4 Light

The current model for the nature of light is one of those areas of modern physics where obfuscation is an integral part of the definition.

Light has the properties of a wave. It reflects and refracts. One of Newton’s arguments against the wave nature of light was that waves were known to bend around obstacles, while light travelled only in straight lines. This is the common observation that shadows are not entirely black.

Yet Newton’s contemporary Huygens[i] created wave theory in 1678 by arguing that each point on a wave can be viewed as the origin of further waves, reasoning now known as Huygens’ Principle. Thomas Young (1773 – 1829) demonstrated this effect in two ways, as diffraction of a shaft of light at a sharp edge and by the interference effect created when light or another wave passes through a pair of openings, as in the critical ‘double-slit’ experiment. At around the same time, Augustin Fresnel (1788 – 1827) used Huygens’ insight to explain these phenomena mathematically, effectively establishing for the following century that light was a wave.

There are therefore four distinguishing characteristics of waves, the phenomena of reflection, refraction, diffraction and interference, and light never fails to demonstrate these properties in any appropriate set of circumstances. There is also a substantial body of evidence that light is a particle. This is more varied and hence could be considered more extensive, but on examination it turns out that much of this is problematic. The evidence most commonly cited is the weakest, and we can deal with this quickly. More complex experiments and arguments must wait until later, when we have the background necessary to address them.

While it is true that Newton, who might normally be considered a reliable authority on matters of physics, supported a ‘corpuscular’ view, his arguments did not stand up to scrutiny, and we saw above an example of this. The particle view became a flat-Earth position during the 1800s, but re-emerged with the work of Planck in 1900 and 1901.

Planck’s work on blackbody radiation[ii] demonstrated what is now widely recognised, that light is emitted in discrete events at specific frequencies. It also strongly suggested that the average energy of emissions – and perhaps the energy of each specific emission – is proportional to the frequency, a proportionality now known as Planck’s constant. The importance of this is clear: if light is emitted in quanta of energy, then it makes the particle hypothesis all the more plausible.

This work established Planck as the founder of modern quantum theory, but he was properly cautious regarding the assumption that this helped to establish that light is essentially particulate, since wave explanations also fit the data. Part of the problem of understanding this work is that there is in the modern mythology of the photon a casual acceptance that discrete emission is somehow the same as discrete existence after emission, but if that were true then every time a bell was struck only one person could hear it.

We also have the problem identified by Niels Bohr, among others:

‘As is well known, the hypothesis introduces insuperable difficulties when applied to the explanation of the phenomena of interference, which constitute our chief means of investigating the nature of radiation. We can even maintain that the picture, which lies at the foundation of the hypothesis of light-quanta excludes in principle the possibility of a rational definition of the conception of a frequency ν, which plays a principle part in this theory.’[iii]

The assumption that particulate light can do what wave light cannot is first seen conspicuously in the paper of Einstein in 1905 on the photoelectric effect, in which he converts Planck’s formula from energy to momentum and argues that only particles of a minimum momentum can knock free an electron from a metallic surface.

This is an interesting argument, but not as strong as it is often made out. A faster vibration is also more able to separate a structure than a slower one, and the ability of wave theory to explain apparently particle-like phenomena was demonstrated later by Schrödinger and others. In contrast, there are no plausible particle models for interference, for example.

There is a definitive experiment, known as ‘single-photon double-slit’, with which the reader may be familiar. If we reduce the radiation incident on a pair of slits to an extremely low level where we believe that photons are arriving singly, then, if they are indeed particle-like, they can travel through only one slit at a time.

We observe them arriving singly, as each absorption creates a tiny dot on our photographic emulsion. A pattern takes a long time to emerge, but when it does, it is a classic interference pattern, with alternating bands of dark and light. Bohr is one of the founders of quantum theory, and we have his word for it, confirmed by many others and by our own logical faculties, that there is no rational explanation of how particle photons can ‘interfere’.

We can check this further, by rapidly alternating the openings so that only one is open at any point in time. In this case light is detected but no interference pattern appears. This rules out the suggestion that a photon can ‘hang about’, as if on a street corner, waiting for another of its kind to interfere with. It is also strong evidence against the notion that a single emission of light is sufficiently localised to pass through a single slit.

In the research literature, and in all teaching material on quantum mechanics – and even enshrined in the very name of the subject – the quantum nature of light is assumed, and we will look shortly at why this is. As a consequence, the experiment just described is known universally as ‘single-photon double-slit’ and presented as confirmation of the mysterious nature of the physical universe. This will not be considered good enough here.

We can ask two other crucial questions. Can we find a way to see what a light emission looks like, and can we detect a single emission of light more than once? The former was answered in the first years of this new millennium when Ferenc Krausz led a team from Austria and Germany that produced individual emissions of light and carefully measured their effect on electrons. By combining this data they demonstrated that a single emission of light is a pulse comprising a few wavelengths of a sinusoid-like wave within an overall envelope in the shape of a bell, confirming what many physicists thought they already knew. With a flair for the poetic they styled this as the ‘first ever photograph’[iv] of a light wave.

On the second question, there has been considerable effort and debate, with the preponderance of evidence and argument on the side of the particulate emission that can only be detected once. We will return to examine this later, but for now it gives us a headache.

Light is clearly a wave of some sort as it possesses all the attributes of a wave and nothing else is known that has these. Interference and the double-slit experiment in particular require a phenomenon that is spread out. We can account for blackbody radiation and the photoelectric effect with waves that are emitted in individual events at specific and well-defined frequencies, but could it be that this ‘wave’ remains in some fashion localised and hence ‘particle-like’, so as to account for our inability to detect it twice? For the moment we will look at the exceptional work of Maxwell in the late nineteenth century on electromagnetism, and re-examine what this had to say about the nature of light.

Esau James…   ©2010

Overview: Physics Rogue Science


[i] Christiaan Huygens (1629-1695), Traite de la Lumiere, 1678.

[ii] Max Karl Ernst Ludwig Planck, 1858 – 1947. Annalen der Physik: 1 (1900) 99 & 719; 4 (1901) 553.

[iii] Niels Bohr, at the 1921 Solvay conference.

[iv] Ferenc Krausz, “Tracking light oscillations: Attosecond spectroscopy comes of age”, Opt. Photon. News 13, 62 (2002). A report, from which the quote is taken, is in the New Scientist of 9 April 2005.

Tags: , , , , , , , , , , , ,

Leave a Reply