

|
c
.
+ div (c
.
v +J
.
) = :
tot
.
(6)
with :
tot
.
:= 1
tot
.
,`
.
and the diusive molar uxes
J
.
:= c
.
(u
.
v). (7)
Below we exploit the important fact that
n

.=1
J
.
= 0. (8)
As explained above we may now assume v = 0 in the isobaric case. In this
case the species equations (6) simplify to a system of reaction-diusion systems
given by

|
c
.
+ div J
.
= :
tot
.
, (9)
where the individual uxes J
.
need to be modelled by appropriate constitutive
equations. The most common constitutive equation is Ficks law which states
that
J
.
= 1
.
grad c
.
(10)
with diusivities 1
.
0. The diusivities are usually assumed to be constant,
while they indeed depend in particular on the composition of the system, i.e.
1
.
= 1
.
(c) with c := (c
1
. . . , c
n
). Even if the dependence of the 1
.
is taken
into account, the above denition of the uxes misses the cross-eects between
the diusing species. In case of concentrated systems more realistic constitutive
equations are hence required which especially account for such mutual inuences.
Here a common approach is the general constitutive law
J
.
=
n

=1
1
.
grad c

(11)
with binary diusivities 1
.
= 1
.
(c). Due to the structure of the driving forces,
as discussed below, the matrix D = [1
.
] is of the form D(c) = L(c) G

(c)
with a positive denite matrix G

(c), the Hessian of the Gibbs free energy.


Then, from general principles of the theory of Irreversible Thermodynamics, it
is assumed that the matrix of transport coecients L = [1
.
] satises
3
L is symmetric (the Onsager reciprocal relations)
L is positive semidenite (the second law of thermodynamics).
Under this assumption the quasilinear reaction-diusion system

|
c + div (D(c) c) = r(c), (12)
satises - probably after a reduction to n1 species - parabolicity conditions suf-
cient for local-in-time wellposedness. Here r(c) is short for (:
tot
1
(c), . . . , :
tot
n
(c)).
A main problem now is how realistic diusivity matrices together with their
dependence on the composition vector c can be obtained.
Let us note in passing that Herbert Amann has often been advocating that
general ux vectors should be considered, accounting both for concentration
dependent diusivities and for cross-diusion eects; see, e.g., [2].
3 The Maxwell-Stefan Equations
The Maxwell-Stefan equations rely on inter-species force balances. More pre-
cisely, it is assumed that the thermodynamical driving force d
.
of species
.
is in local equilibrium with the total friction force. Here and below it is often
convenient to work with the molar fractions r
.
:= c
.
,c
tot
instead of the chemical
concentrations. >From chemical thermodynamics it follows that for multicom-
ponent systems which are locally close to thermodynamical equilibrium (see,
e.g., [18]) the driving forces under isothermal conditions are given as
d
.
=
r
.
1T
grad j
.
(13)
with j
.
the chemical potential of species
.
. Equation (13) requires some more
explanation. Recall rst that the chemical potential j
.
for species
.
is dened
as
j
.
=
G
c
.
, (14)
where G denotes the (volume-specic) density of the Gibbs free energy. The
chemical potential depends on c
.
, but also on all other c

as well as on pressure
and temperature. In the engineering literature, from the chemical potential a
part j
0
.
depending on pressure and temperature is often separated and, depend-
ing on the context, a gradient may be applied only to the remainder. To avoid
confusion, the common notation in use therefore is
j
.
=
T,
j
.
+
j
.
j
j +
j
.
T
T.
Here
T,
j
.
means the gradient taken under constant pressure and tempera-
ture. In the isobaric, isothermal case this evidently makes no dierence. Let us
also note that G is assumed to be a convex function of the c
.
for single phase sys-
tems, since this guarantees thermodynamic stability, i.e. no spontaneous phase
separations. For concrete mixtures, the chemical potential is often assumed to
be given by
j
.
= j
0
.
+1T ln o
.
(15)
4
with o
.
the so-called activity of the i-th species; equation (15) actually implicitly
denes o
.
. In (15), the term j
0
.
depends on pressure and temperature. For a
mixture of ideal gases, the activity o
.
equals the molar fraction r
.
. The same
holds for solutions in the limit of an ideally dilute component, i.e. for r
.
0+.
This is no longer true for non-ideal systems in which case the activity is written
as
o
.
=
.
r
.
(16)
with an activity coecient
.
which itself depends in particular on the full
composition vector x.
The mutual friction force between species i and , is assumed to be propor-
tional to the relative velocity as well as to the amount of molar mass. Together
with the assumption of balance of forces this leads to the relation
d
.
=

=.
)
.
r
.
r

(u
.
u

) (17)
with certain drag coecients )
.
0; here )
.
= )
.
is a natural mechanical
assumption. Insertion of (13) and introduction of the so-called Maxwell-Stefan
(MS) diusivities
.
= 1,)
.
yields the system
r
.
1T
grad j
.
=

=.
r

J
.
r
.
J

c
tot

.
for i = 1, . . . , n. (18)
The set of equations (18) together with (8) forms the Maxwell-Stefan equations
of multicomponent diusion. The matrix [
.
] of MS-diusivities is assumed to
be symmetric in accordance with the symmetry of [)
.
]. Let us note that for
ideal gases the symmetry can be obtained from the kinetic theory of gases; cf.
[7] and [12]. The MS-diusivities
.
will in general depend on the composition
of the system.
Due to the symmetry of [
.
], the model is in fact consistent with the On-
sager reciprocal relations (cf. [16] as well as below), but notice that the
.
are
not to be inserted into (11), i.e. they do not directly correspond to the 1
.
there.
Instead, the MS equations have to be inverted in order to provide the uxes J
.
.
Note also that the Ansatz (17) implies

.
d
.
= 0 because of the symmetry
of [)
.
], resp. of [
.
]. Hence

.
d
.
= 0 is necessary in order for (17) to be
consistent. It in fact holds because of (and is nothing but) the Gibbs-Duhem
relation, see e.g. [10]. The relation

.
d
.
= 0 will be important below.
Example (Binary systems). For a system with two components we have
d
1
(= d
2
) =
1
c
tot

12
(
r
2
J
1
r
1
J
2
)
. (19)
Using r
1
+r
2
= 1 and J
1
+J
2
= 0 one obtains
J
1
(= J
2
) =

12
1T
c
1
grad j
1
. (20)
Writing c and J instead of c
1
and J
1
, respectively, and assuming that the chem-
ical potential is of the form j = j
0
+ 1T ln(c) with the activity coecient
= (c) this nally yields
J =
12
(
1 +
c

(c)
(c)
)
grad c. (21)
5
Inserting this into the species equation leads to a nonlinear diusion equation,
namely

|
c c(c) = :(c), (22)
where the function c : IR IR satises c

(:) =
12
(1 + :

(:),(:)) and,
say, c(0) = 0. Equation (22) is also known as the ltration equation (or, the
generalized porous medium equation) in other applications. Note that well-
known pde-theory applies to (22) and especially provides well-posedness as soon
as c is continuous and nondecreasing; cf., e.g., [19]. The latter holds if : :(:)
is increasing which is nothing but the fact that the chemical potential j of
a component should be an increasing function of its concentration. This is
physically reasonable in systems without phase separation.
4 Inversion of the Flux-Force Relations
In order to get constitutive equations for the uxes J
.
from the Maxwell-Stefan
equations, which need to be inserted into (9), we have to invert (18). Now
(18) alone is not invertible for the uxes, since these are linearly dependent.
Elimination of J
n
by means of (8) leads to the reduced system
c
tot

d
1

d
n1

= B

J
1

J
n1

, (23)
where the (n 1) (n 1)-matrix B is given by
1
.
=

r
.
(
1

1n

.
)
for i = ,,
r
.

.n
+
n

|=.
r
|

.|
for i = , (with r
n
= 1

n<n
r
n
).
(24)
Assuming for the moment the invertibiliy of B and letting j
.
be functions of
the composition expressed by the molar fractions x = (r
1
, . . . , r
n
), the uxes
are given by

J
1

J
n1

= c
tot
B
1

r
1

r
n1

, (25)
where
= [
.
] with
.
= c
.
+r
.
ln
.
r

(26)
captures the thermodynamical deviations from the ideally diluted situation; here
c
.
denotes the Kronecker symbol.
6
Example (Ternary systems). We have
B =

13
+r
2
(
1

12

13
)
r
1
(
1

12

13
)
r
2
(
1

12

23
)
1

23
+r
1
(
1

12

23
)

(27)
and det(BtI) = t
2
tr Bt + det B with
det B =
r
1

12

13
+
r
2

12

23
+
r
3

13

23
min{
1

12

13
,
1

12

23
,
1

13

23
} (28)
and
tr B =
r
1
+r
2

12
+
r
1
+r
3

13
+
r
2
+r
3

23
2 min{
1

12
,
1

13
,
1

23
}. (29)
It is easy to check that (tr B)
2
3 det B for this particular matrix and there-
fore the spectrum of B
1
is in the right complex half-plane within a sector of
angle less than ,6. This implies normal ellipticity of the dierential operator
B
1
(x)(x). Recall that a second order dierential operator with matrix-
valued coecients is said to be normally elliptic if the symbol of the principal
part has its spectrum inside the open right half-plane of the complex plane; see
section 4 in [2] for more details. This notion has been introduced by Herbert
Amann as the appropriate concept for generalizations to more general situations
with operator-valued coecients; see [1].
Consequently, the Maxwell-Stefan equations for a ternary system are locally-
in-time wellposed if = I, i.e. in the special case of ideal solutions. The latter
refers to the case when the chemical potentials are of the form (15) with
.
1
for all i. Of course this extends to any which is a small perturbations of I,
i.e. to slightly non-ideal solutions.
Let us note that Theorem 1 below yields the local-in-time wellposedness also
for general non-ideal solutions provided the Gibbs energy is strongly convex.
Note also that the reduction to n 1 species is the common approach in the
engineering literature, but invertibility of Bis not rigorously checked. For n = 4,
the 3 3-matrix B can still be shown to be invertible for any composition due
to r
.
0 and

.
r
.
= 1. Normal ellipticity can no longer be seen so easily.
For general n this approach is not feasible and the invariant approach below is
preferable.
Valuable references for the Maxwell-Stefan equations and there applications
in the Engineering Sciences are in particular the books [4], [7], [18] and the
review article [10].
5 Wellposedness of the Maxwell-Stefan equations
We rst invert the Maxwell-Stefan equations using an invariant formulation.
For this purpose, recall that

.
n
.
= 0 holds for both n
.
= J
.
and n
.
= d
.
. We
therefore have to solve
J = c
tot
d in 1 = {n IR
n
:

.
n
.
= 0}, (30)
7
where = (x) is given by
=
[
:
1
d
.



d
.
:
n
]
with :
.
=

|=.
r
|

.|
, d
.
=
r
.

.
.
The matrix has the following properties, where x 0 means r
.
0 for all i:
(i) () = span{x} for x = (r
1
, . . . , r
n
).
(ii) 1() = {e}

for e = (1, . . . , 1).


(iii) = [o
.
] is quasi-positive, i.e. o
.
0 for i = ,.
(iv) If x 0 then is irreducible, i.e. for every disjoint partition 1 J of
{1, . . . , n} there is some (i, ,) 1 J such that o
.
= 0.
Due to (i) and (ii) above, the Perron-Frobenius theorem in the version for quasi-
positive matrices applies; cf. [8] or [15]. This yields the following properties of
the spectrum o(): The spectral bound :() := max{Re ` : ` o()} is
an eigenvalue of , it is in fact a simple eigenvalue with a strictly positive
eigenvector. All other eigenvalues do not have positive eigenvectors or positive
generalized eigenvectors. Moreover,
Re ` < :() for all ` o(), ` = :().
>From now on we assume that in the present case x is strictly positive. Then,
since x is an eigenvector to the eigenvalue 0, it follows that
o() {0} {. I C : Re . < 0}.
Unique solvability of (30) already follows at this point. In addition, the same
arguments applied to

:= j(x e) for j IR yield


o(

) {j} {. I C : Re . < j} for all small j 0.


In particular,

is invertible for suciently small j 0 and


J = c
tot
(
j(x e)
)
1
d (31)
is the unique solution of (30). Note that

y = d with d e implies y e
and y = d. A similar representation of the inverted Maxwell-Stefan equations
can be found in [6].
The information on the spectrum of can be signicantly improved by
symmetrization. For this purpose let A = diag(r
1
, . . . r
n
) which is regular due
to x 0. Then
S
:= A

1
2
A
1
2
satises

S
=
[
:
1

d
.

d
.
:
n
]
, :
.
=

|=.
r
|

.|
,

d
.
=

r
.
r

.
,
i.e.
S
is symmetric with (
S
) = span{

x}, where

x
.
:=

r
.
. Hence the
spectrum of
S
and, hence, that of is real. Moreover,

S
(c) =
S
c

x
8
has the same properties as
S
for suciently small c 0. In particular,
S
is
quasi-positive, irreducible and

x 0 is an eigenvector for the eigenvalue c.
This holds for all c < c := min{1,
.
: i = ,}. Hence we obtain the improved
inclusion
o() {0} = o(
S
(c)) {c} for all c [0, c).
Therefore
o() (, c] {0}, (32)
which provides a uniform spectral gap for sucient to obtain normal ellipticity
of the associated dierential operator.
In order to work in a subspace of the composition space IR
n
instead of a
hyperplane, let n
.
= c
.
c
0
tot
,n such that

.
c
.
con:t is the same as n 1 =
{n IR
n
:

.
n
.
= 0}. Above we have shown in particular that
J
: 1 1 is
invertible and
[J
.
] = A
1
2
(
S

J
)
1
A

1
2
[d
.
] =
1
1T
A
1
2
(
S

J
)
1
A
1
2
[j
.
] (33)
with the symmetrized form
S
of and

1 := A
1
2
1 = {

x}

. Note that this


also shows the consistency with the Onsager relations. To proceed, we employ
(14) to obtain the representation
[J
.
] = A
1
2
(
S

J
)
1
A
1
2
G

(x) x. (34)
Inserting (34) into (9) and using c
tot
r
.
= n
.
+ c
0
tot
,n, we obtain the system of
species equations with multicomponent diusion modeled by the Maxwell-Stefan
equations. Without chemical reactions and in an isolated domain IR
n
(with
i the outer normal) we obtain the initial boundary value problem

|
n + div (D(n)n) = 0,
i
n

= 0, n
|=0
= n
0
, (35)
which we will consider in 1

(; 1). Note that A


1
2
(
S

J
)
1
A
1
2
G

(x) from (34)


corresponds to D(n) here. Below we call G C
2
(\ ) strongly convex if G

(x)
is positive denite for all x \ .
Applying well-known results for quasilinear parabolic systems based on 1

-
maximal regularity, e.g. from [3] or [13], we obtain the following result on local-
in-time wellposedness of the Maxwell-Stefan equations in the isobaric, isother-
mal case.
Theorem 1 Let IR

with 1 be open bounded with smooth . Let


j
+2
2
and n
0
\
2
2

(; 1) such that c
0
.
0 in

and c
0
tot
is constant in
. Let the diusion matrix D(n) be given according to (34), i.e. by
D(n) = A
1
2
(
S

J
)
1
A
1
2
G

(x) with c
tot
r
.
= n
.
+c
0
tot
,n,
where G : (0, )
n
IR is smooth and strongly convex. Then there exists - locally
in time - a unique strong solution (in the 1

-sense) of (35). This solution is in


fact classical.
9
Concerning the proof let us just mention that
div (D(n)n) = D(n) (n) + lower order terms,
hence the system of Maxwell-Stefan equations is locally-in-time wellposed if
the principal part D(n) (n) is normally elliptic for all n 1 such that
c(n) := n + c
0
tot
e is close to c
0
. The latter holds if, for some angle 0 (0,
t
2
),
the spectrum of D(n) (1) satises
o(D(n))
0
:= {` I C {0} : arg ` < 0} (36)
for all n 1 such that c(n) c
0

< c for c := min


.
c
0
.
,2, say. For such an
n 1, let ` I C and 1 be such that D(n) = `. Let x := c(n),c
tot
(n)
(0, )
n
and A = diag(r
1
, . . . , r
n
). Then
A
1
2
(
S

J
)
1
A
1
2
G

(x) = `.
Taking the inner product with G

(x) yields
(
S

J
)
1
A
1
2
G

(x) , A
1
2
G

(x) = `, G

(x) .
Note that A
1
2
G

(x) {

x}

, hence the left-hand side is strictly positive


due to the analysis given above. Moreover , G

(x) 0 since G is strongly


convex, hence ` 0. This implies (36) for any 0 (0,
t
2
) and, hence, local-in-
time existence follows.
6 Final Remarks
A straight-forward extension of Theorem 1 to the inhomogeneous case with
locally Lipschitz continuous right-hand side ) : IR
n
IR
n
, say, is possible if
)(n) 1 holds for all n. Translated back to the original variables (keeping the
symbol )) this yields a local-in-time solution of

|
c + div (D(c)c) = )(c),
i
c

= 0, c
|=0
= c
0
for appropriate initial values c
0
. Then a natural question is whether the solution
stays componentwise nonnegative. This can only hold if ) satises
)
.
(c) 0 whenever c 0 with c
.
= 0,
which is called quasi-positivity as in the linear case. In fact, under the considered
assumption, quasi-positivity of ) forces any classical solution to stay nonnegative
as long as it exists. The key point here is the structure of the Maxwell-Stefan
equations (18) which yields
J
.
= 1
.
(c) grad c
.
+c
.
F
.
(c, grad c)
with
1
.
(c) = 1,

=.
r

.
and F
.
(c, grad c) = 1
.
(c)

=.
1

.
J

.
10
Note that 1
.
(c) 0 and J
.
becomes proportional to grad c
.
at points where c
.
vanishes, i.e. the diusive cross-eects disappear. Moreover, it is easy to check
that
div J
.
= 1
.
(c)c
.
0 if c
.
= 0 and grad c
.
= 0.
To indicate a rigorous proof for the nonnegativity of solutions, consider the
modied system

|
c
.
+ div J
.
(c) = )
.
(t, c
+
) +c,
i
c

= 0, c
|=0
= c
0
+ce, (37)
where :
+
:= max{:, 0} denotes the positive part. Assume that the right-hand
side ) is quasi-positive and that (37) has a classical solution c

for all small c 0


on a common time interval [0, T). Now suppose that, for some i, the function
:
.
(t) = min
y

.
(t, y) has a rst zero at t
0
(0, T). Let the minimum of
c

.
(t
0
, ) be attained at y
0
and assume rst that y
0
is an interior point. Then
c

.
(t
0
, y
0
) = 0,
|
c

.
(t
0
, y
0
) 0, grad c

.
(t
0
, y
0
) = 0 and c

.
(t
0
, y
0
) 0 yields a
contradiction since )
.
(t
0
, c

.
(t
0
, y
0
)) 0. Here, because of the specic boundary
condition and the fact that has a smooth boundary, the same argument works
also if y
0
is a boundary point. In the limit c 0+ we obtain a nonnegative
solution for c = 0, hence a nonnegative solution of the original problem. This
nishes the proof since strong solutions are unique.
Note that non-negativity of the concentrations directly implies 1

-bounds
in the considered isobaric case due to 0 c
.
c
tot
c
0
tot
, which is an important
rst step for global existence.
The considerations in Section 5 are helpful to verify that the Maxwell-Stefan
multicomponent diusion is consistent with the second law from thermodynam-
ics. Indeed, (33) directly yields
[J
.
] : [j
.
] =
1
1T
(
(
S

J
)
1
A
1
2
[j
.
]
)
:
(
A
1
2
[j
.
]
)
0,
i.e. the entropy inequality is satised. The latter is already well-known in the
engineering literature, but with a dierent representation of the dissipative term
using the individual velocities; cf. [16].
For suciently regular solutions and under appropriate boundary conditions
the entropy inequality can be used as follows. Let \ (x) =

G(x) dx with G
the Gibbs free energy density. Let
\(x, x) =

[J
.
] : [j
.
] dx 0.
Then (\, \) is a Lyapunov couple, i.e.
\ (x(t)) +

|
0
\(x(:), x(:)) d: \ (x(0)) for t 0
and all suciently regular solutions.
In the present paper we considered the isobaric and isothermal case because
it allows to neglect convective transport and, hence, provides a good starting
point. The general case of a multicomponent ow is much more complicated,
even in the isothermal case. This case leads to a Navier-Stokes-Maxwell-Stefan
11
system which will be studied in future work.
Acknowledgement. The author would like to express his thanks to Jan Prss
(Halle-Wittenberg) for helpful discussions.
References
[1] H. Amann: Dynamic theory of quasilinear parabolic equations - II. Reaction-
diusion systems. Di. Int. Equ. 3, 13-75 (1990).
[2] H. Amann: Nonhomogeneous linear and quasilinear elliptic and parabolic bound-
ary value problems, pp. 9-126 in Function Spaces,Dierential Operators and Non-
linear Analysis. H.J. Schmeisser, H. Triebel (eds), Teubner, Stuttgart, Leipzig,
1993.
[3] H. Amann: Quasilinear parabolic problems via maximal regularity. Adv. Dier-
ential Equations 10, 1081-1110 (2005).
[4] R.B. Bird, W.E. Stewart, E.N. Lightfoot: Transport Phenomena (2
nd
edition).
Wiley, New York 2007.
[5] J.B. Duncan, H.L. Toor: An experimental study of three component gas diusion.
AIChE Journal 8, 38-41 (1962).
[6] V. Giovangigli, Multicomponent Flow Modeling, Birkhuser, Boston 1999.
[7] J.O. Hirschfelder, C.F. Curtiss, R.B. Bird: Molecular Theory of Gases and Liq-
uids (2
nd
corrected printing). Wiley, New York 1964.
[8] R.A. Horn, C.R. Johnson: Matrix Analysis. Cambridge University Press, Cam-
bridge 1985.
[9] P.J.A.M. Kerkhof, M.A.M. Geboers: Analysis and extension of the theory of
multicomponent uid diusion. Chem. Eng. Sci. 60, 3129-3167 (2005).
[10] R. Krishna, J.A. Wesselingh: The Maxwell-Stefan approach to mass transfer.
Chem. Eng. Sci. 52, 861-911 (1997).
[11] J.C. Maxwell: On the dynamical theory of gases, Phil. Trans. R. Soc. 157, 49-88
(1866).
[12] C. Muckenfuss: Stefan-Maxwell relations for multicomponent diusion and the
Chapman Enskog solution of the Boltzmann equations. J. Chem. Phys. 59, 1747-
1752 (1973).
[13] J. Prss: Maximal regularity for evolution equations in

-spaces. Conf. Sem.


Mat. Univ. Bari 285, 1-39 (2003).
[14] K.R. Rajagopal, L. Tao: Mechanics of Mixtures. World Scientic Publishers,
Singapore 1995.
[15] D. Serre: Matrices: Theory and Applications. Springer, NewYork 2002.
[16] G.L. Standart, R. Taylor, R. Krishna: The Maxwell-Stefan formulation of irre-
versible thermodynamics for simultaneous heat and mass transfer. Chem. Engng.
Commun. 3, 277-289 (1979).
12
[17] J. Stefan: ber das Gleichgewicht und die Bewegung insbesondere die Diusion
von Gasgemengen, Sitzber. Akad. Wiss. Wien 63, 63-124 (1871).
[18] R. Taylor, R. Krishna: Multicomponent mass transfer. Wiley, New York 1993.
[19] J.L. Vazquez, The Porous Medium Equation, Mathematical Theory, Clarendon-
Press, Oxford 2007.
13