Technical Comments

Polarized Stellar Light

See allHide authors and affiliations

Science  05 Mar 1999:
Vol. 283, Issue 5407, pp. 1415
DOI: 10.1126/science.283.5407.1415a

In their report “Circular polarization in star-formation regions: Implications for biomolecular homochirality” (1), Jeremy Bailey et al. state that circularly polarized ultraviolet light is not to be expected in the region of a young supernova remnant, such as the Crab Nebula. They cite Roberts, who maintained that the distribution of pitch angles of the electrons will cancel out circular polarization of high-frequency radiation (2). However, circularly polarized synchrotron radiation will be emitted along the axis parallel to the magnetic field, with frequencies that are small integer multiples (harmonics) of the cyclotron frequency ωc =ecB/E e, where eis the electron charge, c is the speed of light, B is the magnetic field, and E e is the total relativistic electron energy (3). The photon energyE p at the n th harmonic is thenE p = enℏ︀cB/ Ee, where ℏ︀ is Planck's constant/2 π. Thus, for photons of 6 eV, which are known to be effective for photolyzing amino acids, and for electrons of 60 MeV volts, B is of the order of 6 × 1010 G. The magnetic field of neutron stars has been estimated to be up to 1012 G (4).

Because the optical synchrotron radiation from the Crab Nebula has been observed to be 40 to 60% linearly polarized and as much as 0.03% circularly polarized, and its 1415-MHz radiofrequency radiation to be as much as 0.05% right circularly polarized (5), it is clear that circularly polarized ultraviolet synchrotron radiation must also be present.

The argument of Bailey et al. (equally applicable to their mechanism) that circularly polarized synchrotron radiation would be ineffective for an enantioselective radiolysis of amino acids is based on the conclusion of Mason (reference 13 in the report) (6) that the “Kuhn-Condon zero sum rule” for circular dichroism bands precludes such effects. This apparent misinterpretation of the Kuhn-Condon rule has recently been corrected (7).


Response: Rubenstein et al. state that “circularly polarized synchrotron radiation will be emitted along the axis parallel to the magnetic field.” This is correct, however, only for low electron energies (cyclotron radiation, as observed astronomically in the AM Herculis binaries or Polars, which have high optical circular polarization). In the highly relativistic case (synchrotron radiation), relativistic beaming means that radiation is not emitted parallel to the magnetic field. The radiation is beamed into a narrow cone around the direction of particle motion (that is, perpendicular to the magnetic field), and predominantly linear polarization is produced.

The quoted figure of 0.03% circular polarization for the optical emission from the Crab Nebula is an upper limit. Circular polarization was not observed (1). Subsequent more sensitive work measured small levels (about 0.1%) of circular polarization, but we attributed these results to birefringence resulting from aligned grains in the interstellar medium; we found nothing intrinsic to the nebula (2).

The 0.05% polarization at 1415 MHz is a small polarization at a frequency a million times less than the circular dichroism bands of interest. The available observations at optical wavelengths (within a factor of 2 of the ultraviolet circular dichroism bands) do not show significant circular polarization. The lack of circular polarization instrumentation in space, however, means that there have been no direct observations at the wavelength of interest.

Rubenstein et al. also dispute the conclusion by Mason (3) that broad-band circularly polarized light cannot be enantioselective. While there may not be complete cancellation, as Mason has stated, there is still likely to be substantial reduction in the enantiomeric excess that can be generated if the spectrum covers bands of alternating sign, as is likely to be the case with a spectrum such as that of the Crab Nebula, which extends into the far ultraviolet and x-ray regions of the spectrum. The problem is much less serious for our mechanism, which uses scattered starlight, because the spectra of most stars falls off steeply below 200 nm, which limits absorption to the longest wavelength circular dichroism band.


Stay Connected to Science

Navigate This Article