absorption of a photon

[derek w]

How close does a photon (of the right frequency),have to be to an atom for absorption to take place?

[Sensei]

http://en.wikipedia.org/wiki/Cross_section_%28physics%29

http://en.wikipedia.org/wiki/Absorption_cross_section

[swansont]

There is no simple answer; cross-section is not actually a physical size and location of a photon is not well-defined until the interaction actually takes place.

[Enthalpy]

If the final state isn't quantized, that is for an ionization, light need only a minimum frequency... Easy, and not your question I suppose.

 

Even if the initial and final states are quantized, there is some frequency tolerance. To my understanding, it means that the photon absorption must be quick if the mismatch is big - or reversed, the photon has less time to be absorbed hence has less chances.

 

This can be seen on gasses. A higher gas density reduces the mean time between molecules' shocks; then the light absorption line gets wider, that is more tolerant to frequency mismatch. Though, I believe absorption gets less efficient then; the center (=frequency accurate) of the absorption peak only gets cut away by the too short time, and the less efficient sides (frequency mismatched) remain.

 

As an extreme case, liquids have much broader absorption lines; and this width in Hz due to the mean time between collisions is the same for all frequencies, so that no resonance (=small relative width) is observed at radio frequencies. For instance the 2.45GHz resonance is sharp in low-pressure vapour but flat in liquid water.

 

An other contribution for line width is that the excited state isn't eternal. In addition to shocks among molecules, the excited states returns to fundamental in a finite time, which can be shorter or longer than the free flight. Absorption couldn't take much longer than the de-excitation, so the absorption peak is limited in strength and narrowness.

 

Still an other contribution is the Döppler effect. The absorbing object (molecule...) has a speed which de-tunes its resonance versus the light source, so if one object doesn't absorb a wavelength, an other may.

 

These three contributions (did I forget some?) compete, often with similar importance.

[timo]

Interactions on the quantum field theory level are usually local. In that picture the wave functions should probably overlap for any interaction term to be non-zero.

[derek w]

Thanks for the replies.

So a photon is not a point like object?It always behaves as a field?

[swansont]

That's actually kind of a loaded question. It's point-like in that there is no internal structure. That doesn't mean that it can be arbitrarily localized.

[Enthalpy]

A photon is much like other particles in that the wave appears and disappears in small packets, that some quantities are conserved (charge if any, angular momentum and more). Because of that, we like to count in a unit called particle.

 

At the interaction (the absorption if applicable), the wave can reduce ("collapse") its diversity. It can be very broad, say many light-years if coming from a star, but be absorbed (with the correspondingly tiny probability) at a small place, say one camera pixel at a telescope. This does not need the particle to be a point before its position gets better known. I prefer to say (as others do) that the photon is as big as the wave.

 

Some particles (photon, electron and more) can be detected in a volume as small as humans can do with present technology, and still carry all their properties; then we can call them "point-like".

 

Take a two-slit experiment: The photon is broad before the screen. There it can be absorbed or not; if it passes through, it has the shape of the slits just downstream the screen. Further, the interference gives it the shape of lines - but at a camera, it takes the size of one pixel if it's detected. I do not feel the need to say "the photon can pass through either slit", because I don't imagine a point photon; I say instead "it passes through both".

 

Similarly, a valence electron has the size of an atom - but more accurate tools can detect it in a smaller volume.

 

Other properties, not only the size or position, can have a big diversity but collapse when the particle interaction needs it. For instance the polarization of a photon, an electron (the magnetic momentum then), and more. Some properties can vary continuously for a free particle, others (like the angular and magnetic momentum) have a finite set of possible values.

[timo]

So a photon is not a point like object?It always behaves as a field?

A photon is an excitation of the electromagnetic field (staying in the particle physics language). Similarly, an electron is an excitation in the electron-field. The proton can, in that image, be considered as a complicated excitation in quite a lot of different fields (since it is not an elementary particle).

[derek w]

Wave/particle duality.

When does a photon ever behave like a particle?Even when a photon is absorbed by an atom,the best we can say is its energy has been funnelled into a smaller area.

[Strange]

Wave/particle duality.

When does a photon ever behave like a particle?

 

Particle is a rather poor analogy. It mainly reflects the fact that a photon is quantized.

[swansont]

Wave/particle duality.

When does a photon ever behave like a particle?Even when a photon is absorbed by an atom,the best we can say is its energy has been funnelled into a smaller area.

 

Localized energy transfer is not a wave behavior — it's much more like a particle. The "funneling" can be quite pronounced, even more that a ~ 1 micron wavelength photon being absorbed by an atom that is perhaps 1/1000 of the size. You have some transitions that have a wavelength of a cm or longer, so that's a reduction of ten million. Viewing that as point-like, i.e. a particle, doesn't seem too unreasonable to me.

[Enthalpy]

I don't understand "particle" as "point".

 

We need particles to remember that some quantities appear, disappear, are exchanged by fixed amounts. For instance electricity comes in multiples of q. Or angular momentum is exchanged in multiples of h/2pi. Or light energy appears and disappears in multiples of hF. Then we say "the photon was absorbed" or "the electron has passed", and the idea of a particle is useful.

 

Even the absorption of a photon does not need to be very local. If a semiconductor absorbs a photon, a quite de-localized valence electron becomes a quite de-localized conduction electron, and the absorption occurs over a wide area. Here the photon is a particle, in that its hF energy suffices to give the electron the necessary energy - and this happens if the frequency suffices even if the power density of light is too small: it just happens less often, hence the useful idea of a light quantum.

 

I wrote "a volume as small as humans can do with present technology" because, as far as I know, we have no means to observe down to a point. Since we measure a particle using an other one, which itself has some volume, I understand that a "point-like" particle can be as small as the interaction needs.

[derek w]

 

Localized energy transfer is not a wave behavior — it's much more like a particle. The "funneling" can be quite pronounced, even more that a ~ 1 micron wavelength photon being absorbed by an atom that is perhaps 1/1000 of the size. You have some transitions that have a wavelength of a cm or longer, so that's a reduction of ten million. Viewing that as point-like, i.e. a particle, doesn't seem too unreasonable to me.

Except there are no photons inside an atom.So we still cannot say that the photon is behaving like a particle.

[swansont]

Except there are no photons inside an atom.So we still cannot say that the photon is behaving like a particle.

 

There is no need to think of a photon inside of an atom. You have a wave that has a wavelength of 1 cm, and yet its energy is all deposited in a single atom that's a nm or so in size. Classical waves deposit their energy over the entirety of their extent, which is not happening — if it was a classical wave, all of the atoms in that region would get the appropriate fraction of the energy of the wave. Particles, on the other hand, interact in a localized region.