Skip to content

Mode Equations on FLRW Spacetimes

Scrollable formula. Use Left and Right arrow keys to reveal hidden content.

Scrollable table. Use Left and Right arrow keys to reveal additional columns.

Scrollable figure. Use Left and Right arrow keys to inspect hidden content.

Expansion turns each Fourier mode of a free field into an oscillator whose frequency and damping vary with time. The conversion is exact, not an approximation: it follows by writing the Klein–Gordon or Dirac equation on a prescribed FLRW geometry and choosing a canonical field variable. What remains nonunique is the state—equivalently, the normalized positive-frequency solution selected as initial data.

Required background. Compactly, a second-order mode equation needs two initial data and its conserved Wronskian fixes their symplectic normalization; distributions assemble the modes into the field two-point function. Partial Differential Equations and Green Functions supplies that initial-value, distribution, and Green-function framework. Helpful background. Flat-Spacetime Quantum Field Theory provides oscillator quantization and Fourier modes; Causality and Global Structure establishes the Cauchy surfaces on which initial data are posed; Quantum Fields in Curved Spacetime provides the state and fixed-background setting.

Take a four-dimensional, spatially flat FLRW spacetime and normalize the scale factor by a(t∗)=1a(t_\ast)=1 at a declared reference time:

ds2=−dt2+a2(t)dx2=a2(η)(−dη2+dx2),dη=dta(t).ds^2=-dt^2+a^2(t)d\mathbf x^2 =a^2(\eta)\left(-d\eta^2+d\mathbf x^2\right), \qquad d\eta=\frac{dt}{a(t)}.

Cosmic-time and conformal-time derivatives are denoted by overdots and primes. Their expansion rates are

H=a˙a,H=a′a=aH,R=6(H˙+2H2)=6a′′a3.H=\frac{\dot a}{a}, \qquad \mathcal H=\frac{a'}a=aH, \qquad R=6(\dot H+2H^2)=\frac{6a''}{a^3}.

The spatial coordinates x\mathbf x and their Fourier labels k\mathbf k are comoving. A comoving observer measures physical momentum p=k/a\mathbf p=\mathbf k/a and physical wavelength 2πa/k2\pi a/k. Consequently, “large kk” is not by itself a short-distance statement at all times; the dimensionless comparison is k/(aH)k/(aH) or k/(am)k/(am).

For nonzero spatial curvature, plane waves are replaced by eigenfunctions of the spatial Laplacian and k2k^2 by the corresponding eigenvalue, with curvature-dependent spectral measure. We keep the flat case so that no discrete-spectrum or topology qualification is hidden.

Scalar reduction to a canonical oscillator

Section titled “Scalar reduction to a canonical oscillator”

For the action and signature used throughout this volume, the real scalar obeys

(□−m2−ξR)ϕ=0.(\Box-m^2-\xi R)\phi=0.

In cosmic time this becomes

ϕ¨+3Hϕ˙−1a2∇2ϕ+(m2+ξR)ϕ=0.\ddot\phi+3H\dot\phi-\frac{1}{a^2}\nabla^2\phi +(m^2+\xi R)\phi=0.

Expand the field in comoving Fourier modes,

ϕ(η,x)=1a∫d3k(2π)3/2[akχk(η)eik⋅x+ak†χk∗(η)e−ik⋅x].\phi(\eta,\mathbf x)=\frac1a\int\frac{d^3k}{(2\pi)^{3/2}} \left[a_{\mathbf k}\chi_k(\eta)e^{i\mathbf k\cdot\mathbf x} +a_{\mathbf k}^\dagger\chi_k^*(\eta)e^{-i\mathbf k\cdot\mathbf x}\right].

Before rescaling, the mode function ϕk=χk/a\phi_k=\chi_k/a satisfies

ϕk′′+2Hϕk′+(k2+a2m2+a2ξR)ϕk=0.\phi_k''+2\mathcal H\phi_k' +\left(k^2+a^2m^2+a^2\xi R\right)\phi_k=0.

Substituting χk=aϕk\chi_k=a\phi_k cancels the first derivative and gives the canonical oscillator

χk′′+ωk2(η)χk=0,ωk2=k2+a2m2+(6ξ−1)a′′a.\chi_k''+\omega_k^2(\eta)\chi_k=0, \qquad \omega_k^2=k^2+a^2m^2+(6\xi-1)\frac{a''}{a}.

This is how expansion enters: the physical mass becomes amam in conformal time, while nonconformal curvature coupling contributes (6ξ−1)a′′/a(6\xi-1)a''/a. The derivation is consistent with the canonical treatment of perturbations in Mukhanov, Feldman, and Brandenberger 1992, Part I, §3, p. 213 onward.

The equal-time commutator [ϕ(η,x),a2ϕ′(η,y)]=iδ3(x−y)[\phi(\eta,\mathbf x),a^2\phi'(\eta,\mathbf y)]=i\delta^3(\mathbf x-\mathbf y) is equivalent to

χkχk∗′−χk∗χk′=i.\chi_k\chi_k^{*\prime}-\chi_k^*\chi_k'=i.

The mode equation conserves this Wronskian because ωk2\omega_k^2 is real, even when it is negative and the solutions cease to oscillate. A measurable family of such data, depending only on k=∣k∣k=|\mathbf k| and with sufficient ultraviolet and infrared control to define the two-point distribution, selects a pure homogeneous, isotropic quasifree state. General mixed quasifree states need occupation data as well. The differential equation alone does not select “the vacuum.” A different normalized solution

χ~k=αkχk+βkχk∗,∣αk∣2−∣βk∣2=1,\widetilde\chi_k=\alpha_k\chi_k+\beta_k\chi_k^*, \qquad |\alpha_k|^2-|\beta_k|^2=1,

defines a different mode basis and generally a different state. Ultraviolet admissibility constrains the large-kk behavior of βk\beta_k, but it does not remove all state freedom.

Choose a conformal tetrad and the Clifford convention {γμ,γν}=2gμν1\{\gamma^\mu,\gamma^\nu\}=2g^{\mu\nu}\mathbf1 used on Dirac and Maxwell Fields. The Dirac equation is then (γμ∇μ−m)ψ=0(\gamma^\mu\nabla_\mu-m)\psi=0: there is no extra ii with these mostly-plus Clifford matrices. The spin connection produces a term proportional to 3H/23\mathcal H/2. Rescaling

Ψ=a3/2ψ\Psi=a^{3/2}\psi

removes it and yields a flat-form equation with a time-dependent mass,

(γ0^∂η+γi^∂i−am)Ψ=0.\left(\gamma^{\hat0}\partial_\eta+\gamma^{\hat i}\partial_i-am\right)\Psi=0.

Hatted matrices are constant tetrad matrices. To compare with the common particle-physics equation iΓa∂a−ami\Gamma^a\partial_a-am, use Γa=−iγa^\Gamma^a=-i\gamma^{\hat a}, so {Γa,Γb}=−2ηab1\{\Gamma^a,\Gamma^b\}=-2\eta^{ab}\mathbf1. With spatial modes eik⋅xe^{i\mathbf k\cdot\mathbf x} and a fixed helicity, absorb the helicity sign in the lower component. The two amplitudes obey

iddη(hkIhkII)=(amkk−am)(hkIhkII),∣hkI∣2+∣hkII∣2=1.i\frac{d}{d\eta} \begin{pmatrix}h_k^I\\ h_k^{II}\end{pmatrix} =\begin{pmatrix}am&k\\ k&-am\end{pmatrix} \begin{pmatrix}h_k^I\\ h_k^{II}\end{pmatrix}, \qquad |h_k^I|^2+|h_k^{II}|^2=1.

The Hermitian matrix conserves this positive norm. Eliminating the opposite component gives hkI′′+[k2+a2m2+i(am)′]hkI=0h_k^{I\prime\prime}+[k^2+a^2m^2+i(am)']h_k^I=0 and the corresponding equation with −i(am)′-i(am)' for hkIIh_k^{II}. The first-order equations are still needed: the separate second-order equations admit pairs that violate the original coupling or normalization. This is the conformal-time translation of Landete, Navarro-Salas, and Torrenti 2014, §III, Eqs. (19)–(26), PDF. For m=0m=0, the rescaled spinor is exactly a flat-space field on the conformally flat patch.

If aa is constant, then a′′=0a''=0 and the scalar frequency reduces to k2+a2m2k^2+a^2m^2, the conformal-time form of p2+m2p^2+m^2. If m=0m=0 and ξ=1/6\xi=1/6, then ωk=k\omega_k=k for every scale factor. A state obtained by conformally transporting the Minkowski vacuum has plane-wave modes and no expansion-induced Bogoliubov mixing. This no-creation statement requires both a conformally invariant field equation and the corresponding conformal state; an excited state, nonconformal boundary condition, nontrivial topology, mass, or different coupling can invalidate it.

Under the harmless coordinate relabeling a↦Caa\mapsto Ca, x↦x/C\mathbf x\mapsto\mathbf x/C, conformal time changes as η↦η/C\eta\mapsto\eta/C and the comoving label changes as k↦Ck\mathbf k\mapsto C\mathbf k. The physical momentum k/a\mathbf k/a and the ratio k/Hk/\mathcal H are invariant. Every term in ωk2\omega_k^2 scales as an inverse conformal time squared. These checks catch the common error of calling kk a physical momentum or mixing HH with H\mathcal H.

Diagnostic: translate a horizon-crossing condition

Section titled “Diagnostic: translate a horizon-crossing condition”

A mode is said to satisfy k=Hk=\mathcal H at conformal time η×\eta_\times. Express the same condition in cosmic-time and physical variables. Then decide whether the massless minimally coupled scalar has the conformal no-creation property.

Solution

Because H=aH\mathcal H=aH and p=k/ap=k/a, the equality k=Hk=\mathcal H is exactly p=Hp=H at the same event; it compares a physical wave number with the cosmic expansion rate. The label kk itself remains comoving. For a massless minimally coupled scalar, m=0m=0 and ξ=0\xi=0, so ωk2=k2−a′′/a\omega_k^2=k^2-a''/a rather than k2k^2. Expansion therefore produces nontrivial mode evolution. Only the massless conformally coupled scalar, m=0m=0 and ξ=1/6\xi=1/6, has the exact flat oscillator after the χ=aϕ\chi=a\phi rescaling, and even then the no-creation claim assumes the conformally transported state and compatible global conditions.

  • Landete, Aitor, José Navarro-Salas, and Francisco Torrenti. “Adiabatic Regularization and Particle Creation for Spin One-Half Fields.” Physical Review D 89 (2014): 044030. DOI; Open PDF.
  • Mukhanov, Viatcheslav F., H. A. Feldman, and Robert H. Brandenberger. “Theory of Cosmological Perturbations.” Physics Reports 215 (1992): 203–333. DOI.
  • Parker, Leonard, and Stephen A. Fulling. “Adiabatic Regularization of the Energy-Momentum Tensor of a Quantized Field in Homogeneous Spaces.” Physical Review D 9 (1974): 341–354. DOI.
  • Parker, Leonard. “Quantized Fields and Particle Creation in Expanding Universes. I.” Physical Review 183 (1969): 1057–1068. DOI.