Zero-Magnetic-Field Splitting in the Excited Triplet States of

Subscriber access provided by University of Texas Libraries

Article

Zero-Magnetic-Field Splitting in the Excited Triplet States of Octahedral Hexanuclear Molybdenum(II) Clusters: [{MoX}(n-CFCOO)] (X = Cl, Br, or I) 6

8

3

7

6

2–

Soichiro AKAGI, Eri Sakuda, Akitaka Ito, and Noboru Kitamura J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.7b06783 • Publication Date (Web): 28 Aug 2017 Downloaded from http://pubs.acs.org on August 29, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry A is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Zero-Magnetic-Field Splitting in the Excited Triplet States of Octahedral Hexanuclear Molybdenum(II) Clusters: [{Mo6X8}(n-C3F7COO)6]2– (X = Cl, Br, or I)

Soichiro Akagi,† Eri Sakuda,§ Akitaka Ito,# and Noboru Kitamura†,‡,*

†

Department of Chemical Sciences and Engineering, Graduate School of Chemical Sciences and Engineering and ‡Department of Chemistry, Faculty of Science, Hokkaido University, Sapporo, 060-0810, Japan §

Division of Chemistry and Materials Science, Graduate School of Engineering, Nagasaki University, Nagasaki, 852-8521, Japan

#

Graduate School of Engineering, School of Environmental Science and Engineering, Kochi University of Technology, Kochi, 782-8502, Japan

Corresponding Author Noboru Kitamura E-mail: [email protected] Phone: +81-(0)11-706-2697 Fax: +81-(0)11-706-4630

1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract Temperature (T)-dependent emission from [{Mo6X8}(n-C3F7COO)6]2– (X = Cl (1), Br (2), and I (3)) in optically transparent polyethyleneglycol dimethacrylate matrices were studied in 3 K < T < 300 K to elucidate the spectroscopic and photophysical properties of the clusters, in special reference to zero-magnetic-field splitting (zfs) in the lowest-energy excited triplet states (T1) of the clusters. The cluster complexes 1 and 2 showed the T-dependent emission characteristics similar to those of [{Mo6Cl8}Cl6]2–, while 3 exhibited emission properties different completely from those of 1 and 2. Such T-dependent emission characteristics of 1, 2, and 3 were explained successfully by the excited triplet state spin-sublevel (Φn, n = 1 – 4) model. The zfs energies between the lowest- (Φ1) and highest-energy spin-sublevels (Φ4), ∆E14, resulted by the first-order spin-orbit coupling, were evaluated to be 650, 720, and 1000 cm–1 for 1, 2, and 3, respectively. The emission spectra of 1, 2, and 3 in CH3CN at 298 K were reproduced very well by the ∆E14 values and the population percentages of Φn at 300 K. We also report that the ∆E14 values of the clusters correlate linearly with the fourth power of the atomic number (Z) of X: ∆E14 ∝{Z(X)}4.

2


Page 2 of 34

Page 3 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction We reported recently that the large temperature (T: 3 – 300 K) dependences of the emission characteristics (maximum energy (νem), band shape/full-width at half-maximum (fwhm) of the spectrum, and lifetime (τem)) of an octahedral hexanuclear metal cluster with the general

formula

of

[{M6(µ3-Q)8}L6]z–

(i.e.,

[{Mo6Cl8}Cl6]2–,

[{Re6S8}Cl6]4–,

or

[{W6Cl8}Cl6]2–, Chart 1) could be explained very well by an excited triplet state (T1) spin-sublevel (Φn) model.1 Since the cluster experiences strong spin-orbit coupling (SOC), the degenerated emissive T1 state splits in energy to the spin-sublevels: zero-magnetic-field splitting (zfs).1-4 Furthermore, it has been reported that [{Mo6Br8}Br6]2– or [{Re6Q8}L6]4– (Q = S or Se, L = X, CN, triethylphosphine, and so forth) at low temperature or in the excited state suffices Jahn-Teller distortion.5-8 According to our previous study, this results in splitting of the lowest spin-sublevel (ϕ1) into two (Φ1 and Φ2) as shown in Chart 2. In total, the four spin-sublevels (Φn, n = 1 – 4) in T1 contribute to the emission of the cluster. Since each Φn possesses own νem and τem, thermal communication between Φn results in large T-dependent νem and τem, reflecting the variation of the Boltzmann population between Φn with T.1,9 This is the origin of the large T-dependent emission characteristics of [{M6(µ3-Q)8}L6]z–. In the cases of [{Mo6Cl8}Cl6]2–, [{Re6S8}Cl6]4–, and [{W6Cl8}Cl6]2–, the splitting energies between the lowest- (Φ1) and highest-energy spin-sublevels (Φ4) in T1, ∆E14, were shown to be as large as 343 – 680 cm–1. Although the T-dependent emission characteristics of these clusters can be analyzed successfully by the Φn model irrespective of the nature of a metal atom (M = Mo(II), Re(III), or W(II)) and a capping ligand (Q = Cl or S),1 the applicability of the Φn model is worth checking for the emission properties of other [{M6(µ3-Q)8}L6]z–. Furthermore, the factor governing the ∆E14 value of [{M6(µ3-Q)8}L6]z– was unclear in the studies on [{Mo6Cl8}Cl6]2–, [{Re6S8}Cl6]4–, and [{W6Cl8}Cl6]2–. In the present paper, we focus our study on the unique emission characteristics of a series of [Mo6X8(n-C3F7COO)6]2– (X = Cl, Br, or I),10 among which the X = I cluster shows emission characteristics different largely from the X = Cl or Br 3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

cluster as well as from other [{M6(µ3-Q)8}L6]z– clusters. In 2011, we reported the emission properties of [Mo6X8(n-C3F7COO)6]2– (X = Cl (1), Br (2), or I (3)) in deaerated CH3CN at 298 K: Figure 1.10 As seen in the figure, 1 and 2 exhibit the broad emission spectrum (i.e., large fwhm) in a near-infrared region with the maximum wavelengths (λem) and fwhm being λem/fwhm = 745 nm/3980 cm–1 and 715 nm/4160 cm–1, respectively, whose emission characteristics are similar to those of other [{Mo6X8}L6]2– (X = Cl or Br; L = Cl, Br, and etc.) clusters.11-17 By contrast, 3 shows the emission at λem = 668 nm with a sharp spectral band width (fwhm = 2390 cm–1).10 It is worth emphasizing that the emission quantum yield (Φem) and τem of 3 in deaerated CH3CN are higher (0.59 – 0.60) and longer (303 – 359 µs),10,18 respectively, than the relevant value of 1 (Φem < 0.01 and τem = 14/0.4 µs (double-exponential decay)) or 2 (0.36 and 206 µs).10 The unique emission properties of a {Mo6I8}-core cluster, showing short λem, long τem, and small fwhm relative to those of the Cl or Br cluster, have been also confirmed for [{Mo6X8}(CF3COO)6]2– (X = Cl, Br, or I) by Kirakci et al.19 as well as for [{Mo6X8}(RCOO)6]2– (X = Br or I; RCOO = a series of aliphatic and aromatic carboxylate).18,20 Due to the intense and long-lived emission from a {Mo6I8}-core cluster, various {Mo6I8}-core based cluster complexes have been reported extensively18-21 and recognized as a possible candidate for strongly luminescent materials.22-25 Therefore, the unique emission properties of 3, differs largely from those of 1 or 2, are worth elucidating in detail. In our previous paper on [{Mo6X8}(n-C3F7COO)6]2–, we have suggested that the large differences in the emission characteristics between 3 and 1 or 2 would be explained by those in the zfs energies (∆E1n, n = 2 – 4) in the emissive T1 states of the clusters.10 We report here T-dependences (3 – 300 K) of the emission properties of 1, 2, and 3 dispersed homogeneously in polymer matrices and elucidate the origin of the unique emission characteristic of 3 on the basis of the analysis of zfs in T1.1-4,9 The present study demonstrates that the Φn model explains successfully the differences in the emission characteristic of 1, 2 and 3 in 3 K < T < 4


Page 4 of 34

Page 5 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


300 K. Furthermore, we report for the first time that the zfs energies of the clusters, ∆E1n (n = 3 or 4), correlate very well with the fourth-power of the atomic number (Z) of the capping ligand (X = Cl, Br, or I) in [{Mo6X8}(n-C3F7COO)6]2–.

Experimental Cluster Complexes. The tetra-n-butylammonium (TBA) salts of 1, 2, and 3 were prepared by similar procedures to those reported in the literatures.10,18-21 Typically, an acetone solution (20 mL, Wako Pure Chemicals Co. Ltd.) of (TBA)2[Mo6X8X6] (X = Cl, Br, or I, 0.19 mmol)26,27 and n-C3F7COOAg (1.35 mmol, Wako Pure Chemicals Co. Ltd.) was stirred at room temperature in the dark for 5 days. Precipitated AgX was filtered off and the yellow filtrate was evaporated to dryness. The crude product was purified by successive recrystallizations from an acetone - diethyl ether (Wako Pure Chemicals Co. Ltd.) mixture (v/v = 1/3): yield 58, 32, or 22% for 1, 2, or 3, respectively. Anal. Calcd for C56H72O12F42N2Mo6X8. 1 (X = Cl): Calcd: C, 25.65; H, 2.77; N, 1.07. Found: C, 25.73; H, 2.52; N, 0.98. 2 (X = Br) Calcd: C, 22.58; H, 2.44; N, 0.94. Found: C, 22.65; H, 2.52; N, 0.99. 3 (X = I) Calcd: C, 20.05; H, 2.17; N, 0.84. Found: C, 20.11; H, 2.16; N, 0.90. ESI-MS (CH3CN) for C56H72O12F42N2Mo6X8. 1 (X = Cl): m/z 1086.7 (M–2TBA). 2 (X = Br): m/z 1246.5 (M–2TBA). 3 (X = I): m/z 1434.5 (M– 2TBA). Preparations of the Cluster Complexes Dispersed in Polymer Matrices. In the present study, we used a polyethyleneglycol dimethacrylate (PEG-DMA) matrix as a medium, since a PEG-DMA matrix was highly transparent in the wavelength region studied and could be prepared

conveniently.28,29

Polyethyleneglycol

dimethacrylate

550,

purchased

from

Sigma-Aldrich Co, Ltd., was purified by passing through an aluminum (Sigma-Aldrich Co, Ltd.) column with acetone as an eluent. A mixture of polyethyleneglycol dimethacrylate 550 (~10 mL), a cluster complex dissolved in a minimum amount of acetone (~5×10–5 mol/dm3), and 2,2-azobis(2,4-dimethylvaleronitrile) (10 wt%, Wako Pure Chemicals Co. Ltd.) as a 5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

polymerization initiator in a Pyrex tube (inner diameter = 1 mm) was degassed in vacuo and, then, the solution was allowed polymerization at 50oC for 5 h in a water bath. The cluster sample in the tube was evacuated thoroughly prior to T-controlled spectroscopic and photophysical measurements to remove volatile chemicals and the resultant PEG-DMA block was isolated from the tube as a sample. Spectroscopic and Photophysical Measurements. The PEG-DMA matrix sample of a cluster complex was placed in a copper block of a liquid-He cryostat system (Oxford Instruments, OptistatCF) to control the sample temperature in ±0.1 K. A pulsed Nd3+:YAG laser (LOTIS TII, 355 nm, fwhm: ~10 ns) was used as an excitation light source. The corrected emission spectra and lifetimes of the complexes were determined by using a red-sensitive multichannel photodetector (Hamamatsu Photonics, PMA-11) and a streak camera (Hamamatsu Photonics, C4334), respectively.

Results and Discussion T-Dependent Emission Characteristics. Figure 2 shows the T-dependences of the emission spectra of 1, 2, and 3 in PEG-DMA matrices, and those of νem and fwhm are summarized in Figure 3. The emission spectrum of 1 was shifted to the lower-energy on going from 3 (νem = 12.9×103 cm–1, λem = 776 nm) to 70 K (12.5×103 cm–1, 800 nm). Subsequently, T-elevation above 70 K gave rise to increase in νem (upto ~13.2×103 cm–1 or ~756 nm at 300 K). Such spectral shifts upon T-elevation accompanied the increase in the fwhm value of the spectrum from 2460 cm–1 (at 3 K) to 3870 cm–1 (at 300 K). The cluster complex, 2, also showed analogous trends to those of 1: typically, νem = 12.7×103 (λem = 787), 12.2×103 (821), and 13.1×103 cm–1 (763 nm) at 3, 70, and 300 K, respectively, and fwhm varied from 2570 to 3790 cm–1 on going from 3 to 300 K. Such T-dependent emission shifts have been commonly observed for (TBA)4[{Re6S8}X6] (X = Cl, Br, and I), (TBA)2[{Mo6Cl8}Cl6], and (TBA)2[{W6Cl8}Cl6] in the crystalline and/or poly(methyl methacrylate) (PMMA) phases as 6


Page 6 of 34

Page 7 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


reported previously.1,9 In contrast, 3 exhibited T-dependent emission characteristics different totally from those of 1 and 2. As seen in Figures 2c and 3, the νem value of 3 was almost constant at 15.3×103 cm–1 (654 nm) in 3 K < T < 100 K, while that was shifted to the lower-energy above 100 K, which was an opposite T-dependent shift to that of 1 or 2 as seen clearly in Figure 3. Furthermore, the fwhm value of 3 (1500 and 2400 cm–1 at 3 and 300 K, respectively) was much smaller than that of 1 or 2 (~2500 cm–1 at 3 K and ~3900 cm–1 at 300 K) in the entire T-range studied (Figure 3b). Therefore, the small fwhm value of 3 compared to that of 1 or 2 is an inherent character. Figures 2 and 3 demonstrate explicitly that the T-dependences of νem and fwhm of 3 are unusual among those of 1, 2, and 3. The emission decay profiles of the complexes were fitted by single exponential functions irrespective of X and T (3 – 300 K), as a typical example being reported in Figure S1 in the Electronic Supporting Information (ESI). The T-dependences of τem observed for 1, 2, and 3 are summarized in Figure 4. In the case of 1 or 2, the τem value decreased sharply upon T-elevation from 3 to ~30 K and, then, decreased gradually above 30 K. From 3 to 300 K, the τem values of 1 and 2 decreased from 670 to 160 µs and from 330 to 115 µs, respectively. On the other hand, 3 showed a relatively monotonous decrease in τem from 340 µs at 3 K to 235 µs at 300 K. It is worth emphasizing that the change in τem observed for 3 upon T-variation from 3 (340 µs) to 300 K (235 µs) is as small as 105 µs: τem(3)/τem(300) = 0.31. Such a small T-dependence of τem in the T range of 3 ~ 300 K has not been observed for other hexanuclear metal clusters: τem(3)/τem(300) = 0.65 (1), 0.76 (2), 0.75 ([{Mo6Cl8}Cl6]2–), 0.93 ([{Re6S8}Cl6]4–), and 0.81 ([{W6Cl8}Cl6]2–). Since 1, 2, and 3 are isostructural and isoelectronic with one another, the T-dependent νem, fwhm, and τem of the clusters (Figures 2, 3, and 4) should be explained by a common emission model. Analysis of T-dependent Emission Characteristics by Spin-sublevel Model. Among the four spin-sublevels (Φn), split in energy by SOC and Jahn-Teller distortion in the energy order of Φ1< Φ2 < Φ3 < Φ4 (see Chart 2), the emission from the Φ4 level to the ground state (S0) is an 7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 34

electronically-allowed dipole transition and, therefore, the emission lifetime is predicted to be short.1 On the other hand, the emission transitions from Φ1, Φ2, and Φ3 to S0 are forbidden and gained allowed characters only by vibronic coupling and, thereby, the τem values of these three levels would be long.1 At low temperature, the emission from Φ1 contributes to an observed emission and observed τem would be long, while the contribution of the emission from Φ4 becomes dominant at a higher T and this leads to a decrease in observed τem.1 Phenomenologically, this explains the experimental results in Figure 4. According to the Φn model,1 the T-dependent τem data (τ(T)) of a cluster are analyzed by eq 1, τ (T ) =

∑g n exp( – ∆E1n / k BT ) g exp( – ∆E1n / k BT ) ∑ n τn

(1)

where gn and τn are the multiplicity (g1 = g3 = 2, g2 = 1, and g4 = 3) and emission lifetime of Φn, respectively. The τ(T) data were then analyzed by eq 1 as the fitting results were shown by the solid curves in Figure 4. As seen in the figure, the τ(T) data of 1, 2, and 3 were fitted very well by the ∆E1n and τn values shown in Figure 5: the correlation coefficients (R2) of the fittings were 0.998 for all of the τ(T) data. The ∆E13 and ∆E14 values were in the range of 55 – 120 cm–1 and 650 – 1000 cm–1, respectively, while the ∆E12 value was fixed at 4 cm–1 for the simulations. The ∆E13 and ∆E14 values of 1 (55 and 650 cm–1, respectively) agree very well with the relevant values of [{Mo6Cl8}Cl6]2– (53 and 653 cm–1, respectively),1 suggesting that the ∆E1n (n = 3, 4) values are almost constant for an {Mo6Cl8}-core cluster. On the other hand, the ∆E13 and ∆E14 values increased in the sequence, 1 (X = Cl, 55 and 650 cm–1) < 2 (X = Br, 68 and 720 cm–1) < 3 (X = I, 120 and 1000 cm–1). These findings demonstrate that one of the important factors governing the zfs energies (∆E1n, n = 3 or 4) is the nature of a {Mo6X8}-core structure. We will discuss further the point in the following sections. On the basis of the ∆E1n values, furthermore, we analyzed the emission spectrum of the cluster at a given T as the sum of the Φn emission spectrum by the following equations.1-4 8


Page 9 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


∆E12 k BT

I (ν , T ) = k r1 F (ν – ν1 ) + k r2 F (ν – ν2 ) exp + k r3 F (ν – ν3 ) exp

∆ E13 ∆ E14 + k r4 F ( ν – ν4 ) exp k BT k BT

(2)

In eq 2, krn is the radiative rate constant of Φn relative to that of Φ1 (i.e., kr1 = 1.0). F(ν – νn) and νn are a spectral Gaussian function and the emission maximum energy of each Φn, respectively, and F(ν – νn) is given by eq 3.1 F (ν – ν n ) =

1 ( fwhm (Φn ))

π

exp[ –

2(ν – ν n ) 2 ] ( fwhm (Φn )) 2

(3)

2

In eq 3, fwhm(Φn) represents fwhm of the emission spectrum of each Φn, and we assume that fwhm(Φn) at a given T is the same for n = 1 – 4, but varies with T. Figure 6 shows the observed (shown by black curves) and simulated emission spectra (shown by given colors) of

1, 2, and 3 at several T. The νn values evaluated are included in Figure 5. Figure 6 demonstrates clearly that the observed spectrum can be fitted very well by eqs 2 and 3, irrespective of X and T with R2 being ~1.00. Although the T-dependent emission shift of 3 is different largely from that of 1 or 2, the present Φn model explains successfully such different T-dependent emission characteristics of 1, 2, and 3.

Origin of the Emission Characteristics of [Mo6X8(n-C3F7COO)6]2–. The present spectral analysis demonstrates that the major contributions to the emission spectra of 1 and 2 at 300 K are the emissions from Φ3 and Φ4 (Φ3 = 45 % and Φ4 = 55 % for 1; Φ3 = 72 % and Φ4 = 28 % for 2), while that of 3 is explained by the large contributions of the emissions from the Φ2 and Φ3: Φ2 = 50 %, Φ3 = 47 %, and Φ4 = 3 %. Thermal population from Φ1 to Φ4 at a given T is governed by ∆E14 and, thereby, the contribution of the Φ4 emission to an observed spectrum at 300 K should become smaller with an increase in ∆E14. In practice, the contribution of the Φ4 emission to the observed spectrum at 300 K decreases in the sequence, 1 (∆E14 = 650 cm–1, Φ4 = 55%) > 2 (720 cm–1, 28%) > 3 (1000 cm–1, 3%). The percentages of Φn contributed to the emission spectrum of each cluster (Φn%) 9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

mentioned above explain very well the emission characteristics of the clusters in CH3CN at 298 K (Figure 1). First, almost equal contributions of Φ3 (ν3 = 13.1×103 cm–1) and Φ4 (ν4 = 13.8×103 cm–1) or Φ2 (ν2 = 15.4×103 cm–1) and Φ3 (ν3 = 14.4×103 cm–1) to the emission spectrum of 1 or 3, respectively, demonstrate the relevant average νem value to be 13.4×103 (λem = 744 nm) or 14.9×103 cm–1 (672 nm), respectively, which agrees very well with the observed νem value in Figure 1: 13.4×103 cm–1 (745 nm) for 1 and 15.0×103 cm–1 (668 nm) for

3. On the basis of the simulated fwhm(Φn) values of 1 (3260 cm–1) and 3 (1460 cm–1) at 300 K shown in Figure 7, furthermore, the fwhm values of the whole spectra of 1 and 3 in CH3CN at 298 K are predicted to be 3980 and 2390 cm–1, respectively, which is also in good agreement with the observed values: 3870 and 2400 cm–1 for 1 and 3, respectively. Similarly, the observed νem and fwhm values of 2 in Figure 1 can be also reproduced very well by the Φn percentage and fwhm(Φn) parameters observed for 2 as reported in Figure 7. It is worth pointing out that the fwhm(Φn) value of 3 is much smaller than that of 1 or 2: 1 (3260 cm–1) ~

2 (3420 cm–1) > 3 (1460 cm–1). The small fwhm(Φn) value of 3 is the origin of the sharp emission band of 3 (fwhm = 2390 cm–1) relative to that of 1 (3980 cm–1) or 2 (4160 cm–1) in CH3CN at 298 K: Figure 1.10 Furthermore, since {Mo6I8}-core clusters hitherto reported show commonly narrow emission bands,10-25 the small fwhm value will be the intrinsic character of an {Mo6I8}-core cluster. The fwhm value of an emission spectrum observed at 3 K will be determined primarily by the slope of the potential energy curve in S0 at the T1 geometry. In the present case, the broad Φn emission spectra of 1 and 2 in Figure 7 will be ascribed to the radiative transition from Φn to the potential curve in the relevant ground state with a rather steep slope, while the sharp Φn spectrum of 3 indicates the radiative transition occurs from T1 to near the bottom of the potential curve in S0.30 Such simple considerations on fwhm/fwhm(Φn) demonstrate that the T1 states of 1 and 2 are considerably distorted probably due to Jahn-Teller effects, while the T1 geometry in 3 is less distorted.30 Second, since the Φ4 emission is an electronically-allowed transition as mentioned 10


Page 10 of 34

Page 11 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


before, a large contribution of the Φ4 emission to an observed emission should give rise to a decrease in τem. In CH3CN at 298 K, the τem value decreases in the sequence, 3 (359 µs) > 2 (206 µs) > 1 (~14 µs),4 which is in good accordance with the increasing order of Φ4% and, thus, the decreasing order of ∆E14. The difference in the Φem value between 1 (< 0.01), 2 (0.36), and 3 (0.59 – 0.60)10,18,19 will be also explained by the similar context to that of τem, since Φem of the cluster is determined by the radiative rate constant (kr) and each Φn possesses an individual krn value (kr1 = 1.0 and krn (n = 2 – 4) is a relative value), whose value evaluated by eq 2 are included in Figure 5. The kr2 values of the clusters were almost constant at 1.5 – 2.3, irrespective X. This will be reasonable since splitting of the lowest-energy spin-sublevel to Φ1 and Φ2 (∆E12 = 4 cm–1) is due to Jahn-Teller distortion. Contrary, kr3 and kr4 increase in

1 (kr3 = 2.3, kr4 = 2.5) < 2 (kr3 = 3.0, kr4 = 64) < 3 (kr3 = 69, kr4 = 2210). Unfortunately, although we are not aware of the absolute krn values in the present stage of the investigation, we suppose the large increase in kr4 in the sequence of 1 < 2 < 3 will explain the increasing order of Φem of the clusters. The characteristic T-dependent emission shifts of 1 and 2, showing lower- and subsequent higher-energy shifts from 3 to 300 K, have been also observed for [{Mo6Cl8}Cl6]2– and, thus, its origin can be explained by the similar arguments to those for [{Mo6Cl8}Cl6]2–.1 Briefly, the emission maximum energy of each Φn (νn in cm–1) does not increase in the sequence, ν1 < ν2 < ν3 < ν4, but increases in the order of ν2 (12.5×103) < ν1 (12.9×103) < ν3 (13.1×103) < ν4 (13.8×103) for 1 or ν2 (12.2×103) < ν3 (12.4×103) < ν1 (12.5×103) < ν4 (13.5×103) for 2. The energy order of νn observed for 1 is essentially the same with that for [{Mo6Cl8}Cl6]2–: common for an {Mo6Cl8}-core structure. Although the main contribution of the emission from 1 or 2 at 3 K is that from Φ1 (ν1 = 12.9×103 or 12.5×103 cm– 1

for 1 or 2, respectively), T-elevation to 70 K gives rise to the increase in the contribution of

the Φ2 emission (ν2 = 12.5×103 or 12.2×103 cm–1 for 1 or 2, respectively) and, thus, the observed νem is shifted to the lower-energy. Since further T-elevation leads to the participation 11



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

of the Φ3 and Φ4 emissions, the observed emission spectrum of 1 or 2 as the sum of the emissions from Φn (n = 1 – 4) shifts to the lower-energy as expected from the ν3 and ν4 values of 1 or 2: see Figure 5. In the case of 3, on the other hand, since the energy order of νn (in cm– 1

) is ν4 (13.5×103) < ν3 (14.4×103) < ν1 (15.3×103) ≤ ν2 (15.4×103) and ∆E13 (120 cm–1) is

relatively large compared to that of 1 (55 cm–1) or 2 (68 cm–1), the main contributions to the emission spectrum of 3 in 3 K < T < 70 K are the Φ1 and Φ2 emissions and, thus, the νem value of the cluster is almost unchanged as seen in Figures 2 and 3. The emission spectrum of 3 shifts to the lower-energy owing to the participations of the Φ3 and Φ4 emissions above 70 K. These discussions described above demonstrate that the Φn model explain satisfactorily the emission characteristic of 1, 2, and 3 in 3 K < T < 300 K. The difference in νn between Φ1 and Φn (∆νn) is governed by ∆E1n and the energy of the Franck-Condon ground state responsible for each Φn emission transition. Knowing ∆E1n and ∆νn, we evaluated the energy differences between the Franck-Condon ground states for the Φn emissions (∆E’) as the data were included in Figure 5. We suppose that ∆E’ will be determined by the displacement of the potential surface between the Φn level and the relevant Franck-Condon ground state along a vibrational coordinate as reported previously.1 Ramirez-Tagle et al. have reported the computational calculations on the vibrational frequencies of [{Mo6X8}Y6]2– (X = halide, Y = halide), and these clusters show active vibrations assigned to the Mo-Mo, Mo-X, and Mo-Y stretching modes in the ranges of 136 – 350, 151 – 399 , and 46 – 355 cm–1, respectively.31 Furthermore, while the stretching vibrational frequency of the Mo-O bond (coordinated O atom in the terminal ligand) has been reported to be 670 cm–1,32 the present clusters show the IR bands at around 670 cm–1 irrespective X. In the case of 1 and 2, the ∆E’ values were evaluated to be ~130, 290, and 670 cm–1, which might correspond roughly to the Mo-Mo, Mo-X, and Mo-O stretching frequencies, respectively. Since the excited state of [{Mo6X8}L6]2– has been reported to be best characterized by the {Mo6X8}-core centered excited state,31,33,34 the observation of ∆E’ 12


Page 12 of 34

Page 13 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


relevant to the Mo-Mo, Mo-X, and Mo-O stretching modes might be reasonable. On the other hand, the ∆E’ values observed for 3 were very large (1090 – 3380 cm–1), though the Mo-I stretching mode in {Mo6I8}4+ is 152 – 154 cm–1.35-37 In the present stage of the investigation, we cannot assign the vibrational modes responsible for the ∆E’ values of 3. Since the T-dependences of νem and τem of 3 are relatively small compared to those of 1 or 2 as seen in Figures 3 and 4, the ∆E’ values evaluated for 3 might be somewhat erroneous. Further detailed vibrational spectroscopy study is absolutely necessary to explain the ∆E’ values of 1,

2, and 3.1 Factors Governing Zero-Magnetic-Field Splitting Energies in the Excited Triplet States of [Mo6X8(n-C3F7COO)6]2–.

The

characteristic

T-dependent

emission

from

[{Mo6X8}(n-C3F7COO)6]2–, [{Mo6Cl8}Cl6]2–, [{Re6S8}Cl6]4–, and [{W6Cl8}Cl6]2– were explained by the Φn model (n = 1 – 4) as demonstrated in the previous1 and present papers. Analysis of the τ(T) and I(ν, T) data of these clusters by the Φn model has provided the zfs parameters: ∆E1n, τn, and νn. Among these zfs parameters, the ∆E1n values (n = 3 and 4) are related directly to the strength of SOC and, thus, are worth discussing in some more detail. The SOC parameter as a measure of the strength of SOC experienced by a molecule, ξ, is related to the atomic number (Z) of a constituted atom through eq 4,38,39

   e2h2  Z4 ξ= 2 2 3  2m c a0  n3 (l + 1)(l + 1 )l  2  

(4)

where e, h, m, a0, n, and l (≥1) are the elementary electric charge, the Planck constant, the mass of an electron, the Bohr radius, the principal and azimuthal quantum numbers of a molecule, respectively. It is very important to note that eq 4 demonstrates that the ξ value is proportional to the fourth power of Z. In the case of the T1 state of [Ru(bpy)3]2+ or [Os(bpy)3]2+ (bpy = 2,2’-bipyridine), the ∆E13 value has been reported to be 61 or 210 cm– 1 40,41

,

respectively. According to the Z and ξ values of a Ru (Z = 44, ξ = 1042 cm–1) or Os 13



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 34

atom (76, 3381 cm–1),39 the relatively small ∆E13 value of [Ru(bpy)3]2+ compared to that of [Os(bpy)3]2+ will be the reasonable consequence. The ∆E14 values of 1, 2, 3, and [{Mo6Cl8}Cl6]2–, corresponding to the zfs energies of the T1 states of the clusters by the first-order SOC, are as large as 650 – 1000 cm–1 and, to the best of our knowledge, the values are the largest among those of various transition metal complexes. The Z and ξ values of an Mo atom are 42 and 678 cm–1,39 respectively, and these values are smaller than the relevant value of a Ru or Os atom mentioned above. The very large ∆E14 value of the cluster would be due to the very heavy atom environments (i.e., {Mo6X8}4+-core centered excited state), although SOC itself is one-atom-center perturbation. It is worth emphasizing again that the ∆E14 value increases in the sequence, X = Cl (650 (1) – 662 cm–1 ([{Mo6Cl8}Cl6]2–) < X = Br (2, 720 cm–1) < X = I (3, 1000 cm–1), whose sequence agrees very well with those of the Z and ξ values of X: Z = 17 (Cl) < 35 (Br) < I (53) and ξ (in cm–1) = 587 (Cl) < 2460 (Br) < 5069 (I).39 Similar tendency to that of ∆E14 can be also seen for ∆E13. To check an applicability of eq 4 to the relationship between ξ and ∆E1n (n = 3 and 4), we plotted the observed ∆E1n values against the Z4 values of X: ∆E1n vs. {Z(X)}4. As the results are shown in Figure 8, we found good correlations between ∆E1n (n = 3 and 4) and {Z(X)}4 with R2 being 0.987 and 0.999 for ∆E13 and ∆E14, respectively. It is worth noting that the linear dependence between ∆E1n (n = 3 or 4) and {Z(X)}4 in Figure 8 is not fortuitous, since our preliminary study indicates that the ∆E14 value of [{Mo6X8}Y6]2– (X, Y = Cl, Br, or I) correlates very well with {Z(X)}4 or {Z(Y)}4 for a given Y or X, respectively, which will be reported in a separate publication (unpublished results). We expect similar relationship between ∆E1n and {Z(X)}4 or {Z(Y)}4 to those of [{Mo6X8}Y6]2– mentioned above will be also obtained for a series of [{W6X8}Y6]2–. An octahedral hexanuclear metal cluster provides good opportunities

to

study

experimentally

and

theoretically

spin-orbit

coupling

zero-magnetic-field splitting of a transition metal complex in the excited triplet state.

14


and

Page 15 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Conclusions We studied the applicability of the excited triplet state spin-sublevel (Φn, n = 1 – 4) model to explain the large T-dependent emission characteristics of 1, 2, and 3 in 3.0 K < T < 300 K. Although 3 showed T-dependent emission properties different largely from those of 1 or 2, it was demonstrated that the Φn model explained almost satisfactorily the emission characteristics of the three clusters. Since we have already shown that the T-dependent emission characteristics of [{Mo6Cl8}Cl6]2–, [{W6Cl8}Cl6]2–, and [{Re6S8}Cl6]4– are elucidated based on the Φn model,1 we are convinced that the Φn model can explain T-dependent emission properties of [{M6(µ3-Q)8}L6]z– irrespective of the nature of M, Q, and L. In the present study, we found that the ∆E13 and ∆E14 values (in cm–1) increased in the sequence, 1 (∆E13 = 55 and ∆E14 = 650) < 2 (∆E13 = 68 and ∆E14 = 720) < 3 (∆E13 = 120 and ∆E14 = 1000). Reflecting the very large ∆E14 value of 3, it was shown that the contribution of the Φ4 emission at 300 K was very small (3%) and the main contributions to the observed emission were Φ2 (50%) and Φ3 (47%), while the emission spectra of 1 and 2 were characterized by the emissions from Φ3 and Φ4: Φ3 =45 % and Φ4 = 55% for 1 and Φ3 = 72% and Φ4 = 28% for 2. The variations of the Φn emission percentages at 300 K with X mentioned above explained very well the νem values of the clusters as the sum of the emission from Φn, indicating that the higher-energy emission of 3 compared to that of 1 or 2 could be also explained by the Φn model. The fwhm values of Φn (fwhm(Φn) in cm–1) at 300 K were evaluated to be 1 (3260) ~ 2 (3420) > 3 (1460) and it was shown that the narrow emission band of 3 relative to those of 1 and 2 was the intrinsic property of a {Mo6I8}-core cluster complex. As the very important result, furthermore, the ∆E13 and ∆E14 values were shown to correlate linearly with {Z(X)}4 as shown in Figure 8. This demonstrates clearly that spin-orbit-coupling is very important to discuss the spectroscopic and photophysical properties of an octahedral hexanuclear metal cluster in a wide range of temperature. Further 15



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 34

work on zero-magnetic-field splitting in the excited triplet states of [{Mo6X8}Y6]2– (X, Y = Cl, Br, or I) will be reported in a separate publication (unpublished results).

Supporting Information Description Emission lifetimes of 1 – 3 and typical example of emission decay profiles in PEG-DMA matrices in 3 K < T < 300 K.

Acknowledgement The authors are indebted to Dr. M. N. Sokolov, Dr. K. A. Brylev, and their group members at Nikolaev Institute of Inorganic Chemistry, the Siberian Branch of Russian Academy of Sciences, for generous gifts of the cluster complexes, 1, 2, and 3, in the initial stage of this research. NK acknowledges a grant-in-aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of the Japanese Government for the support of the research No. 26248022 (Grant-in-Aid for Scientific Research (A)).

Reference (1)

Kitamura, N.; Kuwahara, Y.; Ueda, Y.; Itoh, Y.; Ishizaka, S.; Sasaki, Y.; Tsuge, K.; Akagi, S. Excited Triplet States of [{Mo6Cl8}Cl6]2–, [{Re6S8}Cl6]4–, and [{W6Cl8}Cl6]2– Clusters. Bull. Chem. Soc. Jpn. In press. DOI:10.1246/bcsj.20170168.

(2)

Saito, Y.; Tanaka, H. K.; Sasaki, Y.; Azumi, T. Temperature Dependence of the Luminescence

Lifetime

of

Hexanuclear

Molybdenum(II)

Chloride

Cluster.

Identification of Lower Excited Triplet Sublevels. J. Phys. Chem. 1985, 89, 4413-4415. (3)

Azumi, T.; Saito, Y. Electronic Structures of the Lower Triplet Sublevels of Hexanuclear Molybdenum(II) Chloride Cluster. J. Phys. Chem. 1988, 92, 1715-1721.

(4)

Miki, H.; Ikeyama, T.; Sasaki, Y.; Azumi, T. Phosphorescence from the Triplet Spin 16


Page 17 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Sublevels of a Hexanuclear Molybdenum(II) Chloride Cluster Ion, [Mo6Cl14]2–. Relative Radiative Rate Constants for Emitting Sublevels. J. Phys. Chem. 1992, 96, 3236-3239. (5)

Gray, T. G.; Rudzinski, C. M.; Meyer, E. E.; Nocera, D. G. Excited-State Distortion of Rhenium(III) Sulfide and Selenide Clusters. J. Phys. Chem. 2002, 108, 3238-3243.

(6)

Baudron, S. A.; Deluzet, A.; Boubekeur, K.; Batail, P. Jahn-Teller Distortion of the Open-Shell 23-Electron Chalcogenide Rhenium Cluster Cores in Crystals of the Series, {[Re6Q8]3+(X−)6}3− (Q = S, Se; X = Cl, CN). Chem. Commun. 2002, 2124-2125.

(7)

Deluzet, A.; Duclusand, H.; Sautet, P.; Borsch, S. Electronic Structure of Diamagnetic and Paramagnetic Hexanuclear Chalcogenide Clusters of Rhenium. Inorg. Chem.

2002, 41, 2537-2542. (8)

Costuas, K.; Garreau, A.; Buloi, A.; Fontaine, B.; Cuny, J.; Gautier, R.; Mortier, M.; Molard, Y.; Duvail, J.-L.; Faulques, E. et al. Combined Theoretical and Time-Resolved Photoluminescence Investigations of [Mo6Bri8Bra6]2− Metal Cluster Units: Evidence of Dual Emission. Phys. Chem. Chem. Phys. 2015, 17, 28574-28585.

(9)

Kitamura, N.; Ueda, Y.; Ishizaka, S.; Yamada, K.; Aniya, M.; Sasaki, Y. Temperature Dependent Emission of Hexarhenium(III) Clusters [Re6S8X6]4– (X = Cl–, Br–, and I–): Analysis by Four Excited Triplet-State Sublevels. Inorg. Chem. 2005, 44, 6308-6313.

(10)

Sokolov, M. N.; Mihailov, M. A.; Peresypkina, E. V.; Brylev, K. A.; Kitamura, N.; Fedin, V. P. Highly Luminescent Complexes [Mo6X8(n-C3F7COO)6]2– (X = Br, I). Dalton Trans. 2011, 40, 6375-6377.

(11)

Sokolov, M. N.; Mikhailov, M. A.; Brylev, K. A.; Virovets, A. V.; Vicent, C.; Kompankov, N. B.; Kitamura, N.; Fedin, V. P. Alkynyl Complexes of High-Valence Clusters. Synthesis and Luminescence Properties of [Mo6I8(C≡CC(O)OMe)6]2– – the First Complex with Exclusively Organometallic Outer Ligands in the Family of 17



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Octahedral {M6X8} Clusters. Inorg. Chem. 2013, 52, 12477-12481. (12)

Efremova, O. A.; Shestopalov, M. A.; Chirtsova, N. A.; Smolentsev, A. I.; Mironov, Y. V.; Kitamura, N.; Brylev, K. A.; Sutherland, A. J. A Highly Emissive Inorganic Hexamolybdenum Cluster Complex as a Handy Precursor for the Preparation of New Luminescent Materials. Dalton Trans. 2014, 43, 6021-6025.

(13)

Wilson, W. B.; Stark, K.; Johnson, D. B.; Ren, Y.; Ishida, H.; Cedeño, D. L.; Szczepura, L. F. Photophysical Properties of a Series of Rhenium Selenide Cluster Complexes Containing Nitrogen-Donor Ligands. Eur. J. Inorg. Chem. 2014, 2254-2261.

(14)

Mikhailov, M. A.; Brylev, K. A.; Virovets, A. V.; Gallymov, M. R.; Novozhilov, I.; Sokolov, M. N. Complexes of {Mo6I8} with Nitrophenolates: Synthesis and Luminescence. New J. Chem. 2016, 40, 1162-1168.

(15)

Kirakci, K.; Kubát, P.; Kučeráková, M.; Šícha, V.; Gbelcová, H.; Loveckcá, P.; Grznárová, P.; Ruml, T.; Lang, K. Water-soluble Octahedral Molybdenum Cluster Compounds Na2[Mo6I8(N3)6] and Na2[Mo6I8(NCS)6]: Syntheses, Luminescence, and in vitro Studies. Inorg. Chim. Acta 2016, 441, 42-49.

(16)

Riehl, L.; Ströbele, M.; Enseling, D.; Jüstel, T.; Meyer, H.-J. Molecular Oxygen Modulated Luminescence of an Octahedro-hexamolybdenum Iodide Cluster Having Six Apical Thiocyanate Ligands. Z. Anorg. Allg. Chem. 2016, 642, 403-408.

(17)

Efremova, O. A.; Vorotnikov, Y. A.; Brylev, K. A.; Vorotnikova, N. A.; Novozhilov, I. N.; Kuratieva, N. V.; Edeleva, M. V.; Benoit, D. M.; Kitamura, N.; Mironov, Y. V. et al. Octahedral Molybdenum Cluster Complexes with Aromatic Sulfonate Ligands. Dalton Trans. 2016, 45, 15427-15435.

(18)

Mikhailov, M. A.; Brylev, K. A.; Abramov, P. A.; Sakuda, E.; Akagi, S.; Ito, A.; Kitamura, N.; Sokolov, M. N. Synthetic Tuning of Redox, Spectroscopic, and Photophysical Properties of {Mo6I8}4+ Core Cluster Complexes by Terminal 18


Page 18 of 34

Page 19 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Carboxylate Ligands. Inorg. Chem. 2016, 55, 8437-8445. (19)

Kirakci, K.; Kubát, P.; Langmaier, J.; Polívka, T.; Fuciman, M.; Fejfarová, K.; Lang, K. A Comparative Study of the Redox and Excited State Properties of (nBu4N)2[Mo6X14] and (nBu4N)2[Mo6X8(CF3COO)6] (X = Cl, Br, or I). Dalton Trans.

2013, 42, 7227-7232. (20)

Akagi, S.; Fujii, S.; Horiguchi, T.; Kitamura, N. pKa(L) Dependences of Structural, Electrochemical, and Photophysical Properties of Octahedral Hexamolybdenum(II) Clusters: [Mo6X8L6]2– (X = Br or I; L = carboxylate). J. Clust. Sci. 2017, 28, 757-772.

(21)

Kirakci, K.; Kubát, P.; Dušek, M.; Fejfarová, K.; Šícha, V.; Mosinger, J.; Lang, K. A Highly Luminescent Hexanuclear Molybdenum Cluster – A Promising Candidate toward Photoactive Materials. Eur. J. Inorg. Chem. 2012, 3107-3111

(22)

Cordier, S.; Dorson, F.; Grasset, F.; Molard, Y.; Fabre, B.; Haneda, H.; Sasaki, T.; Mortier, M.; Ababou-Girard, S.; Perrin, C. Novel Nanomaterials Based on Inorganic Molybdenum Octahedral Clusters. J. Clust. Sci. 2009, 20, 9-21.

(23)

Efremova, O. A.; Brylev, K. A.; Vorotnikov, Y. A.; Vejsadová, L.; Shestopalov, M. A.; Chimonides, G. F.; Mikes, P.; Topham, P. D.; Kim, S.-J.; Kitamura, N. et al. Photoluminescent Materials Based on PMMA and Highly-emissive Octahedral Molybdenum Metal Cluster Complex. J. Mater. Chem. C. 2016, 4, 497-503.

(24)

Vorotnikov, Y. A., Efremova, O. A.; Vorotnikova, N. A.; Brylev, K. A.; Edeleva, M.; Tsygankova, A. R.; Smolentsev, A. I.; Kitamura, N.; Mironov, Y. V.; Shestopalov, M. A. On the Synthesis and Characterization of Luminescent Hybrid Particles: Mo6 Metal Cluster Complex/SiO2. RSC Adv. 2016, 6, 43367-43375.

(25)

Amela-Cortes, M.; Molard, Y.; Paofai, S.; Desert, A.; Duvail, J.-L.; Naumov, N. G.; Cordier, S. Versatility of the Ionic Assembling Method to Design Highly Luminescent PMMA Nanocomposites Containing [M6Qi8La6]n– Octahedral Nano-building Blocks. Dalton Trans. 2016, 45, 237-245. 19



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(26)

Page 20 of 34

Kolesnichenko, V.; Messerle, L. Facile Reduction of Tungsten Halides with Nonconventional, Mild Reductants. 2. Four Convenient, High-Yield Solid-State Synthesis

of

the

Hexatungsten

Dodecachloride

Clusters

W6Cl12

and

(H3O)2[W6(µ3-Cl)8Cl6](OH2)x, Including New Cation-Assisted Ternary Routes. Inorg. Chem. 1998, 37, 3660-3663. (27)

Kirakci, S.; Cordier, S.; Perrin, C. Synthesis and Characterization of Cs2Mo6X14 (X = Br or I) Hexamolybdenum Cluster Halides: Efficient Mo6 Cluster Precursor for Solution Chemistry Synthesis. Z. Anorg. Allg. Chem. 2005, 631, 411-416.

(28)

Knight, T. E.; Goldstein, A. P.; Brennaman, M. K.; Cardolaccia, T.; Pandya, A.; De Simone, J. M.; Meyer, T. J. Influence of the Fluid-to-Film Transition on Photophysical Properties of MLCT Excited States in a Polymerizable Dimethacrylate Fluid. J. Phys. Chem. B 2011, 115, 64-70.

(29)

Zanoni, K. P. S.; Kariyazaki, B. K.; Ito, A.; Brennaman, M. K.; Meyer, T. J. Blue-Green Iridium(Ⅲ) Emitter and Conprehensive Photopysical Elucidation of Heteroleptic Cyclometalated Iridium(Ⅲ) Complexes. Inorg. Chem. 2014, 53, 4089-4099.

(30)

We acknowledge one of the reviewers for the comments on the possible explanations of fwhm(Φn).

(31)

Ramirez-Tagle, R.; Arratia-Pérez, R. Electronic Structure and Molecular Properties of the [Mo6X8L6]2–; X = Cl, Br, I; L = F, Cl, Br, I Clusters. Chem. Phys. Lett. 2008, 460, 438-441.

(32)

Hardcastle, F. D.; Wachs, I. E. Determination of Molybdenum–Oxygen Bond Distances and Bond Order by Raman Spectroscopy. J. Raman Spectrosc. 1990, 21, 683-691.

(33)

Ramirez-Tagle, R.; Arratia-Pérez, R. The Luminescent [Mo6X8(NCS)6]2– (X = Cl, Br, I) Clusters?: A Computational Study Based on Time-Dependent Density Functional 20


Page 21 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Theory Including Spin-Orbit and Solvent-Polarity Effects. Chem. Phys. Lett. 2008, 455, 38-41. (34)

Ramirez-Tagle, R.; Arratia-Pérez, R. Pyridine as Axial Ligand on the [Mo6Cl8]4+ Core Switches off Luminescence. Chem. Phys. Lett. 2009, 475, 232-234.

(35)

Preetz, W.; Harder, K. Synthesis, Structure and Properties of the Cluster Anions [{Mo6Cli8}Xa6]2– with Xa ≡ F, Cl, Br, I*. J. Alloys Compds. 1992, 183, 413-429.

(36)

Preetz, W.; Bublitz, D. Darstellung, Kristallstruktur und Spektroskopische Eigenschaften der Clusteranionen [{Mo6Bri8}Xa6]2– mit Xa = F, Cl, Br, I. Z. Anorg. Allg. Chem. 1994, 620, 234-246.

(37)

Brückner,

P.;

Preetz,

W.;

Pünjer,

M.

Darstellung,

Kristallstruktur,

NMR-Schwingungsspektren und Normalkoordinatenanalyse der Clusteranionen [{Mo6Ii8}Xa6]2–, Xa = F, Cl, Br, I. Z. Anorg. Allg. Chem. 1997, 623 8-17. (38)

Condon, E. U.; Shortley, G. H. The Theory of Atomic Spectra; Cambridge University Press: London, U.K., 1953.

(39)

Montalti, M.; Credi, A.; Prodi, L.; Gandolfi, M. T. Handbook of Photochemistry, 3rd ed.; Taylor & Francis: Boca Raton, U.S.A., 2006.

(40)

Crosby, G. A.; Hipps, K. W.; Elfring Jr, W. H. On the Appropriateness of Assigning Spin Labels to Excited States of Inorganic Complexes. J. Am. Chem. Soc. 1974, 96, 629-630.

(41)

Yersin, H.; Strasser, J. Triplets in Metal-Organic Compounds. Chemical Tunability of Relaxation Dynamics. Coord. Chem. Rev. 2000, 208, 331-364.

Synopsis Temperature (T)-dependent emission spectra and lifetimes of [{Mo6X8}(n-C3F7COO)6]2– (X = Cl (1), Br (2), and I (3)) in 3.0 < T < 300 K were analyzed by an excited triplet state spin-sublevel (Φn, n = 1 – 4) model and the Φn model parameters explained successfully the 21



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

emission characteristics of the clusters at given T. The splitting energy between the lowest(Φ1) and third- (Φ3) or fourth-spin-sublevels (Φ4) (∆E13 or ∆E14, respectively) was shown to correlate linearly with the fourth-power of the atomic number (Z) of X: ∆E1n (n = 3 or 4) vs. {Z(X)}4.

Figure Captions Chart 1. Structure of an octahedral hexanuclear metal cluster [M6(µ3-Q)8L6]z–. Chart 2. Excited triplet state spin-sublevel models for the T-dependent emission from [{M6Q8}L6]z–.

Figure 1. Emission spectra of 1 (green), 2 (blue), and 3 (red) in deaerated CH3CN at 298 K. The data compiled from ref. 4.

Figure 2. T-dependences of the emission spectra of 1 (green), 2 (blue), and 3 (red) in PEG-DMA polymer matrices in 3 K < T < 300 K.

Figure 3. T-dependences of νem (a) and fwhm (b) of 1 (green circles), 2 (blue circles), and 3 (red circles) in PEG-DMA polymer matrices.

Figure 4. T-dependences of τem of 1 (a)), 2 (b)), and 3 (c)) in PEG-DMA polymer matrices. Black solid curves represent the best fittings of T-dependent τem in 3.3 K < T < 300 K by eq 1.

Figure 5. zfs parameters in the T1 state of [{Mo6X8}(n-C3F7COO)6]2–. Figure 6. Simulations of the T-dependent emission spectra of 1 (a)), 2 (b)), and 3 (c)) at several temperatures in PEG-DMA polymer matrices by eqs 2 and 3. The spectra shown by the black curves are the observed ones while the simulated spectra are shown by the colors indicated in each panel.

Figure 7. Simulations of the emission spectra of 1 (a)), 2 (b)), and 3 (c)) in CH3CN at 298 K (black curve) by the relevant Φn emission spectra. The color legends are the same with those in Figure 5. 22


Page 22 of 34

Page 23 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 8. The correlation between ∆E1n and {Z(X)}4. The closed circles and squares represent ∆E14 and ∆E13, respectively, and the black broken lines are the linear regression lines. The data on 1, 2, and 3 are shown by green, blue, and red, respectively.

23



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Graphical Abstract

Synopsis Temperature (T)-dependent emission spectra and lifetimes of [{Mo6X8}(n-C3F7COO)6]2– (X = Cl (1), Br (2), and I (3)) in 3.0 < T < 300 K were analyzed by an excited triplet state spin-sublevel (Φn, n = 1 – 4) model and the Φn model parameters explained successfully the emission characteristics of the clusters at given T. The splitting energy between the lowest- (Φ1) and third- (Φ3) or fourth-spin-sublevels (Φ4) (ΔE13 or ΔE14, respectively) was shown to correlate linearly with the fourth-power of the atomic number (Z) of X: ΔE1n (n = 3 or 4) vs. {Z(X)}4.


Page 24 of 34

Page 25 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure Captions Chart 1. Structure of an octahedral hexanuclear metal cluster [M6(μ3-Q)8L6]z–. Chart 2. Excited triplet state spin-sublevel models for the T-dependent emission from [{M6Q8}L6]z–. Figure 1. Emission spectra of 1 (green), 2 (blue), and 3 (red) in deaerated CH3CN at 298 K. The data compiled from ref. 4. Figure 2. T-dependences of the emission spectra of 1 (green), 2 (blue), and 3 (red) in PEG-DMA polymer matrices in 3 K < T < 300 K. Figure 3. T-dependences of νem (a) and fwhm (b) of 1 (green circles), 2 (blue circles), and 3 (red circles) in PEG-DMA polymer matrices. Figure 4. T-dependences of τem of 1 (a)), 2 (b)), and 3 (c)) in PEG-DMA polymer matrices. Black solid curves represent the best fittings of T-dependent τem in 3.3 K < T < 300 K by eq 1. Figure 5. zfs parameters in the T1 state of [Mo6X8(n-C3F7COO)6]2–. Figure 6. Simulations of the T-dependent emission spectra of 1 (a)), 2 (b)), and 3 (c)) at several temperatures in PEG-DMA polymer matrices by eqs 2 and 3. The spectra shown by the black curves are the observed ones while the simulated spectra are shown by the colors indicated in each panel. Figure 7. Simulations of the emission spectra of 1 (a)), 2 (b)), and 3 (c)) in CH3CN at 298 K (black curve) by the relevant Φn emission spectra. The color legends are the same with those in Figure 5. Figure 8. The correlation between ΔE1n and {Z(X)}4. The closed circles and squares represent ΔE14 and ΔE13, respectively, and the black broken lines are the linear regression lines. The data on 1, 2, and 3 are shown by green, blue, and red, respectively.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chart 1. Structure of an octahedral hexanuclear metal cluster, [M6(μ3-Q)8L6]z–.

Chart 2. Excited triplet state spin-sublevel models for the T-dependent emission from [{M6Q8}L6]z–.


Page 26 of 34

Page 27 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Emission spectra of 1 (green), 2 (blue), and 3 (red) in deaerated CH3CN at 298 K. The data compiled from ref. 4.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. T-dependences of the emission spectra of 1 (green), 2 (blue), and 3 (red) in PEG-DMA polymer matrices in 3 K < T < 300 K.


Page 28 of 34

Page 29 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. T-dependences of νem (a) and fwhm (b) of 1 (green circles), 2 (blue circles), and 3 (red circles) in PEG-DMA polymer matrices.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. T-dependences of τem of 1 (a)), 2 (b)), and 3 (c)) in PEG-DMA polymer matrices. Black solid curves represent the best fittings of T-dependent τem in 3.3 K < T < 300 K by eq 1.


Page 30 of 34

Page 31 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. zfs parameters in the T1 state of [{Mo6X8}(n-C3F7COO)6]2–.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 6. Simulations of the T-dependent emission spectra of 1 (a)), 2 (b)), and 3 (c)) at several temperatures in PEG-DMA polymer matrices by eqs 2 and 3. The spectra shown by the black curves are the observed ones while the simulated spectra are shown by the colors indicated in each panel.


Page 32 of 34

Page 33 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 7. Simulations of the emission spectra of 1 (a)), 2 (b)), and 3 (c)) in CH3CN at 298 K (black curve) by the relevant Φn emission spectra. The color legends are the same with those in Figure 5.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 8. The correlation between ΔE1n and {Z(X)}4. The closed circles and squares represent ΔE14 and ΔE13, respectively, and the black broken lines are the linear regression lines. The data on 1, 2, and 3 are shown by green, blue, and red, respectively.


Page 34 of 34

Zero-Magnetic-Field Splitting in the Excited Triplet States of

Recommend Documents