The aim of this proposal is to observe the behaviour of the 6.2 and
 7.7 mu dust features, and the underlying continuum emission, in the
 direction of a well defined edge-on ionization front. This will
 allow to derive new constrains on the nature of the emitting
 particles, PAH molecules or large carbonaceous grains. Ground based
 observations have shown that the associated 3.3, 11.3, and 8.6 mu
 features, generally attributed to C-H peripheral bonds on the same
 particles, are abundant in the photo-dissociation region and absent
 from the HII gas, with a marked transition at the precise location
 of the ionization front. This proves that the particular physical
 conditions in the HII region, either modify the nature of the emitting
 particles (eg ionization or de-hydrogenation) or destroy them (eg photo-
 thermodissociation of PAHs or H+ chemisputtering of small carbonaceous
 grains). In the first hypothesis, the 6.2 mu and 7.7 mu features,
 which are attributed to vibrational modes of the carbon skeleton,
 should survive, and will not show the same behaviour as the C-H
 features. This can be tested by imaging an ionization front, M17-SW
 (Autumn launch) or Orion bar (Spring launch), with ISO-CAM in the
 filters dedicated to the 6.2 and 7.7 PAH features. The transition
 of the 3.3 mu feature has been observed to occur in a very sharp
 region, about 2 arcsec, so that the full angular resolution of the
 camera is needed : 1.5" per pixel (3" resolution).