Semilinear wave equations on the Witten bubble spacetime

Technical Summary: Semilinear Wave Equations on the Witten Bubble Spacetime

Problem Statement
This paper initiates the rigorous mathematical study of nonlinear wave equations on the Witten bubble spacetime, a $(4+1)$-dimensional solution to the Einstein vacuum equations representing the "bubble of nothing." This spacetime models the semiclassical instability of the Kaluza–Klein vacuum $\mathbb{R}^{3+1} \times S^1$. While the Kaluza–Klein vacuum is classically stable, the Witten bubble describes a hole that spontaneously forms, pinching off the compact $S^1$ dimension and expanding with uniform acceleration.

The primary objective is to establish small-data global existence for solutions to semilinear wave equations of the form $\square_g \phi = F(\nabla \phi)$ on this background. A central difficulty arises from the geometry's dual nature: it interpolates between a near-bubble region (effectively $(2+1)$-dimensional with an $S^2$ factor) and an asymptotically flat region (effectively $(3+1)$-dimensional with an $S^1$ factor). Furthermore, the spacetime exhibits "weak trapping" for null geodesics with non-zero angular momentum in the $S^1$ direction, leading to potential non-decay of certain solution modes.

Methodology
The author employs physical space methods, specifically vector field multipliers and energy estimates, rather than the spectral methods used in previous linear studies (e.g., Bachelot [Bac16]). The core strategy involves decomposing the solution into $U(1)$-symmetric and non-$U(1)$-symmetric parts to handle the distinct decay properties of each.

1. Twisted Variables: To derive energy estimates, the author works with the twisted quantity $\psi = r\phi$, where $r = \rho \cosh \tau$ is the area-radius function. This transformation is necessary because standard energy estimates for $\phi$ fail on the Witten bubble due to the expansion of the spacetime. The operator $P$ acting on $\psi$ reveals a structure resembling the wave equation in two space dimensions near the bubble.
2. Mode Decomposition: The solution $\psi$ is split into $\psi_0$ (the $U(1)$-symmetric part, or average over $S^1$) and $\psi_{\ge 1}$ (the non-symmetric part).
   • $\psi_0$: Behaves like a massless wave on Minkowski space, enjoying strong time-decay properties.
   • $\psi_{\ge 1}$: Behaves like a solution to a massive Klein–Gordon equation with an effective mass growing at spatial infinity. Consequently, it does not enjoy improved time decay (it may be periodic in $\tau$) but exhibits faster spatial decay due to the confining effective potential.
3. Energy Estimates:
   • $\partial_\tau$-Energy: A basic energy estimate associated with the timelike vector field $\partial_\tau$ is established for all modes. This controls the solution in the bulk but does not yield decay for $\psi_{\ge 1}$.
   • $r^p$-Type Estimates: A novel $r^p$-weighted energy estimate (inspired by Dafermos–Rodnianski [DR09] and extended to two dimensions in [Gau26]) is derived. Crucially, this estimate is only valid for the $U(1)$-symmetric part $\psi_0$. It captures the improved decay of outgoing null derivatives and allows for exponential decay in the hyperboloidal time $s$.
4. Pointwise Estimates: Using Sobolev embedding and higher-order energies, the author derives pointwise bounds. A key observation is the dichotomy in decay: $\psi_{\ge 1}$ decays faster in space ($\sim \rho^{-1/2}$) than $\psi_0$, while $\psi_0$ decays faster in time.
5. Nonlinear Analysis: For semilinear equations satisfying a version of the null condition, a bootstrap argument is employed. The proof balances the better time decay of the symmetric part against the better spatial decay of the non-symmetric part. Special care is taken to control "dangerous" frequency interactions where the non-symmetric part (with poor time decay) multiplies the symmetric part.

Key Results

• Linear Theory (Theorem 1.1):

  • Quantitative Boundedness: $\psi_{\ge 1}$ decays as $\rho^{-1/2}$ in space, while $\psi_0$ remains bounded. Since $r$ grows in time, this implies time decay for $\phi$.
  • Radiation Field: A radiation field exists at null infinity and is purely $U(1)$-symmetric.
  • Vanishing at the Bubble: Solutions with vanishing $U(1)$-symmetric part vanish along the bubble.
  • Decay Dichotomy: $U(1)$-symmetric solutions enjoy improved time decay (exponential in hyperboloidal time, polynomial in Minkowskian time). Conversely, non-symmetric solutions may fail to decay in time (exhibiting periodicity in $\tau$) due to weak trapping, though they decay faster in space.

• Semilinear Theory (Theorem 1.2):

  • Global Existence (No Symmetry): For semilinear equations satisfying a global null condition (including perturbations of $g^{\alpha\beta}\partial_\alpha\phi\partial_\beta\phi$), small-data global existence is proven without symmetry assumptions. The solution $\psi$ grows at most polynomially in $\tau$ (logarithmically in Minkowskian time $t$).
  • Global Existence (With Symmetry): If both the initial data and the nonlinearity are $U(1)$-symmetric, the solution exists globally and enjoys improved decay properties. Specifically, $\psi$ decays exponentially in the hyperboloidal time $s$ (polynomially in $t$), and a finite radiation field exists.
  • Null Condition: The paper formulates a global version of the null condition suitable for the Witten bubble, which reduces to the classical null condition near null infinity.

Significance and Claims
The paper claims to be the first to study nonlinear wave equations on the Witten bubble spacetime. Its significance lies in:

1. Novel Estimates: Establishing new $r^p$-type energy estimates and quantitative pointwise estimates for the linear wave equation on this non-stationary, non-flat background.
2. Handling Weak Trapping: Successfully addressing the "weak trapping" phenomenon, where non-symmetric modes do not decay in time, by exploiting the spatial decay of these modes to close nonlinear estimates.
3. Symmetry Dichotomy: Demonstrating a clear dichotomy in the behavior of symmetric versus non-symmetric modes, showing that symmetry is essential for obtaining improved time decay and preventing polynomial growth in the solution.
4. Extension of Physical Space Methods: Adapting techniques from the study of wave equations on non-flat backgrounds (specifically the $r^p$ method and two-dimensional wave equation techniques) to a $(4+1)$-dimensional spacetime with a compact dimension.

The author notes that while the solution in the non-symmetric case may grow polynomially in $\tau$ (unlike the bounded solutions on Minkowski space), this growth is logarithmic in the physical Minkowskian time $t$, representing a controlled deviation from classical stability results. The work provides a foundational step toward understanding the classical stability of the Witten bubble as a solution to the Einstein vacuum equations.