- As a kid, I would often connect two ends of a wire to a battery to get the current flowing through it, and then very quickly joined
the ends of the wire together. My hope was that if I
could do it quick enough then I would catch the current, and it would just be going round and round in that loop of wire for ages. I don't think it's possible,
but it is genuinely possible to get electricity flowing round and round in a loop of wire. All you have to do is make one end hot and one end cold like this,
and look using my multimeter, you can see there really
is a small current flowing round and round
in this loop of wire. I say that all you have to do is make one end hot and one in cold. Actually, there's
something a little unusual about the wires, they're
not made of copper like wires normally are. Instead they're made of
two different alloys. One is called Chromel, the
other is called Alumel, and there's nothing particularly special about alloys in general. And alloys just a mix of different metals, in this case, mostly nickel, but with a few different ingredients to differentiate the two. And remarkably, when you
have two different wires made of two different
metals joined in this way, and you heat one of the junctions, while cooling the other, electricity will go round,
and round, and round. It's an incredible effect that turns out to be incredibly useful, especially when you consider
the fact that it's reversible. So in this case, we're using a difference in
temperature to create a voltage, but the opposite is also true. If you present a voltage then it creates a
difference in temperature. Those two phenomenon together are known as the thermoelectric effect. And I'll talk about both in this video, but specifically in this case, where a difference in
temperature is creating a voltage that's called the Seebeck effect. What causes the Seebeck effect? Well, let's look inside
this short bit of wire. You might know that metals have
these free moving electrons. So normally electrons
are bound quite tightly to their atoms, but in the case of metals some of the electrons
are free to move around within the bulk of the metal. That's what makes metal conductive when electricity is moving through metal, it's because those free
electrons are flowing through it. These freely moving electrons
also have some thermal energy, unless the metal's at
absolute zero, which is rare. I'm showing that thermal
energy as a jiggle of the electrons, because that's all thermal energy is, it's atomic scale jiggle. These freely moving electrons
are sometimes referred to as the sea of electrons. But thinking about the electrons as a gas is actually a more useful
analogy for our purposes. And you probably know
this, when you heat a gas the gas expands. In other words, the particles of the gas get further
away from each other. And that's because as you heat the gas they get more thermal energy, the amount of jiggle goes
up, they're moving faster. They're bumping into each other
more often with more energy. So they're pushing themselves
apart from each other. And the same thing happens with the gas in inverted commas of
electrons in our metal. As you heat one end of the bar those electrons will move
apart from each other. And while you're cooling, the other end they're gonna contract. They have less thermal energy. They're bumping into each
other with less energy, less often, they can get
closer to each other. So the overall effect now
is a deficit of electrons on the hot end, and a surplus
of electrons on the cold end. So what that looks like from the outside is a slight positive charge on one end, and a slight negative
charge on the other end. So that's the Seebeck
effect in a nutshell, but crucially the Seebeck
effect is more pronounced in some metals and less
pronounced in others. So imagine taking a different
metal, heating up one end, cooling down the other
in exactly the same way. but that charge separation
is less pronounced, because Seebeck effects isn't as strong, because the lattice of ions of the metal have a stronger grip on those electrons. So a greater charge
separation isn't allowed, but in the case, what happens if you take these two different metal wires, and join the two ends together? Well, the Seebeck effect is strong in the top wire in our diagram here. So the electrons are being
pushed strongly to the right, and in the bottom wire at the
Seebeck effect is less strong. So those electrons are being
weakly pushed to the right. And they're actually
pushing against each other. So the top wire is pushing clockwise, and the bottom wire is
pushing anti-clockwise. But because the Seebeck effect is stronger in the top wire, the top wire wins, or to put that in less
anthropomorphic terms the net effect of the top
wire pushing clockwise, and the bottom wire pushing anti-clockwise is a net clockwise push of the electrons, so you get a flow of electrons
in the clockwise direction. When you have two dissimilar
metals joined together, like this, it's called a thermocouple, and a thermocouple is
the working principle behind a lot of thermometers especially those probe
thermometers that you put in the food that you're cooking, or the one I used when I was calculating absolute zero, that time. If you look inside the tip of one of those probe
monitors there you'll find two dissimilar metals joined together. The voltage you get from the
thermocouple is really small. If you want a decent voltage you need to put a load of
thermocouples in series. And that's actually what's
going on inside here. So in here, you've got loads
of wires going up and down up and down, up and down,
joined altogether in series. So the whole load of the
junctions are all hot. A whole load of junctions
are all much cooler. This is called a thermopile, and the voltage you can get
from it is much greater, as you can see here. A thermopile is an important
component in your boiler. You might call it a
furnace or a water heater. So your boiler has this
thing called a pilot light. Actually modern boilers
don't always have one, but one of the jobs of a pilot light is to ensure unburnt gas
doesn't leak into your house. So the pilot light is always lit. It's right there next to the burner. So if any gas comes out of the burner it's gonna get lit. That's how the pilot light
prevents unburnt gas escaping. But what if the pilot light goes out, you need some way to detect that, and then switch off the gas supply. And that's where the thermopile comes in. If you look just above the
pilot light in your boiler you'll see a thermopile. And that generates enough voltage to power a solenoid that can
hold open the gas valve. So if the pilot light goes out the thermopile cools down,
the voltage goes down, the solenoid shuts off,
and the gas valve closes. You could instead use
a single thermocouple producing a tiny voltage, and use that as a switching mechanism. So you have a separate power
supply for the solenoid, and the thermocouple is
just used as a switch. The issue then is every
time there's a power cut, the gas valve closes, and
you lose heat as well. Conversely, a thermopile system
is entirely self-contained. You've actually experienced
the effect of a thermopile in your boiler, if you've
ever had to relight it, after the pilot light's gone out. And to relight a boiler with
an extinguished pilot light, you have to push the dial in
that overrides the solenoid, when you let go, you give
control back to the solenoid. So if you don't leave the
dial pushed in for long enough after the boiler is relit, the thermopile doesn't have
enough time to heat up. When you let go the solenoid closes again, and you have to try again. I mentioned that the opposites
of the Seebeck effect is also a thing. In other words, if you apply
a voltage to this circuit one junction will get hot, and the other junction will get cold. That's called the Peltier effect. It's analogous, by the
way, to the Sterling engine from my video by Entropy, you can use the difference in
temperature to turn a thing, or you can turn a thing to drive a difference in temperature. Anyway, I'll show you an application of the Peltier effect in a minute, but first what causes it? To see why the Peltier effect is a thing we need to look at metal
in a slightly different way to the way we were when we were explaining
the Seebeck effect. Let's think about the energy
levels of the electrons. So let's imagine we could
extract all the electrons from a metal and then
kind of pour them back in. They would start to occupy
the lowest energy levels that they can, i.e. I, the ones closest to the nuclei of the metal ions, and
the next atoms to come in will go in the next level up from that, and the next level up,
and the next level up. You think of those as the
orbitals of the atoms, but in a metal you get to the point where it's not orbitals anymore, it's energy bands where
all the energy levels are kind of shared amongst the atoms. But in any case it's analogous to pouring balls into a beaker. Gosh, I haven't poured
anything out of a beaker for a long time. At the moment, this analogy
is missing something. It's missing thermal energy. At the moment this looks
like it's at absolute zero. So to add thermal energy we need to put some jiggle in there. And it's only the top most electrons that are able to jiggle. In other words, they're able to jump up and down through higher energy levels, and that's what I've
tried to illustrate here. I want to be clear that when you're filling
a METAL with electrons, it's not like they're sloshing around at the bottom of the metal
with no electrons at the top. It's that they're
filling the energy levels close to the atoms, and
then further and further out until you're looking
at these energy levels that take up the whole bulk of the metal. It's just that we're
representing those energy levels on the vertical axis. Importantly, different
metals have different numbers of electrons, different
strengths of nuclei. and so they'll fill up
two different levels. So here's a different metal here, and you can see those free electrons, that are jumping up and down, they're higher overall. So we can bring these two metals together to form a junction. And in reality, there would
be some movement of electrons between the metals, but
I'm not gonna show that because it confuses what
I'm trying to explain. So now what we're gonna do
to show the Peltier effect we're gonna apply a voltage
and you can think of voltage as just pushing the electrons. In other words we're going
to push the electrons in this case we're gonna push them from the left and see what happens. Well, it's a bit like bouncy
balls on two different shelves, at different levels. The bouncy balls on the lower shelf we're gonna push them across until they get up onto the top shelf. Well, let's look at what
happens to actual bouncy balls when I release them onto the ground, they've got about 60
centimeters of bounds here. But look the ones that happen
to end up on the table, they've only got about 10
centimeters of bounce, if that. And that's because there's
this constant interplay between kinetic energy and
gravitational potential energy. In other words, when a ball is bouncing it's constantly converting
the energy it has between kinetic energy and
gravitational potential energy. At the bottom of the bounce, it's got loads of kinetic energy no gravitational potential energy. At the top of the bounce, all that kinetic energy is converted to gravitational potential energy. And if at that point, you
move it over onto the shelf, well, it's just gonna
sit there on the shelf, because it has no kinetic energy. And the same is true
when you push electrons across onto that high
shelf, in inverted commerce. It's not an interplay
between kinetic energy and gravitational
potential energy we'd like, like with the bouncy balls, it's an interplay between kinetic energy and electrical potential energy. In other words, the
potential energy it has because it's a negatively charged particle near a positively charged nucleus that it's moving away from when it moves up through the levels. The important thing is those
bouncy electrons that make it up onto the shelf, they're not bouncing so much anymore. And that bouncing up
and down, that jiggle, that kinetic energy it's
on the atomic scale. So it's thermal energy. In other words those electrons that
make it up onto the shelf they have less thermal energy, they cool down the metal. So when you apply a
voltage across the junction in such a way as to push
electrons from the low shelf to the high shelf, you
cool down the junction. The opposite is true when you push the electrons off the shelf, they gain kinetic energy because
they have further to fall. And that gaining kinetic
energy is equivalent to a gain in thermal energy,
so the junction heats up. If you make use of the
Peltier effect in a device it's got a Peltier element
like this one here. If I connect these two wires to a nine volt battery,
one side gets really hot. The other side gets really cold. I pulled this one apart so you can see what's going on inside. It's hard to make sense of
it really because, you know, some bits have stayed stuck to the floor, some of it stayed stuck to the roof, but hopefully you can
see there's a whole load of thermocouples all in
series with each other. And it's actually different
semiconductors that are used in here, as opposed to different metals. With a heat sink and a water bath you can even use a Peltier
element to freeze water. ElectroBOOM made a video
about Peltier elements a while back that's worth a watch. And I know that Alpha Phoenix
is using Peltier elements in an upcoming video
where he's trying to grow huge single crystals of water. So subscribed to Alpha Phoenix in advance, so you don't miss it. It's a great channel anyway,
so go and have a look. I'll leave a link to their
channel in the description, and on the end screen. So I thought I'd share some more Blinkist recommendations for you. Blinkist are sponsoring this video. It's an app that does
something remarkable. It condenses non-fiction
titles into 15 minute reads. They're also audio narrated, so you can listen to them in the car. Here are my recommendations "A Short History of Nearly
Everything" by Bill Bryson. Very good, I read the
whole thing after that one. "Talking to Strangers"
by Malcolm Gladwell. "The Book You Wish Your Parents Had Read "and Your Children Will
Be Glad that You Did." It's another parenting book that's been really good
for us and our kids, and "Born a Crime" by Trevor Noah. That's just four recommendations. There's so much to choose from. They've also got some great audio books that you can get a lot
cheaper than anywhere else. The first 100 people to go
to blinkist.com/stevemould will get one week absolutely
free, no strings attached, cancel whenever you like. It's an opportunity to try it out, but if you want to carry on, you'll also get 25% off membership. So the link is also in the description. Check out Blinkist today.
(upbeat instrumental music) I hope you enjoyed this video. If you did, don't forget to hit subscribe and I'll see you next time. (electronic swooshing)
(upbeat instrumental music) (upbeat trumpets begin)
