Bose–Einstein condensation, which demonstrates the wave nature of material particles, now offers further illumination of wave–particle duality: it has been observed in light itself. See Letter p.545
“Art is the Tree of Life. Science is the Tree of Death.” So wrote the visionary English poet and artist William Blake1 in 1826, contrasting the limitless creativity of art with the limiting rigidity of science. Blake understood the science of his day — well enough, for instance, to lump together “The Atoms of Democritus And Newton's Particles of Light”, and compare them both to sand2. In Blake's time, both were conceived of as indivisible and indestructible, their motion governed by Newtonian mechanics. Nothing represented his harsh view of science better than the implication that, if light corpuscles could be neither created nor destroyed, then the Universe contained a fixed amount of light, which could never be increased (Fig. 1). On page 545 of this issue, Weitz and colleagues3 demonstrate that, even if that were true, the wave nature of light would still persist, through the Bose–Einstein condensation of photons. They also demonstrate the creativity that thrives within scientific rigour.
Modern physics teaches that light has wave as well as particle properties. But one of the most basic differences between the wave and particle theories is rarely emphasized in textbooks. Classical light waves are not conserved like the atoms of Democritus, but can easily be excited and absorbed. So, a lamp may run out of battery power, but it does not run out of light. In this respect, Newton's particle theory of light was as false as the caloric theory of heat, according to which heat was a conserved substance held in matter like water in a sponge. In fact, both heat and light are simply convertible forms of energy. And it was the thermodynamics of light that led Planck and Einstein to the quantum unification of wave and particle theories.
Unlike classical particles, quantum particles such as electrons and photons can in general be created and destroyed, and so the issue of whether the amount of light is fixed is now separate from the discussion of wave and particle behaviour. A basic question therefore seems natural: what would light be like if photons were, like atoms, wave-like but conserved? Weitz and colleagues3 have answered this question experimentally. By confining light within the narrow slice of space between two barely separated mirrors, and filling this slab-like cavity with a dye material, they have achieved thermal equilibration of light as a gas of conserved particles, rather than ordinary black-body radiation. The critical step in realizing such unusual thermalization is confining the light in one direction, so that every photon is forced to have a frequency at least as high as that of a standing wave. Because the separation between the mirrors is only a few micrometres, this minimum photon frequency is high — crucially, much higher than the frequency corresponding to the temperature of the dye.
This large frequency difference makes the energy budget of the system resemble the finances of a peculiar commercial firm that sells both skateboards and satellites. In the firm's ledgers, the billions column and the millions column must always be balanced separately, because the total volume of the skateboard business never amounts to a single satellite. In the Weitz group's system of dye and light, the thermal energy and the energy of excitations at the standing-wave frequency are similarly each conserved separately, because their scale discrepancy prevents one from balancing the other. And this means that the number of photons between the mirrors changes as photons are absorbed and re-emitted by the dye, but only in the same way that the number of atoms in a gas changes, locally, as the atoms drift around. In technical terms, the light in the Weitz group's experiment reaches thermal equilibrium with a chemical potential as well as a temperature, just like gases cooled to nanokelvin temperatures in magnetic traps. The textbook example of what this can allow is Bose–Einstein condensation, which confers the properties of classical wave physics on a gas of conserved quantum particles below a critical temperature, and is intimately related to the phenomena of superfluidity and superconductivity. The Weitz group has observed Bose–Einstein condensation of light, in remarkably close analogy to that of atoms.
As well as being a landmark achievement in itself, making photons behave thermodynamically as atoms, even to the point of Bose–Einstein condensation, illustrates a broader theme in current physics. Atomic gases have been made to behave as laser light4, and even as black holes5. The 'holes' left when electrons in graphene sheets are energetically displaced reproduce the behaviour of relativistic positrons6. Quantized spin-wave excitations in magnetic films have been made to behave as quantum gases7, and atomic gases have been made to behave as ferromagnets8. The discernible trend is that everything is becoming everything else. Physics is the art of the interchangeable.
The purely scientific merit in this trend is that demonstrating the interchangeability of physical details clarifies the few universal patterns and principles that really are conserved — the atoms, as it were, not of matter or of light, but of reality. In this sense, the reductionist progress of science proceeds at full tilt. But in the proliferation of startling masquerades, physical science is also taking on more than ever the aspect of a creative art, in a medium that, with the advances of modern technology, is proving far less constraining than it once seemed. Light is unlimited — or not, as we choose. Blake spoke too soon.
Ellis, E. J. (ed.) The Poetical Works of William Blake Vol. 1, 435 (Chatto & Windus, 1906).
Stevenson, W. H. (ed.) The Poems of William Blake 481 (notebook drafts) (Longman, 1971).
Klaers, J., Schmitt, J., Vewinger, F. & Weitz, M. Nature 468, 545–548 (2010).
Taubes, G. Science 275, 617–618 (1997).
Lahav, O., Itah, A., Blumkin, A., Gordon, C. & Steinhauer, J. Preprint at http://arxiv.org/abs/0906.1337v1 (2009).
Novoselov, K. S. et al. Nature 438, 197–200 (2005).
Demokritov, S. O. et al. Nature 443, 430–433 (2006).
Jo, G.-B. et al. Science 325, 1521–1524 (2009).
About this article
Journal of Modern Optics (2018)
Journal of Modern Physics (2014)
Journal of Magnetism and Magnetic Materials (2012)
American Journal of Physics (2011)