In addition to what I just put, any charged particle will be affected by a magnetic field and a metal gas particle may well be charged. I'm not really a chemist so I don't know how such a gas works in reality.

Second mainstream opinion: you can convert heat into other energies, but only a part of the heat. The rest must flow in a cold source. That's called the second principle. The cold source must be colder than the warm one. The maximum theoretical proportion of energy you can extract from the heat of the warm source depends on the ratio of absolute temperatures (the one expressed in Kelvins) between both sources.

Creating a cold source is difficult. For instance, the focus of a telescope pointing upwards is one, but of minute power. Creating a warm source is easy, just burn some wood. That's why the ambient temperature is used as a cold source by heat engines.

Now, you may try to exploit smaller differences in the ambient temperature. The best hope may reside in oceans, where the surface is warm (+25°C in nice places) at the bottom cold (+4°C). However, the efficiency is small because the ratio (273+25)/(273+4) is small, and it is difficult to extract any energy from it after you have paid for all losses.

Steam: 100°C are necessary only at 1 bar. You get steam at 0°C and even less, but at a lower pressure. For instance in any power station, where steam is expanded to much less than 100°C and 1 bar, in order to increase the efficiency. They put a condenser for it, cooled by a river for instance, at a pump that extracts the condensed water from few millibars to 1 bar or back to 200 bars.

Explanation of thermodynamics at the molecular level: this is called the kinetic theory of gases. Rather a good book to read than a short explanation in a forum.

And yes, we know quite a bit about entropy rising, maximum efficiency and related things. That's called thermodynamics.

Macroscopic theories take the second principle as a hypothesis, apparently not what you want. Microscopic theories demonstrate it. I know "Statistical Physics" from Landau&Lifschitz, but it's quite a headache. If you can find a book by a clear teacher, like Feynman, it should help more.

Just a suggestion: incredible numbers of people, some of them brilliant, have spent unconsidered time on trying to build machines or demonstrating ideas that would go against the second principle, and none (= not a single one) is known to have worked. So this might be an intellectually interesting topic, but one's time is, with a huge probability, better invested somewhere else. Growing algae in seawater or in the desert to harvest hydrogen or vegetable oil or cellulose is a better bet - there are more examples.

Any metal can be (and is) vapourized. No metal vapour is ferromagnetic to my knowledge (and not even an electrical conductor) because ferromagnetism is a molecular property, not an atomic one.

Examples: austenitic steels (the common stainless steels, or better the 17-12 versions) are nonmagnetic. But manganese-zinc alloys, called "ferrites" by electronicians, are ferromagnetic though no element is. And CrO2 is a permanent magnet composed of nonmagnetic elements.

I know no gaseous molecule that would be ferromagnetic, but there may be some. If not, just consider a very fine dust of ferromagnetic solid dispersed in a gas (take care, or the dust may catch fire brutally).

Gases don't conduct electricity well. People willing to make MHD devices try hard but get tiny efficiencies because of that.

For an electricity-conducing liquid, take the K-Na alloy for instance. Liquid at room temperature and even below. Flammable, but not as toxic (before the fire...) as Hg. Moving magnetic fields are used to pump them much like a rotating field turns a squirrel-cage rotor.

Ferromagnetic liquid: disperse an adequate powder in the liquid.

Charging a metal gas isn't easy. Ionizing the atoms is more difficult (requires much more energy) than separating them into a gas. So a metal gas is usually composed of neutral atoms.

Thanks a lot for the answers. I am definitely not en expert, so it helped a lot. Plus I don't use English every day, so forgive me my mistakes Anyway, I am definitely not going to argue with laws of physics, that would be a waste of time. I just want to know the "why". So far it seems to me that movement of atoms is what drives the universe, and it cannot be stopped. In anotherwords, entropy is time itself. Maybe a little far-fetched, but why not.

So yet another, also related question.

"Temperature" is in another words "how fast do the atoms move". What processes make the atoms move? I am not talking about simply passing on the kinetic energy, but when do they start to move, from 0 to more than 0. Is it during the nuclear reactions? nuclear fission? Only then?

I hope it's not a confusing question. Well, try to read my mind I know it doesn't fit general physics anymore, sorry.

Whitefire:

"Temperature" is in another words "how fast do the atoms move". What processes make the atoms move? I am not talking about simply passing on the kinetic energy, but when do they start to move, from 0 to more than 0. Is it during the nuclear reactions? nuclear fission? Only then?"

You seem to have a "picture" of atoms in your world view. Like billiard balls maybe?

In fact nobody in the world has ever seen an atom or an aggregate of atoms. Physicists postulate their existence from the traces they leave in the macrocosm, because it is the simplest explanation of the effects studied. One of the traces is temperature, an aggregate trace of many atoms in motion. Atomic kinetic energy is a definition of temperature in a sense. No motion means 0 degrees Kelvin temperature, which is allowed classically but not in quantum mechanics : there is always a basic state of motion.

Physicists have postulated the big bang where everything started billions of years ago with a lot of energy, a lot of it kinetic, so there never was a time when an "atom" had no kinetic energy, i.e. motion. It was created in motion.

One of course should remember that temperature is an average over billions of atoms, and some could be very close to motionless, in their collisions with each other, according to very well defined statistical curves.

Atoms are seen everyday by tunnel microscopes, atomic force microscopes, and even electron microscopes nowadays...

Before, atoms could be observed individually in mass spectroscopes, or through their impact noise on a microphone, and in thousands of other cases, but the first time we could see a nice arrangement of atoms, on a picture observable with one's eyes, was with a tunnel microscope. No that long ago.

Matter and energy were created hot during the big bang and underwent exothermic transformations later, but only 3K remain of that heat. Higher temperatures are created by accretion and nuclear reactions, and some less efficient processes like chemical reactions.

Some movement also exists without heat, because atoms can't stay immobile, as a direct consequence of quantum mechanics. If your system is small enough that its energy quanta are bigger than the temperature, it makes a perfectly observable difference.

Example: the vibration of water molecules isn't excited (or very little) by heat at room temperature, but only by quantum effect (call it delocalization). Deuterium atoms are heavier than normal hydrogen and vibrate less, and this allows stronger forces between molecules, explaining why D2O freezes a few °C eralier than H2O.

"Atoms are seen everyday by tunnel microscopes, atomic force microscopes, and even electron microscopes nowadays..."

Not really seeing, is it? The image you see has passed through many levels of interpretation, many more than a camera picture, and a picture from a camera is not "seeing". I agree that it is economical and practical to call it "seeing", and an extra reason to believe in the existance of atoms and molecules, but that is not what the word " see" really means.