Photoelectron spectroscopy and theoretical investigations have been performed to systematically probe the intrinsic electronic properties of [Mo6X14]2− (X = halogen). All three PE spectra of gaseous [Mo6X14]2− (X = Cl, Br, I) dianions, which were generated by electrospray ionization, exhibit multiple resolved peaks in the recorded binding energy range. Theoretical investigations on the orbital structure and charge distribution were performed to support interpretation of the observed spectra and were further extended onto [Mo6F14]2−, a dianion that was not available for the experimental study. The measured adiabatic (ADE) and vertical detachment energies (VDE) for X = Cl–I were well reproduced by density functional theory calculations (accuracy ∼0.1 eV). Corresponding ADE/VDE values for the dianions were found to be 1.48/2.13 (calc.) and 2.30/2.65, 2.30/2.62, and 2.20/2.42 eV (all expt.) for X = F, Cl, Br, and I, respectively, showing an interesting buckled trend of electron binding energy (EBE) along the halogen series, i.e., EBE (F) ≪ EBE (Cl) ∼ EBE (Br) > EBE (I). Molecular orbital analyses indicate different mixing of metal and halogen atomic orbitals, which is strongly dependent on the nature of X, and suggest that the most loosely bound electrons are detached mainly from the metal core for X = F and Cl, but from halide ligands for X = Br and I. The repulsive Coulomb barrier (RCB), estimated from the photon energy dependent spectra, decreases with increasing halogen size, from 1.8 eV for X = Cl to 1.6 eV for X = I. Electrostatic potential modeling confirms the experimental RCB values and predicts that the most favorable electron detaching pathway should lie via the face-bridging halide ligands.

The [Mo6X14]2− (X = halogen) dianions consist of an octahedral Mo6 metal core surrounded by eight face-bridging inner (Xi) and six apical (Xa) halides (Fig. 1). Previous studies have shown that these clusters exhibit high chemical stability, have long-lived excited electronic states, and can undergo facile ground and excited state electron or energy transfer reactions.1–3 Such properties have been explored to develop luminescent materials, oxygen sensors, and singlet oxygen sensitizers.4–7 For example, high photoluminescence quantum yields and long emission lifetimes of [Mo6X14]2− in the red to near-infrared (NIR) region make them excellent molecular units in solar energy conversion devices.8,9 Additionally, these anions have been used as building blocks for the construction of various functional supramolecular structures, nanomaterials, and multidimensional molecular assemblies.10–12 Also, derivates of these anions have recently been used as x-ray or light activated singlet oxygen sensitizers for photodynamic therapy of cancer or photoinactivation of bacteria.13,14

FIG. 1.

Geometry of octahedral hexanuclear molybdenum [Mo6Xi8Xa6]2− (X = F, Cl, Br, I).

FIG. 1.

Geometry of octahedral hexanuclear molybdenum [Mo6Xi8Xa6]2− (X = F, Cl, Br, I).

Close modal

Investigation of the structural and electronic properties of the octahedral metal clusters has a long standing history and continues to be of ongoing interest.15–25 Brosset was the first to determine the structure of the [Mo6Cl8]4+ core by characterizing [Mo6Cl8](OH)4·14H2O and [Mo6Cl8](Cl4·2H2O)·6H2O crystals using x-ray diffraction.26,27 This study was followed by the characterization of the bond lengths and vibrational modes of other molybdenum halide complexes.28–34 The photochemical and electrochemical properties of [Mo6X14]2− were studied using absorption spectroscopy, emission spectroscopy, and electrochemical measurements in the solution or solid state.2,3,35 In addition, ground and excited electronic states of [Mo6X14]2− have also been investigated using computational chemistry methods.36–39 Molecular orbital (MO) calculations showed that the highest occupied molecular orbital (HOMO) was formed by the 4d atomic orbitals (AOs) of the core Mo6 atoms40,41 and emission occurred via decay of a triplet state.1,42,43

The current study aims to extend the knowledge about the electronic properties of [Mo6X14]2− dianions. To the best of our knowledge, a gas phase study of the intrinsic properties of these compounds without influence of environmental factors has not been reported. In this work, we present photoelectron spectroscopy (PES) results of [Mo6X14]2− (X = Cl, Br, I), which are interpreted with the help of electronic structure calculations. Charge distribution and MO analyses are performed, and the repulsive Coulomb barrier (RCB), derived from photon-energy-dependent spectra, is modeled. All results are discussed in comparison with closo-dodecaborate [B12X12]2− (X = Cl, Br, I),44 since both species are highly symmetric and electronically stable dianions with the negative charge delocalized in the extended molecular framework.

The photoelectron spectroscopy (PES) experiments were conducted on an apparatus consisting of an electrospray ionization source (ESI), a temperature-controlled cryogenic ion trap, and a magnetic-bottle photoelectron spectrometer.45 Briefly, a ∼1 mM acetonitrile solution of the tetrabutylammonium (TBA) salt of [Mo6X14]2− (X = Cl, Br, I) was used to generate the gas phase [Mo6X14]2− dianions. The ions produced from the ESI source were directed by quadrupole ion guides into the temperature-controlled ion trap, where they were accumulated and cooled by collisions with cold buffer gas (20% H2 balanced in helium) at 20 K for 20–100 ms to improve spectral energy resolution and eliminate the possibility of the appearance of peaks in the PE spectra due to hot bands. Then, the cryogenically cooled ions were pulsed out into the extraction zone of a time-of-flight (TOF) mass spectrometer for the mass-to-charge separation and analyses. The [Mo6X14]2− clusters were each mass-selected and maximally decelerated before being photodetached with photons of three different wavelengths: 266 nm (4.661 eV) from an Nd:YAG laser, 193 nm (6.424 eV) from an ArF laser, and 157 nm (7.866 eV) from an F2 laser. All lasers were operated at a 20 Hz repetition rate, with the ion beam off at alternating laser shots, affording shot-to-shot background subtraction. The photodetached electrons were collected with nearly 100% efficiency by the magnetic-bottle and analyzed in a 5.2 m long flight tube. The resultant PE spectra were calibrated with the known PE spectra of iodide46 and osmium hexachloride47 recorded at similar conditions. The energy resolution was approximately 20 meV for electrons with 1 eV kinetic energy.

The structures of [Mo6X14]2− (X = F, Cl, Br, I) were fully optimized with density functional theory (DFT) methods using the B3LYP48,49 functionals. The basis set Def2-tzvppd was used for F, Cl, and Br atoms, and the effective core potential (ECP) basis set Def2-tzvppd50 was used for I and Mo atoms. The theoretical vertical detachment energies (VDEs) were calculated as the energy differences between the singly charged anions and the corresponding dianions, both on the optimized dianions’ structures. Theoretical adiabatic detachment energies (ADEs) were calculated as the energy differences between the geometrically relaxed single charged anion (minimum reached with the dianion geometry as the initial structure) and the dianion. Vibrational frequencies have been computed in order to estimate the zero-point energy correction of ADEs and to confirm that the structures used were energy minima. During this process, the symmetry point groups for the singly charged anions were lowered to C2v, C2v, and C4v for X = Cl, Br, and I, respectively, because of the Jahn-Teller effect. All DFT calculations were performed using the NWChem program package.51 The Natural Population Analysis (NPA) charges were computed using JANPA software.52 

The 157 nm PE spectra of [Mo6X14]2− (X = Cl, Br, I) are shown in Fig. 2. The VDE and ADE of each dianion are determined from the maximum and onset of the first spectral band (X′), respectively, as indicated by dashed (VDE) and dotted (ADE) lines in Fig. 2. The electron binding energy (EBE) of these dianions, as gauged by ADE and VDE, decreases marginally (<0.03 eV) when the halogen changes from Cl to Br, but appreciably by ∼ 0.1–0.2 eV from Br to I (Table I). In order to have a more complete EBE trend for this series of dianions along increasing halogen size, we calculated ADE and VDE of [Mo6F14]2−, a cluster that was not experimentally available. As shown in Table I, both calculated ADE and VDE for X = F are significantly smaller than the corresponding values for X = Cl. Therefore, a clear EBE stabilization by substituting X = F with Cl is found, i.e., EBE (F) <<EBE (Cl), which is opposite to the trend found from Cl to Br to I, EBE (Cl) ∼ EBE (Br) > EBE (I).

FIG. 2.

Photoelectron spectra of [Mo6X14]2− [X = Cl, Br, I] recorded at 157 nm excitation at 20 K. Dashed and dotted lines indicate the maximum and threshold of the first spectral band, from which the VDE and ADE are estimated. The color coded bar below each spectrum assigns the spectral bands to the corresponding calculated MOs (see Fig. 3). (a) [Mo6Cl14]2−; (b) [Mo6Br14]2−; (c) [Mo6I14]2−.

FIG. 2.

Photoelectron spectra of [Mo6X14]2− [X = Cl, Br, I] recorded at 157 nm excitation at 20 K. Dashed and dotted lines indicate the maximum and threshold of the first spectral band, from which the VDE and ADE are estimated. The color coded bar below each spectrum assigns the spectral bands to the corresponding calculated MOs (see Fig. 3). (a) [Mo6Cl14]2−; (b) [Mo6Br14]2−; (c) [Mo6I14]2−.

Close modal
TABLE I.

Experimental and theoretical ADE and VDE (in eV) of [Mo6X14]2− (X = F, Cl, Br, I). Numbers in the parentheses are uncertainties for the last digit.

ADEVDE
Expt.Theo.Expt.Theo.
[Mo6F14]2− n.a. 1.48 n.a. 2.13 
[Mo6Cl14]2− 2.30(5) 2.17 2.65(5) 2.58 
[Mo6Br14]2− 2.30(5) 2.22 2.62(5) 2.55 
[Mo6I14]2− 2.20(5) 2.16 2.42(5) 2.46 
ADEVDE
Expt.Theo.Expt.Theo.
[Mo6F14]2− n.a. 1.48 n.a. 2.13 
[Mo6Cl14]2− 2.30(5) 2.17 2.65(5) 2.58 
[Mo6Br14]2− 2.30(5) 2.22 2.62(5) 2.55 
[Mo6I14]2− 2.20(5) 2.16 2.42(5) 2.46 

The optimized geometries of all clusters [Mo6X14]2− (X = F, Cl, Br, I) have high symmetry of Oh. The clusters consist of an octahedron of Mo6 cores surrounded by eight face-bridging inner (Xi) and six apical (Xa) halides. The bond lengths of Mo-Xa are 2.0, 2.46, 2.65, and 2.93 Å for X = F, Cl, Br, and I, respectively, consistent with the previous available experimental and theoretical data.32,37 The octahedron Mo6 core slightly expands in size as the halide ligand becomes heavier, i.e., the Mo–Mo bond length increases sequentially (2.52 → 2.62 → 2.65 → 2.68 Å for X = F → Cl → Br → I). The six apical Xa halogens form a large octahedron with a side length of 5.35, 6.10, 6.39, and 6.82 Å for X = F, Cl, Br, and I, respectively.

Molecular orbital (MO) analyses indicate that the halogen atomic orbital (AO) coefficients in the HOMOs increase monotonically over the series with halogen size from X = F to I (Table II). Based on this analysis, we first expected a monotonic EBE trend along the halogen series. Therefore, the EBE trend observed experimentally and confirmed theoretically (VDE calculations) is surprising and requires further explanations. Figure 3 shows a Hartree-Fock (HF) orbital energy diagram in which all four dianions [Mo6X14]2− (X = F, Cl, Br, I) are compared. Note that, under Koopmans’ approximation,53 the calculated HF MO energies (relative to the vacuum level 0 eV) usually are 1–2 eV larger than the experimentally measured values due to neglecting large relaxation energies.54 However, the evolution of the orbital energy plot along a series of similar ions is often in good qualitative agreements with band-developments in PE spectra along the same series of ions, as it has been previously observed for several dianions.54,55 As shown in Fig. 3, the measured trend in electronic stability along the series is well reflected in the plotted HF-HOMO energies. The energy levels are labeled with the respective orbital symmetries so that the energy level of a particular type of orbitals along the halogen series can be followed. The seemingly counterintuitive trend along the halogen series may be rationalized by the interplay of several factors. In [Mo6F14]2−, the highest lying orbitals (t1u and t2u) are mainly composed of Mo6 atomic orbitals (blue in Fig. 3, details in Table II). Substitution of X = F with Cl results in a stronger overlap of the high lying Mo6 orbitals with Xa halogen orbitals, now lying higher than the corresponding orbitals in the fluorinated cluster. This results in stabilization of the Mo6 orbitals. However, with the further increase of halogen size, the halogens’ AOs, which overlap with the Mo6 core, become even higher in energy. The extent of mixing and stabilization of the highest lying Mo6 orbitals with halogen AOs depends on the orbital nature. The HOMO with t1u symmetry is significantly stabilized by exchange of X = F with Cl. The stabilization is less pronounced for the HOMO-1 orbital with t2u symmetry. Substitution of X = Cl with Br further stabilizes the t1u-orbital slightly but appears to destabilize the t2u-MO due to the strong influence of high lying Br AOs. For X = Br, the t2u-MO is the HOMO. This trend further continues, and a significant destabilization of the t2u HOMO occurs going from X = Br to I, while the influence on the t1u-MO is less pronounced.

TABLE II.

Atomic orbital contribution of Mo, Xa, and Xi atoms to the highest lying Hartree-Fock MOs. Numbers are for qualitative comparison and have an uncertainty of roughly 3% because only the 100 AOs with largest coefficients have been considered for the calculation.

Halogen XOrbitalSymmetry6*Mo (%)6*Xa (%)8*Xi (%)
HOMO t1u 93 
 HOMO-1 t2u 94 
Cl HOMO t1u 80 15 
 HOMO-1 t2u 68 20 12 
Br HOMO t2u 36 50 14 
 HOMO-1 t1u 62 35 
HOMO t2u 14 70 16 
 HOMO-1 eg 41 18 41 
Halogen XOrbitalSymmetry6*Mo (%)6*Xa (%)8*Xi (%)
HOMO t1u 93 
 HOMO-1 t2u 94 
Cl HOMO t1u 80 15 
 HOMO-1 t2u 68 20 12 
Br HOMO t2u 36 50 14 
 HOMO-1 t1u 62 35 
HOMO t2u 14 70 16 
 HOMO-1 eg 41 18 41 
FIG. 3.

Hartree-Fock molecular orbital energy diagram showing the calculated energy of the highest lying occupied molecular orbitals of [Mo6X14]2− along the halogen series X = F to I. The number of lines accounts for the degree of degeneracy. All energy levels are labeled with the respective orbital symmetry. The color codes are used to associate the groups of orbitals with the observed spectral bands (Fig. 2) and assigned according to the similar energy development features in the F−I series.

FIG. 3.

Hartree-Fock molecular orbital energy diagram showing the calculated energy of the highest lying occupied molecular orbitals of [Mo6X14]2− along the halogen series X = F to I. The number of lines accounts for the degree of degeneracy. All energy levels are labeled with the respective orbital symmetry. The color codes are used to associate the groups of orbitals with the observed spectral bands (Fig. 2) and assigned according to the similar energy development features in the F−I series.

Close modal

The above described transformation of MO levels with X shows some parallels to the previously reported case of [B12X12]2−, in which a stabilization of the HOMO was observed for X = F → Cl → Br, but destabilization occurred for X = I.44 In both cases ([Mo6X14]2− and [B12X12]2−), a break in the electronic stability trend was observed along the halogen series. While the boron centered HOMO in [B12X12]2− continuously stabilizes with increasing halogen overlap, orbitals strongly dominated by halogen atomic orbitals become higher in energy at a certain point along the series, like the t2u-orbital in the Mo6 case.

In Fig. 2, we labeled the first spectral band that is used to determine VDE and ADE with X′. Further resolved bands are labeled alphabetically with increasing EBE. For [Mo6Cl14]2− [Fig. 2(a)], three bands X′, A, and B are seen from EBE = 2.2 to 3.4 eV, followed by a weak peak C at 4.0 eV, a strong sharp peak D at 4.7 eV, and two relatively broad features E at 5.1 eV and F at 6.0 eV (Table S1). A qualitatively similar spectral pattern is observed for [Mo6Br14]2− [Fig. 2(b)], but differences in band positions are evident, i.e. the strong peak D and bands E and F shift to lower EBEs by ∼0.8 eV relative to the X = Cl spectrum, and an additional band G at EBE = 6.1 eV is observed. The [Mo6I14]2− spectrum appears to be significantly different with even better resolved signals. A high intensity sharp band (labeled with B) is observed at 2.9 eV. A color code connects spectral bands in Fig. 2 with the HF orbital energy levels in Fig. 3. The five highest lying orbitals in [Mo6Cl14]2− are associated with the bands X′, A, B, and C in Fig. 2(a). The evolution of HOMO and HOMO-1 upon halogen variation has been discussed in Sec. III A. The substitution of Cl with Br affects the two highest lying orbitals (t1u and t2u symmetry) to exchange their positions in energy. HOMO-3 to -5 do not alter strongly in energy. This is consistent with a similar band shape and position of the blue and violet marked bands (X′, A, B, C) in both spectra (X = Cl, Br). In contrast, orbitals at even higher electron binding energies are much more influenced by the increasing energy of the halogen AOs and rise significantly along the series. The HOMO-5 orbital t1g for X = Cl and Br (color code: red) consists mainly of halogen AOs and is associated with the intensive sharp band D seen in the Cl and Br spectra. In the case of X = I, the energy level of this t1g orbital upshifts to be very similar to the highest lying orbitals. This is well reflected by the position of the associated sharp intense band, which is band D in X = Cl and Br, but becomes band B in the case of X = I. Also, the groups of orbitals are energetically well separated in the case of [Mo6Cl14]2− and [Mo6Br14]2− (color code: green, yellow, brown), but not in the case of [Mo6I14]2−, explaining a different appearance of the spectral bands from X = Cl/Br to I.

Figure 4 demonstrates the difference of the 157 nm PE spectrum of [Mo6Cl14]2− from those taken at 193 nm and 266 nm. Equivalent spectra for X = Br and I are shown in the SI. In the spectra obtained with lower laser energies, several spectral bands at higher binding energies are not observed, although the corresponding binding energy is lower than the photon energy. For example, in the X = Cl case, the 193 nm spectrum shows band F disappeared and the intensity of peak E decreases. The 266 nm spectrum shows that B, C, D, and E peaks all disappeared [Figs. 4(a)–4(c)]. Similar sequential disappearance of the spectral features and suppression of peak intensity at lower detachment photon energies can be found in the systems of X = Br and I. This phenomenon is well known44 and traced back to the existence of the repulsive Coulomb barrier (RCB) in photodetaching multiply charged anions (MCAs), which prevents electrons with kinetic energy smaller than the RCB height being emitted. A more detailed explanation on how RCB can be estimated from energy-dependent PE spectra may be found in Ref. 54 and is given in the SI (Fig. S4 and accompanying text using [Mo6Cl14]2− as an example). We estimated the RCB value of 1.8 eV for X = Cl and a decreasing tendency for RCB values with increasing halogen size, see Table III.

FIG. 4.

Photon-energy dependent spectra of [Mo6Cl14]2− recorded at 266 (a), 193 (b), and 157 nm (c). The signal that is suppressed is shown using the red bar that has a 0.2 eV uncertainty margin (gray); in this case, the kinetic energy of the detached electron is not sufficient to overcome the RCB. Note that electrons with kinetic energies slightly below the RCB may tunnel through the barrier, resulting in only partial signal suppression. (d) Calculated potential energy curve of the {[Mo6Cl14] + e} system for three different electron detachment directions, as illustrated in the insert. The gray line shows the trace of the natural e + e Coulomb potential curve.

FIG. 4.

Photon-energy dependent spectra of [Mo6Cl14]2− recorded at 266 (a), 193 (b), and 157 nm (c). The signal that is suppressed is shown using the red bar that has a 0.2 eV uncertainty margin (gray); in this case, the kinetic energy of the detached electron is not sufficient to overcome the RCB. Note that electrons with kinetic energies slightly below the RCB may tunnel through the barrier, resulting in only partial signal suppression. (d) Calculated potential energy curve of the {[Mo6Cl14] + e} system for three different electron detachment directions, as illustrated in the insert. The gray line shows the trace of the natural e + e Coulomb potential curve.

Close modal
TABLE III.

Experimental and theoretical RCB. Numbers in the parentheses are uncertainties for the last digit.

RCB (eV)
Expt.Theo.
[Mo6Cl14]2− 1.8(1) 2.01 
[Mo6Br14]2− 1.7(1) 1.88 
[Mo6I14]2− 1.6(1) 1.65 
RCB (eV)
Expt.Theo.
[Mo6Cl14]2− 1.8(1) 2.01 
[Mo6Br14]2− 1.7(1) 1.88 
[Mo6I14]2− 1.6(1) 1.65 

For a computational evaluation of the RCB, we explored charge distribution within the [Mo6X14]2−/1− ions. The lowest barrier for a detaching electron will be found along a pathway which crosses the least negative (most positive) surface area of the anion. We used Natural Population Analysis (NPA)56 to assign atomic charges within the molecular frame of these dianionic species. In Table IV, the charge values of the equivalent atoms are summed up. According to this analysis, the Mo6 core holds a positive charge, while the negative charge is, to a large extent, delocalized on the six Xa halogen atoms. The eight Xi atoms are less negative (for X = I, even slightly positive), suggesting that the preferred detachment pathway is generally through the surface of Xi rather than the Xa ligands. Detaching an electron mainly affects the atomic charges of the Mo atoms, which become even more positive, and to a smaller extent the six Xa atoms. The NPA values of the Xi ligands in the singly charged ion are very similar to those of the dianion. Along the halogen series from X = Cl to I, the negative charge moves from the eight Xi halogens to the Mo6 core. This makes the Xi ligands less negative (more positive) which reduces the RCB along this pathway with increasing halogen size. For a more quantitative evaluation of detachment pathways, we calculated the potential energy of the system ([Mo6X14] + e) along different pathways for e using the electrostatic potential (ESP) of the singly charged [Mo6X14] on the geometry of the doubly charged ion (assumption: electron loss faster than structural rearrangement of the cluster). In the case of [B12X12]2− dianions, this approach resulted in a slight overestimation of the RCB for the smaller halogens and appropriate predictions for the larger halogens. (For the smaller halogens, the potential maximum is located very near to the molecular surface so that polarizability has a considerable influence.) We considered three directions for electron loss in our investigation: from the center of the ion through (1) an Xa ligand, (2) an Mo-Mo edge, and (3) a face-bridging Xi ligand. A diagram showing the potential energies for X = Cl is shown in Fig. 4(d). Equivalent diagrams for X = Br and I can be found in the supplementary material. The lowest energy pathway for a detaching electron is—in agreement with the conclusions from atomic charge analysis—through the Xi ligands. The large potential maximum found along the Xa direction confirms that the negative charge is strongly localized on these halogens. Similar to previous investigations on [B12X12]2− ions, the ESP maximum along the preferred pathway is in good agreement with experimental results and only slightly overestimates the experimental values.

TABLE IV.

Atomic charges of [Mo6Xi8Xa6]2− and [Mo6Xi8Xa6] on the geometry of the dianion (X = Cl, Br, I) calculated with NPA. Charges are summed up for the equivalent atoms.

DianionNPA [Mo6X14]2−
6*Mo 8*Xi 6*Xa 
Cl +2.94 −1.60 −3.36 
Br +1.50 −0.40 −3.06 
+0.36 +0.40 −2.76 
DianionNPA [Mo6X14]2−
6*Mo 8*Xi 6*Xa 
Cl +2.94 −1.60 −3.36 
Br +1.50 −0.40 −3.06 
+0.36 +0.40 −2.76 
AnionNPA [Mo6X14]
6*Mo 8*Xi 6*Xa 
Cl +3.56 −1.56 −3.01 
Br +2.11 −0.36 −2.75 
+1.04 +0.52 −2.57 
AnionNPA [Mo6X14]
6*Mo 8*Xi 6*Xa 
Cl +3.56 −1.56 −3.01 
Br +2.11 −0.36 −2.75 
+1.04 +0.52 −2.57 

We have investigated the gas phase electronic structure of halogenated octahedral [Mo6X14]2− clusters. Simple models developed for dianions with separated charges57–63 cannot describe the electronic stability of these symmetric dianions with uniformly delocalized negative charges. No monotonic trend of electron detachment energy was observed upon exchanging the halogens from X = F over Cl to Br and I. The observed trend with energy maximum at X = Cl and Br can be explained by different energetic developments of the two high lying molecular orbitals upon their interaction with several different atomic orbitals of the surrounding halogen shell. The spectral bands and their evolution along the halogen series were qualitatively interpreted with the help from MO analyses and can be tracked back to MO energy level shuffling due to the metal and ligand AO mixings along the whole series. Comparative studies of [Mo6X14]2− and [B12X12]2− cluster ions demonstrate that a consistent trend of electronic stability along the F−I halogen series should not be generally awaited for multiply charged halogenated clusters. The “turning point” (stabilization with increasing halogen size to destabilization with increasing halogen size) is in both cases associated with a change in the HOMO nature.

Calculations of atomic charges indicate that the Mo6 core is very positive for all four halogenated clusters. The negative charge is predominantly localized on the six apical halogen ligands Xa. Therefore, electron detachment has the lowest barrier through the face-bridging ligands Xi, which hold much lower negative charge, and even small positive charge in the case of I. The repulsive Coulomb barrier is decreasing with increasing halogen size in a similar manner with [B12X12]2− ions.

See the supplementary material for the optimized geometries (Fig. S1), spectra at different wavelengths and repulsive Coulomb barrier (RCB) for [Mo6Br14]2− (Fig. S2) and [Mo6I14]2− (Fig. S3), description of how to estimate experimental RCB using [Mo6Cl14]2− as an example (Fig. S4), and higher EBE bands and electronic excited states (Table S1).

The PES work was supported by the U.S. Department of Energy (DOE), Office of Science, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences and Biosciences (X.-B.W.), and was performed at the EMSL, a national scientific user facility sponsored by DOE’s Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory (PNNL). Visiting scholar R.-Z.L. and students Q.Y. and Z.L. acknowledge support by a PNNL alternate sponsored fellowship. R.-Z.L. acknowledges the China Scholarship Council and the National Science Foundation of Shaanxi, China (Grant No. 2019JM-292). J.W. acknowledges support from the Alexander von Humboldt foundation (Feodor Lynen Fellowship and Rückkehrerstipendium). K.K. acknowledges support from the Czech Science Foundation (Grant No. 18-05076S) and the working group Interactions of Inorganic Clusters, Cages, and Containers with Light within the AV21 Strategy of the Czech Academy of Science.

1.
Y.
Saito
,
H. K.
Tanaka
,
Y.
Sasaki
, and
T.
Azumi
,
J. Phys. Chem.
89
,
4413
(
1985
).
2.
A. W.
Maverick
,
J. S.
Najdzionek
,
D.
MacKenzie
,
D. G.
Nocera
, and
H. B.
Gray
,
J. Am. Chem. Soc.
105
,
1878
(
1983
).
3.
A. W.
Maverick
and
H. B.
Gray
,
J. Am. Chem. Soc.
103
,
1298
(
1981
).
4.
K.
Kirakci
,
P.
Kubát
,
J.
Langmaier
,
T.
Polívka
,
M.
Fuciman
,
K.
Fejfarová
, and
K.
Lang
,
Dalton Trans.
42
,
7224
(
2013
).
5.
M. N.
Sokolov
,
M. A.
Mikhailov
,
A. V.
Virovets
,
K. A.
Brylev
,
R. A.
Bredikhin
,
A. M.
Maksimov
,
V. E.
Platonov
, and
V. P.
Fedin
,
Russ. Chem. Bull.
62
,
1764
(
2013
).
6.
J. A.
Jackson
,
M. D.
Newsham
,
C.
Worsham
, and
D. G.
Nocera
,
Chem. Mater.
8
,
558
(
1996
).
7.
J. A.
Jackson
,
C.
Turro
,
M. D.
Newsham
, and
D. G.
Nocera
,
J. Phys. Chem.
94
,
4500
(
1990
).
8.
E.
Rojas-Mancilla
,
A.
Oyarce
,
V.
Verdugo
,
C.
Morales-Verdejo
,
C.
Echeverria
,
F.
Velásquez
,
J.
Chnaiderman
,
F.
Valiente-Echeverría
, and
R.
Ramirez-Tagle
,
Molecules
22
,
1108
(
2017
).
9.
H. B.
Gray
and
A. W.
Maverick
,
Science
214
,
1201
(
1981
).
10.
N.
Nerambourg
,
T.
Aubert
,
C.
Neaime
,
S.
Cordier
,
M.
Mortier
,
G.
Patriarche
, and
F.
Grasset
,
J. Colloid Interface Sci.
424
,
132
(
2014
).
11.
S.
Cordier
,
F.
Dorson
,
F.
Grasset
,
Y.
Molard
,
B.
Fabre
,
H.
Haneda
,
T.
Sasaki
,
M.
Mortier
,
S.
Ababou-Girard
, and
C.
Perrin
,
J. Cluster Sci.
20
,
9
(
2009
).
12.
F.
Grasset
,
F.
Dorson
,
S.
Cordier
,
Y.
Molard
,
C.
Perrin
,
A. M.
Marie
,
T.
Sasaki
,
H.
Haneda
,
Y.
Bando
, and
M.
Mortier
,
Adv. Mater.
20
,
143
(
2008
).
13.
K.
Kirakci
,
J.
Zelenka
,
M.
Rumlová
,
J.
Cvačka
,
T.
Ruml
, and
K.
Lang
,
Biomater. Sci.
7
,
1386
(
2019
).
14.
K.
Kirakci
,
J.
Zelenka
,
M.
Rumlová
,
J.
Martinčík
,
M.
Nikl
,
T.
Ruml
, and
K.
Lang
,
J. Mater. Chem. B
6
,
4301
(
2018
).
15.
N.
Saito
,
S.
Cordier
,
P.
Lemoine
,
T.
Ohsawa
,
Y.
Wada
,
F.
Grasset
,
J. S.
Cross
, and
N.
Ohashi
,
Inorg. Chem.
56
,
6234
(
2017
).
16.
T.
Aubert
,
N.
Nerambourg
,
N.
Saito
,
H.
Haneda
,
N.
Ohashi
,
M.
Mortier
,
S.
Cordier
, and
F.
Grasset
,
Part. Part. Syst. Charact.
30
,
90
(
2013
).
17.
G.
Saito
,
H.
Hosoda
,
Y.
Yoshida
,
J.
Hagiwara
,
K.
Nishimura
,
H.
Yamochi
,
A.
Otsuka
,
T.
Hiramatsu
,
Y.
Shimazaki
, and
K.
Kirakci
,
J. Mater. Chem.
22
,
19774
(
2012
).
18.
A. S.
Mocanu
,
M.
Amela-Cortes
,
Y.
Molard
,
V.
Cîrcu
, and
S.
Cordier
,
Chem. Commun.
47
,
2056
(
2011
).
19.
Y.
Molard
,
F.
Dorson
,
V.
Cîrcu
,
T.
Roisnel
,
F.
Artzner
, and
S.
Cordier
,
Angew. Chem., Int. Ed.
49
,
3351
(
2010
).
20.
S.
Ababou-Girard
,
S.
Cordier
,
B.
Fabre
,
Y.
Molard
, and
C.
Perrin
,
ChemPhysChem
8
,
2086
(
2007
).
21.
K.
Kirakci
,
H.
Hosoda
,
S.
Cordier
,
C.
Perrin
, and
G.
Saito
,
J. Solid State Chem.
179
,
3628
(
2006
).
22.
S.
Cordier
,
K.
Kirakci
,
D.
Méry
,
C.
Perrin
, and
D.
Astruc
,
Inorg. Chim. Acta
359
,
1705
(
2006
).
23.
K.
Kirakci
,
S.
Cordier
, and
C.
Perrin
,
Z. Anorg. Allg. Chem.
631
,
411
(
2005
).
24.
G.
Pilet
,
K.
Kirakci
,
S. C.
de Montigny
,
C.
Lapinte
,
C.
Perrin
, and
A.
Perrin
,
Eur. J. Inorg. Chem.
2005
,
919
.
25.
S.
Akagi
,
S.
Fujii
,
T.
Horiguchi
, and
N.
Kitamura
,
J. Cluster Sci.
28
,
757
(
2017
).
26.
C.
Brosset
,
Ark. Kem. Mineral. Geol.
A22
,
10
(
1946
).
27.
C.
Brosset
,
Ark. Kem. Mineral. Geol.
A20
,
16
(
1945
).
28.
P.
Brückner
,
W.
Preetz
, and
M.
Pünjer
,
Z. Anorg. Allg. Chem.
623
,
8
(
1997
).
29.
J. R.
Schoonover
,
T. C.
Zietlow
,
D. L.
Clark
,
J. A.
Heppert
,
M. H.
Chisholm
,
H. B.
Gray
,
A. P.
Sattelberger
, and
W. H.
Woodruff
,
Inorg. Chem.
35
,
6606
(
1996
).
30.
W.
Preetz
,
D.
Bublitz
,
H.
Von Schnering
, and
J.
Sassmannshausen
,
Z. Anorg. Allg. Chem.
620
,
234
(
1994
).
31.
W.
Preetz
,
K.
Harder
,
H.
von Schnering
,
G.
Kliche
, and
K.
Peters
,
J. Alloys Compd.
183
,
413
(
1992
).
32.
L. J.
Guggenberger
and
A.
Sleight
,
Inorg. Chem.
8
,
2041
(
1969
).
33.
J.
Sheldon
,
J. Chem. Soc.
1960
,
1007
.
34.
P. A.
Vaughan
,
Proc. Natl. Acad. Sci. U. S. A.
36
,
461
(
1950
).
35.
T. C.
Zietlow
,
M. D.
Hopkins
, and
H. B.
Gray
,
J. Solid State Chem.
57
,
112
(
1985
).
36.
K.
Costuas
,
A.
Garreau
,
A.
Bulou
,
B.
Fontaine
,
J.
Cuny
,
R.
Gautier
,
M.
Mortier
,
Y.
Molard
,
J.-L.
Duvail
, and
E.
Faulques
,
Phys. Chem. Chem. Phys.
17
,
28574
(
2015
).
37.
R.
Ramirez-Tagle
and
R.
Arratia-Pérez
,
Chem. Phys. Lett.
460
,
438
(
2008
).
38.
H.
Honda
,
T.
Noro
,
K.
Tanaka
, and
E.
Miyoshi
,
J. Chem. Phys.
114
,
10791
(
2001
).
39.
L.
Robinson
,
R.
Bain
,
D.
Shriver
, and
D.
Ellis
,
Inorg. Chem.
34
,
5588
(
1995
).
40.
F. A.
Cotton
and
G. G.
Stanley
,
Chem. Phys. Lett.
58
,
450
(
1978
).
41.
F. A.
Cotton
and
T.
Haas
,
Inorg. Chem.
3
,
10
(
1964
).
42.
T.
Azumi
and
Y.
Saito
,
J. Phys. Chem.
92
,
1715
(
1988
).
43.
B. E.
Bursten
,
F. A.
Cotton
, and
G. G.
Stanley
,
Isr. J. Chem.
19
,
132
(
1980
).
44.
J.
Warneke
,
G.-L.
Hou
,
E.
Aprà
,
C.
Jenne
,
Z.
Yang
,
Z.
Qin
,
K.
Kowalski
,
X.-B.
Wang
, and
S. S.
Xantheas
,
J. Am. Chem. Soc.
139
,
14749
(
2017
).
45.
X.-B.
Wang
and
L.-S.
Wang
,
Rev. Sci. Instrum.
79
,
073108
(
2008
).
46.
D.
Hanstorp
and
M.
Gustafsson
,
J. Phys. B: At., Mol. Opt. Phys.
25
,
1773
(
1992
).
47.
X. B.
Wang
and
L. S.
Wang
,
J. Chem. Phys.
111
,
4497
(
1999
).
48.
A. D.
Becke
,
J. Chem. Phys.
98
,
1372
(
1993
).
49.
C.
Lee
,
W.
Yang
, and
R. G.
Parr
,
Phys. Rev. B
37
,
785
(
1988
).
50.
F.
Weigend
and
R.
Ahlrichs
,
Phys. Chem. Chem. Phys.
7
,
3297
(
2005
).
51.
M.
Valiev
,
E. J.
Bylaska
,
N.
Govind
,
K.
Kowalski
,
T. P.
Straatsma
,
H. J.
Van Dam
,
D.
Wang
,
J.
Nieplocha
,
E.
Apra
, and
T. L.
Windus
,
Comput. Phys. Commun.
181
,
1477
(
2010
).
52.
T. Y.
Nikolaienko
,
L. A.
Bulavin
, and
D. M.
Hovorun
,
Comput. Theor. Chem.
1050
,
15
(
2014
).
53.
D. J.
Tozer
and
N. C.
Handy
,
J. Chem. Phys.
109
,
10180
(
1998
).
54.
E.
Aprà
,
J.
Warneke
,
S. S.
Xantheas
, and
X.-B.
Wang
,
J. Chem. Phys.
150
,
164306
(
2019
).
55.
J.
Warneke
,
S. Z.
Konieczka
,
G.-L.
Hou
,
E.
Aprà
,
C.
Kerpen
,
F.
Keppner
,
T. C.
Schäfer
,
M.
Deckert
,
Z.
Yang
,
E. J.
Bylaska
, et al.,
Phys. Chem. Chem. Phys.
21
,
5903
(
2019
).
56.
A. E.
Reed
,
R. B.
Weinstock
, and
F.
Weinhold
,
J. Chem. Phys.
83
,
735
(
1985
).
57.
J. R. R.
Verlet
,
D. A.
Horke
, and
A. S.
Chatterley
,
Phys. Chem. Chem. Phys.
16
,
15043
(
2014
).
58.
X. B.
Wang
and
L. S.
Wang
,
Annu. Rev. Phys. Chem.
60
,
105
(
2009
).
59.
A. I.
Boldyrev
,
M.
Gutowski
, and
J.
Simons
,
Acc. Chem. Res.
29
,
497
(
1996
).
60.
M. K.
Scheller
,
R. N.
Compton
, and
L. S.
Cederbaum
,
Science
270
,
1160
(
1995
).
61.
O. T.
Ehrler
,
J.-P.
Yang
,
A. B.
Sugiharto
,
A. N.
Unterreiner
, and
M. M.
Kappes
,
J. Chem. Phys.
127
,
184301
(
2007
).
62.
X.-B.
Wang
and
L.-S.
Wang
,
J. Phys. Chem. A
104
,
4429
(
2000
).
63.
W. E.
Boxford
and
C. E. H.
Dessent
,
Phys. Chem. Chem. Phys.
8
,
5151
(
2006
).

Supplementary Material