We report on the analysis of a BeppoSAX observation of IRAS 13197-1627 .
The complexity of the broad band ( 0.3-100 keV ) spectrum of the source prevents us from selecting from two possible best-fit models , i.e .
one based on a partial covering and the other on a reflection dominated scenario .
Whatever the best fit model , the data imply that : 1 ) the primary spectrum is heavily obscured by at least two absorbers with column densities of the order of 5 \times 10 ^ { 23 } cm ^ { -2 } and \geq 2 \times 10 ^ { 24 } cm ^ { -2 } ; 2 ) the intrinsic 2-10 keV luminosity of the source is at least \sim 2 \times 10 ^ { 44 } erg s ^ { -1 } thus making IRAS 13197-1627 a type-II QSO ( more precisely a type 1.8 QSO ) rather than a Seyfert 1.8 galaxy as previously classified .
Furthermore , there is marginal evidence of an absorption line at E \sim 7.5 keV suggesting a possible outflow with v \sim 0.1c similar to recent findings on other QSOs and Seyferts .