Of Particular Significance

Cerenkov Radiation

Cerenkov (or Cherenkov) radiation is a bit of nineteenth century physics that stumbled into the twentieth. It could have been (and to some degree was, by the physicist Heaviside) predicted in the 1880s, but this effect was discovered by accident, perhaps by the Marie and Pierre Curie.  It was studied in detail by Pavel Cerenkov in the 1930s, and explained a few years later by Ilya Frank and Igor Tamm. The three of them won the 1958 Nobel prize for their work.

What was the basic observation? Cerenkov studied the blue light appearing when radioactive objects (i.e. containing atoms whose nuclei decay to other nuclei by spitting off high-energy particles, including electrons and positrons [anti-electrons]) were placed near water and other transparent materials.  We now know that any electrically charged particle, such as an electron, traveling with sufficiently high energy through water, air, or any other transparent medium, will give off bluish light. This light moves outward from the particle at a particular angle to the particle’s motion.

What’s going on? As Frank and Tamm understood, this is a photonic boom, quite analogous to the sonic boom that is created by a supersonic aircraft traveling through air faster than the speed of sound, or to the bow wave of a ship sailing through water. Light in a transparent medium will traverse that medium at a speed different from the speed of light in vacuum, because of the interactions between the light and the charged particles (electrons and atomic nuclei) that make up the medium. For instance, in water light travels about 25% slower than it does in empty space! And so it is easy for a high-energy electron to travel faster than light travels in water, while remaining slower than light’s speed in vacuum. If such a particle passes through water, it creates an electromagnetic shock wave, analogous to the shock wave that a supersonic jet aircraft makes in the density of air. And that shock wave radiates out from the particle, just as the sonic shock radiates from the airplane, carrying off energy in many different forms (wavelengths) of electromagnetic radiation, including visible light. There’s more energy created at the violet end of the rainbow than at the red end, and that’s why the light looks dominantly blue to our eyes and brains.

This type of radiation is enormously useful in particle physics, since it provides a terrific way to detect high-energy particles! Not only can we observe the presence of a high-energy charged particle by observing the light that it radiates, we can learn more by studying the light in detail. The precise pattern of the light can be used to determine (a) what path did the particle take across the medium, (b) how much energy was it carrying, and even (c) something about its mass (since electrons will scatter in the medium and make a jittery ring, whereas heavier particles will not do so.) A number of very important experiments, including ones that have won Nobel prizes, rely on this type of radiation, including some of those that have played a major role in studying the properties of neutrinos, e.g. SuperKamiokande.

Cerenkov radiation is also very useful in checking whether Einstein’s theory of relativity is an accurate description of nature. Cosmic rays, particles flying in from outer space (which often hit something in the atmosphere and make a large shower of particles that can be detected by observers on the ground), can in rare cases carry extraordinarily high energy, as much as 100 million times more energy than the protons at the Large Hadron Collider. These particles are (as far as we can tell) produced many light years away from earth, in powerful astronomical events such as supernovas. Now suppose the speed of light were not an ultimate speed limit, and those particles traveled faster than the speed of light in the vacuum of outer space. Then these ultra-high-energy particles would undergo Cerenkov radiation too. And because they have so far to travel, they would lose much of their energy to this radiation. It turns out this energy loss can be very rapid, and that these particles could not travel astronomical distances and keep their ultra-high-energy unless their speed remained extremely close to or below the speed of light.

In short, if cosmic rays at ultra-high-energies could move faster than the speed of light, then we shouldn’t observe any cosmic rays at ultra-high-energies at all, because they should all lose most of their energy long before they reach earth. But we do observe them. [Tiny loophole, I think: we are almost certain most of them are charged: their properties indicate they feel the strong nuclear force, and the only stable particles that could travel such distances are protons and more generally atomic nuclei, all of which have electric charge. If you open this loophole a bit, you would reduce the constraint somewhat, but it would still be pretty strong.] So therefore we can conclude: the ultra-high-energy cosmic rays (as well as any cosmic rays at lower energy) cannot exceed the speed of light by more than a very tiny number. Estimates from the late 1990s due to the very famous physicists Sidney Coleman and Sheldon Glashow (who were the first I am aware of to make this argument, but who might well have been preceded by other authors; please let me know if you are familiar with earlier papers) put this tiny number at about ten parts in a trillion trillion. The constraints from experiment have probably improved since then. [Will try to find updated numbers, and ones without any possible loopholes.]

Similarly, the simple fact that we observe high-energy electrons puts limits on their speed relative to that of light.  The most recent claim I’ve read (still trying to figure out its source) is that observation of electrons with energies up to half a TeV imply that electrons cannot exceed the speed of light by more than a part per thousand trillion.

37 Responses

  1. About the Cherenkov effect: how is it possible for a charged particle moving at uniform velocity, to emit photons? shouldn’t it be invariant changing reference frame?

    1. The charged particle does not emit photons. The particle excites the medium (electromagnetic interaction) and the medium relaxes by radiating photons once the particle influence is gone. This is why the medium is called “radiator”.

  2. You said that energy is created at the violet end of the rainbow, however due to the law of conservation that states energy can neither be created or destroyed. Did you mean emitted?

  3. We all know by now, that when you make a habit of connecting
    with potential customers little and often, on topics that they care about, you increase the likelihood of them doing business with you.
    Unfortunately, it’s also introduced a host of ways to violate
    intellectual property laws. With Pinterest, you can pin a
    picture of your customers and include a blurb of the case study or

  4. Thanks , I have recently been searching for info about
    this subject for a long time and yours is the best I’ve discovered till now.

    But, what in regards to the conclusion? Are you
    certain about the supply?

  5. Hostgator is at the moment giving 3 varieties of policies due to the purchaser:
    a child approach, hatching scheme, and opportunity approach.
    Hostgator offers infinite web hosting service to all its buyers.
    They can additionally impart their unique sentiments
    or audits.

  6. Hey there this is kind of of off topic but I was wanting to know
    if blogs use WYSIWYG editors or if you have to manually code with HTML.
    I’m starting a blog soon but have no coding expertise so I
    wanted to get advice from someone with experience. Any help would be greatly appreciated!

  7. Can you get Cerenkov Radiation emitted as particles pass through a curved area of space-time say a gravity well or an intense magnetic field?
    If you had a curved area of space time would c vary enough to facilitate such emissions?

  8. You say that we can proof Einstein by the fact that we observe ultra-high-energy cosmic rays.

    What if there are rays of even higher energy? How would we observe them? They would, as you say, never reach us…

    1. You don’t prove things in science; you check them. Mathematicians prove things. Scientists can only check things, within the ranges that their experiments cover.

      So we have checked that Einstein’s relativity works over a very, very wide range of phenomena. This includes cosmic rays as high as they go. There could be phenomena that lie at even higher energies where Einstein’s relativity fails. That’s true of any principle of science; if you go looking in a regime where a theory hasn’t yet been tested, it may break down. And scientists love to go looking in such places, because they like to make surprising discoveries.

  9. —-Quote————-
    I have two questions:
    1) About the classical Cherenkov radiation: According to the conventional theory, when a charged particle travels through a medium, (some) atoms in the medium become polarized, and emit photons in order to return to
    equilibrium. If the charged particle travels below the speed of light in the medium, the polarizations are not aligned, and no radiation appears.
    If its speed exceeds the speed of light, the polarizations become aligned, and Cherenkov radiation appears. This is a perfect explanation. But I don’t see any difference in the loss of energy for the particle between the two
    situations. Even when no radiation occurs, the atoms are still polarized and that costs energy. Why does the charged particle lose more energy in the case of Cherenkov radiation?
    2) As mentioned above, Cherenkov radiation is caused by interaction
    between traveling charged particles and surrounding atoms. If something travels in vaccum, it cannot interact with anything. How can it produce Cherenkov-like radiation? The claim that it must emit radiation
    at faster than light speed seems to be groundless.

    These two questions i have too.
    About 1, I can only imagine that when we don’t have alligned polarizations, the photons emitted by the depolarization will interfere destructively somehow- and thus transfer no energy away from the particle.

    What would be the mechanism of Cherenkov radiation in vacuo?
    I mean, with what ‘molecule’ would the particle interact to polarize it?

    1. To follow up on those two great! questions from Rick. I am physics grad, but still Cerenkov effect is not that clear to me.
      1) If particle (with v>c in the medium) losses energy to polarize the medium, and medium emits photons “when switches this polarization off”, why this “de-excitation” period is believed to be instantaneous? Say, fastest scintillators, which emit light in the process of de-excitation, still require nanosecond regime to work. What is so special that Cerenkov light emission mechanism takes no time?
      2) Is this related to vacuum polatization? Can it then be explained with virtual particles of the vacuum or quantized fields “between the particle and the medium” ?

  10. I can’t help thinking that all you physicists are missing something(s)
    1) we don’t actually know what the absolute velocity of light is ( because we have only measured it in a gravitational field)
    2) there is no such thing as a vacuum ( if you choose any unit volume, say a one metre diameter sphere, then it will be full of particles because if you observed from the centre of this sphere because there must be at least one photon ( as well as “solid particles such as cosmic rays) from each of the trillions of objects you can see passing through the unit volume every second.
    3) the universe has a refractive index. This results from the way gravity affects light. While the mean path may appear to be a straight line for light to reach our eyes it is actually a very wiggly line, so the absolute velocity of “light” is actually higher than what we think we have measured.

    Maybe neturinos actually do travel faster than the speed of light if the are truly massless & unaffected by gravity.

    1. A few comments:

      Massless particles are affected by gravity. That’s why the sun bends light rays, as Einstein’s theory of gravity predicts and has been shown to be true in experiments. http://undsci.berkeley.edu/article/0_0_0/fair_tests_04

      You can use the GPS system as a test of the speed of light, since it assumes that this speed is constant. This system effectively measures the speed of light out in space, in a location where the gravitational field is much, much smaller than on earth. That the GPS system works at all shows that we do know what we’re doing.

      It doesn’t matter whether there are trillions of particles traveling through a one meter sphere every second. To measure the speed of light requires a region and a time that are vastly smaller and shorter. The number of particles passing through the relevant region while the experiment is taking place is spectacularly smaller than you imagine.

      The speed of neutrinos and of light is known to be the same, to a few parts per billion, across vast distances of the universe and across the large gravitational fields of the galaxy, for neutrinos of energy about 1000 times smaller than those of OPERA. This follows from the observations of the 1987 supernova. See http://profmattstrassler.com/2011/09/20/supernovas-and-neutrinos/

  11. I’m a grad student in computer science and have never had a physics course past high school honors physics. However, I do a lot of reading on my own time about particle physics and find it totally fascinating.

    First, I want to say that I LOVE your site. Most of what I can find on the web is either too simple to be accurate or of use to me, or what I find is way over my head both in terminology as well as mathematics. Your site fits well in between the two and is quite helpful. I just love it. I’ve learned more reading everything on your site this afternoon than I have in months of trolling Wikipedia and the web. Thank you so much.

    “Is it right to say that a photon in a medium is being absorbed and re-emitted, but itself travels at c while in transit?”

    This question was asked above, and is exactly what I’ve wondered about many times in the past. My intuition doesn’t fit well with the idea that light is “slowed” in a medium. The only way for a photon to change is if it feels a “force” by interacting with another particle of some kind [or perhaps by interacting with space-time itself I suppose, though it seems odd to think that water is going to have much impact on the curvature of space time inside it]. So, my mind goes to a Feynman diagram with a photon (wiggly line) coming in on the left of the diagram and going right. At some point it interacts with a particle, so I picture the wiggly line intersecting a straight line representing an electron or some such. But, I thought in these interactions the photon is absorbed? I think I’ve even seen such diagrams given as examples of “invalid” Feynman diagrams: “http://blogs.uslhc.us/wp-content/uploads/2010/02/notallowed-1024×294.png” (The far right in the image). Could you offer any help to a n00b such as myself?

    I thought I even watched a youtube video with Feynman himself saying that a photon in a medium is being absorbed and re-emitted, but itself travels at c while in transit. I can’t find the video, though… it might have been embedded in his “Photons: Corpuscles of Light” lecture somewhere, but I’m not sure. I’ll go watch his video again. I could just be remembering wrong, happens all the time.

  12. @marc gehrig and Professor Strassler, In the analogy of the car and the construction site, it is useful to think, instead of car and mud since the construction workers are safer and we’re freer to not change “speed” up due to our pushing the accelerator. Instead, if we cruise into the mud at 30 mph, the car bends towards the ‘normal’ and we slow down say to 15mph, all the while not changing the amount of force on the accelerator (thus the amount of energy from the engine). Instead, the decrease in speed only comes from the muddy influence. When we move from the muddy medium to the pavement again, with no change of our steady position of the accelerator, won’t we bend away from the normal and resume our original speed? Then, isn’t this “acceleration” only a response to the differing medium, with the car having more resistance and interaction in the mud than on the pavement?

  13. Thank you for the response. And thank you for the Url.
    It clearly shows lights direction altering at refractve boundaries. In the making of compound lenses those changes allow light focus manipulated according to the len’s design.
    But “focusing” on the phenomena of lights speed. (Fastest in vacuum, slower speeds in other media), A slower speed in a denser or more refractive medium seems simple and intuitive. . Unexpressed is a mechanism for speed resumption (to the speed of light) when it re-enters a vacuum.
    Once back in a vacuum wouldn’t light propagate at the speed it originally left the laser?
    If it did revert, is it wrong to impute that it “accelerated” (it’s rate of distance/time increased?)

    If light were a car, the lens might be like a cars slowing down for construction. Coming out of construction we step on the gas to go back to the speed limit.
    But that metaphor relies on metaphoric props of car engine and gas pedal. What does light rely on to recover it’s initial speed?
    In the interest of brevity I didn’t include some paragraphs that I tried to outline how a pulse of light would have a length and a diameter and a brightness/energy. Where I felt that brightness would dim. Diameter would increase (even from the best laser).
    Its the apparant “free” impulse/conversion/transition that seems to violate my sense of the 3 laws of newton.
    This affect or effect ?! I hope to be helpful accounting for something that seems real and implied but may only be my confusion .
    thank you again

    1. What you can see from the picture is that momentum is being absorbed at the two surfaces of the material. I believe, however, that no energy is absorbed (it would be very hard to explain how one surface absorbs energy from the light while the other surface puts it back — as in your question — since if that happened you’d have one surface heating up and the other cooling down, and soon you’d have a temperature gradient across the material!) Instead I believe the light maintains its energy, and that only its momentum is changed inside the material. However, there seems to be some controversy in the literature about how to think about how the momentum is transferred and stored. Will try to track this down; there is something not so simple going on here.

  14. If this is not the right place to ask this question do you have a recommendation where it would be more appropriate?

    I have asked this question before and never improved my understanding by others responses.

    Take a laser. Fire it at a lens. Prior to arrival at the lens the lasers pulsed beam’s speed is c.
    Traversing the lens it’s speed is less than c.
    Exiting the lens (and back to the vacuum) it’s speed is what?
    c or less than c?

    I’m inclined to think that its speed post lens is the same as its speed pre lens.

    If that is so, didn’t the beam of light “accelerate” post lens? Where did the energy come from?

    1. Good question. Let me refocus it. Suppose the laser doesn’t enter straight on. Then you know that the light will “refract” — it will be bent as it enters the lens, and bent again where it departs the lens. See for example this picture: http://www.flickr.com/photos/physicsclassroom/5079912762/in/set-72157625034624199

      So you can see the momentum of the light changes. How did it change on the way in, and how did it change back to what it was on the way out? (What do you learn from the picture?)

  15. I have two questions:
    1) About the classical Cherenkov radiation: According to the conventional theory, when a charged particle travels through a medium, (some) atoms in the medium become polarized, and emit photons in order to return to
    equilibrium. If the charged particle travels below the speed of light in the medium, the polarizations are not aligned, and no radiation appears.
    If its speed exceeds the speed of light, the polarizations become aligned, and Cherenkov radiation appears. This is a perfect explanation. But I don’t see any difference in the loss of energy for the particle between the two
    situations. Even when no radiation occurs, the atoms are still polarized and that costs energy. Why does the charged particle lose more energy in the case of Cherenkov radiation?
    2) As mentioned above, Cherenkov radiation is caused by interaction
    between traveling charged particles and surrounding atoms. If something travels in vaccum, it cannot interact with anything. How can it produce Cherenkov-like radiation? The claim that it must emit radiation
    at faster than light speed seems to be groundless.
    3) Similarly, the new weak interaction Cherenkov radiation theory of Cohen-Glashow seems to be a postulate without solid basis.

    1. Indeed, Rick, I think your reasoning #1 is correct, the external field created by the charged particle polarizes the atoms of the dielectric medium and once the field disappears (that is, the particle is gone) the atom goes back to its original state by emitting light (photons). This process is happening independently of the particle’s speed. In the case of a superluminal particle a constructive interference of all the emitted photons happens for a particular angle (the cherenkov angle) being the interference destructive for the rest of the angles (for an observer far from the origin of radiation) . In short, the energy lost by the particle is the same either it is superluminal or not.

      A plane when travelling through the air makes noise, not just during the sonic boom which just sounds louder because of the constructive interference of the air waves.

      The radiator is the medium. Can the vacuum be polarized ? Well, in principle, yes, but that´s another story.

  16. I have what may be a nonsensical question. A particle can move through a medium faster than the speed of light in that medium, but is that particle actually traveling faster than a given photon in that medium? I ask because I know it shouldn’t be possible ever to boost into a frame in which a photon would appear stationary, but wouldn’t a superluminal-in-medium particle be able to do so? Also in such a case, the photon’s helicity wouldn’t be intrinsic, right? Is it right to say that a photon in a medium is being absorbed and re-emitted, but itself travels at c while in transit?

      1. I’m definitely just a curious layman, so my information comes from casual reading. In this case, http://www.quantumdiaries.org/2011/06/19/helicity-chirality-mass-and-the-higgs/

        And it’s very likely that since I know nothing serious about relativity, I’ve confused myself. But without a preferred frame of reference, mustn’t a photon travel at c in all frames? Otherwise, some observers would see a photon seemingly at rest and be able to measure its rest mass, which it doesn’t have?

        And thank you very much for the reply.

      2. Actually, please disregard (and disregard the reply that appears below). I’m realizing now that this is a pretty basic question, and that I’m confusing things by using terms however I wish to. I’m also thinking about photons as classical objects, and not as components of a wave with a phase velocity. I suppose this is the problem with never really learning the technical details.

        Thank you for a fantastic blog, and for being so accessible with questions.

  17. Question: Why would high energy particles traveling through the vacuum of outer space emit Cerenkov radiation while they travel through the vacuum? According to your explanation, high energy particles need to travel through some medium in order to emit Cerenkov radiation. Please explain. Thanks.

    1. No, the medium is not the point. The medium is just a way of creating a particular situation: that electrons can travel faster than light in the region of space that they both occupy. If that situation holds, then Cerenkov radiation will occur. So if the situation were true in the vacuum also, then Cerenkov radiation would occur there too.

      1. Okay, then here is a follow up question: assume we have a charged particle traveling at some velocity faster that c in the vacuum of space. I always assumed that particles traveling at a constant velocity do not emit radiation. If so, then why would the particle “emit” Cerenkov radiation? Am I incorrect? Is my assumption only true if the particle is traveling at a velocity less than c? Why?

        1. If Einstein’s theory is exactly correct, then in the vacuum of space, there is no preferred reference frame, and then a particle traveling at a constant velocity is the same as a particle at rest and should not radiate. That’s standard relativity.

          If Einstein’s theory is exactly right and a particle travels faster than light does, then that object is a “tachyon” and naively causality breaks down and leads you to these crazy discussions of time travel and other silliness. But typically what actually happens in the equations is that the vacuum of space becomes unstable, and after it stablizes again you no longer have any particles that can travel faster than light. So we don’t get Cerenkov light in this case, but neither can the situation actually be realized.

          If Einstein’s theory isn’t quite correct and there is a preferred reference frame, then it matters how that particle is moving relative to that preferred frame, and your argument need not be right.

          Or if you are in a medium which means there is a preferred reference frame relative to the medium, then it matters how the particle is moving relative to the medium’s frame.

          There are proposed theories where Einstein’s theory is slightly modified, there is no preferred frame, no violation of causality, and no Cerenkov light even though electrons travel faster than light. I don’t understand these theories enough yet to explain them, but in this case your assumption would hold. However, experiments indicate that these modifications have to be extremely small.

          1. What a great group of responses. I was already formulating a “Why do particles in a medium radiate without accelerating?” question in my head… but your comment about relativity having no preferred reference frame nailed that one. Very nice!

            Thanks so much for your excellent posts and great follow-ups!

  18. Nice article, I was wondering why this effect is so ubiquitous and this post explains it very well! One thing I was wondering, and not sure if this is the right place for it, is how the LHC is able to study matter-antimatter asymmetry using proton-protocol collisions rather than the proton-anti-proton collisions that the Tevatron used. A big (if not the main) component of this is the RICH detectors right?

    1. Hmm… the RICH [Ring-Imaging CHerenkov] detectors don’t have much to do with it per se. ATLAS and CMS can do these studies too and have no RICH detectors. The reason proton-proton is just as good as proton-antiproton is that what is important here is that you can make bottom quarks and antiquarks out of proton-proton collisions — and it is in the properties of bottom quarks that most studies of the physics that is vaguely related to the matter-antimatter asymmetry take place at the LHC. In particular, in a two-proton collision, two gluons, or a quark and an antiquark, can collide and make a bottom quark and bottom antiquark. By studying the hadrons that contain the bottom quark and how they behave (in particular how they decay), one can learn something about how the symmetry “CP” is violated in nature, and from there, very indirectly, with many assumptions, something about the matter-antimatter asymmetry of the universe. [Really there is a very long and winding path from CP violation to learning about the universe’s excess matter.]

Leave a Reply


Buy The Book

A decay of a Higgs boson, as reconstructed by the CMS experiment at the LHC