Photon absorption and electron transition levels.

[avicenna]

It is commonly stated in QM that a bound electron may only absorb a photon if only there is a matching energy gap difference that matches the photon energy. This seems ridiculous.

Say a photon emitted by a hydrogen atom with the typical red emission frequency of the Blamer series. If this photon meets a piece of copper, it is unlikely that a copper atom has an exact matching two energy levels matching that of the Balmer spectrum of hydrogen. The photon would just cruise through copper without being absorbed. Well, if a statement is flawed, we then decide to "patch up" our theory and say there are other means that a photon may be absorbed by matter.

So the scientific method nowadays would be a series of ad hoc "patching ups".      
 

[swansont]

25 minutes ago, avicenna said:

It is commonly stated in QM that a bound electron may only absorb a photon if only there is a matching energy gap difference that matches the photon energy. This seems ridiculous.

It’s observed to be true. Gases tend to be transparent, except for select absorption lines, so you can easily test it. 

 

25 minutes ago, avicenna said:

Say a photon emitted by a hydrogen atom with the typical red emission frequency of the Blamer series. If this photon meets a piece of copper, it is unlikely that a copper atom has an exact matching two energy levels matching that of the Balmer spectrum of hydrogen. The photon would just cruise through copper without being absorbed. Well, if a statement is flawed, we then decide to "patch up" our theory and say there are other means that a photon may be absorbed by matter.

It’s not just one copper atom in a piece of copper, you have some reasonable fraction of Avogadro’s number, and the you have a wider band of possible excitation energies (transitions have an energy width, owing to the uncertainty relation). Excitations in a bulk conductor or semiconductor can give an electron kinetic energy, which is not possible in an atom or molecule.

 

25 minutes ago, avicenna said:

So the scientific method nowadays would be a series of ad hoc "patching ups".      
 

Not ad hoc - you have testable models. But if a model doesn’t work, you fix it.

[exchemist]

50 minutes ago, avicenna said:

It is commonly stated in QM that a bound electron may only absorb a photon if only there is a matching energy gap difference that matches the photon energy. This seems ridiculous.

Say a photon emitted by a hydrogen atom with the typical red emission frequency of the Blamer series. If this photon meets a piece of copper, it is unlikely that a copper atom has an exact matching two energy levels matching that of the Balmer spectrum of hydrogen. The photon would just cruise through copper without being absorbed. Well, if a statement is flawed, we then decide to "patch up" our theory and say there are other means that a photon may be absorbed by matter.

So the scientific method nowadays would be a series of ad hoc "patching ups".      
 

No, spectral lines have finite width for a variety of perfectly good reasons. (Finite line width means there is a range of absorbing or emitting frequencies of course.) These include the Doppler effect, from motion of the emitters relative to absorbers, and uncertainty broadening, due to finite lifetime of the excited state leading to uncertainty in its energy, by Heisenberg's Uncertainty Relations. In gases, this can be a function of pressure, cf. "pressure broadening", since collisions may shorten lifetimes of excited states and also alter their energy, due to transient proximity of second atoms, thereby disturbing the potential experienced by the electrons. And for matter in condensed states, atomic lines tend to get broadened into bands anyway, due to the overlay of vibrational and/or rotational fine structure.

If you read up a bit about spectroscopy, there is quite a bit to it besides simple line emission and absorption. 

[avicenna]

3 minutes ago, exchemist said:

No, spectral lines have finite width for a variety of perfectly good reasons. (Finite line width means there is a range of absorbing or emitting frequencies of course.) These include the Doppler effect, from motion of the emitters relative to absorbers, and uncertainty broadening, due to finite lifetime of the excited state leading to uncertainty in its energy, by Heisenberg's Uncertainty Relations. In gases, this can be a function of pressure, cf. "pressure broadening", since collisions may shorten lifetimes of excited states and also alter their energy, due to transient proximity of second atoms, thereby disturbing the potential experienced by the electrons. And for matter in condensed states, atomic lines tend to get broadened into bands anyway, due to the overlay of vibrational and/or rotational fine structure.

If you read up a bit about spectroscopy, there is quite a bit to it besides simple line emission and absorption. 

Thanks. OK. The world of knowledge is vast as the universe itself. 

[exchemist]

8 minutes ago, avicenna said:

Thanks. OK. The world of knowledge is vast as the universe itself. 

Probably, but you can still tackle it in manageable chunks by learning selectively those bits that interest you.

For me spectroscopy was one of them, after a hairy first term at university reading and having stiff tutorials based on Gerhard Herzberg's little green book, which did me a lot of good.😀  

Keeping asking questions: they are not daft.   

[swansont]

1 hour ago, exchemist said:

No, spectral lines have finite width for a variety of perfectly good reasons. (Finite line width means there is a range of absorbing or emitting frequencies of course.) These include the Doppler effect, from motion of the emitters relative to absorbers, and uncertainty broadening, due to finite lifetime of the excited state leading to uncertainty in its energy, by Heisenberg's Uncertainty Relations. In gases, this can be a function of pressure, cf. "pressure broadening", since collisions may shorten lifetimes of excited states and also alter their energy

One should note that these are an effect for an ensemble of atoms or molecules; an individual atom must have its energy match up with the photon, though that energy will have a different value if the atom is moving (Doppler shift) vs an atom at rest. An unperturbed atom still has a finite transition width. 

The bigger picture is that there are a lot of moving parts. Any simple or basic explanation is omitting details. It might seem ad hoc to hear a more complete explanation, but that’s mostly a function of learning things piecemeal. 

[exchemist]

17 minutes ago, swansont said:

One should note that these are an effect for an ensemble of atoms or molecules; an individual atom must have its energy match up with the photon, though that energy will have a different value if the atom is moving (Doppler shift) vs an atom at rest. An unperturbed atom still has a finite transition width. 

The bigger picture is that there are a lot of moving parts. Any simple or basic explanation is omitting details. It might seem ad hoc to hear a more complete explanation, but that’s mostly a function of learning things piecemeal. 

Indeed, QM interactions are inherently probablistic rather than exact no/no go processes. So the probability of interaction goes up progressively as the match gets more exact. I think uncertainty  broadening is also still present for a single atom, if the excited state has a significant spontaneous emission probability, which for electronic transitions it will do, if I recall correctly that it depends on the cube of frequency. But I’m rusty on this and away from my books.

[Mordred]

Just to add to the good answers already posted. One hurdle to overcome is thinking of atoms in accordance to the Bohr model. Which modern physics knows to be incorrect. Instead the atom has a probability cloud with different configurations. All described via the Schrodinger equation.

https://www.khanacademy.org/science/physics/quantum-physics/quantum-numbers-and-orbitals/a/the-quantum-mechanical-model-of-the-atom

This article from Khan University has a decent coverage.

[swansont]

1 hour ago, exchemist said:

Indeed, QM interactions are inherently probablistic rather than exact no/no go processes. So the probability of interaction goes up progressively as the match gets more exact. I think uncertainty  broadening is also still present for a single atom, if the excited state has a significant spontaneous emission probability, which for electronic transitions it will do, if I recall correctly that it depends on the cube of frequency. But I’m rusty on this and away from my books.

The inherent linewidth is indeed from Delta E * delta t > hbar, and electric dipole transitions do vary as frequency cubed

 

[exchemist]

8 hours ago, swansont said:

The inherent linewidth is indeed from Delta E * delta t > hbar, and electric dipole transitions do vary as frequency cubed

 

I think I have read that spontaneous emission processes can be modelled as a special case of normal stimulated emission, but due to interaction with the virtual photons of vacuum fluctuations. We did not go any of that at university, as QED was out of scope for chemists (and my physicist girlfriend at the time preferred to talk about other things). Is it the case?