Thermodynamic Potentials

Why are thermodynamic potentials useful

Consider U=U(T,V)

Complete knowledge of equilibrium properties of a simple thermodynamic

System requires in addition
P=P(T,V) equation of state

U=U(T,V) and P=P(T,V) complete knowledge of equilibrium properties

U(T,V) is not a thermodynamic potential

However

We are going to show: U=U(S,V) complete knowledge of equilibrium properties

U(S,V): thermodynamic potential

The thermodynamic potential U=U(S,V)

Consider first law in differential notation dU  dQ  dW 2nd law

dQ dQ  TdS
inexact differentials expressed by exact differentials
dW dW  PdV

Note: exact refers here to the

coordinate differentials dS and dV.
T dS and PdV are inexact
dU  TdS  PdV as we showed previously.

Legendre Transformations

dU  TdS  PdV dU: differential of the function U=U(S,V)

natural coordinates
 Legendre transformation Special type of coordinate transformation

Partial derivatives of U(S,V) (vector field components)

Example: dU  TdS  PdV

coordinates

Legendre transformation: One (or more) of the natural coordinates becomes

a vector field component
while the
associated coefficient becomes new coordinate.
Click for
graphic example

Back to our example

becomes a coordinate

dU  TdS  PdV
becomes a coefficient in front of dP
 Product
dU  TdS  dPV   VdP rule

easy check:  dPV   VdP  VdP  PdV  VdP  PdV

dU  dPV   TdS  VdP dU  PV   TdS  VdP

=:H (enthalpy)

Enthalpy H=H(S,P) is a thermodynamic potential

dH  TdS  VdP
 Legendre transformation

(T,V): dU  TdS  PdV  d(TS)  SdT  PdV

d(U  TS)  SdT  PdV

from (S,V) to : F Helmholtz free energy

(T,P): dF  SdT  PdV  SdT  d(PV)  VdP

d(F  PV )  SdT  VdP

: G Gibbs free energy

G  F  PV  U  TS  PV  H  TS
equilibrium thermodynamics and potentials

thermodynamics potential complete knowledge of equilibrium properties

Consider Helmholtz free energy F=F(T,V)

Differential reads:
 F   F 
dF  SdT  PdV S   
 T  V
and P   
 V  T
Entropy Equation of state

Response functions from 2nd derivatives

 S    2F 
CV  T   T 2 
 T  V  T  V
 P   V    P     F
2
and  V BT     
 V  T  T  P  T  V TV
 P 
BT  V  V
  2
F
 V  T  2
V  T etc.
 Maxwell relations

dF  SdT  PdV differential of the function F=F(T,V)

dF is an exact differential

 S   2F  2F  P 
      
 V  T VT TV  T  V

 S   P 
   
 V  T  T  V

In general: relations which follow from the exactness of the differentials

of thermodynamic potentials are called Maxwell relations
 Systems in Contact with Reservoirs

Entropy statement of 2nd law: entropy always increased in an

adiabatically isolated system

What can we say about evolution of systems which are not adiabatically isolated

Consider system at constant temperature and pressure

adiabatic wall

 changes from initial state with

G o  U o  TSo  PVo
System to final state with

G f  U f  TSf  PVf
T=const.
G  G f  G 0  U  TS  PV
Heat Reservoir R
remain constant
 U  PV  G
From G  U  TS  PV Entropy change of : S 
T

Aim: Find the total entropy change Stot  S  SR and apply 2nd law

Entropy change SR of the reservoir:

dQ R 1 QR
SR     R
dQ 
L
T T L T
Heat reservoir: T=const.
U  PV  G Q R Q  G Q R
Stot  S  SR    
T T T T
With 1st law:
U  Q  W  Q  PV

Heat QR that, e.g., leaves the reservoir flows into the system  Q = -QR

G
Stot 
T
Entropy statement of 2nd law: S tot  0 for an adiabatically isolated system

G
0 G  0 (T=const, P=const.)
T

Gibbs free energy never increases in a process at fixed pressure

in a system in contact with a heat reservoir.

Gibbs free energy will decrease if it can, since in doing so

it causes the total entropy to increase.

System with V=const. in contact with a heat reservoir

Special case, very important for problems in solid state physics

U  F Q  F
F  U  TS S  
T T
Q = -QR
Q  F Q R F F  0 (T=const, V=const.)
Stot  S  SR   
T T T
 Summary: Thermodynamic potentials for PVT systems

Internal energy Enthalpy Helmholtz free Gibbs free

Potential U(S,V) H(S,P) energy energy
H=U+PV F(T,V) G(T,P)
F=U -TS G=U –TS+PV

differential dU=TdS-PdV dH=TdS+VdP dF=-SdT-PdV dG=-SdT+VdP

Vector field  U   F   G   G 
components T  , P   U  T   H  , V   H  S   F  , P    S   , V   
 S  V  V S  S   P   T   V  T  T  P  P T
P S V

Maxwell  T   P   T   V   S   P   S   V 
                 
relations  V S  S  V  P S  S P  V  T  T  V  P  T  T S

Properties 1st law: Isobaric T=const,V=const T=const,P=const

U  Q  W process
H  Q F  0 G  0
 Open Systems and Chemical Potentials

Open system Particle exchange with the surrounding allowed

Particle reservoir

T=const.
Heat Reservoir R

Thermodynamic potentials depend on variable particle number N

Example: U=U(S,V,N)
 U(2 S,2 V,2 N) = 2 U(S,V,N)


In general: U( S,  V,  N)   U( S, V, N)
(homogeneous function of first order) 

 U  (S)  U  (V)  U  (N)

         U(S, V, N)
  (S)  V,N    (V )  S,N    (N )  S,V 

S V N

holds   and in particular for =1

 U   U   U 
  S  V  N  U(S, V, N)
 S  V, N  V S, N  N S,V
 U   U   U 
  S  V  N  U(S, V, N)
 S  V, N  V S, N  N S,V

keep N constant as in closed systems

 U   U   U 
: 
 
 S  V, N
T    P  
 N S,V
Chemical potential
 V  S , N

U(S, V, N)  TS  PV  N
dU  TdS  PdV  dN

 U   U   U 
dU    dS    dV    dN
 S  V, N  V S, N  N S,V
Intuitive meaning of the chemical potential μ

First law: dU  dQ  dW with dQ  TdS

dU  TdS  dW

mechanical work PdV + work μdN required to change

# of particles by dN

How do the other potentials change when particle exchange is allowed

Helmholtz free energy F=U-TS dU  TdS  PdV  dN

dF  dU  d(TS)  dU  TdS  SdT
dF  SdT  PdV  dN
 F

Gibbs free energy G=U -TS+PV

dG  dF  d(PV)  dF  PdV  VdP

dF  SdT  PdV  dN

dG  SdT  VdP  dN

Properties of μ

 U   F   G 
     
 N S,V  N  T,V  N  T,P

With G  U  TS  PV and U  TS  PV  N
G

N both extensive   (T, P) intensive (independent of N)
 Equilibrium Conditions

Adiabatically isolating
rigid wall

System1: System2:
dQ
T1,P1, 1 T2,P2, 2

From dU  TdS  PdV  dN

dU1 P1 
dS1   dV1  1 dN1
T1 T1 T1
differentials of entropy changes
dU2 P2 
dS2   dV2  2 dN2
T2 T2 T2
Total entropy change S  S1  S2  0

2nd law

In equilibrium dS  dS1  dS2  0

With conservation of

-total internal energy U1  U2  const. dU1  dU2

-total volume V1  V2  const. dV1  dV2

-total # of particles N1  N2  const. dN1  dN2

1 1 P P    
dS     dU1   1  2  dV1   1  2  dN1  0
 T1 T2   T1 T2   T1 T2 
 1 1 P P    
S     dU1   1  2  dV1   1  2  dN1  0  small changes dU1, dV1, dN1
 T1 T2   T1 T2   T1 T2 

0 0 0

Equilibrium conditions

T1 = T2 , P1 = P2 ,1 = 2

Remark: T1 = T2 , P1 = P2 and   (T, P) 1 = 2

1 = 2 no new information for system in a single phase

but

Important information if system separated into several phases