Personal Energy Generation with Thermoelectric Materials

Interview conducted by Will SoutterAug 15 2012

In this interview, Professor Robert Dorey talks to AZoNano about his work on personal energy generation technologies, which will have many applications in the military sector, as well as in consumer devices.

Can you please outline the types of personal energy generation technologies you’ve been working on?

There are many types of different personal energy generation technologies – piezoelectrics, fuel cells, etc. – the main ones working with at the moment are based on thermoelectric materials and solar absorbing materials, which complement each other in a really useful way.

A thermoelectric material can generate electrical power from a temperature difference, and the solar absorbing materials allow us to convert sunlight into heat – they allow our temperature reservoir to heat up more effectively. So when you bring the two together into one device, you can efficiently generate power from the thermoelectric effect.

The thermoelectric material is the one with the most interesting science behind it, and the one that people may not have come across before. If you heat up any material, you excite charge carriers into the conduction band – either electrons or holes. If you imagine heating up one end of a metal bar and keeping the other end cold, the hot end should generate more charge carriers than the cold end, so you end up with more charge at one end of the material than the other – you’ve got a voltage. Nature then tries to create a more uniform charge distribution. That means those charge carriers will migrate through the material, and then you’ve got a current.

Using this general approach, we can generate electrical power just from a temperature difference. All materials will do this to some extent; it just happens that the vast majority of materials do it incredibly poorly, so we don’t notice it. But there is a certain selection of semiconductor materials which do it quite well, and we can start making use of that power.

If we take those semiconductor materials and modify them by putting dopants in, we can make p-type semiconductors which generate a lot of holes (positive charges) when heated, and n-type semiconductors which generate electrons. If we connect these two types in series, and put a temperature difference across both of those materials, we can start getting higher voltages. Just one material will generate 10’s to 100’s of millivolts across it, which is hardly anything. But just like stacking up many batteries, if you start to daisy chain these p- and n-type materials together, you can generate useful voltages.

So how does nanotechnology help to make these materials more efficient?

There are two different ways of conducting energy through your system. One is thermal energy transfer via phonons, which are vibrations of the crystal lattice, and the other one is via electrons travelling through your material. So for a good thermoelectric material you want it to have a very poor thermal conductivity, but a very good electrical conductivity.

However, if you think about copper, for instance, it has a good electrical conductivity, which is why we use it in wires, but we also use it in kettles, because it is a good thermal conductor as well, and that’s actually really bad for the thermoelectric effect. And if you go the other way, to insulators like ceramics and glasses, they are good at insulating from thermal energy, but they are also good electrical insulators. So you’re always fighting a battle trying to find a material which has the right set of properties for the thermoelectric effect.

Fortunately there is a difference in length scales for thermal and electrical conduction. Thermal conduction tends to occur over relatively large distances, so the phonon waves travel a long distance before it is scattered or reflected, but electrons tend to bounce around all over the place. So if you put some nanoscale structure into your material, you can disrupt the phonons, because your material doesn’t have enough of a periodic structure to develop an effective wave of vibrations through the crystal structure, but the electrons aren’t affected, because they are already scattering at a much smaller length scale. So with nanostructures, you can reduce the thermal conductivity, but maintain a good electrical conductivity, so the performance of the thermoelectric material goes up.

With the solar energy absorbers, we’re trying to play a similar trick by structuring the materials so they have really reflectivity, so very little of the incident light radiation bounces off the surface, and it all gets absorbed.

By combining those two aspects of nanotechnology together, using these very small structures, we can get materials that absorb solar energy really well, and materials that have good thermoelectric properties as well. And the materials themselves are not exotic - it’s the nanostructure which gives the materials improved properties, compared with the same materials in the bulk phase.

How will this technology benefit soldiers in the battlefield?

Basically, we’re looking at increasing the range of a soldier – increasing the amount of time he or she can stay away from base, and decreasing the amount they have to carry around. The average soldier has around 70 kg on their back in service or out on patrol, and quite a lot of that is battery weight. Actually, a lot of it is replacements for the batteries that are in the equipment already. So if you can make the batteries last longer, they can stay out for longer, carry more food, more ammunition.

So there are real benefits if you can reduce the load from the stored energy they have to carry around with them. And you can start doing some really exciting things as well, if you scale down from that to more localized energy generation. For example, a sensor, which could be scanning for chemical attacks or other hazards, could be powered by a small thermoelectric device right next to it, so you don’t need to thread cables from the battery pack to power it.

How much current could a portable thermoelectric system feasibly generate?

Well it does depend how big the array is, just like with solar cells. With solar cells we usually talk about “power per square meter”, and it’s exactly the same with thermoelectrics – the bigger the area over which it operates, the more power it generates. So the challenge is to hit the same sort of levels that we are seeing from solar cells – about 100 mW/cm2.

However, with thermoelectric systems, the power that you generate also increases with the temperature difference. So you could have a large area working on a small temperature difference – say room temperature to a soldier’s body temperature, about 16 °C difference.

Alternatively, you could have a smaller, dedicated solar absorbing zone that heats up with exposure to sunlight. Even in Cranfield in March, managed to heat some solar absorbing material up to 70 or 80 °C just by sitting it on the windowsill in the office. So that gives you a temperature difference of 60 °C, and that can start getting you closer to 10’s or 100’s of mW/cm2, which is a useful amount of power.

How long do you think it will be before we start seeing personal energy generation technologies in consumer products, as well as in the military sector?

Probably less time than you think. In the military environment, if it breaks you’re there are serious repercussions, so the devices need to be more robust and withstand different types of environments than they might need to be for civilian applications. And it’s a really cool thing for charging gadgets – just thinking about how many gadgets people own and are carrying around, it’s a typical “boy’s toy” solution for keeping them all charged for longer. So that market is hungry for that kind of application.

In terms of which one is going to come first, the military or commercial applications, I wouldn’t like to put any money on it. I can see there are very clear military applications, but also a lot of strong commercial areas, so I’d think they’ll appear at about the same time. As for the sort of time span, we’re beginning to see near-to-market products coming online already, so it’s not very far away. However, they are based on conventional technologies - existing thermoelectric materials which aren’t that nanostructured at all. They are being integrated with cooking equipment for camping, for example, where large temperature gradients are available. So there is stuff out there already, but it will probably be another couple of years before we see the nano side of it appearing.

What are some of the next big breakthroughs that we will see people working on in the next few years?

One of the really exciting things is trying to get these things integrated into textiles and fabrics, so they become a part of the system – we’re not talking about having a matchbox-sized thing sewn to the side of your coat, these things would become an integral part of the garment.

So, you can imagine, for soldiers, it would become part of their backpack, for instance – the very fabric that their backpack is made from would be the active thermoelectric element. So you’re again trying to reduce the total weight that the soldier is carrying around, and deliver a fully integrated package.

And that is a real challenge – trying to make these things on a flat substrate like a piece of silicon is challenging enough. Trying to make it on a 3D structure, a woven structure is very exciting.

What inspired you to start working on thermoelectric materials and other personal energy generation technologies?

I’m a materials scientist by training, and active nanomaterials have always been an exciting area to work in. The concept of working on personal energy generation in particular came about from a number of different things.

Partly it was frustration from doing lots of travelling, and having phone batteries run out at the most inopportune moment, never having a charging point at the airport, or having the wring type of plug.

Then I read some articles on how telecommunications were developing in Africa – they’ve just missed out the whole landline thing, and gone straight to mobiles. The number of mobile users in Africa is just staggering, and it’s higher than in Europe or America, because we still have the landlines. They’ve missed out that whole concept of putting in a fixed architecture for telephones, and that got me thinking – could we just miss out the concept of putting in a fixed architecture for power distribution?

Even in the UK and Europe, where we have the National Grid which is connected everywhere, it doesn’t really suit our active lifestyles any more. Even in an office, you can have power sockets in the wall next to your desk, but if your desk is in the middle of a large office, getting power to it can still be quite difficult – and that’s just in your office, with power supposedly everywhere.

And it just gets harder as you leave the building and go outside – getting power on the move is quite challenging. Then outside the city, in rural surroundings, power definitely isn’t always around. And that is all in a developed country with a power infrastructure.

So I think there is a real market for power at the watt to kilowatt range – enough to serve individuals or a household, maybe linking up and powering a village.

It’s really a kind of “bottom-up” approach to power - which is very appropriate when we’re talking about nanotechnology and “bottom-up” manufacturing!

About Professor Robert Dorey

Robert Dorey holds a Chair in Nanomaterials at Cranfield University, and is part of the Surface Science and Nanotechnology Institute. He leads an independent research group on nano and microtechnology in energy and the environment. His research work explores three interrelated themes:

• Personal energy generation solutions based on functional materials and devices including solid oxide fuel cells, thermoelectric, ferroelectric, pyroelectric and piezoelectric materials.
• Environmental and structural health monitoring sensors
• Environmentally friendly processing routes for minimising energy, material and chemical usage.