I don't know if this makes any sense, but I came up with this in my head: if each molecular bond in any molecule had a specific *signature wavelength of dissociation", this would be the exact wavelength of the laser beam necessary to excite the electrons involved in that one particular bond and no other bonds in the molecule into dissociating homolytically. Is this theoretically sound in any way?


Obviously if you have a diatomic then yes, you can put energy in and the bond will break. There are three cases; (a) as long as the photon has enough energy it can excite a dissociative excited state, (grey in sketch) that is a state that is only repulsive. The absorption is continuous here so its energy has to be above a certain value. (b) The second case has a curve crossing in the excited state. This means that two excited states cross , a bound state and a dissociative one. Then if you hit the energy at the crossing point, dissociation will occur, but not at higher or lower energy. This is called pre-dissociation. (The read & grey line crossing show pre-dissociation). It occurs in many molecules not just diatomics, e.g. benzene. (c) The third case is when a bound state is excited above its dissociation limit, e.g. above approx 10000 in the blue curve, 5500 on the black curve . However, reaching these energies may be hard because the Franck-Condon factors may be vanishingly small. (roughly speaking the FC factors ensure vertical transitions)


Now the case with polyatomic molecules is tricky. Many experiments have been tried with varying success, the area is sometimes called 'mode selective chemistry' and/or 'coherent control'.

In a molecules there are many vibrations, called normal modes. Although we think generally of one bond vibrating, in actuality many bonds vibrate in unison (i.e. in a fixed phase relationship) in a normal mode. These modes are governed by the symmetry species of the molecule. Each normal mode is supposed to be orthogonal, so that vibration of one does not effect vibration of another. But in fact vibrational anharmonicities allow vibrational quanta to flow from one mode to another. Thus exciting one normal mode soon involves energy spreading out into all other modes and thus even if there is enough energy to break a bond it soon becomes diluted.

In coherent control experiments one uses one or more very short light pulse ( < 100 femtoseconds) to excite the molecules but control the phase of the light (hence pulse shape) to control how long the molecule takes absorb the photon. This is done in the hope that once the energy starts to spread out (as a vibrational wavepacket) it can be controlled to end up in just the bond you want to break, i.e. to try and use the vibrational spreading out tendency to advantage. The pulse shaping is done using an electro-optic or acousto-optic device in a feedback loop using say, a genetic algorithm linked to a detector that observed product formation.

  • 1
    $\begingroup$ The first paragraph concerns electronic excitations (excited states), whereas the rest vibrationally excited states. The dissociation energy of a given bond is the minimum amount of energy you need to put into a molecule to break that bond. There are several ways, including heating, etc... You could even imagine two-photon mechanisms, e.g. first ionizing the system and then breaking the bond, if the ionized state is easier to dissociate. $\endgroup$
    – ssavec
    Aug 8 '16 at 11:00
  • 2
    $\begingroup$ @ ssavec yes heating would work but the question asks about using a specific laser wavelength. Also two photon transitions would work if you wanted to get to a state that selection rules forbid for one photon transitions. Once in that state its the same thing as a one photon transition. Ionisation is not necessary. Electronically excited state also have vibrational normal modes provided that they are not entirely repulsive in nature, i.e have a potential well as in the case of the red and blue curves in the figure. $\endgroup$
    – porphyrin
    Aug 8 '16 at 19:18
  • $\begingroup$ So, Is there a way to imagine a simplified version of reality in which this would be possible. What I mean is, take of all the practical things, and leave just theory, would it then be possible to (no matter how big the molecule is) fine tune a frequency that would disrupt that one particular bond of that one particular molecule if you fired the laser at a bunch of different molecules (for example, you're firing the laser at a solution of several complex molecules in gaseous state). What I mean, take off the technicalities of the real world and leave a theoretical shell, like Bohrs model $\endgroup$ Aug 8 '16 at 21:01
  • $\begingroup$ Just by enthalpy reasoning, I'd guess it is very difficult for larger (more than 4 atoms) molecules. You have to add enough energy to break the bond, approx 100 kcal/mol ~ 4.336 eV ~ 285.88 nm ~ UVB radiation. So if it should be one photon, it has to be UV and therefore electronic excitation, as shown on the plot. Nevertheless, see en.wikipedia.org/wiki/… $\endgroup$
    – ssavec
    Aug 9 '16 at 7:24
  • $\begingroup$ The important point about a polyatomic molecule is that energy flows among vibrational normal modes very quickly indeed so its generally not possible to get it into the one bond you want to dissociate, and then to keep it there long enough to break. $\endgroup$
    – porphyrin
    Aug 9 '16 at 8:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.