Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
CONNECTICUT ACADEMY
OF
VOX^XJJVIE III
t7>
NEW
AYEN
to 1878.
Taylor, Printers,
New Ha
CONTENTS.
PAGE
Art.
I.
Report
on the dredgings
By
H arger.
Plates
Art. IL Descriptions
DROIDS from the
the region of
S.
Smith and O.
I.
-8,
op
NeW EnGLAND
species of Hy-
By
COAST.
S.
F,
58
III.
St.
Plates 9-10,
Clark.
Art.
in
By
E. S.
Tilly-Foster
Dana,
Plates
11-13,
67
Art. IV.
sin
ic
Philips.
Art.
Plates 14-37,
On the
V.
97
108
By
249
stances.
Art. VI.
S.
F. Clark.
By
Y.
Plates 38-41,
Art. VII.
M, Turnbull.
Art. VIII.
Plates 42-44,
265
fins, a contribution to
limbs.
By
J.
Plates 49-60,
K. Thacher,
281
Art. IX.
stances (concluded).
By
J.
W.
Gibbs,
343
President,
ELIAS LOOMIS.
Vice-President,
HUBERT
NEWTON.
A.
/Secretary,
OSCAR HARGER.
Librarian,
Treasurer,
HENRY
KINGSLEY.
C.
PtihlisJdng Committee,
HUBERT
GEORGE
CHESTER
A.
J.
S.
NEWTON,
ELIAS LOOMIS,
E. VERRILL,
ADDISON
WILLIAM
BRUSH,
LYMAN,
D.
WHITNEY.
Auditiiig Committee,
HUBERT
A.
NEWTON,
HENRY
T.
DANIEL
BLAKE.
C.
EATON,
Maximum zu."
CLAfsitrs.*
The comprehension
is
greatly facilitated
entropy
is
the limit of
[dQ denoting
and
all
/
,
by the system
in
it,)
the
that
in all
is
to
another.
trivances, supposed
theoretically perfect,
But
it
predominant importance.
As
is
its
in the case of
ai'e
of
viz.,
the perform-
system, (such as
two
all
is
capable of
S.
ix., S.
44.
W. Gihhs
J.
109
The
from
system which
all
external influences
may
is
isohited
For
the equilibrium
it
and
necessary
is
may be
written
{^V)e ^0.
II.
For
the equilibrium
which do not
(1)
variations in
vanish or be positive.
This condition
m,=
may
it
is
of
its
the system
be written
(2)
appear from the conalways possible to increase both the energy and
will
condition (1)
state of the
is
system in
viz.,
away.
it
by
For,
therefore,
and
necessary
of
0.
it is
the state
>
and 6e
=0;
as a variation
state)
is
not satisfied.
Conversely,
if
condition (2)
for
which
d<
and St/z^ 0;
de=iO and
therefore condition (1)
is
not
(J//
>
satisfied.
of equilibrium, as also
in
is,
infinitesimals
110
tf.
fii-st
relatively to tliose
Substances.
are to be neglected.
We
variations.
we
will
strued strictly,
i.
e.,
in
With
not to be neglected.
(^;/),<0,
for neutral equilibrium there
i.
e.,
(A6),^>0:
(3)
in
the state of
0,
i.
e., {Af),^
0,
(4)
while in general
{^V)e ^0,
and
i.e.,
(Af),^^0;
(5)
for
which
(A;/),>0,
i.
e.,
there must be
some
for
(6)
which
(^f),<0,
(V)
while in general
(d7),^0,i.e., (c^f),^0.
(8)
It is
In the
is
taken only
tliis is
any
finite
although they
may
tions of quantities
For example,
if
its
components, an infinitesimal
is
not to
if
In
is
But,
if
resist-
it
will
be neces-
J.
W. Gihhs
Ill
these parts, as sueli a change can not take place without the passage
then become
there
is
no communication of heat.
When
conduction
no farther consideration.
is
venting change.
ai-e in
of
nature.
its
friction
pressed together,
that which
from another,
that
fluid,
resistance to change
of
the system
and heat
which allow finite variations in the values of all the quantities which express the initial state of the system or the mechanical
or thermal influences acting on it, without producing the change in
The equilibi-ium which is due to such passive properties
question.
limits
is
by the balance
of the
112
J.
W. Gihhs
falls
To
consider
first
In the
first
we
place,
if
the system
two equivalent
is in
forms.
a state in which
must involve
its
it is
entropy
lated system.
either a decrease of
change of
iso-
is
evidently in
no
will
initial
finite
increase of energy.
We
consistent with
with
its
its
entropy, but that there are other states of the same energy
in
its
Avhen
difiiision,
Nor,
difierent
the
components
momenta
in diiferent
of the components
is
transfer of heat
posed
is
by
radiation.
masses and the transfer of heat are concerned, but to show that the
same is true in regard to the motions of difiiision and chemical or
molecular changes, when these can occur without being accompanied
or followed
by the motions
we must
have recourse to considerations of a more general nature. The following considerations seem to justify the belief that the condition is
sufficient for equilibrium in
every respect.
in equilibrium.
J.
W. Gibhs
113
in
in nature
This time must be so chosen that the change does not take place in
infinitely slowly,
which
is
it
in
mencement of the
entropy.
make
all
is
changes
in the state
supposed actually to
may
fied
which
is
thermodynamically impossible.
if
we
As
first
when
it
is
in
its
will evidently
its
is
entropy.
be
energy, or, in
is
always in equilibrium
when
is
it is
we must
it
then, there
supposed,
infinitesimal variation
allowed,
other words,
If,
0,
even
if
infinites-
the entropy
is
(The term possible has here the meaning pi-eviously defined, and the
character A is used, as before, to denote that the equations are to be
Trans. Conn. Acad., Vol.
III.
15
October, 1875.
114
ff.
W. Gibbs
construed
sti'ictly,
i.
e.,
higher orders.)
The only
our notation
6ij
is
that in which
when
is,
the
in
in
equilibrium
we
is
evidently unstable,
it
cannot be realized.
We
have
still
It
is
evidently so in
is
in
every
cases in
all
in the
statement of
(i.
e.,
both
in
the positive and the negative direction,) in states of the system differing infinitely little from the state in question.
may omit
In this case,
we
state of equilibrium
((J;/),
But
0,
i.e.,
(rJ6),
previously enunciated
(10)
is
in every
it
Gihhs
J. TF,
quantity which
EqtiiUhrinm
of Heterogeneons Svhstwtces.
115
is
its
without
or,
entropy,
altering
the
if
entroi)y
of
this
to
Now,
change.
variation
its
passive
as the described
its
energy without
it
forces.
will suffice,
it is
believed, to establish
The
criteria of stability
We
librium.
will
may
now proceed
systems
which
of phenomena.
For
this
purpose we
shall be
make
of the
and
we
We will
our
in a rigid
to heat.
We
will
action of gravity, or
solid portions of the
by any
is
electrical influences,
is
and that
in
the
116
J.
We
Substances.
problem by supposing that the variaand entropy which depend upon the
surfaces separating heterogeneous masses are so small in comparison
with the variations of the parts of the energy and entropy which
will farther simplify the
depend upon the quantities of these masses, that the former may be
neglected by the side of the latter; in other words, we will exclude
the considerations which belong to the theory of capillarity.
It will be observed that the supposition of a rigid and nonconducting envelop enclosing the mass under discussion involves no
would
be
also
so, if
envelop as supposed
mass thus enclosed are the general conditions which must always
be satisfied
As
case of equilibrium.
in
other suppositions
for the
and considerations
the circumstances
all
afterward be
will
made
of
subject
the
sj^ecial discussion.
Conditions relating
Let us
first'
to the Equilihr'nim
Homogeneous Parts of
the given
Mass.
its
part in question
is
its
entropy
and
7/,
its
volume
energy
its
and the
y,
f is
differentials
de
t
dt]
J) dv,
(1 1)
its
pressure.
For
*
dn
during
is
its
de^itdi]
y^ J,
7^2 5
/'n
differential of e
-\- yw^ ^^'"2
+/^"
The substances
aS^,,
^^'"n?
(1-)
taken with
?w.
S21
differen-
J.
W. Gihbs
tials d'm^,
dm2,
dm
shall
llV
from any
initially present.
which do not
initially
may
relating
to
therefore be necessary
component substances
If the conditions
tional) water.
state of
If,
maximum
might possibly
affoi'd
water to the
acid, it
maximum
The quantity of
the components.
this
in the positive
component
and
in
we
concentration as one of
will then be capa-
the quantity of the other component can increase but cannot decrease
0.
118
all
W. Gihhs
J.
may be more
in special cases, it
may be
It
observed that
is
it
equation (12) that the variations of nature and state of the mass to
which the equation refers should be such as do not disturb its homogeneity, provided that in all parts of the mass the variations of
For,
if this last
homogeneous mass
the energy, entropy, etc., of any
cWn
clD)}
- p dlJv
-\- /<
dDm^-\-
We
now suppose
will
all
the infin-
e., if
we write
is
fx,
(12)
cWni^
i.
infinitesimal part,
-f- //
by integrating
is
condition be
clDm^,
for the
(13)
whole
neous, and together occupying the whole space within the envelop.
will at first suppose the case to be such that the component sub-
We
stances are the same for each of the parts, each of the substances
If we
aS being an actual component of each part.
S'2,
/S
.
by
accents,
that
Se' _|-
ds"
etc.
(14)
which do not
of condi-
We
It will
equilibrium that
t'
_|_ t"
-f-
for
-2^'^v' +///(ym,'
Sr,'
S7/"
etc.
-\-
juj 6rn,J
/A2"
6m,"
.+//'
.
//,."
(Jm'
6m ^'
(15)
+ (J7" 4-
for
(5/7'"
which
^'j^Svnj^Sv"'-\-
etc.
etc.
0,
=0,
(16)
(17)
W. Gibbs
J.
and
For
this
it is
Sm^"
dm.^'
dni.^"
+ dm^'" +
+ Sni^'" +
(^m/
dm^'
Si/i"
-\-
-j-
(ym"'
etc.
etc.
-\- etc.
=
=
=r
119
0,
0,
0.
/jg\
= = t"'=
jj
=ip" =p"'
t"
yu
'
A<2'
//,/
etc.
(19)
etc.
(20)
= =
=
= A'2" = /<2"'= etc.
= //" =
=
yu
"
'"
yU
etc.
etc.
/u"'
"]
^21)
J
a quantity
may be
If
as defined
//,,
ditions
we
we
(21)
expressed as follows
TTie potential
be
constant
is no
upon the freedom of motion or combination of the component substances, and that each is an actual component of all parts
It will
restriction
The
state of the
(if
we
it is
we may
call the
is
The number
independent variables,
number of homogeneous
divided.
All
ponents.
its
if
is
is
parts
may be
known
its
as
com-
may
mass, or
its total
If
we
entropy,
also
we
know
will
known
aiford n-\-
addi-
120
tT
W. Gibbs
But
if
require that the potential for that substance in that part should be
equal to the potential for the same substance in the parts of which it
In this
is an actual component, but only that it shall not be less.
case instead of (21)
we may
which
which S^
S^i
is
is
write
ponent,
/'2
which S^
is
which S2
is
^-^2
(22)
Donent,
etc.,
is
only deter-
we now suppose
various homogeneous parts of the given mass are not the same, the
result will
different
any
/S'j,
part, the
S^
term
//
is
be wanting
in the former,
in the
composed of some or of
all
of the same
may
is
no one
which (with
set of substances,
be regarded
expressed as follows:
may be
J.
W. Gibhs
121
we
choose,
we may
these terms.
m=
l)e
is
a possible
We will
all
now suppose
We
out of others.
i.
for each of
component of the
a condition (expressed by ^ ) to
]>ut
e.,
*S\, aSj,
that some of
not an actual
is
composed of S^ and
Sr,
combined
in
6'
are not
the ratio of a to
b,
Let
Sr^
be
S^ and S^
6m
components.
The general
condition
of equilibrium
will
have the form of (15) with certain of the terms of the form
omitted.
:^'{t6if)
It
may
-:^{pSv)-\-:^{/J,dm^)-\-^{;i2d^m.,).
[(23)
,
.-^2:{/./drn)^0,
2 dm = 0, 2 6m = 0,
1
we
shall
:2 6m ^
of condition,
0,
(24)
122
W.
J.
(jrlbbs
For, although
yet
may
it
be that
/z^',
6m
it
Hence,
tive value.
(23)
is satisfied,
if (27)
must
(27)
is satisfied,
(2;3)
^'
is
incapable of a nega-
must
Again,
also be.
if
which
it
may
afiect the
e.,
as the value of
which
is
6m^ may
{a
(28)
0,
{a + b) M^,
M^ -^-bM^
(29)
is
necessary
in this case.
The
relations
may
be
less
simple than in this case, but in any case they will only affect the
equations of condition, and these may always be found without difiiculty, i^nd will enable us to eliminate from the general condition of
equilibrium as
after
many
which the
coefficients of the
remaining variations
may be
set
than zero.
it
be easy to perform these operations in each parbe intei-esting to see the form of the resultant
may
equations in general.
homogeneous parts
*S'i,
S^,
/S,
are considered
is
no
restriction
upon
their
a(Ba+P ;, +
where a,
*
When we come
7c
3,
+ A @, +
as
etc.
(30)
S^,, Sf,,
etc.,
of the equilibrium of
it
=:
and give
<^tc.
by equations such
existing,
we
shall take
(that
etc.
J.
is,
S ^,
denote nnmhers.
So,
We
relating to the
It is
f^,)
it
and
<-(-,
fi,
;(-,
denotes qualitative as
The equations
be observed, equations
will
it
123
Sitbstances.
of con-
easily be derived
will not
reduce to
will
(2.3)
^ 6m + 3I2 2 Sin,^
J/j
-\-
2 (hn^
JI
0.
(31)
This will appear from the same considerations which were used in
2 Srn^, 2
give to
etc,
in equation (30),
making
2 Sm =^
a 31^
or,
It will
Now
(27).
2 dm^.,
Snii,^
etc.
evidently possible to
n,
therefore
- k 31, - X etc. =^
+ etc.
a 31, + p 31, + etc. = h 31, + A 31, + etc.
-f /i 31,,
be observed that
/J,
in
is
it
values proportional to a,
31,,
(33)
this equation
31 taking the
(32)
0,
place of .
coeffi-
It is evident that
which (30) is an example, which may be obtained simWhen these condiply by changing in these equations into 3f.
any
possible values of
tions are satisfied, (31) will be satisfied with
2 6111^. For no values of these quantities are
Sdm^, 2 Sm^,
r equations of
{2 6m ,)(^,
+ {2 6m.,) ^
+ (^' 6m,) =
by the ordinary
from the
(34)
r
equa-
and
(33)
(33)
(19),
(34),
that
is,
is
satisfied, if the
equations like
are
or (23).
satisfied.
124
J.
For
and having
^Substances.
mass when
equilibrium
in
matter
in
r in
number,
quantities of the
the former
by
like
differentiation.
Conditions relating
Previoush/ Existing.
The
we have
variations which
dition
sufficient.
may be
formed entirely
different in state
more convenient to
I'egard a
vacuum
mation of empty spaces within the given mass, the term new parts
will
be used to include
anj^
existed initially.
now be taken
to include
formation of which
components of
we have
all
the
to consider.
for distinction
we
many
parts as
we
if
such exist
As we may
initially,
divide the
W. Gibhs
J.
initial
125
any variation
system are
we may
arl)iti'ary,
so
them
as initially
of
the
is
evidently
:^6-\-:^D,
the
first
summation
to all the
new
6 or
letter
parts.
relating to
(35)
all
D following ^ will
parts.)
Therefore the
is
:^6e^0,
:2de->[-
(36)
[(^V)
we substitute the value of 6e taken from equation (12),
2De+^{txhj) - 2{pr3v) + 2U.i,6, ,)+^{i,^6w.,) +2{i<^hn:)^ 0.
or, if
If
we
S ^,
S^^
*S'
page
122), that
Let these be
^1
^2
^>n
I r equations.
etc.
(38)
2 di] -f 2 By = 0,
2 Sv + ^ Dv 0,
(39)
z=z
and n
r equations of
h^
{^dm^ + ^Dm^)
-j-h^
+
tj
(40)
the form
{2Sm, +2Dm.^)
+i.^
{^(hn^ +
{2 Sm
^ Dm^
{^dm +
2 JDm^J
+ 2 1)m) =
-f /
{2 Sm
JS"
-1
I)m)
.
(41)^
etc.
coefficients in
(41)
and those in
(38),
the
tliat
that this
is
pendent equations
equations
is
the only condition which these coefficients must satisfy, except that the
and that
a reciprocal one.
be independent,
this relation
i.
e.,
shall
between the
coefficients of the
two
sets of
126
J.
Now,
tract
W. Gibhs
using Lagrange's
{^^ di]
-[
21)!/)
of multipliers,"*
P{2 6v + 2 D)j)
from the
yand
we
will sub-
member
iirst
P being constants
is as yet arbitrary.
We might proceed in the
same way with the remaining equations of condition, but we may
obtain the same result more simply in another way.
We will first
observe that
{2 6m^
-f
2Dm^) , + {2 6m + 2D)n^)
+ {2 6m + 2 Dm,,) = 0,
(42)
for
tions (38).
ties,
(Although
(5
J,
if for r
2,
satisfy
let
a^M^ + a^Jf.^
b^M^ + b^M^
3f
-f *
0,
M =
0, 1 r
etc.
equations, (43)
then
This
{^dm^^^Bm^)
+ J/ {2 Sm + 2 Dm,,) = 0. (44)
r of the constants M^, M2,
expression, in which the values of
J/,
{2 6m^-r2I)m^) + M2
11
are
still
arbitrary,
we
from the
first
mem-
become
- 2 Sv) -f 2 {^ (hn,)
2 De + 2
T2 67] + F2 8v - M^2dm^
- T2Dr]-^P2Dv -M^ 2Dm^
{t (J;/)
That
is,
{]>
having assigned to
we may
T,
it is
..
+ 2 (yM (hn)
+ M 2 6m
^3I2Dm>.0.
(45)
P, 31^, J/2,
assert that
equilibrium that (45) shall hold true for any variations in the state
of the system consistent with the equations of condition (39), (40),
(41).
But
it
will
always be possible,
On
iHfg,
in (3T),
3f,
(43),
of negative values, the successive steps in the reasoning will be developed at greater
W. Gihhs
J.
all
127
we
conditions
will represent
by
.4^0,
^^0,
etc.,
(46)
Then it
A, B^ etc. being linear functions of T, P, 31 1, J/g,
31,.
will be possible to deduce from these conditions a single condition of
the form
.
aA+/3B+
a,
/3,
etc,
(47)
0,
being positive constants, which cannot hold true consistBut it is evident from the form of (47)
etc.
that, like
from
directly
by applying
(45)
Now
(40), (41)).
it is
by
in
evident, in the
fore in the
(46), it
this
change
first
place, that
cannot contain
it
in the
T or
P, therewhich (45)
(for
reduces to (47))
2 d?f + :^ 1)1} = 0,
and
:^
60
+ :2 Bv = 0,
Again,
for the
2 dm + 2 Bm ,).,+
1
(>;
dm
, -f- li'
X>>/i,
31^
+ 2 dm 2 Bm) 3/ ~
(48)
-{-
for
which
3f
satisfy
eqs.
(43),
and
therefore
{2^Smj
+ 2 Bm^)
(49)
(S^, ^2,
which satisfy eqs. (38).
This equation (4D) will therefore hold true, if for r of the letters
for
(S
Sg,
Bn we substitute
it
will
we
use
128
J,
W. Gibbs
J.
(2 2,
Equilibrium
of Heterogeneous Substances.
ponents.
Therefore
that
it
necessary, and
it is
shall
it is
be possible to assign to
T^
P, 31^, J/gi
consistent with eqs. (43), that the condition (45) shall hold true for
any change
in the
For
this
it is
p = P,
t=l\
fj.^
Sm.^^
M^ dm ^,
/.I2
'
(50)
Mn^f^n^ MSni
Jje
for
TI))j-\-PI>v
J/,
Dm^ - M^Dn,^
...
(51)
and that
- M,,Drn^^
(52)
0,
each of the new parts as previously defined. If to these condiwe add equations (43), we may treat 1\ P, Jf^, J/g?
-^n
tions
'S'j,
is
distinguished
negative, and
if
a sub-
by
we
a single ac<'ent,
shall
have
/<i'
6m ^' may
^f^
but
be either positive or
if
S^
only a possible
is
posing
it,
If there
is
(not
necessarily as
in
which
this
and
in
W. Gihhs
J.
any
Now,
case.
Dm.^,
for
if
of
Z><^, />//,
7>y,
129
Dm^.,
the same
equivalent to
is
this, that
f
for
_ T)i^Pv -
J/,
-J/2
///^
J/,
'''n
(53)
substances 8^,82
'S'.
But the validity of this last transformation cannot be admitted
.
It
is
dif-
but
in
etc.,
among
substances of different nature or state, and which at their first formation must be infinitely small, the same assumption is evidently
entirely inadmissible, as the surfaces must be regarded as infinitely
We shall see hereafter what
large in proportion to the masses.
modifications are necessary in our formulge in order to include the
etc.,
is
neous
in
is
large in proportion to
all
will
much
smaller than
dimensions
is
if
it
separates.
nature)
when applied
in
between the
it is
shown that
in establishing
III.
17
October, 1ST5.
130
W.
J.
iJribhs
mutual action of the new and the original parts, which can affect the
result.
It will be easy to give such a meaning to the expressions
Dm^,
It will
Dnio^
I>m that
this shall
be evidently the
Even
energy
is
surface
the
if
to
action of
tlie
two masses.
We
none of
tlie
iJi],
Now
new
that
^^^^
if
defined that
parts, shall
not affected by
entropy, volume,
be violated.
These
represented
-^"^n
'^^'^
is
dis-
new
by tbe sums
the vai'ious parts (original and new), so far at least as any of these
quantities are determined or affected
parts.
We
suppose De,
will
TJi],
Dv,
Dm Dm^
^,
done
in various ways.
We may
Dm,, to be
This
may
be
we may supi)Ose this determined in any suitThus we may suppose the total enei'gy in and
new
parts.
Or,
it
imaginary surface which divides any new part fi-om the original is
so placed as to include all the matter which is affected by the
vicinity of the
new
we
W. Gihhs
J.
JEquiUhriuni of Hetercxje.neoxx
Siihsfaiices.
13]
The lioniogeneity of the new parts is of no conwe have made no assumption in that respect. It may
be doubtful whether we can consider the new parts, as thus bonnded,
bounding
surface.
sequence, as
But
way
in
which
tliat in
e.,
i.
otdy
But
apply.
which
it
we
first
may
be considered as
Hence,
finite
differences,
would not
(12)
if
little
For the
varied.
made up of two
dif-
are for a constant nature and state of the mass, and the second
We may
bounded
new
parts to be
as
results.
The condition
(52)
understood
in
either
ways
of these
(or
in
others which will suggest themselves to the reader) will have a per-
and
suffi-
new
when
parts,
and
and
fied,
and
(43)
it is
always
it
may
To
only necessary to show that when (50), (51), and (43) are
and (52) is not, (53) will also not be satisfied.
We
T}/
(50),
pi-ove this,
satis-
- Po -f M^
m^
+ 31^ m^
+ Mm
(54)
denotes the work obtainable by the formation (by a reversible prom are the energy,
cess) of a body of which f, //, v, m^, vy/g,
.
of the
components, within a
of
tlie
if,
after the
J.
1:52
medium and
the
body, their joint volumes and joint quantities of matter, were the
same
medium
body.
may
This consideration
when
f, ;/, //ij,
etc. are
finite
etc. this
T^
P,
Jif^, etc.
do not represent the values of the temperature, pressure, and potenof any real substance.
tials
all
(If
the substances
tS'^,
a>'.,,
S are
actual components
of a
medium
supposed.)
Now by
we
obtain
the equation
which
for
t7/
f-
7'//
pai-ts,
+ Pw- J/,
is
/^
by
J/2?2
and
the condition
not satisfied.
It
is
Therefore
(51),
Mn>^=zO.
(50)
(55)
+//n"?r.
will
all
(56)
possible
new
which
expression
e
JV
be negative, and
is
Avill
Mm
(57)
...
decrease
if
the
number of
iV^,
But
cannot become
size.
it
will be
it
infinite.
Now we
need not
iV^
mass
like 0,
like
it
is
certain that the least possible value of (57) per unit of volume, and
homogeneity.
If the
new
by
is
not
satisfied occurs between two different original jjarts 0' and O", the
may consider the
ai'gument need not be essentially varied.
We
J.
W.
(rihhs
133
value of (57) lor a body consisting of masses like O' and 0" sepaThis value may be decreased by increasing
rated by a lamina N.
may be done
by giving
it
a convoluted form
and
it
will
be evident, as before,
And
be negative.
homogeneous mass,
its least
posa
in
volume is the least possible for the same matter with the
same entropy and volume, hence, if confined in a non-conducting
unit of
vessel,
when
it
will
be
(50), (51),
isfied in
in a state
and
regard to
Therefore
all
possible
new
not sat-
is
be some homo-
aS'j,
iS^,
Therefore,
if
(50), (51), and (43), and condition (53) is satisfied by every homogeneous body which can be formed out of the given matter, there will
be equilibrium.
On
is
For we may easily conceive that the condition (52) shall hold true
(for any very small formations within or between any of the given
masses), while the condition (53)
is
all
large masses
case.
this is
very
etc.,
By
Such an equilibrium
this
is
sufficient to
initial state,
an
perhaps a circumstance which entirely escapes our powers of percej)The presence of a small portion of
tion, will be sufficient to do so.
the substance for which the condition (53) does not hold true,
cient to produce this result,
when
is suffi-
a variable
It will
(50)
little
134
./
TK Gibbs
homogeneous parts
bodies
which can be
any
of the given mass) and (53)
are
always
sufficient
for equiwith
formed of the given matter)
(43)
shall be
equilibrium
which
for
an
librium, and always necessary
fore, the
condition (56)
practically stable.
And,
in regard to equations
with
(38), it is
if
we
choose,
For,
(43).
it
if
is
always immaterial,
consider
it
applying
in
to be
all
com-
bodies to
may
be regarded,
like
T anc\
and
compo
(53).
for equilibrium
sufficient
and
m relate) that it
the component substances to which m^, m^,
shall be possible to give such values to the constants T, F, M^, M^.,
.
the given
Mass.
even when bodies can exist which are compounded in proportions infinitesimally varied from those of the solids considered.
e.
which we have called new). Yet it is worthy of notice that if a homogeneous solid which is a compound of variable proportions is in
contact and equilibrium with a fluid, and the actual components of
the solid (considered as of variable composition) are also actual com-
W. Gihbs
./
135
ponents of the
all
fluid,
is
if it
were
practically-
which would
fluid.
This follows directly from the principles stated on the })recedinoFor in this case the value of (57) will be zero as determined
pages.
either for the solid or for the fluid considered with reference to their
ultimate components, and will not be negative for any body whatever
which can be formed of these components; and these conditions are
sufticient for equilibrium
masses.
a
is
its
*S,
possible
fluid)
then, considering
we
solid as variable,
cW =
shall
t ill]'
dv'
But the
difierentials dm,,'
f.1,,'
dm J
dm,,'
//,/
dm^.',
body by equation
this
-\- //'
this
S^
have for
+
By
in the
-i- fx^:
dm.,:.
(58)
//^.'
(12)
considered as independent,
evi-
dently express variations which are not possible in the sense required
in the criterion of equilibrium.
might, however, introduce them
We
We
may
then write
=t' dt/
de'
t-ij
d}uj.
(59)
In regai'd to the relation of the potential /// to the potentials occurring in equation (58) it will be observed, that as we have by integration of (58)
and
(59)
^'
and
if
//^'
"'
=
mj =
e'
therefore
Now,
- p'
i' '/'
t'
+ /'/ ni,!
//
fij
- p'
mj
components
S,
iS,
we may
v>
//,'
S^
+
mj
//,/
and
(60)
//,; 1)1,;,
(61)
m,'.
S,,
(62)
.
S^.
the actual
J W.
136
6e" =z
a8
p"
t" (h/"
din,,"
l-i,,"
-\-
and
Glbbs
Sv"
+,/'" (^nta'
-^
-\-l-'J' ^^",i"
Sm^:'-\-i.i,'' 6)11,"
ji,:'
l^i"
(hv'\
((53)
by supposition
mj 2r =
m,,' '
ill,'
(64)
equations (43), (50), and (51) will give in this case on elimination of
the constants T, P, etc.,
t'
and
m;
i.>;
u,,; }i:'
may
=p\
t",
+*;
may
But
of S
6
t"
-\-iii,l
:=zm-^
f.1,1
if
)/
S,
-f-
*S',,,
p"
11
S/
aS'x.,
//" nt
.
jS,
and the
The
fluid.
iA"
all
-j-inj
last con-
(67)
p{J'.
we may
write for
m,,
/V'??J^. ///'w?,
...
.
(66)
//".
(65)
all
such bodies
///'
//,/' rn,j
)ij'
m^
(68)
0.
to be observed that
it is
and
member
given
its
by
is
it
Now, by
fluid.)
of this condition
As
state.
is
(60),
zero
solid,
we
shall
0,
(69)
have
(It
t"
f.i,i'
diitj
or,
and
(65),
Mc")
dnij
(/'//-/'/,")
dm,'
by equations
(58)
{l-ij
/<//'
dm,j
f.1,'/
dm,' ...
-{- {i-ij
;j,'')
(/<,'-//,")
///' diiti!
d)iij
dm,'^
0.
(70)
/u,;',
/Y,;
=;/,;',
M,.'= Hi!\
(71)
/V;
which we would
Mt,'^
W. Gibbs
J.
Equilibrium
of Heterogeneous Substances.
137
We
fluid.
Now,
and exceed
in
fluid
which con-
of the solid.
(There
may
number of different
solid bodies,
As
we have made
use, are
number
conditions of
equilibrium
But
is
otherwise immaterial,
system
in
which
may
deter-
Thus, when
by
is
a vari-
parts of
this sub-
stance will be invariable for each of the parts of the system which are
arise
The
solid
In the probThe
effect of
III.
18
Novejiber, 1875.
138
tT.
Suhstcmces.
Effect
each of which
is
homogeneous and
is
fluid,
a diaphragm which
by
and
is
is
per-
6v'=0,
and
for the
6v"=Qi,
(73)
67)1^=0,
dmj'=0,
6mi;=0,
6m,''=0,
etc.,
(74)
Sm/ +
With
rium
doi/ -f 6m/'
drn,''=. 0,
0,
etc.
(75)
these equations of condition, the general condition of equilib(see (15)) will give the following particular conditions:
t'
and
for the
t",
(76)
if
actual com-
biit
///=//;",
=11,!',
p'
not
nor
/.i,!=/.i",
the diaphragm
= p",
=
/,<j'
/-/,,",
etc.,
(77)
etc.
Again, if
proportions only, or in proportions not entirely determined yet subis
may be expressed by
equations
linear
between diti^\
will
be
which
equations of condition,
deduction
of
the
the
particular conknown
6m.,', etc., and if these be
difficulties.
no
We
will however
ditions of equilibrium will present
(being
etc.
actual
components
observe that if the components S^, /S2,
certain conditions, these conditions
ject to
we
6m o",
shall
aj yUj'+ 2
/'2'
as
tc.
= !
/f
,"
+ a2
many independent
//2"+
etc.
(78)
J.
as
TT".
Gibhs
139
t.liere
way upon
fluids,
the variation of the energy of the sources of these forces must appear
in the general condition of equilibrium,
di:'-\-
(U"
I"
(5v'
which will be
P" dv" ^
(79)
0,
P' and P" denoting the external forces per unit of area.
(Compare
From this condition we may evidently derive the same
(14).)
internal conditions of equilibrium as before, and in addition the
external conditions
p'
= r,
is
perfect,
and
p"
its
= F".
(80)
is
it
permeabilit}'^ of
impermeability absolute,
of the
that
e.,
i.
components of the
is
it
fluids in
all.
and
How
of course to
be determined by experiment.
If the
diaphragm
restriction, the
is
permeable to
all
cannot
(in
is
will
it
may
be
the fluid which has the less pressure wdll be practicalhj unstable., in
the sense in which the term has been used on page 133.
For
f"
//"/;/"
0,
(81)
p'
<p"
while
t^
^^t".,
;(j'=;/i",
etc.,
Therefore,
if
140
W. Gihbs
J.
e"
t'
ij" -\-2)'
v"
).i^'
m^"
j.io' n't'2"
tlie fluid
/'n'
(82)
>'n"<0.
accents refer.
in
any case in
which each of the two fluid masses may be regarded as uniform
throughout in nature and state. Let the character d be used with
the variables which express the nature, state, and quantity of the
of the diaphragm, the following relation will hold true in
fluids to
Then, as the
heat received
by
the
d?/',
and
as
is
by
Ttt"
^t'Di/
t"r)>/'
2)'vv' p"Tni",
(83)
(12),
//,'
Dm/ -\-
or
(//,"
It is
-(-
Ml" ^h"
+ /^2'
///) i>m,"-|-
+ /^2" D^'s" +
i^'^'s'
i/io"
- Ms)
i>"'2"+
etc.
etc.
^ 0,
to.
(84)
(85)
made
to
The nature
of
its
energy and
its
for
each
entropy are
both zero. The values of the energy and the entropy of any compound bodv in any particular state will then be fixed. Its energy
will be the sum of the work and heat expended in bringing its components from the states in which their energies and their entropies
and its
are zero into combination and to the state in question
;
entropy
is
for
any
reversible process
141
J.
by which
effected
is
homoge-
if
Or,
tities for a
we may
relation
that
f, ij,
'i
etc.,
To
m^.
that
and
-,
is,
which
ni in m
,
may
m^,
rn^,
-, -,
^,
-,
is
w + 1 independent
mined by experiment.
if f is
it
variables, of
known
function of
may
be more convenient to
In any case,
tion of
Now
we may
v,
our ideas
fix
consider
subsisting
it is
//,
w,
/1 2
m^, m^,
is
?,
as
by equa-
tion (12)
dE=Lt
di]
pdv
^^ dni^
dni2
/<dni,
(86)
t, p, pi-i, 1^21
/^'n ^'"^ functions of the same variables, which may
be derived from the original function by differentiation, and may
therefore be considered as known functions.
This will make n -\- 3
independent known relations between the 2n + 5 variables, e, ;/, v
w?j,
mg,
m, t,p,
of these variables,
yUj,
Uo,
are
//.
These are
all
evidently independent.
Now
upon
compound
considered,
we
may
say
in
general,
all
its
thermal,
which
all
these relations
may
142
W.
J.
Equillhrmm of Heterogeneous
G'lbhs
we
Substances.
will call a
We shall
question.
hereafter consider a
which possess
a fundamental equation.
same property.*
this
Let
t=:e-t
(87)
;/,
we
(If
then,
If,
find
known
is
//'
p, Ml, /'},
1/,
substitute for
we
pdv +
//dt
//'
^2
-|-
t,
v,
f^'^2
+/'f.c?>yi.
^^ ni.^,
/'n ill
in
dm^
as a function of
have again n
shall
//,
obtain
its
m,
If
we can
we then
as before.
2n -\- 5 variables
Let
X=e+pv,
then by
V, /^j,
t,
tdif
dp
X be known
then,
If,
(89)
(86),
dx
find
(88)
/.i^,
/<!
(7m
/'g ^^'"2
as a function of
.
//
?/,
p,
m^,
tn^,
we can
By elimi-
?;?,
in
(90)
/' ^^"^n-
Let
K
then,
by
d'Q=~If,
then,
=s-
t,j
+pv,
(91)
(86)
'C,
is
i]dt
-\-
known
dp
-{
/^i
dm^
-f-
as a function of
j-i2dm
p.,
#,
m^,
-f-
m.^,
/'f?wv
.
(92)
we can
* M. Massieu (Comptes Rendus, T. Ixix, 1869, p. 858 and p. 1057) has shown
all the properties of a fluid " which are considered in thermodynamics" may be
how
deduced from a single function, which he calls a characteristic function of the fluid
In the papers cited, he introduces two different functions of this kind
considered.
viz.,
which
in our notation
would be
_i,
jj'y
of the
rp',
tr/
or
the value of
which
ip,
tp
or -7-.
t
fluid,
is
yWn ^n,
144
J. W^.
'
the subscript numerals refer, (but not excluding the case in which
1 and the composition of the body is invariable,) there is a relabetween the quantities enumerated in any one of the above sets,
from which, if known, with the aid only of general principles and
relations, we may deduce all the i-elations subsisting for such a mass
between the quantities 6,
iii, t, jy, ji^, pi.^,
//, y, ni^^ m.^,
j,
//
tion
i:,
//-,
i-i.
define
^',
It
j,
and
between these
be observed that,
will
one
thei-e is
c,
besides
equations which
the
which subsists
q^^antities
equation.
the
sets of quantities
same property.
The
'Q,
(90),
(88),
convenient
etc.,
=im
=(m
;,,=(*)
letters
=(/L)
(104)
all
the
m^, r/ig,
It will be observed that the quantities in (103) are all independent
of the quantity of the mass considered, and are those which must, in
general, have the same value in contiguous masses in equilibrium.
letters
On
The quantity
//'
the quantities
ip,
x,
'i-
equation
//'
by
the
(105)
ttj.
between
e, ?7, ,
m,, m2,
V,
t,
m,,
m.2,
m.
As. by (86),
^= il)
the second equation
may
first.
But the
first
equation
is
(^)
V,
TO,,
m,,
TO
ldv\
[del vm
^. ^'
"^M 2!
^n-
'
which
is
variables.
//
in
W. Gibbs
J.
Equilibrium
of Heterogeneous Substances.
145
which
cess in
the
is
system, then
t"-W- Q,
= Q.
t{7i" ~
-
e'
and
Hence
and
{101)
(108)
]/)
f-f' = W',
for an
infinitely small
reversible
(109)
change
the
in
state of the
we may
write
-dip = dW.
Therefore,
That
is, if
we
is
//'
temperature, just as
(110)
consider
is
//'
and the
function
by the system
if
//'
is
temperature.
From
this
we may conclude
that
a uniform
and
sufficient for
equilibrium
may
be expressed by
W).^o.*
(Ill)
(2) in
such prob-
lems of equiUbrium as we have considered and others which we shall consider hereafter with evident
advantage in respect to
<
tlie
in (111) applies to
we have
to consider.
sake of deducing
all
in this
the particular
conditions of equilibrium from one general condition, and of having the quantities
mentioned
defined
and because
and
which would be
ff.,
we should adopt
III.
For example,
in the
problem
each
remaining conditions.
we would
if
would be wanting.
19
January, 1876.
146
J.
When
TK Gibhs
it is
and
//;'
'/;"
rehite
by a
be observed that
will
it
it is
not necessary
system
TFand
relate,
provided that the only source of heat or cold used has the
same temperature
external bodies
system
as the
may be
if
It
maybe
tions (111)
in the
show
when applied
(2)
given state
is
may
be restored to the
uniform throughout.
any variations
If there are
Any
aifect-
it
interesting to
and
in
final state.
in its initial or
system which do
If the
and
we may evidently
by supposing heat
And
<
increase
to
its
6t; :=. 0.
not uniform,
pass from
the
warmer
to
For
its
energy
will
S<Q
Hence, as we
may
and
(SjjX).
6e<0
From
this
we may conclude
and
djj
0.
if
we
we
Now
t6t/=z6f-]-t/6t.
(112)
by heating
- t6?j=0
and therefore
rh/-
-f
?/ f)7
0,
(113)
W. Gibhs
J.
(2) is
147
is suffi-
equivalent to
tdi]^0.
is
(114)
equivalent to
6il^-{-if6t^0
(115)
we have
is
in
which
composed
may be
by means
conveniently expressed
when
these
of the func-
tion
= e tt]-\-pv.
(Ii6)
(internal) equilibrium is
m,p^^That
(iiv)
from con-
Hence,
it is
quantities of the
t,
equal.
Or,
sum
for
it is
of the values of
necessary
first
mass
Thus,
salt, it is
is in
true in
more complicated
cases.
of
'Q
The reader
will easily
deduce these
page
In like
combination of masses
using
given on
128.
in
as before
X=t-\-pv.
(118)
48
If
J.
we
W.
Gihhi^
denote by
is
not varied,
the system
by accents
and
we have
x"
and
= ^"
x'
final states of
f'
('""
+i>
"')
= Q-
(iif)
This function may therefore be called the heat function for constant
pressure (just as the energy might be called the heat function for
constant volume), the diminution of the function representing in all
cases in which the pressure
system.
In
all
is
is
by
the
allowed
I'emains michanged.
POTENTIALS.
In the definition of the potentials
/z^,
/y,,
etc.,
the energy of a
Now,
as the
manner
in
does not
fix
mass, until
we have
fixed the
sidered as composed.
by
manner
For example,
in
if
tion,
we may
hydrate and
m^ weight-units
of the
composed
of the hydrate
and water,
ds
viz.,
Xdinjylr/,
would be
sidered as
V,
ms
from the potential for water in the same liquid concomposed of anhydrous salt and water, viz.,
different
dmjr),
V,
m.
./.
W. Gihbs
The value
m^-
is
c?m,
of the
two expressions
not equal to
m,^.,
we may
is,
for,
dni,
in the
two
we may
when
the
of water
although
quantity dm^y or
149
is
sidered.
combination of elements
neous body.)
In the above definition we may evidently substitute for entropy,
volume, and energy, respectively, either temperature, volume, and
the function
or entropy, pressure, and the function j or tempera?/ ;
ture, pressure,
'Q.
same homo-
geneous mass the same equations will subsist as between the units
That
of these substances.
is, if
a, 6, etc.,
/^
the substances,
aS', aS',
etc.,
aS^.,
/S'
etc.,
+ etc. =
(5;.
+ A ,
-|- etc.,
20)
and
a,
etc.
etc., h, A, etc,
/J,
mass,
a /<
4-
fJ
//,,
+ etc. =
/ii
//,
+ etc.
(121)
To show
this,
Then, the
first
150
W. Gihbs
J.
members of
20)
two members
the two
{ 1
But
must be equal.
must be understood that equation (120)
quantity, the
it
of (121)
intended to
is
denote equivalence of the substances represented in the mass conin other words, it is supsidered, and not merely chemical identity
posed that there are no passive resistances to change in the mass
;
example,
in respect to a
by the
For
we may
and oxygen
(at
but water
is
member
other.
ordinary temperatures),
not write
and no neces-
sary relation will subsist between the potential for water and the
potentials for
The reader
by equations
(which are essentially relations between the potentials for actual components in diffei-ent parts of a mass in a state of
equilibrium) are simply those which by (121) would necessary sub-
and
(43)
sist
(51)
as variable
relate.
r=j.im
(122)
tion,
Also
in the case of a
body of
variable composition,
when
all
the
We
J.
W. Gibhs
151
itself,
t.
considered as a potential.
t,
a reversible process,
altered, the entropy
when
But the
not be altered.
the entropies
substance.
is
is
zero
of the energy of the mass and substance taken together, and therefore
equal, as the original energy of the substance
is
zero, to the
increment
The
is
by
homogeneous mass
by
is
equal
a reversible
which
and the temperature is the same
as that of the given mass into combination with this mass, which at
the close of the process must have the same volume and temperature
as at first, and which is supposed so large as not to be sensibly
altered in any part.
A source of heat or cold of the temperature
of the given mass is allowed, with this exception, other bodies are
to be used only on the same conditions as before.
This may be
process from a state
in'
ij:
(109) to the
together.
The
the potential
is
affected
very
easily,
how
the value of
in the
energy and the entropy of each elementary subFor we may imagine the substance brought from the state
definition of the
stance.
in
which
//'
=:
is
the
same
152
W. Gibbs
J.
mass,
first
any
to
Equilibrium
of Heterogeneous Substances.
same temperature, and then
work expended
is
In the
first
denoted by
and
'/',
work expended
in
let
/<
ju z=z
arbitrary,
we may
//''
JC,
of
t
'/,
-[-
as aforesaid
W the
then
W.
(123)
all
6=0
and
//
is
zz:
or e
possible states
f //,
in
for
all
possible states
in (123)
and
The value
to
mass
the substance in
C^
in question,
Now
ip
Let this be
last
state.
term
Applying
in this
this
equation
is
is
to be determined.
term which shall refer solely to the composition and therstate of any such body without regard to its quantity or
modynamic
We
form.
may
call
dif-
which
examples of the same
differ
phase.
plane, in an equilibrium
ances to change,
we
resist-
n+2
-r.
./.
Or,
W. GibbsEquilihrhtm of Heterogeneous
when the
if
we
still
153
r bodies considered
variable components,
Substances.
number
capable will
be n-\-2 r.
still
In this case,
and
(51)
substances.
shall
have by
Let
(97), as
before, r relations
relations
We
is
be necessary to
will
n component
pressure,
it
different
component
substances.
Hence,
if
r z= w
istent) is possible.
2,
no variation
in
It
+ 2. An
liquid,
but
it is
An example
of
7i
=2
and r:=4
seen in a
is
two
Concerning n
We
-\- I
Coexistent Phases.
relation
a whole).
In this case
we have
>i
the quantities
accents.
the same
But
is
//,
v,
mj, m^,
and^
true of
different equations.
etc.
will each
If the total
number
of these potentials
is
+ A,
the h independent relations between the units of the component substances to which the potentials relate,
by means
of
which we
may
III.
20
January, 1876.
154
W. Gihhs
J.
dp=z
7/ dt
+ mj
dj.!^,
let it
m,'
be observed that
//,
etc.,
(124)
become
df-ij,
-\-A,!df.i.
(125)
^?\'a
But the
*S
y8j,
+ ''6'A+etc. = ^/ 1 + ^12' 2
first
member
+^^n'. (126)
the matter in the body to which equations (124) and (125) relate.
we may regard
body
as
quantity
AJ
which
\i'
We
will then
dp=L
rf
dt
m^\
A ^\ ^2?
^^' ^"
become
-^^
m^ d}.i^
-{ rn^
dj-i^,
As
same
this
-{
to
rnj
(12V)
djJ^.
as are the
(97)
and
may be reduced
to a
(124) relate.
The
rest of the
71
equations
in
dp
= if dt
7n^"
df.1^
^m.^'dj-i^
-^m^'dpt^,
(128)
etc.
By elimination
V
of
<:?yu,, df.i^t
dj.4
W. Gibhs
J.
When
11-=. 1,
(/' v'
or, if
we make
ni'
=.
- rn' v")
1
which
dp
{m"
if
tn"
1,
v'-v"
t{v"
(130)
dt
in
155
formula
(131)
-v'Y
'
When
7i>
1, if
the quantities of
are proportional in
all
-Sg,
is less
S^
sufficient
In fact,
will
the condition of the coexistence of the two phases together with the
condition of the equality of the n
with the n
ji>
as a function of
to
if
if
ratios of
w/,
mg',
known
is
may
?'
determine
wi" is sufficient to
for each of
be expressed
form of (130), m' and m" denoting either the quantities of any
components or the total qiiantities of matter in the bodies
in the
one
The
the phases.
of the
which they
relate.
the heat
in that the
is
in this case,
unity.
But
in
letters in the
is
not in
when
and
tlie
i)i
matter to which
it
is
essentially
a crystalline body,
is
letters in
When n =
ature
is
two coexistent phases are capable, when the temperBut as (130) will
this case when m^' m^
m/' ni^'^ it follows that for
2,
hold true in
imum when
is
maximum
in general a
two phases
is
maximum
or a min-
identical.
is in
In like
general a
156
Gibhs
TF.
Equilibrium of
Ffeterogeiieoiis Substa7ices.
is
and
which
2^ foi"
no coexistent phases are possible from those for which there are two
This may be applied to a liquid having
pair of coexistent phases.
two independently variable components in connection with the vapor
which it yields, or in connection with any solid which may be formed
in
it.
When
we have
z=. 3,
m
m
m,
Now
rf
dp=.\?f
we may
m^
m^
tn^"
tn^'
df-
m,"
dM^. (132)
is
when
two.
which the condition is satisfied separates those simultaneous values of t and /> for which three coexistent phases are not
possible, from those for which there are two triads of coexistent
These propositions may be extended to higher values of ?i,
phases.
of
and
and
2^ foi"
by the
illustrated
solutions of
71
INTERNAL
We
will
HOMOGENEOUS FLUIDS
FUNDAMENTAL EQUATIONS.
STABILITY OF
now
envelop which
AS INDICATED BY
is
a rigid
all
the
denote
for
us
its
compounds
all
of S^, S^,
aS',
if
oeneous bodies.
homo-
J.
Now, if
M^, M2,
possible to assign
is
it
157
Substances.
Tt]-\-
Pv ~ J/, mj - M2 m^
3fm
(133)
not identical in nature and state with the given fluid (but composed
entirely of
aS\,
/S'g,
/S),
stable.
For,
in
homogeneous, or
fluid, if
its
is
not
whole mass cannot be negative; and if its value cannot be zero for
any homogeneous part which is not identical in phase with the mass
in its given condition, its value cannot be zero for the whole except
when the whole is in the given condition. Therefore, in the case
supposed, the value of this expression for any other than the given
(That this conclusion cannot be
condition of the mass is positive.
invalidated by the fact that it is not entirely correct to regard a
composite mass as made up of homogeneous parts having the
same properties in respect to energy, entropy, etc., as if they were
parts of larger homogeneous masses, will easily appear from considerations similar to those adduced on pages 131-133.)
If, then, the
value of the expression (133) for the mass considered is less when it
is in the given condition than when it is in any other, the energy of
the mass in its given condition must be less than in any other condition in which it has the same entropy and volume.
The given con(See page 110.)
dition is therefore stable.
Again,
if it is
(133) that the value of the expression shall be zero for the given
fluid mass,
and
shall not
page
110.)
It will
be stable unless
it is
(See
volume and with the given entropy to consist of homofor all of which the value of the expression (133) is zero,
but M^hich are not all identical in phase with the mass in its given condition.
(A mass consisting of such parts would be in equilibrium, as
we have already seen on pages 133, 134.) In this case, if we disre-
in the given
geneous parts
gard the quantities connected with the surfaces which divide the
*
vacuum
We
158
W. Gibhs
J.
homogeneous
Equ'dihfium of Heterdgeneous
we must regard
parts,
But
neutral equilibrium.
Siibstances.
in
which we may evidently consider to be all diiFerent phases, the fol(The accents distinguish the
satisfied.
letters referring to the different parts, and the unaccented letters
lowing conditions must be
whole mass.)
refer to the
rf
+ if + etc. =
v'
?Mj'
v"
-\- etc.
/^,
V,
I
=/>!',,
etc.
Now
the values of
mass
fluid
^-,
etc.
in its
>,
v,
//,
m^, m2,
are determined
,,
etc.,
in fact, tlie
j-.
jj-.
between
v',
whole
set of
we may regard
Flence,
latter is one.
linear equations
Now
(The values of
v" , etc.
fluid
and
it is
than n
is,
the
We
2.
is
mass
will
be
in neutral
of the substances,
if
phases,
it
is
etc. are also subject to the condition that none of them can be
negative.)
will
They form,
v",
by
of
parts.
v'
by the whole
are determined
etc.
etc., etc.,
(134)
)^
fluid, for
which the value of (133) is zero (when the constants are so determined that the value of the expression is zero for the given fluid,
and not negative for any substance) for the tendency of such a formation to be reabsorbed will diminish indefinitely as the mass out of
;
which
it is
When
formed increases.
xS'g,
S are
all
independently vari-
only be
fulfilled
when
it is
M^,
are equal
to the temperature, the pressure, and the several potentials in the given
J.
mass.
If
we
159
Substances.
will necessarily
1S2,
value
its
S are not
positive for
is
it
When
it
but
as
other phases.
will
it
be necessary to give
mass
all
all
and
must be equal to
be easy to obtain as
will
many
its
eqiia"
it is
(133) that the value of the expression shall be zero for the given
and either zero or positive for any phase of the same compowe have already seen (pages 129-134) that if equilibrium
subsists without passive resistances to change, it must be in virtue of
properties which are peculiar to small masses surrounded by masses
of different nature, and which are not indicated bj' fundamental
fluid,
nents,
equations.
we
all
in
degree through-
fundamental equations
it
will
by
in this
sense.*
*
If
we wish
to
know the
when exposed to a
we have only to suppose
enclosed in the same envelop witli the given fluid another body (which cannot combine
with the
Pv.
which
fluid) of
as the case
of course
may
(Tand
be,
must be those
P denoting
When
it is
is e
Ty, or
= ~ Fv.
or
T>)
and
page
fluid
and
positive
every other phase of the same components, the value of (133) for the whole system
will be less when the system is in its given condition than when it is in any other.
for
Hence the
fluid is stable.
When
it
is
fluid are of
stants that the value of (133) shall be zero for the given fluid
fluid
is
of course unstable.
when
it is
possible to assign such values to the constants that the value of (133) shall be zero
for the given fluid
and zero or positive for every other phase, but not without the
value zero for some other phase, the state of equilibrium of the fluid as stable
or neutral will be determined by the possibUity of satisfying, for any other than
fluid,
we
first
60
W. Gihbs
.7.
and
and any
*S', and
if not, whether it is capable of the value zero
any other phase than that of which the stal)ility is in question, it
is only necessary to consider phases having the temperature T and
pressure P.
For we may assume that a mass of matter represented
mis capable of at least one state ot
by any values of w^j, w^,
not unstable equilibrium (which may or may not be a homogeneous
It may easily be shown
state) at this temperature anrl pressure.
that for such a state the value of f T7}-\- Pv must be as small as
The same will therefore be
for any other state of the same matter.
/Sj, JS^,
for
Therefore
if this
expression
is
capable of
We
J/i m, J/2
^"'2
3/m.
(135)
when
T,
whether it is capable of the value zero for any other phase than that
of which the stability is in question, we have only to consider the
Any
least value of which it is capable for a constant value of v.
body giving this value must satisfy the condition that for constant
volume
de
Teh/
J/j
drn^
J/^
din.,
...
3fdm^
0,
(136)
stability of the fluid for constant temperature, or for constant pressure, or for both.
The number
the remaining equations, and then the equilibrium of the fluid will be neutral in
respect to one or
Gibbs
J. TT.
or, if
we
(Is
161
t di]
yM
dm,,
...
Tdi]
Jfj
yM
of
dm,,
dm^
-f-
/</,
dm,,,
M2 dm^
...
dm,^
/^^.
- Mdm^
0.
(137)
is,
potentials of
its
substances, so that
m^.lA^.
and as we have by
= m^
M-^^
-{-
mM;
(138)
(93)
E=.t
7]
pv
-\- /A
m ...-}-
yu,,
body
m^,
(139)
it
[P-p)v,
(140)
the value of
is
(141)
in
fluid of
which
all
the ultimate components are independently variable admit a very simple expression.
If the pressure
of the fluid
is
other phase of the same components which has the same temperature
if its
components, the
pressure
is
not as great
But
if
we should form
arbitrarily
any equation as
a fundamental equation, and ask whether a fluid of which the properTrans. Conn. Acad., Vol.
III.
21
January, 1876.
162
J.
Wl Gibbs
were oiven by that equati<in would be stable, the tests of stabilitylast given would be insufficient, as some of our assumptions might
ties
The
by the equation.
not be fultilled
test,
cases be sufficient.
all
Stability in respect to
may
take place
in
any mass, we
new
in existing
phase of a
may be
fluid
and unstable
In this case
changes
is
it
may
of discontinuous changes.
in regard to continuous
be capable
mencement
phases.
is
unstable
We
change.
to
will
now
may
it
We may
also
is
in question.
In
tliis
little
from
tlie
be
called,
same gen-
use the
be applied
phase of which
component substances
case the
ponents of the
fluid,
il/'j,
J/g?
but
be stable.
if
become
case, in
adjacent phases.
It will
be
in
call the
condition
of stability,
e"
in
in
t'
if
jo'
v"
-/</;,"...-
/4'
m"
> 0,
(142)
J.
W. Gibbs
by double
This condition
accents.
103
by
(93) equiva-
> 0,
(143)
> 0.
(144)
is
lent to
e"^t'if^p'v"-~n,'m,"
f'4-
w'
'//-/
'm/
-{-/<,
//'m"
-j-
j-i'
m'
and to
^t'7f-\-p'v" -fi.'m,"
t"
yf
The condition
- p" v" +
(143)
A'
may be
x"
we use
the character
m,"
~/^:m"
/."
mj'
Ae.>tAn~iy^v +
if
to
i^i^J)ii^
in the
form
-{-jjzlm,
(1^5)
is
first,
but
is
In fact,
when
which
it
it
must be possible
of
(If
is
is
satisfied
But
it
is
to be
remem-
by a
its
the
first
member
of (145)
A->tA--\-i.i.J~^
in
which form
we may
it
+yuJ
(146)
As
d-=itd^ 4- /.i,d^
we
-4-u^d-^,
to
(147)
continuous changes
to the
second
and higher
of
of the density of entropy and the densities of the several components,
which would maJce the density of energy a minimum, if the necessary
differential coefficients
164
TK Gihhs
fZ
in (145)
than
Regarding
v.
of a phase
and higher
in
regarded as a function of
its
if
were
to
which
sufficient condition of
matter, or of
of
any two
tlie
is
it
all
may
as the case
collective
But
relate.
etc.
/>',
t',
fulfilled.
The formula
be, in general.
may
determinations of stability, we
same
1}
Jt-{- V
Jp
m^
...
J/Y J
limits, or,
therefore, of such
in the
form
- m^ J// > 0,
(148)
4-^-^/V
(149)
or
Jp>ljt+'^J,i,
Comparing
bility in
we
(98),
see that
it is
all
given limits, that within those limits the same conditions should be
the second and higher differential coefficients of
fulfilled in respect to
eral potentials,
were
first
if
the
differential coefficients
fulfilled.
By
may
be brought to
the form
^f
For the
and
t'
if
p'
m^'
true of any
+//';/'>
limits
when v'=
it is
y", //</
/></',
m^
=z
t'
(150)
necessary
This evidently
?",
f -f+{t" -t'),/'>i);
and that when
0.
requires that
any given
sufficient that
(151)
t"
f ~ p'
r'
-\- fi
^'
^'
in the
-\-
form
jjj
mj > 0.
(152)
W. Gihhs
J.
[Atl.^
- |X^Am^
[/iil:-\-2)^'0
in
(153)
// Z/>?i],
> 0,
165
(154)
which are to be
m.
standing for all the quantities
regarded as constant,
If these conditions hold true within any given limits, (150) will also
hold true of any two infinitesimally differing phases within the same
To prove
limits.
by
we
this,
t'"
= v",
v"'
Now
by
(155)
t\
= m,",
m,"'
will
the equations
and
m"'
= m".
(156)
(153),
'/'"-'/"+(^"'-O'/"<0;
and by
(157)
(154),
f"
//''
J^ p' v'"
j^v'
-i-//,'/n/
-p'
t'" //"
1^1
v'
m^'"
^'
//
'
m/
-/V >/''"
-\-
>0.
^aJ tnj
(158)
> 0,
//' ///'
(159)
which by (155) and (156) is equivalent to (150). Therefore, the conditions (153) and (154) in respect to the phases within any given
limits are necessary
and
phases
It will
and in volume, and in (154), the condition of mechanand chemical stability of the body considered as maintained at a
Comparing equation (88), we see that the
constant temperature.
in composition
ical
if
if
is
is
-^ <0,
i.
e.,
if-/
When
positive.
or
/^:=
^
1,
we regard
as constant,
i.
(thespee.,
when
(154) will
by which the
{Af
+ p Av\,>().
d7)
v-~-
when
(the
elasticity for
n> 1, (154)
may
if
(160)
-^
>0,
dv^
constant temperature)
i.
e.,
if
L
dv
is
positive.
qy
But
V constant.
(142)
may
be brought to
166
J\
W. Gihhs
Equilibrium of Heter
-
j^t",f-p"v"
^"
-t'
-l;'
+y
if
ix.'m,"
0(1 eneous
-lA^m:
+/<///... +
v"
Substances.
/'n'
ra:
> 0.
(161)
all
it is
7/Jt
V /lp],,<0,
(162)
and
[JC-yU,^mi
may
as
The
easily be
first
-ywJm],,,>0,
(163)
body considered as unchangeable in compoand the second the conditions of chemical stability for a body
considered as maintained at a constant temperature and pressure. If
m^,
n z=. 1, the second condition falls away, and as in this case C
ditions of stability for a
sition,
The foregoing
<^
=:
t' 7}
+p
We
now
will
Let
^^' m^
/V>^n,
(164)
and of
d^={t -
t')
ch]
is
v,
the value of
shall
{p -p') dv +
(/ij
/f,')
be a minimum
Differentiating
dm^
first.
if
(//
^')dm.
we regard
v,
(165)
m^,
as
This condition
may
stability,
for
be written
1^)
>0.
(166)
W. Gibbs
J.
Wlien
of
y,
this condition
is
will
?/',
value of
satisfied, the
for
(Z>,
And
167
therefore, in
We
when
by accenting these
may be any
from the
little
;;,
imum
and
V, m.2
>0.
yu,
A<j'; so that in
(167)
value, for
//
any
limits) for
(-p-X
\nm^!t,
When
l)ut
fixed.
v,
m^,
will
.
',
have a min-
when
t'
t' and /<, = yw/we need only consider the phases for which t
particular conditions
following
the
obtain
also
In this way we may
of stability
(P)
\Zlm2ft,
\JmJt.
When
the n-\-
>0,
V,
m,,
m;5,
(168)
?n
>0.
(169)
V,
v, will
be a
all
satisfied,
minimum when
the value
the tem-
perature and the potentials of the variable phase are equal to those
The
of the fixed.
will
be entirely
will
be completely
identical.
satisfied,
when
satisfied.
From
the
derived,
^,
yUj,
it is
manner
in
evident that
/J.
In this
way we may
might be included
in
the
first
of these
lists,
and
?i
The quan-
in the
168
W. Gihhs
fZ.
Tlieii
tuted for the condition that the volume shall be constant in the state-
minimum
value of ^.
true of any
stability.
(t"
t') (;/"
+ (/^i"
which
may
//')
(/' -i>')
y"i') h'^x"
W-
''H')
limits of
v')
/^')
'-m')>0,
(170)
+ Jfx^/lm^
^tJjj /ip^v
+z/yuz/m>0.
(iVl)
This must hold true of any two infinitesimally difiering phases within
If, then, we give the value zero to one of the
the limits of stability.
difierences in every
completely identical, the values of the two differences in the remaining term will have the same sign, except in the case of Jp and Jv,
which
will
have opposite
signs.
J)
in
(If
two
except p and
we
always
write
^?
for
in (166)-(169),
/}
sufficient for
stability.
ai-e
If
dt
dy
df]
du.
dt
+ dm
dm
when
When
dt=z
-z
d^x^
r^,
dv =:
0,
m^,
we
shall
are
have
v,
dt
d)n
+ am
Let us consider
du
(172)
d^
dfi
d^^=-^-dv\-^^^drn,
d^
+^fm
W. Gibhs
J.
dm
d^E
d-^e
d.T]
d)n
dm
d^E
di]
d^e
^
(173)
dm. dm.
d'^E
dm
dm
dm, j
by
'
dm,'^
(86) the
same
as the coeificients in
i?,
t-'tc.
H^
Then if dt,
by (172)
df-i.^,
f?/'_i,
for the
remaining constituent.
have the value zero, we have
the last
for
and dv
Ii dpi,
that
dm djj
'
'
'
dy}
-\- 1
d^E
d^e
d^8
drf
169
all
= ^+i dni,
(174)
is,
/^\
In like
=^-K
(175)
manner we obtain
\dm_Jt,v,ii,,
f^n^Z^
(176)
etc.
by writing d
for /} in (166)-(169)
are equivalent to this, that the determinant given above with the
it
d^ E
constituent -j-^ shall
all
as
be positive.
Any
]^hase for
remaining
last
which
this con-
Tf,
m, (with
in
responding rows
in
corresponding interchange of
^,
/ij,
//).
by erasing
We will now
of
stability
(i. e.,
III.
22
January, 1876.
170
tT.
Substances.
formed by changing
into
mem-
in the first
(That
it
the
is
in
is
is
phase
capable of variation at
is
by the subscript
ferential
way
all
The same
letters.)
coefficients
/;,
in the
by
(165)
d^ =
{pt
- /tj)dm.
(177)
Now
(")
""-"-(.Irj'c""-"'-')
and
its
accented letters
refer.
and
Therefore,
The
[<^l-'A'r(4/7,
1
i
rn,
same order
different
as {vi
made
i/i^
or
pi^
Now
mj)^.
differential
(the
this value of
/;
constant, etc.
or
letters,
be
will of course
coefficient
v, will
be the least
by accenting these
* The limits of stability with respect to discontinuous changes are formed by phases
which are coexistent with other phases. Some of the properties of such phases have
See pages 152-156.
already been considered.
W. Gibbs
J.
when we give
And
way
interchange in any
way t, /Vj,
when we choose
.
?/,
w,,
in
m,
du.
-J-,
d}]
-f^,
dm^
when we
regard to the
we may
all
these relations
we
It follows that,
/./.
for
if
in
171
diiferential coefficients
dj.in
-7
(181)
^
d>n
'
together with v as those which are to remain constant in differentiathe differential coefficient as thus determined will
be at
as when one or more of the constants in differenfrom the denominators, one being still taken from
least as small
Now we
the limit
differential coefficients, as
determined
in
at that limit.
one at least of these differential coefficients determined by considerinoconstant the quantities occurring in the numerators of the others
But
together with
y, will
value zero,
all
\drnjt^
V, fii,
by
first)
zero.
if
For
nn-i
(98),
;^,
//.
'
That
change the phase without altering any of the quantities
therefore,
if
is,
/->,
we may
yUj,
brium.)
the fractions in
formed
Therefore,
all
may
be expressed, in general, by
Such
172
J.
W. Gihhs
is known, may be
reduced to the form of an equation between the independent variables
members
expressed by
by
may be
The form of
not be altered by the
setting this
?;,
?,,
m^, but
it
will be altered
d^A
=0,
fJ-n
that
is,
(183)
that
drV
(184)
i,
or
by
(98), if
we regard
if,
yw^,
jWn^l
F-\
yw
td^\_
(185)
XdiJ^^I
In like manner
we may obtain
d^p
~dP
Any
d^p
d^p
(186)
one of these equations, (185), (186), may be regarded, in genWe may be certain
the equation of the limit of stability.
eral, as
that at every phase at that limit one at least of these equations will
hold true.
GEOMETRICAL ILLUSTRATIONS.
Surfaces in lohich the Composition of the
Constant.
In vol.
ii,
p.
Body
represented
method
is
is
described of
J.
W. Gibhs
co-ordinates.
first
173
by rectangular
kind of funda-
Any
other kind of
fundamental equation for a substance of invariable composition will
suggest an analogous geometrical method. Thus, if we make m con-
three,
when n =. \.
The method
it
may
more advantageous to select a method in which the properties represented by two of the co-ordinates shall be such as best serve
to identify and describe the difi*erent states of the substance.
This
condition is satisfied by temperature and pressure as well, perhaps, as
by any other properties. We may represent these by two of the
co-ordinates and the potential by the third.
(See page 143.)
It
will not be overlooked that there is the closest analogy between these
often be
energy.)
If
(A
we
give
between volume,
entrojjy,
and
vz=zx,
i]
= y,
z,
174
W. Gibhs
J.
The
surfaces detined
guished as the
which they
In the
v-7]-e
by equations
and the
surface,
(187)
t-p-t surface,
distin-
of the substance to
relate.
t-2>'C
point
by two.
Let
follows:
C^'',
?*^\
C^"^'
may
which
be distinguished as
and
by the three
2>
^(n<KX)<^(S)^
in the
(191)
next space, separated from the former by the line for which
^(r)<^(.?)<^(X)^
(192)
in
by the
line
for
^(S)<^(r)<^W
which
(193)
in the fourth
^^^
in the fifth
^(^)<?(^><?(^
(195)
in the sixth
?(<^(^) <?(*).
(196)
The
(194)
'C,
is
From
which
this it is evident
But the
unstable states.
values of
by
of
t,
may
states
by
represented
the intermediate
the highest.
work obtained
The
differences
^^^^
?^''>,
etc.
represent the
by
amount
a reversible process
from one to the other of the states to which these quantities relate,
in a medium having the temperature and pressure common to the
two states. To illustrate such a process, we may suppose a plane
perpendicular to the axis of temperature to pass through the points
* See the Reports of the British Association for 1871 and 1872
Magazine,
and Philosophical
J.
Substances.
175
and
The geometrical
may be
for the
relations
easily obtained
reversible
path
any
state
states.
stability of
by applying
is
t' 1]
-\- 2)'
ji'
rn^
(197)
If
is in
we
con-
may
'Q-'Q'-[-{t-t')n-{2y-p')v,
(198)
which evidently denotes the distance of the point {t',p\ C') below the
tangent plane for the point {t,p, t), measured parallel to the axis of
t,.
Hence
if
the tangent plane for every other state passes above the
If
any
of the tangent planes pass below the point representing the given
state, that state will
be unstable.
Yet
it is
represented
plane p=.Q
when the
latter
negative.
is
by
which t and jo have the same values and a less value. It follows
from what has been stated, that where the surface is doubly convex
upwards (in the direction in which
is measured) the states reprefor
t,
'Q
surface
is
either of its principal curvatures the states represented will be unstable in respect to adjacent states.
When
it is
We
number of surfaces.
natural extension of either of
the methods already described will give us a series of suifaces in
by an
infinite
176
W. Gibhs
J.
is
the
^7-;/-f
t-jy-t
surface for
Surfaces
sented
When
and Curves in
Variable and
mg
mj, mg,
haps, as follows.
tities
v:ihich
its
is
The body
is
m3
/S'g,
/Sg,
the value of
per-
same straight
If
line.
m,
-f
in the
we suppose masses
If the triangle
is
the height unity, the distances of the point from the three sides will
Z and
(when m^ -j- "'2 "h^s ^= 1)5 ^^ points
form a surface, which may be designated as the
when
curve, which
we
and
(for constant
we imagine
relate,
surface for
if
ni-i;.
surface, or curve,
/.i^
dm^
-f /U2
dm^ +
yWg
dm^,
and denote by
'q'
which these
letters
the plane,
TK Gibbs Equilibrium
J.
oj'
Heterogeneous Substances.
177
/,', /^a? "^3? the distances of tlie foot of this ordinate from
the three sides of the triangle P^ Po P3, we may easily obtain
and by
= yWjm,' +
'Q'
which
may
Ave
fx
m'
(199)
-f //g w^a',
There-
//j,
page
S^, there
indicated
is
Among
149.)
by the
may be formed
of S^,
jS^,
(See
and
will be represented
will
be independent.
The
a
itself,
which
will
viz.,
be equal to
'Q.
common
tangent plane.
is
situated
an angle
in the
edge of a
lies
If
no part of the
it
meets the
plane will represent a stable (or at least not unstable) set of coexistent phases.
the same variables for bodies which consist of parts in different but
coexistent phases.
We may
in
body
The
point which
will evidently
the parts of the body placed at the points representing the phases of
these parts.
III.
23
January, 1876.
ITS
J.
W. Gihhs
may
which are
may
homogeneous.
surface.
The
will
It
This
consist,
may be
in
called
general, of
upon the primitive surface the part of the envelop of its successive
positions which lies between the curves traced by the points of con:
When
has a triple tangent plane or one of higher order, the triangle in the
more phases.
surface as thus constructed for any
thermodynamic
Of the whole
pressure,
that part is especially imporpositive
any
temperature and
value
of
for
any given values of m,, mg,
least
?
tant which gives the
represented
by a point in this part of the
The state of a mass
m^.of
energy would be possible if
surface is one in which no dissipation
impermeable both to
envelop
the mass were enclosed in a rigid
matter and to heat and the state of any mass composed of S^, S^, S^
in any proportions, in which the dissipation of energy has been comconsisting in general of three or
(i.
e., under the
above supposed,) would be
limitations imposed
by such an envelop
as
we
We may
dissipated energy.
It is
evident that
it
Y plane coincides
constructed
is
states
which
for
enero"y,) that
it
The general
it
which the
m-'C,
surface
is
no surface of dissipated
nowhere has any convexity upward, and that the
is
ni-'Q
lines for
now
of
We
proceed to illustrate the use of both the surfaces and the lines
by the discussion
J.
W. Gihhs
We
(C).
may,
179
(B),
for convenience,
Q..
ature
is
re-
If
dt.
(The reader
dt)j j)^
\ai
'^^'
''^'
1.
AC,
Now
if
is
true of the
we denote by
//
If this quantity
of these states.
is
jf) will
first to
the second
ature will evidently cause a part of the (;urve (B) to protrude below
the tangent to (A) and (C), which will no longer form a part of the
line of dissipated energy.
three curves (A), (B), and (C), and of the tangents to (A) and (B)
and
On
and
(C), so that
imstable.
If
all
{if
lie
entirely
by (B)
will
be
be produced by
The
effect of a
constant
tion of
change of
may be found
any ordinate
in a
will
be
pi-essui-e
manner
dp or V
dp.
The
Therefore,
variaif
the
80
J.
W. Gibbs
When
is
the
two
cases which
must be distinguished.
tangent plane
three
in
lies
X-
Y plane
may
which
the other
angles.
(in
by joining
be joined so as to form a
^?^,,
falls
l^lane,
ABC
in the
quadruple tangent
its vertices,
Fig.
sented in the
sheets.
2.
figui-e
Fig.
by ruled
portions of
such that
by
ABC
manner not
entirely determined
by
J.
W. Gihhs
181
C which
is
If the first
(i.
is
the
e.,
the
belongs) to sep-
it,
These
effects will
is
the volume of the same matter divided between the phases A, B, and
C.
If the
pressure will cause the sheet (D) to separate from the plane tangent to
the other sheets, and a diminution of pressure will cause a part of the
sheet (D) to protrude below that tangent plane.
will
And
less
volume.
if
these effects
the homoge-
same considerations which were used in the analogous case for two
component substances.
Now when the sheet (D) rises above the plane tangent to the other
sheets, the general features of the surface of dissipated
by the disappearance
of the point D.
energy are
But when
the sheet (D) protrudes below the plane tangent to the other sheets,
the surface of dissipated energy will take the form indicated in figure
3.
by
182
J.
W. Gihbs
But wlieu the points of contact with the quadruple tangent plane
which represent the four coexistent phases can be joined so as to
form a quadrilateral
ABCD
(fig.
4)
portions of the four sheets of the primitive surface which are tangent
to
it,
Fig. 4.
Pig.
5.
tangent planes rolling upon the four pairs of these sheets which corres-
pond
energy,
let
there
heat
is
first
is
by
the point
Among
all relate to
one which
is
which
of these states
is
the gen-
the
To determine
Now
and
pressure,)
first
which we
(i.
e., if
to the second
may
suppose
(B),
triple
tangent
planes
to
to
(B),
(A),
(D), and
(C),
to
and
(B),
and
(C),
rise
to
above the
(A),
(C),
sipated energy.
into
diminution of temperature
The
quadrilateral
ABCD
lU).
will
The
in
this case
effects
dis-
break
produced by
J.
W. Gihhs
183
sented
by
By considering
the difference
two
states repre-
we may
distin-
So
latter.
may be
at isolated points or
of two component
may be at
Such cases need
in the pre-
And
in the
generally be
left to
the reader to
make
it
will
in
in
cesses.
We will
cal composition,
The
and
first,
represented
is
of components
is
two.
curves.
One
of the
above the other except at the point of contact therefore, when the temperature and pressure remain constant,
one phase cannot be varied in composition without becoming unstable
curves will in general
lie
ponent
may
is
(relatively)
increased.
entirely
By
above
it.
we may
easily determine
temperatures and pressures for which such coexistent phases are possible.
It will be observed that as we pass this limit of temperature
and pressure, the pair of coexistent phases does not simply become
unstable, like pairs and triads of coexistent phases Avhich we have
considered before, but there ceases to be any such pair of coexistent
The same result has already been obtained analytically on
phases.
W. Gibhs
184
J.
page
155.
But on that
be two
thei-e will
and
p, as seen in figure
If the curve
6.
sents vapor,
(represented by)
B may
AA'
repre-
liquid, a liquid
and
(at the
their composition,
Fig.
6.
vapor
component, and
are
made
Therefore,
if
these liquids
the boiling
is
to each other,
If
BB'
we
is
liquids
But
unit of mass, the vapor will then agree with the liquid.
per
if
the
curve BB' (which has the greater curvature) represents vapor, and
AA'
and A'
differ
more
make
the liquids
in composition.
cated in the figure will hold for a temperature higher than that for
which (with the same pressure) the curves are tangent to one another.
When two coexistent phases of three component substances have
the same composition, they are represented
two
just considered.
The upper
is
by the point
of contact of
do not intersect
very similar to that which we have
If these sheets
or pressure are so
varied that a part of the upper sheet protrudes through the lower, the
points of contact of a double tangent plane rolling upon the two
sheets will describe a closed curve on each, and the surface of dissi-
pated energy will include a portion of each sheet of the primitive surface united by a ring-shaped developable siirface.
If the sheet
liquid,
and
the other sheet vapor, the boiling temperature for any given pressure
will be a maximum, and the])ressure of saturated vapor for any given
temperature will be a minimun, when the coexistent liquid and vapor
J. TT.
Gihhs
below
whole primitive
sui-face as
85
seen from
from
may he
represented in tigure
primitive surface, and
united by the
we may make
the other
by the proper
ATB, ETF
sheet to which
belong
is
If the
be merged
developable
surfaces
But
if
ATH, and
It is
also
minimum
is
not a case of
maximum
or
minimum tem-
maximum
two.
BTC,
DTE, FTG.
is
CTD
that to which
is
is
when
In this case, the primitive surface must touch the same plane
same straight
line.
of the primitive surface to which these points belong as the sheets (A),
(B),
The
is
is
intermediate in position.
(B).
It
may
or
it
may
not intersect
it
at the point of
If
it
does not,
which
it
it
must
lie
ture or pressure
is
varied so as to
III.
make
24
Now,
if
the tempera-
186
J.
W. Gihbs
developable surface formed on the sheets (A) and (B), the surface of
dissipated energy will be altered in its general features only by the
is
altered so as to
make a
But
(C).
if
the temperature
through the developable surface formed on (A) and (B), the surface
of dissipated energy will have the
b
form
indicated in figure
include two
will
ABC
8.
Tt
plane triangles
\
^''^-
"
^-
part
CC
two being
last
But
if,
when
the
primitive
surface
it
is
plane intersects the developable surface formed upon the other sheets
(A) and (B), the surface of dissipated energy will not include this
develojjable surface, but will consist of portions of the three primi-
two developable
tri2)le
sink
make
the sheet
relatively to the
de-
and
face
Fig.
9.
formed on (A) and (C) and on (B) and (C) will separate from
one another, and the two parts of the sheet (C) will be merged in
one.
will give a
J.
W. Gihbs
Equllihriunh of Heteror/eneous
is
187
f'iuhstances.
ABC, A'B'C
planes, a portion of the sheet (A) on the left of the line aAxV'a', a por-
two separate
portions
aACc
and a'A'C'c' of the developable surface formed on (A) and (0), two
separate portions bBC;^ and b'B'C';/' of the developable surface
formed on (B) and (C), and the portion A'ABB' of the developable
surface formed on (A) and (B).
From
it
appears that
is
(in
maximum
or
general) the
minimum
for
mum
or
minimum
is
of
on page 150.
are
the nature of the surface of dissipated energy have been only occasionally mentioned, as they are often far
more
distinctly indicated
It will
by
be observed that
upon
surface in
space,
is sufficient for
tn-'C,
the
re-
CRITICAL PHASES.
It
in
Analogous properties may doubtless be exhibited by compounds of variable composition without change of temperature or
state.
''
On
188
W. Glbbs
J.
For
pressure.
if,
at
mg
liquids
than
less
and in any greater than h, a and b being the values of that ratio
for two coexistent phases, while either can form a stable mixture with
a third liquid in all proportions, and any small quantities of the first
and second can unite at once with a great qiiantity of the third to
form a stable mixture, it may easily be seen that two coexistent mix,
may
ture and pressure remaining the same, from initial phases in each of
in
But
interest.
if
from the
different
changes
will in
critical
no respect
it
may
ing discussion, phases in the vicinity of the critical phase will generally be called stable, if they are unstable only in respect to the
Let us
first
consider the
number of independent
the
number
is
capable.
variations of which
If
we denote by
we
by giving
to n
t,
/j.
p^ i.i^, /j^,
of the quantities the
critical
If
we now vary
n l quantities, we shall
stant a
new
is
first series,
we
by the
for the
new
set
have
phase,
Now
for every
J.
little
and
must be
in the first,
from the
mined by a
We
first.
;/
of the quantities
critical phase,
differs,
but
we
^,jtj, /^
j,
//g?
infinitely
differs
we vary
189
arbitrarily
/^n,
as deter-
e.,
i.
a critical phase
capable of
is
two
m,
we
immediate vicinity of a
ties
t,
p,
/<,, //g,
critical phase, if
Mn, are
we may
?/,
This con-
m.
the
=0,
(-j
we have
characterizes, as
m^, mg,
v),
compared
(200)
In fact,
if
we give
values determined
by
to the quantities
^,
yWj, yUg,
/<_i
constant
a series
of values increasing from the less to the greater of the values which
we determine
it
yM
since
it
has the same value in the two coexistent phases, but not a uniform
value throughout the series (for
probable,
all
function of
these phases
^,
or of m,
if it
had, Avhich
would be coexistent)
if
v also
is
is
theoretically im-
must be a decreasing
supposed constant.
Therefore,
the series must contain phases which are unstable in respect to con-
tinuous changes.
phases
may
(See page
68.)
And
critical
phase, the
190
./.
W. Gibbs
which the
in respect to
The
ties
^,
critical
phase
fX2->
t.he
Mn
'
leave out of
unstable.)
is
jo, yUj,
we
account the
phases which
critical
by giving to n
ol the quanti-
either direction
is
it
Hence,
if
remain constant,
it
Therefore,
to discontinuous changes.
rn
when
any
t,
v, jj^, ^Iq,
pin-i
But
critical phase.
yu is
part
the
in
a critical phase
/x^, /U2,
J-^n
an increasing function of
phase, the following conditions must also hold true at that phase
=0,
(^'-^1
\dm^Jt,
'
(20,)
^0.
(202)
V,
satisfied.
It is
may
is
capable of
n1
inde-
pendent variations.
We are not, however, absolutely certain that equation (200) will
always be
satisfied
denominator
in
by a
critical phase.
the fraction
may
For
it
is
some
tial coefficient
different
tinuous changes),
Gihhs
J. TT^
tage
in the use of
//,
mj,*^,,
quantities
t,
ii^, //o,
...//
may be
which Rn+x
^^enotes the
The
independent variables.
as
will
i?+t=^.
in
101
(203)
same determinant
as
To
we observe
when varied
on page 169.
is
critical
^,
volume,
"n+ldtf^^I^^l dm.
dm^
dt}
A'n
viz.,
_j_
dli^,
^j^^^
'
'
(204)
dtrir,
dm,^^ etc.
c?//,
sion (204) has the value zero, \in of the equations (172) are satisfied.
may be
This
expressed by an equation
aS'=0,
which
(205)
in
aS
denotes a determinant
in
we
write
choose
TJ for
^,
/),
Wj, Wg,
wi as
the determinant
^2r
dH_
^?^?_
c?m,2
dm^dm^
dm^_^dm,^
dn
dm 2^
dm,^_^dm^
dn
dnt-ydm,^
dn
dH
d^^
(206)
_dH
drn^dm
and
formed from
this
by substituting
for the
dU
dU
dtn^
dm?^
dU
dn/in^^
''
U=
0,
F=
0.
(208)
192
J.
may
W. Gihhs
it
(i.
in
e.,
is
resembles any phase which has a coexistent phase, but differs from
such phases in that the two parts into which the mass divides when
it ceases to be homogeneous differ infinitely little from each other and
from the original phase, and that neither of these parts is in general
If we consider a change in the mass to be deterinfinitely small.
(///,
dm^,
dv,
dm 2,
dm,
it
evident
is
that the change in question will cause the mass to cease to be homo-
^^1 a,j+
^i^^i
dv
d?]
dv+
^"-' dm,,
dm
c7m
+ ^^^
dm
(209)
be entirely
For,
different.
will consist of
two
if
be independent.
There-
n of the
quantities
fore, for
/^n
will
is
In general, however,
or
if
n of the quantities p,
t,
/.i
//,
j-i.^,
same constant
.
mined
But
if less
J.
W. Gibhs
193
the latter
v, ni^,
//,
in
We
may
cases in which
we keep
its
homogeneity by changing
its
pressure constant
we cannot destroy
volume but if
homogeneity by
its
its
Or,
if
is
main-
its
is
Or,
if
a mass consists of
two coexistent
the mass
is
maintained constant,
or thermal means, or
it
will not
by changing the
be possible by mechanical
whole mass. The stateparagraph and of the preceding have reference only to
ments of
this
infinitesimal changes.*
*
of a
A brief
came
abstract (which
memoir by M. Duclaux,
"
Sur
la
III.
25
February, 1876.
194
W. Gihhs
.7.
we apply equation
homogeneous mass having two indeaS'j, and make <, jo, and mj
(97) to a
we
obtain
,ip\
+,Jp)
^0.
(2,0)
='
(P)
\am2/t,p, m,
>)
(21
^
'
or
(212)
ydm^Jt,
p,
m,
ponent
true
we
/Sg,
shall
general as
irt,
and
in respect to
have
been done with regard to the potential on pages 148, 149) that the
value of a differential coefficient like that in (212) for any given mass,
when
the substance
when
if
by S^ has been
restriction,
it
case.
In fact,
it
is
sists
wholly of
/6'j, i.
e.,
when mo^O.
Therefore,
if
m^ = 0.
mg
is
capable
coefficient in (211)
cannot be positive.
Nor,
if
we
Gihbs
TV.
./.
195
it,
To
that
*S^j
The addition
ular hydrate).
*^'^^t
our ideas,
fix
(either
let
coeffi-
us suppose
anhydrous or any
partic-
in a
state capable of equilibrium with vapor or with ice, will destroy the
possibility of such equilibrium at the
The
brought
that
liquid
//,
ice,
pressure.
is
in contact
salt,
Now
it
when
there seems to be no
We
effect of this
salts is
distinctly indicated
by the
in
zero and
is
hold
true.
But
which mg
in
by
is
this case
For
may
be
illus-
let
trated
this
purpose
IS.y
crystals of
>S,.
in
Pogg. Ann.,
ciii.
(1858j, p.
p.
564.
196
J.
W. Gibbs
stant temperature
constant value as
changed.
we
If
0,
and
that
as having the
same
of
may be regarded
the values of
t,
any con-
tor
p, and
being un-
Then
when m^
ia
we have by
(210), approximately,
m
1.
e.,
(214)
If
we
/^2=^4
j5 like
will
1og-^,
(215)
have a positive value depending only upon the temperaAs this equation is to be applied only to cases in
is
as constant,
ni^,
we may
and write
/i2
= ^log-^,
(216)
body
tities
and
6'
will
temperature and
of the
/S^
the body
/a^,
if
it
is
iS^
composed.
If
W. GihbsEquilihrium of Heterogeneous
J.
Hence,
/S,
as
if
197
Snbstances.
body
are
negative values,
/S'^.,
we may
by equations (subject of course to the uncerof the assumptions which have been made) of the form
potentials for S,
tainties
S,i,
tS^.
/^;,=
(217)
198
JV.
(T.
these
may
cules
which
Gihbs
be more complicated.
by mole-
differ in the
not in the kind of matter, nor in the proportion of the different kinds.
In
still
which
in
body
may be
The
in the con-
These
which depend upon the
which the molecules are combined to form sensible masses.
manner in
The latter do not cause an
fundamental equation
is
increase in the
but they
may
number of
variables in the
set of values
of
perhaps being for a gaseous body, one for a liquid, one for an amorphous solid, and others for different kinds of crystals, and all being
invariable for constant values of the above mentioned independent
variables.
B\it
it
its
thermodynamic
increase in the
any
For example,
if
by the
quantities of
number
we should
mixture of
J.
Substances.
199
vapor of water and free liydrogen and oxygen, which we have just
considered, to a point at which the numbers of the] different sorts of
and the total quantities of hydrogen and of oxygen which are present,
the fundamental equation of such a mass would involve but four inde-
body
as determined
by
its
difliculty
there
indeed nothing in
is
the
physical
and
chemical properties of such bodies, so far as a certain range of experiments is concerned, which is different from what might be, if the
concerned, represented
which
fundamental
satisfy the
has a
maximum
ultimate analysis of the body, with the same energy and volume.
fix
our ideas
let
ultimate components S,
and
and
S,,\
?,
let
w^^j,
aS\,
.
S^
To
and the
m^ denote the
It is evident that w?
.
m.,^
are
* The terms proximate or ultimate are not necessarily to be understood in an absoAll that is said here and in the following paragraphs will apply to many
cases in which components may conveniently be regarded as proximate or ultimate
lute sense.
in a relative sense.
200
J.
W. Gibhs
The phases
ponents.
^,
m,
jo,
ni,,
m,,.
7n,,
or,
?;
as they
a maximum
may also be
is
is
w,
of
we may
is
or for which
readily determined
m,
w,
f,
m,
in question are
values of
for constant
phases by substituting
yUj,
/<
These phases
may
That we
which we
have used a similar term before, with reference to a different kind of
changes, yet in a sense entirely analogous, need not create confusion.
are considering, phases of dissipated energy.
It is characteristic of these
of i
in
any
real
its
phases that
mass
we cannot
volume of
Hence,
But when the proportion of the proximate compoits temperature and pressure
small body.
contact
may
Indeed
it is
by such
the same kinds of matter which also occur in other sorts of molecules),
but which differs from the first fluid in that the quantities of the
various kinds of molecules are entirely determined by the ultimate
composition of the fluid and its temperature and pressure. Or, to
speak without reference to the molecular state of the fluid, the result
considered would doubtless be brought about by contact with another
fluid,
which absorbs
all
the
proximate
components of the
which there
first,
lence in respect to their ultimate analysis), independently, and without passive resistances, but for which the phase is completely deter-
W. Gibbs
J.
mined by
its
its
By
it
is
aS\,
8^ independently
by
possible
change of
An
the
and without
meant that when the absorbing body is in
201
shall
it
be
all
is
The term
When
sidering.
out limitation
it
we
are con-
body has the property of reducing another, withwith respect to the proportion of the two bodies, to a
a
may
be called
a.
in
is
entirely determined
its
by
its
ultimate compo-
by the
quantities
may
be completely defined by the quantities of certain proximate components with the temperature and pressure, and
just mentioned,)
The latter are the phases of dissipated energy (with reference to molecular changes), and when the
more general form of the fundamental equation is known, it will be
to the final phases of equilibrium.
it
phases alone.
Now,
III.
26
February, 1876.
202
J.
W. Gihbs
may
not
still
have a theoretical
significance.
may
If so,
assist
all
is
of experimental demonstration.
is
is
body.
common
temperatures and pressures, likewise regarded as constant, the quantities of these proximate components are not capable of independent
variation,
state.
one of these states from another will not therefore involve any discontinuous change in the values of any of the quantities enumerated
in (99)-(103)
on page 143,
if
m,, m^,
m,
/^j, /<2,
yWn
are
W. Gihbs
J.
Equilihriun)
of Heterogeneuus Substances.
203
Therefore,
equations will agi-ee at the limits dividing these conditions not only
in the values of all the variables
which appear
in the equations,
but
also in all the differential coefficients of the first order involving these
We may
variables.
of
t,
and
I),
Where
by
for a
mass
in
'Q
will not
is not determined
be determined by these variables,
and the points representing connected values of <, jo, and I will form
This solid will be bounded in the direction opposite to that
a solid.
is measured, by a surface which represents the phases of
in which
'Q
In a part of the figure, all the phases thus reprebe permanent, in another part only the phases in the
dissipated energy.
sented
may
bounding
figure
and
surface,
(for
any phases
of which the
may be
existence
is
no such solid
experimentally
its
normal
at the limits
dividing the different parts of the figure which have been mentioned.
etc.,
be
but
if
different
we
sheets
representing liquid
and
We
fundamental equations
of gases, have an example of the derivation of the fundamental equation for phases of dissipated energy (with respect to the molecular
It will
be con-
in
116.
The energy
now
consist of
two
parts, one of
204
W. Gihbs
./
intrinsic
energy of
its intrinsic
Ds the
position in space.
this element,
plane,
its
will
fDe-\-fghJ)m,
in
(219)
all
and
d/De + dfg h Dm ^ 0,
(220)
These
must express that the entropy of the whole mass is constant, that the
surface bounding the whole mass is fixed, and that the total quantiWe shall supties of each of the component substances is constant.
equations
of
condition,
other
and that the
pose that there are no
independently variable components are the same throughout the
whole mass
and we
components.
Then
Dm = Dm^
Z>m,
(221)
+ SI>m.
(222)
and
dDm dBm^
6 Da
By
6I)ij
+ /ij SBm^
dJJv
-\- ,u
6J)m.
(223)
may be
ft 6Dri
-fp 6Dv
+//', 6Bm^
Dm
SDm
Jrfg Sh
Now
it
affect
will
-\-/gh
/// SD^n^
J^fgh dDm ^ 0.
(224)
diflf'erent
we must
have, sepa-
rately,
ftSDtj^O,
if
fdDrf=Q;
(225)
../.
W. Gibbs
Equilibriuiii of Heterogentoiis
fSubstaitces.
dihn
mivaried
is
+ fgh
dlJra^
205
(226)
^ 0,
/6Dm = U n
if
(227)
= Const.
(328)
We may,
motions as miglit
at
of indeterminate form.
we denote by
If
by
v the
volume, included in the part of the tube between one end and a
transverse section of variable position, the condition will take the
form
- fp 6dv + fg 6h dm ^ 0,
(229)
in
if
we denote by y
fg 6h dm =fg -^
By
f6v
is
{dp
-\-
may
(230)
(231
g y dh) ^ 0.
(232)
arbitrary in value,
dp=~gydh,
which
0.
Sv y dv =^fg y Sv dh.
Therefore, since 6v
6v) =z
any point
(233)
taken with respect to the direction of the tube at that point. Therefore, as the form of the tube is indeterminate, this equation must
hold true, without restriction, throughout the whole mass. It evidently requires that the pressure shall be a function of the height
and that the density shall be equal to the first derivative of
alone,
Conditions (227)
equilibrium.
To
by
g.
contain
all
that
is
that
/^i
+ ff^
Const.
Mn
+ 9^=.
Const.
(234)
206
The
J.
W. Gihhs
expressions
pi^,
//
we have called
and which are entirely
nature and state of the
confusion between these
several components,
"the
when
if
by equations
(234)
may
we distinguish
The relations
level,
ence of
the values
of
at two
different levels, being equal to the work done by the force of gravity
xohen a unit of matter falls from the higher to the longer level.
The conditions expressed by equations (228), (233), (234) are
in
phase with any previously existing about the points where they may
be formed. The possibility of such formations at any point is evidently
is
determined entirely by
the phase or phases of the matter about that point. The conditions
of equilibrium in this respect have been discussed on pages 128-134.
But equations (228), (233), and (234) are not entirely independent.
For with respect to any mass in which there are no surfaces of discontinuity (i. e., surfaces where adjacent elements of mass have finite
diiFerences of phase), one of these equations will
the others.
which
Thus by
(228)
and (234), we
be a consequence of
obtain from (97),
may
tion
vdp-g {m
dp= - gy dh
or
+ m) dh
(235)
(236)
which will therefore hold true in any mass in which equations (228)
and (234) are satisfied, and in which there are no surfaces of discon-
(228)
shall
at these surfaces.
./.
W. Gihhs
due to the
in phase, is
made
207
elements of volume
tlie
these elements
Now,
is
as the
volume to vary
elements fixed
and therefore
by
it
will
to
make
these
tinuity, this
surfaces of discontinuity,
mass
would be allowable
it
If the given
in space.
But
plan.
if
It woiild therefore
be necessary to add to
oixr
general con-
elements of
avoided
if
we
surfaces,
necessity
which
we can then
is
sup-
pose that each element remains always on the same side of the sui'face
of discontinuity.
Method of
tJie
elements of
may be
of equilibrium
may
as fixed in position
be obtained
and
size,
if
we
how
volume
possibility of finite as
we
finite differences
sic
of phase,
we may
in the
form
fdDe+fABE,
in
which the
first
finite
(237)
infinitesimally varied,
If
by such
variation.
We may
all
will
we do
and
in intensity,
fT6DmfTADm.
(238)
208
W. Gibbs
J.
We
J'SDe
fSBi)
-\-fAI)i]
(240)
0,
f8Dm + fADm =
We
may
J'6I)m
0,
(239)
be
will
+ fADm^ =
(241)
0.
SDi
T 6Dm
T6Dt]
Jf, 6
Dm
and
-lfdDm^o^
finite,
Jlf 6Dm^
in
(242)
Dm
I,
by a certain change
we do not suppose the same element to experiand an infinitesimal change of phase, we must have
But
of phase.
ence both a
as
finite
6IJ^ rSDiu -
- 3I6J)m^0,
(243)
- MADm^0.
(244)
and
in virtue
{tT)
SD,f
(yu,
- r M,) SBm^
+
for
...
which
it is
{/J
r-
M)
SDm ^
(245)
T,
(246)
V*
//-r=jf.
(247)
intrinsic potentials
W. Gihhs
J.
Condition (244)
form
to the
(r+
to
Ji>f - tJD)] -
/<,
JI)ni^ ...
/JJI)m^0.
If values
209
(249)
ai-e
distin-
may be written
- //2>w'
Be' -
which
may
if
^'/^^
/'
(250)
0,
be reduced by (93) to
De'
Now
Dtf
tJJif
//,
7>>j, ...
Dv is
Dt\
i-iJJniJ
+ pDv^O.
(251)
Drf, lJm^\
J)m^
to be determined
(93) to
p'Bv +pBv^O.
That
(252)
is,
Ba'
- tBif +pBv' -
Z>m,', ...
yu,
u^BmJ^O,
(253)
values determined
the mass
is
128-134.
It
may
shown (see pages 160, 161) that for constant values of ^, /.i^,
//, and of Bv', the first member of (253) will have the least possiBm' are determined by a phase
ble value Avhen Be', />//, Bm^',
for which the temperature has the value t, and tlie potentials the
easily be
.
values //,,...
//.
It
will
be
sufficient, therefore, to
by
it
consider the
may be reduced
(93) to
p^p'^0.
That
is,
(254)
phase of the same components, for which the temperature and the
Tbans. Conn. Acad., Vol.
III.
27
April, 1876.
210
W. Gibhs
J.
is
('240), (247).
when
will
be at
is
not satisfied,
e(|uilil)riuni
if
subsists,
it
mass
itself),
is
completely deter-
at
is
For a
ture.
we may write
a and
^c
By
denoting constants.
=a
d8
t,
dt,
integration,
we
=CJ+E,
in which
ate t and
from
(11),
by these equations we
If
elimin-
we obtain
de =i
f-^ di]
,
s-B
dv,
or
ds
c
The
dv
= di) a
-,
v^
may
z=i
be written
its
of the equation to
volume.
the
volume
its
for t=z\
and p=.\
form
H,
a log V
T)
in the
if
11
we substitute -
in
for
f,
//,
v. respec^
J.
W. Gihbs Biquilibrmm
e
c loo"^
This
Em
cm
Heterogeneous Suhstcmces.
oj^
H^^ ^
If
,-.
a loo-.
_v
(2o5)
"u
in
211
'
is
invariable composition.
pearing
in
if we do not have
which forms the gas as ap-
be observed that
It will
E and
generality give to
solely as constituting
we may without
a state of jiurity),
(in
loss of
values.
which
60
and
components
for
the substance
or, if
which the
in
states of its
is
^"and Zf cannot
We
Avith reference to
i)-=:0
substance appears,
be treated as arbitrary.
in general
=
c^^
'Em
cte
f/f
:=
= man
^
ell]
/
- cE
+ \e~Em,
,^r-
civ
CIV
c+a
H
'/
m^ /
(tni,
(256)
^
'
Em,
p=ia
1.1=1
E^
may
III
s- (c
-\
m+
and
(87), (255),
//
and
= Em + c m
t
f.\ovt=
and eliminating
?/'
we have
i/,
= Em
+m
-\-
c log
"
ti<
Ie +
{-
-{
//-,
t,
v,
and
we have
^ //,
a log
H clog
we
a log
-^
m H
Eliminating
(257).
'
(259)
^
ii).
rii^
from equations
dih =^
(25S)
cv
We
Em
(257)
^
't
t -f
<i
'-^
log
(260)
|.
obtain
m/
IT
dv
cit
log
t -\-
a log
jdm
)
(261)
212
J.
W. Gihhs
?7=
(88),
m (^+
clog^+alog j,
(262)
P = -,
(263)
c-\-a-JIc\ogt-^a\og
From
by
(260),
'C,
= Em
and
(87)
mt
-\-
we
(91),
obtain
c log
we
(263),
(264)
\.
aT'?
a log
^^ v,
tion
?
From
= Em + m tic +
this,
by
differentiation
H - (c+a) log
a log
(265)
j.
(92),
we may
v= amt
if
log
(266)
),
/^
The
E+
tie
last is also a
-\-
^s
(267)
H-
((+ a) \og
fundamental equation.
It
a log
may
(268)
be written
in
the
form
log
or, if
^=
we denote by
+ ^^
a
=:
ae
+a
c
t
fiE
(2<0)
j,
/;,
p, and
-^-//+log^,
logf^4^=
* {c+a)m
m
a
(269)
^
at
may
also
be
it is
{c+a)
^
and
t+'-~-,
Hca
log
'^
(271)
./
W. Gihbs
213
It will
In
stants.
Hca
a
in
"
and another
~\~ O-
for
given
'
the above form, in order that the relations between the constants
The sum
c+a is the
difierentiate (266)
We may
if
we know
may
k.
specific
of a
we
as constant.*
saturated vapor,
regarding^ and
by
V.
we
and
shall
liquid
d?/
dt,
whence
?/
= A-log
Also, from
a constant.
dfi
-(k
log
+ H',
tliis
equation and
(97),
whence
/n
= kt-M log
constant.
t-H'
This
is
t-i-Vp
+ E',
(A)
stance in the gaseous state, the two equations will both hold true of coexistent liquid
and
gas.
Eliminatmg u we obtain
log
If
we
pa = HH' +a kca
A, B, and
C denoting
state, the
If
EE'
logi!
V
a
at
p
^.
t
is
we may
i)=AB\og
constants.
k~ca
write
-,
we make
coexistent solid and gaseous phases will of course have the same form.
tent with
two
and density.
In this case
/I,
and
and
fi.,
//;,
= k't-k'l
=
we may
log
t-H-t+
V'p-^
E,
= E+t(c + a H
(c
+ a)
log
+a
log
j.
214
W. Gihhs
./.
Equillhrium of Heterogeneous
Substances.
Now
if
entirely determined
is
quantity
+/
of the second
it
will
of the
/I
first
of constant
by a
single
that
=(1
/":j+A//,
+>^)/I.2.
we
given above,
fi,,,
^=^-
log
^+
C log ^ -
^,
when
^+^-g'- (l+^
A-
-g''-c-a-Afc' +
B (l+?i)k"-a Wca
(l
+ X)A;^-
E+lE'-{\+l) E"
(1+A)F"-AF'
We may
a.
by
its
specific
it
may be
applied to
C,
and
I)
must
in this case
be under-
stood to denote quantities which vary with the composition of the liquid.
consider the case more in detail,
"^
m
where
H'.
k,
u=kt-kt\os t-H't+
we may
k,
H, V, and
components of the
dU.
dV
dm-i
arw,
to
which
the several
dE
P+ a?n,
T^
-jt+ -,
fflTO,
component
(92),
dk.
dk
= dm^
~jt' tlogt
^
If the
to
Vp + E',
liquid).
//,
.shall
But
(a)
write
C
where
by
E' denote quantities which depend only upon the composition of the
F,
Hence,
liquid.
we have
which
we
have by (269)
log
Eliminating
;"
,
i
we
^pa =r
Hca
+ a^
log
<
H-\E
+ '-i
at
it
log-=^-51og<^+ D^,
a
in
which
A, B, C,
and
Z>
^ dU
dk
dm^
dm
/f^k
B=l'
a dm
\
c-a).
,
\
I
J.
W. Gibbs
We may
variable (m).
215
It is a rule Avhich
if
many
cases very
sub-
equal to the
sum
in the
gas-mixture
same
Now
which
it
yields in the
when the
when it is
liquid or solid
in
in
is
equilibrium with
its
own gas
The
alone.
difference of
the pressure in the two cases will cause a certain difference in the
values of the potential, but that this difference will be small, we may
infer
i^
=('*)
(272)
which may be derived from equation (92). In most cases, there will
be a certain absorption by each liquid of the gases yielded by the
others, but as it is well known that the above rule does not apply to
cases in which sucli absorption takes place to any gi'eat extent, we
may
which we have
to
do
effect of this
is
of secondary importance.
may
If
due
we
neglect the
to these
circum-
be expressed as follows
With respect
may compare
to
some
am
a dm.
of the equations
"On
Saturated Vapors,"
Pliil.
Mag.,
vol. xxxi.
civ. (185.-^), p. 6] 2
(1866), p. 199.
216
J.
W. Gibbs
loops one of the liquids, and above these liquids the gases which they
We
may
Now
it is
same
at the
level,
in
sum of the
in
the form
measured at the same level. Now the experiments by which the rule
has been established relate rather to the gases in the vicinity of the
Yet, although the differences of level in these
may
all
cases
which can be regarded as falling under the laws of ideal gases, for
which very great pressures are not admitted.
If we apply the above law to a mixture of ideal gases and distinguish by subscript numerals the quantities relating to the different
a,
a,
p=^,
,<
]
\i
(273)
it.
In particular,
it
By
differentiation
I,
V
f
H,Cjaj
c,
fi^Ei
^1^
J.
W. Gibhs
217
H,
V
H9C9
(275)
H-i-E-i
V
etc.
I'ehition
the density of any component, and the potential for that component,
not affected by
tlie
is
They may
also be written
/^i
etc.
(276),
(276)
Eliminating
and
we
from
(2 73)
obtain
(277)
(278)
gas-mixture
is
equal to the
if
sum of
same volume at
Equation (278) expresses a similar principle
existing sepai'ately with the
From
(276)
tion between
values of j(?,
ij)=
If
and (277) we may easily obtain the fundamental equat, v, m^, ni^, etc.
For by substituting in (94) the
1/:,
2J^^m^-\-ni^t
Cj ^i-Cilog
-f
we
obtain
a^log
j).
(279)
as constant,
this equation
for
^1 ^1,
for
am,
for
Em
for
losf /n
for
and
The values
Hm, - a m
of
c,
a,
E, and H,
will
As
III.
(c,
m,),
^1 {H^ ni^a^ mj
log m,^).
m will
denote
April. 1876.
218
W.
J.
form of
(xibhs
(260),
it is
when
changed, will
have
the proportion of
its
all
e=^^'^-\
ni
(280)
and
c'
+ a=^\''^-^^l^-^i.
(281;
We
m, and
/<
Cj
j, Xi,
letters
indicate
rately,
and extend
have by
this notation
(273), (274),
P=^iPi,
whence by
'/=^i'U,
quantities p,
we
shall
t=^i'l\\
(282)
l-^^',;,.
(283)
and (91)
(87), (89),
f=^'ifn
The
and (279)
>/,
A'=^'iJi,
//,
f,
j,
'C,
may
be attrib-
if
that
to be
sible for a
is
equal to the
by
sum
itself at
of the
the same
For let
temperature and with the same value of its potential.
relating
to the
be
defined
as
etc.;
etc.
P2i
Pii V\i ^1^ 'I'l^ Xn ^1
l
different
perature,
and potential
by
itself
as in the gas-mixture
= ^iPi,
if
W. Gibhs
J.
and
tlierefore,
by
whence
dt
j/^i,
'\ dtj/i,
f^"
r/^,
nt
in the
relate, is
etc.
^.
219
Also
"
by (93)-(06),
also,
All the same relations will also hold true whenever the value of
for the gas-mixture
equal to the
is
component
sum
//
itself
in
For
etc.
by themselves, we
Xn
'/';,,
?i
P2'>
have
shall
whence
/dif^\
Therefore,
by
mixture and
=(&1l\
in the
gas
6r,
More-
over,
\dt
/ij^ ffi
dt
/v,
whence
when the
*
initial
subscript
is
is
m.^, etc.,
applied, is to
unless
its
in this
paper
to indi-
be regarded as constant
in the differ-
220
W. Gihbs
tT.
are the same, (whether these gas-masses are entirely difterent gases,
is
Avill
initial.
we
If
in equilibrium,
two
different points
^^,
From
is
^''
=e
*''
(284)
which we may regard the quantities distinguished by accents as constant, it appears that the relation between
the density of any one of the components and the height is not
this equation, in
aflfected
process of com-
when
is
used are
all
the same,
final gas-masses.
It is
to the
sum
of the quantities of
We
will
now
ip
equal
which
it
in contact
may
may
first
is
This
may be
*This
It
will
way from
when the
liquid
is
pro-
by Lord Eayleigh,
assumed
W. Gihhs
J.
221
The
substance which
it
same substance
in the
and
there-
of this substance in the gas and the part of the gaseous pressure
e.,
in
remain unchanged.
the potential for S^
the index
(l)
If
we denote
by dp and
we
<?/<,,
have by (272)
shall
\dmjt,p,m
\d2)jt,m
it is
affixed refer
to the liquid.
necessarily the
is
common component
as if the
(//<j
(^(
itself
log
at the
/>j.
Therefore
(dv
\^^^
dp.
dmjt,p,m ^-;
This
may
be integrated at once
approximation.
so accurate, if
member
We may
we
j/^
we regard
as constant,
the differential
which
vrrite the
>S^
p,
and
(287)
ip-p')-
in the gas,
dv Y^^
i^i-Pi riLT
I,
\uni^jt,
pp'
coeffi-
(dv
where
if
(285)
p when
the insoluble
j)y
relates.
is
not neces-
222
W. Gibbs
J.
Equilibriu
common component
we may
are others,
of Heterogeneous Substances.
in
If there
ponent of the
We
liquid.
gas which
effect of a
extent,
may commence by
absorbed to some
is
be regarded as a com-
in strictness
considering in general
like the
previous,
and
d})
;/\^^ d^i^-^-y'i'^
d^i^
whence
{yf^-Yx)
Now
if
the gas
d^'
u
,
is
dp,
a.t
=z
<^Ml
(/2-H'^O
^/'2-
an ideal gas-mixture,
^
-^ dp
,=.
^^
P,
dp
-^ , =
= a^t
-,
-,
and
//
r.
-^-^,
P2
72
therefore
^
We
liquid,
\\dp^=\\-^\
may now
suppose that
and
a gas which
S.^ is
In such cases
extent.
it is
/S',
(288)
is
well
is
dp^.
known
of the substance S2 in the liquid and in the gas is for a given temperature approximately constant. If we denote this constant by ^,
we
shall
have
lU
It
Udp^={\- A)
this
equation regarding
/>j
is
(28P)
dp,.
;/j
as variable,
we
(L)
We
accuracy
if
we regard y^
as well as y^
as con-
(--'-l)(Pt-^i')=(l-^l)
P2^
('^90)
where ^j' denotes the pressure of the saturated vapor of the pure
It will be observed that when .4= 1, the
liquid consisting of S^.
presence of the gas
gas S^.
When
greater than
if
/S'g
J.
The
properties of
jui ide:il
223
As
than the
little less
sum of the
pressure of the
its
make
it
little
more, when
is
insoluble,
it
quantities concerned
is
this circumstance.
It will also be observed that the statement of Dalton's law Avhich we
have adopted, while it serves to complete the theory of gas-mixtures
by the
in
we
show
this,
denote the
aSj
will
common component
gas
is
in equilibrium
by the index
(l).
solid, let
it
quantity which
mass
it
increment
\dmj
and the pressure
dm 2
increment
dp\^^^
\dm2lt,
Now
V.
t,
V,
^'
dp/t,m
ju^
\dmjt,p,m'
and
/:>.
But by
(272)
224
IV.
T.
Gibhs
Therefore,
dm 2
ft, V,
dp yG)
Xdm^Jt,
It
will
first
V,
member
Now
gas.
first
ber of this equation must therefore have the same value for
liquids or solids
which
which
quite inadmissible.
is
solid
identical in
is
it
is in
and liquid
water),
it
the same
when
states (as
Hence, Avhen a
components both
its
a moist gas
is
in equilibrium
solid
in the solid
density.
The foregoing
immaterial whether
(273), or
such
gas
mem-
all
by
we have
chosen.
It is of course
(279), or
and (271)
=7+
};
= 2jB,m^+
may
easily
^\ (. -,
log^.^^^
m, (c,+,
They
are
-^,..,),
(292)
log
(293)
- ^,)
j
- ^\
(c,
m,
+ a,
The components
m,)
log
+^\
(a,
m,
^-^^^^).
We
may
(291) (292), 293) refer, may themselves be gas-mixtures.
binary
gas-mixture
equations
of
a
fundamental
the
apply
for example
W. Gibbs
J.
225
to a gas-mixture of a smaller
The necessary
But the
number of components.
to the gases of
It
air.
is
the
components.
its
vapor of water
all
is
This
is
certainly true
having the
when the
quanti-
of
its
Such equations
as
potentials in pure or
(264),
(268),
(276),
and
Solids.
For
is
quantities
not confined
common
to coexistent liquid
and gas-
eous masses have the same values in each, these expressions will
generally afford the means of determining for liquids, at least approximately, the potential for any independently variable component which
is
gaseous mass,
of the liquid
alone,
its
it
For although
is
volatile,
temperature and
is
to
bring
it
by a
change of pressure
May,
1,s7<;.
226
J.
W. Glhbs
may be
The
volatile
tiie
pressure
general be quite
in
Avill
trifling
as
may moreover be
composition, and
equation (272).
The same
readily estimated
by means
of
solid.
As an
application of
in liquids,
let
absorption of
make
gases by
us
this
m'-^^
between
its potential.
in
the
ni^l'^
common
to the gas
i)art
and
When
let
the
the
liquid, v^^^
and
/<
we have by Henry's
law
,y(L)
(L)
where
^1 is
and
be seen
equation
^^
,,(G)'
y
ni
is
(if
and by (-76)
J^+ (Mog-^,
(295)
Therefore
=S+6Mog A-,-;,,.
y(L)
(296)
we
'
(G)
A.,
It will
'
'^
I<,
jB
which
were regarded
and pressure, but since the potentials
by the
pressure,
we might
anticipate
that these quantities in the case of liquids might be regarded as functions of the temperature alone.
./ TV.
111
Gihhs
(218),
(21'7),
227
that by (204) and (276) they are shown to liold true in ideal gases or
gas-mixtures, not only for components which form only a small
])art
of the whole gas-mixture, but without any such limitation, and not
It is noticeable
when
we may regard
When
small.
question
in
is
is
is
sufficiently
and
C may
not be
mass
we have just
These equations
components.
also
will
of Entropy due
Mixture of Gases hy Diffusion.
From
may
we
equation (278)
easily calculate
the increase
to
ai'e
the
of
mixed by
Let us suppose
that the quantities of the gases are such that each occupies initially
If
we denote
this
volume by
T^,
the
m^
a 2 log
T^
m,^ j log
w/g ^2 ^^g"
or
{ni
XT
JSlow
m,11
a.
=z
(/j -1-
P^
m^
(f^)
and1
2t
may
log
m^^
~?
z
z
2.
a,
^
= ~P^
It
sion
pV
-.- log
It is noticeal)le that the
2.
(297)
if
228
J.
W. Gihbs
of diflferent kinds.
we mean
is
entirely impossible.
We
call
as
tion as at
It is to states of
of thermodynamics relate.
But
if
why
is
such a case as
we have supposed,
is
gases.
Now we may
are embodied in our eqixations suppose other gases to exist than such
as actually do exist, and there does not appear to be any limit to the
resemblance which there might be betAveen two such kinds of gas.
But the increase of entropy due to the mixing of given volumes of
the gases at a given temperature and pressure would be independent
We might
of the degree of similarity or dissimilarity between them.
should
be
absolutely
of
two
gases
which
the
case
identialso imagine
(sensible
and
molecular)
which
properties
come
all
the
into
cal in
gases
either
pure
or
mixed
with
each
exist
as
while
they
other,
play
J.
TT\
Glhhs
229
its
wliicli
any
increase of entropy.
is
tlie
in virtue of their
is
of
separation of a homo-
tlie
geneous gas into the same two parts into Avhicli it has once been
divided, after these have once been mixed.
In other words, the
impossibility of an uncompensated
reduced to improbability.
There
is
perhaps no fact
in the
pV
in
quantity -
(2f7)
And
is
therefore
independent of
is
them.
The
result is of the
If
different kinds
ai'e
we denote by
-y^,
of gas, and by
increase of entropy
may be
if
we denote by
V-
where
Cm^
we
,,
denotes a constant.
V
2^ (m,
loguj.
shall
r,
Cm^
a^,
etc.,
Hence
V: '.m^a^: 2^{ni^ay)
a^
And
volumes of the
V as
may be
r,
^',
?,
written
2,r^\og2,r, - ^\
(r,
log r,)
^
230
TK Gihhs
J.
We
will
now
loith
(see
components the gases G ^, G^, and G^, the units of which are
denoted by @^, (S).^, 3, and that in ultimate analysis
@3
A
= A, ,
+A22,
(299)
Ag
1.
The
phases which we are to consider are those for which the energy of
the gas-mixture
is
minimum
in
ultimate analysis.
^'it^-^O.
(300)
(86),
4- /'o
yUj 6rti^
f^""'o
Ms
/<j
therefore
Ag
= Ms-
//2
(301)
^1 i log
"^+ ^2 2
log
~-
(>s
log
'^
= A^B\ogt-j,
(302)
where
A=\^H^ + ^2^2 ^3 -
'^
^'
1
^ = A,r,
- ^^2^2 + ^3 - ^
-f Ag Cg
we denote by p^ and
(d^
- ^ 32 + 35
C3,
(303)
(304)
^3.
(305)
the gases
gas
6^3,
we
shall
have
/i,
and (302)
will
= -^\
and
/:/,
=r
^,
(306)
log
'
,3
^=r
_i_
logi5
(30/)
J.
W. Gibhs
231
Moreover, as by (277)
= (i rn^
pv
we have on eliminating
^1
;/t3
m^
12
+ a^ m^)
m^
(308)
t,
"'2
m, H-
[a I
+ 3 Wg)
= ^ + ^^^^'-.71'
O
tC o
'-f'o
('*'')
t'
where
B'=X^
+ A3C2
C1
- <'3H-Aii
Agtt^
3.
(310)
be observed that the quantities /S^j, /^g will always be positive and have a simple relation to unity, and that the value of
/?j + /^2 * 1 will be positive or zero, according as gas 6^3 is formed
It will
of Gj and
G^ with
or without condensation.
we should assume,
If
according to the rule often given for the specific heat of compound
of the gas
6^3
quantities which
would be
Gq
equal to the
is
it
zero.
in
and
6^35
+ A\)-\-X2
(cot
+ B.J
Ji
by
and (257)
(283)
A,
(^c,t
or
which
Avill
reduce to
specific heats
is
the second
Bt + C,
C when the above
satisfied.
evolved divided by a^
member
(c^t
E^),
t^ will
t.
Moreover,
the heat evolved in the formation of a unit of the gas G.^ out of the
which
is
G2 imder
t
4- 0-\- A
It
a,
-I-
is
-a^tB't-\- C,
for
constant pressui'e
t,
multiplied
by
any given
finite
values of
;;/j,
a,^
in
m^^
member
of
t'^
the case
tn^,
and
when
y6^j
/ig
(infinitesimal values
gen, and vapor of water as an ideal gas-mixture, for a mixture containing any given quantities of these three gases at any given tern-
232
J.
W. Gibbs
EqiiUihrium
of Heterogeneous Substances.
of platinum.
(If
water into
its
components.)
temperatures, except
oxygen
is
It
may
of dissipated energy
entirely
when
is
short of
what
is
may
it is
is
lie
It is also to
verification.
so unfavorable to
any
But with
is
If
^=
(a case, as
0,
and the
have given
G.
If this
quantity
G^ and G2 and
an infinitesimal value of
is
positive,
also the
t
we
volume
shall
have
(with
W. Gibbs
J.
shall
values of
is
233
finite
The
we have considered
case which
may
is
ture,
fact,
easily
be generalized
this respect.
in
In
in a gas-mixture, if
may
form
A,
+ A2o +
@,
A3
@3
etc.
(311)
0,
where @i, 2? ^^c. denote the units of the various component gases,
and A,, Ag, etc. denote positive or negative constants such that
^\
A,
0.
From
we may
sipated energy,
A,
/<,
Ag
-f
^lo
A3
/<3
:^\.(A, /^)
or
etc.
0,
(312)
0.
Hence, by (276),
^^(A.ilog'^) = ^ +B\ogt~
B and
where A,
(J
= 2,
(313)
by the equations
-4
-,
(314)
B=z^,{\,c,\
G=2,{X,E,).
Also, since
^1
(A, i
j)
logw,)
- 2
v =^
(A^ a,)
H- ^^ (A
(315)
(316)
(a
nt
log^\
,) log
t,
^ )
(a,
m,)
p = A-\-B'\ogt--,
(317)
where
(318)
If there is
ti'ue
It will
dis-
not affected
by the presence
III.
May,
1876.
-234
J.
From
W. Gibhs
it
follows that
if
there
is
a phase
The equations
mixtui-e, the
same
potential,
and
if
we
com-
ci
write
/i
for
volumes of
log
m/
i
m^
where
131
_ A_
d
_^
.'
B
_ lop2^
;;
(J
<(2 f
W. Gihhs
J.
Eqiiilibriuut, of Heterof/eneous
235
Sv^hstance^.
the
that there
in
is
perfect freedom
in
brium
is
(l)
we have
it
follows at
been deduced
Tf
in
it
we
might still suppose the particular phases which alone can exist to be
determined by some other j)rinciple than that of the free convertibility
of the components (as if, perhaps, the case were analogous to one
of constraint in mechanics), it may easily be shown that such a hypothesis is entirely untenable, when the quantities of the proximate
components may be varied independently by suitable variations of the
temperatui'e and pressure, and of the quantities of the ultimate components, and it is admitted that the relations between the energy,
entropy, volume, temperature, pressure, and the quantities of the
several proximate components in the gas-mixture are the
same as
for
by m^, m^,
and
let
etc.,
etc.
{n denoting a
its
s, ?j,
v,
etc.,
is
entirely determined
6, 7,
by
v, t,p.
We
to the quantities
mj,
1112, etc. in
the gas-mixture
us imagine the
let
surface (see
let
us
i.
e.,
for every
body
in2
etc.
It follows
would be
immediately
from our supposition, that every point in the v-rj-s surface relating to
A must coincide with some point of one of the v-t/-e surfaces relating
236
W. Glbbs
./.
B not only in
to
Jqullibrmiii.
of Heterogeneous Substances.
therefore the
must be tangent to the various y-r/- surfaces relating to B, and therefore must be an envelop of these surfaces.
P'rom this it follows that the points which represent phases
common to both gas-mixtures must represent the phases of dissipated
euergy of the gas-mixture B.
i)-t]-e
surface relating to ^1
The
properties of
are
assumed
in
It is usual to
assume
= ^,
in
(c,
;//j
4-
/// ,
E^).
(324)
ponents,
(i. e.,
take place
if
among
the components,)
in
it
will
be
its reverse,
sufficient to
can
assume, in
equation (277) and of the following, which can be derived from (324)
by
differentiation,
and comparison
Avith
a necessary relation,
[tdtji^dv -
We
shall confine
:^^ (Cj
= 0*
m^)dt\,
is
that
It will
if
(325)
be observed
the gas-mixture
is
proximate composition, the heat absorbed or yielded may be calby the same formula as if the components were not conver-
culated
tible.
* This notation
is
etc. are
ulti-
regarded as constant
J.
TT''.
supposed constant).
we use
the same
287
method of geome-
mate,
ai"e
trical
If
v--q-B
sur-
and position of
sition of
M.
occur
when
composition
the limit
is
is
uniform.
same
component?
composition.
It is
convertible
as
before,
tible
components.
tangent to
all
J!/,
in
which
may
be capable of experimental demonstration, the former surface must be an envelop of the latter surfaces,
laws of ideal
vertible.
It
is
its
density
W. Gibhs
Eqii'dlhriiuu
238
J.
NO2
of Heterogeneous
N^O^
is
l^ubstances.
from the depth of the color on the supposition that th^ absorption of
due to one of the components alone, and is proportioned to
the separate density of that component.*
MM. Sainte-Claire Deville and Troostf have given a series of deter-
light is
minations of what
we
it is
viz.,
We
usual in treatises on
air,
or
more
strictly,
an ideal gas which has the same density as air at the zero of the
by these determinations,
it
In order to
be convenient to
will
As
pressure
may by
(263) be expressed
I>=
Now
by
may
by
tiie
formula
the relative
be expressed by
(m, -i-mo)'^.
(326)
(263)
pV
By
tions,
i>i
= -,
1^2
(328)
we
shall
= 2J\,
(329)
have
,=:22-
From
(326)
'
*Salet,
f
(330)
we have
"Sur
la coloration
Comptes Rendus,
J'
J.
and from
(327),
by
2
and
(328)
m,
239
^^vbstances.
(330),
= Z>o ^- 2 i5, ^
hi.,
whence
m.
m,
By
(327), (331),
= (i>,
-i>)|^^,
(331)
= 2(i>-/J.)|^^.
(332)
log
*=
= 2
^^
2 (Z) - Z> J
-r.\
log
(333)
a,
'--^
we
common
introduce
logarithms (denoted
by
log,^)
instead
instead of
2^
a, to
instead
t,
the pressure in
t,,
if
of
'""%i~Jj- = ^ + I -S."
('
+-^)
- M^3-
('^'^
A and C denote constants, the values of which are closely con^ and C
where
D,
14 4. 32
=z-
--
.0691
64
^ = 28 +
= 1.589, and B^
.0691 = 3.178.
2
The determinations
'
of
MM.
(335)
^
2>)2 ,
3118.6
,^
which gives
i>=:3.l78
log^^
where
In the
umns
first
(->
W= 9.4/056
\/<'^
(3.178+0)
3118.6
- fTr^~^^Si oPa,-
MM.
ments of
from these
240
J.
densities.
in
W. Gibhs
It will
Equilibrium
of Heterogeneous Substances.
no case reaching
t,.
J.
W. Gibhs
peroxide of nitrogen by
(3;^3), (334),
and (336)
241
be attected
will not
Wanklyn
of Playfair and
above
The
table.
given
are
the
in
latter
part
of the
We
the pressure due to the free nitrogen from the total pressure.
may
in the
reduction
relative
densities
phere.
The reader
of Playfair and
Wanklyn
the
and
calculated
there
is
observed values of
Z>,
between
these numbers.
fourth observations
density 1.783
is
first
and
We
must
also
by
equation (336) for the temperature of the experiment and the pressure of one atmosphere.
established
For
directly
in seven
Now
by the
the
number
1.687
we may regard
experiments of Deville
by
by experiment,
as
and Troost.
we accept
is
less
than
.01.
the effect of
As
this result is entirely at variance with the facts observed in the case
Wanklyn,
fi'om
it
the three
other determinations
of Playfair
and
III.
?,\
May,
187(5.
242
J.
W. Glbbs
The
observation.
first
cannot thei'efore
series
Similar considerations
experiment.
if
two
tions in the
we admit
first
last
may
By
compar-
experiments
easily convince
the series, which have the greatest weight), the last determination of
relative density 2.588
must be too
In
small.
fact, it
should evidently
we
is
as
good
as could
be desired.
Nor
will
the admission of errors of .152 and .120 (certainly not large in deter-
experiments.
It
not
accurate.
In
the experiments
of Deville
and Troost
the gas was subject to the actual atmospheric pressure at the time of
the experiment.
The
Wauklyn
not given.
is
In the ex-
The numbers
in the
column of
pres-
sures express the fraction of the whole pressure which remains after
given
in the
Now
y'g^^
But no indication
is
in the value of
it
may
easily
be shown that a
vai'ia-
In any
as calculated by equation (336).
than .005 in the value of
Wanklyn
variation
of
more
of
Playfair
and
a
experiments
the
of
is
J.
W. Gibhs
periment of this
sacrifice of
series.
It
seemed
best, however, to
make
243
a trifling
respect,
and there
Any
values of ^, B'
this
in
is
and
in
obtain
B'
C
,^
A+
logjo - 7 = 9-4V056
^
3118.6
may
following conclusions
By
of this relation.)
dift'erentiation
we
obtain
J/:S'C_ 3118.6
where Jf denotes the modulus of the common system of logarithms.
Now by comparing equations (333) and (334) we see that
^=
C
Hence
B'
which
may
t-\-
MC =.43429
C
G 7181 2 = 3590
rtj,
40*^^
or
50*^,
and
mentioned.
Now
B'
version of a unit of
t -\-
NO2
into
NgO^
by the con-
Such
B'
we
shall
But
since
B-\-a^
by equations
= ^ + ii,
40^^^
Now
t -\-
(322)
a,.
transformed into
NgO^
when
a unit of
NOg
is
It therefore
244
J.
W. Glbbs
when
external forces
NOg
would be more
NO^
if
J5zii 0,
^2^4
and
is
i.
e., if
^^
^^'^
same.
This assumption
B' :=
If JB
0.
0,
H'
= u^.
we may
Z>,
If
we
p, and
t,
substitute
6.5228
2977 4
log,
(C
+ 2V3) ^ j^^^Tf:^'
calculated
any
case, or
It is to
by more than
relation expressed
we
by equation
be noticed that
if
which
is
evidently equivalent
to an equation between p, t, v, an<l ni (this letter denoting the quantity of the gas without reference to its molecular condition), or if we
wholly of
is
may be regarded
sufliciently rarefied to
NOg, or,
when
as a constant
be regarded as con-
is
its
but
entropy
/;
entropy
Now
by
//'
(88)
and
real gases,
./
W. Gihbs
^^
245
(387)
therefore
dij
do
7lt
_ d
~ lit
_
~
dr]
dv
drp
dt
~dt^'
dj}^_d_
d_
dv
?/',
we
'
ns8^
^
obtain
dr[
dv
dt
_
~di ~
d dp
dt''
which must hold true within the given limits of temperature and
density.
Now
it is
granted that
^-'#'
(3^0)
^
dt
dt
'
limits,
From
and
have
w^e shall
dt}
dt}'
dv
dv'
may
?/',
is
it
evident that in
calorimetrical
given values of
all
and
v.
in the real
we
relation
de-=.
dr]
p dv,
in like
manner
identical.
Hence
We may
the
use the equations
of the form (312) to eliminate from the equation of the form (273)
phases of equilibrium.
For
this purpose,
we may
246
W. Gihhs
'/
convenient to regard as
is
it
may be denoted
by
//,
Ci+a,
//
p:=za^L^t
will
be
',
C2-\-a.2
-\-a.^Lt
E-i
fi
(342)
where
L,
From
by
this equation,
X.^e
""'
"^
(343)
differentiation
(98),
we
obtain
n-E,
c,
+2-f"
^2(^2
Ci
fi
E,
-=L^t'e
From
(244)
IIE^
^
(345)
following
is
easily obtained,
II Ey
Cj_
- = Z^{c,t+E^)t'''
We
L^t
',
^5
Ci
-\-
may
t,
p-a,t'^={a,-a,)L^t
(i -
a^)
^2
(346)
by eliminating
may proceed
"'
C2
,
and
v,
a^t
""'*
+L^{e2t+E.^)t'''
//
rtl
ii-Ej
Cj_
"'^
as follows-
"'^
+ a-z
(347)
li^E-i
rt,
a,.,t
"
(=^48)
'>
^) "
2 log
^^
'^*
" ^^
)
= (^1 ~
^^s)
jp
4-
a^
~''2)log
^^S (^ 1
2)
-pi
~T^
'
^^^^^
Gihhs
J. IV.
By
lent to (833)).
when a,
(347)
=2 Wg
247
easily eliminate
fx
from
(346).
The reader
will
in
with com-
that the equations derived from (342) do not give the quantities of
any
(350)
_ L,{c^t+E^)^Lo{c^
-J-
-I-
t-\-
E^)t
""
""^
(351)
CjCi El Eft
at
a
By
amount of heat
will
values of J, as defined by equation (89), for the initial and final states
of the gas. From (89), (350), and (351) we obtain
E1E2
C2C;
L,{c^t^ at^E^)^L^(c^t-^at^E^)t
L^^^LJ,
By
differentiation of the
two
last
"
''^
._^.
equations
we may
obtain directly
the specific heats of the gas at constant volume and at constant pres.
sure.
between
its
components
is
248
J.
p = a^
di
+
where A^ and
X.^
Iiave the
a.t
a,
Substances.
0.2
cb.,
'
'
a.,t
230.
{To be co7itinued.)
ERRATA.
Page
"
Page
m, read
to,,
^,.
to_j read//,,
-^.
a,t
//_!.
(35.3)
state
its
of
strain.
A
is
body
is
solid, it is
necessary to
said to be strained
altered,
and
when the
of strain
\^
body
subject of discussion,
we may regard
;, _y,
z as functions of
a;',
y\
z'
possible coincidence.
x'^
y\
z'
The axes
It is
a*,
y,
2,
and
i.
e.,
capable of superposition.
The
state of strain of
x\
;y',
and
z'\
coefficients
.v,
y,
2,
when the
differential
III.
44
May,
187T,
344
./.
IV.
Gibhs
body.
molecular
we
actioji,
niaj^
any element.
state of strain of
differential
coefficients, viz.,
dx
dx
dx
dx'^
dz'"'
^2/
dif
dy
^"
dy'^
dz'^
dz
W
It will
dy
dz
dz
%"
Ih''
^^^^^
strain,
its
which do not
capable of
is
must be capable of
in space,
There-
independent variations,
tliree
considered
six indepen-
dent variations.
The
is
addition or subtraction of
heat.
If
we
write
But
if
we
shall
have
we may
consider
may
8e^,=.t
6,/y,
any constant
for
+A\. S^. +
A-v,
A'x,,
Zy,
dz
witli respect to
-y^,
The
-j-,-
if
in
the
state of strain
(^^
7/v,
and
+ A',, rJ^
dz
(Jjt<j
^,
.^
dz
dz
d'T^
for the
volume
as a function of
dy
dx
where
its
tv(j
dz
produced by
write
CCOC
_-
6v
is
physical
we apply
signiffeation
fv
taken
of these
is
moved
parallel
J. TF.
Gibbs
Eqidlibrium
of Heterogeneous
345
Suhstanceti.
by
Now
the
first
member
= A'x,
(^-77,
dx' dy'
dz'.
member must
same value.
work
the second
we must regard
Since
equal and opposite, the whole work done will be zero except for the
face
which moves
distance
})arallel to
moved by
And
X.
<^'^'
dx
represents
the
com-
since 3-^,
In general,
a;'
is
constant
upon the matter on the negative side of that surface per unit of the
surface measured in the state of reference.
The same may be said,
mutatis mutandis, of the other symbols of the same type.
It will
be convenient to use
X',
V,
Z', respectively.
Sey.
This
is
With
t tf;/v<
:^
this
understanding
we may
write
(356)
y, for
a given element
If
continuous
it
is
in
we must add
the integral
fey,
SX' Ds'
ward
of the intrinsic
is
therefore
variation
346
W. Oibhs
J.
EquUihriuni of Heterogeneous
Suhstances.
y;
This
entirely independent
is
homogeneity of the
any supposition
of
regard to the
in
solid.
in contact,
But
since
conditions of equilibrium for the interior of a solid mass and for the
surfaces
where
it
For
fluids.
this it will
be
only
of the fluids
We may suppose
close proximity to
much
of the fluid as
it
is
in
which
is
by moving normals
to the
equilibrium of the
fluids.
may be
be observed
It will
tliat this
enclosed in a solid.
method
is
is
entirely
necessary to separate the great mass of the fluid from the small
Now
energy of the
- f^^p
f^t SDt]
dIJv
fluid
mass
2^,r
/-' ,
will be,
by equation
(358)
^>'^'>*i
solid
is
composed.
Where
fluid, this
we wish
term
is
of course to be cancelled.
we may suppose
of Z.
It is
that
it
acts
variation of the energy due to gravity for the whole mass considered
is
simply
(359)
where g denotes the force of gravity, and F' the density of the
W. G/M/s
J.
of
E<p',)UI)rliiiii
We have,
347
TIeterorjeneous Suhstdncei^.
solid.
ffft
+fH dBtj
^p'p 6Dv
H- :S , f'li-^
Ds'
Dm
^ 0.
(:H)())
to
total entropy,
(1) that which expresses the constancy of the
(2)
the fluid
a,
how
^N' Ds'
= {a
/i,
surface of the
(h-
/i
+ J'^dDrj =
SDv
the value of
is
dDv
where
-{-fVs,
6y
for
any element of
solid,
Ds
(^z)
y
body
and
volume
its
(3)
in the state to
which
;,
y, z relate,
Ds
how
fluid are
dDni^, dDtn.^,
by
etc.
for
in the solid,
r^'dJV'Ds',
]
r^'
dN' Ds',
(363)
\
J
etc.,
,
state of
the values of
the element
tlie
where
(362)
(3H 1
com-
Now,
We may
from
= const.
(364)
(360).
If
we
multiply (362) by
/>,
the whole surface of the solid and for the fluid in contact with
it,
for
we
y*> SDv
by means of
Avhich
we may
If
(365)
848
J.
TT'T
(xlhhs
2,rM,<Wm,
/Svf/stances.
-/^\(//,r/) SA^'Bs',
(866)
(360).
///^ 2' ( JTx, S ^,) d.c' dy' dz' +ff/g F' Sz dxf dy' dz
+ y fv, dN' Ds' ft SN' Us ^f2y {a8x^ /i 6y \ y 6z) Ds
^fpi\.,SN'Bs' -/^\ (/',^V) QN' Bs'^^, (867)
//v,
in whicli
and
in
fluids are
and
//j,
//o, etc.
Now
variations,
if
we
we have
=fa'
X^, dx Ds
(868)
with similar expressions for the other parts into which the first
may be divided. The condition of equilibrium is
integral in (367)
{a'
+fff;/
2 {a dx) Bs
+f[ey.-tt/y,+pv - ^, {i.iJ\')]6X'Bs'^
It
throughout
therefore the
first
solid
mass
is
(369)
0.
not continuous
surface-integral in (868),
surface-integral in (369),
must be taken
and
to apply
not only to the external surface of the solid, but also to every surface
of discontinuity within it, and that with reference to each of the
surface.
it
is
To
(370)
that throughout the surfaces where the solid meets the fluid
Bs'
2 2' {a
X^, dx) -f
(37
1)
and
[8^,^trfy,+pv-^,{/UJ\')]dX'^0;
and that throughout the internal surfaces of discontinuity
(872)
J.
W. Gihhs
2 2' {a'Xy,,
,+
(Jic)
349
(373)
^I^^
dx'
dY^
dx'
The
first
"^
we
dy'
dY^
(l^\i
"dAf
X, Y, Z
by the
As
it
dz'
'
UjAj'j,
+ ~c' ~
sum of
which
,,
'
neighlioring elements.
we have called
x, y, z relate,
dsy,
which
difi^er
Z',
//,
d-^,
only by a rotation,
and
d~
and
coincide,
represent
we must have
Y^,
(375)
Yy,=
The
i^ince
4^
dx
X^ =
and
arbitrarj^,
dy,
dy
Z,
^z-
d^,
states x,
is
^^
when the
it
X, Y,
dx' dy'
Z.
since,
by
and
''
state to
X, Y,
'
may
dY^
dy'
d'^XI
'
dz'
1^ +
member
^^' n
^
"^
~
^"l'
the axes
obtain
six quantities
^x,
Z^
Zy,
^^y, ^z,
^y
= X,.
(376)
ovX^
are
infiiience of
gravity
may
therefore be expressed
by the equations
350
J.
W. Gihhs
dx
dy
dY^
dl\
where
F denotes
dT,
dZy
dZx
'
dz
dZy^
by the
They
which
not in equilibrium,
is
of
which
witli
equations (377).
cients of
fJir,
we
If
we
obtain
{a'
rx,+/i'
{a'
Now
when
J\.,-\-y' Ty,,)
the a\
ft' ,
normal
a'X^,
represents the component
evident from
the
+ ft'X^y'X,,,
]iarallel to
strained
by
state
measured
in
^Y
(379)
upon
the
This
is
of which
consisting
constant.
solid
(378)
}
J
z' is
any portion of a
when the normal is
solid, or
rest,
drawn outward as usual, the same expression taken negatively repreof the force exerted upon the
sents the component parallel to
surface (per unit of surface measured in the state of reference) by the
surface of discontinuity
its state
force
side
Equations (378)
exerted
ujion
side.
Since
the
and determined by
by
the
Wi Gihbs
J.
{a'),
we may
= - {a'),,
iff),
=-
(/i')
(/),
=-
351
(k%,
also write
etc.,
may
for a surface
equations
a X,,-^
ft'
X +
cx'Y^.^ft' y,.
X=
y'
- ap^^,
+ y'Y^,= -ftp^,
(381)
JDs
the fraction -= denoting the ratio of the areas of the same element
Us
by the
surface,
and determined
by the
upon the same element of surface.
If we wish to replace a and Ds by a\ ft', y\ and the quantities
which express the strain of the element, we may make use of the
(but acting in the opposite direction) to the pressure exerted
fluid
The product a Ds
following considerations.
element
Ds
on the 3 -Z plane.
Now
convenient form.
=. const.,
y'
z'
Let
it
= const.,
is
we may suppose
JJs
-
Ds
-,
is
that
it
x'
and
indepen-
has any
const.,
let
included by the two others with the surface of the body be denoted
by i, M, and X, or by
i',
according as
and
X'
projections of
of
Ds upon
figure.
Z^,
ft'
and
Ds',
y' Ds'
plane
M, and
is
we have
reference
X with
Ds
include a solid
distinguished.
solid figure, the
If the
normal to Ds
normals to X,
III.
M, and
must be
outward from the small
must fall inward, and vice
falls
X
45
May,- 1877,
352
Gihhs
J. 'W.
Noav L'
versa).
may
sides
is
Equillhrium
of H'eterhgeneous Substances.
The
dz'.
perpendicular
tlie
projection of
on the Y-Z
by the co-ordinates
dy
-,
dz
dy
-,
( dy
dz
dz
dy
the area of
fdy_ dz __<h_
-,
dy\
and
and
y', z'
a' in
same plane
this expression
y'.
The
ing
L\
\ dy' dz'
moment
is
for the
dz
into
will
z', ',
if w^e
suppose
be obtained by chang-
and
(i' ,
three expressions
may
'2'
sum of
and into
x' , y',
be substituted for
I)s in (3S1).
We
to denote the
Y\
JT',
Z',
(i. e.,
by changing x
y', z'
into y', z,
;',
and into
z', x',
y',
with similar changes in regard to a', ft', y', and other quantities
to denote the sum of the three terms
relating to these axes,) and
This
Y, Z.
is
symbols.
With
may
be reduced to the
form
etc.
which are
which impede the formation or dissolving of the solid, 6N' may have
either positive or negative values, and we must have
fv,
But
if
some
vv,
+V
Vy,
= 2,
{/I
/
,
',
').
(383)
W. Gibbs
J.
diY'
fluid,
is
353
incapable of positive
and
it
is
iy,-tVy,+Py,^:>\{ft^I\').
To express condition
we may use
(384)
reference,
r^,
ijy,
y,
etc.,
It will
components of the
throughout, or
and the
substance in the
we
fluid,
we
we may
use
F to
and
yu^
composed of
variable state,
reference,)
If
composition
a single substance.
solid is of uniform
If the
solid.
we only
if
shall
have
fv,
tl/y,
+J)Vy,=
H^ r',
(3SG)
and
y ~tl/y-\-J>=:Z l.(J\
To
fix
body which
we denote by s,
case of a solid
If
strain.
(387)
is
//,
let
us apply
it
to the
Now the
1 1/ -\-
l-i
m.
(388)
a solid meets a fluid require that the traction upon the surface deter-
mined by the
This condition
face.
is
always
we
tion,
shall
reference,
lose
which
is
satisfied
known
we make the
nothing in generality,
if
proposistate
of
We
page
349,]
354
W. Gibhs
J.
or,
by
/i
Zv
^z),
(375), (376),
S=a^X^-\-f5^Y^J^y^Z^
a f3Xy~\-2/3y Yj_
-\-2
We
may
+ 2 y a Z^.
also choose
axes.
for the
(389)
so chosen
is
is
as
great as for any other surface, and that the direction of the axis of
X)
passing
,1
through the
,
1^
<
mi
1 hen,
A.
axis oi
as
is
it
is
.>
we write
it
-7
CtOt
and y
fi,
(tfj
by
^-
,,
lor
CCJ^
treating
as independent variables,
dS
-j-
a=l,
and
is,
^77,
dp
A^y
Hence,
-,
,,
a r=
when
^ -
dy
=
y
0,
and
0,
-yr,
'
and
P=
dS
/^
dp
Moreover,
p dp -\- y dy = 0,
;/ = 0.
p = 0,
0,
nrrl,
when
dS
dS
a da
when
That
diS
diS
(iS
-=
Z^ ==
0,
0.
(390)
0.
dS
=
+ ^dy "K
,
^',
'
da =:
0,
0,
pdp-Vydy^(\
and
/:^=1,
Hence
i'z
Therefore,
when
0.
(391)
0.
which are called the principal axes of stress, the rectangular components of the traction across any surface {a,
aXx,
PY^,
/i,
yZ^.
y) are
by
(379)
(392)
when the
surface
is
W. Gihhs
J.
(2), if
surface
is
ing axes,
(3), if
constant for
all
is
equal to
principal tractions)
all
is
normal and
surfaces.
be observed that
It will
A'x,
355
common
the
in
may
be regarded as included
in
the
first,)
but the
always
If,
different.
therefore, a solid
bounded by
strain is
which
homogeneous
is
and
in nature
in state
of
may be
satisfied
[see (381),]
by the contact of
which
will in general
may
fluids
be denoted by
/)',
p'\ p'"
(These pressures
of the solid to dissolve that the potential for the substance of the
solid in the fluids shall
have values
//,', yu/',
fx^" determined
by the
equations
E
These values,
it
will
m^
(393)
tt/ -\-p"' v
{'iM)
= fA^"m.
(395)
nature and state of the solid, and their differences are equal to the
differences of the corresponding pressures divided
by
the density of
the solid.
It
may
temperature
same
solid
write [\,, [f/]j.>, ["]/'? ^"*^ [^'i]/.' for the values which
//, would receive on this supposition, we shall have
[4'
f ['/].'
+P'
L"l'
= bu]r'
?;,
'"
v,
and
(=^96)
Now
it
[f];,-
//-!-
[n],,
+;V
p' [vl.,
=n
;//
[//
J^ m.
(397)
and entropy
356
J.
that the
upon the
W. Gihhs
first
solid in bringing
it
/>'
work spent
stress to the
^:>'
[]^ evi-
If the quantity of
when
the increase of the potential in the fluid on the side of the solid on
Hence, ///
greater than
is
will
be necessary to
is
[yuj^,,
,and
greater than the value of the potential which would be necessary for
equilibrium
yWj'"
if
if
all
when
that
if
/>"'.
is
so
there were present in any one of these fluids any small frag-
such fragment
is
which
facilitate the
is
subject
to
commencement
the
distorting
stresses,
stress
stress.
This
may
will
doubtless
make
in the case of
to hydrostatic
it
a necessary
amorphous solid which can be formed out of it that the solid at the
surface where it meets the fluid shall be sensibly in a state of hydrostatic stress.
J.
t'oi-ni
W. Gihhs
upon
surface
its
(if
distorting
the
be
should
and consequent
stresses
carried
357
too
commence
before
far),
on
to be formed within the
most other bodies, yet within certain limits the
relations expressed by equations (39:3)-(395) must admit of realization, especially when the solutions are such as can be easily super-
would
they
the
or
fluid
of
surface
saturated.*
It
may
in
determines
in
To examine
by equation
may
complete generality, we
(388).
proceed
in
in the state
of reference
the form of a cube, the edges of which are equal to unity, and
We may
homogeneous
reference and
in
generality, since
in
variable
its
we may make
its state
no
(This involves
of
loss
of
we
state.
Let the fluid meet the solid on one or both of the surfaces
for which Z' is constant.
may suppose these surfaces to remain
choose.)
We
edges in which
axis of JT.
and
y'
remain
]iarallel to
It will
body
fix
the
the
and
do not
It
in
in
_=
dz
clz
-^r= const.
==0,
const,
=0,
dy
-^ = const. =
(398)
and
.V
Yy,,
0,
Zy_,
Q,
= 2y~^-,.
(399)
Hence, by (355),
^- 7
1^-7
7
ch...^,.J...+X.4^,
+-\..d^,
^^^
f^''
(iy
dz
<^y
I ,4^-p^ JL, a _.
+ T^
.
(^^
-,
(400)
Again, by (388),
*
The
and
liquid,
xxiv, p. 395.
was
-p.
first
57.5
to the
See Trans. R. S.
PMl.
Mufj., S. 4, vol.
358
W. Gibhs
tT.
de
Now
(hi
p dv
dt
-|- '/
dp
4- in dyi^.
made
dx du
(^01)
require that
dz
and
dii
dv
= -f-,
di/
dz
dx
-r d-^,
dz
dx
dx
dz
-,
dx du
dz
-/-, d
dx dy
dz
dy
.,
+ -yr
"T- d-f-, +
dz dx
dy
^-1
403)
dt
V dp
-f-
'^
Tn
df.i^
dy dz\ ^dx
-.dx
,^
/^^
dx\ ^dy
dz
when
the solid
is
T.rw ("^'
subjected on
all
sides
^,
sion
-^ri
dy
dz
The
case
is
i.
it
by p
x' is
be equal to
will
e., it
JTx,
taken negatively.
d^,
and
A\.,
sure
^>
We
member
by these additional
mass of the
We
solid.
dt
7/f
Vpdp
-\-
m djA^ = 0,
V) dt
/^
Now
(vf
-y)
(405)
fluid.
By
subtrac-
dp
dx
dy dz\ jdx
rTr
if
whicli
forces
of the equation.
pres-
UjQt
^
Uj%I
dz
dx\ dy
W. Gibhs
J.
359
dt^v,-^v^^^^,-^v^
where
should
(407)
VvV
dp
This equation
pressure.
the solid
if
body
change of temperature or
similar to (131), which applies to bodies
2J<'iss
is
-j
ally be the same as if the solid were subject on all sides to the uniform normal pressure p for the quantities v and ?; (and therefore
Q) will in general have different values. But when the pressures on
;
sides are
all
will
dp
equal on
all sides,
But
if
we wish
(A/tJ
know how
(XiJ
the temperature
is
normal and
still
cloC
(AjtL'
be the same,
-j-,
as constant,
Ct'iJ
sure between the solid and fluid remains constant, but the strain of
the solid
is
varied in any
differential coefiicients of
way
when
by equation
all sides
fit
when
-y-,
These
(406).
the pressures on
differential
ffni*
ftnc
fi^i
-=
-^
-j-,
are con-
when the pressures on all sides are normal and equal, vanwhen the density of the fluid is equal to that of the solid.
The case is nearly the same when the fluid is not identical in substance with the solid, if we suppose the composition of the fluid to
stant, or
ishes only
remain unchanged.
We have necessarily
(f) is
/^f =_(^r
But by equation
and
*
etc.,
suflBxed
feV" ={4^T
preceding
may
(409)
equation, transposing
by m, we obtain
except such as
(92)
III.
all
coefficient.
46
May,
1877.
360
J. TFI
Gihbs
\dni J /
By
dt
p,
1,
dv
- m /-
/ dti Y^'^
m.i^-
\am
Y"^^
I
/ 1,
'p,
+ in
and
//f
v^
df-i.
we may
dp
(410)
0.
obtain an equation
we
shall
have the
expressions
/ di]
w(^-- Y^
\dm / p, m
J
The
\am i/t,
i,
Y^
dv
m(I -
^
and
p,
m'
Avill
We may
which
fluid
~,
the quantities
sented by
'C,
-j-, ~^,
we know
If
constant.
m ^^
m^, m^,
yw, in
several components
etc.,
it
Now
dp
TO
We
may
also constant,
M
+ r,
\dm/t,
in
considering variations
we make
will be sufficient if
^,^>,
is
dfi,={~i^)
^
\dp /t,
temperature remaining
the
are varied,
2^ oi*
dm^
we
p, TO
all
therefore give to
shall
but
have
-^1
Xdm^/t,
p,
dm^
+ etc.
containing
dt,
we
obtain
\dph,7n
\dm/t,p.m
/du-,\^^
+ '" (*i)<,
^^
*"=
''"
,, .
^dx
+Av,<?^,+
/ ^^
(lv,
dy dz\ ^dx
/ ^^
= (-^- +
dz
dx\
'"
<^'
<) "ii^'
-.dy
.^^
+ /'^,;^,)"!^,.
,
(411)
This equation shows the variation in the quantity of any one of the
components of the fluid (other than the substance which forms the
solid)
which
/;,
or of
^)%i
(I'X/
(%Q^>
-j-,,
-v- -i
with
re-
^Y.
'T.
Gihhs
361
Solids.
may
he made to de-
pend upon our knowledge of the composition and density of the solid
at every point in some particular state, which we have called the
state of reference, and of the lelation existing between the quantities
which have been represented by
and
When
z.
the solid
fy,,
among
fluids
nature and in
of reference,
'''^
'
>
^''
is
the equilibi-ium of
))r())lems relating to
themselves.
If in
'
is in
^" ^"
Vy<->
its
sofid is capable,
and the
between
relation
it is
we may choose
state of strain,
fy/,
homogeneous
in its
dx
t-/
dx
'/.
dz
?
"i
dz'
will
be
independent of
cc',
y\
z'.
for
however, we
know
in the case of a
reference,
we may
Prince Riipert's
For
if x',
reference,
we
dx dx"
if
we
write
divided by
its
-7-0
y\
z'
denote the
in
and
have necessarily
^ = d^'W-^
and
j,
state of reference,
first
dx
'/v
it
shall
solid,
derive from
fy*
droj).
dx dy"
dy" tc'
dx
dz"
+ d^' dx'
R=z
,
'
y\
iM
dx"^
dx'
dy'
dz'
dy^
(i/
dy"
dx'
dy'
dz'
<M
d2![
dz!^
dx'
dy'
dz'
z'),
we
shall
(,",
(412)
y", z")
have
dx"
(413)
362
Gihhs
J. Tf^
we
If, tlien,
liave ascertained
dz
dm'
of
--^,
7/v,,
we
(414),
,,
shall
(414)
,/,,..
terms
Sy, in
-^,,
by
,;,,= i?
obtam
e^,,
dx
terms of
in
?/v,
dx"
dz
-^,
^o
^,,
dz"
solid of the
dx
,
bring
it
s"),
same composition,
dz
-y
y\
^,
and
if
the
is
known
(;c',
in
terms of
composition, to
its
into a similar
y', z')
//y,,
and
similarly situated state of strain with that of the element of the non-
dx"
for each
We
shall then
have
fy"
ii^i
terms of
the density,
of the
if
?/v,
is
known
in
-^,
known
in
-=^,
as
known
terms of x\ y",
-p,
terms of
z^\
and
and
x",y",z"
of reference
in its state
{x', y',
.-;'),
this is
non-homogeneous
An
body
as
homogeneous body
sufficient for
is,
dz"
.
solid
that
is
fy,,
i/y,,
-^,
...
-^,
involving
which the same relations may be deduced, may be called a pmdamentcd equation for that kind of solid. It will be observed that the
sense in which this term is here used, is entirely analogous to that in
which we have already applied the term to fluids and solids which
are subject only, to hydrostatic pressure.
When
fv,
'/v5
7^/
'
^t
^^
W. Gihhs
J.
in
vvliicli
will
A'x,,
t,
363
^z.
relations
^"' ^^^"
which are
dx"
'
'
^"
^'
di'
'
^^^^^
'"
all
may
is
tion.
^V,=
The equation
(356)
may
6t
we know
Thei'cfore, if
-y-,
-y,
the body,
.
(416)
^7/v,.
;a, St
the value of
if-y,
(417)
in
t,
we may
pendent relations
we
fv,
shall
have
if\,
of equation (416),
we
value of
i/'v,
An
t,
-y^,
determines the
.
,
,
and the
which express the composition of the body when it is capable of continu.ous variation, is a fundamental equation for the kind of
solid to which it relates.
quantities
we
it is
whole mass
(-//")
of the
in question,
same mass
shall
//
for
its
temperature.
We
in
might have
solid
is
364
Gibhs
J. TT\
function of the tem]>eratnre and the state of strain, so that for con-
we
stant temperature
miglit write
"""
the quantities A'x,
= ^^'(-^-4)'
This
equation.
this
^z
and
not aifect their values, for this equation holds true of A'x.
as defined
Zy_,
by
by equation
{'355).
With such
data,
by
transfor-
results.'''
It is
would be that
tions
//-y,
be wanting.
e^,,
Such a method
is
The method
more
distinctly
the role of energy and entropy in the theory of equilibrium, and can
dynamical problems
in
We have
Hence,
dx
;,
-^-,
by
it
must be
is
possi-
six functions of
dz
For these
quantities
we may choose
by the
by A, B,
C, a,
b, c,
we
shall
705.
Memoir
With regard
"
On
If
we denote
these quanti-
have
Thomson and
compare
also
i,
Thomson's
i,
vii.
p.
ligiit
in the
J'ransactions of the
Cambridge
J. TF. (jrlhbs
365
^=^W^'>
^ fdx dx\
'-=^\d^'dy'}
^-,/dxdx\
^=^\d^'dx'}
b, c,
^'''^
The determination
<-)
i/.\,,
t,
A,
f y,
C, a,
i>,
7/v,
thus
is
A,
7i,
b, c.
can
and
t/y,,
or of
i/\,
and
capa-
is
t,
dis-
state.
tion
with
determine
tlie
To demonstrate
is
isotropic,
must with
value of
//y,
fy*,
or
i/\,.
The
lows.
,...,, we may
ratio of elongation r of
any
line of
is
proceed as
which
<t',
/i',
fol-
y' are
evidently given
by
the equation
/dx
Now
(dz
dx
dz
^,
that
is
dx
A2
dz
A2
always possible to give such directions to the two systems of rectangular axes JT', Y\ Z', and X, Y, Z, that
it
is
dx
(421)
We may
to this line in
is
at
any other, and make the axes of -.Tand A"' parthe strained and unstrained states respectively.
J.
.366
Moreover,
if
we
cients obtained
f/fr^)
write -A^-r
da
d{r^\
(:?(r-)
d/y
V-r V-r
dy
/Substances.
for
the
rv',
/J',
difFerential
y'
as
coefii-
itidependent
variables,
'
when
<?^v'
' z=
and
That
r7/f -f
/J'
A.y^O,
IS,
'
when
/i'
1,
^-,=
y'
0,
dx
^.
0,
;=
AJ
dx
Hence,
dy' =z
;/'
0,
and
/i'
1,
;/'
0,
0.
== 0,
0.
gi'eat as for
this line in
other,
is
per-
Of
pendicular to
such lines
(423)
0.
all
least as
Then
= 0.
|,-,
and
it
may
just been
easily
(424)
employed that
1=
be parallel to
,42.)
0.
Y\
JT, Y,
and Z'
and
the unstrained
in the strained
body
body, and
dx
dx'
dy
'
dy'
dz
'
dz'
lines
viz..
J.
W. Gibbs
lines,
367
ceed to find the relations between the ratios of elongation for these
dz
dx
-rj
lines (the Jor^?tc^/:>/ axes of strain) and the quantities y^, ,
.
when
a' da'
strain
mined from
1 c?(r2)
From
(420)
we obtain
a' 2
two
the
ft'2
_^ y'2
last
i^
_
~
d{r^-)
'ft'
"dj^
dft'
we
d{r^)
(426)
'
(427)
dy'
necessary relation
dft'
^^
'
'
^ dy'
r-'
(428)
from (420),
If
l_
y' dy'
we obtain
^
we
_
~
^d{r-^)
or, if
first
directly
da
From
1i^
0,
Therefore,
(420) as before.
a'
we have
368
W. Gibbs
J.
We may
write
/-e
(431)
0.
Then
(432)
Also
(dx dx
\dxi)
2'^
/dx
/dx
z=i2'2
/cfe
~^W/
\dx') \dy')
\2
\dy')
/dx
6fe
may
also
<^y
dx dz dz^\
(^'^
t?.K
f?a;
dx' dy'j
be written
^'
dx
dx
dx'
dy'
dy
dy
dx'
dy'
(434)
it
will
+3
X, y, z, into y,
symbol
dx\\_
dy dy
Y_
~
dx'dy'dx'dy')
\%7
Wv
W'/
dxy
^/dxdy^_dy^
_ ^^
F=i
symbol
~dx'dy'd^'dy'~dx'dy'd^dy']
\dx' dy'
This
c?c
dx\)
\dx' dy'j
(dy y/dx
YY^^y \^
^')
dx'
dx dx
/dx
"
\dx'
c' dy' J
\dy'J
z,
in the
3-3
be taken negatively
to
and
also to use
z'
G= 3-3
/dx dx\
:^'
* The values of
deduced
at length,
and
may be
\dl' d^'J
/dx dx \
\dy' dy'j
\d^' dz')
/dx dx
With
this
(435)
Introductory
Anwendung
der Determi'aanten^
5,
J.
W. Gihhs
In expanding
369
^
3+3
^
(
dx dy dz
/dx dy dz
(dx dy dz
870
W. Gibhs
J.
single-valued function of
//y,
E^ E, and H.
The determination of
therefore reduced to
is
the determination of this function, or (as appears from similar considerations) the determination of
//\
as a function of
t,
E^ E, and H.
E represents
shows that
tion (432)
Again, equa-
eqiiation
(434)
ratios of
and equation
co-ordinate axes
arbitrary,
is
it
shows that
Y\
Y'-Z\ Z'-X'
sum
of the squares of
which
Hence,
^^and ^F
mean squares
are the
There
is
one another,
lines
all
possible directions in
solid.
terms of -^,
in
^,
l)ut
there
E^ F^ H.
X\ Y\
is
or surfaces.
the
of
is
is
also a
If the sys-
it
will
always be
any
possible to give to
shall
This
will give
strain of
a*,
I[t=
1,
x',
11=
1.
= y\
--
(440)
s'.
while
X :=
will give
is
a-.',
y r= y\
=z
(441
z'
\.
Now
although
such a change in the position of the jjarticles of a body as is represented by (440) cannot take place wliil(> the body remains solid, yet
W. Gihhs
J.
a method
of representing
may be
strains
371
considered incomplete,
We may
avoid
the state
X\ Y\
which
all
Let us
sent a strain.
y\
Z\ and
a cube with
z') is
(a?',
call
the edges
cc',
y\ or
~,
to the
shall
-^
it
fZ^c',
dy\
and
dz'
in the state
is
which
edges
ffe',
dy'
1",
2)
continuous changes
in whicli the
(,t, ?/,
as posi-
Whatever
2'
may
c^a?',
its
parallelopiped
may be done
and therefore without changing the sign of H. Now tw^o cases are
possible
the positive end of the edge ffe' may be turned toward the
positive or toward the negative direction of the axis of Z.
In the
;
first
case,
evidently positive;
is
in
The
we may say,
or negative, according
we
(a;',
y\
z')
(a*,
y, z) to the
volume the
its
value zero.
If
we now
U2',
U^
/"g,
and denote by
C/j,
for example,
If
we choose
axes
t^j,
tem
?7j',
coincide.
U^,
f/3
is
or
is
will determine
all
to associate
be positive, the
Or,
if
we
two systems
of axes, so that they shall be capable of superposition, correspondinodirections coinciding, the positive or negative value of
will deter-
In this case
we may
write
372
tT.
Wi Gibhs
= H=
W.
J.
values to
G-ibbs
//v/,
denote by r^ the
rpy,
and
373
etc.,
stresses vanish at
of
by equation
(444) to be expressed in
^2^0.
''i-^o,
it is
(446)
''a-'^o.
That
is,
if.\,
will
dipy,
dlj:y,
dr^
The
'
dr^
dipy,
'
dr^
differences.
Therefore, since
dipy,
<dx
d-^.
_ dipy, dr^
dr.
^
ax
dipy.dr^
dr..
<dx
dr^j-i
jIx
dr^
Ax
dr..
^'
ci^n
<^-^r->
dx
dx
dipy,
dx
whether we regard the true value of ipy, or the value given by equaand since the error in (444) does Jiot affect the values of
tion (444),
dr.
dr^
dr..
-AJjvU
-J.tJC'
~.\A/iMj
dx'
dx'
dx'
etc., etc.
be interesting to see
It will
how
the quantities
e,
/*
and h are
and the
definitions
^'^~''\)t'
* See
Thomson and
^^^"^
= r,^v',
i,
(448)
p. 711.
3*74
where
V
W. Gibhs
j^
volume, and
its
J.
volume
v' its
in
dx
subjected,
and
f&y
^_^_^_,.
when
is
(449)
\
dx ^^dx
dy
di/
dz
dz
dy'
dz'
dx'
dx'
dy'
Now
we
when
the solid
consider so
reference,
dz'
we
is
much
'
of
it
as has the
volume unity
all
sides, if
in the state of
have
shall
=r
r,
=:
r-g
=: y^,
(450)
(430),
if\;
Hence, by equation
~^'
= +3
+ 3/ 0^ + h V.
e /'^
= (M)t == 2
e ^'-^
(451)
//,
+ ifvi + ^
-^,(^)^=-t6^)-* +
(452)
i/.i;
(453)
and by (448);
Frr-t-^ +
1/V,,
(454)
'
(434),
^ must
From
2e
r^, like
differential coefficient
4/r^,2.
when
(455)
= r^^, we have
+ 4/r2+A,.^:^0.
The quantity
vanish in (452)
2e
A, in
(432),
R
i^zz:
Moreover, since
in equation (444),
(449),
R\
may
e, /',
viz.,
and V,
(45G)
^'*
when without
is
Gihhs
J. TF.
It will
case
3V5
is
i]xi
bi
If
+ I/'^o,
F'=~;t--
(458)
= 2e' + 4/'r(,2,
+ 4/' r2 j^ h' r =
7^'
e'
shall
sti-ess, Ave
(459)
(460)
0.
Whence
^,_R'+^r,V'
P^'-^r,V'
^ _ I^
^^'=-^j^,
f'=~~V-l-^'--^
In these equations
the quantity
If
R', and
i/y,.
we wish
changes required
isotropic), the
easily
r,,,
made.
If
in
the fundamental
we change from
E by
a"^
and that of
state of reference,
we must
i/\,,
In
a*.
are
first, it is evi-
first
y,, ?/v,
equation
length in the
its
the
E by
any
of
(461)
are divided
by
a^,
of the
for
7/v,.
tions of
[It will
i]y,,
sul)stitiited,
be remembered that
and that
it is
i',
e',f',
and
represent func-
Jt
ijy,
are
There are certain bodies which are solid with respect to some of
have other components which are fluid.
we
shall suppose
which can
is
i.
e.,
we
shall sup-
III.
48
fluid
June, 1877.
J.W.
376
Gibhs
when the
such a body,
in a
when
/"',
fyi
are constant.
/"/, etc.,
will therefoie
The
be given by an
Now when
this
-\-:2 2'
t di],.,
denoting,
(^Y
divided by
is
its
volume
etc.
(462)
equation
joc^Uvi
body
the
Ix^.
as
in the state
reduce to
elsewhere,
the
:2:e
dx
j> d
=
We have, therefore,
dsy,
dx'
dy'
dy
dy
dy
dx
dy'
dz'
dz
dz
dz
dx'
dy'
dz'
dz'
2^ dvy,.
(464)
d)]y,
2^ <Jvy,
L,,
drj
Z^ dl]! -f
in
etc.,
(465)
ence,
de
di]
J) dv
-\-
L dm^,
(466)
J. TF.
where
f,
//,
Gibbs
denote
tlie
its
if
which
symbols relate
tlie
The varied
matter.
state
It is
body
377
The
finite size.
is
shall
may be
otherwise.
is
de =^
ih]
and the
considered invariable,
is
jo dv
we have by equation
-\-
yi/
dm,,
//,,
dmi,
-f-
solid
(12)
etc.
(467)
It
Since
we have
state of the
former definition.
symbol
may
etc.
we may apply
//,,,
It will also
this
term as we
//,
fliiid
body
is
components
in the
one of hydrostatic
Equation (462)
then be written
f?fv
dih,
pi
drj +
/x,,
dl\'
etc.
(468)
we
evident in the
shall obtain
ditions
first
place that
if
etc.,
constant,
378
TK Gibbs
J.
Again,
solid components.
its
if
we
sup-
it
will easily
substances which form the fluid components of the solid body must
satisfy the
tact with
same conditions
it,
in the solid
body and
may
in like
It is evident that if
the
set
manner
if
body
in
may
is
fluid mass,
and
com-
body or
not,)
is
an
body
Also
if
must be
components
satisfied,
the
as in
The
quantities
t,
X^,,
^^-'
Z^,,
^y.
/<,
/<,
(409)
etc.,
"!" ^-
^^^''
^'''
('^'^
= -^
etc.
(-471)
dx'
This
i-esult
member
may
be generalized as follows.
Not
only
its
is
the second
present form,
W. Gibbs
J.
substitute
iug
^]^i,(Jt
tlie
example,
is
equiA'alent to subtract-
Therefore,
>/y,).
379
if
we
consider the
and (470) which occur in any same term in equation (468) as forming a pair, we may choose as independent variables
either quantity of each pair, and the differential coefficient of the
<puintities in (409)
the remaining quantity of the second pair with respect to the inde-
first,
taken positively,
the independent
if
by the
sign
equation
in
Thus
dx'
^\lx'
WJx - [dxj,,:
where
\di
;;|rx,
-~
\dx;Jr.:
in
(^*'
Ay, or -
It will
ditions
^^'^'
'
either
dz
ry
Zj,,
or y-;, eitiier
or /
//^
= const.,
//,,
= const,
or
//y,
either
etc.
,/,
//.,
^,,
is
body in contact with a fluid of which the phase does not vary, and
which contains the components to which the potentials relate. Also
that
in
when FJ,
any
Ff,',
infinitesimal
etc.,
represented by
is
/^og_A
\
_((IX^\
d^ )q~ UQv.)':^.'
/^oi_\
\ dX^, }q
\dxJt-vi\ogt)x.:
tdi]^,-,.
If
we denote
this
_
-
('1^\
\dQjxJ
\/lt-
\d\o^t)%,^
no
^^^^^
^^^'')
^'dx'
etc.,
must be regarded
either Z^, or
-
,
in
each equation.
380
W. Gihhs
J.
We
in treatiiig
of lieterogeneous masses in
and
such
is
is
its
various components
if it
That
it
could
it is
any mass
is
sensibly affected
by
its
vicinity,
natural
and
We may
method of taking
component substances
possible a simple
is
It
term
may be
in a fluid mass
and uninfluenced by gravity. For the premeasurement of the quantities with which we have to do, it will
which
crise
is
in equilibrium
For
and
all
condition
all
its
Let us
imagine a closed surface cutting the surface S and including a
We will
J.
form of
W. Gihbs
tliis
381
is
is
denoted by
this portion
vS
which
and the area of
lie
the discontinuity in
Let us
its vicinity.
entirely
beyond the
infliience
of
which contains
is
more
precisely
what
is
to be
referable to either
it
will often
be convenient to
in the energies
it
that
of
we may regard
no change
But
is
be considered
be convenient to limit ourselves
in the variations to
the masses
tinuity,
i.
e.,
if
we
382
Wl Gibhs
(T.
is
the surface.
the total
amount of energy.
(juantity,
we have
To show
that
if
it
if it
an infinitesimal discontinuity of
is
phase
first.
If the
considered
is
we may suppose
a similar
method to
With
M, M', and
We
JNI".
shall
suppose
With regard
tinuity,
is
it
to the
sui"fixce
variations
(i.
that
e.,
are
also
of discon-
when its
when
consider
only
and
we
its
various components.
and
this
where
^If,,
A^, A2,
etc.,
It
is
(470)
by the
evident that
A^
initial
is
the
it
will
IJy com])arison
in
yl
,,
homo-
W. Gibhs
J.
383
is
we may
By
which
fact,
is
i-elate.
we
If
true of
This designation
substitue
for
jii^
we may
dm
-f- ;/.
write
(Jw,
less
(477)
etc.
in this equation,
all
will
de =zt6f^-\-
a natural ex-
call
x.\\Q
will
in
tials,
;*
energy
its
is
quantities of
its
its
member
in
of
which
entropy or of the
various components.
It is
rf//,
for
member
is
is
* To illustrate the difference between variations which are reversible, and those
which are
rium
at a
them
as
not,
we may
of mixture
is
mixed.
We may also
conceive of
amount
diminution, and
member
of (477).
it is
the second
all
mixed
but
it
to mix) should
(for a varia-
be greater than
It is
may be
not.
It will
tirely
different
from that
in
which
it
is
the state
is
here used
is
en-
may
also be brought
to a state
from the
state
B is
B to
through the same series of intermediate states taken in the reverse order
when
the system
that
is
bears to B.
49
June, 1877.
384
J.
It will
TT'T
Gihhs
EqxdUhrlum
of Heterogeneous
press this
by an
equation.
value nearest to
of de consistent with
oo)
ISuhstances.
(i.
e.,
the
Sj]
-\- lA^
diti^ -\-
/.(^
+ etc.
Sm.^
(478)
which do not
consists of
+ Se"^0
(479)
we also regard the surfaces separating M, M', and M" as invariable, we may derive from this condition, by equations (478) and (12),
If
+ ^tc.
+ 61/ //j' Sin^' + /'o'
+ etc.
dill/'
/i/
^"i/
+ etc.^0,
+ (h/" + /</'
+
-\- /fj
6?/
d7)i^ -\-
J.I2
Sni^
<Si>i2'
-\-
t'
t"
(480)
dm^
-f
dm/ + dm/ =
61^2
-\-
6111/
0,
6)112" ^^ ^5
-J-
etc.
It
may
etc.,
jji/'
It will
6m/^).i^6m/,
be observed
that,
if
= t'
1.1/'
(482)
t",
67n
/'^
f.12
^^^2%
^tc.
/(,, for
(484)
instance,
first
possible
have
If it is also a
//,'.
we shall have yu,
component of the second homogeneous mass, we shall also
/A^=}.i/'.
W. Gihhs
J.
will not be
/<,
385
determined by
any equation, but cannot be greater than the potential for the same
substance in either of the homogeneous masses in which it may be a
possible component.
conditions of equilibrium
relatiiig
and
tem2)erature
to
which we have
the potentials
befoi-e
by
bounded
like
limits of
homogeneity,
may be
condition which
It
fulfilled
by
sur-
films,
into
by equation
(12), if
=
de"" =
Se'"
t'
we regard
ch/"
6i/"'
t"
whence, by (482)-(484),
Si:"'
(U""
From
//,
Or,
if
rf(w,
we
t'"
(hn^'"
///' (h,i^""
we have
j.i^
//,
6"")
m,'"
t rf(//
m,"")
dm^"'
Srn.J'"
f.1^"
Sm.J"
iA
-\- /.i^
then have,
shall
etc.,
(485)
+ etc.
(48G)
= dif" + dm,'" -f
= t6j/""-{- M,6m^""
S{e
//,'
We
respectively.
we have
//'"
-f /t2
6m2'"'
etc.,
(48V)
+ etc.
(488)
//'")
+ etc, (489)
set*
6'"
f "",
= 7/
//'"
;/"",
m\=zm^ rn^" ^
here used
if
is
m^"",
(490)
etc.,
(491)
The Roman
letter
is
386
J.
we may
W. Gibhs
of Heterof/eneoiis Substances.
write
St^
This
JEquilihrimn
(h/^
ih/f^
//
//2 c^?"|
etc.
(492)
is
fixed.
be observed that
It will
a^
denotes the
excess of the energy of the actual mass which occupies the total
if
it
would
same uniform value quite up to that surface which it has at a sensiand that tf, m^, m|, etc., have analogous significations.
It will be convenient, and need not be a source of any
misconception, to call 6* and ?f the energy and entropy of the surface
in
sities
and entropy),
and
171^
-,
~, etc.,
Now
these quantities
[e^,
if,
m\,
surfaces
The
determined partly by
etc.) are
we
by means
will be
it
remembered, has been regarded as fixed in the variation of the system. It is evident, however, that the form of that portion of these
surfaces,
which
lies in
To
we have
all
necessary to
Let us
its
suppose
fii'st
position in space to
is
varied in position
will
valid
be
if
also, if
the material system and s are both varied in position, while their
relative position
remains unchanged.
But
if
Therefore,
it
will be valid if
its position.
we must add
to the second
member
dtp" ~~ jA
in
Srn^^
the form of
~
s.
f^
Sm% ~
If
we
^^^c.
su]pose s to be
suffi-
W. Gibbs
J.
887
and
may
write
St^ =.
Sif -f
/<, 8tiii\
-\-
a 8s
dm\
//g
6\ 8c
+ etc.
+ C'g
(Jc2,
(493)
or
=
+
8t^
<j,
6ff
4-/^2 ^^"%
yUi Stn^^
^^^-
ff
C,, and
initial state
6*2
+ i{C\
- C,)8{c,-c,),{494)
But
the system.
surface s that 6\
To show
s.
Tlie
above
is
is
it
f**
vanish.
C.^ shall
in
tlie
de
8e"'
=
i.
e.,
8e
/J
8m^
o-8s^i
{C\
C,) 8{c,
by (482)-(484) and
87}
81/"
_ Se"" +
etc.
//g 8711
^"'
^" + etc.
/z,
8711
8m
+
+ //,
^""
8?/"" + //, 8m
+ /1 2 8m.^ "" + etc.
8?/
^x^
8m
this equation
i{C\ -
(J,)
8{c^~c,), (495)
(12),
|<2
^^2
c,)
it
appears
r-^)
^^c.
+H
(^\
- O2)
^(c,
c,).
(496)
the same in the two homogeneous masses separated by a plane surFor let us imagine the material system to
face of discontinuity.
Also
8s
Hence j)'
0,
etc.
//g
//j
0,
0.
0,
8v"".
is
in the
is
plane.
Tn the
388
W. Gibhs
J.
Equilibrium
the system
initial state of
tliis
of Heterogeneous
Siihstances.
In the
parallel.
is
it
it
(Such a variation
may
the system
in
the
as
tively,
whether such
librium
is
surface
may
But
side.
a variation is consistent
Let the
is
make another
us
let
initial
tion
it
same distance
It will
A.
it
be parallel to
its
former posilie
the
s',
of course
unvaried system.
Without
s'.
face s.
by the symbol
lie
s,
Now
by a common normal.
in
equation (496)
is
member
Now
we
the vahie of
the same
is
and
'
if
",
Moreover, ^(Cj
we may
we make
shall
s,
being deter-
Cg) =
both for
marks
Both in the
s".
and s" be traced
we
'
perimeters of
therefore write
Ss"
z=z 0,
s'
and
s'
and
for s"
s".
If
by
the
W. Gihbs
J.
389
Hence
But
tf(e,'+c'./)r=dXc/+C2").
6*/ +
Therefore,
G ^" +
it
tliat
'
+2
tf'.s
must
= 6\
we may give
other side of
plane,
^'3
"
when
s'.
Since this
also
true
C\ ".
sufficient
is
may be regarded
And
nearly plane.
as nearly plane,
is
for
when
the
when the
non-homogeneous
film.
This
size.
discontinuity
so that Cj
plane or curved
is
Now we may
is
the case
is
it
radii of curvature
is
by equation
(498) that
1,
if
the quantity
same order of magnitude as the values of 6^, 7f, ni\, //<|, etc.,
while the values of C, and G2 ai"e of the same order of magnitude
is
of the
as the changes in
by
must
unity.
affected
since
can be very
it
in s, the values of
in general
by
o'.
And
little
C, and G^
hence,
if s'
be
Avill
make Cj"
in
general
be
sensibly
coincident
with the
physical
surface
of
discontiniiity.
We
is
not distinctly
indicated that the surface s, in the unvaried state of the system, has
cUvidmg
is
surface,
of discontinuity.
Cg 1=
0,
It will
be remembered that
and which
We may
is
we have
called the
surface
is
is
390
TK Gibbs
J.
We
may
in (494).
when
is
it
Cj
Now
plane.
will
be observed that
= Cg, which
is
the case
on account of the
as
it
composed of parts which are a})proximately plane. Therefore, without danger of sensible error, we may also cancel the term
Equation (494)
6t^
We may
is
S)f
+ o- 6s+ //
dm^^
(497)
when the dividing surface has its initial position determined in the
manner described.
The above equation is of fundamental importance in the theory
It expresses a relation
of capillarity.
dis-
to
we
we
shall take
in-
at present
we take account
etc.)
is
its
denoted by e\
^
+ '-f f",
will
be
will
be
that
d^iH(^i'+(5f"^0,
(498)
W. Gibhs
J.
when
V^
391
'"n
V'> V"->
by (497) and
'^^''2
'*^^2'i
set =: for
-p
6s
o-
(We may
"4'
"*]'? '"i">
to be all constant,
Tlien
(499)
^t'C-,
Sv'
- p"
dv"
0.
^,
two opposite
if
directions.)
membrane without
rigidity
have chosen
ff
Ss z=
(c
+ C2
may
SN, we
s dJV,
whence
{c^
-\-
This
relates.
jo'
is
of equality of pressure
masses
6v'
= SK
c^)
^p' Jo",
Sv"
=-
.s
6JV;
(500)
the side to
119,
in contact, vvlien
We
faces of discontimiity.
move
have
shall
when
which
flixid
by these
surfaces.
Fundamental
of Discontinuity.
one of equilibrium.
state
is
The only
dt^
Both the
we may
dif' -f
i.
e.,
that an o])po8ite
in equilibrium.
sup[)osed to be
is
-f
/^
d7n\
To
distin-
Mz ^''4
^'t<^'-
(501)
may
be neglected,
we
from
zei'o to
f^
f if^
-\- (7 s
+ n
m^,
-J-
//g
50
n%\ -f etc.,
(502)
July, 1877.
:592
whicli
(in
W.
J.
(irU>hs
may be
Eiuillhrhim
variable,
o",
^,
//,, f^^i
If
of the
is
we
(It -\-
we denote
sdG
m\
-\-
the quantities as
+ m\
f?//j
diA^
etc.
by
fg? Vs?
l\i
^2->
we have
f^
jf
,s
r,
==
-1,
i ,
= -^,
etc.,
(506)
"
t^tc,
8^
f?yu,
7/s
Now
^^tc,
//,
7^2
(505)
any
(503)
0.
ISiihstanees.
equilibrium) which
If
of Heterogeneous
^tc.
(507)
(508)
restriction
the temperatx;re and the potentials for actual components shall have
For however
may vary
(on
always be
satisfied
to the surface of
is
not great.
if
stances.
The values of
variable,
and
it
these quantities.
derive from
number
it
t,
f.i\->
/*3'
^^^- ^^'^
therefore independently
by
If the
form of
this function
is
is
a function of
known, we may
7/s,
/'j,
I\,
which occur
in
that ecpiation.
These
.are
all
';/-f 1
W.
./
<rlJ>hs
we have
that
tities
y/
+3
Or,
we may
occurring in equation
(5ti2),
2^/,
393
consider
+5
quan-
independently
variable.
An
may be
An
(509)
between
may
it is
//,
fs,
also be called a
'h->
-{-
^H
fundamental e([uation
for the
n.
and one
the variables
may
(-510)
^^.c,
^s?
2 of
etc.,
))i\^
)ii\
ft',
etc.,
//g,
equation between
f^
or
yUj,
^,
if this
For
in
may
('51 1)
be regarded as subsist-
be regarded as independent
(viz.,
ii
-|-
-f- 1
differentiation
-}-
we may
2 additional
obtain
by
equations
-\-
may
If
we
we
set
quantities
ij)^
^,
t rf,
(512)
An
-|-
be obtained.
=.
if'
dt
ds-\-
jx^
d)n\
-\-
pi^ dni,2
+ etc.
(513)
equation, and
is
to
The reader
3 94
Gibbs
J. TF.
On
the
When
all
tinuity are
the substances which are found at a surface of disconcomponents of one or the other of the homogeneous
//g,
may
homogeneous masses
small
is
for
is
measurement.
cases in which
is
it is
may
jt>'
=: ji>" (which
since
2^'
and
p>"
we can make,
t,
tion
with increasing
for
we must regard
that
for values of
o"
etc.,
/.i^, /.(2,
t,
which
satisfy
change
in the value of
for variations of
To show
this
more
/>'
t^
which are
=:^>".
distinctly, let
t,
m^,
//g,
etc.
remain constant,
= [\
dp = Yi' d^i^,
dp" = y,"diA
da
denotmg the
f?y
r,
and
But by
ni'
densities
-dp"
r?(;y
d/^i^,
Wi,"
^
and
__
Hence,
(;/,'- y^")dp^,
~ p") =
(;//'_;/,') da.
(50()
(c, -f ('2)
da + a
d{c^
+ C2)
= (Hp'-
j>").
Therefore,
l\
or
{c,
Ir/'
C2) dff
r/
+
^1
r, a d{c,
(i
^2)1
C2)
= ir,"
^^=
^\ ^4^1
;/,') dff,
Cg).
W. Gihbs
J.
Now
Y
i"
To
[\
(Cj
y i'.
will
c^)
compared with
very small
395
we have
;/,',
^"
as constant.
a'
From
plane.
is
this
it
appears that when the radii of curvature have any sensible magnitude, the value of
o'
all
same
as
when
the surface
is
the same values, unless the component for which the potential has
densitj'^ in
the
two
homogeneous masses,
in
is
Accordingly,
cial density
we cannot
l\ from
its
in
value
(^
by measurements of super-
\dj.ii/t,fi
ficial
with
The
tensions.
//s,
The
same with
I\,,
F^,
etc.,
and also
quantities
fg,
to admit of direct
however,
is
sui-face
of discontinuity,
it
may
be
more easy to measure its superticial density than its potential. But
except in this case, which is of secondary interest, it will generally
be easy to determine
ff
in
terms of
t,
//j,
/.i,
etc.,
with considerable
impossible to
difficult or
An
equation giving a
in
terms of
<,
/<i, yWgi
is
plane,
may be
ft
ia
all
and
W. Gibhs
J.
39(3
of TIeterogentous Substances.
JEqidllbrititn
tlie
siu'face of discontinuity.
These
be exliibited
results can
The
position
particular
ado])ted
for
term i
C'j-j-
Cg)
of the surface
ing surface
is
(^
(c^
(which
dividing surface
the
make
the
is
is
It
evident that equation (501) will hold true for plane surfaces (sup-
posed to remain such) without reference to the position of the dividing surfaces, except that
We
tinuity.
None
it
shall
may
for
new symbols,
will
be affected by such a
have
m\, m\,
etc.,
as also
when
different values
The quantity
fg, ?/s,
F^, F^,
however, which we
o",
we
etc.
by
choose,
may
is
changed.
the quantities
Vsj
^s?
^15
A(fv"-fv'),
fv', fv", 'A'',
(//v"
//v'),
^25
etc.,
increments
A(ri"-ri'),
^{y/-y2),
i]\"
o"
etc.,
in the
will receive
the increment
(93)
p"
6v"
p'
Therefore, since
The value
of
ing surface,
o'
jt>'
is
when
ey
t //v'
/'i
ri'
the value of
o'
is
zero.
But when we call this quanwe must remember that it will not have
W. Gibhs
J.
its
Equilihrmrn of Hetefogeneous
chai'acteristic
trary surface.
C/Onsidered as a tension,
position
its
nowhere
we
to
The
else.
in the surface
is
strictly speaking,
however, wliich
make
397
S?d>sfa>ices.
sufficiently
this distinction of
It is
tlie
may
vanish.
etc.,
surface
fail
to
Let us
is
so placed as to
is
make F^
Then equation
0.
(508) reduces to
dff
second
etc.,
^o,
?/si
1\
- Ad) <^M2 -
//s( I) (?
etc.,
(514)
Now we may
0.
member
Tsd) f^/<3
by
consider
a dividing surall
the differen-
given by equation
(98).
e.,
that
i.
/)'
of the
surfaces in terms of
^,
//j,
/.I2,
etc.,
we can
ff
is
known
also
eliminate
yUj
for plane
by means of the relation derived from the equality of presand obtain the value of a for plane surfiices in terms of
From this, by differentiation, we may obtain directly
//3, etc.
pression
sures,
t,
yUg,
the values of
would
l)e
Acd? Ad)?
p\ p>" and
difficulties, we can
elimination oi
braic
?^s(i)
etc., in
terms of
t,
/n.^,
/^j
in
from the
all
finite
But,
if
the
dp\
and thus
differential
This
//g, etc.
e([uations
dp/',
d^.
ol)tain
398
W. Gibhs
J.
differential equation
etc. in
terms of
may
/^o, etc.,
t, f4
^,
^'2(1)5
?/s(i)>
^3(1)5
with (514).*
* If liquid mercury meets the mixed vapors of water and mercury in a plane surface,
and we use
tively,
//,
and
/i.^
i.
e.,
mercury is the same as if the liquid mercury reached this surface on one side and the
mercury vapor on the other without change of density on either side, then r2(i) will
represent the amount of water in the vicinity of this surface, per unit of surface,
above that which there would be, if the water-vapor just reached the surface without
change of density, and this quantity (which we may call the quantity of water condensed upon the surface of the mercury) will be determined by the equation
da
to
dition
may
be regarded as
we
shall
fulfilled in
to the
law of Daiton
where
supposed
2^-2
= 72
df^'i,
denotes the part of the pressure in the vapor due to the water-vapor, and
Hence we obtain
da
For temperatures below 100 centigrade, this will certainly be accurate, since the presmay be neglected.
The value of a forp-^^O and the temperature of 20 centigrade must be nearly the
same as the superficial tension of mercury in contact with air, or 55. CS grammes per
The value of a at the
linear metre according to Quincke (Pogg. Ann., Bd. 139, p. 27).
same temperature, when the condensed water begins
in
sum
+ 8.25,
to
or 50.8.{
its
own
vapor.
grammes per
mercury
in contact
metre,
if
we
As p.^, therefore,
grammes per square metre (when water begins to be
a diminishes from about 55.03 to about 50.83 grammes per linear
its
air.
condensed in mass),
metre.
were
It
does not by any means depend upon inappreciable differences of superficial tension.
The greatest difficulty in the determination would doubtless be that of distinguishing
to the
J.
W. Gibhs
may
relations
;J99
them
If
we make
dp'
= y^'
df.1^
dp"
=y^"
di^t^
4- y^"
da
I\
=.
{,'
/"g
Tlien,
-yr")<h',
d/.t^
if
;/,' f^/'2?
di-t^,
1\
f^yUg?
dp =:dpj\
-r,")<^M2
{y2
^.
and
da
That
= r, -,
~,, du^
I\ du.
is,
vhh/
2/ i'J-i'
P'
The reader
7-Yi
j.
Yi
represents the
Yi
distance
be-
tween the surface of tension and that dividing surface whicli would
make T^j zz
the second number of the last equation is therefore
;
equivalent to
j'^^d)-
any component substance has the same density in the two homogeneous masses separated by a plane surface of discontinuity, the
value of the superficial density for that component is inde})endent
If
of
superficial
we may
Thus
second
member
will reduce to
/"g-
is
is
when
It
in this case
equivalent to
same density
in the
two
homogeneous masses.*
* With respect to questions which concern only the form of surfaces of discontinuity,
such precision as
is
we have employed
in
This precision has not been used for the sake of the
mechanical part of the problem, which does not require the surface to be defined
we
51
July, 1877.
400
W. Gibhs
J.
Concendng
We
Kqii'dibrlum of Heterogeneous
Suhstcmces.
of Surfaces of Discontinuity.
the Stability
in its nature,
will be convenient
it
to
its
homogeposition
For this
suppose that the homogeneous
while
altered.
masses are very large, and thoroughly stable with respect to the
possible formation of any different homogeneous masses out of their
plane and
is
uniform.
Let us distinguish the quantities which relate to the actual components of one or both of the homogeneous masses by the suffixes
relate to
by the
surface of tension.
its
as
changed
may
state
subsist
is
in
supposed to be
still
uniform
between the
equilibrium
and such
homogeneous
in nature,
given
modynamic
state.
The remainder
and on accoxmt of
its infinitely
is
greater
be infinitely less altered in its nature than the first part. Let
As^ denote the increment of the superficial energy of this first part,
size to
Arf^
Ami,
^^iif
ftc, Ainfj,
Amf
etc.,
the increments of
its superficial
determinate values to the superficial densities of energy, entropy, and the component
substances,
which
quantities, as
between the tension of a surface of discontinuity, and the composition of the masses
which
it
separates.
The product
cr
regarded as the available energy due to the surface in a system in which the temperature
//,, /i 2, etc.
or
is
when
the system
is
subject to gravity
entropy and of
tlie
dividing surface.
is
are
The value of a, as well as that of s, is sensibly indewhich we may assign to the dividing surface (so long
used
in
this paper,
lilce
the suiwrficial
localization of the
W. Gihbs
J.
Equilibrium of Heterogeneous
401
jSub.stances.
by
^ /Itf,
-^nr:,
-Anvi,,
Jw.!;
etc.,
tAif"
Hence the
//
Anil
//,,
Jnv),
total increment of
At^
Aif
energy
which Aa^,
to
etc.
relate,* the
Ani\
Anv},
/<!,
its
ii/,Ani.l
De^,
JJif^,
Dntl,
Anif,
etc.
(516)
etc.
positive,
|iart
for
finite
of the film
f-i,,
etc.
necessarily
is
in
etc.,
in
Ami
// Anr; ^
//,,
^'*
will
etc.
ii
If the
its })ositi()n.
Bin^,,
in the
nature of the
i>m^
etc.,
we
write
Bn^,
etc.
For,
,
if
ABe^ -
ADif
-//
A Due: -
//,,
,j,
AJJnil
ADmf,
;/;,
etc.
ADmf
etc.
>
(51V)
0,
A/Dmt=0,
etc.,
we have
z//7>6^
- A/Bift
//
A/Ihn^,
/,,
A/Bml -
etc.
>
0.
(518)
The expression
- AfDif t
//
AfBml -
;/
AfBml -
film,
etc.
the sign
A must be
orders.
zero.
Otherwise, by equation (501), the above expression would have the value
402
Gihhs
J. TT^
of
in
the film, for in considering whether any state of the system will have
less
which
least energy,
state,
we need only
is
is
infini-
tesimal change in the nature of the part of the film to which the
symbols
the film
relate,
obviously unstable.
is
and not
the film
film,
made
in
is
finite
practically unstable*
e.,
i,
if
But
increase.
it
dynamic
it
occurs
for
we cannot
On
parts.
changes which we have been considering are such that every part
remains
of the film
on each side
and
if
in
same
by an
in
may
perhaps be capable
unstable
it
infinitesimal
We
condition.
by
is
practically unstable,
is
or
is
not
stabil-
negative value,
capable.
we need only
Let us write
it
/.;
in the fuller
(mf - mf
it
is
form
,il
{mr - mT] -
etc.,
^^^^
where the single and double accents distinguish the quantities which
*
With respect
term
is
used,
W. Gibhs
J.
film,
403
accents denoting those quantities which have the same value in both
The
states.
when
guished by double accents are alone considered variable, and the area
of the surface
is
To make
(//r
dmr + etc.
i.i;)
unless
mf
=:
unless
wif"
/j"
//;' ,
i.
which
will
e.,
we must have
=1
0,
by equation
0.
(}"s-
ff's,
is
by
g'
(520)
positive or negative.
That is, if the tension of the film is less than that of any other film
which can exist between the same homogeneous masses (which has
therefore the same values of t, yu yu^ etc.), and which moreover has
etc., so far as it contains the
the same values of the potentials jj^
substances to which these relate, then tlie first film will be stable.
But the film will be practically unstable, if any other such film has a
[Comj^are the expression (141), by which the practical
less tension.
stability of homogeneous masses is tested.]
,
It is,
/j,, ,
be positive.
Moreover, since
we have by
and by
e"
if
+ pv' -^ mj
- if + p v" /A nij'
/J
7/'
//,,
/J
etc. = 0,
m/'
I,
m,"
etc.
=.
0,
we denote by
f,
?/,
V,
m,
m^,
etc.,
m,
m,,,
etc.,
404
s
Now
this
W. Gihhs
J.
1 1/ -\-
pv
the value of
if
equation
//
is
/<,,
etc.
/j
>//,,
jj,,
etc.
mi
=:
<J s.
member
first
may
as s increases, and
decrease
will
7n,,
of
therefore be
two kinds
There
will
of
given value of
v, so
is
applicable,
is
i.
e.,
so long
of
i,
Ma,
P,
/'/,,
etc.,
by the
ill
7/,
?',
the
first
r/i/,,
1)1^,,
may
volume
surface of discon-
question,
f,
etc.
//,,
//,
etc.,
?^,
?/,
etc.
member
negative value, and must therefore have a least possible value, which
will
be negative,
if
any value
f,
is
?/,
negative, that
etc.
which
if o" is
is,
negative.
least value
With
if
is
it
can
of the potentials.
if all
all
the component
Therefore, in a
form
s{G"
a')
/'//)
differ
etc.
when the
quanti-
infinitely little
is
W. Gihbs
J.
405
accents relate.
so too,
is
values of
ja^^
jj,,,
same
limits.
condition
may
when
it is
evident
the quantities
be written
sJff-^mfj
Aj.iy
nt^,zij.i,^
etc.
etc.
0,
or
z/ (7
On
< r; J
j.i,j
r,,
jj,,
(521)
make
the
is
the tension a
maximum
if
were
fulfilled.
We
now
at first suppose
to be
and to remain
sj)herica]
and uniform
throughout.
In order that the equilibrium of a spherical mass entirely surrounded by an indefinitely large mass of different nature shall be
neutral with respect to changes in the value of r, the radius of the
sphere,
may
it
is
be written
2G=r{2^'.-p"\
(522)
instability,
it is
necessary that
the equation
2do-={p' -p")dr
be
when
rdp'
shall
satisfied,
406
J.
TK Glbbs
remain
satistied.
(The
i.
e.,
if
Moreover,
if
'%>'%
the equilibrium
ijs
is
dp
necessary to
But
if
-"'^^'^
unstable.
is
stable.
do'
<^^*)
when
,
is
In general, therefore,
is
there-
we may
call
and a cannot vary and condition (524) being satisfied the equilibrium is unstable.
But if either the interior homogeneous mass or the surface of discontinuity contains substances which are not components of the
enveloping mass, the equilibrium may be stable. If there is but one
such substance, and we denote its densities and potential by y\^ 7",,
and
or,
yUj,
by
(98)
and
(508),
(r;//
r,)'^-'
</'-/.
(526)
regarded as constant.
But
W. Glhhs
J.
Dividing by
4;r
and
y ^'
we obtain
dr-\-i
(.,V
By means
dy ^
r^
or, since
by the
4 7tr~ r^.
diiFerentiating,
specified
407
r^
dl\
0,
//,,
dl\
V2r?;//
+ 2r,,*+(^|^ + ,.|-.).,.=o.
is
(627)
brought to
the form
If
we
eliminate r
by equation
we have
(522),
\P'-P"^ ^)
3 ( jo'
If
/:>'
and
member
ff
are
known
in
jt>") fZ//
terms of
"*"
t,
//j
>i
o- djJ.
yUg? ^^.c,
.529)
we may express
the
p".
first
This
the values of
If
//,
'
-^
Smce
we may
(I If
dT,
>
5
(^^^)
2-
also write
r.
dff
d log
o"
'
Again,
if
("'')
Vdr<--2'
m-o^r<-2and - = 0, the condition of stability reduces to
r'j =
*
3 y,'^
y'
Since
du,
= ^,
djji^
we may
also write
_y,'_d
^gjy_y')
p'-p" dy"^^?,'
Trans. Conn. Acad., Vol.
III.
d\ogy^
52
^3
^^'^'^^
Nov., 1877.
408
W. Gibbs
J.
When
is
The two
very small.
Fj, unless
;Ki' is
(533)
r/
(*
+ 2 I\
from which -
1^'
-y-^, etc.
',
r, '+27
(r
may
,) '^^
etc.
<y' - p\
(534)
must be
1\
Two
ra' ^'
s,
+ ^2
etc.
^S
constant.
different
ing a circular
may
homogeneous
orifice, their
by
diaphragm hav-
to
when in equilibrium.
The condition of stability derived from
written
d(p'
;/)
da
,,.
is
(522)
may
in this case
dr
be
/..\
if
jf
and a
all
will
The
surface of tension
To
is less
if
we denote by x
the part of
the axis of the circular orifice intercepted between the center of the
orifice
by
v'
when
is
and
v'
{r
7t x"^.
x)
and
orifice,
plane,
we
shall
W. Gibbs
J.
By
we
diiierentiation
obtain
(r
and
civ' z=. 71
whence
x)
x^ dr
(r
By means
409
clx
-\-
x) dv'
-\-
xdr
=.
0,
rX
{2 TT
n r x^
x~) dx
dr.
(536)
may be reduced
to the form
2da
dp"
dp'
-J-'-^i-r ZT'
dv <
dv
dv
now suppose
liet us
cept one,
(P
- P)
it
//,,
when one of
the
homogeneous masses
i2-
all
'
is
^,
(537)
^
r^ x?
all
of discontinuity which
is
also
is
when
exposed
re-
But
y^'v'
constant
y^" v"
I\s
z=z
dv\
and
ds
When
d}.i^
When
be neglected.
,
dy
-=
suffix)
must be
dy
-z
it is
may
d}x^
by the
= -2 dv',
dfx^
Ki
specified
therefore, since
dv"
By
component
is
and
component of
s
-j
'
may
either
generally
we must not
apply the formula to cases in which the substance spreads over the
diaphragm separating the homogeneous masses.
410
Substances.
equilibrium,
a condition
any to the
result,
when that
which
require a
Let
But problems
it
system
of a
in a
surfaces of tension
It is
supposed that
masses are
satisfied, as well
surfaces of discontinuity
equations for
The question
is
difficulty.
be determined by
to
is in
question.
of the system are not in general states of equilibrium, and the rela-
tions expressed
them.
true of
state of
equilibrium, they
masses are
may
fail
any
state of
the system.
do not
is
entirely determined
but
if
the phases
face of discontinuity
is
we have employed
different,
we cannot
tinuity
For
example,
by assigning
to
it
It
would be
J.
W. Gihhs
41
necessary to give the law by which the temperature passes over from
it
And
if this
tlie
rate of
It is true that
its vicinity.
we
in
different in the
two phases
in
unknown
to us,
is
entirely determined
by
is
we have occasion
to use
surfaces of discontinuity.
may be
It
obsei'ved that
we have
which only
Such equations
initial
tlie
state
may be regarded
as estab-
rium.
neces-
is
when we
dis-
Such equa-
We
is
in question,
it
will
attri-
We
which we may
call the
is
to differ
from the
real system.
(500)
may
not be satisfied.
412
<J.
'tV.
Glhhs
we
be conand that
neither matter nor heat can be transmitted through it.
Other cases
may easily be reduced to this, or treated in a manner entirely
inflnences
It will
is
fixed,
analogous.
Now if the real system in the given state is unstable, there must be
some slightly varied state in which the energy is less, but the entropy
and the quantities of the components the same as in the given state,
and the exterior of the system unvaried. But it may easily be shown
that the given state of the system may be made stable by constraining the surfaces of discontinuity to pass through certain fixed lines
situated in the unvaried surfaces.
Hence,
if
the surfaces of discontinuity belonging to the vai'ied state just mentioned, there
must be a
state
is in
little
little
shown by
imaginary system
The
surface of dis-
state, in
will be the
We
unstable,
to the
imaginary system.
If the
stable.
system
is
is
either neutral or
refei'ence to the
in
is
the case
this determination
by
depends
J.
W. Gihhs
systems do not
diifer.
may
The
result Avhich
(3)-(7) to the
413
by
imaginary system.
is
in
as follows:
the small parts taken separately are stable, the criteria of stable,
all
neutral,
when the
(500)]
It
is
to
pressui'e
[equation
not satisfied.
tions (501)
We may illustrate
the method which has been desci'ibed by applyproblem but slightly different from one already (pp. 408,
409) discussed by a different method. It is required to determine the
conditions of stability for a system in equilibrium, consisting of two
diiferent homogeneous masses meeting at a surface of discontinuity,
the perimeter of which is invariable, as well as the exterior of the
whole system, which is also impermeable to heat.
ing
it
to a
To determine what
minimum
condition of
is-
we need only
consider those varied surfaces of tension which satisfy the same condition.
may therefore regard the surface of tension as deterarea for the surface of tension,
We
mined by
But the
would evidently be completely determined by the
position of the surface of tension and the temperature and potentials,
if the entropy and the quantities of the components were variableand therefore, since the entropy and the quantities of the components
are constant, the state of the system must be completely determined
v',
by
We may
therefore reo-ard
414
all
J.
W. Gibhs
may be
condition of stability
de
(Z2f
-j-,av 4- -
dv
where
TTo dv
dv^
and the
+ etc. -^
> 0,
'
v' ,
written
di
is
that
(541)
,,-^.>0.
Now
we
if
write
f", f^
f',
we have by
(86)
d^
or, since
jo'
dv'
ds,
ds
^_
de
_,=
_y+,/ +
p" dv"
(542)
Hence,
d'^e
dv^^
_ _ dp'
If
we now
da
dp"
d^s
ds
may
stability
'^)
be written
d^s
dj)'
dp"
da
dv'^
dv'
dv'
dv' dv''
v-5
ds
by supposing,
on page 409, that we may disregard the variations of the temperature and of all the potentials except one, the condition will reduce to
d^s ^
The
T^
ds \ du,
Making
this constant,
."
The condition
r.
^y."^"
by the sufRx
is
r\s.
we have
|,)*/+(.
of equilibrium
d^s
<"='
->-+'<??) sf-
y.'v'
(./
''<r.->('''
"^dv'^^
is
g + -|^ +
/^)...=o.,o)
{y^'-^'^"
,dr,'
V -4^ClfX^
+
,
+ ^^%)'
dr/'
V -f-~ -+
dry
ClfX^
-^
(IfX^
^^^'^
W. Gihhs
J.
415
when we can
treat
all
may
F'or if
finite.
we
write
2 As for
sum
may
be
written
2/le
-f-
:^At^
>
(548)
0,
the total entropy and the total quantities of the several components
being constant.
The
z/ signifies,
pression
is
higher orders.
stant, the same will be true of the pressures and surface-tensions, and
by
integration of
mass
/ie
and
for
(Sti)
p -iv +
J/;
-\-
//g -^'i^i
any homogeneous
+ etc.,
These equations
Ajf
will
(7
As
-{- //j
hold true of
A7n\ -f
//|
Am.^ -f
finite differences,
etc.
when
<,
^, a, //,,
and will therefore hold true of infinitesimal difunder the same limitations, without neglect of the infinitesi-
are constant,
/ig, etc.
ferences,
By
:^{pAv)
2{2jAv)
or
That
Am
Hi
ol)tain for
is,
the
sum
form
+ 2{(TAs) >
2{(tAs) <
0,
0.
(549)
by
by the sum of the products of the areas of
of discontinuity by their tensions must be a maximum.
the surfaces
This
is
III.
53
Nov., 1877.
416
J.
sioiis
are constant.
tinuity.
is
iSubstances.
of interest, because
it is
always
by
tubes
if
altered system,
its
parts,
and therefore
On
the PosslhlUty
of
the
Formation of a Fluid of
different
Phase
The study of
homogeneous
fluid
masses
some of them) and having the same temperature and the same
method of treating
we must
first
of all inquire
surfaces of discontinuity
is
how
far
our
applicable to cases in
which the radii of curvature of the surfaces are of insensible magniThat it should not be applied to such cases without limitation
tude.
is evident from the consideration that we have neglected the term
i(Ci C^8{c C2) i" equation (494) on account of the magnitude
^
of the radii of curvature compaied with the thickness of the nonhomogeneous film. (See page 390). When, however, only spherical
masses are considered, this term will always disappear, since Cj and
C2 will necessarily be equal.
Ao^ain, the surfaces of discontinuity have been regarded as separat-
homogeneous masses. But we may easily conceive that a globumass (surrounded by a large homogeneous mass of diflerent
nature) may be so small that no part of it will be homogeneous, and
that even at its center the matter cannot be regarded as having any
ino-
lar
vve
is
in a less
J.
W. Gihbs
417
with
This phase
it.
sidered,
and
is
contact with
it
at a plane surface.
we
ing surface.*
With
this
no ambiguity
fictitious
in
it
non-homogeneous film
But
of curvature which are large compared with its thickness.
in regard to globular masses too small for this theorem to have any
application, it will be worth while to examine how far we may be
certain that the radius of the dividing surface will have a real and
I'adii
it
is
natural application.
(500)
globule in
may
and
equilibrium with
(502),
surrounding
may
If
we
homogeneous
G from equations
(497),
and contain
be written
2
0'
{p'
-p")r,
(550)
If
we
write
[f],
[//],
[m.^],
[m^],
etc.
for the
actual density or pressure at the center of the globule, but the density of liquid water
(in kirge quantities)
which has
tlie
418
J.
W. Gibhs
excess of the total energy, entropy, etc. in and about the globuhir
same space
in the
if it
were uniformly
filled
,s
^^1
where
^,]
[''^il
' (,^.'
^''
iVi
^ _
"4 = D^s] ,^s
,/),
Yi")^
y y"
y^',
^,
[-,^j
etc.
(,^^/
^''
,^^,.)^
iVi
r/)^
etc.,
We may
= [e] -
aS
V' (fy'
6y")
t [/;]
V' (//y'
//y")
-^i,[m,'\-\-fi,v'{y,'-y,")-ti2[m^-]+^i.,v'{y.,--y.,")-^^tc.
But by
(551)
(93),
p'
p"
W=
^x"+t ;/y"+
//j
/V2
ri"+
^2" +
etc.
for brevity
[f]
t [;/]
[m,]
//,
1.1,
[m.^
etc.
(552)
is entirely determined by
be observed that the value of
and that the notion of
considered,
system
physical
the nature of the
(It will
We
shall then
have
a
or, substituting for s
and eliminating
ff
by
and
r^
W+
v'
we
v' ip'
enter into
2)"),
a z=
W+
irrr^
(^V
its definition.)
(553)
r,
(554)
(550),
^rrr^
If
way
eliminate r instead
ff,
{jy'
-p")
= W,
(555)
we have
Now, if we first suppose the difference of the pressures in the homogeneous masses to be very small, so that the surface of discontinuity is nearly plane, since without any important loss of generality
Gihhs
J. AV.
Equ'dibruim of Heterogeneous
certainly
fulfilled
is,
in
}>'
o',
p\
and r
be positive.
TF. that
when
p'-=.p",
the
regard to position, as
stable in
perimeter), we see by (550) that the presmust be the greater; i. e., we may regard
its
By
as all positive.
But
it is
and p'^zp".
will
not positive
419
with respect to
also
o' is
/Substances.
is
necessarily real, in
become
any possible
when r becomes
as p' p" increases
infinite
from very small values, W, r, and o' have single, real, and positive
values imtil they simultaneously reach the value zero.
Within this
limit,
our method
such exist,
equaiions.
it
will
is
evidently
applicable
beyond
this
limit,
if
radius of the
however,
(see pp. 387-389,) that the globule must bemagnitude before r can vanish.
It may easily be shown that the quantity denoted by
is the
work which w^ould be required to form (by a reversible process) the
It is evident,
come
insensible in
heterogeneous globule
in
For
work
is
is
initially the
this
the globule
is
But
formed.
increment of energy
in
[f]
all
Therefore, by
(55-2),
W denotes
the total
re-
these considerations
with respect to the stability of the homogeneous mass of the pressure p" (supposed less than ^>', the pressure belonging to a difi^erent
phase of the same temperature and
jiotentials)
420
W.
J.
(iibbs
Within those
limits within
justified,
work required
the
which would be
in
Nor can
W,
surrounding mass,
of that to which
size
relates.
size
(page 406) been shown to be unstable, and with the least excess in
size,
value of
is
given
in
We may
relates.
If the three
funda-
mental equations which give o', ^j*', and jt>" in terms of the temperature and the potentials were known, we might regard the stability
(
TF) as
that
It will
If p'
be observed
p"
increases
without greater changes of the phases than are necessary for such
will vary at first very nearly inversely as the square of
increase,
p'
p\
p'^p" continues
\i
to increase,
may
it
TFreachcs the value zero; but until this occurs the phase is certainly
Another kind
stable with respect to the kind of change considered.
of change
be great
is
is
may
homogeneous mass of
But with respect to the kind of
which the
chano-es here considered, which are initially small in extent but great
in
degree,
it
how we
can
fix
with the same precision. But it is safe to say that if there is such a
This
limit it must be at or beyond the limit at which o' vanishes.
equation
of
fundamental
the
by
the
entirely
determined
latter limit is
stability
the
is
of
which
phase
the
between
discontinuity
surface of
is
in
question.
formation
possible
which
the
of
that
in question and
We have
ing surface
when
o'
W vanish
with
at the
it.
If the fault in
same time,
(it
the
evidently
421
limit.
W. Gihbs
J.
But
if
vanish with
r,
a and W,
and
no
sufficient reason
work necessary
of
mal, this
this
amount
the
is infinitesi-
not enough to
is
therefore that TF
It
why
appears
although
is
must be remembered
fundamental
is
in the
phase extends
in
on pages 160, 161), which may be called the limit of practical stability, at which the phase can exist in contact with another at a plane
surface,
It
when supposed
to extend indefi-
limits,
would accommodate
By
may
Wz=z
we
and
positive,
is
work
ffs
be written
- (y - p")
W consists of
v',
(559)
tw^o parts, of
which one
is
always
and the area of the surface of tension, and the other is always negative, and is numerically equal to the product of the diiFerence of pres-
* To make the
two processes
to be
performed separately
above more
clear,
in the following
manner.
pose a large mass of the same phase as that which has the volume
initially in
Of course,
it
must be surrounded by
We
like
to exist
a resistino-
v'
that within.
of such proper-
We may
allow the
422
J. VT. Glbhfi
On
Formation
the Possible
Let A, B, and
We
v'^r^^'s.
W ^as=h^
we
if
neglect
may
- p")
{p'
(560)
v'.
a.
he three
difl:erent fliiid
by a very
its
be the same or
and B to be separated
This sheet
principal curvatures
not necessarily
will
must be
We
zero.
and
The value of
of curvature.
be Cac+o'bc,
tlie
will
B may
Homo-
either.
and
or B.
may
The components of
C must have no
but
diffei'cnt,
is
therefore write
at plane surfaces.
its sign,
first,
if wt"
in
its
thickness
comparison with
its radii
tlie
tensions of
tlie
which
may
not be affected
a.
its
the
its
If this
superficial area.
film, until
be done
phase
mass within will remain constant. (See page 139.) A homogeneous mass of
the volume v' and of the desired phase has thus been produced, and the work gained
of the
is
volume
//
of the
is
and
by
this
amount.
to the edge of the orifice, but not elsewhere to the outside surface of the envelop.
this
and the
in
will be resisted
will
by the
capillary tension.
^{'p' p")v'
W. Gibbs
J.
limit
is
Now
if
fihii
o'ac+ ^bc
is
and B, such a
423
We
film will
cannot sup-
pose that 0'ab^<5'ac+ <^BCi for this would make the ordinary surface
A and B unstable and difficult to realize. If o'ab=o"'ac+ Cbc,
between
we may assume,
relation
we have
and
is
is
of the kind
which
described.
Let us
now
and
to vary, so as
to
still
C determined by
and B. A
A and B will be entirely stable with
respect to the formation of any phase like C.
(The case is not quite
identical with that considered on page 161, since the system in question contains two different phases, but the principles involved are
tials
of
and
shall
become
less
With
direction
we must
consider
two
and
cases separately.
in
the opposite
be conven-
It will
to
make jii9A=Jt)B=^:>c5
it
will not
and Pc^pA,
foi"
and
so that
by
if o'ab<^<5'ac+<5'bc5
so that ^;<A=i^B5
III.
54
Nov., 1877.
424
J.
The
W. Gibbs
Eq^dUhrinrn of Heterogeneous
system
stability of the
in
Substances.
must therefore
C commences to be
same result if we use
than that of
less
We
and B.
arrive at the
when the
equal,
all
it
must
make
it
become unstable,*
it is shown that the surface of contact
under certain circumstances, with respect
of phases
and
stable
is
To complete the
it is
it is
if
at
all,
when
o'ab<^<3'ac+
o'bc,
This
in lentiform masses.
manner,
in this
for-
C would
be
likely to
is
evidently be unstable
antl
masses.
It will
be convenient to consider
in
meet
in a
figiire
and
which
10 represent a section
mass of phase
faces, the
Let
plane surface.
of such a system
Jjj*
first
lying on the
on the right of
left
of
sur-
D E H' F G,
D E H" F G.
Let
and
Fig.
10.
in
may
r'
Cac
* It
is
to,
which
x'
we
are
ffhc
now
is
x"
considering
is
-,
(561)
Gak,
homogeneous masses.
-J =
r"
-,
discussion referred
(520)
we
any
is
different
perfectly valid in this case with respect to the possible formation of a thin
<
',
which, as
we have
seen,
may
J.
which
(Tac
W. GihhsEquiUhrhim. of Heterogeneous
components parallel
and
<Tbc
EF
to
of the tension
we denote by
If
(T^b-
425
Substances.
tlie
TF
J/
iV,
(562)
where
denotes the work expended in replacing the surface between A and B by the surfaces between A and C and B and C, and
A^ denotes the work gained in replacing the masses of phases A and
B by the mass of phase C. Then
31=
where
and
6'ac, -Sge,
(T AC
Sac
'S'bc^bc
(568)
^-ab ab,
Sab
]V=
(564)
Pn),
where F' and V" denote the >'olumes of the masses of the phases
A and B which are replaced. Now by (500),
2 <5^AC
Pc-~Pa= ,
We
'^
(565)
F' r= f
V" =
By
^BC
pc-ps^'
-I
and
substitution
N= I
and by
we
;r
r'2 x'
%7i r"2
tt
R''
TT
7^2 (r"
- x),
(r'
(566)
x!').
obtain
o'ac r' x!
7t
c"
-^
\7i
+t
;r o-Bc /"
^"
+|
R^
Oac
-^
r
^'inR^
ff^,
''~^.
(567)
(561),
N=
tt (Tbc
r"
x!'
-%7T R^
(Tab.
(568)
Since
2
we may
7T r'
Sac,
'^
T?"
r" x"
Ssc,
TT
R^
Sab,
write
^= I
(o'ac *^ac
O'bc 6'bc
(569)
O'ab Sab).
is
the same as
W=
This value
is
ij
We
(o-Ac Sac
positive so long as
have therefore
O'bc ^bc
^-ab 5ab).
(570)
42(3
Gihhs
J. ^V.
E(j[vilihritim
since
But
^s^c
at the limit,
>
by
see
and
s^b,
6^
>
6\vb.
when
G.KC
we
of Heterogeneous Substances.
O'Br
Cab,
(561) that
<*AC ^^^
anCl
^ABj
and therefore
TF
Sbc
i'*AB?
0,
It
circle in
intersect,
We
the
by the
ble magnitude.
We may
formed
jOa
becomes so
is one
ing values of
PcPh
is
by equation
indicated
Hence, no
is
of sufficiently small.
The
is
easily
proved
we
If
phases
same
(in
point,
positions of
If the
disregarded,
phases
all
is
and allow ourselves, for brevity, to speak of the phases as having the
the points by which they are represented, we may say that three coex-
where three
intersect,
series of
all
fluid,
to be distinguished.
this is
when each
or
when
meet or
may be
two others, or one of the series terminates where the two others intersum
The series
this is where one surface tension is equal to the sum of the others.
gect
of coexistent phases will be represented by lines or surfaces, according as the phases
of the
have one or two independently variable components. Similar relations exist when
the number of components is greater, except that they are not capable of geometrical
representation without some limitation, as that of constant temperature or pressure or
certain constant potentials.
W. Gibbs
J.
and
have moderately
42*7
radii of curva-
C may
be
in
equilibrium
satisfied
where
and
Cj
C2
film,
the
Eliminating
we have
c^-\-C2,
Cbc {Pk
It is
evident that
- Pc) =
(Jkc
{Pc
- Pb\
Obc
the phases
Again,
between
(^ac
if
tials, it will
dis-
con-
sidered.
of the surface
o'ab
when equation
(571)
the phase
this.
If (J'ab=<5'ac+0'bc5
we may presume
make p^
it
C.
But
if
o'AB<C^Ac+crBc
and B,
mark the
for the
when equation
between
finite
428
J.
W. Gihbs
we
of phase C, or (as
TF=
(Tac
'*>ac
G^c >%c
Cab ^ab
- Pc Vc + Pa T^a + Pb Fb,
(573)
where /Sac, i^bc denote the areas of the surfaces formed between A and
C, and B and C, S^n the diminution of the area of the surface between
A and B, F^ the volume formed of the phase C, and V^, Fg the
diminution of the volumes of the phases A and B. Let us now suppose (Tac, O'bc, o'ab, Pa, Pb to remain constant and the external boundary of the surface between A and B to remain fixed, while pc
increases and the surfaces of tension receive such alterations as are
It
is
we may suppose
the changes
Cab
dW=z Cac diS^c + ^BC
-PcdVc + pj,dVj,-\-p^dV^c^'S'bc
C'bc
f^'^^'AB
Vcdpc.
(574)
requii'e that
dS^c
Cab
^^'^ab
pcdVc-^Pj,dVj,+pjidVB
(575)
0.
Hence,
dW=
Now
it
evident that
is
Vq.
will
Vcdpc.
(576)
diminish as
2^c
W'
1F"=0.
and
Hence
W"
W" denote
is
original
a negative quantity,
(577)
positive.
But
work necessary
its
W=
where
Let us
increases.
But
in the
and B.
It is
J.
W. Gibbs
Equilibrium
We
of Heterogeneous Substances.
429
insensible magnitude.
these differences
may
It
may
of
room on the
a mass of phase C
and
for
In this case
we may
consider a
A
mass of phase C which is
which
the
three
and B in virtue of a constraint applied to the line in
surfaces of discontinuity intersect, which will not allow this line to
become longer, although not preventing it from becoming shorter.
We may prove that the value of is positive by such an integrain equilibrium upon the surface between
tion as
we have used
Substitution
before.
by the use
or, it
may
formed,
direct
Now
We
ai-e
two
is
a gas-
430
TK Gibhs
J.
mixture which
satisfies
may
_2;iven
be expressed
in
we may
By the
many
eliminate as
simple gases
An
the gas-mixture.
in
may be regarded as a
fundamental equation for the surface of discontinuity to which it
equation obtained by such substitutions
are
as an indispensable preliminary, or
It is evident,
however,
we cannot
same
relations as
by the
tensions, temperatures,
inquire, at least in the
described alone.
If there
is
aftbrd
It
is
necessary
may be convenient
to regard
Now we
d{p'-p")
=.
7/s
dt
Fd/u
= - [iH - -yp,
p".
Vv'
//v")) dt
differential coetticients of
with respect to
evident that .
y
sion
j.
plane,
(5 V8)
and
it
is
-r
to a dividing
surface located so as to
make
the superficial
when the
W. Gihhs
./
latter surface is
431
e.,
by the
by
When
7/^,
on page 397.
may
to these variables
be represented
a simple
in
form
we choose
if
such substances for the components that in the particular state considered each mass shall consist of a single component.
This will
and
We
coefficients.
and
//s
dt
F^
dfj^
d2)'
dp"
= Vv" dt + y"
7/v'
dt
-\-
y'
F^^
dfj.^
dfj.^,
component
the
homogeneous masses
dj-t^^
we
d/A^^
sjiecified
specified
by
and
'
and
to indi-
".
obtain
dff
= (?k -;
We may
7v'
^ W) dt - ^ dp' - ^
y
(//,
^' ;a'
It is
is
and
would make
that 7
two
/"^^,
surfaces, or the
dis-
(without
(580)
represent the
;'
two
of dt
and write
(5 7 9)
strictly to
evident that
/)",
y
F^
djy".
da = -
cient
^^
entropy
not
then have
da =z
ume due
is
represents
system consisting of a unit of the surface of discontinuity with a part of each of the adjacent masses above that
which the same matter would have if it existed in two homogeneous
masses of the same phases but without any surface of discontinuity.
in a
III.
5.5
Nov.. ISTI.
432
W. Gibhs
J.
(A mass thus
JEquilihrium of Heterogeneous
existingf
without any
Siibsta)ices.
must
of
in
surfjice of discontinuity
The form
in
\dtjp
equation (580)
is
these quantities to the particular state of the system for which they
For
of the system.
state
ordinary components.
If
we eliminate
df/^
* If
we
and
=
=
dp
df
do'
//s
dp
r]y'
dt
//v"
dt
I\
~\-
d/.(
+ r/
+ ;//'
+ / d/./^,
+ y2 ^^/'a?
+ 7/3", c?/<2,
f//<i
'
manner
)'
easily obtain,
by means
equation (580)
may be
Hs +
)f
^'
of equations (93)
Es
Now
g-
set
in like
we may
and
'3
fZ/'i
and
(T
and
(507),
- i9 F.
(d)
written
d(y=
-K.di +
dE,
Vdp.
(e),
dHs -i)d V.
(e)
we
obtain
(/)
is
of energy and
when
the
and
fact,
H are much more simple in their definition than s and ??s, and would probably be
more naturally suggested by the terms superficial density of energy and of entropy. It
would also be natural in this case to regard the quantities of the homogeneous masses
as determined by the total quantities of matter, and not by
surface.
But such
tlie
surface of tension or
be extended so as to treat cases of more than two components with entire generality.
In the treatment of surfaces of discontinuity in this paper, the definitions and
The object of
in some
cases with advantage, and to show the precise relations between the quantities which
are used in this paper and others which might be confounded with them, and which
may be made more prominent when the subject is treated differently.
how
strictly
a different
adhered
to.
J.
we
W. Gibhs
433
obtain
do'
= j^ dt
(J
-{-
-^ dp,
(581)
.A.
Ji.
where
-A
= rt"r2'-ri'r2",
B=
c= i\
ir,"
Ki'
;/v'
r2')
r^'
^^2
(582)
(583)
(ri'-
(584)
ri")-
'
less
With regard
to gas-mixtures
suppose that there is not more than one component in the liquid
which does not appear in the gas-mixture. We have already seen
that in limiting the fundamental equation to plane surfaces we can
get rid of one potential by choosing such a dividing surface that the
shall
components vanishes.
if
there
is
no such component,
Let this be
if
such there
We
may
the equations
434
J.
W. Gihbs
da
ih^^.dt
dp 2
dp^
r2, o
= ^V2
=
f^/^2
<^<^
7/v3 ^^^
^^/^3
^zi\^
^tc.,
+ K2 ^t^2l
+
Ks
^/^35
etc.,
where the
suffix
relates to the
made
surface-
to vanish,
gases,
da
we
obtain
= (//sd)
?i'-'
-~
/;v2
'/V3
etc.
dt
-AiA^dp,-^^-l-^dp,-etc.
/ 2
This equation
aftbi-ds
(585)
wdth
ff
t,
to the
Extension of a
Surface of Discontinuity.
Let us first consider the case in which there is only a single component substance. We may treat the surface as plane, and place
the dividing surface so that the surface density of the single com-
and by
= ^%(,,,
(586)
(514),
*^
d(J
dff
dt
d\ogt
^.
^
ll;ie
total
volume
W. Gibbs
./.
Again,
the surface
if
is
435
the pressure in the liquid and vapor remains constant, the tempera-
by condensation
vapor.
If
of the
surface formed, and by a/m' and //m" the entropies of the liquid and
vapor per unit of mass, the condition of no addition of heat will
require that
^/('/m"-Vm')
The
= %(,).
(588)
(589)
^mT
/(W-'/mT
Hence, for the work done (per unit of surface formed) by the exter-
nal bodies
vm
shall
y/
\/
'/u
have
da
dt
dt
dp
da
d log p'
'
a.
Let us
now
not
dG
dp
make
is
Since
we
can-
it
will be best to
We
by
mass
shall
Quantities relating to
is
added
extended until
its
area
is
increased
by
If the sur-
is
by
'
?/v'
and
by
r
~
/"
r/y",
Their
respect-
436
J.
W. Gibbs
ively.
'/s
r
p
Vv
r
//
'/v
equal to
is
7j^,
by
have
We
\dt/p
difference
is
\d\ogtJp
between
this
'
In
In
these
(587)
constant.
considering
it is
is
necessary to
differ-
This
y^r.
r.
be
/do-
<^-'
\dp
-'+7'=-(i).
y"
;/'
will
and the work done (per unit of surface formed) by the external
bodies which maintain the pressure constant will be
W may
by
(581),
we
By
substitution in (593)
differential coefficients
obtain
1858)
W. Gihbs
./
W=-pj.
Q=-tj^,
where A, B^ and
It will
(584).
which
differ infiuitesimally in
When
quantity A.
the
composition,
by (582)-
and TTare
in general
for the
infinite
(596)
(J
43V
is
it
the
and
composition,
phases
absolutely identical in
are
by any supply
For
of heat.
the matter at the surface will not in general have the same composition as the
increased surface
W are
by the
fact that
for the
is
explained
in composition,
accompanied by the
is
is
necessary
we
changes sign
ments on page
When
i.
e.,
with y
vanishes
may be
This
extended,
by the extension of
to keep the
is
when
it
^" y2'
'^Yi'y/
>
^^^
the composition of
is
initially abso-
two
perature
The
most
is
will
effect
maximum
or
minimum
easily verified
befoi-e
either
by experiment
is
is
adjacent masses.
originally
superficial
tension
in
equilibrium
will evidently
between
coexistent
have sensibly
its
phases,
the
original value,
unless there are substances at the surface which are either not found
438
./
W. Gihhs
found only
But a surface
different tension.
component substance,
com-
in quantities
only a single
is
When there
are
of which
is
confined to the
tended,
i.
e.,
when the
is
surface
to cool
it;
is
ex-
but
if
the tension of any surface increases with the temperature, the effect
of extending the surface will be to raise
case,
face
it
is
will
its
temperature.
effect of
In either
effect.
must
in
short.
two
more than two components, there are two ex-
it
is
desirable to notice.
is
When
the superficial
normal
surface can be taken from the immediate vicinity of the surface with-
But
if
it is
same component
is
very small
in
each
mation or extension of the normal surface must come from a considerable distance.
Especially
if
we
ponding potential
make
in
the
J. TF. Gihhs
it
439
])lace
of raj)idity.
duced.
which
may
be
When
its
normal value
is
will
the extension or contraction of the surface, the phenomenon resembling that of viscosity, except that the variations of tension arising
from variations
same
in
in all directions,
property of the surface really analogous to viscosity would be greatest in the direction of the
We may here
homogene-
its
it.
When
surface-density
may have
a considerable positive
For the
by the
* When, however, homogeneous masses which have not coexistent phases are
brought into contact, the superficial tension
The
may
is
in a large
room
is
will
dimin-
we have chosen
for
incapable of resolution into other components which are independentlj' variable in the
homogeneous masses.
III.
56
Jan., 1878.
440
J.
W. Gihhs
Each of these quantities is exceedingly small. The surfacewhen positive is of the same order of magnitude as the thickness of the non-homogeneous film, but is not necessarily small com-
film.
density
homo-
The
is
sensibly afiected.
But the
[See (98).]
increased.
(21*7).]
be
not
efi:ect
and
will
positive.
be to diminish
[See (508).]
It
when
the tension
is
the surface-density
therefore
is
very
diminish the tension, but not possible that such a trace should greatly
increase
it.*
Impermeahle Vilms.
We have
which
is
impermeable to some or
The
a film of matter
is
all
of
de Physique, Ser. 4,
et
appears that one per cent, of alcohol in water will diminish the
The experiments do
not extend to pure alcohol, but the difference of the tensions for mixtures of alcohol
and 20 per
cent,
water
is
cited),
of soap
pui'e water,
He
same
but
if
solution)
the tension
is
sensibly
tension of the same solution as well as tlie great effect on the superficial tension
which a very small quantity of soap or other trifling impurity may produce, by the
tendency of the soap or other substance to form a film on the surface of the liquid.
(See Annales de Chiinie et de Physiiiup, Ser. 4, vol. vii, p. 409, and vol. ix, ]). .'i79.)
ficial
W. Gihbs
J.
Equilibrhim
of Heterogeneous
when
when
the film
a film of
oil is
may
Such
Stthstances.
be the case,
is
441
oil in
mass.
there
if
is
regarded as an immaterial circumstance, so far as states of equilibrium are concerned, and our formula^ will require no change. But
when there is no such indirect communication, the potential for any
component for which the film is impermeable may have entirely
difterent values on opposite sides of the film, and the case evidently
requires a modification of our usual method.
single consideration
Avill
component which
If a certain
cases.
is
cannot pass from either side to the other, the fact that the part of the
component which is on one side is the same kind of matter with the
must hold
true,
We
stances.
film,
ni\ for
j-i^
were a
different substance)
same
on the
first
and
if it
tension or otherwise)
if
we may
use 1\ and
siirface,
and
and
its
m%
and dividing
I^
to denote
sm-face.
tlie
On
values of
m/, y ^^
densities in the
two homogeneous
??ii',
masses.
a notation, which may be extended to cases in which
impermeable to any number of components, the equations
relating to the surface and the contiguous masses will evidently have
the same form as if the siibstances specified by the different suffixes
With such
the film
is
were
all
of
and
//j
really different.
//g,
components, and
/',, i
The
be a function
/^j
^^'il^
and
//g.
equal to
^^^
its
In a word,
may
all
differential coeffi-
if
442
J.
we remember always
according as
to treat the
component
^Substances.
as a different substance
it is
film.
When
fixes
there
and
in the film,)
is
component
specified
we
have
shall
If
we wish
we may
set a. single
symbol
for /<i
and
//g in
form of the fundamental equation. Cases may occur of an impermeability which is not absolute, but which renders the transmission of
In such cases,
it
may be
may
be
complete equilibrium.
latter
by
T7ie
Let us
now
deduce from it the conditions of internal equilibrium for such a sysIn accordance with the method which has been developed, the
tem.
intrinsic energy, (/. e., the part of the energy which is independent of
of gravity,) the entrojDy, and the quantities of the several components must each be divided into two parts, one of which we regard as
belonging to the surfaces which divide approximately homogeneous
masses, and the other as belonging to these masses. The elements
Ds
etc.,
-Z^f^,
etc.
We
Ds and Dv
resjiectively.
is,
Dm^
Din''
The
That
etc.,
volume Dv, by
use Dm^ or FDs and Diti'^ or
to an element of
shall also
Dm\, Dm%,
is
etc.,
(597)
etc.
(598)
J.
into
W. Gihhs
two
parts,
443
d/Be"
in
It will
z tlie
(599)
height of the
ele-
of reversible variations.
masses or surfaces.
We may
therefore regard
any infinitesimal
If
we then
in the
to
its
elements, and
several
by equations
(i;5),
we
shall
have
ft dBif
+./"//2
f^Uml
etc.
SBml + etc.
4- fg 6z Bm"+fg z dBm\ + /// z 8Bm\ + etc.
+ yV ^^ Dm^ + fg z dBni\ -^fg z 6Bra\ -\- etc. =
0,
(600)
initially in equilibrium, it
particular conditions of equilibrium relating to every possible deformation of the system, or reversil)le variation
between the
difl:erent
in the
distribution
of
The
from the nature of the system and from the supposition that the
changes in the system are not such as to affect external bodies. This
supposition
is
necessary, unless
we
444
J.
W. Gihhs
tern.*
we may
in detail,
tions
ft
diJt,''
SlJs
+ft dlJif =
+ fg
+//', 6Dm\
(601)
0,
6z Bni'-^fg 6z Bnv^
-\-fgz6I)rn^\
0,
(602)
+fgz6Dm\
+ etc. =
(603)
0.
in
will relate
in
rately.
The
This
maybe
fdJJif -i-fSBtf
To
it is
(604)
0.
= const.
(605)
is
thermal
the condition of
equilibrium.
The
may
also be written
be observt>d that
this condition
by heavy and
elastic
We
(606)
if
the
membranes with-
all direc-
The
tions in
*
0.
was
to
dis-
impermeable envelop.
to give the
influence of the
we wish
to limit ourselves at
we
shall give
up the
is
The
variations of the
it.
J. IV.
Gibhs
it,
are fixed.
siil)ject
and the
445
to the condi-
lines in
which
Avill
any
in (606) relate
the
we may
assign to
possible.
is
of the system, the element I)v will in general have a diflferent position.
Let the variation 6}) be determined solely by the change in position
of the element
Dv.
This
may
i:
which J^? are determined by the function mentioned
ax
ay
dz
and f^a', ^y, 6z by the variation of the position of the element Dv.
Again, in the initial state of the system the tension o", in each of
in
'
GJ.^,
which determine
is
the' position
two co-ordinates
of the element Ds.
In the
a function of
varied state of the system, this element will in general have a diffei-ent position.
The change of position may be resolved into a com-
ponent lying
in the surface
it.
r>^
da
P.
da
of the
446
W. Gihhs
J.
in wliieli
doo^
by
doo
(5^Cc5j,
lies in
the
With
by
divided
is
write
= Sfp Dv f dp Dv.
/"p 6I)v
But
we may
evident that
it is
6j'pI)v=fp6NDs,
where the second integral
relates to the
surfaces of discontinuity
bounding the volume considered, and 6]Sf denotes the normal component of the motion of an element of the surface, measured outward.
Hence,
fp 6Dv
Since this equation
system
is
divided,
fp
where
j)'
Ds, and
is
we may
dJDv =f{p'^2^")
(609)
<^iV^is
/
and
dfaJJs
where
c^
(positive,
(J
when
of the surface,
when
it
fG6nsz=fa{c^
-f-Cg)
Hence we have
for the
whole system
SNDsi-f2{(rST)Dl-fd(rDs,
(6 JO)
By
equilibrium
is
J.
W. Glhbs
447
+C2) 6NIJS
+fg rSzUs
0.
y 6z -\-
d]))
Dv
this condition,
Dl
it is
shall vanish
by the
Dv
shall vanish,
it is
necessary and
surfaces of tension,
(611)
shall
be a function of
is
z alone,
divided
such that
'-t=-gy.
(612)
dz
In oi-der that the coefficient of
movements of the
it
Ds
surface.
P^'or
it is
normal movements
we may
write
and
6 (1=0,
3 dN,
6z =z cos
The
first
y_y' =
(T(.-.-f.',)+^rcos5,
in
is
coefficient
= .,1.
of
Dl
(614)
in
any
in
-^
In order that the
(613)
respect to tangential
must have,
vertical
(611)
shall
we
vanish,
:^{g6T)=0.
(615)
This condition evidently expresses the same relations between the tensions of the surfaces meeting in the line
per-
which hold
for the
foi-ces in
equilibrium
in a plane.
III.
57
Jan., IS'ZS.
in
448
W. GIbbs
J.
is
The same
is
true
f6I>ml-J^fdDm\=0,
fdDml^fSI)ml
0,
etc.
(616)
(It is
i. e.,
that none can be formed out of others, and that the parts of the sys-
tem
in
which any component actually occurs are not entirely sepain which it does not occur.)
To satisfy the condition
by parts
]-ated
it
is
necessary and
fx^-\-gz=z3I^,
etc.,
(71/ J,
(617)
J/gj 6tc. denoting- constants,) shall each hold true in those parts
We
may
an actual component.
is
it
is
by parts
viz.,
that the
expression jA^gz must not have a less value in such parts of the
system than
contiguous part
in a
in
an actual
is
component.
From
we may
easily obtain
ff
Cg
first
and second
For
Cj
are
known
in
terms of the
differential coefficients of z
in
with respect
We
first
and second
differential coefficients
A'^ariables
d{p'-.p")
*
Tbe term
and the
that
if
definition
may be extended
a substance
is
face of discontinuity,
well as
when
= g{y"^y')dz,
it
to surfaces of discontinuity.
It will
117,
be observed
sur-
W. Oibhs
J.
449
j/^p"
= c/{y"-/)z,
(018)
z
to be measured from the horizontal plane for which p'=p".
Substituting this value in (<>i:^), and neglecting the term containing
where
is
we have
/ \
c,+c,=^iz'---Zl).,
where the
coefficient of 2 is to be
(619)
regarded as constant.
y"~}/'
is
We
very small.
may
we
If
sum
of the curvatures
its
dz'^\ d'^z
.,
.,
dz dz d'z
dxdydxdy
dy^jdx'-
dx^
A
\
dz'^\
dx'^
(/(y"7')
z.
(G20)
to the sign
fraction,
to be observed that, if
is
d'z
) dy''-
dy'^ I
dy'
With regard
it
we have
we always
value of the root, the value of the whole fraction will be positive or
is turned upward or
But we wish the value of the fraction to be positive
when the greater concavity is turned toward the mass specified by a
downward.
single accent.
We
The
last
is
surface.
may be regarded
is
number
of comisonents
is
and the potentials. For the conditions (612) and (614) may be regarded as consequences of (605) and (617) in virtue of the necessary
relations (98) and (508).
* Concerning another kind of conditions of chemical equilibrium, which relate to
the molecular arrangement of the components, and not to their sensible distribution in
space, see pages 197-203.
f
Compare page
206,
where
a similar
problem
is
450
J. TFT Cxlhhs
The mechanical
cial
importance, since
mass
In such a mass,
are observable.
in
when
as satisfied in
any
mechanical equilibrium will take place very soon; thermal and chemical
time for
its
complete equilibrium
When
in
may
is.
reached.
components
in
mobility
in
all,
want of sufficient
But when this surface
for
has only one component which does not occur in the contiguous
masses, and the temperature and potentials in these masses satisfy
the conditions of equilibrium, the potential for the component peculiar to the surface will
since this
(617),
is
mechanical equilibrium
relating to temperature
to be satisfied.
connection
the
of
conditions
of the surface.
with
The necessary
in
(614)
is
mass containing
this
by
the surface, the potential for this component in the surface will of
course be determined
The
by
tliat in
it
meets.
which may be regarded as expressing the relations 'which must subbetween contiguous portions of a fluid system in a state of
mechanical equilibrium, are serviceable in determining whether a
But the mechanical theogiven system is or is not in such a state.
sist
rems which relate to finite parts of the system, although they may
be deduced from these conditions by integration, may generally be
more
easily obtained
by a
by the application of
ordi-
will
by
this equation.
temperature and
tlic
iiotciitials
arc
not affecteil by
tlie
surfaces of
J.
W. Gihhs
Equilihriiitii
[Compare
discontinuity.
(228)
<>f
and
451
Ileterogeiieous Siih.stances.
(2:U).] *
it
(viz.,
level.
The
by any part of
the system of substances which are not actual components of that part
ing to such changes, and are not always distinguishable from them.
They
We
have
mation of new
surface
fluid
(see
fluid
and at the
a surface of discontinuity.
considered
is
surfaces.)
We shall
may
easily be
hereafter consider, in
extended to curved
and
])oints in
lines
meet.
to the
whole system.
Beside
the
permeable to
all
librium of the fluids relating to temperature and the potentials wUl not be affected.
(Compare page
extended
139.)
to this case.
in the
above paragraph
may
be
452
W. Gihbs
J.
tion of the
when
gravity,
all
which relate to small parts of the sysuse the method desci'ibed on page
413, the demonstration of which (pages 411, 412) will not require
any essential modification on account of gravity.
When the variations of temperature and of the quantities 3/j, M,
etc. [see (617)] involved in the changes considered are so small that
they may be neglected, the condition of stability takes a very simple
as those conditions of stability
tem taken
form, as
tem
separately,
we may
we have already
uninfluenced by gravity.
We
total entropy
in which,
perature and the potentials are satisfied, and the relations expi-essed
satisfied,
complete equilibrium.
is
not one of
in
'^1
and
form
(62
that of
o'
at
any point
in
(614),
dp-=.
and
gy
dz.,
da = g Fdz
W. Gibbs
J.
453
If
we
commencing
given
at the
obtain
AE=
where
f/,
(622)
(623)
tem diminished by
variations,
and
is
its
This
is
without
The condition of
(6 24)
fpDv-^j'ffDs
(625)
has a
maximum
p and
o',
ary cases
0"
may
z.
mass
(In ordin-
tinuity,
tinuity, even
when
the variations of
t,
M^, M2,
etc.
cannot be neg-
0)1
the Possibility
of
the Forination
When more
may
line, Ave
conceive of a
new
surface
being formed between any two of the masses which do not meet
sui'face in the original state of the
system.
The condition of
in a
stability
by
exemplified
by a system which
To
our ideas,
fix
is
equilibrium when
a very small
homogeneous
is in
formed.
let
which we
along a
may
line L.
call
in four
This
is
indicated in figure 11
by
all
surfaces,
meeting
454
J.
ISubstances.
and C
(to
U.
Fig.
infini-
be called A-C),
In an
we may conceive of a
by the same
Fig. 12.
value to the tensions of the several surfaces A-B, B-C, CD, D-A,
and are parallel to the normals to these surfaces at the point O
in
we can
mals at
two
and yd^ da
as
We
may
meeting
as
at O".
two
meeting at
O',
Therefore,
if
we
direction of the normal to the surface A-C, and the value of its tenIf the tension of a surface between such masses as A and C had
sion.
it is
would have
been stable with respect to the possible formation of any such surIf the tension had been less, the state of the system would
face.
have been
unstable
ill
at least
practically unstable.
To determine whether
it is
it is
prop-
W. Gihbs
J.
analysis than
The
result
we have used.*
which we have obtained may be
455
refined
generalized as follows.
When more
and equal
surface,
new
With
to its tension.
less,
when
the diagonal
is
in
which surfaces
lines in
We
all
We may here
stability
manner
But when,
of discontinuity meet.
surfaces
structed for
is
in the
in
of discontinuity meet.
might recognize linear densities of energy, of entropy, and of the several sub-
With respect
stances which occur about the line, also a certain linear tension.
to
these quantities and the temperature and potentials, relations would hold analogous to
sum
tension of the line L, this line will be in strictness stable (although practically unstable)
The
is
little less
and
C,
when
system
is
is
paper need
its
In the former part of this paper, in which the influence of surfaces of discontinuity
was
neglected, a system
would be produced by
was regarded
as practically unstable
when such
a result
a disturbance of the
relating to surfaces of
is
III.
58
March,
1878.
456
J.
W. Gibhs
siirface, biit
a certain diagonal
in
For
page 413
is
may be
similar proposition
enunciated in
method described on
evidently applicable.
If these
many
is
the lines in which these edges meet, the edges being perpendicular
to the corresponding surfaces, and equal to their tensions, and the
Now
may
if
the
be subtended by
the sides and bases of a triangular prism enclosing the vertical point,
or can be derived from such
by deformation,
the tensions will have the form of two triangular pyramids on oppo-
same base, and the system will be stable or practicwith respect to the formation of a surface between the
ally unstable
in a point,
is
sions.
in
which
With regard
to the formation of
new phases
We may
which
there are three such surfaces, this being the only one of frequent occur-
rence,
in a straight line.
It
will
be
which such a
line is replaced
by a filament of a
being
by
A, B, and C,
meet in a straight
less
difterent phase.
fluid masses,
line,
W. Gibhs
J.
system
which
is
is
457
cylindrical surfaces
straiglit lines.
surfaces.
brium
will
Fig. 14.
considerations.
Fig. 16.
Fig. 15.
Let Vd denote the volume of the mass D per unit of length or the area
ab c. Equilibrium is evidently possible for
any values of the surface-tensions (if only Cab, o'bc? ^ca satisfy the conmentioned above, and the tensions of the three surfaces meeting at each of the edges of D satisfy a similar condition) with any
dition
by only two
We
cylindrical surfaces.)
system as maintained
in
equilibrium
by
may
will
be bounded
and acting
at right angles to
C-D.
It will
be observed that
O'dc-
[See (615).]
all
the angles at ,
b, c,
and d
in figure 14
458
W. Gibhs
J.
(Tab, o^bc?
be supplements
To
d.
our ideas,
fix
2i
oTda,
^CA,
<5'ab,
ft
a,
<-!'
/i
cTda.
<3'db5
y,
antl iipon
Cdc,
ft,
ft
lines
The angles
y.
of
discontinuity at a,
Now
a
ft
6'"
y,
we
case
and d"
if S'
will
ft
curvilinear triangle
in
(3',
ft
6", 6'"
y,
do not
ft
y d'=.da
in a single
b,
and a y S"=zdb
a.
Thus
c.
together
fall
have
shall
and
b,
fall
In this
angles of the
right angles.
But
fall
it is
all
<5'dcj
eflfect
points
6',
a ft
y,
we may suppose
the
greatest of the tensions Cd^, Cdb, ^dc the two greatest, when these
are equal, and all three when they all are equal to diminish until
first
case
6',
S", 6'"
we may
is
by the distances of an
new
ft
y.)
surfaces are
bdc,
a ft 6'",
(Tda, Cdb,
1
6,
cda
ft
in
yd", y a 6"
little
when
the ten-
W. Gihhs
J.
Let us denote by
v^y
459
Sj,a, ^db?
the
dc
and
will
ahd', and
ah, c
d,
"'^S ^^^
ad,b d.
Also
^DB ^DB
and
TFv
The general
let
"T" '^'dC
=^^d ^d
*DC
^AB ^AB
<5'bc
S^q
Oqa
^Ca?
(626)
(627)
of
is
fixed,
may
be written
2{(}6s)-:S{pdv)
[See (606).]
If
we apply
tliis
Q.
(628)
by the
intro-
duction of the mass D, and take the difference of the results, suppos-
have
^da
(^da
we
if
we
differentiate (626)
0.
we
obtain
now suppose
(629)
size,
(630)
remaining always
<?'db?
tc.
S^B (^^mi
SDA<?<?'DA=if^(^DA*'DA),
We
TFs
Since
all
the quantities
ratio,
=^
d{(^I,B Sdb),
CtC.
(631)
TTv
O'bcSbc
- C^ca^ca),
(632)
whence, by (626),
TFs
The
= 2 Wy,
(633)
when
460
J.
W. Gibbs
[See (549).]
pj^^ jOg,
p^,
system remaining
W^
Now
he a minimum under
any constant values of
we may make Wp so small that when it
in
we may
d a, d b, d c, and regard
the figure
For the
itself.
total curvature
shall
for
abed
(^. e.,
b,
as remaining similar to
be, c
may
be regarded as con-
Therefore,
when
Vi,
is
adb,bdc,cda
and
all
Wy
so
is
if j^d
the tensions
ab c d, and
This shows that the system cannot be stable for constant pressures
D.
The components
determined by
tliat
of
phase of
which
satisfies
to
temperature and
the
by means
and of the surfaces D-A,
D-B, D-C, from the temperature and potentials of the given system.
Let us first consider the case in which the tensions, thus determined, can be represented as in figure 15, and
sistent with the equilibrium of a small
considering.
It
pj)
is
W. Gibbs
J.
that
its
'^c,
we may suppose
figure
abed
it is
y^,
We
manner.
jOa? i>B5
We
may
make
jOp
this quantity
supposed that
pressures, since
of
and
alone,
ratios,
ratios
ab
461
^c,
?'b
triangle
bdc.
be sin bed
sin
sin
ySp
da/J sin
Sfi
sin
y 6 fi
triangle
[D
udc
ac sin acd
triangle
yda 6a
triangle
y 6a
B C] [D C A].
:
Hence,
yA:yB:^^c:^D::[DBC]:[DCA]:[DAB]:[ABC],
(634)
where
4V[(0'ab+<5'bc+0'ca)(0'ab+0'bc 0'ca)(0'bc+0-ca~0-ab) (o'ca+O'ab-
may be
written for [A
O'ec)]
symbols, the sign \/ denoting the positive root of the necessarily positive expression
which
case of equilibrium,
follows.
when the
and v-D is sufficiently small. Now if pA=i-*B =/>c5 I^d will have the
same value, and we shall have by (627) Wy := 0, and by (633) Ws
0.
But when Vq is very small, the value of TFg is entirely determined by
the tensions and Vo.
Therefore, whenever the tensions satisfy the
condition supposed, and ?Jd is very small (whether pj^, p^, p^ are
equal or unequal,)
=: TFs
i?A ?^a
- IH Vb - Pc Vc,
(635)
^^ ""
Since this
is
[D B C]
[D C A]
[D
A B]
'
is
possible
when
462
W. Gihhs
J^.
when
jy^
small,
it
follows
C meet
is
When p^
a greater value,
if
practically unstable,
should be formed
Let
new
us
next
must be
only a very small mass of the
it
if
e.,
/.
would tend to
all, it
has
at least
fluid
increase.
in
If
the pressures and tensions are consistent with equilibrium for any
v-q,
and the
a b,b
lines
be together
will
tri-
For
c , will
c,
less
given values of the pressures and tensions, it will be easy to deterFor the tensions will give the total
mine the magnitude of v-^.
curvatures (in degrees) of the lines ab, be, ca; and the pressures
will give the radii of curvature.
In order that
determined.
necessary that
p-o shall
tensions of the
new
be
p-r,
case
is
only a
when the
These
shall
lines
be very small
evidently
is
it
Yet
less
if
the
represented as in figure
of
v-d
15,
may be
Vp
quite small
when
the value
little less
new
tensions of the
In any
sented as in figure 15, and v^ is small, W^ is negative, and the equiMoreover, TFg
is stable.
W^, which reprelibrium of the mass
sents the work necessary to form the mass D with its surfaces in
tensions of the
member
differ
of (636),
more
more or
less smaller
according as
values of
tensions)
which
will
we may
meeting
in
/)o,
we may
For
still
C-A
as
smaller values of
C-A
as capable
W. Gibbs
J.
determined
as
?'d,
the conception of a
may
It
be
in
But
463
in
the
The
When
much
new
Let us sup-
make
to
which
relates
it
when
somewhat greater than the second member of (636), more or
greater according as the tensions differ more or less from such as
is
Pj)
less
C-A meet
will
of such a mass as
tive.
The same
we have
W^ Wy
considered, since
will be posi-
jOr,.
of jOp, the value of TFg - TFy, which measures the stability with respect
to the kind of change considered, diminishes.
ai*e
finite
values of
It
p-^.
as
we now
suppose,
we must
also
which the surfaces D-A, D-B, D-C meet together (with the surfaces
in two opposite points.
If such a figure is to be in
equilibrium, the six tensions must be such as can be represented by
may have
We may here
a negative value.
three surfaces of discontinuity are regarded as meeting, but where nevertheless there
reaUy exists in stable equilibrium a filament of different phase from the three surrounding masses.
The value
W^
it
would be
would be necessary
The reader
(For the
We may
to
linear
we must
recog-
nize a certain negative tension in the line of intersection of the three surfaces of
discontinuity.
III.
59
March,
1878.
464
'L
ISubstances.
condition
we denote by mv
and form
?0s
it
may
D (of such
new
If
size
and by
easily
that wJs=f^vpositive
when p-^ is
with increase of
^s- Hence
B-C, C-A is
be
it
diminishes
garded
But
jf^,
all
as accurate,
in equilibrium
values of
which
will
px))
be so long as
has a sensible
so long as our
size.
new
much
surfaces are
in
which
It is
formed, since
it
(^TiYit
would take
if
it
by
should be
is
we have
obtained.
pair the stability of the line considered, as deditced from our equa-
Nor
tions.
demonstrated
may
fail
to be realized
when
is
(See
page
426.)
fluid
masses A, B, C,
meeting
mass E.
in other respects,
tical point
W. Gibbs
J.
system can be
If the
in
equilibrium
when the
465
it
is
evident that
will
have
and
will
The
E.)
meet,
if
a mass
should be formed,
it
l)e in
pended
in
by W^ the work
may
ex-
easily
W^,- Wy=^W.,;
whence
also, that
when
the
volume
(638)
is
will
be
W^ and W^
(637)
when
The ten
magnitude and direction by the
tensions
may
all
then be represented in
/?,
y,
by the
6, , viz.,
a /5,
line
the tension of
etc.
are suppressed to
make room
for E,
we have evidently
all
(639)
Wg,=
0.
Now
is
v^
Pb,
is
W^ =
0, if
p^
But when
Pc
Pt>i
may be
neglected.
jo^,
by the
466
J.
W. Gibbs
relation
critical
and
satisfied
is
v^.
This
small.
gives
P^.
Then
may
be the volumes
of
E
is
we may suppose
a, b, c, d,
it
^^b,
The
e.
ai'e
a fd y S e,
common to the
^d will
divided
^c?
may be
its
the
all
will
be called
mass which
after the
?.>a,
surfaces to be plane.
named
(640)
we
is
common
yS
to the lines
recollect
y6.,6s,ey,
i.e.,
to the surface
ular to 6
^,
f, etc.
etc.
the tetrahedron or the area of the triangle specified, sin (ab, be),
ein (abc, dbc), sin (abc, ad), etc. the sines of the angles
lines
and surfaces
specified,
and [B
CD E],
[C
D E A],
made by
etc.,
the
the vol-
t'B
bcde
tetr
acde
tetr
be
pade
tetr yfide
[B
C D E]
tetr
:
[C
aSe
tetr
'
sin {ySs,
aSs)
sin (/i^^f,
(i^f:i)
trian
ade
trian fJSe
yade
D E A].
Hence,
^-'a:Wb:^>c:'-'d::[BCDE]:[CDEA]:[DEAB]:EABC],(641)
and (640)
^'^
may be
written
_ [BCDE];>^+[CDEA]/>B+[DE AB]pc+[EABClgg
[BCDE]-f[CDEAJ + [DEAB] + [EABC]
,^^^.
J.
Wi Gihhs
is less
than
this,
when the
467
D meet is
any mass of the nature of E.
But if the value of p-^, is greater, either the masses A, B, C, I) cannot
meet at a point in equilibrium, or the equilibrium will be at least
point where vertices of the masses A, B, C,
relation, the
practically unstable.
When
new
critical relation
vex toward
when
with the other tensions, these surfaces will be contheir tensions are too great for that relation,
In the
case,
first
71 y
is
is stable,
is
when
which
at a point
will
form
Avill
/>e is
be so small that
when p-^
is
smaller
it
in
meeting
This
may
be neglected.
sub-
is
mass
This will
than the second number of (642). In the second case, the equilibrium
of tlie five masses A, B, C, D, E will be unstable, but the equilibrium
of the four masses A, B, C,
jt>E is
member
will
is
of insensible magnitude.
greater
in general
This will
considerably greater
of (642).
JAquid Films.
When
its
between other
even when
its
thickness
films,
fluids
we
shall
are
gaseous.
The
fre-
give
is sufficient
fluids,
Avill
The reader
is
will
To
fix
our
is
supposition.
its
of
the film.., \t
is
such portions
approximate equilibrium
between the difierent parts of any such element and the other fluids
which are immediately contiguous, than for the attainment of equi-
468
W. Gihbs
J.
librium between
all
There
will
may be regarded
as satisfying the
may
not satisfy
It is M^ien the
all
changes due to
this
want
oi
rest,
except so far as
it
accommodates
is
it
itself to
any change
in the
properties of the film are most striking and most sharply defined.
all
its
elements
each element
is
The occurrence
may
somewhat
sion of this
principal
films
in
general.
Let us
first
ele-
The same
surfaces.
will
is
is
one com-
face alone.
If
we regard
its sur-
by
dividing surfaces which make the surface-density of this component vanish, the thickness will vary inversely as the area of the ele-
ment of the
film,
in the
nature or
in those cases in
itself
J.
When
W. Gibhs
on
its sui-faces,
be found
469
will
in greater proi>ortion
When
on the surfaces.
the film
is
ex-
The value
the film.,
{i.
e.,
of the elasticity
by the
surfaces divided
of
its
may
face),
We may
known.
illustrate this
by
a simple example.
Let us suppose that the two surfaces of a plane film are entirely
alike, that
Let us
except two.
call these
Let us denote by y
by such dividing
surfaces
and
j/g ^^he
densities
by A the thickness of
as make the surface-
of the film
by ^the
elasticity
From
of
/Nj
the definition of JS
we have
2dff
= B-,
(643)
ds
)/j)
d{X y^
^ri
film
2 -^2(1))
(644)
^X3+2/'2(i)
Hence we obtain
\ y^
=. y
+ 2 ^2( 1)) =
A dy^.,
d\
ds
(^
r2
72
<^^
'^
(h'2
2 f?^2(i)
470
J.
W.
(xibbs
-^Vi'^^^^d^rJ^2^v-=-^ri(b'2-\-^r-i^b'i
s
If
we
set
we have
.j,
= ll,
(646)
'
^ y^dy^nXldy,
(647)
y.'
ds
and
= \ y^dr 2d^2^^,.
2 I\^^^
With
tliis
equation
eliminate dcf
we may
by the necessary
Avill
We may
also
dff =^
This
-(648)
^2(1)^^2-
give
4 / 2^2 ^^,^
^ (A ^^ dr-^2 dP^,,,),
(649)
or
~+
^^4^= A y,
where the
are constant,
and
all
-%^\
(650)
may he
//g
The
latter con-
ir i-Vi') '^Mi
in
and
/u^
-f-
tiguous gas-masses.
densities of
[See (98).]
When
its interior
/S^
(65i)
o,
and S^
in
the con-
and
in
masses are known in terms of the temperature and potentials, equation (650) will give the value of
together with
If
E in
A.
we write G^ and G^
and
S.^
per
we have
G,=:ly
(652)
G,=Ay,-\-2r,,,,^,
(653)
Therefore,
G^=G,r+2l\^,
=A;.,i^+2^i^lll,
(p)
(654)
is
constant.
Therefore,
J.
W. Gibbs
471
2(1)
\ dfi^
k-
the last differential coefficient being determined by the same condiIt will be observed that the
value of
other components,
yu,
and
ju.^,
The
considered.
increase
of
The
Nor
rior drains
away
same
be altered,
stance
/S'j,
tension.)
in
Yet
the case
it
is
we have
may
easily be
just considered,
if
may
film
it.
When
a plane film
may be reduced to
By restoring the
tion.
film
equality
original position
It
is
we may
Nor
can this be
show
III.
60
movements
The
differ-
March,
1818.
472
J.
W. Gihhs
we account
when held
power
for the
increase of tension
by
vertically.
viscosity,
it
will
its
its
This
is
its
phenomena
of the
is
phenomena
is
the immediate
an increase of
its
tension,
by which
it is
The phenomena
to
described
by M. Plateau,* an
increased tension
is
manifested in a
The warmth
thin enough to
show
When
a diminution of thickness.
returns to
would be
its
original color.
resisted
by any
the finger
is
of a
which
appear indicating
We have
so far supposed that the film is thick enough for its intehave the properties of matter in mass. Its properties are then
entirely determined by those of the three phases and the two surfaces
rior to
of discontinuity.
From
these
we can
and contraction of
its
limit.
Yet a
which we
may
* " Statique experimentale et theorique des liquides soumis aux seules forces moleculaires," vol.
i,
p.
294.
W. Gihhs
J.
it
473
relates to
is
bution of
its
not
With
satisfied.
components, a film
will
general be stable,
when
its
interior has the properties of matter in mass, with the single exception
of variations afiecting
its
With
equilibrium
is
neutral.
it
is
become unstable
interior ceases to
in this respect.
For
it is
not likely
reduced
sufficiently
the
first effect
in tliickness,
it is
most natural
to
when
it
suppose that
(We
liquid films
when
sufficiently
Let us
enough
now
reduced
air,
in thickness.)
is
in
is
thick
mass, and
tiguous gas-masses
and
such a film
may
its
several elements,
With
is
That
is,
the
when by a we understand
equivalent to the
by
/" its
474
W. Gihbs
J.
application of these
equations.
in
which requires
(612),
it
z,
and condi-
which makes
it
a certain
Nor can
which
is
free
is
contrary to what
to
move independently
of the surfaces,
we have supposed.
is
were
tions
is
horizontal.
For
if
these condi-
we
impossible that
all
if
when
the temperature
is
is
not hori-
uniform,
is
it
not due
it
must have
nioi-e
(617).
J.
The
W. Gihbs
difficulties of
475
filled.
effect
entirely ful-
is
made
for such
most symmetrically
The
rises to the top of a mass of the liquid.
motion of the liquid, as it is displaced by the bubble, is evidently
such as to stretch the two surfaces in which the liquid meets the air,
This will cause an
where these surfaces approach one another.
increase of tension, which will tend to restrain the extension of the
The extent to which this effect is produced will vary with
surfaces.
The formation of a
when a bubble of air
motion of the
surfaces
increasing
in
more permanent
film,
their
by
its
and the
But
this
running out of
viscosity as soon as
it is
tension will
able for supplying the substances which go to form the increased surfaces.
We
may form
is
by
we
obtain
V= 581 Z)3,
where
V denotes
the
mean
(656)
[i. e.,
that velocity
The
xl, S.
calculation of formula (656) and that of the factor (5) applied to the formula
of Poiseuille,
476
W. Gihbs
J.
which,
if it
fixed planes,
D denotes
in
is
water
in
equation
between
difi^ers
sup-
This
337 Z>3
F=r
for descent
is
is
equivalent to
899 Z>2
parallel planes.
(657)
The numerical
coeflicient in this
is
derived from
we may
at least con-
fifth
order in a film of
water), and .036'" per Jiour for a thickness of .0001"' (which corre-
On
tenuity.
tlie
is
in
1"'"'
A little
sometimes be extended so as to form a large bubble. Since the elasticity {i. e., the increase of the tension with extension) is greater in
the thinner parts, the thicker parts will be most extended, and the
effect of this process (so far as it is not modified by gravity) will be
to diminish the ratio of the gre'atest to the least thickness of the film.
During this extension, as well as at other times, the increased elas-
due to imperfect communication of heat, etc., will serve to protect the bubble from fracture by shocks received from the air or the
ticity
* Ibid., p. 653
ix, p. 5.32.
W. Gibbs
J.
bubble
pipe.
If the
tion
(613) will
477
is
The bubble
will then
ing to the top, more or less slowly, according to the viscosity of the
The
film.
and
is
may
components S^ and S.^, the latter denoting (as on page 469) that
which exists in excess at the surface, one element of the film will
tend toward the same level with another, or a higher, or a lower
level,
quantity of
aSj
in the first
element as
less ratio.
When
and
its
process
have the
be subject to the
we have already described, if it is truly fluid, howmay be. It seems probable, however, that
viscosity
is
still
in virtue of
it
may have
those which are caused by gravity in the interior of a vejy thin film
which
satisfies
However
potent
in
this
may
be, there
producing changes
is
a film,
in
when the
often
is
more
interior.
where the
film is terminated.
if
its
bounded by concave
less
surfaces,
and
in
film.
which
which the pressure is therefore
This liquid mass therefore
thickness
is
rapidly reduced.
This
efiect
is
best seen
when
its
a film
Unless
diminished thickness near the edge causes
a rapid upwai'd current on each side, Avhile the central portion slowly
the film
is
very viscous,
its
478
W. Gibbs
J.
film, where the edge is nearlywhich have become thinned escape from their
position of unstable equilibrimn beneath heavier portions, and pass
upwards, traversing the central portion of the film until they find a
position of stable equilibrium.
By these processes, the whole film is
rapidly reduced in thickness.
descends.
horizontal, portions
suction which
produces these
following considerations.
effects
may be
The pressure
is
the
in
of course variable,
being greater
in the lower portions than in the upper, but it is everywhere less than the pressure of the atmosphere. Let us take a point
where the pressure is less than that of the atmosphere by an amount
represented by a column of the liquid one centimetre in height.
(It
is
much
probable that
point near
by
At
limited
space must be
its effect in
producing a current
in a
that of
gravity.
liquid
in
the
There
when
the
is
a third
principal
through the
air.
is
component of the
interior
is
volatile,
viz.,
if
we
may
be
and
condensation in the lower. These processes, if the atmosphere were
otherwise undisturbed, would occasion currents of diffusion and other
currents, the general effect of which would be to carry the moisture
downward. Such a precise adjustment would be hardly attainable,
and the processes described would not be so rapid as to have a practhe same value in the atmosphere as in the film,
tical
it
easily
importance.
dry atmosphere, or a
W. Gihhs
J.
atmospliere,
tlie
In the
neglected.
evaporation or condensation
cftfct of
tirst
is
479
not to be
densed cannot
in
all
down by
the
internal current but that the two processes together will tend to
,
wash oiit
But when a com])onent which greatly diminishes the tension of the
although forming but a small fraction of
film,
its
mass, (therefore
evaporation
is
of
and condensation may
the
sui-the
film
as
in
the potential for that component is the same in
rounding atmosphere. To illustrate this, let us take the simple case
of two components /S'j and xS'^, as before. (See page 469.) It appears
from equation (508) that the potentials must vary in the film with
the height 3, since the tension does, and from (98) that these varia-
be considerable, even
tions
;/j
The
film.
and
densities of S^
S.^
to another
is
therefore as
much more
is
aS'j,
level
as its
and condensation of the two substances are the same, these processes
It will be
will go on much more rapidly with respect to S^.
observed that the values of
-^
and
to pass up.
So,
it
will
may
easily be
These
efi'ects
same kind.
which
/Sj.
sigiis,
effect
S^.
have opposite
down through
Moreover,
evaporation or condensation of
will
dz
dz
For
if
will cause a
condensation
diminution of
its
We may
infer that
it is
it
But apart
volatile.
III.
61
we have
seen that a
April, 1878.
48C
filiii
J.
W. Gibbs
which
is
by the
is in
and the suction at its edge. Sooner or, later, the interior wall somewhere cease to have the properties of matter in mass. The film will
then probably become unstable with respect to a flux of the interior
page 473), the thinnest parts tending to become still more thin
any external cause) very much as if there were an
attraction between the surfaces of the film, insensible at greater dis(see
(apart from
tances, but
ciently reduced.
We
is
film is
suffi-
In a film
of soap-water, however, the rupture does not take place, and the
processes which go on can be watched.
superficial observation that a film of
It is
which the
is
interrupted
black spots.
That
by
tint is
approaching
in
shown by the
is
made
when a
single
This
is
film,
black spot breaks out and spreads rapidly over a considerable area
Avhich was before of a nearly uniform tint approaching the black.
The edge of the black spot as it sj^reads is marked as it were by a
becoming
spots.
ently
larger, glide
They
many
down
times thicker
Now
than
if
much
thicker
appar-
it is
J.
W. Gibbs
That which
black spots
ner.
most
is
difficult to
account for
is tlie
by
fonuatiou of the
iu the
It
this,
481
grows
if possible.,
thin-
by passive
For
unstable by the
the film.
it
flux of
matter from
its interioi-,
film.
is
necessary for
is
connected with
an
its
stability
excess
of
the
its
film.
With
the interior has ceased to have the properties of matter in mass, and
not be applicable,
Avill
arrest
of
soap or some of
still
its
components
The preceding
still exists.
currents
by
If therefore
the increasing
in the interior of
interior
gas.
internal
is
it
extension
reference to
diftei-ence of
we account
density
the film,
of
we must
composition
in the
been obliterated.
liquids or
to films of gas
between
* The experiments of M. Plateau (chapter VII of the work already cited) show that
this is the case to a very remarkable degree with respect to a solution of saponine.
With respect
to soap-water,
we
are considering,
is
anj^ greater
superficial
to.
482
J.
W. Gibbs
masses of
liquid.
This
liquid.
may
latter
may
the surface of a
be done
mass
of the
same or a
is
is
not broken,
is
formed by a bubble of
volume
different
(If
air ris-
drop.)
carry
it
momentum
which
air,
of the
is
the counter-
drop
appears
it falls, it
Since,
how-
is
component of
tlie
in
side of the air film, the necessary differences of the potential of this
persistence
compared with
is
in
greatly reduced.
Compare page
their tension
is
479.
We
In this
and
Fluids.
This
is
by
far the
for
On
is
But we
must take account of the nature of the surfaces of discontinuity
tinuity in determining the state of strain of solid masses.
They
are distinguish-
able from ordinary air-bubbles by their general behavior and by their appearance.
one appear-
Tliis is of
tlie li(j\iid.
course an
J. TF.
between
Gihhs
Equillbriinti of
and
solids
fluids
483
IIeteroge)ieoiis Stihstances.
and a
line structure,
stress
compatible with a
We
shall
{i.
we
of the non-hoiiio-
e.^
entirely determined
is
it
(see
The
vanish.
suj)erficial
may
and of the
page 397),
^S(l)5
ff
VhDI
^ 2(1)J
^ 3(1)5
6tc.
^=^S(i)-^Vs(i,-/^2 ^2U)-/^3^^3(l)-etC.,
in
which
yUg? /^s?
^^c.
(659)
the potentials
As
0'
in the case of
as expressing the
of discontinuity
specify
but
it
of the surface.
The
latter quantity
spent
m forming
the surface.
With
G depends upon
the
work
these processes are not distinguishable, unless the surface of discontinuity has com2:)onents which are not found in the contiguous masses,
and even
must be supposed
to be
formed
out of matter supplied at the same potentials which belong to the matter in the surface,) the
work
484
J.
nially
by stretching
is
Substances.
masses
there
is
and
is solid,
its states
surface
tlie
With
we now
meets a
it
fluid,
when
sf>lid
at
It will
be necessary
for us to
is
may be
it
by the
influenced
state of strain
tlie
in
minima.f
may
This
line
discussion.
if
we
By applying
alike.
vanish in
forces,
fluid),
and
let
may
The
its interior.
Let us consider
tensions of the
two
we
can
make
all stress
evidently have different values in different directions, and are entirely different
may
from the quantity which we denote by a, which represents the work required to form
a unit of the surface by any reversible process, and is not connected with any idea of
direction.
it
not greatly
differ.
if its
fluid,
is
numerous
as solids,
its
insensible gradations,
ficial
This
we may
fluid to the
solid state
has been by
was decidedly solid, and that they will only differ so far
by subsequent variations of temperature and of tlie
Moreover, when an amorphous solid is in a state of
the solid.
as they
may be
differently affected
stresses applied to
it
may have no
it
seems not improbable that the particles at its surface, which have a greater degree of
mobility, may so arrange themselves that the value of a will coincide with the superficial
The
differential coefficients of a
latter quantities.
W.
J.
Gihhi<
may
with crystals.
Also,
on
its effect
quite otherwise
it is
solids are
in
fluid in
Avdiile
485
which
can dissolve
it
bounded
is
in
For
as follows.
we suppose
If
is
may
be deduced
infinitesimal portion
fluid,
will
where
Ds
])Osition
is
its
dissolved),
and
e^
Cg
principal
its
curvatures
(positive
when
%!, the surfacedensity of energy, e^' and y" the volume-densities of energy in the
solid and fluid respectively, and the sign of integration relates to the
elements Ds. In like manner, the inci-ements of entropy and of the
quantities of the several components in the vicinity of the surface
their centers
will
lie
solid),
be
.fVh'-Vv"
(<''l4-''3) '/SU)]
S^^J^S,
etc.
the
space
wdiole
occupied
by the
solid.
fluid
except
that
which
may be
may be
yw^",
components by
* The potential
is
//
"
is
by t, and those
|^2^ tc.,*
relating
to
the
If to this
and to distinguish
it
fi
486
W. Gibbs
J.
we add
tlie
Now
by
(93)
we have
r=
//'
By
f,"
t ,1,"
+ ///' ;//' +
//^
y/ +
etc.
this equation
energy
in
/[fv'
tlie
6N^
it
(660)
is
necessary
tion of equilibrium
^,,^eV-tv.J+if + i^e,+e.^a^
y
This equation
containing
(7,
(061)
is
that
when
tlie solid
-|(<^'i
c^) ^
^^'^^^
plane.*
is
has no
by
sti'esses
ff is
except an
member
represent [see equation (93)] the A'alue of the potential in the solid
for the substance of
which
it
consists.
equation reduces to
that
is, it
the same
if
o"
fluid.
all solids is
when
(Cj -\-<-2)
jo"
+ (c^ +^2) ^
shall see,
may
mass (when
with
if calcu-
member
The other
o"
potentials
fi^, etc.,
we
have
the same values as in (659), and consist of two classes, one of which relates to sub-
stances which are components of the fluid mass, (these might be marked by the double
accents.)
The expressions
to
relates to substances
all
have the
is
therefore
value zero.
7/
in (661).
is
denoted by
T,
which
W. Gibbs
J.
E<piilibrhiiii
when the
of eciuatioii (661),
tropic,
of Il'eterogeneovs
is
487
Stibstauces.
-\- {ci -{
may be
isotropic stresses
accuracy by saying
that the potential for the substance of the solid in the fluid must
have this value. In like manner, when the solid is not in a state of
isotropic stress, the difterence of the
two
aud
second
member
of the ecpiation
v'
may be
c^)
ff
represented the true pressure in the solid in the direction of the norTherefore,
we had taken
for
ff
fluid, as it
if
does
when both
would
not have led us into any })ractical error in determining the value of
the potential
hand,
if
which
/<,"
is
On
the other
differs
0'
in (661) will
/./
," in
-\-
c^
0.
When
may
crystalline bodies,
in a state
of isotropic stress.
jj^"
here.)
The value of
new
matter
is
is
all
if
is
here
left
new
It
solid matter
out of account.
might
differ in
134-137, but without reference to the state of strain of the solid or the influence of
the curvature of the surface of discontinuity.
III.
62
488
if
tT.
W. Gibhs
is
and then,
solid,
isotropic stress
is
fluid,
if
would be
commence and go on
likely to
The
equilibrium for any single point at which the fluid meets the solid,
but
it
by
most important
case, since
it is
is in
least
that
is
at
stance of the solid which would belong to the solid mass at the
temperature
Or,
if
we denote by
temperature
{p'^
{Ci-^-c.^)
G must be equal
to ///'.
/ij",
the condition
may be
(y)z=y'+(c,+
Now
if Ave
write
;/"
(662)
C2)(7.
we have by
(612)
dp)"=-gy"dz.
By
d{p')^y^di.i^\
(98)
and by (617)
d^i^' =z
d {p')
whence
=1
</
dz
</
^^^
'
Accordingly we have
d{p')^dp"
= g{y"~.y,')dz,
and
z being
we
obtain
c,+c^= ^^^'"~^''h
generalized so as to hold true of the formation of
of
new matter
is
new
solid matter of
any kind
if
(663)
will not
the second
any kind on
be formed upon
member
of (661) cal-
J. TF.
Gibhs
common
their
489
[Compare (6 19).
The obstacles
is
now
fluid will
may
which the
in
solid is
consist
be regarded as invari-
of the
some
of pieces of ice.
Proceedings of
Royal
vol. xxi, p.
tlie
takes place
vi'hich
440
Society, vol. x, p.
Although
146.)
ser.,
by the directions
phenomena have not been observed
crystalline structure,
is
far as
body of
this is a
general character
is
concerned,
is
ice, v?e
may
vsfith
effect,
of
to
so
an isotropic
body.
we may
neglect the differences in the values of ciw (the suffixes are used to indicate
that the symbol relates to the surface between ice and water) for different orientations
of the axes of crystallization,
if
measured by a quantity
when one
and
In
when
surface.
of
ctu,
which
is
is fluid,
tendency of pieces of
case
of
faces
is
ffn
it
evidently not
is
suffi-
when meeting
in
2(Tiw,
and
It vdll
With
Royal Society
is
in
407.
490
J.
W. Gibhs
If the crystal
able.
I'
cot
oo') SJST,
where fy' and fy" denote the volume-densities of energy in the crystal
and fluid respectively, s the area of the side on which the crystal
grows, (1) the surface-density of energy on that side, s(i) the surfacedensity of energy on an adjacent side, go' the external angle of these
two sides, I' their common edge, and the symbol 2' a summation
with respect to the different sides adjacent to the
first.
The
incre-
equal to zero,
we
condition of equilibrium
{es'-tn.'-li,"y,'+p")s6N
+ :^"
where G and
= fy'
will be
It
cosec Gj'(f
7k'
I'
cot
same
go')
6N=
sides as
fg.^,
(G64)
0,
and
fsd)' i"
This gives
{ p"
'2'i a'
I'
cosec
vr^^-
gV cot go'\
go
(')
jy-'
small or narrow,
which
0-' /'
/'.
we may
is
same value of
//," as equation
(387), or as
is
ing some other body, the condition of equilibrium for the crystalline
foi-m
is
gl)')
/rfifi^
shall
G and
s ai-e
in
value
when
tal
its
similar crystals
in
first side.)
this condition, as
of
mea-
W. Gihbs
J.
491
be
preceding equation.
If
we
of the areas of
all
its
The value
respectively.
of the fraction
is
d2{o' s)
dv
as determined
ai-e fixed.
sides
tal
is
that the
value of this differential coeflicient must be independent of the particular side which
is
supposed to be displaced.
condition of equilibrium
satisfied, as
is
constant volume
Foi- a
minimum
has therefore a
s)
may
value
easily be
when the
proved more
directly.
When
represents the
crystal,
and the
resents the
fluid
indefinitely
is
work required
coeflicient of s
work gained
to
its
in
We may
like the
W^
and
Wy
the
the surfaces.
work gained
in
\J.e., ^'((Ts)],
Equation (664)
may
then be written
6Wy+l:{0 6s)z::zO.
Now
(667)
d TFs =
and that of
Wy by
(667).
d:^{ff s)
= I 2{0 ds),
Hence,
dW^=^dWy,
The
variation of
492
J.
W. Gibhs
Wy
vanish together,
TTs- TFv
= iWs==iTFv,
(668)
the same relation which we have before seen to snbsist with respect
to a spherical
mass of
425, 465.)
coincide
(when the
Yx'i
^y'l Vv'
gravity
is
with
The values of
Now
concerned.
by
since
= -gr"dz,
=. g dz,
d{r,'M^"-2n = ^{r"~ri')(^^'
dp"
and
d/.i^"
we have
Comparing
(664),
we
crystal will have the greater tendency to grow, (other things l)eing
equal,) according as the crystal
When
on
is
two masses
may
be neglected.
If
we
is
regarded as
in
a state of
and tempera-
make
the
d(r,'^i,"-p")
Q^
dz
wlien the value of the differential coefficient
is
is
determined
in
accord-
dp))t,m
}//'
equivalent to
{X
=J_.
\drn^Jt,p,m
(669)
y^'
may
suffixed
is
used to represent
coefficient.
all
etc.,
except such as
W. Gibbs
J.
The tendency of
a crystal to
grow
be greater^in the
will
493
or
upijei*
fluid,
may suppose
Again, we
the composition
by the
and
?/v'
fluid
in
no
?;v'
foi'
will
in
mean temperature
the
may
=.
same crystal
These condifluid is
we cannot
we may regard
t dti^'.
its
then
Therefore,
their
if
we
treat
e^'
-t
shall then
may
make
We
z.
and
fluid
Owing
gently stirred.
fy'
the
to be determined
regard
with
is,
of
The temperature
t/y'
as constant.
effect of
We
gravity shall
vanish
dz
which
or,
by
dp Jii,m~ ^i"
(90),
=^.
i^)'
Since the entropy of the crystal
is
(670)
sure
is
is
The manner
paragraph.
in
This
is
When
when the
crystal
is
growing
is
consider-
or dissolving rapidly.
ably supersaturated, the action of the crystal keeps the part immediately contiguous to
it
The
most exposed
grow most rapidly.
The same
flixid,
and
will
Bd.
ii,
S. 1.
i,
S.
453
Wiedemann's
494
J. TF.
Gibbs
fluid
is
supersaturated
grow
as
is
not to be
all, it is
minimum
has a
s)
may
much
only so
at
value (or
result.
fluid.
is
(664)
intended unless
6N
is
we can only
con-
clude with certainty that sensible changes cannot take place for
it is
may
equilibrium
subsist
explained (see Clausius " Ueber die Art der Bewegung, welche wir
Fogg. Ann., Bd.
c, S.
353
to evaporation,
Warme
XIV,) by suppos-
ing that a passage of individual molecules from the one mass to the other
many
the solid.
continually
between amorphous
is
may
is
nennen,"
fluids,
there
may be
mass of
would be
sufficient to
and
it
make
the probability
equilibrium that the theoretical conditions deduced above should be precisely satisfied.
But
this supposition
to a
very small
side.
The following view of the molecular state of a crystal when in equilibrium with
Since the molecules at
respect to growth or dissolution appears as probable as any.
the corners and edges of a perfect crystal would be less firmly held in their places
than those in the middle of a
theoretical equilibrium (6G5)
on each side
side,
is satisfied
we may
The boundaries
of these
b}-
may
indeed attach themselves to the side of the crystal but they will speedily be dislodged,
and
if
W. Gihhs
J.
495
which three
relates to a line at
is
we apply
diiferent
If
solid.
it
the
evident that
is
to the surfaces, a
f^{GST)Dl
to
definition of
ff
in its
in (611),
be made good
greatly to affect the general smoothness of the surfaces, except near the edges
the surfaces
fall off
/i
new
except
when
the side
will
is
however be
the whole,
it
The value
of
/^
"
which
the case
is
2((7 s) shall
finite quantity.
commencement
and
is least.
principally determined
is
by equation
(665),
by the
[i. e.,
the
by a
(665)
On
very small.
where
"
I
side,
Now
value of
In
neces-
it is
new matter on
the deposition of
is
surrounding surfaces.
As
necessary to
at
more
Vjy
by
their size
and
relations to the
least readily,
new
If
bounded by
satisfying this condition cannot form a closed figure, the crystal will be
of surfaces
The kinds
of
surface thus determined will probably generally be those for v/hich g has the least
values.
But the
relative
It
crystal
wiU
finally
sides,
even
if
be such as to make
unmodi2(cts)
does not appear that any part of the operation of removing a layer of molecules
new
layer
yet
the values of ^ " which wiU just allow the different stages of the process to go on
must be slightly different, and therefore, for the continued dissolving of the crystal
,
the value of
It
//
"
must be
is
The
this
(665).
which a has
done as slowly
sides for
when
it is
III.
ca
growing
crystal.
June, 1878.
496
J.
W. Gihhs
2(o-fJr)^0
(671)
incapable of motion.
At the
surfaces
between
solid
and
fluid masses,
and impossible
cases,
The
simplest case
in others.
is
solid
solid
fluids
tively.
fluids,
respec-
is
a=
(Tbs
condition of equilibrium
and
fluids
/5
it
If the line
by S, the
by a and
is
by
is
- G k^.
at an
is
(672)
edge of the
solid, the
that
Cab cos
^V
Cab cos
/i
^ Cbs O'as,
^ Cas O'bs
((573)
which reduces to the preceding when a-\-fi=z7t. Since the displacement of the line can take place by a purely mechanical process,
this condition is capable of a more satisfactory experimental verification than those conditions
and
Yet the
dissolution.
the line
is
solidification
resistance to a displacement of
in
fluids,
enormously greater.
fluid masses.
is
arbitrary
to a greater extent than in the case of fluid masses, (in which a single
is
uniformly
distri-
make the
If one or
line
such a surface
is
The condition
If the soluble
W. Gibbs
J.
497
adjacent surfaces, since a displacement of the line will involve a displacement of the whole side of the crystal which is terminated at this
line.
may be
General Relations.
of the solid,
we may
write
f?fs(,)=:d^/;s(,)H-/'2^^^^2Ci)
since
relation
tliis
From
involved.
implied
is
in
the
+ A<3f^^'3(i)+
etc.,
(674)
of the quantities
definition
this
which
is
(675)
the state
But
may
limitation
in
(If the
quantity
<J
this
repre-
between
wool and
of
ice
may be mentioned
here.
The
fact that
a fiber of wool which remains in contact with a block of ice under water will become
attached to
it
seems to be
strictly
if
it
water will
generally rise up about the point of contact so as to touch the solid over a surface of
some
extent.
The condition
"S A
where the
like
suffixes
s,
a,
and
+ (TwA ^ Csw
if
we
In
neglect
ice, is
where the
suffixes
p,
w; and
and
to ice, respectively.
See
498
W. Gibbs
./.
nature and state of these masses together with the quantities of the
may be
Problems relating
to processes of
solidification
which appears in general allowable with respect to the surfaces produced by these processes. But in considering the equilibrium of
an unchangeable
solid,
such a limitation
is
treating
the subject will be found more simple and at the same time more
general.
by the excess of energy in the vicinity of the surface over that which
would belong to the solid, if (with the same temperature and state
of strain) it were bounded by a vacuum in place of the fluid, and to
the
fluid, if it
any case
solid, or, if in
by the
some
definite
tJtus defined.
way
(fg)
to
manner
entirely analogous,
and be denoted by (?/). In like manner also, for all the components
of the fluid, and for all foreign fluid substances which may be present
be determined, and denoted
These superficial densities of the fluid components
All matter
relate solely to the matter which is fluid or movable.
which is immovably attached to the solid mass is to be regarded as a
Moreover, let ? be defined by the equation
part of the same.
at the surface, let the su2:)erficial densities
by
{l\), {l\),
t'tc.
?=(fs)-^(A;s)-/<o(ro)~/^3(r3)-etc.
d{e^)
d?=z-
d{t/s)
(7s)
+yM3
(676)
These quantities
(7(i 2)
+ Ms (Kf\) +
etc-,
(Qll)
etc.
(678)
and
(675).
But
by
this
?,
limitation
(fg),
etc.
may
must
generally be neglected.
in general
than those of
(7,
The quantity
It
less
fg^,, etc.
?
be much
may be
Its value
may be
is
of
J.
It will
Wl Glbbs
be observed that for the same solid surface and for the same
499
by a constant,
quantity
this
viz.,
is
(in
all
cases to
the value of
a vacuum.
we may
O-AB
cos
a line on
difiei'cnt fluids at
easily obtain
?BS
(679)
^AS,
etc.,
the solid.
must
It
at the
also
if there are any foreign subA-S and B-S, the potentials for these subhave the same value on both sides of the line or, if
is
that the
line,
less value
on the other
and that the potentials for the components of the mass A, for
example, must have the same values in the surface B-C as in the
mass A, or, if they are not actual components of the surface B-C, a
side
two
fluids in contact
solid,
by
art;
necessary to pre-
The
for a fluid
masses,
dissolution.
in
is
may
easily
be
bounded by or includes
be of the form
y> 6Dv
-I-
(680)
where the first four integrals relate to the fluid masses and the surfaces which divide them, and have the same signification as in
equation (606), the fifth integral relates to the movable solid masses,
and the sixth and seventh to the surfaces between the solids and
fluids, (r) denoting the sum of the quantities (i^g), (^3), etc.
It
should be observed that at the surfiice where a fluid meets a solid
500
J.
dz and
W. Gibhs
6z,
and the
fluid,
rium,
cU
= g{r)dz,
(681)
[compare (614),] which expresses the law governing the distribuon the surface of a solid, when there are no
passive resistances to
By
its
motion.
fluids in contact
matter
in the
is
if
the boundary
between the fluids were in the horizontal plane at which their presIn this theorem, we may either include or
sures would be equal.
exclude the weight of a film of fluid matter adhering to the tube.
is usually applied to the column of fluid in mass
between the horizontal plane for which p':^p" and the actual
boundary between the two fluids. The superficial tensions 5' and s"
The proposition
But
Ave
may
which would
fill
by the difference
plane p'=: p".
If
is
the values of
is infinite,
the value of
at
the top vanishes, and the w^eight of the film of water adhering to the
tube and of the mass of liquid water above the plane p'=.2')" diminished by the weight of vapor which would fill the same space is
equal in numerical value but of opposite sign to the product of the
perimeter of the internal surface of the tube multiplied by ?", the
superficial tension of liquid water in contact with the tube at the
pressure at which the water and
a plane surface.
its
W. Gibhs
J.
501
We
know by
ductors) there
The quantitative
(electrolytic
con-
is
may
relation
between
Dm^
De
Dm^,
H
Dm
+ etc.
Dm,.
etc.,
((,82)
o-i,,
etc.,
rt-g,
Dm^,
regard
o-h,
For,
another.
etc.
JDrn^,,
if
We may
etc.,
Dm^, Dm^,,
so,
etc.,
evidently
as independent of one
Since
tlie
motion of the
fixes
must
~+ +
Yi
These
fluid as a
Vh
densities, therefore,
densities of the
etc.
y
Vh
=^
+^
+
etc.
(683)
in other cases.
account for the relation (682) by supposing that electricity (positive or negative) is inseparably attached to the diflerent
kinds of molecules, so long as they remain in the interior of the fluid,
in such a way that the quantities a^, 0-5, etc. of the substances speci-
We may
fied are
each charged with a unit of positive electricity, and the quanix^, etc. of the substances specified by these suflixes are each
tities a^,
The
cr^,
o-^,
relation (683)
etc.
is
are so small
that the electrical charge of any sensible portion of the fluid varying
sensibly from the law expressed in (683) would be enormously great,
so that the formation of such a
mass would be
resisted
by a very
great force.
It will
same physical
relations
may
is
to
some extent
arbitrary,
and
502
TK Gibhs
J.
form
tions of the
(682), in
we
tution of the fluid, those of which the fluxes appear in the equation of
the form (682) are called the lons^ and the constants of the equation
tution,
ient,
and
call
a set of components as
may be
conven-
Now,
components of
do not necessitate any electrical currents, all the
conditions of equilibrium which relate to the movements of these
components will be the same as if the fluid were incapable of the
electrolytic process. Therefore all the conditions of equilibrium which
since the fluxes of the independently variable
an electrolytic
fluid
But we have
nents.
and
its
still
For
simplicity,
we
shall
is
without internal
may
be slightly aftected
conductors {electrodes)
parts of
in diflTerent
its surface,
being other-
which
at another.
it
If all
by
d^a^ of tlie
component
specified,
energy
in the
in
the
in
which
V"
V)
V, V"
6e
(//,"
IJ,!)
Sm^
(2"- T")
dm,,
two
two
electrodes,
and
The
/y/,
///',
first tei*m
the intrinsic
represents the
W. Gihbs
J.
ment of
tlie intrinsic
But by
(682)
=z a^ 6e
It is
V"
To extend
-V' +
a^ (//."
^^:
cx^ (//,'
T')
V"
a^
{ia:'
a,
by
and
ilar condition,
V-
a^ (/.;
V'"
two
^")
we may
- T")
(//,'"
etc.)
(684)
0-
write
F'")
etc.
T')
= V"
- a,
V'"
may
(/.,"
(685)
When
T"
V'
For
503
form
T")
a^ (Ms"
- r")
(686)
etc.
we have
for
any
cation
V" and
for
a,
{/J.J
a^
(/V
or electrolytic
(687)
V'
- mA
(688)
conditio7is
cell,
of equilibrium,
i7i
Let us apply
i7%
the surfaces
and the
of a
catio7i.
by
a galvanic
of
trode, or
fulfilled
ai-e
the values
(in
/<;'),
where V"
That is:
Whe7i all the
(I.)
any anion
V"
of
V=
itself constitutes
component of an
elec-
resist-
ances to change) will have the same value within the electrode as on
its surface,
and
trode with
its
by a
cell
will be
It is
elec-
When
64
June, 1878.
5C4
zinc)
tT.
W. Gibhs
aud an
is
fluid
and
electrical current,
may
necessary to treat
as a different substance.
if
the process
electrode,
may
be entirely
changed.
it
is
it
would be
But
of electricity
we
(687) or (688)
electrolyte.
as
dissolved in the
move independently
can
somewhat barren
we only
obtain the
may be
of diffusion.
we may regard
such cases,
In
fluid
when
electrodes identical.
is
zero,
equilibrium
of the
in differ-
ent proportions, that which has the greater potential for zinc will have the less potential for
mercury.
if
This
is
W. Gihhs
J.
505
not satisfied.
This
may be
is
the case with hydrogen and oxygen as ions (or apparent ions)
first
When
considered.
the ion
is
a non-conductor, a continuous
of an electrical current.
conductor
is
disengaged
in
by hydrogen appearing
an important one.
it, is
bubbles at a cathode.
in
In case of perfect
may
Yet the
is
to
it is
electrolytic current
is
at the line
ing mass, and the electrolytic fluid meet, so that the electrolytic process
is
But even
may
greatly
And
by the
may
act in the
by the
is
supposed
same way.
When
in the
surface of a
the ion
is
absorbed
comes
under the heads which we have already considered, yet the fact that
the ion is set free in mass is important, since it is in such a mass that
electrode, or
When
the ion
is
it
may
still
by
be
J.W.Gibbs
506
may
take place before the deposit will have the properties of matter
in mass.
is
absorbed by the
electrolytic fluid
of
it.
or not,
The
electrode
if it is
The
effect
o'
(see
which we have
if
the electrode
V V"
pages 482-500)
liquid electrode.
combination
afl^ect
varied,
when
it is
too
if
applied to an electrolytic
weak to produce
a lasting current,
and the electrodes are thereby brought into a new state of polarization, in which they make eqiiilibrium with the altered value of the
electromotive force, without change in the nature of the electrodes or
of the electrolytic fluid, then by (508) or (675)
- r; d^l,
d(j'
do"
and by
- r:
dj.ll'
(687),
,/(
F' - V")
Hence
d{
If
we suppose
trodes
(689)
is aflfected (as
when
its
surface
is
very small
d(y'=^d{V'V").
The
superficial tension of
(690)
is
then a function of
which bears his name.* In applying equaand (690) to dilute sulphuric acid between electrodes of
mercury, as in a Lippmann's electrometer, we may suppose that the
tion of the electrometer
tions (689)
serie,
phenom^nes
t. v,
p.
494.
electriques et capillaires,"
W. Gibbs
J.
hydrogen.
suffix refers to
It will
507
may
last
mann's determinations,
state
(i.
e.,
is
is
iell's cell,
tive force
V"
it is
equation (690).
applied),
since
increases
a'
mum
V"
value,
is
V" V.
with
When
and there
This
is
is
remaining
in its natural
surface.
in its natural
is
it
new
The
when
it
flux of elec-
in
in
its
it
is
extended,
direction,
is
when F^'
represented
is
negative,
by
is
from
We have so far
supposed,
in
resist.
The
with
There
by which an ion
is
For, in any
electrolytic current.
is
fluid.
external
electromotive force
may
often
vary
within wide limits, without creating any current by which the ion
transferred from one of the masses considered to the other.
V V"
may
we determine
I, II,
and
III.
from that
We
when the
is
In other
may, however,
508
J.
W. Gihbs
by an
it
But
cesses.
in a
of an electrode
it
may
surface-densities
in this
which
it
however,
necessary
is
The
No
such distinction,
which
in
it
many
ions.
anion in
that this
may
component of the
electrolytic fluid.
When
If
all
in
changes
which accompany the current can be reversed by reversing the current, the cell may be called a perfect electro-chemical apparatus.
The electromotive
tions Avhich
to
is
subject
may
be conveniently stated
in
subject to
electricity
(2)
The supply
gases.
The
which
is
We
the
equation
dE
= {V'-
V") de
(691)
W. Gibhs
J.
ill
intrinsic
energy of the
cell,
V and
V"
509
it,
of the
dWa
the
apparatus.
The
we suppose
it
is
is
is
equal to
The only
exter-
communicated to the
cell
Eliminating c?,
c?
dii
F") de
we
obtain
F'
-^.
d,i
(692)
^dW^ + d TFp,
(693)
or
It is
if
we give up
no longer supposed
is
But,
if
we
still
we
is
all
di]^ j^,
and instead of
(
is
cell, will
subjected.
de-^
is
ds +
dr/
+ d Wo + d TFp.
member
(696)
of (694), for
cell
generally to be neglected.
it is
If
is
The term
no heat
gravity
(695)
(693), (694)
V" - V)
still
parts of the
But
is
relating to
supplied or with-
in the calculation of
510
'J.
The
W. Gihbs
by
unless
it
in
is
justified
In fact,
force
it is
electrolysis.
by any
sufficient
reason.
dii
is
any
may be
member
in (694), all
the other
This
is
true
In
a process which
temperature.
its
initial
volume to the volume occupied by the two gases together. The same
work is equal, as appears from equations (278), (279) on page 217,
(see also page 220,) to the increase of the entropy of the system
multiplied by the temperature.
It is
gen
at
Let the
way
cell
each pole under very unequal pressures (as of one and two
in the
The
difference of
be balanced by the
diff*erence
acidulated water.
It will
two columns of
* Philosophical Magazine,
W. Glhbs
J.
511
pump the
And
if
by any
by
we could reduce
to the other without finite variation of the electromotive force, the only
j)Ossible
expression
^^
as a
It will
be observed
by maintaining the
Again,
by
it is
is
to
negative.
salt
being of
in
by the diminution of energy in the cell. And when the mixsame rule wouhl make any electromotive force
impossible except in the direction which would tend to increase the
difference of concentration.
Such conclusions as would be quite
irreconcilable with the theory of the phenomena given by Professor
simplj^
Helmholtz.
motive force
is
Since heat
III.
is
Neue
65
Band
iii,
is
dissolved
February, 1878.
June, 1878.
512
in
IV.
tT.
Gihhs
Equilibrium of Heterogeneous
to the mercury,
when
cell is
increased
the temperature
is
by
Siibstayices.
a transfer of zinc
maintained constant.
Yet
mercury.
case
The electromotive
lasually cited as
is
motive force
cell,
i. e.,
are accompanied
certain
is in
energy of the
is
directly.
of M. Gaugain
is
reverse direction
these anomalies can hardly affect the general conclusions with which
alone
we
zinc and an
If
amalgam containing
amount
without losing
mercury,
it is
is
amalgam by
such a
is
dissolved
dii-ection
cell.
by
*
j-
J. Regnaiild,
Compks Eendus,
t. li,
t. xlii,
p. 778.
p.
430.
W. Gibbs
J.
Eqaillbrmni of Heterogoteous
513
iSub{>ta?i,ces.
entropy,
liut
general seem to indicate any possihility of obtaining from the combination of snlistances
cal
work which
is
Ijy
of the substances.
We
any temperature
desired.
effect
elements.
much
if
to
Now
If,
work
C, and
in
This, of course,
ie,
Bd.
ii,
p.
290.
same pressure.
514
W. Gibbs
tT.
that
the
all
is
or
combustion as obtainable)
is
water as considerably
the supposition that
supply
all
The
less
is
it
case
is
essentially the
is
The phenomenon
of dissocia-
is
than half of the gas being dissociated at 1400 C.f And the heat
which is obtained by the combination of hydrochloric acid gas with
water, especially with water which already contains a considerable
is
comparatively low.
(/.
the elec-
e..
less
all
submitted to electrolysis.
ena exhibited
rated
by
in the direct
who
is
corrobo-
was
The
electrolyzed.J
* Unless the received ideas concerning the behavior of gases at high temperatures
are quite erroneous,
it
is
by
other expenditure than that of an amount of heat equal to the difference of energy of
the matter in the two states and supplied at a temperature far below 2500 C.
essential parts of the process
would be
(1)
continuous process
filtration
all
by
is
filtration,
condensed.
and
(3)
The
to a
little
calculation will
show
tlie
that in a
products of
t.
(2)
it
Ixxiii, p. 97.3.
The
figures obtained
same nature.
by M. Favre
1, p.
42
or Comptes Bendus,
W. Gibbs
J.
515
electromotive work expended must therefore have been less than the
increase of energj'^ in the
cell.
we have
compound has more entropy than its elements, and
the difl'erence is by no means inconsiderable. This appears to be the
rule rather than the exception with respect to compounds which have
less energy than their elements.
Yet it would be rash to assert that
it is an invai-iable rule.
And when one substance is substituted for
considered, the
another
tions of
rela-
unit of electricity
a striking correspondence between the elecand the rate of diminution of its energy per
transmitted, the temperature remaining constant.
tromotive force of a
Daniel I's
cell is
is
cell
common
use,
it
all
may
It
cells in
we apply our
If
we have
M. Favre,* estimat-
kilogrammes)
of zinc dissolved
V" - V) Ae
It will
is
24327"''-,
Ae
=-
JQ= -
25394"-''-,
lOoV""'-.
cent, less
cell
cell.f
of
in the cell
been corrected
It
V" V)
Ae, which
is
dif-
and
produced
in a Grove's cell.
sometimes cold
is
produced.^
When
loc. cit., p.
90
neither
is
produced, of course
much
ii,
1117, 1118.
\ Mem. Sa rants
Ixxiii, p.
closer correspondence.
893.
Etrang., loc.
cit.,
p. 9.^
or Comptes Rendus,
etc.,
t.
2''^
Ixix,
Auflage, Bd.
p.
37,
and
t.
516
W. Glbbs
tT.
is
exactly equal to
is
With
observed.
its
dimiinition
(HNOg + mO), M.
Z/
46V81''"-,
= -41824^^'-, J^ = 4957^"'-;
(HNOg+HO),
^ = -2867'^''^
In the
first
example,
will
it
Jf=: 52714
"',
z/
absorbed
^"',
cell.
is
by the diminu-
cell
his
experiments give*
(
-p"_
9256
A TFp = 290'-,
V')
Ae
8258^"'-,
AQz=^
1288^'"-.
Z/6 z=
'-''',
is
For electrodes of
zinc
and platinum
cell
in the
with the
sei'ies
of
experiments givesf
(F" V) Ae=:
A= -
16950'^'-,
AWp = -290'^\
and a
16189
'^^',
J^z=1051'-;
later series,J
(V"- V)
Ae=i
Ae=16738'''';
17702'^'"-,
J^ = -674''''-.
J Wp = -290'-,
he has found
* Comptes Bendus,
(inchidiug the cell)
t.
Ixviii, p.
when
all
V'' V)
is
Ae
AQ, also by
The
Ae + A Wp.
whole
work
is
[See (691).]
Ae by
adding
is
evidently represented
The value
of
A Wp, which
F"
is
Mem. Savants
Ibid,
p. 142.
Etrang.. loc.
cit., p.
145.
circuit
1305.
the electromotive
F'
by
Ae
easily esti-
W. Gihhs
J.
Equllihriuni of Heterogeneous
V V") Ae 34825
Suhstancei^.
AQIW?,
'-
517
"'',
whence
Z/f-JTFp = 36938.
We
rine
zJ
value of
TFp
must
in the
lie
form of gas
between
and
290*"'-
is
Hut the
not stated.
580'-,
probably nearer
to the former.
The great
two
series of
experiments
in
is
the condi-
That which
it is
is
all
these cases
absorbed
in
is
that
a galvanic or
by the diminution
would be calculated from the increase of energy in the cell, especially when the work done against the pressure of the atmosphei'e is
taken into account.
It
in all these
as insignificant in comparison,
made
in
due to
this circumstance,
absorbed
this resistance.
which would
With exception
in all cases
of the error
AQ
* It should perhaps be stated that in his extended memoir published in 1877 in the
Memoires des Savants Strangers, in which he has presumably collected those results
of his experiments
ments.
may have
This
zinc.
t.
showed an evolution
been,
may
Ixviii, p. ISO.'S.
it
of heat.
Whatever
518
Gihbs
VT.
-T.
we
remember
when hydrogen
that
same changes
tus, the
much
less
amount.
In either case,
set free at
is
in
in the cell
A Q would
is
should,
is
by the
part absorbed
electrolytic fluid,
and
and
in pai't as gas,
in
is in
electrode.
which would not sensibly affect the electromotive force, would cause
all of the ion to be disposed of in one of the three ways mentioned, if
the current were sufficiently weak. This would make a considerable
* Except in the case of the Grove's
the absorption of heat
is
cell, in
marked
nfost
The
tlie
electrolysis
The
In addition
was performed
in a cell
with a porous
chlorine and hydrogen dissolved in the liquid from coming in contact with each
other.
It
had appeared
in a previous series
or Comptes Eendus,
t.
Sarants Mrang.,
of experiments {Mem.
a very considerable
amount of
In a cell
form
calories.
If,
A Q.
A.
lytic fluid.
ment
It is
by the
electro-
part of
the chlorine to be evolved as gas, without essentially affecting the electromotive force.
The
it
t.
Ixxvi, p. 1514.)
cell
was greatly
dimini.shed
If
in solution.
Under
certain circumstances,
chloric acid.
It
experiments which
regard
it
oxygen
is
does not appear that this took place to any considerable extent in the
we
are considering.
But so far as
absorption of heat
is
it
may have
occurred,
we may
W. Gibbs
J.
519
alone in
all
The
these cases.
work performed
is
set free as
gas
appear on
It will
is
Nor does
distinctly
it
to be regarded as normal
and which
it
to be
liquid body.
variation of energy,
force
we
make
in
equation (694),
It will
make any
and
\
it is
apparent
being both
we do
when we take
not have to
The only
limitation
muth
t.
Ixviii,
p.
kilogramme or 885
643.)
force
tion.
cell.
But
was manifested
In
fact,
above to about
(TO''''),
in melting 12.64
while a Daniell's
cell
energy of the
bis-
in contact
in
at the
cell, if
moment when
its
state of aggrega-
tin
gave similar
III.
cell.
results.
66
July, 1878.
520
Gihhs
J. iPT
affected
by the same
would be easy
by
In fact,
if
to devise
be
will
sufficient, it is believed, to
its
temperature
its
cell
show
determining the
in
is
pressures or
by
gravity.
we
t //,
df
td i}\
=. de
y_y,^_d^
dW^
dW^
de
de
de
and
for
In a
cell
V"
V) de^ dip +
tZ
TTg
f?
is
maintained uni-
TFp
(698)
work done by
gravity, as
be neglected.
we
C=
where p denotes the pressure
in
if
//
the
and
de
as the temperature
set, as
is
may
main-
on page 147,
+ jo w,
and
cell,
its
total
volume
(in-
we have
dif -\-
p dv,
F'-F'=-f,
or for any
cell,
(
V"
V) de ^
(699)
dl.
(700)
PREr.iMiNARY
REMARK on
thermodynamic systems
the
theory of
108
.-
100
Meaning
110
HI
112
existing,..
116
118
119
120
121
124
--
..-
116
116
116
of con-
129
siderable size,
may
129
but not always necessary, 131
A mass in which this condition is not satisfied, is at least practically unstable, 133
See p. 156).
(Tills condition is farther discussed under the head of Stability.
134
Effect of solidity of any part of the system
137
Effect of additional equations of condition,
equilibrium of osmotic forces,
138
Effect of a diaphragm,
sought,
Condition (53)
is
always
FUNDAMENTAL EQUATIONS.
140
Definition and properties,
Concerning the quantities ip, Xi C -144
Expression of the criterion of equilibrium by means of the quantity i/a
145
Expression of the criterion of equilibrium in certain cases by means of the
quantity
...
C,
147
POTENTIALS.
The value
522
J.
W. Gihhs
In the same homogeneous mass we may distinguish the potentials for an indefinite
number of substances, each of which has a perfectly determined value. Between
the potentials for different substances in the same homogeneous mass the same
equations will subsist as between the units of these substances
149
The values of potentials depend upon the arViitrary constants involved in the definition of the energy and entropy of each elementary substance
151
COEXISTENT PHASES.
Definition of phages
of coexistent phases,
152
Number of the independent variations which are possible in a system of coexistent
phases,
152
Case of w + 1 coexistent phases, ..
153
Cases of a less number of coexistent phases,
155
INTERNAL
STABILITY
OF
BY
FUNDAMENTAL
EQUATIONS.
156
160
162
169
GEOMETRICAL ILLUSTRATIONS.
Surfaces in which the composition of the body represented is constant,
172
Surfaces and curves in which the composition of the body represented is variable
pressure
are
constant,
and its temperature and
:
176
CRITICAL PHASES.
188
Definition,
Number
mass
191
BODIES.
catalytic agent,
perfect
equation for phases of
-.
197
200
201
A fundamental
THE CONDITIONS OF EQUILIBRIUM FOR HETEROGENEOUS MASSES UNDER THE INFLUENCE OF GRAVITY.
The problem is treated by two different methods
203
The elements of volume are regarded as variable,
.- 207
The elements of volume are regarded as fixed,
Dalton's Law,
...
Inferences in regard to potentials in liquids and solids,
Considerations relating to the increase of entropy due to the mixture of gases by
^27
diffusion,
The phases
210
215
225
components which
230
J.
W. Gibhs
523
Page
234
237
245
SOLIDS.
The conditions
375
fluids,
THEORY OF CAPILLARITY.
SURFACES OF DtSCONTINUITY BETWEEN FLUID MASSES.
Preliminary notions. Surfaces of discontinuity.
Dividing surface,
The particular conditions of equilibrium for contiguDiscussion of the problem.
ous masses relating to temperature and the potentials which have already been
obtained are not invalidated l\v the influence of the surface of discontinuity.
Superficial energy and entropy.
Superficial densities of the component subGeneral expression for the variation of the superficial energy. Constances.
dition of equilibrium relating to the pressures in the contiguous masses,
Fundamental equations for surfaces of discontinuity between fluid masses,
Experimental determination of the same,
Fundamental equations for plane surfaces,
geneous
On
380
400
405
homo-
"_
fluid,
380
391
394
395
42g
the possible formation at the surface where two different homogeneous fluids
Impermeable
434
44Q
films,
The conditions
453
of a
new phase
at a
of a
new phase
at a
455
"
films,
454
4gij
^g-j
Each element may generally be regarded as in a state of equilibrium. Properties of an element in such a state and sufficiently thick for its interior to
have the properties of matter in mass. Conditions under which an exten-
524
J.
W. Gihhs
Preliminary notions,
Conditions of equilibrium for isotropic solids,
'
--
Effect of gravity,
which
which three
different
_-
is solid,
482
485
488
489
492
493
General relations,
49'7
49t
ELECTROMOTIVE FORCE.
Modification of the conditions of equilibrium by electromotive force,
Equation of fluxes. Ions. Pllectro-chemical equivalents,
Conditions of equilibrium,
501
501
502
503
Four cases,
Lippmann's electrometer,
Limitations due to passive resistances,
General properties of a perfect electro-chemical apparatus,
Reversibility the test of perfection,
506
507
508
508
Determination of the electromotive force from the changes which take place
Modification of the formula for the case of an imperfect
in the cell.
apparatus,
When the temperature of tlie cell is regarded as constant, it is not allowable
This
to neglect the variation of entropy due to heat absorbed or evolved.
is shown by a Grove's gas battery charged with hydrogen and nitrogen,
by the currents caused by differences in the concentration of the electrolyte,
and by electrodes of zinc and mercury in a solution of sulphate of zinc,
That the same is true when the chemical processes take place by definite
proportions is shown by a priori considerations based on the phenomena
exhibited in the direct combination of tlie elements of water or of hydro,
509
510
51
511
513
chloric acid,
and by the absorption of heat which M. Favre has in many cases observed
516
in a galvanic or electrolytic cell,
different physical states in which the ion is deposited do not affect the
Experiments
value of the electromotive force, if the phases are coexistent.
The
of M. Raoult,
518
520
ERRATA TO VOL.
Page
1,
Page
5,
Page
9,
Page
Page
read capillare.
Paga
read Samythella.
Page
Page
read 1874.
read
verticillata
Page
Page
Page
m,, read
Page
m,,
Page
read
Page
295, note
X, read IX.
z^,.
mn_i, read
/'n-i-
a.it
f,
fj-,,
massen.
Page
Page 356,
last line but two, for crystalline solid, read solid of continuous crystalline
structure.
Page 405,
Page 432,
is
11,
read H,.
a like
series of
JON 9
m\
Q
11
C9
V.3
Pltyiical ft
Applied Sd.
Serials
PLEASE
CARDS OR
DO NOT REMOVE
SLIPS
UNIVERSITY
FROM
THIS
OF TORONTO
LIBRARY
STORAGE
2009
University of Toronto
http://www.archive.org/details/transactions03conn