Abstract
We consider the torsional rigidity and the principal eigenvalue related to the p-Laplace operator. The goal is to find upper and lower bounds to products of suitable powers of the quantities above in various classes of domains. The limit cases \(p=1\) and \(p=\infty \) are also analyzed, which amount to consider the Cheeger constant of a domain and functionals involving the distance function from the boundary.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
In this paper we consider the problem of minimizing or maximizing the quantity
on the class of open sets \(\Omega \subset {\mathbb {R}}^d\) having a prescribed Lebesgue measure, where \(\alpha ,\beta \) are two real parameters, and \(\lambda _p(\Omega )\), \(T_p(\Omega )\) are respectively the principal eigenvalue and the torsional rigidity, which are defined below, relative to the p-Laplace operator
In all the paper, we use the following notation:
-
\(p'\) is the conjugate exponent of p given by \(p':=p/(p-1)\);
-
\(\Omega \subset {\mathbb {R}}^d\) is an open set with finite Lebesgue measure \(|\Omega |\);
-
\(d_\Omega \) is the distance function from \(\partial \Omega \)
$$\begin{aligned} d_\Omega (x):=\inf \big \{|x-y|\ :\ y\in \partial \Omega \big \}; \end{aligned}$$ -
\(\rho (\Omega )\) is the inradius of \(\Omega \)
$$\begin{aligned} \rho (\Omega ):=\Vert d_\Omega \Vert _{L^\infty (\Omega )}, \end{aligned}$$corresponding to the maximal radius of a ball contained in \(\Omega \);
-
\({{\,\mathrm{diam}\,}}(\Omega )\) is the diameter of \(\Omega \)
$$\begin{aligned} {{\,\mathrm{diam}\,}}(\Omega ):=\sup \big \{|x-y|\ :\ x,y\in \Omega \big \}; \end{aligned}$$ -
\(P(\Omega )\) is the distributional perimeter of \(\Omega \) in the De Giorgi sense, defined by
$$\begin{aligned} P(\Omega ):=\sup \left\{ \int _\Omega {{\,\mathrm{div}\,}}\phi \,dx\ :\ \phi \in C^1_c({\mathbb {R}}^d;{\mathbb {R}}^d),\ \Vert \phi \Vert _{L^\infty ({\mathbb {R}}^d)}\le 1\right\} ; \end{aligned}$$ -
\(h(\Omega )\) is the Cheeger constant of \(\Omega \), that we define in Sect. 5;
-
\(B_r\) is the open ball in \({\mathbb {R}}^d\) centered at the origin with radius r and \(\omega _d:=|B_1|\);
-
\({\mathcal {H}}^{d-1}\) is the \(d-1\) dimensional Hausdorff measure.
Given \(1<p<\infty \), \(T_p(\Omega )\) denotes the p-torsional rigidity of \(\Omega \), defined by
where \(W^{1,p}_0(\Omega )\) stands for the usual Sobolev space obtained as the completion of the space \(C^{\infty }_c(\Omega )\) with respect to the norm \(\Vert \nabla u\Vert _{L^p(\Omega )}\). Equivalently, if \(w_p\) is the unique weak solution of the nonlinear PDE
we can define \(T_p(\Omega )\) as (see [7, Proposition 2.2]):
Note that \(w_p\) is a nonnegative function and (1.2) is the Euler–Lagrange equation of the variational problem
where
Multiplicating by \(w_p\) in (1.2) and integrating by parts gives
When \(\Omega =B_1\), the solution \(w_p\) to the boundary problem (1.2) is explicit and given by
which leads to
The p-principal eigenvalue \(\lambda _p(\Omega )\) is defined through the Rayleigh quotient
Equivalently, \(\lambda _p(\Omega )\) denotes the least value \(\lambda \) such that the nonlinear PDE
has a nonzero solution; we recall that in dimension 1 we have (see for instance [22])
while in higher dimension the following estimate holds true, see [20, Theorem 3.1]:
It is easy to see that the two quantities above scale as
By using a symmetrization argument and the so-called Pólya–Szegö principle (see [18]) it is possible to prove that balls maximize \(T_p\) (respectively minimize \(\lambda _p\)) among all sets of prescribed Lebesgue measure, which can be written in a scaling free form as
where B is any ball in \({\mathbb {R}}^d\). The inequalities (1.10) are known respectively as Faber–Krahn inequality and Saint-Venant inequality.
Moreover, we have:
To prove (1.11) and (1.12) it is enough to take into account of the scaling properties (1.9) and use the fact that if \(\Omega \) is the disjoint union of a family of open sets \(\Omega _i\) with \(i\in I\), then
Then, choosing \(\Omega _n\) as the disjoint union of n balls with measure 1/n each and taking the limit as \(n\rightarrow \infty \), gives
and
Thus, a characterization of \(\inf /\sup \) of the quantity \(\lambda _p^\alpha (\Omega )T_p^\beta (\Omega )\), among the domains \(\Omega \subset {\mathbb {R}}^d\) with unitary measure, when \(\alpha =0\) or \(\beta =0\) or \(\alpha \beta <0\), follows by (1.10), (1.11) and (1.12).
It remains to consider the case \(\alpha >0\) and \(\beta >0\). Setting \(q=\beta /\alpha >0\) we can limit ourselves to deal with the quantity
Using the scaling properties (1.9) we can remove the constraint of prescribed Lebesgue measure on \(\Omega \) by normalizing the quantity \(\lambda _p(\Omega )T_p^q(\Omega )\), multiplying it by a suitable power of \(|\Omega |\). We then end up with the scaling invariant shape functional
that we want to minimize or maximize over the class of open sets \(\Omega \subset {\mathbb {R}}^d\) with \(0<|\Omega |<\infty \).
The limit cases, when \(p=1\) and \(p=+\infty \), are also meaningful. When \(p\rightarrow 1\) the quantities \(\lambda _p(\Omega )\) and \(T_p(\Omega )\) are related to the notion of Cheeger constant \(h(\Omega )\), see definition (5.1). In particular we obtain as a natural “limit” functional
whose optimization problems are well studied in the literature. Concerning the case \(p=+\infty \), we show that the family \(F_{p,q}^{1/p}\) pointwise converges, as \(p\rightarrow \infty \), to the shape functional
and we study the related optimization problems in the class of all domains \(\Omega \) and in that of convex domains.
The study of the functionals \(F_{p,q}\) has been already considered in the literature. The case when \(p=2\) has been extensively discussed in [30,31,32,33] (see also [9]) and our results can be seen as natural extensions. An interesting variant, where the shape functionals involve the \(L^\infty \) norm of the function \(w_p\) solution of (1.2) has been considered in [19] in the case \(p=2\).
The paper is organized as follows. In the first three sections we study the optimization problems for \(F_{p,q}\), when \(1<p<\infty \) and in different classes of domains. More precisely: in Sect. 2 we consider the class of all open sets of \({\mathbb {R}}^d\) with finite Lebesgue measure, in Sect. 3 we consider the class of bounded convex open sets and in Sect. 4 that of thin domains which will be suitable defined. The analysis of the optimization problems in the extremal cases (respectively when \(p=1\) and \(p=+\infty \)) are contained in Sects. 5 and 6. Finally Sect. 7 contains a list of several open problems which we believe may be interest for future researches. For the sake of completeness we add an appendix section devoted to clarify the assumptions we use for the limit case of Sect. 6.
2 Optimization for general domains
The crucial inequality to provide a lower bound to \(F_{p,q}\) is the Kohler-Jobin inequality, first proved for \(p=2\) in [25, 26] and then for a general p in [7], which asserts that balls minimize principal frequency among all sets of prescribed torsional rigidity. More precisely we have
Proposition 2.1
Let \(1<p<+\infty \). Then
where B is any ball in \({\mathbb {R}}^d\).
Proof
We denote for the sake of brevity \({{\bar{q}}}=p'/(p'+d)\). Notice that
and thus
By Kohler-Jobin inequality (2.1) and Saint-Venant inequality (1.10) we get the thesis for \(0<q\le {{\bar{q}}}\). Now, let \(\Omega \) be the disjoint union of \(B_1\) and N disjoint balls of radius \({\varepsilon }\in (0,1]\). Taking into account (1.13) we have
Taking now \(N{\varepsilon }^{d+p/(p-1)}=1\) gives
which vanishes as \({\varepsilon }\rightarrow 0\) as soon as \(q>{{\bar{q}}}\). \(\square \)
In dealing with the supremum of \(F_{p,q}\) a natural threshold arises from the Polya inequality whose brief proof we recall.
Proposition 2.2
For every \(\Omega \subset {\mathbb {R}}^d\) with \(0<|\Omega |<+\infty \) and every \(1<p<+\infty \) we have
Proof
Let \(w_\Omega \) be the solution to (1.2). By the definition of \(\lambda _p(\Omega )\) and by Hölder inequality we have
The conclusion follows by (1.3). \(\square \)
Proposition 2.3
Let \(1<p<\infty \). Then
Proof
Let \(\Omega _N\) be the disjoint union of N balls of unitary radius. By (1.13) we have
Taking the limit as \(N\rightarrow \infty \) we have \(F_{p,q}(\Omega _N)\rightarrow +\infty \) whenever \(0<q<1\). Moreover, when \(q\ge 1\), using Proposition 2.2 and the Saint-Venant inequality (1.10), we have
which concludes the proof. \(\square \)
When \(p=2\) and \(q=1\) the upper bound given in the Proposition 2.2 is sharp as first proved in [32]. Using the theory of capacitary measures, a shorther proof was given in [30]. The latter extends, naturally, to the case when \(p\le d\) and \(q=1\) as we show in the proposition below.
Proposition 2.4
Let \(1<p\le d\). Then
Proof
By repeating the construction made in [13] (see also Remark 4.3.11 and Example 4.3.12 of [10], and references therein) we have that for every p-capacitary measure \(\mu \) (that is a nonnegative Borel measure, possibly taking the value \(+\infty \), with \(\mathrm {cap}_p(E)=0\Longrightarrow \mu (E)=0\)) there exists a sequence \((\Omega _n)\) of (smooth) domains such that
where
Taking the ball \(B_1\) and for every \(c>0\), we have
Clearly \(\lambda _p(\mu _c)=c+\lambda _p(B_1)\). Now, consider for \(\delta >0\) the function
We have
Therefore
By letting \(c\rightarrow +\infty \) and then \(\delta \downarrow 0\) we obtain the thesis. \(\square \)
3 Optimization in convex domains
We now deal with the optimization problems in the class of convex domains. Notice that adding in (1.11) and in (1.12) a convexity constraint on the admissible domains \(\Omega \) does not change the values of \(\inf \) and \(\sup \). To see this one can take a unit measure normalization of the following convex domains (slab shape)
being A a convex \(d-1\) dimensional open set with finite \(d-1\) dimensional measure and use the following Lemma, which will be proved in a slightly more general version in Proposition 4.1 of Sect. 4.
Lemma 3.1
Let \(A\subset {\mathbb {R}}^{d-1}\) be a bounded open set and let \({\varepsilon }>0\). Let \(C_{A,{\varepsilon }}:=A\times (-{\varepsilon },{\varepsilon })\). Then we have
where \(\pi _p\) is given in (1.7). In addition, as \({\varepsilon }\rightarrow 0\), we have
By using the previous lemma we have also
Hence the only interesting optimization problems in the class of convex domains are the following ones
We denote respectively by \(m_{p,q}\) and \(M_{p,q}\) the two quantities above.
With the convexity constraint, the so called Hersch–Protter inequality holds (for a proof see for instance [8, 22]):
Moreover, the p-torsional rigidity of a bounded convex open set satisfies the following generalization of Makai inequality (see [28, Theorem 4.3]):
Both inequalities are sharp and the equality is asymptotically attained by taking, for instance, the sequence \(C_{A,{\varepsilon }}\) of Lemma 3.1. Taking advantage of (3.2) and (3.3) we can show the following bounds.
Proposition 3.2
Let \(1<p<+\infty \). Then
Proof
Let \(\Omega \subset {\mathbb {R}}^{d}\) be any bounded convex set. Without loss of generality, we can suppose \(0\in \Omega \). We denote by \(j_{\Omega }(x)\) the Minkowski functional (also known as gauge function) of \(\Omega \), that is
The main properties of \(j_{\Omega }\) are summarized in Lemma 2.3 of [9]. In particular we recall that \(j_\Omega \) is a convex, Lipschitz, 1-positively homogeneous function, \({\mathcal {H}}^{d-1}\)-a.e. differentiable in \(\partial \Omega \), and satisfies
being \(\nu _{\Omega }(x)\) the outer normal unit versor at the point \(x\in \partial \Omega \). We consider
By using coarea formula (3.6) and the divergence theorem it is easy to prove that
and
where the last inequality follows by the fact that
see Lemma 2.1 in [9]. Hence by testing (1.1) with the function u we have
Taking into account (3.2), we obtain
which proves (3.4).
To prove the second inequality we use (3.3) and the inequality
to obtain
which, together with Proposition 2.2, gives (3.5). \(\square \)
Remark 3.3
We stress here that inequality (3.7) has been already proved in [12, 28]. However, their results are given in the more general anisotropic setting where the proofs become more involved.
Remark 3.4
Combining inequalities (3.5) and (1.8), we obtain
Thereby, as soon as p is large enough, we have \(M_{p,1}<1\).
When \(q\ne 1\) the values \(m_{p,q}\) and \(M_{p,q}\) are achieved by some optimal domains, as shown in the next theorem.
Theorem 3.5
Let \(1<p<+\infty \). Then
Moreover, there exist convex domains \(\Omega ^m_{p,q}\) and \(\Omega ^M_{p,q}\) such that
Proof
The first part follows at once using Saint-Venant inequality (1.10) together with the equality
Concerning the existence of optimal convex domains, we can repeat the argument used in [30]. First we notice that
Moreover, any convex open set \(\Omega \) contains a two-sided cone with base area equal to a \(d-1\) dimensional disk of radius \(\rho (\Omega )\) and total height equal to \({{\,\mathrm{diam}\,}}(\Omega )\), hence
Thus, suppose \(0<q<1\) and let \((\Omega _n)\) be a minimizing sequence for \(F_{p,q}\) made up of convex domains. By scaling invariance we can suppose \(\rho (\Omega _n)=1\). For n large enough we have \(F_{p,q}(\Omega _n)\le F_{p,q}(B)\). Using (3.2) and (3.8) we have
Combining the last estimate with (3.9) we have
Hence, up to translations, the whole sequence \((\Omega _n)\) is contained in a compact set and we can extract a subsequence \((\Omega _{n_k})\) which converges in both Hausdorff and co-Hausdorff distance to some \(\Omega ^m_{p,q}\) (see [17], for details about these convergences). Using the well-known continuity properties for \(\lambda _p\), \(T_p\) and Lebesgue measure with respect to Hausdorff metrics on the class of bounded convex sets, we conclude that
If \(q>1\) we can follow the similar strategy and consider a maximizing sequence \((\Omega _n)\) with unitary inradius. By (3.8) and (3.2) we have, for n large enough,
which, thanks to (3.9), implies again \(\sup _n{{\,\mathrm{diam}\,}}(\Omega _n)<+\infty \). \(\square \)
4 Optimization for thin domains
In this section we study the optimization problems for the functionals \(F_{p,1}\) in the class of the so-called thin domains, which has been already considered in [30] for \(p=2\). By a thin domain we mean a family of open sets \((\Omega _{\varepsilon })_{{\varepsilon }>0}\), of the form
where A is \((d-1)\)-dimensional open set, \(h_-,h_+\) are real bounded measurable functions defined on A and \({\varepsilon }\) is a small parameter. We assume \(h_+\ge h_-\) and we denote by h(x) the local thickness function
Moreover we say that the thin domain \((\Omega _{\varepsilon })_{{\varepsilon }>0}\) is convex if the corresponding domain A is convex and the local thickness function h is concave. The volume of \(\Omega _{\varepsilon }\) is clearly given by
while we can compute the behaviour (as \({\varepsilon }\rightarrow 0\)) of \(T_p(\Omega _{\varepsilon })\) and \(\lambda _p(\Omega _{\varepsilon })\) by means of the following proposition (in the case \(p=2\) a more refined asymptotics can be found in [5, 6]). From now on, we write the norms \(\Vert \cdot \Vert _{p}\), omitting the dependence on the domain.
Proposition 4.1
Let \(A\subset {\mathbb {R}}^{d-1}\) be an open set with finite \({\mathcal {H}}^{d-1}\)-measure and \(h_-,h_+\in C^{1}(A)\) with \(h_+>h_-\). Let \(\Omega _{\varepsilon }\) be defined by (4.1). We have
where \(\pi _p\) is given in (1.7). In addition, as \({\varepsilon }\rightarrow 0\), we have
Proof
First we deal with inequalities (4.2). Let \(\phi \in C^{\infty }_c(\Omega _{\varepsilon })\); since the function \(\phi (x,\cdot )\) is admissible to compute \(T_p\left( {\varepsilon }h_-(x),{\varepsilon }h_+(x)\right) \), by (1.1) we obtain
Taking into account (1.5) we have
and thus, integrating on A in (4.4), we deduce
Hölder inequality now gives
Since \(\phi \) is arbitrary and \(p'+1=(2p-1)/(p-1)\), we conclude that
To get the second inequality in (4.2) we notice that, by (1.6), for every \(\phi \in C^\infty _c(\Omega _{\varepsilon })\) we have
Since
integrating on A and minimizing on \(\phi \), we obtain
We now prove (4.3) for \(T_p(\Omega _{\varepsilon })\). To this end we consider the function
where w denotes the solution to (1.2) when \(\Omega =(0,1)\) and \(d=1\) (for the sake of brevity we omit the dependence on p). Notice that \(w_{\varepsilon }(x,\cdot )\) solves (1.2) in the interval \(({\varepsilon }h_-(x),{\varepsilon }h_+(x))\). In particular, by using (1.3) and (4.5), we have
A simple computation shows that
where
and
In particular
By exploiting the change of variable \(z=\frac{y-{\varepsilon }h_-(x)}{{\varepsilon }h(x)}\) in the latter identity, we conclude that, as \({\varepsilon }\rightarrow 0\),
Let \(\phi \in C^\infty _c(A)\). Since the function \(v(x,y)=\phi (x)w_{\varepsilon }(x,y)\) is admissible in (1.1), we get
Moreover, by using basically the same argument as above, we have also that
By combining (4.6) and (4.7) we obtain
Finally, by taking \(\phi \) which approximates \(1_A\) in \(L^p(A)\) in the right hand side of the inequality above, we conclude that
and the thesis is achieved taking into account (4.2). The asymptotics in (4.3) for \(\lambda _p\) can be treated with similar arguments. \(\square \)
Actually, by means of a density argument, we can drop the regularity assumptions on \(h_+\) and \(h_-\) and extend the formulas (4.2) and (4.3) to any family \((\Omega _{\varepsilon })_{{\varepsilon }>0}\) defined as in (4.1), with \(h_+\) and \(h_-\) bounded and measurable functions. We thus have:
where
We then define the functional \(F_{p,1}\) on the thin domain \((\Omega _{\varepsilon })_{{\varepsilon }>0}\) associated with the \(d-1\) dimensional domain A and the local thickness function h by
Our next goal is to give a complete solution to the optimization problems for the functional \(F_{p,1}\) in the class of convex thin domains. To this aim we recall the following result (see Theorem 6.2 in [4]).
Theorem 4.2
Let \(E\subset {\mathbb {R}}^N\) be a bounded open convex set, such that \(0\in E\) and let \(1\le s<r<\infty \). Then for every continuous function \(h:E\rightarrow {\mathbb {R}}^+\) satisfying
and such that \(\Vert h\Vert _{L^\infty (E)}=1\), it holds
where
In addition, equality occurs if E is a ball of radius 1 and \(h(x)=1-|x|\).
As an application we obtain the following lemma, which generalizes Proposition 5.2 in [30].
Lemma 4.3
Let \(E\subset {\mathbb {R}}^N\) be a bounded open convex set and let \(1<r<\infty \). Then for every concave function \(h:E\rightarrow {\mathbb {R}}^+\) with \(\Vert h\Vert _{L^\infty (E)}=1\) we have
In addition, the inequality above becomes an equality when E is a ball of radius 1 and \(h(x)=1-|x|\).
Proof
First we assume that \(E\subset {\mathbb {R}}^N\) is a ball centered in the origin and h is a radially symmetric, decreasing, concave function \(h:E\rightarrow [0,1]\) with \(h(0)=1\). Then h satisfies (4.9) and we can apply Theorem 4.2 with \(s=1\), to get
where
In order to get the inequality (4.10) in the general case, let \(h^*:B\rightarrow [0,1]\) be the radially symmetric decreasing rearrangement of h, defined on the ball B centered at the origin and with the same volume as E. The standard properties of the rearrangement imply that
Moreover, it is well-known that \(h^*\) is concave. Since \(h^*\) satisfies all the assumptions of the previous case, we get that \(h^*\) (hence h) satisfies (4.10). Finally, it is easy to show that the inequality in (4.10) holds as an equality for every cone function \(h(x)=1-|x|\). \(\square \)
We are now in a position to show the main theorem of this section.
Theorem 4.4
Let \(1<p<\infty \). Then
In addition, the first equality is attained taking h(x) to be any constant function while the second equality is attained taking as A the unit ball and as the local thickness function h(x) the function \(1-|x|\).
Proof
Using definition (4.8) it is straightforward to prove that
and to verify that, if h is constant, then
Finally, by applying Lemma 4.3 with \(N=d-1\), \(E=A\) and \(r=p'+1\) we obtain the second part of the theorem. \(\square \)
5 The case \(p=1\)
Given an open set \(\Omega \subset {\mathbb {R}}^d\) with finite measure we define its Cheeger constant \(h(\Omega )\) as
where \(E\Subset \Omega \) means that \({\bar{E}}\subset \Omega \). Notice that in definition (5.1), thanks to a well-known approximation argument, we can evaluate the quotient P(E)/|E| among smooth sets which are compactly contained in \(\Omega \). Following [24] we have
for every open set \(\Omega \) with finite measure.
Remark 5.1
A caveat is necessary at this point: the usual definition of Cheeger constant as
is not appropriate to provide the limit equality (5.2), which would hold only assuming a mild regularity on \(\Omega \) (for instance, it is enough to consider \(\Omega \) which coincides with its essential interior, see [27]). To prove that in general \(h(\Omega )\ne c(\Omega )\), one can consider \(\Omega =B_1{\setminus }\partial B_{1/2}\). Then \(c(\Omega )=c(B_1)=d\), while \(h(\Omega )=2d\). The latter follows from the fact that, if \(E\subset \Omega \), then \(E=E_1\cup E_2\) where \(E_1\Subset B_1{\setminus } {{\bar{B}}}_{1/2}\) and \(E_2\Subset B_{1/2}\), together with the equality
By the same argument used in [24] to prove (5.2) we can show that \(T_p(\Omega )\rightarrow h^{-1}(\Omega )\) as \(p\rightarrow 1\). For the sake of completeness we give the short proof below (see also Theorem 2 in [11]).
Proposition 5.2
Let \(\Omega \subset {\mathbb {R}}^d\) be an open set with finite measure. Then, as \(p\rightarrow 1\),
Proof
First we notice that for any \(u\in C^{\infty }_c(\Omega )\), it holds:
Indeed, by assuming without loss of generality that \(u\ge 0\), by coarea formula and Cavalieri’s principle, we have that
Since the sets \(\{u>t\}\Subset \Omega \) are smooth for a.e. \(t\in u(\Omega )\), (5.4) follows straightforwardly from (5.1). By combining (1.1) with (5.4) and Hölder inequality we then have
for any \(1<p<\infty \).
Now, let \(E_k\Subset \Omega \) be a sequence of smooth sets of \(\Omega \) such that \(P(E_k)/|E_k|\rightarrow h(\Omega )\). For a fixed k and any \({\varepsilon }>0\) small enough, we can find a Lipschitz function v compactly supported in \(\Omega \), such that,
where \(E_{k,{\varepsilon }}=E_k+B_{\varepsilon }\). Hence, by (1.1), we have
By first passing to the limit as \(p\rightarrow 1\), and then as \({\varepsilon }\rightarrow 0\) we get
which implies, as \(k\rightarrow \infty \),
Finally we conclude, taking into account (5.5). \(\square \)
The limits (5.2) and (5.3) justify the following definition:
Notice that \(F_{p,q}(\Omega )\rightarrow F_{1,q}(\Omega )\) as \(p\rightarrow 1\). In the next proposition we solve the optimization problems for \(F_{1,q}\) in the class of general domains and in that of convex domains.
Proposition 5.3
For \(0<q<1\), we have
For \(q>1\), we have
Proof
The minimality (respectively maximality) of B, for \(0<q<1\) (respectively for \(q>1\)), is an immediate consequence of the well known inequality
which holds for any \(\Omega \subset {\mathbb {R}}^d\) with finite measure. To prove the other cases we use the inequality
which holds for any \(\Omega \subset {\mathbb {R}}^d\) open, bounded, convex set (see [8, Corollary 5.2]). Then taking \(C_{A,{\varepsilon }}\) as in (3.1) we get
from which the thesis easily follows. \(\square \)
6 The case \(p=\infty \)
The limit behaviour of the quantities \(\lambda _p(\Omega )\), \(T_p(\Omega )\), as \(p\rightarrow \infty \), are well known for bounded open sets \(\Omega \subset {\mathbb {R}}^d\): in [14] and in [21] the authors prove that
while, following [3] (see also [23]) it holds \(w_p\rightarrow d_{\Omega }\) uniformly in \(\Omega \), which implies
Actually, in all these results, the boundedness assumption on \(\Omega \) is not needed, as it is only used to provide the compactness of the embedding \(W_0^{1,p}(\Omega )\) into the space \(C_0(\Omega )\) defined as the completion of \(C_c(\Omega )\) with respect to the uniform convergence. Indeed, this holds under the weaker assumption that \(|\Omega |<+\infty \) (see Appendix A for more details and for a \(\Gamma \)-convergence point of view of both limits (6.1) and (6.2)).
According to (6.1) and to (6.2) we define the shape functional \(F_{\infty ,q}\) as
Proposition 6.1
Let \(\Omega \subset {\mathbb {R}}^d\) be an open convex set. Then
Moreover, both inequalities are sharp. In particular
For its proof, we recall the following result, for which we refer to [4, 15].
Theorem 6.2
Let \(1\le q\le p\). Then for every convex set E of \({\mathbb {R}}^N\) \((N\ge 1)\) and every nonnegative concave function f on E we have
where the constant \(C_{p,q}\) is given by
In addition, the inequality above becomes an equality when E is a ball of radius 1 and \(f(x)=1-|x|\).
Proof of Proposition 6.1
In order to prove the right-hand side inequality in (6.4), for every \(t\ge 0\), we denote by \(\Omega (t)\) the interior parallel set at distance t from \(\partial \Omega \), i.e.
and by \(A(t):=|\Omega (t)|\). Moreover we set
Then for a.e. \(t\in (0,\rho (\Omega ))\) there exists the derivative \(A'(t)\) and it coincides with \(-L(t)\). Moreover, being \(\Omega \) a convex set, L is a monotone decreasing function. Then A is a convex function such that and \(A(0)=|\Omega |\). As a consequence we have
Integrating by parts, we get
The value 1/2 is asymptotically attained in (6.4) by considering a sequence of slab domains
as \({\varepsilon }\rightarrow 0\). Indeed, we have \(\rho (\Omega _{{\varepsilon }})={\varepsilon }/2\) and \(|\Omega _{{\varepsilon }}|={\varepsilon }\). Being
we get
Now we prove the left-hand side inequality in (6.4). Since \(\Omega \) is convex, the distance function \(d_\Omega \) is concave (see [2]); then, applying Theorem (6.2) to \(d_\Omega \), we obtain
Since (6.5) is an identity when \(\Omega =B\), \(C_{p,1}\) satisfies
As \(p\rightarrow \infty \) in (6.5), we obtain
which is an equality when \(\Omega =B\). \(\square \)
Remark 6.3
The proof of the right-hand side of (6.4) relies on the convexity properties of the function A(t). In the planar case a general result, due to Sz. Nagy (see [29]), ensures that, if \(\Omega \) is any bounded k-connected open set, (i.e. \(\Omega ^c\) has k bounded connected components), then the function
is convex. Therefore, for such an \(\Omega \), with the same argument as above it is easy to prove that
Hence, it is interesting to notice how, even when \(k=0,1\), the upper bound given in (6.4) remains sharp. In other words, in the maximization of \(F_{\infty ,1}\) on planar domains, there is no gain in replacing the class of convex domains by the larger one consisting of simply-connected domains or even more in allowing \(\Omega \) to have a single hole.
In the general case \(q\ne 1\) the optimization problems for the functional \(F_{\infty ,q}\) defined in (6.3) are studied below.
Corollary 6.4
If \(0<q<1\), then
If \(q>1\), then
Proof
Notice that
and that the inequality \(|\Omega |\ge \omega _d\rho (\Omega )^d\) holds for every open set \(\Omega \subset {\mathbb {R}}^d\) with equality when \(\Omega =B\). Thus, if \(0<q<1\), by (6.4) we have
while if \(q>1\), using again (6.4) we have
Finally, let \(\Omega _{\varepsilon }\) be the slab domain as in Proposition 6.1. Then
from which the thesis is achieved. \(\square \)
If we remove the convexity assumption on the admissible domains \(\Omega \), the picture is similar to those provided by Proposition 2.1 and Proposition 2.3. More precisely, if \(q>1/(d+1)\), the minimization problem for \(F_{\infty ,q}\) is ill posed. When \(q>1\), this follows directly by Corollary 6.4, while, in the case \(1/(d+1)<q\le 1\), by taking \(\Omega _n\) to be the union of n disjoint balls of radius \(r_j=j^{-1/d}\) with \(j=1,\dots ,n\), one can verify that \(F_{\infty ,q}(\Omega _n)\rightarrow 0\), as \(n\rightarrow \infty \). On the contrary, when \(q\le 1/(d+1)\), the minimum of \(F_{\infty ,q}\) is attained by any ball. Indeed, since \(1/(d+1)\le p'/(p'+d)\) for every \(p>1\), by using Proposition 2.1, we have
Hence, passing to the limit as \(p\rightarrow +\infty \), we obtain
Concerning the upper bound, Corollary 6.4 implies that the maximization problem is ill posed in the case \(q<1\), while, when \(q\ge 1\), using (6.6), we obtain
However, working with general domains provides an upper bound larger than in (6.4); for instance, in the two-dimensional case, taking as \(\Omega _N\) the unit disk where we remove N points as in Fig. 1, gives
where E is the regular exagon with unitary sides centered at the origin, as an easy calculation shows.
7 Further remarks and open questions
Several interesting problems and questions about the shape functionals \(F_{p,q}\) are still open; in this section we list some of them.
Problem 1
The characterization of the infimum of \(F_{p,q}\) in the class of all domains is well clarified in Proposition 2.1; on the contrary, for the supremum of \(F_{p,q}\), Proposition 2.3 only says it is finite for \(q\ge 1\). It would be interesting to know if the supremum can be better characterized, if it is attained, and in particular if it is attained for a ball when the exponent q is large enough (see also Problems 1 and 2 in [30]).
Problem 2
Concerning Problem 1 above, the case \(q=1\) is particularly interesting. The Polya inequality gives \(\sup F_{p,1}\le 1\), and Proposition 2.4 gives \(\sup F_{p,1}=1\) whenever \(p\le d\). It would be interesting to prove (or disprove) that \(\sup F_{p,1}<1\) for all \(p>d\).
Problem 3
In the convex setting, Proposition 3.2 provides some upper and lower bounds to \(F_{p,1}\) that however are far from being sharp. Even in the case \(p=2\), sharp values for the infimum and the supremum of \(F_{2,1}\) in the class of convex sets are unknown (see Conjecture 4.2. in [30]). It seems natural to conjecture that the right sharp inequalities are those given in Theorem 4.4 for \(F_{p,1}\) on the class of thin domain.
Problem 4
In the two-dimensional case with \(p=\infty \) we have seen that the domains \(\Omega _N\) in Fig. 1 give the asymptotic value \(\frac{1}{3}+\frac{\log 3}{4}\) for the shape functional \(F_{\infty ,1}\). It would be interesting to prove (or disprove) that this number is actually the supremum of \(F_{\infty ,1}(\Omega )\) when \(\Omega \) varies in the class of all bounded open two-dimensional sets. In addition, in the case of a dimension \(d>2\), it is not clear how a maximizing sequence \((\Omega _n)\) for \(F_{\infty ,1}\) has to be.
Change history
28 July 2022
Missing Open Access funding information has been added in the Funding Note.
References
Adams, R.A., Fournier, J.J.F.: Sobolev spaces. In: Pure and Applied Mathematics, vol. 140 (2nd edn). Elsevier/Academic Press, Amsterdam (2003)
Armitage, D.H., Kuran, U.: The convexity of a domain and the superharmonicity of the signed distance function. Proc. Am. Math. Soc. 93(4), 598–600 (1985)
Bhattacharya, T., DiBenedetto, E., Manfredi, J.: Limits as \(p\rightarrow \infty \) of \(\Delta _pu_p=f\) and related extremal problems. Rend. Sem. Mat. Univ. Politec. Torino, Special Issue 1989, 15–68 (1991)
Borell, C.: Integral inequalities for generalized concave or convex functions. J. Math. Anal. Appl. 43, 419–440 (1973)
Borisov, D., Freitas, P.: Asymptotics of Dirichlet eigenvalues and eigenfunctions of the Laplacian on thin domains in \({\mathbb{R}}^d\). J. Funct. Anal. 258, 893–912 (2010)
Borisov, D., Freitas, P.: Asymptotics for the expected lifetime of Brownian motion on thin domains in \({\mathbb{R}}^n\). J. Theor. Probab. 26, 284–309 (2013)
Brasco, L.: On torsional rigidity and principal frequencies: an invitation to the Kohler-Jobin rearrangement technique. ESAIM Control Optim. Calc. Var. 20, 315–338 (2014)
Brasco, L.: On principal frequencies and inradius in convex sets. Bruno Pini Math. Anal. Semin. 9, 78–101 (2018)
Brasco, L., Mazzoleni, D.: On principal frequencies, volume and inradius in convex sets. NoDEA Nonlinear Differ. Equ. Appl. 27(2), 26 (2020)
Bucur, D., Buttazzo, G.: Variational methods in shape optimization problems. In: Progress in Nonlinear Differential Equations, vol. 65. Birkhäuser, Basel (2005)
Bueno, H., Ercole, G.: Solutions of the Cheeger problem via torsion functions. J. Math. Anal. Appl. 381, 263–279 (2011)
Buttazzo, G., Bianco, S.G.L., Marini, M.: Sharp estimates for the anisotropic torsional rigidity and the principal frequency. J. Math. Anal. Appl. 457, 1153–1172 (2017)
Dal Maso, G., Malusa, A.: Approximation of relaxed Dirichlet problems by boundary value problems in perforated domains. Proc. R. Soc. Edinb. Sect. A 125, 99–114 (1995)
Fukagai, N., Ito, M., Narukawa, K.: Limit as \(p\rightarrow \infty \) of \(p\)-Laplace eigenvalue problems and \(L^\infty \)-inequality of the Poincaré type. Differ. Integral Equ. 12, 183–206 (1999)
Gardner, R.J., Zhang, G.: Affine inequalities and radial mean bodies. Am. J. Math. 120(3), 505–528 (1998)
Garroni, A., Nesi, V., Ponsiglione, M.: Dielectric breakdown: optimal bounds. Proc. R. Soc. Lond. A–457, 2317–2335 (2001)
Henrot, A., Pierre, M.: Variation et Optimisation de Formes. In: Une Analyse Géométrique. Mathématiques and Applications, vol. 48. Springer, Berlin (2005)
Henrot, A.: Extremum Problems for Eigenvalues of Elliptic Operators. Birkhäuser, Basel (2006)
Henrot, A., Lucardesi, I., Philippin, G.: On two functionals involving the maximum of the torsion function. ESAIM Control Optim. Calc. Var. 24(4), 1585–1604 (2018)
Huang, Y.X.: On the eigenvalues of the \(p\)-Laplacian with varying \(p\). Proc. Am. Math. Soc. 125(11), 3347–3354 (1997)
Juutinen, P., Lindqvist, P., Manfredi, J.J.: The \(\infty \)-Eigenvalue Problem. Arch. Ration. Mech. Anal. 148, 89–105 (1999)
Kajikiya, R.: A priori estimate for the first eigenvalue of the \(p\)-Laplacian. Differ. Integral Equ. 28, 1011–1028 (2015)
Kawohl, B.: On a family of torsional creep problems. J. Reine Angew. Math. 410, 1–22 (1990)
Kawohl, B., Fridman, V.: Isoperimetric estimates for the first eigenvalue of the \(p\)-Laplace operator and the Cheeger constant. Comment. Math. Univ. Carolin. 44(4), 659–667 (2003)
Kohler-Jobin, M.T.: Une méthode de comparaison isopérimétrique de fonctionnelles de domaines de la physique mathématique. II. Seconde partie: cas inhomogène: une inégalité isopérimétrique entre la fréquence fondamentale d’une membrane et l’énergie d’équilibre d’un problème de Poisson. Z. Angew. Math. Phys. 29, 767–776 (1978)
Kohler-Jobin, M.T.: Une méthode de comparaison isopérimétrique de fonctionnelles de domaines de la physique mathématique. I. Première partie: une démonstration de la conjecture isopérimétrique \(P\lambda ^2\ge \pi j_0^4/2\) de Pó1ya et Szegö. Z. Angew. Math. Phys. 29, 757–766 (1978)
Leonardi, G.P.: An overview on the Cheeger problem. In: New Trends in Shape Optimization. International Series of Numerical Mathematics, vol. 166, pp. 117–139. Birkhäuser/Springer, Cham (2015)
Pietra, F.D., Gavitone, N., Bianco, S.G.L.: On functionals involving the torsional rigidity related to some classes of nonlinear operators. J. Differ. Equ. 265, 6424–6442 (2018)
Sz.-Nagy, B.: Über Parallelmengen nichtkonvexer ebener Bereiche. Acta Sci. Math. (Szeged) 20, 36–47 (1959)
van den Berg, M., Buttazzo, G., Pratelli, A.: On the relations between principal eigenvalue and torsional rigidity. Commun. Contemp. Math. 23(8), 28 pages (2021)
van den Berg, M., Buttazzo, G., Velichkov, B.: Optimization problems involving the first Dirichlet eigenvalue and the torsional rigidity. In: “New Trends in Shape Optimization”, pp. 19–41. Birkhäuser, Basel (2015)
van den Berg, M., Ferone, V., Nitsch, C., Trombetti, C.: On Pólya’s inequality for torsional rigidity and first Dirichlet eigenvalue. Integral Equ. Oper. Theory 86, 579–600 (2016)
van den Berg, M., Ferone, V., Nitsch, C., Trombetti, C.: On a Pólya functional for rhombi, isosceles triangles, and thinning convex sets. Rev. Mat. Iberoam. 36, 2091–2105 (2020)
Acknowledgements
The work of GB is part of the Project 2017TEXA3H “Gradient flows, Optimal Transport and Metric Measure Structures” funded by the Italian Ministry of Research and University. The authors are member of the Gruppo Nazionale per l’Analisi Matematica, la Probabilità e le loro Applicazioni (GNAMPA) of the Istituto Nazionale di Alta Matematica (INdAM).
Funding
Open access funding provided by Università di Pisa within the CRUI-CARE Agreement.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by A. Mondino.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix A
Appendix A
We devote this Appendix to give a proof of the known asymptotics (6.1) and (6.2) by means of \(\Gamma \)-convergence when \(0<|\Omega |<+\infty \). We recall that if \(p>d\) and \(\Omega \subset {\mathbb {R}}^d\) is any (possibly unbounded) open set with finite measure we have the compact embedding:
A quick proof of it can be obtained by combining the Gagliardo–Niremberg inequality in \(W^{1,p}({\mathbb {R}}^d)\):
together with the well known facts that the inclusion \(W^{1,p}_0(\Omega )\hookrightarrow L^p(\Omega )\) is compact (thanks to the Riesz–Frechét–Kolmogorov Theorem) and the embedding (A.1) is continuous. Note that, in (A.2), C(d, p) denotes a positive constant depending on p and d, and \(\alpha =1-d/p\) (we refer to [1], Chapter 6, for a comprehensive discussion on necessary and sufficient conditions for the compactness of several embeddings of Sobolev spaces).
In particular, if we denote by \(W_0^{1,\infty }(\Omega )\) the closure of \(C_c^\infty (\Omega )\) with respect to the weak* convergence of \(W^{1,\infty }(\Omega )\), we have that \(u\in W^{1,\infty }_0(\Omega )\) if and only if \(u\in C_0(\Omega )\) and u is a Lipschitz continuous function on \(\Omega \). Moreover \(W_0^{1,\infty }(\Omega )\) can be easily characterized as:
Proposition A.1
Let \(\Omega \subset {\mathbb {R}}^d\) be an open set with finite measure and let \(\Psi _p,\Psi _\infty : L^{1}(\Omega )\rightarrow {{\bar{{\mathbb {R}}}}}\) be defined by
Then, as \(p\rightarrow \infty \), the sequence \(\Psi _p\) \(\Gamma \)-converges to \(\Psi _\infty \) with respect to the \(L^1\)-convergence.
Proof
Let \(p_n\rightarrow \infty \). The \(\Gamma \)-\(\limsup \) inequality is trivial since, for every \(u\in W^{1,\infty }_0(\Omega )\) with \(\Vert u\Vert _\infty =1\), the sequence \(u_{p_n}=\Vert u\Vert _{p_n}^{-1}u\) converges to u in \(L^1\) and satisfies
To prove the \(\Gamma \)-\(\liminf \) inequality, without loss of generality, let \(u\in L^{\infty }(\Omega )\), \((u_{p_n})\subseteq W^{1,p_n}_0(\Omega )\) be such that \(u_{p_n}\rightarrow u\) in \(L^1(\Omega )\), \(\Vert u_{p_n}\Vert _{p_n}=1\), and \(\liminf _{n\rightarrow \infty }\Psi _{p_n}(u_{p_n})=C<\infty \). Since for every \(q\ge 1\) and for n large enough it holds
we get that \(u_{p_n}\rightarrow u\) in \(L^q(\Omega )\), \(u\in W^{1,q}_0(\Omega )\) and
Moreover
which yields, as \(q\rightarrow \infty \), \(\Vert u\Vert _{\infty }\le 1\). Combining this estimate with (A.4), we get that \(u\in W^{1,\infty }(\Omega )\); hence, by (A.3), \(u\in W^{1,\infty }_0(\Omega )\). Thanks to the compact embedding of \(W_0^{1,q}(\Omega )\) in \(C_0(\Omega )\) when \(q>d\), we obtain that \(\Vert u_{p_n}- u\Vert _{\infty }\rightarrow 0\) as \(n\rightarrow \infty \) and, since
we get that \(\Vert u\Vert _\infty =1\). Finally, by letting \(n\rightarrow \infty \) in (A.4), it follows
\(\square \)
Corollary A.2
Let \(\Omega \subset {\mathbb {R}}^d\) be an open set with finite measure. Then, as \(p\rightarrow \infty \),
Proof
Using \(d_{\Omega }\) as a test function for (1.6) we have \(\limsup _{p\rightarrow \infty }\lambda _p(\Omega )\le \rho (\Omega )^{-1}\). Moreover we notice that, for any \(\phi \in W^{1,\infty }_0(\Omega )\), it holds
Let \(u_p\) be the (only) minimum of \(\Psi _p\), and let \(p_n\rightarrow \infty \). With the same argument of Proposition A.1 we can assume \(u_{p_n}\rightarrow u_\infty \) uniformly in \(L^{1}(\Omega )\), where \(u_\infty \in W^{1,\infty }_0(\Omega )\) and \(\Vert u_{\infty }\Vert _{\infty }=1\). Then, by Proposition A.1 and by (A.5), we have
The thesis follows by the arbitrariness of the sequence \(p_n\). \(\square \)
Next Proposition generalizes Proposition 2.1 in [16].
Proposition A.3
Let \(\Omega \subset {\mathbb {R}}^d\) be an open set with finite measure and let \(\Phi _p,\Phi _\infty :L^1(\Omega )\rightarrow {{\bar{{\mathbb {R}}}}}\) be defined by
Then, as \(p\rightarrow \infty \), the functionals \(\Phi _p\) \(\Gamma \)-converge to \(\Phi _\infty \) with respect to the \(L^1\)-convergence. Moreover, we have
Proof
Let \(p_n\rightarrow \infty \). The \(\Gamma \)-limsup inequality is trivial since for any \(u\in W^{1,\infty }_0(\Omega )\) with \(\Vert \nabla u\Vert _\infty \le 1\) we have \(u\in W_0^{1,p_n}(\Omega )\) for every \(n\in {\mathbb {N}}\) and thus
To prove the \(\Gamma \)-liminf inequality, we can assume \(u_n,u\in L^1(\Omega )\), \(u_n\rightarrow u\) in \(L^1(\Omega )\), \(u_n\in W_0^{1,p_n}(\Omega )\), and
If \(q>1\) and \(p_n>q\) we have
which forces \(u\in W^{1,q}_0(\Omega )\). Moreover, since
we have also \(\Vert \nabla u\Vert _\infty \le 1\). Therefore, by (A.3), \(u\in W_0^{1,\infty }(\Omega )\) and
The thesis follows by the arbitrariness of the sequence \(p_n\). \(\square \)
Corollary A.4
Let \(\Omega \subset {\mathbb {R}}^d\) be an open set with finite measure, let \(w_p\) be the solution to (1.2). Then, as \(p\rightarrow \infty \),
Proof
It is sufficient to show that \(w_p\rightarrow d_\Omega \) uniformly in \(\Omega \). First we notice that by (2.2) we get
By Corollary A.2 we have
Moreover, for every fixed \(q\ge 1\) and p large enough, by Hölder inequality, we have that
Let \(p_n\rightarrow \infty \). By applying (A.7) we can show that there exists \(w_{\infty }\in W^{1,\infty }_0(\Omega )\), such that \(w_{p_n}\) converges uniformly to \(w_\infty \) and weakly in \(W^{1,q}(\Omega )\) for every \(q\ge 1\). Notice that (A.6) combined with Corollary A.2 shows also
for every \(q\ge 1\), which implies \(\Vert w_\infty \Vert \le 1\).
Now let \(J_p\) be the functional defined in (1.4) and \(J_\infty \) be the functional given by
Since the functional \(u\mapsto \int _\Omega u\,dx\) is continuous with respect to the \(L^1\)-convergence, thanks to Proposition A.3, we have that
Moreover, by using (A.5) we have \(w_\infty (x)\le d_\Omega (x)\). In addition, since \(d_\Omega \in W^{1,\infty }_0(\Omega )\), we have also \(J_\infty (w_\infty )\le J_\infty (d_\Omega )\), i.e. \(\int _\Omega (w_\infty -d_\Omega )\,dx\ge 0\). Hence \(d_\Omega =w_\infty \). By the arbitrariness of the sequence \(p_n\), we get that \(w_p\rightarrow d_{\Omega }\) uniformly as \(p\rightarrow \infty \). \(\square \)
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Briani, L., Buttazzo, G. & Prinari, F. Inequalities between torsional rigidity and principal eigenvalue of the p-Laplacian. Calc. Var. 61, 78 (2022). https://doi.org/10.1007/s00526-021-02129-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00526-021-02129-9