This article reports interesting experiments and calculations, but in my opinion also an incorrect interpretation of the results. Namely, that they result from photon-induced symmetry breaking.
Fig 3.A shows that the
H2 + photon → H + proton + electron
process involves correlation between the recoil directions of the three fragments. For example, Fig. 3.A.d reveals "orientation" (1): a preference for ejection of the electron along a particular direction ("up" rather than "down").
I do not dispute the data in the figure, but I disagree with the attribution of its asymmetry to photon-induced symmetry breaking.
The photoionization process involves a correlation between two vectors: those specifying the directions of light polarization and electron ejection. Its probability is a function of only one angle and must have cylindrical symmetry around the light polarization direction (2).
This means that, in the absence of further considerations, all the results in Fig. 3.A should have cylindrical symmetry around the light polarization direction (the horizontal line in panels a-f). Therefore, no preference for electron ejection towards the upper or lower half of each panel should be observed.
The cylindrical symmetry is not broken by anything intrinsic to the photon absorption or to the ionization and dissociation dynamics. Instead, it is broken by consideration of a third vector (the proton recoil direction). What the results in Fig. 3.A show is that if the proton is found to go "up," then the probability that the electron is found to also go "up" may differ from the probability that the electron is found to go "down." This orientation effect becomes observable not because of photon-induced symmetry breaking, but rather because of the simultaneous consideration of more than two vectors. Had the measurement involved further vectors, say, the photofragment spins, further symmetry breaking would have been found.
The relation between the number of correlated directions and symmetry reduction is discussed in many articles and textbooks on vector correlations; Refs. 2-3 are classic sources.
Practical examples of this effect are easily found in the stereodynamics literature. For instance, a recent review (4) reports several cases of photofragment orientation resulting from the photodissociation of homonuclear diatomics.
References
1. A. J. Orr-Ewing, R. N. Zare, Annu. Rev. Phys. Chem. 45, 315 (1984).
2. J. D. Barnwell, J. G. Loeser, D. R. Herschbach, J. Phys. Chem. 87, 2781 (1983).
3. R. N. Zare, Angular Momentum (Wiley, New York, 1981).
4. A. P. Clark et al., Phys. Chem. Chem. Phys. 8, 5591 (2006).
E-Letters: ![[Read E-Letter]](/icons/misc/sectionDOWN.gif){kind=icon}
