betaFeedback

WorksPhysics and Astronomy

Elastic Planetoids

Abstract

Modeling the internal structure of self-gravitating solid and liquid bodies presents a challenge, as existing approaches are often limited to either overly simplistic constant-density approximations or more complex numerical equations of state. We present a detailed analysis of a tractable and physically motivated model for perfectly elastic, spherically symmetric self-gravitating bodies in hydrostatic equilibrium. The model employs a logarithmic equation of state (logotropic EOS) with a non-zero initial density and constant bulk modulus. Importantly, scaling properties of the model allow all solutions to be derived from a single, universal solution of an ordinary differential equation, resembling the Lane-Emden and Chandrasekhar models. The model provides new insights into stability issues and reveals oscillatory asymptotic behavior in the mass-radius relation, including the existence of both a maximum mass and a maximum radius. We derive useful, simple analytical approximations for key properties, such as central overdensity, moment of inertia, binding energy, and gravitational potential, applicable to small, metallic bodies like asteroids and moons. These new approximations could aid future research, including space mining and the scientific characterization of small Solar System bodies.

Introduction

The hydrostatic equilibrium of spherically symmetric, self-gravitating bodies is a cornerstone of research and education in astronomy, astrophysics, and planetary science. Simple analytical models are particularly valuable because they are widely applicable, easy to understand, and facilitate quick calculations and discussions grounded in established concepts. A prime example is the Lane-Emden equation (Lane [48]), which, with its power-law Equation Of State (thereafter EOS), describes a wide range of objects, ranging from constant density planetoid models and gas giants (Knopik et al. [43]) to stars via e.g. the Eddington model (Eddington [25]) and even globular clusters (Plummer [64]). Its general relativistic generalization (Tooper [68]) has been applied to model cosmological voids (Odrzywołek 2009). The Chandrasekhar white dwarf model (Chandrasekhar [8]; Kippenhahn et al. [42]), based on the degenerate fermion EOS (Bisnovatyi-Kogan et al. [5]), provides another key example. However, while polytropes with small indices 0<n<1/20<n<1/2 offer useful crude approximations for planetary objects, they inherently assume a zero density at zero pressure, which is unrealistic for solid and liquid bodies. This limitation restricts their use in planetary science, especially for exoplanets and small Solar System bodies, where non-zero densities are fundamental. Consequently, researchers often face a stark choice: either employ the overly simplistic constant-density model or resort to complex analytical formulations like the Birch-Murnaghan EOS (Birch [4]; Murnaghan [57]) or numerical solutions with tabulated data (Smith et al. [67]).

To address this challenge, we introduce a simple, analytical model for perfectly elastic, self-gravitating bodies, characterized by an uncompressed density ρ0\rho_0 and constant bulk modulus KK. This leads to a logarithmic EOS that is only approximate, valid for small compression ratios, but is treated as exact within the framework of our study.

Corresponding author: Andrzej Odrzywołek

[email protected] The proposed model provides a crucial bridge between the overly simplistic constant-density approximation and the complexities of realistic solid-state EOSs, enabling a smooth transition between these two regimes.

EOS mathematically similar to ours have been studied under the name ”logotropic” model in the context of alternative cosmology and Dark Matter halos by Chavanis [10] in a series of articles [16, 10, 15] (Chavanis 2019; Chavanis 2016; Chavanis 2022). The term ”logotropic” itself was introduced by McLaughlin & Pudritz [52] (1997) in their study of giant molecular clouds (GMCs). Although the term ’logotropic’ (’logatropic’) originates in astronomy and cosmology [52, 10], here it describes elastic matter with a fixed bulk modulus, distinct from ISM or dark matter models. These previous studies, like ours, feature the density logarithm ln⁡ρ\ln\rho, but our work is distinguished by its specific EOS formula, physical scales, and, crucially, interpretation. Previous authors did not recognize the fact that logarithmic EOS can arise naturally from the assumption of perfectly elastic matter with non-zero density and fixed bulk modulus – a common characteristic of materials like metals and liquids. Such an approximation is plausible for small compressions and ’cold EOS’ conditions where thermal pressure is not dominant. For technical simplicity, our model also omits phase transitions, which may result in abrupt changes of EOS parameters, here treated as constants.

Additional motivation comes from the recent detection of interstellar objects like 1I/’Oumuamua [54, 56] and 2I/Borisov [30] (Jewitt & Luu 2019) entering our Solar System. They provide direct samples of material formed around other stars. Their sometimes puzzling characteristics, along with the highly anomalous composition of presolar materials found within meteorites, such as the Hypatia stone [45], suggest that the composition and physical properties of extrasolar planetesimals may differ significantly from those familiar within our own system. Rapidly assessing the nature and structure of future discoveries, potentially composed of unexpected materials, necessitates simple and flexible physical models. The elastic model presented here, relying only on the fundamental parameters of uncompressed density (ρ0\rho_0) and bulk modulus (KK), offers such a tool. It allows for straightforward analytical estimates of internal structure, stability limits, and the mass-radius relation across a vast parameter space, providing crucial first-order insights when detailed equations of state are unavailable or uncertain.

The article is organized as follows. § 2 presents the derivation of the elastic model and the conversion of the equilibrium equations into a dimensionless form with appropriate scaling. § 3 details the properties of the special ”elastic function” U(x)U(x) including its mathematical characteristics, approximations, and asymptotic behavior. Physical quantities expressed using appropriate scaling and elastic function U(x)U(x) are presented in Sect. 4, along with illustrative examples. § 5 explores the astrophysical properties of the model, focusing on the mass-radius M(R)M(R) curve, the existence of a maximum mass, and stability considerations. A collection of useful simple approximate analytical formulas is provided in Sect. 6. The limits of applicability of the model are established in § 7 through a comparison with a realistic iron equation of state. Finally, § 8 summarizes the key findings. The appendix provides a detailed, simultaneous derivation of the equations for several analytical equations of state commonly used in astronomy and planetary science.

Elastic Model

We consider a perfectly elastic body with bulk modulus KK compressed due to its own gravity. With spherical symmetry, other elastic properties of the matter can be neglected. We also assume uncompressed density1^{1} ρ0\rho_0. For example, one could imagine an iron ball with progressively increasing mass (and radius). Initially, it will grow in size at nearly constant density, transitioning from an asteroid to a planetoid, moon/planet, super-Earth, etc. Eventually, gravity will compress the matter. The subsequent behavior, and whether this process can continue indefinitely within Newtonian gravity and assuming perfectly elastic matter, will be determined from the analysis of our model.

From the definition of the bulk modulus measured at ρ0\rho_0

K=ρdPdρ(1)K = \rho\frac{dP}{d\rho} \tag*{(1)}

we obtain a perfectly elastic equation of state (EOS) in logarithmic form

P(ρ)=Kln⁡(ρ/ρ0).(2)P(\rho) = K \ln(\rho/\rho_0). \tag*{(2)}

This EOS is valid for ρ≥ρ0\rho\ge\rho_0, ensuring P≥0P \ge0. The above equation belongs to the barotropic class, along with other important examples such as polytropic and degenerate electron gas EOSs. In non-standard cosmology [10] and interstellar medium research (McLaughlin & Pudritz 1997) the formula (2) is known as the logotropic EOS. Using the dimensionless compression ratio (or simply compression)

η=ρρ0(3)\eta= \frac{\rho}{\rho_0} \tag*{(3)}

the elastic EOS can be written as P=Kln⁡ηP = K \ln\eta.

In engineering, the term dilatation, denoted by

μ=ρρ0−1=η−1,(4)\mu= \frac{\rho}{\rho_0} - 1 = \eta- 1, \tag*{(4)}

is often used. This casts the elastic EOS in the form P=Kln⁡(1+μ)P = K \ln(1+\mu). The elastic approximation is valid when the dilatation is small, i.e., when ∣μ∣≪1|\mu| \ll1.

We will start with the hydrostatic equilibrium equation in the form

h+Φg=C,(5)h + \Phi_g = C, \tag*{(5)}

where h=∫dp/ρh = \int\mathrm{d}p/\rho is specific enthalpy and CC is an integration constant. Gravitational potential Φg\Phi_g obeys Poisson’s equation

ΔΦg=4πGρ.\Delta\Phi_g = 4\pi G\rho.

For EOS (2) the specific enthalpy is

h(ρ)=−Kρ+const.(6)h(\rho) = -\frac{K}{\rho} + \mathop{\mathrm{const}}. \tag*{(6)}

If we demand that h(ρ)=0h(\rho) = 0 at the surface, the constant in (6) is equal to K/ρ0K/\rho_0. Taking spherical 3D Laplacian of (5) yields

r−2∂r(r2∂rh)+4πGρ=0.(7)r^{-2}\partial_r\left(r^2\partial_r h\right) + 4\pi G\rho= 0. \tag*{(7)}

Using a more convenient form of the enthalpy,

h=K∫dρρ2=K∫ρ′ρ2dr,(8)h = K\int\frac{\mathrm{d}\rho}{\rho^2} = K\int\frac{\rho'}{\rho^2}\mathrm{d}r, \tag*{(8)}

from eqns. (7) and (8) we obtain

ρ′′+2ρ′r−2(ρ′)2ρ+4πGKρ3=0.(9)\rho'' + \frac{2\rho'}{r} - \frac{2(\rho')^2}{\rho} + \frac{4\pi G}{K}\rho^3 = 0. \tag*{(9)}

Substituting ρ(r)→ρcu(r)\rho(r) \to\rho_c u(r) and dividing through by ρc\rho_c gives a differential equation in u(r)u(r):

u′′+2u′r−2(u′)2u+4πGρc2Ku3=0.(10)u'' + \frac{2u'}{r} - \frac{2(u')^2}{u} + \frac{4\pi G\rho_c^2}{K}u^3 = 0. \tag*{(10)}

Using the dimensionless variable x=r/λx = r/\lambda, where the scale length λ\lambda is defined as

λ2=K4πGρc2,(11)\lambda^2 = \frac{K}{4\pi G\rho_c^2}, \tag*{(11)}

we remove physical constants and arrive at a crucial ordinary differential equation (Elastic ODE)

u′′+2xu′−2u′2u+u3=0.(12)u'' + \frac{2}{x}u' - \frac{2u'^2}{u} + u^3 = 0. \tag*{(12)}

This form isolates the nonlinear elastic term −2u′2/u-2u'^2/u, distinguishing it from the Lane-Emden equation. Without this negative term, (12) would be the n=3n=3 Lane-Emden equation. The first two terms represent the spherical Laplacian in variable xx, allowing us to write Eqn. (12) as

Δu−2u′2u+u3=0,\Delta u - \frac{2u'^2}{u} + u^3 = 0,

or, even more simply, noting that the reciprocal appears in the expression for the specific enthalpy (Eq. (6)), we can rewrite the equation as:

Δ(1u)−u=0.(13)\Delta\left(\frac{1}{u}\right)-u=0. \tag*{(13)}

The standard initial conditions for (12) are

u(0)=1,u′(0)=0.(14)u(0)=1,\quad u'(0)=0. \tag*{(14)}

Similar analysis can be performed for the continuity equation

dmdr=4πr2ρ.\frac{dm}{dr}=4\pi r^2\rho.

Substituting

m(r)=4πλ3ρc μ(r/λ)m(r)=4\pi\lambda^3\rho_c\,\mu(r/\lambda)

leads to the dimensionless form

μ′=x2u.\mu'=x^2u.

Moreover, due to Eqn. (13), or by integrating the continuity equation, we obtain the first integral

μu2+u′x2=0.(15)\mu u^2+u'x^2=0. \tag*{(15)}

Mathematical Properties of the Elastic Function

Because all physical properties of our model are described by a single special function, it is worthwhile to study it in detail. This function is defined as the specific solution to Eqn. (12) with the initial conditions given by (14). To avoid confusion with the general solution, u(x)u(x), we denote this specific solution as the elastic function, using the capital letter U(x)U(x).

The initial conditions (14) cannot be applied directly at x=0x=0 due to the division by zero in the spherical Laplacian. The standard procedure in such a case is to begin integration with a series expansion:

U(x)≃1−16x2+13360x4−253024x6+⋯(16)U(x) \simeq1-\frac{1}{6}x^2+\frac{13}{360}x^4-\frac{25}{3024}x^6+\cdots \tag*{(16)}

Comparing this series with that of the Lane-Emden function, we find that the best match is for the polytropic index n=13/3≃4.33n=13/3 \simeq4.33. However, starting with the x6x^6 term, the elastic and Lane-Emden functions diverge. Like for the Lane-Emden equation ([35]), the series expansion, Eqn. (16), has a small radius of convergence, therefore providing a useful solution only for small xx, but it is not sufficient to cover the entire domain of interest. Nevertheless, the series expansion is important for initiating the numerical integration of Eqn. (12), and for deriving a variety of approximate analytical formulas (Sect. 6) such as for the density profile.

The elastic function U(x)U(x), obtained as a numerical solution to the ODE, is presented in Figure 1. From numerical evidence, we infer that U(x)>0U(x)>0, i.e., the solution is positive everywhere (see also [10]). The solution asymptotes to

Figure 1. Universal elastic (logotropic) function U(x)U(x). Cutoff value 1/ηc=ρ0/ρc1/\eta_c=\rho_0/\rho_c defines surface of the body and corresponding dimensionless radius xtx_t. Important one-to-one property is U(xt)=1/ηcU(x_t)=1/\eta_c. Inset illustrates oscillatory asymptotic, which is too small to be visible in the main plot.

U(x)∼2x,(17)U(x) \sim\frac{\sqrt{2}}{x}, \tag*{(17)}

which is also the exact singular solution of Eqn. (12). Therefore, in analogy to the n=5n=5 Plummer solution, θ5(x)=1/1+x2/3\theta_5(x)=1/\sqrt{1+x^2/3} ([50]), we attempted to find a simple analytical formula preserving both Eqns. (16) and (17). The best result, obtained using the integral form, Eqn. (21), reads

U≃13+1/y−2(1+y)ln⁡(1+1/y),y=2x.(18)U \simeq\frac{1}{3+1/y-2(1+y)\ln(1+1/y)}, \qquad y=\frac{\sqrt{2}}{x}. \tag*{(18)}

See Figure 2 to assess the quality of this simple formula, for which error relative to U(x)U(x) is plotted as a solid blue line. As we will demonstrate, such a simple formula is not generally applicable, and the behavior of the model depends strongly on the actual solution in the range not covered by either the series or asymptotic formulas. A high-precision numerical solution to Eqn. (12) for the elastic function is used as a reference in Figure 2. Possible approximations include: the Lane-Emden function with n=3n=3 (Figure 2, purple), n=13/3n=13/3 (Figure 2, magenta, dot-dashed) and the asymptotic formula (Figure 2, black). These approximations are summarized in Figure 2. For completeness, we recall that the n=0n=0 Lane-Emden function, θ0(x)=1−x2/6\theta_0(x)=1-x^2/6 (Figure 2, red, dashed), corresponds to a constant density ρ=ρ0\rho=\rho_0 ball in polytropic2, but not in the elastic model. For the latter, it is still a well-behaved approximation for U(x)U(x).

Figure 2. Numerical solution to (12) and various attempts to approximate it using known formulas. Vertical grid line shows ϵ=1/128\epsilon= 1/128 where initial conditions for numerical integration were imposed using 6th order series expansion (16). Curves display absolute value of the error relative to U(x)U(x) for simple asymptotics (black, solid), 2nd order series expansion at zero (red, dashed), n=13/3n = 13/3 Lane-Emden function θ13/3\theta_{13/3} (magenta, dot-dashed), analytical formula (18) preserving both 2nd series at zero and simple asymptotics (blue, solid), n=3n = 3 Lane-Emden θ3\theta_3 (purple, solid) and full asymptotic formula (20) (black, dotted).

Function UU is monotonic, so it makes sense to define its inverse U−1U^{-1}, which is uniquely defined for arguments in the interval (0,1)(0,1). Although conceptually useful, the inverse function can sometimes be challenging to work with in practice; therefore, we primarily rely on the parameters ηc\eta_c and xtx_t (see Figure 1) in subsequent derivations.

Elastic equation, Eqn. (12), has many interesting properties. If we introduce the reciprocal function w=1/uw=1/u, a particularly appealing form of it is obtained:

Δw=1w.(19)\Delta w=\frac{1}{w}. \tag*{(19)}

This can be considered as a Lane-Emden (A5b) equation with index n=−1n=-1 (McLaughlin & Pudritz 1997; Chavanis 2022). Kyczia & Filipjuk [47] classify it as a generalized Emden-Fowler equation with parameters, in their notation: α=2\alpha=2, δ=−1\delta=-1, n=0n=0 and p=−1p=-1. Despite its simplicity and many applications in astrophysics ([52, 10]), the logotropic equation, Eqn. (19), has not been studied in the detail required for this work. The simpler form of (19) compared to (12) is not surprising, given the function ww being a linear function of the specific enthalpy, Eqn. (6), which is known to be the optimal function for computations related to barotropic configurations ([26] (Odryzwollek 2003; Odryzwollek 2004)).

Initial conditions for ww, w(0)=1/u(0)=1w(0) = 1/u(0) = 1, w′(0)=−(1/u(0)2)u′(0)=0w'(0) = -(1/u(0)^2)u'(0) = 0, are identical to (14). A formal series solution for ww is

w(x)=∑n=0∞cn(2n+1)!x2n,w(x) = \sum_{n=0}^{\infty} \frac{c_n}{(2n+1)!}x^{2n},

where the coefficients obey the recurrence relation:

cn=12n2+n∑k=1n−12n+12k+1ck(k−n)cn−k.c_n = \frac{1}{2n^2+n}\sum_{k=1}^{n-1}\frac{2n+1}{2k+1}c_k(k-n)c_{n-k}.

The first coefficients are c0=1,c1=1,c2=−1,c3=13/3,c4=−125/3,…c_0 = 1, c_1 = 1, c_2 = -1, c_3 = 13/3, c_4 = -125/3, \ldots The radius of convergence can be estimated numerically from

lim⁡n→∞cncn+1(2n+3)(2n+2)≃4.\lim_{n\to\infty}\frac{c_n}{c_{n+1}}(2n+3)(2n+2) \simeq4.

The function

w(x)=x2,w(x) = \frac{x}{\sqrt{2}},

is an exact analytical solution to Eqn. (19), which, remarkably, also describes the asymptotic behavior of solutions to this equation as x→∞x \to\infty. However, the above asymptotics provide only a crude approximation for large xx. In the next section, we will need more terms, which can be found as follows. First, we substitute

w(x)=x2+ϵ ϕ(x),w(x) = \frac{x}{\sqrt{2}} + \epsilon\,\phi(x),

into (19) where ϵ\epsilon is a small perturbative parameter. A series expansion with respect to ϵ=0\epsilon= 0 gives:

(ϕ′′+2ϕ′x+2ϕx2)ϵ−22x3ϕ2ϵ2+O(ϵ3)=0,\left(\phi^{\prime\prime}+\frac{2\phi^{\prime}}{x}+\frac{2\phi}{x^2}\right)\epsilon-\frac{2\sqrt{2}}{x^3}\phi^2\epsilon^2+O(\epsilon^3)=0,

where the zeroth-order term in ϵ\epsilon vanishes. Ignoring the secular ϵ2\epsilon^2 term, we obtain a linear ODE:

ϕ′′+2xϕ′+2x2ϕ=0,\phi^{\prime\prime}+\frac{2}{x}\phi^{\prime}+\frac{2}{x^2}\phi=0,

with the analytical solution

ϕ(x)=Acos⁡(72ln⁡(x)+ϕ0)/x.(20)\phi(x)=A\cos\left(\frac{\sqrt{7}}{2}\ln(x)+\phi_0\right)/\sqrt{x}. \tag*{(20)}

The coefficient A≃0.4164A \simeq0.4164 and phase ϕ0≃0.1395\phi_0 \simeq0.1395 can be determined numerically to match the numerical solution (Figure 2, black dotted curve). Attempts to find these values analytically using textbook methods (Olver [63]) were unsuccessful. This subtle oscillation described by Eqn. (20) is numerically so small that it is barely visible on standard plots (cf. Figure 1, inset, or Fig. 18 in Chavanis [10]). Nevertheless, it plays a crucial role in explaining the unexpected spiraling behavior observed in the mass-radius relation (Figure 3), discussed further in Sect. 5. Due to its subtlety and unexpected nature, we explored its origin more deeply.

Figure 3. Mass-radius relation for elastic model: (a) full curve showing the relationship between mass and radius, with grid line crossing marking the asymptotic point; (b) zoomed view of the unstable branch.

The oscillatory behavior, which at first seems to emerge out of nowhere, can be understood in the following way. Changing the independent variable in (19) to x=etx=e^t we obtain:

W′′+W′=e2tW,W^{\prime\prime}+W^{\prime}=\frac{e^{2t}}{W},

where W(t)≡w(et)W(t)\equiv w(e^t). Rescaling W(t)=etz(t)W(t)=e^tz(t) leads to the autonomous system

z′′+3z′+2z−1/z=0.z^{\prime\prime}+3z^{\prime}+2z-1/z=0.

This equation can be cast in an easily recognizable energy-conservation form from classical mechanics:

{ddt[12v2+V(z)]=−κv2dzdt=v\begin{cases} \frac{d}{dt}\left[\frac{1}{2}v^2+V(z)\right]=-\kappa v^2\\ \frac{dz}{dt}=v \end{cases}

where the potential VV is

V(z)=z2−ln⁡z.V(z)=z^2-\ln z.

Therefore, the solution of the logotropic equation (19) is equivalent to classical one-dimensional motion in a potential well, V(z)V(z), with energy dissipation, resembling a typical damped oscillator system. The quadratic drag coefficient, κ=3\kappa=3, is very close to the threshold of 4, where oscillations vanish in the linear regime. This explains why oscillations were difficult to detect in the numerical results. The specific solution required for our problem is the one with infinite initial energy, falling down the potential well until the minimum, Vmin⁡=ln⁡2eV_{\min}=\ln\sqrt{2e}, is reached at z=1/2z=1/\sqrt{2}. The trajectory equivalent to function UU on {z,v}\{z,v\} plane is the one starting from the straight line v=−zv=-z.

We also note the Volterra integral equation equivalent to our elastic problem

w(x)=1+∫0x(s−s2x)dsw(s).(21)w(x)=1+\int_0^x\left(s-\frac{s^2}{x}\right)\frac{ds}{w(s)}. \tag*{(21)}

which can be used to find iteratively a variety of approximations, including Eqn. (18).

Physical Properties of the Elastic Model

Using U(x)U(x) and its inverse, we can derive formulas for all the interesting properties of the self-gravitating elastic body. We denote the scaling radius as

R02=K4πGρ02.(22)R_0^2=\frac{K}{4\pi G\rho_0^2}. \tag*{(22)}

Using sound speed, csc_s, at ρ0\rho_0

cs2=∂P∂ρ∣ρ0=Kρ0,(23)c_s^2=\left.\frac{\partial P}{\partial\rho}\right|_{\rho_0}=\frac{K}{\rho_0}, \tag*{(23)}

we can express the radial scale in terms of the sound crossing time: R0=1/(π3/8)csτ≃0.52csτR_0=1/(\pi\sqrt{3/8})c_s\tau\simeq0.52c_s\tau, where τ=3π/(32Gρ0)\tau=\sqrt{3\pi/(32G\rho_0)} is the dynamical free-fall collapse timescale of a uniform ball, assuming zero pressure. Note that comparing Eqns. (11) and (22), we get R0=ληcR_0=\lambda\eta_c, where the dimensionless central compression ratio is

ηc=ρcρ0.(24)\eta_c=\frac{\rho_c}{\rho_0}. \tag*{(24)}

The related dimensionless radius, xtx_t, is connected to ηc\eta_c (Figure 1) via the elastic function, U(x)U(x). The essential relation

1ηc=U(xt),(25)\frac{1}{\eta_c}=U(x_t), \tag*{(25)}

allows us to use ηc\eta_c or xtx_t interchangeably. Physically, this relation explicitly defines the surface of the body as the radius where density reaches the uncompressed density ρ0\rho_0. The mass scale is simply

M0=43πρ0R03.(26)M_0=\frac{4}{3}\pi\rho_0R_0^3. \tag*{(26)}

Example values of the uncompressed density ρ0\rho_0, bulk modulus KK, and the resulting scaling parameters, R0R_0 and M0M_0, for selected materials are given in Table 1. For liquids, the bulk modulus is directly proportional to the sound speed squared cs2c_s^2, via Newton-Laplace equation, Eq. (23). For isotropic solids, KK can be derived from any two other elastic moduli, e.g., K=13E/(1−2ν)K=\frac{1}{3}E/(1-2\nu), where EE is the Young modulus and ν\nu - Poisson ratio. Alternatively, using sound speeds, we recall that longitudinal cLc_L and shear cSc_S wave speeds are not identical and the proper formula is

NamePhaseSymbolρ0\rho_0 [g cm−3^{-3}]KK [GPa]K′K'R0R_0 [km]M0ML\frac{M_0}{M_L}Why incl.?Ref.
IronsolidFe7.874159.35.8617802.5Inner SSHuang et al. (2022)
MagnesiumsolidMg1.7374542207.4Inner SSKaye & Laby (1995)
SiliconsolidSi2.329100464013.3Inner SSKaye & Laby (1995)
QuartzsolidSiO2_22.6537.46.225202.4Inner SSWang et al. (2015)
NitrogeniceN2_20.852.1618880.32Outer SSYamashita et al. (2010)
NitrogenliquidN2_20.8070.711320.07Outer SSLide (2005)
WatericeH2_2O0.9178.835302.3Outer SSYoung et al. (2012)
MethaneliquidCH4_40.440.822200.27Outer SSLide (2005)
DiamondsolidC3.5154393.6653056Max. M0M_0Hu et al. (2021)
HeliumliquidHe0.1250.0045501.2e-3Min. M0M_0Lide (2005)
MercuryliquidHg13.534284250.06Min. R0R_0Young et al. (2012)
Boronα\alpha solidB2.362253.0694045Max. R0R_0Chuvashova et al. (2017)
OsmiumsolidOs22.594004.09701.2Max. ρ0\rho_0; KKDubrovinsky et al. (2015)
Marshmallowfoam-0.368e-5274e-7IllustrativeKelly (2014)
Quark matterspeculated-4.7e141.3e250?8.41.6e7Exotic/StrangeKutschera et al. (2020)

Table 1. Selected sample of matter with example values of the uncompressed density ρ0\rho_0, bulk modulus KK, and its pressure derivative K′K'. Resulting scale length (22) and mass scale (26) are provided. Mass scale is given relative to Moon (lunar) mass, denoted by MLM_L. Only liquid and solid state of matter is considered, and temperature dependence ignored. Due to assumed barotropic form of the EOS and various possible phases, especially for solids, the given values are only rough indicators of the possible measurement ranges. Selected materials represent astrophysically relevant, theoretically interesting, or extreme cases. Values of ρ0\rho_0 and KK are generally well-measured or inferred from sound-speed experiments, while reliable data for the pressure derivative K′K' are sparse due to experimental challenges and theoretical uncertainties. Thus, K′K' values are provided only where trustworthy references exist.

K=ρ0(cL2−43cS2).K=\rho_0\left(c_L^2-\frac{4}{3}c_S^2\right).

The materials selected for Table 1 illustrate the wide range of parameters relevant to astrophysical bodies and engineering applications, as well as theoretical extremes. Typical Solar System (SS) materials like iron, silicates (quartz), common ices (water, nitrogen ice), and liquids (liquid nitrogen, liquid methane) are included. The potential relevance of nitrogen ice is highlighted by hypotheses concerning ’Oumuamua ([21]). Exploring the parameter space further, Table 1 shows elements resulting in extreme scaling values: mercury (Hg, high ρ0\rho_0, low KK) and liquid helium (very low KK) are on the lower-left corner of the R0R_0, M0M_0 plane. Boron (B, [19]) and diamond (C, Hu et al. 2021) provide extremes on the opposite corner (large R0R_0, M0M_0), due to their very large values of KK and small density. Osmium, the common-sense record-holder for density ρ0\rho_0, surprisingly also possesses the highest known bulk modulus KK of any element ([24]). Notably, the calculated mass scale M0M_0 for many typical elements and compounds naturally yields values comparable to the mass of Earth’s Moon (MLM_L). This makes the lunar mass a convenient and physically relevant unit for expressing these scales, which vary between 10−3≲M0/ML≲5610^{-3}\lesssim M_0/M_L\lesssim56 (almost 5 orders of magnitude). The scaling radius across the periodic table of elements is within the range 400km≲R0≲7000km400\mathrm{km}\lesssim R_0\lesssim7000\mathrm{km} (a factor of ∼17\sim17). We also provided an amusing, illustrative example of marshmallow. Last but not least, strange quark matter is speculated ([46]) to exist in astrophysical environments. For the MIT bag model ([17]), as considered in astrophysical contexts ([74]), K=4B/3K=4B/3, where B=60 MeV fm−3B=60\ \mathrm{MeV}\ \mathrm{fm}^{-3} is a typical bag constant value. The value of KK can be only marginally affected by additional factors, such as the inclusion of a non-zero strange quark mass ([7]).

The presented values are intended to provide insight into the possible parameter ranges, using recent experimental and theoretical references where available, rather than a comprehensive survey, and values may depend on specific conditions (temperature, pressure, phase).

The quantities R0R_0 and M0M_0 provide the characteristic length and mass units (kilometers and masses comparable to the Moon), confirming their suitability for analyzing planetoid-sized objects (Table 1). We denote the total radius and mass of the body as RR and MM, respectively. The corresponding dimensionless radius and mass will be defined as R~=R/R0\widetilde{R}=R/R_0 and M~=M/M0\widetilde{M}=M/M_0, respectively.

The function U(x)U(x), with the appropriate length scale (11), leads to formulas for the physical description of the self-gravitating body, in analogy to the Lane-Emden and Chandrasekhar models (Appendix § 10). For the density stratification, we have

ρ(r)=ρcU(r/λ).(27)\rho(r)=\rho_c U(r/\lambda). \tag*{(27)}

The enclosed mass is

m(r)=κ μ(r/λ),κ=4πρcλ3,(28)m(r)=\kappa\,\mu(r/\lambda),\qquad\kappa=4\pi\rho_c\lambda^3, \tag*{(28)}

where μ\mu can be expressed using UU via Eqn. (15). Therefore, for the gravitational acceleration g=Gm/r2g=Gm/r^2, we have

g(r)=−4πGρcλU′(r/λ)U(r/λ)2.(29)g(r)=-4\pi G\rho_c\lambda\frac{U'(r/\lambda)}{U(r/\lambda)^2}. \tag*{(29)}

The maximum gg-force is always at the surface of the elastic body because g(r)g(r) is a monotonically increasing function.

The uncompressed density, ρ0\rho_0, provides a cutoff and defines the physical radius, RR, and mass, MM, of the body. Using Eq. (25) the dimensionless radius of the body is therefore explicitly given as:

R~=xtηc=xtU(xt).(30)\widetilde{R}=\frac{x_t}{\eta_c}=x_tU(x_t). \tag*{(30)}

Correspondingly, using Eqns. (25) and (15), the dimensionless mass can be succinctly expressed as:

M~=−3(xtηc)2U′(xt)U(xt)2=−3xt2U′(xt).(31)\widetilde{M}=-3\left(\frac{x_t}{\eta_c}\right)^2\frac{U'(x_t)}{U(x_t)^2}=-3x_t^2U'(x_t). \tag*{(31)}

The density contrast is

ρcρˉ=−xt3U(xt)2U′(xt)(32)\frac{\rho_c}{\bar{\rho}}=-\frac{x_t}{3}\frac{U(x_t)^2}{U'(x_t)} \tag*{(32)}

where the mean density ρˉ=M/(43πR3)\bar{\rho}=M/(\frac{4}{3}\pi R^3). The latter can be expressed in units of ρ0\rho_0:

ρˉρ0=−3U′(xt)xtU(xt)3(33)\frac{\bar{\rho}}{\rho_0}=-\frac{3U'(x_t)}{x_tU(x_t)^3} \tag*{(33)}

The gravitational potential at the center is given by:

Φg(0)=−GMR{1+[U(xt)−1]U(xt)xtU′(xt)}.\Phi_g(0)=-\frac{GM}{R}\left\{1+\frac{[U(x_t)-1]U(x_t)}{x_tU'(x_t)}\right\}.

Coefficient in curly braces increases from the well-known result of 3/23/2 for a uniform ball to 22 as xt→∞x_t\to\infty. The gravitational binding energy Ω\Omega is:

Ω=−GM22R{1−U(xt)xtU′(xt)−U(xt)43U′(xt)2}.(34)\Omega=-\frac{GM^2}{2R}\left\{1-\frac{U(x_t)}{x_tU'(x_t)}-\frac{U(x_t)^4}{3U'(x_t)^2}\right\}. \tag*{(34)}

To calculate the limit of the coefficient in curly braces for xt→0x_t\to0, which is, as expected from a well-known result for uniform ball, 6/56/5, at least the 4-th order expansion (16) of UU is required. For xt→∞x_t\to\infty, using (17), we obtain 4/34/3.

Mass-Radius Relation and Stability of Elastic Bodies

A crucial tool in the analysis of self-gravitating bodies with a given EOS is the mass-radius relation, M(R)M(R). For the elastic model, it is given in parametric form by equations (30) and (31) and is plotted in Figure 3. The callout points on the mass-radius curve in Figure 3 indicate the values of the central overdensity parameter, ηc\eta_c. As a sanity check, we note that for a body of constant density, the dimensionless relation is simply

M~=R~3.\widetilde{M}=\widetilde{R}^{3}.

In the main panel (a) of Figure 3 the above relation (a cubic function) is shown as a dot-dashed line. The constant density approximation is visibly inaccurate even at 10% compression of the initial density. Surprisingly, the elastic M(R)M(R) curve (Figure 3, (a), solid line) does not extend indefinitely, as one might expect from the constant density approximation, but terminates.

From the asymptotic formula (17), using (30, 31) we can easily find the terminating point, which is achieved in the limit of the central compression parameter ηc→∞\eta_c\to\infty:

R~∞=2,M~∞=32.\widetilde{R}_{\infty}=\sqrt{2},\qquad\widetilde{M}_{\infty}=3\sqrt{2}.

In other words, in the asymptotic case, the mass-radius relation degenerates to a single point, shown as the intersection of grid lines in Figure 3 (a).

The behavior near the tip of the M(R)M(R) curve is non-trivial. Before the terminating point, we first reach the maximum radius (Figure 3, inset panel) at

R~1≃1.462,M~1=4.387.\widetilde{R}_1\simeq1.462,\qquad\widetilde{M}_1=4.387.

This is attained at xt≃5.21x_t\simeq5.21, which is equivalent to ηc≃3.562\eta_c\simeq3.562. Not far from that point, at ηc≃5.134\eta_c\simeq5.134, xt≃7.46x_t\simeq7.46, we achieve the maximum mass:

R~2≃1.453,M~2=4.481,\widetilde{R}_2\simeq1.453,\qquad\widetilde{M}_2=4.481,

cf. panel (b) of Figure 3.

The M(R)M(R) curve then begins to spiral towards the terminating point, forming a sequence (possibly infinite) of alternating RR and MM extrema. This conclusion is supported by the asymptotic behavior described in equation (20), and the distance from the terminating point decreases proportionally to ηc−3/2\eta_c^{-3/2}.

The shape of the M(R)M(R) curve, Figure 3 (b), allows us to discuss some aspects of stability for the elastic model (Bisnovatyi-Kogan et al. [5], see Sect. 12.3.1). The turning point theorem (e.g., Zel’dovich [79]; Harrison et al. [32]; Hadžić & Lin [31]) indicates that the first mass peak marks the onset of dynamical instability. A body with mass above the maximum cannot exist in hydrostatic equilibrium and is, therefore, dynamically unstable. For the branch where the solution for a given mass is not unique (see Figure 3b), secular instability occurs. A process dissipating the binding energy gap can bring the body to a state of smaller radius (the ground state). This is possible for masses above the second extremum (1st local minimum) of mass, at ηc≃55.97\eta_c \simeq55.97, xt≃79.1x_t \simeq79.1, where we have

R~3≃1.413,M~3=4.236.\widetilde{R}_3 \simeq1.413,\qquad\widetilde{M}_3 = 4.236.

The secular instability limit is shown as the horizontal grid line labeled M~3\widetilde{M}_3 in Figure 3b, just below the line denoting the mass of the terminating point at M~=32\widetilde{M} = 3\sqrt{2}.

Remarkably, the instability leading to a maximum mass (M2M_2) is of purely Newtonian nature. To confirm this, we compare the Schwarzschild black hole radius RgR_g associated with the characteristic mass scale M0M_0, to the model’s characteristic radius R0R_0:

RgR0=2GM0/c2R0=23K/ρ0c2=23(csc)2,\frac{R_g}{R_0}=\frac{2GM_0/c^2}{R_0}=\frac{2}{3}\frac{K/\rho_0}{c^2}=\frac{2}{3}\left(\frac{c_s}{c}\right)^2,

where csc_s is the sound speed (23) in the medium and cc is the speed of light. Using iron as an example, this ratio is extremely small, Rg/R0≃1.6×10−10R_g/R_0 \simeq1.6 \times10^{-10}, confirming that general relativistic effects are entirely negligible in determining the structure and stability limits found in this model.

One of the motivations for initiating this study was a question, whether one could manufacture spherical bodies with average density much higher than those found in heavy precious metals (platinum, gold, tungsten, osmium, cf. Table 1). Such dense, spherical bodies could be then used (Hills [33]; Feldman et al. [28]) in deep space experiments (Armano et al. [2]) to measure the gravitational constant, GG (Rothleitner & Schlamminger [65]). Naively, without our analysis, for large ρ0\rho_0 and small KK one would expect significant compression under self-gravity. The elastic model can answer this question precisely. Using Eqn. (33), the maximum average density of 1.508ρ01.508\rho_0 is achieved for xt≃22.1x_t \simeq22.1 equivalent to ηc≃15.6\eta_c \simeq15.6. This is deep into the unstable branch of the M(R)M(R) curve, cf. inset (b) of Figure 3. Therefore, in practice, the maximum ρˉ≃1.46ρ0\bar{\rho} \simeq1.46\rho_0 is found for maximum mass threshold M2M_2. The question is therefore answered negatively within this model: stable configurations cannot achieve an average density more than about 50% higher than the initial density ρ0\rho_0, regardless of how small the bulk modulus is assumed to be.

In summary, the elastic body exhibits secular instability when its dimensionless mass exceeds a critical threshold, specifically M~>M~3≈4.236\widetilde{M} > \widetilde{M}_3 \approx4.236. Furthermore, it transitions to a state of dynamic instability when the dimensionless mass surpasses a higher critical value, namely M~>M~2≈4.481\widetilde{M} > \widetilde{M}_2 \approx4.481. These findings, illustrated by the spiraling behavior of the mass-radius curve in Figure 3b, encapsulate the primary conclusions drawn from the preceding detailed analysis.

Universal Quadratic Profile

In Section 3, we announced various analytical approximations (Figure 2) to the elastic function U(x)U(x) with the goal of deriving simple analytical formulas. Using the asymptotic formula, Eq. (17), yields only the trivial result R~=R~∞=2\widetilde{R}=\widetilde{R}_{\infty}=\sqrt{2}, M~=M~∞=32\widetilde{M}=\widetilde{M}_{\infty}=3\sqrt{2}, i.e., the mass-radius curve degenerates to a single point.

However, a more useful approximation is obtained by replacing the elastic function, U(x)U(x), with the n=0n=0 Lane-Emden function, θ0(x)\theta_0(x), which is equivalent to the series expansion (16) truncated at the quadratic term. Substituting U(x)→1−x2/6U(x)\to1-x^2/6 into equations (30) and (31) leads to

R~=61−1/ηcηc,M~=66(1−1/ηc)3/2.(35)\widetilde{R}=\sqrt{6}\frac{\sqrt{1-1/\eta_c}}{\eta_c},\qquad\widetilde{M}=6\sqrt{6}(1-1/\eta_c)^{3/2}. \tag*{(35)}

The formula for compactness,

M~R~=6μc\frac{\widetilde{M}}{\widetilde{R}}=6\mu_c

is particularly simple.

The corresponding density profile (27), expressed in terms of the radial variable, becomes:

ρ(r)=ρc(1−r2R2)+ρ0r2R2.(36)\rho(r)=\rho_c\left(1-\frac{r^2}{R^2}\right)+\rho_0\frac{r^2}{R^2}. \tag*{(36)}

From this, the mass is

M=43πR3(25ρc+35ρ0)=43πR3ρ0(1+25μc),(37)M=\frac{4}{3}\pi R^3\left(\frac{2}{5}\rho_c+\frac{3}{5}\rho_0\right)=\frac{4}{3}\pi R^3\rho_0\left(1+\frac{2}{5}\mu_c\right), \tag*{(37)}

allowing for a quick estimate of the central density given MM, RR and ρ0\rho_0. The moment of inertia is given by the similarly elegant formula:

I=815πR5(27ρc+57ρ0)=25MR21+2μc/71+2μc/5.(38)I=\frac{8}{15}\pi R^5\left(\frac{2}{7}\rho_c+\frac{5}{7}\rho_0\right)=\frac{2}{5}MR^2\frac{1+2\mu_c/7}{1+2\mu_c/5}. \tag*{(38)}

Due to central compression, the moment of inertia for an elastic body is smaller than that of a uniform ball:

I≃25MR2(1−435μc+8175μc2+⋯ )I\simeq\frac{2}{5}MR^2\left(1-\frac{4}{35}\mu_c+\frac{8}{175}\mu_c^2+\cdots\right)

We also present formulas for gravitational potential at the center,

Φg(0)=−πG(ρc+ρ0)R2,(39)\Phi_g(0)=-\pi G(\rho_c+\rho_0)R^2, \tag*{(39)}

and binding energy,

Ω=−16π2GR5315(4ρc2+10ρcρ0+7ρ02).(40)\Omega=-\frac{16\pi^{2}GR^{5}}{315}\left(4\rho_{c}^{2}+10\rho_{c}\rho_{0}+7\rho_{0}^{2}\right). \tag*{(40)}

which might become useful for quick estimates of the energy required to dismantle potentially hazardous near-Earth asteroids, e.g., the recently tracked 2024 YR4 (NASA Center for Near-Earth Object Studies 2025).

Eliminating ηc\eta_{c} from Eqns. (35) yields the dimensionless M~(R~)\widetilde{M}(\widetilde{R}) curve, described by a cubic polynomial:

M~3+18M~2R~+M~(108R~3−216)+216R~3=0.\widetilde{M}^{3}+18\widetilde{M}^{2}\widetilde{R}+\widetilde{M}\left(108\widetilde{R}^{3}-216\right)+216\widetilde{R}^{3}=0.

A notable property of the simple profile given by Eq. (36) is its universality. As we will demonstrate in the next § 7, for small overdensities, this profile matches the quadratic expansion common to many physical EOS at low compression. Furthermore, as an approximation, its applicability extends far beyond theoretical expectations. Analogous formulas could be derived for higher-order expansion, but they became too complicated to be practical.

Comparison with Realistic Iron EOS Models

To compare the elastic model with more realistic cases, we begin with a survey of EOSs. A common starting point for a higher-order analytical description of elastic matter is the Birch–Murnaghan [4, 57] EOS

P(ρ)=32K(x7−x5)×{1+34(K′−4)(x2−1)}.(41)P(\rho)=\frac{3}{2}K\left(x^{7}-x^{5}\right)\times\left\{1+\frac{3}{4}\left(K^{\prime}-4\right)\left(x^{2}-1\right)\right\}. \tag*{(41)}

where

x=(ρρ0)1/3.x=\left(\frac{\rho}{\rho_{0}}\right)^{1/3}.

The series expansion of the equation (41) at ρ=ρ0\rho=\rho_{0} agrees with (2) up to quadratic order only when K′=0K^{\prime}=0. Dimensionless coefficient K′K^{\prime} is

K′=∂K∂P∣P=0.K^{\prime}=\left.\frac{\partial K}{\partial P}\right|_{P=0}.

At high densities, Eq. (41) becomes polytropic with an index n=1/2n=1/2:

P(ρ)∼9K8(K′−4)(ρρ0)3.P(\rho)\sim\frac{9K}{8}\left(K^{\prime}-4\right)\left(\frac{\rho}{\rho_{0}}\right)^{3}.

Therefore, the n=1/2n=1/2 Lane-Emden function is sometimes used to find a crude approximation for Earth-like planetary core. An exceptional case for the Birch-Murnaghan EOS occurs when K′=4K^{\prime}=4, where the asymptotic is different:

P(ρ)∼3K2(ρρ0)7/3,P(\rho)\sim\frac{3K}{2}\left(\frac{\rho}{\rho_{0}}\right)^{7/3},

i.e., n=3/4n=3/4. Note that the EOS (41) becomes unphysical for K′<4K^{\prime}<4 if densities are larger than

ρmax⁡=ρ0(−29K′2−84K′+241+21K′−9827(K′−4))3/2,\rho_{\max}=\rho_{0}\left(\frac{-2\sqrt{9K^{\prime2}-84K^{\prime}+241}+21K^{\prime}-98}{27\left(K^{\prime}-4\right)}\right)^{3/2},

because pressure starts to decrease with increasing density.

Another option for incorporating stiffening of matter (increasing bulk modulus), especially for small K′<4K^{\prime}<4, is to generalize logotropic EOS, in a manner similar to how we derived it from Eqn. (1). The generalized elastic EOS is then obtained by replacing K→K+K′PK\to K+K^{\prime}P in definition (1) and solving the resulting ODE, producing:

P(ρ)=KK′[(ρρ0)K′−1].(42)P(\rho)=\frac{K}{K^{\prime}}\left[\left(\frac{\rho}{\rho_{0}}\right)^{K^{\prime}}-1\right]. \tag*{(42)}

The limit of equation (42) as K′→0K^{\prime}\to0 [51], obtained, for example, using L’Hospital’s rule, is equal to formula (2).

We remark that in the limit of small central compression ηc≃1\eta_c \simeq1 (dilatation μc≪1\mu_c \ll1), the elastic EOS is equivalent to the linear EOS:

P=K(ρ/ρ0−1)=Kμ.(43)P = K(\rho/\rho_0 - 1) = K\mu. \tag*{(43)}

The generalized elastic EOS (42) is also linear when K′=1K' = 1.

If we limit analysis to a specific, well-studied material, like iron, a large set of EOSs for small overdensities is used in engineering – for example, in crash simulations. Among these, we can list the Compaction EOS (a cubic polynomial in μ\mu), Gruneisen EOS [27], and the linear EOS (43). Notably, engineering EOSs usually extend below ρ0\rho_0, due to possible tensile stretching of matter. In astrophysics, this type of behaviour is very unlikely, unless we need to model unusual situations, such as those involving central cavities. The natural occurrences of large spherical central cavities seems unlikely, except in very exotic scenarios, for instance, those involving micro black holes [20]. However, this might be interesting for futuristic scenarios like drilling space habitats or space mining. We do not go into this discussion further and assume ρ>ρ0\rho> \rho_0 for the remainder of this work.

A recent article by Wu et al. (2023) discusses iron EOS over a wide range of temperatures, phases and densities. For our purposes, it is sufficient to use their fit to the cold AP2 model of Holzapfel [34]:

P=3K1−xx5ea0(1−x)(1+a2x(1−x)),x=(ρ0/ρ)1/3,(44)P = 3K\frac{1-x}{x^5}e^{a_0(1-x)}(1+a_2x(1-x)),\quad x=(\rho_0/\rho)^{1/3}, \tag*{(44)}

where a0=−ln⁡(3K/PFGr)a_0=-\ln(3K/P_{\mathrm{FGr}}), a2=(3/2)(K′−3)−a0a_2=(3/2)(K'-3)-a_0, PFGr=aFG(Z/V0)5/3P_{\mathrm{FGr}}=a_{\mathrm{FG}}(Z/V_0)^{5/3}. In SI units, aFG=2.3336×10−38a_{\mathrm{FG}}=2.3336\times10^{-38} [J m2^2] and V0=mFe/ρ0V_0=m_{\mathrm{Fe}}/\rho_0 where the atomic mass for 56Fe^{56}\mathrm{Fe} is mFe=55.934936m_{\mathrm{Fe}}=55.934936 u.

State-of-the-art (SOTA) experimental data for high-density 56Fe^{56}\mathrm{Fe} EOS have been presented by Smith et al. [67]. Their Supplementary Table 1 provides both the EOS and the corresponding mass-radius relation. The data starts at compression η≃1.3\eta\simeq1.3 (Figure 4). Comparing various analytical formulas with numerical data (Figure 4) we found AP2 EOS (44) performs exceptionally well for high densities, with Birch-Murnaghan (41) being the next best. Surprisingly, the simple quadratic EOS,

Figure 4. Comparison of high-density iron EOS measured at National Ignition Facility and various analytical formulas. Data points and errors are from Smith et al. [67]. Parameters for analytical formulas, from Wu et al. (2023), Table V, α\alpha-phase are: ρ0=7.927 g cm−3\rho_0 = 7.927\ \mathrm{g\ cm^{-3}}, K=164.8 GPaK = 164.8\ \mathrm{GPa}, K′=5.55K' = 5.55. Respective formulas are: Elastic EOS (2), Birch–Murnaghan (41), Generalized elastic (Power-law) (42), Linear (43), Quadratic (45) and Holzapfel AP2 (44).

P(μ)=Kμ−12K(1−K′)μ2,(45)P(\mu)=K\mu-\frac{1}{2}K(1-K')\mu^2, \tag*{(45)}

provides a fit that is almost as good (Figure 4, purple) as these more sophisticated approaches. Notably, equations (41), (42) and (44) all3^{3} have an identical second-order expansion given by (45) and begin to differ only at the third-order expansion in μ\mu or higher. Therefore, we will use EOS of the form (45) to assess the validity of our elastic model. It is worth noticing that for ρ<11\rho< 11 g cm−3^{-3}, the experimental data (Figure 4, data points with error bars) tend towards the linear and elastic EOS predictions. If this trend is more pronounced for materials other than iron, then the elastic model may, in fact, have a wider range of applicability than indicated by subsequent analysis.

To quantitatively measure the quality of the elastic model approximation, we perform a direct comparison of the internal density profiles (Figure 5). For this, we require a reference model valid from η=1\eta=1 upwards. Available high-density experimental data ([67]) (Figure 4) exhibits significant error bars and starts at η≈1.3\eta\approx1.3. Constructing a smooth numerical reference across the full range by ’gluing’ data with analytical formula is troublesome. We therefore adopt the quadratic EOS (45) as a physically grounded and computationally convenient reference proxy. It provides a single, smooth analytical form valid from η=1\eta=1, correctly reflects expected low-compression behavior via its second-order expansion, and reasonably approximates the high-density data trend seen in Figure 4.

Figure 5. Comparison of profile densities. Legend shown in panel (a) is common for all four plots.

The specific parameters used to generate the reference (quadratic EOS) profiles in Figure 5 correspond to α\alpha-phase iron (bcc), taken from Table V in Wu et al. (2023): ρ0=7.927\rho_0=7.927 g cm−3^{-3}, K=164.8K=164.8 GPa, and K′=5.55K'=5.55.

Four representative cases are presented in Figure 5. First, we consider an example with a central overdensity 0.1% (dilatation μc=10−3\mu_c=10^{-3}) shown in the upper-left panel (a). In this regime, all the compared models produce essentially identical, overlapping curves, demonstrating that the elastic model performs well. The total mass, M(a)=8.55×1019M_{(a)}=8.55\times10^{19} kg, falls in the range of objects size like Mimas or Vesta (0.32 M(a)M_{(a)}). The dimensionless radius, R~(a)<0.1\widetilde{R}_{(a)}<0.1 (with R0≃1768R_0\simeq1768 km, cf. Table 1) is well in the elastic regime. For smaller objects, representing the majority of asteroids and smaller moons, the elastic model is therefore a very good choice, and offers an improvement over the constant density approximation.

Increasing the central overdensity to 1% (Figure 5, panel (b)), we begin to observe the differences between the models. Two groups emerge: elastic and linear EOS tend to underestimate density (and consequently, the radius) to a similar extent, while generalized elastic EOS still matches the reference model closely. However, discrepancy remains small. We are now in the mass range of M(b)=2.75×1021 kgM_{\mathrm{(b)}} = 2.75 \times10^{21}\ \mathrm{kg}, comparable to objects like Ceres or Pluto (0.21 M(b)0.21\ M_{\mathrm{(b)}}). The elastic model remains an adequate approximation, particularly when the primary goal is to perform quick, back-of-the-envelope analytical calculations. The dimensionless radius is still less than 1 (Figure 5 (b), upper-right panel, upper axis).

The third example, with an overdensity at the core of 10%, equivalent to central dilatation μc=10−1\mu_{\mathrm{c}} = 10^{-1}, brings us into the realm of larger objects. The mass, M(c)=1023 kgM_{\mathrm{(c)}} = 10^{23}\ \mathrm{kg}, is equivalent to 0.31 of Mercury’s mass, or 1.4 times the Moon’s mass. Dimensionless radius R~(c)∼1\widetilde{R}_{\mathrm{(c)}} \sim1 marks the approximate limit of the elastic regime. As we see in Figure 5 (c) (lower-left panel, red dot-dashed curve), the elastic model begins to significantly diverge from the reference result.

Finally, we presented a case with a large central overdensity, double the uncompressed density ρ0\rho_0. None of the proposed analytical EOSs ((2), (43), (42)) can reproduce the reference profile obtained with the quadratic EOS (45). This is particularly surprising for the generalized elastic EOS (42), which has a second-order expansion identical to that of Eq. (45). However, the quadratic fit (36) remains surprisingly accurate as evident in Figure 5 (d) (blue dashed curve). This case falls within the true planetary range, with a mass of M(d)=5.17×1024 kgM_{\mathrm{(d)}} = 5.17 \times10^{24}\ \mathrm{kg}, or 0.87 M⊕0.87\ M_{\oplus}. The radius is approximately 2.5R02.5R_0, whereas we showed in Sect. 5 that the maximum possible radius for the elastic model is slightly below 1.5R01.5R_0.

We emphasize that this comparison is based on a specific study (Smith et al. [67]) of the iron EOS, which is well approximated using (45). For generic matter with arbitrary K′K', especially for small, say K′<4K' < 4, the range of validity might be different. Nevertheless, as a general guideline, values of dimensionless radius R~>1\widetilde{R} > 1 or dimensionless mass

M~>1\widetilde{M} > 1 clearly indicate that the elastic model is no longer applicable. Typical values for a range of materials are gathered in Table 1.

In summary, even for a well-studied material like iron, the cold barotropic EOS remains remarkably complex topic and subject to significant uncertainties. The modeling of objects such as Mercury’s core (composed of up to 95% Fe) ([69, 29, 66, 71]) and iron asteroids ([1, 72, 59]) continues to be an area of active research. The theoretical challenges involved in modeling even these relatively well-understood objects are considerable. It is therefore easy to envision the difficulties that future researchers will encounter when confronted with a mysterious interstellar object, like ’Oumuamua ([22]) or Borisov (Jewitt & Luu 2019), of unusual composition similar to, for instance, the suspected interstellar meteoroid Hypatia ([45]), which unexpectedly enters the Solar System. Rapidly evaluating a multitude of hypotheses in such a scenario necessitates a simple, analytical tool capable of encompassing a broad range of parameters. The elastic model, with its derived formulas, provides such a capability.

Conclusions

We have derived a solution for the internal structure and properties of a perfectly elastic body characterized by a constant bulk modulus, KK, and an initial density, ρ0\rho_0. All properties are determined by a single univariate special function, U(x)U(x), which is the solution to the ODE given by Eq. (12). This function exhibits well-defined behavior at x→0x \to0, eq. (16), and x→∞x \to\infty, eq. (17). It can be approximated by simple analytical formulas, e.g., (18). The properties of the model for specific astrophysical bodies are recovered through appropriate scaling of the radius (22) and mass (26), and depend on a single dimensionless parameter, ηc\eta_c, Eqns. (24),(25), which represents the ratio of central density to initial density. The scaling formula (22) for the radius is structurally identical to the well-known limits of asphericity Melosh [55], up to a multiplicative factor of the order of unity. However, the interpretation of the constant KK, measured in pressure units (usually GPa), is different: it represents the bulk modulus for our model, whereas in the aspherical case it is the tensile strength. From dimensional analysis, the combination K1/2G−1/2ρ0−1K^{1/2}G^{-1/2}\rho_0^{-1} is the only one that leads to the length units, cf. (22).

Our theoretical analysis has uncovered a complex and nuanced behavior in the model. Initially, for radii r≪R0r \ll R_0, the elastic model behaves as one would anticipate for incompressible matter, i.e., the body’s size increases while maintaining a constant density. Subsequently, the effects of elasticity become significant, and the model remains a valid approximation up to a dimensionless radius of R~<0.25\tilde{R} < 0.25, corresponding to ηc<1.01\eta_c < 1.01. Beyond this point, the coefficient K′K' can no longer be neglected, particularly for materials similar to iron, where K′∼5K' \sim5 (Table 1). In such cases, the elastic model breaks down when ηc≳2\eta_c \gtrsim2 (Figure 5d). However, theoretically, for hypothetical matter with K′≪1K' \ll1, the elastic model’s applicability could extend to higher compressions, allowing us to explore its behavior even beyond. When the central density, ρc\rho_c, exceeds 3ρ03\rho_0, the behavior of the system becomes particularly intricate. At ηc≃3.562\eta_c \simeq3.562 (Figure 3b) the object reaches its maximum radius of 1.462R01.462R_0. With a further increase in mass, at ηc≃5.134\eta_c \simeq5.134, the maximum mass of 4.481M04.481M_0 is attained. The mass-radius curve then deflects, entering the unstable branch. Ultimately, the mass-radius curve spirals towards the asymptotic point of the M(R)M(R) relationship. This asymptotic state is approached as ηc→∞\eta_c \to\infty, and is characterized by the coordinates R∞=2R0R_\infty= \sqrt{2}R_0, M∞=32M0M_\infty= 3\sqrt{2}M_0.

To evaluate the practical applicability of the elastic model, we employed a quadratic fit to the experimentally determined equation of state for iron. Remarkably, the simple EOS given by Eqn. (45) reproduces the numerical data with comparable accuracy to the established (Eq. (41)) and modern AP2 (Eq. (44)) analytical approximations for the iron EOS (Figure 4). This analysis indicates that the scale R0R_0 (22) provides a reasonable estimate for the range of validity of the elastic model, see Figure 5.

From an astronomy and astrophysics education perspective, a new analytically solvable model is invaluable in the era of Large Language Models (LLMs). Traditional exercises (e.g. Lane-Emden and Chandrasekhar models) are readily solved in numerous textbooks and online resources, therefore, they are already present in training data for these AI systems. These tasks can be solved instantly by any sufficiently advanced AI. This presents a growing challenge for university-level instruction, particularly for homework assignments, making them less effective as tools for genuine learning. The model presented in this work is new, and for a time, it will pose a genuine challenge to these systems, compelling students to engage in substantial intellectual effort to grasp fundamental concepts, such as the mass-radius M(R)M(R) relationship, by considering readily visualized examples like liquid bodies or iron spheres (representing asteroids). We believe our article provides a refreshing bird’s-eye view and serves as a modern introduction to the topic of simple, spherically symmetric, self-gravitating bodies in hydrostatic equilibrium.

Several simple formulas derived in this work, e.g., Eqs. (36)-(40), could be valuable for future asteroid mining endeavors, offering analytical estimates of central compression and internal density profiles that go beyond the trivial ρ=const\rho= const approximation. More precise results can be obtained using Eqns. (30)-(32).

For possible follow-up, we highlight the extraordinary effectiveness of simple quadratic EOS (45) in approximating high-pressure experimental data. While the focus of this work was on the elastic (logotropic) model, the superior performance of the quadratic EOS in fitting real-world data suggests that the resulting ODE (A5f) warrants a similarly detailed mathematical investigation. Such an analysis could potentially lead to an even more accurate and widely applicable model for the internal structure of liquid/solid self-gravitating bodies.

Future work could extend this model to piecewise EOS, with e.g. polytropic core and elastic mantle (Weppner et al. [73]), which might be particularly useful for objects with homogeneous outer layer, like icy moons or exoplanets (Kozak & Wojnar [44]). Extending to rotating bodies (Chowdhury et al. [18]) following Chandrasekhar & Milne [9] would improve asteroid breakup predictions (Drahus et al. [23]) due to YORP effect (Bottke et al. [6]) beyond constant-density models.

Acknowledgments

We are grateful to Prof. Piotr Bizoń for the derivation of the autonomous system equivalent to the logotropic equation, and Prof. P. Mach for the discussion on the existence of its solution. We also would like to thank Prof. Ł. Bratek for clarification on the bulk modulus of the quark matter. We also thank the anonymous referee for valuable insights.

Software: Julia [3], Wolfram Mathematica [75]

Simultaneous Derivation of Elastic, Lane-Emden, and Chandrasekhar Models

This appendix presents a simultaneous derivation of the internal structure and properties for an elastic model, alongside several other analytical EOSs commonly used in astrophysical modeling. The well-known Lane-Emden and Chandrasekhar theories are included for comparison, allowing the derivations to proceed in a familiar manner. This approach highlights the similarities and differences between our model and established calculations. We begin with the following EOSs:

P=Kln⁡η(46)P = K \ln\eta \tag*{(46)}
P=Kη1+1/n(47)P = K\eta^{1+1/n} \tag*{(47)}
P=Kf(η1/3)(48)P = K f\left(\eta^{1/3}\right) \tag*{(48)}
P=Kb(η1/3)(49)P = K b\left(\eta^{1/3}\right) \tag*{(49)}
P=Kμ(50)P = K\mu \tag*{(50)}
P=Kμ−12K(1−K′)μ2(51)P = K\mu- \frac{1}{2}K(1-K')\mu^2 \tag*{(51)}
P=KK′(ηK′−1)(52)P = \frac{K}{K'}\left(\eta^{K'}-1\right) \tag*{(52)}

where the compression, η\eta, and dilatation, μ\mu, are defined in Eqns. (3) and (4), respectively. The core function for the Chandrasekhar EOS is:

f(x)=x(2x2−3)1+x2+3arsinh⁡xf(x) = x(2x^2-3)\sqrt{1+x^2}+3\operatorname{arsinh}x

and for the Birch-Murnaghan EOS, it is:

b(x)=32(x7−x5)[1+34(K′−4)(x2−1)]b(x) = \frac{3}{2}\left(x^7-x^5\right)\left[1+\frac{3}{4}(K'-4)(x^2-1)\right]

The specific enthalpy h=∫dp/ρh = \int dp/\rho is

h=Kρ0(1−1η)(53)h = \frac{K}{\rho_0}\left(1-\frac{1}{\eta}\right) \tag*{(53)}
h=Kρ0(1+n)η1/n(54)h = \frac{K}{\rho_0}(1+n)\eta^{1/n} \tag*{(54)}
h=8Kρ01+η2/3(55)h = \frac{8K}{\rho_0}\sqrt{1+\eta^{2/3}} \tag*{(55)}
h=316Kρ0(7η4/3(14−3K′)+5η2/3(3K′−16)+9η2(K′−4)−3(K′−6))(56)h = \frac{3}{16}\frac{K}{\rho_0}\left(7\eta^{4/3}(14-3K')+5\eta^{2/3}(3K'-16)+9\eta^2(K'-4)-3(K'-6)\right) \tag*{(56)}
h=Kρ0ln⁡η(57)h = \frac{K}{\rho_0}\ln\eta \tag*{(57)}
h=Kρ0(K′−1)(η−1−K′−2K′−1ln⁡η)(58)h = \frac{K}{\rho_0}(K'-1)\left(\eta-1-\frac{K'-2}{K'-1}\ln\eta\right) \tag*{(58)}
h=Kρ0ηK′−1−1K′−1(59)h = \frac{K}{\rho_0}\frac{\eta^{K'-1}-1}{K'-1} \tag*{(59)}

Before proceeding to scale and convert the hydrostatic equilibrium equations into dimensionless form, we discuss the general approach. For each EOS, three distinct approaches are possible: scaling the density by ρc\rho_c (Case I), scaling by ρ0\rho_0 (Case II), or shifting by ρ0\rho_0 and scaling by ρc−ρ0\rho_c-\rho_0 to the interval {0,1}\{0,1\} (Case III). In Case I, a single family of special functions is obtained, subject to the standard initial conditions given by (14). The solutions are determined by the properties at the terminating point, defined by the intersection with 1/ηc1/\eta_c. Ideally, ηc\eta_c would not appear in the resulting differential equation. In Case II, a family of solutions is obtained with varying initial conditions. Finally, in Case III, all driving parameters could, in principle, appear in the ODE itself. The primary criterion for selecting among Cases I-III was to obtain the simplest form of the final ODE, minimizing the number of EOS parameters that appear within the ODE.

Substituting the following expressions for ρ(r)\rho(r) into the hydrostatic equilibrium equation (Eq. (5)) and Poisson’s equation (or, equivalently, into Eq. (7)):

ρ(r)=ρcu(r/λ)(60)\rho(r)=\rho_c u(r/\lambda) \tag*{(60)}

(Elastic, Case I)

ρ(r)=ρcθ(r/λ)n(61)\rho(r)=\rho_c\theta(r/\lambda)^n \tag*{(61)}

(Polytropic, Case III)

ρ(r)=ρczc3(φ(r/λ)2−1zc2)3/2(62)\rho(r)=\rho_c z_c^3\left(\varphi(r/\lambda)^2-\frac{1}{z_c^2}\right)^{3/2} \tag*{(62)}

(Chandrasekhar, Case I)

ρ(r)=ρ0B(r/λ)(63)\rho(r)=\rho_0 B(r/\lambda) \tag*{(63)}

(Birch-Murnaghan, Case II)

ρ(r)=ρcL(r/λ)(64)\rho(r)=\rho_c L(r/\lambda) \tag*{(64)}

(Linear, Case I)

ρ(r)=ρ0Q(r/λ)(65)\rho(r)=\rho_0 Q(r/\lambda) \tag*{(65)}

(Quadratic, Case II)

ρ(r)=ρcR(r/λ)(66)\rho(r)=\rho_c R(r/\lambda) \tag*{(66)}

(Generalized Elastic, Case I)

Equation not extracted
(67)

where u(x)u(x) is the elastic function, θ(x)\theta(x) - Lane-Emden polytropic function and φ(x)\varphi(x) - Chandrasekhar white dwarf function, B(x)B(x) is Birch-Murnaghan function, L(x)L(x) is the function for linear EOS, Q(x)Q(x) - quadratic EOS function and R(x)R(x) denotes the function resulting from generalized elastic EOS. After some algebra (see attached Wolfram Language code) we obtain the following scaling relations:

λ2=K4πGρc2(68)\lambda^2=\frac{K}{4\pi G\rho_c^2} \tag*{(68)}

(Elastic)

λ2=K(n+1)4πGρcρ0(ρcρ0)1/n(69)\lambda^2=\frac{K(n+1)}{4\pi G\rho_c\rho_0}\left(\frac{\rho_c}{\rho_0}\right)^{1/n} \tag*{(69)}

(Polytropic)

λ2=2KπGρ021zc2(70)\lambda^2=\frac{2K}{\pi G\rho_0^2}\frac{1}{z_c^2} \tag*{(70)}

(Chandrasekhar)

λ2=K4πGρ02(71)\lambda^2=\frac{K}{4\pi G\rho_0^2} \tag*{(71)}

(Birch-Murnaghan)

λ2=K4πGρ0ρc(72)\lambda^2=\frac{K}{4\pi G\rho_0\rho_c} \tag*{(72)}

(Linear)

λ2=K4πGρ02(73)\lambda^2=\frac{K}{4\pi G\rho_0^2} \tag*{(73)}

(Quadratic)

λ2=K4πGρc2(ρcρ0)K′(74)\lambda^2=\frac{K}{4\pi G\rho_c^2}\left(\frac{\rho_c}{\rho_0}\right)^{K'} \tag*{(74)}

(Generalized Elastic)

The resulting dimensionless ordinary differential equations are listed below. The elastic equation is

Δu−2u′2u+u3=0.(75)\Delta u-\frac{2u'^2}{u}+u^3=0. \tag*{(75)}

(Elastic)

Lane-Emden polytropic equation [35] is

Δθ+θn=0.(76)\Delta\theta+\theta^n=0. \tag*{(76)}

(Polytropic)

Chandrasekhar’s white dwarf equation [42 37.1 Chandrasekhar’s Theory] is

Δφ+(φ2−1zc2)3/2=0.(77)\Delta\varphi+\left(\varphi^2-\frac{1}{z_c^2}\right)^{3/2}=0. \tag*{(77)}

(Chandrasekhar)

Equation related to Birch-Murnaghan EOS is too complicated, so we skip it. For linear EOS (43) we get

ΔL−L′2L+L2=0.(78)\Delta L-\frac{L^{\prime2}}{L}+L^2=0. \tag*{(78)}

For quadratic EOS (45) we got

ΔQ+(1−δ)Q3+δQ′2Q(Q−δ)=0,δ=K′−2K′−1.(79)\Delta Q+\frac{(1-\delta)Q^3+\delta Q^{\prime2}}{Q(Q-\delta)}=0,\qquad\delta=\frac{K^\prime-2}{K^\prime-1}. \tag*{(79)}

Finally, generalized elastic EOS leads to

ΔR+(K′−2)R′2R+R3−K′=0.(80)\Delta R+(K^\prime-2)\frac{R^{\prime2}}{R}+R^{3-K^\prime}=0. \tag*{(80)}

As intended, setting K′=0K^\prime=0 in the equation for the generalized elastic EOS recovers the elastic equation. Furthermore, in the limit K′→1K^\prime\to1 both the generalized elastic and quadratic EOSs reduce to the linear EOS, demonstrating that the linear EOS is a special case of both.

The initial conditions for the respective ODEs are:

u(0)=1,u′(0)=0,(Elastic)(81)u(0)=1,\qquad u^\prime(0)=0,\qquad\text{(Elastic)} \tag*{(81)}
θ(0)=1,θ′(0)=0,(Polytropic)(82)\theta(0)=1,\qquad\theta^\prime(0)=0,\qquad\text{(Polytropic)} \tag*{(82)}
φ(0)=1,φ′(0)=0,(Chandrasekhar)(83)\varphi(0)=1,\qquad\varphi^\prime(0)=0,\qquad\text{(Chandrasekhar)} \tag*{(83)}
B(0)=ηc,B′(0)=0,(Birch-Murnaghan)(84)B(0)=\eta_c,\qquad B^\prime(0)=0,\qquad\text{(Birch-Murnaghan)} \tag*{(84)}
L(0)=1,L′(0)=0,(Linear)(85)L(0)=1,\qquad L^\prime(0)=0,\qquad\text{(Linear)} \tag*{(85)}
Q(0)=ηc,Q′(0)=0,(Quadratic)(86)Q(0)=\eta_c,\qquad Q^\prime(0)=0,\qquad\text{(Quadratic)} \tag*{(86)}
R(0)=1,R′(0)=0.(Generalized Elastic)(87)R(0)=1,\qquad R^\prime(0)=0.\qquad\text{(Generalized Elastic)} \tag*{(87)}

The integration is terminated at a dimensionless radius xtx_t, defined by the following conditions:

u(xt)=1/ηc,(Elastic)(88)u(x_t)=1/\eta_c,\qquad\text{(Elastic)} \tag*{(88)}
θ(xt)=0,(Polytropic)(89)\theta(x_t)=0,\qquad\text{(Polytropic)} \tag*{(89)}
φ(xt)=1/zc,(Chandrasekhar)(90)\varphi(x_t)=1/z_c,\qquad\text{(Chandrasekhar)} \tag*{(90)}
B(xt)=1,(Birch-Murnaghan)(91)B(x_t)=1,\qquad\text{(Birch-Murnaghan)} \tag*{(91)}
L(xt)=1/ηc,(Linear)(92)L(x_t)=1/\eta_c,\qquad\text{(Linear)} \tag*{(92)}
Q(xt)=1,(Quadratic)(93)Q(x_t)=1,\qquad\text{(Quadratic)} \tag*{(93)}
R(xt)=1/ηc,(Generalized Elastic)(94)R(x_t)=1/\eta_c,\qquad\text{(Generalized Elastic)} \tag*{(94)}

respectively.

References

  1. [1]Anders, E. 1964, Space Science Reviews, 3, 583, doi: 10.1007/BF00177954DOI
  2. [2]Armano, M., Audley, H., Baird, J., et al. 2019, PhRvD, 100, 062003, doi: 10.1103/PhysRevD.100.062003DOI
  3. [3]Bezanson, J., Karpinski, S., Shah, V. B., & Edelman, A. 2012, arXiv preprint arXiv:1209.5145arxiv.org/abs/1209.5145
  4. [4]Birch, F. 1947, Phys. Rev., 71, 809, doi: 10.1103/PhysRev.71.809DOI
  5. [5]Bisnovatyi-Kogan, G., Blinov, A., & Romanova, M. 2011, Stellar Physics: 2: Stellar Evolution and Stability, Astronomy and Astrophysics Library (Springer Berlin Heidelberg). https://books.google.pl/books?id=dPS&xPp_66QC
  6. [6]Bottke, W. F., Vokrouhlický, D., Rubincam, D. P., & Nesvorný, D. 2006, Annual Review of Earth and Planetary Sciences, 34, 157, doi: https://doi.org/10.1146/annurev.earth.34.031405.125154
  7. [7]Bratek, L., Jalocha, J., & Kutschera, M. 2024, Phys. Rev. D, 110, 083041, doi: 10.1103/PhysRevD.110.083041DOI
  8. [8]Chandrasekhar, S. 1935, MNRAS, 95, 207, doi: 10.1093/mnras/95.3.207DOI
  9. [9]Chandrasekhar, S., & Milne, E. A. 1933, Monthly Notices of the Royal Astronomical Society, 93, 390, doi: 10.1093/mnras/93.5.390DOI
  10. [10]Chavanis, P.-H. 2015a, The European Physical Journal Plus, 130, 130, doi: 10.1140/epjp/i2015-15130-5
  11. [15]Chavanis, P.-H., & Kumar, S. 2017, Journal of Cosmology and Astroparticle Physics, 2017, 018, doi: 10.1088/1475-7516/2017/05/018
  12. [16]Chavanis, P.-H., & Sire, C. 2007, Physica A: Statistical Mechanics and its Applications, 375, 140, doi: https://doi.org/10.1016/j.physa.2006.08.076
  13. [17]Chodos, A., Jaffe, R. L., Johnson, K., Thorn, C. B., & Weisskopf, V. F. 1974, Phys. Rev. D, 9, 3471, doi: 10.1103/PhysRevD.9.3471
  14. [18]Chowdhury, S., Banerjee, P., & Wojnar, A. 2024, Phys. Rev. D, 109, 104049, doi: 10.1103/PhysRevD.109.104049
  15. [19]Chuvashova, I., Bykova, E., Bykov, M., et al. 2017, Journal of Solid State Chemistry, 245, 50, doi: 10.1016/j.jssc.2016.10.002
  16. [20]Dai, D.-C., & Stojkovic, D. 2024, Physics of the Dark Universe, 46, 101662, doi: https://doi.org/10.1016/j.dark.2024.101662
  17. [21]Desch, S. J., & Jackson, A. P. 2021, Journal of Geophysical Research: Planets, 126, e2020JE006807, doi: https://doi.org/10.1029/2020JE006807
  18. [22]Drahus, M., Guzik, P., Waniak, W., et al. 2018, Nature Astronomy, 2, 407, doi: 10.1038/s41550-018-0440-1
  19. [23]Drahus, M., Waniak, W., Tendulkar, S., et al. 2015, The Astrophysical Journal Letters, 802, L8, doi: 10.1088/2041-8205/802/1/L8
  20. [24]Dubrovinsky, L., Dubrovinskaia, N., Bykova, E., et al. 2015, Nature, 525, 226, doi: 10.1038/nature14681DOI
  21. [25]Eddington, A. S. 1926, The Internal Constitution of the Stars (Dover)
  22. [26]Eriguchi, Y., & Mueller, E. 1985, A&A, 146, 260
  23. [27]Fazio, D., Mulargia, F., & Boschi, E. 1978, Geophysical Journal International, 53, 531, doi: 10.1111/j.1365-246X.1978.tb03757.x
  24. [28]Feldman, M. R., Anderson, J. D., Schubert, G., et al. 2016, Classical and Quantum Gravity, 33, 125013, doi: 10.1088/0264-9381/33/12/125013
  25. [29]Fricker, P. E., Reynolds, R. T., Summers, A. L., & Cassen, P. M. 1976, Nature, 259, 293, doi: 10.1038/259293a0DOI
  26. [30]Guzik, P., Drahus, M., Rusek, K., et al. 2020, Nature Astronomy, 4, 53, doi: 10.1038/s41550-019-0931-8DOI
  27. [31]Hadžić, M., & Lin, Z. 2021, Communications in Mathematical Physics, 387, 729, doi: 10.1007/s00220-021-04197-6
  28. [32]Harrison, B. K., Thorne, K. S., Wakano, M., & Wheeler, J. A. 1965, Gravitation Theory and Gravitational Collapse
  29. [33]Hills, J. G. 1986, AJ, 92, 986, doi: 10.1086/114229DOI
  30. [34]Holzapfel, W. B. 2010, High Pressure Research, 30, 372, doi: 10.1080/08957959.2010.494845
  31. [35]Horedt, G. P. 2004, Astrophysics and Space Science Library, Vol. 307, Polytropes: Applications in Astrophysics and Related Fields (Springer Netherlands), doi: 10.1007/1-4020-2351-0
  32. [42]Kippenhahn, R., Weigert, A., & Weiss, A. 2012, Stellar Structure and Evolution, 2nd edn. (Springer). https://link.springer.com/book/10.1007/978-3-642-30304-3
  33. [43]Knopik, J., Mach, P., & Odrzywołek, A. 2017, Monthly Notices of the Royal Astronomical Society, 467, 4965, doi: 10.1093/mnras/stx164DOI
  34. [44]Kozak, A., & Wojnar, A. 2021, Phys. Rev. D, 104, 084097, doi: 10.1103/PhysRevD.104.084097
  35. [45]Kramers, J. D., Belyanin, G. A., Przybyłowicz, W. J., Winkler, H., & Andreoli, M. A. 2022, Icarus, 382, 115043, doi: https://doi.org/10.1016/j.icarus.2022.115043
  36. [46]Kutschera, M., Jalocha, J., Bratek, L., Kubis, S., & Kedziorek, T. 2020, The Astrophysical Journal, 897, 168, doi: 10.3847/1538-4357/ab97ab
  37. [47]Kycia, R. A., & Filipuk, G. 2015, Applied Mathematics and Computation, 265, 1003, doi: https://doi.org/10.1016/j.amc.2015.05.140
  38. [48]Lane, H. J. 1870, American Journal of Science, s2-50, 57, doi: 10.2475/ajs.s2-50.148.57DOI
  39. [50]Mach, P. 2012, Journal of Mathematical Physics, 53, 062503, doi: 10.1063/1.4725414DOI
  40. [51]Mahajan, S. 2020, American Journal of Physics, 88, 261, doi: 10.1119/10.0000841DOI
  41. [52]McLaughlin, D. E., & Pudritz, R. E. 1996, ApJ, 469, 194, doi: 10.1086/177771DOI
  42. [54]Meech, K. J., Weryk, R., Micheli, M., et al. 2017, Nature, 552, 378, doi: 10.1038/nature25020DOI
  43. [55]Melosh, H. J. 2011, Strength versus gravity, Cambridge Planetary Science (Cambridge University Press), 49–103DOI
  44. [56]Micheli, M., Farnocchia, D., Meech, K. J., et al. 2018, Nature, 559, 223, doi: 10.1038/s41586-018-0254-4DOI
  45. [57]Murnaghan, F. D. 1944, Proceedings of the National Academy of Sciences, 30, 244, doi: 10.1073/pnas.30.9.244DOI
  46. [59]Ockert-Bell, M., Clark, B., Shepard, M., et al. 2010, Icarus, 210, 674, doi: https://doi.org/10.1016/j.icarus.2010.08.002
  47. [63]Olver, F. W. J. 1974, Asymptotics and Special Functions (New York: Academic Press), doi: https://doi.org/10.1016/B978-0-12-528550-0.50003-1
  48. [64]Plummer, H. C. 1911, Monthly Notices of the Royal Astronomical Society, 71, 460, doi: 10.1093/mnras/71.5.460DOI
  49. [65]Rothleitner, C., & Schlamminger, S. 2017, Review of Scientific Instruments, 88, 111101, doi: 10.1063/1.4994619
  50. [66]Smith, D. E., Zuber, M. T., Phillips, R. J., et al. 2012, Science, 336, 214, doi: 10.1126/science.1218809DOI
  51. [67]Smith, R. F., Fratanduono, D. E., Braun, D. G., et al. 2018, Nature Astronomy, 2, 452, doi: 10.1038/s41550-018-0437-9DOI
  52. [68]Tooper, R. F. 1964, ApJ, 140, 434, doi: 10.1086/147939DOI
  53. [69]Urey, H. C. 1951, Geochimica et Cosmochimica Acta, 1, 209, doi: https://doi.org/10.1016/0016-7037(51)90001-4
  54. [71]Wardinski, I., Amit, H., Langlais, B., & Thé bault, E. 2021, Journal of Geophysical Research: Planets, 126, e2020JE006792, doi: https://doi.org/10.1029/2020JE006792
  55. [72]Wasson, J. T., Kallemeyn, G. W., Runcorn, S. K., Turner, G., & Woolfson, M. M. 1988, Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences, 325, 535, doi: 10.1098/rsta.1988.0066DOI
  56. [73]Weppner, S. P., McKelvey, J. P., Thielen, K. D., & Zielinski, A. K. 2015, Monthly Notices of the Royal Astronomical Society, 452, 1375, doi: 10.1093/mnras/stv1397DOI
  57. [74]Witten, E. 1984, Phys. Rev. D, 30, 272, doi: 10.1103/PhysRevD.30.272DOI
  58. [75]Wolfram Research, Inc. 2024, Mathematica, Version 14.2. https://www.wolfram.com/mathematica
  59. [79]Zel’dovich, Y. B. 1963, Voprosy Kosmonogii, 9, 157

Paper details

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

views · 0

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