Entropy

Introduction to Entropy: Creation of Exact Differentials from Inexact Ones

PV-w rev
1. đw = –pexdV = –PindV
đwrev/P = –dV (exact differential)  đwrev/P is an exact differential
 
2 2
1 đwrev /P = – 1 dV = V1 – V2 = f(initial & final states)
1/P is an integrating factor for đwrev
PV-w rev
2. đw = –pexdV = –PindV can assume an i.g. without “loss of generality”
i.g.
đwrev/T = –(P/T)dV = –(nR/V)dV = –nRdℓnV (exact differential)
 đwrev/T is an exact differential
1/T is also an integrating factor for đwrev
PV-w i.g.
3. đqrev = dU – đwrev = dU + PindV = nCVdT + (nRT/V)dV
đqrev/T = nCVdT/T + nRdV/V
can assume [CV] without “loss of generality”
[CV]
 2
1 đqrev/T = nCVℓn(T2/T1) + nRℓn(V2/V1) = f(initial & final states)
Also,  dq rev/T  0  đqrev/T is an exact differential
1 1/T is also an integrating factor for đqrev (cont’d) © Prof. Zvi C. Koren 20.07.2010
 Rudolf Julius Emmanuel Clausius
1822 [Koslin, Prussia (now Koszalin, Poland)] – 1888 [Bonn, Germany]
http://www-groups.dcs.st-and.ac.uk/~history/Mathematicians/Clausius.html

The Clausius Entropy Equation

–
dq rev
dS  (note the units)
T
S  entropy (Greek: entropê, meaning “turning towards”)
dq rev [T] 1 q rev
S  
T
 dq rev 
T T
 for any ISOTHERMAL + REVERSIBLE process: ST  q rev
T
Entropy: The Origin of the Word
http://www.es.ucsc.edu/~smf/entropy.html
Quoting from Rudolf Clausius (1880):
I prefer going to the ancient languages for the names of important scientific
quantities, so that they mean the same thing in all living tongues. I propose,
accordingly, to call S the entropy of a body, after the Greek word "transformation". I
have designedly coined the word entropy to be similar to energy, for these two
quantities are so analogous in their physical significance, that an analogy of
denominations seems to me helpful.
[The above quote is found in a book by Leon Cooper: An Introduction to the Meaning and Structure of Physics.]
2 © Prof. Zvi C. Koren 20.07.2010
 Reversible vs. Irreversible Processes – A Reminder
Recall:
A reversible process is synonymous with a (quasi-)equilibrium process.
At each stage, an infinitesimal change can reverse the process.
Examples:
chemical rxn at equilibrium (recall Le Chatelier’s Principle);
physical transformation (fusion, vaporization, sublimation, transition);
infinitely slow expansion/compression.

An irreversible process is synonymous with the following:

Spontaneous, natural, real

3 © Prof. Zvi C. Koren 20.07.2010

 Ssystem vs. Ssurroundings
SURROUNDINGS
• qsurroundings = –qsystem
• Surroundings are so big that any q flowing into or out
does not affect the equilibrium in the surroundings
•  Tsurr does not change
–
dq •  the surroundings are always in equilibrium
dS  rev • No matter how the system process proceeds, whether
T rev (equil) or spont, the surroundings always absorb (or
q release) q from (or to) the system in an equilibrium
SYSTEM (reversible) manner.
[T]
• q for the surroundings is always an equilibrium one:
For rev:
Ssys,T = qsys,rev/T Recall for an isothermal reversible process ΔS  q rev
T
(for system or surroundings): T
For irrev: q surr,equil.  q sy stem,actual q sys,actual
Ssys,T  qsys,irrev/T  ΔS surr,T   ΔS surr,T  
T T T
 ΔH sy s ΔH sys
ΔS surr,T, P  
[P]: ΔH  q P
Ifalso  
T T
4 (continued) © Prof. Zvi C. Koren 20.07.2010
 Entropy of Irreversible Change or qrev vs. qirrev
(U is state fn.)
dUirrev = dUrev
đqirrev + đwirrev = đqrev + đwrev
đqrev – đqirrev = đwirrev – đwrev > 0
But, đwirrev > đwrev
Recall:
wcompression > 0, wrev,comp = wminimum
wexpansion < 0, wrev,exp = wmaximum
 đqrev > đqirrev
dq rev dqirrev
dS  
Clausius Inequality: T T
dS = đq1,rev/T
–
dq irrev: spontaneous, natural, real đq2/T
dS 
T rev: equilibrium đq3/T
5 (continued) © Prof. Zvi C. Koren 20.07.2010
 Entropy for an Isolated System (Universe)
System SURROUNDINGS

dq irrev
1. From Clausius Inequality: dS  But, đqIsol. Sys. = 0
T rev
 dSI.S.  0 I.S. = universe = total
2. Also from Definition of Isolated System:
dSIsol. Sys. = dSsystem + dSsurroundings
= ( đqsys/T) + (–đqsys/T)  0
(See the example of the freezing
Again dSI.S.  0 of supercooled water – later)
2nd Law of Thermodynamics: Entropy of the universe is increasing!!!
1st Law of Thermodynamics: Energy of the universe is constant!!!
Entropy is Nature’s Arrow
6
Problems 1 – Entropy: 1-4. © Prof. Zvi C. Koren 20.07.2010
 The Second Law of Thermodynamics (continued)
Stotal (universe) 0 spontaneous

equilibrium
Spontanoeus = Natural = Irreversible
Equilibrium = Reversible
Criterion for spontaneity:
If Stotal > 0, NOT just Ssys (process), then process is spontaneous!
(Not the most convenient of criteria, but a criterion nevertheless.)
So, to determine if a process would be spontaneous, we must calculate
Stotal (and not just Ssys) according to:
ΔS total  ΔS sy s  ΔS surr
If system process is a chemical rxn, then recall
the equation for Srxn from General Chemistry:
 H rxn
0
Stotal,T,P   S 0

Products
  S 0

Reactants

T
If Stotal > 0, then rxn is spontaneous!
7 © Prof. Zvi C. Koren 20.07.2010
 Some Thoughts About the 2nd Law
2nd Law indicates direction of spontaneous process
1st Law (Conservation of Energy) is not enough.
Is the 1st Law disobeyed if the opposite processes occur?

Thi Tlo ball

surface

Phi Plo

Kelvin’s Statement Q: What do all of these

of the 2nd Law: spontaneous rxns have in
No cyclical engine common?
process is possible in
which the sole result A: Spontaneous changes are
is the absorption of always accompanied by a
heat from a reservoir dispersal of energy into a more
and its complete
conversion into disordered form!
work.
8 © Prof. Zvi C. Koren 20.07.2010
Is the criterion for spontaneity one that requires the energy of a
system to reach a minimum?
Consider the following:
(1) an ideal gas expanding into a vacuum,
(2) Also remember that a decreasing system energy results in an increasing energy in
the surroundings

Disordered Ordered
‘Money’ ‘Money’
Energy
Bank:

Disordered Ordered
‘Money’ ‘Money’

Total Energy Value is Constant, but the Distribution of Energy is Different

9 © Prof. Zvi C. Koren 20.07.2010

 S for Physical Transformations
Physical Processes:
fusion (melting): s  ℓ, Sfus
opposite:solidification
vaporization (boiling): ℓ  v, Svap
opposite:condensation
sublimation: s  g, Ssub
transition: s()  s(), Str
Reversible Phase Change for a Pure Substance (occurs at constant P & T)
T Heating/Cooling Curve

ℓv

sℓ


time (continued)
10 © Prof. Zvi C. Koren 20.07.2010
 REVERSIBLE Physical Process

dq rev
dS  Clausius equation
T
dq rev [T] 1 q T,rev [P] q P,T,rev ΔH phy s
S     dq rev   
T T T T T

ΔH phy s
SREV ,phy s 
T
Ssub > Svap >> Sfus > 0

Trouton’s Rule:
For MANY liquids,
Svap  constant  85 J/Kmol Nike Problem
What do you predict would be the Svap for water relative to Trouton’s rule constant?
(Hint: What is S a measure of?)
11 Problems 1 – Entropy: 30, 31. © Prof. Zvi C. Koren 20.07.2010
 First Law in Terms of Entropy
dU = đq + đw = đqrev + đwrev,PV = TdS – PdV
dU = đq + đw (1st Law in terms of system/surroundings interactions)
dU = TdS – PdV (1st Law in terms of system variables – state functions)
Fundamental Equation of Thermodynamics
for a Closed System (Constant Composition):
Energy-explicit 1st Law: dU = TdS – PdV

Entropy-explicit 1st Law: dS  1 dU  P dV  S = S(U,V)

T T
However, it’s often more convenient to express S=S(T,V) and S=S(T,P)
(see following discussion)
Nike Questions/Exercises: (“PD” = Partial Derivative)
(1) “1/T” is equal to which PD? (2) “P/T” is equal to which PD?
 U   U  (5) Find more PD’s from dU and dS above
(3)    ? (4)   ? based on the condition of exactness.
 S V  V S (6) Note that a number of PD’s can be created
from an exact differential equation.
12 © Prof. Zvi C. Koren 20.07.2010
 S=S(T,V) For any pure substance: dS   S  dT   S  dV  ?
U=U(T,V)  T V  V T
1 P 1  U 
dS  dU  dV  
 U 
 dT  
T  T  V
 P 1
 P
 dV  dV  n CV dT  π T dV  dV 
T T  V T  T T T
 S  CV  S 
dS 
nCV
dT  P  π T dV
1
    ,   
1
P  πT 
T T  T V T  V T T
For an ideal gas: Si.g.,[CV] = nCVℓn(T2/T1) + nRℓn(V2/V1) Nike Problem

S=S(T,P) For any pure substance: dS   S  dT   S  dP  ?

U=H–PV  T  P  P T

dS  dU  dV  dH  PdV  VdP   dV  dH  dP

1 P 1 P 1 V
T T T T T T
1 V 1  H   H   V Proved  H   V1  αT 
dS  dH  dP    dT    dP  dP. Later:  
T T T  T  P  P T  T  P T
H=H(T,P)
nCP  S  C  S 
dS  dT  αVdP     P ,    αV
T  T  P T  P T
For an ideal gas: Si.g.,[CP] = nCPℓn(T2/T1) – nRℓn(P2/P1) Nike Problem
13 Problems 1 – Entropy: 5, 6, 11-23, 28, 29. © Prof. Zvi C. Koren 20.07.2010
 Any irreversible process can be broken down into
two (or more) reversible steps for calculating S
T
SB
2

SA

1
V or P
For example, Ideal gas, [CV,CP]:
S = nCVℓn(T2/T1) + nRℓn(V2/V1) = f(initial & final states)

SA SB
 T  P  γ 1)/γ  
S = nCPℓn(T2/T1) – nRℓn(P2/P1)  n C P n  2  1   Nike
 1  2 
T P 
 γ 1)/γ 
T2  P2 
 For adiabatic + reversible:    (as before) Nike
14 T1  P1  © Prof. Zvi C. Koren 20.07.2010
 Entropy of Mixing Ideal Gases
Mixing process is effectively one of diffusion or expansion.
For isothermal mixing, for each ideal gas i: Si,T = nRℓn(V2i/V1i)  Smixing = Si

Consider a system of two different gases “A” and “B”, both at same P & T
A B
nA, VA nB, VB mixing nA+nB, VA+VB
(P,T) (P,T) irreversible (P,T)
Proof that P of gas mixture = P of each gas
Dalton’s Law of Partial Pressures for Gas Mixture: Pmix = PA + PB
 VA  VB 
n B  
n RT n B RT n A  n B
RT    B  RT   n BRT  Ppure B or A
V
Pmix  PA  PB  A  
in mix Vtot Vtot VA  VB VA  VB VB
proved
P of each lone gas before mixing: (see below)

n A RT n B RT n n V V
=P=  A  B  nA  nB A  nA  nB  nB A  nB
VA VB VA VB VB VB
 VA  VB  Problems 1 – Entropy: 25-27.
 nA  nB  n B  
15  VB  © Prof. Zvi C. Koren 20.07.2010
 P (slowly)
IRREVERSIBLE Physical Process ℓ ℓ
The Freezing of Supercooled Water @ 1 atm: Stotal > 0
S = S1 + S2 + S3 T
H2O (ℓ) H2O (s)
–10 oC 0.01oC
273 263
n 263 (CP,ℓ/T)dT = S1,P S3,P = n 273 (CP,s/T)dT

0 oC
H2O (ℓ) H2O (s)
S2 = Hsolidification/T
Data for H2O: Cℓ  75.3 J/degmol, Cs  36.8 J/degmol,
Hofus,0 C = 6.004 kJ/mol, Hofus,-10 C = 5.619 kJ/mol
Ssystem = S1 + S2 + S3 = –20.54 J/Kmol Nike Problem
(why?)
Question: Why isn’t S = H/T for the main freezing step at –10oC?
From before: Ssurr,-10 C = –Hsys,-10 C/T = 21.37 J/Kmol
 Stotal,-10 C = Ssys + Ssurr > 0, as required by the 2nd Law
Note: Stotal,0 C = Ssys + Ssurr = 0, as required by the 2nd Law
 Though Ssys < 0, process is spontaneous.
16 (see next slide for another example) © Prof. Zvi C. Koren 20.07.2010
 Cooling Off a Hot Object: Stotal > 0

Question:
A steel casting [CP = 0.5 kJ/kg-1K-1] weighing 40 kg and at a temperature of 450oC is
quenched in 150 kg of oil [CP = 2.5 kJ/kg-1K-1] at 25oC. If there are no heat losses, what
is the isobaric change in entropy of (a) the casting, (b) the oil, and (c) both considered
together?

Answer:
First: Heat Balance Equation: qi = 0  For each qP = H = mCPdT 
(mCPT)casting + (mCPT)oil = 0  tfinal = 46.52oC.
Second: For each, SP = mCPln(T2/T1) (conditions?)

 (a) Scasting = –16.33 kJ/K, (b) Soil = 26.13 kJ/K, (c) Stotal = 9.80 kJ/K > 0.

(why?)

 Once again, though Ssys < 0, process is spontaneous.

Problems 1 – Entropy: 32-36. Problems 2 – Entropy & Heat Engines: 1, 2.
17 © Prof. Zvi C. Koren 20.07.2010
 Mathematical Proofs with S
 U   P 
Prove that the internal pressure coefficient πT     T   P
 V T  T  V
 S  CV  S 
From before: dS 
nCV
dT  P  π T dV
1
    ,   
1
P  πT 
T T  T V T  V T T
 U   S   U   S 
So,    C  T  and    π  T   P
 T V  T V  V T  V T
V T

2U  CV     U      S  
 V        T   
VT   T  V  T  V  T  V  T  V  T
2U  π T     U      S    S   P 
 T        T       
TV V  T  V T  V  T  V T  V  V T  T  V
 S   P   U   P 
     and πT     T  P
 V T  T  V  V T  T  V
Prove the following identity:
 H 
μ
Isothermal J-T Coefficient T     V(1  αT) (used before)
 P T
18 Problems 1 – Entropy: 7-10. © Prof. Zvi C. Koren 20.07.2010