Could one build "thermal power lines"?

[rumborak]

It just occurred to me that akin to electric power lines, which in the end just connect a voltage differential over a long distance and facilitate power exchange between the two end points, one could build "thermal power lines". That is, a well-insulated cable that internally consists of a material with very high thermal conductivity. This way, one could facilitate energy (heat) exchange over long distances, between two points of vastly different temperatures (e.g. a desert and the bottom of the ocean).

Question is, how feasible would it be? It would certainly not travel at the speed of light like electricity, but for heat exchange it probably wouldn't need to.
Has something like this been tried before?

 

[Drakkith]

The problem is that thermal energy isn't transported very well through even a good thermal conductor compared to the transfer of electrical energy through an electrical conductor. Try heating up the end of a 1 meter long copper wire and seeing how hot the other ends gets. Even with very good thermal insulation and a temperature difference of several hundred degrees, the rate that energy can be transferred from the hot end to a heat sink attached to the cold end is extremely low compared to the rate of energy transfer in even crude electrical circuits.

You could try setting up lines that were hollow tubes filled with a working fluid, but you'll lose so much energy through the insulation and in pumping the fluid that you're probably better off just using standard electrical lines. I'm sure it is sometimes feasible to use a working fluid inside hollow tubes if you can't connect your heat engine directly to the heat source, though. But you want those lines to be as short as possible so you lose as little heat as possible to the surrounding environment.

 

[DrClaude]

You could try setting up lines that were hollow tubes filled with a working fluid, but you'll lose so much energy through the insulation and in pumping the fluid that you're probably better off just using standard electrical lines.

Not necessarily. This is, after all, how district heating works. On a smaller scale, it is also the basis of a geothermal heat pump.

 

[Drakkith]

Not necessarily. This is, after all, how district heating works. On a smaller scale, it is also the basis of a geothermal heat pump.

You're right. I was thinking in terms of energy transfer for conversion into electrical power.

 

[jtbell]

And a lot of old buildings in places where it gets cold in the winter probably still have steam-heating systems. My office in grad school had one of those old style cast-iron radiators in it.

 

[Nugatory]

The big problems with using heat to transmit power the way we do with electricity are
- There is no such thing as a really good insulator for heat.
- There is no equivalent of the step-up/step-down transformer for heat.
- Changes in the potential difference propagate much more slowly, which complicates control.

As Dr Claude's examples show, the idea works just fine in the applications where these limitations don't matter.

 

[ZapperZ]

And a lot of old buildings in places where it gets cold in the winter still have steam-heating systems.

Argonne is still being heated by steam lines. Most of the buildings on the lab site do not have their own boiler. You'll see these insulated pipe lines running into many buildings from their main heating station. See the right side of the street in the photo below:

As for the question being asked in this thread, which is heating wires to transfer heat, not only is this highly inefficient, but it is also very, very slow when compare to charge transport.

Zz.

 

[Nidum]

There is a well developed technology of Heat Pipes .

These work well over short to medium distances .

I first heard of them in connection with the Apollo project .

 

[russ_watters]

I suspect most corporate/industrial campuses in northern latitudes use campus wide steam and chilled water. Steam is more flexible than I think most people realize: you can generate it at "high" pressure and then "step it down" to lower pressure or even use it to generate hot water. Heck, if you have an excess of steam, you can even use it to generate chilled water!

But no, direct thermal conduction wouldn't be very useful.

 

[anorlunda]

It just occurred to me that akin to electric power lines, which in the end just connect a voltage differential over a long distance and facilitate power exchange between the two end points, one could build "thermal power lines". That is, a well-insulated cable that internally consists of a material with very high thermal conductivity. This way, one could facilitate energy (heat) exchange over long distances, between two points of vastly different temperatures (e.g. a desert and the bottom of the ocean).

I think you are asking about the heat pipe. They are very clever things, but they are not useful for large power or long distances.

https://en.wikipedia.org/wiki/Heat_pipe

 

[russ_watters]

Heat pipes don't transmit via conductivity, though. They are basically passive (self powered) heat pumps.

 

[rumborak]

Interesting comments, thanks.

The big problems with using heat to transmit power the way we do with electricity are
- There is no such thing as a really good insulator for heat.

This one struck me as the key. I actually looked a bit at tables of thermal conductivity of different materials, and they all seemed to hover with maybe 3 orders of magnitude. Looking at a table of electrical resistance of materials, that spans about 30 orders.

 

[sophiecentaur]

Interesting comments, thanks.
This one struck me as the key. I actually looked a bit at tables of thermal conductivity of different materials, and they all seemed to hover with maybe 3 orders of magnitude. Looking at a table of electrical resistance of materials, that spans about 30 orders.

It strikes me that the reason for the differences in 'range' of conductivities is that electrical and thermal conductivity in metals are by the same mechanism (electron motion) and they are both high values. Their values tend to track each other from metal to metal. But electrical conductivity in a non-metal will be poor because there are no delocalised electrons. However, heat can still be transferred by vibrations (phonons). That mechanism takes over and imposes a much higher lower limit over the whole range of materials.

 

[rumborak]

Yeah, and when you look at an effect that doesn't rely on delocalized electrons, like dielectric constants, those also only have a spread of 3 orders of magnitude.

 

[sophiecentaur]

Yeah, and when you look at an effect that doesn't rely on delocalized electrons, like dielectric constants, those also only have a spread of 3 orders of magnitude.

You have done the footwork on this one and I'm being lazy but have you spotted some substances 'in between'? Or is there a big gap between metals and others?

I remember being told / having read somewhere that the quantity Resistivity shows the biggest range of values of any quantity. I wonder if the difference between electrical and thermal conductivity could be because of the statistics involved in the transfer of energy by a random process as opposed to a process like electrical conduction which has an imposed order to it. Very arm waving, I know but statistics is responsible for some pretty powerful force-like effects. I wonder if Prof Kittel could help us there. Must try and dig him out.