MOLECULAR-HYDROGEN IN THE DIRECTION OF ZETA-ORIONIS-A

A spectrum of zeta Ori A over the wavelength interval 950-1150 Angstro m recorded by the Interstellar Medium Absorption Profile Spectrograph (IMAPS) on the ORFEUS-SPAS I mission shows Lyman and Werner band absor ption features from molecular hydrogen in rotational levels J = 0, 1, 2, 3, and 5. Most of the molecules are found in two distinct velocity components. One is at a heliocentric radial velocity of about -1 km s( -1) with log N(H-2) = 14.5 and a rotational temperature T-rot = 950 K, while the other is at +25 km s(-1) with log N(H-2) = 15.9 and T-rot = 320 K. Some extra H-2 exists in a much weaker component [log N(H-2) = 14.0] between the two main peaks. The H-2 component at -1 km s(-1) ex hibits profile shapes that become broader and show small displacements toward more negative velocities as J increases. These changes are inc onsistent with a simple interpretation that UV optical pumping in an o ptically thin, uniform medium creates the H-2 in excited rotational le vels. Differential shielding of the UV radiation at certain velocities does not appear to be a satisfactory explanation for the effect. Evid ence from atomic features at other velocities may offer some insight i nto the origin of this unusual behavior exhibited by the H-2 profiles. Absorption features from moderately ionized atoms at -94 km s(-1) and more highly ionized species at about -36 km s(-1) suggest that along the line of sight to zeta Ori A, there may be a standing bow shock wit h an initial compression ratio of 2.6. This shock is probably created when a negative-velocity gas flow collides with an obstruction, in thi s case a neutral cloud at 0 km s(-1). If this interpretation is correc t, the H-2 with the changing profiles may represent molecules forming in the postshock gas that is undergoing further compression as it reco mbines and cools. We suggest that molecules can form initially by asso ciative detachment of H- in a moving, warm, partly ionized medium behi nd the front. The H-2 in this area is most conspicuous in the higher J levels. Later, when the gas becomes very cool, neutral, and more comp ressed as it comes nearly to a halt, it is more easily seen in the low est J levels. In this part of the medium, the principal way of produci ng H-2 should be from reactions on the surfaces of dust grains, as one expects for quiescent interstellar clouds.