Isoperimetric Inequalities For Eigenvalues of The Laplacian and The Schrödinger Operator

Bull. Math. Sci.
(2012) 2:1–56
DOI 10.1007/s13373-011-0017-0
Isoperimetric inequalities for eigenvalues

of the Laplacian and the Schrödinger operator
Rafael D. Benguria · Helmut Linde ·

Benjamín Loewe
Received: 18 April 2011 / Revised: 18 October 2011 / Accepted: 20 November 2011 /

Published online: 10 January 2012
© The Author(s) 2012. This article is published with open access at SpringerLink.com
Abstract The purpose of this manuscript is to present a series of lecture notes on

isoperimetric inequalities for the Laplacian, for the Schrödinger operator, and related
problems.
Keywords Eigenvalues of the Laplacian · Isoperimetric inequalities
Mathematics Subject Classification (1991) Primary 35P15; Secondary 35J05 ·

49R50
1 Introduction
They came to this place, and bought land, where you now see
the vast walls, and resurgent stronghold, of new Carthage,
as much as they could enclose with the strips of hide
from a single bull, and from that they called it Byrsa.
Book I: “The Trojans reach Carthage”, in The Aeneid.
Communicated by A. Laptev.
R. D. Benguria (B) · B. Loewe

Departamento de Física, P. Universidad Católica de Chile,
Casilla 306, Santiago 22, Chile
e-mail: [email protected]
B. Loewe
H. Linde
Global Management Office, Field Services and Support,
SAP Deutschland AG. & Co. KG, Walldorf, Germany
123
2 R. D. Benguria et al.
Isoperimetric inequalities have played an important role in mathematics since the

times of ancient Greece. The first and best known isoperimetric inequality is the clas-
sical isoperimetric inequality
L2
A≤ ,
4π
relating the area A enclosed by a planar closed curve of perimeter L. In mathematics,

the oldest existing recorded appearance of this problem is in Book V, The Sagacity
of Bees, in the “Collection” of Pappus of Alexandria (IV century AD). The whole
of Book V in Pappus Collection is devoted to isoperimetry. The first section follows
closely the previous exposition of Zenodorus (now lost) as recounted by Theon (see
[101]), where Pappus proposes the two problems: i) Which plane figure of a given
perimeter encloses the largest area? and also ii) Which solid figure with given sur-
face area encloses the largest volume? [59]. The first problem proposed by Pappus is
widely known as Queen Dido’s problem in connection to the foundation of Carthage,
described in Virgilio’s epic poem “The Aeneid” [102]). Previously, Euclid in his Ele-
ments (Book VI) had solved the related simpler problem, of all rectangles of a given
perimeter, the one that encloses the largest area is the square. This problem, solved
geometrically by Euclid, is equivalent to the well known inequality between the geo-
metric and the arithmetic mean. After the introduction of Calculus in the XVII century,
many new isoperimetric inequalities have been discovered in mathematics and physics
(see, e.g., the review articles [20,80,82,87]; see also the essay [16] for a panorama on
the subject of Isoperimetry). The eigenvalues of the Laplacian are “geometric objects”
in the sense they do depend on the geometry of the underlying domain, and to some
extent (see Section 2) the knowledge of the domain characterizes the geometry of the
domain. Therefore it is natural to pose the problem of finding isoperimetric inequali-
ties for the eigenvalues of the Laplacian. The first one to consider this possibility was
Lord Rayleigh in his monograph The Theory of Sound [88]. In these lectures we will
present some of the problems arising in the study of isoperimetric inequalities for the
Laplacian, some of the tools needed in their proof and many bibliographic discussions
about the subject. We start our review with the classical problem of Mark Kac, Can
one hear the shape of a drum. In Section 3 we review the definitions and basic facts
about rearrangements of functions. Section 4 is devoted to the Rayleigh–Faber–Krahn
inequality. In Section 5 we review the Szegö–Weinberger inequality, which is an iso-
perimetric inequality for the first nontrivial Neumann eigenvalue of the Laplacian.
In Section 6 we review the Payne–Pólya–Weinberger isoperimetric inequality for the
quotient of the first two Dirichlet eigenvalues of the Laplacian, as well as several
recent extensions. In Section 7, we review the “Fundamental Gap Formula” between
the lowest two Dirichlet eigenvalues for a bounded, convex domain in R N , which
was recently proved by Andrews and Clutterbuck [1]. In Section 8, we review an
interesting conjectured isoperimetric property for ovals in the plane, which arises in
connection with the Lieb–Thirring inequalities. Finally, in Section 9 we review three
different isoperimetric problems for fourth order operators. There are many interesting
results that we have left out of this review. For different perspectives, selections and
emphasis, please refer for example to the reviews [3–5,15,60], among many others.
123
Isoperimetric inequalities for eigenvalues of the Laplacian 3
The contents of this manuscript are an extended version of the lectures that one of
us (RB) gave as an intensive course for graduate students in the Tunis Science City,
Tunisia (May 21–22, 2010), in connection with the International Conference on the
isoperimetric problem of Queen Dido and its mathematical ramifications, that was
held in Carthage, Tunisia, May 24–29, 2010. We would like to thank all the local
organizers of the conference, especially Najoua Gamara and Lotfi Hermi, for their
great hospitality. Previous versions of this intensive course have been given by RB in
the Arizona School of Analysis with Applications, University of Arizona, Tucson, AZ,
March 15–19, 2010, and previously in the IV Escuela de Verano en Análisis y Física
Matemática at the Unidad Cuernavaca del Instituto de Matemáticas de la Universidad
Nacional Autónoma de México, in the summer of 2005 [25]. Preliminary versions of
these lectures were also given in the Short Course in Isoperimetric Inequalities for
Eigenvalues of the Laplacian, given by one of us (RB) in February of 2004, as part of
the Thematic Program on Partial Differential Equations held at the Fields Institute,
in Toronto, and also as part of the course Autovalores del Laplaciano y Geometría
given at the Department of Mathematics of the Universidad de Pernambuco, in Recife,
Brazil, in August 2003.
2 Can one hear the shape of a drum?
…but it would baffle the most skillful

mathematician to solve the Inverse Problem,
and to find out the shape of a bell by means
of the sounds which is capable of sending out
Sir Arthur Schuster (1882).
In 1965, the Committee on Educational Media of the Mathematical Association of

America produced a film on a mathematical lecture by Mark Kac (1914–1984) with
the title: Can one hear the shape of a drum? One of the purposes of the film was to
inspire undergraduates to follow a career in mathematics. An expanded version of that
lecture was later published [62]. Consider two different smooth, bounded domains, say
1 and 2 in the plane. Let 0 < λ1 < λ2 ≤ λ3 ≤ . . . be the sequence of eigenvalues
of the Laplacian on 1 , with Dirichlet boundary conditions and, correspondingly,
0 < λ1 < λ2 ≤ λ3 ≤ . . . be the sequence of Dirichlet eigenvalues for 2 . Assume
that for each n, λn = λn (i.e., both domains are isospectral). Then, Mark Kac posed
the following question: Are the domains 1 and 2 congruent in the sense of Euclid-
ean geometry?. A friend of Mark Kac, the mathematician Lipman Bers (1914–1993),
paraphrased this question in the famous sentence: Can one hear the shape of a drum?
In 1910, H. A. Lorentz, at the Wolfskehl lecture at the University of Göttingen,
reported on his work with Jeans on the characteristic frequencies of the electromag-
netic field inside a resonant cavity of volume in three dimensions. According to the
work of Jeans and Lorentz, the number of eigenvalues of the electromagnetic cavity
whose numerical values is below λ (this is a function usually denoted by N (λ)) is
given asymptotically by
|| 3/2
N (λ) ≈ λ , (2.1)
6π 2
123
for large values of λ, for many different cavities with simple geometry (e.g., cubes,
spheres, cylinders, etc.) Thus, according to the calculations of Jeans and Lorentz,
to leading order in λ, the counting function N (λ) seemed to depend only on the
volume of the electromagnetic cavity ||. Apparently David Hilbert (1862–1943),
who was attending the lecture, predicted that this conjecture of Lorentz would not be
proved during his lifetime. This time, Hilbert was wrong, since his own student, Her-
mann Weyl (1885–1955) proved the conjecture less than two years after the Lorentz’
lecture.
Remark There is a nice account of the work of Hermann Weyl on the eigenvalues of
a membrane in his 1948 J. W. Gibbs Lecture to the American Mathematical Society
[105].
In N dimensions, (2.1) reads,
||
N (λ) ≈ C N λ N /2 , (2.2)
(2π ) N
where C N = π (N /2) / ((N /2) + 1)) denotes the volume of the unit ball in N
dimensions.
Using Tauberian theorems, one can relate the behavior of the counting function
N (λ) for large values of λ with the behavior of the function
∞

Z (t) ≡ exp{−λn t}, (2.3)
n=1
for small values of t. The function Z (t) is the trace of the heat kernel for the domain
, i.e., Z (t) = tr exp(t). As we mention above, λn () denotes the n Dirichlet
eigenvalue of the domain .
An example: the behavior of Z (t) for rectangles.
With the help of the Riemann Theta function (x), it is simple to compute the trace of
the heat kernel when the domain is a rectangle of sides a and b, and therefore to obtain
the leading asymptotic behavior for small values of t. The Riemann Theta function is
defined by
∞

e−π n x ,
2
(x) = (2.4)
n=−∞
for x > 0. The function (x) satisfies the following modular relation,

1 1
(x) = √ . (2.5)
x x
Remark This modular form for the Theta Function already appears in the classical
paper of Riemann [91] (manuscript where Riemann puts forward his famous Riemann
Hypothesis). In that manuscript, the modular form (2.5) is attributed to Jacobi.
123
The modular form (2.5) may be obtained from a very elegant application of Fou-
rier Analysis (see, e.g., [43], pp. 75–76) which we reproduce here for completeness.
Define
∞

e−π(n+y) x .
2
ϕx (y) = (2.6)
n=−∞
Clearly, (x) = ϕx (0). Moreover, the function ϕx (y) is periodic in y of period 1.

Thus, we can express it as follows,
∞

ϕx (y) = ak e2π ki y , (2.7)
k=−∞
where, the Fourier coefficients are

1
ak = ϕk (y)e−2π ki y dy. (2.8)
0
Replacing the expression (2.6) for ϕx (y) in (2.9), using the fact that e2π ki n = 1, we
can write,
1 ∞

e−π(n+y) x e−2πik(y+n) dy.
2
ak = (2.9)
0 n=−∞
Interchanging the order between the integral and the sum, we get,
∞
1
e−π(n+y) x e−2πik(y+n) dy.
2
ak = (2.10)
n=−∞ 0
In the nth summand we make the change of variables y → u = n + y. Clearly, u runs

from n to n + 1, in the nth summand. Thus, we get,
∞
e−π u x e−2πik u du.
2
ak = (2.11)
−∞
i.e., ak is the Fourier transform of a Gaussian. Thus, we finally obtain,
1
ak = √ e−π k /x .
2
(2.12)
x
Since, (x) = ϕx (0), from (2.7) and (2.12) we finally get,

∞
∞
1 −π k 2 /x 1 1
(x) = ak = √ e =√ . (2.13)
x x x
k=−∞ k=−∞
123
Remarks i) The method exhibited above is a particular case of the Poisson Sum-
mation Formula (see, e.g., [43], pp. 76–77); ii) It should be clear from (2.4) that
lim x→∞ (x) = 1. Hence, from the modular form for (x) we immediately see that
√
lim x (x) = 1. (2.14)
x→0
Once we have the modular form for the Riemann Theta function, it is simple to get
the leading asymptotic behavior of the trace of the heat kernel Z (t), for small values
of t, when the domain is a rectangle. Take to be the rectangle of sides a and b.
Its Dirichlet eigenvalues are given by

n2 m2
λn,m = π 2 + , (2.15)
a2 b2
with n, m = 1, 2, . . .. In terms of the Dirichlet eigenvalues, the trace of the heat kernel,
Z (t) is given by
∞

Z (t) = e−λn,m t . (2.16)
n,m=1
and using (2.15), and the definition of (x), we get,

1 πt πt
Z (t) = θ −1 θ −1 . (2.17)
4 a2 b2
Using the asymptotic behavior of the Theta function for small arguments, i.e., (2.14)
above, we have

1 a b 1 1 1
Z (t) ≈ √ −1 √ −1 ≈ ab− √ (a + b) + + O(t).
4 πt πt 4π t 4 πt 4
(2.18)
In terms of the area of the rectangle A = ab and its perimeter L = 2(a + b), the
expression Z (t) for the rectangle may be written simply as,
1 1 1
Z (t) ≈ A − √ L + + O(t). (2.19)
4π t 8 πt 4
Remark Using similar techniques, one can compute the small t behavior of Z (t) for
various simple regions of the plane (see, e.g., [73]).
The fact that the leading behavior of Z (t) for t small, for any bounded, smooth
domain in the plane is given by
1
Z (t) ≈ A (2.20)
4π t
123
was proven by Hermann Weyl [104]. Here, A = || denotes the area of . In fact,
what Weyl proved in [104] is the Weyl Asymptotics of the Dirichlet eigenvalues, i.e.,
for large n, λn ≈ (4π n)/A. Weyl’s result (2.20) implies that one can hear the area of
the drum.
In 1954, the Swedish mathematician Åke Pleijel [86] obtained the improved asymp-
totic formula,
A L
Z (t) ≈ − √ ,
4π t 8 πt
where L is the perimeter of . In other words, one can hear the area and the perimeter
of . It follows from Pleijel’s asymptotic result that if all the frequencies of a drum
are equal to those of a circular drum then the drum must itself be circular. This follows
from the classical isoperimetric inequality (i.e., L 2 ≥ 4π A, with equality if and only
if is a circle). In other words, one can hear whether a drum is circular. It turns out
that it is enough to hear the first two eigenfrequencies to determine whether the drum
has the circular shape [9].
In 1966, Mark Kac obtained the next term in the asymptotic behavior of Z (t) [62].
For a smooth, bounded, multiply connected domain on the plane (with r holes)
A L 1
Z (t) ≈ − √ + (1 − r ). (2.21)
4π t 8 πt 6
Thus, one can hear the connectivity of a drum. The last term in the above asymptotic
expansion changes for domains with corners (e.g., for a rectangular membrane, using
the modular formula for the Theta Function, we obtained 1/4 instead of 1/6). Kac’s
formula (2.21) was rigorously justified by McKean and Singer [74]. Moreover, for
domains having corners they showed that each corner with interior angle γ makes an
additional contribution to the constant term in (2.21) of (π 2 − γ 2 )/(24π γ ). (Notice
that for a rectangle, i.e., when all the corners have γ = π/2, this factor is 1/16, which
multiplied by 4, i.e., by the number of corners, yields the third term in (2.19)).
A sketch of Kac’s analysis for the first term of the asymptotic expansion is as follows
(here we follow [62,73]). If we imagine some substance concentrated at ρ = (x0 , y0 )
diffusing through the domain bounded by ∂, where the substance is absorbed at
the boundary, then the concentration P ( p r ; t) of matter at r = (x, y) at time t
obeys the diffusion equation
∂ P
= P
∂t
with boundary condition P ( p r ; t) → 0 as r → a, a ∈ ∂, and initial condition
P ( p r ; t) → δ(r − p) as t → 0, where δ(r − p) is the Dirac delta function. The
concentration P ( p r ; t) may be expressed in terms of the Dirichlet eigenvalues of
, λn and the corresponding (normalized) eigenfunctions φn as follows,
∞

P ( p r ; t) = e−λn t φn ( p)φn (r ).
n=1
123
For small t, the diffusion is slow, that is, it will not feel the influence of the boundary
in such a short time. We may expect that
P ( p r ; t) ≈ P0 ( p r ; t),
ar t → 0, where ∂ P0 /∂t = P0 , and P0 ( p r ; t) → δ(r − p) as t → 0.P0 in fact

represents the heat kernel for the whole R2 , i.e., no boundaries present. This kernel is
explicitly known. In fact,
1
P0 ( p r ; t) = exp(−|r − p|2 /4t),
4π t
where |r − p|2 is just the Euclidean distance between p and r . Then, as t → 0+ ,

∞
1
P ( p r ; t) = e−λn t φn ( p)φn (r ) ≈ exp(−|r − p|2 /4t).
4π t
n=1
Thus, when set p = r we get

∞
1
e−λn t φn2 (r ) ≈ .
4π t
n=1
Integrating both sides with respect to r , using the fact that φn is normalized, we finally
get,
∞
||
e−λn t ≈ , (2.22)
4π t
n=1
which is the first term in the expansion (2.21). Further analysis gives the remaining
terms (see [62]).
Remark In 1951, Mark Kac proved [61] the following universal bound on Z (t) in
dimension d:
||
Z (t) ≤ . (2.23)
(4π t)d/2
This bound is sharp, in the sense that as t → 0,
||
Z (t) ≈ . (2.24)
(4π t)d/2
Recently, Harrell and Hermi [58] proved the following improvement on (2.24),
||
Z (t) ≈ e−Md ||t/I () . (2.25)
(4π t)d/2
123
where I = mina∈Rd |x − a|2 d x and Md is a constant depending on dimension.

Moreover, they conjectured the following bound on Z (t), namely,
||
e−t/|| .
2/d
Z (t) ≈ (2.26)
(4π t)d/2
Recently, Geisinger and Weidl [52] proved the best bound up to date in this direction,
||
e−M d t/|| ,
2/d
Z (t) ≈ (2.27)
(4π t) d/2
where M d = [(d + 2)π/d](d/2 + 1)−2/d Md (in particular M 2 = π/16. In gen-

eral M d < 1, thus the Geisinger–Weidl bound (2.27) falls short of the conjectured
expression of Harrell and Hermi.
2.1 One cannot hear the shape of a drum
In the quoted paper of Mark Kac [62] he says that he personally believed that one cannot
hear the shape of a drum. A couple of years before Mark Kac’ article, John Milnor [77],
had constructed two non-congruent sixteen dimensional tori whose Laplace–Beltrami
operators have exactly the same eigenvalues. In 1985 Toshikazu Sunada [94], then at
Nagoya University in Japan, developed an algebraic framework that provided a new,
systematic approach of considering Mark Kac’s question. Using Sunada’s technique
several mathematicians constructed isospectral manifolds (e.g., Gordon and Wilson;
Brooks; Buser, etc.). See, e.g., the review article of Robert Brooks (1988) with the
situation on isospectrality up to that date in [36]. Finally, in 1992, Carolyn Gordon,
David Webb and Scott Wolpert [54] gave the definite negative answer to Mark Kac’s
question and constructed two plane domains (henceforth called the GWW domains)
with the same Dirichlet eigenvalues.
2.2 Proof of isospectrality using transplantation
The most elementary proof of isospectrality of the GWW domains is done using
the method of transplantation. For the method of transplantation see, e.g., [27,28].
See also the expository article [29] by the same author. The method also appears
briefly described in the article of Sridhar and Kudrolli cited in the Bibliographical
Remarks, (iii) at the end of this chapter.
To conclude this chapter we will give the details of the proof of isospectrality of the
GWW domains using transplantation (for the interested reader, there is a recent review
article [53] that presents many different proofs of isospectrality, including transplan-
tation, and paper folding techniques). For that purpose label from 1 to 7 the congruent
triangles that make the two GWW domains (see Figure 1). Each of this isosceles right
triangles has two cathets, labeled A and B and the hypothenuse, labeled T . Each of
the pieces (triangles) that make each one of the two domains is connected to one or
more neighboring triangles through a side A, a side B or a side T . Each of the two
isospectral domains has an associated graph, which are given in Figure 2.
123
Fig. 1 GWW Isospectral

Domains D1 and D2
Fig. 2 Sunada Graphs

corresponding to Domains D1
and D2
These graphs have their origin in the algebraic formulation of Sunada [94]. The
vertices in each graph are labeled according to the number that each of the pieces
(triangles) has in each of the given domains. As for the edges joining two vertices
in these graphs, they are labeled by either an A, a B or a T depending on the type
of the common side of two neighboring triangles in Figure 1. In order to show that
both domains are isospectral it is convenient to consider any function defined on each
domain as consisting of seven parts, each part being the restriction of the original func-
tion to each one of the individual triangles that make the domain. In this way, if ψ is a
function defined on the domain 1, we will write it as a vector with seven components,
i.e., ψ = [ψi ]i=1
7 , where ψ is a scalar function whose support is triangle i on the
i
domain 1. Similarly, a function ϕ defined over the domain 2 may be represented as a
seven component vector ϕ = [ϕi ]i=1 7 , with the equivalent meaning but referred to the
second domain.
In order to show the isospectrality of the two domains we have to exhibit a mapping
transforming the functions defined on the first domain into functions defined in the
second domain. Given the decomposition we have made of the eigenfunctions as vec-
tors of seven components, this transformation will be represented by a 7 × 7 matrix.
In order to show that the two domains have the same spectra we need this matrix to
be orthogonal. This matrix is given explicitly by
⎛ ⎞
−a a a −a b −b b
⎜ a −b −a b −a a −b ⎟
⎜ ⎟
⎜ a −a −b b −b a −a ⎟
⎜ ⎟
TD = ⎜
⎜ −a b b −a a −b a ⎟⎟. (2.28)
⎜ −b a b −a a −a b ⎟
⎜ ⎟
⎝ b −a −a b −a b −a ⎠
−b b a −a b −a a
123
For the matrix TD to be orthogonal, we need that the parameters a and b satisfy the
following relations: 4a 2 + 3b2 = 1, 2a 2 + 4ab + b2 = 0, and 4a + 3b = 1. Although
we do not need the numerical values of a and b in the sequel, it is good √ to know
that there√is a solution to this system of equations, namely a = (1 − 3 8/4)/7 and
b = (1 + 8)/7. The matrix TD is orthogonal, i.e., TD TDt = 1. The label D used here
refers to the fact that this matrix TD is used to show isospectrality for the Dirichlet
problem. A similar matrix can be constructed to show isospectrality for the Neumann
problem. In order to show isospectrality it is not sufficient to show that the matrix
TD is orthogonal. It must fulfill two additional properties. On the one hand it should
transform a function ψ that satisfies the Dirichlet conditions in the first domain in a
function ϕ that satisfies Dirichlet boundary conditions on the second domain. More-
over, by elliptic regularity, since the functions ψ and its image ϕ satisfy the eigenvalue
equation −u = λu, they must be smooth (in fact they should be real analytic in the
interior of the corresponding domains), and therefore they must be continuous at the
adjacent edges connecting two neighboring triangles. Thus, while the function ψ is
continuous when crossing the common edges of neighboring triangles in the domain
1, the function ϕ should be continuous when crossing the common edges of neigh-
boring triangles in the domain 2. These two properties are responsible for the peculiar
structure of a’s and b’s in the components of the matrix TD .
To illustrate these facts, if the function ϕ is an eigenfunction of the Dirichlet problem
for the domain 2, it must satisfy, among others, the following properties,
ϕ2A = ϕ7A , (2.29)
and
ϕ5T = 0. (2.30)
Here ϕ2 denotes the second component of ϕ, i.e., the restriction of the function ϕ to
the second triangle in Domain 2 (see Figure 2). On the other hand, ϕ2A denotes the
restriction of ϕ2 on the edge A of triangle 2. Since on the domain 2, the triangles 2
and 7 are glued through a cathet of type A, (2.29) is precisely the condition that ϕ
has to be smooth in the interior of 2. On the other hand, ϕ must be a solution of the
Dirichlet problem for the domain 2 and as such it must satisfy zero boundary condi-
tions. Since the hypothenuse T of triangle 5 is part of the boundary of the domain 2, ϕ
must vanish there. This is precisely the condition (2.30). Let us check, as an exercise
that if ψ is smooth and satisfies Dirichlet boundary conditions in the domain 1, its
image ϕ = TD ψ satisfies (2.30) over the domain 2. We let as an exercise to the reader
to check (2.29), and all the other conditions on “smoothness” and boundary condition
of ϕ (this is a long but straightforward task). From (2.28) we have that
ϕ5T = −bψ1T + aψ2T + bψ3T − aψ4T + aψ5T − aψ6T + bψ7T . (2.31)
Since all the sides of type T of the pieces 1, 3 and 7 in the domain 1 are part of the
boundary of the domain (see Figure 1), ψ1T = ψ3T = ψ7T = 0. On the other hand,
since 2 and 4 are neighboring triangles in the domain 1, glued through a side of type
123
T, we have ψ2T = ψ4T . By the same reasoning we have ψ5T = ψ6T . Using these three
conditions on (2.31) we obtain (2.30). All the other conditions can be verified in a
similar way. Collecting all these facts, we conclude with
Theorem 2.1 (P. Bérard) The transformation TD given by (2.28) is an isometry from
L 2 (D1 ) into L 2 (D2 ) (here D1 and D2 are the two domains of Figures 1 and 2), which
induces an isometry from H01 (D1 ) into H01 (D2 ), and therefore we have,
Theorem 2.2 (C. Gordon, D. Webb, S. Wolpert) The domains D1 and D2 of Figures 1
and 2 are isospectral.
Although the proof by transplantation is straightforward to follow, it does not shed

light on the rich geometric, analytic and algebraic structure of the problem initiated by
Mark Kac. For the interested reader it is recommendable to read the papers of Sunada
[94] and of Gordon, Webb and Wolpert [54].
In the previous paragraphs we have seen how the answer to the original Kac’s ques-
tion is in general negative. However, if we are willing to require some analyticity of
the domains, and certain symmetries, we can recover uniqueness of the domain once
we know the spectrum. During the last decade there has been an important progress
in this direction. In 2000, S. Zelditch, [108] proved that in two dimensions, simply
connected domains with the symmetry of an ellipse are completely determined by
either their Dirichlet or their Neumann spectrum. More recently [109], he proved a
much stronger positive result. Consider the class of planar domains with no holes and
very smooth boundary and with at least one mirror symmetry. Then one can recover
the shape of the domain given the Dirichlet spectrum.
2.3 Bibliographical remarks
i) The sentence of Arthur Schuster (1851–1934) quoted at the beginning of this chap-
ter is cited in Reed and Simon’s book, volume IV [90]. It is taken from the article
A. Schuster, The Genesis of Spectra, in Report of the fifty–second meeting of the Brit-
ish Association for the Advancement of Science (held at Southampton in August 1882),
Brit. Assoc. Rept., pp. 120–121, 1883. Arthur Schuster was a British physicist (he was
a leader spectroscopist at the turn of the XIX century). It is interesting to point out that
Arthur Schuster found the solution to the Lane–Emden equation with exponent 5, i.e.,
to the equation,
−u = u 5 ,
in R3 , with u > 0 going to zero at infinity. The solution is given by
31/4
u= .
(1 + |x|2 )1/2
(A. Schuster, On the internal constitution of the Sun, Brit. Assoc. Rept. pp. 427–429,
1883). Since the Lane–Emden equation for exponent 5 is the Euler–Lagrange equation
123
for the minimizer of the Sobolev quotient, this is precisely the function that, modulo
translations and dilations, gives the best Sobolev constant. For a nice autobiography of
Arthur Schuster see A. Schuster, Biographical fragments, Mc Millan & Co., London,
(1932).
ii) A very nice short biography of Marc Kac was written by H. P. McKean [Mark
Kac in Bibliographical Memoirs, National Academy of Science, 59, 214–235 (1990);
available on the web (page by page) at http://www.nap.edu/books/0309041988/html/
214.html]. The reader may want to read his own autobiography: Mark Kac, Enigmas
of Chance, Harper and Row, NY, 1985 [reprinted in 1987 in paperback by The Univer-
sity of California Press]. For his article in the American Mathematical Monthly, op.
cit., Mark Kac obtained the 1968 Chauvenet Prize of the Mathematical Association
of America.
iii) For a beautiful account of the scientific contributions of Lipman Bers (1993–1914),
who coined the famous phrase, Can one hear the shape of a drum?, see the article by
Cathleen Morawetz and others, Remembering Lipman Bers, Notices of the AMS 42,
8–25 (1995).
iv) It is interesting to remark that the values of the first Dirichlet eigenvalues of the
GWW domains were obtained experimentally by S. Sridhar and A. Kudrolli [93],
Experiments on Not “Hearing the Shape” of Drums, Physical Review Letters, 72,
2175–2178 (1994). In that article one can find the details of the physics experiments
performed by these authors using very thin electromagnetic resonant cavities with
the shape of the Gordon–Webb–Wolpert (GWW) domains. This is the first time that
the approximate numerical values of the first 25 eigenvalues of the two GWW were
obtained. The corresponding eigenfunctions are also displayed. A quick reference to
the transplantation method of Pierre Berard is also given in this article, including the
transplantation matrix connecting the two isospectral domains. The reader may want
to check the web page of S. Sridhar’s Lab (http://sagar.physics.neu.edu/) for further
experiments on resonating cavities, their eigenvalues and eigenfunctions, as well as
on experiments on quantum chaos.
v) The numerical computation of the eigenvalues and eigenfunctions of the pair of
GWW isospectral domains was obtained by Tobin A. Driscoll, Eigenmodes of iso-
spectral domains, SIAM Review 39, 1–17 (1997).
vi) In its simplified form, the Gordon–Webb–Wolpert domains (GWW domains) are
made of seven congruent rectangle isosceles triangles. Certainly the GWW domains
have the same area, perimeter and connectivity. The GWW domains are not convex.
Hence, one may still ask the question whether one can hear the shape of a convex
drum. There are examples of convex isospectral domains in higher dimension (see
e.g. C. Gordon and D. Webb, Isospectral convex domains in Euclidean Spaces, Math.
Res. Letts. 1, 539–545 (1994), where they construct convex isospectral domains in
Rn , n ≥ 4). Remark: For an update of the Sunada Method, and its applications see
the article of Robert Brooks [The Sunada Method, in Tel Aviv Topology Conference
“Rothenberg Festschrift” 1998, Contemporary Mathematics 231, 25–35 (1999); elec-
tronically available at: http://www.math.technion.ac.il/~rbrooks].
vii) There is a vast literature on Kac’s question, and many review lectures on it. In
particular, this problem belongs to a very important branch of mathematics: Inverse
123
Problems. In that connection, see the lectures of R. Melrose [76]. For a very recent
review on Kac’s question and its many ramifications in physics, see [53].
3 Rearrangements
3.1 Definition and basic properties
For many problems of functional analysis it is useful to replace some function by an

equimeasurable but more symmetric one. This method, which was first introduced
by Hardy and Littlewood, is called rearrangement or Schwarz symmetrization [55].
Among several other applications, it plays an important role in the proofs of isoperi-
metric inequalities like the Rayleigh–Faber–Krahn inequality, the Szegö–Weinberger
inequality or the Payne–Pólya–Weinberger inequality (see Sections 4, 5 and 6 below).
In the following we present some basic definitions and theorems concerning spheri-
cally symmetric rearrangements.
We let be a measurable subset of Rn and write || for its Lebesgue measure,
which may be finite or infinite. If it is finite we write for an open ball with the
same measure as , otherwise we set = Rn . We consider a measurable func-
tion u : → R and assume either that || is finite or that u decays at infinity, i.e.,
|{x ∈ : |u(x)| > t}| is finite for every t > 0.
Definition 3.1 The function
μ(t) = |{x ∈ : |u(x)| > t}|, t ≥ 0
is called distribution function of u.
From this definition it is straightforward to check that μ(t) is a decreasing (non-

increasing), right-continuous function on R+ with μ(0) = |sprt u| and sprt μ =
[0, ess sup |u|).
Definition 3.2
• The decreasing rearrangement u : R+ → R+ of u is the distribution function of
μ.
• The symmetric decreasing rearrangement u : → R+ of u is defined by u (x) =
u (Cn |x|n ), where Cn = π n/2 [(n/2 + 1)]−1 is the measure of the n-dimensional
unit ball.
Because μ is a decreasing function, Definition 3.2 implies that u is an essentially

inverse function of μ. The names for u and u are justified by the following two
lemmas:
Lemma 3.3
(a) The function u is decreasing, u (0) = esssup |u| and sprt u = [0, |sprt u|)
(b) u (s) = min {t ≥ 0 : μ(t) ≤ s}
∞
(c) u (s) = 0 χ[0,μ(t)) (s) dt
123
(d) |{s ≥ 0 : u (s) > t}| = |{x ∈ : |u(x)| > t}| for all t ≥ 0.
(e) {s ≥ 0 : u (s) > t} = [0, μ(t)) for all t ≥ 0.
Proof Part (a) is a direct consequence of the definition of u , keeping in mind the
general properties of distribution functions stated above. The representation formula
in part (b) follows from
u (s) = |{w ≥ 0 : μ(w) > s}| = sup{w ≥ 0 : μ(w) > s} = min{w ≥ 0 : μ(w) ≤ s},
where we have used the definition of u in the first step and then the monotonicity
and right-continuity of μ. Part (c) is a consequence of the ‘layer-cake formula’, see
Theorem 10.1 in the appendix. To prove part (d) we need to show that
{s ≥ 0 : u (s) > t} = [0, μ(t)). (3.1)
Indeed, if s is an element of the left hand side of (3.1), then by Lemma 3.3, part (b),
we have
min{w ≥ 0 : μ(w) ≤ s} > t.
But this means that μ(t) > s, i.e., s ∈ [0, μ(t)). On the other hand, if s is an element
of the right hand side of (3.1), then s < μ(t) which implies again by part (b) that
u (s) = min{w ≥ 0 : μ(w) ≤ s} ≥ min{w ≥ 0 : μ(w) < μ(t)} > t,
i.e., s is also an element of the left hand side. Finally, part (e) is a direct consequence
of part (d).
It is straightforward to transfer the statements of Lemma 3.3 to the symmetric

decreasing rearrangement:
Lemma 3.4
(a) The function u is spherically symmetric and radially decreasing.
(b) The measure of the level set {x ∈ : u (x) > t} is the same as the measure of
{x ∈ : |u(x)| > t} for any t ≥ 0 (Figure 3).
From Lemma 3.3 (c) and Lemma 3.4 (b) we see that the three functions u, u and u
have the same distribution function and therefore they are said to be equimeasurable.
Quite analogous to the decreasing rearrangements one can also define increasing ones:
Definition 3.5
• If the measure of is finite, we call u (s) = u (|| − s) the increasing rearrange-
ment of u.
• The symmetric increasing rearrangement u : → R+ of u is defined by u (x) =
u (Cn |x|n ).
123
Fig. 3 The level sets of the functions u (left) and of its rearrangement u ∗ (right)
In his lecture notes on rearrangements (see the reference in the Bibliographical

Remarks, i) at the end of this chapter), G. Talenti, gives the following example, illus-
trating the meaning of the distribution and the rearrangement of a function: Consider
the function u(x) ≡ 8 + 2x 2 − x 4 , defined on the interval −2 ≤ x ≤ 2. Then, it is a
simple exercise to check that the corresponding distribution function μ(t) is given by
√
21 + 9 − t if 0 ≤ t ≤ 8,
μ(t) = √
2 2 − 2 t − 8 if 8 < t ≤ 9.
Hence,
√
9 − x 2 + x 4 /4 if x ≤ √ 2,
u (x) =
u(x) if |x| > 2.
This function can as well be used to illustrate the theorems below.
3.2 Main theorems
Rearrangements are a useful tool of functional analysis because they considerably

simplify a function without changing certain properties or at least changing them in
a controllable way. The simplest example is the fact that the integral of a function’s
absolute value is invariant under rearrangement. A bit more generally, we have:
Theorem 3.6 Let be a continuous increasing map from R+ to R+ with (0) = 0.

Then

(u (x)) dx = (|u(x)|) dx = (u (x)) dx.

Proof The theorem follows directly from Theorem 10.1 in the appendix: If we choose
m( dx) = dx, the right hand side of (10.1) takes the same value for v = |u|, v = u
and v = u . The conditions on are necessary since (t) = ν([0, t)) must hold for
some measure ν on R+ .
123
For later reference we state a rather specialized theorem, which is an estimate on the
rearrangement of a spherically symmetric function that is defined on an asymmetric
domain:
Theorem 3.7 Assume that u : → R+ is given by u (x) = u(|x|), where u :

R+ → R+ is a non-negative decreasing (resp. increasing) function. Then u (x) ≤
u(|x|) (resp. u (x) ≥ u(|x|)) for every x ∈ .
Proof Assume first that u is a decreasing function. The layer–cake representation for
u is
∞
u (x) = u (Cn |x|n ) = χ[0,|{x∈:u (x)>t}|) (Cn |x|n ) dt
0 ∞
≤ χ[0,|{x∈Rn :u(|x|)>t}|) (Cn |x|n ) dt
0
∞
= χ{x∈Rn :u(|x|)>t} (x) dt
0
= u(|x|).
The product of two functions changes in a controllable way under rearrangement:
Theorem 3.8 Suppose that u and v are measurable and non-negative functions
defined on some ⊂ Rn with finite measure. Then

u (s) v (s) ds ≥ u(x) v(x) dx ≥ u (s) v (s) ds (3.2)
R+ R+
and

u (x) v (x) dx ≥ u(x) v(x) dx ≥ u (x) v (x) dx. (3.3)

Proof We first show that for every measurable ⊂ and every measurable v :
→ R+ the relation
| | | |
v (s) ds ≥ v dx ≥ v (s) ds (3.4)
0 0
holds: We can assume without loss of generality that v is integrable. Then the layer-
cake formula (see Theorem 10.1 in the appendix) gives
∞ ∞

v= χ{x∈:v(x)>t} dt and v = χ[0,μ(t)) dt. (3.5)
0 0
123
Hence,
∞
v dx = | ∩ {x ∈ : v(x) > t}| dt,
0
| | ∞
v (s) ds = min(| |, |{x ∈ : v(x) > t}|) dt.
0 0
The first inequality in (3.4) follows. The second inequality in (3.4) can be established
with the help of the first:
| | || || ||−| |
v ds = v ds − v ds = v dx − v ds
0 0 | | 0

≤ v dx − v dx = v dx.
\
Now assume that u and v are measurable, non-negative and—without loosing

generality—integrable. Since we can replace v by u in the equations (3.5), we have
∞
u(x)v(x) dx = dt v(x) dx,
0
{x∈:u(x)>t}
∞ ∞ μ(t)

u (s)v (s) ds = dt v (s) ds,
0 0 0
where μ is the distribution function of u. On the other hand, the first inequality in (3.4)
tells us that
μ(t)
v(x) dx ≤ v (s) ds
0
{x∈:u(x)>t}
for every non-negative t, such that the first inequality in (3.2) follows. The second part
of (3.2) can be proven completely analogously, and the inequalities (3.3) are a direct
consequence of (3.2).
3.3 Gradient estimates
The integral of a function’s gradient over the boundary of a level set can be estimated
in terms of the distribution function:
Theorem 3.9 Assume that u : Rn → R is Lipschitz continuous and decays at infinity,
i.e., the measure of t := {x ∈ Rn : |u(x)| > t} is finite for every positive t. If μ is
the distribution function of u then

2/n μ(t)
2−2/n
|∇u|Hn−1 ( dx) ≥ −n 2 Cn . (3.6)
∂t μ (t)
123
Remarks i) Here Hn (A) denotes the n-dimensional Hausdorff measure of the set A
(see, e.g., [51]); ii) it is worth to point out that (3.6) is an equality for radial functions.
Proof On the one hand, by the classical isoperimetric inequality we have

1/n 1/n
Hn−1 ( dx) ≥ nCn |t |1−1/n = nCn μ(t)1−1/n . (3.7)
∂t
On the other hand, we can use the Cauchy–Schwarz inequality to get

√
|∇u|
Hn−1 ( dx) = √ Hn−1 ( dx)
∂t ∂t |∇u|
1/2 1/2
1
≤ |∇u|Hn−1 ( dx) Hn−1 ( dx) .
∂t ∂t |∇u|
The last integral in the above formula can be replaced by −μ (t) according to Federer’s
coarea formula (see, [51]). The result is
1/2
1/2
Hn−1 ( dx) ≤ |∇u|Hn−1 ( dx) −μ (t) . (3.8)
∂t ∂t
Comparing the equations (3.7) and (3.8) yields Theorem 3.9.
Integrals that involve the norm of the gradient can be estimated using the following
important theorem:
Theorem 3.10 Let : R+ → R+ be a Young function, i.e., is increasing and

convex with (0) = 0. Suppose that u : Rn → R is Lipschitz continuous and decays
at infinity. Then

(|∇u (x)|) dx ≤ (|∇u(x)|) dx.
Rn Rn
For the special case (t) = t 2 Theorem 3.10 states that the ‘energy expectation
value’ of a function decreases under symmetric rearrangement, a fact that is key to the
proof of the Rayleigh–Faber–Krahn inequality (see Section 4.1).
Proof Theorem 3.10 is a consequence of the following chain of (in)equalities, the

second step of which follows from Lemma 3.11 below.
∞
d
(|∇u|) dx = ds (|∇u|) dx
Rn 0 ds
{x∈Rn :|u(x)|>u ∗ (s)}
∞
du ∗
(|∇u |) dx.
1/n
≥ ds −nCn s 1−1/n (s) =
0 ds Rn
123
Lemma 3.11 Let u and be as in Theorem 3.10. Then for almost every positive s
holds

d 1/n du ∗
(|∇u|) dx ≥ −nCn s 1−1/n (s) . (3.9)
ds ds
{x∈Rn :|u(x)|>u ∗ (s)}
Proof First we prove Lemma 3.11 for the special case of being the identity. If
s > |sprt u| then (3.9) is clearly true since both sides vanish. Thus we can assume that
0 < s < |sprt u|. For all 0 ≤ a < b < |sprt u| we show that

|∇u(x)| dx ≥ nCn a 1−1/n (u ∗ (a) − u ∗ (b)).
1/n
(3.10)
{x∈Rn :u ∗ (a)>|u(x)|>u ∗ (b)}
The statement (3.10) is proven by the following chain of inequalities, in which we

first use Federer’s coarea formula, then the classical isoperimetric inequality in Rn
and finally the monotonicity of the integrand:
u ∗ (a)
l.h.s. of (3.10) = Hn−1 {x ∈ Rn : |u(x)| = t} dt
u ∗ (b)
u ∗ (a)
1/n
≥ nCn |{x ∈ Rn : |u(x)| ≥ t}|1−1/n dt
u ∗ (b)
≥ nCn |{x ∈ Rn : |u(x)| ≥ u ∗ (a)}|1−1/n · (u ∗ (a) − u ∗ (b))

1/n
≥ r.h.s. of (3.10).
In the case of being the identity, Lemma 3.11 follows from (3.10): Replace b by
a + with some > 0, multiply both sides by −1 and then let go to zero.
It remains to show that equation (3.9) holds for almost every s > 0 if is not the
identity but some general Young function. From the monotonicity of u ∗ follows that
∗
for almost every s > 0 either du ∗
ds is zero or there is a neighborhood of s where u is
continuous and decreases strictly. In the first case there is nothing to prove, thus we
can assume the second one. Then we have
|{x ∈ Rn : u ∗ (s) ≥ |u(x)| > u ∗ (s + )}| = (3.11)
for small enough > 0. Consequently, we can apply Jensen’s inequality to get

1
(|∇u(x)|) dx

{x∈Rn :u ∗ (s)≥|u(x)|>u∗(s+h)}
⎛ ⎞

⎜1 ⎟
≥ ⎝ |∇u(x)| dx ⎠ .

{x∈Rn :u ∗ (s)≥|u(x)|>u∗(s+h)}
123
Taking the limit ↓ 0, this yields

⎛ ⎞

d ⎜ d ⎟
(|∇u(x)|) dx ≥ ⎝ |∇u(x)| dx ⎠ .
ds ds
{x∈Rn :|u(x)|>u∗(s)} {x∈Rn :|u(x)|>u∗(s)}
Since we have already proven Lemma 3.11 for the case of being the identity, we
can apply it to the argument of on the right hand side of the above inequality. The
statement of Lemma 3.11 for general follows.
i) Rearrangements of functions were introduced by G. Hardy and J. E. Littlewood.

Their results are contained in the classical book of G.H. Hardy, J. E. Littlewood,
and G. Pólya [55]. The fact that the L 2 norm of the gradient of a function decreases
under rearrangements was proven by Faber and Krahn (see references below). A more
modern proof as well as many results on rearrangements and their applications to
PDE’s can be found in [100]. See also Chapter 3, pp. 79-ff in reference [69]. The
reader may want to see also the article by E.H. Lieb, Existence and uniqueness of the
minimizing solution of Choquard’s nonlinear equation, Studies in Appl. Math. 57,
93–105 (1976/77), for an alternative proof of the fact that the L 2 norm of the gradient
decreases under rearrangements using heat kernel techniques. An excellent expository
review on rearrangements of functions (with a good bibliography) can be found in G.
Talenti, Inequalities in rearrangement invariant function spaces, in Nonlinear anal-
ysis, function spaces and applications, Vol. 5 (Prague, 1994), 177–230, Prometheus,
Prague, 1994. (available at the website: http://www.emis.de/proceedings/Praha94/).
The Riesz rearrangement inequality is the assertion that for nonnegative measurable
functions f, g, h in Rn , we have

f (y)g(x − y)h(x) d x d y ≤ f (y)g (x − y)h (x) d x d y.
R n ×R n R n ×R n
For n = 1 the inequality is due to F. Riesz, Sur une inégalité intégrale, Journal of
the London Mathematical Society 5, 162–168 (1930). For general n is due to S.L.
Sobolev, On a theorem of functional analysis, Mat. Sb. (NS) 4, 471–497 (1938) [the
English translation appears in AMS Translations (2) 34, 39–68 (1963)]. The cases
of equality in the Riesz inequality were studied by A. Burchard, Cases of equality
in the Riesz rearrangement inequality, Annals of Mathematics 143 499–627 (1996)
(this paper also has an interesting history of the problem). In addition to Burchard’s
thorough analysis, there is a strictness statement in Lieb’s paper on the Choquard’s
nonlinear equation, cited above, which is useful in applications.
ii) Rearrangements of functions have been extensively used to prove symmetry prop-
erties of positive solutions of nonlinear PDE’s. See, e.g., B. Kawohl, Rearrangements
and convexity of level sets in Partial Differential Equations, Lecture Notes in Mathe-
matics, 1150, Springer-Verlag, Berlin (1985), and references therein.
123
iii) There are different types of rearrangements of functions. See, e.g., the review arti-
cle of Al Baernstein [19]. For an interesting approach to rearrangements see also [35].
This approach goes back through Baernstein–Taylor (A. Baernstein and B. A. Taylor,
Spherical rearrangements, subharmonic functions and ∗–functions in n–space, Duke
Math. J. 43, 245–268 (1976)), who cite Ahlfors (L. V. Ahlfors, Conformal Invariants,
McGraw Hill, NY, 1973), who in turn credits Hardy and Littlewood.
4 The Rayleigh–Faber–Krahn inequality
4.1 The Euclidean case
Many isoperimetric inequalities have been inspired by the question which geometrical
layout of some physical system maximizes or minimizes a certain quantity. One may
ask, for example, how matter of a given mass density must be distributed to minimize
its gravitational energy, or which shape a conducting object must have to maximize
its electrostatic capacity. The most famous question of this kind was put forward at
the end of the 19th century by Lord Rayleigh in his work on the theory of sound
[88]: he conjectured that among all drums of the same area and the same tension the
circular drum produces the lowest fundamental frequency. This statement was proven
independently in the 1920s by Faber [50] and Krahn [64,65].
To treat the problem mathematically, we consider an open bounded domain ⊂ R2
which matches the shape of the drum. Then the oscillation frequencies of the drum are
given by the eigenvalues of the Laplace operator − D on with Dirichlet boundary
conditions, up to a constant that depends on the drum’s tension and mass density. In
the following we will allow the more general case ⊂ Rn for n ≥ 2, although the
physical interpretation as a drum only makes sense if n = 2. We define the Laplacian
− D via the quadratic-form approach, i.e., it is the unique self-adjoint operator in
L 2 () which is associated with the closed quadratic form

h[] = |∇|2 dx, ∈ H01 ().

Here H01 (), which is a subset of the Sobolev space W 1,2 (), is the closure of C0∞ ()
with respect to the form norm
| · |2h = h[·] + || · || L 2 () . (4.1)
For more details about the important question of how to define the Laplace operator
on arbitrary domains and subject to different boundary conditions we refer the reader
to [46,32].
The spectrum of − D is purely discrete since H0 () is, by Rellich’s theorem, com-
1
pactly imbedded in L () (see, e.g., [32]). We write λ1 () for the lowest eigenvalue
2
of − D.
Theorem 4.1 (Rayleigh–Faber–Krahn inequality) Let ⊂ Rn be an open bounded

domain with smooth boundary and ⊂ Rn a ball with the same measure as . Then,
123
λ1 (∗ ) ≤ λ1 (),
with equality if and only if itself is a ball.

Proof With the powerful mathematical tool of rearrangements (see Chapter 3) at hand,
the proof of the Rayleigh–Faber–Krahn inequality is actually not difficult. Let be
the positive normalized first eigenfunction of − D . Since the domain of a positive
self-adjoint operator is a subset of its form domain, we have ∈ H01 (). Then we
have ∈ H01 ( ). Thus we can apply first the min–max principle and then the
Theorems 3.6 and 3.10 to obtain
|∇ |2 dn x |∇| d x
2 n
λ1 ( ) ≤ ∗ 2 n
≤ = λ1 ().
| | d x d x
2 n
The Rayleigh–Faber–Krahn inequality has been extended to a number of different set-

tings, for example to Laplace operators on curved manifolds or with respect to different
measures. In the following we shall give an overview of these generalizations.
4.2 Schrödinger operators
It is not difficult to extend the Rayleigh–Faber–Krahn inequality to Schrödinger oper-

ators, i.e., to operators of the form − + V (x). Let ⊂ Rn be an open bounded
domain and V : Rn → R+ a non-negative potential in L 1 (). Then the quadratic
form

h V [u] = |∇u|2 + V (x)|u|2 dn x,

defined on

Dom h V = H01 () ∩ u ∈ L 2 () : (1 + V (x))|u(x)|2 dn x < ∞

is closed (see, e.g., [45,46]). It is associated with the positive self-adjoint Schrödinger
operator HV = − + V (x). The spectrum of HV is purely discrete and we write
λ1 (, V ) for its lowest eigenvalue.
Theorem 4.2 Under the assumptions stated above,
λ1 (∗ , V ) ≤ λ1 (, V ).
Proof Let u 1 ∈ Dom h V be the positive normalized first eigenfunction of HV . Then

we have u 1 ∈ H01 ( ) and by Theorem 3.8

(1 + V )u 1 2 dn x ≤ (1 + V )u 21 dn x < ∞.

123
Thus u 1 ∈ Dom h V and we can apply first the min–max principle and then Theo-
rems 3.6, 3.8 and 3.10 to obtain

|∇u 1 |2 + V u 1 2 dn x

λ1 ( , V ) ≤ 2 n
|u | d x
2 1
|∇u 1 | + V u 21 dn x
≤ 2 n
= λ1 (, V ).
u1 d x
4.3 Gaussian space
Consider the space Rn (n ≥ 2) endowed with the measure dμ = γ (x) dn x, where
|x|2
γ (x) = (2π )−n/2 e− 2 , (4.2)
is the standard Gaussian density. Since γ (x) is a Gauss function we will call (Rn , dμ)
the Gaussian space. For any Lebesgue-measurable ⊂ Rn we define the Gaussian
perimeter of by

Pμ () = sup ((∇ − x) · v(x))γ (x) dx : v ∈ C01 (, Rn ), ||v||∞ ≤1 .

If ∂ is sufficiently well-behaved then

Pμ () = γ (x) dH n−1 ,
∂
where H n−1 is the (n − 1)-dimensional Hausdorff measure [51]. It has been shown by
Borell that in Gaussian space there is an analog to the classical isoperimetric inequal-
ity. Yet the sets that minimize the surface (i.e., the Gaussian perimeter) for a given
volume (i.e., Gaussian measure) are not balls, as in Euclidean space, but half-spaces
[33]. More precisely:
Theorem 4.3 Let ⊂ Rn be open and measurable. Let further be the half-space
{x ∈ Rn : x1 > a}, where a ∈ R is chosen such that μ() = μ( ). Then
Pμ () ≥ Pμ ( )
with equality only if = up to a rotation.
Next we define the Laplace operator for domains in Gaussian space. We choose an
open domain ⊂ Rn with μ() < μ(Rn ) = 1 and consider the function space

1,1
H 1 (, dμ) = u ∈ Wloc () such that (u, |∇u|) ∈ L 2 (, dμ) × L 2 (, dμ) ,
123
endowed with the norm
||u|| H 1 (, dμ) = ||u|| L 2 (, dμ) + ||∇u|| L 2 (, dμ) .
We define the quadratic form

h[u] = |∇u|2 dμ

on the closure of C0∞ () in H 1 (, dμ). Since H 1 is complete, Dom h is also com-
plete under its form norm, which is equal to || · || H 1 (, dμ) . The quadratic form h
is therefore closed and associated with a unique positive self-adjoint operator −G .
Dom h is embedded compactly in L 2 (, dμ) and therefore the spectrum of −G
is discrete. Its eigenfunctions and eigenvalues are solutions of the boundary value
problem

n
∂
− ∂x j γ (x) ∂∂x j u = λγ (x)u in ,
j=1 (4.3)
u=0 on ∂.
The analog of the Rayleigh–Faber–Krahn inequality for Gaussian Spaces is the

following theorem.
Theorem 4.4 Let λ1 () be the lowest eigenvalue of −G on and let be a
half-space of the same Gaussian measure as . Then
λ1 ( ) ≤ λ1 ().
Equality holds if and only if itself is a half-space.
4.4 Spaces of constant curvature
Differential operators can not only be defined for functions in Euclidean space, but also
for the more general case of functions on Riemannian manifolds. It is therefore natural
to ask whether the isoperimetric inequalities for the eigenvalues of the Laplacian can
be generalized to such settings as well. In this section we will state Rayleigh–Faber–
Krahn type theorems for the spaces of constant non-zero curvature, i.e., for the sphere
and the hyperbolic space. Isoperimetric inequalities for the second Laplace eigenvalue
in these curved spaces will be discussed in Section 6.7.
To start with, we define the Laplacian in hyperbolic space as a self-adjoint oper-
ator by means of the quadratic form approach. We realize Hn as the open unit ball
B = {(x1 , . . . , xn ) : nj=1 x 2j < 1} endowed with the metric
4|d x|2
ds 2 = (4.4)
(1 − |x|2 )2
123
and the volume element

2n d n x
dV = , (4.5)
(1 − |x|2 )n
where | · | denotes the Euclidean norm. Let ⊂ Hn be an open domain and assume
that it is bounded in the sense that does not touch the boundary of B. The quadratic
form of the Laplace operator in hyperbolic space is the closure of

h[u] = g i j (∂i u)(∂ j u) dV, u ∈ C0∞ (). (4.6)

It is easy to see that the form (4.6) is indeed closeable: Since does not touch the
boundary of B, the metric coefficients g i j are bounded from above on . They are
also bounded from below by g i j ≥ 4. Consequently, the form norms of h and its
Euclidean counterpart, which is the right hand side of (4.6) with g i j replaced by δ i j ,
are equivalent. Since the ‘Euclidean’ form is well known to be closeable, h must also
be closeable.
By standard spectral theory, the closure of h induces an unique positive self-adjoint
operator −H which we call the Laplace operator in hyperbolic space. Equivalence
between corresponding norms in Euclidean and hyperbolic space implies that the
imbedding Dom h → L 2 (, dV ) is compact and thus the spectrum of −H is dis-
crete. For its lowest eigenvalue the following Rayleigh–Faber–Krahn inequality holds.
Theorem 4.5 Let ⊂ Hn be an open bounded domain with smooth boundary and
⊂ Hn an open geodesic ball of the same measure. Denote by λ1 () and λ1 ( ) the
lowest eigenvalue of the Dirichlet-Laplace operator on the respective domain. Then
λ1 ( ) ≤ λ1 ()
with equality only if itself is a geodesic ball.
The Laplace operator −S on a domain which is contained in the unit sphere Sn
can be defined in a completely analogous fashion to −H by just replacing the metric
g i j in (4.6) by the metric of Sn .
Theorem 4.6 Let ⊂ Sn be an open bounded domain with smooth boundary and
⊂ Sn an open geodesic ball of the same measure. Denote by λ1 () and λ1 ( ) the
lowest eigenvalue of the Dirichlet-Laplace operator on the respective domain. Then
λ1 ( ) ≤ λ1 ()
with equality only if itself is a geodesic ball.
The proofs of the above theorems are similar to the proof for the Euclidean case and
will be omitted here. A more general Rayleigh–Faber–Krahn theorem for the Laplace
operator on Riemannian manifolds and its proof can be found in the book of Chavel
[42].
123
4.5 Robin boundary conditions
Yet another generalization of the Rayleigh–Faber–Krahn inequality holds for the

boundary value problem

n
∂2
− ∂ x 2j
u = λu in ,
j=1 (4.7)
∂u
∂ν + βu =0 on ∂,
on a bounded Lipschitz domain ⊂ Rn with the outer unit normal ν and some con-
stant β > 0. This so-called Robin boundary value problem can be interpreted as a
mathematical model for a vibrating membrane whose edge is coupled elastically to
some fixed frame. The parameter β indicates how tight this binding is and the eigen-
values of (4.7) correspond the the resonant vibration frequencies of the membrane.
They form a sequence 0 < λ1 < λ2 ≤ λ3 ≤ . . . (see, e.g., [75]).
The Robin problem (4.7) is more complicated than the corresponding Dirichlet
problem for several reasons. For example, the very useful property of domain monoto-
nicity does not hold for the eigenvalues of the Robin-Laplacian. That is, if one enlarges
the domain in a certain way, the eigenvalues may go up. It is known though, that a
very weak form of domain monotonicity holds, namely that λ1 (B) ≤ λ1 () if B is ball
that contains . Another difficulty of the Robin problem, compared to the Dirichlet
case, is that the level sets of the eigenfunctions may touch the boundary. This makes it
impossible, for example, to generalize the proof of the Rayleigh–Faber–Krahn inequal-
ity in a straightforward way. Nevertheless, such an isoperimetric inequality holds, as
proven by Daners:
Theorem 4.7 Let ⊂ Rn (n ≥ 2) be a bounded Lipschitz domain, β > 0 a constant
and λ1 () the lowest eigenvalue of (4.7). Then λ1 ( ) ≤ λ1 ().
For the proof of Theorem 4.7, which is not short, we refer the reader to [44].
i) The Rayleigh–Faber–Krahn inequality is an isoperimetric inequality concerning the

lowest eigenvalue of the Laplacian, with Dirichlet boundary condition, on a bounded
domain in Rn (n ≥ 2). Let 0 < λ1 () < λ2 () ≤ λ3 () ≤ . . . be the Dirichlet
eigenvalues of the Laplacian in ⊂ Rn , i.e.,
−u = λu in ,
u = 0 on the boundary of .
If n = 2, the Dirichlet eigenvalues are proportional to the square of the eigenfre-

quencies of an elastic, homogeneous, vibrating membrane with fixed boundary. The
Rayleigh–Faber–Krahn inequality for the membrane (i.e., n = 2) states that
π j0,1
2
λ1 ≥ ,
A
123
where j0,1 = 2.4048 . . . is the first zero of the Bessel function of order zero, and
A is the area of the membrane. Equality is obtained if and only if the membrane is
circular. In other words, among all membranes of given area, the circle has the lowest
fundamental frequency. This inequality was conjectured by Lord Rayleigh (see [88],
pp. 339–340). In 1918, Courant (see R. Courant, Math. Z. 1, 321–328 (1918)) proved
the weaker result that among all membranes of the same perimeter L the circular one
yields the least lowest eigenvalue, i.e.,
2
4π 2 j0,1
λ1 ≥ ,
L2
with equality if and only if the membrane is circular. Rayleigh’s conjecture was proven
independently by Faber [50] and Krahn [64]. The corresponding isoperimetric inequal-
ity in dimension n,

1 2/n 2/n
λ1 () ≥ Cn jn/2−1,1 ,
||
was proven by Krahn [65]. Here jm,1 is the first positive zero of the Bessel function
Jm , || is the volume of the domain, and Cn = π n/2 / (n/2 + 1) is the volume of
the n-dimensional unit ball. Equality is attained if and only if is a ball. For more
details see, R.D. Benguria, Rayleigh–Faber–Krahn Inequality, in Encyclopaedia of
Mathematics, Supplement III, Managing Editor: M. Hazewinkel, Kluwer Academic
Publishers, pp. 325–327, (2001).
ii) A natural question to ask concerning the Rayleigh–Faber–Krahn inequality is the
question of stability. If the lowest eigenvalue of a domain is within (positive
and sufficiently small) of the isoperimetric value λ1 (∗ ), how close is the domain
to being a ball? The problem of stability for (convex domains) concerning the
Rayleigh–Faber–Krahn inequality was solved by Antonios Melas (see, A. D. Melas,
The stability of some eigenvalue estimates, J. Differential Geom. 36, 19–33 (1992)).
In the same reference, Melas also solved the analogous stability problem for convex
domains with respect to the PPW inequality (see Chapter 6 below). The work of Melas
has been extended to the case of the Szegö–Weinberger inequality (for the first non-
trivial Neumann eigenvalue) by Y.-Y. Xu, The first nonzero eigenvalue of Neumann
problem on Riemannian manifolds, J. Geom. Anal. 5, 151–165 (1995), and to the case
of the PPW inequality on spaces of constant curvature by A. Avila, Stability results
for the first eigenvalue of the Laplacian on domains in space forms, J. Math. Anal.
Appl. 267, 760–774 (2002). In this connection it is worth mentioning related results
on the isoperimetric inequality of R. Hall, A quantitative isoperimetric inequality in
n-dimensional space, J. Reine Angew Math. 428, 161–176 (1992), as well as recent
results of Maggi, Pratelli and Fusco (recently reviewed by F. Maggi in Bull. Amer.
Math. Soc. 45, 367–408 (2008)).
iii) The analog of the Faber–Krahn inequality for domains in the sphere Sn was proven
by Sperner, Emanuel, Jr. Zur Symmetrisierung von Funktionen auf Sphären, Math. Z.
134, 317–327 (1973).
iv) For isoperimetric inequalities for the lowest eigenvalue of the Laplace–Beltrami
operator on manifolds, see, e.g., the book by Chavel, Isaac, Eigenvalues in
123
Riemannian geometry. Pure and Applied Mathematics, 115. Academic Press, Inc.,
Orlando, FL, 1984, (in particular Chapters IV and V), and also the articles, Chavel, I.
and Feldman, E. A. Isoperimetric inequalities on curved surfaces. Adv. in Math. 37,
83–98 (1980), and Bandle, Catherine, Konstruktion isoperimetrischer Ungleichungen
der mathematischen Physik aus solchen der Geometrie, Comment. Math. Helv. 46,
182–213 (1971).
5 The Szegö–Weinberger inequality
In analogy to the Rayleigh–Faber–Krahn inequality for the Dirichlet–Laplacian one

may ask which shape of a domain maximizes certain eigenvalues of the Laplace opera-
tor with Neumann boundary conditions. Of course, this question is trivial for the lowest
Neumann eigenvalue, which is always zero. In 1952 Kornhauser and Stakgold [63]
conjectured that the ball maximizes the first non-zero Neumann eigenvalue among all
domains of the same volume. This was first proven in 1954 by Szegö [97] for two-
dimensional simply connected domains, using conformal mappings. Two years later
his result was generalized general domains in any dimension by Weinberger [103],
who came up with a new strategy for the proof.
Although the Szegö–Weinberger inequality appears to be the analog for Neumann
eigenvalues of the Rayleigh–Faber–Krahn inequality, its proof is completely different.
The reason is that the first non-trivial Neumann eigenfunction must be orthogonal to
the constant function, and thus it must have a change of sign. The simple symmetri-
zation procedure that is used to establish the Rayleigh–Faber–Krahn inequality can
therefore not work.
In general, when dealing with Neumann problems, one has to take into account that
the spectrum of the respective Laplace operator on a bounded domain is very unstable
under perturbations. One can change the spectrum arbitrarily much by only a slight
modification of the domain, and if the boundary is not smooth enough, the Laplacian
may even have essential spectrum. A sufficient condition for the spectrum of − N to
be purely discrete is that is bounded and has a Lipschitz boundary [46]. We write
0 = μ0 () < μ1 () ≤ μ2 () ≤ . . . for the sequence of Neumann eigenvalues on
such a domain .
Theorem 5.1 (Szegö–Weinberger inequality) Let ⊂ Rn be an open bounded

domain with smooth boundary such that the Laplace operator on with Neumann
boundary conditions has purely discrete spectrum. Then
μ1 () ≤ μ1 ( ), (5.1)
where ⊂ Rn is a ball with the same n-volume as . Equality holds if and only if
itself is a ball.
Proof By a standard separation of variables one shows that μ1 ( ) is n-fold degen-
erate and that a basis of the corresponding eigenspace can be written in the form
{g(r )r j r −1 } j=1,...,n . The function g can be chosen to be positive and satisfies the
differential equation
123

n−1 n−1
g + g + μ1 ( ) − 2 g = 0, 0 < r < r1 , (5.2)
r r
where r1 is the radius of . Further, g(r ) vanishes at r = 0 and its derivative has its
first zero at r = r1 . We extend g by defining g(r ) = limr ↑r1 g(r ) for r ≥ r1 . Then
g is differentiable on R and if we set f j (r ) := g(r )r j r −1 then f j ∈ W 1,2 () for
j = 1 . . . , n. To apply the min–max principle with f j as a test function for μ1 () we
have to make sure that f j is orthogonal to the first (trivial) eigenfunction, i.e., that

f j dn r = 0, j = 1, . . . , n. (5.3)

We argue that this can be achieved by some shift of the domain : Since is bounded
we can find a ball B that contains . Now define the vector field b : Rn → Rn by its
components

b j (v) = f j (r ) dn r, v ∈ Rn .
+v
For v ∈ ∂ B we have

v ·r
v · b(v) = g(r ) dn r
+v r

v · (r + v)
= g(|r + v|) dn r
|r + v|

|v|2 − |v| · |r |
≥ g(|r + v|) dn r > 0.
|r + v|
Thus b is a vector field that points outwards on every point of ∂ B. By an application

of the Brouwer’s fixed-point theorem (see Theorem 10.3 in the Appendix) this means
that b(v0 ) = 0 for some v0 ∈ B. Thus, if we shift by this vector, condition (5.3) is
satisfied and we can apply the min-max principle with the f j as test functions for the
first non-zero eigenvalue:
|∇ f j | d r
n
μ1 () ≤ 2 n
fj d r

g 2 (r )r 2j r −2 + g 2 (r )(1 − r 2j r −2 )r −2 dn r
= .
g 2 r 2j r −2 dn r
We multiply each of these inequalities by the denominator and sum up over j to obtain
B(r ) dn r
μ1 () ≤ (5.4)
g (r ) d r
2 n
123
with B(r ) = g 2 (r ) + (n − 1)g 2 (r )r −2 . Since r1 is the first zero of g , the function g

is non-decreasing. The derivative of B is
B = 2g g + 2(n − 1)(rgg − g 2 )r −3 .
For r ≥ r1 this is clearly negative since g is constant there. For r < r1 we can use
equation (5.2) to show that
B = −2μ1 ( )gg − (n − 1)(rg − g)2 r −3 < 0.
If the following we will use the method of rearrangements, which was described in
Chapter 3. To avoid confusions, we use a more precise notation at this point: We intro-
duce B : → R , B (r ) = B(r ) and analogously g : → R, g (r ) = g(r ).
Then equation (5.4) yields, using Theorem 3.7 in the third step:
B (r ) dn r (r ) dn r
B B(r ) dn r

μ1 () ≤
= ≤ = μ1 ( ) (5.5)
g (r ) d r g (r ) d r
2 n 2 (r ) dn r
g 2 n

Equality holds obviously if is a ball. In any other case the third step in (5.5) is a
strict inequality.
It is rather straightforward to generalize the Szegö–Weinberger inequality to

domains in hyperbolic space. For domains on spheres, on the other hand, the cor-
responding inequality has not been established yet in full generality. At present, the
most general result is due to Ashbaugh and Benguria: In [12] they show that an ana-
log of the Szegö–Weinberger inequality holds for domains that are contained in a
hemisphere.
i) In 1952, Kornhauser and Stakgold [63] conjectured that the lowest nontrivial Neu-
mann eigenvalue for a smooth bounded domain in R2 satisfies the isoperimetric
inequality
π p2
μ1 () ≤ μ1 (∗ ) = ,
A
where ∗ is a disk with the same area as , and p = 1.8412 . . . is the first positive
zero of the derivative of the Bessel function J1 . This conjecture was proven by G.
Szegö in 1954, using conformal maps [97]. The extension to n dimensions was proven
by H. Weinberger [103].
ii) For the case of mixed boundary conditions, Marie–Helene Bossel [Membranes
élastiquement liées inhomogénes ou sur une surface: une nouvelle extension du théo-
reme isopérimétrique de Rayleigh–Faber–Krahn, Z. Angew. Math. Phys. 39, 733–742
(1988)] proved the analog of the Rayleigh–Faber–Krahn inequality.
123
iii) Very recently, A. Girouard, N. Nadirashvili and I. Polterovich proved that the sec-
ond positive eigenvalue of a bounded simply connected planar domain of a given area
does not exceed the first positive Neumann eigenvalue on a disk of a twice smaller area
(see, Maximization of the second positive Neumann eigenvalue for planar domains,
J. Differential Geom. 83, 637–662 (2009)). For a review of optimization of eigenvalues
with respect to the geometry of the domain, see the recent monograph of A. Henrot
[60].
iv) In the Bibliographical Remarks of Section 4 (see Section 4.6, ii)) we discussed
the stability results of A. Melas for the Rayleigh–Faber–Krahn inequality. In the same
vein, recently L. Brasco and A. Pratellli, Sharp Stability of some Spectral Inequali-
ties, preprint (2011), have proven related stability results for the Szegö–Weinberger
inequality. Moreover, these authors have also proven stability results for the E. Krahn–
P. Szego inequality, which says that among all sets of a given measure (in Euclidean
Space) the disjoint union of two balls with the same radius minimizes the second
eigenvalue of the Dirichlet Laplacian.
6 The Payne–Pólya–Weinberger inequality
6.1 Introduction
A further isoperimetric inequality is concerned with the second eigenvalue of the

Dirichlet–Laplacian on bounded domains. In 1955 Payne, Pólya and Weinberger
(PPW) showed that for any open bounded domain ⊂ R2 the bound λ2 ()/λ1 () ≤
3 holds [83,84]. Based on exact calculations for simple domains they also conjectured
that the ratio λ2 ()/λ1 () is maximized when is a circular disk, i.e., that
λ2 ( )
2
λ2 () j1,1
≤ = ≈ 2.539 for ⊂ R2 . (6.1)
λ1 () λ1 ( ) 2
j0,1
Here, jn,m denotes the m th positive zero of the Bessel function Jn (x). This conjecture
and the corresponding inequalities in n dimensions were proven in 1991 by Ashbaugh
and Benguria [9–11]. Since the Dirichlet eigenvalues on a ball are inversely propor-
tional to the square of the ball’s radius, the ratio λ2 ( )/λ1 ( ) does not depend on
the size of . Thus we can state the PPW inequality in the following form:
Theorem 6.1 (Payne–Pólya–Weinberger inequality) Let ⊂ Rn be an open bounded

domain and S1 ⊂ Rn a ball such that λ1 () = λ1 (S1 ). Then
λ2 () ≤ λ2 (S1 ) (6.2)
with equality if and only if is a ball.
Here the subscript 1 on S1 reflects the fact that the ball S1 has the same first Dirichlet
eigenvalue as the original domain . The inequalities (6.1) and (6.2) are equivalent
in Euclidean space in view of the mentioned scaling properties of the eigenvalues.
123
Yet when one considers possible extensions of the PPW inequality to other settings,
where λ2 /λ1 varies with the radius of the ball, it turns out that an estimate in the form
of Theorem 6.1 is the more natural result. In the case of a domain on a hemisphere, for
example, λ2 /λ1 on balls is an increasing function of the radius. But by the Rayleigh–
Faber–Krahn inequality for spheres the radius of S1 is smaller than the one of the
spherical rearrangement . This means that an estimate in the form of Theorem 6.1,
interpreted as
λ2 () λ2 (S1 )
≤ , , S1 ⊂ Sn ,
λ1 () λ1 (S1 )
is stronger than an inequality of the type (6.1).

On the other hand, we will see that in the hyperbolic space λ2 /λ1 on balls is a strictly
decreasing function of the radius. In this case we can apply the following argument
to see that an estimate of the type (6.1) cannot possibly hold true: Consider a domain
that is constructed by attaching very long and thin tentacles to the ball B. Then the
first and second eigenvalues of the Laplacian on are arbitrarily close to the ones on
B. The spherical rearrangement of though can be considerably larger than B. This
means that
λ2 () λ2 (B) λ2 ( )

≈ > , B, ⊂ Hn ,
λ1 () λ1 (B) λ1 ( )
clearly ruling out any inequality in the form of (6.1).

The proof of the PPW inequality (6.2) is somewhat similar to that of the Szegö–
Weinberger inequality (see Chapter 5), but considerably more difficult. The additional
complications mainly stem from the fact that in the Dirichlet case the first eigenfunc-
tion of the Laplacian is not known explicitly, while in the Neumann case it is just
constant. We will give the full proof of the PPW inequality in the following three
sections. Since it is quite long, a brief outline is in order:
The proof is organized in six steps. In the first one we use the min–max principle to
derive an estimate for the eigenvalue gap λ2 () − λ1 (), depending on a test function
for the second eigenvalue. In the second step we define such a function and then show
in the third step that it actually satisfies all requirements to be used in the gap formula.
In the fourth step we put the test function into the gap inequality and then estimate the
result with the help of rearrangement techniques. These depend on the monotonicity
properties of two functions g and B, which are to be defined in the proof, and on a
Chiti comparison argument. The later is a special comparison result which establishes
a crossing property between the symmetric decreasing rearrangement of the first ei-
genfunction on and the first eigenfunction on S1 . We end up with the inequality
λ2 () − λ1 () ≤ λ2 (S1 ) − λ1 (S1 ), which yields (6.2). In the remaining two steps
we prove the mentioned monotonicity properties and the Chiti comparison result. We
remark that from the Rayleigh–Faber–Krahn inequality follows S1 ⊂ , a fact that
is used in the proof of the Chiti comparison result. Although it enters in a rather subtle
manner, the Rayleigh–Faber–Krahn inequality is an important ingredient of the proof
of the PPW inequality.
123
6.2 Proof of the Payne–Pólya–Weinberger inequality
First step: We derive the ‘gap formula’ for the first two eigenvalues of the Dirichlet–
Laplacian on . We call u 1 : → R+ the positive normalized first eigenfunction of
− D . To estimate the second eigenvalue we will use the test function Pu , where
1
P : → R is is chosen such that Pu 1 is in the form domain of − D and

Pu 21 dr n = 0. (6.3)

Then we conclude from the min–max principle that

|∇(Pu 1 )|2 − λ1 P 2 u 21 dr n
λ2 () − λ1 () ≤ 2 2 n
P u 1 dr
n
|∇ P| u 1 +(∇ P )u 1 ∇u 1 + P |∇u 1 | −λ1 P u 1 dr
2 2 2 2 2 2 2
= (6.4)
P 2 u 21 dr n
If we perform an integration by parts on the second summand in the numerator of

(6.4), we see that all summands except the first cancel. We obtain the gap inequality
|∇ P| u 1 dr
2 2 n
λ2 () − λ1 () ≤ 2 2 n
. (6.5)
P u 1 dr
Second step: We need to fix the test function P. Our choice will be dictated by
the requirement that equality should hold in (6.5) if is a ball, i.e., if = S1 up to
translations. We assume that S1 is centered at the origin of our coordinate system and
call R1 its radius. We write z 1 (r ) for the first eigenfunction of the Dirichlet Laplacian
on S1 . This function is spherically symmetric with respect to the origin and we can
take it to be positive and normalized in L 2 (S1 ). The second eigenvalue of − SD1 in n
dimensions is n-fold degenerate and a basis of the corresponding eigenspace can be
written in the form z 2 (r )r j r −1 with z 2 ≥ 0 and j = 1, . . . , n. This is the motivation
to choose not only one test function P, but rather n functions P j with j = 1, . . . , n.
We set
P j = r j r −1 g(r )
with
z
2 (r )
z 1 (r ) for r < R1 ,
g(r ) = (r )
limr ↑R1 zz21 (r ) for r ≥ R1 .
We note that P j u 1 is a second eigenfunction of − D if is a ball which is centered
at the origin.
Third step: It is necessary to verify that the P j u 1 are admissible test functions. First,
we have to make sure that condition (6.3) is satisfied. We note that P j changes when
123
(and u 1 with it) is shifted in Rn . Since these shifts do not change λ1 () and λ2 (),
it is sufficient to show that can be moved in Rn such that (6.3) is satisfied for all
j ∈ {1, . . . , n}. To this end we define the function

r
b(v) = u 21 (|r − v|) g(r ) dr n for v ∈ Rn .
+v r
Since is a bounded domain, we can choose some closed ball D, centered at the
origin, such that ⊂ D. Then for every v ∈ ∂ D we have

r +v
v · b(v) = v · u 21 (r ) g(|r + v|) dr n
|r + v|

|v|2 − |v| · |r |
> u 21 (r ) g(|r + v|) dr n > 0
|r + v|
Thus the continuous vector-valued function b(v) points strictly outwards everywhere
on ∂ D. By Theorem 10.3, which is a consequence of the Brouwer fixed-point theorem,
there is some v0 ∈ D such that b(v0 ) = 0. Now we shift by this vector, i.e., we
replace by − v0 and u 1 by the first eigenfunction of the shifted domain. Then the
test functions P j u 1 satisfy the condition (6.3).
The second requirement on P j u 1 is that it must be in the form domain of − D,
i.e., in H0 (): Since u 1 ∈ H0 () there is a sequence {vn ∈ C ()}n∈N of func-

1 1 1
tions with compact support such that | · |h − limn→∞ vn = u 1 , using the definition
(4.1) of | · |h . The functions P j vn also have compact support and one can check
that P j vn ∈ C 1 () (P j is continuously differentiable since g (R1 ) = 0). We have
| · |h − limn→∞ P j vn = P j u 1 and thus P j u 1 ∈ H01 ().
Fourth step: We multiply the gap inequality (6.5) by P 2 u 21 dx and put in our
special choice of P j to obtain
r2 r 2
j j
(λ2 − λ1 ) 2
g 2 (r )u 21 (r ) dr n ≤ ∇ g(r ) u 21 (r ) dr n
r r
r 2j

rj 2

= ∇ g (r ) +
2
g (r ) 2
u 21 (r ) dr n .
r r2
Now we sum these inequalities up over j = 1, . . . , n and then divide again by the
integral on the left hand side to get
B(r )u 21 (r ) dr n
λ2 () − λ1 () ≤ (6.6)
g 2 (r )u 21 (r ) dr n
with
B(r ) = g (r )2 + (n − 1)r −2 g(r )2 . (6.7)
123
If the following we will use the method of rearrangements, which was described in
Chapter 3. To avoid confusions, we use a more precise notation at this point: We intro-
duce B : → R, B (r ) = B(r ) and analogously g : → R, g (r ) = g(r ).
Then equation (6.6) can be written as
B (r )u 21 (r ) dr n
λ2 () − λ1 () ≤ . (6.8)
g (r )u 1 (r ) dr
2 2 n
Then by Theorem 3.8 the following inequality is also true:

(r )u (r )2 dr n
B
1
λ2 () − λ1 () ≤ 2 (r )u (r )2 dr n
. (6.9)
g 1
Next we use the very important fact that g(r ) is an increasing function and B(r ) is
a decreasing function, which we will prove in step five below. These monotonicity
(r ) ≤ B(r ) and g (r ) ≥ g(r ). Therefore
properties imply by Theorem 3.7 that B
B(r )u 1 (r )2 dr n
λ2 () − λ1 () ≤ . (6.10)
g (r )u 1 (r ) dr
2 2 n
Finally we use the following version of Chiti’s comparison theorem to estimate the
right hand side of (6.10):
Lemma 6.2 (Chiti comparison result) There is some r0 ∈ (0, R1 ) such that
z 1 (r ) ≥ u 1 (r ) for r ∈ (0, r0 ) and

z 1 (r ) ≤ u 1 (r ) for r ∈ (r0 , R1 ).
We remind the reader that the function z 1 denotes the first Dirichlet eigenfunction for
the Laplacian defined on S1 . Applying Lemma 6.2, which will be proven below in
step six, to (6.10) yields
B(r )z 1 (r )2 dr n
λ2 () − λ1 () ≤ = λ2 (S1 ) − λ1 (S1 ). (6.11)
g (r )z 1 (r ) dr
2 2 n
Since S1 was chosen such that λ1 () = λ1 (S1 ) the above relation proves that λ2 () ≤
λ2 (S1 ). It remains the question: When does equality hold in (6.2)? It is obvious that
equality does hold if is a ball, since then = S1 up to translations. On the other
hand, if is not a ball, then (for example) the step from (6.10) to (6.11) is not sharp.
Thus (6.2) is a strict inequality if is not a ball.
6.3 Monotonicity of B and g
Fifth step: We prove that g(r ) is an increasing function and B(r ) is a decreasing func-
tion. In this step we abbreviate λi = λi (S1 ). The functions z 1 and z 2 are solutions of
the differential equations
123
n−1
− z 1 − z 1 − λ1 z 1 = 0,
r (6.12)
n−1 n−1
−z 2 − z2 + − λ 2 z2 = 0
r r2
with the boundary conditions
z 1 (0) = 0, z 1 (R1 ) = 0, z 2 (0) = 0, z 2 (R1 ) = 0. (6.13)
We define the function

⎧ rg (r )
⎨ g(r ) for r ∈ (0, R1 ),
q(r ) := limr ↓0 q(r ) for r = 0, (6.14)
⎩
limr ↑R1 q(r ) for r = R1 .
Proving the monotonicity of B and g is thus reduced to showing that 0 ≤ q(r ) ≤ 1

and q (r ) ≤ 0 for r ∈ [0, R1 ]. Using the definition of g and the equations (6.12), one
can show that q(r ) is a solution of the Riccati differential equation
(1 − q)(q + n − 1) z
q = (λ1 − λ2 )r + − 2q 1 . (6.15)
r z1
It is straightforward to establish the boundary behavior

2 2
q(0) = 1, q (0) = 0, q (0) = 1+ λ1 − λ2
n n
and
q(R1 ) = 0.
Lemma 6.3 For 0 ≤ r ≤ R1 we have q(r ) ≥ 0.
Proof Assume the contrary. Then there exist two points 0 < s1 < s2 ≤ R1 such that
q(s1 ) = q(s2 ) = 0 but q (s1 ) ≤ 0 and q (s2 ) ≥ 0. If s2 < R1 then the Riccati equation
(6.15) yields
n−1 n−1
0 ≥ q (s1 ) = (λ1 − λ2 )s1 + > (λ1 − λ2 )s2 + = q (s2 ) ≥ 0,
s1 s2
which is a contradiction. If s2 = R1 then we get a contradiction in a similar way by
n−1 n−1
0 ≥ q (s1 ) = (λ1 − λ2 )s1 + > (λ1 − λ2 )R1 + = 3q (R1 ) ≥ 0.
s1 R1
In the following we will analyze the behavior of q according to (6.15), considering

r and q as two independent variables. For the sake of a compact notation we will make
use of the following abbreviations:
123
p(r ) = z 1 (r )/z 1 (r )
Ny = y2 − n + 1
Q y = 2yλ1 + (λ2 − λ1 )N y y −1 − 2(λ2 − λ1 )
M y = N y2 /(2y) − (n − 2)2 y/2
We further define the function

(n − 2)y + N y
T (r, y) := −2 p(r )y − − (λ2 − λ1 )r. (6.16)
r
Then we can write (6.15) as
q (r ) = T (r, q(r )).
The definition of T (r, y) allows us to analyze the Riccati equation for q considering
r and q(r ) as independent variables. For r going to zero, p is O(r ) and thus
1
T (r, y) = ((n − 1 + y)(1 − y)) + O(r ) for y fixed.
r
Consequently,
limr →0 T (r, y) = +∞ for 0 ≤ y < 1 fixed,

limr →0 T (r, y) = 0 for y = 1 and
limr →0 T (r, y) = −∞ for y > 1 fixed.
The partial derivative of T (r, y) with respect to r is given by
∂ (n − 2)y Ny
T = T (r, y) = −2yp + + 2 − (λ2 − λ1 ). (6.17)
∂r r 2 r
In the points (r, y) where T (r, y) = 0 we have, by (6.16),
n−2 Ny (λ2 − λ1 )r
p|T =0 = − − − . (6.18)
2r 2yr 2y
From (6.12) we get the Riccati equation
n−1
p + p2 + p + λ1 = 0. (6.19)
r
Putting (6.18) into (6.19) and the result into (6.17) yields
My (λ2 − λ1 )2 2
T |T =0 = 2
+ r + Qy. (6.20)
r 2y
Lemma 6.4 There is some r0 > 0 such that q(r ) ≤ 1 for all r ∈ (0, r0 ) and q(r0 ) < 1.
123
Proof Suppose the contrary, i.e., q(r ) first increases away from r = 0. Then, because
q(0) = 1 and q(R1 ) = 0 and because q is continuous and differentiable, we can find
two points s1 < s2 such that q̂ := q(s1 ) = q(s2 ) > 1 and q (s1 ) > 0 > q (s2 ). Even
more, we can chose s1 and s2 such that q̂ is arbitrarily close to one. Writing q̂ = 1 +
with > 0, we can calculate from the definition of Q y that
Q 1+ = Q 1 + n(λ2 − (1 − 2/n)λ1 ) + O( 2 ).
The term in brackets can be estimated by
λ2 − (1 − 2/n)λ1 > λ2 − λ1 > 0.
We can also assume that Q 1 ≥ 0, because otherwise q (0) = n22 Q 1 < 0 and
Lemma 6.4 is immediately true. Thus, choosing R1 and r2 such that is sufficiently
small, we can make sure that Q q̂ > 0.
Now consider T (r, q̂) as a function of r for our fixed q̂. We have T (s1 , q̂) > 0 >
T (s2 , q̂) and the boundary behavior T (0, q̂) = −∞. Consequently, T (r, q̂) changes
its sign at least twice on [0, R1 ] and thus we can find two zeros 0 < ŝ1 < ŝ2 < R1 of
T (r, q̂) such that
T (ŝ1 , q̂) ≥ 0 and T (ŝ2 , q̂) ≤ 0. (6.21)
But from (6.20), together with Q q̂ > 0, one can see easily that this is impossible,
because the right hand side of (6.20) is either positive or increasing (depending on
Mq̂ ). This is a contradiction to our assumption that q first increases away from r = 0,
proving Lemma 6.4.
Lemma 6.5 For all 0 ≤ r ≤ R1 the inequality q (r ) ≤ 0 holds.
Proof Assume the contrary. Then, because of q(0) = 1 and q(R1 ) = 0, there are
three points s1 < s2 < s3 in (0, R1 ) with 0 < q̂ := q(s1 ) = q(s2 ) = q(s3 ) < 1
and q (s1 ) < 0, q (s2 ) > 0, q (s3 ) < 0. Consider the function T (r, q̂), which
coincides with q (r ) at s1 , s2 , s3 . Taking into account its boundary behavior at
r = 0, it is clear that T (r, q̂) must have at least the sign changes positive-nega-
tive-positive-negative. Thus T (r, q̂) has at least three zeros ŝ1 < ŝ2 < ŝ3 with the
properties
T (ŝ1 , q̂) ≤ 0, T (ŝ2 , q̂) ≥ 0, T (ŝ3 , q̂) ≤ 0.
Again one can see from (6.20) that this is impossible, because the term on the right
hand side is either a strictly convex or a strictly increasing function of r . We conclude
that Lemma 6.5 is true.
Altogether we have shown that 0 ≤ q(r ) ≤ 1 and q (r ) ≤ 0 for all r ∈ (0, R1 ), which
proves that g is increasing and B is decreasing.
123
6.4 The Chiti comparison result
Sixth step: We prove Lemma 6.2: Here and in the sequel we write short-hand λ1 =
λ1 () = λ1 (S1 ). We introduce a change of variables via s = Cn r n , where Cn is the

volume of the n-dimensional unit ball. Then by Definition 3.2 we have u 1 (s) = u 1 (r )

and z 1 (s) = z 1 (r ).

Lemma 6.6 For the functions u 1 (s) and z 1 (s) we have
s
du 1 −2/n n/2−2
− ≤ λ1 n −2 Cn s u 1 (w) dw, (6.22)
ds 0
s
dz −2/n n/2−2
− 1 = λ1 n −2 Cn s z 1 (w) dw. (6.23)
ds 0
Proof We integrate both sides of −u 1 = λ1 u 1 over the level set t := {r ∈ :

u 1 (r ) > t} and use Gauss’ Divergence Theorem to obtain

|∇u 1 |Hn−1 ( dr ) = λ1 u 1 (r ) dn r, (6.24)
∂t t
where ∂t = {r ∈ : u 1 (r ) = t}. Now we define the distribution function μ(t) =

|t |. Then by Theorem 3.9 we have

2/n μ(t)
2−2/n
|∇u 1 |Hn−1 ( dr ) ≥ −n 2 Cn . (6.25)
∂t μ (t)
The left sides of (6.24) and (6.25) are the same, thus

2/n μ(t)
2−2/n
−n 2
Cn
≤ λ1 u 1 (r ) dn r
μ (t) t
(μ(t)/Cn )1/n
= nCn r n−1 λ1 u 1 (r ) dr.
0
Now we perform the change of variables r → s on the right hand side of the above chain

of inequalities. We also chose t to be u 1 (s). Using the fact that u 1 and μ are essentially

inverse functions to one another, this means that μ(t) = s and μ (t)−1 = (u 1 ) (s).
The result is (6.22). Equation (6.23) is proven analogously, with equality in each step.

Lemma 6.6 enables us to prove Lemma 6.2. The function z 1 is continuous on

(0, |S1 |) and u 1 is continuous on (0, | |). By the normalization of u 1 and z 1 and

because S1 ⊂ it is clear that either z 1 ≥ u 1 on (0, |S1 |) or u 1 and z 1 have at least
one intersection on this interval. In the first case there is nothing to prove, simply
123
setting r0 = R1 in Lemma 6.2. In the second case we have to show that there is no

intersection of u 1 and z 1 such that u 1 is greater than z 1 on the left and smaller on the
right. So we assume the contrary, i.e., that there are two points 0 ≤ s1 < s2 < |S1 |

such that u 1 (s) > z 1 (s) for s ∈ (s1 , s2 ), u 1 (s2 ) = z 1 (s2 ) and either u 1 (s1 ) = z 1 (s1 )
or s1 = 0. We set
⎧ s1 s1
⎪
⎪ u 1 (s) on [0, s1 ] if u 1 (s) ds > z 1 (s) ds,
⎪
⎨ 0
s1
0
s1
z 1 (s) on [0, s1 ] if u 1 (s) ds ≤ z 1 (s) ds,
v (s) =
0 0 (6.26)
⎪
⎪ u (s) on [s1 , s2 ],
⎪
⎩ 1
z 1 (s) on [s2 , |S1 |].
Then one can convince oneself that because of (6.22) and (6.23)

dv −2/n n/2−2
s
− ≤ λ1 n −2 Cn s v (s ) ds (6.27)
ds 0
for all s ∈ [0, |S1 |]. Now define the test function v(r ) = v (Cn r n ). Using the Ray-
leigh–Ritz characterization of λ1 , then (6.27) and finally an integration by parts, we
get (if z 1 and u 1 are not identical)
|S1 |

λ1 v (r ) dx <
2
|∇v| dx = 2
(nCn r n−1 v (s))2 ds
S1 S1 0
|S1 | s

≤− v (s)λ1 v (s ) ds ds
0 0
|S1 | s S1
= λ1 v (s)2 ds − λ1 v (s) v (s ) ds
0 0 0
≤ λ1 v (r ) dx
2
S1
Comparing the first and the last term in the above chain of (in)equalities reveals a
contradiction to our assumption that the intersection point s2 exists, thus proving
Lemma 6.2.
6.5 Schrödinger operators
Theorem 6.1 can be extended in several directions. One generalization, which has
been considered by Benguria and Linde in [23], is to replace the Laplace operator
on the domain ⊂ Rn by a Schrödinger operator H = − + V . In this case the
question arises which is the most suitable comparison operator for H . In analogy to
the PPW inequality for the Laplacian, it seems natural to compare the eigenvalues of
H to those of another Schrödinger operator H̃ = − + Ṽ , which is defined on a
ball and has the same lowest eigenvalue as H . The potential Ṽ should be spherically
symmetric and it should reflect some properties of V, but it will also have to satisfy
123
certain requirements in order for the PPW type estimate to hold. The precise result is
stated in Theorem 6.7 below, which can be considered as a natural generalization of
Theorem 6.1 to Schrödinger operators.
We assume that is open and bounded and that V : → R+ is a non-negative
potential from L 1 (). Then we can define the Schrödinger operator HV = − + V
on in the same way as we did in Section 4.2, i.e., HV is positive and self-adjoint in
L 2 () and has purely discrete spectrum. We call λi (, V ) its i-th eigenvalue and, as
usual, we write V for the symmetric increasing rearrangement of V .
Theorem 6.7 Let S1 ⊂ Rn be a ball centered at the origin and of radius R1 and let
Ṽ : S1 → R+ be a radially symmetric non-negative potential such that Ṽ (r ) ≤ V (r )
for all 0 ≤ r ≤ R1 and λ1 (, V ) = λ1 (S1 , Ṽ ). If Ṽ (r ) satisfies the conditions
a) Ṽ (0) = Ṽ (0) = 0 and
b) Ṽ (r ) exists and is increasing and convex,
then
λ2 (, V ) ≤ λ2 (S1 , Ṽ ). (6.28)
If V is such that V itself satisfies the conditions a) and b) of the theorem, the best
bound is obtained by choosing Ṽ = V and then adjusting the size of S1 such that
λ1 (, V ) = λ1 (S1 , V ) holds. (Note that S1 ⊂ by Theorem 4.2). In this case
Theorem 6.7 is a typical PPW result and optimal in the sense that equality holds in
(6.28) if is a ball and V = V . For a general potential V we still get a non-trivial
bound on λ2 (, V ) though it is not sharp anymore.
For further reference we state the following theorem, which is a direct consequence
of Theorem 6.7 and Theorem 3.7:
Theorem 6.8 Let Ṽ : Rn → R+ be a radially symmetric positive potential that satis-

fies the conditions a) and b) of Theorem 6.1. Further, assume that ⊂ Rn is an open
bounded domain and that S1 ⊂ Rn be the open ball (centered at the origin) such that
λ1 (, Ṽ ) = λ1 (S1 , Ṽ ). Then
λ2 (, Ṽ ) ≤ λ2 (S1 , Ṽ ).
The proof of Theorem 6.7 is similar to the one of Theorem 6.1 and can be found
in [23]. One of the main differences occurs in step five (see Section 6.3), since the
potential Ṽ (r ) now appears in the Riccati equation for p. It turns out that the condi-
tions a) and b) in Theorem 6.7 are required to establish the monotonicity properties
of q. A second important difference is that a second eigenfunction of a Schrödinger
operator with a spherically symmetric potential can not necessarily be written in the
form u 2 (r )r j r −1 . It has been shown by Ashbaugh and Benguria [7] that it can be
written in this form if r V (r ) is convex. On the other hand, the second eigenfunction
is radially symmetric (with a spherical nodal surface) if r V (r ) is concave. This fact,
which is also known as the Baumgartner–Grosse–Martin Inequality [22], is another
reason why the conditions a) and b) of Theorem 6.7 are needed.
123
6.6 Gaussian space
Theorem 6.8 has direct consequences for the eigenvalues of the Laplace operator −G
in Gaussian space, which had been defined in Section 4.3. In this section we write
λi− () for the i-th eigenvalue of −G on some domain .
Theorem 6.9 Let ⊂ Rn be an open bounded domain and assume that S1 ⊂ Rn is
a ball, centered at the origin, such that λ− −
1 () = λ1 (S1 ). Then
λ− −
2 () ≤ λ2 (S1 ).
Proof If is an eigenfunction of −G on then e−r /2 is an eigenfunction of the

2
Dirichlet–Schrödinger operator − + r 2 on , and vice versa. The eigenvalues are

related by
λi− () = λ(, r 2 − n),
where we keep using the notation from Section 6.5. Theorem 6.9 now follows directly
from Theorem 6.8, setting Ṽ (r ) = r 2 .
6.7 Spaces of constant curvature
There are generalizations of the Payne–Pólya–Weinberger inequality to spaces of con-

stant curvature. Ashbaugh and Benguria showed in [14] that Theorem 6.1 remains valid
if one replaces the Euclidean space Rn by a hemisphere of Sn and ‘ball’ by ‘geodesic
ball’. Similar to the Szegö–Weinberger inequality, it is still an open problem to prove
a Payne–Pólya–Weinberger result for the whole sphere. Although there seem to be
no counterexamples known that rule out such a generalization, the original scheme of
proving the PPW inequality is not likely to work. One reason is that numerical studies
show the function g to be not monotone on the whole sphere.
For the hyperbolic space, on the other hand, things are settled. Following the gen-
eral lines of the original proof, Benguria and Linde established in [24] a PPW type
inequality that holds in any space of constant negative curvature.
7 The fundamental gap
The gap between the first two Dirichlet eigenvalues of the Laplacian, λ2 − λ1 and its
dependence on the domain , in particular for compact, convex domains in Rn , has
been considered by several authors. Moreover, many people have considered the more
general problem for Schrödinger operators of the form
H = − + V (7.1)
on compact, convex domains in Rn , with Dirichlet boundary conditions and semicon-

vex potentials (i.e., potentials V such that V (x) + c|x|2 is convex for some c). Such
operators have an increasing sequence of eigenvalues
123
λ1 (; V ) < λ2 (; V ) ≤ λ3 (; V ) ≤ · · · → ∞
For a convex domain ⊂ Rn , and a convex potential, Singer, et al [92] proved the
following lower bounds on the gap λ2 − λ1 between the two lowest eigenvalues of the
Schrödinger operator H :
π2
≤ λ2 (; V ) − λ1 (; V )
4d 2
where d is the diameter of , i.e., d = supx,y∈ |x − y|. For the case V = 0, they
also proved the upper bound
4nπ 2
λ2 (; 0) − λ1 (; 0) ≤ ,
D2
where D is the diameter of the largest inscribed ball in .
It was observed by Michael van den Berg [30] that for many exactly solvable cases
involving convex domains, the gap λ2 − λ1 is bounded from below by 3π 2 /d 2 (which
should be asymptotically achieved by a domain in the shape of a cigar and a constant
potential V . The value 3π 2 /d 2 is the actual gap of the Dirichlet Laplacian on an inter-
val of length d in one dimension. This was also independently suggested by Ashbaugh
and Benguria [8] and S.-T. Yau [106]. That this bound actually holds for any bounded,
convex, domain in Rn and for any semiconvex potential was recently proven by Ben
Andrews and Julie Clutterbuck [1]. In fact, we have
Theorem 7.1 (Sharp Lower Bound on the Gap; Andrews and Clutterbuck [1]) Let
⊂ Rn be a bounded, convex, domain of diameter d, and V a semiconvex potential.
The the Dirichlet eigenvalues of the Schrödinger operator H, satisfy,
3π 2
λ2 (; V ) − λ1 (; V ) ≥ .
d2
Remarks i) In the one dimensional case, for the Schrödinger operator with a more
restrictive class of potentials, namely symmetric single well potentials (i.e., centering
the interval of length d around 0 and imposing V (−x) = V (x) and V (x) increasing
in [0, d/2]) this result was proven by Ashbaugh and Benguria [8]. Later M. Horvath
proved a similar result for single well potentials but removing the symmetry condition.
In the one dimensional case, for the Schrödinger operator, with a convex potential, this
theorem was proven by R. Lavine [66]. Concerning the situation in Rn , with n ≥ 2,
in 1986, Yu and Zhong [107] proved the estimate
π2
λ2 − λ1 ≥ .
d2
More recently, for V = 0, Davis [47], and Bañuelos and Kröger [21] proved van den
Berg’s conjecture for ⊂ R2 symmetric with respect to the x and y axes, and convex
in both x and y. The fundamental gap for the Neumann problem with V = 0 is just the
123
first non trivial Neumann eigenvalue. The sharp lower bound for the first nontrivial
Neumann eigenvalue on a convex domain was proven by Payne and Weinberger [85].
For a complete survey on the history of the Fundamental Gap problem up to 2006, see
[6].
ii) The proof of Andrews and Clutterbuck relies on two different ingredients, both
of which involve analysis of the heat equation. They first introduce the modulus of
continuity, η, of a function f : → R, defined on R+ , in such a way that,

|y − x|
| f (y) − f (x)| ≤ 2η ,
2
for all x, y ∈ . It turns out (see [1], and references there in) that if u is any smooth
solution of the heat equation with Neumann data on a convex domain ⊂ Rn of
diameter d, then u(·, t) has modulus of continuity ϕ(·, t), where ϕ is the solution to
the one-dimensional heat equation on [−d/2, d/2] with Neumann boundary data and
an initial condition given in terms of the initial condition of the original equation (see
[1] for details). Using this method Andrews and Clutterbuck immediately recover the
result of [85], which is not strong enough to prove the Fundamental Gap theorem.
Thus, they need a second ingredient which is based on a very beautiful generaliza-
tion of a known technique in this trade, which stems from the log-concavity of the
ground state of the Schrödinger Operator for convex potentials (a result that goes back
to Brascamp and Lieb [34]). Their generalization is based on the introduction of the
so-called modulus of concavity: assume that a one dimensional potential Ṽ is even in
the interval [−d/2, d/2], and assume that the potential V of the Schrödinger operator
defined in is more convex than Ṽ in the sense that for any y = x in one has

(y − x) |y − x|
(∇V (y) − ∇V (x)) · ≥ 2 Ṽ . (7.2)
|y − x| 2
(in their terms, Ṽ is thus a modulus of convexity for V ). Then, Andrews and Clutter-
buck proved the following improvement on the Brascamp and Lieb result:
Theorem 7.2 Let φ0 be the groundstate of the Schrödinger operator H . If V satisfies

(7.2) then, φ0 satisfies the following log-concavity estimate,
y−x
(∇ ln φ0 (y) − ∇ ln φ0 (x)) · ≤ 2(ln φ̃0 ) , (7.3)
|y − x| z=|y−x|/2
for every y = x in , where φ̃0 is the ground state of the one dimensional Schrödinger
operator −d 2 /d x 2 + Ṽ in the interval [−d/2, d/2] with Dirichlet boundary condition.
Using these two ingredients, Andrews and Clutterbuck prove the following:
Theorem 7.3 If V and Ṽ are related by (7.2), then the fundamental gap for the
Schrödinger operator H = − + V, on a bounded, convex domain is bounded
below by the fundamental gap of the one dimensional operator −d 2 /d x 2 + Ṽ on the
interval [−d/2, d/2].
123
In particular, if the potential V is convex, one can choose Ṽ = 0 and van den Berg’s
Conjecture follows.
Moreover, Andrews and Clutterbuck prove the following extension:
Theorem 7.4 If V is semiconvex, then, the fundamental gap satisfies the bound,
3π 2
λ2 − λ1 ≥ .
d2
For the proofs of these results, references and extensions see [1].
8 An isoperimetric inequality for ovals in the plane
The next problem we will consider is a conjectured isoperimetric inequality for closed,
smooth curves in the plane. It has attracted considerable attention in the literature
during the last decade (see, e.g., [37,48,49,56,57]), and it has many interesting con-
nections in geometry and physics. In particular, Benguria and Loss [26] have shown a
connection between this problem and a special case of the Lieb–Thirring inequalities
[68,71], inequalities which play a fundamental role in Lieb and Thirring’s proof of
the stability of matter (see, in particular, [70] and the review article [67]).
Denote by C a closed curve in the plane, of length 2π, with positive curvature κ,
and let
d2
H (C) ≡ − + κ2 (8.1)
ds 2
acting on L 2 (C) with periodic boundary conditions. Let λ1 (C) denote the lowest
eigenvalue of H (C). Certainly, λ1 (C) depends on the geometry of the curve C. It has
been conjectured that
λ1 (C) ≥ 1, (8.2)
with equality if and only if C belongs to a one-parameter family of ovals which include
the circle (in fact the one-parameter family of curves is characterized by a curvature
given by κ(s) = 1/(a 2 cos2 (s) + a −2 sin2 (s))[26]). It is a simple matter to see that if
C is a circle of length 2π, the lowest eigenvalue of H (C) is precisely 1. The fact that
there is degeneracy of the conjectured minimizers makes the problem much harder.
The conjecture (8.2) is still open. Concerning (nonoptimal) lower bounds, Benguria
and Loss proved [26]
λ1 (C) ≥ 1/2, (8.3)
and more recently Linde [72] found the best lower bound to date,
−2
π
λ1 (C) > 1 + ≈ 0.60847 (8.4)
π +8
123
Even though this is the best bound to date, it is still some distance from the conjectured
optimal value 1.
Burchard and Thomas [38] have shown that the ovals that give λ1 (C) = 1, alluded
to before, minimize λ1 (C) at least locally, i.e., there is no small variation around these
curves that reduces λ1 (C). This is an important indication that the conjecture (8.2) is
true, though, of course, it is not enough to prove it and thus it remains open.
In recent years several authors have obtained isoperimetric inequalities for the low-
est eigenvalues of a variant of H (C), and we give a short summary of the main results
in the sequel. Consider the Schrödinger operator
d2
Hg (C) ≡ − + gκ 2 (8.5)
ds 2
defined on L 2 (C) with periodic boundary conditions. As before, C denotes a closed

curve in R2 with positive curvature κ, and length 2π . Here, s denotes arclength. If
g < 0, the lowest eigenvalue of Hg (C), say λ1 (g, C), is uniquely maximized when
C is a circle [48]. When g = −1, the second eigenvalue, λ2 (−1, C), is uniquely
maximized when C is a circle [57]. If 0 < g ≤ 1/4, λ1 (g, C) is uniquely minimized
when C is a circle [49]. It is an open problem to determine the curve C that minimizes
λ1 (g, C) in the cases, 1/4 < g ≤ 1, and g < 0, g = −1. If g > 1 the circle is not a
minimizer for λ1 (g, C) (see, e.g., [49,56] for more details on the subject).
Remark Very recently, Bernstein and Breiner [31] have introduced a new twist on
this problem (i.e., on the conjectured isoperimetric inequality for ovals on the plane).
Bernstein and Breiner have considered the Catenoid
Cat = {x12 + x22 = cosh2 (x3 )} ⊂ R3 ,
which is a minimal surface of revolution introduced by Euler in 1744. They showed

how the conjectured inequality of Benguria and Loss could be used to prove that
subsets of the Catenoid minimize area within a geometric natural class of minimal
annuli.
9 Fourth order differential operator
In continuum mechanics, the vibrations of more rigid objects, like plates, rods, etc.,
are governed by wave equations involving higher order operators on the spatial vari-
ables. The normal modes of oscillations of these equations give rise to an eigenvalue
problem associated to fourth order operators. There are isoperimetric inequalities for
these eigenvalues, which are analogous to the ones that we have been discussing for
vibrating membranes. In this section we will briefly review three of these isoperimet-
ric inequalities, arising in connection to i) the vibrations of the clamped plate, ii) the
buckling problem, and iii) the vibrations of the free vibrating plate. In connection to
fourth order operators, there is also a vast literature (in particular in the last few years)
involving universal inequalities for the eigenvalues of these spectral problems. We
will not discuss these universal inequalities here.
123
9.1 The clamped plate
Consider a bounded, smooth domain ⊂ R2 . The eigenvalue problem that determines

the eigenfrequencies of a clamped plate is given by
2 u = u, in (9.1)
together with the clamped boundary conditions,
u = |∇u| = 0, in ∂ (9.2)
(the eigenfrequencies are proportional to the square root of the eigenvalues). The
boundary value problem (9.1), (9.2) has a countable sequence of eigenvalues
0 < 1 () ≤ 2 () ≤ . . .
and n () → ∞ as n → ∞. Because we are dealing with the operator 2 (and not
just the Laplacian) two nasty things may occur:
i) The principal eigenfunction u 1 of the boundary value problem defined by (9.1) and
(9.2) is not necessarily of one sign (say positive).
ii) The lowest eigenvalue 1 for the clamped plate may be degenerate.
There is an extensive literature on these two facts (see, in particular the Biblio-
graphical Remarks i), ii) and iii) at the end of this section). These two facts make
it impossible to use the standard techniques that we have discussed in previous sec-
tions in connection with the proof of isoperimetric inequalities for the Laplacian and
Schrödinger operators.
9.2 Rayleigh’s conjecture for the clamped plate
In the first edition of his book The Theory of Sound, Lord Rayleigh conjectured [88] that
1 () ≥ 1 (∗ ) (9.3)
where 1 () denotes the first eigenvalue for the vibrations of a clamped plate, and
∗ is a disk with the same area as .
In 1950, G. Szegö [95] (see also [96], and the Erratum [98]) proved Rayleigh’s con-
jecture for the clamped plate under the assumption that u 1 is of one sign (assumption
which we now know does not always hold).
A significant step towards the proof of Rayleigh’s conjecture was done by G. Tal-
enti [99] in 1981. Finally, in 1995, the conjecture was proven by N.S. Nadirashvili
[79]. The analog of Rayleigh’s conjecture for the clamped plate in three dimensions
was proven by M.S. Ashbaugh and R.D. Benguria [13] (see also, [17]). The analog of
Rayleigh’s conjecture for the clamped plate in dimensions larger than 3 (i.e., n ≥ 4)
is still an open problem. Although not sharp, the best results to date for n ≥ 4 have
been obtained by M. S. Ashbaugh and R. S. Laugesen, [18].
123
The proof of Rayleigh’s conjecture for the clamped plate, i.e., the proof of (9.3) is
based on several steps. The first step, as usual, is the variational characterization of
1 (). In the second step, taking into account that the ground state, say u, of (9.1)
and (9.2) is not necessarily positive, one defines the sets + = {x u + > 0} and
− = {x u − > 0}, where u + = max(u, 0) and u − = max(−u, 0) are the positive
and negative parts of u, respectively. The third step is to consider the positive and
negative parts of u, in other words we write u = (u)+ − (u)− . Then, one
considers the rearrangement,
g(s) = (u)∗+ (s) − (u)∗− (μ() − s),
where s = Cn |x|n , and Cn is the volume of the unit ball in n dimensions (here n = 2 or 3,
as we mentioned above), and μ() is the volume of . The next step is to consider
the solutions, v and w of some Dirichlet problem in the balls ∗+ and ∗− respectively.
Then one uses a comparison theorem of Talenti [99], namely u ∗+ ≤ v in ∗+ and
u ∗− ≤ w in ∗− , respectively. The functions v and w can be found explicitly in terms
of modified Bessel functions. The final step is to prove the necessary monotonicity
properties of these functions v, and w. We refer the reader to [13], for details.
9.3 Rayleigh’s conjecture for the buckling of a clamped plate
The buckling eigenvalues for the clamped plate for a domain in R2 , correspond to
the eigenvalues of the following boundary value problem,
− 2 u = u, in (9.4)
together with the clamped boundary conditions,
u = |∇u| = 0, in ∂ (9.5)
The lowest eigenvalue, 1 say, is related to the minimum uniform load applied in the
boundary of the plate necessary to buckle it. There is a conjecture of L. Payne [81],
regarding the isoperimetric behavior of 1 , namely,
1 () ≥ 1 (∗ ).
To prove this conjecture is still an open problem. For details see, e.g., [81,18], and
the review articles [82,3]. Recently, Antunes [2] has checked numerically Payne’s
conjecture for a large class of domains (mainly families of triangles or other simple
polygons). Also, in [2] Antunes has studied the validity of other eigenvalue inequalities
(mainly relating the 1 with different Dirichlet eigenvalues for the same domain .
9.4 The fundamental tones of free plates
The analog of the Szegö–Weinberger problem for the “free vibrating plate” has been
recently considered by L. Chasman, in her Ph.D. thesis [39] (see also [40,41]). As
discussed in [39], the fundamental tone for that problem, say ω1 (), corresponds to
the first nontrivial eigenvalue for the boundary value problem
123
u − τ u = ωu,
in a bounded region in the d dimensional Euclidean Space, with some natural bound-
ary conditions, where τ is a positive constant. In fact ω1 () is the fundamental tone of
a free vibrating plate with tension (physically τ represents the ratio of the lateral ten-
sion to the flexural rigidity). In [39,40], Chasman proves the following isoperimetric
inequality,
ω1 () ≤ ω1 (∗ ),
where ∗ is a ball of the same volume as (here, equality is attained if and only
if is a ball). This result is the natural generalization of the corresponding Szegö–
Weinberger result for the fundamental tone of the free vibrating membrane. Here, we
will not discuss the exact boundary conditions appropriate for this problem (we refer
the reader to [39] for details). In fact, the appropriate “free” boundary conditions are
essentially obtained as some transversality conditions of the Direct Calculus of Vari-
ations for this problem. In [39], Chasman first derives the equivalent of the classical
result of F. Pöckels for the membrane problem in this case, i.e., the existence of a
discrete sequence of positive eigenvalues accumulating at infinity. Then, she carefully
discusses the free boundary conditions both, for smooth domains, and for domains
with corners (in fact she considers as specific examples the rectangle and the ball,
and, moreover the analog one dimensional problem). Then, universal upper and lower
bounds are derived for ω1 in terms of τ . As for the proof of the isoperimetric inequality
for ω1 , she uses a similar path as the one used in the proof of the Szegö–Weinberger
inequality Since in the present situation the operator is more involved, this task is not
easy. She starts by carefully analyzing the necessary monotonicity properties of the
Bessel (and modified) Bessel functions that naturally appear in the solution for the
(d-degenerate) eigenfunctions corresponding to the fundamental tone, ω1 , of the ball.
Then, the Weinberger strategy takes us through the standard road: use the variational
characterization of ω1 () in terms of d different trial functions (given as usual as a
radial function ρ times the angular part xi /r, for i = 1, . . . , d) and averaging, to get
a rotational invariant, variational upper bound on ω1 . As usual, a Brower’s fixed point
theorem is needed to insure the orthogonality of these trial functions to the constants.
Then, one chooses the right expression for the variational function ρ guided by the
expressions of the eigenfunctions associated to the fundamental tone of the ball. As
in the proof of many of the previous isoperimetric inequalities, Chasman has to prove
monotonicity properties of ρ (chosen as above) and of the expressions involving ρ and
higher derivatives that appear in the bound obtained after the averaging procedure in
the previous section. Finally, rearrangements and symmetrization arguments are used
to conclude the proof of the isoperimetric result (see [39] for details).
i) There is a recent, very interesting article on the sign of the principal eigenfunc-
tion of the clamped plate by Guido Sweers, When is the first eigenfunction for the
clamped plate equation of fixed sign, USA–Chile Workshop on Nonlinear Analy-
sis, Electronic J. Diff. Eqns., Conf. 06, 2001, pp. 285–296, available on the web at
123
http://ejde.math.swt.edu/conf-proc/06/s3/sweers.pdf], where the author reviews the

status of this problem and the literature up to 2001
ii) For general properties of the spectral properties of fourth order operators the reader
may want to see: Mark P. Owen, Topics in the Spectral Theory of 4th order Elliptic
Differential Equations, Ph.D. Thesis, University of London, 1996. Available on the
Web at http://www.ma.hw.ac.uk/~mowen/research/thesis/thesis.ps
iii) Concerning the two problems mentioned in the introduction of Section 9, the
reader may want to check the following references: R. J. Duffin, On a question of Had-
amard concerning super–biharmonic functions, J. Math. Phys. 27, 253–258 (1949); R.
J. Duffin, D. H. Shaffer, On the modes of vibration of a ring–shaped plate, Bull. AMS
58, 652 (1952); C.V. Coffman, R. J. Duffin, D. H. Shaffer, The fundamental mode of
vibration of a clamped annular plate is not of one sign, in Constructive approaches
to mathematical models (Proc. Conf. in honor of R. Duffin, Pittsburgh, PA, 1978),
pp. 267–277, Academic Press, NY (1979); C.V. Coffman, R. J. Duffin, On the funda-
mental eigenfunctions of a clamped punctured disk, Adv. in Appl. Math. 13, 142–151
(1992).
Acknowledgments We would like to thank the anonymous referee for many helpful suggestions and
remarks. The work of RB and BL has been partially supported by Iniciativa Científica Milenio, ICM
(CHILE), project P07-027-F. The work of RB has also been supported by FONDECYT (Chile) Project
1100679.
Open Access This article is distributed under the terms of the Creative Commons Attribution License
which permits any use, distribution and reproduction in any medium, provided the original author(s) and
source are credited.
10 Appendix
10.1 The layer-cake formula
Theorem 10.1 Let ν be a measure on the Borel sets of R+ such that (t) := ν([0, t))
is finite for every t > 0. Let further (, , m) be a measure space and v a non-negative
measurable function on . Then
∞
(v(x))m( dx) = m({x ∈ : v(x) > t})ν( dt). (10.1)
0
In particular, if m is the Dirac measure at some point x ∈ Rn and ν( dt) = dt then

(10.1) takes the form
∞
v(x) = χ{y∈:v(y)>t} (x) dt. (10.2)
0
123
Proof Since m({x ∈ : v(x) > t}) = χ{v>t} (x)m( dx) we have, using Fubini’s
theorem,
∞ ∞
m({x ∈ : v(x) > t})ν( dt) = χ{v>t} (x)ν( dt) m( dx).
0 0
Theorem 10.1 follows from observing that

∞ v(x)
χ{v>t} (x)ν( dt) = ν( dt) = (v(x)).
0 0
10.2 A consequence of the Brouwer fixed-point theorem
Theorem 10.2 (Brouwer’s fixed-point theorem) Let B ⊂ Rn be the unit ball for
n ≥ 0. If f : B → B is continuous then f has a fixed point, i.e., there is some x ∈ B
such that f (x) = x.
The proof appears in many books on topology, e.g., in [78]. Brouwer’s theorem can
be applied to establish the following result:
Theorem 10.3 Let B ⊂ Rn (n ≥ 2) be a closed ball and b(r ) a continuous map from
B to Rn . If b points strictly outwards at every point of ∂ B, i.e., if b(r ) · r > 0 for every
r ∈ ∂ B, then b has a zero in B.
Proof Without losing generality we can assume that B is the unit ball centered at
the origin. Since b is continuous and b(r ) · r > 0 on ∂ B, there are two constants
0 < r0 < 1 and p > 0 such that b(r ) · r > p for every r with r0 < |r | ≤ 1. We show
that there is a constant c > 0 such that
| − cb(r ) + r | < 1
for all r ∈ B: In fact, for all r with |r| ≤ r0 the constant c can be any positive number
below (supr ∈B |b(r )|)−1 (1 − r0 ). The supremum exists because |b| is continuously
defined on a compact set and therefore bounded. On the other hand, for all r ∈ B with
|r | > r0 we have
| − cb(r ) + r |2 = c2 |b(r )|2 − 2cb(r ) · r + |r|2

≤ c2 sup |b|2 − 2cp + 1,
r ∈B
which is also smaller than one if one chooses c > 0 sufficiently small. Now set
g(r ) = −cb(r ) + r for r ∈ B.
Then g is a continuous mapping from B to B and by Theorem 10.2 it has some fixed
point r1 ∈ B, i.e., g(r1 ) = r1 and b(r1 ) = 0.
123
References
1. Andrews, B., Clutterbuck, J.: Proof of the fundamental gap conjecture. J. Am. Math. Soc. 24, 899–
916 (2011)
2. Antunes, P.R.S.: On the buckling eigenvalue Problem. J. Phys. A Math. Theor. 44, 215205(13) (2011)
3. Ashbaugh, M.S.: Isoperimetric and universal inequalities for eigenvalues. In: Davies B., Safarov Yu.
(eds.) Spectral Theory and Geometry (Edinburgh, 1998), E. London Math. Soc. Lecture Notes, vol.
273, pp. 95–139. Cambridge University Press, Cambridge (1999)
4. Ashbaugh, M.S.: Open problems on eigenvalues of the Laplacian. In: Rassias, Th.M., Srivastava, H.M.
(eds.) Analytic and Geometric Inequalities and Applications, Mathematics and Its Applications, 478,
pp. 13–28. Kluwer Academic Publishers, Dordrecht (1999) [see also M. S. Ashbaugh contribution
(pp. 2–10) to ESI-Workshop on Geometrical Aspects of Spectral Theory (Matrei, Austria, 1999), L.
Friedlander and T. HoffmannOstenhof, editors, 33 pp., available electronically at http://www.esi.ac.
at/ESIPreprints.html, preprint no. 768]
5. Ashbaugh M.S.: The universal eigenvalue bounds of Payne–Pólya–Weinberger, Hile–Protter, and
H. C. Yang, In: Spectral and Inverse Spectral Theory (Goa, 2000). Proc. Indian Acad. Sci. Math. Sci.,
vol. 112, 3–30 (2002)
6. Ashbaugh M.S.: The Fundamental Gap (2006) http://www.aimath.org/WWN/loweigenvalues/
7. Ashbaugh, M.S., Benguria, R.D.: Log-concavity of the ground state of Schrödinger operators: a new
proof of the Baumgartner–Grosse–Martin inequality. Phys. Lett. A 131, 273–276 (1988)
8. Ashbaugh, M.S., Benguria, R.D.: Optimal lower bound for the gap between the first two eigenvalues of
one-dimensional Schrödinger operators with symmetric single well potentials. Proc. Am. Math. Soc.
105, 419–424 (1989)
9. Ashbaugh, M.S., Benguria, R.D.: Proof of the Payne–Pólya–Weinberger conjecture. Bull. Am. Math.
Soc. 25, 19–29 (1991)
10. Ashbaugh, M.S., Benguria, R.D.: A sharp bound for the ratio of the first two eigenvalues of Dirichlet
Laplacians and extensions. Ann. Math. 135, 601–628 (1992)
11. Ashbaugh, M.S., Benguria, R.D.: A second proof of the Payne–Pólya–Weinberger conjecture. Com-
mun. Math. Phys. 147, 181–190 (1992)
12. Ashbaugh, M.S., Benguria, R.D.: Sharp upper bound to the first nonzero Neumann eigenvalue for
bounded domains in spaces of constant curvature. J. Lond. Math. Soc. 52(2), 402–416 (1995)
13. Ashbaugh, M.S., Benguria, R.D.: On Rayleigh’s conjecture for the clamped plate and its generaliza-
tion to three dimensions. Duke Math. J. 78, 1–17 (1995)
14. Ashbaugh, M.S., Benguria, R.D.: A sharp bound for the ratio of the first two Dirichlet eigenvalues
of a domain in a hemisphere of Sn . Trans. Am. Math. Soc. 353, 1055–1087 (2001)
15. Ashbaugh, M.S., Benguria, R.D.: Isoperimetric inequalities for eigenvalues of the Laplacian. In:
Gesztesy, F., Deift, P., Galvez, C., Perry, P., Schlag, W. (eds.) Spectral Theory and Mathematical
Physics: A Festschrift in Honor of Barry Simon’s 60th Birthday, Proceedings of Symposia in Pure
Mathematics, vol. 76, Part 1, pp. 105–139. Amer. Math. Soc., Providence, RI (2007)
16. Ashbaugh, M.S., Benguria, R.D.: El problema de la Reina Dido: Panorama sobre los problemas de
la isoperimetría, Joven Matemático, vol. 1, 3–8 (2010). [The original English version of this essay,
Dido and Isoperimetry, can be found on the web page of the International Conference on the isoperi-
metric problem of Queen Dido and its mathematical ramifications, that was held in Carthage, Tunisia,
May 24–29, 2010; see, http://math.arizona.edu/~dido/didon.html; a french translation, by Jacqueline
Fleckinger–Pellé, can be found in the same web page]
17. Ashbaugh, M.S., Benguria, R.D., Laugesen R.S.: Inequalities for the first eigenvalues of the clamped
plate and buckling problems. in General inequalities 7, (Oberwolfach, 1995), pp. 95–110. Internat.
Ser. Numer. Math., vol. 123. Birkhäuser, Basel (1997)
18. Ashbaugh, M.S., Laugesen, R.S.: Fundamental tones and buckling loads of clamped plates. Ann.
Scuola Norm. Sup. Pisa Cl. Sci. 23, 383–402 (1996)
19. Baernstein II, A.: A unified approach to symmetrization, In: Partial Differential Equations of
Elliptic Type, Cortona, 1992, Sympos. Math., vol. XXXV, pp. 47–91. Cambridge University Press,
Cambridge, UK (1994)
20. Bandle, C.: Isoperimetric inequalities and applications, Pitman monographs and studies in mathe-
matics, vol. 7. Pitman, Boston (1980)
21. Bañuelos, R., Kröger, P.: Gradient estimates for the ground state Schrödinger eigenfunction and
applications. Commun. Math. Phys. 224, 545–550 (2001)
123
22. Baumgartner, B., Grosse, H., Martin, A.: The Laplacian of the potential and the order of energy
levels. Phys. Lett. B 146, 363–366 (1984)
23. Benguria, R.D., Linde, H.: A second eigenvalue bound for the Dirichlet Schrödinger operator. Com-
mun. Math. Phys. 267, 741–755 (2006)
24. Benguria, R.D., Linde, H.: A second eigenvalue bound for the Dirichlet Laplacian in hyperbolic
space. Duke Math. J. 140, 245–279 (2007)
25. Benguria, R.D., Linde, H.: Isoperimetric inequalities for eigenvalues of the Laplace operator. In:
Villegas-Blas, C. (ed.) Fourth summer school in analysis and mathematical physics: topics in spectral
theory and quantum mechanics. Contemporary Mathematics (AMS), vol. 476, pp. 1–40 (2008)
26. Benguria, R.D., Loss, M.: Connection between the Lieb–Thirring conjecture for Schrödinger opera-
tors and an isoperimetric problem for ovals on the plane. In: Conca, C., Manásevich, R., Uhlmann,
G., Vogelius, M.S. (eds.) Partial Differential Equations and Inverse Problems. Contemp. Math., vol.
362, pp. 53–61. Amer. Math. Soc., Providence, R.I. (2004)
27. Bérard, P.: Transplantation et isospectralité I. Math. Ann. 292, 547–559 (1992)
28. Bérard, P.: Transplantation et isospectralité II. J. Lond. Math. Soc. 48, 565–576 (1993)
29. Bérard, P.: Domaines plans isospectraux a la Gordon–Web–Wolpert: une preuve elementaire. Afr.
Math. 1, 135–146 (1993)
30. van den Berg, M.: On condensation in the free-boson gas and the spectrum of the Laplacian. J. Stat.
Phys. 31, 623–637 (1983)
31. Bernstein, J., Breiner, C.: A variational characterization of the catenoid. (2010) (preprint)
32. Birman, M.S., Solomjak, M.Z.: Spectral theory of self-adjoint operators in Hilbert Space. D. Reidel
Publishing Company, Dordrecht (1987)
33. Borell, C.: The Brunn-Minkowski inequality in Gauss space. Invent. Math. 30, 207–211 (1975)
34. Brascamp, J., Lieb, E.H.: On extensions of the Brunn–Minkowski and Prékopa–Leindler theorems,
including inequalities for log concave functions, and with an application to the diffusion equation. J.
Funct. Anal. 22, 366–389 (1976)
35. Brock, F., Solynin, A.Y.: An approach to symmetrization via polarization. Trans. Am. Math. Soc.
352, 1759–1796 (2000)
36. Brooks, R.: Constructing isospectral manifolds. Am. Math. Mon. 95, 823–839 (1988)
37. Burchard, A., Thomas, L.E.: On the Cauchy problem for a dynamical Euler’s elastica. Commun.
Partial Differ. Equ. 28, 271–300 (2003)
38. Burchard, A., Thomas, L.E.: On an isoperimetric inequality for a Schrödinger operator depending on
the curvature of a loop. J. Geom. Anal. 15, 543–563 (2005)
39. Chasman, L.M.: Isoperimetric problem for eigenvalues of free plates. Ph. D thesis, University of
Illinois at Urbana–Champaign (2009)
40. Chasman, L.M.: An isoperimetric inequality for fundamental tones of free plates. Commun. Math.
Phys. 303, 421–429 (2011)
41. Chasman, L.M.: Vibrational modes of circular free plates under tension. Appl. Anal. 90, 1877–
1895 (2011)
42. Chavel, I.: Eigenvalues in Riemannian geometry. Academic Press, NY (1984)
43. Courant, R., Hilbert, D.: Methods of Mathematical Physics, vol. 1. Interscience Publishers, New York
(1953)
44. Daners, D.: A Faber–Krahn inequality for Robin problems in any space dimension. Math. Ann.
335, 767–785 (2006)
45. Davies, E.B.: Heat kernels and spectral theory, paperback edn. Cambridge University Press,
Cambridge (1990)
46. Davies, E.B.: Spectral theory and differential operators. Cambridge University Press, Cambridge
(1996)
47. Davis, B.: On the spectral gap for fixed membranes. Ark. Mat. 39, 65–74 (2001)
48. Duclos, P., Exner, P.: Curvature-induced bound states in quantum waveguides in two and three dimen-
sions. Rev. Math. Phys. 7, 73–102 (1995)
49. Exner, P., Harrell, E.M., Loss, M.: Optimal eigenvalues for some Laplacians and Schrödinger oper-
ators depending on curvature. In: Dittrich, J., Exner, P., Tater, M. (eds.) Mathematical Results in
Quantum Mechanics (Prague, 1998). Oper. Theory Adv. Appl., vol. 108, pp. 47–58 (1999)
50. Faber, G.: Beweis, dass unter allen homogenen Membranen von gleicher Fläche und gleicher
Spannung die kreisförmige den tiefsten Grundton gibt. In: Sitzungberichte der mathematisch-
physikalischen Klasse der Bayerischen Akademie der Wissenschaften zu München Jahrgang,
pp. 169–172 (1923)
123
51. Federer, H.: Geometric Measure Theory. Springer Verlag, New York (1969)
52. Geisinger, L., Weidl, T.: Universal bounds for traces of the Dirichlet Laplace operator. J. Lond. Math.
Soc. 82, 395–419 (2010)
53. Giraud, O., Thas, K.: Hearing shapes of drums: Mathematical and physical aspects of isospectrali-
ty. Rev. Mod. Phys. 82, 2213–2255 (2010)
54. Gordon, C., Webb, D., Wolpert, S.: Isospectral plane domains and surfaces via Riemannian orbi-
folds. Invent. Math. 110, 1–22 (1992)
55. Hardy, G.H., Littlewood, J.E., Pólya, G.: Inequalities. Cambridge Univ. Press, Cambridge, UK (1964)
56. Harrell, E.M.: Gap estimates for Schrödinger operators depending on curvature talk delivered at the
2002 UAB International Conference on Differential Equations and Mathematical Physics. Available
electronically at http://www.math.gatech.edu/~harrell/
57. Harrell, E.M., Loss, M.: On the Laplace operator penalized by mean curvature. Commun. Math.
Phys. 195, 643–650 (1998)
58. Harrell, E.M., Hermi, L.: Differential inequalities for Riesz means and Weyl type bounds for eigen-
values. J. Funct. Anal. 254, 3173–3191 (2008)
59. Heath, T.L.: A History of Greek Mathematics. vol. 2. The Clarendon Press, Oxford (1921)
60. Henrot, A.: Extremum Problems for Eigenvalues of Elliptic Operators. Collection Frontiers in Math-
ematics, Birkhauser (2006)
61. Kac M.: On some connections between probability theory and differential and integral equations. In:
Neyman, J. (ed.) Proceedings of the Second Berkeley Symposium on Mathematical Statistics and
Probability. University of California Press, Berkeley, CA, pp. 189–215 (1951)
62. Kac, M.: Can one hear the shape of a drum?. Am. Math. Mon. 73, 1–23 (1966)
63. Kornhauser, E.T., Stakgold, I.: A variational theorem for ∇ 2 u + λu = 0 and its applications. J. Math.
Phys. 31, 45–54 (1952)
64. Krahn, E.: Über eine von Rayleigh formulierte Minimaleigenschaft des Kreises. Math. Ann. 94, 97–
100 (1925)
65. Krahn, E. :Über Minimaleigenschaften der Kugel in drei und mehr Dimensionen, Acta Comm. Univ.
Tartu (Dorpat), vol. A9, 1–44 (1926) [English translation: Minimal properties of the sphere in three
and more dimensions. In: Edgar Krahn 1894–1961: A Centenary Volume, Ü. Lumiste and J. Peetre,
editors, IOS Press, Amsterdam, The Netherlands, pp. 139–174 (1994)]
66. Lavine, R.: The eignevalue gap for one-dimensional convex potentials. Proc. Am. Math. Soc.
121, 815–821 (1994)
67. Lieb, E.H.: The stability of matter Rev. Mod. Phys. 48, 553–569 (1976)
68. Lieb, E.H.: Lieb–Thirring inequalities. In: Encyclopaedia of Mathematics, Suppl. II, Kluwer, Dordr-
echt, pp. 311–312 (2000)
69. Lieb, E.H., Loss, M.: Analysis, Graduate Studies in Mathematics, vol. 14. American Mathematical
Society, Providence, RI (1997)
70. Lieb, E.H., Thirring, W.: Bounds for the kinetic energy of fermions which proves the stability of
matter. Phys. Rev. Lett. 35, 687–689 (1975) (Errata: PRL 35, 1116 (1975))
71. Lieb, E.H., Thirring, W.: Inequalities for the moments of the eigenvalues of the Schrödinger hamilto-
nian and their relation to Sobolev inequalities. In: Lieb, E.H., Simon, B., Wightman, A.S. Studies in
Mathematical Physics, Essays in Honor of Valentine Bargmann, Princeton University Press, Prince-
ton (1986)
72. Linde, H.: A lower bound for the ground state energy of a Schrödinger operator on a loop. Proc. Am.
Math. Soc. 134, 3629–3635 (2006)
73. McHale, K.P.: Eigenvalues of the Laplacian, “Can you Hear the Shape of a Drum?”, Master’s Project.
Mathematics Department University of Missouri, Columbia, MO (1994)
74. McKean, H.P., Singer, I.M.: Curvature and the eigenvalues of the Laplacian. J. Differ. Geom. 1, 662–
670 (1967)
75. Maz’ja V.G. : Sobolev spaces, Springer Series in Soviet Mathematics. Springer-Verlag, Berlin (1985)
(Translated from the Russian by T.O. Shaposhnikova)
76. Melrose, R.B.: The inverse spectral problem for planar domains. In: Proceedings of the Centre for
Mathematics and its Applications, Australian National University, vol. 34 (1996)
77. Milnor, J.: Eigenvalues of the Laplace operator on certain manifolds. Proc. Nat. Acad. Sci.
51, 542 (1964)
78. Munkres, J.R.: Topology, A first course. Englewood Cliffs, Prentice-Hall (1975)
79. Nadirashvili, N.: Rayleigh’s conjecture on the principal frequency of the clamped plate. Arch. Ration.
Mech. Anal. 129, 1–10 (1995)
123
80. Osserman, R.: Isoperimetric inequalities and eigenvalues of the Laplacian. In: Proceedings of the
International Congress of Mathematicians (Helsinki, 1978), pp. 435–442, Acad. Sci. Fennica, Hel-
sinki (1980)
81. Payne, L.E.: A note on inequalities for plate eigenvalues. J. Math. Phys. 39, 155–159 (1960/1961)
82. Payne, L.E.: Isoperimetric inequalities and their applications. SIAM Rev. 9, 453–488 (1967)
83. Payne, L.E., Pólya, G., Weinberger, H.F.: Sur le quotient de deux fréquences propres consécu-
tives. Comptes Rendus Acad. Sci. Paris 241, 917–919 (1955)
84. Payne, L.E., Pólya, G., Weinberger, H.F.: On the ratio of consecutive eigenvalues. J. Math.
Phys. 35, 289–298 (1956)
85. Payne, L.E., Weinberger, H.F.: An optimal Poincaré inequality for convex domains. Arch. Ration.
Mech. Anal. 5, 286–292 (1960)
86. Pleijel, Å.: A study of certain Green’s functions with applications in the theory of vibrating mem-
branes. Ark. Math. 2, 553–569 (1954)
87. Pólya, G., Szegö, G.: Isoperimetric Inequalities in Mathematical Physics. Princeton University
Press, Princeton, NJ (1951)
88. Rayleigh, J.W.S.: The Theory of Sound, 2nd edn. revised and enlarged (in 2 vols.), Dover Publications,
New York (1945) (republication of the 1894/1896 edition)
89. Reed, M., Simon, B.: Methods of Modern Mathematical Physics, vol. 3. Academic Press, NY (1979)
90. Reed, M., Simon, B.: Methods of Modern Mathematical Physics Vol. 4. Analysis of Operators. Aca-
demic Press, NY (1978)
91. Riemann, B.: Über die Anzahl der Primzahlen unter einer gegebenen Grösse, Monatsberichte der
Berliner Akademie, pp. 671–680 (1859)
92. Singer, I.M., Wong, B., Yau, S.-T., Yau, S.S.-T.: An estimate of the gap of the first two eigenvalues
in the Schrödinger operator. Ann. Scuola Norm. Sup. Pisa Cl. Sci. 12(4 ), 319–333 (1985)
93. Sridhar, S., Kudrolli, A.: Experiments on not “Hearing the Shape” of drums. Phys. Rev. Lett. 72, 2175–
2178 (1994)
94. Sunada, T.: Riemannian coverings and isospectral manifolds. Ann. Math. 121, 169–186 (1985)
95. Szegö, G.: On membranes and plates. Proc. Nat. Acad. Sci. USA 36, 210–216 (1950)
96. Szegö, G. : On the vibrations of a clamped plate. In: Atti del Quarto Congresso dell’Unione Matem-
atica Italiana, Taormina, 1951, vol. II, pp. 573–577. Casa Editrice Perrella, Roma (1953)
97. Szegö, G.: Inequalities for certain eigenvalues of a membrane of given area. J. Ration. Mech.
Anal. 3, 343–356 (1954)
98. Szegö, G.: Note to my paper “On membranes and plates”. Proc. Nat. Acad. Sci. USA 44, 314–
316 (1958)
99. Talenti, G.: On the first eigenvalue of the clamped plate. Ann. Mat. Pura Appl. 129(4), 265–280 (1981)
100. Talenti, G.: Elliptic equations and rearrangements. Ann. Scuola Norm. Sup. Pisa 3(4), 697–718 (1976)
101. Thomas, I.: Greek Mathematics. In: Newman J.R. (ed.) The World of Mathematics, vol. 1,
pp. 189–209, Dover, NY (2000)
102. Vergil: The Aeneid, [English translation by Sarah Ruden], Yale University Press, New Haven, CT
(2008)
103. Weinberger, H.F.: An isoperimetric inequality for the n-dimensional free membrane problem. J.
Ration. Mech. Anal. 5, 633–636 (1956)
104. Weyl, H.: Über die asymptotische Verteilung der Eigenwerte. Nachr. Akad. Wiss. Göttingen Math.
Phys., Kl. II, 110–117 (1911)
105. Weyl, H.: Ramifications, old and new, of the eigenvalue problem. Bull. Am. Math. Soc. 56, 115–
139 (1950)
106. Yau, S.-T.: Nonlinear analysis in geometry, Monographies de LEnseignement Mathématique, vol. 33.
Série des Conférences de lUnion Mathématique Internationale, vol. 8, Geneva (1986)
107. Yu, Q.-H., Zhong, J.-Q.: Lower bounds of the gap between the first and second eigenvalues of the
Schrödinger operator. Trans. Am. Math. Soc. 294, 341–349 (1986)
108. Zelditch, S.: Spectral determination of analytic bi-axisymmetric plane domains. Geom. Funct. Anal.
10, 628–677 (2000)
109. Zelditch, S.: Inverse spectral problems for analytic domains II: Z2 -symmetric domains. Ann. Math.
170, 205–269 (2009)
123

Isoperimetric Inequalities For Eigenvalues of The Laplacian and The Schrödinger Operator

Uploaded by

Copyright:

Available Formats

Isoperimetric Inequalities For Eigenvalues of The Laplacian and The Schrödinger Operator

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Isoperimetric Inequalities For Eigenvalues of The Laplacian and The Schrödinger Operator

Uploaded by

Copyright:

Available Formats

Bull. Math. Sci.

Isoperimetric inequalities for eigenvalues

Rafael D. Benguria · Helmut Linde ·

Received: 18 April 2011 / Revised: 18 October 2011 / Accepted: 20 November 2011 /

Abstract The purpose of this manuscript is to present a series of lecture notes on

Keywords Eigenvalues of the Laplacian · Isoperimetric inequalities

Mathematics Subject Classification (1991) Primary 35P15; Secondary 35J05 ·

R. D. Benguria (B) · B. Loewe

Isoperimetric inequalities have played an important role in mathematics since the

relating the area A enclosed by a planar closed curve of perimeter L. In mathematics,

2 Can one hear the shape of a drum?

…but it would baffle the most skillful

In 1965, the Committee on Educational Media of the Mathematical Association of

Clearly, (x) = ϕx (0). Moreover, the function ϕx (y) is periodic in y of period 1.

where, the Fourier coefficients are

In the nth summand we make the change of variables y → u = n + y. Clearly, u runs

i.e., ak is the Fourier transform of a Gaussian. Thus, we finally obtain,

Since, (x) = ϕx (0), from (2.7) and (2.12) we finally get,

and using (2.15), and the definition of (x), we get,

ar t → 0, where ∂ P0 /∂t = P0 , and P0 ( p r ; t) → δ(r − p) as t → 0.P0 in fact

where |r − p|2 is just the Euclidean distance between p and r . Then, as t → 0+ ,

Thus, when set p = r we get

This bound is sharp, in the sense that as t → 0,

where I = mina∈Rd |x − a|2 d x and Md is a constant depending on dimension.

where M d = [(d + 2)π/d](d/2 + 1)−2/d Md (in particular M 2 = π/16. In gen-

2.1 One cannot hear the shape of a drum

2.2 Proof of isospectrality using transplantation

Fig. 1 GWW Isospectral

Fig. 2 Sunada Graphs

ϕ2A = ϕ7A , (2.29)

ϕ5T = −bψ1T + aψ2T + bψ3T − aψ4T + aψ5T − aψ6T + bψ7T . (2.31)

Although the proof by transplantation is straightforward to follow, it does not shed

2.3 Bibliographical remarks

in R3 , with u > 0 going to zero at infinity. The solution is given by

3.1 Definition and basic properties

For many problems of functional analysis it is useful to replace some function by an

Definition 3.1 The function

μ(t) = |{x ∈ : |u(x)| > t}|, t ≥ 0

is called distribution function of u.

From this definition it is straightforward to check that μ(t) is a decreasing (non-

Because μ is a decreasing function, Definition 3.2 implies that u  is an essentially

{s ≥ 0 : u  (s) > t} = [0, μ(t)). (3.1)

min{w ≥ 0 : μ(w) ≤ s} > t.

u  (s) = min{w ≥ 0 : μ(w) ≤ s} ≥ min{w ≥ 0 : μ(w) < μ(t)} > t,

It is straightforward to transfer the statements of Lemma 3.3 to the symmetric

In his lecture notes on rearrangements (see the reference in the Bibliographical

This function can as well be used to illustrate the theorems below.

3.2 Main theorems

Rearrangements are a useful tool of functional analysis because they considerably

Theorem 3.6 Let  be a continuous increasing map from R+ to R+ with (0) = 0.

Theorem 3.7 Assume that u : → R+ is given by u (x) = u(|x|), where u :

The product of two functions changes in a controllable way under rearrangement:

Now assume that u and v are measurable, non-negative and—without loosing

3.3 Gradient estimates

Proof On the one hand, by the classical isoperimetric inequality we have

On the other hand, we can use the Cauchy–Schwarz inequality to get

Comparing the equations (3.7) and (3.8) yields Theorem 3.9.

Theorem 3.10 Let  : R+ → R+ be a Young function, i.e.,  is increasing and

Proof Theorem 3.10 is a consequence of the following chain of (in)equalities, the

Clearly, (x) = ϕx (0). Moreover, the function ϕx (y) is periodic in y of period 1.

Since, (x) = ϕx (0), from (2.7) and (2.12) we finally get,

and using (2.15), and the definition of (x), we get,

ar t → 0, where ∂ P0 /∂t = P0 , and P0 ( p r ; t) → δ(r − p) as t → 0.P0 in fact

where M d = [(d + 2)π/d](d/2 + 1)−2/d Md (in particular M 2 = π/16. In gen-

Because μ is a decreasing function, Definition 3.2 implies that u is an essentially

{s ≥ 0 : u (s) > t} = [0, μ(t)). (3.1)

u (s) = min{w ≥ 0 : μ(w) ≤ s} ≥ min{w ≥ 0 : μ(w) < μ(t)} > t,

Theorem 3.6 Let be a continuous increasing map from R+ to R+ with (0) = 0.

Theorem 3.10 Let : R+ → R+ be a Young function, i.e., is increasing and

|{x ∈ Rn : u ∗ (s) ≥ |u(x)| > u ∗ (s + )}| = (3.11)

Taking the limit ↓ 0, this yields

with equality only if = up to a rotation.

μ1 () ≤ μ1 ( ), (5.1)

B = −2μ1 ( )gg − (n − 1)(rg − g)2 r −3 < 0.

λ2 () λ2 (B) λ2 ( )

We note that P j u 1 is a second eigenfunction of − D if is a ball which is centered

z 1 (r ) ≥ u 1 (r ) for r ∈ (0, r0 ) and

Q 1+ = Q 1 + n(λ2 − (1 − 2/n)λ1 ) + O( 2 ).