One thing that would seem to be a concern to me is how to keep the attitude steady while the ion engine is firing. Yes, not much thrust, but to make the most of it, you want it pointing in as close to the ideal direction as possible. I don't know how the various probes like Hayabusa and Dawn do it.. Cold gas? Gyro? I would not think a bar magnet would be good enough (certainly not for deep space probes, but even for earth orbit). Starting to get more complicated.
Another thought about the really cool energy analysis done by KK6MC: Besides the duty cycle imposed by wanting to be out of eclipse, there is another duty cycle imposed by the starting orbit and the desired eccentricity. For example if you were starting from a highly elliptical GTO and you want to get to high circular, you need to thrust mainly at the apogee in order to raise the perigee. If you are circular and you want to be elliptical with perigee equal to the starting height, you thrust at what will be the perigee to raise the apogee. (Of course you probably want to raise both ends some, and you may want to change the plane too but that's the general idea.)
Thanks for bringing this up...great thought experiments, and that's how real projects begin!