WorksPhysics and Astronomy
Nuclear matrix elements for neutrinoless double beta decays and spin-dipole giant resonances
Abstract
Nuclear matrix elements (NMEs) for neutrinoless double beta decays (DBDs) are required for studying neutrino physics beyond the standard model by DBD. The experimental spin-dipole (SD) giant resonance energy and the SD strength are shown for the first time to be closely related to the DBD-NME, and are used for studying the spin-isospin correlation and the quenching of the axial-vector coupling. So they are used to help the theoretical model calculation of the DBD-NME.
Neutrinoless double beta decay (DBD), which violates the lepton-number conservation law, is a sensitive and realistic probe for studying the neutrino () nature (Majorana or Dirac) relevant to the origin of matter in the universe, the absolute -mass scale, and other -properties beyond the standard model [1, 2, 3]. The nuclear matrix element (NME) for the neutrinoless DBD is crucial to extract the effective -mass and other -properties of the particle physics interests from the decay rate. The NME is also needed to design DBD detectors [4, 5, 6]. Thus the NMEs are of great interest from astro-, particle- and nuclear-physics view points. Theoretical works on DBD-NMEs by using various nuclear models are discussed in [7, 8, 9, 10, 11, 12], and recent DBD experiments in [13, 14, 15].
Actually, accurate theoretical calculations for the DBD-NME are very hard since they are sensitive to nucleonic and non-nucleonic correlations and nuclear medium effects. Consequently, calculated DBD-NMEs, including the effective axial-vector coupling (), scatter over an order of magnitude [3, 6], depending on the nuclear models, the interaction parameters and the effective coupling () used in the models. Thus experimental inputs are useful to check the theoretical models and the nuclear parameters to be used for the models [1, 6].
The present letter aims to show for the first time that the experimental energy and the strength of the spin-dipole (SD) giant-resonance in the intermediate nucleus are closely related to the DBD-NME () based on the pnQRPA (proton-neutron Quasi-particle Random Phase Approximation) model, and reflect the nuclear structure and the quenching of the , which are involved in the DBD-NME. Here the SD strength is given as with being the SD NME. We note that is associated with the SD component of the DBD-NME, which is one of the major components of the DBD-NME, and the quenching of is one of key parameters for the theoretical model calculation. Thus experimental information on the SD giant-resonance energy and the SD strength are used to help and check the theoretical model calculation of the DBD-NME.
Recently, the experimental values were shown to depend on the isospin -component with and being the neutron and proton numbers [6], and pnQRPA calculations for were performed by adjusting the particle-hole parameter to the in [12], where is assumed. Single- SD NMEs for the ground states in the medium-heavy nuclei (mostly non-DBD nuclei) were studied in the framework of the pnQRPA, and were found to be reduced by , depending much on the individual states [16].
The present paper puts emphasis on studying and discussing universal features on these , the GT and SD strengths, and the as a function of the mass number . For , we use a common value for the as derived by referring to the experimental summed GT strengths for the DBD nuclei and the SD NMEs for the ground state transitions in the medium-heavy nuclei. Then we discuss impact of the present findings on the DBD experiments.
We discuss double decays of the ground-state-to-ground-state transition of . The neutrinoless DBD () for is schematically shown in Figure 1. The intermediate nucleus is . The process to be discussed is the Majorana -mass process of the current interest. Here, a light Majorana is virtually exchanged between two neutrons in the DBD nucleus. DBD nuclei discussed are medium-heavy (

Figure 1. DBD transition scheme for with the Majorana exchange in the intermediate nucleus . and : Isospin raising and lowering operators. and : and K-mode NMEs associated with the K-mode DBD NME. : Intermediate state. : Charge-exchange reaction. 2QP: Two-quasiparticle state. GR: Giant resonance. DGR: Double giant resonance. See text.
nuclei with the high -mass sensitivity [14].
The DBD rate is expressed as [1, 3, 6]
where is the phase space factor, is the axial-vector coupling for a free nucleon in units of the vector coupling of , is the effective -mass and is the DBD NME. The NME is given as [3, 6, 7, 12]
where , and are the axial-vector (GT: Gamow-Teller), tensor (T) and vector (F: Fermi) DBD-NMEs, respectively, and is the effective axial-vector coupling introduced to incorporate the quenching effect. DBDs involve mainly axial-vector spin-isospin () and vector isospin () transitions. The NMEs depend much on nucleonic and non-nucleonic and interactions and their correlations. The pnQRPA model, which includes explicitly the nucleonic interactions and their correlations, has been widely used. Then the NMEs with are given by the pnQRPA model NMEs, and stands for the re-normalization (quenching) coefficient due to the non-nucleonic correlations, nuclear medium effects and others which are not explicitly included in the pnQRPA model. Then depends on the model to be used and the experimental data to be compared with, as discussed in [4, 6, 16, 17, 18, 19, 20]. In particular, the present pnQRPA is the one which is standardly used in the beta and double-beta calculations, and it is explained in great detail in [21]. The for a free nucleon is assumed in Eq. (2).
The NME is given by the sum of the NMEs for the intermediate states as [3, 6, 7, 12]
where the transition operator with and are given by , , and . The operator includes the neutrino potential with being the distance between the two neutrons involved in the exchange and being the excitation energy of the intermediate state , is the spin tensor operator and is the isospin operators for proton neutron.
The momentum of the virtual- is of the order MeV/, the NME involves mainly intermediate states in the wide ranges of the spins of – and -30 MeV. Thus, the DBD-NME may reflect gross properties of the nuclear core.
The axial-vector SD () component of the in Eq. (2) is one of the major components since the orbital angular momentum matches the medium momentum of the virtual neutrino. The SD DBD-NME is associated with the product of the single SD NMEs of and , as shown in Figure 1. The single- SD NMEs are expressed as with being the spherical harmonics.
On the other hand, the two-neutrino DBD within the standard model is followed by the emission of the two real s-wave () neutrinos with low momentum, and the NME () involves exclusively the for GT intermediate states at the low excitation-energy region. Thus is very sensitive to properties of the valence-nucleons at the proton neutron Fermi surfaces. So far, most theoretical DBD models use these valence-nucleon GT properties.
Recent experimental studies of DBD-NMEs are in [6, 15]. Among them the charge-exchange reactions provide useful information on with , , and so on for DBD nuclei with - in a wide excitation-energy and momentum regions, which are relevant to DBD-NMEs [19, 22, 23, 24, 25, 26, 27, 28, 29, 30]. We discuss the experimental results on the DBD nuclei of , , , , , , and . As an example of the reactions, Figure 2 shows the spectrum [26].

Figure 2. The energy spectrum of the reaction on [26]. F, GT, and SD are Fermi , GT and SD resonances, respectively. The energy scale is enlarged to make the sharp peaks visible. The and giant resonances are at the higher energy region of SD. 2QP: Two-quasiparticle states. Yields at the emission-angles of 0-0.5, 0.5-1, 1-1.5, 1.5-2, 2-2.5, and 2.5-3, each in deg, are shown by red, yellow, pink, light blue, blue, and green, respectively. F and GT transitions are enhanced at the forward angles (red, yellow), while the SD ones at larger angles (blue, green).
The pnQRPA calculations for [12] were extended to study relation of the to the experimental EGR(SD) and the GT and SD strength distributions. The calculations were made for and by referring to the experimental data [6, 16, 19, 31] (see (v)). The modified single-particle energies so as to reproduce the observed energies for low-lying states were used for all nuclei, except for where the level energies based on the Woods-Saxon potential were used. The particle-particle interaction parameter is divided in isoscalar and isovector parts in order to recover the isospin symmetry (vanishing of the two-neutrino DBD Fermi NME) and to reproduce the observed [20]. The dependence of is visualized, e.g., in [32] The particle-hole interaction derived so as to reproduce the experimental EGR(SD) in [6] was used for all multipole transitions except for the GT one, for which the that fits the experimental EG(GT) was used. Actually, we first adjust using . Then using the adjusted , we adjust to fit the GR energy. This is a common way to get the pp and ph interaction parameters to achieve a procedure which is the least prone to errors [20]. The parameter adjustments for the pnQRPA calculation are well described in [12].
Interesting universal features on the experimental SD data and the calculated NMEs are found as given below in (i)-(vi):
The GT strength and the SD strength are mostly concentrated, respectively, in the GT and SD giant-resonances at the high excitation-energy region in all nuclei (see Figure 2). Several sharp peaks in the low (0-6 MeV) energy region are two-quasiparticle states (2QP). The giant-resonance energies in units of MeV are found to be expressed as
as shown in Figure 3 A. They are given also as and [6]. Note that for the present nuclei. The and increase gradually as and increase. This is in accord with the GT and SD giant resonances in other nuclei [33].
Figure 3. (a): Experimental GT and SD giant-resonance energies of and . Solid lines: Fits by Eq. (4). Squares: The observed energies. Inverse triangles: The energies from the experimental MeV and MeV [6] as used in the pnQRPA. (b): The GT strengths in logarithmic scale. GT strength . S: for the nucleon-based sum-rule limit. T: for the observed total strength up to MeV. The strength beyond is corrected for the small contribution from the quasi-free scattering. 2QP: for the sum of the observed 2QP strengths up to 6 MeV. Straight lines are fits to the data.
The energies are found to be shifted higher above the GT and SD 2QP states by and . They increase with increase of and , which reflect the giant-resonance strengths. The SD giant-resonance energies are higher than the GT ones by around (harmonic oscillator energy) because of the jump involved in the SD excitation.
The giant-resonance energies depend on and , thus following the gross properties of nuclei.
The observed total GT strengths increase as and increase, and they are around 0.55 of the nucleon-based sum-rule limit of as shown in Figure 3 (b) [6]. These are consistent with GT strengths in other nuclei [33]. The reduction is incorporated by the quenched axial-vector coupling of . This may stand for such non-nucleonic effects that are not included in the simple sum-rule. The 2QP GT states are expected in the low excitation-energy region. The summed strength up to 6 MeV is around 10-5 % of the total strength of , and decrease as and increase (see Figure 3(b)).
The GT and SD cross sections for the () reactions show the maximum at the - and -wave scattering-angles of deg. and around 2 deg., respectively. The GT and SD cross sections there are the same within 10%. The SD cross section for the 2QP states below the SD giant resonance is also nearly same as the GT one below the GT giant resonance. The SD cross sections for the 2QP region are found to be also of the order of 10-5 % of the total SD cross sections. Thus the SD strengths at the giant resonance and the 2QP region show the similar trends as the GT strengths there. The 2QP GT and 2QP SD strengths for the very low excitation-energy (a few MeV) region depend on individual nuclei, reflecting the valence nucleon configurations containing single particle states near the proton and neutron Fermi surfaces of each nucleus.
TABLE I. The pnQRPA NMEs. and and (See Eq. (2)). a) and b) are the NMEs with and 0.55, respectively. Note that for the 0g-shell effect on the is huge, but looks little on the .
| a) | b) | a) | b) | ||
| -1.16 | 2.59 | 2.02 | 3.75 | 3.18 | |
| -1.03 | 2.12 | 1.29 | 3.14 | 2.31 | |
| -1.51 | 2.11 | 1.81 | 3.62 | 3.32 | |
| -1.01 | 2.03 | 1.43 | 3.03 | 2.44 | |
| -0.95 | 1.88 | 1.29 | 2.82 | 2.24 | |
| -0.81 | 1.57 | 1.08 | 2.37 | 1.89 | |
| -0.63 | 1.57 | 1.05 | 2.19 | 1.68 |
Table 1. TABLE I. The pnQRPA NMEs.
(iii). The pnQRPA NMEs of and for (see Eq. (2)) are shown in Table 1. They are found to decrease gradually as and increases, reflecting the gross properties of the nuclear core, as shown in Figure 4. This is in contrast to the SD and GT giant-resonance energies and the strengths, which increase as and (see Figure 3). The DBD-NMEs are found to behave like the 2QP GT and the 2QP SD strengths of (GT) and (SD) (see Figure 3 B). , which is associated with the GT NMEs for low-lying 2QP states, changes more than a factor 10 among the DBD nuclei, depending on the valence nucleon configurations of states near the Fermi surfaces [6, 34, 35, 36].
Figure 4. (blue squares) and (light blue square) with are plotted against (panel (a)) and (panel (b)). Solid lines are fits. See text.
(iv). The repulsive interaction pushes up the GT and SD strengths to the GT and SD giant resonances, and reduces the low-lying GT and SD NMEs with respect to the 2QP NMEs due to the nuclear polarization effects [4, 37, 38]. The reduction rate is given by with being the susceptibility. Likewise the DBD-NME may be doubly reduced by the factor . Then, the NME may be expressed as with and for the present DBD nuclei. As and increase, the susceptibility increases, and decreases. The value for is around 0.4 for DBD nuclei with around 16, and .
The particle-hole interaction () pushes up the giant resonances in energy, shifting the spin-isospin strengths from the low-lying states to the giant resonances, and reduces the DBD-NME . The susceptibility in (iv) is proportional to [4, 38]. Thus increases as increases and decreases as increases, 10% increase of leading to around 5% decrease of in accord with the pnQRPA calculations [12].
(v). We use for the present DBD NMEs and 0.55, which are suggested, respectively, by comparing the observed total GT strength of (GT) and the GT and SD strengths for low-lying states with those of the sum rule and the pnQRPA model [6, 16, 19, 30, 31]. The charge-exchange reaction data show the uniform quenching factor over the momentum region of 30-120 MeV/c of the DBD interest [6, 30]. The coupling of (i.e. ) is used in [12]. The quenching of the for is discussed in [6, 17, 19, 39], and that for GT and the in [6, 18, 40, 41, 42]. The similar quenching factor is obtained for the low-momentum GT NMEs from the chiral two-body current [43].
(vi). The NMEs with are smaller than those with by factors around 0.7. The reduction is less effective than 0.62 for the ratio of the . The difference is due to the dependence of the computed NME on the ratio and hence on . As decreases the magnitude of the experimental NME, derived from the experimental half-life, increases and hence also the magnitude of the computed NME has to increase. Since the magnitude of the computed increases with decreasing value of [32] the value of has to decrease. This, in turn, means that the NMEs are computed with different values of for different values of the ratio (larger ratio means larger value of ). behaves in a similar way in terms of as (less sensitively, though, since the higher multipoles are less dependent on the ratio than the multipole). The computed NMEs with are smaller by factors around 0.8 than those with .
Let us discuss the impact of the present findings on DBD studies. On the bases of the above discussions, we use . with this region corresponds approximately to the region around
The DBD NMEs for -136 are considered to be around 3-2. Since they do not depend much on individual nuclei, selection of the DBD nucleus for the experiment may be made from experimental requirements such as the availability of ton-scale DBD isotopes, the large value, the low-background and the high energy-resolution.
The present analyses show that the pnQRPA is closely related to the SD giant resonance and the distribution of the strength with , SD in the intermediate nucleus. Various nuclear models are being used for calculating . Then, it is interesting to compare the calculated strength distribution of in by using the model wave function for with the observed giant resonance and the strength. Actually, the SD () giant resonance is accompanied at the higher energy side by the resonances with , and the quasi-free scattering. Further experimental studies for their strengths and also for higher multipole strengths with and are interesting. The low and high multipole giant resonances are studied in the pnQRPA [44].
OMC (ordinary muon capture) of is used to study [6, 45]. The large strength is found in the giant resonances [45, 46]. The quenching of in the muon strengths is under discussions [47, 48]. Actually, strengths of are also studied by using various nuclear reactions [6, 14, 15].
The strengths in the intermediate nucleus are mostly in the giant resonances, and little strengths at the 2QP and ground states. Likewise, the DBD () strengths are considered to be mostly in the double giant resonances, and little DBD strengths at the 2QP and ground states (Figure 1). Double charge-exchange reactions on are used to study the DBD strengths in [6, 49]. The double GT and SD giant-resonance energies measured from the ground state of are around -32 MeV and -48 MeV for DBD nuclei with -136. The RCNP () data [50] show that most of the strengths are at the high excitation-energy region. The double GT strengths for the ground states are shown to have a positive correlation with the DGT-state DBD-NMEs [51].
The delta isobar () is strongly excited by the quark flip in a nucleon to form the giant resonance. Then, the axial-vector and the axial-vector with and T are reduced with respect to the model NMEs without the isobar effect, which is incorporated by using the effective coupling of [6, 52, 53, 54, 55]. Studies of the effects are interesting.
In conclusion, the present work shows for the first time (i) a clear relation between the experimental SD strength distribution in the intermediate nucleus and the DBD NME in the pnQRPA formalism based on the SD data, and (ii) a simple expression of as a function of on the basis of the pnQRPA with the derived experimentally. Thus such experimental inputs are useful for pinning down the values for the DBD NMEs.
References
- [1]H. Ejiri, J. Phys. Soc. Jpn. 74, 2101 (2005).
- [2]F. Avignone, S. Elliott, and J. Engel, Rev. Mod. Phys. 80, 481 (2008).
- [3]J. Vergados, H. Ejiri, and F. Šimkovic, Rep. Prog. Phys. 75, 106301 (2012).
- [4]H. Ejiri and J.I. Fujita, Phys. Rep. 38, 85 (1978).
- [5]H. Ejiri, Phys. Rep. 338, 265 (2000).
- [6]H. Ejiri, J. Suhonen and K. Zuber, Phys. Rep. 797, 1 (2019).
- [7]J. Suhonen and O. Civitarese, Phys. Rep. 300, 123 (1998).
- [8]A. Faessler and F. Simkovic, J. Phys. G Nucl. Part. Phys. 24, 2139 (1998).
- [9]J. Suhonen and O. Civitarese, J. Phys. G Nucl. Part Phys. 39, 085105 (2012).
- [10]J. Barea, J. Korita and F. Iachello, Phys. Rev. C 87, 014315 (2013).
- [11]J. Engel and J. Menéndez, Rep. Prog. Phys. 80, 046301 (2017).
- [12]J. Jokiniemi, H. Ejiri, D. Frekers, and J. Suhonen, Phys. Rev. C 98, 024608 (2018).
- [13]J. Detwiler, Proc. Neutrino conference 2020 (2020), doi:10.3389/fphy.2018.00160.
- [14]H. Ejiri, Universe, 6, 225 (2020).
- [15]H. Ejiri, Front. Phys. 9, 650421 (2021).
- [16]H. Ejiri, N. Soukouti, and J. Suhonen, Phys. Lett. B 729, 27 (2014).
- [17]J. Suhonen, Phys. Rev. C 96, 055501 (2017).
- [18]J.T. Suhonen, Front. Phys. 5, 55 (2017).
- [19]H. Ejiri, Front. Phys 7, 30 (2019).
- [20]F. Simkovic, V. Rodin, A. Faessler and P. Vogel, Phys. Rev. C 87 045501 (2013).
- [21]J. Suhonen, From Nucleons to Nucleus: Concepts of Microscopic Nuclear Theory, Springer, Berlin (2007).DOI
- [22]H. Akimune, et al., Phys. Lett. B 394 23 (1997), 665 424 (2011).
- [23]J.H. Thies, et al., Phys. Rev. C 86, 014304 (2012).
- [24]J. H. Thies et al., Phys. Rev. C 86, 054323 (2012).
- [25]P. Puppe et al., Phys. Rev. C 84, 051305R (2011).
- [26]J.H. Thies, et al., Phys. Rev. C 86, 044309 (2012).
- [27]P. Puppe, et al., Phys. Rev. C 86, 044603 (2012).
- [28]H. Ejiri and D. Frekers, J. Phys. G Part. Phys. 43, 11LT01 (2016).
- [29]H. Akimune, et al., J. Phys. G Nucl. Part Phys. 47, 05LT01 (2020).
- [30]H. Ejiri, J. Phys. G Nucl. Part Phys. 46, 125202 (2019).
- [31]H. Ejiri, J. Phys. G 42, 055201 (2015).
- [32]J. Suhonen, Physics letters B 607 87 (2005).
- [33]D.J. Horen et al., Phys. Lett. B 99, 383 (1981).
- [34]N. Yoshida and F. Iachello, Prog. Theor. Exp. Phys, 2013, 043D01 (2013).
- [35]P. Pirinen and J. Suhonen, Phys. Rev. C 91, 054309 (2015).
- [36]H. Ejiri, J. Phys. G 44, 115201 (2017).
- [37]A. Bohr and B. Mottelson, Nuclear Structure, II , W.A. Benjamin, INC., Amsterdam, (1975).
- [38]H. Ejiri and M. A. deVoigt, Electron and Gamma -ray Spectroscopy in Nuclear Physics, Oxford, (1988).
- [39]P. Gysbers, et al., Nat. Phys. 15, 428 (2019).
- [40]E. A. Coello Perez, J. Menéndez, and A. Schwenk, Phys. Rev. C 98, 045501 (2018).
- [41]F. Simkovic, R. Dvornicky, D. Stefanik, and A. Faessler, Phys. Rev. C 97, 034315 (2018).
- [42]L. Coraggio, L. De Angelis, T. Fukui, A. Gargano, N. Itaco, and F. Nowacki, Phys. Rev. C 100, 014316 (2019).
- [43]J. Menéndez, D. Gazit and A. Schwenk, Phys. Rev. Lett. 107 062501 (2011).
- [44]L. Jokiniemi and J. Suhonen, Phys. Rev. C 96, 034308 (2017).
- [45]H. Hashim, H. Ejiri, T. Shima, K. Takahisa, A. Sato, Y. Kuno, K. Ninomiya, N. Kawamura, and Y. Miyake, Phys. Rev. C 97, 014617 (2018).
- [46]L. Jokiniemi, J. Suhonen, H. Ejiri H, and I. Hashim, Phys. Lett. B 794, 143 (2019).
- [47]L. Jokiniemi and J. Suhonen. Phys. Rev. C 100, 014619 (2019).
- [48]F. Simkovic, R. Dvornieky, P. Vogel, Phys. Rev. C 102, 034301 (2020).
- [49]F. Cappuzzello, et al., Eur. Phys. J. 51, 145 (2015).
- [50]K. Takahisa, H. Ejiri, et al., arXiv 1703.08264 (2017).arxiv.org/abs/1703.08264
- [51]N. Shimizu, J. Menendez, K. Yako, Phys. Rev. Lett. 120 142502 (2018).
- [52]A. Bohr, A. and B. Mottelson, Phys. Lett. B 100, 10 (1981).
- [53]H. Ejiri, Phys. Rev. C 26, 2628 (1982).
- [54]G. Cattapan and L. S. Ferreira, Phys. Rep. 362, 303 (2002).
- [55]H. Ejiri, Few-body Syst. 62, 60 (2021).