Content-Type: multipart/mixed; boundary="-------------9904161716868" This is a multi-part message in MIME format. ---------------9904161716868 Content-Type: text/plain; name="99-120.keywords" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="99-120.keywords" Schr\"{o}dinger operator, periodic potential, embedded eigenvalues ---------------9904161716868 Content-Type: application/x-tex; name="kuch_vai.tex" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="kuch_vai.tex" %% This document created by Scientific Word (R) Version 2.0 \documentclass{article} \usepackage{amsfonts} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%\usepackage{sw20jart} %TCIDATA{TCIstyle=article/art4.lat,jart,sw20jart} %TCIDATA{Created=Thu Jan 09 20:09:52 1997} %TCIDATA{LastRevised=Fri Apr 16 16:45:11 1999} %TCIDATA{Language=American English} \newtheorem{theorem}{Theorem} \newtheorem{lemma}[theorem]{Lemma} \newtheorem{proposition}[theorem]{Proposition} \newtheorem{conjecture}[theorem]{Conjecture} \newtheorem{example}[theorem]{Example} \newtheorem{definition}[theorem]{Definition} \newtheorem{corollary}[theorem]{Corollary} \newtheorem{exercise}{Exercise} \newtheorem{problem}{Problem} \newtheorem{remark}{Remark} \newtheorem{condition}[theorem]{Condition} \def\stackunder#1#2{\mathrel{\mathop{#2}\limits_{#1}}} %%\def\bf{\@subst@obsolete{amsfonts}\bf\mathbb} \begin{document} \title{ON ABSENCE OF EMBEDDED EIGENVALUES FOR SCHR\~{O}DINGER OPERATORS WITH PERTURBED PERIODIC POTENTIALS} \author{Peter Kuchment \\ %EndAName Department of Mathematics and Statistics\\ Wichita State University\\ Wichita, KS 67260-0033\\ kuchment@twsuvm.uc.twsu.edu\\ http://www.math.twsu.edu/Faculty/Kuchment/ \and Boris Vainberg \\ %EndAName Mathematics Department\\ University of North Carolina\\ Charlotte, NC 28223\\ brvainbe@uncc.edu} \date{} \maketitle \begin{abstract} The problem of absence of eigenvalues imbedded into the continuous spectrum is considered for a Schr\"{o}dinger operator with a periodic potential perturbed by a sufficiently fast decaying ``impurity'' potential. Results of this type have previously been known for the one-dimensional case only. Absence of embedded eigenvalues is shown in dimensions two and three if the corresponding Fermi surface is irreducible modulo natural symmetries. It is conjectured that all periodic potentials satisfy this condition. Separable periodic potentials satisfy it, and hence in dimensions two and three Schr\"{o}dinger operator with a separable periodic potential perturbed by a sufficiently fast decaying ``impurity'' potential has no embedded eigenvalues. \textbf{1991 Subject Classification: }35P99, 47A55, 47F05 \textbf{Keywords}: Schr\"{o}dinger operator, periodic potential, embedded eigenvalues \end{abstract} \section{Introduction} Consider the stationary Schr\"{o}dinger operator \begin{equation} H_0=-\Delta +q(x) \label{H0} \end{equation} in $L^2(\bf{R}^n)$ ($n=2$ or $3$), where the real potential $q(x)\in L^\infty (\bf{R}^n)$ is periodic with respect to the integer lattice $\bf{Z% }^n$: $q(x+l)=q(x)$ for all $l\in \bf{Z}^n$, $x\in \bf{R}^n$. The spectrum of this operator is absolutely continuous and has the well known band-gap structure (see \cite{E}, \cite{Gelf}, \cite{K}, \cite{OK}, \cite{RS}, \cite {T}, \cite{W}): \[ \sigma (H_0)=\stackunder{i\geq 1}{\cup }[a_i,b_i], \] where $a_i0 . \] The Fermi variety $F_{\,\lambda }(q)$ is the set of all zeros of a non-zero entire function of order $n$ on $\bf{C}^n$. \end{lemma} Lemma \ref{analyt} implies in particular that both Bloch and Fermi varieties are examples of what is called in complex analysis \textbf{analytic sets} (see for instance \cite{Ch}, \cite{GR}, and \cite{N}). Moreover, these are \textbf{principal} analytic sets in the sense that they are sets of all zeros of single analytic functions, while general analytic sets might require several analytic equations for their (local) description. Analyticity of these varieties (without estimates on the grows of the defining function) was obtained in \cite{W}. \begin{definition} An analytic set $A\subset \bf{C}^m$ is said to be \textbf{irreducible}, if it cannot be represented as the union of two proper analytic subsets. \end{definition} Irreducibility of the zero set of an analytic function can be understood as absence of non-trivial factorizations of this function (i.e., of a factorization into analytic factors that have smaller zero sets). \begin{definition} A point of an analytic set $A\subset \bf{C}^m$ is said to be \textbf{regular% }, if in a neighborhood of this point the set $A$ can be represented as an analytic submanifold of $\bf{C}^m$. The set of all regular points of $A$ is denoted by $regA$. \end{definition} We collect in the following lemma several basic facts about analytic sets that we will need later. The reader can find them in many books on several complex variables. In particular, all these statements are proven in sections 2.3, 5.3, 5.4, and 5.5 of Chapter 1 of \cite{Ch}. \begin{lemma} \label{property}Let $A$ be an analytic set. a) The set $regA$ is dense in $A$. Its complement in $A$ is closed and nowhere dense in $A$. b) The set $A$ can be represented as a (maybe infinite) locally finite union of irreducible subsets \[ A=\stackunder{i}{\cup }A_i \] called its \textbf{irreducible components}. c) Irreducible components are closures of connected components of $regA$. In particular, set $A$ is irreducible if and only if $regA$ is connected. d) Let $A$ be irreducible and $A_1$ be another analytic set such that $A\cap A_1$ contains a non-empty open portion of $A$. Then $A\subset A_1$. In particular, if $f$ is an analytic function that vanishes on an open portion of $A$, then $f$ vanishes on $A$. e) Any analytic set $A$ has a stratification $A=\cup A_j$ into disjoint complex analytic manifolds (strata) $A_j$ such that the union $\cup A_j$ is locally finite, the closure $\overline{A_j}$ of each $A_j$ and its boundary $% \overline{A_j}\backslash A_j$ are analytic subsets, and such that if the intersection $A_j\cap \overline{A_k}$ of two different strata is not empty, then $A_j\subset \overline{A_k}$ and $dimA_j4/3 \label{estimate} \end{equation} almost everywhere in $\bf{R}^n$, then the spectrum of $H$ contains no embedded eigenvalues. In other words, \[ \left\{ \lambda _j\right\} \cap \stackunder{i\geq 1}{\cup }% (a_i,b_i)=\emptyset , \] where $\left\{ \lambda _j\right\} $ is the impurity point spectrum of $H$, and \[ \stackunder{i\geq 1}{\cup }[a_i,b_i]=\sigma (H_0) \] is the band structure of the essential spectrum of $H$. \end{theorem} \textbf{Proof}. Let us assume that there exists a $\lambda $ that belongs to some $(a_i,b_i)$ and to the point spectrum of $H$ simultaneously. Then there exists a non-zero function $u(x)\in L_2(\bf{R}^n)$ (an eigenfunction) such that \[ -\Delta u+qu+vu=\lambda u, \] or \[ (H_0-\lambda )u=-vu. \] Let us denote the function in the right hand side by $\psi (x):$% \[ \psi (x)=-v(x)u(x). \] Consider the fundamental domain $\mathcal{K}$ of the group $\bf{Z}^n$ of periods: \[ \mathcal{K}=\{(x_1,...,x_n)\in R^n|\;0\leq x_i\leq 1,\;i=1,...,n\}. \] Then (\ref{estimate}) implies that the function $\psi (x)$ satisfies the estimate \begin{equation} \left| \left| \psi \right| \right| _{L^2(\mathcal{K}+l)}\leq Ce^{-\left| l\right| ^r} \label{decay} \end{equation} for all $l\in \bf{Z}^n$, where $r>4/3$. Consider the following Floquet transform of functions defined on $\bf{R}^n$ (see section 2.2 in \cite{K} about properties of this transform): \[ \mathcal{F}:\;f(x)\longrightarrow \widehat{f}(k,x)=\sum\limits_{l\in \bf{Z}% ^n}f(x-l)e^{-ik\cdot (x-l)}. \] The result is a function, which is $\bf{Z}^n$-periodic with respect to $x$. We can consider the resulting function as a function of $k$ with values in a space of functions on the $n$-dimensional torus $\bf{T}^n=\bf{R}^n/\bf{Z}% ^n$. If we apply this transform to the function $\psi (x)$, then we get a function of $k\in \bf{C}^n$ \begin{equation} \phi (k)=\widehat{\psi }(k,x) \label{funct} \end{equation} with values in the space $L^2(\bf{T}^n)$. \begin{lemma} \label{order}Function $\phi (k)$ is an entire function on $\bf{C}^n$ of the order $s=r/(r-1)<4$. \end{lemma} \textbf{Proof of the lemma}. From (\ref{decay}) we get \[ \left| \left| \psi (x-l)e^{-ik\cdot (x-l)}\right| \right| _{L^2(\mathcal{K}% )}\leq Ce^{C\left| k\right| +Im(k\cdot l)-\left| l\right| ^r}. \] Then \[ \left| \left| \phi (k)\right| \right| _{L^2(\mathcal{K})}\leq Ce^{C\left| k\right| }\stackunder{l\in \bf{Z}^n}{\sum }e^{-0.5\left| l\right| ^r}e^{% Im(k\cdot l)-0.5\left| l\right| ^r}. \] A simple Legendre transform type estimate (finding extremal values of the exponent) shows that \[ e^{Im(k\cdot l)-0.5\left| l\right| ^r}\leq Ce^{C\left| k\right| ^{r/(r-1)},} \] which proves that the function $\phi (k)$ is an entire function of the order $s=r/(r-1)$ in $\bf{C}^n$. The lemma is proven. As it is well known (see e.g. Chapter 2 of \cite{K} or XIII in \cite{RS}), the transform $\mathcal{F}$ leads to the following operator equation with a parameter $k$: \begin{equation} \left( H_0(k)-\lambda \right) \widehat{u}(k)=\phi (k), \label{operator} \end{equation} where \[ \widehat{u}(k)=\left( \mathcal{F}u\right) (k,x) \] is defined for $k\in \bf{R}^n$ and \[ H_0(k)=(i\nabla -k)^2+q(x) \] is considered as an operator on the torus $\bf{T}^n$. According to Theorem 2.2.5 in \cite{K}, \[ \widehat{u}(k)\in L_{loc}^2(\bf{R}^n,L_2(\bf{T}^n)). \] In fact, standard interior elliptic estimates show that \[ \widehat{u}(k)\in L_{loc}^2(\bf{R}^n,H^2(\bf{T}^n)), \] where $H^2(\bf{T}^n)$ is the Sobolev space of order two on $\bf{T}^n$. Theorem 3.1.5 of \cite{K} implies that the operator \[ \left( H_0(k)-\lambda \right) :H^2(\bf{T}^n)\rightarrow L_2(\bf{T}^n) \] is invertible if and only if $k\notin F_{\bf{R},\lambda }(q)$. As it is shown in the proof of Theorem 3.3.1 and in Lemma 1.2.21 of \cite{K} (see also comments in Section 3.4.D on reducing requirements on the potential, in particular Theorem 3.4.2), the inverse operator can be represented as follows: \begin{equation} \left( H_0(k)-\lambda \right) ^{-1}=B(k)/\zeta (k), \label{represent} \end{equation} where $B(k)$ is a bounded operator from $L_2(\bf{T}^n)$ into $H^2(\bf{T}% ^n) $ and $B(k)$ and $\zeta (k)$ are correspondingly an operator and a scalar entire functions of order $n$ in $\bf{C}^n$. Besides, the zeros of $% \zeta (k)$ constitute exactly the Fermi variety $F_{\,\lambda }(q)$. We conclude now that the following representation holds on the set $\bf{R}% ^n\backslash F_{\bf{R},\lambda }(q)$: \[ \widehat{u}(k)=\frac{B(k)\phi (k)}{\zeta (k)}=\frac{g(k)}{\zeta (k)}, \] where $g(k)$ is a $H^2(\bf{T}^n)$-valued entire function of order $n$. Now we need the following auxiliary result: \begin{lemma} \label{divis}Let $Z$ be the set of all zeros of an entire function $\zeta (k) $ in $\bf{C}^n$ and $Z_j$ be its irreducible components. Assume that the real part $Z_{j,\bf{R}}=Z_j\cap \bf{R}^n$ of each $Z_j$ contains a submanifold of real dimension $n-1$. Let also $g(k)$ be an entire function in $\bf{C}^n$ with values in a Hilbert space $H$ such that on the real subspace $\bf{R}^n$ the ratio \[ \widehat{u}(k)=\frac{g(k)}{\zeta (k)} \] belongs to $L_{2,loc}(\bf{R}^n,H)$. Then $\widehat{u}(k)$ extends to an entire function with values in $H$. \end{lemma} \textbf{Proof of the lemma}. Applying linear functionals, one can reduce the problem to the case of scalar functions $g$, so we will assume that $g(k)\in \bf{C}$. According to Lemma \ref{property}, the sets of regular points of components $% Z_j$ are disjoint. Hence, the traces of these sets on $\bf{R}^n$ are also disjoint. Consider one component $Z_j$. The intersection of $regZ_j$ with the real space $\bf{R}^n$ contains a smooth manifold of dimension $(n-1)$. Namely, we know that $Z_{j,\bf{R}}$ contains such a manifold, which we will denote $M_j$. The only alternative to our conclusion would be that the whole $M_j$ sits inside the singular set of $Z_j$. The proof of Corollary \ref{cor} shows that this is impossible, since lower dimensional strata cannot contain any open pieces of $M_j$. Let us denote by $m_j$ the minimal order of zero of function $\zeta (k)$ on $% Z_j$. (We remind the reader that the order of zero of an analytic function at a point is determined by the order of the first non-zero term of the function's expansion at this point into homogeneous polynomials.) Since the condition that an analytic function has a zero of order higher than a given number can be written down as a finite number of analytic equations, one can conclude that the order of zero of $\zeta (k)$ equals $m_j$ on a dense open subset of $Z_j$, whose complement is an analytic subset of lower dimension. As it was explained in the proof of Corollary \ref{cor}, lower dimensional strata cannot contain $(n-1)$-dimensional submanifolds of $\bf{R}^n$. This means that one can find a point $k^j\in Z_{j,\bf{R}}$ such that $Z_{j,\bf{R% }}$ is a smooth hypersurface in a neighborhood $U$ of $k^j$ in $\bf{R}^n$ and such that the order of zero of $\zeta (k)$ on $Z_{j,\bf{R}}$ equals $% m_j $ in this neighborhood. Let us now prove that $g(k)$ has zeros on $Z_{j,% \bf{R}}\cap U$ of at least the same order as $\zeta (k)$. If this were not so, then in appropriate local coordinates $k=(k_1,...,k_n)$ the ratio $% g/\zeta $ would have a singularity of at least the order $k_1^{-1}$, which implies local square non-integrability of the function $\widehat{u}(k)$. This contradicts our assumption, so we conclude that $g(k)$ has zeros on $% Z_{j,\bf{R}}\cap U$ of at least the same order as $\zeta (k)$. Consider the (analytic) set $A$ of all points in $\bf{C}^n$ where $g(k)$ has zeros of order at least $m_j$. We have just proven that the intersection of $A$ with $% Z_j$ contains an $(n-1)$-dimensional smooth submanifold of $R^n$. Then Corollary \ref{cor} implies that $Z_j\subset A$, or $g(k)$ has zeros on $Z_j$ of at least the order $m_j$. Hence, according to Proposition 3 in section 1.5 of Chapter 1 in \cite{Ch} the ratio $g/\zeta $ is analytic everywhere in $\bf{C}^n$ except maybe at the union of subsets of $Z_j$, where the function $\zeta $ has zeros of order higher than $m_j$. This subset, however, is of dimension not higher than $n-2$. Now a standard analytic continuation theorem (see for instance Proposition 3 in Section 1.3 of the Appendix in \cite{Ch}) guarantees that $g/\zeta $ is an entire function. This concludes the proof of the lemma. Returning to the proof of the theorem and using the result of the Lemma \ref {big} and the assumption that the Condition \ref{condition} is satisfied for the potential $q(x)$, we conclude that $\widehat{u}(k)$ is an entire function and that it is the ratio of two entire functions of order at most $% w=\max (n,s)<4$, where $s$ is defined in Lemma \ref{order}. Using (in the radial directions) the estimate of entire functions from below contained in section 8 of Chapter 1 in \cite{Le} (see also Theorem 1.5.6 and Corollary 1.5.7 in \cite{K} or similar results in \cite{Bo}), we conclude that $% \widehat{u}(k)$ is itself an entire function of order $w$ with values in $% H^2(\bf{T}^n)$. Now Theorem 2.2.2 of \cite{K} claims that the solution $u(x) $ satisfies the decay estimate: \[ \left| \left| u\right| \right| _{H^2(K+a)}\leq C_p\exp (-c_p\left| a\right| ^p) \] for any $p4/3. \] This contradiction concludes the proof of the theorem.% %TCIMACRO{\TeXButton{End Proof}{\endproof}} \begin{remark} The condition $p>4/3$ (and hence the dimension restriction $n<4$) is essential for the validity of the result of \cite{FH} and \cite{M}, so this is the place where our argument breaks down for dimensions four and higher even if we require compactness of support of the perturbation potential $v(x) $. The rest of the arguments stay intact (the embedding theorem argument, which also depends on dimension, is not really necessary). \end{remark} \section{Separable potentials} The result of the previous section leads to the problem of finding classes of periodic potentials that satisfy the Condition \ref{condition}. As we stated in conjectures \ref{conj1} and \ref{conj2}, we believe that all (or almost all) of periodic potentials satisfy this condition. Although we were not able to prove these conjectures, as an immediate corollary of the Lemma \ref{irred} and of the Theorems \ref{cond} and \ref{sep} we get the following result: \begin{theorem} \label{combin}If for $n<4$ the background periodic potential $q(x)\in L^\infty (\bf{R}^n)$ is separable for $n=2$ or $q(x)=q_1(x_1)+q_2(x_2,x_3)$ for $n=3$, and the perturbation potential $v(x)$ satisfies the estimate \ref {estimate}, then there are no eigenvalues of the operator $H$ in the interior of the bands of the continuous spectrum. \end{theorem} \section{Comments} 1. We have only proven the absence of eigenvalues embedded into the interior of a spectral band. It is likely that eigenvalues cannot occur at the ends of the bands either (maybe except the bottom of the spectrum), if the perturbation potential decays fast enough. This was shown in the one-dimensional case in \cite{RB} under the condition \[ \int (1+\left| x\right| )\left| v(x)\right| dx<\infty \] on the perturbation potential. On the other hand, if $v(x)$ only belongs to $% L^1$, then the eigenvalues at the endpoints of spectral bands can occur \cite {RB2}. 2. Most of the proof of the conditional Theorem \ref{cond} does not require the unperturbed operator to be a Schr\"{o}dinger operator. One can treat general selfadjoint periodic elliptic operators as well. The only obstacle occurs at the last step, when one needs to conclude the absence of fast decaying solutions to the equation. Here we applied the results of \cite{FH} and \cite{M}, which are applicable only to the operators of the Schr\"{o}dinger type. Carrying over these results to more general operators would automatically generalize Theorem \ref{cond}. The restriction that the dimension $n$ is less than four also comes from the allowed rate of decay stated in the result of \cite{FH} and \cite{M}. 3. It might seem that the irreducibility condition \ref{condition} arises only due to the way the proof is done. We believe that this is not true, and that the validity of the condition is essentially equivalent to the absence of embedded eigenvalues. To be more precise, we conjecture that existence for some $\lambda $ in the interior of a spectral band of an irreducible component $A$ of the Fermi surface such that $A\cap \bf{R}^n=\emptyset $ implies existence of a localized perturbation of the operator that creates an eigenvalue at $\lambda $. As a supporting evidence of this one can consider fourth order periodic differential operators, where the Fermi surface contains four points. In this case one can have $\lambda $ in the continuous spectrum, while some points of the Fermi surface being complex. Then one can use these components of the Fermi surface ``hidden'' in the complex domain to cook up a localized perturbation that does create an eigenvalue at $\lambda $ \cite{P}. \begin{center} ACKNOWLEDGMENTS \end{center} The authors express their gratitude to Professors A. Figotin, T. Hoffmann-Ostenhof, H. Kn\"{o}rrer, S. Molchanov, and V. Papanicolaou for helpful discussions and information. The work of P. Kuchment was partly supported by the NSF Grant DMS 9610444 and by a DEPSCoR Grant. P. Kuchment expresses his gratitude to NSF, ARO, and to the State of Kansas for this support. The work of B. Vainberg was partly supported by the NSF Grant DMS-9623727. B. Vainberg expresses his gratitude to NSF for this support. The content of this paper does not necessarily reflect the position or the policy of the federal government, and no official endorsement should be inferred. \begin{thebibliography}{99} \bibitem{AADH} S. Alama, M. Avellaneda, P. A. Deift, and R. Hempel, On the existence of eigenvalues of a divergence form operator $A+\lambda B$ in a gap of $\sigma (A)$, Asymptotic Anal. 8(1994), no.4, 311-314. \bibitem{ADH} S. Alama, P. A. Deift, and R. Hempel, Eigenvalue branches of the Schr\"{o}dinger operator $H-\lambda W$ in a gap of $\sigma (H)$, Commun. Math. Phys. 121(1989), 291-321. \bibitem{BKT} D. B\"{a}ttig, H. Kn\"{o}rrer, and E. Trubowitz, A directional compactification of the complex Fermi surface, Compositio Math. 79(1991), no.2, 205-229. \bibitem{B} M. S. Birman, The discrete spectrum of the periodic Schr\"{o}dinger operator perturbed by a decreasing potential, Algebra i Analiz 8(1996), no.1, 3-20. \bibitem{B2} M. S. Birman, On the spectrum of singular boundary value problems, Mat. Sbornik 55(1961), no.2, 125-173. \bibitem{B3} M. S. Birman, On the discrete spectrum in the gaps of a perturbed periodic second order operator, Funct. Anal. Appl. 25(1991), 158-161. \bibitem{Bo} R. Boas, Jr., Entire Functions, Acad. Press, NY 1954. \bibitem{Ch} E. M. Chirka, Complex analytic sets, Mathematics and its Applications (Soviet Series), 46. Kluwer Academic Publishers Group, Dordrecht, 1989. \bibitem{D} P. A. Deift and R. Hempel, On the existence of eigenvalues of the Schr\"{o}dinger operator $H-\lambda W$ in a gap of $\sigma (H)$, Commun. Math. Phys. 103(1986), 461-490. \bibitem{E} M. S. P. Eastham, The Spectral Theory of Periodic Differential Equations, Scottish Acad. Press, Edinburgh - London 1973. \bibitem{EK} M. S. P. Eastham and H. Kalf, Schr\"{o}dinger-type Operators with Continuous Spectra, Pitman. Boston 1982. \bibitem{FH} R. Froese, I. Herbst, M. Hoffmann-Ostenhof, and T. Hoffmann-Ostenhof, $L^2$-lower bounds to solutions of one-body Schr\"{o}dinger equations, Proc. Royal Soc. Edinburgh, 95A(1983), 25-38. \bibitem{Gelf} I.M. Gelfand, Expansion in eigenfunctions of an equation with periodic coefficients, Dokl. Akad. Nauk SSSR 73(1950), 1117-1120. \bibitem{GS} F. Gesztesy and B. Simon, On a theorem of Deift and Hempel, Commun. Math. Phys. 116(1988), 503-505. \bibitem{G} I. M. Glazman, Direct Methods of Qualitative Spectral Analysis of Singular Differential Operators, I.P.S.T., Jerusalem 1965. \bibitem{GR} R. C. Gunning and H. Rossi, Analytic Functions of Several Complex Variables, Prentice-Hall, Englewood Cliffs, NJ 1965. \bibitem{He} R. Hempel, Eigenvalue branches of the Schr\"{o}dinger operator $H\pm \lambda W$ in a spectral gap of $H$, J. Reine Angew. Math. 399(1989), 38-59. \bibitem{H} L. H\"{o}rmander, The Analysis of Linear Partial Differential Operators, v. 2, Springer Verlag, Berlin 1983. \bibitem{KT} H. Kn\"{o}rrer and J. Trubowitz, A directional compactification of the complex Bloch variety, Comment. Math. Helv. 65(1990), 114-149. \bibitem{K} P. Kuchment, Floquet Theory for Partial Differential Equations, Birkh\"{a}user, Basel 1993. \bibitem{KV} P. Kuchment and B. Vainberg, On embedded eigenvalues of perturbed periodic Schr\"{o}dinger operators, in Spectral and scattering theory (Newark, DE, 1997), 67--75, Plenum, New York,1998. \bibitem{Lel} P. Lelong and L. Gruman, Entire Functions of Several Complex Variables, Springer Verlag, Berlin 1986. \bibitem{L} S. Z. Levendorskii, Asymptotic formulas with remainder estimates for eigenvalue branches of the Schr\"{o}dinger operator $H-\lambda W$ in a gap of $H$. To appear in Transactions of American Mathematical Society. \bibitem{L2} S. Z. Levendorskii and S. I. Boyarchenko, An asymptotic formula for the number of eigenvalue branches of a divergence form operator $% A+\lambda B$ in a spectral gap of $A$. To appear in Communications in Part. Differ. Equat. \bibitem{L3} S. Z. Levendorskii, Lower bounds for the number of eigenvalue branches for the Schr\"{o}dinger operator $H-\lambda W$ in a gap of $H$: the case of indefinite $W$, Comm. partial Diff. Equat. 20(1995), no.5-6, 827-854. \bibitem{Le} B. Levin, Distribution of Zeros of Entire Functions, Transl. Math. Monogr. v.5, Amer. Math. Soc., Providence, RI 1964. \bibitem{M} V. Meshkov, On the possible rate of decay at infinity of solutions of second order partial differential equations, Mat. Sbornik, 182(1991), no.3, 364-383. English translation in Math. USSR Sbornik 72(1992), no.2, 343-351. \bibitem{N} R. Narasimhan, Intorduction to the Theory of Analytic Spaces, Lect. Notes in Math., v. 25, Springer Verlag, Berlin 1966. \bibitem{OK} F. Odeh, J.B. Keller, Partial differential equations with periodic coefficients and Bloch waves in crystals, J. Math. Phys. 5(1964), 1499-1504. \bibitem{P} V. Papanicolaou, private communication, 1998. \bibitem{Ra} G. D. Raikov, Eigenvalue asymptotics for the Schr\"{o}dinger operator with perturbed periodic potential, Invent. Math. 110(1992), 75-93. \bibitem{RS} M. Reed and B. Simon, Methods of Modern Mathematical Physics v. 4, Acad. Press, NY 1978. \bibitem{RB} F. S. Rofe-Beketov, A test for the finiteness of the number of discrete levels introduced into the gaps of a continuous spectrum by perturbations of a periodic potential, Soviet Math. Dokl. 5(1964), 689-692. \bibitem{RB2} F. S. Rofe-Beketov, Spectrum perturbations, the Knezer-type constants and the effective mass of zones-type potentials, in ``Constructive Theory of Functions'84'', Sofia 1984, p.757-766. \bibitem{S} A. V. Sobolev, Weyl asymptotics for the discrete spectrum of the perturbed Hill operator, Adv. Sov. Math. 7(1991), 159-178. \bibitem{T} E.C. Titchmarsh, Eigenfunction Expansions Associated with Second-Order Differential Equations, Part II, Claredon Press, Oxford 1958. \bibitem{V} B. Vainberg, Principles of radiation, limiting absorption and limiting amplitude in the general theory of partial differential equations, Russian Math. Surveys 21(1966), no 3, 115-193. \bibitem{W} C. Wilcox, Theory of Bloch waves, J. Analyse Math. 33(1978), 146-167. \end{thebibliography} \end{document} ---------------9904161716868--