@ Michael : thank you Michael for comments, some of cited cookies are very suspicious also :-) I hope that news about P2O5 as dessicant (not for ALL alcohols) is good for you. Question was not about purification - but about availability of dry CF3CH2OH, so I am sure.
1) CF3CH2OH is acidic like phenol so drying over CaH2 should be very interesting experiment.
2) NaHCO3 is also like on profanation, sorry - it decomposes to Na2CO3 and H2O at 50oC - Bp of trifluoroethanol is 77oC
3) Molecular sieves as Na/K/Ca salts of very weak acids (aluminosilicates) are also unlikely suitable because of relative acidic nature of CF3CH2OH; however, molecular sieves 4A were used for this purposes (48 hrs , 250ml of 4A molecular sieves granules per 3L of CF3CH2OH ) http://pubs.acs.org/doi/abs/10.1021/ac60294a046journalCode=ancham .
Activation method of molecular sieves noted in this publication is mistrustful ( 16hrs@150oC) as minimal temperature ~200oC is needed :Trent, R.E., 1993, Fundamentals of molecular sieve design, Presented at the American Institute of Chemical Engineers Spring National Meeting,Huston, Texas, April.
ALDRICH recommend to use 200-315oC at normal pressure for 4A-type : http://www.sigmaaldrich.com/chemistry/chemicalsynthesis/learning-center/technical-bulletins/al-1430/molecular-sieves.html . I prefer 200oC (silicone oil or Wood's alloy bath) and oil-pump vacuum (~0.001Torr).
@Michael : it is not surprise. Please do not use dessicants in stoichiometric amounts (it is common student fail) - and it will be OK.
P2O5 is best dessicant. Some of acid phosphate will act as dessicant but will not distilled with alcohol because of large difference in boiling points.
Ester bond in P-O-CH2CF3 is very sensitive to hydrolysis : http://pubs.acs.org/doi/abs/10.1021/jo01313a008 . Same paradigm is used in common method of absolutation of EtOH with diethyl phthalate.
"Vladimir Dvoryanchikov · 5.84 · 0.62 -> IMHO, P2O5 is poor idea. It - and larger losses of a main product, and probability its pollution by (CF3CH2)2O. How about CaO or CaC2?" - bug fix for possible removal of wrong answer
Comment:
This humble opinion based on lack of hands-on with fluorinated alcohols. IMHO :-)
1) pKa of CF3CH2OH is 12.46 ( 104 more acidic than CH3OH and 104.5 than H2O) - so it will react with Ca(OH)2 (from CaO+H2O) to CF3CH2O-Ca-OH with recovery of water.
2) CaC2 react with acidic fluorinated alcohols in the first place but not with water; technical CaC2 contain large quantity of impurities; some crystals of pure CaC2 are available worldwide only in laboratories but not commercially.
3) Protonation of CF3CH2OH need much stronger acid in comparison to alkanols (protonation pKa~ -2); pyrophosphoric acid pKa1 ~ -1.5. By this reason it is impossible to get (CF3CH2)2O in this simple conditions but it is possible to have absolute CF3CH2OH (H2O ~100 ppm or less) with minimal loss of material
Valid synthesis routes for (CF3CH2)2O look at http://www.molbase.com/en/synthesis_333-36-8-moldata-160051.html#tabs