A significant point is that the phosphors on a CRT screen have their "persistence" designed to support a particular fairly narrow range of refresh rates. The phosphors could be made to have really long persistence (seconds), so there would be no serious flicker down to even maybe a 5 second refresh interval, but then, since the phosphors can only be "turned on" and not "turned off", you wouldn't be able to see motion much faster than that. (Some early CRT terminals used long-persistence phosphors, with the characters "drawn" on the screen instead of scanned. This didn't provide very fast "refresh", but it only had to be as good as a 10 CPS Teletype.)
LCDs have the property that they can be turned on or off, at some relatively high rate, and once set one way or the other they have a relatively long persistence, on the order of a second or so. For this reason they can support a wide range of refresh rates.
The upper limit on refreshing an LCD is a function of capacitance and of the fact that the L in LCD stands for "liquid" --
LCDs are "scanned" via an X-Y matrix of wires, with a pixel at each point where two wires cross. Only one pixel can be manipulated at a time. The voltage on a pixel must be maintained long enough to "charge" the pixel, so that it will hold the charge until refreshed, and all pixels must be visited on each refresh cycle.
And, in addition to the charge time, the liquid inside needs time to mechanically reorient its crystal structure (though, at a physics level, this reorientation is tied at least partially to the "charge" time). Both of these factors place an upper limit on refresh rate.