Prescription of the multiplicity of the eigenvalues of the Hodge-de Rham Laplacian (Q643321)

The goal of this article is to extend the work of \textit{Y. Colin de Verdière} [Comment. Math. Helv. 61, 254--270 (1986; Zbl 0607.53028); Ann. Sci. Éc. Norm. Supér. (4) 20, No. 4, 599--615 (1987; Zbl 0636.58036)] who proved that it is always possible to prescribe a finite part of the spectrum of the Laplacian on any compact Riemannian manifold of dimension at least 3, and in dimension 2 subject to certain constraints. The present article demonstrates similar results for the Hodge-de Rham Laplacian (on forms) in dimensions 6 and higher. The first main result, Theorem 1, shows that one may prescribe the volume and the (positive) spectrum for \(p\)-forms for all degrees except the middle degree, and the first part of the spectrum in the middle degree. There is one constraint: the first eigenvalue on 1-forms must have multiplicity one. The second main result, Theorem 4, states that on any compact, connected, orientable closed manifold of dimension at least 5, there exists a Riemannian metric such that the multiplicity of the first eigenvalue of the Hodge-de Rham Laplacian on 1-forms is equal to three. The methods used to prove these results are similar to the techniques in [loc. cit., Zbl 0607.53028, Zbl 0636.58036]. In \S 2.1 the author discusses the boundary conditions which shall be assumed for the model domain (there appears to be a very minor typo in condition (R) on p. 970; we believe that the second condition should read \(j^* (\delta \omega) = 0\).) In \S 2.2, a variational principle for the Hodge-de Rham Laplacian is presented in Proposition 7 (due to [\textit{J. Dodziuk}', Proc. Am. Math. Soc. 85, 437--443 (1982; Zbl 0502.58038)], [\textit{J. McGowan}, Math. Ann. 297, No. 4, 725--745 (1993; Zbl 0801.53034)]), and the quadratic form associated to the Hodge-de Rham Laplacian restricted to exact forms is identified in Proposition 8. The proof of spectral convergence comprises \S 2.3. This section begins with a heuristic description of the proof of Theorem 11, which shows that the spectrum of the manifold can be made to converge to the spectrum of a domain in the manifold. This is followed by Theorem 13, which contains a volume estimate proven in [\textit{P. Jammes}, Manuscr. Math. 123, No. 1, 15--23 (2007; Zbl 1127.35027)]. Two rather general functional analytic Lemmas (16 and 17) from [loc. cit., Zbl 0607.53028] together with Remarks 18 and 19 preceed the proof of Theorem 11 which completes this section. The last section of the paper \S 3 presents the construction of the domain with the desired spectrum (Lemmas 21, 22, 23), and completes the proof of Theorems 1 and 4. The introduction contains a brief survey on the subject and raises some interesting open problems: Is it possible that the multiplicity of the first positive eigenvalue of the Hodge-de Rham Laplacian on 1-forms may be arbitrarily large? The same question for the \(k^{th}\) eigenvalue on middle degree forms. In dimension 3, does there exist a bound on the multiplicity of the \(k^{th}\) eigenvalue of the Hodge-de Rham Laplacian on 1-forms which depends uniquely on \(k\) and the topology of \(M\)?