betaFeedback

WorksPhysics and Astronomy

Atomic corrections for the unique first forbidden ββ transition of 187^{187}Re

Abstract

In this paper, we reexamine one of the most promising candidates for determining the neutrino mass scale – the unique first forbidden ββ transition from 187^{187}Re(5/2+5/2^+) to 187^{187}Os(1/2−1/2^-). With the lowest-known ground-state to ground-state QQ-value for a ββ transition at 2.47092.4709 keV, rhenium's ββ decay can offer insights into the neutrino mass scale puzzle. However, understanding its electron spectrum is a complex task. Besides involving a mixture of s1/2s_{1/2}-state and p3/2p_{3/2}-state electrons, the rhenium ββ spectrum could be strongly influenced by various atomic corrections. In addition to our previous paper, we have incorporated finite nuclear size, diffuse nuclear surface, screening, and exchange corrections into the rhenium ββ decay model. We have accounted for the last two effects within the framework of the Dirac-Hartree-Fock-Slater self-consistent method. We have discovered that both screening and exchange effects significantly alter the partial decay rates for the s1/2s_{1/2}- and p3/2p_{3/2}-state emission channels, while still maintaining the experimentally confirmed dominance of the p3/2p_{3/2}-state emission. The ratio between the respective decay rates has been found to be approximately 10410^4. When compared to the other corrections, the exchange effect stands out due to the modification it induces in the spectrum shape. We demonstrate that calculations with and without the exchange effect lead to entirely different shape factors for the decay spectrum. Finally, we illustrate that to preserve the linearity of the Kurie plot, it is essential to include the exchange correction in its definition. We conclude that atomic effects, especially the exchange effect, should be taken into account in current and future investigations of the neutrino mass scale from ββ decays.

Introduction

An open question in particle physics is related to the absolute values of the light neutrinos masses. The information gathered from solar neutrinos, atmospheric neutrinos, nuclear power reactors, and experiments at accelerators has established that at least two of the three neutrinos have a non-zero mass. However, the result is based on neutrino oscillations, which depends on neutrino mass differences, not on their absolute values, so the absolute neutrino mass scale is uncertain. Besides the cosmological observations, which recently yielded an upper limit for the sum of the neutrino masses of ∼0.12\sim0.12 eV [1], other puzzle pieces can be obtained from nuclear β\beta decay and neutrinoless double β\beta decay [2, 3, 4].

The distortion in the endpoint measurements of the spectrum of electrons emitted in a β\beta decay offers a direct means of determining the values of neutrino masses, denoted as mkm_k (k=1,2,3k = 1, 2, 3). However, the number of events emitted near the endpoint, within an interval ΔTe\Delta T_e, is proportional to (ΔTe/Q)3(\Delta T_e/Q)^3 [5, 6]. Therefore, a low QQ-value β\beta transition is desirable to enhance sensitivity. Consequently, some experiments are based on the ground-state to ground-state β\beta decays of tritium (3H{}^{3}\mathrm{H}) and rhenium (187Re{}^{187}\mathrm{Re}) with QQ-values of 18.6 keV [7] and 2.4709 keV [8], respectively.

Other experiments, such as HOLMES [9], NuMEcs [10], and ECHo [11], aim to utilize the lowest energy electron capture of 163Ho{}^{163}\mathrm{Ho}, which has a ground-state to ground-state QQ-value of 2.833 keV [12]. Recently, there has also been a growing interest in ultra-low QQ-value (under 1 keV) ground-state-to-excited-state β\beta transitions [13, 14, 15], which represent potential candidates for future neutrino mass scale determination experiments. The current best upper limit on effective neutrino mass,

mβ=∑k=13∣Uek∣2mk2(1)m_{\beta} = \sqrt{\sum_{k=1}^{3} \lvert U_{ek} \rvert^{2}m_{k}^{2}} \tag*{(1)}

was recently fixed from the tritium β\beta decay, measured by the KATRIN experiment [16], mβ≤0.8m_{\beta} \leq0.8 eV. This limit far exceeds the previous investigations by Troitsk experiment, mβ≤2.2m_{\beta} \leq2.2 eV [17], and Mainz experiment, mβ≤2.3m_{\beta} \leq2.3 eV [18], also based on tritium β\beta decay. Even more recently, Project 8 has demonstrated that cyclotron radiation emission spectroscopy (CRES) can constrain mβ≤155m_{\beta} \leq155 eV from the tritium β\beta spectrum using only a cm3^{3}-scale physical detection medium [19]. This suggests that CRES is an attractive technique for next-generation direct neutrino mass experiments and for measuring the β\beta spectra in general.

While the number of events near the endpoint is favored by the QQ-value of rhenium decay, which is nearly one order of magnitude smaller than that of tritium decay, understanding its spectrum shape is a complex task. Firstly, the emission of low-energy electrons suggests that various atomic and molecular effects could have a significant impact on the spectrum shape of the rhenium ground-state to ground-state β\beta transition. An example of this is the oscillations observed in the rhenium β\beta spectrum due to environmental fine structure from the target crystal [20]. Secondly, in contrast to the decay of tritium, which is an allowed transition, the transition 187Re(5/2+)→187Os(1/2−){}^{187}\mathrm{Re}(5/2^+) \to{}^{187}\mathrm{Os}(1/2^-) is classified as a unique first forbidden transition. Consequently, its spectrum is a mixture of s1/2s_{1/2}-state and p3/2p_{3/2}-state electrons. The exceptionally low QQ-value of rhenium decay results in a clear dominance of the p3/2p_{3/2}-state electron emission channel over the s1/2s_{1/2}-state electron emission channel. This dominance has been experimentally confirmed [21], and our previous investigation calculated the ratio between the channels to be around 10410^4 [6].

The experimental efforts to achieve sub-eV sensitivity to the neutrino mass from the rhenium β\beta spectrum culminated with the combination of the MANU and MI-BETA groups into the Microcalorimeter Arrays for a Rhenium Experiment (MARE) [22]. The MARE project eventually transitioned to the holmium efforts of ECHo and HOLMES [23]. One possible reason for closing the project is the lack of theoretical knowledge about the rhenium β\beta spectrum. In contrast, the theoretical description of tritium β\beta decay involves many more corrections, including screening, exchange with atomic electrons, finite nuclear size effects, radiative corrections, recoil effects, etc. [24, 25].

In this paper, we investigate the β\beta decay of 187Re{}^{187}\mathrm{Re}, incorporating all relevant corrections to its spectrum. We begin with the same relativistic wave functions for the emitted electrons as used in our previous study [6]. Next, we enhance the precision of the theoretical rhenium decay spectrum by including finite nuclear size, diffuse nuclear surface, and screening corrections. The latter is calculated using the self-consistent Dirac-Hartree-Fock-Slater description of the atomic bound electrons surrounding the electron emitted during rhenium decay. We have observed significant differences in the decay rates for both emission channels compared to our previous paper, but negligible modifications to the spectral shape resulting from the aforementioned corrections. The primary focus of this paper lies in the inclusion of the so-called exchange correction, which considers the possible interchange between emitted electrons and the atomic bound electrons. It turns out that the exchange correction not only alters the decay rates for the p3/2p_{3/2}- and s1/2s_{1/2}-state channels but also affects the shapes of the spectra for these channels. While the most significant shape modification occurs in the low-energy region of the total spectrum, the exchange correction remains substantial near the endpoint, potentially impacting the analysis for neutrino mass scale determination from rhenium β\beta decay. When studying the deviation of the rhenium spectrum from an allowed one, we discover that the exchange correction strongly transforms the shape factor, shifting it from an increasing linear behavior to a decreasing quadratic one. We present the best-fit parameters in both cases. To preserve the linearity of the Kurie plot in the scenario of zero effective neutrino mass, it is necessary to revise its definition to incorporate the exchange correction. We illustrate how varying effective neutrino masses impact the Kurie plots near the endpoint of the β\beta decay of 187Re{}^{187}\mathrm{Re}. While this paper is primarily aimed on the rhenium β\beta decay, we emphasize that the corrections introduced here could potentially have a significant effect on any low and ultra-low QQ-value β\beta transition.

Electron Wave Functions

Continuum states

The relativistic wave function for an electron emitted from a β\beta decay in a continuum state can be expanded in spherical waves,

ψ(Ee,r)=ψ(s1/2)(Ee,r)+ψ(p1/2)(Ee,r)+ψ(p3/2)(Ee,r)+⋯(2)\psi(E_e,\mathbf{r})=\psi^{(s_{1/2})}(E_e,\mathbf{r})+\psi^{(p_{1/2})}(E_e,\mathbf{r})+\psi^{(p_{3/2})}(E_e,\mathbf{r})+\cdots \tag*{(2)}

where EeE_e is the total electron energy. The superscript in the spherical waves represents the orbital and total angular momentum (ℓj=01/2,11/2,13/2,…\ell_j=0_{1/2},1_{1/2},1_{3/2},\ldots) written in the spectroscopic notation (ℓj=s1/2,p1/2,p3/2,…\ell_j=s_{1/2},p_{1/2},p_{3/2},\ldots). The relevant spherical waves for the first-unique forbidden β\beta transition of 187Re{}^{187}\mathrm{Re} are as follows [6]:

ψ(s1/2)(Ee,r)=(g−1(Ee,r)χμf+1(Ee,r)(σ⋅p^)χμ),ψ(p3/2)(Ee,r)=i(g−2(Ee,r)[3(r^⋅p^)(σ⋅p^)−(σ⋅r^)]χμf+2(Ee,r)[(r^⋅p^)(σ⋅p^)−(σ⋅r^)]χμ)(3)\psi^{(s_{1/2})}(E_e,\mathbf{r})= \begin{pmatrix} \dfrac{g_{-1}(E_e,r)\chi_\mu}{f_{+1}(E_e,r)(\boldsymbol{\sigma}\cdot\hat{\mathbf{p}})\chi_\mu} \end{pmatrix}, \qquad \psi^{(p_{3/2})}(E_e,\mathbf{r})= i\begin{pmatrix} \dfrac{g_{-2}(E_e,r)[3(\hat{\mathbf{r}}\cdot\hat{\mathbf{p}})(\boldsymbol{\sigma}\cdot\hat{\mathbf{p}})-(\boldsymbol{\sigma}\cdot\hat{\mathbf{r}})]\chi_\mu}{f_{+2}(E_e,r)[(\hat{\mathbf{r}}\cdot\hat{\mathbf{p}})(\boldsymbol{\sigma}\cdot\hat{\mathbf{p}})-(\boldsymbol{\sigma}\cdot\hat{\mathbf{r}})]\chi_\mu} \end{pmatrix} \tag*{(3)}

where p^=pe/pe\hat{\mathbf{p}}=\mathbf{p}_e/p_e is defined by the momentum of the electron, pe=Ee2−me2p_e=\sqrt{E_e^2-m_e^2}. Here, r\mathbf{r} stands for the position vector of the electron, r=∣r∣r=|\mathbf{r}| and r^=r/r\hat{\mathbf{r}}=\mathbf{r}/r. Throughout the paper, we are using the units ℏ=c=1\hbar=c=1.

The functions gκ(Ee,r)g_\kappa(E_e,r) and fκ(Ee,r)f_\kappa(E_e,r) are the large- and small-component radial wave functions, respectively, which satisfy the following system of coupled differential equations [26]

(ddr+κ+1r)gκ−(Ee−V(r)+me)fκ=0,(ddr−κ−1r)fκ+(Ee−V(r)−me)gκ=0,(4)\begin{aligned} \left(\frac{d}{d r}+\frac{\kappa+1}{r}\right)g_{\kappa}-\left(E_{e}-V(r)+m_{e}\right)f_{\kappa}&=0,\\ \left(\frac{d}{d r}-\frac{\kappa-1}{r}\right)f_{\kappa}+\left(E_{e}-V(r)-m_{e}\right)g_{\kappa}&=0, \tag*{(4)} \end{aligned}

where V(r)V(r) is the atomic potential of the final system.

The relativistic quantum number, κ\kappa, takes positive and negative integer values, and specifies both the total angular momentum, jj, and the orbital angular momentum, ℓ\ell, by

j=∣κ∣−1/2,ℓ={κif κ>0,∣κ∣−1if κ<0.(5)j=|\kappa|-1/2,\quad\ell=\begin{cases}\kappa&\text{if }\kappa>0,\\|\kappa|-1&\text{if }\kappa<0.\end{cases} \tag*{(5)}

The large- and small-component radial functions satisfy, for large values of perp_{e}r, the following asymptotic conditions

{gκ(Ee,r)fκ(Ee,r)}∼1per{Ee+me2Eesin⁡(per−ℓπ2+δκ−ηln⁡(2per))Ee−me2Eecos⁡(per−ℓπ2+δκ−ηln⁡(2per))}(6)\begin{Bmatrix}g_{\kappa}(E_{e},r)\\f_{\kappa}(E_{e},r)\end{Bmatrix}\sim\frac{1}{p_{e}r}\begin{Bmatrix}\sqrt{\frac{E_{e}+m_{e}}{2E_{e}}}\sin\left(p_{e}r-\ell\frac{\pi}{2}+\delta_{\kappa}-\eta\ln(2p_{e}r)\right)\\\sqrt{\frac{E_{e}-m_{e}}{2E_{e}}}\cos\left(p_{e}r-\ell\frac{\pi}{2}+\delta_{\kappa}-\eta\ln(2p_{e}r)\right) \tag*{(6)}\end{Bmatrix}

where δκ\delta_{\kappa} is the phase shift and η=αZEe/pe\eta=\alpha Z E_{e}/p_{e} is the Sommerfeld parameter. Here, ZZ is the atomic number of the final nucleus and α\alpha is the fine-structure constant. The energy-dependent normalization ensures the correct behavior of the Fermi functions (see § 3) when the electrostatic interaction is switched off, i.e., V(r)=0V(r)=0.

To obtain the electron continuum states in the potential of the final positive ion, 187Os+{}^{187}\mathrm{Os}^{+}, we employ the RADIAL subroutine package [27]. A comprehensive manual of the package can be found in the Supplementary Material of [27]. In what follows, we briefly present different approximation schemes for the calculation of gκ(Ee,r)g_{\kappa}(E_{e},r) and fκ(Ee,r)f_{\kappa}(E_{e},r).

The approximation scheme A. We assume the final nucleus as a uniformly charged sphere, generating the following potential,

V(r)={−αZrfor r≥R,−αZ2R[3−(rR)2]for r<R.(7)V(r)=\begin{cases}-\dfrac{\alpha Z}{r}&\text{for }r\ge R,\\-\dfrac{\alpha Z}{2R}\left[3-\left(\dfrac{r}{R}\right)^{2}\right]&\text{for }r<R.\end{cases} \tag*{(7)}

Here, RR is the radius of the final nucleus, R=r0A1/3R=r_{0}A^{1/3} with r0=1.2r_{0}=1.2 fm. By keeping the lowest power of the expansion of rr, the radial wave functions for the s1/2s_{1/2} wave and p3/2p_{3/2} wave states are given by [28]

\begin{pmatrix}g_{-1}(E_{e},r)\\f_{+1}(E_{e},r)\end{pmatrix}=\begin{pmatrix}A_{-1}\\A_{+1}\end{pmatrix}, \tag*{(8)}\end{pmatrix}

and

\begin{pmatrix}g_{-2}(E_{e},r)\\f_{+2}(E_{e},r)\end{pmatrix}=\frac{p_{e}r}{3}\begin{pmatrix}A_{-2}\\A_{+2}\end{pmatrix}, \tag*{(9)}\end{pmatrix}

respectively.

The normalization constant can be expressed in a good approximation as

A±k≃Fk−1(Z,Ee)Ee+me2Ee(10)A_{\pm k}\simeq\sqrt{F_{k-1}(Z,E_{e})}\sqrt{\frac{E_{e}+m_{e}}{2E_{e}}} \tag*{(10)}

where k=∣κ∣k=|\kappa| and the Fermi function Fk−1(Z,Ee)F_{k-1}(Z,E_{e}) is given by

Fk−1(Z,Ee)=[Γ(2k+1)Γ(k)Γ(2γk+1)]2(2pR)2(γk−k)eπη∣Γ(γk+iη)∣2.(11)F_{k-1}(Z,E_{e})=\left[\frac{\Gamma(2k+1)}{\Gamma(k)\Gamma(2\gamma_{k}+1)}\right]^{2}(2pR)^{2(\gamma_{k}-k)}e^{\pi\eta}\left|\Gamma(\gamma_{k}+i\eta)\right|^{2}. \tag*{(11)}

The remaining quantity is given by

γk=k2−(αZ)2,(12)\gamma_{k}=\sqrt{k^{2}-(\alpha Z)^{2}}, \tag*{(12)}

and Γ(z)\Gamma(z) is the Gamma function. We mention that this approximation scheme, was also used in our previous investigation [6].

The approximation scheme B. In this approximation scheme, we consider the case, where the final nucleus generates a point-like potential, V(r)=−αZ/rV(r)=-\alpha Z/r. The radial wave functions can be expressed analytically as [29]

gκ(Ee,r)=κkprEe+me2Ee∣Γ(1+γk+iη)∣Γ(1+2γk)(2pr)γkeπη/2×Im⁡{ei(pr+ζ)1F1(γk−iη,1+2γk,−2ipr)},fκ(Ee,r)=κkprEe−me2Ee∣Γ(1+γk+iη)∣Γ(1+2γk)(2pr)γkeπη/2×Re⁡{ei(pr+ζ)1F1(γk−iη,1+2γk,−2ipr)},(13)\begin{aligned} g_{\kappa}(E_{e},r)&=\frac{\kappa}{k p r}\sqrt{\frac{E_{e}+m_{e}}{2E_{e}}}\frac{\left|\Gamma(1+\gamma_{k}+i\eta)\right|}{\Gamma(1+2\gamma_{k})}(2pr)^{\gamma_{k}}e^{\pi\eta/2}\\ &\quad\times\operatorname{Im}\left\{e^{i(pr+\zeta)}{}_{1}F_{1}(\gamma_{k}-i\eta,1+2\gamma_{k},-2ipr)\right\},\\ f_{\kappa}(E_{e},r)&=\frac{\kappa}{k p r}\sqrt{\frac{E_{e}-m_{e}}{2E_{e}}}\frac{\left|\Gamma(1+\gamma_{k}+i\eta)\right|}{\Gamma(1+2\gamma_{k})}(2pr)^{\gamma_{k}}e^{\pi\eta/2}\\ &\quad\times\operatorname{Re}\left\{e^{i(pr+\zeta)}{}_{1}F_{1}(\gamma_{k}-i\eta,1+2\gamma_{k},-2ipr)\right\}, \tag*{(13)} \end{aligned}

with

eiζ=κ−iηme/Eeγk−iη.(14)e^{i\zeta}=\sqrt{\frac{\kappa-i\eta m_{e}/E_{e}}{\gamma_{k}-i\eta}}. \tag*{(14)}

Here, 1F1(a,b,z){}_{1}F_{1}(a,b,z) is the confluent hypergeometric function. We mention that the numerical solutions from the RADIAL package, with the input rV(r)=−αZrV(r)=-\alpha Z, are equivalent to the analytical solutions presented in Eq. (13), if we fix rr on the nuclear surface.

The approximation scheme C. We consider the final nucleus as a sphere filled with protons following a Fermi distribution [30]

ρp(r)=ρ01+e(r−crms)/a,(15)\rho_{p}(r)=\frac{\rho_{0}}{1+e^{(r-c_{\mathrm{rms}})/a}}, \tag*{(15)}

where we chose for the half-way radius crms=1.07A1/3=6.118c_{\mathrm{rms}}=1.07A^{1/3}=6.118 fm and for the surface thickness a=0.546a=0.546 fm. ρ0\rho_{0} is determined from the normalization to ZZ.

Thus, the electrostatic interaction of an electron at rr with the final nucleus is described by

Vnuc(r)=−α∫ρp(r′)∣r−r′∣ dr′.(16)V_{\mathrm{nuc}}(r)=-\alpha\int\frac{\rho_p(r')}{\lvert r-r'\rvert}\,\mathrm{d}r'. \tag*{(16)}

The approximation scheme D. In our final approximation scheme, we consider for the final positive ion, 187Os+{}^{187}\mathrm{Os}^{+}, the following electrostatic potential,

V(r)=Vnuc(r)+Vel(r)+VexSlater(r),(17)V(r)=V_{\mathrm{nuc}}(r)+V_{\mathrm{el}}(r)+V_{\mathrm{ex}}^{\mathrm{Slater}}(r), \tag*{(17)}

which is a sum of the nuclear, electronic and exchange potentials. The nuclear potential, Vnuc(r)V_{\mathrm{nuc}}(r), is presented in the approximation scheme C, i.e. Eq. (16). The electronic potential, Vel(r)V_{\mathrm{el}}(r), describes the interaction energy of an electron at rr with the atomic cloud, and it is found from integrating over the volume of the atomic electron density, ρ(r)\rho(r),

Vel(r)=α∫ρ(r′)∣r−r′∣ dr′.(18)V_{\mathrm{el}}(r)=\alpha\int\frac{\rho(r')}{\lvert r-r'\rvert}\,\mathrm{d}r'. \tag*{(18)}

Considering Slater’s approximation [31], we can write the exchange potential in terms of the atomic electron density,

VexSlater(r)=−32α(3π)1/3[ρ(r)]1/3.(19)V_{\mathrm{ex}}^{\mathrm{Slater}}(r)=-\frac{3}{2}\alpha\left(\frac{3}{\pi}\right)^{1/3}[\rho(r)]^{1/3}. \tag*{(19)}

The atomic electronic density, ρ(r)\rho(r), is discussed in the following section, where we present the calculation for the electron bound states.

Our choice to include a local exchange component in the potential for the continuum states can be arguable, but, as motivated later, it ensures the orthogonality between the continuum states of the emitted electron and the bound states of the atomic electrons of 187Os+{}^{187}\mathrm{Os}^{+}. Orthogonal continuum and bound electron wave functions for the final nucleus are crucial ingredients for correctly calculating the exchange correction [32, 33, 34].

In Figure 1, we plot all quantities rV(r)rV(r) as functions of rr, for each approximation scheme considered above. In the simplest case of a point-like nucleus (scheme B), we can see that rV(r)=−76αrV(r)=-76\alpha is a straight line. In the most complex case (scheme D), where we include the finite nuclear size, diffuse nuclear surface, and the atomic screening corrections, for large values of rr, V(r)=−α/rV(r)=-\alpha/r. It is an expected asymptotic result because the potential describes the interaction of the emitted electron with a positive ion of charge +1+1. We can see that when moving from a uniformly charged sphere (scheme A) to one filled with protons following a Fermi distribution (scheme C), the difference between potentials is slightly visible. Still, the electron wave functions A are approximated by keeping the lowest power in the expansion in rr. By solving the radial Dirac equation numerically for the potentials A and C, we confirm that there are no differences between the wave functions due to the diffuse nuclear surface correction. Thus, any differences between the electron wave

Electrostatic potentials for ${}^{187}\mathrm{Os}^{+}$

Figure 1. The electrostatic potential for 187Os+{}^{187}\mathrm{Os}^{+} as function of rr, where the emitted electron is located, in four different approximation schemes: (A) The final nucleus as an uniform charged sphere. (B) A point-like final nucleus. (C) The final nucleus as a charged sphere filled with protons following a Fermi distribution. (D) The same as the preceding case but the atomic electron screening is taken into account.

functions A and C are associated with missing terms in the expansion.

In Figure 2, we depict the radial wave functions of an electron in the s1/2s_{1/2} wave state, g−1(Ee,r)g_{-1}(E_e,r) and f+1(Ee,r)f_{+1}(E_e,r) (top panel), and in the p3/2p_{3/2} wave state, g−2(Ee,r)g_{-2}(E_e,r) and f+2(Ee,r)f_{+2}(E_e,r) (bottom panel), evaluated on the nuclear surface of the final nucleus, r=Rr=R, and as functions of the electron kinetic energy, Ee−meE_e-m_e. We present the results for the β\beta emitter 187Re{}^{187}\mathrm{Re} using all the approximation schemes. In the case of the large component radial wave functions, g−1(Ee,r)g_{-1}(E_e,r) and g−2(Ee,r)g_{-2}(E_e,r), there are no shape deviations between different approximation schemes. Their amplitudes progressively decrease from scheme A to D. The missing terms in the expansion in rr from scheme A strongly influence the small component radial wave functions, f+1(Ee,r)f_{+1}(E_e,r) and f+2(Ee,r)f_{+2}(E_e,r). Instead of overlying the wave functions C, they are almost zero and constants for any electron kinetic energy. As a final remark, the effect of the atomic screening for an electron emitted in the p3/2p_{3/2} wave state is substantial, as can be seen in the wave functions D. The emission of electrons in the p3/2p_{3/2} wave state is dominant for the β\beta decay of 187Re{}^{187}\mathrm{Re}, so we expect a considerable effect when the screening correction is switched on.

Figure 2. The radial wave functions for an electron emitted from the β\beta decay of 187Re^{187}\mathrm{Re}, evaluated on the nuclear surface of the final nucleus RR, as functions of the electron kinetic energy Ee−meE_e-m_e. In the top panel, the components of an electron emitted in s1/2s_{1/2} state, g−1(Ee,r)g_{-1}(E_e,r) and f+1(Ee,r)f_{+1}(E_e,r), and in the bottom panel, the components associated with the emission in p3/2p_{3/2} state, g−2(Ee,r)g_{-2}(E_e,r) and f+2(Ee,r)f_{+2}(E_e,r). We consider four different approximation schemes in the evaluation of the radial wave functions (see text for details).

Bound states

For atomic bound states (Ee<meE_e<m_e), each discrete energy level is characterized by its quantum number kk, its principal quantum number nn, its binding energy ϵenk\epsilon_e^{nk}, and its total energy Eenk=me−∣ϵenk∣E_e^{nk}=m_e-\lvert\epsilon_e^{nk}\rvert. The relativistic wave function for a bound electron in a spherically symmetric atomic potential, V(r)V(r), can be written as [26]

ψnκm(r)=(gn,κ(r)Ωκ,m(r^)ifn,κ(r)Ω−κ,m(r^)),(20)\psi_{n\kappa m}(\mathbf{r}) = \begin{pmatrix} g_{n,\kappa}(r)\Omega_{\kappa,m}(\hat{\mathbf{r}}) \\ if_{n,\kappa}(r)\Omega_{-\kappa,m}(\hat{\mathbf{r}}) \end{pmatrix}, \tag*{(20)}

Here, the spherical spinors, Ωκ,m(r^)\Omega_{\kappa,m}(\hat{\mathbf{r}}), are defined by [35, 36]

Ωκ,m(r^)=∑μ=±1/2⟨ℓ,12,m−μ,μ|j,m⟩Yℓ,m−μ(r^)χμ(21)\Omega_{\kappa,m}(\hat{\mathbf{r}}) = \sum_{\mu=\pm1/2} \left\langle\ell,\frac{1}{2},m-\mu,\mu\middle|j,m\right\rangle Y_{\ell,m-\mu}(\hat{\mathbf{r}})\chi_\mu \tag*{(21)}

where ⟨j1,j2,m1,m2|j,m⟩\left\langle j_1,j_2,m_1,m_2\middle|j,m\right\rangle are Clebsch-Gordan coefficient, Yℓ,m(r^)Y_{\ell,m}(\hat{\mathbf{r}}) are the spherical harmonics, and χμ\chi_\mu are the usual Pauli spinors. The large- and small-component radial functions for bound orbitals are referred as gn,κ(r)g_{n,\kappa}(r) and fn,κ(r)f_{n,\kappa}(r), respectively, and are obeying,

(ddr+κ+1r)gn,κ−(Eenκ−V(r)+me)fn,κ=0,\left(\frac{d}{dr}+\frac{\kappa+1}{r}\right)g_{n,\kappa}-(E_e^{n\kappa}-V(r)+m_e)f_{n,\kappa}=0,
(ddr−κ−1r)fn,κ+(Eenκ−V(r)−me)gn,κ=0,(22)\left(\frac{d}{dr}-\frac{\kappa-1}{r}\right)f_{n,\kappa}+(E_e^{n\kappa}-V(r)-m_e)g_{n,\kappa}=0, \tag*{(22)}

For the calculation of the bound orbitals, we employed the program DHFS.f, included in the RADIAL subroutine package [27]. It solves the Dirac-Hartree-Fock-Slater (DHFS) equations for the ground-state electron configuration of neutral atoms or positive ions with NeN_e bound electrons and ZZ protons in the nucleus. The procedure uses almost the same potential as described in Eq. (17), but with the difference that the Latter’s tail correction [37] is imposed for the exchange potential. The electron density, ρ(r)\rho(r), entering the electronic and exchange potentials, is obtained self-consistently [38, 39]. The procedure starts with an approximate electron density obtained from the Molière parametrization of the Thomas-Fermi potential [40]. Then, the electron density is renewed iteratively from the obtained bound orbitals until neither the DHFS potential nor the binding energies change. We call the procedure from the program DHFS.f a true DFHS self-consistent method.

Because we want to ensure orthogonality between continuum and bound states, we take the last iterated atomic electron density, ρ(r)\rho(r), obtained from the program DHFS.f, and construct the potential from the approximation scheme D, i.e., Eq. (17). When solving for bound states in this potential, we call the procedure a modified DHFS self-consistent method. More about the differences between the true and modified DHFS self-consistent frameworks can be found in our previous paper about the exchange correction for allowed β\beta decay [34]. In Table I, we compare the binding energies obtained with the true and modified DHFS methods with the experimental values for the neutral atom 187Re^{187}\mathrm{Re}. We can see that the deviation from the conventional (true) DHFS method subtly influences the binding energies of the last three occupied orbitals.

Orbital (nℓj)(n\ell_j)ϵenκ\epsilon_e^{n\kappa} (true)ϵenκ\epsilon_e^{n\kappa} (modified)ϵenκ\epsilon_e^{n\kappa} (exp) [42]
1s1/21s_{1/2}−71857.5-71857.5−71857.5-71857.5−71681±2-71681 \pm2
2s1/22s_{1/2}−12508.4-12508.4−12508.4-12508.4−12532±2-12532 \pm2
2p1/22p_{1/2}−11993.7-11993.7−11993.7-11993.7−11963±2-11963 \pm2
2p3/22p_{3/2}−10537.7-10537.7−10537.7-10537.7−10540±2-10540 \pm2
3s1/23s_{1/2}−2911.9-2911.9−2911.9-2911.9−2937±2-2937 \pm2
3p1/23p_{1/2}−2677.7-2677.7−2677.7-2677.7−2686±2-2686 \pm2
3p3/23p_{3/2}−2360.0-2360.0−2360.0-2360.0−2371±2-2371 \pm2
3d3/23d_{3/2}−1961.4-1961.4−1961.4-1961.4−1953±2-1953 \pm2
3d5/23d_{5/2}−1891.9-1891.9−1891.9-1891.9−1887±2-1887 \pm2
4s1/24s_{1/2}−615.7-615.7−615.7-615.7−629±2-629 \pm2
4p1/24p_{1/2}−516.7-516.7−516.7-516.7−522±2-522 \pm2
4p3/24p_{3/2}−442.2-442.2−442.2-442.2−450±2-450 \pm2
4d3/24d_{3/2}−277.8-277.8−277.8-277.8−278±2-278 \pm2
4d5/24d_{5/2}−263.9-263.9−263.9-263.9−264±2-264 \pm2
5s1/25s_{1/2}−91.2-91.2−91.2-91.2−86±2-86 \pm2
5p1/25p_{1/2}−60.6-60.6−60.6-60.6−56±2-56 \pm2
4f5/24f_{5/2}−55.0-55.0−55.0-55.0−47±2-47 \pm2
4f7/24f_{7/2}−52.3-52.3−52.3-52.3−45±2-45 \pm2
5p3/25p_{3/2}−49.0-49.0−49.0-49.0−45±2-45 \pm2
5d3/25d_{3/2}−9.28-9.28−9.24-9.24−9.6±1-9.6 \pm1
5d5/25d_{5/2}−8.24-8.24−8.20-8.20−9.6±1-9.6 \pm1
6s1/26s_{1/2}−7.98-7.98−7.67-7.67−7.9±1-7.9 \pm1

Table I. Binding energies for neutral atom 187Re^{187}\mathrm{Re} in eV. In the first column, we indicate all occupied shells using the spectroscopic notation [27, 41]. In the second and third columns, we present the true DHFS self-consistent method binding energies and the results obtained with the modified DHFS self-consistent method, respectively. In the last column, we present the experimental values taken from [42].

First Unique Forbidden β\beta Decay of 187Re^{187}\mathrm{Re}

The β\beta transition from the ground-state 5/2+5/2^+ of 187Re^{187}\mathrm{Re} to the ground-state 1/2−1/2^- of 187Os^{187}\mathrm{Os} is classified as a first unique forbidden β\beta transition. Due to the difference in the angular momentum and parity, ΔJπ=2−\Delta J^\pi= 2^-, the electron and neutrino are emitted, respectively, in p3/2p_{3/2} and s1/2s_{1/2} states or vice versa. An interesting feature of first unique forbidden β\beta decay of 187Re^{187}\mathrm{Re} is that due to its low transition energy, Q=2.4709 keVQ = 2.4709\ \mathrm{keV} [8], the emission of p3/2p_{3/2} state electrons is favored with four orders of magnitude then the emission of s1/2s_{1/2} state electrons [6]. The theoretical differential decay rate is a sum of two contributions associated with emission of the s1/2s_{1/2} and the p3/2p_{3/2} state electrons,

state electrons,

dΓdEe=dΓp3/2dEe+dΓs1/2dEe=∑k=13∣Uek∣2GF2Vud22π3peEe(E0−Ee)(E0−Ee)2−mk2 θ(E0−Ee−mk)×[Bp3/2(Ee,pν)+Bs1/2(Ee,pν)](23)\frac{d\Gamma}{dE_e} = \frac{d\Gamma^{p_{3/2}}}{dE_e} + \frac{d\Gamma^{s_{1/2}}}{dE_e} = \sum_{k=1}^{3} |U_{ek}|^2 \frac{G_F^2 V_{ud}^2}{2\pi^3} p_e E_e (E_0-E_e) \sqrt{(E_0-E_e)^2-m_k^2}\,\theta(E_0-E_e-m_k) \times\left[B^{p_{3/2}}(E_e,p_\nu)+B^{s_{1/2}}(E_e,p_\nu)\right] \tag*{(23)}

where

Bs1/2=12gA2×(∣⟨f∥∑nτn+g−1(Ee,r)j1(pνr){σn⊗r^n}2∥i⟩∣2+∣⟨f∥∑nτn+f+1(Ee,r)j1(pνr){σn⊗r^n}2∥i⟩∣2)(24)B^{s_{1/2}} = \frac{1}{2}g_A^2 \times \left( \left|\langle f\Vert\sum_n \tau_n^+ g_{-1}(E_e,r)j_1(p_\nu r)\{\sigma_n\otimes\hat r_n\}_2\Vert i\rangle\right|^2 + \left|\langle f\Vert\sum_n \tau_n^+ f_{+1}(E_e,r)j_1(p_\nu r)\{\sigma_n\otimes\hat r_n\}_2\Vert i\rangle\right|^2 \right) \tag*{(24)}

and,

Bp3/2=12gA2×(∣⟨f∥∑nτn+g−2(Ee,r)j0(pνr){σn⊗r^n}2∥i⟩∣2+∣⟨f∥∑nτn+f+2(Ee,r)j0(pνr){σn⊗r^n}2∥i⟩∣2).(25)B^{p_{3/2}} = \frac{1}{2}g_A^2 \times \left( \left|\langle f\Vert\sum_n \tau_n^+ g_{-2}(E_e,r)j_0(p_\nu r)\{\sigma_n\otimes\hat r_n\}_2\Vert i\rangle\right|^2 + \left|\langle f\Vert\sum_n \tau_n^+ f_{+2}(E_e,r)j_0(p_\nu r)\{\sigma_n\otimes\hat r_n\}_2\Vert i\rangle\right|^2 \right). \tag*{(25)}

Here, GFG_F is the Fermi constant and VudV_{ud} is the element of the Cabbibo-Kobayashi-Maskawa (CKM) matrix. E0E_0 is maximal endpoint energy (in the case of zero neutrino mass) of the electron. pν=(E0−Ee)2−mk2p_\nu= \sqrt{(E_0-E_e)^2-m_k^2} is the neutrino momentum and θ(x)\theta(x) is a theta (step) function. gAg_A denotes an axial-vector coupling constant. ∣f⟩\lvert f\rangle is the initial (final) state of 187Re^{187}\mathrm{Re} (187Os^{187}\mathrm{Os}) with Jπ=5/2+J^\pi=5/2^+ (Jπ=1/2−J^\pi=1/2^-). rn\mathbf r_n (r=∣r∣r=|\mathbf r| and r^=r/r\hat r=\mathbf r/r) is a coordinate of the nn-th nucleon.

The differential decay rate in Eq. (23) depends on four different squared matrix elements incorporated in the Bs1/2B^{s_{1/2}} and Bp1/2B^{p_{1/2}} terms. For the purpose of factorization of calculation of squared nuclear matrix element and phase-space factor, the large- and small-component electron radial functions are approximated as follows [6]:

g−1(Ee,r)≃g−1(Ee,R)j0(peR)j0(per)≃g−1(Ee,R),g_{-1}(E_e,r) \simeq\frac{g_{-1}(E_e,R)}{j_0(p_eR)}j_0(p_er) \simeq g_{-1}(E_e,R),
f+1(Ee,r)≃f+1(Ee,R)j0(peR)j0(per)≃f+1(Ee,R),f_{+1}(E_e,r) \simeq\frac{f_{+1}(E_e,R)}{j_0(p_eR)}j_0(p_er) \simeq f_{+1}(E_e,R),
g−2(Ee,r)≃g−2(Ee,R)j1(peR)j1(per)≃rRg−2(Ee,R),g_{-2}(E_e,r) \simeq\frac{g_{-2}(E_e,R)}{j_1(p_eR)}j_1(p_er) \simeq\frac{r}{R}g_{-2}(E_e,R),
f+2(Ee,r)≃f+2(Ee,R)j1(peR)j1(per)≃rRf+2(Ee,R).(26)f_{+2}(E_e,r) \simeq\frac{f_{+2}(E_e,R)}{j_1(p_eR)}j_1(p_er) \simeq\frac{r}{R}f_{+2}(E_e,R). \tag*{(26)}

where RR is a nuclear radius. For bound states, required in the atomic exchange correction (see § 4), the same approximation holds leading to gn,−1(r)≃gn,−1(R)g_{n,-1}(r) \simeq g_{n,-1}(R), fn,1(r)≃fn,1(R)f_{n,1}(r) \simeq f_{n,1}(R), gn,−2(r)≃(r/R)gn,−2(R)g_{n,-2}(r) \simeq(r/R)g_{n,-2}(R) and fn,2(r)≃(r/R)fn,2(R)f_{n,2}(r) \simeq(r/R)f_{n,2}(R). For continuum states, only the leading terms in the expansion of spherical Bessel functions j0(per)j_0(p_{\mathrm{e}}r) and j1(per)j_1(p_{\mathrm{e}}r) were considered. In this way one ends up with the energy distribution, which depends only on a single squared matrix element [6]:

dΓdEe=dΓp3/2dEe+dΓs1/2dEe=∑k=13∣Uek∣2GF2Vud22π3BR2peEe(E0−Ee)×13[F1(Z,Ee)pe2+F0(Z,Ee)((E0−Ee)2−mk2)]×(E0−Ee)2−mk2 θ(E0−Ee−mk),(27)\begin{aligned} \frac{d\Gamma}{dE_{\mathrm{e}}} = \frac{d\Gamma^{p_{3/2}}}{dE_{\mathrm{e}}} + \frac{d\Gamma^{s_{1/2}}}{dE_{\mathrm{e}}} \\ &= \sum_{k=1}^{3} |U_{\mathrm{e}k}|^2 \frac{G_{\mathrm{F}}^2 V_{\mathrm{ud}}^2}{2\pi^3} B R^2 p_{\mathrm{e}} E_{\mathrm{e}} (E_0-E_{\mathrm{e}}) \\ &\times\frac{1}{3}\left[F_1(Z,E_{\mathrm{e}})p_{\mathrm{e}}^2 + F_0(Z,E_{\mathrm{e}})\left((E_0-E_{\mathrm{e}})^2-m_k^2\right)\right] \\ &\times\sqrt{(E_0-E_{\mathrm{e}})^2-m_k^2}\ \theta(E_0-E_{\mathrm{e}}-m_k), \tag*{(27)} \end{aligned}

with

B=gA26R2∣⟨f∥∑nτn+{σn⊗rn}2∥i⟩∣2,(28)B = \frac{g_{\mathrm{A}}^2}{6R^2}\left|\left\langle f\left\|\sum_n \tau_n^+ \{\boldsymbol{\sigma}_n \otimes\boldsymbol{r}_n\}_2\right\|i\right\rangle\right|^2, \tag*{(28)}

and

F0(Z,Ee)=g−1(Ee)g−1(Ee)+f1(Ee)f1(Ee)j0(peR)j0(peR),F_0(Z,E_{\mathrm{e}}) = \frac{g_{-1}(E_{\mathrm{e}})g_{-1}(E_{\mathrm{e}})+f_1(E_{\mathrm{e}})f_1(E_{\mathrm{e}})}{j_0(p_{\mathrm{e}}R)j_0(p_{\mathrm{e}}R)},
F1(Z,Ee)=g−2(Ee)g−2(Ee)+f2(Ee)f2(Ee)j1(peR)j1(peR).(29)F_1(Z,E_{\mathrm{e}}) = \frac{g_{-2}(E_{\mathrm{e}})g_{-2}(E_{\mathrm{e}})+f_2(E_{\mathrm{e}})f_2(E_{\mathrm{e}})}{j_1(p_{\mathrm{e}}R)j_1(p_{\mathrm{e}}R)}. \tag*{(29)}

Here, gκ(Ee)≡gκ(Ee,R)g_\kappa(E_{\mathrm{e}}) \equiv g_\kappa(E_{\mathrm{e}},R) and fκ(Ee)≡fκ(Ee,R)f_\kappa(E_{\mathrm{e}}) \equiv f_\kappa(E_{\mathrm{e}},R). In the limit the Coulomb interaction is switched-off F0(Z,Ee)=1F_0(Z,E_{\mathrm{e}})=1 and F1(Z,Ee)=1F_1(Z,E_{\mathrm{e}})=1. Different approximation schemes, for the electron wave functions, are indicated in the Fermi functions as Fk−1I(Z,Ee)F^I_{k-1}(Z,E_{\mathrm{e}}), where I = A, B, C or D.

Exchange correction

Besides the screening correction, another prominent atomic correction, significant for low Q-value β\beta transitions, is the exchange effect. It arises from creating a β\beta electron in a bound orbital of the final atom corresponding to one occupied in the initial atom. An atomic electron from the bound orbital simultaneously makes a transition to a continuum state of the final atom [32, 33]. We have extended the exchange effect formalism for allowed β\beta transitions, presented in [32, 33], for the unique first forbidden β\beta transitions. We found that two components of the exchange correction independently modify the spectra associated with the emission of electrons in the s1/2s_{1/2} and p3/2p_{3/2} states,

dΓs1/2dEe⇒dΓs1/2dEe×(1+η1T(Ee))\frac{d\Gamma^{s_{1/2}}}{dE_{\mathrm{e}}} \Rightarrow\frac{d\Gamma^{s_{1/2}}}{dE_{\mathrm{e}}}\times(1+\eta^T_1(E_{\mathrm{e}}))
dΓp3/2dEe⇒dΓp3/2dEe×(1+η2T(Ee))(30)\frac{d\Gamma^{p_{3/2}}}{dE_{\mathrm{e}}} \Rightarrow\frac{d\Gamma^{p_{3/2}}}{dE_{\mathrm{e}}}\times(1+\eta^T_2(E_{\mathrm{e}})) \tag*{(30)}

The result is consistent with the one obtained in [43], for the unique first forbidden β\beta-decay of 85Kr^{85}\mathrm{Kr}. The total exchange correction for each partial spectrum is given by,

ηκT(Ee)=fκ(2T−κ+T−κ2)+(1−fκ)(2T+κ+T+κ2)=η−κ(Ee)+η+κ(Ee)(31)\begin{aligned} \eta^T_\kappa(E_{\mathrm{e}}) = f_\kappa(2T_{-\kappa}+T_{-\kappa}^2)+(1-f_\kappa)(2T_{+\kappa}+T_{+\kappa}^2) \\ &= \eta_{-\kappa}(E_{\mathrm{e}})+\eta_{+\kappa}(E_{\mathrm{e}}) \tag*{(31)} \end{aligned}

where k=∣κ∣k=|\kappa| can take the values 1 or 2. Here,

fk=g−k′2(Ee,R)g−k′2(Ee,R)+f+k′2(Ee,R),(32)f_k = \frac{g_{-k}^{\prime2}(E_{\mathrm{e}},R)}{g_{-k}^{\prime2}(E_{\mathrm{e}},R)+f_{+k}^{\prime2}(E_{\mathrm{e}},R)}, \tag*{(32)}

and the dimensionless quantities TκT_\kappa depend on the overlaps between the bound states of the initial atom and the continuum states of the final atom with energy EeE_{\mathrm{e}},

Tκ=∑(nκ)′Tnκ=−∑(nκ)′⟨ψEeκ′∣ψnκ⟩⟨ψnκ′∣ψnκ⟩gn,κ′(R)gκ′(Ee,R),(33)T_\kappa= \sum_{(n\kappa)'}T_{n\kappa}=-\sum_{(n\kappa)'}\frac{\langle\psi'_{E_{\mathrm{e}}\kappa}|\psi_{n\kappa}\rangle}{\langle\psi'_{n\kappa}|\psi_{n\kappa}\rangle}\frac{g'_{n,\kappa}(R)}{g'_\kappa(E_{\mathrm{e}},R)}, \tag*{(33)}

for electrons in s1/2s_{1/2} (κ=−1\kappa=-1) and p3/2p_{3/2} (κ=−2\kappa=-2) states, and

Tκ=∑(nκ)′Tnκ=−∑(nκ)′⟨ψEeκ′∣ψnκ⟩⟨ψnκ′∣ψnκ⟩fn,κ′(R)fκ′(Ee,R),(34)T_\kappa= \sum_{(n\kappa)'}T_{n\kappa}=-\sum_{(n\kappa)'}\frac{\langle\psi'_{E_{\mathrm{e}}\kappa}|\psi_{n\kappa}\rangle}{\langle\psi'_{n\kappa}|\psi_{n\kappa}\rangle}\frac{f'_{n,\kappa}(R)}{f'_\kappa(E_{\mathrm{e}},R)}, \tag*{(34)}

for electrons in p1/2p_{1/2} (κ=+1\kappa=+1) and d3/2d_{3/2} (κ=+2\kappa=+2) states. All primed continuum and bound states refer to the final atom. The sums are running over all occupied orbitals of the final atom, which, in the sudden approximation, correspond to the parent electronic configuration.

Taking into account the sums inside the TκT_\kappa quantities, we can write

ηκ(Ee)=∑nηnκ+f∣κ∣∑n,mn≠mTnκTmκ(35)\eta_\kappa(E_{\mathrm{e}})=\sum_n\eta_{n\kappa}+f_{|\kappa|}\sum_{\substack{n,m\\n\ne m}}T_{n\kappa}T_{m\kappa} \tag*{(35)}

for negative values of κ\kappa, and

ηκ(Ee)=∑nηnκ+(1−fκ)∑n,mn≠mTnκTmκ(36)\eta_\kappa(E_{\mathrm{e}})=\sum_n\eta_{n\kappa}+(1-f_\kappa)\sum_{\substack{n,m\\n\ne m}}T_{n\kappa}T_{m\kappa} \tag*{(36)}

for positive values of κ\kappa. In this way, we can define the partial exchange correction contributions, ηnκ\eta_{n\kappa}, given by

ηnκ=f∣κ∣(2Tnκ+Tnκ2)(37)\eta_{n\kappa}=f_{|\kappa|}(2T_{n\kappa}+T_{n\kappa}^2) \tag*{(37)}

for negative values of κ\kappa, and

ηnκ=(1−fκ)(2Tnκ+Tnκ2),(38)\eta_{n\kappa}=(1-f_\kappa)(2T_{n\kappa}+T_{n\kappa}^2), \tag*{(38)}

for positive values of κ\kappa. The essential quantities in computing the exchange correction are the overlaps between the continuum state electron wave function, with energy EeE_{\mathrm{e}}, in the final atom and the bound orbitals electron wave functions in the initial atom, i.e., ⟨ψEeκ′∣ψnκ⟩\langle\psi'_{E_{\mathrm{e}}\kappa}|\psi_{n\kappa}\rangle. This overlap is given explicitly by

⟨ψEeκ′∣ψnκ⟩=∫0∞r2gκ′(Ee,r)gn,κ(r) dr+∫0∞r2fκ′(Ee,r)fn,κ(r) dr.(39)\begin{aligned} \langle\psi'_{E_{\mathrm{e}}\kappa}|\psi_{n\kappa}\rangle= \int_0^\infty r^2 g'_\kappa(E_{\mathrm{e}},r)g_{n,\kappa}(r)\,dr \\ &+\int_0^\infty r^2 f'_\kappa(E_{\mathrm{e}},r)f_{n,\kappa}(r)\,dr. \tag*{(39)} \end{aligned}

and its numerical calculation requires good knowledge of the continuum wave function over a wide region of space, from the nuclear center to where the bound wave function for the initial atom ends.

In the calculation of the exchange correction, the most important condition is to ensure that the final-state electron continuum wave function is orthogonal to the wave functions of the final-state bound orbitals, i.e., ⟨ψEeκ′∣ψnκ′⟩=0\langle\psi'_{E_e\kappa}|\psi'_{n\kappa}\rangle=0. They are eigenfunctions of the same Hamiltonian, so they must be orthogonal. A non-zero value of ⟨ψEeκ′∣ψnκ′⟩\langle\psi'_{E_e\kappa}|\psi'_{n\kappa}\rangle may lead to significant error in ⟨ψEeκ′∣ψnκ⟩\langle\psi'_{E_e\kappa}|\psi_{n\kappa}\rangle [32]. The implications in the exchange correction of non-orthogonal continuum and bound final-states wave functions were discussed in detail in [34]. Our chose of a modified DHFS self-consistent method for bound states and of the potential presented in Eq. (17) for continuum states, imposes automatically the orthogonality condition, ⟨ψEeκ′∣ψnκ′⟩=0\langle\psi'_{E_e\kappa}|\psi'_{n\kappa}\rangle=0, without any further numerically expensive orthogonalization procedures.

In the case of 187Re^{187}\mathrm{Re}, there is an abundance of possible switches between the β\beta emitted electrons and the bound electrons from the atomic cloud. For the emission of s1/2s_{1/2}-state electrons, ten orbitals contribute to η1T\eta_1^{\mathrm{T}}, and for the dominant emission in p3/2p_{3/2}-state, seven orbitals contribute to η2T\eta_2^{\mathrm{T}}. In Figure 3, we present all partial exchange corrections for the β\beta decay of 187Re^{187}\mathrm{Re}, i.e., Eqs. (37) and (38), as functions of the kinetic energy of the emitted electrons, i.e., Ee−meE_e-m_e. We note that the contributions from p1/2p_{1/2} (top right panel) and d3/2d_{3/2} (bottom right panel) states are much smaller than the ones from s1/2s_{1/2} (top left panel) and p3/2p_{3/2} (bottom left panel) states because they are associated with the small components of the Fermi functions f1(Ee)f_1(E_e) and f2(Ee)f_2(E_e), respectively (see Eq. (29)). They can not be neglected for a precise calculation.

Figure 3. All partial exchange corrections for the decay of 187Re^{187}\mathrm{Re}, i.e., Eqs. (37) and (38). The results are presented for atomic electrons in s1/2s_{1/2} (κ=−1\kappa=-1), p3/2p_{3/2} (κ=−2\kappa=-2), p1/2p_{1/2} (κ=+1\kappa=+1) and d3/2d_{3/2} (κ=+2\kappa=+2) states.

The total exchange correction for s1/2s_{1/2} (p3/2p_{3/2}) electron spectrum of rhenium β\beta decay is depicted with a solid (dashed) line in Figure 4 as a function of the kinetic energy of the emitted electron. We can see that the exchange corrections decrease with the increasing energy of the emitted electron, starting from around 73% for s1/2s_{1/2} electrons and 37% for p3/2p_{3/2} electrons, at 2 eV kinetic energy. The exchange correction remains significant even in the QQ-value region, around 12% for s1/2s_{1/2} electrons and 5.5% for p3/2p_{3/2} electrons. Considering the large amplitude and the shape of the exchange correction, we expect that it will induce considerable modifications in both the decay rate and spectrum shape of the β\beta-decay of rhenium. Those modifications are discussed in what follows.

Total exchange correction for electrons emitted in $s_{1/2}$ and $p_{3/2}$ wave states

Figure 4. The total exchange correction for electrons emitted in s1/2s_{1/2} wave state, η1T(Ee)\eta^{\mathrm{T}}_1(E_e) (solid black line), and for electrons emitted in p3/2p_{3/2} wave state, η2T(Ee)\eta^{\mathrm{T}}_2(E_e) (dashed blue line).

Results and discussions

In Table Table II, we present the partial decay rates, Γs1/2\Gamma^{s_{1/2}} (second column) and Γp3/2\Gamma^{p_{3/2}} (fourth column), from which we excluded the square matrix element, BB. The partial decay rates, presented in MeV, have been obtained from the integration over the full energy range of Eq. (27). In each integration, we have used different approximation schemes (indicated in the first column) displayed in Section § 2 A. The last line corresponds to the approximation scheme D, where we also included the atomic exchange correction, presented in Section § 4. In what follows, this combination is indicated as D + ex. We consider the approximation scheme A, used in our previous investigation, as a reference for a percent deviation with the other approximation schemes. The deviations are presented in the third (fifth) column for the partial decay rate Γs1/2\Gamma^{s_{1/2}} (Γp3/2\Gamma^{p_{3/2}}). Using the experimental half-live, 4.33×10104.33 \times10^{10} y [44], we also present our prediction for the squared matrix elements in the last column of Table Table II.

As a general result, in all the approximation schemes, the ratios between p3/2p_{3/2}-state electron emission channel and the s1/2s_{1/2}-state electron emission one are always around 10410^4, so different corrections do not change this particular feature of the rhenium decay. Still, in comparison with scheme A, there are considerable differences in the decay rates due to the screening and exchange corrections. There is a 24% (18.5%) decrease in the partial decay rate Γp3/2\Gamma^{p_{3/2}} (Γs1/2\Gamma^{s_{1/2}}) due to screening correction alone (scheme D). If we also include the exchange correction

w. f.1041B×Γs1/2\frac{10^{41}}{B}\times\Gamma^{s_{1/2}} [MeV]δs1/2\delta^{s_{1/2}} %1037B×Γp3/2\frac{10^{37}}{B}\times\Gamma^{p_{3/2}} [MeV]δp3/2\delta^{p_{3/2}} %B×104B\times10^4
A9.30-9.19-3.63
B8.33-10.418.92-2.953.74
C7.88-15.238.88-3.353.76
D7.58-18.486.98-24.024.78
D+ex9.461.757.92-13.844.22

Table 2. The partial decay rates for 187^{187}Re, excluding the squared matrix element, associated with the emission of electrons in s1/2s_{1/2} wave state (second column) and p3/2p_{3/2} wave state (fourth column). The approximation scheme for the electron wave functions is indicated in the first column. In the last line, the addition of the exchange correction over the scheme D is indicated as D + ex. In the third and fifth column, we present the decay rate percent deviation between the scheme A and the other schemes, δs1/2=100(ΓAs1/2−ΓXs1/2)/ΓXs1/2\delta^{s_{1/2}} = 100(\Gamma^{s_{1/2}}_{\mathrm{A}}-\Gamma^{s_{1/2}}_{\mathrm{X}})/\Gamma^{s_{1/2}}_{\mathrm{X}}, and δp3/2=100(ΓAp3/2−ΓXp3/2)/ΓXp3/2\delta^{p_{3/2}} = 100(\Gamma^{p_{3/2}}_{\mathrm{A}}-\Gamma^{p_{3/2}}_{\mathrm{X}})/\Gamma^{p_{3/2}}_{\mathrm{X}}, respectively, where X can be B, C, D or D + ex. The last column presents the experimental squared matrix elements.

(scheme D + ex), the decay rate Γs1/2\Gamma^{s_{1/2}} returns close to the one from scheme A (just a 1.7% increase). It is not the case of Γp3/2\Gamma^{p_{3/2}}, which in scheme D + ex, is 13.8% lower than the value from scheme A.

The most striking modification due to the exchange correction is the change in the spectrum shape of the rhenium decay. In Figure 5, we plot the normalized single electron spectrum for both s1/2s_{1/2}- and p3/2p_{3/2}-state emissions, using the approximation schemes D and D + ex. We mention that there are no considerable shape differences between approximation schemes A, B, C, and D. In the lower portion of Figure 5, we also present the residuals between spectra. Although the dominant p3/2p_{3/2} spectrum is not as strongly influenced by the exchange correction as the s1/2s_{1/2} spectrum, the modification is big enough to change the shape of the total spectrum of the decay.

Figure 5. The single electron differential decay rate normalized to the particular decay rate (Γs1/2\Gamma^{s_{1/2}} and Γp3/2\Gamma^{p_{3/2}}) for the emission of s1/2s_{1/2} and p3/2p_{3/2} electrons as functions of the electron kinetic energy Ee−meE_e-m_e for β\beta-decay of 187^{187}Re. We indicate with D the approximation scheme D and with ex the atomic exchange correction. The lower portion of the figure gives the difference between the spectra with and without the exchange correction for s1/2s_{1/2} wave state electrons with a dashed line and p3/2p_{3/2} wave state electrons with a solid line. Spectra are normalized over the full energy range.

We present the total spectrum of rhenium β\beta decay in Figure 6 with the same convention as in Figure 5, but now the electron kinetic energy takes values from 700 keV up to the QQ-value, and the residuals are in percentages. Note that the spectra are normalized to unity over the full range of the kinetic energy. One can see a different shape of the total electron spectrum of the rhenium decay when the exchange correction is switched on. The considerable spectrum shape modification induced by the exchange correction indicates that this correction should be included in the neutrino mass investigations from rhenium decay, particularly, and other low QQ-value β\beta transitions, in general.

Figure 6. The differential decay rate normalized to the total decay rate (Γ\Gamma) as function of the electron kinetic energy Ee−meE_e-m_e for β\beta decay of 187^{187}Re. We indicate with D the approximation scheme D and with ex the atomic exchange correction. The lower portion of the figure gives the percentage residuals between the spectrum with and without the exchange correction. Spectra are normalized over the full energy range.

To evaluate the deviation from an allowed spectrum, we write the β\beta spectrum of 187^{187}Re as,

dΓdEe=GF2Vud22π3BpeEeF0I(Z,Ee)(E0−Ee)2AFI(40)\frac{d\Gamma}{dE_e} = \frac{G_{\mathrm{F}}^2 V_{ud}^2}{2\pi^3} B p_e E_e F_0^I(Z,E_e)(E_0-E_e)^2 A_{\mathrm{F}}^I \tag*{(40)}

where, for the moment, the neutrino masses are neglected and I = A, B, C, or D. The shape factor AFI=1A_{\mathrm{F}}^I = 1 for the allowed transitions, but for unique first forbidden,

AFI=R23[peF1I(Z,Ee)F0I(Z,Ee)+(E0−Ee)2],(41)A_{\mathrm{F}}^I = \frac{R^2}{3}\left[p_e\frac{F_1^I(Z,E_e)}{F_0^I(Z,E_e)}+(E_0-E_e)^2\right], \tag*{(41)}

where we did not include the exchange with bound electrons. If we want to take into account this effect, the shape factor becomes,

AFD+ex=R23{peF1D(Z,Ee)F0D(Z,Ee)(1+η2T(Ee))+(E0−Ee)2(1+η1T(Ee))}.(42)A_{\mathrm{F}}^{\mathrm{D+ex}} = \frac{R^2}{3}\left\{p_e\frac{F_1^{\mathrm{D}}(Z,E_e)}{F_0^{\mathrm{D}}(Z,E_e)}(1+\eta_2^{\mathrm{T}}(E_e))+(E_0-E_e)^2(1+\eta_1^{\mathrm{T}}(E_e))\right\}. \tag*{(42)}

We can see from Figure 7, that all AFIA_{\mathrm{F}}^I are linear increase with energy in the experimentally available range, from 700 eV to QQ-value. We reproduce perfectly AFIA_{\mathrm{F}}^I if we consider the following model fit,

Figure 7. The shape factor AFIA_{\mathrm{F}}^I, i.e. Eq. (41), for different approximation schemes, I = A, B, C and D, and the shape factor AFD+exA_{\mathrm{F}}^{\mathrm{D+ex}}, i.e. Eq. (42).

AFI=aI(1+b1Te+b2Te2)(43)A_{\mathrm{F}}^I = a^I(1+b_1T_e+b_2T_e^2) \tag*{(43)}

with Te=Ee−meT_e=E_e-m_e in eV. The best fit parameters are b1=1.50×10−5 eV−1b_1=1.50\times10^{-5}\,\mathrm{eV}^{-1} and b2=4.82×10−11 eV−2b_2=4.82\times10^{-11}\,\mathrm{eV}^{-2}. The parameters aIa^I simply scale through different approximations of the Fermi functions (for example, aA=1.44×10−5a^{\mathrm{A}}=1.44\times10^{-5}), but they do not change the linear increasing behavior of AFIA_{\mathrm{F}}^I. In contrast, AFD+exA_{\mathrm{F}}^{\mathrm{D+ex}} decreases

with energy, and its shape is more complex due to the exchange correction. The best fit in the range from 700 eV to the QQ-value is obtained with the following assumption,

AFD+ex=aD+ex(b−1Te+1+b1Te+b2Te2),(44)A_{\mathrm{F}}^{\mathrm{D+ex}}=a^{\mathrm{D+ex}}\left(\frac{b_{-1}}{T_e}+1+b_1T_e+b_2T_e^2\right), \tag*{(44)}

where we found the best fit parameters b−1=19.95b_{-1}=19.95 eV, b1=−6.80×10−6 eV−1b_1=-6.80\times10^{-6}\,\mathrm{eV}^{-1} and b2=3.05×10−9 eV−2b_2=3.05\times10^{-9}\,\mathrm{eV}^{-2} with aD+ex=1.46×10−5a^{\mathrm{D+ex}}=1.46\times10^{-5}. Important to notice is the sign change of b1b_1 when the exchange correction is included.

The choice of a fit model with terms proportional to Te−1T_e^{-1} is based on the experimental shape factors that can also involve such terms [45].

The best upper limit on effective neutrino mass, mβ≤0.8m_\beta\leq0.8 eV, recently reported by the KATRIN experiment [16], holds in the degenerate neutrino mass region, i.e., m1≃m2≃m3≃mβ=∑k=13∣Uek∣2mkm_1 \simeq m_2 \simeq m_3 \simeq m_\beta= \sum_{k=1}^{3} \lvert U_{ek}\rvert^2 m_k. So, we replace all the neutrino masses mkm_k (k=1,2,3k = 1, 2, 3) with the effective neutrino mass mβm_\beta. For the following discussion we consider just the approximation scheme D for the electron wave functions and we also include the atomic exchange correction.

Kurie plots for different effective neutrino masses

Figure 8. The Kurie plots in arbitrary units (a.u.) for the β\beta decay of 187Re{}^{187}\mathrm{Re} with different values of the effective neutrino mass: mβ=0.0m_\beta= 0.0, 0.2, 0.4, 0.6, and 0.8 eV. The Q-value considered is 2470.9 eV [8].

The Kurie functions for the unique first forbidden transitions is given by

K(Ee,mβ)=dΓ/dEepeEe(peR)2F1(Z,Ee)(1+η2T(Ee))=GFVudB6π3(E0−Ee)1−mβ2(E0−Ee)2×[1+pνpeF0(Z,Ee)F1(Z,Ee)1+η1T(Ee)1+η2(Ee)]1/2.(45)\begin{aligned} K(E_e,m_\beta) &= \sqrt{\frac{d\Gamma/dE_e}{p_eE_e(p_eR)^2F_1(Z,E_e)(1+\eta_2^{T}(E_e))}} \\ &= G_FV_{ud}\sqrt{\frac{B}{6\pi^3}}(E_0-E_e)\sqrt{1-\frac{m_\beta^2}{(E_0-E_e)^2}} \\ &\quad\times\left[1+\frac{p_\nu}{p_e}\frac{F_0(Z,E_e)}{F_1(Z,E_e)}\frac{1+\eta_1^{T}(E_e)}{1+\eta_2(E_e)}\right]^{1/2}. \tag*{(45)} \end{aligned}

In the case of mβ=0m_\beta= 0, the Kurie plot, which is a plot of K(Ee,mβ)K(E_e,m_\beta) versus EeE_e, is ensured to be linear by keeping the exchange correction for p3/2p_{3/2} wave state electrons in the denominator of its definition. The term that can change the linear behavior is the last term in squared brackets, but its deviation from unity is very small. This deviation is below 6×10−56 \times10^{-5} after 1000 eV kinetic energy, and in the region of interest from 2300 eV to the Q-value, the deviation is even lower under 10−610^{-6}.

Conclusions

The distortion in the endpoint measurements of low QQ-value β\beta decay spectra is a direct tool to measure the values of neutrino masses. Current experiments are based on the ground-state to ground-state β\beta transitions, and for next-generation investigations, some suitable candidates could be the ultra-low QQ value (under 1 keV) ground-state-to-excited-state β\beta transitions. With the increasing interest in measuring neutrino masses from β\beta decays, accurate theoretical spectra predictions should be provided. However, their description can be challenging considering the β\beta transition classification and the plurality of atomic effects in the low-energy region.

We investigate the ground-state to ground-state unique first forbidden β\beta transition from 187Re(5/2+){}^{187}\mathrm{Re}(5/2^+) to 187Os(1/2−){}^{187}\mathrm{Os}(1/2^-), considering all relevant corrections to its spectrum and decay rate. In addition to our previous paper, we incorporate corrections for finite nuclear size, diffuse nuclear surface, screening, and exchange in the rhenium β\beta decay model. The last two effects are calculated using the self-consistent Dirac–Hartree–Fock–Slater description for the atomic bound electrons of the final atom. As the rhenium β\beta emission involves a mixture of s1/2s_{1/2}-state and p3/2p_{3/2}-state electrons, our exchange correction calculation accounts for all possible contributions from exchanges with s1/2s_{1/2}, p3/2p_{3/2}, p1/2p_{1/2}, and d3/2d_{3/2} bound orbitals. Our findings reveal significant alterations in the partial decay rates for both s1/2s_{1/2}- and p3/2p_{3/2}-state emission channels due to the screening and exchange effects while maintaining the experimentally confirmed dominance of p3/2p_{3/2}-state emission.

The key feature of this paper is that, in addition to changing the partial decay rates, the exchange correction modifies the shape of the total electron spectrum for rhenium β\beta decay. By studying the deviations from an allowed spectrum, we demonstrate that calculations with and without the exchange effect lead to entirely different shape factors, transitioning from an increasing linear behavior to a decreasing quadratic one. We present the best-fit parameters for both cases. The extent of this shape modification is so pronounced that we need to include the exchange correction in the definition of the Kurie plot to maintain its linearity in the case of zero effective neutrino mass. We also demonstrate how different effective neutrino masses affect the Kurie plots near the endpoint of the β\beta decay of 187Re{}^{187}\mathrm{Re}. Our conclusion underscores the importance of considering atomic effects, especially the exchange effect, in current and future neutrino mass scale investigations from β\beta decays.

Acknowledgments

F.Š. acknowledges support by the Slovak Research and Development Agency under Contract No. APVV-22-0413, and by the Ministry of Education, Youth and Sports of Czechia under the INAFYM Grant No. CZ.02.1.01/0.0/0.0/16 019/0000766. O.N. acknowledges support support of the Romanian Ministry of Research, Innovation, and Digitalization through Project CNCS-UEFISCDI No. 99/2021 within PN-III-P4-ID-PCE-2020-237 and the project CNCS-UEFISCDI No. TE12/2021 within PN-III-P1-1.1-TE-2021-0343.

The figures for this article have been created using the SciDraw scientific figure preparation system [46].

References

  1. [1]Aghanim, N. et al. (Planck Collaboration), Planck 2018 results - VI. Cosmological parameters, A&A 641, A6 (2020).DOI
  2. [2]F. T. Avignone, S. R. Elliott, and J. Engel, Double beta decay, majorana neutrinos, and neutrino mass, Rev. Mod. Phys. 80, 481 (2008).DOI
  3. [3]H. Ejiri, J. Suhonen, and K. Zuber, Neutrino-nuclear responses for astro-neutrinos, single beta decays and double beta decays, Physics Reports 797, 1 (2019).
  4. [4]F. Šimkovic, Neutrino masses and interactions and neutrino experiments in the laboratory, Physics-Uspekhi 64, 1238 (2021).
  5. [5]E. Ferri, D. Bäginaini, M. Biasotti, G. Ceruti, D. Corsini, M. Faverzani, F. Gatti, A. Giachero, C. Gotti, C. K. Kilbourne, A. Kling, M. Maino, P. Manfrinetti, A. Nucciotti, G. Pessina, G. Pizzigoni, M. R. Gomes, and M. Sisti, The status of the mare experiment with 187Re and 163Ho isotopes, Physics Procedia 61, 227 (2015), 13th International Conference on Topics in Astroparticle and Underground Physics, TAUP 2013.DOI
  6. [6]R. Dvornický, K. Muto, F. Šimkovic, and A. Faessler, Absolute mass of neutrinos and the first unique forbidden β decay of 187Re, Phys. Rev. C 83, 045502 (2011).DOI
  7. [7]E. G. Myers, A. Wagner, H. Kracke, and B. A. Wesson, Atomic masses of tritium and helium-3, Phys. Rev. Lett. 114, 013003 (2015).
  8. [8]P. Filianin, C. Lyu, M. Door, K. Blaum, W. J. Huang, M. Haverkort, P. Indelicato, C. H. Keitel, K. Kromer, D. Lange, Y. N. Novikov, A. Rischka, R. X. Schüssler, C. Schweiger, S. Sturm, S. Ulmer, Z. Harman, and S. Eliseev, Direct q-value determination of the β− decay of 187Re, Phys. Rev. Lett. 127, 072502 (2021).
  9. [9]M. Faverzani et al., The HOLMES experiment, Journal of Low Temperature Physics 184, 10.1007/s10909-016-1540-x (2016).
  10. [10]M. P. Croce et al., Development of holmium-163 electron-capture spectroscopy with transition-edge sensors, Journal of Low Temperature Physics 184, 10.1007/s10909-015-1451-2 (2016).DOI
  11. [11]L. Gastaldo et al., The electron capture in 163Ho experiment – ECHo, The European Physical Journal Special Topics 226, 10.1140/epjst/e2017-70071-y (2017).
  12. [12]S. Eliseev, K. Blaum, M. Block, S. Chennarev, H. Dorrer, C. E. Düllmann, C. Enss, P. E. Filianin, L. Gastaldo, M. Goncharov, U. Köster, F. Lautenschläger, Y. N. Novikov, A. Rischka, R. X. Schüssler, L. Schweikhard, and A. Türler, Direct measurement of the mass difference of 163Ho and 163Dy solves the q-value puzzle for the neutrino mass determination, Phys. Rev. Lett. 115, 062501 (2015).
  13. [13]N. D. Gamage, R. Bhandari, M. Horana Gamage, R. Sandler, and M. Redshaw, Identification and investigation of possible ultra-low Q value β decay candidates, Hyperfine Interactions 240, 10.1007/s10751-019-1588-5 (2019).
  14. [14]M. Redshaw, Precise q value determinations for forbidden and low energy β-decays using penning trap mass spectrometry, The European Physical Journal A 59, 18 (2023).DOI
  15. [15]D. K. Kebleck, R. Bhandari, N. D. Gamage, M. Horana Gamage, K. G. Leach, X. Mougeot, and M. Redshaw, Updated evaluation of potential ultralow q-value β-decay candidates, Phys. Rev. C 107, 015504 (2023).DOI
  16. [16]M. Aker, A. Beglarian, J. Behrens, A. Berlev, U. Besserer, F. Bieringer, B. Block, et al. (The KATRIN Collaboration), Direct neutrino-mass measurement with sub-electronvolt sensitivity, Nature Physics 18, 160 (2022).
  17. [17]V. N. Aseev, A. I. Belyaev, A. I. Berlev, E. V. Geraskin, A. A. Golubev, N. A. Likhovid, V. M. Lobashev, A. A. Nozik, V. S. Pantuev, V. I. Parfenov, A. K. Skasyrskaya, F. V. Tkachov, and S. V. Zadorozhny, Upper limit on the electron antineutrino mass from the tritium experiment, Phys. Rev. D 84, 112003 (2011).DOI
  18. [18]C. P. Kraus et al., Final results from phase ii of the mainz neutrino mass search in tritium βdecay, The European Physical Journal C - Particles and Fields 40, 10.1140/epjc/s2005-02139-7 (2005).
  19. [19]A. Ashtari Esfahani, S. Böser, N. Buzinsky, M. C. Carmona-Benitez, C. Claessens, L. de Viveiros, P. J. Doe, M. Fertl, J. A. Formaggio, J. K. Gaison, L. Gladstone, M. Grando, M. Guigue, J. Hartse, K. M. Heeger, X. Huyan, J. Johnston, A. M. Jones, K. Kazkaz, B. H. LaRocque, M. Li, A. Lindman, E. Machado, A. Marsteller, C. Matthé, R. Mohiuddin, B. Monreal, P. Mueller, J. A. Nikkel, E. Novitski, N. S. Oblath, J. I. Peña, W. Pettus, R. Reimann, R. G. H. Robertson, D. Rosa De Jesús, G. Rybka, L. Saldaña, M. Schram, P. L. Slocum, J. Stachurska, Y.-H. Sun, P. T. Surukuchi, J. R. Tedeschi, A. B. Telles, F. Thomas, M. Thomas, L. A. Thorne, T. Thümmel, L. Tvrznikova, W. Van De Pontseele, B. A. VanDevender, J. Weintroub, T. E. Weiss, T. Wendler, A. Young, E. Zayas, and A. Ziegler (Project 8 Collaboration), Tritium beta spectrum measurement and neutrino mass limit from cyclotron radiation emission spectroscopy, Phys. Rev. Lett. 131, 102502 (2023).
  20. [20]F. Gatti, F. Fontanelli, M. Galeazzi, A. M. Swift, and S. Vitale, Detection of environmental fine structure in the low-energy β-decay spectrum of 187Re, Nature 397, 137 (1999).
  21. [21]C. Arnaboldi, G. Benedek, C. Brofferio, S. Capelli, F. Capozzi, O. Cremonesi, A. Filipponi, E. Fiorini, A. Giuliani, A. Monfardini, A. Nucciotti, M. Pavan, M. Pedretti, G. Pessina, S. Pirro, E. Previtali, and M. Sisti, Measurement of the p to s wave branching ratio of 187Re β decay from beta environmental fine structure, Phys. Rev. Lett. 96, 042503 (2006).
  22. [22]A. Nucciotti, The MARE project, Journal of Low Temperature Physics 151, 597 (2008).
  23. [23]J. A. Formaggio, A. L. C. de Goûvêa, and R. H. Robertson, Direct measurements of neutrino mass, Physics Reports 914, 1 (2021), direct measurements of neutrino mass.
  24. [24]D. Wilkinson, Small terms in the beta-decay spectrum of tritium, Nuclear Physics A 526, 131 (1991).DOI
  25. [25]M. Kleesiek, J. Behrens, D. Drexlin, K. Eitel, M. Erhard, J. A. Formaggio, F. Glück, S. Groh, M. Hötzel, S. Mertens, A. W. P. Poon, C. Weinheimer, and K. Valerius, β-decay spectrum, response function and statistical model for neutrino mass measurements with the katrin experiment, The European Physical Journal C 79, 204 (2019).
  26. [26]M. E. Rose, Relativistic Electron Theory (John Wiley and Sons, 1961).
  27. [27]F. Salvat and J. M. Fernández-Varea, RADIAL: A fortran subroutine package for the solution of the radial Schrödinger and Dirac wave equations, Computer Physics Communications 240, 165 (2019).
  28. [28]M. Doi, T. Kotani, and E. Takasugi, Double Beta Decay and Majorana Neutrino, Progress of Theoretical Physics Supplement 83, 1 (1985), https://academic.oup.com/ptps/article-pdf/doi/10.1143/PTPS.83.1/5220783/ptps83-1.pdf.
  29. [29]V. B. Berestetskij, E. M. Lifshitz, and L. P. Pitaevskij, Quantum Electrodynamics (Nauka, 1989).
  30. [30]B. Hahn, D. G. Ravenhall, and R. Hofstadter, High-energy electron scattering and the charge distributions of selected nuclei, Phys. Rev. 101, 1131 (1956).
  31. [31]J. C. Slater, A simplification of the hartree-fock method, Phys. Rev. 81, 385 (1951).
  32. [32]M. R. Harston and N. C. Pyper, Exchange effects in β decays of many-electron atoms, Phys. Rev. A 45, 6282 (1992).
  33. [33]N. C. Pyper and M. R. Harston, Atomic effects on β-decay, Proc. R. Soc. Lond. A 420, 277 (1998).
  34. [34]O. Niţescu, S. Stoica, and F. Şimkovic, Exchange correction for allowed β decay, Phys. Rev. C 107, 025501 (2023).
  35. [35]M. E. Rose, Elementary Theory of Angular Momentum (Dover, 1995).
  36. [36]D. A. Varshalovich, A. N. Moskalev, and V. K. Khersonskii, Quantum Theory of Angular Momentum (World Scientific, 1995).
  37. [37]R. Latter, Atomic energy levels for the thomas-fermi and thomas-fermi-dirac potential, Phys. Rev. 99, 510 (1955).
  38. [38]D. Liberman, D. Cromer, and J. Waber, Relativistic self-consistent field program for atoms and ions, Computer Physics Communications 2, 107 (1971).
  39. [39]D. Liberman, J. T. Waber, and D. T. Cromer, Self-consistent-field dirac-slater wave functions for atoms and ions, i. comparison with previous calculations, Phys. Rev. 137, A27 (1965).
  40. [40]G. Moliere, Theorie der streuung schneller geladener teilchen i. einzelstreuung am einzelnen coulombfeld, Zeitschrift für Naturforschung A 2, 133 (1947).
  41. [41]C. Froese Fischer, G. Gaigalas, P. Jönsson, and J. Bieron, Grasp2018—a fortran 95 version of the grasp atomic structure package, Computer Physics Communications 237, 184 (2019).
  42. [42]T. A. Carlson, Photoelectron and Auger Spectroscopy (Plenum Press, 1975).
  43. [43]S. J. Haselschwardt, J. Kostensalo, X. Mougeot, and J. Suhonen, Improved calculations of β decay backgrounds to new physics in liquid xenon detectors, Phys. Rev. C 102, 065501 (2020).
  44. [44]M. Basunia, Nuclear data sheets for A = 187, Nuclear Data Sheets 110, 999 (2009).
  45. [45]X. Mougeot, Reliability of usual assumptions in the calculation of β and ν spectra, Phys. Rev. C 91, 055504 (2015).
  46. [46]M. Caprio, Levelscheme: A level scheme drawing and scientific figure preparation system for mathematica, Computer Physics Communications 171, 107 (2005).

Paper details

vlnc:W_zw42rgje
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 11

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