betaFeedback

WorksPhysics and Astronomy

Synthesis of elements in compact stars in pycnonuclear reactions with Carbon isotopes: Quasibound states versus states of zero-points vibrations

Abstract

(1) Purpose: Conditions of formation of compound nuclear system needed for synthesis of heavy nuclei in pycnonuclear reactions in compact stars are studied on a quantum mechanical basis. (2) Methods: Method of multiple internal reflections is generalized for pycnoreactions in compact stars with new calculations of quasibound spectra and spectra of zero-point vibrations. (3) Results: Peculiarities of the method are analyzed for reaction with isotopes of Carbon. The developed method takes into account continuity and conservation of quantum flux (describing pycnonuclear reaction) inside the full spacial region of reaction including nuclear region. This gives appearance of new states (called as quasibound states), in which compound nuclear systems of Magnesium are formed with the largest probability. These states have not been studied yet in synthesis of elements in stars. Energy spectra of zero-point vibrations and spectra of quasibound states are estimated with high precision for reactions with isotopes of Carbon. At the first time influence of plasma screening on quasibound states and states of zero-point vibrations in pycnonuclear reactions has been studied. (4) Conclusion: The probability of formation of compound nuclear system in quasibound states in pycnonuclear reaction is essentially larger than the probability of formation of this system in states of zero-point vibrations studied by Zel'dovich and followers. So, synthesis of Magnesium from isotopes of Carbon is more probable through the quasibound states than through the states of zero-point vibrations in compact stars. Energy spectra of zero-point vibrations are changed essentially after taking plasma screening into account. Analysis shows that from all studied isotopes of Magnesium only \isotope[24]{Mg} is stable after synthesis at energy of relative motion of 4.881~MeV of incident nuclei \isotope[12]{C}.

I Introduction

Pycnonuclear burning occurs in dense and cold cores of white dwarfs [1] and crusts of accreting neutron stars [2, 3]. Such a phenomenon known as pycnonuclear reaction [4] is reaction at sufficiently high densities in stars where zero-point vibrations of nuclei in the lattice sites leads to an essential increasing rate of formation of more heavy nuclei. Insight to this phenomenon was given by Zel’dovich who estimated zero-point energy as in the ground state of discrete energy spectrum of the harmonic oscillator potential which is formed near middle point between two nuclei in lattice sites [5]. Rates of reactions derived at such zero-point energies are estimated for some atomic nuclei in lattices in compact stars [6].

Key process in pycnonuclear reactions is fusion of nuclei producing new nucleus with larger mass. In Ref. [7] this question was analyzed for reactions with nuclei of different charges and masses, where authors derived the astrophysical SS-factors for carbon-oxygen and oxygen-oxygen fusion reactions on the microscopic basis. Phenomenological expressions for reaction rates containing several fit parameters were found. In Ref. [8] the astrophysical SS-factors were derived for 946 fusion reactions involving stable and neutron-rich isotopes of C, O, Ne, and Mg for center-of-mass energies from 2 to ≈\approx 18–30 MeV. Results in that paper can be converted to thermonuclear or pycnonuclear reaction rates to simulate stellar burning at high temperatures and nucleosynthesis in high density environments. Large collection of astrophysical SS factors and their compact representation for isotopes of Be, B, C, N, O, F, Ne, Na, Mg, and Si were presented in Ref. [9]. Database of SS factors was created for about 5000 nonresonant fusion reactions in those researches. Structure of the multi-component matter (a regular lattice, a uniform mix, etc.) in these reactions, plasma screening [10], rates of reactions in a wide range of temperatures and stellar densities [7, 11] have been studied by many researchers.

It has been known that cross sections of reactions are changed essentially after taking into account conservation of quantum fluxes in the internal region of nuclear system [12, 13, 14]. This question has been studied in details for α\alpha decays of nuclei and captures of α\alpha-particles by nuclei. For example, the study of captures of α\alpha particles by nuclei [13, 14] shows the importance to take into account internal nuclear processes during capture that depends on the shape of the nuclear potential. After such an inclusion, different scenarios of capture (before fusion) appear, those processes give resulting cross sections with difference up to 4 times at the same beam energies of α\alpha particles used in experiments. Such changes are controlled by additional independent parameters appeared from fully quantum study. At the same time, such processes are completely missed in the WKB-approximation where such quantum parameters do not exist. It is important to note that this dependence of cross sections in fully quantum study is not small. For example, it can be larger essentially than inclusion of nuclear deformations to the calculation of cross sections without such quantum parameters. Up to now, method in Ref. [13] is the most accurate approach in description of experimental data for the α\alpha-capture (this calculation is in Fig. 3 (b) in Ref. [14] for α+44Ca\alpha+ {}^{44}\mathrm{Ca} in comparison with experimental data [15]).

In the fully quantum study the accuracy in determination of penetrability of barrier and cross section is about of 10−1410^{-14}, while such an accuracy in the WKB-approximation is about 10−110^{-1}–10−310^{-3}[13, 14]. Energies for the pycnonuclear reactions are low and there are only processes of deep tunneling under the barrier where the semiclassical approximation is not applicable. This reinforces our interest to this problem. These quantum effects have not been studied yet by other researchers for pycnonuclear reactions in stars. In Ref. [16] investigation of these questions on the fully quantum basis was initiated on example of reaction of 12C+12C{}^{12}\mathrm{C}+{}^{12}\mathrm{C}. Interest to that reaction is explained by its impact on nucleosynthesis, energy production and other questions in stellar evolution [11, 17]. Also this reaction has a significant impact on the evolution and structure of massive stars with M≥M⊙M \ge M_{\odot} (M⊙M_{\odot} is the Solar mass). 12C+12C{}^{12}\mathrm{C}+{}^{12}\mathrm{C} fusion is known as pycnonulear reaction that reignites carbon-oxygen white dwarf into type Ia supernova explosion. However, it could be useful to obtain more complete picture for the systematic analysis for nuclear processes and fusion for reactions with isotopes of Carbon. So, in this paper we perform such an investigation for pycnonuclear reactions with Carbon.

The paper is organized in the following way. In § 2 a new generalized formalism of the multiple internal reflections is reviewed with focus on new elements for fusion and quasibound states in pycnonuclear reactions. In § 3 reactions with isotopes of Carbon on the basis of the method are studied, with calculations of penetrabilities of the potential barriers, probabilities of formation of the compound nucleus, estimation of energies for zero-point vibrations and quasibound states, etc.. In § 4 influence of plasma screening on properties of the pycnonuclear reaction is studied on example of 12C+12C{}^{12}\mathrm{C}+{}^{12}\mathrm{C}. Conclusions are summarized in § 5.

Method of multiple internal reflections in description of nucleus-nucleus scattering with possibility of fusion

Let us consider a process of capture of the particle by the nucleus with the radial barrier of arbitrary shape, which has successfully been approximated by a sufficiently large number NN of rectangular steps:

V(r)={V1at rmin⁡<r≤r1(region 1),⋯⋯⋯VNcapat rNcap−1≤r≤rcap(region Ncap),⋯⋯⋯VNat rN−1≤r≤rmax⁡(region N).(1)V(r) = \begin{cases} V_1 & \text{at } r_{\min} < r \le r_1 \quad\text{(region 1)},\\ \cdots& \cdots\quad\cdots\\ V_{N_{\mathrm{cap}}} & \text{at } r_{N_{\mathrm{cap}-1}} \le r \le r_{\mathrm{cap}} \quad\text{(region }N_{\mathrm{cap}}\text{)},\\ \cdots& \cdots\quad\cdots\\ V_N & \text{at } r_{N-1} \le r \le r_{\max} \quad\text{(region }N\text{)}. \end{cases} \tag*{(1)}

where VjV_j are constants (j=1…Nj = 1\ldots N). We denote the first region with a left boundary at point rmin⁡r_{\min}. In addition to study in Ref. [13], in this paper we shall assume that the capture of the particle by the nucleus takes place most probably in region with number NcaptureN_{\mathrm{capture}} after its tunneling through the barrier, and further propagation of waves inside the internal region exists and should be studied. This logic follows from the condition of continuity of fluxes in quantum mechanics, that is strict condition, and it should be included to formalism. Solution of the radial wave function (up to its normalization) for the above barrier energies has the following form:

χ(r)={α1eik1r+β1e−ik1r,at rmin⁡<r≤r1(region 1),α2eik2r+β2e−ik2r,at r1≤r≤r2(region 2),⋯⋯⋯αN−1eikN−1r+βN−1e−ikN−1r,at rN−2<r<rN−1(region N−1),e−ikNr+AReikNr,at rN−1≤r≤rmax⁡(region N).(2)\chi(r) = \begin{cases} \alpha_1 e^{ik_1r}+\beta_1 e^{-ik_1r}, & \text{at } r_{\min}<r\le r_1 \quad\text{(region 1)},\\ \alpha_2 e^{ik_2r}+\beta_2 e^{-ik_2r}, & \text{at } r_1\le r\le r_2 \quad\text{(region 2)},\\ \cdots& \cdots\quad\cdots\\ \alpha_{N-1}e^{ik_{N-1}r}+\beta_{N-1}e^{-ik_{N-1}r}, & \text{at } r_{N-2}<r<r_{N-1} \quad\text{(region }N-1\text{)},\\ e^{-ik_Nr}+A_R e^{ik_Nr}, & \text{at } r_{N-1}\le r\le r_{\max} \quad\text{(region }N\text{)}. \end{cases} \tag*{(2)}

where αj\alpha_j and βj\beta_j are unknown amplitudes, ARA_R is unknown amplitude of the full reflection, and kj=1ℏ2m(E~−Vj)k_j=\frac{1}{\hbar}\sqrt{2m(\widetilde{E}-V_j)} are complex wave numbers. We fix the normalization so that the modulus of amplitude of the starting wave e−ikNre^{-ik_Nr} equals to unity. We shall find a solution of this problem by the multiple internal reflections approach. This method was described in details in Ref. [13] (see also Refs. [12, 14, 18, 19]). In this paper we will indicate only the main aspects of this approach in order to explain logic of obtaining solution.

Each wave in an arbitrary step can be represented as contribution of the exponential factor e±ikjre^{\pm i k_j r} with unknown amplitude. The amplitude of the wave transmitted through the boundary with number jj is written down as the product of the amplitude of the corresponding wave incident on this boundary and the new factor Tj±T_j^{\pm} (i. e., the amplitude of transition through the boundary with number jj). The bottom index indicates the number of the boundary, upper sign “+” or “−” is the direction of the incident wave to the right or left, respectively. The amplitude of the reflected wave from the boundary with number jj is associated with the amplitude of the wave incident on this boundary via new factors Rj±R_j^{\pm}. The coefficients T1±…TN−1±T_1^{\pm}\ldots T_{N-1}^{\pm} and R1±…RN−1±R_1^{\pm}\ldots R_{N-1}^{\pm} are found from the recurrence relations as (see Ref. [20])

Tj+=2kjkj+kj+1ei(kj−kj+1)rj,Tj−=2kj+1kj+kj+1ei(kj−kj+1)rj,Rj+=kj−kj+1kj+kj+1e2ikjrj,Rj−=kj+1−kjkj+kj+1e−2ikj+1rj.(3)\begin{aligned} T_j^+ &= \frac{2k_j}{k_j+k_{j+1}}e^{i(k_j-k_{j+1})r_j}, \qquad T_j^- = \frac{2k_{j+1}}{k_j+k_{j+1}}e^{i(k_j-k_{j+1})r_j},\\ R_j^+ &= \frac{k_j-k_{j+1}}{k_j+k_{j+1}}e^{2ik_jr_j}, \qquad R_j^- = \frac{k_{j+1}-k_j}{k_j+k_{j+1}}e^{-2ik_{j+1}r_j}. \tag*{(3)} \end{aligned}

Crucial point in the formalism of multiple internal reflections for a barrier with really large number of steps is determination of summed waves for an arbitrary region. Wave propagating to the left in the region with number j−1j-1 is formed after transmission through the boundary at point rj−1r_{j-1} of all possible incident waves, produced after all reflections and transmissions of waves in the right part of potential from this boundary. Amplitude of this wave is determined as the sum of the amplitudes of all waves incident on the boundary at point rj−1r_{j-1} multiplied on factor Tj−1−T_{j-1}^- as

T~j−1−=T~j−Tj−1−(1+∑m=1+∞(Rj−1−R~j+)m)=T~j−Tj−1−1−Rj−1−R~j+.(4)\widetilde{T}_{j-1}^-=\widetilde{T}_j^-T_{j-1}^-\left(1+\sum_{m=1}^{+\infty}(R_{j-1}^- \widetilde{R}_j^+)^m\right)=\frac{\widetilde{T}_j^-T_{j-1}^-}{1-R_{j-1}^-\widetilde{R}_j^+}. \tag*{(4)}

Summed reflection amplitude R~j−1+\widetilde{R}_{j-1}^+ includes waves transferred through boundary at rjr_j, propagated to the right, then after reflections and transmissions returned back to region with number jj:

R~j−1+=Rj−1++Tj−1+R~j+Tj−1−1−R~j+Rj−1−.(5)\widetilde{R}_{j-1}^+=R_{j-1}^++\frac{T_{j-1}^+\widetilde{R}_j^+T_{j-1}^-}{1-\widetilde{R}_j^+R_{j-1}^-}. \tag*{(5)}

Recurrent relations (4) and (5) calculate consequently all amplitudes R~N−2+…R~Ncap+\widetilde{R}_{N-2}^+\ldots\widetilde{R}_{N_{\mathrm{cap}}}^+, and T~N−2−…T~Ncap−\widetilde{T}_{N-2}^-\ldots\widetilde{T}_{N_{\mathrm{cap}}}^- if at start we use

R~N−1+=RN−1+,T~N−1−=TN−1−.(6)\widetilde{R}_{N-1}^+=R_{N-1}^+,\qquad\widetilde{T}_{N-1}^-=T_{N-1}^-. \tag*{(6)}

Amplitude of summed wave of reflection combining all waves reflected from the boundary at point rj+1r_{j+1} and propagating to the left, is

R~j+1−=Rj+1−+Tj+1−R~j−Tj+1+1−Rj+1+R~j−.(7)\widetilde{R}_{j+1}^-=R_{j+1}^-+\frac{T_{j+1}^-\widetilde{R}_j^-T_{j+1}^+}{1-R_{j+1}^+\widetilde{R}_j^-}. \tag*{(7)}

Summed amplitude AT,barA_{T,\mathrm{bar}} of transition through the barrier (summed amplitude AR,barA_{R,\mathrm{bar}} of reflection from the barrier) is determined via all waves transmitted through (reflected from) the potential region with the barrier from rcapr_{\mathrm{cap}} to rN−1r_{N-1} as

AT,bar=T~Ncap−,AR,bar=R~N−1−,where R~Ncap−=RNcap−.(8)A_{T,\mathrm{bar}}=\widetilde{T}_{N_{\mathrm{cap}}}^-,\quad A_{R,\mathrm{bar}}=\widetilde{R}_{N-1}^-,\quad\text{where }\widetilde{R}_{N_{\mathrm{cap}}}^-=R_{N_{\mathrm{cap}}}^-. \tag*{(8)}

Amplitudes, coefficients of penetrability and reflection concerning to barrier, test

We can also find the summed amplitude AR,extA_{R,\mathrm{ext}} of all waves reflected from the external barrier region (from the external turning point rtp,extr_{\mathrm{tp,ext}} to rN−1r_{N-1}) and propagated outside (which can characterize a potential scattering) as

AR,ext=R~N−1−,where R~Ntp,ext−=RNtp,ext−.(9)A_{R,\mathrm{ext}}=\widetilde{R}_{N-1}^-,\quad\text{where }\widetilde{R}_{N_{\mathrm{tp,ext}}}^-=R_{N_{\mathrm{tp,ext}}}^-. \tag*{(9)}

and the summed amplitude AR,tunA_{R,\mathrm{tun}} of all waves which are reflected just inside the potential region from rcapr_{\mathrm{cap}} to the external turning point rtp,extr_{\mathrm{tp,ext}} [i.e. they are propagated through the external barrier region (without any reflection), tunnel under the barrier, may propagate up to the boundary at point rcapr_{\mathrm{cap}} and then are reflected back from this boundary] as

AR,tun=AR,bar−AR,ps.(10)A_{R,\mathrm{tun}}=A_{R,\mathrm{bar}}-A_{R,\mathrm{ps}}. \tag*{(10)}

We estimate the amplitude of oscillations AoscA_{\mathrm{osc}} near capture with number NcapN_{\mathrm{cap}} as

Aosc(Ncap)=11−R~Ncap−1−R~Ncap+.(11)A_{\mathrm{osc}}(N_{\mathrm{cap}})=\frac{1}{1-\widetilde{R}_{N_{\mathrm{cap}}-1}^{-}\widetilde{R}_{N_{\mathrm{cap}}}^{+}}. \tag*{(11)}

We define coefficients of penetrability TbarT_{\mathrm{bar}} and reflection RbarR_{\mathrm{bar}} concerning to the whole barrier (i.e. the potential region from rcapr_{\mathrm{cap}} to rN−1r_{N-1}), and add also definitions for coefficient RextR_{\mathrm{ext}} of reflection from the external part of the barrier (i.e. region from rtp,extr_{\mathrm{tp,ext}} to rN−1r_{N-1}), coefficient RtunR_{\mathrm{tun}} of reflection from the barrier region (i.e. region from rcapr_{\mathrm{cap}} to rtp,extr_{\mathrm{tp,ext}}) as

Tbar=kcapkN∥AT,bar∥2,Rbar=∥AR,bar∥2,Rext=∥AR,ext∥2,Rtun=∥AR,tun∥2,Rtun=∥AR,tun∥2.(12)\begin{aligned} T_{\mathrm{bar}}&=\frac{k_{\mathrm{cap}}}{k_N}\lVert A_{T,\mathrm{bar}}\rVert^2,\qquad R_{\mathrm{bar}}=\lVert A_{R,\mathrm{bar}}\rVert^2,\\ R_{\mathrm{ext}}&=\lVert A_{R,\mathrm{ext}}\rVert^2,\qquad R_{\mathrm{tun}}=\lVert A_{R,\mathrm{tun}}\rVert^2,\qquad R_{\mathrm{tun}}=\lVert A_{R,\mathrm{tun}}\rVert^2. \tag*{(12)} \end{aligned}

We check the property

Tbar+Rbar=1,(13)T_{\mathrm{bar}}+R_{\mathrm{bar}}=1, \tag*{(13)}

which is used as test to indicate whether the MIR method gives the proper solution for the wave function. Let us calculate sums of the amplitudes αj(i)\alpha_j^{(i)} and βj(i)\beta_j^{(i)}. For the sums of amplitudes for the region with number jj, we get

βj≡∑i=1jβj(i)=T~j−(1+∑i=1j−1(R~j−1−R~j+)i)=T~j−1−R~j−1−R~j+,αj≡∑i=1jαj(i)=R~j−1−∑i=1jβj(i)=R~j−1−T~j−1−R~j−1−R~j+.(14)\begin{aligned} \beta_j\equiv\sum_{i=1}^{j}\beta_j^{(i)}&=\widetilde{T}_j^{-}\left(1+\sum_{i=1}^{j-1}\left(\widetilde{R}_{j-1}^{-}\widetilde{R}_{j}^{+}\right)^i\right)=\frac{\widetilde{T}_j^{-}}{1-\widetilde{R}_{j-1}^{-}\widetilde{R}_{j}^{+}},\\ \alpha_j\equiv\sum_{i=1}^{j}\alpha_j^{(i)}&=\widetilde{R}_{j-1}^{-}\sum_{i=1}^{j}\beta_j^{(i)}=\frac{\widetilde{R}_{j-1}^{-}\widetilde{T}_j^{-}}{1-\widetilde{R}_{j-1}^{-}\widetilde{R}_{j}^{+}}. \tag*{(14)} \end{aligned}

Probability of existence of the compound nucleus

To determine the probability of the existence of a compound nucleus, we calculate the integral of the square of the modulus of the wave function over the inner region to the barrier. We define this region as the region between two internal turning points rint,1r_{\mathrm{int},1} and rint,2r_{\mathrm{int},2}. In the region of above-barrier energies, we have:

∫rint,1rint,2∥χ(r)∥2 dr=∑j=1nint∫rj−1rj∥∑i=1nintαj(i)eikjr+∑i=1nintβj(i)e−ikjr∥2dr=∑j=1nint{(∥αj∥2+∥βj∥2)Δr+∣αjβj∣kjsin⁡(θαj−θβj+2kjr)∣rj−1rj}.(15)\begin{aligned} \int_{r_{\mathrm{int},1}}^{r_{\mathrm{int},2}}\lVert\chi(r)\rVert^2\,\mathrm{d}r &=\sum_{j=1}^{n_{\mathrm{int}}}\int_{r_{j-1}}^{r_j}\left\lVert\sum_{i=1}^{n_{\mathrm{int}}}\alpha_j^{(i)}e^{ik_jr}+\sum_{i=1}^{n_{\mathrm{int}}}\beta_j^{(i)}e^{-ik_jr}\right\rVert^2\mathrm{d}r\\ &=\sum_{j=1}^{n_{\mathrm{int}}}\left\{\left(\lVert\alpha_j\rVert^2+\lVert\beta_j\rVert^2\right)\Delta r+\frac{\lvert\alpha_j\beta_j\rvert}{k_j}\sin\left(\theta_{\alpha_j}-\theta_{\beta_j}+2k_jr\right)\Big|_{r_{j-1}}^{r_j}\right\}. \tag*{(15)} \end{aligned}

where θαj\theta_{\alpha_j} and θβj\theta_{\beta_j} are phases of the amplitudes αj\alpha_j and βj\beta_j, respectively. If the spatial region includes tunneling, then the formulas above should be rewritten in the complex form (where kjk_j are complex numbers):

∫0rint,max∥χ(r)∥2 dr=∑j=1nint,max∫rj−1rj∥αjeikjr+βje−ikjr∥2dr=∑j=1nint,max{(∥αj∥2+∥βj∥2)Δr+αjβj∗2ikje2ikjr∣rj−1rj−αj∗βj2ikje−2ikjr∣rj−1rj}.(16)\begin{aligned} \int_{0}^{r_{\mathrm{int,max}}}\lVert\chi(r)\rVert^2\,\mathrm{d}r &=\sum_{j=1}^{n_{\mathrm{int,max}}}\int_{r_{j-1}}^{r_j}\left\lVert\alpha_je^{ik_jr}+\beta_je^{-ik_jr}\right\rVert^2\mathrm{d}r\\ &=\sum_{j=1}^{n_{\mathrm{int,max}}}\left\{\left(\lVert\alpha_j\rVert^2+\lVert\beta_j\rVert^2\right)\Delta r+\frac{\alpha_j\beta_j^*}{2ik_j}e^{2ik_jr}\Big|_{r_{j-1}}^{r_j}-\frac{\alpha_j^*\beta_j}{2ik_j}e^{-2ik_jr}\Big|_{r_{j-1}}^{r_j}\right\}. \tag*{(16)} \end{aligned}

In this paper we define the probability of the existence of the compound nucleus through the integral (16) over the spatial region between two internal turning points (where the larger internal turning point is determined as the turning point of the barrier for sub-barrier energies, or as the coordinate of the barrier maximum for above-barrier energies):

Pcn≡∫rint,1rint,2∥χ(r)∥2 dr=∑j=1nint{(∥αj∥2+∥βj∥2)Δr+αjβj∗2ikje2ikjr∣rj−1rj−αj∗βj2ikje−2ikjr∣rj−1rj}.(17)P_{\mathrm{cn}} \equiv\int_{r_{\mathrm{int},1}}^{r_{\mathrm{int},2}} \lVert\chi(r) \rVert^{2} \, \mathrm{d}r = \sum_{j=1}^{n_{\mathrm{int}}} \left\{ \left(\lVert\alpha_j \rVert^{2} + \lVert\beta_j \rVert^{2}\right) \Delta r + \left. \frac{\alpha_j \beta_j^{*}}{2\mathrm{i}k_j} e^{2\mathrm{i}k_j r} \right|_{r_{j-1}}^{r_j} - \left. \frac{\alpha_j^{*}\beta_j}{2\mathrm{i}k_j} e^{-2\mathrm{i}k_j r} \right|_{r_{j-1}}^{r_j} \right\}. \tag*{(17)}

In case of the simplest barrier defined by Eqs. (1) in Ref. [14] (and studied in Sect. II in that paper) one can write down Pcn(E)P_{\mathrm{cn}}(E) as [see Eqs. (6), (7), Ref. [14]]

Pcn(without fusion)=PoscTbarPloc,P_{\mathrm{cn}}^{(\mathrm{without\ fusion})} = P_{\mathrm{osc}} T_{\mathrm{bar}} P_{\mathrm{loc}},
Posc=∥Aosc∥2=(k+k1)22k2(1−cos⁡(2k1r1))+2k12(1+cos⁡(2k1r1)),P_{\mathrm{osc}} = \lVert A_{\mathrm{osc}} \rVert^{2} = \frac{(k+k_1)^2}{2k^2(1-\cos(2k_1r_1))+2k_1^2(1+\cos(2k_1r_1))},
Tbar=k1k2∥T−1∥2,T_{\mathrm{bar}} = \frac{k_1}{k_2} \lVert T_{-1} \rVert^{2},
Ploc=2k22k12(r1−sin⁡(2k1r1)2k1).(18)P_{\mathrm{loc}} = 2\frac{k_2^2}{k_1^2}\left(r_1-\frac{\sin(2k_1r_1)}{2k_1}\right). \tag*{(18)}

Comparing formula (17) with Eq. (18), we see that Eq. (17) is much more complicated. It is not possible to represent it exactly as a simple product of three coefficients of penetrability, oscillations and localization shown in Eqs. 18 (that corresponds to idea of Gamow for inverse processes, i.e. decays, as just direct multiplication of factor of oscillations and penetrability). We also add formula for fast fusion for the simplest barrier as

Pcn(fast fusion)=∥∑i=1nβ1(i)∥2∫0r1∥R0eik1r+e−ik1r∥2 dr=∥T−1∥2r1=k2r1k1Tbar.(19)P_{\mathrm{cn}}^{(\mathrm{fast\ fusion})} = \left\lVert\sum_{i=1}^{n} \beta_1^{(i)} \right\rVert^{2} \int_{0}^{r_1} \left\lVert R_0 e^{\mathrm{i}k_1r}+e^{-\mathrm{i}k_1r} \right\rVert^{2}\,\mathrm{d}r = \lVert T_{-1} \rVert^{2}r_1 = \frac{k_2r_1}{k_1}T_{\mathrm{bar}}. \tag*{(19)}

Cross-section of fusion and coefficients of fusion

In the description of capture of nuclei by nuclei, there is a standard definition of the fusion cross section σ\sigma based on the barrier penetrability Tbar,lT_{\mathrm{bar},l} and fusion probabilities PlP_l (for example, see Ref. [15], where only Pl=1P_l=1 or P0=0P_0=0 according to the sharp angular momentum cutoff in that paper, see also Eq. (3), (4) in Ref. [13] for details):

σfus(E)=∑l=0+∞σl(E),σl(E)=πℏ22mE(2l+1)Tbar,l(E)Pl,(20)\sigma_{\mathrm{fus}}(E)=\sum_{l=0}^{+\infty}\sigma_l(E), \qquad\sigma_l(E)=\frac{\pi\hbar^2}{2mE}(2l+1)T_{\mathrm{bar},l}(E)P_l, \tag*{(20)}

where σl\sigma_l is the partial cross-section of fusion at ll, EE is the energy of the relative motion of the incident nucleus relative to another nucleus. To study the compound nucleus, we introduce a new definition of the partial fusion cross section in terms of the probability of the existence of a compound nucleus (17) as

σl=πℏ22mE(2l+1)fl(E)Pcn(E),(21)\sigma_l=\frac{\pi\hbar^2}{2mE}(2l+1)f_l(E)P_{\mathrm{cn}}(E), \tag*{(21)}

where fl(E)f_l(E) is an additional factor that is needed to connect probability Pcn(E)P_{\mathrm{cn}}(E), penetrability Tbar,l(E)T_{\mathrm{bar},l}(E) and the old factor of fusion PlP_l. To find the explicit form of this coefficient, we consider a case of complete fusion, described by the old formula. A similar result should give a coefficient at coefficients of fusion equal to one. We obtain:

f(E)=kcapkN∥rcap−rtp,in,1∥.(22)f(E)=\frac{k_{\mathrm{cap}}}{k_N\left\lVert r_{\mathrm{cap}}-r_{\mathrm{tp,in,1}}\right\rVert}. \tag*{(22)}

Now, to study formation of the compound nucleus with slow fusion (i.e., without instantaneous fusion), we vary these fusion coefficients in the spatial region between the capture point rcapr_{\mathrm{cap}} and the internal second turning point rint,2r_{\mathrm{int},2}.

Analysis

We will choose reactions XC+XC=2XMg{}^{X}\mathrm{C}+{}^{X}\mathrm{C}={}^{2X}\mathrm{Mg} [11] (X=10,12,14,18,20,22,24X=10,12,14,18,20,22,24) for the analysis. The first indications on the possibility to synthesize more heavy elements from Carbon isotopes can be found in researches of Hamada and Salpeter, who estimated that 12C{}^{12}\mathrm{C} would be converted to 24Mg{}^{24}\mathrm{Mg} via pycnonuclear reactions above density of 6⋅109 g⋅cm−36 \cdot10^{9}\ \mathrm{g} \cdot\mathrm{cm}^{-3} [21]. That study was based on estimations of pycnonuclear reaction rates calculated by Cameron [4]. Then estimates of densities of stellar medium for those reactions were improved (for example, see estimates of Salpeter and Van Horn [1]). However, the densities quoted here are still quite uncertain. Besides difficulty of an accurate calculation, finite temperatures and crystal imperfections can increase the rates significantly (the critical density for carbon can be about 5⋅1010 g⋅cm−35 \cdot10^{10}\ \mathrm{g} \cdot\mathrm{cm}^{-3}). In our research we will focus on understanding of new quantum phenomena which exist in pycnonuclear reactions and have not been studied yet by other researchers. As inclusion of such effects can significantly change the rates of reactions and even the picture of participating mechanisms, for brevity of calculations we will choose the density given by Hamada and Salpeter above.

Potential of interaction between nuclei in lattice sites

We define the potential of interactions between nuclei of carbon XC{}^{X}\mathrm{C} as

V(r)=vC(r)+vN(r)+vl=0(r),(23)V(r) = v_C(r) + v_N(r) + v_{l=0}(r), \tag*{(23)}

where vC(r)v_C(r), vN(r)v_N(r) and vl(r)v_l(r) are Coulomb, nuclear and centrifugal components have the form

vN(r)=−VR1+exp⁡{r−RRaR},vl(r)=l(l+1)2mr2,v_N(r) = -\frac{V_R}{1+\exp\left\{\frac{r-R_R}{a_R}\right\}}, \qquad v_l(r) = \frac{l(l+1)}{2mr^2},
vC(r)={Z1Z2e2r,at r≥RC,Z1Z2e22RC{3−r2RC2},at r<RC.(24)v_C(r) = \begin{cases} \frac{Z_1Z_2e^2}{r}, & \text{at } r \ge R_C, \\ \frac{Z_1Z_2e^2}{2R_C}\left\{3-\frac{r^2}{R_C^2}\right\}, & \text{at } r < R_C. \end{cases} \tag*{(24)}

Here, VRV_R is strength of nuclear components defined in MeV as

VR=−75.0 MeV,(25)V_R=-75.0\ \mathrm{MeV}, \tag*{(25)}

RCR_C and RRR_R are Coulomb and nuclear radiuses of nuclear system, aRa_R is diffusion parameter, mm is reduced mass defined in Eq. (34). We define parameters as [16, 23]

RR=rR(A11/3+A21/3),RC=rC(A11/3+A21/3),aR=0.44 fm,R_R=r_R\left(A_1^{1/3}+A_2^{1/3}\right), \qquad R_C=r_C\left(A_1^{1/3}+A_2^{1/3}\right), \qquad a_R=0.44\ \mathrm{fm},
rR=1.30 fm,rC=1.30 fm.(26)r_R=1.30\ \mathrm{fm}, \qquad r_C=1.30\ \mathrm{fm}. \tag*{(26)}

These potentials for isotopes of Carbon are shown in Figure 1. One can see a shape of internal well of the potential (this

Potentials of interaction between two nuclei of carbon

Figure 1. (Color online) Potentials of interaction between two nuclei of carbon XC{}^{X}\mathrm{C} [potentials and parameters are defined in Eq. (23)–(25)].

internal well is absent in potentials used in Ref. [6], for example). For brevity, we include maximums of barriers and minimums of wells for potentials of interaction between nuclei to Tabl. I.

reactionrmin⁡r_{\min}, fmVmin⁡V_{\min}, MeVrmax⁡r_{\max}, fmVmax⁡V_{\max}, MeVR0R_0, fmnAn_A, 10−7 fm−310^{-7}\,\mathrm{fm}^{-3}
10C+10C{}^{10}\mathrm{C}+{}^{10}\mathrm{C}3.36−62.157-62.1577.98+6.249+6.24987.063.617 027 31
12C+12C{}^{12}\mathrm{C}+{}^{12}\mathrm{C}3.64−63.018-63.0188.33+5.972+5.97292.523.014 189 41
14C+14C{}^{14}\mathrm{C}+{}^{14}\mathrm{C}3.92−63.702-63.7028.68+5.743+5.74397.402.583 590 92
16C+16C{}^{16}\mathrm{C}+{}^{16}\mathrm{C}4.20−64.258-64.2588.96+5.552+5.552101.832.260 642 06
18C+18C{}^{18}\mathrm{C}+{}^{18}\mathrm{C}4.48−64.726-64.7269.24+5.386+5.386105.912.009 459 61
20C+20C{}^{20}\mathrm{C}+{}^{20}\mathrm{C}4.62−65.133-65.1339.52+5.242+5.242109.691.808 513 65
22C+22C{}^{22}\mathrm{C}+{}^{22}\mathrm{C}4.90−65.483-65.4839.80+5.115+5.115113.231.644 103 31
24C+24C{}^{24}\mathrm{C}+{}^{24}\mathrm{C}5.04−65.792-65.79210.08+5.001+5.001116.571.507 094 70

Table 1. Minimums of wells and maximums of barriers of potentials of interactions between two isotopes of carbon, also distance R0R_0 between nuclei and their concentration nAn_A [isotopes of carbon are chosen in accordance with Ref. [8] on systematic study of astrophysical SS-factors in fusion reactions for C, O, Ne, Mg, parameters are determined for density ρ0=6⋅109gcm3\rho_0 = 6 \cdot10^9 \frac{\mathrm{g}}{\mathrm{cm}^3}].

Distance between nuclei in lattice sites

Let us determine at what distance the nuclei are in the stellar matter, taking the density of the matter into account. We denote the distance between two close nuclei fixed in the lattice as 2R02R_0, and put a “scattering” nucleus between these nuclei (in analogy to Ref. [6], p. 90, Fig. 3.5). The density ρ0\rho_0 in the sphere surrounding one nucleus of the lattice can be found as the ratio of mass mAm_A of the nucleus to the volume VAV_A inside this sphere as

ρ0=mAVA=Amu4/3πR03(27)\rho_0 = \frac{m_A}{V_A} = \frac{A m_u}{4/3 \mathop{\mathrm{\pi}} R_0^3} \tag*{(27)}

or

R0=(Amu4/3πρ0)1/3,(28)R_0 = \left(\frac{A m_u}{4/3 \mathop{\mathrm{\pi}} \rho_0}\right)^{1/3}, \tag*{(28)}

where mum_u is mass of nucleon, AA is mass number of nucleus. Concentration of nuclei nAn_A is derived as

nA=ρ0Amu.(29)n_A = \frac{\rho_0}{A m_u}. \tag*{(29)}

For analysis of the pycnonuclear reactions XC+XC=2XMg{}^{X}\mathrm{C}+{}^{X}\mathrm{C}=2{}^{X}\mathrm{Mg} we choose density, estimated in Ref. [6]

ρ0=6⋅109gcm3.(30)\rho_0 = 6 \cdot10^9 \frac{\mathrm{g}}{\mathrm{cm}^3}. \tag*{(30)}

The derived distance R0R_0 and concentration nAn_A for different isotopes of carbon at such a density are given in Tabl. I.

Energy of zero-point vibrations of nuclei in lattice sites

A nucleus located in a lattice site and located between two nuclei with adjacent sites can oscillate and has a discrete spectrum of energy from such oscillations. The approach to determine the energy levels of such a spectrum was proposed by Zel’dovich and investigated by other researchers. In this approach the energy of zero-point vibrations of nucleus in lattice site is calculated as [see Eqs. (3.7.19), (3.7.20), [6]]:

E0(zero)=ℏω2=ℏZemR03,ΔE=2Z2e2R0,Efull=E0(zero)+ΔE.(31)E_0^{(\mathrm{zero})} = \frac{\hbar\omega}{2} = \frac{\hbar Z e}{\sqrt{m R_0^3}},\quad\Delta E = \frac{2Z^2e^2}{R_0},\quad E_{\mathrm{full}} = E_0^{(\mathrm{zero})} + \Delta E. \tag*{(31)}

Here, E0(zero)E_0^{(\mathrm{zero})} is energy of the ground state of harmonic oscillator relative to the potential minimum of this oscillator, ΔE\Delta E is shift of oscillator relative to zero value of the potential of interaction between nuclei (i.e., distance between minimum of oscillator and zero value of potential of the interaction), EfullE_{\mathrm{full}} is energy value of the ground state in system relative to zero value of the potential. For example, for reaction 12C+12C=24Mg{}^{12}\mathrm{C}+{}^{12}\mathrm{C}={}^{24}\mathrm{Mg} we obtain

E0(zero)=0.02180806 MeV,ΔE=0.56787237 MeV,Efull(zero mode)=0.58968043 MeV.(32)E_0^{(\mathrm{zero})} = 0.02180806\ \mathrm{MeV},\quad\Delta E = 0.56787237\ \mathrm{MeV},\quad E_{\mathrm{full}}^{(\mathrm{zero\ mode})} = 0.58968043\ \mathrm{MeV}. \tag*{(32)}
No.Energy, En(zero)E_n^{(\mathrm{zero})}, MeVEnergy, Efull(zero)E_{\mathrm{full}}^{(\mathrm{zero})}, MeV
10.0218080618337360.589680437522993
20.0654241855012080.633296561190465
30.1090403091686800.676912684857937
40.1526564328361530.720528808525410
50.1962725565036260.764144932192882
60.2398886801710980.807761055860354
70.2835048038385700.851377179527827
80.3271209275060430.894993303195299
90.3707370511735150.938609426862772
100.4143531748409870.982225550530244

Table 2. Energy levels for the first 10 states of zero-point vibtations calculated by Eqs. (31) for reactions XC+XCX^{C} + X^{C}.

For brevity of analysis, we call such a state (and the corresponding energy) as state of zero mode (or state of zero-point vibrations of nuclei).

However, harmonic oscillator has not only the ground state, but the full discrete energy spectrum, which are calculated as

En(zero)=(2n+1)⋅ℏω2=(2n+1)⋅En=0(zero)=(2n+1)ℏZemR03.(33)E_n^{(\mathrm{zero})} = (2n+1)\cdot\frac{\hbar\omega}{2} = (2n+1)\cdot E_{n=0}^{(\mathrm{zero})} = (2n+1)\frac{\hbar Z e}{\sqrt{m R_0^3}}. \tag*{(33)}

Energy spectrum can be written down via density of matter ρ0\rho_0, instead of distance R0R_0. Using Eq. (28) and formula for reduced mass

R0=(Amu4/3 πρ0)1/3,m=mpA1A2A1+A2(34)R_0 = \left(\frac{A m_u}{4/3\ \pi\rho_0}\right)^{1/3},\qquad m = m_p\frac{A_1 A_2}{A_1+A_2} \tag*{(34)}

from Eq. (31) we obtain (let us consider case of the same nuclei in lattice: A1=A2A_1=A_2, and A=A1A=A_1):

E0(zero)=c1⋅ZAρ0,c1=ℏe8π3mump,ΔE=c2⋅Z2(ρ0A)1/3,c2=2e2(4π3mu)1/3.(35)\begin{aligned} E_0^{(\mathrm{zero})} &= c_1\cdot\frac{Z}{A}\sqrt{\rho_0},\qquad c_1=\hbar e\sqrt{\frac{8\pi}{3m_u m_p}},\\ \Delta E &= c_2\cdot Z^2\left(\frac{\rho_0}{A}\right)^{1/3},\qquad c_2=2e^2\left(\frac{4\pi}{3m_u}\right)^{1/3}. \tag*{(35)} \end{aligned}

We find new interesting property for nuclei of type 2Z=A2Z=A:

E0(zero)=c12ρ0,ΔE=c2⋅Z2(ρ02Z)1/3.(36)E_0^{(\mathrm{zero})}=\frac{c_1}{2}\sqrt{\rho_0},\qquad\Delta E=c_2\cdot Z^2\left(\frac{\rho_0}{2Z}\right)^{1/3}. \tag*{(36)}

Thus, according to this property, the spectra En(zero)E_n^{(\mathrm{zero})} are the same for nuclei 8Be{}^{8}\mathrm{Be}, 10B{}^{10}\mathrm{B}, 12C{}^{12}\mathrm{C}, 14N{}^{14}\mathrm{N}, 16O{}^{16}\mathrm{O}, 18F{}^{18}\mathrm{F}, 20Ne{}^{20}\mathrm{Ne}, 22Na{}^{22}\mathrm{Na}, 24Mg{}^{24}\mathrm{Mg}, 26Si{}^{26}\mathrm{Si}, etc.. Those depend only on the chosen density in the stellar medium. In Tabl. II energy values are presented for the first 10 states of zero-point vibrations calculated by Eqs. (31) for reactions XC+XCX^{C}+X^{C}.

D Energy spectra of zero-point vibrations: Method of Multiple internal reflections versus Zel'dovich's approach

Formalism of the method of Multiple Internal Reflections can be further investigated, with aim to calculate spectrum of energy in the states of zero-point vibrations. Radial wave function in the asymptotic region has the form as

χ(r)=e−ikr+ARe+ikr.(37)\chi(r)=e^{-ikr}+A_R e^{+ikr}. \tag*{(37)}

In particular, the wave function at point R0R_0 has the same form (where the scattered nucleus is at the minimum of potential between two nuclei in the lattice sites). However, discreteness of the spectrum of energy imposes an additional requirement that the total wave function should be equal to zero (for odd states) or take the maximum value in module (for even states) at that point.We find

1)χ(R0)=e−ikR0+ARe+ikR0=e−ikR0+e+ikR0,AR=+1,2)χ(R0)=e−ikR0+ARe+ikR0=e−ikR0−e+ikR0,AR=−1.(38)\begin{aligned} 1)\quad\chi(R_0)&=e^{-ikR_0}+A_R e^{+ikR_0}=e^{-ikR_0}+e^{+ikR_0},\quad A_R=+1,\\ 2)\quad\chi(R_0)&=e^{-ikR_0}+A_R e^{+ikR_0}=e^{-ikR_0}-e^{+ikR_0},\quad A_R=-1. \tag*{(38)} \end{aligned}

The amplitude of reflection calculated by this way at point R0R_0 for 12C+12C{}^{12}\mathrm{C}+{}^{12}\mathrm{C} is shown in Figure 2. One can see that each function has its own maxima and minima at close energy values. According to Eq. (38), energy levels can be determined from equality to zero of the imaginary part of such an amplitude [see Figure 2 (b)].

The real and imaginary parts of the amplitude of reflection

Figure 2. (Color online) The real and imaginary parts of the amplitude of reflection, AR(r)A_R(r), calculated for 12C+12C{}^{12}\mathrm{C}+{}^{12}\mathrm{C} at point R0R_0 in the region of energy of relative motion Ekin=0.55−1.0E_{\mathrm{kin}}=0.55-1.0 MeV [potential and parameters are defined by Eqs. (23)–(25), R0R_0 is determined by formula (28), its value is specified in Tabl. I].

1) even states:AR=+1, Re⁡(AR)=+1, Im⁡(AR)=0,2) odd states:AR=−1, Re⁡(AR)=−1, Im⁡(AR)=0.(39)\begin{aligned} 1)\ &\text{even states:}\quad A_R=+1,\ \operatorname{Re}(A_R)=+1,\ \operatorname{Im}(A_R)=0,\\ 2)\ &\text{odd states:}\quad A_R=-1,\ \operatorname{Re}(A_R)=-1,\ \operatorname{Im}(A_R)=0. \tag*{(39)} \end{aligned}

In Tabl. III the energy values calculated in the discrete energy spectrum for reactions with carbon isotopes XC+XC→2XMg{}^{X}\mathrm{C}+{}^{X}\mathrm{C}\to{}^{2X}\mathrm{Mg} are presented. Parity of each state is determined by sign of the real part of amplitude of reflection.

Probabilities of formation of compound nuclei in pycnonuclear reactions

Our analysis has shown that the coefficients of penetrability and reflection increases monotonously with the energy of the incident nucleus [16]. There are no maximums and minimums in such dependencies. This means that penetrability and reflection themselves cannot indicate on possible existence of some definite states of more heavy nuclei synthesized in pycnonuclear reactions in stars. Such behaviour of these characteristics is in full agreement with previous studies on the capture of α\alpha particles by nuclei [13, 14].

Another important quantum characteristic is probability of formation of a compound nucleus which can be created during the studied reactions with nuclei. In Figure 3 we present such probabilities for isotopes of Carbon calculated by our method. In these figure one can clearly see presence of maxima in probabilities at certain definite energies for all studied isotopes of Carbon. This means that at such energies compound nuclei are formed with maximum probability. Another conclusion from such calculations is that at such energies the formation of the new XMg{}^{X}\mathrm{Mg} nuclei is much more probable, than at energies of zero mode vibrations determined by the theory of pycnonuclear reactions in compact stars. This result gives an entirely new picture of pycnonuclear reactions in compact stars. These maxima are explained by the need to take into account the further propagation of quantum fluxes in the potential region, in contrast to the existing modern description of pycnonuclear reactions, where these fluxes are interrupted at internal turning point and disappeared in nuclear region. Quantum mechanics requires such a consideration, which strictly requires the continuity of the wave function in the full region of its definition and the conservation of fluxes in the full region of definition of the wave function.

Probabilities of formation of compound nucleus for carbon isotope reactions

Figure 3. (Color online) Probabilities of formation of compound nucleus PcnP_{\mathrm{cn}} in dependence on energy for incident isotopes of Carbon in reactions 10C+10C{}^{10}\mathrm{C}+{}^{10}\mathrm{C}, 12C+12C{}^{12}\mathrm{C}+{}^{12}\mathrm{C} (a), 14C+14C{}^{14}\mathrm{C}+{}^{14}\mathrm{C}, 16C+16C{}^{16}\mathrm{C}+{}^{16}\mathrm{C} (b), 18C+18C{}^{18}\mathrm{C}+{}^{18}\mathrm{C}, 20C+20C{}^{20}\mathrm{C}+{}^{20}\mathrm{C} (c) and 22C+22C{}^{22}\mathrm{C}+{}^{22}\mathrm{C}, 24C+24C{}^{24}\mathrm{C}+{}^{24}\mathrm{C} (d) in lattice [potentials and parameters are defined in Eq. (23)–(25)]. One can clearly see presence of maxima of such probabilities for all studied isotopes. Energies corresponding to maxima of such probabilities are given in Tabl. V.

In Tabl. V we present energies of the quasibound states for reactions with isotopes of Carbon up to 150 MeV. Only first quasibound energies for 10C+10C{}^{10}\mathrm{C}+{}^{10}\mathrm{C}, 12C+12C{}^{12}\mathrm{C}+{}^{12}\mathrm{C}, 24C+24C{}^{24}\mathrm{C}+{}^{24}\mathrm{C} are smaller than the barrier maximums for these nuclear systems. This means that at such energies compound nuclear systems are the most stable and are transformed to new synthesized isotope of Magnesium 20Mg{}^{20}\mathrm{Mg}, 24Mg{}^{24}\mathrm{Mg} and 48Mg{}^{48}\mathrm{Mg} with large probability. There is a simple way to estimate half-lives of these obtained heavier nuclei using Gamow’s approach (well developed in the problem of nuclear decays) or the method of Multiple internal reflections for higher precision (we omit these calculations in this paper).

No.10C+10C^{10}\mathrm{C}+{}^{10}\mathrm{C}12C+12C^{12}\mathrm{C}+{}^{12}\mathrm{C}14C+14C^{14}\mathrm{C}+{}^{14}\mathrm{C}16C+16C^{16}\mathrm{C}+{}^{16}\mathrm{C}
10.5174348697394790.5174348697394790.5174348697394790.527054108216433
20.5366733466933870.5366733466933870.5366733466933870.536072144288577
30.5462925851703410.5462925851703410.5462925851703410.545090180360721
40.5655310621242490.5559118236472950.5559118236472950.554108216432866
50.5847695390781560.5751503006012020.5655310621242490.572144288577154
60.6136272545090180.5943887775551100.5751503006012020.581162324649299
70.6424849699398800.6136272545090180.5943887775551100.599198396793587
80.6809619238476950.6424849699398800.6232464929859720.626252505010020
90.7386773547094190.6809619238476950.6521042084168340.653306613226453
100.8156312625250500.7290581162324650.6905811623246490.689378757515030
110.9503006012024050.8060120240480960.7290581162324650.734468937875752
121.277354709418840.9118236472945890.7963927855711420.797595190380762
132.239278557114231.113827655310620.8925851703406810.878757515030060
143.691783567134272.768336673346691.046492985971941.02304609218437
15—4.086172344689381.642885771543091.39278557114228
16——3.047294589178361.99699398797595
17——4.288176352705413.20541082164329
18———4.37775551102204

Table III. Energy levels of zero-point vibrations EC(mir)E_{C}^{(\mathrm{mir})} for reactions XC+XC^{X}\mathrm{C}+{}^{X}\mathrm{C} (presented data are in MeV, below than 5 MeV) calculated by the method of Multiple internal reflections (see § 3 D, for details). Distances between each two adjacent energy levels are radically different from the spectrum of harmonic oscillator, obtained in the approach of Zel’dovich and his colleagues (see Tabl. II). The amplitude Re⁡(AR)\operatorname{Re}(A_{R}) for each level takes positive or negative value that determines even or odd state in the discrete energy spectrum. The energy levels in table belong to the discrete spectrum for each studied system and there are few energy levels below energy of the zero-point vibrations in the ground state Efull,0(zero)E_{\mathrm{full},0}^{(\mathrm{zero})} calculated in the approach of Zel’dovich and his colleagues (see Eq. (31)). Accuracy of determination of energy levels is estimated from ∣Re⁡(AR)∣≈1\lvert\operatorname{Re}(A_{R}) \rvert\approx1 (it can be improved without limit). Summation of [Re⁡(AR)]2+[Im⁡(ArmR)]2[\operatorname{Re}(A_{R})]^{2}+[\operatorname{Im}(A_{r_mR})]^{2} is additional estimation of accuracy of the method MIR in determination of obtained digits of the amplitude.

No.18C+18C^{18}\mathrm{C}+{}^{18}\mathrm{C}20C+20C^{20}\mathrm{C}+{}^{20}\mathrm{C}22C+22C^{22}\mathrm{C}+{}^{22}\mathrm{C}24C+24C^{24}\mathrm{C}+{}^{24}\mathrm{C}
10.5174348697394790.5270541082164330.5366733466933870.517434869739479
20.5270541082164330.5462925851703410.5462925851703410.527054108216433
30.5366733466933870.5559118236472950.5559118236472950.536673346693387
40.5462925851703410.5655310621242490.5655310621242490.584769539078156
50.5559118236472950.5847695390781560.5751503006012020.594388777555110
60.5655310621242490.6040080160320640.5943887775551100.613627254509018
70.5751503006012020.6232464929859720.6040080160320640.632865731462926
80.5847695390781560.6424849699398800.6232464929859720.661723486938788
90.5943887775551100.6713426853710740.6521042084168340.690581162324649
100.6136272545090180.7098196392785570.6809619238476950.729058116232465
110.6328657314629260.7482965931836730.7194387775555110.767535070140281
120.6617234468937880.8060120240480960.7579158316633270.825250501002004
130.7386773547094190.8829659318637270.8156312625250500.911823647294589
140.7963927855711421.008016032064130.8925851703406811.02725450901804
150.8829659318637271.277354709418841.017635270541081.29659318637275
161.008016032064132.248897795591181.277354709418842.20080160320641
171.306212424849703.268537074148302.248897795591183.14348697394790
182.181563126252514.297755951182373.220440881763534.08617234468938
193.26853707414830—4.21122244488978—

Table IV. (Continuation of Tabl. III)

Plasma screening in nuclear reactions

It is well known that nuclear reactions in compact stars, which contain matter of high density, can be strongly modified by plasma physics effects [1]. So, a natural question appears in how much of the results presented above are changed after taking effects of plasma screening into account. We will follow Ref. [10], where effects of plasma screening in thermonuclear fusion reactions in dense nuclear matter in stars were studied. Here, in addition of physical analysis, authors provided clear formalism for use and implementations to other researches. So, we will estimate influence of plasma screening on pycnonuclear reactions on the basis of isotope 12C{}^{12}\mathrm{C} and we use that research as a basis for our analysis.

Following Ref. [10] [10] [see Eq. (7) in that paper], we define Coulomb potential UC(r)U_C(r) for colliding nuclei in the standard form

UC,full(r)=vC(r)+H(r),(40)U_{C,\mathrm{full}}(r)=v_C(r)+H(r), \tag*{(40)}

where vC(r)v_C(r) is pure Coulomb potential without screening, H(r)H(r) is the mean-field plasma screening potential. In contrast to Ref. [10] [10], we calculate pure Coulomb potential vC(r)v_C(r) on the basis of Eqs. (24)–(26) (here, Coulomb potential in the nuclear region at r<RCr<R_C is different from corresponding potential in Eq. (7) in Ref. [10] [10]), and we use nuclear potential vN(r)v_N(r) in Eq. (24) (we set l=0l=0). In definition of the screening part of potential we follow Ref. [10] [10] and use (see Eqs. (10), (11) in that paper):

H(r)=E12h(x),x=ra12,(41)H(r)=E_{12}h(x),\quad x=\frac{r}{a_{12}}, \tag*{(41)}

where

h(x)=b0+b2x2+b4x4+….(42)h(x)=b_0+b_2x^2+b_4x^4+\ldots. \tag*{(42)}

At Z1/Z2=1Z_1/Z_2=1 parameters b0b_0, b2b_2, b4b_4 are derived in Ref. [10] [10] as

b0=1.0573,b2=−0.25,b4=0.0394.(43)b_0=1.0573,\quad b_2=-0.25,\quad b_4=0.0394. \tag*{(43)}
No.10C+10C^{10}\mathrm{C}+{}^{10}\mathrm{C}12C+12C^{12}\mathrm{C}+{}^{12}\mathrm{C}14C+14C^{14}\mathrm{C}+{}^{14}\mathrm{C}16C+16C^{16}\mathrm{C}+{}^{16}\mathrm{C}18C+18C^{18}\mathrm{C}+{}^{18}\mathrm{C}20C+20C^{20}\mathrm{C}+{}^{20}\mathrm{C}22C+22C^{22}\mathrm{C}+{}^{22}\mathrm{C}24C+24C^{24}\mathrm{C}+{}^{24}\mathrm{C}
10.634714.881769.062127.270546.374755.478965.180364.58317
215.3326711.4509016.5270513.8396811.7495010.555119.062128.46493
326.3807620.4088225.7835721.9018018.3186416.5270514.4368713.24248
440.1162331.4569136.2344730.8597226.3807623.6933920.7074119.21443
555.0460943.6994047.5811640.7134334.7414831.1583227.5751525.48497
671.7675457.1362759.8236551.4629343.9980039.5190435.0400832.35271
789.6833771.7675472.9619262.8096253.8517048.4769542.8036139.81764
8109.3907887.2945986.9959974.7535164.3026157.7334751.1643347.58116
9130.29259104.01603101.6272587.5931975.3507067.8857760.1222455.64329
10–121.93186117.45291101.0300686.9959978.0380869.3700064.30261
11––134.17435115.3627398.9398889.0861779.2324672.96192
12–––130.29259112.07816100.4328789.0861782.51703
13–––146.11824125.21643112.3767599.8356792.07214
14––––139.25050124.91784110.88377102.22445
15–––––137.75752122.23046112.67535
16––––––134.17435123.42485
17––––––146.11824134.77154
18–––––––146.11824

Table V. Energies of the quasibound states of the compound nuclear systems formed in reactions with isotopes of Carbon 10C^{10}\mathrm{C}, 12C^{12}\mathrm{C}, 14C^{14}\mathrm{C}, 16C^{16}\mathrm{C}, 18C^{18}\mathrm{C}, 20C^{20}\mathrm{C}, 22C^{22}\mathrm{C}, 24C^{24}\mathrm{C}, calculated by the method of multiple internal reflections up to 150 MeV [for each calculation, we check test of ∣Tbar+Rbar∣=1|T_{\mathrm{bar}}+R_{\mathrm{bar}}|=1 and obtain accuracy of 10−1410^{-14} of its confirmation]. Comparing these energies with maximums of the potential barriers for all studied systems given in Tabl. I, we find that only first quasibound energies for 10C+10C^{10}\mathrm{C}+{}^{10}\mathrm{C}, 12C+12C^{12}\mathrm{C}+{}^{12}\mathrm{C}, 24C+24C^{24}\mathrm{C}+{}^{24}\mathrm{C} are smaller than barrier maximums for these nuclear systems. That means that at such energies the compound nuclear systems have barriers which prevent decays going through tunneling phenomenon.

Other parameters are (see Eqs. (2), (4) in Ref. [10])

ae=(34πne)1/3,aj=Zj1/3ae,(44)a_e=\left(\frac{3}{4\pi n_e}\right)^{1/3},\quad a_j=Z_j^{1/3}a_e, \tag*{(44)}
a12=a1+a22,E12=Z1Z2e2a12,(45)a_{12}=\frac{a_1+a_2}{2},\quad E_{12}=\frac{Z_1Z_2e^2}{a_{12}}, \tag*{(45)}

where nen_e is concentration of electrons.

On the basis of Eqs. (29) we calculate concentration of nuclei nAn_A at the studied density as

12C+12C, ρ0=6⋅109gcm3: nA=3.01418⋅10−7 fm−3.(46){}^{12}\mathrm{C}+{}^{12}\mathrm{C},\ \rho_0=6\cdot10^9\frac{\mathrm{g}}{\mathrm{cm}^3}:\ n_A=3.01418\cdot10^{-7}\,\mathrm{fm}^{-3}. \tag*{(46)}

This can be understood as each nucleus 12C^{12}\mathrm{C} is in volume with size about 200 fm. From here we find concentration of electrons

ne=Z⋅nA,ne=1.80851⋅10−6 fm−3(47)n_e=Z\cdot n_A,\quad n_e=1.80851\cdot10^{-6}\,\mathrm{fm}^{-3} \tag*{(47)}

and from Eqs. (42)–(43) we obtain

a12=92.52241 fm.(48)a_{12}=92.52241\ \mathrm{fm}. \tag*{(48)}

As it is indicated in Ref. [10], Eq. (42) should be used at x≪2x\ll2. We estimate that this condition is fulfilled in the full region of study of reaction 12C+12C^{12}\mathrm{C}+{}^{12}\mathrm{C} at the chosen density, so we use Eq. (42) in description of the screening part of the potential.

Potential of interactions with taking into account screening, calculated by such an approach, is shown in Figure 4. From this figure one can see that the screening does not change the potential much at the density of matter under consideration. So, one can suppose that the screening does not influence essentially on results of quasibound states and energies above. But, energies of zero point vibrations are essentially smaller than quasibound energies, and one can suppose that energy spectrum of zero-point vibrations will be changed after inclusion of plasma screening to calculations.

Potential of interaction between two nuclei $^{12}\mathrm{C}$ with inclusion of screening, showing full and unscreened potentials and the screening part

Figure 4. (Color online) Potential of interaction between two nuclei 12C^{12}\mathrm{C} with inclusion of screening in comparison with the same potential without screening (a) and part of potential describing screening (b) [potential and parameters are defined in Eq. (23)–(25) screening part of the potential is defined in Eqs. (40)–(42)].

In Tabl. VI values of energies of zero-point vibrations calculated for reaction 12C+12C^{12}\mathrm{C}+{}^{12}\mathrm{C} are presented taking into account screening and without it. From this table one can see that energy spectrum of zero-point vibrations for reaction 12C+12C^{12}\mathrm{C}+{}^{12}\mathrm{C} is changed essentially after taking into account plasma screening.

No.12C+12C^{12}\mathrm{C} + ^{12}\mathrm{C} with screening12C+12C^{12}\mathrm{C} + ^{12}\mathrm{C} without screening
10.2096192384769540.517434869739479
20.2673346693386770.536673346693387
30.3442885771543090.546292585170341
40.4789579158316630.555911823647295
50.7867735470941880.575150300601202
62.210420841683370.594388777555110
73.836072144288580.613627254509018
80.642484969939880
90.680961923847695
100.729058116232465
110.806012024048096
120.911823647294589
131.11382765531062
142.76833667334669
15—4.08617234468938

Table 6. Energies for zero-point vibrations Ezero(mir)E_{zero}^{(\mathrm{mir})} (values are presented in MeV, below 5 MeV) calculated for reaction 12C+12C^{12}\mathrm{C} + ^{12}\mathrm{C}.

Conclusions and perspectives

Question of conditions needed for the most probable formation of compound nuclei (as the first stage needed for synthesis of more heavy elements) in pycnonuclear reactions in compact stars is investigated in this paper. The method is developed on the basis of formalism of the multiple internal reflections, previously constructed for study of quantum processes and phenomena with high precision and testing in nuclear decays [12, 18, 19, 22], nuclear captures by nuclei [13, 14], as well as in problems of quantum cosmology where the idea of tunneling was investigated [26]. In this paper we continue investigations of pycnonuclear reactions with isotopes of Carbon, started in Ref. [16] for 12C+12C^{12}\mathrm{C} + ^{12}\mathrm{C}. Conclusions of our analysis are the following.

  • In this research, pycnonuclear processes are studied taking the nuclear part of potential of interactions between nuclei into account. Requirement of continuity of quantum flux (describing pycnonuclear reaction on the basis of quantum mechanics) gives appearance of new states, in which compound nuclear system of 2XMg^{2X}\mathrm{Mg} is formed with the highest probability (see Figure 3). Following to logic in Refs. [13, 14, 16], we call such states as quasi-bound states in pycnonuclear reactions. Note that these states have not been studied yet by other researchers in study of synthesis of elements in stars.

  • As shown in Figure 3, probability of formation of compound nuclear system in quasibound states is essentially

higher than probability of formation of this system in states of zero-point vibrations studied by Zel’dovich [5] and followers of that idea. Therefore, synthesis of more heavy nuclei of Magnesium from isotopes of Carbon is essentially more probable in quasibound states than in states of zero-point vibrations. This leads to revision (reconsideration) of pictures of formation of heavy elements in compact stars, to use quasibound states as the basis for synthesis. This leads to the essential changes in estimation of the rates of pycnonuclear reactions in stars. One can note the perspective to test the method presented in this paper on the basis of experimental measurements in Ref. [27].

  • Only the first quasibound energies for 10C+10C{}^{10}\mathrm{C} + {}^{10}\mathrm{C}, 12C+12C{}^{12}\mathrm{C} + {}^{12}\mathrm{C}, 24C+24C{}^{24}\mathrm{C} + {}^{24}\mathrm{C} (see Tabl. V) are smaller than the barrier maximums for these nuclear systems (see Tabl. I). So, at such energies the compound nuclear systems have barriers which prevent their decays going through tunneling phenomenon. At such energies the compound nuclear systems are the most probable and the most longer lived. These systems are transformed to new synthesized isotopes of Magnesium 20Mg{}^{20}\mathrm{Mg}, 24Mg{}^{24}\mathrm{Mg} and 48Mg{}^{48}\mathrm{Mg} with large probabilities. There is a simple way to estimate half-lives of these obtained more heavy nuclei using Gamow’s approach or the method of Multiple internal reflections for higher precision. Note that other approaches cannot estimate the quasibound energies needed for prediction of synthesis of more stable nuclear systems by such a way described above. At the same time, the method of Multiple internal reflections calculates such energies with high precision, providing also tests to check calculations. But, analysis of binding energies for the obtained isotopes of Magnesium shows that only 24Mg{}^{24}\mathrm{Mg} will be stable after synthesis.

  • At the first time influence of plasma screening on quasibound states and states of zero-point vibrations in pycnonuclear reactions has been studied. It is found that energy spectrum of zero-point vibrations is changed essentially after taking into account plasma screening (see Tabl. VI for reaction 12C+12C{}^{12}\mathrm{C} + {}^{12}\mathrm{C}).

Acknowledgements

S. P. M. thanks the Wigner Research Centre for Physics in Budapest for warm hospitality and support. Authors are highly appreciated to Prof. V. S. Vasilevsky for fruitful discussions concerning to mechanisms of formation of the compound nuclear system and scattering of nuclei, peculiarities of the method of multiple internal reflections, Prof. A. G. Magner for fruitful discussions concerning to the aspects of nuclear matter in conditions of compact stars and in Earth, Prof. V. M. Ambrus for fruitful discussions concerning to physical issues of astrophysical SS-factors and applicability of quantum methods in stars. G. W. and S. P. M. thank the support of OTKA grant K138277.

References

References

  1. [1]E. E. Salpeter and H. M. Van Horn, Nuclear reaction rates at high densities, Astrophys. J. 155, 183 (1969).DOI
  2. [2]S. Schramm and S. E. Koonin, Astrophys. J. 365, 296 (1990); erratum: 377, 343 (1991).
  3. [3]P. Haensel and J. L. Zdunik, Astron. Astrophys. 229, 117 (1990); P. Haensel and J. L. Zdunik, Astron. Astrophys. 404, 133 (2003).
  4. [4]A. G. W. Cameron, Pycnonuclear reactions and nova explosions, Astrophys. J. 130, 916 (1959).DOI
  5. [5]Ya. B. Zel’dovich and O. H. Guseynov, Collapsed stars in binaries, Astrophys. J. 144, 840 (1965).DOI
  6. [6]S. L. Shapiro and S. A. Teukolsky, Black Holes, White Dwarfs, and Neutron Stars: The Physics of Compact Objects (Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2004), 645 pp.
  7. [7]D. G. Yakovlev, L. R. Gasques, M. Beard, M. Wiescher, and A. V. Afanasjev, Fusion reactions in multicomponent dense matter, Phys. Rev. C 74, 035803 (2006); arXiv:astro-ph/0608488.DOI
  8. [8]M. Beard, A. V. Afanasjev, L. C. Chamon, L. R. Gasques, M. Wiescher, and D. G. Yakovlev, Astrophysical S factors for fusion reactions involving C, O, Ne, and Mg isotopes, At. Dat. Nucl. Dat. Tabl. 96, 541–566 (2010); arXiv: 1002.0741 [astro-ph.SR].DOI
  9. [9]A. V. Afanasjev, M. Beard, A. I. Chugunov, M. Wiescher, and D. G. Yakovlev, Large collection of astrophysical S-factors and its compact representation, Phys. Rev. C 85, 054615 (2012); arXiv: 1204.3174 [astro-ph.SR].DOI
  10. [10]P. A. Kravchuk and D. G. Yakovlev, Strong plasma screening in thermonuclear reactions: Electron drop model, Phys. Rev. C 89, 015802 (2014).DOI
  11. [11]L. R. Gasques, A. V. Afanasjev, E. F. Aguilera, M. Beard, L. C. Chamon, P. Ring, M. Wiescher, and D. G. Yakovlev, Nuclear fusion in dense matter: Reaction rate and carbon burning, Phys. Rev. C 72, 025806 (2005).DOI
  12. [12]S. P. Maydanyuk, S. V. Belchikov, Joun. Phys. Stud. 14 (4), 40 (2011).
  13. [13]S. P. Maydanyuk, P.-M. Zhang, and S. V. Belchikov, Quantum design using a multiple internal reflections method in a study of fusion processes in the capture of alpha-particles by nuclei, Nucl. Phys. A 940, 89–118 (2015); arXiv:1504.00567.DOI
  14. [14]S. P. Maydanyuk, P.-M. Zhang, and L.-P. Zou, New quasibound states of the compound nucleus in α-particle capture by the nucleus, Phys. Rev. C96, 014602 (2017); arXiv:1711.07012.DOI
  15. [15]K. A. Eberhard, Ch. Appel, R. Bangert, L. Cleemann, J. Eberth, and V. Zobel, Fusion cross sections for α+⁴⁰,⁴⁴Ca and the problem of anomalous large-angle scattering, Phys. Rev. Lett. 43 (2), 107–110 (1979).DOI
  16. [16]S. P. Maydanyuk, K. A. Shaulskyi, Quantum design in study of pycnonuclear reactions in compact stars: Nuclear fusion, new quasibound states and spectroscopy, Europ. Phys. J. A58, 220 (2022), arXiv: 2205.13895.
  17. [17]L. H. Chien, D. T. Khou, D. C. Cuong, and N. H. Phuc, Consistent mean-field description of the ¹²C + ¹²C optical potential at low energies and the astrophysical S factor, Phys. Rev. C 98, 064604 (2018).
  18. [18]V. S. Olkhovsky, S. P. Maydanyuk, Method of multiple internal reflections in description of tunneling evolution through barriers, Ukr. Phys. Journ. 45 (10), 1262–1269 (2000); nucl-th/0406035.
  19. [19]S. P. Maydanyuk, V. S. Olkhovsky, A. K. Zaichenko, The method of multiple internal reflections in description of tunneling evolution of nonrelativistic particles and photons, Journ. Phys. Stud. 6 (1), 1–16 (2002), nucl-th/0407108.
  20. [20]S. P. Maydanyuk, S. V. Belchikov, Problem of nuclear decay by proton emission in fully quantum consideration: Calculations of penetrability and role of boundary condition, Journ. Mod. Phys. 2 (6), 572–585 (2011) [open access].DOI
  21. [21]T. Hamada and E. E. Salpeter, Models for zero-temperature stars, Astrophys. J. 134, 683 (1961).
  22. [22]F. Cardone, S. P. Maidanuyuk, R. Mignani, V. S. Olkhovsky, Multiple internal reflections during particle and photon tunneling, Found. Phys. Lett. 19 (5), 441–457 (2006).
  23. [23]S. P. Maydanyuk, V. S. Olkhovsky, G. Mandaglio, M. Manganaro, G. Fazio, and G. Giardina, Bremsstrahlung emission of high energy accompanying spontaneous of ²⁵²Cf, Phys. Rev. C82, 014602 (2010).DOI
  24. [26]S. P. Maydanyuk, Resonant structure of the early-universe space-time, Europ. Phys. J. Plus 126, 76 (2011), arXiv:1005.5447.DOI
  25. [27]X. Fang, W. P. Tan, M. Beard, R. J. deBoer, G. Gilardy, et al., Experimental measurement of ¹²C + ¹⁶O fusion at stellar energies, Phys. Rev. C 96, 045804 (2017).

Paper details

vlnc:W_2k9j5v5w
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 2

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