Let's consider a function $f\in L^2(\mathbb{R})$ for which the second weak derivative exists and lie in $L^2(\mathbb{R})$, i.e. there exists $f''\in L^2(\mathbb{R})$ such that for all $\varphi\in C_0^\infty(\mathbb{R})$ the following integral equation stands: $$ \int\limits_\mathbb{R}f(x)\varphi''(x)dx=\int\limits_\mathbb{R}f''(x)\varphi(x)dx. $$
My question is, having this can we assume that weak $f'$ also exists in $L^2(\mathbb{R})$?
Suppose we found a normal (not generalized) function $g:\mathbb{R}\to\mathbb{C}$ such that for all $\varphi\in C_0^\infty(\mathbb{R})$ $$ \int\limits_\mathbb{R}f(x)\varphi'(x)dx=-\int\limits_\mathbb{R}g(x)\varphi(x)dx. $$ Then $$ \langle -f'', f \rangle_{L^2}=-\int\limits_\mathbb{R}f''(x)f(x)dx=\int\limits_\mathbb{R}g(x)g(x)dx=\|g\|_{L^2}^2 $$ which means, that $g$ is in $L^2(\mathbb{R})$. But what guarantees us the existence of such $g$?