please explain it...
Further than what? Wikipedia isn't perfect, but it might help you get the general idea:
thanx but i have got it. and the answer is that Probably the easiest way to think about this is with an analogy: the free electrons in a conductor can be roughly treated as an ideal gas. If you create a temperature gradient between two sides of the conductor, then one is at a higher temperature than the other and thus has a lower density of electrons. Since most conductors are solids backed by a crystal lattice, the positive charges must remain while the electrons are more dense on one side than another. What occurs is then an equating of currents: the thermal gradient current which creates the charge discrepancy must balance out the backcurrent induced by Ohm's law -- in a conductor, those electrons will repel each other if they get too dense on one side; conductors prefer to stay at one potential all the way through. An equilibrium voltage will develop between the two sides.
We now apply this with two different materials, both conductors or semiconductors, both are wires that have been joined to be a loop. We hold the junctions at different temperatures: then a current flows through the conductors as the difference between equilibrium voltages results in an electromotive force around the loop.
You can think of it also this way: the "hot" electrons move faster through one of the conductors than they move through the other, creating a net motion "down" the fast conductor's gradient from hot to cold. But Kirchoff's rules forbid this sort of net current in a loop unless it's felt all the way across the loop, so that the difference in speeds propagates into a full current effect.
This also helps you see that the electrons pick up thermal energy at the hot reservoir and dissipate it into the cold reservoir with their fluid flow.
You still need to introduce different behaviours of different metals. In this first attempt, junctions of identical materials would give a current, which doesn't happen. More, the loop has two paths from hot to cold, which can't behave identically if you're to observe a current or voltage.
Beware also that electrons in a metal don't resemble a gas. For some properties maybe, for others not at all - for instance the heat capacity of a metal.
how can you say like that electron do not resemble a gas. i have studied that they do so but i m myself fictitious about it. have you any evidence which shows this nature of electrons, then specify please!
i use a different analogy.....
Electrons in a metal are loosely bound to their atoms, at least the outer ones. Hence the term "Sea of electrons"....
Well, different metals have diferent affinities for those outer electrons.....
hold that thought.
Now, temperature is atomic motion . I use the mental picture of shaking electrons off the metal atoms. This actually happens in vacuum tube cathodes, read up on 'thermionic emission'.
Now, if you heat a wire at one end only you will shake electrons off that end and they will be pushed toward opposite end. A small potential will exist between the ends.
If you join two different wires and heat them at the joined end, at the opposite end you will find two different potentials. That's because of their different affinities for electrons and the temperature gradient along the wires.
Sorta like shaking fruit out of trees. Apples come loose easier than acorns.....
Separate names with a comma.