Atomic Frequency Standards

One of the wonders of the quantum mechanics that governs the world of atoms is the fact that every atom in the universe is exactly the same as every other one. Of course, oxygen atoms are different from nitrogen atoms, and not even all oxygen atoms are the same, since some of them have an extra neutron -- but when they are the same element and when they have the same number of neutrons they are absolutely identical. This means that if we could make a clock using atoms, it would be an absolute clock, the same everywhere.

Well, it turns out that we can make a clock using atoms. Why not? you may ask. Well, because most internal atomic vibrations die away very quickly. Here's why.

Most atomic resonances are due to electron motion, and the electrons are very light and the electron clouds are very stiff since the electrons all strongly repel each other. (The hardness of all matter is due to the electron clouds from neighboring atoms bumping into each other.)

Because they are light and stiff, the frequencies are very high, typically at optical frequencies, and the wavelength of the electromagnetic radiation (light) is not very much bigger than the atom.

If you make waves in a tub with your hand, it is easy to make waves that are the size of your hand.
But, try to make waves as big as your hand with your finger! Nothing happens! In order to make waves with your finger you find yourself beating very fast in order to effectively make small waves.

Because these atomic resonant frequencies are so high, their wavelengths are short, and the atoms couple strongly to the light. Thus the internal atomic vibrations die away after only a few thousand cycles, and so they don't make good clocks. (See How Does it Work -- Oscillators to see why a high Q factor is important to keeping time.)

What this means is that if we could find an atomic process that was not so stiff the frequency would be very much lower, the long light wavelength could not be excited by the small atom, and so the internal vibrations would not die away so quickly.

Atomic hyperfine transitions correspond to vibrations at much lower frequencies and they are due to a weak magnetic interaction between the electron cloud and the spin of the nucleus, the heavy atomic core which is made of protons and neutrons. Not all atoms have a net spin in the nucleus, and these are not candidates for atomic clocks. However, the wavelengths of hyperfine radiation range from 1 centimeter to 10 centimeters, lengths that are much bigger than an atom. Thus (remember the water waves?) these internal vibrations radiate their energy away very slowly, and so they have a very high Q. Instead of microseconds, calculated relaxation times for real atoms can be 10,000 years or longer. (10,000 years at 10 billion cycles per second is a lot of cycles.)

What this means is that these internal vibrations form almost perfect clocks. An atomic frequency standard must then:

Isolate the atoms so that their internal vibrations are not modified by unwanted effects such as:

Collisions with other atoms
Electric and magnetic fields
Thermally induced velocities (remember the twin paradox? This relativistic effect is measured every day by atomic frequency standards all around the world.)

Interrogate the internal vibrations in a way that does not upset their inherent accuracy.

The success of atomic standards such as JPL's Linear Ion Trap Standard (LITS) is such that time and frequency are today measured with far higher accuracy than any other physical quantity.

References