If d(dx) = d²x then why not possible dt(dt) = d²t²? Still I don't understand. Please simplify.
This is entirely a matter of notation -- how we choose to write the second derivative. Ultimately, you just have to learn that this is how it is written.
But the motivation is that, using
dxd to represent the operation of taking the derivative of a function, the second derivative is represented as
dxd(dxdy), and we write that
as if d were a number,
even though it isn't:
dxdxddy=(dx)2d2y=dx2d2y. We don't bother to use parentheses around the dx because in some sense the d, as a "differential operator", is thought of as tightly bound to what follows it; and we don't distribute and write d
2x
2 because that would suggest we are doing something to x
2, which we are not.
Another way to look at it is in terms of the definition. The derivative is a limit of a fraction,
dxdy=limΔx→0ΔxΔy, so the second derivative is the limit of
ΔxΔΔxΔy, and this can
almost be thought of as
(Δx)2Δ(Δy). Here, again,
Δ is not a number, but an operator, so distribution is not valid.
But, again, this is simply a notation that reminds us of these relationships, and happens to work well; if you try to analyze it too closely, it falls apart. Just accept it as the way we write these things.