Efficiency of radial transport of ices in protoplanetary disks probed with infrared observations: the case of CO2

Context. Radial transport of icy solid material from the cold outer disk to the warm inner disk is thought to be important for planet formation. However, the efficiency at which this happens is currently unconstrained. Efficient radial transport of icy dust grains could significantly alter the composition of the gas in the inner disk, enhancing the gas-phase abundances of the major ice constituents such as H2O and CO2. Aims. Our aim is to model the gaseous CO2 abundance in the inner disk and use this to probe the efficiency of icy dust transport in a viscous disk. From the model predictions, infrared CO2 spectra are simulated and features that could be tracers of icy CO2, and thus dust, radial transport efficiency are investigated. Methods. We have developed a 1D viscous disk model that includes gas accretion and gas diffusion as well as a description for grain growth and grain transport. Sublimation and freeze-out of CO2 and H2O has been included as well as a parametrisation of the CO2 chemistry. The thermo-chemical code DALI was used to model the mid-infrared spectrum of CO2, as can be observed with JWST-MIRI. Results. CO2 ice sublimating at the iceline increases the gaseous CO2 abundance to levels equal to the CO2 ice abundance of similar to 10(-5), which is three orders of magnitude more than the gaseous CO2 abundances of similar to 10(-8) observed by Spitzer. Grain growth and radial drift increase the rate at which CO2 is transported over the iceline and thus the gaseous CO2 abundance, further exacerbating the problem. In the case without radial drift, a CO2 destruction rate of at least 10(-11) s(-1) or a destruction timescale of at most 1000 yr is needed to reconcile model prediction with observations. This rate is at least two orders of magnitude higher than the fastest destruction rate included in chemical databases. A range of potential physical mechanisms to explain the low observed CO2 abundances are discussed. Conclusions. We conclude that transport processes in disks can have profound effects on the abundances of species in the inner disk such as CO2. The discrepancy between our model and observations either suggests frequent shocks in the inner 10AU that destroy CO2, or that the abundant midplane CO2 is hidden from our view by an optically thick column of low abundance CO2 due to strong UV and/or X-rays in the surface layers. Modelling and observations of other molecules, such as CH4 or NH3, can give further handles on the rate of mass transport.