As background for this Idea, see the link. It describes an "interferometric modulator" or "IMOD" device, which can affect ordinary ambient white light in an unusual way: It causes some of the light-waves to interfere with each other and "disappear", leaving the rest of the light to reflect as a pure
color.
In Nature, it is the mechanism by which butterflies get their wing-colors.
A single IMOD is a "two-state" device: it has a sort-of transparent "flap" that flips between a "high" position and a "low" position, and the result is that the device either reflects a pure color, or it shows black (the bottom of the IMOD cell is colored black).
Therefore, to reflect white, at least three IMODs are needed, one each for red, green and blue. This Idea is about a variation on the theme, which, if it works as described, should be better in several ways.
A Rotary Interferomodulation pixel has a white bottom and a glass top. This cell is cylindrical in shape, like a classic well, where a bucket is used to fetch water. The diameter of the cylinder is maybe 100 microns, but there are reasons why you might want it either smaller or larger, so let's say that 100 microns (1/10 millimeter) is an estimate.
The glass top is thick enough to have screw-threads on its edge. (Look at the metal screw-top lid of an ordinary jar, and even on the outside edge, you can see hints of the screw-threads that are on the inside of the lid.) The whole well-depth also has screw-threads. Therefore the glass top of our Rotary Interferomodulation pixel can be rotated/screwed all the way down to the bottom of the well/cylinder, or all the way out of the well/cylinder.
When the glass top is at the top of the well, NO interference happens. So ambient light passes through the glass, reflects off the white bottom, and passes through the glass again, and an observer will see a white pixel.
Screw the glass top partway down, and now some light-wave interference begins to happen, so the observer will now see a colored pixel. Which color? Depending on how far down the well the glass is screwed, it can be any color of the spectrum, from red to violet!
And if the glass is screwed down just a bit farther (NOT all the way to the bottom), then all the visible-light-waves interfere with each other, and the result is a black pixel, despite the bottom of the cell being colored white.
Now, how do we rotate that tiny circular piece of glass? I'm going to suggest that when it is made, it is made as a multi-charged electret (see link). Imagine it being invisibly divided up like a pizza, and alternate slices have alternate permanent-static-electric charges --"Plus" followed by "Minus" followed by "Plus" followed by "Minus" ...
The wall of the cylindrical well can now contain electronic circuits and capacitors, which when appropriately charged can attract or repel certain of the glass pizza slices. Note that lots of today's circuitry, in computer chips, has features smaller than 0.1 micron, so quite a lot of electronics can be fit in the walls of our well/cylinder. The circular piece of glass is, essentially, the rotor of a simple electric motor. And we can run that motor either forward or backward, so that the piece of glass either screws closer to the bottom of the well, or farther away from it.
Note that when the power is off, there is nothing to cause that piece of glass to rotate, so no power is needed for it to continue to display whatever color (from ambient light, remember).
Now build a nice large array of those pixels (1024x1024 is more than a million of them), and you will have an extremely versatile display screen. Note that since each pixel is circular, they can naturally fit together fairly closely, much like a grid of hexagons.
Note that the pixel-size of a typical computer monitor these days is approximately 250 microns (1/4 millimeter). And the pixels of a large-screen TV are even larger.
The reason we might want each of the glass pieces to be smaller than 100 microns is, each will have less mass, and can be rotated faster, using less energy. A certain minimum rotation speed is essential, if we want this display to handle full-motion video. However, this will mean that a number of those small pixels are needed to equal one ordinary monitor-pixel.
The reason we might want each of the glass pieces to be larger than 100 microns is, the overall display will be less complex (fewer components), and thus be less expensive. This variation on the theme might best be used in e-book readers, where full-motion video is seldom needed.