[syhprum (OP)]

Radioactive thermal generators produce power from the heat produced by radioactive materials such P238 with the aid of thermal junctions or in later designs sterling engines.
They have been extensively used to power space craft where the long life and absence of any need for maintenance is an advantage.
The cost of the fuel is very high so I wonder how the cost of power produced compares with the cost of power produced by fission or fossil fuel power stations

 

[Janus]

A couple of factors come into play.  Pu-239 and fuel used for RTGs has to be synthezied, and has a purity of ~85% while fuel for a typical commercial fission reactor can be mined and then enriched to ~ 20% U-325. (Naval Nuclear vessel reactors can use fuel enriched to about the same degree as RTG fuel)
The Galileo probe RTG could produce 5.4 watts/kg of fuel.
A a 1,000 Mw commercial reactor needs about 75,000 kg of low enriched nuclear fuel, which works out to over 13,000 watts/kg.
I assume that it cost less to mine and enrich Uranium than it would to synthesize the same amount of PU-239, and the nuclear fission reactor is much more efficient in terms of wattage per kg of fuel,  so I'd have to hazard a quess and say that cost-wise, the RTG doen't stack up very well.

[Bored chemist]

The 239 Pu is pretty much a by-product of producing power in a nuclear reactor.
So, there's never going to be much of it. If you used lots of reactors to make it then there would be  lots of reactors producing power and the cost of electricity would fall.
The cost of RTG power is huge.

[alancalverd]

The cost per watt is significant, but the cost per joule is not! The halflife of Pu28 is 87.7 years, so you can extract a continuous 5W/kg initially, 2.5 W/kg into the next century, and 1.25W/kg when your great-grandchildren  have lost all interest in the project. Additionally, the system is extremely reliable since the thermocouple has no moving parts or consumables. Increasing the power output is not a problem as the inherent heat generation is 500W/kg of fuel, and doubling the size of the the fuel pellet does not require significant expansion of the thermocouple system.

[syhprum (OP)]

I understand that the thermocouple material is damaged by radiation does this limit the life ?.
as with many heat engines the source of cold is more of a problem than the source of heat, could one operate on Venus ?.

[alancalverd]

Pu238 emits mostly alpha radiation, so I guess the trick is to find an intermediate material (lead, ceramic or even polymer?) that can be expected to outlast the design life of the battery. Most of the heat actually comes from self-absorption within the plutonium billet, so the larger the billet the more power per unit radiation damage.

Of the 250 plutonium-powered pacemakers Medtronic manufactured, twenty-two were still in service more than twenty-five years later, a feat that no battery-powered pacemaker could achieve.

....where the "cold source" is the human body. I can't imagine what power sources could work for any length of time on Venus at 480 °C.

In terms of energy density, Pu238 is 2.2 x 10¹² J/kg, some 2,000,000 times that of a lithium battery, so "a few dollars more" may be well spent.

[Bored chemist]

I can't imagine what power sources could work for any length of time on Venus at 480 °C.

How about an RTG?
OK, you would need slightly different materials.
Plutonium oxide melts around 2700C
So you can run the "hot" source at say 2000C with no problem.
That gives you a temperature difference of  roughly 1500C to work with.
That's comfortably in the range of type D and G thermocouples.
https://en.wikipedia.org/wiki/Thermocouple#Type_G

[evan_au]

What is the cost of electricity produced by radiothermal generators (RTGs)?

Maybe we could reverse the question: If you are putting a piece of equipment into a difficult location, how much money will you save by using an RTG, compared to the alternatives?

The New Horizons spacecraft carried 10kg of Plutonium to Pluto; the initial 250W dropped to 200W by the time it reached Pluto.
To generate 200W of electricity at Pluto by means of solar cells:
- Pluto orbits at an average of 49 times the Earth-Sun distance, so sunlight at Pluto is around 2000 times dimmer than at Earth
- Assuming high-efficiency solar cells (eg 20% electrical efficiency), you would need about 2000 m² of solar cells.
- If you made this strong enough to withstand the vibration of liftoff, I am sure it would weigh far more than the RTG, and introduces a lot more mechanical complexity (eg motors to unfold and steer the solar cells)
- And minimising mass at liftoff is a major constraint for a space mission

The New Horizons RTG is now dropping below 200W - but it still has enough power to support a 4-year extension of its mission. It is planned to fly past Kuiper Belt Object 2014 MU₆₉ on 1st January 2019. This would not really have been possible using solar cells.

Similarly, the Mars Curiosity Rover has 5kg of Plutonium in its RTG. It provides about 100W of electricity, and 2,000W of heating to keep the electronics at working temperature. So the electrical efficiency is not high, but the "waste" heat is actually very useful (as it would be at Pluto).

Again, solar cells would be impractically large, and would require big batteries to store electrical power during the Martian nights.
See: https://en.wikipedia.org/wiki/Curiosity_(rover)#Specifications

[KevinSalyer]

Thinking of using radiothermal generator for cars which may result in running without adding oil. So go to google for the WTF nuclear car out of Cadillac. Using the thorium (thorium) drive claimed 100 years without refueling. But not yet on sale. It looks like it happened three years ago-but he's got a big flaw. If he's in a car accident, it's like a micronucleus leak. Imagine Japan's nuclear leak.