So far I’ve had a lot of posts describing the electronics of this project, but the real meat of the thing is the physics and optics. It’s time to start digging in.
Very Abbreviated Laser Theory
First, a very brief overview of what a laser is. LASER is an acronym for Light Amplification by Stimulated Emission of Radiation. The Stimulated Emission part is the hardest to understand, but probably the most important part. To understand it we need to talk a bit about atoms. I can use the Bohr model of an atom here as it’s accurate enough to understand what’s going on.
An atom is normally in what we call the “ground state”. It can become “excited” if it is given enough outside energy. An atom won’t stay in the excited state for long. When it drops back down to the ground state, the energy it loses can emit a photon of light. If you’ve ever seen a neon sign, you’ve seen this process in action. Electricity is used to excite neon atoms, and when they fall back to their ground state they emit visible light. That’s the “emission” part of the acronym.
Light coming out of a neon sign goes randomly in all directions. It’s “spontaneous” emission. If one of those excited atoms is hit by a photon emitted from a neighboring neon atom something very interesting happens: the excited neon atom releases its photon too (it’s “stimulated” by the first photon). This newly released photon has the same wavelength, phase and direction as the photon that hit it. In a laser, these are the majority of photons.
In order to get the majority of photons to be created by stimulated emission you need to have a lot of atoms in the excited state. In fact, you need more atoms excited than there are at the ground state. This is called a “population inversion” and until you get it you will never get a laser to work. There are two things required for a population inversion:
The material you’re using to create photons has to be able to hold an atom in its excited state for some time. Different laser materials have different “upper state lifetimes”. The longer the lifetime the easier it is to get a population inversion.
You need to be providing energy to the lasing material at a rate that’s faster than the atoms can drop back to their ground state. So a shorter upper state lifetime means more energy must be put into the system.
This isn’t a laser yet. With stimulated emission you’re now generating photons in phase with other photons, but they are still completely random. To fix this you need a resonator. By placing mirrors at either end of the lasing material and very carefully aligning them you reinforce photons traveling in the direction you want. Stimulated emission guarantees that these photons continue to grow in number. One of these mirrors should only be partially reflective so a small amount of light gets out. This is the exiting laser beam.
That’s the theory anyway and it’s pretty simple. I’ve left out a ton of complicated details and unfortunately, it’s these details that cause all the trouble.
Resonator Designs
The laser I’m making is a “solid state” laser. This means the lasing material (referred to as the gain medium) is a solid, usually a crystal. Solid state lasers get their input energy — their “pump” energy — optically. This needs to be a really bright source of light in order to maintain a population inversion. You can use an arc lamp, a flash tube, or another laser. I’m going to be using another laser because it is the most efficient. But why would you use a laser to create another laser? Beam quality. The laser I’m using to pump my gain medium is very bright — up to 80 watts of laser light. But it’s not really one laser. It’s 36 small laser diode chips all focused into a fiberoptic cable. The output of this cable at best will form a beam that diverges broadly after just a few feet.
There are two ways to pump a solid state laser crystal: side pumping and end pumping. With side pumping you use mirrored reflectors to try to illuminate the crystal as evenly and efficiently as possible. It takes some good optics design to make this work well and you typically buy laser diodes mounted in “donuts” that you can stack together:
Pump laser diode donut. The laser crystal goes through the center hole. The holes around the center are for water cooling.
Big lasers used for cutting metal usually use these things. You can channel a ton of light into the rod. The downside is that side pumping excites a lot of the crystal area, but only a small part of it in the center makes the laser go. The rest of the energy is wasted.
The other technique is end pumping. In this configuration the high reflector (HR) mirror is transparent to the frequency of light used by the pump laser. The laser can be a direct diode with a lens to focus it or it can be provided by an optical fiber. The advantage here is you can design the shape of this beam to overlap with the laser beam inside the crystal which makes things more efficient. You can also keep the pump diodes somewhere else so you don’t need to bring sophisticated cooling into the resonator. The downside is your mirror setup just got more complicated and expensive.
With end pumping there are also several types of resonators to choose from. The three that are common and can be reasonably implemented by a hobbyist are the Linear, L-Fold and Z-fold.
Linear Resonator
A linear resonator is the simplest. It has just two mirrors and goes in a straight line.
Linear cavity has just two mirrors.
This cavity design is very simple and many lasers use this design. It has some limitations that make it unsuitable for what I want to do.
L-Fold Resonator
The L-Fold resonator has one additional mirror that allows the cavity to be a bit longer can keep the curvature of the end mirror from interfering with the pump beam.
The L-Fold has one additional mirror to “fold” the beam at an angle.
The L-Fold provides a little more length in the cavity for other elements. It takes a little more work to align.
Z-Fold Resonator
The Z-Fold resonator adds an additional folding mirror.
The Z-Fold allows for pumping at both ends of the laser crystal.
There is more complexity to the Z-Fold cavity but the big advantage is you can pump the crystal from both ends. This helps prevent crystal damage at high powers. The alignment of a Z-Fold cavity is quite a bit more complicated.
That’s enough physics and background. I’ll fill in more as needed. Next up, getting the linear cavity to lase.