Hey - ask all the questions you want! I just hope we can come up with the answers!
The target for "flexible" means to be able to be rolled up like sheets of PVC - the coloured plastic people use to cover lights sometimes to give them colours. I think if the epaper were to be actually folded almost all would fail, but that's also dependant on the cell technology and architecture.
The limit for this isgoing to be set by the frontplane and backplane technology combination. The current frontplane is quite flexible really, but this is limited by materials and by the thickness of the frontplane. This is the easy part in my opinion.
The backplane is harder to make flexible - that breakthrough will probably come when organic electronics really hits as current backplanes are mostly relaint on using non-flexible materials and keeping them aligned using something solid like ITO.
For the particles being misaligned, that won;t really be the problem, but fracturing of the microcapsules could happen, though they are also made from flexible materials too and I'm not sure what the limit of this will really be! They will find that certain frontplanes will lend themselves to flexibility better.
The other problem may be that if you bend and the front and backplanes become slightly misaligned, you will get cross-talk between the 2 meaning cells which shouldn;t be switched, might be because they are now in contact with an electrode they shouldn't have been in contact with. Also, if the electrophoretic display uses a colour filter and this become misaligned when flexed....well....you get the picture...or not ;)