entropy Qrev/T?


by aniketp
Tags: entropy, qrev or t
aniketp
aniketp is offline
#1
Jun16-08, 11:04 AM
P: 79
Hey can neone tell me why this is so? I have not got the ans in any book i read...
Thnx 4 replying
Phys.Org News Partner Physics news on Phys.org
Better thermal-imaging lens from waste sulfur
Scientists observe quantum superconductor-metal transition and superconducting glass
New technique detects microscopic diabetes-related eye damage
Mapes
Mapes is offline
#2
Jun16-08, 05:20 PM
Sci Advisor
HW Helper
PF Gold
Mapes's Avatar
P: 2,532
Hi aniketp,

One way of writing the First Law is

[tex]dU=q+w[/tex]

which means that the differential change in energy equals the heat transferred to a system plus the work done on a system. But another way is

[tex]dU=T\,dS-p\,dV[/tex]

where the work is expressed as a generalized force (pressure) times a generalized displacement (change in volume). The heat transfer term is expressed in an analogous way: a generalized force (temperature) times a generalized displacement (change in entropy).

If you haven't heard the terms generalized force and generalized displacement before, they're just ways to classify variables. Generalized forces are intensive, and they drive processes; a change in temperature drives heat flow, and a change in pressure drives mass flow. Generalized displacements, which are extensive, are the "stuff" that is transferred: entropy, volume.

[itex]S=Q_\mathrm{rev}/T[/itex] arises because at constant volume and temperature, and if the process is reversible (no excess entropy generated) we can integrate [itex]q=T\,dS[/itex] to give [itex]Q_\mathrm{rev}=T\Delta S[/itex].

Does this make sense?
aniketp
aniketp is offline
#3
Jun16-08, 11:43 PM
P: 79
Hey mapes,
But isn't dU=TdS-PdV actually derived from dS=dQ/T?

nicksauce
nicksauce is offline
#4
Jun17-08, 12:51 AM
Sci Advisor
HW Helper
nicksauce's Avatar
P: 1,276

entropy Qrev/T?


No...

Entropy is defined as S = k*log(omega)
Temperature is defined as 1/T = dS/dU at constant V
Pressure is given by P = T * dS/dV at constant U

From the previous two relationships it follows that
dU = TdS - PdV, and S = Q/T is a special case.

For an intuitive discussion see, for example, An Introduction to Thermal Physics by Schroeder, sections 2&3.
nicksauce
nicksauce is offline
#5
Jun17-08, 12:58 AM
Sci Advisor
HW Helper
nicksauce's Avatar
P: 1,276
I should mention that these expressions for T and P are derived (in Schroeder) by consideration of equilibrium conditions.
aniketp
aniketp is offline
#6
Jun17-08, 01:03 AM
P: 79
Hi nicksauce, thnx 4 the help. So T & P are actually defined on the basis of entropy....
But in k*log(omega) what do 'k' n 'omega represent?
nicksauce
nicksauce is offline
#7
Jun17-08, 01:21 AM
Sci Advisor
HW Helper
nicksauce's Avatar
P: 1,276
k is the boltzmann constant, and omega is the "multiplicity" of the system, or the number of possible microscopic configurations the system can have.

Simple example:
Say I have a system of 3 "oscillators" (that are quantized), with a total energy of "q" units. Say q = 0, then there is only one possible arrangement (Omega=1). Say q = 1, then there are three possible arrangements (Omega = 2). Say q =2, then there are 6 possible arrangements, ie (2,0,0) three times, and (1,1,0) 3 times. So Omega = 6. Say q = 3, then there are 10 possible arrangements, (3,0,0) 3 times, (2,1,0) 6 times, and (1,1,1) once. So Omega = 10.
aniketp
aniketp is offline
#8
Jun17-08, 01:30 AM
P: 79
Oh ,ok. got it now...thanks once more !
Count Iblis
Count Iblis is offline
#9
Jun17-08, 09:37 AM
P: 2,159
See here for rigorous derivation of dE = T dS - P dV


Register to reply

Related Discussions
Is the Entropy of the Universe Zero?:(Entropy as Entanglement) Quantum Physics 10
the entropy when T=>0 Advanced Physics Homework 2
Entropy (information entropy) Advanced Physics Homework 5
Entropy Introductory Physics Homework 1
thermodynamics and was looking at the second law Classical Physics 7