Knowledge Bank

(This is a last minute paper, not in the preliminary program, and replaces the original N4.) We have demonstrated by photofragment spectroscopy that the long accepted assignment of the main iodine visible continuum as being almost entirely due to the $B3\Pi ou+\leftarrow x1\Sigma og+$ parallel transition, which gives one excited and one ground state atom, is incorrect. In reality, a perpendicular transition to a $Iu$ state (most probably $\mathrm{<mrow\; data-mjx-texclass="ORD">1</mrow>}\Pi 1u$) dissociating to ground state atoms is of comparable importance. By crossing a molecular beam of $I2$ with pulses of polarized light from a laser-pumped tunable dye laser and measuring the time of flight to a mass-spectrometer detector as a function of angle of the recoiling I $atoms,1$ we have separated at several wavelengths the $B3\Pi o+o$ and $Iu$ components of the continuum absorption. The results, which clearly distinguish between the two components both by angular and translational energy distributions, allow us to determine directly the ratios of absorption probability to the two states. Our conclusions are consistent with recent $studies2$ of the $I2$ visible banded region by Steinfeld. Campbell and Weiss, by Chutjian, and by Brewer and Tellinghuisen which also indicate the importance of another state in addition to $B3\Pi og+$. The additional state is again probably $\mathrm{<mrow\; data-mjx-texclass="ORD">1</mrow>}\Pi 1u$, bringing the new interpretation of the $I2$ visible spectrum into close parallel with the accepted interpretations of the $Cl2$ and $Br23$ visible spectra. We thank NSF for support of these experiments and NIH for support of the on-line computer system we used.