I don’t think this is a terribly useful question to examine, because you have to decide what you mean by “theoretical”, and because clock speed isn’t typically a goal unto itself unless you work in CPU marketing. Under some definitions, we’re nowhere close to the fastest theoretical clock speed you can achieve with electronics, even though we might be getting there as far as the materials we know about are concerned.