Journal of Inequalities in Pure and Applied Mathematics LINEAR ELLIPTIC EQUATIONS AND GAUSS MEASURE G. DI BLASIO Dipartimento di Matematica e Applicazioni “R. Caccioppoli” Università degli Studi di Napoli “Federico II” Complesso Monte S. Angelo - Via Cintia. Napoli-ITALY EMail: giuseppina.diblasio@dma.unina.it volume 4, issue 5, article 106, 2003. Received 20 February, 2003; accepted 30 June, 2003. Communicated by: A. Fiorenza Abstract Contents JJ J II I Home Page Go Back Close c 2000 Victoria University ISSN (electronic): 1443-5756 020-03 Quit Abstract In this paper we study a Dirichlet problem relative to a linear elliptic equation with lower-order terms, whose ellipticity condition is given in terms of the n function ϕ(x)=(2π)− 2 exp(−|x|2 /2) , the density in the Gaussian measure. Using the notion of rearrangement with respect to the Gauss measure, we prove a comparison result with a problem of the same type defined in a half space, with data depending only on the first variable. 2000 Mathematics Subject Classification: 35B05, 35J70 Key words: Linear elliptic equation, Comparison theorem, Rearrangements of functions, Gauss measure Contents Linear Elliptic Equations and Gauss Measure G. di Blasio Title Page Contents 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2 Notations and Preliminary Results . . . . . . . . . . . . . . . . . . . . . . 6 3 Comparison Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 References JJ J II I Go Back Close Quit Page 2 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au 1. Introduction The object of this paper is to give comparison results for the solution of the problem in Ω − (aij (x)uxi )xj + bi (x)uxi + c(x)u = f (x)ϕ(x) (1.1) u=0 on ∂Ω n where ϕ(x) = (2π)− 2 exp − |x|2 /2 is the density in the Gaussian measure, Ω is an open set of Rn (n ≥ 2) which has Gauss measure less than one, aij (x), bi (x) and c(x), i, j = 1, . . . , n, are measurable functions on Ω such that (i) aij (x)ξi ξj ≥ ϕ(x) |ξ|2 , P 1 (ii) ( b2i ) 2 ≤ ϕ(x)B, (iii) c (x) ≥ c0 (x) ϕ(x), ∀ξ ∈ Rn , a.e. x ∈ Ω, G. di Blasio Title Page Contents c0 (x) ∈ L∞ (Ω), and f is taken in a suitable weighted Lp space in order to guarantee the existence of a solution u of the problem (1.1). We recall that u ∈ H01 (ϕ, Ω) is a weak solution of the problem (1.1), if JJ J II I Go Back Close Quit Z (1.2) Linear Elliptic Equations and Gauss Measure (aij (x)uxi ψxj + bi (x)uxi ψ + c(x)u(x)ψ)dx Ω Z = f (x)ϕ(x)ψdx, ∀ψ ∈ H01 (ϕ, Ω). J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 Ω http://jipam.vu.edu.au Page 3 of 24 Let us observe that the operator in (1.1) is uniformly elliptic if Ω is bounded. It is well known that when Ω is bounded, comparison results for elliptic problems have been obtained via Schwarz symmetrization, for a simpler problem which is defined in a ball and has spherically symmetric data (see for example [1], [2], [4], [3], [6], [17], [20], [19], [18], [21]). In our case Ω can be not bounded and ellipticity condition (i) is given in terms of the density in the Gaussian measure. We consider solutions of problem (1.1) in the weighted Sobolev space H01 (ϕ, Ω) (see § 2) and we compare the solution of the problem (1.1) with the solution of a problem in which the data depend only on the first variable and the domain is a half-space which has the same Gauss measure as Ω. More precisely let Ω? = {x = (x1 , x2 , . . . , xn ) ∈ Rn : x1 > λ} such that γn (Ω? ) = γn (Ω), and let f ? (x) be the rearrangement with respect to the Gauss measure of the function f (x) (see § 2 for the definition). To give an example of the obtained results we consider the case c0 (x) = 0. Let w (x) = w (x1 ) be the solution of in Ω? , − (wx1 ϕ (x))x1 − Bwx1 ϕ (x) = f ? (x1 )ϕ (x) (1.3) w=0 on ∂Ω? . We prove that the pointwise comparison (1.4) u? (x) = u? (x1 ) ≤ w (x1 ) = w? (x) for a.e x = (x1 , x2 , . . . , xn ) ∈ Ω? holds, where u? and w? are the rearrangements with respect to the Gauss measure of u and w respectively. In this case the comparison also gives an explicit Linear Elliptic Equations and Gauss Measure G. di Blasio Title Page Contents JJ J II I Go Back Close Quit Page 4 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au estimate of u? that is also a condition for the existence of the solution for the problem (1.1) in terms of the existence of the solution of (1.3). If c0 (x) 6= 0, a comparison with a problem which also takes the term “c(x) u(x)” into account can be found. In this case, depending on the sign of c0 (x) , pointwise comparison (1.4) is false in general, but a comparison between the concentrations can be found (see Theorem 3.2). In the proofs of our results we use tools similar to the classical methods based on the isoperimetric inequality and Schwarz symmetrization (see [2]). In our case a fundamental rule is played by the isoperimetric inequality with respect to the Gauss measure. The problem (1.1) has been studied using rearrangement with respect to the Gauss measure in [7], when bi (x) = c(x) = 0. Existence results for weak solutions of problem (1.1) can be obtained for aij (x) c(x) example via the Lax Milgram theorem when (i) and (ii) hold and ϕ(x) , ϕ(x) ∈ ∞ 2 L (Ω) and f (x) ∈ L (ϕ, Ω) (see also [15], [22]). Linear Elliptic Equations and Gauss Measure G. di Blasio Title Page Contents JJ J II I Go Back Close Quit Page 5 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au 2. Notations and Preliminary Results In this section we recall some definitions and results which will be useful in what follows. Let Ω be an open set of Rn and let γn be the n−dimensional Gauss measure on Rn defined by ! 2 n |x| dx, x ∈ Rn γn (dx) = ϕ (x) dx = (2π)− 2 exp − 2 normalized by γn (Rn ) = 1. One of the main tools used to prove the comparison result is the isoperimetric inequality with respect to the Gauss measure, to recall this result we define the perimeter with respect to the Gauss measure as (see [13]) ! Z 2 |x| −n exp − Hn−1 (dx) , P (E) = (2π) 2 2 ∂E where E is a (n−1)−rectificable set and Hn−1 denotes the (n−1)−dimensional Hausdorff measure. For all λ ∈ R, we denote by H (ξ, λ) the half-space defined by H (ξ, λ) = {x ∈ Rn : (x, ξ) > λ} and we set H (ξ, λ) = Rn if λ = −∞ and H (ξ, λ) = ∅ if λ = +∞. It is well known (see [11]) that among all measurable sets of Rn with prescribed Gauss measure, the half-spaces take the smallest perimeter. In other words, the half-spaces are extremal in the isoperimetric problem for the Gauss Linear Elliptic Equations and Gauss Measure G. di Blasio Title Page Contents JJ J II I Go Back Close Quit Page 6 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au measure. From an appropriate form of the Brunn - Minkowski inequality, the isoperimetric inequality follows (see [8], [11], [10], [12]) P (E) ≥ P (H (ξ, λ)) for all subsets E ⊂ Rn such that γn (E) = γn (H (ξ, λ)). For the sake of simplicity we shall consider ξ = (1, 0, . . . , 0). Now we consider the notion of rearrangement with respect to the Gauss measure. If u is a measurable function in Ω, we define the distribution function of u, denoted by µ, as the Gauss measure of the level set of u, i.e. µ : t ∈ [0, ∞[ −→ µ (t) ∈ [0, 1] , Linear Elliptic Equations and Gauss Measure G. di Blasio where µ (t) = γn ({x ∈ Ω : |u| > t}) . ∗ We denote by u the decreasing rearrangement of u (with respect to the Gauss measure), i.e. u∗ (s) = inf {t ≥ 0 : µ (t) ≤ s} , s ∈ ]0, 1] , the functions µ and u∗ are decreasing and right - continuous. Moreover the increasing rearrangement of u is the function defined as follows ∗ u∗ (s) = u (γn (Ω) − s) , s ∈ [0, 1[ . Finally we define the rearrangement of u with respect to the Gauss measure, denoted by u? , as the function whose level sets are half-spaces having the same Gauss measure of the level sets of u. More precisely we define 2 Z +∞ t 1 n exp − dτ Φ (τ ) = γn ({x ∈ R : x1 > τ }) = √ 2 2π τ Title Page Contents JJ J II I Go Back Close Quit Page 7 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au for all τ ∈ R. Then u? is a map from Ω? into [0, +∞[ defined by u? (x) = u∗ (Φ (x1 )) , where Ω? = {x = (x1 , x2 , . . . , xn ) ∈ Rn : x1 > λ} such that γn (Ω? ) = γn (Ω). For statements about the properties of rearrangement with respect to a positive measure see, for example, [9], [20], [16]. We recall that if f (x) , g (x) are measurable functions, a Hardy type inequality (see [9]) Z Z Z Linear Elliptic Equations and Gauss Measure γn (Ω) f ∗ (s) g ∗ (s) ds G. di Blasio and a Polya-Szëgo inequality for a Lipschitz continuous function u (x) (see Title Page ? |f (x) g (x)| γn (dx) ≤ ? f (x) g (x) γn (dx) = Ω? Ω 0 Contents [20]) Z |∇u | γn (dx) ≤ (2.1) JJ J Z ? Rn |∇u| γn (dx) Rn II I Go Back holds. Moreover we have Close Z Ω |f (x)|p γn (dx) = Z Ω? |f ? (x)|p γn (dx) = Z γn (Ω) f ∗ (s)p ds, 0 that is, the Lp weighted norm is invariant under the rearrangement. Quit Page 8 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au Now define the weighted Sobolev space H01 (ϕ, Ω) as the closure of C0∞ (Ω) under the norm Z 12 2 (2.2) kukH 1 (ϕ,Ω) = |∇u (x)| dγn (x) . 0 Ω We remark that the following Poincarè type inequality can be proved. Proposition 2.1. Let Ω be a open subset of Rn with γn (Ω) < 1. For each function f ∈ H01 (ϕ, Ω) we have Linear Elliptic Equations and Gauss Measure kf kL2 (ϕ,Ω) ≤ C k∇f kL2 (ϕ,Ω) , G. di Blasio (2.3) where C is a constant depending on n and Ω. Proof. By (2.1) we have the following inequality (2.4) kf k2L2 (ϕ,Ω) k∇f k2L2 (ϕ,Ω) ≤ kf ? k2L2 (ϕ,Ω? ) k∇f ? k2L2 ? (ϕ,Ω ) 2 R +∞ ? x 2 f (x1 ) exp − 21 dx1 λ , =R 2 2 +∞ x1 d ? (x ) f exp − dx 1 1 dx1 2 λ where λ is such that γn (Ω? ) = γn (Ω) with Ω? = {x = (x1 , x2 , . . . , xn ) ∈ Rn : x1 > λ}. The ratio in (2.4) is bounded (see e.g. [14, Theorem 1.3.1./2]). Title Page Contents JJ J II I Go Back Close Quit Page 9 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au The previous Poincaré type inequality ensures the equivalence between the norm (2.2) and the following one 12 Z Z 12 2 2 |u (x)| dγn (x) kukH 1 (ϕ,Ω) = |∇u (x)| dγn (x) . + 0 Ω Ω Now, we recall the following lemmas which will be useful in the following. Lemma 2.2. Let f (x) , g (x) be measurable, positive functions such that Z α Z α f (x) dx ≤ g (x) dx, α ∈ [0, a] . 0 0 G. di Blasio If h(x) ≥ 0 is a decreasing function then Z α Z α f (x) h(x)dx ≤ g (x) h(x)dx, 0 Title Page α ∈ [0, a] . 0 Lemma 2.3. Let z be a bounded function, K a nonnegative integrable function, and ψ a function with bounded variation that vanishes at +∞. If Z ∞ z (t) ≤ K (s) z (s) ds + ψ (t) t for almost every t > 0, then Z z (t) ≤ t for almost every t > 0. Linear Elliptic Equations and Gauss Measure Contents JJ J II I Go Back Close Quit ∞ Z exp s K (τ ) dτ [−dψ (s)] Page 10 of 24 t J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au 3. Comparison Results In this section we prove some comparison results for the solution of the problem (1.1). We consider first the case c0 (x) = 0 and then the case c0 (x) 6= 0. Theorem 3.1. Let Ω be an open set of Rn with γn (Ω) < 1 and let u ∈ H01 (ϕ, Ω) be the solution of (1.1) with the assumptions (i), (ii) and (iii). Let c0 (x) = 0 and 2 Z +∞ 2 Z +∞ σ2 τ ? (3.1) exp f (σ) exp B (σ − τ ) − dσ dτ < +∞, 2 2 λ τ where λ is such that Ω? = {x1 > λ} . Then we have u? (x1 ) ≤ w(x) (3.2) for a.e. x ∈ Ω? Linear Elliptic Equations and Gauss Measure G. di Blasio and Z Z q |∇u| ϕ (x) dx ≤ (3.3) |∇w|q ϕ (x) dx for all 0 < q ≤ 2, Ω? Ω where 2 Z +∞ τ σ2 ? w (x)=w (x1 )= exp f (σ) exp B (σ − τ ) − dσ dτ 2 2 λ τ is the solution of the problem (1.3). Z x1 Remark 3.1. Condition (3.1) ensures the existence of a solution w (x1 ) = w? (x) ∈ H01 (ϕ, Ω) of (1.3). It is satisfied for a wide class of functions, for instance for functions f such that 1 σ2 ? f ≤ exp −B (σ − τ ) + (1 + |τ |) 2 − ∀τ ≥ λ 4 for some constants C > 0, > 0. Title Page Contents JJ J II I Go Back Close Quit Page 11 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au Remark 3.2. The assumption γn (Ω) < 1 is made to guarantee that the Poincarè type inequality (2.3) holds. Proof. If we choose in (1.2), for h > 0 and t ∈ [0, sup |u|[ h sgn u if |u| > t + h, (|u| − t) sgn u if t < |u| ≤ t + h, ψ(x) = 0 otherwise, we get Z 1 aij (x)uxi uxj dx h t<|u|≤t+h Z Z 1 + bi (x)uxi (|u| − t) sgn udx + bi (x)uxi sgn udx h t<|u|≤t+h |u|>t+h Z Z 1 + c(x) (|u| − t) sgn udx + c(x)u (x) sgn udx h t<|u|≤t+h |u|>t+h Z Z 1 = f (x)ϕ(x) (|u| − t) sgn udx + f (x)ϕ(x) sgn udx. h t<|u|≤t+h |u|>t+h By the ellipticity condition (i), (ii) and (iii) and letting h go to 0, we obtain Z d (3.4) − |∇u|2 ϕ(x)dx dt |u|>t Z Z ≤B |∇u| ϕ (x) dx + f (x) sgn uϕ(x)dx |u|>t |u|>t Linear Elliptic Equations and Gauss Measure G. di Blasio Title Page Contents JJ J II I Go Back Close Quit Page 12 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au On the other hand, the coarea formula (see [13]) and the isoperimetric inequality with respect to the Gauss measure give Z Z d (3.5) − |∇u| ϕ(x)dx ≥ ϕ(x)Hn−1 (dx) dt |u|>t ∂{|u|>t}? ! Φ−1 (µ (t))2 1 , = √ exp − 2 2π where {|u| > t}? is the half space having Gauss measure µ(t). Then using (3.5) and the Hölder inequality we obtain ! 2 −1 1 Φ (µ (t)) (3.6) 1 ≤ (2π) 2 exp 2 12 Z 1 d 2 0 2 × (−µ (t)) − |∇u| ϕ(x) dx . dt |u|>t Using (3.6) and the Hölder inequality, (3.4) becomes Z d − |∇u|2 ϕ(x)dx dt |u|>t Z ∞ Z 1 d 2 2 ≤ B (2π) − |∇u| ϕ(x)dx exp ds |u|>s t Linear Elliptic Equations and Gauss Measure G. di Blasio Title Page Contents JJ J II I Go Back Close ! Φ−1 (µ (s))2 2 Z µ(t) × (−µ0 (s)) ds + f ∗ (s) ds. 0 Quit Page 13 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au By Gronwall’s Lemma 2.3 we obtain Z d (3.7) − |∇u|2 ϕ(x)dx dt |u|>t " Z µ(t) ≤ exp B (2π) 0 Z = 1 2 Z µ(t) exp r Φ−1 (τ )2 2 ! # dτ f ∗ (r) dr µ(t) exp B(Φ−1 (r) − Φ−1 (µ (t))) f ∗ (r) dr. Linear Elliptic Equations and Gauss Measure 0 Using again (3.6) we have G. di Blasio (3.8) 1 ≤ 2π exp Φ−1 (µ (t))2 (−µ0 (t)) Z µ(t) × exp B(Φ−1 (r) − Φ−1 (µ (t))) f ∗ (r) dr. Title Page Contents 0 Then using (3.8) and integrating between 0 and t, (3.7) becomes Z γn (Ω) t ≤ 2π µ(t) exp Φ−1 (σ)2 Z σ exp B(Φ−1 (s) − Φ−1 (σ)) f ∗ (s) dsdσ, 0 which, putting µ (t) = s and s = Φ (x1 ) , gives (3.9) u? (x) = u? (x1 ) Z x1 2 Z +∞ τ σ2 ? ≤ exp f (σ) exp B (σ − τ ) − dσ dτ, 2 2 λ τ JJ J II I Go Back Close Quit Page 14 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au where λ is such that γn (Ω? ) = γn (Ω) with Ω? = {x = (x1 , x2 , . . . , xn ) ∈ Rn : x1 > λ}. This completes the proof of (3.2) observing that the right-hand side of (3.9) is the solution of (1.3). Let us prove now (3.3). Using the Hölder inequality and (3.7) we have Z d (3.10) − |∇u|q ϕ (x) dx dt |u|>t ! 2q Z µ(t) q 1− ≤ (−µ0 (t)) 2 . exp B(Φ−1 (r) − Φ−1 (µ (t))) f ∗ (r) dr 0 By (3.6) we have (3.11) Linear Elliptic Equations and Gauss Measure G. di Blasio − 2q (−µ0 (t)) h q i q (2π)− 2 exp − Φ−1 (µ (t))2 2 2q Z d 2 ≤ − |∇u| ϕ (x) dx . dt |u|>t Using (3.11), (3.7) and integrating between 0 and +∞, (3.10) becomes q Z Z γn (Ω) Z s ∗ q q −1 −1 |∇u| ϕ (x) dx ≤ (2π) 2 exp B(Φ (r) − Φ (s)) f (r) dr Ω 0 0 h q i 2 −1 × exp Φ (s) ds q Z +∞ Z 2+∞ 2 1 σ ? = (2π) 2 f (σ) exp B (σ − τ ) − dσ 2 λ τ q 2 τ2 × exp τ − dτ, 2 2 Title Page Contents JJ J II I Go Back Close Quit Page 15 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au that is, (3.3). In the next theorem we consider the case c0 (x) 6= 0 and we compare problem (1.1) with a “symmetrized” problem which also takes account of the influence of the term “c (x) u (x)”. Theorem 3.2. Let Ω be an open set of Rn with γn (Ω) < 1 and let u ∈ H01 (ϕ, Ω) be a solution of (1.1) with the assumptions (i), (ii) and (iii). Moreover, let − + + c+ 0 (x) = max {c0 (x), 0} , c0 (x) = max {−c0 (x), 0} , c0? (x) = c0∗ (Φ(x1 )). We will assume that the problem − (wx1 ϕ (x))x1 − Bwx1 ϕ (x) −? ? ? + c+ (3.12) 0? (x1 ) − c0 (x1 ) wϕ(x) = f (x1 )ϕ (x) in Ω , w=0 on ∂Ω? , has a solution w(x) = w? (x1 ). Then u∗ (s) ≤ w∗ (s) holds in [0, s01 ] , where s01 = inf s : c+ (s) > 0 , and 0∗ Z s Z s ∗ u (r) dr ≤ w∗ (r) dr 0 holds in ]s01 , γn (Ω)]. Linear Elliptic Equations and Gauss Measure G. di Blasio Title Page Contents JJ J II I Go Back Close Quit Page 16 of 24 0 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au Proof. Using (iii), the Schwartz inequality and the Hardy inequality we have, Z Z (3.13) c(x) |u| dx ≤ − c0 (x) |u| ϕ(x)dx − |u|>t |u|>t µ(t) Z + ∗ c0∗ (s) − c−∗ 0 (s) u (s)ds, ≤− 0 c+ 0∗ (s) where is the increasing rearrangement of c+ 0 (x) and µ(t) is the distribution function of u(x). Proceeding as in Theorem 3.1 and using (3.13) we obtain Z d (3.14) − |∇u|2 ϕ(x)dx dt |u|>t 21 Z Z ∞ 1 d 2 |∇u| ϕ(x)dx (−µ0 (s)) 2 ds ≤B − ds |u|>s t Z µ(t) Z µ(t) + ∗ −∗ − c0∗ (s) − c0 (s) u (s)ds + f ∗ (s)ds. 0 0 By (3.6) we have (3.15) Z −1 2 1 d 2 ≤ 2π exp Φ (s) − |∇u| ϕ(x)dx . −µ0 (t) dt |u|>t Following the same steps as in Theorem 3.1 and using (3.14), (3.15) becomes Z −1 2 s ∗ 0 (3.16) (−u ) (s) ≤ 2π exp Φ (s) exp B(Φ−1 (r) − Φ−1 (s)) 0 + ∗ × [f ∗ (r) + [c−∗ 0 (r) − c0∗ (r)]u (r)]dr. Linear Elliptic Equations and Gauss Measure G. di Blasio Title Page Contents JJ J II I Go Back Close Quit Page 17 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au Now if we consider the problem (3.12) we can proceed in the same way except that the inequalities should be replaced by equalities. Then if we call w(x) = w? (x) the solution of the problem (3.12), we obtain the following equality (3.17) Z ∗ 0 (−w ) (s) = 2π exp Φ−1 (s)2 s exp B(Φ−1 (r) − Φ−1 (s)) 0 + ∗ × [f ∗ (r) + [c−∗ 0 (r) − c0∗ (r)]w (r)]dr. ∗ ∗ To prove the comparison result, we put v(s) = u (s) − w (s). By (3.16) and (3.17) we have (3.18) 0 (−v) (s) ≤ 2π exp Φ−1 (s)2 Z exp B(Φ−1 (r) − Φ−1 (s)) + × [c−∗ 0 (r) − c0∗ (r)]v (r) dr. + Let us suppose c−∗ 0 (x) , c0∗(x) 6= 0 and and s00 = sup s : c−∗ 0 (s) > 0 . We write c+ 0∗ exp B(Φ−1 (r) − Φ−1 (s)) c−∗ 0 (r) v(r)dr, s∈ [0, s00 ] exp B(Φ−1 (r) − Φ−1 (s)) c+ 0∗ (r) v(r)dr, s ∈ ]s01 , γn (Ω)] . let us put s01 = inf s : (s) > 0 Z0 s V2 (s) = s01 Title Page Contents JJ J II I Go Back s V1 (s) = G. di Blasio s 0 Z Linear Elliptic Equations and Gauss Measure 0 We assume initially that c−∗ 0 (s) is continuous at s0 , we have to prove, now, that V1 (s) ≤ 0. Arguing as in [1] we observe that the existence of a solution w (x) = Close Quit Page 18 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au w? (x) of (3.12), that implies (3.17), guarantees that the problem: 0 −1 2 −1 2 −1 0 −∗ − c (s) Z (s) = (2π) exp Φ (s) Z (s) + 2π exp Φ (s) 0 Rs × 0 exp [B(Φ−1 (r) − Φ−1 (s))] f ∗ (r) dr, Z (0) = Z 0 (s0 ) = 0, 0 has the following positive solution Z s ∗ Z (s) = exp B(Φ−1 (r) − Φ−1 (s)) c−∗ 0 (r) w (r)dr. 0 This allows us to state (see [5]) that the problem 0 −1 0 c−∗ ξ (s) + λ (2π) exp Φ−1 (s)2 ξ (s) = 0, 0 (s) (3.19) ξ (0) = ξ 0 (s00 ) = 0, has the first eigenvalue λ1 > 1, and consequently in the following problem 0 −1 0 c−∗ V1 (s) + 2π exp Φ−1 (s)2 V1 (s) ≥ 0, 0 (s) Linear Elliptic Equations and Gauss Measure G. di Blasio Title Page Contents JJ J II I Go Back Close Quit V1 (0) = V10 (s00 ) = 0, Page 19 of 24 we have V1 (s) ≤ 0 and V10 (s) ≤ 0, s ∈ [0, s00 ] , J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au that is, u∗ (s) ≤ w∗ (s), (3.20) s ∈ [0, s00 ] . On the other hand, from (3.18) and (3.20), we have 0 −1 2 −1 0 c+ (s) V (s) − 2π exp Φ (s) V2 (s) ≥ 0, 0∗ 2 V2 (s01 ) = V20 (γn (Ω)) = 0. Here we can use the maximum principle to obtain V2 (s) ≤ 0. For Lemma 2.2 −1 0 with h(r) = c+ 0∗ (r) , we have for each s ∈ ]s1 , γn (Ω)] , Z Linear Elliptic Equations and Gauss Measure G. di Blasio s (3.21) s01 exp B(Φ−1 (r) − Φ−1 (s)) u∗ (r)dr Z s ≤ exp B(Φ−1 (r) − Φ−1 (s)) w∗ (r)dr, s01 that is, Title Page Contents JJ J II I u∗ (s01 ) ≤ w∗ (s01 ). Go Back On the other hand, from (3.16), (3.17), (3.20) and the definition of v (s) we have Close −v 0 (s) ≤ 0, Now since u ∗ (s01 ) ≤w ∗ (s01 ), Quit s ∈ [s00 , s01 ] . integrating between s and s01 Page 20 of 24 we obtain J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 (3.22) u∗ (s) ≤ w∗ (s), s ∈ [s00 , s01 ] . http://jipam.vu.edu.au From (3.20) and (3.22) we have u∗ (s) ≤ w∗ (s), s ∈ [0, s01 ] . Moreover, for s ∈ s01, γn (Ω) , (3.21) becomes Z s 0≥ exp B(Φ−1 (r) − Φ−1 (s)) [u∗ (r) − w∗ (r)]dr, 0 that is, Z 0 s u∗ (r)dr ≤ Z s w∗ (r)dr, s ∈ [s01 , γn (Ω)] . 0 0 Finally we can remove the hypothesis about the continuity of c−∗ 0 (s) at s0 proceeding by approximations. −∗ 0 0 If c+ 0∗ (x) = 0 or c0 (x) = 0 then s1 = γn (Ω) or s0 = 0 and the result follows with the obvious modifications. Linear Elliptic Equations and Gauss Measure G. di Blasio Title Page Contents JJ J II I Go Back Close Quit Page 21 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au References [1] A. ALVINO, P.L. LIONS AND G. TROMBETTI, Comparison results for elliptic and parabolic equations via Schwarz symmetrization, Ann. Inst. H. Poincaré Anal. Non Linéaire, 7 (1990), 37–65. [2] A. ALVINO AND G. TROMBETTI, Sulle migliori costanti di maggiorazione per una classe di equazioni ellittiche degeneri, Ricerche Mat., 27 (1978), 413–428. [3] A. ALVINO AND G. TROMBETTI, Equazioni ellittiche con termini di ordine inferiore e riordinamenti, Atti Accad. Naz. Lincei Cl. Sci. Fis. Mat. Nater. Rend. Lincei (8), 66 (1979), 194–200. [4] A. ALVINO AND G. TROMBETTI, Su una classe di equazioni ellittiche non lineari degeneri, Ricerche Mat., 29 (1980), 193–212. [5] C. BANDLE, Isoperimetric Inequalities and applications, Pitman London, 1980. [6] M.F. BETTA, F. BROCK, A. MERCALDO AND M.R. POSTERARO, A weighted isoperimetric inquality and application to symmetrization, J. Inequal. Appl., 4 (1999), 215–240. [7] M.F. BETTA, F. BROCK, A. MERCALDO AND M.R. POSTERARO, A comparison related to Gauss measure, C.R. Acad. Sci. Paris, 334, Serie I, (2002), 451–456. [8] C. BORELL, The Brunn-Minkowski inequality in the Gauss space, Invent. Math., 30 (1979), 194–200. Linear Elliptic Equations and Gauss Measure G. di Blasio Title Page Contents JJ J II I Go Back Close Quit Page 22 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au [9] K.M. CHONG AND N.M. RICE, Equimisurable rearrangements of Sobolev functions, J. Reine Angew. Math., 384 (1988), 153–179. [10] A. EHRHARD, Symmétrisation dans l’space de Gauss, Math. scand., 53 (1983), 281–301. [11] A. EHRHARD, Inégalités isoperimétriques et intégrales de Dirichlet Gaussiennes, Ann. Scient. Éc. Norm. Sup., 4e sèrie, 17 (1984), 317–332. [12] A. EHRHARD, Éléments extremaux pour les inégalitès de BrunnMinkowski gaussennes, Ann. Inst. H. Poincarè Anal. non linèaire, C(1986), 149–168. [13] H. FEDERER, Geometric measure theory, Grundlehren der Mathematischen Wissenschaften, 153, Springer-Verlag, 1969. [14] G. MAZ’JA , Sobolev Spaces, Springer-Varlag, Berlin Heidelberg New York Tokyo (1980). [15] M.K.W. MURTY AND G. STAMPACCHIA, Boundary value problems for some degenerate elliptic operators, Ann. Mat. Pura Appl., 80 (1968), 1– 122. [16] J. M. RAKOTOSON AND B. SIMON, Relative rearrangement on a measure space. Application to the regularuty of weighted monotone rearrangement. Part1, Part2, Rev. R. Acad. Ciend. Exact. Fis. Nat. (Esp), 91 (1997), 17–31. [17] G. TALENTI, Elliptic Equations and Rearrangements, Ann. Scuola Norm. Sup. Pisa Cl. Sci., 3 (1976), 697–718. Linear Elliptic Equations and Gauss Measure G. di Blasio Title Page Contents JJ J II I Go Back Close Quit Page 23 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au [18] G. TALENTI, Nonlinear elliptic equation, rearrangements of functions and Orlicz spaces, Ann. Mat. Pura Appl., 120 (1979), 156–184. [19] G. TALENTI, Linear Elliptic P.D.E.’s:Level Set, Rearrangements and a priori Estimates of Solutions, Boll. Un. Mat. Ital. B, 6 4-B (1985), 917– 949. [20] G. TALENTI, A weighted version of a rearrangement inequality, Ann. Univ. Ferrara, Sez. VII (NS), 43 (1997), 121–133. [21] G. TROMBETTI, Metodi di simmetrizzazione nelle equazioni a derivate parziali, Boll. Un. Mat. Ital. B, 3 (2000), 601–634. [22] N.S. TRUDINGER, Linear elliptic operators with measurable coefficients, Ann. Scuola Norm. Sup. Pisa, 27 (1973), 265–308. Linear Elliptic Equations and Gauss Measure G. di Blasio Title Page Contents JJ J II I Go Back Close Quit Page 24 of 24 J. Ineq. Pure and Appl. Math. 4(5) Art. 106, 2003 http://jipam.vu.edu.au