---
title: 'Bounce Radius: Definitions and Interpretations'
url: https://www.emergentmind.com/topics/bounce-radius
type: topic
---

# Bounce Radius: Definitions and Interpretations

“Bounce radius” is not a standardized invariant across the literature. In nonsingular cosmology it can denote the minimum scale factor at the bounce, a physical or comoving Hubble radius, or a curvature scale; in black-bounce geometries it is the minimum areal radius of the spacetime; in kinetic plasma theory it denotes the minimum transverse size required to sustain trapped orbits. The unifying idea is a finite radius-like scale associated with reversal of collapse or contraction, but the precise quantity is model-dependent and must be read from the dynamical variables used in each framework [2602.02576].

## 1. Terminological scope and principal definitions

A recurrent feature of the literature is that many papers do **not** introduce a unique symbol explicitly called “bounce radius.” Instead, they use radius-like quantities adapted to the system under study. In FLRW bounce cosmology, the most common candidates are the minimum scale factor \(a_{\min}=a(t_b)\), the physical Hubble radius \(R_H=1/|H|\), and the comoving Hubble radius \(r_H=(a|H|)^{-1}\) [1601.04112]. In Simpson–Visser-type regular spacetimes, the bounce radius is the minimum areal radius, e.g. \(\Sigma_{\min}=a\) or \(\rho_{\text{bounce}}=q_H\), attained at the throat or interior bounce surface [2404.14816]. In finite-gyro-radius electron-hole equilibria, the same phrase refers to the minimum transverse size compatible with sustained trapping, bounded by the thermal gyro-radius \(r_g\) [2101.10180].

An important source of confusion is that some works call the comoving Hubble radius a “Hubble radius,” while others reserve that name for \(1/|H|\). The unimodular \(F(R)\) study explicitly writes \(R_H=1/(aH)\), i.e. its \(R_H\) is comoving rather than physical [1601.04112]. The Weyl-type \(f(Q)\) study likewise distinguishes the minimum size \(a_{\min}\) from the Hubble-type radii, and notes a caption-level ambiguity in the plotted “cosmic Hubble radius” \(r_h\) [2602.02576].

| Context | Radius-like quantity | Representative expression |
|---|---|---|
| Weyl-type \(f(Q)\) bounce cosmology | minimum size and Hubble-type radii | \(a_{\min}=\gamma^{n/3}\), \(R_H=3(\beta t^2+\gamma)/(2n\beta|t|)\), \(R_c=3(\beta t^2+\gamma)^{1-n/3}/(2n\beta|t|)\) [2602.02576] |
| Unimodular and related bounce cosmologies | physical or comoving Hubble radius | \(R_H=1/|H|\), \(r_H=1/(a|H|)\) [1601.04112] |
| LQC matter/deformed matter bounce | minimal scale factor at the bounce | \(a_B\equiv a(t_B)\) [1606.03689] |
| Black-bounce spacetimes | minimum areal radius | \(\Sigma_{\min}=a\), \(\rho_{\text{bounce}}=q_H\) [2404.14816] |
| Electron-hole equilibria | minimal transverse size | \(a\gtrsim \sqrt{2}\,r_g\) [2101.10180] |

## 2. Minimum-size definitions in bounce cosmology

In FLRW bounce models, the cleanest finite notion of bounce radius is often the minimum value of the scale factor. The Weyl-type \(f(Q)\) model with quintom signature adopts
\[
a(t)=(\beta t^2+\gamma)^{n/3},
\]
with \(\beta>0\), \(\gamma>0\), \(n>0\), so the minimum size is
\[
a_{\min}=a(0)=\gamma^{n/3}.
\]
This is the direct “radius at the bounce” in a spatially flat background, and in that reconstruction it depends only on \(\gamma\) and \(n\), not directly on \(\alpha\), \(\xi\), or \(m\) [2602.02576]. The same paper quotes the representative values \(a_{\min}=1\) for \((n,\gamma)=(0.35,1)\), \(a_{\min}=2^{1/6}\approx1.122\) for \((0.50,2)\), and \(a_{\min}=3^{1/4}\approx1.316\) for \((0.75,3)\) [2602.02576].

Loop Quantum Cosmology formulations often package the same idea as \(a_B\equiv a(t_B)\). In the deformed matter-bounce scenario,
\[
a(t)=\left(\frac{3}{4}\rho_c t^2+1\right)^{1/3}\exp\!\left[\frac{f_0}{1-\alpha}(t-t_s)^{1-\alpha}\right],
\]
and \(a_B=a(t_B)\) is the minimal radius. Because the deformation is negligible near the bounce for the parameter regime studied, the paper concludes that \(a_B=1\) to an excellent approximation in its normalization [1606.03689]. The \(\Lambda\)CDM bounce scenario uses the same normalization idea: the LQC bounce occurs at \(a_b=1\), while the curvature scale is encoded by the critical density \(\rho_c\) [1412.2914].

The ghost-condensate matter-bounce literature makes the same conceptual distinction explicit: since \(H(t_b)=0\), neither \(R_H\) nor \((aH)^{-1}\) is finite at the bounce, so the operationally meaningful “radius” is \(a_b\equiv a(t_b)\), together with the curvature scale set by \(\dot H(t_b)\) [1007.2654]. This suggests that, in cosmological usage, “bounce radius” most often denotes the minimum physical size whenever the Hubble radii diverge at the bounce.

## 3. Hubble-radius formulations and horizon dynamics

A second, widely used meaning of bounce radius is the Hubble radius. Here the distinction between physical and comoving quantities is essential:
\[
R_H(t)=\frac{1}{|H(t)|}, \qquad r_H(t)=\frac{1}{a(t)|H(t)|}.
\]
These scales control horizon crossing and the perturbative chronology of contracting, bouncing, and expanding phases [1601.04112].

The behavior of these radii is strongly model-dependent. In the Weyl-type \(f(Q)\) reconstruction,
\[
H(t)=\frac{2n\beta t}{3(\beta t^2+\gamma)},
\]
so
\[
R_H(t)=\frac{3(\beta t^2+\gamma)}{2n\beta|t|}, \qquad
R_c(t)=\frac{3(\beta t^2+\gamma)^{1-n/3}}{2n\beta|t|}.
\]
Both diverge at \(t\to0\), and both are even in time:
\[
R_H(-t)=R_H(t), \qquad R_c(-t)=R_c(t).
\]
The paper identifies this as the symmetric behavior of the radius scales around the bounce [2602.02576].

By contrast, the superbounce in unimodular \(F(R)\) gravity has
\[
R_H(t)=\frac{c^2}{2}|t-t_s|,
\]
so the physical Hubble radius vanishes at the bounce because \(|H|\to\infty\), while the comoving radius also goes to zero for \(c^2>2\) [1601.04112]. Matter-bounce models show yet another pattern: the comoving Hubble radius decreases during a long contracting phase and increases during expansion, which is the standard exit-and-re-entry structure required for nearly scale-invariant perturbations [1206.4196].

Type-IV singular and symmetric bounces do not share that perturbative advantage. In the singular Type-IV case,
\[
R_H(t)=\frac{1}{|f_0||t-t_s|^\alpha}, \qquad
r_H(t)=\frac{\exp[-f_0(t-t_s)^{\alpha+1}/(\alpha+1)]}{|f_0||t-t_s|^\alpha},
\]
so both radii diverge at the bounce and vanish far away from it, making perturbation generation near the bounce non-scale-invariant in the model as studied [1512.04787]. The symmetric bounce has \(R_H,r_H\to\infty\) at \(t=0\) and both fall to zero for large \(|t|\), which prevents the usual exit-and-re-entry story [1601.04112].

Asymmetric bounce-to-dark-energy constructions sharpen this point further. In the ghost-free \(f(R,\mathcal G)\) model, \(r_H(-t)\neq r_H(t)\) because the exponential late-time factor is negligible in contraction but important in expansion; the comoving Hubble radius diverges both in the far contracting past and at the bounce, while late-time acceleration makes it decrease again [2202.02695]. The Chern–Simons-corrected \(F(R)\) scenario uses the same logic: \(r_h=(aH)^{-1}\) diverges in deep contraction and at the bounce, then decreases during the late accelerating era [2109.00345].

## 4. Weyl \(f(Q)\) gravity and the quintom bounce

The most explicit recent treatment of bounce radius in this sense appears in the Weyl-type \(f(Q)\) model with a massive Weyl vector and power-law non-metricity sector
\[
f(Q)=\alpha Q^\xi, \qquad Q=6H^2(t)
\]
after imposing the FLRW ansatz and the simplifying condition \(\psi(t)=H(t)\) [2602.02576]. The effective density and pressure are
\[
\rho(t)= - \frac{1}{2} m^2 H^2 - \alpha 6^\xi (2 \xi - 1) (H^2)^\xi,
\]
\[
p(t)= - \frac{1}{2} (24 + m^2) H^2 - \alpha 6^\xi (H^2)^\xi - 4 \dot H,
\]
with
\[
\omega(t)=\frac{p}{\rho}.
\]

For the bounce ansatz
\[
a(t)=(\beta t^2+\gamma)^{n/3},
\]
the Hubble rate and its derivative are
\[
H(t)=\frac{2 n \beta t}{3 (\beta t^2 + \gamma)}, \qquad
\dot H(t)=\frac{2 n \beta (\gamma - \beta t^2)}{3 (\beta t^2 + \gamma)^2}.
\]
The bounce conditions follow immediately:
\[
H(0)=0,\qquad \dot H(0)=\frac{2n\beta}{3\gamma}>0,\qquad \dot a(0)=0,\qquad \frac{\ddot a(0)}{a(0)}=\dot H(0)>0.
\]
The minimum-size identification is therefore
\[
R_{\text{bounce}}\equiv a_{\min}=\gamma^{n/3},
\]
while the causal scales are encoded in the divergences
\[
R_H(t)\simeq \frac{3\gamma}{2n\beta|t|}, \qquad
R_c(t)\simeq \frac{3\gamma^{1-n/3}}{2n\beta|t|}
\quad (|t|\ll1).
\]
The amplitude of the divergence scales as \(\gamma/(n\beta)\) for \(R_H\) and \(\gamma^{1-n/3}/(n\beta)\) for \(R_c\) [2602.02576].

The dynamical interpretation is quintom-like. Near the bounce, the null energy condition is violated, \(\omega\) crosses the phantom divide \(\omega=-1\), and \(\rho\to0\) while
\[
p(0)=-4\dot H(0)=-\frac{8n\beta}{3\gamma}.
\]
Hence \(\omega\) is ill-defined exactly at \(t=0\) but crosses \(-1\) on either side [2602.02576]. The same work reconstructs an effective two-scalar description in which the quintessence-like kinetic term becomes negative near the bounce and the phantom-like kinetic energy becomes maximally positive, matching the NEC-violating regime. Stability analysis via the adiabatic index indicates instability near the bouncing point, whereas the energy conditions indicate dark-energy dominance [2602.02576].

## 5. Black-bounce geometries and black-hole interior bounces

In black-bounce spacetimes, bounce radius is not a cosmological scale but the minimum areal radius of the geometry. The Simpson–Visser prescription replaces the spherical radius by
\[
R(r)=\sqrt{r^2+a^2},
\]
so the minimum occurs at \(r=0\) and is
\[
R_{\min}=a.
\]
If this minimum lies in a static region it is a wormhole throat; if it lies inside an event horizon it is a regular bounce in the black-hole interior [2404.14816]. This is the canonical geometric meaning of “bounce radius” in the regular-black-hole literature.

The same structure persists in halo-embedded solutions. In the M60-calibrated Simpson–Visser metric,
\[
\rho(r)=\sqrt{r^2+q_H^2},
\]
and the dark-matter halo modifies only the lapse \(A(r)\), not the areal radius. Consequently,
\[
\rho_{\text{bounce}}=q_H
\]
remains the bounce radius even in the presence of the halo [2606.24917]. In generalized \(k\)-\(n\) black-bounce metrics,
\[
\Sigma(r)=\sqrt{r^2+a^2},
\]
so \(a\) is again the bounce radius, while the thresholds
\[
a_{\rm hor}=2m\left(\frac{k}{k+1}\right)^{\frac{k}{2n}}\left(\frac{1}{k+1}\right)^{\frac{1}{2n}}
\]
and \(a_\star\) separate regular-black-hole, horizonless double-ring, and no-photon-sphere regimes [2510.23748].

A charged version appears in the Reissner–Nordström geometry corrected by a bounce parameter. There
\[
h(r)^2=r^2+a^2, \qquad h_0=h(0)=a,
\]
so the bounce radius is again the minimal areal radius. The coordinate horizon becomes
\[
r_h=\sqrt{\left(m+\sqrt{m^2-Q^2}\right)^2-a^2},
\]
and the coordinate photon-sphere radius becomes
\[
r_{\rm ph}=\sqrt{H_{\rm ph}^2-a^2},
\qquad
H_{\rm ph}=\frac{3m+\sqrt{9m^2-8Q^2}}{2},
\]
showing that the bounce parameter lowers the coordinate radii while leaving the areal photon-sphere radius unchanged [2301.01855].

A distinct but related usage appears in the semiclassical Schwarzschild-interior analysis. In Kantowski–Sachs form, the areal radius is \(r(T)=e^{\beta(T)}\), and the bounce ansatz imposes
\[
r_b\equiv r(0)=e^{\beta_0}\equiv a>0.
\]
This \(r_b\) is the minimum areal radius reached inside the black hole, with an explicit example giving \(r_b\approx10^5\ell_{\rm pl}\) for suitable curvature-quadratic couplings [1808.03717]. By contrast, regular-center alternatives explicitly reject the bounce/throat interpretation and instead keep \(R=r\) with \(r=0\) as a regular center rather than a minimum-radius surface [2404.14816].

## 6. Other specialized uses and interpretive cautions

Outside gravitation, “bounce radius” can refer to a kinetic size threshold. In multidimensional electron-hole equilibria with finite gyro-radius, the relevant quantity is the minimal transverse size that can sustain trapped parallel motion. For a Gaussian profile, the practical bound is
\[
a\gtrsim \sqrt{2}\,r_g,
\]
equivalently \(\sqrt{\langle r^2\rangle}\gtrsim 2r_g\), and widths \(\lesssim r_g\) do not persist beyond roughly a quarter of a bounce period [2101.10180]. Here the word “bounce” comes from parallel bounce motion, not cosmological reversal or spacetime regularization.

A related conceptual shift occurs in black-hole evolution through a cosmological bounce. There the operational “bounce radius” is the **minimum proper Hubble radius during contraction**,
\[
R_{H,\min}=\min_t \frac{1}{|H(t)|}=-\frac{1}{H_{\min}},
\]
not the minimum scale factor [2206.08466]. The ratio \(R_{H,\min}/r_{BH,0}\) controls horizon behavior: if \(R_{H,\min}/r_{BH,0}\gtrsim 3.5\), the black-hole apparent horizon persists; if \(R_{H,\min}/r_{BH,0}<3.5\), it merges with the cosmological horizon and temporarily disappears during contraction [2206.08466].

A third gravitational usage appears in relativistic spherical collapse with a vacuum-like ground state. There the gravitational bounce radius is
\[
R_B=\left(\frac{8\pi G\rho_G}{3}\right)^{-1/2}
=\sqrt{\frac{3}{\Lambda_{\rm eff}}},
\]
the turning-point radius induced by the transition to a constant ground-state density \(\rho_G\) [2505.23877]. This is neither a Hubble radius nor a scale factor, but the minimum physical radius of a collapsing closed FLRW patch.

The central interpretive caution is therefore straightforward: the phrase “bounce radius” does not by itself specify a unique observable. In FLRW cosmology it often means \(a_{\min}\) because \(H=0\) makes \(R_H\) and \(r_H\) diverge at the bounce; in horizon-based analyses it may instead denote \(1/|H|\) or \((a|H|)^{-1}\); in black-bounce geometries it is the minimum areal radius; and in kinetic plasma theory it is a minimum trapping width. Any technical use of the term is meaningful only after the underlying dynamical variable has been fixed by the model.

Source: https://www.emergentmind.com/topics/bounce-radius