Let $\Omega \subset \mathbb{R}^n$ be a domain. Consider the space of test functions $\mathcal{D}((0,T)\times \Omega)$ and the space of distributions $\mathcal{D}^*((0,T)\times \Omega).$

Is it true that $L^2(0,T;H^1(\Omega))^* = L^2(0,T;H^{-1}(\Omega))$ is a subset of $\mathcal{D}^*((0,T)\times \Omega)$?

I've not been able to prove this. Given $f \in L^2(0,T;H^{-1}(\Omega))$, we have that $f \in \mathcal{D}^*((0,T)\times \Omega)$ if for $v_n \to v$ in $\mathcal{D}((0,T)\times \Omega)$, it holds that $f(v_n) \to f(v)$ in $\mathbb{R}$ (linearity is easy to check).

How to show this? We know that $D^\alpha v_n \to D^\alpha v$ uniformly for all multiindices $\alpha$, but I can't get $L^p$ convergence from this. Help appreciated.


$D((0,T)\times \Omega)\subset L^2((0,T);H^1(\Omega))\subset L^1_{loc}((0,T);H^1(\Omega))$.

Define $$f(v) = \int_0^Tdt \langle f(t,\cdot),v (t,\cdot)\rangle_{H^{-1};H^1},$$ then $$|f(v)|\le \int_0^Tdt \| f(t,\cdot)\|_{H^{-1}}\|v (t,\cdot)\|_{H^1} \le | K |\int_0^Tdt \| f(t,\cdot)\|_{H^{-1}}\sqrt{\|v (t,\cdot)\|_{L^\infty}^2+\|\nabla_x v (t,\cdot)\|_{L^\infty}^2 },$$ where $K$ is a projection of the support of $v$ on $\Omega$. Now we need only to deal with technical details $$|f(v)|\le |K|T \|v\|_{C^1((0,T)\times\Omega)} \int_0^Tdt \| f(t,\cdot)\|_{H^{-1}} \le |K|T^{3/2} \|f\|_{L^2((0,T);H^{-1}(\Omega))}\|v\|_{C^1((0,T)\times\Omega)} . $$ This proves that $f(v)$ is a distribution.

  • $\begingroup$ Thanks but I don't understand why this means that $f$ is a distribution. You've shown that $f \in (C^1((0,T)\times\Omega))^*$ but this doesn't mean $f \in \mathcal{D}^*((0,T)\times \Omega).$ $\endgroup$ – maximumtag Mar 5 '14 at 19:11
  • $\begingroup$ $C^1((0,T)\times\Omega)\supset\mathcal{D} ((0,T)\times \Omega) $, hence $(C^1((0,T)\times\Omega))^\ast\subset\mathcal{D}^\ast ((0,T)\times \Omega) $. $\endgroup$ – TZakrevskiy Mar 5 '14 at 20:48
  • $\begingroup$ Is that obvious? I know when $X \subset Y$ are normed spaces with continuous embedding that $Y^* \subset X^*$ but $\mathcal{D}^*$ is not a normed space.. it still holds? $\endgroup$ – maximumtag Mar 5 '14 at 22:05
  • $\begingroup$ @maximumtag $D^\ast$ are linear functionals continuous in the topology of multinorm given by seminorms $p_\alpha(\phi)=\sup_{\beta\le\alpha}\sup|D^\alpha\phi|$, $\alpha,\,\beta\in \Bbb N^d$. We proved that $f$ is a linear functional which is continuous in one of that seminorms (namely, $p_{1,1,\dots,1}$), hence it is continuous in the topology of multinorm. $\endgroup$ – TZakrevskiy Mar 6 '14 at 9:59
  • 1
    $\begingroup$ @maximumtag the inclusion of topological duals bases on the inclusion of topologies, the presence of the norm is irrelevant. $\endgroup$ – TZakrevskiy Mar 6 '14 at 10:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.