PDA

View Full Version : Visibility of dihydrogen



chornedsnorkack
2011-May-24, 07:31 AM
Is there any way to detect hydrogen molecules?

Protons recombine with electrons and emit Lyman and Balmer series in the process. Hydrogen atoms have lowlying, longlived spin states and emit radio waves, from 21 cm to 210 m according to redshift. But does a hydrogen molecule have any states making it visible?

There are supposed to be other molecules. Like carbon monoxide or hydrogen fluoride. But these obviously require high metallicity!

In which forms is primordial lithium supposed to occur? Li atoms, LiH molecules or solid LiH grains? How can a low metallicity, high redshift molecular hydrogen cloud be seen?

EigenState
2011-May-24, 03:10 PM
Greetings,


Is there any way to detect hydrogen molecules?


I assume you mean within an astronomical context.

Rotational transitions can be observationally resolved within the X 1Σg+ ― B 1Σu+ absorption band. That band lies in the UV near 1050.

I have seen one spectrum of ζ Ophiuchi in which a rotational transition of HD was observed.

Homonuclear diatomics lack a dipole moment and therefore will show neither a pure rotational spectrum within the microwave domain, nor a pure vibrational spectrum within the IR.

Best regards,
EigenState

Cobra1597
2011-May-24, 03:59 PM
nor a pure vibrational spectrum within the IR.

Not in the IR, but I think it has been observed in the violet/UV range via Raman scattering.
http://www.lpl.arizona.edu/~yelle/eprints/Betremieux99a.pdf

chornedsnorkack
2011-May-24, 04:02 PM
Greetings,
Homonuclear diatomics lack a dipole moment and therefore will show neither a pure rotational spectrum within the microwave domain, nor a pure vibrational spectrum within the IR.



What is the lifetime of rotationally excited hydrogen molecules?

The transition energy is quite high compared to orto and para hydrogen atom. But rotational transitions are very strongly forbidden - not only because of lack of diplole moment, but because allowed rotational states are separated by 2h: parahydrogen molecule can have 0, 2, 4 etc, while ortohydrogen can have 1, 3 etc. Parahydrogen and ortohydrogen are separated from each other by 2h as well - 1 from molecular rotation and 1 from spin. And that means the only allowed transitions should be 2 photon emission (or 4, 6 or bigger even number). Which, for one, has low probability and, for another, gives continuum rather than spectral line.

Is there any way to detect molecular hydrogen with no source of UV or light behind it, just relict radiation?

Amber Robot
2011-May-24, 06:03 PM
Molecular hydrogen is detected in all kinds of environments, both in absorption and in emission. I know people who have written dissertations about it. You mentioned carbon monoxide. That is also observed in all kinds of environments in both absorption and emission and does not require any higher metallicity than normal.

Vibrational transitions of H2, though very weak, have been detected. For example, see Lacy et al. (1994: ApJ 428, L69).

EigenState
2011-May-24, 07:03 PM
Greetings,


Molecular hydrogen is detected in all kinds of environments, both in absorption and in emission. I know people who have written dissertations about it. You mentioned carbon monoxide. That is also observed in all kinds of environments in both absorption and emission and does not require any higher metallicity than normal.

Vibrational transitions of H2, though very weak, have been detected. For example, see Lacy et al. (1994: ApJ 428, L69).

I was not aware of that. Thank you!

Best regards,
EigenState

Amber Robot
2011-May-24, 07:38 PM
I was not aware of that. Thank you!


You're welcome.

And here's a sample reference to detection of rotational and vibrational emission lines from H2:

Timmermann et al. (1996: A&A 315, L281)

EigenState
2011-May-24, 07:43 PM
Greetings,


What is the lifetime of rotationally excited hydrogen molecules?

In the lowest vibrational state of the ground electronic state, spontaneous radiative relaxation of the J levels occurs with lifetimes of the order of 107 (high J) to 1010 (low J) seconds. Those transitions are electric quadrupole allowed such that the transition probability scales with the fifth power of J.

However, the rotational state populations are driven by several additional mechanisms: shock wave heating; spontaneous radiative decay of excited rovibronic levels; collisional redistribution; and possibly H2 formation on grain surfaces.


The transition energy is quite high compared to orto and para hydrogen atom. But rotational transitions are very strongly forbidden - not only because of lack of diplole moment, but because allowed rotational states are separated by 2h: parahydrogen molecule can have 0, 2, 4 etc, while ortohydrogen can have 1, 3 etc. Parahydrogen and ortohydrogen are separated from each other by 2h as well - 1 from molecular rotation and 1 from spin. And that means the only allowed transitions should be 2 photon emission (or 4, 6 or bigger even number). Which, for one, has low probability and, for another, gives continuum rather than spectral line.

Emphasis added. The rigorous selection rule for electric quadrupole transitions is ∆J = 0, 1, 2. Two-photon emission is not required.

It is also interesting to note that the evidence indicates that ortho-H2 and para-H2 come into effective thermodynamical equilibrium via a proton exchange reaction:

H+|β +ortho-H2|αα = para-H2|αβ + H+|α

where |αβ, etc., specifies the nuclear spin state.

Best regards,
EigenState

EigenState
2011-May-24, 07:46 PM
You're welcome.

And here's a sample reference to detection of rotational and vibrational emission lines from H2:

Timmermann et al. (1996: A&A 315, L281)

Keeping me busy I see. :cool: