>Could you, please, give me reliable reference regarding parameters to use in D-PCM for acetonitrile? Unfortunatelly, this solvent is not given as a preset of its parameters for SOLVNT variable of $PCM group (as manual says), and the only parameter I have is its dielectric permitivity.
a dielectric constant is quite enough information for basic PCM
computations. The PCM code in Firefly is smart enough and does not
request those constants which are not required for particular job.
You are encouraged to try and check if it just works
>And one more question: is DPCM the only option for geometry optimization within a solution model? What about possibility to do this within CPCM (and COSMO as a particular case) as of today and in future?
At present, DPCM is the only option. In the future, there will be an option for CPCM gradients.