betaFeedback

WorksPhysics and Astronomy

Alleviating CMB-BAO tension with a large cosmic birefringence angle during reionization induced by energy injection to axion-like particles

Abstract

The recent measurement of baryon acoustic oscillations by the Dark Energy Spectroscopic Instrument reveals a mild tension with observations of the cosmic microwave background (CMB) within the standard ΛΛCDM cosmological model. Another interesting hint beyond standard cosmology arises from recent analyses of Planck polarization data which has been interpreted as cosmic birefringence. One possible solution to explain both observations is to introduce cosmic birefringence induced by axion-like particles (ALPs) that has a much larger rotation angle during the reionization epoch than that during the recombination: the large rotation suppresses the EEEE reionization signal, allowing an elevated value of the CMB optical depth ττ that alleviates the CMB-BAO tension. In this paper, we explore a scenario where the energy transfers from matter to ALPs before the reionization. Using Planck low-ℓ\ell EEEE, BBBB, and EBEB data, we show that, for energy injection at 10≲z≲10010 \lesssim z \lesssim 100, there exist parameter regions where τ≃0.08−0.10τ\simeq 0.08-0.10 is compatible with the data.

Alleviating CMB-BAO tension with a large cosmic birefringence angle during reionization induced by energy injection to a

Abstract

Introduction

Recent measurements of the cosmic microwave background (CMB) and baryon acoustic oscillations (BAO) exhibit a mild tension within spatially flat Λ\Lambda cold dark matter (Λ\LambdaCDM) cosmology [1, 2]. In particular, cosmological analyses prefer a sum of neutrino masses below the minimum allowed by oscillation experiments [3]. When the effective neutrino mass is extended to negative values, the minimal mass is disfavored at approximately 3σ3\sigma, although the precise significance depends on the adopted Planck likelihood [3, 4, 5, 6, 7, 8, 9]. One possible explanation is a shift in the CMB reionization optical depth, τ\tau [10, 11, 12, 13]. If the Planck low-ℓ\ell polarization data are excluded, a larger value, such as τ=0.09\tau= 0.09, substantially reduces the neutrino-mass tension and weakens the preference for dynamical dark energy, as well as the lensing and curvature anomalies [10, 12]. However, this value is inconsistent with the Planck low-ℓ\ell measurement at 3.8σ3.8\sigma [12].

Another intriguing hint of beyond the Λ\LambdaCDM model is cosmic birefringence, a rotation of the polarization plane of light during propagation (see Refs. [14, 15] for reviews). Minami & Komatsu [16] first reanalyzed Planck polarization data and reported a hint for isotropic birefringence at 2.4σ2.4\sigma statistical significance. See also Refs. [17, 18, 19] for other analyses of the birefringence angle with Planck data. The analysis is improved by adding Wilkinson Microwave Anisotropy Probe polarization data, finding a rotation angle β≃0.34∘\beta\simeq0.34^\circ at a 3.6σ3.6\sigma preference [20]. The latest constraint is obtained by a joint analysis of the Planck and ACT data that yields a birefringence angle of (0.277±0.057)∘(0.277 \pm0.057)^\circ, which is non-zero with a statistical significance of 4.8σ4.8\sigma [21]. Because cosmic birefringence violates parity, it provides a sensitive probe of new physics beyond both Λ\LambdaCDM and the Standard Model [22]. Cosmic birefringence can be generated by a pseudoscalar field ϕ\phi, such as an axion-like particle (ALP), coupled to electromagnetism through a Chern–Simons interaction,

L⊃−14gϕFμνF~μν,(1)\mathcal{L} \supset-\frac{1}{4}g\phi F_{\mu\nu}\widetilde{F}^{\mu\nu}, \tag*{(1)}

where gg is the coupling constant, FμνF_{\mu\nu} is the electromagnetic field-strength tensor, and F~μν\widetilde{F}^{\mu\nu} is its dual. This mechanism has been studied in models of axion dark energy [23, 24, 25, 26, 27, 28, 29, 30, 31, 32], early dark energy [27, 33, 34, 35], and axion dark matter [36, 37, 38, 39, 40, 41]. Other proposed sources include topological defects [42, 43, 44, 45, 46] and quantum-gravity effects [47, 48, 49].

One possible scenario to explain the above two anomalies is to introduce a large rotation of cosmic birefringence during the reionization epoch [50]. If cosmic birefringence is induced by a single ALP, the ambiguity of the phase of the observed cosmic birefringence [51] could lead to a large rotation angle during the reionization while keeping the recombination birefringence signal unchanged. Alternatively, we can also consider a scenario originally proposed in Graham et al. (2025) [52] where part of the energy of matter transfers to ALPs.

In this paper, we develop the scenario of Ref. [52] to quantitatively constrain that scenario. We use Planck data to search for parameter regions that reproduce the observed data and to show whether this model explains the recent CMB-BAO tension.

This paper is organized as follows. In Sec. II, we review cosmic birefringence induced by ALPs and introduce the energy injection model. In Sec. III, we present constraints on cosmic-birefringence scenarios using data from Planck. Finally, Sec. IV is devoted to a summary of our results and conclusions.

Cosmic Birefringence and Energy Injection Model

CMB angular power spectra

Let us first introduce the CMB angular power spectra induced by cosmic birefringence from ALPs [24, 33, 36, 53, 54, 55, 56, 57, 58, 59]. When cosmic birefringence is sourced by an ALP field, the rotation angle becomes time-dependent. The net rotation angle for a photon observed today and emitted at conformal time η\eta is given by [60, 61, 62]

β(η)=g2[ϕ(η0)−ϕ(η)]=gϕini2ϕ(η0)−ϕ(η)ϕini,(2)\beta(\eta) = \frac{g}{2}[\phi(\eta_0)-\phi(\eta)] = \frac{g\phi_{\mathrm{ini}}}{2}\frac{\phi(\eta_0)-\phi(\eta)}{\phi_{\mathrm{ini}}}, \tag*{(2)}

where η0\eta_0 is the present conformal time and ϕini\phi_{\mathrm{ini}} is the initial value of ϕ\phi. Throughout this paper, the overall amplitude gϕini/2g\phi_{\mathrm{ini}}/2 is fixed so that the present-day rotation angle of the CMB photons emitted at recombination is β0≃0.34∘\beta_0 \simeq0.34^\circ [20]. Since β(η)\beta(\eta) scales linearly with gϕini/2g\phi_{\mathrm{ini}}/2, varying β0\beta_0 within its uncertainty would only shift the allowed values of the energy injection, without changing our results qualitatively. To compute the impact of this time-dependent rotation on the CMB polarization, we solve the Boltzmann equation for the polarized photon distribution [24, 36, 53, 54, 63]:

±2ΔP′+iqν ±2ΔP=aneσT[−±2ΔP+6π5 ±2Y20(ν)Π(η,q)]±igϕ′ ±2ΔP,(3)\pm2\Delta'_P + iq\nu\,\pm2\Delta_P = an_e\sigma_T\left[-\pm2\Delta_P+\sqrt{\frac{6\pi}{5}}\,\pm2Y_{20}(\nu)\Pi(\eta,q)\right]\pm ig\phi'\,\pm2\Delta_P, \tag*{(3)}

where ±2ΔP\pm2\Delta_P are the Fourier modes of the linear polarization Q±iUQ \pm iU and are functions of conformal time η\eta, wavevector magnitude qq, and angle cosine between line-of-sight direction and wavevector ν\nu. The other symbols are: aa (scale factor), nen_e (electron number density), σT\sigma_T (Thomson scattering cross-section), Π\Pi (polarization source term) [64], and ±2Y20\pm2Y_{20} (spin-two spherical harmonics). Primes denote derivatives with respect to conformal time.

To derive the angular power spectra, ±2ΔP\pm2\Delta_P is expanded as [64]:

±2ΔP(η0,q,ν)≡−∑ℓiℓ4π(2ℓ+1)[ΔE,ℓ(q)±iΔB,ℓ(q)] ±2Yℓ0(ν),(4)\pm2\Delta_P(\eta_0,q,\nu)\equiv-\sum_{\ell}i^\ell\sqrt{4\pi(2\ell+1)}\left[\Delta_{E,\ell}(q)\pm i\Delta_{B,\ell}(q)\right]\,_{\pm2}Y_{\ell0}(\nu), \tag*{(4)}

where the formal solution for the EE- and BB-mode sources is given as a line-of-sight integral [24]:

ΔE,ℓ(q)±iΔB,ℓ(q)=34(ℓ+2)!(ℓ−2)!∫0η0dη aneσTe−τ(η)Πjℓ(x)x2e±2iβ(η).(5)\Delta_{E,\ell}(q)\pm i\Delta_{B,\ell}(q)=\frac{3}{4}\sqrt{\frac{(\ell+2)!}{(\ell-2)!}}\int_0^{\eta_0}d\eta\,an_e\sigma_Te^{-\tau(\eta)}\Pi\frac{j_\ell(x)}{x^2}e^{\pm2i\beta(\eta)}. \tag*{(5)}

Here, jℓ(x)j_\ell(x) is the spherical Bessel function, x=q(η0−η)x=q(\eta_0-\eta), and τ(η)=∫ηη0dη1aneσT\tau(\eta)=\int_\eta^{\eta_0}d\eta_1an_e\sigma_T. Note that the EE-mode generated at time η\eta is multiplied by cos⁡[2β(η)]\cos[2\beta(\eta)]. Thus, a large variation of β(η)\beta(\eta) during reionization leads to cancellations of signals among different epochs, resulting in a suppression of the reionization bump signal in the EEEE power spectrum and mimicking a lower τ\tau [50]. The CMB angular power spectrum is then calculated as

CℓXY=4π∫d(ln⁡q) Ps(q)ΔX,ℓ(q)ΔY,ℓ(q),(6)C_\ell^{XY}=4\pi\int d(\ln q)\,\mathcal{P}_s(q)\Delta_{X,\ell}(q)\Delta_{Y,\ell}(q), \tag*{(6)}

where X,Y∈E,BX,Y\in E,B and Ps(q)\mathcal{P}_s(q) is the dimensionless primordial scalar curvature power spectrum. We systematically compute all CMB polarization auto- and cross-power spectra in Eq. (6) by appropriately modifying the polarization transfer function, ΔX,ℓ\Delta_{X,\ell}, within the CLASS code [65, 66].

Energy injection from CDM to ALPs

We now incorporate a large rotation angle realized by energy injection to ALPs from CDM. We consider the following phenomenological model approach often used for CDM / dark energy interactions [52, 67, 68, 69, 70, 71]; we modify the continuity equation that transfers part of the DM energy density to ALPs:

ρˉdm′+3Hρˉdm=−Q,(7)\bar{\rho}'_{\mathrm{dm}}+3\mathcal{H}\bar{\rho}_{\mathrm{dm}}=-Q, \tag*{(7)}
ρˉϕ′+3H(ρˉϕ+pˉϕ)=Q,(8)\bar{\rho}'_{\phi}+3\mathcal{H}(\bar{\rho}_{\phi}+\bar{p}_{\phi})=Q, \tag*{(8)}

where ρˉdm\bar{\rho}_{\mathrm{dm}} is the energy density of CDM, ρˉϕ\bar{\rho}_{\phi} is the energy density of ALPs, pˉϕ\bar{p}_{\phi} is the pressure of ALPs, H=a′/a\mathcal{H}=a'/a, QQ is the rate of energy transfer.

We rewrite the continuity equation in terms of the pseudo-scalar field as

ϕ′a2(ϕ′′+2Hϕ′+a2V,ϕ)=Q,(9)\frac{\phi'}{a^2}(\phi''+2\mathcal{H}\phi'+a^2V_{,\phi})=Q, \tag*{(9)}

where V,ϕ(ϕ)V_{,\phi}(\phi) is the partial derivative of the potential V(ϕ)V(\phi) with respect to ϕ\phi. In the following, we assume a cosine potential V(ϕ)=mϕ2fϕ2[1−cos⁡(ϕ/fϕ)]V(\phi)=m_\phi^2f_\phi^2[1-\cos(\phi/f_\phi)] where fϕf_\phi is the decay constant. We assume that ALP is a spectator field and that the energy density of ALPs is negligible in the background evolution. Thus, even a small amount of energy transfer from matter can significantly enhance the energy density of ALPs, and the amplitude of ϕ\phi. Note that in the absence of the energy transfer, we obtain the usual Klein-Gordon equation:

ϕ′′+2Hϕ′+a2V,ϕ=0.(10)\phi''+2\mathcal{H}\phi'+a^2V_{,\phi}=0. \tag*{(10)}

We assume the following energy transfer rate (e.g., [69, 70])

Q=Hξρˉdm.(11)Q = H \xi\bar{\rho}_{\rm dm}. \tag*{(11)}

Instead of using a specific functional form for ξ\xi, we constrain ξ\xi in a model independent way; we assume that ξ\xi is a constant during a specific short period starting at the redshift of z=zstartz=z_{\rm start} and ending at zendz_{\rm end}. Substituting the above equation into Eq. (9) yields

ϕ′′+2Hϕ′+a2V,ϕ=a2Hξρˉdmϕ′,(12)\phi^{\prime\prime}+2H\phi^{\prime}+a^2V_{,\phi}=\frac{a^2H\xi\bar{\rho}_{\rm dm}}{\phi^{\prime}}, \tag*{(12)}

when ϕ′≠0\phi^{\prime}\ne0. In our calculation, ϕ′≠0\phi^{\prime}\ne0 is always satisfied during the energy injection period.

To realize a large rotation angle, the ALP field is initially located near the bottom of the potential. We specifically assume the initial condition that satisfies

θini≡ϕinifϕ=0.01≪1,(13)\theta_{\rm ini}\equiv\frac{\phi_{\rm ini}}{f_\phi}=0.01\ll1, \tag*{(13)}
ϕini′=0.(14)\phi_{\rm ini}^{\prime}=0. \tag*{(14)}

This choice, together with β0≃0.34∘\beta_0\simeq0.34^\circ, implies gfϕ≃1gf_\phi\simeq1, i.e., an anomaly coefficient C=2πgfϕ/α∼103C=2\pi gf_\phi/\alpha\sim10^3 for g=αC/(2πfϕ)g=\alpha C/(2\pi f_\phi). Such large values have been considered in previous models of cosmic birefringence [28]. In this case, the initial ALP energy density is ρˉϕ,ini=V(ϕini)≃mϕ2ϕini2/2\bar{\rho}_{\phi,\rm ini}=V(\phi_{\rm ini})\simeq m_\phi^2\phi_{\rm ini}^2/2, where the last equality holds at the bottom of the potential well. Considering that ρˉdm=ρˉdm,0a−3\bar{\rho}_{\rm dm}=\bar{\rho}_{\rm dm,0}a^{-3}, and denoting x≡ϕ/ϕinix\equiv\phi/\phi_{\rm ini}, the evolution equation is rewritten as

x′′+2Hx′+a2V,x(x)=Hax′mϕ22ξρˉdm,0ρˉϕ,ini,(15)x^{\prime\prime}+2Hx^{\prime}+a^2V_{,x}(x)=\frac{H}{ax^{\prime}}\frac{m_\phi^2}{2}\xi\frac{\bar{\rho}_{\rm dm,0}}{\bar{\rho}_{\phi,\rm ini}}, \tag*{(15)}

where ρˉdm,0\bar{\rho}_{\rm dm,0} is the present energy density of DM and V(x)=mϕ2ϕini2[1−cos⁡(θinix)]V(x)=m_\phi^2\phi_{\rm ini}^2[1-\cos(\theta_{\rm ini}x)].

It is convenient to define

F≡ρˉϕ,iniρˉdm,0,(16)F\equiv\frac{\bar{\rho}_{\phi,\rm ini}}{\bar{\rho}_{\rm dm,0}}, \tag*{(16)}
ξ~≡ξF=ξρˉdm,0ρˉϕ,ini.(17)\widetilde{\xi}\equiv\frac{\xi}{F}=\xi\frac{\bar{\rho}_{\rm dm,0}}{\bar{\rho}_{\phi,\rm ini}}. \tag*{(17)}

For fixed θini=0.01\theta_{\rm ini}=0.01, Eq. (15) depends on ξ\xi and FF only through ξ~\widetilde{\xi}: the evolution of x=ϕ/ϕinix=\phi/\phi_{\rm ini}, the birefringence angle, and the CMB spectra are all functions of ξ~\widetilde{\xi} alone. FF (and also ξ\xi) is instead a free parameter that the CMB spectra cannot determine. FF only rescales the physical injection rate ξ=ξ~F\xi=\widetilde{\xi}F and the initial ALP energy, without affecting ξ~\widetilde{\xi}.

Figure 1 shows an example of the evolution of the ALP field. We choose ξ~∼O(10)\widetilde{\xi}\sim O(10) that leads to a large birefringence angle. In the following, we focus on the rescaled energy injection rate ξ~\widetilde{\xi} instead of ξ\xi.

Figure 1. Time evolution of the cosmic birefringence angle β(z)\beta(z) for log⁡10(mϕ/eV)=−30.6\log_{10}(m_\phi/{\rm eV})=-30.6 and an energy-injection of ξ~=22\widetilde{\xi}=22 (blue curve), with injection occurring in the redshift interval 50≲z≲6050\lesssim z\lesssim60 (red band). The blue band marks the reionization epoch as reference. The rotation angle is null before the injection takes place; once the injection begins, ϕ\phi is kicked out of its potential minimum and starts to oscillate, producing the large swing in β(z)\beta(z). The oscillations are subsequently damped by Hubble friction, and β(z)\beta(z) asymptotically approaches the present-day birefringence angle β0≃0.34∘\beta_0\simeq0.34^\circ. The phase model (black curve) used in [50] is shown for comparison.

We note that ξ~\widetilde{\xi} has an upper bound to realize a large birefringence angle. To see this, let us consider the case

if too much energy is injected into ALPs by increasing ξ~\widetilde{\xi}, and ϕ\phi crosses over the peak of the cosine potential, ϕ=kπfϕ\phi=k\pi f_\phi, k∈Zk\in\mathbb{Z}. In this situation, the ALP no longer oscillates around ϕ=0\phi=0 and instead settles into oscillations around a different minimum, ϕmin\phi_{\rm min}. Then, the resulting cosmic birefringence angle for the CMB photons emitted at recombination becomes approximately gfϕmin/2≃0.34∘gf_{\phi_{\rm min}}/2\simeq0.34^\circ. Since the field value varies within ∼ϕmin\sim\phi_{\rm min}, we cannot realize a large rotation angle. Thus, there is a critical value of the energy transfer efficiency, ξ~=ξ~crit\widetilde{\xi}=\widetilde{\xi}_{\rm crit}, beyond which this kick-out occurs. Because the driving term is time-dependent and the injection acts only over a finite redshift interval, ξ~crit\widetilde{\xi}_{\rm crit} cannot be obtained in closed form and is instead determined numerically for each ALP mass and each injection window. Figure 2 shows ξ~crit\widetilde{\xi}_{\rm crit} as a function of the injection redshift window for log⁡10(mϕ/eV)=−30.4,−30.5,−30.6\log_{10}(m_\phi/{\rm eV})=-30.4,-30.5,-30.6. For the three representative masses shown, ξ~crit(z)\widetilde{\xi}_{\rm crit}(z) is non-monotonic, exhibiting a minimum around z∼30z\sim30–5050. This minimum marks the redshift bin where energy injection couples most efficiently to the field, corresponding to the epoch when the field starts to be released from Hubble friction, H(z)∼mϕH(z)\sim m_\phi, shortly before the onset of coherent oscillations at mϕ≃3H(zosc)m_\phi\simeq3H(z_{\rm osc}): injecting energy at this epoch requires the smallest ξ~\widetilde{\xi} to kick the field over the potential barrier, while injection well before (field still frozen by Hubble friction) or well after (field already oscillating and decaying) is less efficient and requires larger ξ~\widetilde{\xi}. Consistently, the minimum shifts to lower redshift for the lighter ALP, tracking the mass dependence of the epoch H(z)∼mϕH(z)\sim m_\phi.

Figure 2. Critical energy-injection efficiency ξ~crit\tilde{\xi}_{\mathrm{crit}} as a function of the injection redshift window, for log⁡10(mϕ/eV)=−30.4,−30.5,−30.6\log_{10}(m_\phi/\mathrm{eV})=-30.4,-30.5,-30.6. The non-monotonic trend, with a minimum around z∼30z\sim30–5050 that shifts to lower redshift for lighter masses, reflects the efficient coupling between the injection and the epoch H(z)∼mϕH(z)\sim m_\phi, when the field starts to be released from Hubble friction.

Constraints on ξ\xi and FF

As discussed in Sec. II B, the CMB spectra depend on the energy injection only through ξ~=ξ/F\tilde{\xi}=\xi/F, where F=ρˉϕ,ini/ρˉdm,0F=\bar{\rho}_{\phi,\mathrm{ini}}/\bar{\rho}_{\mathrm{dm},0} is a free parameter. Consequently, ξ\xi and FF are degenerate and cannot be independently constrained by the CMB spectra alone. However, the physical energy exchanged between CDM and ALPs depends on both ξ=ξ~F\xi=\tilde{\xi}F and FF. Since the energy densities of CDM and ALPs are constrained by CMB and large-scale structure observations, these constraints impose additional restrictions on the allowed values of ξ\xi and FF. We consider the following two conditions:

  1. CDM depletion: The energy transferred to the ALP is extracted from CDM. Since the CDM energy density is tightly constrained to be close to that in the best-fit Λ\LambdaCDM cosmology, the fraction of CDM depleted must be smaller than the observational uncertainty in the CDM energy density. As the CDM energy density is proportional to the physical density parameter Ωdmh2\Omega_{\mathrm{dm}}h^2, we conservatively require

1−ρˉdmρˉdmΛCDM∣a=1<σ(Ωdmh2)≃10−2.(18)1-\left.\frac{\bar{\rho}_{\mathrm{dm}}}{\bar{\rho}_{\mathrm{dm}}^{\Lambda\mathrm{CDM}}}\right|_{a=1}<\sigma(\Omega_{\mathrm{dm}}h^2)\simeq10^{-2}. \tag*{(18)}

Here, the 1 σ1\,\sigma uncertainty is taken from Ref. [72].

  1. Spectator condition: The ALP energy density must remain negligible compared to that of CDM; otherwise, it could significantly modify the background expansion history, potentially conflicting with large-scale structure observations. Following Ref. [27], we impose Ωϕh2<0.006\Omega_\phi h^2<0.006, where Ωϕ\Omega_\phi denotes the present-day ALP density parameter. This condition can be more conservatively expressed as

ρˉϕ,0ρˉdm,0=Ωϕh2Ωdmh2≲10−2,(19)\frac{\bar{\rho}_{\phi,0}}{\bar{\rho}_{\mathrm{dm},0}}=\frac{\Omega_\phi h^2}{\Omega_{\mathrm{dm}}h^2}\lesssim10^{-2}, \tag*{(19)}

where we have used Ωdmh2≃0.1\Omega_{\mathrm{dm}}h^2\simeq0.1 [72].

These two conditions do not alter the predicted CMB spectra at fixed ξ~\tilde{\xi}, but they restrict the values of ξ\xi and FF for which a given ξ~\tilde{\xi} is physically viable.

Satisfying these conditions also justifies the approximations adopted in our numerical calculations: the background CDM energy density is fixed to its best-fit Λ\LambdaCDM value, and the ALP is treated as a spectator field whose contribution to the background expansion is negligible.

In the following, we derive constraints on ξ\xi and FF from these two conditions.

CDM depletion. Integrating Eq. (7) with QQ of Eq. (11) gives ρˉdm∝a−(3+ξ)\bar{\rho}_{\mathrm{dm}}\propto a^{-(3+\xi)} for zend≤z≤zstartz_{\mathrm{end}}\leq z\leq z_{\mathrm{start}} and ρˉdm∝a−3\bar{\rho}_{\mathrm{dm}}\propto a^{-3} otherwise. The ratio to the Λ\LambdaCDM density after the injection is therefore

R(ξ)≡ρˉdmρˉdmΛCDM=(1+zstart1+zend)−ξ,(20)R(\xi)\equiv\frac{\bar{\rho}_{\mathrm{dm}}}{\bar{\rho}_{\mathrm{dm}}^{\Lambda\mathrm{CDM}}}=\left(\frac{1+z_{\mathrm{start}}}{1+z_{\mathrm{end}}}\right)^{-\xi}, \tag*{(20)}

and Eq. (18) becomes a limit on ξ\xi only,

ξ<ξdep=−ln⁡(0.99)ln⁡(1+zstart1+zend),(21)\xi<\xi_{\mathrm{dep}}=-\frac{\ln(0.99)}{\ln\left(\frac{1+z_{\mathrm{start}}}{1+z_{\mathrm{end}}}\right)}, \tag*{(21)}

which depends only on the injection window and is independent of mϕm_\phi and FF.

Spectator condition. The ALP energy density has two contributions: the initial energy density ρˉϕ,ini\bar{\rho}_{\phi,\mathrm{ini}}, frozen in the potential until the field starts to oscillate, and the energy injected by the source term QQ. The ratio ρˉϕ/ρˉdm\bar{\rho}_\phi/\bar{\rho}_{\mathrm{dm}} grows while the field is frozen, since ρˉdm\bar{\rho}_{\mathrm{dm}} dilutes as a−3a^{-3}, once the field starts to oscillate, ρˉϕ\bar{\rho}_\phi also dilutes as a−3a^{-3} and the ratio settles to a plateau. We therefore evaluate Eq. (19) today, a=1a=1, where the plateau value is reached and which is representative of the maximum of the ratio over the whole evolution. Using the results in Appendix A, we find

ρˉϕ,0ρˉdm,0=F1R(ξ)[a3ρˉϕ(a)ρˉϕ,ini]a=1.(22)\frac{\bar{\rho}_{\phi,0}}{\bar{\rho}_{\mathrm{dm},0}}=F\frac{1}{R(\xi)}\left[a^3\frac{\bar{\rho}_\phi(a)}{\bar{\rho}_{\phi,\mathrm{ini}}}\right]_{a=1}. \tag*{(22)}

This ratio is not simply proportional to FF at fixed ξ\xi: Eq. (15) depends on FF and ξ\xi only through ξ~\tilde{\xi}, so that ρˉϕ(a)/ρˉϕ,ini\bar{\rho}_\phi(a)/\bar{\rho}_{\phi,\mathrm{ini}} is a function of ξ~\tilde{\xi}. Defining the transfer function

G(ξ,F)≡[a3ρˉϕ(a)ρˉϕ,ini]a=1=G(ξ~),(23)\mathcal{G}(\xi,F)\equiv\left[a^3\frac{\bar{\rho}_\phi(a)}{\bar{\rho}_{\phi,\mathrm{ini}}}\right]_{a=1}=\mathcal{G}(\tilde{\xi}), \tag*{(23)}

which depends on ξ\xi and FF only through their ratio ξ~=ξ/F\tilde{\xi}=\xi/F, Eq. (22) reads

ρˉϕ,0ρˉdm,0=FG(ξ~)R(ξ).(24)\frac{\bar{\rho}_{\phi,0}}{\bar{\rho}_{\rm dm,0}}=F\frac{\mathcal{G}(\tilde{\xi})}{R(\xi)}. \tag*{(24)}

The function G(ξ~)\mathcal{G}(\tilde{\xi}) includes non-linear effects of the injection, such as the kinetic phase that follows a strong kick and the delay in the onset of oscillations, which analytical approximations cannot reproduce (Appendix A). We therefore compute it numerically, solving Eq. (15) once for each ξ~\tilde{\xi} on a grid 0≤ξ~≤ξ~crit0\le\tilde{\xi}\le\tilde{\xi}_{\rm crit}, and then map it onto the (ξ,F)(\xi,F) plane using ξ~=ξ/F\tilde{\xi}=\xi/F. The allowed region is the intersection of:

  1. the kick-out limit, ξ~≤ξ~crit(mϕ)\tilde{\xi}\le\tilde{\xi}_{\rm crit}(m_\phi);

  2. the spectator limit, FG(ξ~)/R(ξ)≤10−2F\mathcal{G}(\tilde{\xi})/R(\xi)\le10^{-2};

  3. the depletion limit ξ<ξdep\xi<\xi_{\rm dep}.

Figure 3 shows the result for log⁡10(mϕ/eV)=−30.5\log_{10}(m_\phi/\mathrm{eV})=-30.5. Since the CMB spectra depend only on ξ~\tilde{\xi}, the choice of FF only moves the physical value of ξ=ξ~F\xi=\tilde{\xi}F. For F≲10−3F\lesssim10^{-3} the whole range ξ~≤ξ~crit\tilde{\xi}\le\tilde{\xi}_{\rm crit} satisfies both conditions for the three masses considered, and all the results of Sec. III are therefore presented in terms of ξ~\tilde{\xi}, up to ξ~crit\tilde{\xi}_{\rm crit}.

Parameter Constraints

Having established the background physics of the injection term, here we want to present a systematic study on how the energy injection and the large cosmic birefringence angle generated from it is actually a model able to suppress the reionization bump in the case of a high value of the optical depth τ\tau. We have performed this analysis for axion masses −31≤log⁡10(mϕ/eV)≤−30-31\le\log_{10}(m_\phi/\mathrm{eV})\le-30, in steps of 0.10.1, and for all the injection redshift windows shown in Figure 2, finding qualitatively similar results. In the following, we present the results for three representative mass values, log⁡10(mϕ/eV)={−30.4,−30.5,−30.6}\log_{10}(m_\phi/\mathrm{eV})=\{-30.4,-30.5,-30.6\}, injecting energy only within the redshift window 50≤z≤6050\le z\le60.

Data and methodology

In this work we have considered the Planck NPIPE low-ℓ\ell lollipop data [73, 74, 75], considering the EEEE, BBBB, and EBEB spectra and the corresponding NPIPE low-ℓ\ell covariance matrix. To compute the theoretical angular power spectra including the energy-injection mechanism described in Sec. II B, we have developed birefINJ2, a modified version of the Boltzmann code CLASS [65, 66]. birefINJ implements the birefringence-induced source term of Eq. (5) in the computation of the angular power spectra, Eq. (6), following the approach of birefCLASS [56], together with the axion-field dynamics under energy injection described in Eq. (12).

Once the field dynamics and the resulting angular power spectra are computed, we evaluate the results through a χ2\chi^2 statistics performed in a EEEE-only case or in a joint case involving EEEE, BBBB, and EBEB spectra. We adopt as fiducial cosmology the best-fit Λ\LambdaCDM model of Ref. [12], obtained without low-ℓ\ell polarization data. When varying τ\tau, we keep Ase−2τA_s e^{-2\tau} fixed to its fiducial value, where AsA_s is the scalar amplitude, so that the high-ℓ\ell CMB spectra are unchanged and only the low-ℓ\ell reionization signal is affected. We do not include the TTTT spectrum in the analysis. Cosmic birefringence does not affect the temperature anisotropies, so that TTTT is independent of ξ~\tilde{\xi}; moreover, with Ase−2τA_s e^{-2\tau} fixed, increasing τ\tau from 0.0540.054 to 0.090.09 changes the low-ℓ\ell TTTT spectrum by less than 10%10\%, well below the cosmic variance at these multipoles. We also do not include the low-ℓ\ell TETE spectrum, whose constraining power on τ\tau is limited by its large uncertainties.

Denoting by S\mathcal{S} the set of spectra entering a given analysis — either S={EE}\mathcal{S}=\{EE\} for the EEEE-only case or S={EE,BB,EB}\mathcal{S}=\{EE,BB,EB\} for the joint analysis — we define the data vector

D⃗X(ξ~,τ)≡DℓX,th(ξ~,τ)−DℓX,obs,X∈S,(25)\vec{D}^{X}(\tilde{\xi},\tau)\equiv D_{\ell}^{X,\mathrm{th}}(\tilde{\xi},\tau)-D_{\ell}^{X,\mathrm{obs}},\qquad X\in\mathcal{S}, \tag*{(25)}

and the concatenated vector D⃗(ξ~,τ)=⨁X∈SD⃗X(ξ~,τ)\vec{D}(\tilde{\xi},\tau)=\bigoplus_{X\in\mathcal{S}}\vec{D}^{X}(\tilde{\xi},\tau), obtained by stacking each spectrum in S\mathcal{S}. Here DℓX=ℓ(ℓ+1)2πCℓXD_{\ell}^{X}=\frac{\ell(\ell+1)}{2\pi}C_{\ell}^{X} is the angular power spectrum in the standard DℓD_\ell convention. The χ2\chi^2 is then

χ2(ξ~,τ)=D⃗(ξ~,τ)T⋅Cov−1⋅D⃗(ξ~,τ),(26)\chi^2(\tilde{\xi},\tau)=\vec{D}(\tilde{\xi},\tau)^{T}\cdot\mathrm{Cov}^{-1}\cdot\vec{D}(\tilde{\xi},\tau), \tag*{(26)}

where Cov⁡\operatorname{Cov} is the NPIPE low-ℓ\ell covariance matrix restricted to the spectra in S\mathcal{S}: for the EEEE-only case Cov⁡\operatorname{Cov} reduces to the single EEEE block, while for the joint analysis Cov⁡\operatorname{Cov} is the full 3Nℓ×3Nℓ3N_{\ell} \times3N_{\ell} block covariance correlating EEEE, BBBB, and EBEB, with NℓN_{\ell} the number of multipoles. In this analysis Nℓ=29N_{\ell}=29, having restricted the analysis to the multipole range 2≤ℓ≤302 \le\ell\le30. The corresponding likelihood is L(ξ~,τ)∝e−χ2(ξ~,τ)/2\mathcal{L}(\tilde{\xi},\tau) \propto e^{-\chi^{2}(\tilde{\xi},\tau)/2}. We evaluate χ2(ξ~,τ)\chi^{2}(\tilde{\xi},\tau) using two independent methods: a Markov Chain Monte Carlo (MCMC) sampler and a direct evaluation on a two-dimensional (ξ~,τ)(\tilde{\xi},\tau) grid at fixed ALP mass. The two approaches have been cross-validated against each other, yielding consistent results; in the following we present the results obtained from the grid-based likelihood evaluation.

Power spectra

As recently proposed in the literature [10, 11, 12, 13], raising the optical depth to τ∼0.09\tau\sim0.09 can help alleviate the CMB-BAO tension. However, at this value of τ\tau and in the absence of injection, the resulting EEEE reionization bump clearly overshoots the Planck data (blue curve in Figure 4). Introducing a large cosmic birefringence angle sourced by the energy-injection mechanism of Eq. (12) suppresses the bump back down to a level consistent with the data (green curve in Figure 4), closely tracking the standard, low-τ\tau case (orange curve in Figure 4). The same qualitative behavior, but with different values of ξ~\tilde{\xi} and τ\tau, is observed for the other masses considered in this work and is therefore not shown separately.

Figure 4. EEEE (upper left), BBBB (upper right), and EBEB (center) angular power spectra for log⁡10(mϕ/eV)=−30.5\log_{10}(m_\phi/\mathrm{eV})=-30.5, compared to the Planck NPIPE low-ℓ\ell data (black points). Orange: τ=τPlanck=0.054\tau=\tau_{\mathrm{Planck}}=0.054 with no injection. Blue: τ=0.09>τPlanck\tau=0.09>\tau_{\mathrm{Planck}} with no injection where the resulting EEEE bump overshoots the observed reionization signal. Green: τ=0.09\tau=0.09 but with an injection of ξ~=11.3\tilde{\xi}=11.3, below the kick-out limit for this mass, ξ~crit=16.3\tilde{\xi}_{\mathrm{crit}}=16.3; in this case, the injection suppresses the EEEE bump back to a level consistent with the data. The same injection also imprints a distinctive dip in DℓEBD_{\ell}^{EB} and a corresponding bump in DℓBBD_{\ell}^{BB} around ℓ∼5\ell\sim5–88, the direct signature of the late-time rotation sourced by the axion dynamics.

The BBBB and EBEB panels of Figure 4 show that the injection produces a genuine, visible change in these two channels as well. However, given the current NPIPE low-ℓ\ell uncertainties, all three curves remain statistically consistent with the data in both channels: no single spectrum, taken individually, visually excludes either scenario. This suggests that EBEB and BBBB can carry complementary information to that of EEEE alone, which tightly constrains the shape of the reionization bump but leaves (ξ~,τ)(\tilde{\xi},\tau) highly degenerate. As shown quantitatively in Sec. III C, combining EEEE, BBBB, and EBEB reshapes the allowed (ξ~,τ)(\tilde{\xi},\tau) region into a set of distinct sub-regions rather than a single, simply-connected area; a residual degeneracy nonetheless persists, as we discuss in Sec. III C.

Altogether, Figure 4 shows that there exist values of (ξ~,τ)(\tilde{\xi},\tau) for which a large optical depth remains compatible with the observed EEEE, BBBB, and EBEB spectra, thereby strongly motivating the quantitative parameter analysis presented in the remainder of this section. We note that all these results closely resemble those obtained by Ref. [50] within the nπn\pi-phase model.

Contour plots

Having shown in Sec. III B that a large optical depth can, for suitable values of (ξ~,τ)(\tilde{\xi},\tau), remain qualitatively compatible with the observed EEEE, BBBB, and EBEB spectra, we now turn to a quantitative likelihood analysis of the full (ξ~,τ)(\tilde{\xi},\tau) parameter space. We compare the results with the EEEE-only and joint EE+BB+EBEE+BB+EB cases under both a uniform and a Gaussian prior on τ\tau.

Top block of Figure 5 shows the 1σ1\sigma and 2σ2\sigma confidence contours in the (ξ~,τ)(\tilde{\xi},\tau) plane obtained assuming a uniform prior on τ\tau. In the EEEE-only case (top row), all three masses display a common qualitative structure: a broad, continuous region around τ∼0.05\tau\sim0.05–0.070.07, together with one or more narrow features extending toward higher τ\tau at specific, discrete values of ξ~\tilde{\xi}. This multimodal structure is not a numerical artifact — it reflects the oscillatory nature of the ALP dynamics under a finite-window energy injection, whereby the accumulated rotation phase β(η)\beta(\eta) can reach similar values for several distinct values of ξ~\tilde{\xi}, producing comparably good fits to the EEEE reionization bump at those discrete points.

Figure 5. 1σ1\sigma (dark) and 2σ2\sigma (light) confidence contours in the (ξ~,τ)(\tilde{\xi},\tau) plane for the EE-only (top row of each block) and joint EE+BB+EB (bottom row of each block) analyses, for the three ALP masses log⁡10(mϕ/eV)=−30.4,−30.5,−30.6\log_{10}(m_{\phi}/\mathrm{eV})=-30.4,-30.5,-30.6 (columns). Top block: uniform prior τ∈(0.04,0.12)\tau\in(0.04, 0.12). Bottom block: Gaussian prior τ∼N(τ0=0.0899,σ=0.011)\tau\sim\mathcal{N}(\tau_0=0.0899,\sigma=0.011), motivated by high-ℓ\ell CMB+DESI-BAO constraints [12]. Note that the ξ~\tilde{\xi} range differs across columns, as ξ~crit\tilde{\xi}_{\mathrm{crit}} depends on the ALP mass (cf. Figure 2).

Combining EEEE, BBBB, and EBEB (bottom row) does not simply narrow this structure into a single, better-constrained region. Instead, for all three masses the allowed parameter space is reorganized into a set of distinct, sometimes disconnected sub-regions, some of which persist at values of τ\tau higher than the dominant low-τ\tau feature of the EEEE-only case. This is consistent with the fact that EEEE, BBBB, and EBEB are not independent probes: all three spectra are sourced by the same underlying rotation history where β=β(η,ξ~,τ)\beta=\beta(\eta,\tilde{\xi},\tau), so the additional information carried by BBBB and EBEB reshapes rather than simply shrinks the EEEE-only degeneracy.

Bottom block of Figure 5 shows the same analysis with a Gaussian prior on τ\tau centered at τ0=0.0899\tau_{0}=0.0899, motivated by external CMB+BAO constraints. As expected, for all three masses, this prior consistently selects and reinforces the higher-τ\tau sub-regions that were only marginal features under a uniform prior, while suppressing the low-τ\tau region that dominates the EEEE-only, uniform-prior case. In all three cases, the surviving sub-regions cluster around τ∼0.08\tau\sim0.08–0.100.10, directly tracking the external prior value τ0=0.0899\tau_{0}=0.0899. This result is robust across the three representative masses considered in this work: there consistently exist regions of the (ξ~,τ)(\tilde{\xi},\tau) parameter space, compatible with the joint EE+BB+EBEE+BB+EB likelihood, in which a large optical depth is achieved through the energy-injection mechanism.

Because the allowed parameter space is multimodal rather than unimodal, a single symmetric credible interval does not adequately summarize the joint constraint; we therefore refrain from quoting a single best-fit value with symmetric errors, and instead rely on the sub-regions shown in the two blocks of Figure 5 to convey the structure of the allowed parameter space. To further characterize this multimodal structure, Figure 6 shows the profile likelihood in τ\tau, obtained by maximizing the likelihood over ξ~\tilde{\xi} at each value of τ\tau. Unlike marginalization, this procedure is independent of the prior volume in ξ~\tilde{\xi} and is therefore not affected by volume effects in the degenerate regime. In the joint analysis the profile exhibits several distinct peaks, including some at τ≳0.08\tau\gtrsim0.08, consistent with the large-τ\tau regions of Figure 5 being genuine features of the likelihood.

Figure 6. Profile likelihood in τ\tau, Lprofile(τ)=max⁡ξ~L(ξ~,τ)\mathcal{L}_{\mathrm{profile}}(\tau)=\max_{\widetilde{\xi}}\mathcal{L}(\widetilde{\xi},\tau), normalized to its maximum, for the EE-only (blue) and joint EE+BB+EB (red) analyses and the three ALP masses of Figure 5.

Summary and Discussion

We quantitatively explored the energy injection scenario in which a tiny fraction of the CDM energy density is transferred to the ALP energy density before the reionization epoch. Using a specific model of energy injection, we numerically showed that the ALP field value is enhanced after the energy injection. We then showed that the EEEE power spectrum at low-ℓ\ell is suppressed due to a large rotation angle of cosmic birefringence after the reionization epoch. We found a parameter region of the model that simultaneously explains the Planck low-ℓ\ell polarization data, a nonzero value of the rotation angle of cosmic birefringence obtained from EBEB power spectrum at high-ℓ\ell, and a large value of τ\tau suggested as a possible solution to the CMB-BAO tension. Although we have shown results for a single injection window and three representative masses, we have verified that similar parameter regions exist for all ALP masses −31≤log⁡10(mϕ/eV)≤−30-31\le\log_{10}(m_\phi/\mathrm{eV})\le-30 and injection windows within 10≤z≤10010\le z\le100.

In our constraints on the model parameters, we did not include the instrumental miscalibration angle. The miscalibration angle could introduce a nonzero EBEB power spectrum. However, both the EEEE-only and EE+EB+BBEE+EB+BB cases allow a larger value of τ\tau at a similar value of

ξ~\widetilde{\xi}, suggesting that the impact of the miscalibration angle is not significant. The miscalibration angles at each frequency channel constrained from Ref. [20] are also within sub-degree, indicating that the impact of the miscalibration angle is much smaller than the observational errors of the EBEB power spectra at low-ℓ\ell.

Our model predicts a large rotation angle of cosmic birefringence at low redshift, z<10z<10. Such a large rotation angle can be tested by observations of radio galaxies [76, 77]. Measurements of the low-ℓ\ell polarization by future CMB experiments, such as LiteBIRD [78], are also important to validate the measurement of low-ℓ\ell polarization power spectra.

Acknowledgments

We thank Elisa G. M. Ferreira for valuable suggestions at the early stage of this project. We also thank Luca Visinelli for useful comments on the draft. We acknowledge support from JSPS KAKENHI Grant No. JP22K03682, No. JP24KK0248, and No. JP25K00996, and from JST EXPERT-J, Japan Grant No. JPM-JEX2508. The Kavli IPMU is supported by World Premier International Research Center Initiative (WPI Initiative), MEXT, Japan. This work made use of the idark computing cluster at the Kavli IPMU.

Data availability

The data that support the findings of this article are publicly available at the following GitHub repository: https://github.com/SirlettiSS/birefINJ.

Energy injection and the spectator condition

In this appendix, we derive the evolution of the CDM and ALP energy densities during and after the injection, which are the basis for the depletion and spectator conditions discussed in Sec. II C. We denote by astart=1/(1+zstart)a_{\mathrm{start}}=1/(1+z_{\mathrm{start}}) and aend=1/(1+zend)a_{\mathrm{end}}=1/(1+z_{\mathrm{end}}) the scale factors at the beginning and at the end of the injection window, and by aini≪astarta_{\mathrm{ini}}\ll a_{\mathrm{start}} the scale factor at which the initial conditions are set.

Dark matter evolution and depletion limit

The CDM density obeys Eq. (7) with the source QQ of Eq. (11), i.e.

ρˉ˙dm+3Hρˉdm=−Q,Q=Θ(a)ξHρˉdm={ξHρˉdmastart≤a≤aend,0otherwise.(27)\dot{\bar{\rho}}_{\mathrm{dm}}+3H\bar{\rho}_{\mathrm{dm}}=-Q,\qquad Q=\Theta(a)\xi H\bar{\rho}_{\mathrm{dm}}= \begin{cases} \xi H\bar{\rho}_{\mathrm{dm}} & a_{\mathrm{start}}\le a\le a_{\mathrm{end}},\\ 0 & \text{otherwise}. \end{cases} \tag*{(27)}

where Θ(a)\Theta(a) is the step-function, equal to 1 inside the injection window and 0 outside. Since Q∝HQ \propto\mathcal{H}, the Hubble rate cancels when the equation is written in terms of the scale factor,

dρˉdmda=−3+ξΘ(a)aρˉdm.(28)\frac{d\bar{\rho}_{\mathrm{dm}}}{da}=-\frac{3+\xi\Theta(a)}{a}\bar{\rho}_{\mathrm{dm}}. \tag*{(28)}

The equation can then be solved separately in the three intervals, requiring continuity of ρˉdm\bar{\rho}_{\mathrm{dm}} at astarta_{\mathrm{start}} and aenda_{\mathrm{end}}. Before the injection, there is no source, and ρˉdm\bar{\rho}_{\mathrm{dm}} coincides with the Λ\LambdaCDM solution ρˉdmΛCDM(a)=ρˉdm,0a−3\bar{\rho}^{\Lambda\mathrm{CDM}}_{\mathrm{dm}}(a)=\bar{\rho}_{\mathrm{dm},0}a^{-3}, where ρˉdm,0\bar{\rho}_{\mathrm{dm},0} is the present-day Λ\LambdaCDM value. We obtain

ρˉdm(a)=ρˉdm,0a−3×{1,a≤astart,(aastart)−ξ,astart≤a≤aend,(aendastart)−ξ,a≥aend.(29)\bar{\rho}_{\mathrm{dm}}(a)=\bar{\rho}_{\mathrm{dm},0}a^{-3}\times \begin{cases} 1, & a\leq a_{\mathrm{start}},\\ \left(\dfrac{a}{a_{\mathrm{start}}}\right)^{-\xi}, & a_{\mathrm{start}}\leq a\leq a_{\mathrm{end}},\\ \left(\dfrac{a_{\mathrm{end}}}{a_{\mathrm{start}}}\right)^{-\xi}, & a\geq a_{\mathrm{end}}. \end{cases} \tag*{(29)}

In the window, the density therefore scales as a−(3+ξ)a^{-(3+\xi)}, while after the injection it dilutes again as a−3a^{-3}, but with a reduced normalization.

Dividing by the Λ\LambdaCDM density, the depletion factor after the injection is (cf. Eq. (18))

R(ξ)≡ρˉdmρˉdmΛCDM∣a≥aend=(aendastart)−ξ=(1+zstart1+zend)−ξ,(30)R(\xi)\equiv\left.\frac{\bar{\rho}_{\mathrm{dm}}}{\bar{\rho}^{\Lambda\mathrm{CDM}}_{\mathrm{dm}}}\right|_{a\geq a_{\mathrm{end}}}=\left(\frac{a_{\mathrm{end}}}{a_{\mathrm{start}}}\right)^{-\xi}=\left(\frac{1+z_{\mathrm{start}}}{1+z_{\mathrm{end}}}\right)^{-\xi}, \tag*{(30)}

which depends only on ξ\xi and on the width of the injection window, and is independent of mϕm_\phi and FF. Imposing the upper bound from the data as discussed in Sec. II C, 1−R(ξ)<10−21-R(\xi)<10^{-2}, i.e. R(ξ)>0.99R(\xi)>0.99, gives Eq. (21). For the window 50≤z≤6050\leq z\leq60 used throughout the paper, ξdep≃0.056\xi_{\mathrm{dep}}\simeq0.056.

ALP energy density

The energy density of the ALP obeys the continuity equation with the source QQ of Eq. (11);

ρˉ˙ϕ+3H(1+wϕ)ρˉϕ=ξHρˉdm,(31)\dot{\bar{\rho}}_\phi+3\mathcal{H}(1+w_\phi)\bar{\rho}_\phi=\xi\mathcal{H}\bar{\rho}_{\mathrm{dm}}, \tag*{(31)}

with wϕ=pˉϕ/ρˉϕw_\phi=\bar{p}_\phi/\bar{\rho}_\phi. Using the scale factor as a time variable, this reads

adρˉϕda+3(1+wϕ)ρˉϕ=ξρˉdm(a),(32)a\frac{d\bar{\rho}_\phi}{da}+3(1+w_\phi)\bar{\rho}_\phi=\xi\bar{\rho}_{\mathrm{dm}}(a), \tag*{(32)}

with ρˉdm(a)\bar{\rho}_{\mathrm{dm}}(a) given by Eq. (29), and the initial condition ρˉϕ(aini)=ρˉϕ,ini=Fρˉdm,0\bar{\rho}_\phi(a_{\mathrm{ini}})=\bar{\rho}_{\phi,\mathrm{ini}}=F\bar{\rho}_{\mathrm{dm},0}. Assuming wϕw_\phi constant, the equation can be solved in the three intervals.

a. Interval 1, a≤astarta\leq a_{\mathrm{start}}. There is no source and

ρˉϕ(a)=ρˉϕ(aini)(aaini)−3(1+wϕ).(33)\bar{\rho}_\phi(a)=\bar{\rho}_\phi(a_{\mathrm{ini}})\left(\frac{a}{a_{\mathrm{ini}}}\right)^{-3(1+w_\phi)}. \tag*{(33)}

b. Interval 2, astart<a≤aenda_{\mathrm{start}}<a\leq a_{\mathrm{end}}. The equation to solve is

dρˉϕda+3(1+wϕ)aρˉϕ=ξaρˉdm(aini)(astartaini)−3(aastart)−(3+ξ).(34)\frac{d\bar{\rho}_\phi}{da}+\frac{3(1+w_\phi)}{a}\bar{\rho}_\phi=\frac{\xi}{a}\bar{\rho}_{\mathrm{dm}}(a_{\mathrm{ini}})\left(\frac{a_{\mathrm{start}}}{a_{\mathrm{ini}}}\right)^{-3}\left(\frac{a}{a_{\mathrm{start}}}\right)^{-(3+\xi)}. \tag*{(34)}

It is linear and can be solved exactly with the integrating factor a3(1+wϕ)a^{3(1+w_\phi)}, which turns the left-hand side into a−3(1+wϕ)d[a3(1+wϕ)ρˉϕ]/daa^{-3(1+w_\phi)}d\left[a^{3(1+w_\phi)}\bar{\rho}_\phi\right]/da. The solution is

ρˉϕ(a)=ρˉϕ(astart)(aastart)−3(1+wϕ)+ξ3wϕ−ξρˉdm(astart)[(aastart)−(3+ξ)−(aastart)−3(1+wϕ)].(35)\bar{\rho}_\phi(a)=\bar{\rho}_\phi(a_{\mathrm{start}})\left(\frac{a}{a_{\mathrm{start}}}\right)^{-3(1+w_\phi)}+\frac{\xi}{3w_\phi-\xi}\bar{\rho}_{\mathrm{dm}}(a_{\mathrm{start}})\left[\left(\frac{a}{a_{\mathrm{start}}}\right)^{-(3+\xi)}-\left(\frac{a}{a_{\mathrm{start}}}\right)^{-3(1+w_\phi)}\right]. \tag*{(35)}

with the boundary values

ρˉϕ(astart)=ρˉϕ(aini)(astartaini)−3(1+wϕ),ρˉdm(astart)=ρˉdm(aini)(astartaini)−3.(36)\bar{\rho}_{\phi}(a_{\mathrm{start}})=\bar{\rho}_{\phi}(a_{\mathrm{ini}})\left(\frac{a_{\mathrm{start}}}{a_{\mathrm{ini}}}\right)^{-3(1+w_{\phi})},\qquad\bar{\rho}_{\mathrm{dm}}(a_{\mathrm{start}})=\bar{\rho}_{\mathrm{dm}}(a_{\mathrm{ini}})\left(\frac{a_{\mathrm{start}}}{a_{\mathrm{ini}}}\right)^{-3}. \tag*{(36)}

At a=aenda=a_{\mathrm{end}} this gives

ρˉϕ(aend)=ρˉϕ(aini)(aendaini)−3(1+wϕ)+ρˉdm(aini)(astartaini)−3ξ3wϕ−ξ[(aendastart)−(3+ξ)−(aendastart)−3(1+wϕ)].(37)\begin{aligned} \bar{\rho}_{\phi}(a_{\mathrm{end}})={}&\bar{\rho}_{\phi}(a_{\mathrm{ini}})\left(\frac{a_{\mathrm{end}}}{a_{\mathrm{ini}}}\right)^{-3(1+w_{\phi})}\\ &+\bar{\rho}_{\mathrm{dm}}(a_{\mathrm{ini}})\left(\frac{a_{\mathrm{start}}}{a_{\mathrm{ini}}}\right)^{-3}\frac{\xi}{3w_{\phi}-\xi}\left[\left(\frac{a_{\mathrm{end}}}{a_{\mathrm{start}}}\right)^{-(3+\xi)}-\left(\frac{a_{\mathrm{end}}}{a_{\mathrm{start}}}\right)^{-3(1+w_{\phi})}\right]. \tag*{(37)} \end{aligned}

c. Interval 3, a>aenda>a_{\mathrm{end}}. The source vanishes again, and

ρˉϕ(a)=ρˉϕ(aend)(aaend)−3(1+wϕ).(38)\bar{\rho}_{\phi}(a)=\bar{\rho}_{\phi}(a_{\mathrm{end}})\left(\frac{a}{a_{\mathrm{end}}}\right)^{-3(1+w_{\phi})}. \tag*{(38)}

Spectator ratio and the function GG

The ratio between the ALP and CDM densities follows from writing ρˉdm(a)=ρˉdm,0 a−3R(a)\bar{\rho}_{\mathrm{dm}}(a)=\bar{\rho}_{\mathrm{dm},0}\,a^{-3}R(a) and ρˉϕ,ini=Fρˉdm,0\bar{\rho}_{\phi,\mathrm{ini}}=F\bar{\rho}_{\mathrm{dm},0},

ρˉϕ(a)ρˉdm(a)=F1R(a)a3ρˉϕ(a)ρˉϕ,ini,(39)\frac{\bar{\rho}_{\phi}(a)}{\bar{\rho}_{\mathrm{dm}}(a)}=F\frac{1}{R(a)}a^{3}\frac{\bar{\rho}_{\phi}(a)}{\bar{\rho}_{\phi,\mathrm{ini}}}, \tag*{(39)}

which for a=1a=1 is Eq. (22). The dependence of G≡[a3ρˉϕ/ρˉϕ,ini]a=1G\equiv[a^{3}\bar{\rho}_{\phi}/\bar{\rho}_{\phi,\mathrm{ini}}]_{a=1}, defined in Eq. (23), on FF and ξ\xi follows from the solution in the third interval, Eq. (38), once the constant term of Eq. (37) is expressed in terms of FF and ξ\xi. Evaluating it at a=1a=1 and using ξ~=ξ/F\tilde{\xi}=\xi/F we obtain

G={(aendaini)−3(1+wϕ)+ξ~astart−33wϕ−ξ[(aendastart)−(3+ξ)−(aendastart)−3(1+wϕ)]}aend3(1+wϕ),(40)G=\left\{\left(\frac{a_{\mathrm{end}}}{a_{\mathrm{ini}}}\right)^{-3(1+w_{\phi})}+\tilde{\xi}\frac{a_{\mathrm{start}}^{-3}}{3w_{\phi}-\xi}\left[\left(\frac{a_{\mathrm{end}}}{a_{\mathrm{start}}}\right)^{-(3+\xi)}-\left(\frac{a_{\mathrm{end}}}{a_{\mathrm{start}}}\right)^{-3(1+w_{\phi})}\right]\right\}a_{\mathrm{end}}^{3(1+w_{\phi})}, \tag*{(40)}

where for the second term we have used ρˉdm(aini)=ρˉdm,0aini−3\bar{\rho}_{\mathrm{dm}}(a_{\mathrm{ini}})=\bar{\rho}_{\mathrm{dm},0}a_{\mathrm{ini}}^{-3} (there is no depletion before astarta_{\mathrm{start}}) and ρˉϕ(aini)=ρϕ,ini=Fρˉdm,0\bar{\rho}_{\phi}(a_{\mathrm{ini}})=\rho_{\phi,\mathrm{ini}}=F\bar{\rho}_{\mathrm{dm},0} to obtain

ρˉdm(aini)ρˉϕ(aini)(astartaini)−3=astart−3F.(41)\frac{\bar{\rho}_{\mathrm{dm}}(a_{\mathrm{ini}})}{\bar{\rho}_{\phi}(a_{\mathrm{ini}})}\left(\frac{a_{\mathrm{start}}}{a_{\mathrm{ini}}}\right)^{-3}=\frac{a_{\mathrm{start}}^{-3}}{F}. \tag*{(41)}

The first term in Eq. (40) is the free evolution of the initial energy and does not depend on FF or ξ\xi, while the injected energy is proportional to ξ~\tilde{\xi}. Therefore, GG depends on FF and ξ\xi only through ξ~=ξ/F\tilde{\xi}=\xi/F, up to the explicit dependence on ξ\xi in 3wϕ−ξ3w_{\phi}-\xi and (aend/astart)−ξ(a_{\mathrm{end}}/a_{\mathrm{start}})^{-\xi}, which is at the percent level for ξ≤ξdep\xi\leq\xi_{\mathrm{dep}}. The ratio at a=1a=1 is then FG(ξ~)/R(ξ)FG(\tilde{\xi})/R(\xi), which is the expression used in Eq. (22) to compute the ALP spectator limit in Eq. (19).

Limits of the analytical approximation

Equation (A14) is obtained for constant wϕw_{\phi}. To first approximation, the evolution of the field can be described by a sharp and instantaneous transition, at a scale factor atrana_{\mathrm{tran}}, from wϕ=−1w_{\phi}=-1, when the field is frozen in the potential by Hubble friction, to wϕ=0w_{\phi}=0, when it oscillates around the minimum and behaves as matter. A natural choice is to place the transition at the end of the injection, atran=aenda_{\mathrm{tran}}=a_{\mathrm{end}}. For wϕ=−1w_{\phi}=-1 up to atrana_{\mathrm{tran}} and wϕ=0w_{\phi}=0 afterwards, Eq. (40) becomes

G=atran3{1+ξ~astart−33+ξ[1−(aendastart)−(3+ξ)]}.(42)G=a_{\mathrm{tran}}^{3}\left\{1+\tilde{\xi}\frac{a_{\mathrm{start}}^{-3}}{3+\xi}\left[1-\left(\frac{a_{\mathrm{end}}}{a_{\mathrm{start}}}\right)^{-(3+\xi)}\right]\right\}. \tag*{(42)}

This is a good approximation of the exact result. As an example, Figure 7 compares the analytical ratio ρˉϕ/ρˉdm\bar{\rho}_{\phi}/\bar{\rho}_{\mathrm{dm}} with the transition at aenda_{\mathrm{end}} to the numerical solution.

Evolution of the ratio $\bar{\rho}_{\phi}/\bar{\rho}_{\rm dm}$ for three values of $\tilde{\xi}$

Figure 7. Evolution of the ratio ρˉϕ/ρˉdm\bar{\rho}_{\phi}/\bar{\rho}_{\rm dm} for log⁡10(mϕ/eV)=−30.5\log_{10}(m_{\phi}/{\rm eV})=-30.5 and F=10−4F=10^{-4}, for ξ~=0.3\tilde{\xi}=0.3 (left), 0.50.5 (center) and 0.8ξ~crit0.8\tilde{\xi}_{\rm crit} (right). The shaded band marks the injection window 50≤z≤6050\le z\le60, the dash-dotted line marks zosc≃25z_{\rm osc}\simeq25, and the dotted horizontal line the spectator limit, Eq. (19). The thick blue line is the direct numerical solution of Eq. (15) at the given (ξ,F)(\xi,F). The dotted yellow line is obtained by solving the equation once at fixed ξ~\tilde{\xi} and rescaling the result as FG/RFG/R, as done to compute the contours of Figure 3. Its agreement with the direct numerical solution (thick blue) validates this construction. The dashed orange line is the analytical solution with an instantaneous transition from wϕ=−1w_{\phi}=-1 to wϕ=0w_{\phi}=0 at aenda_{\rm end}. It reproduces the numerical result well for small ξ~\tilde{\xi}, but progressively underestimates it as ξ~\tilde{\xi} increases. This is due to the intermediate phase between aenda_{\rm end} and aosca_{\rm osc}, in which wϕ=wϕ(a)w_{\phi}=w_{\phi}(a) varies and is not captured by the sharp transition: its duration grows with ξ~\tilde{\xi}, and for ξ~\tilde{\xi} approaching ξ~crit\tilde{\xi}_{\rm crit} it extends beyond aosca_{\rm osc}, as shown by the oscillations of the numerical curve, which are larger and last longer in the right panel.

In reality, the transition from wϕ=−1w_{\phi}=-1 to wϕ=0w_{\phi}=0 is smooth. After the end of the injection, the field is still frozen in the potential until the Hubble friction becomes comparable to the mass, 3H(aosc)≃mϕ3H(a_{\rm osc})\simeq m_{\phi}. In the matter era, H(a)≃H0Ωma−3/2H(a)\simeq H_{0}\sqrt{\Omega_{\rm m}}a^{-3/2}, which gives

aosc=(3H0Ωmmϕ)2/3.(43)a_{\rm osc}=\left(\frac{3H_{0}\sqrt{\Omega_{\rm m}}}{m_{\phi}}\right)^{2/3}. \tag*{(43)}

Between aenda_{\rm end} and aosca_{\rm osc}, the equation of state is not constant but varies with redshift wϕ=wϕ(a)w_{\phi}=w_{\phi}(a). This transition phase can end or extend even beyond aosca_{\rm osc}, depending on how close ξ~\tilde{\xi} is to its critical value. A strong injection gives the field a large velocity, such that its kinetic energy 12ϕ˙2\frac{1}{2}\dot{\phi}^{2} dominates over the potential energy, and wϕw_{\phi} can reach +1+1 (kination). In this regime, the kinetic energy dilutes as a−6a^{-6}, faster than matter, so that the ALP energy density decreases quickly until the field is again frozen by Hubble friction (wϕ→−1w_{\phi}\to-1), and the onset of the oscillations is delayed. Since this behavior is not captured by a constant wϕw_{\phi}, the sharp-transition approximation can both overestimate and underestimate the ratio, depending on ξ~\tilde{\xi}.

Although the analytical approximation works well in most of the parameter space, we generate the exclusion plot shown in Figure 3 from the numerical solution of Eq. (15), which treats the evolution of wϕw_{\phi} exactly for every ξ~\tilde{\xi}, following the procedure described in Sec. II C.

Figure 3. Allowed region in the (ξ,F)(\xi,F) plane for log⁡10(mϕ/eV)=−30.5\log_{10}(m_\phi/\mathrm{eV})=-30.5 and 50≤z≤6050\le z\le60. Black: spectator limit, ρˉϕ/ρˉdm=10−2\bar{\rho}_\phi/\bar{\rho}_{\rm dm}=10^{-2}; dashed red: kick-out limit ξ=ξ~critF\xi=\tilde{\xi}_{\rm crit}F with ξ~crit=16.3\tilde{\xi}_{\rm crit}=16.3 (gray: broken dynamics); dash-dotted blue: CDM depletion limit ξ=ξdep\xi=\xi_{\rm dep}.

References

  1. [1]DESI Collaboration , M. Abdul Karim et al. Phys. Rev. D 112 (2025), no. 8 083515, arXiv:2503.14738.
  2. [2]SPT-3G Collaboration , E. Camphuis et al. Phys. Rev. D 113 (2026), no. 8 083504, arXiv:2506.20707.
  3. [3]W. Elbers et al. Phys. Rev. D 112 (2025), no. 8 083513, arXiv:2503.14744.arxiv.org/abs/2503.14744
  4. [4]N. Craig, D. Green, J. Meyers, and S. Rajendran J. High Energy Phys. 09 (2024) 097, arXiv:2405.00836.arxiv.org/abs/2405.00836
  5. [5]D. Green and J. Meyers Phys. Rev. D 111 (2025), no. 8 083507, arXiv:2407.07878.arxiv.org/abs/2407.07878
  6. [6]W. Elbers, C. S. Frenk, A. Jenkins, B. Li, and S. Pascoli Phys. Rev. D 111 (2025), no. 6 063534, arXiv:2407.10965.arxiv.org/abs/2407.10965
  7. [7]P. G. Lynch and L. Knox Phys. Rev. D 112 (2025), no. 8 083543, arXiv:2503.14470.arxiv.org/abs/2503.14470
  8. [8]L. Herold and M. Kamionkowski Phys. Rev. D 111 (2025), no. 8 083518, arXiv:2412.03546.DOI
  9. [9]D. Naredo-Tuero, M. Escudero, E. Fernández-Martínez, X. Marcano, and V. Poulin Phys. Rev. D 110 (2024), no. 12 123537, arXiv:2407.13831.
  10. [10]W. Giarè, E. Di Valentino, and A. Melchiorri Phys. Rev. D 109 (2024), no. 10 103519, arXiv:2312.06482.DOI
  11. [11]M. Loverde and Z. J. Weiner J. Cosmol. Astropart. Phys. 12 (2024) 048, arXiv:2410.00090.DOI
  12. [12]N. Sailer, G. S. Farrens, S. Ferraro, and M. White Phys. Rev. Lett. 136 (2026), no. 8 081002, arXiv:2504.16932.DOI
  13. [13]T. Jhaveri, T. Karwal, and W. Hu Phys. Rev. D 112 (2025), no. 4 043541, arXiv:2504.21813.
  14. [14]S. di Serego Alighieri Proceedings of the XI Multifrequency Behaviour of High Energy Cosmic Sources Workshop (MULTIF15) (2015) 009, arXiv:1507.02433.
  15. [15]E. Komatsu Nature Rev. Phys. 4 (2022), no. 7 452–469, arXiv:2202.13919.DOI
  16. [16]Y. Minami and E. Komatsu Phys. Rev. Lett. 125 (2020), no. 22 221301, arXiv:2011.11254.DOI
  17. [17]A. Gruppuso, G. Maggio, D. Molinari, and P. Natoli JCAP 05 (2016) 020, arXiv:1604.05202.DOI
  18. [18]Planck Collaboration , N. Aghanim et al. Astron. Astrophys. 596 (2016) A110, arXiv:1605.08633.
  19. [19]M. Ballardini, A. Gruppuso, S. Paradiso, S. S. Sirletti, and P. Natoli JCAP 09 (2025) 075, arXiv:2507.16714.arxiv.org/abs/2507.16714
  20. [20]J. R. Eskilt and E. Komatsu Phys. Rev. D 106 (2022), no. 6 063503, arXiv:2205.13962.DOI
  21. [21]J. R. Eskilt arXiv:2608.06480.
  22. [22]Y. Nakai, R. Namba, I. Obata, Y.-C. Qiu, and R. Saito JHEP 01 (2024) 057, arXiv:2310.09152.DOI
  23. [23]S. M. Carroll Phys. Rev. Lett. 81 (1998) 3067–3070, astro-ph/9806099.
  24. [24]G.-C. Liu, S. Lee, and K.-W. Ng Phys. Rev. Lett. 97 (2006) 161303, astro-ph/0606248.
  25. [25]S. Panda, Y. Sumitomo, and S. P. Trivedi Phys. Rev. D 83 (2011) 083506, arXiv:1101.5877.
  26. [26]T. Fujita, Y. Minami, K. Murai, and H. Nakatsuka Phys. Rev. D 103 (2021), no. 6 063508, arXiv:2008.02473.DOI
  27. [27]T. Fujita, K. Murai, H. Nakatsuka, and S. Tsujikawa Phys. Rev. D 103 (2021), no. 4 043509, arXiv:2011.11894.DOI
  28. [28]G. Choi, W. Lin, L. Visinelli, and T. T. Yanagida Phys. Rev. D 104 (2021), no. 10 L101302, arXiv:2106.12602.DOI
  29. [29]I. Obata J. Cosmol. Astropart. Phys. 09 (2022) 062, arXiv:2108.02150.DOI
  30. [30]S. Gasparrotto and I. Obata J. Cosmol. Astropart. Phys. 08 (2022), no. 08 025, arXiv:2203.09409.
  31. [31]M. Galaverni, F. Finelli, and D. Paoletti Phys. Rev. D 107 (2023), no. 8 083529, arXiv:2301.07971.DOI
  32. [32]W. Lin, L. Visinelli, and T. T. Yanagida JCAP 10 (2025) 023, arXiv:2504.17638.DOI
  33. [33]K. Murai, F. Naokawa, T. Namikawa, and E. Komatsu Phys. Rev. D 107 (2023) L041302, arXiv:2209.07804.DOI
  34. [34]J. R. Eskilt, L. Herold, E. Komatsu, K. Murai, T. Namikawa, and F. Naokawa Phys. Rev. Lett. 131 (2023), no. 12 121001, arXiv:2303.15869.
  35. [35]J. Kochappan, L. Yin, B.-H. Lee, and T. Ghosh Phys. Rev. D 112 (2025), no. 6 063562, arXiv:2408.09521.
  36. [36]F. Finelli and M. Galaverni Phys. Rev. D 79 (2009) 063002, arXiv:0802.4210.DOI
  37. [37]G. Sigl and P. Trivedi (11, 2018) arXiv:1811.07873.arxiv.org/abs/1811.07873
  38. [38]G.-C. Liu and K.-W. Ng Phys. Dark Univ. 16 (2017) 22–25, arXiv:1612.02104.arxiv.org/abs/1612.02104
  39. [39]M. A. Fedderke, P. W. Graham, and S. Rajendran Phys. Rev. D 100 (2019) 015040, arXiv:1903.02666.DOI
  40. [40]D. Zhang, E. G. M. Ferreira, I. Obata, and T. Namikawa Phys. Rev. D 110 (2024), no. 10 103525, arXiv:2408.08063.arxiv.org/abs/2408.08063
  41. [41]T. Namikawa, K. Murai, and F. Naokawa Phys. Rev. D 112 (2025), no. 12 123513, arXiv:2506.20824.arxiv.org/abs/2506.20824
  42. [42]F. Takahashi and W. Yin J. Cosmol. Astropart. Phys. 04 (2021) 007, arXiv:2012.11576.DOI
  43. [43]N. Kitaima, F. Kozai, F. Takahashi, and W. Yin J. Cosmol. Astropart. Phys. 10 (2022) 043, arXiv:2205.05083.
  44. [44]M. Jain, R. Hagimoto, A. J. Long, and M. A. Amin J. Cosmol. Astropart. Phys. 10 (2022) 090, arXiv:2208.08391.
  45. [45]D. Gonzalez, N. Kitajima, F. Takahashi, and W. Yin Phys. Lett. B 843 (2023) 137990, arXiv:2211.06849.DOI
  46. [46]J. Lee, K. Murai, F. Takahashi, and W. Yin Phys. Rev. D 112 (2025), no. 4 043538, arXiv:2503.18417.DOI
  47. [47]R. C. Myers and M. Pospelov Phys. Rev. Lett. 90 (2003) 211601, hep-ph/0301124.
  48. [48]K. R. S. Balaji, R. H. Brandenberger, and D. A. Easson J. Cosmol. Astropart. Phys. 12 (2003) 008, hep-ph/0310368.
  49. [49]A. Arvanitaki, S. Dimopoulos, S. Dubovsky, N. Kaloper, and J. March-Russell Phys. Rev. D 81 (2010) 123530, arXiv:0905.4720.DOI
  50. [50]T. Namikawa Phys. Rev. Lett. 135 (2025), no. 16 161004, arXiv:2506.22999.arxiv.org/abs/2506.22999
  51. [51]F. Naokawa, T. Namikawa, K. Murai, I. Obata, and K. Kamada Phys. Rev. Lett. 136 (2026), no. 4 041003, arXiv:2405.15538.arxiv.org/abs/2405.15538
  52. [52]P. W. Graham, D. Green, and J. Meyers Phys. Rev. D 113 (2026), no. 4 043514, arXiv:2508.20999.DOI
  53. [53]G. Gubitosi, M. Martinelli, and L. Pagano J. Cosmol. Astropart. Phys. 12 (2014), no. 2014 020, arXiv:1410.1799.DOI
  54. [54]S. Lee, G.-C. Liu, and K.-W. Ng Phys. Rev. D 89 (2014), no. 6 063010, arXiv:1307.6298.DOI
  55. [55]B. D. Sherwin and T. Namikawa Mon. Not. R. Astron. Soc. 520 (2021) 3298–3304, arXiv:2108.09287.DOI
  56. [56]H. Nakatsuka, T. Namikawa, and E. Komatsu Phys. Rev. D 105 (2022) 123509, arXiv:2203.08560.DOI
  57. [57]F. Naokawa and T. Namikawa Phys. Rev. D 108 (2023), no. 6 063525, arXiv:2305.13976.arxiv.org/abs/2305.13976
  58. [58]K. Murai Phys. Rev. D 111 (2025), no. 4 043514, arXiv:2407.14162.DOI
  59. [59]S. Arcari, N. Bartolo, A. Greco, A. Gruppuso, M. Lattanzi, and P. Natoli JCAP 10 (2024) 101, arXiv:2407.02144.
  60. [60]S. M. Carroll, G. B. Field, and R. Jackiw Phys. Rev. D 41 (1990) 1231.
  61. [61]S. M. Carroll and G. B. Field Phys. Rev. D 43 (Jun, 1991) 3789–3793.
  62. [62]D. Harari and P. Sikivie Phys. Rev. B 289 (1992) 67–72.
  63. [63]M. Kamionkowski and R. R. Caldwell arXiv:2609.36000.
  64. [64]M. Zaldarriaga and U. Seljak Phys. Rev. D 55 (1997) 1830–1840, astro-ph/9609170.
  65. [65]J. Lesgourgues arXiv e-prints (2011) arXiv:1104.2932.arxiv.org/abs/1104.2932
  66. [66]D. Blas, J. Lesgourgues, and T. Tram JCAP 07 (2011) 034, arXiv:1104.2933.DOI
  67. [67]M. B. Gavela, D. Hernandez, L. Lopez Honorez, O. Mena, and S. Rigolin JCAP 07 (2009) 034, arXiv:0901.1611. [Erratum: JCAP 05, E01 (2010)].
  68. [68]V. Salvatielli, A. Marchini, L. Lopez-Honorez, and O. Mena Phys. Rev. D 88 (2013), no. 2 023531, arXiv:1304.7119.DOI
  69. [69]E. Di Valentino, A. Melchiorri, and O. Mena Phys. Rev. D 96 (2017), no. 4 043503, arXiv:1704.08342.DOI
  70. [70]B. Wang, E. Abdalla, F. Atrio-Barandela, and D. Pavon Rept. Prog. Phys. 79 (2016), no. 9 096901, arXiv:1603.08299.DOI
  71. [71]E. Di Valentino, R. Z. Ferreira, L. Visinelli, and U. Danielsson Phys. Dark Univ. 26 (2019) 100385, arXiv:1906.11255.DOI
  72. [72]Planck Collaboration Astron. Astrophys. 641 (2020) A6, arXiv:1807.06209.
  73. [73]Planck Collaboration, R. Adam et al. Astron. Astrophys. 596 (2016) A108, arXiv:1605.03507.
  74. [74]M. Tristram et al. Astron. Astrophys. 647 (2021) A128, arXiv:2010.01139.DOI
  75. [75]M. Tristram et al. Phys. Rev. D 105 (2022), no. 8 083524, arXiv:2112.07961.DOI
  76. [76]S. M. Carroll and G. B. Field Phys. Rev. Lett. 79 (1997) 2394–2397, astro-ph/9704263.
  77. [77]F. Naokawa Phys. Rev. Lett. 136 (2026), no. 4 041004, arXiv:2504.06709.arxiv.org/abs/2504.06709
  78. [78]LiteBIRD Collaboration Prog. Theor. Exp. Phys. (2, 2022) arXiv:2202.02773.

Paper details

vlnc:W_pgmh57p6
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 1

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents