Thread: Does mercury conduct or convect heat?

I'm just wondering because it's a metal but in liquid form. Would it conduct or convect heat?

Barry

i thought you had your answer in this thread

you could have asked admin to move it for you

yes it would conduct in liquid or solid form and convect in gas form like any other substance

Originally Posted by Red

yes it would conduct in liquid or solid form and convect in gas form like any other substance

Doesn't convection occur in liquids too?

Convection is the primary mopde of heat transfer in liquids. My recollection is that conduction occurs initally and if the temperature gradient is sufficient to overcome the liquid viscosity then convection will set in. (And it won't, typically, take much of a gradient. However, mercury is presumably a good conductor (as a metal) and its density makes it more likely to resist convection to a degree (no pun intended).

So, is it convection or conduction that gives mercury the property of being highly sensitive to temperature changes? (used in thermometers)

Interesting question. You might think that the definitions of conductive and convective heat transfer are somewhat blurry with a very dense liquid like mercury around room temperature. Both forms of heat transfer involve molecular motion. Heat is really just molecular kinetic energy transported by moving molecules and molecular interaction. Conduction is usually associated with relatively small random molecular motion (not in any particular direction), while convection is associated with motion of fluid particles (lumps of many molecules) on microscopic to macroscopic scales. In essence, convection is a term applied in "continuum" mechanics, where terms like density and pressure make sense, while the motions responsible for conduction are really small scale.

A strict definition could regard convection (from Latin "carry together") as heat transfered from A to B simply because of a fluid particle traveling from A to B and carrying a certain internal energy. Conduction, on the other hand, involves an exchange of kinetic energy on very small scale from one molecule to another. Both processes appear concurrently, and at any given time, in fluids (liquids and gases), and that's the answer to your question.

You could, of course, also create a definition based on our current idea of the laws of physics by saying that conduction is the part of heat transfer goverened purely by Fourier's law, while convection is described by the Navier-Stokes equations. In fluid mechanics, both relations are used in combined form. Because they appear in combined form. Not just in mercury.

So, is it convection or conduction that gives mercury the property of being highly sensitive to temperature changes? (used in thermometers)

The thermal expansion coefficient is what makes the mercury in a thermometer work. This is not the same as convection but it is related. If a pool of mercury is heated at the bottom it will expand and become less dense and the heated mercury at the bottom will tend to rise to the top, which is the mechanism for convection. Conduction would tend to transfer the heat from the hot "bubble" to the main mass of liquid so it would work against the convection effect. So I don't know if there would be much convection or not.

The thermal expansion coefficient is what makes the mercury in a thermometer work.

Yes. For expansion to take effect uniformly, the temperture has to reach a steady state, hence the question "how do we get there", i.e. through heat transfer.

Conduction would tend to transfer the heat from the hot "bubble" to the main mass of liquid so it would work against the convection effect.

Don't quite understand your point here. Can you elaborate? Conduction serves heat transfer (from the "hot bubble" to the rest), just like convection. They act in unison towards a thermodynamic equilibrium (temperature matching the outside of the tube) not against each other. A solid rod of steel could be used as a thermometer (with obvious disadvantages such as high response time and small magnitude) purely on the basis of conduction (and resulting thermal expansion). In a liquid thermometer convection serves the same purpose as conduction, just much faster.

Don't quite understand your point here. Can you elaborate? Conduction serves heat transfer (from the "hot bubble" to the rest), just like convection.

Yes, you are right. What I was trying to say was that if the conduction is efficient enough, the temperature will quickly equalize through the pool of mercury and that would remove the thermal gradient needed to drive the convection. I imagine if the heat input was enough and sufficiently localized you would probably be able to make a blob of hot fluid rise through the mass of cooler fluid. Would this actually happen in a thermometer? I don't know.

The reason mercury works better than a steel rod is because you have a ball of mercury at the bottom and it rises in a thin capillary tube so the volume change is magnified into a large linear displacement that is easier to see and measure. This would not work for a solid metal, but there are other ways like a bimetallic strip type of thermometer. My guess is that the conductivity of the glass tube containing the ball of mercury has more effect on the time response than the rate of heat transfer within the mercury itself.

It would depend on the Rayleigh number. http://en.wikipedia.org/wiki/Rayleigh_number