Did you know that two-thirds of the energy that goes into an average family car is wasted, primarily through heat loss? Thermoelectric materials have been hailed as one solution to this energy problem. But what exactly are they? Well, as the name might suggest, thermoelectric materials have some interesting thermal and electrical properties. Generators based on these materials can use wasted heat to produce electricity, and without the need for complex mechanical parts.
To generate electricity, you apply heat to one side of a block of thermoelectric material. This heat gives energy to the electrons inside the material, causing them to flow from the hot side to the cold side. A flow of electrons is otherwise known as current electricity. So, by applying heat, we can produce electricity – simple right? Not so much. The trick is that you want the heat to stay at the hot end and only the electrons to flow.
But, most of the time, those materials that like to move electricity also like to move heat – and this is the problem. Designing materials to do one but not the other – to have a high electrical conductivity and low thermal conductivity – is a difficult task. Add to that, the fact that existing thermoelectric generators have low efficiencies, and you can see that there are big challenges surrounding the use of thermoelectrics.
The efficiency of an average thermoelectric generator is about 7%. In comparison, some domestic solar cells, like the ones seen on the roofs of houses, may be expensive but can easily reach efficiencies of 35 – 40%. The numbers don’t work. So, to make thermoelectrics viable, we need to be clever. Enter nanotechnology.
There is firm evidence to suggest that engineering these materials on the nanoscale can produce thermoelectrics with high efficiencies. Nanotechnology is all about size; think about filling a box with large particles or small particles – you need many more of the small particles to fill it. In a material, the size of particles – or more specifically, the interfaces between them – defines how heat flows through it. Using nanotech, the materials can be engineered to “trap” some of the heat and slow it down, reducing its thermal conductivity, while allowing electricity to flow. So, by being a bit clever with size and interfaces, traditional, inefficient thermoelectric materials could be used to produce a new generation of energy harvesters.
NO LAB COAT NEEDED 
I loooooooooove it! Thanks for this post! I hope there is more to come!
Oh! Oh! tell people about the thermal electric materials on spaceships!
bumped into my twitter feed!
Well, consider me subscribed. Hope it goes well. I’ll try not to pull holes in every single thing. Who’s the blog really for though, Laurie? Other researchers? random Guardian readers? If you work that out it will make the writing a bit easier.
But I couldn’t resist. What’s this “wasted”? Aren’t all thermodynamic systems going to have some waste heat, as in not converted to mechanical work? But you seem to be connoting wasted as in (Lean, Toyota etc ) Muda (無駄) : resource not fully utilised in an economic sense.
Its more for me at the moment. I’m just using it as a place to collect all my non-work/PhD writing.
The “Life etc.” posts are pulled from my journal.
The science-y articles are pitched at a general (non-technical) audience – both of the current ones will be appearing on a website near you soon, albeit not in their current form.
I’ll write even if no-one reads it…. thanks for subscribing tho 🙂
Yes, of course there will always be wasted heat. But upwards of 60% is excessive in anyone’s language! I mean that you put fuel in, and of the energy that fuel can offer, only ~40% goes towards moving the car forward, operating the brakes etc. TE materials can capture a small percentage of this and use it to recharge a battery for example – in fact, we’re involved in large European project to do just that. NPL’s emphasis is on the characterisation of the materials themselves.
Hi Laurie.
I have thought about this a few times and if I have understood this correctly, I think that solar cells and thermoelectric devices shouldn’t really be compared like this.
Solar cells, like fuel cells, are energy conversion devices and can in principle be 100% efficient. The second law of thermodynamics places no limit on their efficiency.
Thermoelectric devices are heat engines: they derive some useful work from a heat flow across a temperature drop. Their efficiency is fundamentally limited by the second law of thermodynamics. For a temperature difference of 100 °C this limits the possible conversion efficiency to 27% ish before one even begins to worry about how to make the devices work. Actually 7% sounds pretty good to me.
However I do look forward to new super duper materials which push the laws of thermodynamics to their limit!
M
Hi Michael 🙂 It was more to give an idea of how much more work we still need to do to make them viable – we’ve “only” managed to reach 7% efficiencies and we’ve known about the Seebeck effect since the 1800’s. Solar cells are comparatively new and yet we’ve managed to massively improve on their operating efficiencies. But yes, you’re absolutely right (of course!) they certainly should not be directly compared, as they are entirely different technologies.
7% is indeed pretty impressive, but we also have a big problem with cost and availability – Tellurium, the element that is used in the most efficient TE generators is the rarest stable solid elements in the Earth’s crust. So there are two parallel efforts to broach this:
1. We’re looking at a number of alternative materials, such as skutterudites, bismuth selenide and some oxides
2. Making the tellurium based materials more efficient, to get the full “worth” from their high cost, e.g. nanostructuring the materials to improve the figure of merit.
So we’ve a lot to do!