QUASI-SPHERICAL ACCRETION OF OPTICALLY THIN CLOUDS AS A MODEL FOR THEOPTICAL UV/SOFT X-RAY-EMISSION OF AGN/

We examine four models for the optical/UV/soft X-ray emission of AGN: a mechanically heated plasma, either optically thin or thick for elect ron scattering and a radiatively heated plasma, either optically thin or thick for electron scattering. We present detailed computations of the emitted spectra, and we discuss the total energy budget (including the requirement for the mass accretion flow), and the interaction wit h the hot optically thin medium responsible for hard X-ray emission. W e show that in no case the spectrum can be considered as due to a pure free-free process: lines and/or free-bound continuum are always impor tant. We can eliminate mechanical heating on the basis of variability time scales and of a detailed fit of the spectrum in the optical/UV/X- ray range. If the system of clouds is radiatively heated, the energy b udget requires that it is optically thick to electron scattering. Fina lly, we conclude that a system of radiatively heated clouds, optically thick for electron scattering and having a covering factor close to u nity, is a viable model for Seyfert galaxies. The same model is applic able for quasars although one cannot rule out in their case a small co ntribution from mechanical heating. Therefore we propose a new model f or the emission of the optical to X-ray spectrum in AGN, made of a qua si-spherical (or a thick disk) distribution of mildly optically thick clouds. Smaller covering factor, 0.5 - 0.9, and smaller optical depth for electron scattering, similar to 50, are more appropriate for Seyfe rt galaxies while larger values (> 0.9 and similar to 500 respectively ) are deduced for quasars. As a consequence, we expect higher Eddingto n ratios in quasars than in Seyfert galaxies. One peculiarity of the m odel is that UV is due to pure emission, and X-rays to a combination o f reflection and primary spectrum and that absorption features are wea k in the X-ray range in spite of the large covering factor. Although i t is attractive, this cloud scenario is not unique for the UV-X emissi on of AGN, with two others being: reprocessing clouds radiating as bla ck bodies and accretion disks with hot coronae. We discuss the prospec ts of differentiation between these possibilities.