Physicists make first observation of the pushing pressure of light

June 2, 2015 by Lisa Zyga feature
When light impinges on the surface of a liquid, part of the light is reflected and the remaining fraction is transmitted. The new experiments show for the first time that the liquid surface bends inward, meaning that the light is pushing on the fluid in agreement with the Abraham momentum of light. Credit: Zhang, et al.

(Phys.org)—For more than 100 years, scientists have debated the question: when light travels through a medium such as oil or water, does it pull or push on the medium? While most experiments have found that light exerts a pulling pressure, in a new paper physicists have, for the first time, found evidence that light exerts a pushing pressure.

The scientists suggest that this apparent contradiction is not a fundamental one, but can be explained by the interplay between the light and the fluid medium: if the light can put the fluid in motion, it exerts a pushing force; if not, it exerts a pulling force.

The researchers, Li Zhang, Weilong She, and Nan Peng at Sun Yat-Sen University in Guangzhou, China, and Ulf Leonhardt at the Weizmann Institute of Science in Rehovot, Israel, have published a paper on the first evidence for the pushing of light in a recent issue of the New Journal of Physics.

Minkowski vs. Abraham

The debate on the nature of the pressure, or , of light goes back to 1908, when Hermann Minkowski (best known for developing the four-dimensional "Minkowski spacetime" used in Einstein's theory of relativity) predicted a pulling force. In 1909, physicist Max Abraham predicted the exact opposite, that light exerts a pushing force.

"Scientists have argued for more than a century about the momentum of light in materials," Leonhardt told Phys.org. "Is it Abraham's, is it Minkowski's? We discovered that momentum is not a fundamental quantity, but it is made in the interplay between light and matter, and it depends on the ability of the light to move the material. If the medium does not move, it is Minkowski's, and if it moves, Abraham's. This was not understood before."

(a) Minkowski’s momentum of light: the surface bulges out, indicating that light is pulling on the medium. This regime occurs when the light is not able to put the fluid in motion (the light is too focused or the container of fluid too shallow). (b) Abraham’s momentum of light: the surface bends inward, indicating a pushing force. This regime occurs when the light is able to move the fluid. In both figures, the surface deformations are exaggerated for making them visible. Credit: Zhang, et al.

The two different types of pressures can be experimentally distinguished by illuminating the surface of a liquid with a light beam and seeing whether the liquid rises or falls. If the liquid's surface bulges out, then the light is pulling the liquid in agreement with Minkowski's theory. If the surface bends inward, the light is pushing in agreement with Abraham's theory. While the predictions of the two theories agree in empty space (which has a refractive index of 1), they differ in any medium with a refractive index greater than 1.

In the new study, the scientists showed that they could make the surface bend inward, corresponding to the pushing pressure, by using a relatively wide light beam and a relatively large container—two factors that cause the light to create a flow pattern in the fluid. The researchers demonstrated this pushing force in both water and oil, which have different refractive indices, in agreement with Abraham's theory.

In previous experiments, which found that light exhibits a pulling pressure, researchers had used narrower light beams and smaller containers than those in the current experiment, so the researchers here modified their original experiment by using a narrower beam. Their results in this new regime now revealed a pulling pressure, in agreement with the experiments from previous studies, suggesting that the nature of the pressure depends not only on the light, but on the fluid as well.

Light and snooker balls

Taking a step back, we might ask, why does light have momentum in the first place? Leonhardt explains that the momentum of light is slightly different than its energy, and can be understood as a pressure that causes motion, in analogy to snooker (i.e., billiards) balls.

comet
The momentum of light (along with the solar wind) creates the tails of comets by pushing material off the comets. Credit: European Southern Observatory

"Imagine a snooker game," he explained. "The player kicks one ball and this ball kicks another one. In all these kicks, the momentum the player initially gives to the cue stick is setting things in motion. Light may kick materials as well, just like the snooker balls, but these kicks are minuscule. In some circumstances, however, the kicks of light make a dramatic appearance. One example is the tail of a comet. Johannes Kepler speculated a long time ago that comet tails are caused by light pushing material off the comets, because they always point away from the Sun; we know now that he was partly right (the rest of the pushing is done by the solar wind). The ability of setting mechanical objects into motion is called momentum. It is not the same as energy, but often closely related to it."

He went on to explain that the controversy of the pushing vs. pulling nature of light's momentum only concerns situations in which light is not completely reflected off an object, but at least partially transmitted through the material.

"There is no conceptual problem with the momentum of light if the light is reflected, for example from a mirror or the dust particles of a comet, because here the momentum balance is very simple: twice the incident momentum causes motion, the incident and the reflected one," Leonhardt said. "If, however, part of the light is transmitted, then the transmitted light in the material needs to be taken into account. There it matters whether the Abraham or Minkowski momentum is carried by the transmitted light, as it affects the net balance of momentum, whether it is positive or negative. In Abraham's case the net balance leads to a push, in Minkowski's to a pull."

The findings have both fundamental and practical significance. Fundamentally, the results help scientists gain a better understanding of the nature of light. While it has long been known that light carries both energy and momentum, and that the energy of a photon is quantified by its frequency f times Planck's constant h, the momentum of light has not been so easy to describe. Does the momentum increase or decrease as the of the medium increases? The results here suggest that the answer depends on whether or not the light can put the fluid into motion: if it can, its momentum decreases and it exerts Abraham's pushing force; otherwise, its momentum increases and it exerts Minkowski's pulling force.

This distinction may prove very useful, as scientists have recently begun to develop applications that take advantage of light's momentum, or pressure. One such application, called inertial confinement fusion, uses the power of light's momentum to ignite nuclear fusion. Physicists can also use the momentum exchange between light and an oscillating mirror to cool the mirror to its quantum-mechanical ground state. Optical manipulation techniques, such as optical tweezers, use the gentle pressure of light to hold and manipulate cells for biomedical and nanoengineering applications. The researchers here hope that a better understanding of the momentum of will contribute to these developments.

Explore further: Study resolves century-long debate over how to describe electromagnetic momentum density in matter

More information: Li Zhang, et al. "Experimental evidence for Abraham pressure of light." New Journal of Physics. DOI: 10.1088/1367-2630/17/5/053035

Related Stories

Recommended for you

Probe for nanofibers has atom-scale sensitivity

January 20, 2017

Optical fibers are the backbone of modern communications, shuttling information from A to B through thin glass filaments as pulses of light. They are used extensively in telecommunications, allowing information to travel ...

Magnetic recording with light and no heat on garnet

January 19, 2017

A strong, short light pulse can record data on a magnetic layer of yttrium iron garnet doped with Co-ions. This was discovered by researchers from Radboud University in the Netherlands and Bialystok University in Poland. ...

47 comments

Adjust slider to filter visible comments by rank

Display comments: newest first

docile
Jun 02, 2015
This comment has been removed by a moderator.
mytwocts
not rated yet Jun 02, 2015
The pushing effect is explained by the fact that part of the light is reflected.
The pulling effect is caused by the tractor beam phenomenon: the liquid is attracted to the beam region. I don't see the connection with the Abraham-Minkowski debate about the momentum of the light _inside_ the medium.
mytwocts
not rated yet Jun 02, 2015
I predict that when the light exits the fluid the surface is pushed outward, again by reflection.
Still, the work is a beautiful demonstration of a fundamental effect.
rufusgwarren
1 / 5 (5) Jun 02, 2015
Seems obvious when one understands the effect of a time varying field and two particles, the proton and electron. Molecular response is similar to the gravitational field, but slightly different. Here the response is not the interaction of two similar E fields; it is the particular atom, or molecule motion. The applied electromagnetic energy and its dispersion, i.e. phonon energy and photon energy. I do not like to think of the later as a particle, rather the field, plane wave, or beam, then spot size, frequency, power, etc.. All may be properly rationalized either with the necessary measurements empirically, or a model built from fact. Minkowski space is a fantasy space. Seeking a fit is an exercise in science fiction!
mytwocts
5 / 5 (2) Jun 02, 2015
@rufusgwarren
You seem to be off-topic in spite of an abundance of physics keywords.
billpress11
5 / 5 (1) Jun 02, 2015
These two experiments are not done under identical circumstances.

Quotes:
"In the new study, the scientists showed that they could make the surface bend inward, corresponding to the pushing pressure, by using a relatively wide light beam and a relatively large container—two factors that cause the light to create a flow pattern in the fluid. The researchers demonstrated this pushing force in both water and oil, which have different refractive indices, in agreement with Abraham's theory."

"In previous experiments, which found that light exhibits a pulling pressure, researchers had used narrower light beams and smaller containers than those in the current experiment, so the researchers here modified their original experiment by using a narrower beam. Their results in this new regime now revealed a pulling pressure, in agreement with the experiments from previous studies, suggesting that the nature of the pressure depends not only on the light, but on the fluid as well."
billpress11
5 / 5 (1) Jun 02, 2015
There is a lot of missing information, such as the sizes of the container, the dept of the liquids or even if the the containers are made from the same material. Even the width of the light beams are not the same.
Without more information I am not willing to accept the pulling nature of light.
pauljpease
5 / 5 (3) Jun 02, 2015
This article must not be describing the history of the field and this experiment well. People use optical traps daily and they work by direct transfer of momentum from the photons to the object the photons pass through. The trapping force is caused by the object refracting light. Since the focused light has a 2D guassian intensity profile (normal bell-curve shape, more photons passing through the center and diminishing as you move from focal point), any object that is not symmetric and centered in the beam will have more photons passing through one side than the other. More photons = more momentum transfer as the light passes through object, creating a net force on the object that tends to drive the object towards the center of the beam (for an object with a higher refractive index than the medium, reverse if otherwise). It is very well understood. The article does a poor job of explaining the details of this particular experiment that presumably show something that wasn't already done
mytwocts
5 / 5 (1) Jun 02, 2015
There is a lot of missing information, such as the sizes of the container, the dept of the liquids or even if the the containers are made from the same material. Even the width of the light beams are not the same.
Without more information I am not willing to accept the pulling nature of light.

Tractor beams, once science fiction, are now a reality.
http://en.wikiped...tor_beam
arom
Jun 02, 2015
This comment has been removed by a moderator.
DarkLordKelvin
5 / 5 (3) Jun 02, 2015
I see (at least) a couple of red flags here. First this rather odd quote from one of the PI's:
We discovered that momentum is not a fundamental quantity, but it is made in the interplay between light and matter, and it depends on the ability of the light to move the material.


And then this surprising statement further on:
The results here suggest that the answer depends on whether or not the light can put the fluid into motion: if it can, its momentum decreases and it exerts Abraham's pushing force; otherwise, its momentum increases and it exerts Minkowski's pulling force.


Momentum is most certainly a fundamental quantity in physics; the conservation of momentum arises from the translational symmetry of space (c.f. Noether's theorem). If the light is gaining momentum from the liquid, why is the liquid expanding? Shouldn't it be cooling locally? I think they have neglected molecular level phenomena related to optical tweezer effect, as noted below by pauljpease.
adam_russell_9615
5 / 5 (2) Jun 02, 2015
Imagine light being reflected back and forth between 2 perfect mirrors floating in space. If light pushes the mirrors the mirrors gain energy and the light must then lose energy, correct? Does it lose photons or does the light reduce in energy through red-shift?
DarkLordKelvin
4.8 / 5 (4) Jun 02, 2015
Imagine light being reflected back and forth between 2 perfect mirrors. If light pushes the mirrors the mirrors gain energy and the light must then lose energy, correct?
It depends ... if the mirrors are fixed and perfect, then there will be no net transfer of momentum ... if the mirrors are free to move, say if they were mounted on a perfectly aligned frictionless rail, then yes, there would be net transfer of momentum to the mirrors.
Does it lose photons or does the light reduce in energy through red-shift?
The photon count would not change, but there would be a progressive red-shifting of the light to longer wavelengths as the separation between the mirrors increased.
TimLong2001
2.5 / 5 (2) Jun 02, 2015
Light pressure is old news! The almost negligible mass of the photon ("defined" as zero to simplify atomic calculations) still has momentum due to its high velocity.
MarkmBha
not rated yet Jun 03, 2015
On the lighter side - COOL !
billpress11
5 / 5 (1) Jun 03, 2015
Imagine light being reflected back and forth between 2 perfect mirrors floating in space. If light pushes the mirrors the mirrors gain energy and the light must then lose energy, correct? Does it lose photons or does the light reduce in energy through red-shift?

I think it would lose momentum and energy due it being redshifted upon every reflection,
billpress11
5 / 5 (1) Jun 03, 2015
Maybe someone can explain this to me. What is the origin of the side pressure toward the light beam as it travels through the liquid? Or maybe it is the other way, what is the origin of the reduction of pressure in the liquid as the light beam travels through the liquid?
Benni
1 / 5 (2) Jun 03, 2015
Imagine light being reflected back and forth between 2 perfect mirrors floating in space. If light pushes the mirrors the mirrors gain energy and the light must then lose energy, correct? Does it lose photons or does the light reduce in energy through red-shift?


If the mirrors are free to move & they are not perfectly reflecting electromagnetic energy, then they will move based upon the ability of the mirrors surfaces to absorb the wavelength (s) of energy that is focused on those surfaces, then MOMENTUM is generated as an aftereffect of energy being imparted to the mirrors. The resulting effect of absorption of a photon by the imperfect mirror is that the absorbed energy may be re-emitted at a lower frequency of wavelength (often referred to as redshifting). It is possible too that energy transformation to mass may occur through electron pair production.

There are no perfect mirrors, if there were energy could never be absorbed by those surfaces resulting in MOMENTUM.
docile
Jun 03, 2015
This comment has been removed by a moderator.
swordsman
4.7 / 5 (3) Jun 03, 2015
"The pressure of light was first discovered and measured experimentally by P. N. Lebedev (1899)." And:
"The assertion that light, as electromagnetic radiation, has the property of momentum and thus exerts a pressure upon any surface exposed to it was published by James Clerk Maxwell in 1862, and proven experimentally by Russian physicist Pyotr Lebedev in 1900 and by Ernest Fox Nichols and Gordon Ferrie Hull in 1901.".

The Minkowski interpretation is flawed, as I have shown. Einstein's theory was based on the Minkowski interpretation, and is also flawed. Physicists are beginning to learn that Planck's original quantum theory was correct, and the modern quantum mechanics is flawed.
DarkLordKelvin
5 / 5 (1) Jun 04, 2015
There are no perfect mirrors, if there were energy could never be absorbed by those surfaces resulting in MOMENTUM.
I don't think that's correct. While energy transfer of some kind must clearly accompany momentum transfer, I don't think the only mechanism for that is from "imperfections" in the mirror. A theoretically "perfect" mirror is analogous to a "perfectly elastic" collision (which also is impossible in nature) .. the reflected photon reverses direction, requiring transfer of p=2hf/c of momentum to the mirror. If the mirror is free to move, then it will acquire a tiny amount [KE=(p^2)/(2m) ] of kinetic energy, and an equivalent amount of energy will be lost from the photon, redshifting its characteristic frequency slightly. I have worked out the relevant amounts before, and they are REALLY tiny ... on the order of 10^-55 J per reflection transferred for a photon in the visible region of the spectrum.
docile
Jun 04, 2015
This comment has been removed by a moderator.
shavera
4.7 / 5 (3) Jun 04, 2015
@TimLong: Light has precisely zero mass unless some experiment shows us otherwise in the future. Not just "defined to be zero, but really small." Zero mass. Period. Full stop. It is simply that the definition of momentum as p=mv is an incomplete definition.

Furthermore, yes, we all know light exerts pressure... this article summarising the science somewhat misses the point of the fact that the question is "What *kind* of pressure does light exert on a *fluid interface*?" It's a matter of the known concept (radiation pressure) being applied to a specific case (fluid materials).
earthling98765
not rated yet Jun 04, 2015
PERFECT TIME to have this article published...when the Planetary Society has their Solar Sail satellite in orbit around Earth. It uses the momentum of light to accelerate - hopefully to Light Speed itself. I also am interested in the "Push-pull" effect as it relates to my design of the world's first "ION Telescope" which uses ionized gas to reflect light to a focus. Will there be a distortion effect on the gas?? to what degree??
docile
Jun 04, 2015
This comment has been removed by a moderator.
mytwocts
5 / 5 (1) Jun 04, 2015
Light pressure is old news! The almost negligible mass of the photon ("defined" as zero to simplify atomic calculations) still has momentum due to its high velocity.

Why would zero photon mass make atomic calculations easier ?
On the contrary, in QED the photon mass has to be defined as infinitesimally small, but not zero, to make calculation even possible.
mytwocts
5 / 5 (2) Jun 04, 2015
Light has precisely zero mass unless some experiment shows us otherwise in the future
We already know from mass spectrometer experiments, that the photons transfer mass - not just momentum. Without radiative transfer of matter the stars couldn't evaporate their matter via radiation. How could you explain the E=mc^2 formula, after then?

Photons transport energy, momentum and angular momentum, not mass.
Stars lose very little mass by fusion. Over its lifetime the sun will lose less that a thousandth of its mass by radiation (assuming todays radiation rate).
docile
Jun 04, 2015
This comment has been removed by a moderator.
mytwocts
5 / 5 (2) Jun 04, 2015
If the mass of photons would be zero, it couldn't lose anything.

The mass of the sun can decrease by emission of massless energy. There is conservation of energy, but there is no conservation of mass.
Andrew Palfreyman
not rated yet Jun 04, 2015
Get your geek on.
If you want to calculate the force F Newton on a light sail due to incident power P Watts, assuming perfect reflectivity, it's simply
F = 2*P/c
where c is the speed of light in vacuo.

Until things get relativistic. Then it gets a lot hairier.
docile
Jun 05, 2015
This comment has been removed by a moderator.
Benni
1 / 5 (2) Jun 05, 2015
Light has precisely zero mass unless some experiment shows us otherwise in the future


We already know from mass spectrometer experiments, that the photons transfer mass - not just momentum. Without radiative transfer of matter the stars couldn't evaporate their matter via radiation. How could you explain the E=mc^2 formula, after then?
........because he does not comprehend Special relativity & the Energy/Mass Equivalence Principle that Energy & Mass are simply transformed products of one another.

Photons transport energy, momentum and angular momentum, not mass.
Stars lose very little mass by fusion. Over its lifetime the sun will lose less that a thousandth of its mass by radiation (assuming todays radiation rate).
.......as I just stated, you simply do not comprehend the Energy/Mass Equivalence Principle in SR. And to help you along a little bit here, Energy Flux Fields also have conservation of gravity effects upon transformation from Mass.
mytwocts
5 / 5 (1) Jun 05, 2015
[ the mass-energy equivalence would be violated: E=mc^2.

Calling E/c^2 mass is a bad idea. Mass always means rest mass.
mytwocts
5 / 5 (1) Jun 05, 2015
you simply do not comprehend the Energy/Mass Equivalence Principle in SR [\q]
Let me give you a lesson in SR.
Apply m=E/c^2 to E^2=m^2c^4+p^2c^2.
This gives m^2=m^2+p^2/c^2, thus p=0.
So E/c^2=m only holds in the rest frame.
indio007
not rated yet Jun 05, 2015
what happens when the index of refraction is negative hmmm?
Benni
1 / 5 (2) Jun 05, 2015
you simply do not comprehend the Energy/Mass Equivalence Principle in SR
Let me give you a lesson in SR.
Apply m=E/c^2 to E^2=m^2c^4+p^2c^2.
This gives m^2=m^2+p^2/c^2, thus p=0.
So E/c^2=m only holds in the rest frame.


......then simply include the Lorentz factor if you want to be so picky, but until you are very near lightspeed it is an almost pointless calculation to include.
mytwocts
5 / 5 (1) Jun 05, 2015
Well ok then. E=m\gamma c^2 so that E/c^2=m\gamma which is only equal to m if \gamma=1, which implies v=0.
Benni
1 / 5 (4) Jun 05, 2015
Well ok then. E=m\gamma c^2 so that E/c^2=m\gamma which is only equal to m if \gamma=1, which implies v=0.


...........why are you on a science site? You don't really imagine you're teaching us anything do you?
mytwocts
5 / 5 (1) Jun 06, 2015
You just dropped out of class.
mytwocts
5 / 5 (2) Jun 06, 2015
Benni, with statements like
"then MOMENTUM is generated as an aftereffect of energy "
you are definitely in need of teaching.
Benni
1 / 5 (4) Jun 06, 2015
Benni, with statements like
"then MOMENTUM is generated as an aftereffect of energy "
you are definitely in need of teaching.


.............after my six years of Engineering School education in Electrical/Nuclear Engineering, what is it you think you can teach me?
mytwocts
5 / 5 (1) Jun 06, 2015
Probably nothing on Electrical/Nuclear Engineering. Still, momentum (small caps, please !) is not generated. Reflection is not absorption followed be reemission. Your conclusion is otherwise correct. Defining mass to be identical, up to a constant factor, to energy is confusing, hence the confusion above.
Benni
1 / 5 (2) Jun 06, 2015
Probably nothing on Electrical/Nuclear Engineering. Still, momentum (small caps, please !) is not generated. Reflection is not absorption followed be reemission. Your conclusion is otherwise correct. Defining mass to be identical, up to a constant factor, to energy is confusing, hence the confusion above.


Quite simply, you still do not comprehend the science of Special Relativity.
pepe2907
not rated yet Jun 06, 2015
First? Really? Then how they manipulate cold atoms with light in so many experiments? How hydrogen gets compressed in an H-bomb? /well, I admit after hydrogen gets compressed in an H-bomb there's nobody left to report on the observation, but still the results are pretty visible/
hudres
not rated yet Jun 06, 2015
Optical pressure, or more correctly radiation pressure, is well known and has been measured for decades. Consider the case of optical tweezers, which use pressure to manipulate molecular scale structures and whose force has been measured for many years. One can also consider thermonuclear bomb physics where radiation pressure has been measured and successfully used since the early 1950's. Nothing new here.
mytwocts
5 / 5 (2) Jun 07, 2015
Benni
You are an interesting case of Dunning Kruger.
casualjoe
5 / 5 (2) Jun 07, 2015
Benni, with statements like
"then MOMENTUM is generated as an aftereffect of energy "
you are definitely in need of teaching.


.............after my six years of Engineering School education in Electrical/Nuclear Engineering, what is it you think you can teach me?


It seems you still need help with the simple stuff, like the question of where the centre of the universe is, and how to properly think about that question. (Quickhint - it's everywhere)

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.