1

I realised that I have a fundamental confusion in understanding Joule's experiment. When we say the paddles' motion causes a rise in temperature of the fluid, what do we mean-

(1) the paddles' motion imparts a greater momentum to the fluid molecules, which increases their energy and this reflects as temperature rise, or

(2) the molecules closest to the paddle heat up more than those farther away, due to friction, and this creates a temperature difference between the layers of fluid, transferring heat and thus the temperature increases

Now, which of these is the correct? I think the answer to this lies in doing the same experiment with a non viscous fluid. Can someone please explain what happens in this case? Or are the possible explanations I gave for heating both wrong?

valerio
  • 16,231
GRrocks
  • 2,748
  • Very much related because these answers explain viscosity on a molecular level. https://physics.stackexchange.com/questions/129676/how-is-viscosity-described-on-the-molecular-level?answertab=votes#tab-top – Farcher Feb 12 '18 at 11:45
  • It cannot be just $(1)$ because you have collisions with the back of the paddles where there is a decrease in momentum. – Farcher Feb 12 '18 at 11:46

1 Answers1

3

Both of your mechanisms are correct. There is no such thing as an inviscid fluid; even ideal gases are viscous. So, even the smallest amount of viscous behavior is sufficient to convert coordinated kinetic energy (non-random) to random kinetic energy.

The rate of deformation of the fluid (which gives rise to viscous heat generation) is largest near the paddles and, to a lesser extent at the walls of the tank. So this is the region where, during the mixing, the temperature rise is greatest. But, the heat generated in these regions is transferred to other regions of the container by convection and conduction. Also, the fluid in these regions is replaced by fresh fluid as a result of the circulation.

Chet Miller
  • 33,586
  • 3
  • 21
  • 48
  • Good answer. Being pedantic, there technically are inviscid fluids, superfluids, but from a practical perspective, they aren't really common enough that it was worth the focus. Just for reference in case anyone reads this and then thinks "superfluids are impossible". – JMac Feb 12 '18 at 18:47
  • Thanks a lot @ChesterMiller! So basically it would be wrong to segregate the source of temperature rise into work and heat conduction; both are in operation. Is that correct? – GRrocks Feb 17 '18 at 09:09
  • @ChesterMiller also is it wrong to say that "we can heat up a body using friction", because in the strict thermodynamic sense we are only doing work on it and increasing its internal energy(hence temperature, which we loosely call heat here?), or does a similar reasoning apply and we can say that the temperature gradient established along the volume of the body causes a flow of 'heat'.? – GRrocks Feb 17 '18 at 10:12
  • Or is there a difference between heat generated by friction and the heat that flows because of a temperature gradient? This seems plausible because the former can occur irrespective of the body temperature – GRrocks Feb 17 '18 at 10:17
  • In my judgment, you are correct that, strictly speaking, when we say "viscous heat generation," what we really mean is internal energy increase (i.e., conversion of work to internal energy). However, dry friction is a little different. Since this occurs at the interface between two bodies (and the interface has no mass), it more properly represents a conversion work to heat (which then is conducted into the bodies). – Chet Miller Feb 17 '18 at 12:46