Thermodynamics question

1. Apr 1, 2008

a9211l

So, I am looking a question about adiabatic expansions and the associated entropy changed.

Why exactly is the change in entropy for a reversible adiabatic expansion/compression equal to zero? The book says its because there is no heat transfer, but for irreversible adiabatic processes, there is also no heat transfer. So no matter what, shouldn't the entropy of the surroundings for an adiabatic process be zero? In terms of the system, since entropy is a state function, why would it matter whether its reversible or irreversible?

2. Apr 1, 2008

Mapes

Hi a9211l, these are great questions. First, the adiabatic requirement just ensures that no entropy enters or leaves the system by way of heating. However, in an irreversible process we create entropy. Or it might be better to say it in reverse: when we create entropy through a mechanism like turbulence, the process is irreversible. The system will never spontaneously go back to its orginal state, because entropy always tends to increase in a closed system.

Thus, one answer to your question is that no entropy enters or leaves the system (work is a transfer of energy without an accompanying transfer of entropy), and no entropy is created. Therefore, the system entropy is constant.

So what would create entropy? Entropy is created whenever a process occurs due to a gradient (in charge, temperature, material concentration, etc.). This is how all real process occur: irreversibly. But we might imagine making the gradient very small to reduce the creation of entropy. This idea taken to its limit gives us the reversible process, an idealization in which we imagine the gradient to be exactly zero.

3. Apr 1, 2008

a9211l

yes, that is very helpful. it made sense to me theoretically (idk if thats the right term...), but just needed that physical explanation.

4. Apr 1, 2008

a9211l

A second question then, if I do a constant pressure heating of some non-adiabatic system, that still changes the entropy in the surroundings, correct?

5. Apr 2, 2008

Mapes

Yes, heating a system increases the entropy of the system, reduces the entropy of the surroundings, and creates entropy at the interface due to the temperature gradient.