Photons Are Particles

Over at Dot Physics, Rhett is taking another whack at photons. If you recall, the last time he did this wasn’t too successful, and this round fares no better:

So back to the photon. In my original post I made the claim that the photoelectric effect is not a great experiment to show photons. Maybe that is not how it came off, but that is what I meant. The photoelectric effect can be explained quite well with the classical electromagnetic waves model and a quantum nature of matter. Of course there is a quantum nature to light as well.

I think the biggest problem with the photon is that the manner it is introduced encourages students to think of it as an actual particle. One thing about particles is that they are localized. I am pretty sure that even quantized light (the real photon) is not confined to a set space.

I’m no field theorist, but I can’t figure out what that last sentence is supposed to mean. I think it’s probably intended to be a reference to the idea of a photon as an excitation of a particular mode of the electromagnetic field, with those modes necessarily being extended over some region of space. But if you want to say that that disqualifies photons from being “actual particles,” then an electron isn’t an “actual particle,” either– the same field-excitation language works for electrons, as well as photons.

The fact is, photons are particles, in every way that matters.

When you detect photons with a CCD camera, they show up as discrete localized spots on the CCD. When you put photons in a cavity (two mirrors facing one another), you can see discrete, localized excitations of that cavity– you can “count” the number of photons in the cavity. When you look at the absorption and emission of light by single atoms, you see discrete jumps in the energy of both the light field and the atom itself.

Any test you can devise for particle-like properties of light, the photon will pass.

Now, it’s true that you can explain the photoelectric effect in semi-classical terms, with light as a continuous classical electromagnetic wave. Given that, why do we teach about photons and the photoelectric effect in introductory modern physics classes? Three reasons:

1) The semi-classical explanation of the photoelectric effect is much more complicated than the photon model. It requires the idea of energy quantization of electrons in a solid (basically, you treat it as an excitation from a discrete quantum state in the solid to a continuum of free-particle states outside the metal), and some understanding of the coupling between light and atoms. These concepts are well beyond students in introductory classes, and the math involved (the Fermi “Golden Rule” for transition probabilities) is way out of reach for those students.

2) The photoelectric effect gets at the idea of energy quantization for light, and the relationship between energy and frequency. This is essential when talking about atomic states– you need to know that light carries energy proportional to the frequency to explain the observed spectra of atoms, whether that’s via the Bohr model or a full quantum treatment of hydrogen. Spectroscopy is the best tool we have for studying the internal structure of atoms, and understanding spectroscopy requires understanding the energy quantization of light.

3) It’s a pedagogically useful experiment. Not only can you do the photoelectric effect in lab, but you can easily use it to set up two-equations-two-unknowns problems, which are great practice for students at the intro level. You also use the same photon concept to understand the Compton effect (another experiment that has a very difficult semi-classical explanation, and a really simple photon-based explanation), which makes another excellent sophomore-level lab.

And the idea of quantized light is all over modern physics. The notion of photons as particles with discrete energy and momentum turns up in atomic physics, condensed matter physics, nuclear physics, particle physics– just about any active field of physics research will use the photon concept at some level.

Photons are particles, in every meaningful sense of the word, and students need to know that photons are particles. You cannot claim to have even the most rudimentary grasp of modern physics without knowing that photons are particles.

Comments

  1. #1 gg
    February 25, 2009

    “Given that, why do we teach about photons and the photoelectric effect in introductory modern physics classes?”

    One other reason, which shows my bias, is that historically the photoelectric effect did convince people that light has particle properties. Subtleties aside, this was a turning point for quantum mechanics.

  2. #2 Matt Springer
    February 25, 2009

    Precisely so. The fact that the photoelectric effect can be explained semiclassically is similar in practice to the fact that Rutherford scattering can be explained classically. Sure it’s possible, but why would you want to go through so much trouble to create an epicycle-style explanation for a more basic underlying concept?

    That said, I’m willing to make a point in his defense – one sense in which photons are somewhat different from your average massive particle is that it’s not necessarily possible to cleanly define a photon wavefunction depending on what you mean by “wavefunction”.

  3. #3 Moshe
    February 25, 2009

    The question is ambitious, because when you look close enough neither electrons nor photons can be described as particles. Rather, they are excitations of a continuous field, which at certain limits can be arranged in definite number of localizable excitations. That number is usually not preserved by interactions, so when those are important, the definite particle number basis is not very useful.

    The reason why we can sometimes describe the electron by single particle QM, and not the photon, is because there is a non-relativistic limit for the former, but not for the latter. If your energy is lower than the electron mass, the number of electrons will be (approximately) conserved, so you can work with definite number of them (say one). There is no such limit for photons, and for general processes the particle number basis is inconvenient. Maybe this is a more elaborate version of that paragraph, maybe not.

  4. #4 Moshe
    February 25, 2009

    I meant ambiguous in the first sentence, that typo really makes that sentence mysterious.

  5. #5 D. C. Sessions
    February 25, 2009

    Sometimes there are multiple isomorphic abstractions for the same thing, with the “best” being application specific.

    Pedagogically, “best” is usually the simplest. Alas, that is not always the one used.

  6. #6 Greg Laden
    February 25, 2009

    Best pun of the day:

    The fact is, photons are particles, in every way that matters.

  7. #7 Alex Besogonov
    February 25, 2009

    I remember Millikan’s (?) experiment – they took a small charged piece of dust and suspended it in the air using electric field. And then shone light on it.

    About once a half hour photons knocked off electron from piece of dust. If light behaved only as a wave – it wouldn’t have happened. I’m too tired to explain why :)

  8. #8 Mary
    February 25, 2009

    I sympathize strongly with the views expressed at Dot Physics. I find that thinking of a photon as a particle hurts more than it helps when I am trying to reason about the behavior of light. You may well say that if photons are not particles than neither are electrons — I also find thinking of electrons as waves (or at any rate as clouds of charge shaped by a wavefunction) more helpful than thinking of them as particles, in most circumstances.

    You say “Photons are particles, in every meaningful sense of the word” but there is an important sense in which they are not — they do not usually behave anything like particles of dust or sand. In fact neither photons nor electrons are “particles” or “waves” in the usual senses of those words, associated with dust and oceans. Both words are potentially helpful metaphors, but I find that for most aspects of photon (and single electron) behavior, the wave metaphor is really much better at helping me intuit how a system will evolve. The particle metaphor confused me for a long time as an undergraduate and new grad student, and gave me very misleading ideas about the location and velocity of photons…

    I generally think light as a blurry wave, like one drawn in pencil and then smudged horizontally and vertically. In the single photon case, the wave becomes so smeary (the blur has a gaussian profile…) that it doesn’t really look like a wave at all any more because the phase is impossible to distinguish. Nevertheless it is still in some sense a wave, with an associated wavelength and frequency, and the phase uncertainty can be collapsed by a measurement…

    The particle-like properties of photons (like “clicks” on a detector) can frequently be understood in terms of the quantized states of the atoms they’re interacting with. When the quantization of the field itself becomes relevant, I find the picture of the smeary-wave much more helpful than a picture of a particle, for reasoning about what will happen. How can a particle have a squeezed state? How can it have a frequency at all? How does a particle make an electron oscillate? And how can a particle go through two slits at once, or both ways at a beamsplitter? All of these phenomena are better in terms of waves where the amplitude and phase are described by wavefunctions like those that describe a harmonic oscillator. Quantized energy states, sure, but no more a particle than a standing wave on a drum head is.

    I recognize that it’s a matter of taste, though, since duality says they’re neither.

  9. #9 Rhett
    February 25, 2009

    I hate it when I suck. But really, Maxwell’s equations are awesome in so many ways, do we have to destroy them? I don’t think so. In classical EM radiation, fields are everywhere and not confined. When you think of a particle, you think confined to a certain space.

    Either way, I think that using photoelectric effect to talk about photons isn’t needed.

  10. #10 Ambitwistor
    February 25, 2009

    This brief post by John Baez describes some ways in which photons might be considered somewhat less “particle-like” than ordinary matter. (As alluded to, there was a whole sci.physics thread on this question about 10 years back.) However, I agree, as far as QFT is concerned, both photons and electrons are just field excitations.

  11. #11 lfmorgan
    February 25, 2009

    MatterAsElectrons

    Ode to Massive Denial Of Reality(MDOR)

    To be uniquely human is to want to know;
    Anything else is just putting on a good show.
    Does common ignorance binds us into a group?
    Can we be really smart and still be in the loop?
    The way we are going now it seems to be clear,
    Evolution does not favor Survival Dear.
    Whole civilizations have come and gone before
    This one may have plenty of reason to feel sore.

    Science and technology today are Physics-Ignorant beyond belief

    On NASA TV today, the best of our scientists confess ignorance about what mass is—yet it fills the overwhelming majority of what we call The Universe. They call it “Dark Matter”- -meaning that it is matter we cannot see, yet shares the property we call “mass” –whose movement causes energy and force. The energy and force of visible matter is a given. Dark matter has to be (1) locally too small in magnitude to register, (2) else not available to our limited view using reflected light to see by. Neither today nor in 1946 when deciding to major in Physics has this author found public answers to these burning questions about physical reality. Profoundly curious about what defines the outer “visible” surface of the atom, his imagination insisted on a “minds eye’’ that could travel to the surface of an atom to see what it must be made of. In studying Math and Physics from 1946-1950, the best of prof’s were of little help in this need to visualize physical reality. Only in early (1987) retirement as a Lockheed Senior Scientist and Systems engineer, did on one begin to see that the atom had to be a spherical array of invisible circulating dark matter whose speed of motion must be said to equal to the speed of light so Einstein’s E = Mc2 made good sense. The proper picture is one of ever-traveling electrons: at the outer surface of the atom, inside, and in open space away from atoms. The success of Einstein’s predictions make it necessary! Visible matter is dark matter bound and shaped uniquely by closed loop electrons whose numerical count equals the visible matter mass. Since there can be no “nothing” anywhere at any time, electrons as mass in motion have to fill all space with what we call mass, energy and force. That means gravity force has to be due to the systematic inward travel of electrons far beyond the number coming upward. Only an electron-sucking black hole (BH) at the center of every gravity field fills the bill! As long as the gravity field is visible and intact, the BH must be growing in size due to prodigious electron absorption to support the concentric-arrayed force of gravity that reaches a maximum at the BH outer surface. This new physics theory has been posted on the internet since 1990. Please click on the following link for more info.
    http://www.geocities.com/CapeCanaveral/Hall/2638/1MrMorganNewPhysics.doc

    ElectronReality

    Ode to the electron as physical reality

    The word “anthropic” now needs to include it all,
    Of how we are made, see, feel, run fast or just crawl.
    Our Brain best uses electrons in great number,
    Whether fully awake or in deepest slumber.
    All forms of force must use electrons in motion;
    They really are worthy of Mind’s best devotion.

    Moving mass is frequency-pulsed electrons per E = nhf, where n = 1/h

    Where the speed of electron flow can be said to be constant at v, E = Mv2 º (nh)f where n is the number of electrons being pulsed in parallel, h is Planck’s constant, f is the frequency of pulsing per second. By letting the electron mass be h grams and n = 1/h , nh = 1 unit of mass so that f and v2 ate numerical identities! It is incredible to contemplate, but that is what our consistent measure has always shown —that frequency of pulsing, f, is always the direct measure also of the linear energy of the pulsing, v2! Now see that radiation energy at any frequency, f, moves along a line of spinning R-size material points that are touching so that they form a ray of radiation , where R is the replacement sphere radius of the ray-line defined by the touching R’s. If the pulse speed of movement of the electron in going from one material point R to another spin touching next door is set to v, then the linear energy of the ray of radiation as it moves from R to R in a world line way is v2. We never bothered to explain this elementary physical quality of a ray of radiation along with why f is always numerically equal to v2! It seems likely that this huge oversight is because we never properly related mass and force as applied at the frequency pulse level. The h in the Planck relation is constant because it is the constant mass of an electron being pulse accelerated in open space where v = c to make the Einstein relation, E = Mc2 º (nh)f work! The applied force of electron acceleration in open space is F = Ma = ha where h is h grams and a is the pulse acceleration of the electron in going from one R position to the next R in line in a given direction. For a 3-D reality, there are 6 degrees of freedom and therefore 6 R’s that are spin touching a given R. Since v is both the pulse speed and pulse acceleration of the electron in one of six possible directions, F = Ma º hc, the Hubble constant for light as continuous electron energy transfer! If we assume spin speed v = c for the electron that defines the outer spin boundary of the atom as an R-size material point, then we begin to understand E = Mc2 as it applies to nuclear energy! Existing explanations for nuclear energy and for light travel are both complexly wrong! New very elementary physics must replace secrets now guarded at great expense! More importantly, new physics can be adequately taught in elementary school so even an 8 year old around the globe knows all the great secrets about physical reality! Current PhD’s will need “unlearning” courses in elementary new physics. For much more “Totally Free New Physics”, click on:
    http://www.geocities.com/CapeCanaveral/Hall/2638/1MrMorganNewPhysics.doc


    You got time to go, you got time to google!

  12. #12 onymous
    February 25, 2009

    But really, Maxwell’s equations are awesome in so many ways, do we have to destroy them? I don’t think so.

    Then you’re in luck, because they’re still the equations that describe photons! Just, you know, quantized.

  13. #13 Peter Morgan
    February 25, 2009

    What Mary said, what John Baez said (thanks Ambitwistor). Photons are problematic as particles. The Unruh effect should be added to the list of delicate questions. Try Brit. J. Phil. Sci. 52 (2001) 417-470, also at http://arxiv.org/abs/quant-ph/0008030 for a good account that crosses well between math and philosophy of physics.

  14. #14 Bouncing Bosons
    February 25, 2009

    Wave-particle duality is, I think a terrible idea that encourages students to cling to classical analogies that are wholly inadequate for the quantum nature of things. A “particle”, in the quantum sense, is something which, though it is not necessarily localized, is indivisible and the laws governing it’s propagation through spacetime give wave equations as the stationary action (classical) equations of motion. Any discussion of quantum particles as classical particles or as classical waves is incorrect, or at best an approximation that one must be very careful generalizing.

  15. #15 Markk
    February 26, 2009

    It seems like there is an underlying disagreement here, in that the concept of “particle” means something different to different people. This is a social not a technical issue really. If you use the quantum excitation of a field definition then Chad’s view falls out naturally. If you use a classical definition then nothing in physics at the atomic or smaller level is a particle “like a grain of sand” and so of course photons are not like grains of sand to a high degree, but neither is anything else.

    Almost all people discussing the Standard Model all do seem to use the word particle in for these fundamental things at the base – quarks with gluons, leptons with photons and Z and W’s, neutrino’s. So I think that should be the normative use of the word. It is even used when things get a little higher level – quasiparticles, quantized excitations of different fields.

    Yes photons are different than electrons in SM physics and we precisely know the difference in how the model treats them, but they are still generally called particles.

  16. #16 rob
    February 26, 2009

    BEDEMIR: And that, my liege, is how we know the gravity force has to be due to the systematic inward travel of electrons far beyond the number coming upward.
    ARTHUR: This new learning amazes me, Sir Bedemir. Explain again how existing explanations for nuclear energy and for light travel are both complexly wrong!
    BEDEMIR: Oh, certainly, sir.
    LAUNCELOT: Look, my liege!
    ARTHUR: Photons!
    GALAHAD: Photons!
    LAUNCELOT: Photons!
    PATSY: It’s only quantized excitations of particular modes of the electromagnetic field.

  17. #17 DA
    February 26, 2009

    I have to agree with “bouncing bosons:”

    One of my issues with discussions like this is that we are using words like wave and particle which were developed with the classical world in mind. And they carry all the connotations that classical objects have. This is silly.

    At the quantum level, these things stop being applicable, but we attempt to use the same words to describe it. Is it a particle? Is it a wave? Neither. It is a quantum object represented by a wavefunction and/or state vector and how it acts depends on the POVMs of your measurement device.

    Does that sound complicated? It is. Feel free to use particles/waves to describe it to undergraduates, but don’t be surprised when there is confusion and there is no perfect way of describing it in those terms — there isn’t a perfect way.

  18. #18 MarkH
    February 26, 2009

    Photons are wavicles. If you measure them like particles they’re particles, and if you measure them like waves they’re waves. Wavicles. Or partaves. But wavicles sound tastier.

  19. #19 DCBob
    February 27, 2009

    Ladies and gentlemen, do not forget the words of the Amazing Feynman: “Things on a very small scale behave like nothing that you have any direct experience about. They do not behave like waves, they do not behave like particles, they do not behave like clouds, or billiard balls, or weights on springs, or like anything that you have ever seen.”

  20. #20 CCPhysicist
    February 28, 2009

    Question:
    Why don’t you start your own class in your lab, given that you have photon particles and particle waves in all their glory right there in the building? What better way to show your hot shot classical physicists that everything they know is wrong?

    Comment:
    I was with you right up until you said the Compton effect could be explained semi-classically. I don’t know of any way to explain the angular distribution based on classical radiation by a free particle. More importantly, since there is a lot I don’t know, neither did Compton. Reference?

    I think Compton scattering is a cleaner starting point, and suspect it loses out due to the historical sequence rather than pedagogy. What is simpler than the scattering of one free particle by another? Similarly, what is simpler than the diffraction of electrons by a lattice if you want to show that electrons are as non-particle as photons are as regards going through two slits? That helps address how a particle can have a frequency, just as the former shows that a photon can have energy and momentum without having mass (no matter what lfmorgan asserts to the contrary).

  21. #21 Chad Orzel
    February 28, 2009

    Why don’t you start your own class in your lab, given that you have photon particles and particle waves in all their glory right there in the building?

    Because there are only so many hours in the day.
    It’s certainly possible to do single-photon demonstrations and experiments at the undergrad level, but getting that going takes time, and I have a lot of other things on my plate.

    I was with you right up until you said the Compton effect could be explained semi-classically. I don’t know of any way to explain the angular distribution based on classical radiation by a free particle. More importantly, since there is a lot I don’t know, neither did Compton. Reference?

    That assertion was based on a comment in the Quantum Challenge by Greenstein and Zajonc. There were some papers cited in the book, but the book itself is elsewhere at the moment, so I can’t look them up.

    I looked at the papers in question once, and they didn’t make a great deal of sense on a quick glance. I didn’t have time to do more than glance quickly, so I can’t say much about the quality of the explanation.

  22. #22 moshe
    February 28, 2009

    I am still not clear how can you tell experimentally there is only a single photon involved. Unlike electrons, photons can come with arbitrarily low momentum and energy, how can you tell (given that any physical quantity is measured with finite precision) that you don’t have 27 additional such “soft” photons involved in your experiment, in addition to the one you are interested in?

  23. #23 Chad Orzel
    February 28, 2009

    I am still not clear how can you tell experimentally there is only a single photon involved. Unlike electrons, photons can come with arbitrarily low momentum and energy, how can you tell (given that any physical quantity is measured with finite precision) that you don’t have 27 additional such “soft” photons involved in your experiment, in addition to the one you are interested in?

    In a strict sense, you can’t, but you can tell that there is only a single photon in a particular mode that interests you. There might be a billion photons of arbitrarily low energy hanging around, but most experiments in quanum optics are concerned with photons of a particular energy and momentum, not just any random photon that might happen along.

    You can select a particular mode of interest in a number of different ways, either by filtering thermal sources, by using sources that are known to produce only photons in particular modes (single-atom fluorescence, atomic cascades, parametric down-conversion sources), or by using a cavity to limit the possible modes of the field. They’re all interesting in their own way.

  24. #24 moshe
    February 28, 2009

    OK, I agree, this is a situation I’d also call a single-photon experiment, though I think it’s also important to keep in mind this term cannot be taken too literally. I think the soft-photon issue is one of the reasons you cannot have single-particle QM description of the photon (e.g. write wavefunctions for photons), because their number is indefinite in any physical state.

  25. #25 Sean
    May 31, 2009

    PHYSICISTS are not looking at the entirety of reality.

    We exist within an open stationary 4 dimensional Time-Space environment. All objects have a fixed magnitude of motion. All that can be changed is the direction of travel within the open stationary 4 dimensional Time-Space environment. See http://www.outersecrets.com/real/forum_againstum2.htm

    This constant motion also determins the fastest speed that one may move across time or space.

    This structure also creates two kinds of events. 1) Real-Time Events. 2) Cross-Time Events.

    Picture a Photon having the shape of a string, or something like a stick or pen perhaps. This photon spins end over end, and moves across space. Due to its shape and spin, its path across space is of a corkscrew-like shape. From the open stationary 4 dimensional Time-Space environment point of view, one is also looking across time, thus the entirety of a photons complete corkscrew-like shaped path extending from source to destination is ALL seen.

    Thus if you perform an experiment such as the two slit light experiment, the complete experiment is being governed as a Cross-Time Event and thus the complete corkscrew-like shaped paths of the photons are interfering with each other, thus creating a wave-like interference pattern at the destination wall.

    If, however real time events are added to the experiment, such as detecting photons as they pass through the slits, then the experiment is now governed via Real-Time Events and thus one photon can cross another photons path as long as the other photon is now out of the way of this second photons, thus this becomes particle interference rather than path interference.

    These Cross-Time Events also explain other strange events such as Action at a Distance, Entangled Pairs, etc.

The site is currently under maintenance and will be back shortly. New comments have been disabled during this time, please check back soon.