From the results, it is evident that the Cu(I)-PLs performed at least
equally well as the CuI powder or the Cu-ligand catalyst. Furthermore,
as expected, the reaction toward Rufinamide (employing propiolamide) was
considerably faster than the reaction to product 2 with
phenylacetylene. Only the Cu(I)-PLs experiment with water as co-solvent
was not successful. This may be caused by the increased solvent polarity
that makes the bilayer membrane less accessible to reagents. After the
reaction was completed, the reaction mixture was cooled down to r.t.
(around 10 mins) and gratifyingly, for all reactions performed in pure
ACN, the product crystallized out and could be conveniently collected
from the bottom of the reaction vials. This demonstrates a good
agreement with the COSMO-RS prediction. Furthermore, based on analysis
of Entry 6, the polymersomes were effectively retrieved in the ACN
layer. In case of Rufinamide, due to the faster reaction time and lower
solubility compared to 2 , product precipitation already
occurred after 2 minutes during reaction and only 15 mins were needed
for full conversion from R1. However, the Rufinamide
crystals were contaminated to some extent with polymersomes. This could
be circumvented by addition of ACN after the reaction to wash and filter
the polymersomes out. In Entry 10, the isolated Rufinamide yield was
93%, with 4% of product retained in the organic layer. In addition,
although the conversions in DMSO were high, no product precipitated at
r.t. and Cu(I)-PLs were not observed by TEM after the reaction.