Kinds of lasers
So you want laser guns, and you also want to know how they work. One place to start is by looking at how people have already got lasers to work. There are a lot of very clever people in this world, and they’ve tried a lot of inventive ways to get lasers to do their thing. I’ll cover some of the more well known and promising high powered laser technologies here.  Note that if you have some futuristic society where laser pistols are effective sidearms, they will have developed some new technology not on this list. We’ve got some pretty good lasers these days - but not that good.
Carbon dioxide lasers
CO2 lasers were one of the first practical high power lasers. As you might have guessed from the name, they use carbon dioxide gas as their laser medium, typically excited into lasing by an electric discharge although the very high powered versions work by being compressed to a high pressure and then allowed to expand to a low pressure in the optical cavity.  The carbon dioxide in one of these lasers might not cause much global warming, but it sure can cause a lot of local warming! They are useful in industry because they are relatively cheap and efficient, commonly turning upwards of 20% of the supplied electricity into laser light. However, they emit beams in the long-wave infrared. This is problematic for using them as any kind of weapon, because those long wavelengths don’t focus very well at typical combat ranges. The long wavelength light is also easily absorbed by plasma, and at high intensity it is all too easy to get a runaway cascade ionization going that stops your beam. So while they might be used to weld fighter jets together, CO2 lasers are not the best choice for burning fighter jets out of the skies.
Back in the 20th century, people were trying to figure out how to make really high powered lasers. Rocket engines are really high powered, right? So how about taking a rocket engine, burn some special chemicals as fuel in the rocket, run an optical cavity across the nozzle, and use that energized burned fuel shooting out for the laser? And you know what? It worked.  You get a nightmare of tubes and plumbing like any rocket engine, tanks of highly volatile toxic corrosive and flammable chemicals, lots of noise, flaming jets of toxic corrosive exhaust, and megawatt beams of poorly focused and poorly focusable laser death. Some of these lasers were used as proving grounds to advance important aspects of laser technology or exploring its applications    , but no one wanted to deal with the chemicals. So they were abandoned as soon as possible in favor of electric lasers.
The two most promising chemical lasers from a “blow all the bad guys up” perspective were the deuterium fluoride (DF) laser, which emitted light in the mid-wave infrared part of the spectrum at 3.8 μm in the atmospheric transparency window, and the chemical oxygen-iodine laser (COIL), which produced near infrared light at 1.315 μm wavelength.
Semiconductors do this weird thing where they conduct electricity both with electrons and with the lack of electrons. A particle of a lack-of-electron is called a hole, and you can think of it as a missing molecular bond if you wish, and the missing bond can move around the crystal by borrowing the electron of a neighboring chemical bond - moving the hole to that chemical bond that was borrowed from and repairing the previously broken bond  . This isn’t the place for a lecture in solid state physics, but the main idea is that in a diode you mix an electron current with a hole current and when an electron falls into a hole, it takes the place of that missing molecular bond and can emit a particle of light in the process. If you do this inside an optical cavity, you can get lasing.
Now during the 20th century we got very good at working with semiconductors. So perhaps it is not surprising that we can make very nice diode lasers. They’re tiny little things, perhaps the size of a grain of rice, and sometimes packed together in parallel into large bricks. By adjusting what we dope the semiconductor with and how we layer and arrange the semiconductor layers, we can get beam colors ranging from ultraviolet-C to far infrared. The efficiency can be crazy high - 60% electricity to light or better. They are cheap and robust and found all over the place in modern consumer electronics.  
For making laser death beams, however, they have one annoying limit. You can get high powers out of them, but when you do so they start to lase on all kinds of different modes and their beam quality goes to hell. If you can’t focus them better than a flashlight, you’re not going to be using them to burn your enemies out of the sky.
The usual way to get around this is to use cheap, highly efficient diode lasers to pump other kinds of lasers that need light to get their laser action going. That way we can use a fiber laser coil, for example, to convert poorly focusable diode laser light into extremely focusable light from the fiber laser.
Recent advances suggest it may be possible to make high powered diode lasers that produce a single, clean, well defined mode of output light.    This could allow laser death rays to be made out of diode lasers alone, eliminating a conversion step, increasing the efficiency, and making the laser more compact, cheaper, and more robust. Laser diodes can be made to operate over a wide range of wavelengths, by choosing the right material, ranging across the infrared on the low end to 193 nm ultraviolet light for aluminum nitride
Solid state lasers
The first laser ever made was a solid state laser. It was made with a ruby crystal. While it might seem neat to use lasers with gems as the laser generator, there were better options. Today, solid state lasers are made with slabs or rods of garnet crystal called yttrium-aluminum-garnet doped with the rare earth element neodymium, abbreviated Nd:YAG  . Early Nd:YAG lasers were pumped with xenon flash lamps, and had abysmal efficiency (around 1% or less). Then mankind invented semiconductor technology, made laser diodes, and used the laser diodes to pump the garnet crystals instead  . Now the efficiency got to higher than 30%. The Nd:YAG turned out to be excellent at taking badly focused diode laser light and efficiently turning it into near diffraction limited focused laser light at 1.064 μm wavelength in the near infrared. You could even shoot the beam through a nonlinear optical crystal and upconvert two 1.064 μm photons into a single 0.532 μm photon, and get green light at 80% conversion efficiency. You can even get higher harmonic conversion to 0.355 μm, 0.266 μm, and 0.213 μm wavelengths in the near ultraviolet.
Meanwhile, a different kind of solid state laser was gaining popularity. If you dope a sapphire crystal with titanium, you can get it to lase on a very wide band of colors spanning the near infrared and even a bit into the red part of the spectrum. This ends up being useful because a single frequency technically only defines an infinite duration wave. In order to make a waveform that starts, operates for a while, and then stops you need to broaden that spectrum a bit, mixing in waves with slightly different frequencies that add up together at times when the beam is on but cancel out when the beam is off. And the shorter the pulse of the beam, the wider the range of frequencies need to be in that pulse to get it short enough. Normally you can ignore this effect, but if you want to get very short pulses - picoseconds or femtoseconds long - the spread in frequency starts to be important. And if your laser can’t amplify those frequencies you can’t get such short pulses. So titanium sapphire lasers were used to make these incredibly short laser pulses. By compressing what would normally be a fairly moderate amount of energy into crazy-short time spans, titanium sapphire lasers could reach powers and intensities that were off the charts. They don’t give pulses that have as much energy as a Nd:YAG laser, and their average power is overall lower, but for instantaneous power during their pulse they can’t be beat.
Soon, Nd:YAG lasers became the workhorses for just about any application that needed a high energy laser; and Ti-saphhire lasers were in common use for producing ultra-short, extreme power pulses. They became common in medicine, machining, science, and all sorts of other fields. Many were investigated for military laser weapons. They could direct tens of kilowatts of infrared death onto incoming missiles and mortar shells and other flying things     . The main annoyance was that with high power came heat, and heat warped and expanded the crystals, which degraded the quality of the beam. Various clever designs were used to cool the crystals. But eventually they were replaced by …
Way back in the good old days of the 20th century, the telecommunications industry discovered that they could send long distance signals better using laser pulses down optical fibers than they could over copper cables. And so a multi-billion dollar industry poured huge amounts of money into developing all the technology around these new-fangled optical fibers. One thing they tried was doping the fiber material with rare earth elements that could undergo lasing. If you shine a diode laser into the fiber, the diode light is confined to inside the fiber where it very efficiently couples to the rare earth elements doping the fiber. These dopants then begin to lase. A spool of fiber may be as thin as a hair, but run for kilometers. This gives plenty of room for the initial light to be amplified, and filters out annoying side modes that can’t be focused very well. It also gives an incredible surface area for shedding heat. And the fibers are flexible - they’re not going to crack on you like a Nd:YAG crystal. Soon, people were pulling kilowatts of power out of fibers. And then the manufacturing industry stood up and took notice, and started replacing their Nd:YAG lasers with cheaper, simpler, more robust, more compact, more efficient fiber lasers with better beam quality. By the 2020’s, continuous beams of up to 30 kW were being produced from a single fiber. Efficiencies on the order of 40% are common and around 50% have been reported, and there are ways of pulsing the fibers to get even higher momentary powers. Wavelength is generally somewhere in the short-wave to near infrared, depending on the rare earth dopant used and what modes are selected. 
Needless to say, the military watched these developments with some interest. In the early 2000’s, power levels from individual fibers weren’t good enough to do much - only a few kW per fiber. The first thing they tried was just bundling a whole bunch of fibers together and letting them shoot their beams out side by side. This is a terrible way to do things, the beam quality is going to be awful and it was, but it also was able to shoot down rockets and mortar shells. Then the telecommunications industry came in again. They had been developing ways to send multiple signals at slightly different wavelengths down the same fiber. Turns out, the same trick could be used to combine laser beams. And so fiber lasers are being built at up to hundreds of kilowatts. These are the first true laser weapons to enter service (as opposed to experimental platforms). We still don’t know how they will perform against enemy action, but today ships and trucks are going around armed with laser guns made out of fiber lasers.     
An excimer laser does some crazy stuff to get forbidden chemistry to happen between a halogen and a noble gas, and uses that unholy spawn of a reaction product to produce ultraviolet laser light. Excimer lasers can produce fairly high powered pulses of ultraviolet light, usually in the near ultraviolet to near vacuum ultraviolet (to as short as 126 nm wavelength). Their efficiency isn’t terrible - modern ones can get around 10% conversion of electricity to laser light  - but not as good as solid state or fiber lasers. If you are looking for a modern technology that can explain ultraviolet lasers, excimers might be the lasers for you.  
Excimers are usually forced to undergo their blasphemous reactions using intense pulses of electricity or electron beams, making them inherently pulsed. However, there has been some research on directly pumping an excimer gas mixture with the ionizing radiation from nuclear reactions. In principle, you could put your laser inside your nuclear reactor and run it without electricity at all!  
Free electron lasers
Some purists might argue that a free electron laser (or FEL) isn’t technically a laser at all. It does amplify light, and produces coherent monochromatic and highly directional beams like any other laser. But the world laser itself is an acronym for Light Amplification by Stimulated Emission of Radiation. All the other kinds of lasers described here use stimulated emission to get their beams. Free electron lasers don’t. Instead, they use a particle accelerator to shoot a beam of energetic and relativistic electrons through a series of magnets. This produces a kind of instability that makes the electrons shed their energy as light, and then the light acts on the electrons to bunch the electrons together so they emit their light in phase. And now you get light output that acts like a laser without any actual lasing being involved. 
In addition to annoying pedants, FELs have other benefits. They can be tuned to produce light of nearly any wavelength. FELs have operated across the spectral range from microwaves to x-rays. Any energy in the electron beams that is not turned into light remains in the electron beam, allowing it to be recycled or used again. And in principle they can be very high power. These are one of the contenders for multi-megawatt or multi-tens-of-megawatt lasers that the U.S. Navy looked into.   
An FEL can extract something like 10% of the energy of an electron beam in a single pass through its resonator   . However, all the unused energy stays in the particle beam. So you can, for example, turn the beam around and run it backwards through your particle accelerator to pump up the RF fields in the cavities for the next electron pulse, recycling all that energy  , although this only really works for linac acclerators. If you can keep your electron bunches from spreading out over time, you can also recirculate them in a synchrotron. So in principle you can extract nearly all of the electrical energy you pump in as laser energy coming out (although reference  describes many of the other losses you can expect, from running cryogenic compressors to overhead for building lighting and computers).
However, a 10% extraction efficiency per pass assumes that you can run the FEL inside an optical cavity. Currently, x-ray FELs operate at a much lower extraction efficiency because it is hard to make x-ray resonant cavities. So if you can't solve this problem, while you might still have a high wallplug efficiency for your laser you will have a much higher circulating particle beam power than x-ray beam output. This raises the question of why not just shoot the electron beam at your enemies? you will be able to dump a couple orders of magnitude more power into them. But - new technologies have been proposed for x-ray resonant cavities. If these pan out, you'll be able to get high efficiency and high power x-ray beams for your war spacecraft.
Their main limitation is that they are big. Those electron accelerators don’t come small. You might fit one into a naval vessel or jumbo jet. A number of new technologies for drastically shrinking electron beams are being developed, but these are also likely to take a hit on efficiency as well. That’s not to say that traditional electron linac technology can’t be made even more compact. There’s been a considerable amount of progress made on that as well. But in the foreseeable future you’re still probably going to be looking at a really big device.
Science fiction lasers
We can’t predict what the laser of the future will be. It might be some development of any of the previously mentioned lasers. Or maybe it will be a completely new technology not yet dreamed up by human minds.
If you want to include advanced laser weapons in your science fiction world, you don’t need to figure out all the details (after all, if you could you could also patent it, sell the technology to Boeing or Raytheon, and retire with more money than you know what to do with). But if you want a consistent world you should figure out some details about their performance. For example:
- What color beams do they emit?
- What is their specific power (beam power emitted for a given mass of the system)?
- What is their specific energy (pulse energy emitted for a given mass of the system)?
- What is the specific cost (price for a given mass)?
- Can you get diffraction limited performance?
- How heavy and expensive are your focusing optics?
- How efficient is the laser at turning input energy (either electricity or something else) into beam power?
- What is the specific power and energy of the energy storage systems (batteries and the like) available to you?
- Are there different varieties that give you different trade-offs?
- Can the technology be made robust or do you always need to be careful not to break delicate components or throw the optical alignment out?
Author: Luke Campbell
- Clifford L. Laurence, “The Laser Book: A new technology of Light”, Prentice Hall Press, New York (1986)
- “Carbon Dioxide Laser – How Does CO2 Laser Work?”
- M. C. Lin, M. E. Umstead, and N. Djeu, "Chemical Lasers", Ann. Rev. Phys. Chem. 1983. 34:557-591
- [ https://www.globalsecurity.org/space/systems/miracl.htm GlobalSecurity.org, “Mid-Infrared Advanced Chemical Laser (MIRACL)”]
- Defense Advanced Research Projects Agency, “MIRACL”
- Northrop Grumman, “Chemical High-Energy laser Systems”
- William G. Schulz, “Building a Chemical Laser Weapon: Under fire, Airborne Laser program director confronts challenges of revolutionary weapons system”, Chemical & Engineering News
- Neil W. Ashcroft and N. David Mermin, "Solid State Physics", Saunders College Publishing (1976)
- Laserline, “Actually, what is a diode laser?”
- RP Photonics Encyclopedia, “Laser Diodes”
- New single-mode semiconductor laser delivers power with scalability
- Contractor, R., Noh, W., Redjem, W. et al. Scalable single-mode surface emitting laser via open-Dirac singularities. Nature (2022). https://doi.org/10.1038/s41586-022-05021-4
- Semiconductor Lasers Hit Steel-Slicing Levels: PCSEL devices now blast brawny beams comparable to bulkier laser systems, IEEE Spectrum
- Circuit Globe, “Nd:YAG Laser”
- “Laser Machining Processes”
- Defense Science Board Task Force on Directed Energy Weapons, December 2007
- Jason D. Ellis, “Directed-Energy Weapons: Promise and Prospects”, April 2015
- “Directed Energy Futures 2060: Visions for the next 40 years of U.S. Department of Defense Directed Energy Technologies”, Public Affairs release approval # AFRL-2021-1152
- Ronald O’Rourke, “Navy Shipboard Lasers for Surface, Air, and Missile Defense: Background and Issues for Congress”, Congressional Research Service, CRS Report for Congress, December 9, 2010
- Alphanov, “Tutorial: fiber laser basics”
- Jeff Hecht, “Fiber lasers mean Ray Guns are Coming”, IEEE Spectrum, 27 March 2018
- S. P. Obenschain et al., “Direct drive with the argon fluoride laser as a path to high fusion gain with sub-megajoule laser energy”, Philos Trans A Math Phys Eng Sci. 2020 Nov 13; 378(2184): 20200031.
- M.C. Rao, “A Brief Introduction to Excimer Lasers: Fundamental Study”, International Journal of Advances in Pharmacy, Biology and Chemistry, Vol. 2(3), pp. 533-536, Jul-Sep, 2013
- RP Photonics Encyclopedia, “Excimer Lasers”
- “Nuclear-Pumped Lasers”, NASA Conference Publication 2107, Proceedings of a workshop held at NASA Langley Research Center, Hampton, Virginia, July 25-26, 1979
- S.P. Melnikov, A.A. Sinyanskii, A.N. Sizov, and G.H. Miley, “Lasers with Nuclear Pumping”
- Peter Schmüser, “FEL Theory for Pedestrians”
- Roy Whitney, David Douglas, and George Neil, “Airborne megawatt class free-electron laser for defense and security”, SPIE Proceedings Volume 5792, Laser Source and System Technology for Defense and Security; (2005) 
- Heishun Zen, Hideaki Ohgaki, and Ryoichi Hajima, "Record high extraction efficiency of free electron laser oscillator", Applied Physics Express 13, 102007 (2020) 
- Physics & Applications of High Efficiency Free-Electron Lasers Workshop April 11-13, 2018 at the UCLA California NanoSystem Institute
- C. Emma, K. Fang, J. Wu, and C. Pellegrini, "High efficiency, multiterawatt x-ray free electron lasers", Phys. Rev. Accel. Beams 19, 020705 – Published 26 February 2016
- G. A. Krafft, "ENERGY RECOVERED LINACS", Proceedings of LINAC08, Victoria, BC, Canada, 688-692
- Glennda Chui, "Researchers take important step toward developing cavity-based X-ray laser technology"
- Margraf, R., Robles, R., Halavanau, A. et al. "Low-loss stable storage of 1.2 Å X-ray pulses in a 14 m Bragg cavity". Nat. Photon. (2023)