Have we managed to make a device without any atom inside it on earth? I was reading about vacuum here, and I found in the examples part here that even on the best man made vacuum devices, there are still millions of "molecules per $cm^3$". So my other question here is, what are these molecules and what is the mass of them?
As zonk said, there is no perfect vacuum. Even the 'vacuum' of space contains a few atoms per cubic meter on average.
In the lab, the lack of a high vacuum usually results from not having a pump that can effectively extract enough of the particles inside the chamber you're trying to evacuate.
There are several different kinds of pumps used, depending on how 'good' of a vacuum you want: mechanical roughing pumps, ion pumps, turbo-molecular pumps among others. Each works either by transferring gas particles or by capturing them. Since transfer works by preferentially forcing gas molecules in a certain direction, it cannot remove ALL the particles (as in the electrons inside a wire: on average, you can force the current preferentially in the direction of the applied voltage but each individual electron will zip around in all directions). A capture pump is better at removing particles, but it is still limited by how many materials they have already captured and can saturate.
Even if we had a perfect pump, you would still need to be careful about what is inside the chamber or what the chamber is made of. Fingerprint grease, zinc (and brass), and plastics actually out-gas under high enough vacuums. This means that the particles normally trapped under atmospheric pressures can break loose and rattle around inside your vacuum if you're not careful.
Also, pumps such as mechanical pumps use lubricating oil to function (though some fancier ones such as the ion pumps do not) and that can get into the chamber as well and increase the number of particles.
That's probably more than you wanted to know, but there you have it.
I do not think the tag 'particle physics' is accurate. Particle physics generally refers to subatomic physics and encompasses things like quarks, QED, and the Higgs boson. Perhaps 'experimental physics' would be better?
(Adding another answer as my response to Hurricane's question is too long for comments)
Glad it helped. Richard Terrett is correct, (charged) anti-matter is confined in a magnetic trap in as high of a vacuum as we can get. Uncharged anti-matter must be contained using laser traps ('optical tweezers' is something to look into if you're curious).
There will still be some particles in the antimatter container but, contrary to the DaVinci Code and popular perception, when one particle of anti-hydrogen meets a particle of hydrogen, the energy released (according to E=mc^2, which I believe is correct) is 3*10^-10 Joules or 0.3 nanojoules.
To give this a sense of scale, Wikipedia says a nanojoule (10^-9 Joules) is equal to 'about 1/160 of the kinetic energy of a flying mosquito'. So an antihydrogen annihilating with a stray hydrogen in the vessel would be have less of an effect than a mosquito bumping the outside of the container, energetically speaking.
Now, we can also determine how many particles would be in a container. Let's take the highest vacuum rating; ultra-high vacuum, which is classified as having a pressure under 10^-7 pascal, with 10^-10 pascal (10^-12 torr) being the golden standard among people who care about excellent vacuum vessels. A lot of UHV systems operate at very low temperatures, but let us take the temperature of the inside of the vessel to be room-temperature, T=293 Kelvin. We can use the ideal gas equation if we know the volume of the container, PV=nRT. Plugging in the pressure, room temperature, and assuming a container size of a rubik's cube [V=(5.5*10^-2 meters)^3=1.66*10^-4 meters^3], we arrive at a number of particles of n=6.83*10^-18 moles or 4.113*10^6 molecules. To put this in perspective, at standard pressure and temperature in the same size vessel (STP, 101325 pascal, 293K), we would have 4.155×10^21 molecules. So UHV is a pretty damn good vacuum.
The current record vacuum, made at CERN, is around 1000atoms/cubic cm
For comparison intersteller space (well away from the solar system) is around 1 atom/cc and deep intergalactic, really empty, space is around 0.001 atom/cc
The bits of space we can reach, in Earth orbit, are only about the sort of vacuum you can get in an ordinary lab vac pump.
I'd like to add another dimension to this discussion. Even if we were to hypothetically evacuate all atoms from a certain region of space, quantum field theory tells us that there would still be something there, namely the "zero point" energy corresponding to the ground state of fields. So in this sense, at least as far as I am aware, our current model of fundamental physics renders the existence of a region of space with "nothing" there an impossibility, even in principle. http://en.wikipedia.org/wiki/Zero-point_energy