For a spherical shell a<r<ba<r<b filled by an inviscid fluid with radially varying radial mass density ρ(r)\rho(r), tangential mass density s(r)s(r), and compressibility γ(r)\gamma(r), the question is whether there exist positive triplets {ρ(r),s(r),γ(r)}\{\rho(r),s(r),\gamma(r)\} that produce exactly zero scattering of incident plane waves for all incident directions at a fixed wavenumber while keeping the inner-boundary radial density ρ(a)\rho(a) finite — and hence the total radial mass ∫abρ(r)r2dr\int_a^b \rho(r) r^2 dr finite — and that genuinely cloak a nontrivial interior. All known exact solutions arise from transformation acoustics and require a mapping R(r)R(r) with R(a)=0R(a)=0, which forces ρ(a)=∞\rho(a)=\infty and infinite total radial mass; A. N. Norris conjectured in 2015, without proof, that no other solutions exist. The problem asks for either an explicit finite-mass zero-scattering construction (a counterexample) or a proof of impossibility.

Background

Acoustic cloaking aims to make an object undetectable to sound waves by surrounding it with a structured medium. In transformation acoustics, one designs the medium by a coordinate change: a virtual homogeneous space is mapped onto the physical shell so that incident waves are guided around the cloaked region and re-emitted without distortion. For a fluid, the resulting medium is an 'inertial cloak': an inviscid fluid whose effective mass density is anisotropic, described by a radial density ρ(r)\rho(r) (governing motion along the radius) and a tangential density s(r)s(r) (governing motion along the sphere), together with a scalar compressibility γ(r)\gamma(r) describing how much the fluid compresses under pressure. Norris ('Acoustic cloaking theory', 2008) and Cummer et al. (2008) showed that the exact transformation solution for a spherical cloak corresponds to a radial map R(r)R(r) from the physical coordinate rr to the virtual coordinate, with parameters ρ=ρ0r2R′/R2\rho=\rho_0 r^2 R'/R^2, s=ρ0/R′s=\rho_0/R', γ=(R2/r2)R′γ0\gamma=(R^2/r^2)R'\gamma_0 (primes denote radial derivatives; ρ0,γ0\rho_0,\gamma_0 are the uniform exterior values). Perfect cloaking requires R(a)=0R(a)=0: the map collapses the virtual origin onto the inner boundary. This degeneration makes ρ(r)→∞\rho(r)\to\infty as r↓ar\downarrow a, so the total radial mass ∫abρ(r)r2dr\int_a^b \rho(r) r^2 dr of the cloak is infinite — a fundamental obstacle to physical realization. Norris (2008) also showed a related design using a pentamode material (a solid whose stress is isotropic but whose stiffness tensor is anisotropic, outside the inertial-fluid class) that achieves finite mass, equal to the mass of background fluid filling the sphere of radius bb, and truncated or regularized 'near-cloaks' following Kohn, Shen, Vogelius and Weinstein hide only approximately.

Dassios and Fokas ('Characterization of an acoustic spherical cloak', Inverse Problems 31:035001, 2015) posed the inverse problem within this framework: given the radial density ρ(r)\rho(r) in the shell, are the remaining two parameters s(r)s(r) and γ(r)\gamma(r) uniquely determined by the requirement of zero scattering, and how can they be computed? They proposed a solution method based on a Riemann–Hilbert problem (the task of reconstructing a function analytic in domains of the complex plane from prescribed jumps across contours, the engine of the Fokas unified-transform method for solving linear partial differential equations) applied to spectral data of a one-dimensional inhomogeneous Schrödinger equation equivalent to the radial mode equations, but gave no constructive examples. In a 2015 arXiv comment (rejected by the journal and unpublished beyond arXiv), Norris supplied explicit solutions — for any ρ(r)\rho(r) compatible with transformation acoustics, ss and γ\gamma follow from the map RR defined by ρ\rho — corrected the n=0n=0 compatibility condition of Dassios–Fokas (the mode-wise condition at the inner boundary r=ar=a expressing continuity of pressure and radial velocity, valid for all spherical-harmonic mode numbers n≥0n\ge 0), and concluded that the transformation solutions require R(a)=0R(a)=0, which makes ρ(r)\rho(r) singular as r↓ar\downarrow a and the total radial mass infinite, via the identity expressing the enclosed radial mass as a term proportional to 1/R(r)→∞1/R(r)\to\infty. He then asked whether this is the end of the story: do other, non-transformation positive triplets exist with finite ρ(a)\rho(a), or is the mass singularity unavoidable? He conjectured (intuitively, without proof) that no finite-ρ(a)\rho(a) solutions exist. The broader mathematical framework (Greenleaf, Kurylev, Lassas and Uhlmann's survey of invisibility and inverse problems) establishes that exact passive cloaking for Helmholtz-type equations is achieved only by singular (degenerate) transformation structures, while nonsingular parameters yield at best approximate cloaking — a strong indication that Norris's conjecture is right, but not a published theorem for this fixed-frequency, radial, inertial-cloak statement, and no construction of a perfect finite-mass inertial cloak exists in the literature through 2025, where weight-reduction studies explicitly treat the exact finite-mass perfect cloak as unattainable under known strategies.

Problem Statement

Consider a spherical shell a<r<ba<r<b embedded in a homogeneous acoustic background fluid with uniform mass density ρ0\rho_0, compressibility γ0\gamma_0, and sound speed c=(ρ0γ0)−1/2c=(\rho_0\gamma_0)^{-1/2}; time-harmonic fields with dependence e−iωte^{-i\omega t} are assumed, so the background wavenumber is k=ω/ck=\omega/c. The shell surrounds an interior ball r<ar<a containing prescribed nontrivial acoustic content: a concentric uniform fluid with density ρ1\rho_1 and compressibility γ1\gamma_1 (sound speed c1=(ρ1γ1)−1/2c_1=(\rho_1\gamma_1)^{-1/2}, wavenumber k1=ω/c1k_1=\omega/c_1) that is not itself a nonsingular image of the homogeneous background fluid — in this uniform-fluid class that means excluding exactly the one-parameter family γ1=(ρ0/ρ1)3γ0\gamma_1=(\rho_0/\rho_1)^3\gamma_0. This spherically symmetric interior is the class for which the per-mode conditions below apply; a counterexample need only be exhibited for one such interior, while an impossibility proof must cover every such interior. The shell is occupied by an inviscid fluid with anisotropic effective mass density: a positive radial density ρ(r)\rho(r) (governing motion along the radius), a positive tangential density s(r)s(r) (governing motion along the sphere), and a positive scalar compressibility γ(r)\gamma(r). Expanding the acoustic pressure in spherical harmonics YnmY_n^m (with radial mode functions independent of the order mm by spherical symmetry), each mode fnf_n satisfies, for every mode number n≥0n\ge 0,

ddr[r2ρ(r)dfndr]+[ω2r2γ(r)−n(n+1)s(r)]fn(r)=0,a<r<b,\frac{d}{dr}\left[\frac{r^2}{\rho(r)}\frac{df_n}{dr}\right]+\left[\omega^2 r^2\gamma(r)-\frac{n(n+1)}{s(r)}\right]f_n(r)=0,\qquad a<r<b,

the radial mode equation of Dassios–Fokas (their eq. (19), as quoted in the source comment). Zero scattering for plane-wave incidence at the fixed wavenumber kk means that the exterior field equals the incident plane wave exactly, with no outgoing scattered part for any incident direction; in modes this requires, at the outer boundary,

fn(b)=jn(kb),1ρ(b)fn′(b)=kρ0jn′(kb),f_n(b)=j_n(kb),\qquad \frac{1}{\rho(b)}f_n'(b)=\frac{k}{\rho_0}j_n'(kb),

where primes denote radial derivatives and jnj_n is the spherical Bessel function of order nn, together with the shell–interior matching at r=ar=a (continuity of pressure and radial velocity across the interface), whose compatibility condition valid for every mode number n≥0n\ge 0 — including n=0n=0, where the original condition of Dassios–Fokas fails — is the corrected condition derived in the comment: the exterior scattering coefficient of the two-sided mode expansion vanishes for every n≥0n\ge 0 (the comment's eq. (26)). Equivalently, the far-field scattering amplitude is zero for all incident and observation directions.

Within this class of spherically symmetric inertial acoustic cloaks, determine whether there exist parameter triplets {ρ(r),s(r),γ(r)}\{\rho(r),s(r),\gamma(r)\} such that:

  1. the composite (shell plus interior) produces exactly zero scattering for plane-wave incidence at the fixed wavenumber kk, for every incident direction, in the precise sense of the mode conditions above; and
  2. the inner-boundary radial density is finite, ρ(a)<∞\rho(a)<\infty as r↓ar\downarrow a, so that the total radial mass ∫abρ(r) r2 dr\int_a^b \rho(r)\,r^2\,dr is finite.

A resolution is either an explicit positive triplet with a proof of zero scattering and finite total radial mass that genuinely cloaks the interior (a counterexample), or a mathematical proof that no such triplet exists (impossibility, or a classification of all zero-scattering triplets and their mass properties). The impossibility branch would establish Norris's conjecture that every such exact solution is a transformation-acoustics solution whose radial map R(r)R(r) collapses the inner boundary, R(a)=0R(a)=0, which forces ρ(a)=∞\rho(a)=\infty and infinite total radial mass; a classification of zero-scattering triplets counts as a resolution only insofar as it fixes the mass verdict for every zero-scattering triplet. Solutions that are mere nonsingular relabelings of the homogeneous background fluid — gauge-equivalent to empty space, the image of the homogeneous background under a single non-collapsing change of variables that fixes the outer boundary r=br=b and has positive derivative throughout — satisfy the mode conditions but cloak nothing, whether or not the interior parameters equal the background, and they do not answer the question.

References

  1. A. N. Norris, Comment on "Characterization of an acoustic spherical cloak", Inverse Problems 31(3):035001, 2015 — unpublished comment (rejected by Inverse Problems; exists only as arXiv:1511.02908), DOI: 10.48550/arXiv.1511.02908, https://arxiv.org/abs/1511.02908
  2. G. Dassios, A. S. Fokas, Characterization of an acoustic spherical cloak, Inverse Problems 31(3):035001, 2015, DOI: 10.1088/0266-5611/31/3/035001, https://doi.org/10.1088/0266-5611/31/3/035001
  3. A. N. Norris, Acoustic cloaking theory, Proceedings of the Royal Society A 464(2097):2411–2434, 2008, DOI: 10.1098/rspa.2008.0076, https://doi.org/10.1098/rspa.2008.0076
  4. S. A. Cummer, B. I. Popa, D. Schurig, D. R. Smith, J. Pendry, M. Rahm, A. Starr, Scattering theory derivation of a 3D acoustic cloaking shell, Physical Review Letters 100(2):024301, 2008, DOI: 10.1103/PhysRevLett.100.024301, https://doi.org/10.1103/PhysRevLett.100.024301
  5. R. V. Kohn, H. Shen, M. S. Vogelius, M. I. Weinstein, Cloaking via change of variables in electric impedance tomography, Inverse Problems 24(1):015016, 2008, DOI: 10.1088/0266-5611/24/1/015016, https://doi.org/10.1088/0266-5611/24/1/015016
  6. A. Greenleaf, Y. Kurylev, M. Lassas, G. Uhlmann, Invisibility and inverse problems, Bulletin of the American Mathematical Society 46(1):55–97, 2009, DOI: 10.1090/S0273-0979-08-01232-9 (preprint arXiv:0810.0263), https://doi.org/10.1090/S0273-0979-08-01232-9
  7. D. E. Quadrelli, S. Cominelli, G. Cazzulani, F. Braghin, Reduced-weight near-cloaks for underwater invisibility via transformation acoustics, Journal of the Acoustical Society of America 158(5):3861–3869, 2025, DOI: 10.1121/10.0039884, https://doi.org/10.1121/10.0039884
  8. A. Arjunan, A. Baroutaji, J. Robinson, C. Wang, Characteristics of Acoustic Metamaterials, in Encyclopedia of Smart Materials, Elsevier, pp. 35–45, 2022, DOI: 10.1016/B978-0-12-815732-9.00090-5, https://doi.org/10.1016/B978-0-12-815732-9.00090-5