Published online 23 December 2008 | Nature | doi:10.1038/news.2008.1332


Optical fibres feel light's recoil

Experiment claims to resolve an old debate about how light behaves.

fibre opticsCan light give fibre-optic wire a kick?PUNCHSTOCK

Does light push or pull? For 100 years, scientists have disagreed over whether light increases or decreases its momentum as it passes from glass into air. Now an experiment by researchers in China may have finally settled the matter.

Weilong She and his colleagues at Sun Yat-Sen University in Guangzhou have shown that a very thin, dangling optical fibre actually bends as light comes out of the end, similar to the way that a garden hose writhes as it spurts water (see video)1.

This, they say, is evidence that the light gains momentum when it travels from glass into air, supporting the idea proposed in 1909 by German physicist Max Abraham.

But the question might not be resolved so easily. Peter Knight of Imperial College in London says that past studies have been dogged by delicate questions about the way light behaves at the interface between one medium and another. He says that the answer might in fact depend on how the experiment is done.

Beam time

Abraham's argument contradicted the theory proposed a year earlier by Hermann Minkowski, the German mathematical physicist better known for his exploration of curved space-time.

Minkowski's theory was based on quantum mechanics. When light travels from air to transparent substances such as water or glass, it slows down, a fact that explains the bending of light rays in refraction. At the same time, its wavelength gets shorter. According to quantum theory, that should increase its momentum, the reverse being true for the passage from glass to air.

Abraham, by contrast, said that light should behave like ordinary objects, the momentum of which decreases as their speed decreases.

Both arguments seem reasonable — but which is right? Earlier experiments have failed to settle the issue. For example, in 1973 two scientists at Bell Telephone Laboratories in New Jersey looked at what happens when a light beam shines onto water2. Some of the light enters the water, and if it gains momentum in doing so, as Minkowski claimed, the water surface should bulge outward. This is indeed what the researchers found. But later work showed that the water's bulge was mostly caused by other effects.

She and his colleagues have now used a similar principle, but without the complications of a wobbly water surface. They reason that as light leaves the end of an optical fibre, the tip should be pulled forward if Minkowski is right, but pushed backwards — like a recoiling gun barrel — if Abraham is right.

In both cases, the fibre would bend, and a detailed study of its shape should reveal whether it was caused by a push or a pull.

Everyone's a winner

To see the effect, the researchers needed to use a very thin, lightweight fibre just half a micrometre wide, dangling freely in air. When they shone pulses of laser light down the fibre, they saw it wiggle and swing like a pendulum, in a manner consistent with an increase in momentum as the light exits — exactly as Abraham predicted.

The movement of the fibre's tip could potentially be caused by other factors, such as heating of the glass by the light. But She and his colleagues calculate that heating alone wouldn't be enough to explain the effects they see.

All the same, physicist Miles Padgett of the University of Glasgow, UK, says that "a lot of things could make the fibre bend". What's more, other experiments looking at the effects of light on ultracold gases of atoms3 seem to support Minkowski's theory.

In fact, it is possible that both of them might be right, says Knight, depending on the kind of experiment conducted. He says that the quantum wave-particle duality of light might mean that it behaves as Abraham predicted when it is probed in a 'particle-like' way, and as Minkowski predicted when its wave nature is examined4.

The issue is not merely academic, Knight adds, because scientists are now using light to probe and control tiny mechanical devices, such as resonating beams, to measure small distances or stresses. The way momentum is transferred between light and materials could have a profound effect on such experiments, he suggests. 

  • References

    1. She, W., Yu, J. & Feng, R. Phys. Rev. Lett. 101, 243601 (2008). | Article | ChemPort |
    2. Ashkin, A. & Dziedzic, J. M. Phys. Rev. Lett. 30, 139–142 (1973).
    3. Campbell, G. K. et al., Phys. Rev. Lett. 94, 170403 (2005). | Article | ChemPort |
    4. Leonhardt, U. Nature 444, 823–824 (2006).
Commenting is now closed.