So what is a Laser, really?
J. L. Doty
copyright © 2009–2016 by J. L. Doty
released: June 10, 2016
The acronym LASER stands for Light Amplification by Stimulated Emission of Radiation. Simply put, a laser is just a light bulb, but less, and more. An incandescent bulb generates light by using electricity to heat up a tungsten filament, not unlike heating a piece of metal until it’s red hot. But a tungsten filament is much hotter, so it radiates at much shorter wavelengths. That’s why the light it emits appears almost white. In a sense, a laser is nothing more than a light bulb, because both generate light (electromagnetic radiation), and once the light leaves either type of source, it’s just light.
To understand how a laser works, consider the analogy of a speaker system as depicted below:
Figure 1. Speaker system
Here we have a microphone, amplifier and speaker. Sound goes into the microphone where it is converted to an electrical signal, which is amplified by the amplifier, then converted back to even more sound by the speaker. Now we all know what happens if we do something like this:
Figure 2. Acoustic oscillator
We get a very loud, screeching noise that can be quite irritating. From an engineering standpoint, this is a positive feedback loop that’s oscillating in an uncontrolled fashion: sound goes into the microphone, more sound comes out of the speaker and goes into the microphone, even more sound comes out of the speaker, ad infinitum. Positive feedback is a good thing in rearing children and in interpersonal relationships, but in engineering it can be disastrous. On the other hand, electronic oscillators are used to generate the timing in wrist watches, and there are several in a modern PC to control timing, processor events, memory access, etc. So sometimes we want an oscillator, but we need to control it carefully.
There are two things wrong with the uncontrolled acoustic oscillator above:
- it’s irritatingly loud, a problem that can be solved easily by limiting the output of the amplifier, and
- the output is loaded with harmonics.
The figure below depicts two clean, pure waveforms.
Figure 3. Pure waveforms, fundamental and 3rd harmonic.
The upper waveform is referred to as the fundamental frequency, while the lower has a frequency three times that of the fundamental, and a wavelength one third that of the fundamental, so the lower one is referred to as the third harmonic. When the two are combined, we get:
Figure 4. Combined fundamental and 3rd harmonic.
This waveform contains a combination of the fundamental and the third harmonic. In fact, the screeching sound coming out of the speaker contains the fundamental and a whole series of odd harmonics: 3rd, 5th, 7th, etc. It’s these harmonics that make the difference between the pure tone one hears from a tuning fork, and that emitted by our uncontrolled oscillator. So for a nicely controlled oscillator, we must limit the frequencies at which the system is allowed to oscillate.
A laser is simply an electromagnetic oscillator, but where our uncontrolled acoustic oscillator is emitting sound frequencies in the range of 20–20,000 Hz, a laser emits light frequencies in the neighborhood of 600,000,000,000,000 Hz, or 6 x 1014 Hz.
To make a laser we need to induce positive feedback in an optical amplifier, just as we induced positive feedback in the acoustically amplified system. The figure below illustrates the concept of an optical amplifier:
Figure 5. Conceptual optical amplifier
Optical amplifiers are bidirectional, meaning that when light enters either end, more light comes out the other. Feedback is created by putting mirrors at both ends of the amplifier to reflect the amplified light back into it, as in the illustration below.
Figure 6. Conceptual optical oscillator (laser).
The mirrors are a resonant cavity for electromagnetic radiation at optical frequencies. The pipes on a pipe organ are a simple example of acoustic resonant cavities. The large pipes resonate at lower frequencies, and the small pipes at higher frequencies, limiting the frequencies that can be emitted by any single cavity. The mirrors in a laser do the same, and are an important element in controlling the oscillation frequency of the device.
If both mirrors are fully reflecting, then all of the light is trapped inside the laser cavity and there is none coming out of it to make use of. Hence, we make one of the mirrors partially-reflecting so we can tap some of the energy out of the cavity. We laser geeks have really fancy names for these mirrors, like rear mirror and front mirror. The front mirror is also referred to as the output coupler.
Note that the two mirrors must be aligned parallel to each other to within a tiny fraction of an arc-second, and if they drift about with changes in room temperature, or other environmental factors, the laser won’t operate, or its output power will fluctuate wildly. This mechanical constraint is almost always a serious factor in the cost of a commercial laser system.
Interestingly enough, the optical amplifier by itself produces Light Amplification by Stimulated Emission of Radiation, but without the mirrors for feedback we don’t have oscillation, which means we don’t have the device we call a laser. Hence, a more appropriate acronym might be Light Oscillation by Stimulated Emission of Radiation, or
I’m trying to picture myself at the pentagon standing in front of a bunch of admirals, generals and senators making a pitch for funding, and one of them asks, “Now what is it you said you’re going to build, Dr. Doty.”
And I respond with, “Why, with this funding, I’m going to build you a bunch of losers.”
Even us laser geeks didn’t need the marketing department to tell us that pitch won’t sell.
On the subject of being acronymically correct, the term infrared laser is disjoint since the “L” in laser refers to light, which is implicitly visible radiation, so the words infrared and light are mutually exclusive. To be acronymically correct, we should call it an IRaser, and an ultraviolet laser would be called a UVaser. Or, to be truly acronymically correct, we would use IRoser and UVoser. But that’s just too complicated, so even us laser geeks don’t bother.
The figure below illustrates the configuration of a simple neon sign.
Figure 7. Neon sign configuration
In its simplest form a neon sign is a sealed glass tube containing neon gas at low pressure, with cathodes at either end, and a high-voltage power supply. Once a plasma is sparked and the gas is ionized, free electrons travel from the negative cathode to the positive at very high velocity. The electron configuration—essentially the electron orbits—of the neon molecules are initially at the lowest energy level possible, which is referred to as the ground state. When an electron smashes into an ionized neon molecule it gives up a tiny amount of its kinetic energy to the molecule, which absorbs the energy by elevating the molecule’s electron configuration to a higher energy state. This is referred to as the excited state.
Note that the means of elevating a large number of molecules to an excited state is called the pumping mechanism. The means described here for the HeNe laser is electrical pumping. Some lasers use optical pumping where an intense pulse of white light is directed into the cavity from the sides. And the largest weapons lasers use gas dynamic pumping, which is described in more detail in the article The Laser Weapons Myth.
Here’s where quantum mechanics kicks in: the possible energy states that a molecule can occupy are quantized and discrete. The electrons can’t inhabit just any configuration; they must jump to a very specific level, or not jump at all.
Excited states are unstable, and after a brief period of time—nanoseconds to milliseconds, depending upon the type of molecule—it spontaneously decays back to the ground state, giving up the energy it absorbed in the form of a photon of light. The frequency (color) of that photon is proportional to the difference in energy between the two states, and the direction it travels is random. This is called spontaneous emission, and the light we see from a neon sign is just all the millions of photons being spontaneously emitted.
In the early part of the 20th century Albert Einstein proposed the concept of stimulated emission. He theorized that while a molecule is in an elevated energy state, if a photon collides with it before it spontaneously decays, and that photon has a frequency that exactly matches the frequency determined by the difference in energy levels, then the collision will stimulate the molecule to release its energy and emit the photon. He further theorized that:
- the original photon will not be deflected by the collision, and will continue on in the same direction;
- The emitted photon will:
- travel in the same direction as the original photon,
- have exactly the same frequency (wavelength), and
- be in phase with the original.
One photon in, two photons out; we have an optical amplifier. So now let’s modify the simple neon sign configuration so it can be used in a laser.
Figure 8. A real HeNe laser configuration.
Here, the tube configuration has been changed so windows can be added to either end, the gas is changed to a combination of He and Ne, and we’ve added the mirrors.
That’s really all it takes to make a laser. In fact, as an undergraduate electrical engineering student in the late 60’s we made our own lasers because the cheapest commercially available HeNe laser cost about $6,000, and that’s in late 60’s dollars, which is about $20,000 in today’s dollars. We bought cheap neon sign cathodes; got a whole bag of them for a few dollars. We had to learn a little glass blowing and welding. We cut up fused silica microscope slides for the windows on the ends of the tube. And since we were electrical engineers, we designed and built the power supply ourselves. The mirrors were more complex, but the commercial laser manufacturer sold them to us at cost. They weren’t losing any market share because we couldn’t afford to buy their lasers anyway, and they were helping to foster the next generation of laser engineers, so they considered it a good investment on their part.
We had a complete production line going, with an abundance of lasers for experiments. They worked well in that they produced excellent laser output, but they were really crappy lasers because we had to constantly tweak the mirrors to realign them, and we used epoxy to glue the windows on the ends of the tube. Epoxy outgasses a number of volatile elements that contaminate the laser gas, so they only had a lifetime of 800 to 1000 hours, whereas 10,000 hours is a minimum for a commercial device. But when the output of a laser began to decline, we simply opened up the tube, pumped it out, refreshed the gas, resealed it, and got another 800 hours out of it.
To understand what makes the laser so unique, let’s compare it to an incandescent light bulb.
Figure 9. Source characteristics of an incandescent bulb.
The light bulb on the left in the above figure illustrates how every point on the surface of a bulb is an infinitesimal point source of light. That means that the bulb consists of an infinite number of infinitesimal point sources. In optical engineering we refer to that as a distributed source. Unfortunately, in most optical experiments we need a single, intense point source. Prior to the laser we achieved that by placing a small aperture in front of a distributed source, as is illustrated on the right side of the above figure. Apertures of this kind are usually on the order of about 10–30 µm in diameter and are quite expensive.
When using a distributed source a scientist must play a balancing act between the size of the aperture and the amount of measurable energy passing through it. Ideally, the aperture will be infinitesimally small, but that means the amount of energy available for experimentation is infinitesimally small as well, and our signal-to-noise ratio is in the toilet. Then along came the laser.
Figure 10. Laser output focused to an infinitesimal spot.
An important characteristic of a laser is that the rays in the output beam are highly parallel. And with modern design programs and fabrication techniques we can make lenses that are immeasurably close to perfect. So with the advent of the laser, experimenters finally had true infinitesimal spots with unheard of amounts of energy for measurements.
In this case, one might say that a light bulb is to a laser as an axe is to a scalpel.
Another important characteristic is the spectral character of a laser. The figure below compares the spectral output of an incandescent bulb to that of a Nd:YAG (Neodymium-doped: Yttrium-Aluminum-Garnet) laser, a workhorse laser used extensively in industry and research. It also includes red, green and blue lines to indicate the location of the visible portion of the spectrum.
Figure 11. Comparison of laser and bulb spectral characteristics.
The bulb emits a small amount of ultraviolet light, then a bit of blue, more green, more red, and it finally peaks in the near-infrared. It then trails off, emitting optical energy all the way into the mid-IR. This illustrates why incandescent bulbs are so inefficient. A 100 Watt light bulb produces about 5 Watts of visible light, and 95 Watts of heat. Also note that it produces much more red than blue, which is why such bulbs appear yellowish, and why the photodetector arrays in modern cameras must be carefully balanced to prevent photos from exhibiting a decidedly yellow cast.
We characterize the spectral output of optical sources with the Full Width at Half Maximum (FWHM) by measuring the width of the curve at 50% of the peak. The incandescent bulb has an FWHM of 1200 nm, while the Nd:YAG laser has a FWHM of 0.5 nm. In fact, it’s so narrow that if truly drawn to scale in Figure 11, the line for the laser’s spectral characteristic would be so thin it wouldn’t appear at all.
One might say that a light bulb is to a laser as a hammer-hitting-an-anvil is to a tuning fork.
In the figure below, the spectral output of the bulb has been replaced with that of the sun, with a blue curve demonstrating how one might use a narrow-band filter to pass only the laser output.
Figure 11. Comparison of laser and solar spectral characteristics.
Narrow-band filters are actually much narrower than the curve in the illustration implies. In fact, they’re so narrow that if you were to look through a visible laser filter directly at the sun at noon on a clear day in the Mojave Desert, you’d see nothing but absolute blackness. And yet such a filter passes almost all of the laser energy.
The figure below is a photo of a Reaper drone firing a Hellfire missile.
Figure12. Reaper drone firing a Hellfire missle
Note the small turret beneath the “chin” of the drone. It contains controls for both azimuth and elevation, optics for receiving visual images, and an IR laser for targeting. When launching a missile, the operator aims the laser at the target where it’s focused to a small spot, and keeps it locked there until the missile impacts the target. The missile’s guidance optics employ a narrow-band filter, like that previously described, to block out all light but that coming from the laser spot. Basically, the missile sees a black background with nothing present in the field of view but the spot, so it doesn’t have to attempt to distinguish the target from the other parts of the image. Systems like this are used on all types of military aircraft, as well as some ground based vehicles. This is how our military can guide a smart bomb or missile right through the door of a bunker. It’s the scalpel effect that allows the laser beam to focus to a small spot, and the tuning fork effect that allows the filter to eliminate all the optical background and pass only the laser light.