If you interact light with electricity, you have to deal with parameters. If the electricity you're using doesn't mutually affect light, you don't have the right kind of electricity. The voltage may be way too low. Or it might need a poly-phasic characteristic. (Two signals at 120 degrees gives the sine wave 1.73 of its normal height, which equates to a sharp rise time - but not as good as a compound click.) Also, where the interaction occurs is another factor. The electrical signal must get there at the same time as the photon. So the electrical speed might need to be adjustable. Minto non-resonance transmits trans luminaly, until 80 to 100 kHz, then quickly drops off, substantially slower. (Even like molasses, at a high enough frequency.)
From another perspective, even normal electricity can go a specified distance, then curve. If the completed semi circle would have a resonant length (e.g.: one half wavelength) the curve would throw off an energy beam at some point, and this might interact with light. Atleast with a longitudinal electrical vibration.
A gradient of these curves should be able to really pump some thermal electrons.
And there's always the option to use rf em as the photon, with the longitudinal force being analogous to the pressure wave given off by a Van deGraaff each time it sparks. Or maybe just curve the pressure wave at some distance. Out there in the atmosphere.
