Trip the Light Fantastic

By John McCormick

ALL THE COLORS OF THE RAINBOW. Why the sky is blue. How to slow a light beam to a stop. Scientific advances often come when we look at some minor change and wonder how to explain it, or think about whether if it can be expanded upon.

Even moreso, children who don’t know better often ask what seems like a silly or obvious question such as “Why is the sky blue?” and thereby instantly flummox both the average parent and even scientists.

The simple answer that blue light scatters more than other frequencies really doesn’t satisfy most kids and it really shouldn’t since it actually makes no sense. Brighter children will go on to ask why light scatters and once that is explained or, more often, glossed over because few adults really know why it happens, they often again ask “So why is it blue?” They rightly feel they haven’t been given a sensible answer.

So, why is the sky blue? For example, if you look straight up, what light is being scattered? From what? Why don’t we see black, the “color” of outer space?

Given that the small amount of light being reflected by the atmosphere is obviously brighter than no-light of outer space, blue might make sense except for one glaring problem best illustrated by the simple acronym, Roy G Biv.

Most of you will instantly recognize this as the colors of the spectrum or a rainbow, with the letters standing for Red, Orange, Yellow, Green, Blue, Indigo, and Violet.

When you explain to a child that the sky is blue because blue bends more than the other colors of light, were you just repeating something you had been told or did you for some reason skip over the fact that if blue light bends (refracts) more than red, orange, yellow, and green, then indigo and violet should both bend more than blue?

So why is the sky blue instead of purple? Point a twelve dollar spectroscope at the sun and you will see that sunlight, the source of all the light impinging on our atmosphere, including the reflection from the moon, actually includes violet as well as the other colors of the rainbow.

About the time of the Civil War (or War between the States) John Tyndall showed that light is scattered in water that has small particles in suspension—for example, a tank of water with some milk mixed in. Therefore, the blue sky is technically said to be caused by the Tyndall effect. A bit later Lord Rayleigh (John William Strutt) explored this in more detail and showed that light is scattered inversely proportional to the fourth power of the wavelength; blue light is scattered more than red light by a factor of (700/400)4 =9.3789.

Tyndall scattering explains why the sky has some color rather than appearing white, but that still ignores the problem of violet light that has a wavelength of about 380 to 400 nm (nanometers) and therefore is deflected even more than the blue light.

The complete answer as to why the sky is blue comes in several parts which are actually easy enough to explain to a child.

First, although the sun’s light covers the entire spectrum except for small Fraunhofer absorption lines, not all frequencies are equally bright and the violet component of sunlight is considerably less intense than the blue or yellow.

(Fraunhofer lines are the black lines seen in spectra of sunlight. A few decades after Joseph von Fraunhofer described the lines, Robert Bunsen—the Bunsen who invented the burner we commonly use to heat elements to view their spectral lines—and a fellow physicist, Gustav Kirchhoff, used bright line emission spectra to identify new elements. Those pioneers of spectroscopy also showed that the dark lines in solar spectra corresponded to the emission lines of elements when heated, making it possible to analyze the composition of the sun and other stars.)

So violet is less intense to begin with, but violet is also absorbed in the upper atmosphere and even if it were absorbed by the same percentage amount as blue is absorbed, violet starts out with the disadvantage of being less intense.

Finally, and most importantly, the human eye is not very sensitive to violet light; the color receptor cones of the retina are actually more sensitive to about 555 yellow-green (the color of the African jungle canopy). Therefore, with blue and violet both scattered most because they are short wavelengths, the human eye would see the blue as brighter than all the rest.

Although it is technically correct that yellow-green is the most easily seen, the difference between yellow-green and blue-cyan sensitivity is only about five to ten percent, which is difficult for humans to differentiate. With long wavelengths the most bent, the question is whether we see blue or violet. Due to the very different sensitivities, we see blue.

The difference between blue-cyan and violet is about ten to fifteen times greater, not a few percent.

There are also different ways of measuring sensitivity to various frequencies. No two methods seem to completely agree, but all show that blue-cyan is most easily seen by the human eye.

According to the Rensselaer Polytechnic Institute chart, violet is far less visible. Looking at the logarithmic scale on the left shows there is very little difference between sensitivity to yellow-green and blue-cyan. Other charts additionally show that the blue sensors in the retina are more sensitive to blue than the red and green sensors are to their respective peak frequencies.

This trifecta of causes—refraction, less violet intensity, and human eyes being less sensitive to violet—finally explains why the noonday sky usually appears blue to humans. It might not to other creatures (or visiting aliens) depending on their color sensitivity.

You might ask why Mars has a red sky as seen by various probes looking up from the surface. The answer is depressingly simple; Mars is rusty and at times there is a lot of iron oxide in the atmosphere whipped up by tremendous winds. After an extended period of calm weather, the sky on Mars would also appear a very pale blue to us—although perhaps not to the hordes of Martians still hiding in their River View condos in the walls of the Valles Marineris rift valley.

Light Motif

The basic story gimmick in one of Terry Pratchett’s novels is the nature of light, which is described as not being something seen but something we see by. In “Feet of Clay,” the Patrician is being poisoned by arsenic in his candles and the plot turns on the fact we tend to ignore light. And it’s true, for the most part we do ignore light in the same manner we ignore time.

But a basic difference is that we all think we know what light is while no one who actually gives it serious thought, not even Stephen Hawking, believes he or she understands what time actually is.

I was really working on an article about variable time; the idea that time even in the same reference frame might speed up or slow down, but that article is taking more (ahem) time than I had expected, so in this issue I want to look at what we know and, more importantly, what we think we know about light, including the troubling questions children ask—which most adults find annoyingly difficult to answer in any real depth.

First, we all know that light moves at a constant speed “c.”

Next, we know that physicist James Clerk Maxwell showed light is a combination of electric and magnetic fields, sinusoidal in shape, perpendicular to each other —as shown in the most common diagrams describing light as an electromagnetic wave phenomenon. As the magnetic field collapses, the electric field grows, and vice versa.

Finally, we know that light is also, and at the same time, both a wave when measured in one way, and a discrete particle when measured another way.

Now, for the trick question, which of those three statements is incorrect?

Unfortunately for most people’s view of post-classical science, it is the first of those three factoids that is wrong. The speed of light itself, “c,” isn’t constant.

As stated above, that first statement is incomplete, light is only supposed to have the constant speed of “c” in a vacuum.

The problem isn’t with the speed of light itself but with the definition of vacuum. Quantum mechanics totally changed the meaning of vacuum. Due to quantum probability, no vacuum can ever be perfect.

Unlike the popular conception of outer space, which is devoid of matter except in large clumps of atomic or larger size, it turns out that at the quantum level a “perfect” vacuum in interstellar space is potentially still filled with charged particle pairs.

Recent papers by March Urban of the University of Paris-Sud, along with Gerd Leuchs and Luis L. Sanchez-Soto from the Max Planck Institute for the Physics of Light in Erlangen, Germany, explore this phenomenon.

In fact, measuring variations in the speed of light from the ultimate “c” allows physicists to estimate the number of electron-positron and quark-antiquark pairs in a given volume of space.

Leuchs and Sanchez-Soto discovered that combining variations in the speed of light with the value of impedance in a vacuum allows them to calculate the number of charged elementary particles in any given space if it can be measured using ultra-fast lasers capable of resolutions in the order of one-fifth the time it takes electrons to transfer between atoms.

That is an important application for the variation of light speeds in various vacuums.

Slow Light

Another example of something a child might question relates to the speed of light. After being told that light can be slowed down, causing the rainbow, or the blue sky, a curious child might ask, quite innocently, if light can be slowed by passing through air, slowed more by passing through air into water, and even more by passing from air through a glass prism, why can’t it be slowed much more, or even perhaps stopped completely and stored, ready for use when needed (the way Galadriel stored starlight in her gift to Frodo of a vial of light)?

Two decades ago I would have made light (so to speak) of this question and gone on to explain that as mediums get denser and denser they not only slow down light, they also become opaque and therefore absorb the light rather than store it. Also, even when slowed by a glass prism, light still travels at about 180,000 miles per second, so stopping light or even slowing it down to human attainable speeds such as that of a satellite in Earth orbit at say, 22,000 miles per hour, is completely impossible.

In giving this explanation, I along with virtually every other physicist in the world who would probably have given a similar answer, would have been wrong.

Some researchers ignored the fact that such a childish question about stopping light was utter nonsense and looked at the implications of the fact that light is slowed when it travels through various media such as glass.

Instead of saying that no transparent media can slow light very much, they set about developing a material that might significantly slow light.

It turns out that light can not only be slowed a great deal; it can be brought to a complete stop.

Lene Hau, a physicist at one of my alma maters, led this field of study more than a decade ago when she and her team slowed light from 186,282 mps to 38 mph by 1998. By late 2000 they had succeeded in stopping light completely and then restarting it. [Professor Hau, below right. Photo courtesy of Justin Ide/Harvard News Office.]

Remember, Einstein demonstrated that nothing carrying information could move faster than light in a vacuum. Relativity says nothing about light moving slower.

At 38 mph, the Harvard researchers had succeeded in slowing down light by a factor of 200 million times when just a few years (not decades, mind you) earlier no scientist would have considered this even remotely possible.

Their trick was to create a new state of matter by cooling down a near vacuum to one-millionth the temperature of outer space.

(Note: does this bring into question all the claims that red shift occurs because stars are moving away from us? If light can be brought to a complete stop in an Earth laboratory in just a few meters distance, is it not being slowed at all by passing through light centuries of “empty” space? Not only that, but wouldn’t it be slowed by different amounts in different regions depending on the density of the vacuum in that particular part of space?)

In fact, Hau’s apparatus was meters in size; the actual medium that stops light is only eight-thousandths of an inch thick.

The atoms, which are few and far between in the near total vacuum, are cooled to within a tiny fraction of minus 459.7 degrees F or absolute zero.

Other researchers elsewhere at Harvard independently stopped a beam of light at almost the same time.

But how did either group know they had actually stopped light? Light that isn’t moving can’t be seen, but by exciting the particles with a laser the original beam of light could be restarted and exit the apparatus at the original frequency and direction.

The second group of experimenters used a completely different method to create an exotic material, using a cloud of rubidium and helium atoms at higher than room temperature and normal pressures instead of incredibly low temperatures and low pressures, which are both difficult to produce.

If you could refine either or both of these methods to where they can slow light for relatively long periods, wouldn’t they essentially work just like a computer core memory did in the early days of computing? (Yes, I’m old enough that the computer I supervised for Wang Labs had actual core memory in that then new IBM 360-65.)

A cubic inch of this kind of quantum light memory would store more data than available with one of today’s supercomputers, and both storage and access of data (read/write times) would occur at quantum speeds.

This is one of the building blocks of so-called quantum computers, but isn’t related to the quantum processors being built today, which can barely multiply small whole numbers. Both technologies are at about the same primitive level of development and real world usefulness.

The ability to stop and restart light beams could be used as memory for conventional as well as quantum computers.

Today

Those original experiments circa 2000 were able to stop light for a tiny fraction of a second, on average about 1/1000 second, before the “trapping” mechanism was perturbed enough to either release or completely degrade the signal beam.

By the end of 2013, researchers had extended that period from a tiny fraction of a second to a minute, a million-fold improvement in just over a decade.

Thomas Halfmann at the Institute of Applied Physics of the Technische Universität Darmstadt and colleagues combined several techniques, including a crystal doped with electrically charged praseodymium atoms.

As with most similar efforts, the process involves two lasers, one to excite the atoms and the other to generate the “signal” light beam. When the excitation laser is turned off, the other beam is trapped in the crystal, i.e. slowed to a complete stop.

Shining the control laser on the atoms again frees up the original beam to continue. This can now be done for such an incredibly long time (light would normally travel more than 11 million miles—half the distance between Earth and Venus at closest approach) that the time involved no longer has to be measured with an oscilloscope; it can be done by an observer counting his or her own heartbeats, or with a simple stop watch. In other words, unlike most quantum events, this isn’t something seen only using extremely expensive and sensitive instruments, but something that happens at the macro scale of human senses.

In fact, this is a very dramatic event. The crystal is normally opaque and the control beam turns it transparent. Turning that beam off turns the crystal opaque again, trapping the signal beam until a second control beam turns the crystal transparent again, releasing the signal beam.

Being quantum phenomena, the how and why of how all this happens is not simple to explain, but essentially the signal beam is acting like a wave rather than a particle; it excites the praseodymium electrons to alter their spin (magnetic field). As Maxwell showed, light can be treated as an alternating electric and magnetic field. The electrons act as tiny magnets absorbing the light as a change in spin, and later releasing the original light when stimulated again by the control laser.

In essence, you can say the signal beam is trapped in the crystal lattice until the lattice is “shaken” by a second exposure to the excitation laser and the beam released.

A minute may seem like a brief time for a signal to be held if your aim is to use or to store information, but consider the fact that short-term memory storing calculations in progress need only store and release information over a few millionths of a second and even that relatively long period of time is as long as it is because of the speed limitations of semiconductor memory.

Scharnhorst Effect—Light Faster Than Light?

If light can be slowed to a complete stop, something completely unthinkable two decades ago, what about the other end of the velocity scale? Can light be caused to travel faster than the currently recorded speed of light?

Theoretically, getting light to move faster than “c” would not necessarily violate relativity since Einstein didn’t actually say “nothing” could move faster than light, and certainly not light itself; instead, it would simply reset “c” at a higher value.

What the Theory of Relativity actually forbids is superluminary signaling—that is, something can move faster than light as long as it doesn’t convey any information.

(You may be wondering how the relativistic speed limit fits with the quantum entanglement events that show action at a distance as nearly instantaneously as it is possible to measure. Don’t worry; physicists are having fits explaining that also, especially with experimenters demonstrating quantum entanglement on a daily basis. Essentially, it boils down to the question of just what constitutes information exchange because it looks a lot as if the best entanglement can do is let two observers in different frames of reference “see” what is happening elsewhere, but they don’t actually exchange information between each other.)

Not surprisingly, because no idea is completely new, there is actually a theoretical description of the theory behind and process of accelerating photons beyond the normal “c.” This is known as the Scharnhorst Effect, named for, unsurprisingly, a University of Berlin physicist, Klaus Scharnhorst, who speculated that between two uncharged plates a photon could move faster than “c” because the quantum electrodynamics analysis of the setup predicts an effective refractive index less than one.

(The index of refraction is a measure of how much a beam of light is bent/slowed down in a particular medium. It is 1.00 for a vacuum, 1.33 for water at room temperature, and so forth. That is, by the way, how refractometers can easily measure how much sugar there is in a liquid—the amount of disolved sugar alters the index of refraction. Or if a cup of coffee is properly brewed. The amount of coffee solids alters the index of refraction. —Ed.)

Obviously an index of refraction lower than 1.00 would cause light to move faster than it does in a vacuum (the index is defined by the speed of light, not vice versa, but you get the idea).

This, in itself, doesn’t violate relativity because it applies to individual photons, not the wave front that makes up an actual light beam.

Light in a vacuum actually encounters virtual particles as required by quantum physics and Heisenberg’s Uncertainty Principle, which both state that it is impossible to fix the position of a particle; in other words, that a particular particle has a tiny but finite probability of being anyplace! Recall that in quantum physics particles aren’t a physical point but rather a wave, a probability wave in point of fact, one that shows where a particle is not by definitely indicating the position but by showing how probable it is to be be located in a particular place.

As photons are absorbed by these virtual particles creating electron-positron pairs that (almost) instantly decay into the original photon, the “almost” part causes light to slow slightly.

Two uncharged parallel plates separated by a tiny space (on the order of one hundred times the size of an atom) in vacuum will exhibit a quantum dynamic force.

But the reason photons could move more quickly than the speed of light in a normal vacuum is because Casimir Effect plates are so close together that many of the particles making up the virtual particles that slow down photons in a normal vacuum will have wavelengths larger than the space between the plates, eliminating them from the mix of virtual particles that could fit in the vacuum space and thereby permitting photons to travel more quickly because they won’t encounter as many virtual particles.

So, although far from as dramatically or as usefully (for the time being) as slowing light to a complete stop, it is also possible to speed up light beyond the usual value of “c.”

If a science article could have a moral this one would be—ignore childish questions at the risk of never meeting the Nobel Committee. 

Further Reading

New Studies Suggest the Speed of Light is Variable.
Researchers Now Able to Stop, Restart Light.
Bringing Light to a Halt.

John McCormick is a trained physicist, science/technology journalist, and widely-published author with more than 17,000 bylines to his credit. He is a member of The National Press Club and the AAAS. His full bibliography can be accessed online.