Electroproduction of medium- and heavy-mass hypernuclei P. Bydžovský 1 , D. Denisova 1 , F. Knapp 2 , and P. Veselý 1 Abstract The DWIA formalism for computing the cross sections in electroproduction of hypernuclei used before for lighter ( p p -shell and s ​ d sd -shell) systems is utilized in studying electroproduction on heavier targets such as 52 Cr, and 208 Pb. First, the effects from kaon distortion and kinematics are re-examined and then results for the heavy targets are discussed in view of various forms of the Λ ​ N \Lambda N effective interaction, the elementary amplitude, and kinematics. Apparent sensitivity of the hypernucleus excitation spectra to various forms of the Λ ​ N \Lambda N effective interaction is shown. Predictions of the excitation spectrum in electroproduction of Λ 208 {}^{208}_{~~\Lambda} Tl are also given in kinematics of the experiment E12-20-013 in preparation at Jefferson Lab. I Introduction Investigation of hypernucleus properties provides significant information about the baryon-baryon interaction with a non zero strangeness. Processes involving production of hypernuclei play important role as they allow to study effective forms of the interaction. The measured excitation spectra and cross sections give information about details of the interaction. To this end, one should have a good understanding of the reaction mechanism to be able to draw firm conclusions from a data analysis. The hypernuclei have been already studied for several decades in various reactions GHM2016 ; HT2006 . The quality of obtained data depends on the reaction where the best precision was achieved in the measurement of (\pi^{+},K^{+}\gamma) HT2006 . These high-quality \gamma -ray data for p -shell hypernuclei allowed performing a detailed analysis of the spin-dependent part of the effective hyperon-nucleon (YNG) interaction JohnM . The hypernuclei were also studied in the (\pi^{+},K^{+}) reaction Pile ; HT2006 providing mainly information about the binding energies of \Lambda in various shells and the cross sections for production of the ground state. Reaction spectra with a better energy resolution than in (\pi^{+},K^{+}) were obtained from the (e,e^{\prime}K^{+}) reaction measured for the first time at Jefferson Lab HallCexp . Data from the successive measurements of electroproduction of p -shell hypernuclei performed at Jefferson Lab then allowed studying the YNG interaction archiv . Heavier hypernuclei beyond the p -shell were investigated in the past utilizing mainly the (\pi^{+},K^{+}) reaction Pile . However, new experiments for electroproduction of hypernuclei on the medium-mass, 40 Ca and 48 Ca (E12-15-008 E12-15-008 ), and heavy, 208 Pb (E12-20-013 E12-20-013 ), targets are planned at Jefferson Lab. Electroproduction of \Lambda hypernuclei provides not only a test ground for various effective YNG interactions but it also allows studying the reaction mechanism and the elementary amplitudes at very small kaon angles. The theoretical photoproduction excitation spectra obtained in the distorted-wave impulse approximation (DWIA) for the p -shell and medium-mass hypernuclei were discussed by Motoba et.al. in Refs. Motoba2010 and NPA881 . It was shown that the spectra for typical medium-mass hypernuclei such as {}^{28}_{~\Lambda} Al and {}^{40}_{~\Lambda} K provide interesting opportunities of spectroscopic study beyond the p shell. Another analysis of the cross sections in photoproduction of light ( {}^{12}_{~\Lambda} B) as well as very heavy ( {}^{208}_{~~\Lambda} Tl) hypernuclei was performed in Ref. MotobaJPS showing the wide applicability of the ( e,e^{\prime}K^{+} ) reaction. In our previous analysis HYP2018 ; fermi ; structure of hypernucleus electroproduction we studied the kinematical and Fermi motion effects in the case of p -shell hypernuclei ( {}^{12}_{~\Lambda} B, {}^{16}_{~\Lambda} N, {}^{9}_{\Lambda} Li) fermi and verified that the nucleus and hypernucleus structure of medium-mass systems ( {}^{28}_{~\Lambda} Al, {}^{40}_{~\Lambda} K) can be realistically described in the Tamm-Dancoff approach structure . Here we continue our study focusing on the medium- and heavy-mass hypernuclei and providing predictions for the experiments E12-15-008 and E12-20-013. The paper is organized as follows: in Section II we give a brief overview of basic formalism. More details can be found in Refs. fermi and structure . In Section III, the effects from kaon distortion, kinematics, and Fermi motion are discussed in the case of heavy hypernuclei. Here we also elaborate the optimum on-shell approximation introduced in Ref. fermi . The excitation spectra in electroproduction of medium- and heavy-mass hyperncuclei, {}^{40}_{~\Lambda} K, {}^{48}_{~\Lambda} K, {}^{52}_{~\Lambda} V, and {}^{208}_{~~\Lambda} Tl, are discussed in Sec. IV. The summary and conclusions from our study are given in Sec. V. II Basic formalism The cross section in electroproduction of hypernuclei is calculated in the impulse approximation (IA) considering the optimal factorization approximation where the amplitude for elementary production is evaluated for an effective proton momentum fermi . The general form of the two-component elementary amplitude, constructed in Ref. fermi , allows for an arbitrary value of this momentum. In the previous analysis we had suggested an optimum value of the effective proton momentum which fulfills the necessary kinematical conditions and the elementary amplitude is on-shell. In the many-particle matrix element, the photon plane wave and kaon distorted wave are decomposed into partial waves ( LM ) and the target proton and final \Lambda are supposed to occupy the single-particle states \alpha ( =n,l,j ) and \alpha^{\prime} , respectively. The reduced amplitude, used to compute the cross sections, then reads \displaystyle A^{\lambda}_{Jm}= \displaystyle\frac{1}{[J]}\,\sum_{S\eta}\,{\cal F}^{S}_{\lambda\eta}\,\sum_{LM}\;{\sf C}^{Jm}_{LMS\eta}\sum_{\alpha^{\prime}\alpha}{\cal R}^{LM}_{\alpha^{\prime}\alpha}\;{\cal H}^{LSJ}_{l^{\prime}j^{\prime}lj} (1) \displaystyle\times(\,\Phi_{H}\,||\,[b_{\alpha^{\prime}}^{+}\otimes a_{\alpha}]^{J}\,||\;\Phi_{A}\,)\,, where {\cal F}^{S}_{\lambda\eta} are spherical elementary amplitudes with a baryon spin-flip S=0,1 and photon helicity \lambda=\pm 1,0 . The nucleus( \Phi_{A} )-hypernucleus( \Phi_{H} ) transition is described by the radial integral {\cal R}^{LM}_{\alpha^{\prime}\alpha} and the one-body density matrix element (OBDME), where a_{\alpha} and b^{+}_{\alpha^{\prime}} are the proton annihilation and \Lambda particle creation operators. The term {\cal H}^{LSJ}_{l^{\prime}j^{\prime}lj} includes the Racah algebra and {\sf C}^{Jm}_{LMS\eta} is the Clebsch-Gordan coefficient. Derivation of this formula and expressions for the cross sections can be found in Ref. fermi . The radial integrals are explicitly {\cal R}^{LM}_{\alpha^{\prime}\alpha}=\int_{0}^{\infty}\!\!d\xi\,\xi^{2}\,{R}_{\alpha^{\prime}}(\xi)^{*}\,F_{LM}(\Delta B\xi)\,{R}_{\alpha}(\xi)\,, (2) where R_{\alpha^{\prime}} and R_{\alpha} are the \Lambda and proton radial wave functions, respectively, and F_{LM} comes from the partial-wave decomposition of the photon and kaon waves, see appendix C in Ref. fermi . This function depends on the momentum transfer \Delta=|\vec{P_{\gamma}}-\vec{P}_{K}| and parameter B=(A-1)/(A-1+m_{\Lambda}/m_{p}) . Note that in the case without a kaon distortion in PWIA the function F_{LM} corresponds to the spherical Bessel function. Kaon distortion is in our approach included by means of the eikonal approximation with the first-order optical potential. The kaon distorted wave then reads as DThesis \chi^{+}_{\rm K}(\vec{r}\,)={\sf exp}\left[-b\,\sigma^{tot}_{\rm KN}(s)\!\left(\!\frac{}{}1-i\alpha(s)\right)\!\!\int_{0}^{\infty}\!\!\!dt\,\rho(\vec{r}+\hat{p}\,t)\right], (3) where \sigma^{tot}_{\rm KN} is the kaon-nucleon (KN) total cross section and \alpha(s)={\rm Re}f_{\rm KN}(s,0)/{\rm Im}f_{\rm KN}(s,0) , with the KN amplitude at zero angle and invariant energy squared s . The parameter b stands for a kinematical factor DThesis and the integrand \rho(\vec{r}+\hat{p}\,t) describes the nucleon density where \hat{p} is a unit vector in direction of the relative kaon-hypernucleus momentum. The nucleus and hypernucleus structure included in OBDME, (\,\Phi_{H}\,||\,[b_{\alpha^{\prime}}^{+}\otimes a_{\alpha}]^{J}\,||\;\Phi_{A}\,) , and the radial single-particle wave functions R_{\alpha} are calculated using a many-particle formalism with specific NN and YN effective interactions. In the present analysis we use the Hartree-Fock (HF) and Tamm-Dancoff (TD Λ ) approaches Ves1 ; structure . In Refs. Ves1 ; structure the HF was solved for hypernuclear Hamiltonian which included two-body ( NN , N\Lambda ), and three-body ( NNN ) interactions. In this paper we do not include the three-body NNN interaction term. Instead, to mimic the many-nucleon interactions in the nuclear Hamiltonian, we complement the NN interaction with a phenomenological density-dependent (DD) term DDterm v_{\rho}=\frac{C_{\rho}}{6}(1+P_{\sigma})\rho\left(\frac{\vec{r}_{1}+\vec{r}_{2}}{2}\right)\delta(\vec{r}_{1}-\vec{r}_{2}), (4) where C_{\rho} is a coupling constant which plays role of a free parameter, P_{\sigma} is the operator of the spin exchange, and \rho is the nucleon density of the calculated nucleus. Applying the HF method for description of 40 Ca, 48 Ca, 52 Cr, and 208 Pb we obtain the proton and neutron single-particle states. The \Lambda single-particle states are obtained by solving the additional Schrödinger-like equation (see Eq. (4) in Ves1 ). In the case of 52 Cr, which consists of 24 protons and 28 neutrons, we solve the HF equations by using the approximation of partial occupation of the proton valence 0f_{7/2} level by 4 instead of 8 protons. The hypernuclei {}^{40}_{~\Lambda} K, {}^{48}_{~\Lambda} K, {}^{52}_{~\Lambda} V, and {}^{208}_{~~\Lambda} Tl are described within TD Λ approach Ves1 ; structure in which we diagonalize the hypernuclear Hamiltonian in the space of \Lambda -particle proton-hole excitations on top of the HF states of 40 Ca, 48 Ca, 52 Cr, and 208 Pb, respectively. The HF + TD Λ calculations are performed within a harmonic oscillator (HO) basis, which should be large enough in order to reach the convergence with respect to a HO frequency. It is sufficient to use the space up to N_{\rm{max}}=12 major shell. The nucleon-nucleon interaction is described by the Daejeon 16 Dae16-ref ; Dae16-code NN interaction complemented with the phenomenological DD term ( 4 ) which we denote as D16+DDT. The Daejeon 16 interaction itself is derived from the NN component of chiral N3LO potential N3LO , subsequently softened by a SRG transformation SRG with flow parameter \lambda = 1.5 fm -1 . The phenomenological DD term mimic the effects of the 3-body and higher-body nucleon interactions and its coupling constant C_{\rho} was tuned for each nucleus separately in order to obtain a realistic description of nuclear radii and nucleon single-particle energies. The N\Lambda interaction is described using various effective G-matrix potentials based on the YN interactions Nijmegen F YNG , Jülich A YNG , and chiral LO YNLO . The G-matrix derived from the Nijmegen F, and Jülich A interactions is parametrized as a sum of Gaussian-like terms V_{\Lambda N}=\sum^{3}_{i=1}(a_{i}+b_{i}k_{\rm{F}}+c_{i}k^{2}_{\rm{F}})\,\rm{exp}(-r^{2}/\beta^{2}_{i}), (5) which depend on the Fermi momentum k_{F} . The coefficients a_{i} , b_{i} , c_{i} , and \beta_{i} are given in Ref. YNG . The interaction matrix elements of the G-matrix derived from the chiral LO YN potential YNLO G_{ijkl}(\Omega)=\langle ij|\hat{G}(\Omega)|kl\rangle are obtained by solving a first approximation of the Bethe-Goldstone equation \displaystyle G_{ijkl}(\Omega)= \displaystyle V^{N\Lambda-N\Lambda}_{ijkl} (6) \displaystyle- \displaystyle\sum_{mn}V^{N\Lambda-N\Sigma}_{ijmn}\frac{1}{(\epsilon^{N}_{m}+\epsilon^{\Sigma}_{n})-\Omega}V^{N\Lambda-N\Sigma}_{klmn}\,, where \Omega is a parameter of the G-matrix, V^{N\Lambda-N\Lambda} ( V^{N\Lambda-N\Sigma} ) is the N\Lambda-N\Lambda ( N\Lambda-N\Sigma ) channel of the LO YN interaction, and \epsilon^{N}_{m} ( \epsilon^{\Sigma}_{m} ) are nucleon ( \Sigma ) single-particle energies. The sum over single-particle states in Eq. ( 6 ) runs only over unoccupied nucleon states. The single-particle energies \epsilon^{\Sigma}_{m} were obtained by using the N\Sigma-N\Sigma channel of the LO YN interaction in the equation analogous to that which we solve to obtain the single-particle levels of \Lambda (see Eq. (4) in Ves1 ). The parameter \Omega in Eq. ( 6 ) is defined as \Omega=E_{\rm{av}}-(m_{\Sigma^{0}}-m_{p})\,, (7) where m_{\Sigma^{0}} ( m_{p} ) is mass of \Sigma^{0} (proton), respectively, and E_{\rm{av}} is either a free parameter or we can determine it as an average energy of occupied nucleons, i.e. E_{\rm{av}}=\sum_{i}\epsilon^{N}_{i}v_{i} /A, where the coefficient v_{i}=1 for the occupied single-particle states, and v_{i}=0 otherwise. III Effects from kaon distortion and kinematics Before we will show results for the heavy hypernuclei we discuss in more detail effects from re-scattering of kaons in the final state and from shifting the hypernucleus mass due to an excitation energy. The latter was used in our approach archiv ; fermi ; structure but has not been discussed yet. As shown in Eq. ( 3 ) the kaon distorted wave depends on the nucleon density and kaon-nucleon ( KN ) scattering amplitude. For the latter we utilize a separable form constructed in Ref. mesons98 . This form was also used in description of kaon scattering on nuclei mesons95 ; CJP97 . Here and in the previous hypernucleus DWIA calculations archiv ; HYP2018 we have utilized an improved version of the separable KN amplitude which includes the partial waves up to l=7 and which can describe the available KN data up to the invariant energy of about 2.4 GeV. This is especially important for the hypernucleus calculations at photon energies of about 2 GeV archiv as other KN amplitudes, e.g. that in Ref martin , fail to describe the KN scattering at those energies. Note also that this new separable KN amplitude is isospin dependent and therefore the distortion differs for various targets according to a number of the protons and neutrons, e.g. in 40 Ca and 48 Ca. The nucleon density \rho in Eq. ( 3 ) is described by the Hartree-Fock (HF) form which is taken from the HF many-particle calculations consistently with OBDME. However, in our previous calculations for the p -shell hypernuclei archiv ; HYP2018 ; fermi ; structure we used a phenomenological parametrization motivated by the harmonic oscilator model (HO) DThesis \rho_{\rm HO}(r)=(a_{0}+a_{1}\alpha^{2}r^{2})\exp{(-\alpha^{2}r^{2})}\, (8) with the parameters a_{0} , a_{1} and \alpha depending on the nucleus mass number A and the HO parameter b_{\rm HO} DThesis . In next discussion we compare influence of the HF and HO forms of the nucleon density on the cross sections. Remind that in our approach the kaon distortion is included via the eikonal approximation with the first-order kaon-nucleus optical potential. In the case of light p-shell hypernuclei like {}^{12}_{~\Lambda} B the electroproduction cross section is decreased due to the kaon final-state re-scattering by 45% for the ground-state doublet with \Lambda in the s orbit and by 34% for the multiplet at about 11 MeV with \Lambda in the p orbit. In this case, the distortion is calculated using the HO density with b_{\rm HO}=1.621 fm -1 and the suppression of the cross sections for the multiplets depends only very weakly on the many-particle approach and the effective YN interaction, particularly in the considered case of the shell model with the p -shell optimized interaction JohnM ; archiv and the TD Λ approach with the Nijmengen F interacion structure . Figure 1: Effects from kaon distortion for {}^{40}_{\;\Lambda} K. The panel (a) shows radial nucleon densities from the harmonic-oscilator parametrization (HO) with b_{\rm HO}=1.939 fm -1 and the Hartree-Fock calculation (HF), both normalized to A = 40. The panel (b) compares the excitation spectra calculated in PWIA and DWIA. In the latter the kaon-nucleus optical potential is calculated using the HO and HF nucleon densities shown in (a). Calculations are performed in the Tamm-Dancoff (TD Λ ) formalism using the Nijmegen F YNG interaction with the Fermi momentum k_{F}= 1.30 fm -1 and the BS3 amplitude in the optimum on-shell approximation. The curves are plotted with FWHM = 800 keV. In Figure 1 we show effects of kaon distortion in the case of {}^{40}_{~\Lambda} K calculated with two different nucleon densities. The harmonic-oscilator (HO) and Hartree-Fock (HF) forms of densities are shown in part (a) where it is apparent that the HO density, calculated with b_{\rm HO}=1.939 fm -1 , is not a realistic description, mainly in the central region of the nucleus. However, as shown in part (b) the suppression of the excitation spectrum caused by the HO and HF density does not differ too much, suggesting that the effect is given mainly by a behavior of the density in the peripheral region of the nucleus whereas its behavior inside the nucleus is irrelevant for the kaon distortion accounted for by the eikonal approximation. Note that the mean nucleus radius calculated with the HF density is 3.07 fm and that with the HO density is 2.83 fm which do not differ too much. The failure of the HO form in describing the nucleon density in the central region of the nucleus can be attributed to the simple form considered in Eq. ( 8 ) which is inadequate for the sd -shell and still heavier nuclei. A detailed study of distortion effects in electroproduction of p - and sd -shell hypernuclei was also performed in Ref. DThesis . Similar effects are shown for a heavy hypernucleus {}^{208}_{\;\;\;\Lambda} Tl in Fig. 2 . Here we show distortion effects calculated with two distinct forms of HO with b_{\rm HO}=3.51 fm -1 and b_{\rm HO}=2.25 fm -1 . The latter, being an extremely unrealistic description, was chosen just to demonstrate that even such very bad description of the density in the central region has only mild effects in the excitation spectrum. The mean radius calculated with these HO densities are \langle R\rangle=5.22 fm and 3.36 fm, respectively and when we compare them with the HF radius \langle R_{\rm HF}\rangle=5.26 fm we see that the former agrees well with \langle R_{\rm HF}\rangle . As this former form of the HO density is also very close to the HF density in the peripheral region, see Fig. 2 (a), it is not surprising that the distortion effects are very similar to those of the HF density, see part (b). The version of HO with b_{\rm HO}=2.25 fm -1 differs quite a lot in the peripheral region and therefore the distortion effect is smaller as we see in Fig. 2 (b). We can conclude that the distortion effects amount to 40–60% for {}^{40}_{~\Lambda} K and 50–80% for {}^{208}_{~~\Lambda} Tl and that they depend on behavior of the nucleus density in the peripheral region rather than in the nuclear interior. Another observation is that the hypernucleus states with a more deeply bound \Lambda hyperon, e.g. in the s orbit, are more strongly affected with the kaon distortion than the states with \Lambda in higher lying orbits, e.g. d . Note also that the different strength of the kaon distortion for various states changes the shape of multiplets in the spectrum as observed in Fig. 2 . In our present calculations we use the HF densities which provide a more realistic optical potential and which are obtained in the many-particle calculations being consistent with the values of OBDMEs. Figure 2: The same as in Fig. 1 but for {}^{208}_{\;\;\;\Lambda} Tl. As we mentioned above, in determination of kinematics we use the hypernucleus mass shifted by excitation energy of given state: M^{*}_{H}=M_{H}^{0}+E^{*} where M_{H}^{0} and E^{*} are the hypernucleus ground-state mass and excitation energy, respectively. Even if this mass shift is relatively very small it can change values of some particle momenta even by a few percent and therefore the elementary amplitude and the radial integrals can notably change their values. In Table 1 we show how a relatively tiny shift of the mass propagates into variations of the momenta, the elementary amplitude, the radial integral, and finally how it changes the cross section. The results, calculated in PWIA at photon energy E_{\gamma}=1.5 GeV, are shown for two states of {}^{208}_{~~\Lambda} Tl with the same spin to see variations of the effect in dependence on E^{*} . Whereas changes of the momenta are proportional to the excitation energy, the radial integral reveals different effects and therefore the cross section can be less suppressed for the larger shift. The value of the radial integral is given by the overlap of the proton and \Lambda single-particle wave functions and the function proportional to the spherical Bessel function in PWIA, F_{LM}(\Delta B\xi)\sim j_{L}(\Delta B\xi) , see Eq. ( 2 ). A few percent change of \Delta can vary the overlap of the functions resulting in a quite different value of {\cal R}^{LM}_{\alpha^{\prime}\alpha} . It is also remarkable in the table that the particle momenta decrease due to the mass shift but the momentum transfer \Delta rises and that the proton momentum |\vec{p}_{opt}| significantly changes which then markedly affects the value of the elementary amplitude. The changes are also present in the transversal part (d \sigma_{T} ) of the full cross section (d \sigma ). Note that dependence of the cross sections on the momentum transfer \Delta was also studied in the DWIA calculations of photoproduction of {}^{208}_{\;\;\;\Lambda} Tl in Ref. MotobaJPS observing noticeable effects. states (11.964, 8^{-} ) (18.201, 8^{+} ) M_{H}^{*} 0.006 0.009 |\vec{P}_{K}| -1.0 -1.6 \Delta 3.0 4.6 |\vec{p}_{opt}| -15.7 -23.5 Re {\cal F}^{1}_{00} -1.2 -4.5 Im {\cal F}^{1}_{00} -31.4 -44.4 {\cal R}^{LM}_{\alpha^{\prime}\alpha} {\cal R}^{70}_{\alpha^{\prime}\alpha}:\,-3.7 {\cal R}^{80}_{\alpha^{\prime}\alpha}:\,20.2 d \sigma -16.0 -10.6 d \sigma_{T} -13.2 -9.6 Table 1: Changes, in % with respect to the ground-state values, are due to the mass shift for two states of {}^{208}_{~~\Lambda} Tl: ( E^{*} [MeV], J_{\rm H}^{P} ). The changes are shown for the kaon momentum |\vec{P}_{K}| , momentum transfer \Delta=|\vec{\Delta}| , proton optimum momentum |\vec{p}_{opt}| , the spin-flip elementary amplitude with zero projections and the relevant radial integrals for dominant transitions 0h_{11/2}\to 0d_{5/2} and 0h_{11/2}\to 0f_{5/2} . The full (d \sigma ) and transversal (d \sigma_{T} ) cross sections fermi decrease by 10–16%. Despite the fact that the cross sections for various states can change in different way, as one can see in Table 1 , the overall effect in the excitation spectrum is monotonously rising with the excitation energy as shown in Fig. 3 . Therefore, both the d_{\Lambda} and f_{\Lambda} peaks change similarly by about 16% even if the cross section for the state (18.201, 8^{+} ) changes only by 10.6%. This is because of dominance of the first state (11.964, 8^{-} ) in the d_{\Lambda} peak whereas the f_{\Lambda} peak is dominated by the (18.316. 9^{+} ) state with the cross section changed by 16.3%. Figure 3: The excitation spectra of {}^{208}_{\;\;\;\Lambda} Tl calculated in PWIA without and with the shift of hypernucleus mass due to the excitation energy. The calculations are for E_{\gamma}=1.5 GeV and \theta_{K\gamma}=8^{\circ} with the BS3 elementary amplitude in the optimum on-shell approximation. We can conclude that the effects from the hypernucleus-mass shift are quite important correction to the cross section of high-lying hypernucleus states. In the considered kinematical region the mass shift suppresses the cross sections. III.1 On-shell approximation In previous calculations we used the optimum on-shell approximation fermi with the optimum proton momentum calculated from the two-body energy conservation in the laboratory frame (the nucleus is in rest) \displaystyle E_{\gamma}+\sqrt{m_{p}^{2}+(\vec{p}_{opt})^{2}} \displaystyle= \displaystyle\sqrt{m_{\Lambda}^{2}+|\vec{\sf P}_{\sf K}|_{mb}^{2}}\ + (9) \displaystyle+ \displaystyle\sqrt{m_{\Lambda}^{2}+(\vec{\Delta}-\vec{p}_{opt})^{2}}\,, where m_{p} and m_{\Lambda} are the proton and \Lambda masses, respectively, E_{\gamma} is the photon energy, and |\vec{P}_{K}|_{mb} is magnitude of the kaon lab momentum computed from the many-body energy conservation. This equation assures that the elementary amplitude is on shell but we have to chose a value of the angle \theta_{\Delta p} between the proton momentum and the momentum transfer. Kinematics in the coplanar case ( \Phi_{K}=\Phi_{p} ) is depicted in Fig. 4 . In previous calculations fermi ; structure , we had chosen \theta_{\Delta p}=180^{\circ} as then both proton and \Lambda momentum received the minimum value at given kinematics. Figure 4: Kinematics of the elementary process in the coplanar case with the axis \hat{\sf y} oriented rearwards the page. The angle \theta_{\Delta p} between the proton optimum momentum \vec{p}_{opt} and the momentum transfer \vec{\Delta}=\vec{P}_{\gamma}-\vec{P}_{\rm K} has a general value. In the present calculations we elaborate this approach assuming that the magnitude of \vec{p}_{opt} equals the mean momentum of the proton in the single-particle state of the dominant transition. This mean value, p_{mean}=\sqrt{2\mu\,\langle\,T_{kin}\,\rangle} with the reduced mass \mu , is calculated for each hypernucleus state from the proton single-particle wave function that corresponds to the dominant value of OBDME. The mean value of the kinetic-energy operator \langle\;T_{kin}\,\rangle is evaluated directly from its definition differentiating numerically the radial wave functions. Having fixed the magnitude of the proton momentum in Eq. ( 9 ), |\vec{p}_{opt}|=p_{mean} , we can calculate the angle \theta_{\Delta p} and if it acquires a physical value, particularly \cos\theta_{\Delta p}\geq-1.0 , we can use this value of the proton momentum in the calculations. In the case the angle is nonphysical, \cos\theta_{\Delta p}<-1.0 , we can calculate |\vec{p}_{opt}| with \cos\theta_{\Delta p}=-1.0 as before, which is the minimal value at given kinematics, and because in this case p_{mean} is still smaller this value of |\vec{p}_{opt}| is a good substitution for p_{mean} . Since many hypernucleus states, especially the low lying states, have nonphysical values of \theta_{\Delta p} and the values of p_{mean} are in general close to |\vec{p}_{opt}| with \cos\theta_{\Delta p}=-1.0 the new and old variants of the optimum on-shell approximation do not differ too much. To summarize the variants of the optimum on-shell approximation: a) previous version: the angle is chosen as \theta_{\Delta p}=180^{\circ} and the magnitude of \vec{p}_{opt} calculated; b) elaborated version: the magnitude is specified as |\vec{p}_{opt}|=p_{mean} and the angle \theta_{\Delta p} calculated. If the angle is physical we use this otherwise we use the variant a). In the following we are going to use the variant b) and denote it as “on-shell approximation”. III.2 Fermi motion effects Dependence of the cross sections on various values of the proton effective momentum was shown and discussed for the p -shell hypernuclei in Ref. fermi . In Figures 5 and 6 we show these Fermi motion effects for several states ( E^{*} [MeV], J_{\rm H}^{P} ) of the heavy hypernucleus {}^{208}_{~~\Lambda} Tl in dependence on the kaon angle with respect to the electron beam \theta_{\rm Ke} and on the photon laboratory energy E_{\gamma} , respectively. Figure 5: Angular dependence of the cross sections for selected states ( E^{*} [MeV], J_{\rm H}^{P} ) in electroproduction of {}^{208}_{\;\;\;\Lambda} Tl. The results are shown for various values of the effective proton momentum, \vec{p}_{eff}=0 (frosen p), \vec{p}_{eff}=-\vec{\Delta} (frosen \Lambda ), and \vec{p}_{eff}=\vec{p}_{opt} (optimum). The calculations were performed with the elementary amplitude BS3 SB at E_{\gamma}=1.5 GeV. The biggest differences of the cross sections, mostly observed between the boundary cases of “frosen” proton and \Lambda , are at small kaon angles amounting about 20–30%. The energy dependence of the effects reveals a resonant-like structure in some cases and more pronounced effects are observed for energies below 2 GeV. Similarly as in the case of lighter hypernuclei one can see that the nature of the effects is very similar for the states 6 + , 9 - , and 8 + and for 7 + , 10 - , and 7 + , particularly the sequence of the curves is almost identical. The similarity of the effects inside these two groups of states is given by the selection rule discussed in Ref. fermi . This rule says that contributions from the longitudinal mode of the virtual photon ( \lambda=0 ) are important only for the states with specific spin and parity. Let us show this in the example: both states (5.794, 6 + ) and (5.804, 7 + ) are dominated by the proton \to\Lambda transition 0h\to 0p which means that L=6 in Eq. ( 1 ). As the leading contribution to the reduced amplitude with \lambda=0 comes from the spin-flip elementary amplitude with S=1 and \eta=0 and the biggest value of the radial integral is for M=0 the corresponding Clebsch-Gordan coefficients in Eq. ( 1 ) are {\sf C}^{60}_{6010}=0 and {\sf C}^{70}_{6010}=\sqrt{7/13} . The longitudinal contributions in the case of 7^{+} then make the frosen- \Lambda cross section mostly larger than that for the case with the stationary proton contrary to the case of the state 6^{+} . This longitudinal contribution also makes the pronounced resonant structures in the energy dependent cross sections in Fig. 6 . These observations demonstrate importance of the contributions from the longitudinal and transverse-longitudinal interference terms in the electroproduction cross section fermi . Figure 6: The same as in Fig. 5 but for dependence on the photon lab energy at \theta_{\rm Ke}=11^{\circ} . The results with the optimum momentum mostly lie within the limit cases with the frosen proton and \Lambda which is given by values of the proton and \Lambda momenta. For example, at E_{\gamma}=1.5 GeV and \theta_{Ke}=11^{\circ} the proton and \Lambda momenta for the state (5.804, 7 + ) are 0 and 392 MeV/c for the frosen proton, 292 and 0 MeV/c for the frosen \Lambda , and 207 and 162 MeV/c for the optimum proton momentum case, respectively. Note that the results in Figs. 5 and 6 are calculated with the elementary amplitude BS3 SB which has a more abundant energy structure than the Saclay-Lyon amplitude as discussed in Ref. fermi . IV Discussion of excitation spectra for medium- and heavy-mass hypernuclei IV.1 {}^{40}_{~\Lambda} K and {}^{48}_{~\Lambda} K The excitation spectra in electroproduction of {}^{40}_{~\Lambda} K and {}^{48}_{~\Lambda} K were already discussed in Ref. structure . It was shown how much the spectra depend on the Fermi momentum k_{F} , used to parametrize the YNG Nijmegen F interaction, and on the elementary amplitudes SLA and BS3. We have also compared the results obtained in two various approaches, namely in TD Λ and the Equation of Motion Phonon Method (EMPM Λ ). These results were obtained using the chiral NNLO sat NN+NNN potential NN-NNN and the kaon distortion was computed with the HO nucleus density and b_{\rm HO}=1.939 fm -1 structure . In the present paper we show results calculated in the TD Λ approach using the D16+DDT interaction with C_{\rho}=2000 MeV \cdot fm 6 in eq. ( 4 ) and the kaon distortion is included using the optical potential constructed with the HF densities of the target nuclei obtained in the HF approach. The upgraded results calculated with the BS3 amplitude in the on-shell approximation and with two YNG interactions, Jülich A (JA) and Nijmegen F (NF), are compared with the previous calculations in Figs. 7 and 8 . Note that the old results were calculated with k_{F}= 1.25 fm -1 and the new ones with 1.30 and 1.34 fm -1 for {}^{40}_{~\Lambda} K and {}^{48}_{~\Lambda} K, respectively. Differences in the NN interaction, the value of k_{F} and using the JA YNG interaction make changes in the excitation spectra where the production strength is re-distributed into diverse structures. However, the magnitudes of the peaks, which are sensitive especially to the elementary amplitude and kinematics, mostly do not differ too much for various interactions. The results presented in Figs. 7 and 8 demonstrate a variety of predictions for the experiment E12-15-008 in preparation at Jefferson Lab. Figure 7: The previous result for {}^{40}_{~\Lambda} K with the BS3 amplitude in the optimum on-shell approximation structure (NN+NF125) is compared with the upgraded calculations using the D16+DDT interaction and the NF (NNDaej-NF130) and JA (NNDaej-JA130) YNG interactions, both with k_{F}=1.30 fm -1 . The kaon distortion is included using the HF density of 40 Ca. The curves are plotted with FWHM = 800 keV. Figure 8: The same as in Fig. 7 but for {}^{48}_{~\Lambda} K and k_{F}=1.34 fm -1 . IV.2 {}^{52}_{~\Lambda} V We also provide predictions of the excitation spectrum in the reaction 52 Cr( e,e^{\prime}K^{+} ) {}^{52}_{~\Lambda} V and compare them with the older calculation in Ref. NPA881 . This older result presented in Ref. NPA881 , Fig. 7, is for photoproduction in kinematics E_{\gamma}=1.3 GeV and \theta_{K\gamma}=3^{\circ} and with the SLA amplitude in the frosen-proton approximation. The spectrum clearly shows separated multiplets with the \Lambda hyperon in the s , p , d , and f orbit. The major peaks are based on the conversion of protons in the 0f_{7/2} orbit into the \Lambda bound in the given l_{\Lambda} orbit. Note that the dominant unnatural-parity peaks ( 4^{-} , 5^{+} , 6^{-} , 7^{+} ) are with J_{max}=L_{max}+1 . The other well separated and important peaks are based on the proton hole d_{3/2}^{-1} NPA881 . Figure 9: Excitation spectra for electroproduction of {}^{52}_{~\Lambda} V calculated in the TD Λ approach with the Jülich A (JA) and Nijmegen F (NF) YNG interactions and with the elementary amplitude SLA in the frosen-proton approximation. The results presented here are for electroproduction in kinematics E_{i}=1.6 GeV, E_{f}=0.3 GeV, \theta_{e}=4.5^{\circ} , \theta_{Ke}=4.04^{\circ} , and \Phi=180^{\circ} giving photoproduction kinematics E_{\gamma}=1.3 GeV, Q^{2}=0.003 (GeV/c) 2 , \varepsilon=0.362 , \theta_{\gamma e}=1.04^{\circ} , and \theta_{K\gamma}=3.0^{\circ} , quite close to that in Ref. NPA881 . The kaon distortion is included using the HF density for 52 Cr and the OBDMEs are determined within the HF+TD Λ approach using the NN interaction D16+DDT with C_{\rho}=2000 MeV \cdot fm 6 in eq.( 4 ) and JA and NF YNG interactions with k_{F}=1.34 fm -1 . In Figure 9 we show results calculated with the SLA amplitude in the frosen-proton approximation ( p_{eff}=0 ) and the JA and NF YNG interactions, which can be compared with the results in Fig. 7 of Ref. NPA881 . One can observe that the main three peaks in the JA result are of similar magnitude and position as in Ref. NPA881 . In the new calculations, these main peaks are especially formed by the following states ( E^{*} [MeV], J_{H}^{P} ) dominantly populated by the single-particle transitions nl_{j}\to n^{\prime}l^{\prime}_{j^{\prime}} : (0.0 MeV, 4^{-} ) with 0f_{7/2}\to 0s_{1/2}\, and (0.078 MeV, 3^{-} ) with 0f_{7/2}\to 0s_{1/2}\, for the ground-state doublet; (9.497 MeV, 5^{+} ) with 0f_{7/2}\to 0p_{3/2}\, and (9.633 MeV, 4^{+} ) with 0f_{7/2}\to 0p_{3/2} and 0p_{1/2}\, for the second main peak; (18.181 MeV, 6^{-} ) with 0f_{7/2}\to 0d_{5/2}\, and (18.313 MeV, 5^{-} ) with 0f_{7/2}\to 0d_{3/2} and 0d_{5/2}\, for the third peak; and (24.619 MeV, 7^{+} ) with 0f_{7/2}\to 0f_{7/2}\, and (24.687 MeV, 6^{+} ) with 0f_{7/2}\to 0f_{7/2} and 0f_{5/2}\, for the fourth main peak. Exceptions are the states: (11.942 MeV, 1^{+} ) with dominant transition 1s_{1/2}\to 0s_{1/2}\, , (13.327 MeV, 2^{+} ) with 0d_{3/2}\to 0s_{1/2}\, , (27.814 MeV, 4^{-} ) with 0d_{5/2}\to 0p_{3/2}\, , and (20.008 MeV, 3^{-} ) with 0d_{5/2}\to 0p_{3/2} and 0p_{1/2} . Moreover the part of the spectrum with \Lambda in f orbit reveals quite different distribution of the production strength than in Fig. 7 of Ref. NPA881 . Also the strength in between the main peaks, based on the proton hole in d^{-1}_{3/2} , is missing in our result based on the TD Λ approach. Possible explanation is that we would need to use an approach beyond TD Λ , which includes coupling to the nuclear core excitation, in order to obtain this strength. An example of such approach is EMPM Λ , see Ref. structure for more detailed discussion. Figure 10: The same as in Fig. 9 but with the BS3 amplitude in the on-shell approximation (JA-3 and NF-3). The result NF-3 is compared with previous “optimum on-shell approximation” (NF-2). In Figure 10 we compare the results calculated with the BS3 amplitude in the on-shell approximation for the JA and NF YNG interactions. The only difference of shapes of spectra in Figs. 9 and 10 for a given YNG interaction is in magnitudes of the peaks which depend on the elementary amplitude and kinematics (on-shell or frosen-proton approximation). Note that in the frosen-proton approximation the proton and \Lambda momenta are 0 and 362 MeV/c whereas in the on-shell approximation they are 181 and 105 MeV/c (for the ground state), respectively. This difference might seem small but it does play a role as shown in Figs. 5 and 6 and in Ref. fermi . In case of NF we also compare the new on-shell result (NF-3) with the previous optimum on-shell approximation (NF-2) used before in Refs. fermi and structure . As we have already mentioned the difference between these spectra is almost negligible. This is because although the magnitudes of the cross sections for separate hypernucleus states differ by about 1% the sign of the differences for various states in a multiplet is scattered on both sides (positive/negative) and therefore the overall shape of the peaks composed from several states remains almost unchanged. The small difference of the cross sections can be attributed to a small difference between the proton momentum p_{opt} with \cos\theta_{\Delta p}=-1 and p_{mean} for the proton single-particle wave function of the dominant transition, see Subsect. 3.1. The results in Fig. 10 demonstrate a measure of sensitivity of the spectra to using different forms of the effective YNG interactions in the DWIA calculations which is comparable to the width of the peaks (FWHM = 800 keV). We therefore presume that analysis of a good quality experimental spectrum obtained in the reaction 52 Cr( e,e^{\prime}K^{+} ) {}^{52}_{~\Lambda} V could help in a more precise determination of the spin-dependent part of the YNG interaction. IV.3 {}^{208}_{\;\;\;\Lambda} Tl In this subsection we discuss results for the reaction 208 Pb (e,e^{\prime}K^{+})^{208}_{\;\;\;\Lambda} Tl where both 208 Pb and {}^{208}_{\;\;\;\Lambda} Tl are very heavy systems with 126 neutrons which act as spectators in the impulse approximation still affecting kinematics of the reaction. The mass of the ground state of 208 Pb is taken to be M_{Pb}^{N}=193.68769 GeV and that of {}^{208}_{\;\;\;\Lambda} Tl is estimated to be M_{Tl}^{H}=M_{Pb}^{N}-m_{p}+m_{\Lambda}+\varepsilon_{p}-\varepsilon_{\Lambda}\doteq 193.8482 GeV considering that the difference of the binding energies is m_{\Lambda}-m_{p}+\varepsilon_{p}-\varepsilon_{\Lambda}\approx 160.5 MeV. Determining precise values of the masses is quite important, remind the discussion of sensitivity of the cross sections to the hypernucleus mass in Sect. 3. The cross sections of {}^{208}_{\;\;\;\Lambda} Tl are calculated in the HF+TD Λ approach using the NN interaction D16+DDT with C_{\rho}=4000 MeV \cdot fm 6 and various YNG interactions and elementary amplitudes. The aim is to show a variety of predictions for the planned experiment E12-20-013 at JLab E12-20-013 . The results are for the new kinematics in Hall C Hall C , E_{i}=2.24 GeV, E_{f}=0.74 GeV, \theta_{e}=8^{\circ} , \theta_{Ke}=11^{\circ} , and \Phi_{K}=180^{\circ} , which gives E_{\gamma}=1.5 GeV, |\vec{p}_{\gamma}|=1.511 GeV/c, Q^{2}=0.0323 (GeV/c) 2 , \theta_{\gamma e}=3.9^{\circ} , \varepsilon=0.591 , \Gamma=0.0140 sr -1 GeV -1 , \theta_{K\gamma}=7.1^{\circ} , and in the case of the ground state: |\vec{P}_{K}|_{mb}=1.245 GeV/c, |\vec{\Delta}|=315 MeV/c, \theta_{\Delta\gamma}=29.2^{\circ} , |\vec{p}_{opt}|=205 MeV/c, and |\vec{p}_{\Lambda}|=110 MeV/c. The kinematics is in the laboratory frame. The ground state of the target nucleus 208 Pb is J_{\sf A}^{P_{\sf A}}=0^{+} and its wave function is represented by the HF Slater determinant. We assume that {}^{208}_{\;\;\;\Lambda} Tl can be understood as \Lambda bound in the core nucleus 207 Tl which is a spectator in the impulse approximation. Therefore, it is important to describe well also the core nucleus in our approach. Within the HF method, energy spectrum of 207 Tl can be estimated from the proton single-particle energies \varepsilon_{p} . If we set the last occupied proton level (i.e. 0h 11/2 ) to 0.0 MeV we can take the relative energies of the following occupied levels (2s 1/2 , 1d 3/2 , 1d 5/2 , 0g 7/2 ) with respect to 0h 11/2 as an estimate of the low-lying energy spectrum of \frac{11}{2}^{-} , \frac{1}{2}^{+} , \frac{3}{2}^{+} , \frac{5}{2}^{+} , \frac{7}{2}^{+} states in 207 Tl. Such spectrum of the core nucleus obtained within our HF calculation is shown in the left column in Fig. 11 and denoted as “original” because it was obtained with the original form of the NN interaction. This original spectrum is compared with the phenomenological one in the right column which was obtained from experiments Wilson ; DataSheets ; WittH . Note that the fifth single-particle state \frac{7}{2}^{+} was observed in the 208 Pb( \gamma,p ) and 208 Pb( e,e^{\prime}p ) reactions DataSheets ; WittH and was also considered in the calculations by Motoba and Millener E12-20-013 . However, this state was not seen in the collision of two 208 Pb nuclei Wilson . Instead the state \frac{17}{2}^{+} at 3.813 MeV was observed. Figure 11: The spectrum of the core nucleus 207 Tl calculated in the HF method using the D16+DDT NN interaction with C_{\rho} = 4000 MeV \cdot fm 6 and denoted as “original” in the left column is compared with phenomenological values obtained in experiments Wilson ; DataSheets ; WittH (the right column). The energies are in MeV. The comparison demonstrates that the empirical spectrum is quite significantly squeezed down with respect to the theoretical (original) result and the ground-state in the theoretical spectrum \frac{11}{2}^{-} lies above the state \frac{3}{2}^{+} in the phenomenological spectrum. The phenomenological spectrum reflects that, in fact, the effective NN interaction should be modified. In this study, we mimic such effect by changing the proton 2s 1/2 , 1d 3/2 , 1d 5/2 , 0g 7/2 single-particle energies \varepsilon_{p} in order to reproduce the phenomenological \frac{1}{2}^{+} , \frac{3}{2}^{+} , \frac{11}{2}^{-} , \frac{5}{2}^{+} , and \frac{7}{2}^{+} energies of 207 Tl. These values of \varepsilon_{p} enter the TD Λ calculation (see Eq. (11) in structure ) and, therefore, modify OBDME. The excitation spectra in electroproduction of {}^{208}_{~~\Lambda} Tl calculated with the original and phenomenological spectrum of 207 Tl are shown in Fig. 12 . Both results were obtained using the BS3 amplitude in the on-shell approximation and the Nijmegen F YNG interaction with k_{\sf F}=1.34 fm -1 . The results are for kinematics of the experiment E12-20-013 in Hall C. One can see that changing the spectrum of 207 Tl results in a systematical shift of positions of the main peaks to larger energies. In general, the shape of the new excitation spectrum has changed with the phenomenological single-particle energies suggesting that ordering of some hypernucleus states has changed as well as the production cross sections. Note that assuming the phenomenological spectrum of 207 Tl means a modification of the cross sections via the values of OBDME. Figure 12: The excitation spectrum in 208 Pb (e,e^{\prime}K^{+})^{208}_{~~\Lambda} Tl obtained consistently with the original and phenomenological spectrum of 207 Tl shown in Fig. 11 . A strength of the proton( \alpha ) \to \Lambda(\alpha^{\prime}) transition is given by the reduced matrix element of the one-body transition operator (OBDME) (\Phi_{\sf H}||\left[b_{\alpha^{\prime}}^{+}\otimes a_{\alpha}\right]^{J}||\Phi_{\sf A}) and the radial integral {\cal R}^{LM}_{\alpha^{\prime}\alpha} where the sum in Eq. ( 1 ) runs over a model space given by \alpha\equiv(n\,l\,\,j)_{\sf p} and \alpha^{\prime}\equiv(n^{\prime}l^{\prime}j^{\prime})_{\Lambda} . In case of heavy hypernuclei the model space is big and its dimension is restricted by a considered accuracy. We compared the results for {}^{208}_{~~\Lambda} Tl calculated only with the transitions satisfying |{\rm OBDME}|\leq 0.001 and |{\rm OBDME}|\leq 0.0001 and have found a tiny differences. Therefore we use the former restriction in our calculations. Moreover, in the sum typically one up to five transitions dominate and therefore one can say that the given hypernucleus state is populated by a particular transition \alpha\to\alpha^{\prime} with J , L , and S . Note that the following triangle conditions must be fulfilled for the quantum numbers: \Delta(l^{\prime},l,\,L) , \Delta(J_{\sf H},J_{\sf A},J) , \Delta(j^{\prime},j,\,J) , and \Delta(J,\,L,\,S) with J_{\sf A}=0 and S=0,1 for the transitions without and with spin flip, respectively. Moreover, the orbital momenta must respect the parity of the final state, i.e. P_{\sf H}\cdot P_{\sf A}=(-1)^{L} and l+l^{\prime}+L must be an even number. The right single-particle quantum numbers to couple to the core nucleus must also satisfy \Delta(J_{\sf c},J_{\sf H},j^{\prime}) , \Delta(J_{\sf c},J_{\sf A},j) , P_{\sf c}\cdot P_{\sf A}=(-1)^{l} , and P_{\sf c}\cdot P_{\sf H}=(-1)^{l^{\prime}} . The dominant transitions for selected hypernucleus states with the corresponding quantum numbers are shown in Tables 2 and 3 . We also show the cross sections and the states in the “original” and “phenomenological” spectrum of the core 207 Tl which can be used to build up the ground state of 208 Pb and the states of {}^{208}_{~~\Lambda} Tl. In Table 2 we compare the first and second doublet based on the ground and first excited states of the core nucleus. In the case of the original spectrum the first hypernucleus doublet is based on the ground state \frac{11}{2}^{-} of 207 Tl and the second doublet on the excited state \frac{1}{2}^{+} at 1.317 MeV. This excitation energy then gives the splitting about 1.52 MeV of the doublets. Analogous situation is in the case of the phenomenological spectrum with the states \frac{1}{2}^{+} and \frac{3}{2}^{+} split only by 0.351 MeV giving the moderate doublet splitting \approx 0.28 MeV. original spectrum of 207 Tl core proton \Lambda hypernucleus cross sections E_{\sf c} J_{\sf c}^{P_{\sf c}} n l j n^{\prime} l^{\prime} j^{\prime} E_{\sf H} J_{\sf H}^{P_{\sf H}} J L S [nb/sr] 0.000 \frac{11}{2}^{-}\ \ 0 5 \frac{11}{2}\ \ 0 0 \frac{1}{2}\ \ 0.000 6^{-} 6 5 1 3.66 0.000 \frac{11}{2}^{-}\ \ 0 5 \frac{11}{2}\ \ 0 0 \frac{1}{2}\ \ 0.003 5^{-} 5 5 0,1 1.74 1.317 \frac{1}{2}^{+}\ \ 2 0 \frac{1}{2}\ \ 0 0 \frac{1}{2}\ \ 1.485 0^{+} 0 0 0 0.03 1.317 \frac{1}{2}^{+}\ \ 2 0 \frac{1}{2}\ \ 0 0 \frac{1}{2}\ \ 1.519 1^{+} 1 0 1 2.04 phenomenological spectrum of 207 Tl core proton \Lambda hypernucleus cross sections E_{\sf c} J_{\sf c}^{P_{\sf c}} n l j n^{\prime} l^{\prime} j^{\prime} E_{\sf H} J_{\sf H}^{P_{\sf H}} J L S [nb/sr] 0.000 \frac{1}{2}^{+} 2 0 \frac{1}{2} 0 0 \frac{1}{2} 0.000 0^{+} 0 0 0 0.03 0.000 \frac{1}{2}^{+} 2 0 \frac{1}{2} 0 0 \frac{1}{2} 0.034 1^{+} 1 0 1 2.07 0.351 \frac{3}{2}^{+} 1 2 \frac{3}{2} 0 0 \frac{1}{2} 0.313 2^{+} 2 2 0,1 2.08 0.351 \frac{3}{2}^{+} 1 2 \frac{3}{2} 0 0 \frac{1}{2} 0.319 1^{+} 1 2 1 1.16 Table 2: Doublets of hypernuclear states are shown with the quantum numbers of dominant transitions p\to\Lambda , the cross sections, and the corresponding states of the core nucleus. The cases with the “original” and “phenomenological” spectrum of the core 207 Tl are compared. In Table 2 we can also observe that the hypernucleus state with the larger spin in the doublet possesses the larger cross section in accord with the rule that the state with the maximum spin in the multiplet is populated most strongly. Another well known fact is that at small kaon angles, considered in electroproduction of hypernuclei, the elementary production is dominated by its spin-flip ( S=1 ) part. Therefore, the hypernucleus states with mere S=0 are only weakly populated whereas those with S=1 have significant cross sections. This dominance of the spin-flip amplitude is clearly seen in the second and first doublet of the upper and lower part of the table, respectively. This dominance also makes the difference of the ground-state cross sections for the original and phenomenological spectrum where the former is dominantly populated by the core-nucleus state \frac{11}{2}^{-} allowing S=1 but the latter is given by the state \frac{1}{2}^{+} requiring S=0 . In Table 3 we show the multiplets with \Lambda in the p and d orbits, see also Fig. 12 . It is apparent again that the maximum spin states 7 + and 8 - dominate the p_{\Lambda} and d_{\Lambda} multiplets, respectively, in both computational approaches. All transitions shown in Table 3 are based on the proton hole in the 0h_{11/2} orbit which corresponds to the ground state of the core nucleus in the approach with the original spectrum but to the excited state at 1.348 MeV in the phenomenological spectrum. The difference 1.348 MeV then makes the shift of the main p_{\Lambda} and d_{\Lambda} peaks observed in Fig. 12 . Remind that due to the selection rule the longitudinal parts of the cross section obtain significant contributions from the spin-flip amplitude ( S=1 ) only for the states 7^{+} , 8^{-} , and 6^{-} whereas the other states 6^{+} , 7^{-} , and 5^{-} receive their leading contributions only from the non spin-flip part of the elementary amplitude which is weaker in the studied kinematical region. This is also one of the reasons why the cross sections for the states 7^{+} and 8^{-} are so big. original spectrum of 207 Tl core proton \Lambda hypernucleus cross sections E_{\sf c} J_{\sf c}^{P_{\sf c}} n l j n^{\prime} l^{\prime} j^{\prime} E_{\sf H} J_{\sf H}^{P_{\sf H}} J L S [nb/sr] 0.000 \frac{11}{2}^{-} 0 5 \frac{11}{2} 0 1 \frac{3}{2} 5.804 7^{+} 7 6 1 17.82 0.000 \frac{11}{2}^{-} 0 5 \frac{11}{2} 0 1 \frac{1}{2} , \frac{3}{2} 5.794 6^{+} 6 6 0,1 7.88 0.000 \frac{11}{2}^{-} 0 5 \frac{11}{2} 0 2 \frac{3}{2} , \frac{5}{2} 11.875 7^{-} 7 7 0,1 18.02 0.000 \frac{11}{2}^{-} 0 5 \frac{11}{2} 0 2 \frac{3}{2} , \frac{5}{2} 11.934 5^{-} 5 5 0,1 0.20 0.000 \frac{11}{2}^{-} 0 5 \frac{11}{2} 0 2 \frac{3}{2} , \frac{5}{2} 11.946 6^{-} 6 5,7 1 1.27 0.000 \frac{11}{2}^{-} 0 5 \frac{11}{2} 0 2 \frac{5}{2} 11.964 8^{-} 8 7 1 41.78 phenomenological spectrum of 207 Tl core proton \Lambda hypernucleus cross sections E_{\sf c} J_{\sf c}^{P_{\sf c}} n l j n^{\prime} l^{\prime} j^{\prime} E_{\sf H} J_{\sf H}^{P_{\sf H}} J L S [nb/sr] 1.348 \frac{11}{2}^{-} 0 5 \frac{11}{2} 0 1 \frac{1}{2} , \frac{3}{2} 6.974 6^{+} 6 6 0,1 7.44 1.348 \frac{11}{2}^{-} 0 5 \frac{11}{2} 0 1 \frac{3}{2} 6.984 7^{+} 7 6 1 16.62 1.348 \frac{11}{2}^{-} 0 5 \frac{11}{2} 0 2 \frac{3}{2} , \frac{5}{2} 13.109 7^{-} 7 7 0,1 10.65 1.348 \frac{11}{2}^{-} 0 5 \frac{11}{2} 0 2 \frac{3}{2} , \frac{5}{2} 13.110 5^{-} 5 5 0,1 0.63 1.348 \frac{11}{2}^{-} 0 5 \frac{11}{2} 0 2 \frac{3}{2} , \frac{5}{2} 13.122 6^{-} 6 5,7 1 1.37 1.348 \frac{11}{2}^{-} 0 5 \frac{11}{2} 0 2 \frac{5}{2} 13.144 8^{-} 8 7 1 39.94 Table 3: The same as in Table 2 but for the \Lambda in the p ( l^{\prime}=1 ) and d ( l^{\prime}=2 ) orbit. Predictions of excitation spectra in electroproduction of {}^{208}_{~~\Lambda} Tl are also compared with the previous result by Motoba nad Millener utilized in the proposal of the Jlab experiment E12-20-013 E12-20-013 . This older calculation was done for photoproduction at E_{\gamma}=1.5 GeV and \theta_{K\gamma}=0.5^{\circ} using the Saclay-Lyon amplitude A (SLA) in the frosen-proton approximation ( p_{eff}=0 ). The \Lambda was assumed to be weakly coupled to the proton-hole states of 207 Tl, which are strongly populated in the ( e,e^{\prime}p ) and (d, 3 He) reactions on 208 Pb, and the \Lambda single-particle energies were calculated from the Woods-Saxon potential. Our calculation for electroproduction was performed in the HF + TD Λ approach using the NN interaction D16+DDT with C_{\rho} = 4000 MeV \cdot fm 6 and the SLA amplitude in the frosen-proton approximation assuming kinematics E_{i}=4 GeV, E_{f}=2.5 GeV, \theta_{e}=2^{\circ} , \theta_{Ke}=3.83^{\circ} , and \Phi_{K}=180^{\circ} giving photoproduction kinematics close to that by Motoba and Millener, E_{\gamma}=1.5 GeV, |\vec{q}\,|=1.504 GeV/c, Q^{2}=0.012 (GeV/c) 2 , and \theta_{K\gamma}=0.5^{\circ} . The relatively small value of Q^{2} allows a comparison with the photoproduction calculation. Figure 13: Comparison of the result by Motoba and Millener E12-20-013 with our calculations in the HF + TD Λ approach using the NN interaction D16+DDT with C_{\rho} = 4000 MeV \cdot fm 6 and the SLA amplitude in the frosen-proton approximation. The results were calculated with the Nijmegen F and Jülich A (JA) YNG interactions where the former was used with the original (NF) and phenomenological (NFphen) spectrum of 207 Tl in Fig. 11 . The curves are plotted with the width 800 keV for all states. The results obtained with the Nijmegen F (NF) and Jülich A (JA) YNG interactions are compared with the result by Motoba and Millener in Fig. 13 . In the case of Nijmegen YNG interaction we show both variants of the calculation, with the original (NF) and phenomenological (NFphen) spectrum of 207 Tl (see Fig. 11 ). One can see that in the new results the peaks with \Lambda in the d , f , g and h (for NFphen also p ) orbits are shifted to larger values of the excitation energy with respect to Motoba-Millener’s result where the shift is larger for the Nijmegen interaction. Moreover, the calculated cross sections for the high-laying states are systematically larger than the older values but they are a bit smaller for the s_{\Lambda} states. The best agreement is achieved with the NF and JA interactions for the second p_{\Lambda} peak composed mainly of the very close states (5.792 MeV, 6 + ), (5.804 MeV, 7 + ), and (5.831 MeV, 5 + ) with the NF and (5.425 MeV, 7 + ), (5.456 MeV, 6 + ), and (5.463 MeV, 5 + ) with the JA interaction. These states are dominantly populated by the 0h_{11/2}\to 0p_{3/2} and 0h_{11/2}\to 0p_{1/2} transitions. In the case of the s_{\Lambda} peaks in Fig. 13 the NFphen result agrees quite well with that by Motoba and Millener. This is because the phenomenological spectrum of 207 Tl is consistent with the spectrum used by Motoba and Milenner E12-20-013 . Therefore, the left s_{\Lambda} peak is based on the proton holes 2s^{-1}_{1/2} and 1d^{-1}_{3/2} and the right peak on 1d^{-1}_{5/2} and 0h^{-1}_{11/2} where contribution of the latter is negligible in the Motoba-Millener results but it is quite important in the NFphen result. On the contrary, in the NF and JA results the left peak is given by the 0h^{-1}_{11/2} and 2s^{-1}_{1/2} holes whereas the right one by the 1d^{-1}_{3/2} hole. Note that using the BS3 amplitude in the on-shell approximation changes mainly magnitudes of the main peaks at small excitation energies but at high energies the shape of the excitation spectrum is also changed which is due to the large density of hypernuclear states populated with different strength for the SLA and BS3 amplitudes. One of the noticeable differences between our and Motoba-Millener’s results is presence of a rising background in our result which causes that the peaks in the high-energy region of the spectrum are systematically rising. This phenomenon can be attributed to a high density of relatively weakly populated states in our calculations. In the JA calculation, there are about 630 states in the energy region 0–33 MeV, where most of the states are located above 10 MeV, and only 67 states are produced with the cross section larger than 5 nb/sr. On the other hand the older result includes only about 160 states. Note that in the Motoba-Millener’s result presented in Ref. E12-20-013 the background is partially simulated assuming a larger width of the peaks above the quasi-free threshold but in the result presented here we plotted the spectrum with the uniform width 800 keV. Finally, in Figure 14 we show predictions for the planned experiment E12-20-013 obtained with various forms of the effective YN interaction. The DWIA calculations for the Hall C kinematics Hall C were performed in the HF + TD Λ approach using the Daejeon 16 NN potential with the phenomenological DD term ( C_{\rho}=4000 MeV \cdot fm 6 ) and the BS3 amplitude in the on-shell approximation. The results are for the YNG interactions Jülich A (JA), chiral in the leading order ( \chi LO), and Nijmegen F. The latter was considered with the original (NF) and phenomenological (NFphen) spectrum of the core nucleus 207 Tl. The G-matrix for the \chi LO potential was calculated with E_{\rm{av}}=-20.75 MeV in Eqs. ( 6 ) and ( 7 ). The results in Fig. 14 show variety of predictions coming from different forms of the YNG interactions. A good quality experimental data, allowing determination of positions of the main peaks in the spectrum, could prefer one of the considered YNG interactions. Moreover, measured magnitudes of the peaks related to averaged cross sections can shed more light on dynamics of the process, the elementary amplitude and kaon distortion. Figure 14: Predicted spectrum of {}^{208}_{~~\Lambda} Tl in the Hall C kinematics of E12-20-013 calculated with the BS3 amplitude in the on-shell approximation for various forms of the baryon-baryon interaction: Jülich A (JA), Nijmegen F with original (NF) spectrum of 207 Tl, Nijmegen F with phenomenological (NFphen) spectrum of 207 Tl, and chiral leading order ( \chi LO). For the G-matrix of the \chi LO potential, Eqs. ( 6 ) and ( 7 ) with E_{\rm{av}}=-20.75 MeV were used. V Summary and conclusions We studied various effects in DWIA calculations of the cross sections in electroproduction of hypernuclei beyond the p shell. We analyzed the results in view of kaon distortion, kinematical effects, and various forms of the effective hyperon-nucleon interaction. The latter was aimed at illustrating uncertainties of predictions of the reaction cross sections due to variety of possible forms of the effective hyperon-nucleon interactions. We showed that the damping effect of the kaon re-scattering is rising with the hypernucleus mass, amounting 40–60% for {}^{40}_{~\Lambda} K and 50–80% for {}^{208}_{~\Lambda} Tl where the states with deeply bound \Lambda are more affected. The kaon distortion, accounted via the eikonal approximation with the first-order optical potential, depends mainly on behavior of the nucleon density in the peripheral region. We have found that the momenta and the cross sections apparently depend on a tiny modification of the hypernucleus mass due to excitation energy, denoted as the hypernucleus-mass shift. This mass shift modifies the proton, kaon, and \Lambda momenta as well as the momentum transfer by a few per cents which changes values of the elementary amplitude and radial integrals. These changes then suppress the cross section by several per cents which is important mainly for the high-excited states. We have elaborated our previously suggested optimum on-shell approximation fermi specifying yet free value of the angle \theta_{\Delta p} . In the new variant we suggest equate the magnitude of the proton effective momentum to a mean momentum of the proton in the single-particle orbit which corresponds to the dominant transition given by OBDME. The angle is then calculated from the energy conservation in the elementary vertex. This we denote as “on-shell approximation” where the elementary amplitude is on-shell with a specified non zero value of the proton momentum. We demonstrated influence of various values of the proton effective momentum on the cross sections for electroproduction of {}^{208}_{~~\Lambda} Tl. In the given kinematics, differences between the proton- and \Lambda - frozen approximations amount up to 30% in the photon energy region 1.2–2.2 GeV. Similarly as for the p -shell hypernuclei fermi the effects are more apparent at small kaon angles. Influence of the selection rule fermi , which controls contributions from the longitudinal mode of the virtual photon to the reduced amplitude, was also discussed for selected states of {}^{208}_{~~\Lambda} Tl. We updated and discussed spectra in electroproduction of {}^{40}_{~\Lambda} K and {}^{48}_{~\Lambda} K using a new effective NN interaction, Daejeon 16 NN interaction complemented with the phenomenological DD term, and two forms of the YNG interaction, Nijmegen F and Jülich A. The results were calculated in kinematics of the planned experiment E12-15-008 at Jefferson Lab and they extend the discussion of predictions for the experiment initiated in Ref. structure . We also discussed predictions for the electroproduction spectrum of {}^{52}_{~\Lambda} V. This spectrum clearly shows separated multiplets with \Lambda in the s , p , and d orbits which might be suitable for an experimental investigation. In the case of the 208 Pb target we discussed the results in view of various forms of the baryon-baryon interaction. We showed predictions with the Jülich A, Nijmegen F, and chiral leading-order YNG interactions and compared them with older calculations by Motoba and Millener utilized in preparing the experimental proposal E12-20-013 at Jefferson Lab. We also modified our calculations by using a phenomenological spectrum of the core nucleus 207 Tl and utilized this modified description in predicting the excitation spectra for {}^{208}_{~~\Lambda} Tl in kinematics of the experiment E12-20-013. The results show that a good quality data could discriminate between various forms of the effective YNG interactions.