Context. Chemical models predict the presence of S-bearing molecules such as hydrogen sulfide (H 2 S) in interstellar icy grain mantles in dense molecular clouds. Up to now only two S-bearing molecules, namely sulfur dioxide (SO 2 ) and carbonyl sulfide (OCS), have been detected in the solid phase towards young stellar objects (YSOs), while upper limits for solid H 2 S have been reported towards the same lines of sight. The estimated abundance of S-bearing molecules in icy grain mantles is not able to account for the cosmic S abundance. Aims. In this paper we studied the effects of ion irradiation on different icy targets formed by carbon monoxide (CO) and SO 2 or H 2 S as mixtures and, for the first time, as layers. Methods. We carried out several irradiation experiments on ices containing SO 2 or H 2 S mixed or layered with CO. The samples were irradiated with 200 keV protons in a high-vacuum chamber (P < 10 −7 mbar) at a temperature of 16-20 K. IR spectra of the samples were recorded after various steps of irradiation and after warm-up. Results. We have found that the column density of H 2 S and SO 2 , as well as CO, decreases after irradiation, and the formation of new molecular species is observed. In the case of CO:SO 2 samples, OCS, sulfur trioxide (SO 3 ), ozone (O 3 ), and carbon dioxide (CO 2 ) are the most abundant species formed. In the case of CO:H 2 S samples the most abundant species formed are OCS, SO 2 , carbon disulfide (CS 2 ), hydrogen persulfide (H 2 S 2 ), and CO 2 . The profile of the OCS band formed after irradiation of the CO:H 2 S mixture compares well with the profile of the OCS band detected towards the high mass YSO W33A. Conclusions. Our results show that on a time scale comparable to the molecular cloud lifetime, the column density of H 2 S is strongly reduced and we suggest that this could explain the failure of its detection in the solid phase in the lines of sight of YSOs. We suggest that the solid OCS and SO 2 detected in dense molecular clouds are formed after ion irradiation of icy grain mantles.

Sulfur appears to be depleted by an order of magnitude or more from its elemental abundance in star-forming regions. In the last few years, numerous observations and experiments have been performed in order to to understand the reasons behind this depletion without providing a satisfactory explanation of the sulfur chemistry towards high-mass star-forming cores. Several sulfur-bearing molecules have been observed in these regions, and yet none are abundant enough to make up the gas-phase deficit. Where, then, does this hidden sulfur reside? This paper represents a step forward in our understanding of the interactions among the various S-bearing species. We have incorporated recent experimental and theoretical data into a chemical model of a hot molecular core in order to see whether they give any indication of the identity of the sulfur sink in these dense regions. Despite our model producing reasonable agreement with both solid-phase and gas-phase abundances of many sulfur-bearing species, we find that the sulfur residue detected in recent experiments takes up only ∼6 per cent of the available sulfur in our simulations, rather than dominating the sulfur budget.