26

I was studying spectra and suddenly a question popped up relating to the absorption spectra. When we say that the electron absorbs certain wavelengths(photons) so we are implying that white light is a collection of infinite photons of many many wavelengths and the electron simply eats it.

My question is what exactly is white light and how is it different from the monochromatic ones. Is it a bag of infinite photons of different wavelengths or is it a single photon? If it is a single photon then how can electron take up a photon from single photon and still the photon continues with other wavelengths?

Ankit
  • 8,306
chittaranjan rout
  • 547
  • Related: https://physics.stackexchange.com/q/122747/123208 & https://physics.stackexchange.com/a/427865/123208 – PM 2Ring Jan 08 '21 at 18:01

6 Answers6

45

There is some confusion of terms in the question.

  1. A photon is an elementary particle in the standard model of particle physics, see table. Its mass is equal to zero, it is a point particle, and its energy is equal to $h*ν$, where $h$ is planck's constant, $ν$ is the frequency for the classical electromagnetic wave, light, that emerges from a large number of such photons. As far as the photon is concerned the term "frequency" has no meaning other to identify its energy.

  2. the electron is also a point particle in the same table with a fixed invariant mass of 0.51099895 MeV, which is invariant. In no way a free electron can absorb a photon, a photon can scatter off an electron, its energy becoming less. Absorption of photons can only happen in scatters of photons with bound electrons in energy levels, in atoms, molecules and lattices . It is the whole atom that absorbs the photon, the electron changing energy levels due to the absorption. The energy levels have a width, and that is reflected in the ability of atoms to absorb photons with a $Δ(E)$ in energy, which width is directly related to the frequency of the light of multitudes of photons.

  3. The colors of the spectrum are not one to one with the colors our eyes have defined. The spectrum from a crystal have specific frequencies that we have named with the color we see, and there, there is a one to one correspondence, frequency to color. Note there is no "white" in the spectrum:

visible spectrum

But our eyes can see the same named colors with a combination of light frequencies, called color perception:

colorperc

The color perceived at point T , comes from a combination of frequencies, and many different pairs give the same perceived color. White in this plot is around the achromatic point. Please read the link for details.

In summary, white is not a color in the visible light spectrum, many frequencies could make up the perception of white color, which means that photons of a large variation in energies make up the white color.

Is it a bag of infinite photons of different wavelengths

The figure shows how the frequencies combine to give the perception of white. One needs many photons for our eyes to be able to perceive them, but even a few hundreds can give a signal to the brain, this link might interest you.

or is it a single photon.

A single photon cannot give the perception of white.

Hope this helps.

Edit: Since comments might disappear if there are too many, I copy here a significant comment by @PhysicsTeacher:

but it should be noted that when speaking generally of "white light" one often means light that contains all the spectrum to a significant degree, rather than just a combination of a few frequencies. This is because the context is often that of illumination, and illumination with a weird and tiny frequency combination will result in distorted, "artificial" colors ratther than the "real" colors (i.e. the colors seen in daylight). –

anna v
  • 233,453
  • 5
    Your answer made me realize that all the fuss about pink not being a real (spectral) color is (should be) overshadowed by the fact that white isn't either! – Michael Jan 08 '21 at 06:36
  • @Michael in fact, all colors that you can actually see are real. But there does exist the concept of "unreal" colors — impossible colors. – Ruslan Jan 08 '21 at 14:07
  • 5
    "A single photon cannot give the perception of white." — this is not quite true. Some of the cone cells (in fact, more than half of the population) in the retina generate achromatic percepts, despite having the usual spectral sensitivity, see e.g. this paper. – Ruslan Jan 08 '21 at 14:16
  • 2
    Thank you for this really good answer. Until now, I've never understood that color perception diagram. – fool4jesus Jan 08 '21 at 14:31
  • @anna v, in your comment of the figure you say that "The combination of three pairs as described in the figure, gives the perception of white". Unless I'm missing something, these three pairs give the perception of color T, which is not white but light green... – Vorbis Jan 08 '21 at 15:16
  • 1
    @Ruslan if you read the link I gave last, it is still under research if one photon can excite enough the cones to give a flash let alone a color perception.,The link you give speaks of "flashes of light", light is hundreds of thousands of photons. – anna v Jan 08 '21 at 15:54
  • I'd upvote this twice if I could, It's really helped my understanding of photons and colour perception. Given that I work in the field of video encoding, the last is quite important! – SiHa Jan 08 '21 at 16:31
  • A key point is that there are many ways of arriving at white. Most white LEDs combine a blue LED with a yellow phosphor, so you don't even need all 3 RGB primaries. – Mark Ransom Jan 09 '21 at 05:37
  • @MarkRansom I finally think that is what the plot shows for white light that is why I edited last.time. the triplets are a matter of illustration. – anna v Jan 09 '21 at 05:45
  • Sorry if my comment sounded like a criticism, it wasn't - rather it was meant to bolster your point. There are infinite combinations of wavelengths that look like "white" to us, and the diagrams we use to interpret color can sometimes obscure that fact. – Mark Ransom Jan 09 '21 at 06:05
  • @MarkRansom not to worry, we agree :). I had received a negative vote overnight (just woke up) and I thought your comment was related in my mistaken, before edit, statement that as with the T point in the colored sections white would also need three lines, then I reread the link and saw that two would be enough for the achromatic (white) region. – anna v Jan 09 '21 at 07:25
  • @Vorbis please see my edit, for the white region two lines are enough, which I understood rereading the link. The T is for the chromatic regions, the E is the center of the achromatic region – anna v Jan 09 '21 at 07:28
  • I have to disagree with you, you say photons do not have mass, it is only true if photons were static, it is true they don't have rest mass, but there are not static photons; Photons do have energy, and according to relativity, energy is mass so mass of photons is E/c^2 and hf/c^2 however small it is... –  Jan 09 '21 at 13:22
  • @XeнεiΞэnвϵς You are not correct in your opinions , main stream physics photons have zero mass and obey four vector algebra , their energy equal to momentum in the frame c=1 http://hyperphysics.phy-astr.gsu.edu/hbase/Relativ/vec4.html . – anna v Jan 09 '21 at 14:08
  • So this equation:E=m*c^2, it is incorrect, as photons do not conform to it? –  Jan 09 '21 at 14:18
  • 1
    it is correct only at the center of mass rest vrame of a massive particle.zero mass particles do not have a rest frame as they always move with velocity c. . A particle moving with velocity close to the velocity of light displays an inertial mass according to $E-mc^2$ , a relativistic mass which is a function of velocity, so is no longer in use in studying particles. Ist only real use would be for spaceships traveling close to the velocity of light to see how much extra fuel they would need to go faster. – anna v Jan 09 '21 at 15:20
  • 2
    Excellent answer, but it should be noted that when speaking generally of "white light" one often means light that contains all the spectrum to a significant degree, rather than just a combination of a few frequencies. This is because the context is often that of illumination, and illumination with a weird and tiny frequency combination will result in distorted, "artificial" colors ratther than the "real" colors (i.e. the colors seen in daylight). – PhysicsTeacher Jan 09 '21 at 20:18
  • 1
    @PhysicsTeacher the concept you're talking about is called Metamerism. Light bulb makers have a measurement called CRI that attempts to quantify the effect, but it's not perfect. I insist on using incandescent bulbs to illuminate product pictures because they generate true white light, even though they're getting harder and harder to find. – Mark Ransom Mar 28 '21 at 02:43
22

It is very important to understand that white is not a spectral color, but a perceived color. Why?

enter image description here

White is not a spectral color. It's a perceived color. The human eye has three kinds of color receptors, commonly called red, green, and blue. There's no point on the spectrum that you could label as "white". White is a mixture of colors such that our eyes and brain can't distinguish which of red, green, or blue is the winner.

How much red, blue, and green does white light have?

There is no point in the spectrum that you could label as white. White is perceived as white color by our brain, made up of a combination of different colors, that out eyes' receptors sense. It is a commonly believed that there are three different types of cones for red, green and blue. In reality, these cones are sensing short (peaked at 445nm, we call it as blue), mid range (peaked at 535nm, we call it green), and long wavelength range (peaked at 575nm, we call it red) photons.

These receptors have a sensitivity range, that peaks at those wavelengths, but they are sensitive almost through the entire visible spectrum.

enter image description here

It is very important that there can be many different combinations of (different wavelength) photons that can give the perception of white. It is our brain that combines these signals coming from the receptors into a perception of white, and we cannot distinguish how this combination is reached, they will all produce the perception of white.

Yes, a photon by itself can be in a quantum superposition of different frequencies, which one might call "white". No, such a photon probably can't be produced by a simple natural process. No, such a photon would not look white, because the superposition collapses upon measurement, giving only one frequency. (Only one of your cone cells could possibly fire in response, assuming that any even fire at all.) However, a collection of many such photons would collectively look white.

Does a single white photon exist?

Now you are asking about a single photon, but a single photon cannot create the perception of white, because you need multiple photons, with certain different wavelengths to be able to create the perception of white in our brain. Please note, that however, a single photon is a QM entity, and it is possible for a single photon to be in a superposition of states so, that it could be interpreted as a combination of colors that could create white, but as the single photon interacts with the cones in the eye, its superposition collapses into an eigenstate with a single wavelength and thus cannot create the perception of white.

Árpád Szendrei
  • 28,452
17

At first glance, the question has a trivial answer: white light is light that contains a roughly uniform mixture of photons of all visible wavelengths. Light can appear white when it has a non-uniform mixture of wavelengths that excite the three color receptors in the human retina the same way a uniform mixture does.

HOWEVER, that answer does not address the question that the OP seems really to be asking: "Can a photon be 'white', or must it be only a single wavelength?"

In fact, the wave function of any photon has a finite spectral width. A suitably constructed light source can produce photons having very large spectral width, spanning the entire visible spectrum. If the wavelength of any one such photon is measured, of course only one wavelength will be obtained; but repeated measurements will obtain wavelengths that span the full spectrum of the source. Absorption of a photon by an atom or molecule is equivalent to a measurement of the photon.

S. McGrew
  • 24,774
  • What are the spectral widths of the photons emitted by the Sun? Are those mostly narrowband photons of mixed frequencies or broadband homogenous “white” photons? – Prof. Legolasov Jan 07 '21 at 16:00
  • @Prof.Legolasov may this answershttps://en.wikipedia.org/wiki/Sunlight#Composition_and_power – anna v Jan 07 '21 at 16:37
  • 3
    @Prof.Legolasov Unfortunately it is not possible to measure the spectral width of a single photon from the Sun, or from any other source. The spectral widths of the wavefunctions of a large number of identically produced photons from a source can be determined statistically. – S. McGrew Jan 07 '21 at 16:55
  • @S.McGrew I find it hard to believe that a measurement is impossible in principle for sunlight. Even if it is impossible in practice, we should be able to predict this using astrophysics – Prof. Legolasov Jan 08 '21 at 00:55
  • @Prof.Legolasov The light we receive from the Sun is mostly blackbody radiation, and blackbody radiation is not a bunch of "white" photons. Such a distribution would have low entropy, since all the photons would be in the same quantum state, whereas blackbody radiation has high entropy. – Brian Bi Jan 08 '21 at 01:31
  • @BrianBi still, there must be a way to characterize the widths of individual photons, or at least the distribution of widths – Prof. Legolasov Jan 08 '21 at 01:43
  • @Prof.Legolasov I think there are people on this site who know the properties of blackbody radiation. I don't know anything other than what I said in my previous comment. Your question will get more attention if you post it as a new question. – Brian Bi Jan 08 '21 at 01:44
  • 1
    I daresay “any photon has a finite spectral width” is, if not just wrong, then at least more confusing than helpful. Photons don't contain the information of how they were created in any meaningful (measurable) sense. They're just excitations of the EM field. Spectral uncertainty comes from certain properties of the whole system under consideration, not from anything about the individual photons. – leftaroundabout Jan 08 '21 at 11:49
  • @leftaroundabout, My statement was that *the wavefunction of any photon has a finite spectral width", not that the photon itself does (unless "photon" refers to the photon wavefunction). There is a whole fascinating and enlightening conversation to be had around this concept. If you'd like, we can continue in chat. – S. McGrew Jan 08 '21 at 14:56
  • The most popular way of expressing the concept of photon is that it is the excitation of an EM mode. This implies single frequency. I'm aware that there are expressions of photon that allow a distribution of frequencies, but they are not commonly used, and are not as useful as the single-frequency expression. – garyp Jan 09 '21 at 00:32
  • It's pretty easy to set up an experiment that shows photons to have a finite spectral width, such as doing single-photon interferometry in which the relative path lengths are varied. E.g., if the wavefunction of a single photon is monochromatic, its spatial extent along its path must be infinite, but in fact the experiment will simply produce a result that echoes the temporal coherence (which is essentially the inverse of the bandwidth) of the source. – S. McGrew Jan 09 '21 at 01:18
  • @garyp A preferred basis then, huh? To me a single excitation means a+ applied on vacuum, and that a+ can be any nontrivial linear combination of an arbitrary number of other creation operators. In fact, it's the monochromatic excitations that can't exist, because they aren't normalizable! – The Vee Jan 09 '21 at 17:46
  • @TheVee Then $E = h\bar{\nu}$? How can $a+$ deliver the fraction of energy needed to make up the energy of the photon? When the white light interacts with an isolated atom, what happens? One particular $a-$ operates, the one at the transition energy? What happens to all the other modes. If you are considering an ensemble I might understand better. I have it in my head that $a+$ raises the excitation of a mode by $h\nu$, and a mode is monochrormatic by definition. We shouldn't discuss it here, but if you can point me to a resource I'd be grateful. – garyp Jan 09 '21 at 21:01
  • Keep in mind that the wavefunction represents a probability distribution. If a detector (e.g., an atom) is in the path of a broad-spectrum wavefunction, and the detector is only sensitive to a very narrow frequency range, the likelihood that it will detect the photon depends on the fraction of the wavefunction's spectrum that falls within that range. – S. McGrew Jan 09 '21 at 22:23
  • @garyp E is an observable. It will come with a probability distribution. So does any other observable in quantum mechanics, but as opposed to many other examples, E does not have a discrete spectrum so in reality it can never have an exact value (eigenvectors). In interaction with matter, e.g., absorption, a wavefunction collapse may happen, so seemingly an "exact" energy may be observed, but nothing prevents it from having an arbitrarily broad statistics, as nicely summed by S. McGrew's comment below mine. – The Vee Jan 10 '21 at 17:36
  • Really, it's nothing that wouldn't be said in any QM textbook, if you accept the view that $E = ħω$ is a relation between observables rather than classical numbers. – The Vee Jan 10 '21 at 17:37
13

There's a bit of a definition or situational problem here. White light MIGHT be an even distribution of visible frequencies, but it might not. From a perceptual perspective, white light is a mixture of frequencies that stimulate the cones of the human visual system in such a manner that they produce the sensation we call "white".

Unlike the other answers, this mixture does not have to be a uniform mixture of frequencies. If you have a good* magnifying glass handy, look at a white area on your computer screen. You will see that it is composed of tiny red, green, and blue dots, with none of the other spectral colors.

It's also a matter of perception. If you have ever taken pictures of a snowy landscape near sunset, you have probably noticed that what you see as white snow, the camera sees as reddish-orange. The brain adjusts (within limits) what you see to what you expect to see - white snow.

*Has to be a good one, or an older low dot pitch display. With my display and desk magnifier, it's barely possible to make out the dots.

jamesqf
  • 1,308
  • 1
    I daresay that the physics meaning of "white light" necessitates all frequencies. – user253751 Jan 08 '21 at 17:05
  • @user253751: But that is just linguistic shorthand for "evenly distributed over all (visible?) frequencies". Indeed, if we consider sunlight as white light, it's not evenly distributed in the visible light range, being rather more intense in the blue than the red: https://www.fondriest.com/environmental-measurements/parameters/weather/photosynthetically-active-radiation/ – jamesqf Jan 08 '21 at 19:07
  • In fact, it's sufficient to have two spectral lines (and nothing else) in the spectrum to see the light as white. E.g. $571,\mathrm{nm}$ and $460,\mathrm{nm}$ (or near these) in appropriate proportions. – Ruslan Feb 14 '21 at 14:19
  • @Ruslan: Interesting. I would have thought it would take at least three, one for each type of cone cell. Though if the light's dim enough to activate only rods, it's always perceived as white. Which just goes to show that physics and perception are quite different. – jamesqf Feb 15 '21 at 17:29
  • Just look at the chromaticity diagram. Choose a white point (e.g. D65, as for sRGB) and draw a straight line through it, trying not to intersect the line of purples (you'll still have some room to choose the slope). You'll get two intersections of this line with the gamut boundary, which are the two wavelengths you need to achieve your chosen white point. Then it's just the matter of weighing the powers of light sources to move the mixture along the line to achieve the white point. – Ruslan Feb 15 '21 at 17:42
  • @Ruslan you might be interested to know that's how most white LEDs work. A blue LED is coated with a phosphor that absorbs some of the blue and emits yellow. Now the yellow isn't a single frequency, but it's balanced to be complimentary to the blue. And different formulations will give different white points, which is why there are both warm and cool white LEDs. – Mark Ransom Jan 18 '22 at 04:47
  • @MarkRansom yeah, I know it pretty well, thanks :) – Ruslan Jan 18 '22 at 08:01
6

White light can result from combination of different monochromatic sources, as used in TV screens for example.

But that doesn't mean in my opinion, that all white light results from a mix like that. It is only a trick to get some sensorial effect, as real movement are simulated by sequences of pictures in movies.

The sunlight results from a chaotic movement of charges of the ionized H and He at the surface of the sun. That plane EM waves coming to earth produce in our eyes a "white" sensation.

Colours result from the interaction of that light with matter (diffraction gratings, prisms, or simply selected absortion by material surfaces).

Claudio Saspinski
  • 16,319
6

Your main confusion seems to be whether "whiteness" is a property of a particular photon, or of a collection of them, and the answer to that question is that it is a property of a collection of photons. However, you contrast a single photon to an infinite number of photons. While the number of photons emitted by a typical light source is vast (for instance, in one second, a typical light bulb will emit more than a billion times as many photons as there are people on Earth), it's not infinite.

As for your literal question title, light being "white" is more of a biological matter than a physics matter. When light is described as "white", that refers to it having a distribution of wavelengths that humans perceive as "white". And human perception of "white" is context dependent; the human brain actually has a tendency to "normalize" ambient light to white, so that you'll be able to recognize that the same object has the "same" color regardless of light source. "White" light will generally refer to an "even" distribution of light, but the exact definition of "even" varies by context. When dealing with light sources that can be reasonably modeled as black body radiation, "white" is often defined as corresponding to a particular temperature range. (It may seem weird to say that white light is light that comes from black body radiation, but that's a matter for a different question.)

Acccumulation
  • 8,745