Chemical Physics 482 (2017) 244–248

Contents lists available at ScienceDirect

Chemical Physics

journal homepage: www.elsevier .com/locate /chemphys

Interatomic Coulombic decay and electron-transfer-mediated decayfollowing triple ionization of Ne2 and NeAr

http://dx.doi.org/10.1016/j.chemphys.2016.09.0220301-0104/� 2016 Elsevier B.V. All rights reserved.

⇑ Corresponding author.E-mail address: ueda@tagen.tohoku.ac.jp (K. Ueda).

Fig. 1. Schematic diagrams of (a) direct ICD (interatomic Coulombic deexchange ICD, and (c) ETMD (electron-transfer-mediated decay).

T. Ouchi a, H. Fukuzawa a, K. Sakai a, T. Mazza a,b, M. Schöffler c, K. Nagaya d, Y. Tamenori e, N. Saito f,K. Ueda a,⇑a Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japanb European XFEL, Albert-Einstein-Ring 19, 22761 Hamburg, Germanyc Institute for Nuclear Physics, Johann Wolfgang Goethe University Frankfurt, Frankfurt 60438, GermanydDepartment of Physics, Kyoto University, Kyoto 606-8502, Japane Japan Synchrotron Radiation Research Institute, Sayo, Hyogo 679-5198, JapanfNational Institute of Advanced Industrial Science and Technology, NMIJ, Tsukuba 305-8568, Japan

a r t i c l e i n f o

Article history:Available online 15 September 2016

a b s t r a c t

We report observations of the interatomic Coulombic decay (ICD) and electron-transfer-mediated decay(ETMD) from the triply charged states in Ne2 and NeAr dimers. The ICD processes leading to fragmenta-tion of Ne3+-Ne into Ne3+-Ne+ and Ne3+-Ar into Ne3+-Ar+, and ETMD processes leading to fragmentation ofNe3+-Ne into Ne2+-Ne2+ are unambiguously identified by electron–ion-ion coincidence spectroscopy inwhich the kinetic energy of the ICD or ETMD electron and the kinetic energy release between the twofragment ions are measured in coincidence.

� 2016 Elsevier B.V. All rights reserved.

1. Introduction

In 1997, Cederbaum et al. [1] theoretically suggested a newrelaxation mechanism of the electronically excited species in thepresence of the loosely bound neighboring species. In this mecha-nism, the excited species decays electronically by ionizing theneighboring species. This new relaxation mechanism is calledinteratomic/intermolecular Coulombic decay (ICD). Six years later,the first experimental proof for occurrence of ICD was reported forneon clusters [2]. Since then, a number of theoretical and experi-mental studies on ICD have been carried out in many different sys-tems, as recently reviewed [3]. These studies elucidated ICD playsan important role in chemistry, transferring the energy and thecharge from the excited species to the environment surroundingit. It is worth noting also that ICD in hydrogen bonded systemsand its role as a source of low energy electrons in biological med-ium have been discussed in several recent publications [4–9].

Prototype systems where ICD have been studied in details, bothexperimentally and theoretically, are rare-gas dimers. Let us con-sider the ICD in a dimer AB illustrated in Fig. 1. An atom A withan inner-valence vacancy transfers its energy to a neighboringatom B which subsequently releases its energy by emitting an elec-tron from its outer-valence orbital (Fig. 1(a)). This energy transfer

process corresponds to the direct Coulomb term of the two-elec-tron integral. This term causes an asymptotic behavior of the ICDrate of R�6 (R being the internuclear distance) characteristic ofdipole–dipole interaction and thus can be viewed as virtual photonexchange [10]. The ICD rate caused by the exchange term (Fig. 1(b)), on the other hand, drops exponentially as a function of R.There is another class of ICD-related processes, as illustrated byFig. 1(c), where the outer-valence electron of the neighboring atomB fills the inner-valence vacancy of atom A and another outer-valence electron of the atom B is emitted. This process is calledelectron-transfer-mediated decay (ETMD) [11]. ETMD is differentfrom ICD in the final charge states of both atoms. In ETMD, thecharge state of the atom A that had the initial vacancy decreasesby one while the originally neutral neighboring atom B nowbecomes doubly charged. The ETMD rate drops exponentially asa function of R. ETMD is much slower than ICD at equilibrium

cay), (b)

T. Ouchi et al. / Chemical Physics 482 (2017) 244–248 245

internuclear distances of van der Waals clusters in general. Forexample, in the case of decays from Ne+(2s�1)-Ar, ICD is at least103 times faster than ETMD at the equilibrium internuclear dis-tance of NeAr and thus the lifetime of �30 fs for Ne+(2s�1)-Ar issolely determined by the ICD rate [11]. ETMD, however, becomesdominant if ICD is energetically closed [12,13].

The initial states of ICD and ETMD can be not only inner-valenceionized states [1,2] but also resonant neutral states [14], ionicsatellite states [15], Auger final dicationic states [16,17], triply ion-ized states [12], etc. In the present work, we have investigated ICDand ETMD from the triply ionized states using homo-nucleardimers Ne2 and hetero-nuclear dimers NeAr. In the case of atomicNe, �6% of Ne+(1s�1) produced by Ne 1s photoionization leads toNe3+ by double Auger decay [18,19]. This double Auger decaywas recently investigated by Hikosaka et al. in further detail [20].We detected slow electrons in coincidence with a pair of twoatomic ions using momentum-resolved electron–ion-ion coinci-dence spectroscopy and extracted the correlation between thekinetic energy of the ICD or ETMD electron and kinetic energyrelease (KER), i.e., the sum of the energies of two dissociating ions.

Fig. 2. Ion-ion coincidence TOF spectra.

2. Experiment

The experiment was carried out on the C-branch of the beamline BL27SU [21–23] at SPring-8. The photon beam was focusedto a size of less than 0.2 mm in height and 0.5 mm in width atthe point crossing with the cluster beam. The measurements wereperformed with the linearly polarized light having the electric vec-tor E in the horizontal direction at a photon energy of 888.7 eV, i.e.,18.5 eV above the atomic Ne 1s ionization threshold [24]. The stor-age ring was operated in several-bunches mode providing 53 sin-gle bunches (4/58 filling bunches) separated by 82.6 ns.

The heteronuclear dimers NeAr were produced by expanding amixture of neon and argon gases at a flow-rate ratio of 70:1 and atotal stagnation pressure of 0.12 MPa at temperature of 103 Kthrough a pinhole of 80 lm in diameter. Under these conditionsthe cluster beam contains monomers (Ne and Ar), homonucleardimers (Ne2 and Ar2), heteronuclear dimers (NeAr), and a smallfraction of larger clusters. The cluster beam was directed verticallyand crossed the incident radiation at right angles.

Our momentum-resolved electron–ion multicoincidence[16,25] is equivalent to cold-target recoil-ion momentum spec-troscopy or reaction microscope [26] and is based on recordingtimes of flight (TOFs) for electrons and ions with two positionand time sensitive multihit-capable detectors (Roentdek HEX120for electrons and HEX80 for ions). Knowledge of position and arri-val time on the particle detectors, (x, y, t), allows us to extractinformation about the 3Dmomentum of each particle. The electronand ion TOF spectrometers were placed face to face. The spectrom-eter axis was horizontal and perpendicular to both the incidentradiation and the cluster beam. Detailed geometric descriptionsand typical field conditions of the spectrometers were given else-where [25]. The TOFs of electrons and ions were recorded withrespect to the bunch marker of the light source using multi-hittime-to-digital converters (Roentdek TDC8HP), selecting by logicgating only electron signals synchronized with the single bunches.

3. Results and discussions

3.1. Ion-ion coincidence map

Fig. 2 depicts the ion-ion coincidence TOF spectra. The horizon-tal and vertical coordinates correspond to the TOFs of the first andthe second ions of the coincidence pair. Pairs of Ne and/or Aratomic ions satisfying the momentum conservation laws within

the plane perpendicular to the TOF axis are shown in the figure.There are correlation lines of ion pairs indicating that the sum ofmomenta of two ions parallel to the TOF axis are zero. Findingthese correlation lines excludes existence of the third atom whichcan take some momentum and thus evidences that ion pairs arereleased from dimers. Coincidence events originating from the22Ne isotopes are also seen in Fig. 2 but are not included in the fol-lowing analysis. Although some correlation lines due to NeAr areoverlapped with those of Ne2 or Ar2, those can be easily distin-guished by the fact that ratio of velocities of Ne and Ar releasedfrom NeAr must be 2:1 whereas that of Ne and Ne or Ar and Armust be 1:1 to satisfy the momentum conservation law.

The ion pairs Ne+-Ne+, Ar+-Ar+ and Ne+-Ar+ are formed mostly asa result of radiative and/or non-adiabatic charge transfer afterAuger decay [27–29]. The ion pairs Ne2+-Ne+, Ar2+-Ar+, Ne2+-Ar+

and Ne+-Ar2+ are formed mostly as a result of ICD after Auger decay[16,17,27,30,31]. The ion pairs Ar3+-Ar+ and Ar2+-Ar2+ are formedmostly as a result of ICD and ETMD after triple ionization of Ar2,respectively [12].

Below, we will discuss in detail the ion pair formation channelsvia ICD and ETMD from triply ionized states, after double Augerdecay following Ne 1s photoionization, using electron and ionenergy distributions in the Ne3+-Ne+, Ne2+-Ne2+ and Ne3+-Ar+ for-mation channels found in Fig. 2.

3.2. Ne3+-Ne+ and Ne2+-Ne2+ coincident channels

Fig. 3(a) depicts the energy distribution of electrons recorded incoincidence with Ne3+-Ne+ ion pairs. The peak that appears at�19 eV corresponds to the Ne 1s photoelectron. There are anotherstructures at the lower energy region which are considered to cor-respond to the ICD electron emission as we discuss below. Fig. 3(b)depicts KER for the two dissociating ions Ne3+ and Ne+ recorded incoincidence. For describing fragmentation of a multiply chargedvan der Waals dimer, it is a very good approximation to take onlythe Coulomb repulsion between two ions [3]. Using this commonlyused approximation, we find that the measured KER peak of 14 eVcorresponds to an internuclear distance of 3.1 Å. This value agreeswith the equilibrium distance 3.09 Å[32] of the neutral groundstate of Ne2.

10 15 20 25 30 350

5

10

15

20

25

30

10 15 20 25 30 35 40 450

5

10

15

β

Cou

nts

KER + Electron Energy (eV)

α

δ

γ

ζ

Cou

nts

KER + Electron Energy (eV)

ε

Fig. 4. Energy distributions for the sum of the emitted electron energy and the KERin (a) Ne3+-Ne+ and (b) Ne2+-Ne2+.

25

20

15

10

5

KE

R (

eV)

20151050Electron energy (eV)

100

30

20

10

Cou

nts

500Counts

(a)

(b)(c)

Fig. 3. (a) Kinetic energy distribution of electrons detected in coincidence withNe3+-Ne+ pairs. (b) Kinetic energy release (KER) for Ne3+-Ne+fragmentation channel.(c) Correlation map between emitted electron energy and KER.

246 T. Ouchi et al. / Chemical Physics 482 (2017) 244–248

Fig. 3(c) shows a correlation between the electron kineticenergy and the KER recorded in coincidence. In this correlation dia-gram, one can see two islands labeled a and b in coincidence withlow energy electrons. The islands appear as a line of slope �1, sug-gesting that the sum of KER and electron kinetic energy is constant.Since the energy difference between ICD initial state and final stateis shared by the KER and ICD electron kinetic energy, the slope of�1 provides evidence that Ne3+-Ne+ ion pair formation is via theICD transition. Because Ne 1s photoelectrons are detected also incoincidence with these ion pairs, the triply charged ICD initialstates (Ne3+-Ne) are produced by double ionization of Ne+(1s�1),i.e. double Auger decay following the Ne 1s photoionization.

The distribution of the sum of KER and electron kinetic energyfor Ne3+-Ne+ is depicted in Fig. 4(a). The peaks a and b in Fig. 4(a) correspond to the lines a and b in Fig. 3(c), respectively. Theseenergy sums should agree with the energy differences of initial andfinal states of the ICD transitions. Thus, we can identify the transi-tions a and b as given below;

a: Ne3+(2s�12p�2 2S)-Ne ? Ne3+(2p�3 4S)-Ne+(2p�1 2P) + eICD� ,b: Ne3+(2s�12p�2 2P)-Ne ? Ne3+(2p�3 4S)-Ne+(2p�1 2P) + eICD� .

The energy differences of these initial and final states for thetransitions a and b are 15.6 eV and 18.1 eV, respectively [24], ingood agreement with the measurements. Those energies areshown in Fig. 4(a) as broken lines.

In a similar way, we can also identify ETMD transitions notingthat final charge state is Ne2+-Ne2+. Since the Ne 1s photoelectronis detected at �18 eV in the electron spectrum, in coincidence withthe Ne2+-Ne2+ ion pair, the ETMD channel yielding Ne2+-Ne2+ isconsidered to take place after double Auger decay following Ne1s photoionization. Fig. 4(b) is the distribution of the sum of KERand electron energy for Ne2+-Ne2+. The broad structure around42 eV is due to coincidence with Ne 1s photoelectron. Besides,we can recognize four peaks at �19, �24, �29 and �32 eV, labeledby c, d, e and f, respectively. Comparing these values with the

energy differences between possible candidates of initial and finalstates, we can identify the transitions as given below;

c: Ne3+(2s�12p�2 4P)-Ne ? Ne2+(2p�2 1D)-Ne2+(2p�2 3P) + eETMD� ,

d: Ne3+(2s�12p�2 4P)-Ne ? Ne2+(2p�2 3P)-Ne2+(2p�2 3P) + eETMD� .

e: Ne3+(2s�12p�2 2D)-Ne? Ne2+(2p�2 1D)-Ne2+(2p�2 3P) + eETMD� ,

f: Ne3+(2s�12p�2 2D)-Ne ? Ne2+(2p�2 3P)-Ne2+(2p�2 3P) + eETMD� .

The energy differences of these initial and final states for thetransitions c, d, e and f are 20.5, 23.7, 29.2 and 32.4 eV, respectively[24], in good agreement with the measurements. Those energiesare shown in Fig. 4(b) as broken lines.

Fig. 5 is a schematic energy diagram for the initial and finalstates of ICD and ETMD. Energies of the initial states Ne3+-Ne(labeled by arabic numbers 1–4) are approximated by the horizon-tal lines given by the relevant atomic energies [24]. Energies of thefinal states Ne2+-Ne2+ (labeled by alphabets A–G) and Ne3+-Ne+ (H–J) are approximated by the sum of the atomic energies [24] andCoulomb repulsion energies. Let us focus on Ne3+(2s�12p�2 4P)-Ne(line 1) and Ne3+(2s�12p�2 2D)-Ne (line 2). Because these statesare below the ICD final states Ne3+-Ne+ (H–J) in the Franck–Condonregion (�3.09 Å), these states are not subject to ICD but can decayto the states Ne2+-Ne2+ below them via ETMD. The measured KERfor Ne2+-Ne2+ peaked at �28 eV corresponds to internuclear dis-tance of �2.1 Å assuming the Coulomb repulsion. This internucleardistance likely corresponds to the inner repulsive wall of the vandear Waals potential. Because only Ne3+(2s�12p�2 2D)-Ne (line 2)can decay to Ne2+(2p�2 3P)-Ne2+(2p�2 3P) (line A) or Ne2+(2p�2 1D)-Ne2+(2p�2 3P) (line B) around internuclear distance of 2.1 Å, thoseETMD channels (e and f) are observed when Ne3+(2s�12p�2 2D)-Ne (line 2) is populated. In the case Ne3+(2s�12p�2 4P)-Ne (line 1)is populated, again only decays to Ne2+(2p�2 3P)-Ne2+(2p�2 3P)(line A) or Ne2+(2p�2 1D)-Ne2+(2p�2 3P) (line B) are energeticallypossible. Thus, those ETMD channels (c and d) are also observed.The internucler distances where the ETMD occurs are, however,likely close to the crossing points of the potential energy curves

Fig. 5. Schematic energy diagram for the initial (labeled by 1–4) and final (labeledby A–G for ETMD and H–J for ICD) states of the interatomic processes. Initial statesdecay to final states which are energetically below them. 1, Ne3+(2s�12p�2 4P)-Ne;2, Ne3+(2s�12p�2 2D)-Ne; 3, Ne3+(2s�12p�2 2S)-Ne; 4, Ne3+(2s�12p�2 2P)-Ne; A,Ne2+(2p�2 3P)-Ne2+(2p�2 3P); B, Ne2+(2p�2 1D)-Ne2+(2p�2 3P); C, Ne2+(2p�2 1D)-Ne2+(2p�2 1D); D, Ne2+(2p�2 1S)-Ne2+(2p�2 3P); E, Ne2+(2p�2 1S)-Ne2+(2p�2 1D); F,Ne2+(2p�2 1S)-Ne2+(2p�2 1S); G, Ne2+(2s�12p�1 3P)-Ne2+(2p�2 3P); H, Ne3+(2p�3 4S)-Ne+(2p�1 2P); I, Ne3+(2p�3 2D)-Ne+(2p�1 2P); J, Ne3+(2p�3 2P)-Ne+(2p�1 2P). The ver-tical dotted line labelled as Re indicate the equilibrium bond length of the neutralNe2.

Fig. 6. (a) Energy distribution for the sum of the electron energy and the KER inNe3+-Ar+. The blue bars are estimates (the energy differences of initial and finalstates of ICD). (b) Schematic energy diagram for the initial (labeled by 1–3) and final(labeled by A–C) states of the ICD processes. Initial states decay to final states whichare energetically below them. 1, Ne3+(2s�12p�2 2D)-Ar; 2, Ne3+(2s�12p�2 2S)-Ar; 3,Ne3+(2s�12p�2 2P)-Ar; A, Ne3+(2p�3 4S)-Ar+(3p�1 2P); B, Ne3+(2p�3 2D)-Ar+(3p�1 2P);C, Ne3+(2p�3 2P)-Ar+(3p�1 2P).The vertical dotted line labelled as Re indicate theequilibrium bond length of the neutral NeAr.

T. Ouchi et al. / Chemical Physics 482 (2017) 244–248 247

of the ETMD initial and final states, i.e., the shortest possible dis-tances but longer than 2.1 Å.

3.3. Ne3+-Ar+ coincident channels

Let us discuss the ICD channels for the hetero dimers NeAr.Fig. 6(a) is the energy distribution for the sum of the electronkinetic energy and the KER in Ne3+-Ar+. The broad structure around30 eV is due to coincidence with Ne 1s photoelectrons; KER peakedat 11.8 eV corresponds to NeAr equilibrium internuclear distance�3.48 Å[33]. Based on the energy scheme in Fig. 6(b), we canexpect that ICD transitions may take place from Ne3+(2s�12p�2 2D,2S, 2P)-Ar to Ne3+(2p�3 4S, 2D, 2P)-Ar+(3p�1 2P). The energy differ-ences for the initial and final states of these transitions are in therange between 10.6 and 23.9 eV, as illustrated in Fig. 6(a) by thethe vertical blue bars.

4. Conclusion

We have carried out electron–ion-ion coincidence spectroscopyon homonuclear dimers Ne2 and heteronuclear dimers NeAr. Mea-suring energy of electrons and the kinetic energy release betweenthe two fragment atomic ions in coincidence, we could identify,without any ambiguity, the ICD processes leading to fragmentationof Ne3+-Ne into Ne3+-Ne+ and Ne3+-Ar into Ne3+-Ar+, and ETMD pro-cesses leading to fragmentation of Ne3+-Ne into Ne2+-Ne2+. ETMD

processes are usually much slower than ICD processes. However,for the lower energy states for which ICD is energetically forbid-den, ETMD becomes dominant decay processes. The present studyillustrates that ETMD is not an unusual process but can be com-monly seen for loosely bound systems as ICD does.

Acknowledgments

The experiments were performed at SPring-8 with the approvalof JASRI. We are grateful to L.S. Cederbaum and A.I. Kuleff for help-ful discussion. The work was supported by Grant-in-Aid(21244062) from JSPS, by the Management Expenses Grants forNational Universities Corporations fromMEXT, and by IMRAM pro-ject funding.

References

[1] L.S. Cederbaum, J. Zobeley, F. Tarantelli, Phys. Rev. Lett. 79 (1997) 4778.[2] S. Marburger, O. Kugeler, U. Hergenhahn, T. Möller, Phys. Rev. Lett. 90 (2003)

203401.[3] T. Jahnke, J. Phys. B: At. Mol. Opt. Phys. 48 (2015) 082001.[4] T. Jahnke, H. Sann, T. Havermeier, K. Kreidi, C. Stuck, M. Meckel, M. Schöffler, N.

Neumann, R. Wallauer, S. Voss, A. Czasch, O. Jagutzki, A. Malakzadeh, F. Afaneh,T. Weber, H. Schmidt-Böcking, R. Dörner, Nat. Phys. 6 (2010) 139.

[5] M. Mucke, M. Braune, S. Barth, M. Förstel, T. Lischke, V. Ulrich, T. Arion, Y.Becker, A. Bradshaw, U. Hergenhahn, Nat. Phys. 6 (2010) 143.

[6] G.A. Grieves, T.M. Orlando, Phys. Rev. Lett. 107 (2011) 016104.[7] S. Stoychev, A.I. Kuleff, L.S. Cederbaum, J. Am. Chem. Soc. 133 (2011) 6817.[8] V. Stumpf, K. Gokhberg, L.S. Cederbaum, Nat. Chem. 8 (2016) 237.[9] P. Slavícek, N.V. Kryzhevoi, E.F. Aziz, B. Winter, J. Phys. Chem. Lett. 7 (2016)

234.

248 T. Ouchi et al. / Chemical Physics 482 (2017) 244–248

[10] R. Santra, L.S. Cederbaum, Phys. Rep. 368 (2002) 1.[11] J. Zobeley, R. Santra, L.S. Cederbaum, J. Chem. Phys. 115 (2001) 5076.[12] K. Sakai, S. Stoychev, T. Ouchi, I. Higuchi, M. Schöffler, T. Mazza, H. Fukuzawa,

K. Nagaya, M. Yao, Y. Tamenori, A.I. Kuleff, N. Saito, K. Ueda, Phys. Rev. Lett. 106(2011) 033401.

[13] M. Förstel, M. Mucke, T. Arion, A.M. Bradshaw, U. Hergenhahn, Phys. Rev. Lett.106 (2011) 033402.

[14] T. Aoto, K. Ito, Y. Hikosaka, E. Shigemasa, F. Penent, P. Lablanquie, Phys. Rev.Lett. 97 (2006) 243401.

[15] T. Jahnke, A. Czasch, M. Schöffler, S. Schössler, M. Käsz, J. Titze, K. Kreidi, R.E.Grisenti, A. Staudte, O. Jagutzki, L.Ph. Schmidt, Th. Weber, H. Schmidt-Böcking,K. Ueda, R. Dörner, Phys. Rev. Lett. 99 (2007) 153401.

[16] Y. Morishita, X.-J. Liu, N. Saito, T. Lischke, M. Kato, G. Prümper, M. Oura, H.Yamaoka, Y. Tamenori, I.H. Suzuki, K. Ueda, Phys. Rev. Lett. 96 (2006) 243402.

[17] T. Ouchi, K. Sakai, H. Fukuzawa, I. Higuchi, Ph.V. Demekhin, Y.-C. Chiang, S.D.Stoychev, A.I. Kuleff, T. Mazza, M. Schöffler, K. Nagaya, M. Yao, Y. Tamenori, N.Saito, K. Ueda, Phys. Rev. A 83 (2011) 053415.

[18] N. Saito, I.H. Suzuki, Phys. Scr. 49 (2000) 80.[19] B. Kanngießer, M. Jainz, S. Brünken, W. Benten, Ch. Gerth, K. Godehusen, K.

Tiedtke, P. van Kampen, A. Tutay, P. Zimmermann, V.F. Demekhin, A.G. Kochur,Phys. Rev. A 62 (2000) 014702.

[20] Y. Hikosaka, T. Kaneyasu, P. Lablanquie, F. Penent, E. Shigemasa, K. Ito, Phys.Rev. A 92 (2015) 033413.

[21] H. Ohashi, E. Ishiguro, Y. Tamenori, H. Kishimoto, M. Tanaka, M. Irie, T. Tanaka,T. Ishikawa, Nucl. Instrum. Methods A 467-468 (2001) 529.

[22] H. Ohashi, E. Ishiguro, Y. Tamenori, H. Okumura, A. Hiraya, H. Yoshida, Y.Senba, K. Okada, N. Saito, I.H. Suzuki, K. Ueda, T. Ibuki, S. Nagaoka, I. Koyano, T.Ishikawa, Nucl. Instrum. Methods A 467-468 (2001) 533.

[23] K. Ueda, J. Phys. B: At. Mol. Opt. Phys. 36 (2003) R1.

[24] A. Kramida, Yu. Ralchenko, J. Reader, NIST ASD Team, NIST Atomic SpectraDatabase (ver. 5.1). Available: http://physics.nist.gov/asd.

[25] K. Ueda, H. Fukuzawa, X.-J. Liu, K. Sakai, G. Prümper, Y. Morishita, N. Saito, I.H.Suzuki, K. Nagaya, H. Iwayama, M. Yao, K. Kreidi, M. Schöffler, T. Jahnke, S.Schössler, R. Dörner, T. Weber, J. Harries, Y. Tamenori, J. Electron Spectosc.Relat. Phenom. 166-167 (2008) 3.

[26] J. Ullrich, R. Moshammer, A. Dorn, R. Dörner, L.Ph.H. Schmidt, H. Schmidt-Böcking, Rep. Prog. Phys. 66 (2003) 1463.

[27] K. Kreidi, T. Jahnke, T. Weber, T. Havermeier, X. Liu, Y. Morisita, S. Schössler, L.P.H. Schmidt, M. Schöffler, M. Odenweller, N. Neumann, L. Foucar, J. Titze, B.Ulrich, F. Sturm, C. Stuck, R. Wallauer, S. Voss, I. Lauter, H.K. Kim, M. Rudloff, H.Fukuzawa, G. Prümper, N. Saito, K. Ueda, A. Czasch, O. Jagutzki, H. Schmidt-Böcking, S. Stoychev, P.V. Demekhin, R. Dörner, Phys. Rev. A 78 (2008) 043422.

[28] N. Saito, Y. Morishita, I. Suzuki, S. Stoychev, A. Kuleff, L. Cederbaum, X.-J. Liu, H.Fukuzawa, G. Prümper, K. Ueda, Chem. Phys. Lett. 441 (2007) 16.

[29] T. Ouchi et al., to be submitted.[30] K. Kreidi, T. Jahnke, T. Weber, T. Havermeier, R.E. Grisenti, X. Liu, Y. Morisita, S.

Schössler, L.P.H. Schmidt, M. Schöffler, M. Odenweller, N. Neumann, L. Foucar,J. Titze, B. Ulrich, F. Sturm, C. Stuck, R. Wallauer, S. Voss, I. Lauter, H.K. Kim, M.Rudloff, H. Fukuzawa, G. Prümper, N. Saito, K. Ueda, A. Czasch, O. Jagutzki, H.Schmidt-Böcking, S.K. Semenov, N.A. Cherepkov, R. Dörner, J. Phys. B: At. Mol.Opt. Phys. 41 (2008) 101002.

[31] T. Ouchi, K. Sakai, H. Fukuzawa, X.-J. Liu, I. Higuchi, Y. Tamenori, K. Nagaya, H.Iwayama, M. Yao, D. Zhang, D. Ding, A.I. Kuleff, S.D. Stoychev, Ph.V. Demekhin,N. Saito, K. Ueda, Phys. Rev. Lett. 107 (2011) 053401.

[32] A. Wüest, F. Merkt, J. Chem. Phys. 118 (2003) 8807.[33] J.-U. Grabow, A.S. Pine, G.T. Fraser, F.J. Lovas, R.D. Suenram, T. Emilsson, E.

Arunan, H.S. Gutowsky, J. Chem. Phys. 102 (1995) 1181.