betaFeedback

WorksPhysics and Astronomy

Nuclear stability and the Fold Catastrophe

Abstract

A geometrical analysis of the stability of nuclei against deformations is presented. In particular, we use Catastrophe Theory to illustrate discontinuous changes in the behavior of nuclei with respect to deformations as one moves in the N - Z space. We construct a minimalistic deformation model using the microscopic-macroscopic approach. A third-order phase transition is found in the liquid-drop model, which translates to a complete loss of stability (using the Fold catastrophe) when shell effects are included. The analysis is found to explain the instability of known fissile nuclei and also justify known decay chains of heavy nuclei.

Introduction

Developed in the 1970s by René Thom, Catastrophe theory[1] is a geometrical framework that explains discontinuous changes in dynamical systems. Thom showed that any function of r≤5r \le5 parameters can be mapped to one (and only one) of 11 known families of functions (catastrophes). One can thus study a wide range of systems by mapping their potential to one of these catastrophes. These catastrophes have been well-studied for discontinuous features, which can be mapped back to the system of interest to find where these changes occur. The proof of the theorem can be found in [2] and [3]; we shall take it as a given.

The theory’s application to quantum systems is trickier since one must use semi-classical methods to write a classical version of the quantum Hamiltonian. In particular, nuclei present a suitable testing ground for this since nuclear decays (especially fission) are well-studied discontinuous phenomena that would provide experimental verification for any analysis conducted. In this work, we consider the liquid-drop model and the shell model.

We note here that the expressions for deformation energy in most nuclear deformation models, including ours, are in the form of an infinite power series. Owing to the analytical nature of Catastrophe Theory, such a power series must be suitably truncated to obtain a polynomial. Thus, in this work we construct a minimalistic model and incorporate only the most important deformation parameter, and truncate our expressions so as to obtain reasonable agreement with known fission barriers, while still retaining the simplicity needed for an analytical treatment.

In Section II, we outline the generalization of the liquid-drop model to deformed shapes as presented in [4]. In Section III, we then discuss the stability analysis of the model and find a third-order phase transition. In Section IV, we outline the shell model as presented in [4], and how it is incorporated into the liquid-drop model as a correction term. In § 0.5, we map the final deformation energy to the fold catastrophe and find a complete loss of stability as one moves in the N−ZN - Z space, and see how this explains observed fissile nuclei and decay chains of heavy nuclei in Section VI.

The deformed liquid drop model

We begin with the spherical liquid-drop model as presented in [4]:

B=avA[1−kvI2]−asA2/3[1−ksI2]−aCZ2A1/3(1)B = a_v A \left[1-k_v I^2\right] - a_s A^{2/3} \left[1-k_s I^2\right] - a_C \frac{Z^2}{A^{1/3}} \tag*{(1)}

where BB is the binding energy of the nucleus, ava_v, asa_s, aCa_C, kvk_v and ksk_s are constants, ZZ is the atomic number, NN is the neutron number and

I=N−ZN+Z(2)I = \frac{N-Z}{N+Z} \tag*{(2)}

The values of the constants are obtained through best fits and have been stated in [4] and [5] as

aC=0.7053,as=17.944MeV,ks=1.782(3)a_C = 0.7053,\quad a_s = 17.944MeV,\quad k_s = 1.782 \tag*{(3)}

The parameter kvk_v is in the range 1.5 - 2.0 [4], however, the parameter is not relevant in the analysis reported here.

The model has a volume term (proportional to the volume), a surface term (proportional to the surface area), and a coulomb term respectively. This model can be generalized to a deformed nucleus by simply accounting for the change in the surface area and coulomb interactions and rewriting BB as

B=asA2/3(1−ksI2)Es(d)Es0+aCZ2A1/3EC(d)EC0(4)B = a_s A^{2/3} (1-k_s I^2)\frac{E_s(d)}{E_s^0} + a_C\frac{Z^2}{A^{1/3}}\frac{E_C(d)}{E_C^0} \tag*{(4)}

where we assume that any deformations conserve the volume of the nucleus, making the volume term irrelevant. EC(d)E_C(d), Es(d)E_s(d) are the surface and coulomb energies of the deformed nucleus, and Es0E_s^0, EC0E_C^0 are the surface and coulomb energies of the spherical nucleus

Es0=asA2/3(1−ksI2)(5)E_s^0 = a_s A^{2/3}(1-k_s I^2) \tag*{(5)}
EC0=aCZ2A1/3(6)E_C^0 = a_C\frac{Z^2}{A^{1/3}} \tag*{(6)}

The shape of a deformed nucleus can be parameterised using spherical harmonics as

R(θ,ϕ)=Rα[1+∑λ=1∞∑μ=−λλαλμYλμ(θ,ϕ)](7)R(\theta,\phi) = R_\alpha\left[1+\sum_{\lambda=1}^{\infty}\sum_{\mu=-\lambda}^{\lambda}\alpha_{\lambda\mu}Y_{\lambda\mu}(\theta,\phi)\right] \tag*{(7)}

To simplify this, we assume that the deformations of the nucleus are symmetric about an axis (taken to be the zz axis). Thus, we can work with Legendre Polynomials instead of spherical harmonics; we only need to consider μ=0\mu= 0 in the summation. We define βλ=αλμ\beta_\lambda= \alpha_{\lambda\mu} and use

Yλ0(θ,ϕ)=(2λ+14π)1/2Pλ(cos⁡(θ))(8)Y_{\lambda0}(\theta,\phi) = \left(\frac{2\lambda+ 1}{4\pi}\right)^{1/2} P_\lambda(\cos(\theta)) \tag*{(8)}

We thus obtain

R(θ)=Rβ[1+(2λ+14π)1/2∑λ=1∞βλPλ(θ)](9)R(\theta) = R_\beta\left[1 + \left(\frac{2\lambda+ 1}{4\pi}\right)^{1/2} \sum_{\lambda=1}^{\infty} \beta_\lambda P_\lambda(\theta)\right] \tag*{(9)}

It turns out only β2\beta_2 (up to its third power) is relevant for small deformations [4]. As shown in [4], the Coulomb and the surface energies for small deformations can be calculated as

EC=EC0(1−15a22−4105a23)(10)E_C = E_C^0 \left(1 - \frac{1}{5}a_2^2 - \frac{4}{105}a_2^3\right) \tag*{(10)}
Es=Es0(1+25a22−4105a23)(11)E_s = E_s^0 \left(1 + \frac{2}{5}a_2^2 - \frac{4}{105}a_2^3\right) \tag*{(11)}

where

a2=54πβ2(12)a_2 = \sqrt{\frac{5}{4\pi}}\beta_2 \tag*{(12)}

Thus, the additional energy due to deformations is

ΔE=(Es+EC)−(Es0+EC0)\Delta E = (E_s + E_C) - (E_s^0 + E_C^0)

giving us

ΔE=Vd=Es0(25(1−x)a22−4105(1+2x)a23)(13)\Delta E = V_d = E_s^0\left(\frac{2}{5}(1-x)a_2^2 - \frac{4}{105}(1+2x)a_2^3\right) \tag*{(13)}

where

x=EC02Es0=aCasZ2A11−ksI2(14)x = \frac{E_C^0}{2E_s^0} = \frac{a_C}{a_s}\frac{Z^2}{A}\frac{1}{1-k_s I^2} \tag*{(14)}

xx is the so-called fissility parameter; as [4] shows, it is a measure of how easy it is for a nucleus to undergo fission. We plot xx as a function of NN for Z=92Z = 92 in Figure 1.

Figure 1. xx as a function of NN for Z=92Z = 92. The point at which x=1x = 1 is marked.

We now examine the deformation energy VdV_d (Eq. (13)) for instabilities and phase transitions.

The third-order phase transition

We note that both xx and Es0E_s^0 are functions of ZZ and NN. Let us consider a fixed ZZ, for which we slowly vary NN from smaller to larger values. Since ZZ is constant, we can say that xx and EsE_s are functions of NN. We consider NN to be a parameter, and treat a2a_2 as a coordinate. We assume that the ground state of the system dwells at the lowest energy minimum. We study how the system’s energy (the energy the lowest energy minimum) varies with NN. We have

Vd(N)=Es0(N)(a(N)a23+b(N)a22)(15)V_d(N) = E_s^0(N)\left(a(N)a_2^3 + b(N)a_2^2\right) \tag*{(15)}

where

a(N)=−4105(1+2x(N))b(N)=25(1−x(N))(16)\begin{aligned} a(N) &= -\frac{4}{105}\left(1+2x(N)\right) \\ b(N) &= \frac{2}{5}\left(1-x(N)\right) \tag*{(16)} \end{aligned}

It can be easily shown from Eq. (5) that Es(N)E_s(N) is a smooth and positive function of NN in the region of interest. Further, a(N)<0a(N)<0 for all N>0N>0.

The fixed points of the system a20a_2^0 are given by

∂Vd∂a2=0⇒a20=0, −2b3a(17)\frac{\partial V_d}{\partial a_2}=0 \Rightarrow a_2^0=0,\,-\frac{2b}{3a} \tag*{(17)}

Further

∂2V∂a22(0)=2b\frac{\partial^2 V}{\partial a_2^2}(0)=2b
∂2V∂a22(−2b3a)=−2b(18)\frac{\partial^2 V}{\partial a_2^2}\left(-\frac{2b}{3a}\right)=-2b \tag*{(18)}

For b>0b>0 the equilibrium at a2=0a_2=0 is stable and the equilibrium at a2=−2b3aa_2=-\frac{2b}{3a} is unstable.

For b<0b<0, the equilibrium at a2=0a_2=0 in unstable and the equilibrium at a2=−2b3aa_2=-\frac{2b}{3a} is stable.

Thus, at b=0b=0, the fixed points meet, and the stability of the two fixed points swap. For b>0b>0 the system dwells at a20=0a_2^0=0. The energy of the system is

Vd(N,a20=0)=0(19)V_d(N,a_2^0=0)=0 \tag*{(19)}

This is completely independent of NN. Thus

b>0⇒dnVd(a20=0)dNn=0∀n∈N(20)b>0\Rightarrow\frac{d^nV_d(a_2^0=0)}{dN^n}=0\quad\forall n\in\mathbb{N} \tag*{(20)}

Now, for b<0b<0, the system dwells at a20=−2b3aa_2^0=-\frac{2b}{3a}. The energy of the system is given by

Vd(N,a20=−2b3a)=4b327a2Es0(21)V_d\left(N,a_2^0=-\frac{2b}{3a}\right)=\frac{4b^3}{27a^2}E_s^0 \tag*{(21)}

We now evaluate the derivatives of this energy with respect to NN, at b=0b=0. We note that aa, bb, Es0E_s^0 and their derivatives with respect to NN are smooth in our region of interest (where most observed nuclei are found in the N−ZN-Z space). Also, as noted earlier, a(N)<0∀ Na(N)<0\quad\forall\ N.

On computing the derivatives, one obtains multiple terms containing derivatives of aa, bb, Es0E_s^0. However, since we are evaluating these derivatives at b=0b=0, most of these terms vanish. Using this, one finds easily that, at b=0b=0

dVd(a20)dN=d2Vd(a20)dN2=0(22)\frac{dV_d(a_2^0)}{dN}=\frac{d^2V_d(a_2^0)}{dN^2}=0 \tag*{(22)}

However

d3Vd(a20)dN3=(89Es027a2(dbdN)3)∣b=0≠0(23)\frac{d^3V_d(a_2^0)}{dN^3}=\left(\frac{8}{9}\frac{E_s^0}{27a^2}\left(\frac{db}{dN}\right)^3\right)\bigg|_{b=0}\ne0 \tag*{(23)}

One can verify that xx is a decreasing function of NN (for a fixed ZZ) in the region of interest (see Figure 1). From Eq. (16), we see that we cross b=0b=0 at x=1x=1. As NN is increased from lower to larger values, xx keeps decreasing, and at some N=N0N = N_0, xx falls below 1. Before this point, b<0b < 0, and the system’s lowest minimum is always given by a20=−2b3aa_2^0 = -\frac{2b}{3a} (note this is a function of NN). The first three derivatives of this energy are given by Eq. (22) and Eq. (23).

At x=1x = 1, this minimum stops changing with NN and remains fixed at a20=0a_2^0 = 0. Now, all the derivatives of the system’s energy turn to 0. From Eq. (22) and Eq. (23), we see that the first two derivatives are continuous, while the third derivative is discontinuous, implying a third-order phase transition.

One can find the N=N0N = N_0 at which this phase transition occurs using

x(N0)=1=aCasZ2A11−ksI2(24)x(N_0) = 1 = \frac{a_C}{a_s}\frac{Z^2}{A}\frac{1}{1-k_s I^2} \tag*{(24)}

As per this analysis, as one varies NN from lower to higher values (for a fixed ZZ), the nucleus simply goes from preferring a deformed shape to a spherical shape via a third-order phase transition. This in itself does not imply a catastrophic instability one would imagine is needed for fission. We thus now consider a more microscopic nuclear model: the shell model.

The shell model and shell correction

The liquid-drop model has had success in describing the ‘average’ behavior of nuclei but fails to account for the existence of magic numbers (neutron and proton numbers for which a nucleus is unusually stable). This extra stability can naturally be attributed to the filling of energy levels.

One considers each nucleon to be in an attractive potential caused by all the other nucleons. A typical choice of this potential (for an undeformed nucleus) is the modified-harmonic oscillator potential given by Nilsson ([6] and [4])

VMO(r)=12mω02r2−κℏω0(2l⋅s+μ(l2−⟨l2⟩N))(25)V_{\mathrm{MO}}(r) = \frac{1}{2}m\omega_0^2r^2-\kappa\hbar\omega_0\left(2\mathbf{l}\cdot\mathbf{s}+\mu\left(l^2-\langle l^2\rangle_{\mathcal{N}}\right)\right) \tag*{(25)}

where N=2n+l\mathcal{N} = 2n + l is the principal quantum number. To account for Coulombic interactions and the Pauli exclusion principle, neutrons and protons are considered to experience separate potentials with different natural frequencies ([4])

ℏω0N,Z=41A−13(1±13N−ZA)(26)\hbar\omega_0^{N,Z}=41A^{-\frac{1}{3}}\left(1\pm\frac{1}{3}\frac{N-Z}{A}\right) \tag*{(26)}

The energy levels for the modified harmonic oscillator potential are analytically obtained as ([4])

ENls=ℏω0(N+32−κ[2s(l+12)−12]−μ′[l(l+1)−N(N+3)2])(27)E_{Nls}=\hbar\omega_0\left(\mathcal{N}+\frac{3}{2}-\kappa\left[2s\left(l+\frac{1}{2}\right)-\frac{1}{2}\right]-\mu'\left[l(l+1)-\frac{\mathcal{N}(\mathcal{N}+3)}{2}\right]\right) \tag*{(27)}

To account for deformations, one simply considers deformations to the spherical potential ([7] and [4]):

V=12mω2(ε⃗)r2(1−23εP2(cos⁡θt)+∑λ=3,4,…ελPλ(cos⁡θt))(28)V=\frac{1}{2}m\omega^2(\vec{\varepsilon})r^2\left(1-\frac{2}{3}\varepsilon P_2(\cos\theta_t)+\sum_{\lambda=3,4,\ldots}\varepsilon_\lambda P_\lambda(\cos\theta_t)\right) \tag*{(28)}

Here, the frequency is considered as a function of the deformations to account for the volume conservation of an equipotential surface, and the factor of 2/32/3 allows for simpler calculations later. One can find the shifts in the energy levels due to the deformations using perturbation theory. Let these energy levels be eν(ε)e_\nu(\varepsilon) (where ν\nu iterates over each nucleon, not each energy level, so degenerate energy levels are counted multiple times). To obtain the total energy of the nucleus, one can’t simply add up the individual nucleon energies because of overcounting of inter-nucleon interactions. Since the liquid-drop model accounts well for the average behavior of nuclei, one incorporates the shell effects as a correction to the liquid-drop model.

This correction procedure is described in [4]: since the ‘average’ behavior of nuclei is reproduced well by the liquid-drop model, one adds the shell model’s deformation energy to the liquid-drop model and subtracts an ‘averaged’ shell model energy (so that the new corrected energy still describes the average behavior of nuclei correctly).

To compute this ‘averaged’ shell model energy, consider the shell model’s energy density

g(e)=1γ∑νδ(ξν)(29)g(e)=\frac{1}{\gamma}\sum_\nu\delta(\xi_\nu) \tag*{(29)}

where γ\gamma is an energy scale to make the argument of the Dirac-delta dimensionless, and

ξν=e−eνγ\xi_\nu=\frac{e-e_\nu}{\gamma}. The total energy is then

E=∫eg(e) de=∑νeν(30)E=\int eg(e)\,de=\sum_\nu e_\nu \tag*{(30)}

irrespective of γ\gamma. To obtain an averaged energy density, one smears the energy levels by expanding out the Dirac-delta functions in a Hermite polynomial expansion

δ(x)=∑n=0∞1πcke−x2H2k(x)(31)\delta(x)=\sum_{n=0}^{\infty}\frac{1}{\sqrt{\pi}}c_k e^{-x^2}H_{2k}(x) \tag*{(31)}

where

ck=(−1)k(2k)!!2k(32)c_k = \frac{(-1)^k}{(2k)!!2^k} \tag*{(32)}

and truncating this expansion at some order nn to obtain

g~(e)=1γπ∑ν∑k=0ncke−ξν2H2k(ξν)(33)\tilde{g}(e) = \frac{1}{\gamma\sqrt{\pi}} \sum_{\nu} \sum_{k=0}^{n} c_k e^{-\xi_\nu^2} H_{2k}(\xi_\nu) \tag*{(33)}

Then, γ\gamma decides the scale of this smearing, and is typically chosen as ℏω0\hbar\omega_0. One chooses an nn so that the resulting energies are relatively independent of γ\gamma, and [4] shows this is observed at n=3n = 3.

The difference of these energy densities is then

g(e)−g~(e)=1γπ∑ν∑k=n+1∞cke−ξν2H2k(ξν)(34)g(e) - \tilde{g}(e) = \frac{1}{\gamma\sqrt{\pi}} \sum_{\nu} \sum_{k=n+1}^{\infty} c_k e^{-\xi_\nu^2} H_{2k}(\xi_\nu) \tag*{(34)}

Then, as per the correction outlined earlier, the total energy is

E=ELD+Ecorr=ELD+∑ν∑k=n+1∞∫eckγπe−ξν2H2k(ξν) de(35)\begin{aligned} E &= E_{LD} + E_{corr} \\ &= E_{LD} + \sum_{\nu} \sum_{k=n+1}^{\infty} \int\frac{e c_k}{\gamma\sqrt{\pi}} e^{-\xi_\nu^2} H_{2k}(\xi_\nu)\,de \tag*{(35)} \end{aligned}

where ELDE_{LD} is the liquid-drop energy. To find the energy due to deformation, we simply need to include the deformation dependence of the energy levels

V(ε⃗)=Vd(a2)+∑ν∂Ecorr∂eν∣ε⃗=0Δeν(36)V(\vec{\varepsilon}) = V_d(a_2) + \sum_{\nu} \left.\frac{\partial E_{corr}}{\partial e_\nu}\right|_{\vec{\varepsilon}=0} \Delta e_\nu \tag*{(36)}

where Vd(a2)V_d(a_2) is given by Eq. (15). We shall later link the deformation parameters ε\varepsilon and a2a_2.

Stability analysis

Since the shell model is being incorporated with the liquid-drop model, we consider only ε\varepsilon in the expansion given by Eq. (28):

V=12mω2(ε⃗)r2(1−23εP2(cos⁡θt))(37)V = \frac{1}{2}m\omega^2(\vec{\varepsilon})r^2\left(1-\frac{2}{3}\varepsilon P_2(\cos\theta_t)\right) \tag*{(37)}

We can then write the perturbed (deformed) Hamiltonian as

H=H0+εH′(38)H = H_0 + \varepsilon H' \tag*{(38)}

where

H′=−23mω02r2P2(cos⁡θ)(39)H'=-\frac{2}{3}m\omega_0^2r^2P_2(\cos\theta) \tag*{(39)}

The shifts in energy levels are then computed in [4] as

ΔE=⟨ϵH′⟩=16ϵmω(ϵ)2⟨r2⟩3mj2−j(j+1)j(j+1)(40)\Delta E=\langle\epsilon H'\rangle=\frac{1}{6}\epsilon m\omega(\epsilon)^2\langle r^2\rangle\frac{3m_j^2-j(j+1)}{j(j+1)} \tag*{(40)}

where j=l+sj=l+s and a 2j+12j+1 degeneracy is broken. We assume that upon deformation, a nucleon will pick the mjm_j state with the least energy. Thus, the energy shift for deformed energy levels is given by

ΔE=⟨ϵH′⟩=16ϵmω(ϵ)2⟨r2⟩3mj02−j(j+1)j(j+1)(41)\Delta E=\langle\epsilon H'\rangle=\frac{1}{6}\epsilon m\omega(\epsilon)^2\langle r^2\rangle\frac{3m_{j0}^2-j(j+1)}{j(j+1)} \tag*{(41)}

where mj0m_{j0} is the least energetic mjm_j state available while respecting the Pauli Exclusion principle. In the spherical case, each nucleon has unique quantum numbers N,l,j\mathcal{N},l,j. Thus, once a nucleon occupies any mjm_j state, the set of quantum numbers N,l,j,mj\mathcal{N},l,j,m_j is still unique for any mjm_j. Thus, the least energetic state each nucleon can choose is

mj0=0(42)m_{j0}=0 \tag*{(42)}

Finally, we impose volume conservation of the equipotential surface to obtain [4]

ω(ϵ)=ω0(1+19ϵ2⋯ )(43)\omega(\epsilon)=\omega_0\left(1+\frac{1}{9}\epsilon^2\cdots\right) \tag*{(43)}

Plugging Eq. (43) and Eq. (41) in Eq. (36), we obtain (up to third order in ϵ\epsilon)

V(ϵ⃗)=Vd(ϵ⃗)+c(N,Z)(ϵ+2ϵ39)(44)V(\vec{\epsilon})=V_d(\vec{\epsilon})+c(N,Z)\left(\epsilon+\frac{2\epsilon^3}{9}\right) \tag*{(44)}

where c(N,Z)c(N,Z) incorporates all the proportionality coefficients and is given by

c(N,Z)=∑ν16mω02⟨r2⟩3mj02−j(j+1)j(j+1)∂Ecorr∂ϵν∣ϵ=0(45)c(N,Z)=\sum_\nu\frac{1}{6}m\omega_0^2\langle r^2\rangle\frac{3m_{j0}^2-j(j+1)}{j(j+1)}\left.\frac{\partial E_{\mathrm{corr}}}{\partial\epsilon_\nu}\right|_{\epsilon=0} \tag*{(45)}

One can easily calculate ⟨r2⟩\langle r^2\rangle to be [4]

⟨r2⟩=(N+32)ℏmω0(46)\langle r^2\rangle=\left(\mathcal{N}+\frac{3}{2}\right)\frac{\hbar}{m\omega_0} \tag*{(46)}

leading to

c(N,Z)=∑ν16ℏω0(N+32)3mj02−j(j+1)j(j+1)∂Ecorr∂ϵν∣ϵ=0(47)c(N,Z)=\sum_\nu\frac{1}{6}\hbar\omega_0\left(\mathcal{N}+\frac{3}{2}\right)\frac{3m_{j0}^2-j(j+1)}{j(j+1)}\left.\frac{\partial E_{\mathrm{corr}}}{\partial\epsilon_\nu}\right|_{\epsilon=0} \tag*{(47)}

To link the deformation ε\varepsilon with the deformation a2a_2 considered in the liquid-drop model, we simply consider an equipotential surface for the deformed potential (Eq. (37)), given by

r(θ)=r0(1+2ε3P2(cos⁡θ))(48)r(\theta) = r_0 \left(1 + \frac{2\varepsilon}{3}P_2(\cos\theta)\right) \tag*{(48)}

Comparing this to the spatial deformations considered in the liquid-drop model (Eq. (9)), we immediately see

ε=3254πβ2=32a2(49)\varepsilon= \frac{3}{2}\sqrt{\frac{5}{4\pi}}\beta_2 = \frac{3}{2}a_2 \tag*{(49)}

Substituting this and Vd(a2)V_d(a_2) in Eq. (44), we obtain

V(a2)=Aa23+Ba22+Ca2(50)V(a_2) = Aa_2^3 + Ba_2^2 + Ca_2 \tag*{(50)}

where

A=Es0(N)a(N,Z)+3c(N,Z)4B=Es0(N,Z)b(N,Z)C=32c(N,Z)(51)\begin{aligned} A &= E_s^0(N)a(N,Z) + \frac{3c(N,Z)}{4} \\ B &= E_s^0(N,Z)b(N,Z) \\ C &= \frac{3}{2}c(N,Z) \tag*{(51)} \end{aligned}

We now appeal to Catastrophe theory. An obvious choice of a catastrophe to map V(a2)V(a_2) is the fold catastrophe:

Vt(u)=u3+tu(52)V_t(u) = u^3 + tu \tag*{(52)}

where tt is a parameter and uu is a coordinate. Notably, the fold catastrophe has a pitchfork bifurcation at t=0t = 0. For t<0t < 0, the potential has a maximum and a minimum, which meet and annihilate at t=0t = 0. For t>0t > 0, no stable point exists in the system, leading to a drastic change in the system’s behavior.

One can easily check that Ax3+Bx2+CxAx^3 + Bx^2 + Cx can be mapped to u3+tu+su^3 + tu + s via the mapping

x=uA13+p(53)x = \frac{u}{A^{\frac{1}{3}}} + p \tag*{(53)}

where

p=−B3A,t=3AC−B23A43,s=2B3−9ABC27A2(54)p = -\frac{B}{3A}, \qquad t = \frac{3AC-B^2}{3A^{\frac{4}{3}}}, \qquad s = \frac{2B^3-9ABC}{27A^2} \tag*{(54)}

Here, ss is a additive term with no coordinate dependence and can be safely discarded. Under this mapping, one observes the loss of stability at t>0t > 0, which translates to

t>0⇒3AC−B2A4/3>0(55)t > 0 \Rightarrow\frac{3AC-B^2}{A^{4/3}} > 0 \tag*{(55)}

Defining this expression as f(N,Z)f(N,Z), we get

f(N,Z)=[Es0(N)a(N,Z)+34c(N,Z)]−4/3×(92[34c(N,Z)+Es0(N)a(N,Z)]c(N,Z)[Es0(N,Z)b(N,Z)]2)>0(56)f(N,Z)=\left[E_s^0(N)a(N,Z)+\frac{3}{4}c(N,Z)\right]^{-4/3}\times\left(\frac{9}{2}\left[\frac{3}{4}c(N,Z)+E_s^0(N)a(N,Z)\right]c(N,Z)\left[E_s^0(N,Z)b(N,Z)\right]^2\right)>0 \tag*{(56)}

Experimental verification

To compare with observables, first let we compare the fission barriers of heavy nuclei in our model to their experimental values [8]. This comparison is tabulated in Table I. Because of the minimalistic nature of the model, we only observe one barrier. However, we obtain reasonable agreement in spite of our minimalistic approach.

More intuitively, we plot f(N,Z)f(N,Z) in the N−ZN-Z space. The values of the shell model parameters κ\kappa and μ\mu for various N\mathcal{N} levels are taken from [4] and are shown in Table II.

N=2n+l\mathcal{N} = 2n + lμ\muκ\kappa
200.8
30.02630.075
4,5,6,70.0240.06

Table II. μ\mu and κ\kappa values for various N=2n+l+1\mathcal{N} = 2n + l + 1 levels

NucleusCalculated (MeV)Experimental (EbE_b) (MeV)Experimental (EaE_a) (MeV)
Po20815.2919.9
Po20915.5521.1
Po21015.8021.2
Th2309.296.806.10
Th2319.446.706.00
Th2329.576.705.80
U2347.285.504.80
U2357.406.005.25
U2387.755.506.30

Table 1. Calculated and experimental fission barriers[8] for heavy nuclei. We obtain reasonable agreement in spite of us considering only a single deformation parameter.

c(N,Z)c(N, Z) is symbolically evaluated in python. We plot f(N,Z)f(N, Z) as a heat map in the N−ZN-Z space, leading to Figure 2. We observe a band-like structure, with nuclei in the interior of the

Figure 2. f(N,Z)f(N, Z) is plotted in the N−ZN-Z plane. A band-like structure is seen in which f(N,Z)f(N, Z) is positive or close to 0, indicating a loss of stability. Various contours are added, and nuclei satisfying Ronen’s fissile rule are scattered in red. The contour of f(N,Z)=0f(N, Z) = 0 is marked in white; these nuclei favor spontaneous deformations from a spherical shape.

band having f(N,Z)<0f(N, Z) < 0 and being stable and those near the periphery having higher values of f(N,Z)f(N, Z). We note here that any nuclei with a positive f(N,Z)f(N, Z) are inherently unstable, and we would not expect to see them at all. Instead, we can use f(N,Z)f(N, Z) as a measure of how unstable a nucleus is, with higher values implying less stability. We also add some

contours and mark nuclei satisfying Ronen’s fissile rule [9]: nuclei satisfying

90≤Z≤100,2Z−N=43±2(57)90 \le Z \le100,\qquad2Z-N=43\pm2 \tag*{(57)}

are fissile. We see that all the nuclei lie close to the band periphery, where there is a sudden sharp increase in f(N,Z)f(N,Z) (implying a sudden increase in instability). Another test is to see how f(N,Z)f(N,Z) varies across decay chains. We show this for various decay chains in Figure 3.

Figure 3. Evolution of f(N,Z)f(N,Z) in various decay chains is shown. Note the jumps at each β\beta decay. Decay chains: 238U→α234Th→β234Pa→β234U→α230Th→α226Ra→α222Rn→α218Po→α214Pb→β214Bi→β214Po→α210Pb→β210Bi→β210Po→α206Pb{}^{238}\mathrm{U}\xrightarrow{\alpha}{}^{234}\mathrm{Th}\xrightarrow{\beta}{}^{234}\mathrm{Pa}\xrightarrow{\beta}{}^{234}\mathrm{U}\xrightarrow{\alpha}{}^{230}\mathrm{Th}\xrightarrow{\alpha}{}^{226}\mathrm{Ra}\xrightarrow{\alpha}{}^{222}\mathrm{Rn}\xrightarrow{\alpha}{}^{218}\mathrm{Po}\xrightarrow{\alpha}{}^{214}\mathrm{Pb}\xrightarrow{\beta}{}^{214}\mathrm{Bi}\xrightarrow{\beta}{}^{214}\mathrm{Po}\xrightarrow{\alpha}{}^{210}\mathrm{Pb}\xrightarrow{\beta}{}^{210}\mathrm{Bi}\xrightarrow{\beta}{}^{210}\mathrm{Po}\xrightarrow{\alpha}{}^{206}\mathrm{Pb}; 237Pu→α233U→α229Th→α225Ra→β225Ac→α221Fr→β221Ra→α217Rn→α213Po,209Pb→β209Bi→α205Tl{}^{237}\mathrm{Pu}\xrightarrow{\alpha}{}^{233}\mathrm{U}\xrightarrow{\alpha}{}^{229}\mathrm{Th}\xrightarrow{\alpha}{}^{225}\mathrm{Ra}\xrightarrow{\beta}{}^{225}\mathrm{Ac}\xrightarrow{\alpha}{}^{221}\mathrm{Fr}\xrightarrow{\beta}{}^{221}\mathrm{Ra}\xrightarrow{\alpha}{}^{217}\mathrm{Rn}\xrightarrow{\alpha}{}^{213}\mathrm{Po},{}^{209}\mathrm{Pb}\xrightarrow{\beta}{}^{209}\mathrm{Bi}\xrightarrow{\alpha}{}^{205}\mathrm{Tl}; 260Lr→α256Md→α252Es→α248Bk→α244Am→β244Cm→α240Pu→α236U→α232Th→α228Ra→β228Ac→β228Th→α224Ra→α220Rn→α216Po→2β216Rn→α212Po→α208Pb{}^{260}\mathrm{Lr}\xrightarrow{\alpha}{}^{256}\mathrm{Md}\xrightarrow{\alpha}{}^{252}\mathrm{Es}\xrightarrow{\alpha}{}^{248}\mathrm{Bk}\xrightarrow{\alpha}{}^{244}\mathrm{Am}\xrightarrow{\beta}{}^{244}\mathrm{Cm}\xrightarrow{\alpha}{}^{240}\mathrm{Pu}\xrightarrow{\alpha}{}^{236}\mathrm{U}\xrightarrow{\alpha}{}^{232}\mathrm{Th}\xrightarrow{\alpha}{}^{228}\mathrm{Ra}\xrightarrow{\beta}{}^{228}\mathrm{Ac}\xrightarrow{\beta}{}^{228}\mathrm{Th}\xrightarrow{\alpha}{}^{224}\mathrm{Ra}\xrightarrow{\alpha}{}^{220}\mathrm{Rn}\xrightarrow{\alpha}{}^{216}\mathrm{Po}\xrightarrow{2\beta}{}^{216}\mathrm{Rn}\xrightarrow{\alpha}{}^{212}\mathrm{Po}\xrightarrow{\alpha}{}^{208}\mathrm{Pb}.

We observe a drop in f(N,Z)f(N,Z) (implying an increase in stability) as each decay progresses, consistent with our analysis. Notably, all α\alpha decays are seen to decrease f(N,Z)f(N,Z). However, we observe that each β\beta decay increases f(N,Z)f(N,Z), with a drop again in consequent α\alpha decays. Understanding this feature requires further investigation, but this naturally paints a picture of a β\beta decay being an ‘intermediate’ step that allows the decay chain to end at a more stable nucleus (as compared to the one it would have ended at without a β\beta decay), at the expense of some temporary instability in the form of an increase in f(N,Z)f(N,Z).

Conclusion

We have presented a geometric analysis of the stability of nuclei against deformations using Catastrophe theory, by constructing a minimalistic microscopic-macroscopic deformation model. The model accounts for a single deformation parameter to preserve the analytical nature of this work, while also retaining enough complexity to account for known fission barriers. We find a third-order phase transition in the liquid-drop model as one moves in the N−ZN-Z space, in which the nucleus simply goes from preferring a spherical shape to preferring a deformed shape with no characteristic loss of stability. Upon incorporating shell effects, this translates from a phase transition to a complete loss of stability characteristic of the Fold catastrophe.

Experimentally, our analysis is seen to explain the instability of fissile nuclei and also explain various decay chains of heavy nuclei, especially α\alpha decays. We also find good agreement with experimentally observed fission barriers.

Acknowledgements

We would like to thank Vikram Rentala and Kumar Rao for their valuable insights.

Funding: No funding was received for conducting this study.

Conflict of interest: The authors of this work declare that they have no conflicts of interest.

References

References

  1. [1]R. Thom. Stabilite Structurelle et Morphogenese. Benjamin, New York, 1972.
  2. [2]T. Poston and I. Stewart. Catastrophe Theory and its Applications. Surveys and Reference Works in Mathematics, 1978.
  3. [3]V. I. Arnold. Catastrophe Theory. Springer Berlin, Heidelberg, 2012.
  4. [4]Ingemar Ragnarsson and Sven Gvsta Nilsson. Shapes and Shells in Nuclear Structure. Cambridge University Press, 1995.
  5. [5]William D. Myers and Wladyslaw J. Swiatecki. Nuclear masses and deformations. Nuclear Physics, 81(1):1–60, 1966.DOI
  6. [6]S. G. Nilsson. Binding states of individual nucleons in strongly deformed nuclei. Kong. Dan. Vid. Sel. Mat. Fys. Med., 29N16:1–69, 1955.
  7. [8]T. Belgya et.al. Handbook for calculations of nuclear reaction data., RIPL-2. IAEA-TECDOC-1506 (IAEA, Vienna, 2006). Available online at http://www-nds.iaea.org/RIPL-2/, 2003.
  8. [9]Yigal Ronen. A rule for determining fissile isotopes. Nuclear Science and Engineering, 152(3):334-335, 2006.DOI

Paper details

vlnc:W_kvj5uuww
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