## Homework Statement:

A horizontal cylindrical container is divided into two equal chambers by an airtight movable partition. The walls of the cylinder and the partition are adiabatic. Each chamber is filled with 1 mol of a monoatomic ideal gas. The gas in the left chamber is hotter, the one in the right chamber is colder.
The partition moves until the pressure in the two chambers is the same. Find the final temperatures in the two chambers.

## Relevant Equations:

Equation of state for an ideal gas. Relations between p, V and T for an adiabatic process. Internal energy for a monoatomic ideal gas. First law of thermodynamics.
The actual data for the problem and my (and my friend's) attempt at a solution are in the attached file.
In a nutshell, this is what happened.

I obtained a solution based on the fact that the system is isolated. Thus the initially hot gas moves the partition doing work onto the initially cold gas, but the total amount of internal energy stays the same. I thought I used the conservation of the total volume, but somehow this does not appear to be the case.

A friend of mine obtained a different solution explicitly requiring that the total volume of the cylindrical container should be the same for the initial and final states of the systems.

My solution has the problem that the final volume is not the same as the initial one.
My friend's solution has the problem that the total internal energy of the final state is not the same as in the initial state.

So, probably both of us are making a (different) mistake.
Possibly, both of us are introducing different assumptions on how the process comes about (although we both initially assumed that the system is isolated and that the volume does not vary).

#### Attachments

• 107.1 KB Views: 52

Related Introductory Physics Homework Help News on Phys.org
Chestermiller
Mentor
Homework Statement:: A horizontal cylindrical container is divided into two equal chambers by an airtight movable partition. The walls of the cylinder and the partition are adiabatic. Each chamber is filled with 1 mol of a monoatomic ideal gas. The gas in the left chamber is hotter, the one in the right chamber is colder.
The partition moves until the pressure in the two chambers is the same. Find the final temperatures in the two chambers.
Homework Equations:: Equation of state for an ideal gas. Relations between p, V and T for an adiabatic process. Internal energy for a monoatomic ideal gas. First law of thermodynamics.

The actual data for the problem and my (and my friend's) attempt at a solution are in the attached file.
In a nutshell, this is what happened.

I obtained a solution based on the fact that the system is isolated. Thus the initially hot gas moves the partition doing work onto the initially cold gas, but the total amount of internal energy stays the same. I thought I used the conservation of the total volume, but somehow this does not appear to be the case.

A friend of mine obtained a different solution explicitly requiring that the total volume of the cylindrical container should be the same for the initial and final states of the systems.

My solution has the problem that the final volume is not the same as the initial one.
My friend's solution has the problem that the total internal energy of the final state is not the same as in the initial state.

So, probably both of us are making a (different) mistake.
Possibly, both of us are introducing different assumptions on how the process comes about (although we both initially assumed that the system is isolated and that the volume does not vary).

Your mistake was assuming that the equations for an adiabatic reversible process (involving $\gamma$) apply to this problem. They do not. The process involved here is highly irreversible. For the details on how to solve this exact problem for the case of an insulated partition, see this thread: https://www.physicsforums.com/threads/irreversible-expansion-of-gas-against-gas.981339/#post-6270858. If the partition is not insulated, you can find the final temperature easily by requiring that the internal energy does not change. This then tells you the final pressure and volumes.

Hi Chestermiller,
thanks a lot for your help!

I took a look at the post you linked. Funny that someone had exactly the same problem as me, one day before... I do not think that the OP in that thread was my friend. Of course I searched the forums, but must have used the wrong keywords, because I did not find this thread. It would have saved me a lot of time.

Anyway, there is something I do not understand in your solution, and I have a lot of questions.

Your key assumption in this post seems to be that the work involved in the transformation is the same as in a isobaric process at the final pressure.

What is the reason for this assumption, other than it actually seems to provide reasonable and self-consistent results? The process is not isobaric. Pressure varies in both containers.

Are there different reasonable assumptions that produce a (possibly different) reasonable result?

Also, how do I know that the process is not reversible, other than from the lack of consistency of the results I get by using the equation for reversible adiabatic processes?
Perhaps because if I move the partition back to its original position the gases can be in different states depending on how I acted on the partition?

Chestermiller
Mentor
Hi Chestermiller,
thanks a lot for your help!

I took a look at the post you linked. Funny that someone had exactly the same problem as me, one day before... I do not think that the OP in that thread was my friend. Of course I searched the forums, but must have used the wrong keywords, because I did not find this thread. It would have saved me a lot of time.

Anyway, there is something I do not understand in your solution, and I have a lot of questions.

Your key assumption in this post seems to be that the work involved in the transformation is the same as in a isobaric process at the final pressure.

What is the reason for this assumption, other than it actually seems to provide reasonable and self-consistent results? The process is not isobaric. Pressure varies in both containers.
The process is not isobaric, but this does not rule out the possibility that the force per unit area on both sides of the partition changes to the final pressure virtually immediately after the partition is released, and stays at that value on both sides of the partition throughout the expansion/compression. This is what happens in an analogous situation where a single gas is compressed in a cylinder with an external atmosphere at a constant pressure and the massless frictionless piston is released at time zero.

Are there different reasonable assumptions that produce a (possibly different) reasonable result?
Not in my judgment. This is one of those situations in which the problem cannot be solved accurately solely using thermodynamics. Without this assumption, we would have to solve the full compressible viscous fluid Navier Stokes partial differential equations as a function of time and spatial position within the chambers in conjunction with the continuity (local mass balance) equation and the thermal energy balance differential equation.
Also, how do I know that the process is not reversible, other than from the lack of consistency of the results I get by using the equation for reversible adiabatic processes?
Perhaps because if I move the partition back to its original position the gases can be in different states depending on how I acted on the partition?
The process that takes place is spontaneous and rapid. To make it reversible, you would have to manually control the motion of the partition so that the deformation takes place quasi-statically. This would involve your doing work of the combined system which is not a feature of what is happening physically.

Hi, thanks again for your help
The process is not isobaric, but this does not rule out the possibility that the force per unit area on both sides of the partition changes to the final pressure virtually immediately after the partition is released, and stays at that value on both sides of the partition throughout the expansion/compression.

But why should the partition move (or, for that matter, stop moving) if the pressure (and hence the force) on both its sides is the same?
I find this hard to understand.

Perhaps this has to do with the fact that during (this) irreversible process the pressure and temperature are not really well defined?

The process that takes place is spontaneous and rapid.
But how is one (especially one high school student) supposed to know this? It should be at least stated in the text of the problem, right?

I found this problem in two high-school textbooks, and in both cases the solution is the one I derived in section 3 of the document I uploaded. I looked further into that just because a more mathematically inclined friend worked out a different solution that seemed equally reasonable at first (but we found problems for both of our solutions).

To make it reversible, you would have to manually control the motion of the partition so that the deformation takes place quasi-statically. This would involve your doing work of the combined system which is not a feature of what is happening physically.
I sort of understand this, although not fully.

Without this assumption, we would have to solve the full compressible viscous fluid Navier Stokes partial differential equations as a function of time and spatial position within the chambers in conjunction with the continuity (local mass balance) equation and the thermal energy balance differential equation.
This is interesting, I wish I had the time, skill-set and computational resources to try that.
Do you know whether such an analysis has been performed?
Would it make sense to track the evolution of the pressure and temperature while the process is happening?
Would the final state be the same as the one obtained via the "reasonable assumption"?

I mean, if this is the case I would agree that the assumption is actually reasonable.
Still, I won't include this problem in a high-school textbook (or, at least, not in its present form).

Andrew Mason
Homework Helper
Would it make sense to track the evolution of the pressure and temperature while the process is happening?
Would the final state be the same as the one obtained via the "reasonable assumption"?

I mean, if this is the case I would agree that the assumption is actually reasonable.
Still, I won't include this problem in a high-school textbook (or, at least, not in its present form).
This is just a straight forward first law problem. Since there is no work done on the surroundings, and no heat flow into or out of the gases, the total internal energy of the gases does not change. The process details are irrelevant.

AM

Chestermiller
Mentor
Hi, thanks again for your help

But why should the partition move (or, for that matter, stop moving) if the pressure (and hence the force) on both its sides is the same?
I find this hard to understand.
The reason you are struggling with this is because you don't have the full picture of what is happening. So let me fill you in:

In a rapid irreversible expansion or compression, the behavior of an ideal gas is not described by the ideal gas law (even though it is still called an ideal gas). The pressure of the gas (or, more precisely, the normal stress at the piston face) is not equal to the RT/v, where v is the molar volume. In addition to the pressure being a function of temperature and molar volume, it is also a function of the rate at which the molar volume is changing dv/dt. A better approximation to the behavior would be $$P=\frac{RT}{v}-\eta\frac{1}{v}\frac{dv}{dt}$$ where $\eta$ is proportional to the viscosity of the gas. So, on the originally higher pressure side of the partition, the volume is increasing with time, and the pressure is less than you would calculate from the ideal gas law (for that compartment), and on the originally lower pressure side of the partition, the volume is decreasing with time, and the pressure is higher than you would calculate from the ideal gas law (for that compartment). It is the rate of change of volume on the two sides of the partition that enables the pressures on the two sides to be virtually equal. Since the partition is massless and frictionless, even the slightest differential between the pressures on the two sides will cause the partition to move at a finite rate. According to the law I wrote down, the piston must be moving (and the volumes must be changing) in order for the pressures to be essentially equal.
But how is one (especially one high school student) supposed to know this? It should be at least stated in the text of the problem, right?

I found this problem in two high-school textbooks, and in both cases the solution is the one I derived in section 3 of the document I uploaded. I looked further into that just because a more mathematically inclined friend worked out a different solution that seemed equally reasonable at first (but we found problems for both of our solutions).
I can't account for why a problem like this would be presented to high school students. The only thing I can conclude is that the person to posed the problem did not himself understand the irreversible nature of this process, and what it entailed. The same goes for the person who derived the solution in the textbooks.

I sort of understand this, although not fully.
What I am saying is that the solution you derived would be correct if supplementary external work were being applied to the partition. However, in that case, the internal energy of the system would have to change.

This is interesting, I wish I had the time, skill-set and computational resources to try that.
Do you know whether such an analysis has been performed?
Would it make sense to track the evolution of the pressure and temperature while the process is happening?
Would the final state be the same as the one obtained via the "reasonable assumption"?
Calculations of this type can routinely be applied by engineers using Computational Fluid Dynamics (CFD). This usually involves commercially available computer software that automatically solves the complicated partial differential equations once the input parameters and computational grid is specified. If this particular problem were solved using CFD, the results for the final state would not be exactly the same as the "reasonable assumption," but my experience tells me it would be pretty close.
I mean, if this is the case I would agree that the assumption is actually reasonable.
Still, I won't include this problem in a high-school textbook (or, at least, not in its present form).
I would never dream of including a problem like this in a high-school textbook. However, I would consider including a version in which the partition movement is controlled quasi-statically by an external agent. Of course, in that case, part of the problem solution would be to determine the change in internal energy of the system (equal to the work done by the external agent).

Thanks Chestermiller,
I would never dream of including a problem like this in a high-school textbook. However, I would consider including a version in which the partition movement is controlled quasi-statically by an external agent. Of course, in that case, part of the problem solution would be to determine the change in internal energy of the system (equal to the work done by the external agent).
You actually anticipated a further question I'd like to ask, if you are available.

The wording of the textbook problem might actually leave room for that. It actually says "the partition is moved quasi-statically".
My friend and I supposed that all the job was done by the pressure differential alone. One reason for doing so was that we thought we needed to know how the partition was moved to answer the question.

In light of your clarifications I would say that the solution should be the one in section 4 of the document I uploaded (that is my friend's "second attempt").
I guess that the system loses energy because the external agent is countering the rapid expansion of the gas, and hence the latter has to perform extra work that lowers its internal energy.

My first attempt, i.e. the solution provided in the textbook, seems to be kind of hopeless. While the internal energy is conserved, the system has a final volume that is less than the initial one, which makes no sense.

Do you agree with these two comments?

Thanks again for the really precious insight you gave me.

Chestermiller
Mentor
I agree with your two comments. The sum of the two volumes will not change while the partition does external work so that the internal energy of the system decreases. Nice reasoning.

Yes. Also, it just occurred to me that in the "first attempt" it makes little sense requiring that the internal energy be conserved.
The fact that the partition is "controlled" by an external agent (as stated in the problem) very likely means that the internal energy is going to change.
This is what confused me in the text of the problem. I checked that my (initial) solution was the same as the one provided by the book. Then I concluded that all the work should be ascribed to the pressure differential alone, and the external agent had no role.

Now I definitely agree also with your comment
The only thing I can conclude is that the person to posed the problem did not himself understand the irreversible nature of this process, and what it entailed. The same goes for the person who derived the solution in the textbooks.
A partial excuse for the latter is that he/she probably was very young and underpaid :D
Thanks again!

Andrew Mason
Homework Helper
Yes. Also, it just occurred to me that in the "first attempt" it makes little sense requiring that the internal energy be conserved.
The fact that the partition is "controlled" by an external agent (as stated in the problem) very likely means that the internal energy is going to change.
The internal energy cannot change because there is no external work done on or by the system and no heatflow into/out of the system. It does not matter how fast or slow the partition moves.

So the decrease in internal energy of one side has to equal the increase of internal energy of the other side. Since they are monatomic gases of equal measure (1 mole), this means that the temperature decrease of the high temperature side will be equal to the temperature increase of the low temperature side. Since the total volume is fixed, the increase in volume of the high temperature side has to equal the decrease in volume of the low temperature side.

AM

Work was done (negatively) by an external agent to keep the partition from accelerating. I believe the solution arrived at is correct.

Chestermiller
Mentor
The internal energy cannot change because there is no external work done on or by the system and no heatflow into/out of the system. It does not matter how fast or slow the partition moves.

So the decrease in internal energy of one side has to equal the increase of internal energy of the other side. Since they are monatomic gases of equal measure (1 mole), this means that the temperature decrease of the high temperature side will be equal to the temperature increase of the low temperature side. Since the total volume is fixed, the increase in volume of the high temperature side has to equal the decrease in volume of the low temperature side.

AM
Andrew,

In order to cause the change to be quasi-static, you are going to need to apply an external force to the partition, say by passing a rod through the wall of the enclosure. Otherwise, if you allow the partition to move on its own, the change will not be quasi-static.

Last edited:
Andrew Mason
Homework Helper
It doesn't matter whether the change is quasi-static or not. It is still an unbalanced pressure so the work done by the left side on the right side is determined by the pressure of the gas in the right side (we will say the initial high pressure side is on the left). It doesn't matter whether the system is in equilibrium while the process occurs.

Let's suppose that there are a thousand little notches and the partition moved in 1000 steps (somebody pulls a pin out to move to the next notch). No work is done by stopping the partition (we are ignoring the kinetic energy of the partition itself as there is no mention of its mass, so assume its mass is negligible i.e. the notches do not do any work in stopping the partition). All the work is being done within the system. There is no change in total internal energy. The work done is determined entirely by the pressure of the gas in the right side whether it happens quickly or very slowly.

AM

Chestermiller
Mentor
It doesn't matter whether the change is quasi-static or not. It is still an unbalanced pressure so the work done by the left side on the right side is determined by the pressure of the gas in the right side (we will say the initial high pressure side is on the left). It doesn't matter whether the system is in equilibrium while the process occurs.

Let's suppose that there are a thousand little notches and the partition moved in 1000 steps (somebody pulls a pin out to move to the next notch). No work is done by stopping the partition (we are ignoring the kinetic energy of the partition itself as there is no mention of its mass, so assume its mass is negligible i.e. the notches do not do any work in stopping the partition). All the work is being done within the system. There is no change in total internal energy. The work done is determined entirely by the pressure of the gas in the right side whether it happens quickly or very slowly.

AM
In my judgment, for whatever it is worth, the process you describe with 1000 pins being pulled would be every bit as dissipative (during each pin interval) as the process of simply releasing the partition at time zero and allowing nature to take its course. So again, even for this process, in my judgment, the $pv^{\gamma}$ equations for a reversible process would not apply.

Andrew Mason
Homework Helper
The adiabatic condition would apply to the right side because the work done in each step is against the internal pressure of the compressing gas. Since the work done on the right side is by the gas in the left side and there is no external work or heatflow, the $\Delta Us$ are equal and opposite.

A similar effect would occur with an adiabatic free expansion. If there was no gas in the right half, it would not matter if the partition was allowed to move all at once or in small increments. No work would be done so there would be no temperature change. If you add some gas to the right side the internal energy, hence temperature of the left side would decrease by exactly the amount that the internal energy of the right side increased. It is just first law: conservation of energy.

If the partition is allowed to move freely, it would oscillate back and forth until everything settled down in reaching equilibrium. A similar thing happens with a sound wave in air. Each compression or rarefaction can be closely approximated by reversible adiabatic processes.

AM

Last edited:
Chestermiller
Mentor
The adiabatic condition would apply to the right side because the work done in each step is against the internal pressure of the compressing gas. Since the work done on the right side is by the gas in the left side and there is no external work or heatflow, the $\Delta Us$ are equal and opposite.

A similar effect would occur with an adiabatic free expansion. If there was no gas in the right half, it would not matter if the partition was allowed to move all at once or in small increments. No work would be done so there would be no temperature change. If you add some gas to the right side the internal energy, hence temperature of the left side would decrease by exactly the amount that the internal energy of the right side increased. It is just first law: conservation of energy.

If the partition is allowed to move freely, it would oscillate back and forth until everything settled down in reaching equilibrium. A similar thing happens with a sound wave in air. Each compression or rarefaction can be closely approximated by reversible adiabatic processes.

AM
I'm not arguing that if the partition is released and the system is allowed to re- equilibrate, the change in internal energy will be zero. What I am saying is that this process will not be even close to reversible, to the extent that the equations for a reversible process can not be used. And I am also arguing that, if the pins-and-notches version that you suggested will also not be even close to reversible.

Last edited:
Andrew Mason
Homework Helper
I'm not arguing that if the partition is released and the system is allowed tore- equilibrate, the change in internal energy will be zero. What I am saying is that this process will not be even close to reversible, to the extent that the equations for a reversible process can not be used. And I am also arguing that, if the pins-and-notches version that you suggested will also not be even close to reversible.
The left side expansion is definitely not reversible. But the right side compression would be very close to reversible because the right side would be close to equilibrium during the compression if it occurred where the speed of the partition was much less than the average speed of the gas molecules (so that the right side gas responds more rapidly than the applied change initiated by the moving partition.

AM

Last edited:
Chestermiller
Mentor
The left side expansion is definitely not reversible. But the right side compression would be very close to reversible because the right side would be close to equilibrium during the compression if it occurred where the speed of the partition was much less than the average speed of the gas molecules.

AM
I guess we are just going to have to agree to disagree.

Andrew Mason
Homework Helper
I guess we are just going to have to agree to disagree.
All I am saying is that the work done on the right side is $-\int P_{int}dV = \int C_vdT$ . What do you say it would be?

AM

Chestermiller
Mentor
All I am saying is that the work done on the right side is $-\int P_{int}dV = \int C_vdT$ . What do you say it would be?

AM
This comes down to just a judgment call. The approximation that I thought would probably be pretty accurate would be the one I suggested using in post #9 of this thread: https://www.physicsforums.com/threads/irreversible-expansion-of-gas-against-gas.981339/#post-6270858. Based on this approximation, there would be entropy increase in both chambers. As I said, it really comes down to a judgment call. At least this approach would be guaranteed to be consistent with the exact final pressure.

Last edited:
Andrew Mason
Homework Helper
All one has to do is assume that the right side reacts very quickly to the moving adiabatic partition so that $P_{int}V=RT$ and $\Delta (P_{int}V) = R\Delta T$ applies, to a reasonable approximation, on the right side during compression. This means that the work done on the right side will be $W_R = \Delta U_R = -\int P_{R_{int}}dV = C_v \Delta T$. This relation does not apply on the left side because the work done BY the left side is $W_L = -\Delta U_L \ne \int P_{L_{int}}dV$

As a result, the change in entropy of the left side is positive (there is no reversible adiabatic process by which it can be returned to its original state) but the change in entropy of the right side is very close to 0.

AM

Last edited:
Chestermiller
Mentor
All one has to do is assume that the right side reacts very quickly to the moving adiabatic partition so that $P_{int}V=RT$ and $\Delta (P_{int}V) = C_v\Delta T$ applies, to a reasonable approximation, on the right side during compression. This means that the work done on the right side will be $W_R = \Delta U_R = -\int P_{R_{int}}dV = C_v \Delta T$. This relation does not apply on the left side because the work done BY the left side is $W_L = -\Delta U_L \ne \int P_{L_{int}}dV$

As a result, the change in entropy of the left side is positive (there is no reversible adiabatic process by which it can be returned to its original state) but the change in entropy of the right side is very close to 0.

AM
Are you saying that, if you have a massless frictionless partition, the work done by the gas on the right side is not equal to minus the work done by the gas on the left side and that the magnitude of the force exerted by the gas on the right side of the partition is not equal to the magnitude of the force exerted by the gas on the left side of the partition.

If the two works are equal in magnitude, as your equations seem to imply, then the condition of constancy of the total internal energy requires that, for an ideal gas, the final pressure is the volume weighted average of the two initial pressures. Does your analysis automatically predict this?

Let's compare results. What does your analysis predict for the example in the link I provided for:

1. The final temperatures
2. The final pressure
3. The final volumes
4. The final changes in entropy

Last edited:
Andrew Mason
Homework Helper
Are you saying that, if you have a massless frictionless partition, the work done by the gas on the right side is not equal to minus the work done by the gas on the left side and that the magnitude of the force exerted by the gas on the right side of the partition is not equal to the magnitude of the force exerted by the gas on the left side of the partition.
The work done by the gas by the left is exactly equal to the (negative) work done by the right side.

As far as the forces are concerned, the forces on the partition are not equal if the partition has mass, which of course it would have. If the gases on each side were in equilibrium during the process, the difference in forces (the respective pressures x area of the partition) would be equal to the mass x acceleration of the partition. The kinetic energy acquired by the partition would be imparted back to the gases when it settled down.

If the partition mass is negligible, the partition will move with a speed that results in the pressures on both immediate sides of the partition being equal during the process, but with neither side in equilibrium. In other words, there is a pressure gradient in the gases both sides of the partition. [* In the extreme case of a free expansion, the pressure on both sides would be zero.] But, at the end of the process when both sides return to equilibrium, the right side will be compressed and the left side will have expanded and the loss of internal energy by the left side will be equal to the gain in internal energy of the right side.

The actual pressure in the immediate vicinity of the partition is really not material if the reaction time of the gases is rapid (which for a small cyclinder at 300K it would be: the average speed of helium atoms at 300K is 1370 m/sec after all) because the rest of the gas - i.e. all but the atoms in the immediate vicinity of the partition - will be at pressure P = RT/V, the internal pressure of the gas ($P_{int}$). So the work on the right side is going to be very close to $\int P_{int}dV$

If the two works are equal in magnitude, as your equations seem to imply, then the condition of constancy of the total internal energy requires that, for an ideal gas, the final pressure is the volume weighted average of the two initial pressures. Does your analysis automatically predict this?

Let's compare results. What does your analysis predict for the example in the link I provided for:

1. The final temperatures
2. The final pressure
3. The final volumes
4. The final changes in entropy
I'll have to work it out.

AM

Chestermiller
Mentor
The work done by the gas by the left is exactly equal to the (negative) work done by the right side.

As far as the forces are concerned, the forces on the partition are not equal if the partition has mass, which of course it would have. If the gases on each side were in equilibrium during the process, the difference in forces (the respective pressures x area of the partition) would be equal to the mass x acceleration of the partition. The kinetic energy acquired by the partition would be imparted back to the gases when it settled down.

If the partition mass is negligible, the partition will move with a speed that results in the pressures on both immediate sides of the partition being equal during the process, but with neither side in equilibrium. In other words, there is a pressure gradient in the gases both sides of the partition. [* In the extreme case of a free expansion, the pressure on both sides would be zero.] But, at the end of the process when both sides return to equilibrium, the right side will be compressed and the left side will have expanded and the loss of internal energy by the left side will be equal to the gain in internal energy of the right side.
I agree with most this if we replace in your wording the "pressures on both sides" with the "normal stresses on both sides" which include the viscous contribution to the stresses. This all is consistent with the approximation I described as well.
The actual pressure in the immediate vicinity of the partition is really not material if the reaction time of the gases is rapid (which for a small cyclinder at 300K it would be: the average speed of helium atoms at 300K is 1370 m/sec after all) because the rest of the gas - i.e. all but the atoms in the immediate vicinity of the partition - will be at pressure P = RT/V, the internal pressure of the gas ($P_{int}$). So the work on the right side is going to be very close to $\int P_{int}dV$
This is the part that I'm not convinced about. The argument would be more convincing to me if it addressed quantitatively the reason that the viscous dissipation on the right side of the partition would be negligible compared to the viscous dissipation on the left side. In any event, I don't think it is going to matter much when we look at the actual final temperatures and volumes predicted by the two approximations. I still feel that, at this point, it is a judgment call.

I'll have to work it out.

AM
I've worked it out already, and the final predicted temperatures and volumes from the two approaches are relatively close to one another. I'm anxious to see the numbers you come up with. I think it will also be interesting to look at the predicted entropy changes for the two compartments.