Hydrido copper clusters supported by dithiocarbamates: Oxidative hydride removal and neutron diffraction analysis of [Cu7(H){S 2C(aza-15-crown-5)}6]

Article abstract of DOI:10.1021/ic300135w

Reactions of Cu(I) salts with Na(S₂CR) (R = NⁿPr ₂, NEt₂, aza-15-crown-5), and (Bu₄N)(BH ₄) in an 8:6:1 ratio in CH₃CN solution at room temperature yield the monocationic hydride-centered octanuclear Cu^I clusters, [Cu₈(H){S₂CR}₆](PF₆) (R = N ⁿPr₂, 1_H; NEt₂, 2_H; aza-15-crown-5, 3_H). Further reactions of [Cu₈(H){S ₂CR}₆](PF₆) with 1 equiv of (Bu ₄N)(BH₄) produced neutral heptanuclear copper clusters, [Cu₇(H){S₂CR}₆] (R = NⁿPr ₂, 4_H; NEt₂, 5_H; aza-15-crown-5, 6_H) and clusters 4-6 can also be generated from the reaction of Cu(BF₄)₂, Na(S₂CR), and (Bu₄N) (BH₄) in a 7:6:8 molar ratio in CH₃CN. Reformation of cationic Cu^I₈ clusters by adding 1 equiv of Cu^I salt to the neutral Cu₇ clusters in solution is observed. Intriguingly, the central hydride in [Cu₈(H){S₂CN ⁿPr₂}₆](PF₆) can be oxidatively removed as H₂ by Ce(NO₃)₆^2- to yield [Cu^II(S₂CNⁿPr₂)₂] exploiting the redox-tolerant nature of dithiocarbamates. Regeneration of hydride-centered octanuclear copper clusters from the [Cu^II(S ₂CNⁿPr₂)₂] can be achieved by reaction with Cu(I) ions and borohydride. The hydride release and regeneration of Cu^I₈ was monitored by UV-visible titration experiments. To our knowledge, this is the first time that hydride encapsulated within a copper cluster can be released as H₂ via chemical means. All complexes have been fully characterized by ¹H NMR, FT-IR, UV-vis, and elemental analysis, and molecular structures of 1_H, 2_H, and 6_H were clearly established by single-crystal X-ray diffraction. Both 1_H and 2_H exhibit a tetracapped tetrahedral Cu ₈ skeleton, which is inscribed within a S₁₂ icosahedron constituted by six dialkyl dithiocarbamate ligands in a tetrametallic- tetraconnective (μ₂, μ₂) bonding mode. The copper framework of 6_H is a tricapped distorted tetrahedron in which the four-coordinate hydride is demonstrated to occupy the central site by single crystal neutron diffraction. Compounds 1-3 exhibit a yellow emission in both the solid state and in solution under UV irradiation at 77 K, and the structureless emission is assigned as a ³metal to ligand charge transfer (MLCT) excited state. Density functional theory (DFT) and time-dependent density functional theory (TDDFT) calculations on model compounds match the experimental structures and provide rationalization of their bonding and optical properties.

Full text of DOI:10.1021/ic300135w

Article
pubs.acs.org/IC
Hydrido Copper Clusters Supported by Dithiocarbamates: Oxidative
Hydride Removal and Neutron Diffraction Analysis of [Cu7(H)
{S2C(aza-15-crown-5)}6]
Ping-Kuei Liao,† Ching-Shiang Fang,† Alison J. Edwards,‡ Samia Kahlal,§,⊥ Jean-Yves Saillard,*,§,⊥
and C. W. Liu*,†
†Department of Chemistry, National Dong Hwa University, Hualien, Taiwan 97401, R. O. C.
‡Bragg Institute, Australian Nuclear Science and Technology Organisation, Lucas Heights, NSW Australia
§UMR-CNRS 6226 “Institut des Sciences Chimiques de Rennes”, Universite
́
de Rennes 1, 35042 Rennes cedex, France
⊥Universite
́ ́
Europeenne de Bretagne, 5 bd. Laennec, 35000 Rennes, France
̈
S
* Supporting Information
ABSTRACT: Reactions of Cu(I) salts with Na(S2CR) (R = NnPr2, NEt2, aza-15-crown-5), and (Bu4N)(BH4) in an 8:6:1 ratio
in CH3CN solution at room temperature yield the monocationic hydride-centered octanuclear CuI clusters, [Cu8(H){S2CR}6]-
(PF6) (R = NnPr2, 1H; NEt2, 2H; aza-15-crown-5, 3H). Further reactions of [Cu8(H){S2CR}6](PF6) with 1 equiv of (Bu4N)(BH4)
produced neutral heptanuclear copper clusters, [Cu7(H){S2CR}6] (R = NnPr2, 4H; NEt2, 5H; aza-15-crown-5, 6H) and clusters
4−6 can also be generated from the reaction of Cu(BF4)2, Na(S2CR), and (Bu4N)(BH4) in a 7:6:8 molar ratio in CH3CN.
Reformation of cationic CuI8 clusters by adding 1 equiv of CuI salt to the neutral Cu7 clusters in solution is observed. Intriguingly,
the central hydride in [Cu8(H){S2CNnPr2}6](PF6) can be oxidatively removed as H2 by Ce(NO3)62− to yield [CuII(S2CNnPr2)2]
exploiting the redox-tolerant nature of dithiocarbamates. Regeneration of hydride-centered octanuclear copper clusters from the
[CuII(S2CNnPr2)2] can be achieved by reaction with Cu(I) ions and borohydride. The hydride release and regeneration of CuI8
was monitored by UV−visible titration experiments. To our knowledge, this is the first time that hydride encapsulated within a
copper cluster can be released as H2 via chemical means. All complexes have been fully characterized by 1H NMR, FT-IR, UV−
vis, and elemental analysis, and molecular structures of 1H, 2H, and 6H were clearly established by single-crystal X-ray diffraction.
Both 1H and 2H exhibit a tetracapped tetrahedral Cu8 skeleton, which is inscribed within a S12 icosahedron constituted by six
dialkyl dithiocarbamate ligands in a tetrametallic-tetraconnective (μ2, μ2) bonding mode. The copper framework of 6H is a
tricapped distorted tetrahedron in which the four-coordinate hydride is demonstrated to occupy the central site by single crystal
neutron diffraction. Compounds 1−3 exhibit a yellow emission in both the solid state and in solution under UV irradiation at 77
3
K, and the structureless emission is assigned as a metal to ligand charge transfer (MLCT) excited state. Density functional
theory (DFT) and time-dependent density functional theory (TDDFT) calculations on model compounds match the
experimental structures and provide rationalization of their bonding and optical properties.
the chemistry of metal hydride species is thus central to the
design of new materials for hydrogen storage.
The metallic hydrides are frequently referred to as interstitial
hydrides due to the fact that hydrogen occupies interstitial sites
in the metal host lattice. Four-coordinate hydrogen in
INTRODUCTION
■
The chemistry of metal hydrides has been heavily pursued
because these molecules are key intermediates in various
catalytic reactions.1 Metal hydrides have promising applications
as new materials for hydrogen storage.2 In this regard,
conversion of dihydrogen into a metal hydride and liberation
of the hydride as dihydrogen via chemical means represents an
attractive and safe method of hydrogen storage.3 Understanding
Received: January 18, 2012
Published: June 4, 2012
© 2012 American Chemical Society
6577
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
Scheme 1. Preparation of Complexes 1−7
tetrahedral sites and six-coordinate hydrogen in octahedral sites
are those most commonly observed in intermetallic hydrides.2b
In molecular compounds, six-coordinate hydrogen in an
octahedral cavity has been well studied for example [Cu6(H)-
(Cl){Me3(NCH2)3}3]+ ,4 [Cr6(μ3-S)8(H)(PEt3)6],5
[Ru7(CO)19(μ-CNMe2)(μ6−H)],6 [HRu6(CO)18]−,7
[HCo6(CO)15]−,8 and [HNi12(CO)21]3−.9 The latter three
structures were elucidated by single-crystal neutron analysis.
Surprisingly it was not until 2008 that the first molecular
species containing a 4-coordinate hydrogen in the tetrahedral
cavity of [(C5Me4SiMe3)YH2]4(THF) was characterized via
neutron diffraction.10 Other examples of 4-coordinate hydrogen
in a tetrahedral hole are [(C5Me4SiMe3)LuH2]4,11 [(TpMe2)-
LnH2]4 (Ln = Y, Nd, Sm, Lu),12 [{Rh2(PNNP)(CO)2}2(μ4-
H)] (PNNP = 3,5-bis(diphenylphosphinomethyl)pyrazolate],13
[(C5Me4SiMe3Lu)4(μ-NH2)7(μ4-H)],14 [{PhP(CH2)3Fe}4(μ4-
H)]−,15 and [(tBuCH2O)12Mo4(μ4-H)]−.16
Despite numerous reports of transition metal clusters with an
interstitial hydride located at either tetrahedral or octahedral
sites, surprisingly, very few compounds from which interstitial
hydride can be released as dihydrogen by chemical methods
have been established. The lability of interstitial hydrides is
controversial. It has been reported that the oxidation of
[Cr6(μ3-S)8(H)(PEt3)6]5 with ferrocenium tetrafluoroborate
affords [Cr6(μ3-S)8(H)(PEt3)6](BF4) while the hydride re-
mains in the center of Cr6 octahedron.17 On the other hand,
the interstitial hydrides in some octahedral complexes have
been noted to be mobile between the center of the octahedron
and the cluster surface. For instance, the central hydride in
[Ru6(μ6-H)(CO)18]− is stable in both a methanol solution of
KOH and a THF solution of KH, but protonation of [Ru6(μ6-
H)(CO)18]− affords H2Ru6(CO)18, with two bridging
hydrides.18 The interstitial hydride in [HCo6(CO)15]− under-
goes an intriguing rapid migration from the inside to the
outside of the cluster.19 The 4-coordinate hydrogen atom in a
tetrahedral cavity identified in [(C5Me4SiMe3)LnH2]4(THF)
(Ln = Y, Gd, Dy, Lu) seems to be quite stable despite the
observation that these rare-earth polyhydrido complexes show
excellent olefin-polymerization activity.20 The reaction of
tetranuclear rare-earth metal octahydrido complexes
[(C5Me4SiMe3)LuH2]4(THF) with ammonia yields selectively
the heptaamido-monohydrido complexes [{(C5Me4SiMe3)-
Lu}4(μ-NH2)6(μ3-NH2)(μ4-H)], in which a μ4-H ligand
remains at the center of the tetrahedral metal framework.14
Exposure of this heptaamido-monohydrido complex to an
ammonia atmosphere at room temperature for two days affords
the octaamido complex [(C5Me4SiMe3)Lu(μ-NH2)]4 quantita-
tively.14 Apparently H2 was released on ammonolysis.
Recently reported hydride-centered octanuclear copper/
silver clusters, [M8(H){E2P(OR)2}6]+ (M = Cu, Ag; E = S,
Se)21 and [M8(H){S2CC(CN)2}6]5− (M = Cu, Ag),22 are a
new class of 4-coordinate H atom lying in the center of a
tetracapped tetrahedral (or triakis-tetrahedral)23 metallic cavity
inscribed within a distorted E12 icosahedron. The encapsulated
hydride is extraordinarily stable, and this stability can be
attributed to strong bonding interactions between the central
hydride and encapsulating Cu(I)/Ag(I) ions. The strong
interaction and small radius of hydride induces a tetrahedral
contraction of the cubic metallic framework. In the search for
hydride activation within copper clusters, the redox-tolerant
dithiocarbamate ligands were pursued with the hope that the
central hydride can be abstracted oxidatively to release
dihydrogen. Herein the syntheses, redox, and photophysical
properties of a series of hydrido copper clusters stabilized by
dialkyl dithiocarbamates as well as their hydride abstraction and
addition reactions are described. Also reported is the single
crystal neutron diffraction analysis of [Cu7(H){S2C(aza-15-
crown-5)}6] from which the central hydride is unequivocally
proven to occupy a central tetrahedral site.
RESULTS AND DISCUSSION
■
The synthetic details of hydrido-copper clusters stabilized by
dithiocarbamate (dtc) ligands as well as their redox reactions en
route to the formation of either heptanuclear copper clusters or
mononuclear Cu(II) species from octanuclear Cu(I) complexes
are described in Scheme 1. The hydride (or deuteride)
encapsulated compounds (1, 2, and 3) can be produced by
the reaction of [Cu(CH3CN)4](PF6), Na(S2CR) (R = NnPr2,
1; NEt2, 2; aza-15-crown-5, 3), and (nBuN4)(BH4) or
(nBuN4)(BD4) in an 8:6:1 molar ratio in CH3CN solution.
Unlike other hydrido-copper clusters stabilized by six
dichalcogenophosphate ligands which can also be synthesized
from the hydride insertion into the empty copper homocubane,
the formation of compounds 1−3 via this method is not an easy
task. In 2006, Hogarth et al. reported the structure of dicationic
[Cu8{S2CNnPr2}6]2+,24 which contains an empty CuI8 cubic
cage capped by six dipropyl dithiocarbamate ligands
6578
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
Figure 1. (a) Molecular structure of the cationic [Cu4(μ4-H)(μ3-Cu)4{S2CNnPr2}6]+ in 1H. H-atoms omitted for clarity except the central hydride.
(b) Central unit of [Cu8H] in 1H. (c) Tetracapped tetrahedron with numbers in the vertices indicating their degrees.
(S2CNnPr2)−, in a μ4 bridging pattern on each square face of a
drides. To our great surprise, the hydride-centered (or
deuteride-centered) CuI7 clusters surrounded by six dithiocar-
bamate ligands, 4, 5, and 6 can be isolated from the reaction of
1−3 with (Bu4N)(BH4) within 1 h, respectively (Scheme 1).
Compounds 4, 5, and 6 can also be isolated by the reaction of
Cu(BF4)2, Na(S2CR), and (nBuN4)(BH4) in a 7:6:8 molar ratio
in CH3CN in 10 min. Here each CuII is reduced to CuI by 1
equiv of BH4 anion and a further equivalent of BH4 anion
provides the hydride to generate CuI7 clusters. More
importantly, the CuI7 clusters can be easily converted to CuI8
clusters by adding 1 equiv of a CuI salt to the CuI7 compounds
in solution.
We have demonstrated that the hydride-centered copper
clusters stabilized by dichalcogenophosphate ligands are
exceedingly stable even on addition of stoichiometric amounts
of strong acid or base such as HBF4 or NaOH with prolonged
stirring.21a,b No decomposition was detected as the line shape
cube. In contrast, the octanuclear copper clusters [Cu8{S2P-
21a
(OR)2}6]2+
and [Cu8{Se2P(OR)2}6]2+,21b can be readily
prepared by the direct reaction of copper salts and
dichalcogenophosphate ligands in a 4:3 ratio, whereas
[Cu8{S2CNnPr2}6]2+ could not be generated by this method-
ology. The difficulty in the isolation of pure cluster
[Cu8{S2CNnPr2}6]2+ discouraged synthesis of 1−3 via the
hydride insertion reaction. Nevertheless the empty cubane,
[Cu8{S2CNnPr2}6]2+, can be reasonably proposed as a labile
intermediate, en route to the final products 1−3 on the basis of
the known reaction pathways in the formation of dichalcoge-
nophosphato copper clusters with an encapsulated hydride.
Since the dithiocarbamate-type ligands are both innocent and
redox-tolerant, they have the potential to stabilize high
oxidation state metal complexes25 so we were curious about
the further reaction products of clusters 1−3 with borohy-
−
−
6579
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
Figure 2. (a) Molecular structure of the [Cu4(μ4-H)(μ3-Cu)3{S2C(aza-15-crown-5)}6] in 6H. H-atoms omitted for clarity except the interstitial
hydride. (b) Cu7 skeleton of 6H displays a tricapped pyramid. (c) Cu7H core drawing from neutron diffraction data.
of broad hydride resonance in 1H NMR spectra remains
unchanged. Gratifyingly, the hydride of Cu8H clusters (1H, 2H,
and 3H) can be oxidatively removed only after adding the
oxidizing agent, (NH4)2[Ce(NO3)6], with conversion of the
cluster to [CuII(S2CR)2], 7, and vice versa. The titration
reactions which demonstrate the hydride uptake and release
can be monitored by UV−vis spectroscopy. More importantly,
the hydrido CuI8 clusters can be regenerated from CH3CN
solutions containing [CuII(S2CR)2] and CuI sources (in the
form of [Cu(CH3CN)4](PF6) in a 3:5 molar ratio with 4 equiv
All the spectroscopic characteristics and main structural
features of the new compounds 1−7 as well as their redox and
photophysical properties are described below:
Crystallography: Compound 1H. The monocationic
octanuclear copper cluster crystallizes in the trigonal space
¯
group, P31c, with a solvating CHCl and a PF anion per
̅
3
6
cluster. Unlike previously reported hydrido copper com-
plexes,21 none of the copper atoms of this cluster are
¯
disordered, but the PF6 anion is disordered over two equivalent
positions. The central Cu8 core is arranged in a tetracapped-
tetrahedral (or triakis-tetrahedral) geometry23 about a 3-fold
axis on which the hydrogen must lie and is stabilized by twelve
μ2-S atoms from six dithiocarbamate (dtc) ligands in a distorted
−
of BH4 anion; Scheme 1).
6580
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
icosahedral geometry (Figure 1a). One Cu(3) atom and three
Cu(2) (abbreviated as Cuv) atoms form an inner tetrahedron
and are directly connected to the inferred central hydride. Each
face of the inner tetrahedron is capped either by the Cu(1) or
by one of the three Cu(4) atoms (abbreviated as Cucap) of an
outer tetrahedron (Figure 1b). The Cuv−Cucap [2.577(1)−
2.669(1) Å] distances are much shorter than the sum of the van
der Waals radii for copper atoms (2.80 Å)26 and are shorter
than the Cuv−Cuv distances 2.894(2)−2.989(2) Å. The average
Cu−Cu distance [2.806(1) Å] in the Cu8 empty cluster,
[Cu8{S2CNnPr2}6]2+,24 lies between Cuv−Cucap and Cuv−Cuv
of complex 1H. The average Cuv-inferred central μ-H distance
[1.80(12) Å] is slightly shorter than that observed in
[Cu8(H){S2P(OiPr)2}6]+ [1.831(3) Å]20 and longer than the
sum of their covalent radii [1.63(5) Å].27 Six dtc ligands, each
adopting a tetrametallic, tetraconnective (μ2-S, μ2-S) bonding
mode, are located on the top of Cu4 butterflies where the hinge
positions are the edge of the inner tetrahedron and the wingtips
are the four capping Cu atoms (Figure 1a). The inner Cu
atoms in 1H are proposed to possess a tetrahedral S3H
coordination, while the capping Cu atoms have trigonal planar
S3 coordination. The dihedral angles of Cu4 butterflies range
from 153.2° to 157.8°. Two kinds of Cu−S distances, namely,
Cuv−S and Cucap−S, each averaging 2.358(2) and 2.270(2) Å,
are revealed in 1H with the shorter being to the wingtip Cu
atoms (Cucap).21,22 The Cucap−S distances lie in the same range
as those observed in the octanuclear cubic copper cluster
[Cu8{S2CNnPr2}6]2+ [2.246(2)Å],24 and the average intraligand
S···S bite distance, 3.066(2) Å, is likewise comparable to
3.082(1) Å.24 The C−S distances in the range of 1.741(5)−
1.747(5) Å are marginally shorter than those observed in
[Cu8{S2CNnPr2}6]2+ [1.752(3)−1.754(3) Å].24
mode to cap a Cu4 butterfly formed by two Cucap and two Cuv
atoms. The average dihedral angle of Cu4 butterflies
[156.23(11)°] is analogous to those in 1H. The presence of a
central hydride is inferred from spectroscopic measurements
and is proposed to lie at the center of the cube 1.795(2) Å from
the surrounding Cu atoms for 2H.
Compound 6H. The neutral heptanuclear copper cluster
crystallizes in the triclinic P1 space group, with one cluster and
̅
one acetone molecule per asymmetric unit. The cluster
contains a distorted, tricapped tetrahedral copper core centered
about an interstitial hydride anion and is surrounded by six aza-
15-crown-5-dtc ligands (Figure 2a). Compared with the
tetracapped tetrahedral copper skeleton of 1H, this tricapped-
tetrahedral copper skeleton 6H (Figure 2b) appears to arise
from losing one capping copper atom of 1H. The central
tetrahedron, which approaches a triangular pyramid arising
from a pseudo-C3 elongation, is formed by Cu2, Cu4, Cu6, and
Cu7 atoms (abbreviated as Cuv), and the three capping atoms
are Cu1, Cu3, and Cu5 (abbreviated as Cucap). The basal plane
of atoms Cu2, Cu4, and Cu6 lacks a capping Cu atom. The
elongated edges of the pyramid range from 3.373(2)−3.403(1)
Å, the average length of the basal plane edge is 2.726(1) Å,
which is slightly shorter than the sum of the van der Waals radii
for two copper atoms, 2.8 Å, and is longer than Cuv−Cucap [av.
2.616(2)] of compound 1H. Of the six aza-15-crown-5-dtc
ligands, only two achieve a tetrametallic−tetraconnective (μ2-S;
μ2-S) bonding mode and are located on the top of Cu4
butterflies where hinge positions are elongated edges of the
triangular pyramid and wingtips are capping Cu atoms. The
other four dtc ligands exist in a trimetallic, triconnective (μ2-S;
μ1-S) bonding mode. One is capping the remaining Cu4
butterfly above the basal plane with a nonbonding Cu6
S11, 3.067(2) Å, and the other three are located about the
uncapped Cu3 basal plane linking two vertices and one capping
Cu atom. Thus the Cu6 presents a trigonal planar S2H
coordination. Twelve sulfur atoms form a distorted icosahedral
cage with an average, intraligand S···S bite distance of 3.041(2)
Å. This constitutes the first Cu7 cluster within an S12 icosahedron.
The central hydride atom position of 6H is unequivocally
proven by single-crystal neutron diffraction (Figure 2c), and it
is well-defined when refined anisotropically. This unambiguous
determination of H atom positions reveals bond lengths of
Cuv−H as Cu2−H1 = 1.81(2) Å, Cu4−H1 = 1.84(2) Å, Cu6−
H1 = 1.78(2) Å, Cu7−H1 = 2.02(2) Å, and a Cucap−H distance
(Cu3−H1) of 2.21(2) Å. This clear differentiation of the Cuv
atoms from the Cucap atoms is in stark contrast to the negligible
resolution of these particular features by X-ray diffraction.
Clearly, only the four Cuv (Cu2, Cu4, Cu6, Cu7) atoms display
bonding interactions with the central hydride. The distortion
away from ideal tetrahedral geometry is observed in the bond
angles at H(1) which range from 94.3(13) to 120.2(15)°. The
mean Cu−(μ4-H) distance [1.86(2) Å] of 6H is somewhat
longer than the 1.76(3) Å (average), observed in a three-
coordinate hydride in [Cu6(P(p-tolyl)3)(μ3-H)]6 determined
by neutron diffraction analysis.33
Heteroelemental cages having a tetracapped-tetrahedron or
distorted-cubic framework are not unusual. The [Fe4S4]
moiety,28 ubiquitously present in the core unit of ferredoxin,
is one of the most studied metal chalcogenide skeletons in
bioinorganic chemistry. Nevertheless homometallic clusters
containing this geometry have only been reported in metal-
locarbohedranes such as Ti8C12,29 observed in the gas phase
only, and no guest atom was observed in the center. The
tetracapped tetrahedral core does exist in the heterometallic
cluster, [Os4(CO)12{Pd(PtBu3)}4],30a and [E4(PdL2)4]2+ (L =
PPh2Me; E = Sb or Bi)30b involving a tetrahedral core E4 (E =
Sb or Bi) with four PdL2 as capping groups. Examples also
31
include [Cu4Mn4(SnPr)12S]2−
and [Re4(CO)12{μ3-InRe-
(CO)5}4].32 A schematic drawing of a tetracapped tetrahedron
with the degree of vertex represented by number is displayed in
Figure 1c.
Compound 2H. The ethyl homologue, 2H, crystallizes in the
cubic Pn3n space group. The geometry of the copper skeleton
̅
is a tetracapped tetrahedron of Td symmetry corresponding to a
cube in which alternate corners are displaced along the 3-fold
toward the center, and the eight CuI ions are further
surrounded by six dtc ligands (Supporting Information Figure
S1). The whole copper skeleton is disordered over two
positions. The vertices of the tetrahedron are formed by the
Cu1, Cu4C, Cu4D, and Cu4E atoms, and the capping atoms
are Cu3A, Cu3, Cu3B, and Cu2A (or vice versa). The edge
lengths of the tetrahedron (Cuv−Cuv), 2.854(4) and
3.003(4)Å, are larger than Cuv−Cucap [2.492(4)2.659(3)
Å]; however, those distances are slightly less than those (Cuv−
Cuv and Cuv−Cucap) observed in 1H. Similar to 1H, the dtc
ligands adopt a tetrametallic−tetraconnective (μ2, μ2) bonding
Considering the geometry of the Cu7 skeleton only, this
tricapped tetrahedral core does occur in several homometallic
and heterometallic clusters. Examples are [Cu7Te{As(Te2)-
34
(Te2)}3]4−
,
[Co7S6(PPh3)5Cl2],35 [Au3Ru4(μ-H)-
(CO)1 2 (PPh3 )3 ],36 [Au3 CoRu3 (CO)1 2 (PPh3 )3 ],3 7
[Bi4Fe4(CO)13]−,38 and [Os5(CO)12Ru(η5-C5H5)(AuPPh3)].39
NMR Studies of Compounds 1−6. The 1H NMR spectra
display a set of chemical shifts corresponding to the propyl,
6581
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
Figure 3. (left) Absorption spectral changes of 1H (6.6 × 10−5 M) upon addition of the (NH4)2[Ce(NO3)6] [(0−5.28) × 10−4 M] in 50% (v/v)
CH3CN−CH2Cl2. (right) Back titrations of (Bu4N)(BH4) [(0.44−5.28) × 10−4 M] in 50% (v/v) CH3CN−CH2Cl2.
ethyl, and aza-15-crown-5 groups respectively for each of the
compounds in the series. Compound 1H shows a broad peak
centered at 7.05 ppm that integrates to 1H relative to 12
methylene protons of the propyl groups. Peak broadening is
attributed to coupling to adjacent quadrupolar copper nuclei (I
= 3/2). In a similar manner 2H and 3H also gave a resonance for
encapsulated hydride at 7.02 and 7.03 ppm, respectively. The
presence of deuteride in the analogous deuterium species (1D−
3D) was confirmed via 2H NMR. 1D exhibits a peak at 7.07 ppm
corresponding to the entrapped deuteride, while 2D and 3D
gave the same deuteride chemical shifts at 7.02 ppm. These
hydride resonance at 7.05 ppm, 1H NMR spectra were
recorded at 5 min intervals and a new peak was observed at
6.55 ppm, corresponding to the hydride resonance of the
neutral cluster [Cu7(H){S2CNnPr2}6], 4H. The integration ratio
of these two peaks centered at 7.05 and 6.55 ppm relative to the
methylene protons of the six propyl groups is also
approximately 1−12 at each time step. Thus, in this process,
no other hydrido copper species were detected. The CuI8 can
be completely converted to CuI7 cluster over 50 min. The
hydride resonance corresponding to [Cu8(H){S2CNnPr2}6]+
1
species was detected in the H NMR spectrum immediately
observed values are markedly different from hydride chemical
upon addition of 1 equiv [Cu(CH3CN)4](PF6) to the NMR
tube containing [Cu7(H){S2CNnPr2}6]. It is thus demonstrated
that a copper ion can be released from the cluster cation,
[Cu8(H)(S2CR2)6]+, or taken up by the neutral Cu7 species,
mediated by the reducing agent or CuI salt (Scheme 1),
respectively. The redox-tolerant nature of the dtc ligands where
the whole copper skeleton encapsulated can stabilize both
Cu8H and Cu7H frameworks and facilitates interconversion.
Mass Spectrometry. Molecular ion peaks of 1−3 were
readily identified from the positive ESI-mass spectrometry
[1568.1 (1569.2) for 1H (1D), 1400.8 (1402.5) for 2H (2D), and
2275.4 (2276.7) for 3H (3D)]. The experimental isotopic
distribution pattern of both hydride and deuteride compounds
are in good agreement with the theoretical ones and
representative patterns of 1H and 1D are given in Supporting
Information Figure S3. The molecular ion peak, [Cu7(H)(dtc)6
+ H+], of Cu7H(dtc)6 clusters, 1504.8 for 4H, 1336.5 for 5H,
and 2213.3 for 6H, are not observed by positive ESI-mass
spectrometry. Instead, peaks for which m/z correspond to the
whole molecular species with an additional copper ion to form
the adduct ion, [{Cu7(H)(dtc)6}(Cu)]+, are identified. This is
quite unexpected and implies that the neutral Cu7H clusters are
not being stable in the ESI gas phase conditions. Indeed the
reaction of the heptanuclear copper cluster with a copper ion
forms the cationic, hydrido CuI8 cluster as reported above.
Hydride Abstraction and Addition Monitored by UV−
vis. The UV−vis titration experiments of a series of
[Cu8(H){S2CR}6]+ complexes with (NH4)2[Ce(NO3)6] and
the further backtitration with (Bu4N)(BH4) have been studied.
All the complexes exhibit similar phenomena and representative
spectra of compound 1H are given in Figure 3. The UV−vis
spectrum of 1H in 50% (v/v) CH3CN−CH2Cl2 solution
21b
shifts of [Cu8(H){Se2P(OR)2}6]+
and [Cu8(H){S2P-
(OR)2}6]+,21a at −0.58 and 3.69 ppm, respectively, but
comparable to [Cu8(H){S2CC(CN)2}6]5−,22 at 7.60 ppm.
−
The PF6 counteranion in compounds 1−3 is clearly observed
by 31P NMR spectroscopy. The hydride in the Cu7 (4H−6H)
clusters and their deuterium analogues (4D−6D) is observed.
Compounds 4H and 5H both gave encapsulated hydride
resonances at 6.55 and 6.49 ppm. 4D, 5D, and 6D show
deuteride peaks at 6.40, 6.50, and 6.51 ppm, respectively. No
PF6 counteranion was detected in the 31P NMR spectrum of
−
compounds 4−6, and these compounds are thus demonstrated
to be neutral species. A slightly downfield shift of ∼0.5 ppm for
the hydride resonance of octanuclear copper(I) clusters (1H,
2H, and 3H) is revealed by comparison with the hydride
resonance of heptanuclear copper(I) clusters (4H, 5H, and 6H).
Since the absorption spectra of hydride-encapsulated CuI8 or
CuI7 clusters are almost the same (vide infra), NMR
spectrometry is the best spectroscopic tool to differentiate
these two clusters. The hydride downfield shift of the
octanuclear species relative to the heptanuclear one is
reproduced by our DFT calculations for models of [Cu4(μ4-
H)Cu4{S2CNH2}6]+ and [Cu4(μ4-H)Cu3(S2CNH2)6] (vide
infra).
Copper Uptake by the Clusters [Cu7(H){S2CR2}6]. The
cationic CuI8 clusters, 1−3, can be converted to the neutral
CuI7 clusters, 4−6, upon the reaction of (Bu4N)(BH4)/
(Bu4N)(BD4) with the octanuclear copper-hydrido complexes.
1
The reactions were monitored by H NMR spectroscopy and
representative spectra are given in Supporting Information
Figure S2. After the addition of 1 equiv of (Bu4N)(BH4) to the
NMR tube containing [Cu8(H){S2CNnPr2}6](PF6), 1H, with a
6582
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
showed intense bands centered at 276 and 345 nm. Figure 3
illustrates the absorption spectral changes of 1H upon the
addition of (NH4)2[Ce(NO3)6] (forward reaction, left) and the
addition of (Bu4N)(BH4) (backward reaction, right). The
absorbance of the visible band at 345 nm gradually decreases
upon stepwise addition of (NH4)2[Ce(NO3)6] while a new
band maximum at 430 nm and a low-energy tail at 600 nm with
low extinction coefficient, progressively increase. The new
absorption bands at ∼430 nm can be tentatively assigned to a
ligand to metal charge transfer (LMCT) band40 and at ∼600
nm, a symmetry forbidden LMCT transition (vide infra). The
reaction sequence can be represented as eqs 1−3 shown below:
reaction of Cu(BF4)2 and Na(S2CNnPr2) in a 7:6 ratio in
CH3CN followed by the addition of 8 equiv of (Bu4N)(BH4)
produces pure 4H in high yield (eq 4). These studies thus reveal
a new synthetic route for the preparation of pure hydride-
centered CuI8 and CuI7 clusters.
−
7Cu2+ + 6(S2CR)− + 8BH4
→ [Cu7(H)(S2CR)6] + 4(B2H6) + 3.5H2
(4)
Cyclic Voltammograms of 1−6. Cyclic voltammograms
of the hydride-centered CuI8 and CuI7 clusters surrounded by
six dithiocarbamate ligands display similar irreversible CuI/CuII
oxidation and they are summarized in Table 2. The
electrochemical oxidation of 1H has been investigated at
room temperature in 50% (v/v) CH3CN−CH2Cl2 solution
[0.1 M (Bu4N)(ClO4)] at 100 mV s−1 for the oxidation of 5
mM of compound 1H using a glassy carbon working electrode.
1H exhibits an irreversible wave at 0.36 V which indicates the
formation of the CuII species (Figure 4). The small peak at 0.06
V suggests the reduction of CuII species to CuI species.
[Cu8(H)(S2CR)6]+ + Ce4+
→ [Cu8(S2CR)6]2+ + 0.5H2 + Ce3+
(1)
[Cu8(S2CR)6]2+ + 3Ce4+
→ 3[CuII(S2CR)2] + 5Cu+ + 3Ce3+
(2)
−
3[CuII(S2CR)2] + 5Cu+ + 4BH4
When the scan rate was decreased, the CuII/CuI reduction
was decreased. The peak position shifts upon variation of the
scan rate. This suggests that oxidation occurs alongside a rapid
chemical process. Appearance of a brown coloration is observed
near the working electrode, which suggests decomposition of
the compound during the redox process. After running several
cycles of the redox process, the UV−vis spectra of the
experimental solution show a new visible band at 430 nm
consistent with the formation of [CuII(dtc)2] species. Owing to
the decomposition of the CuI8 hydride cluster and the gradual
loss of hydride during the oxidation, the hydrido clusters
cannot be regenerated in the cyclic voltammetry experiments.
Photophysical Properties of 1−3. The photophysical
data including UV−vis, excitation, and emission of compounds
1H−3H are summarized in Table 3. The spectral profiles of all
compounds are quite similar and representative spectra of 1H
are displayed in Figure 5. The electronic absorption spectrum
of 1H in CH2Cl2 displays an intense absorption band at 276 (ε
= 68 000 cm−1 M−1) nm and a less intense band at about 345
(ε = 37 000 cm−1 M−1) nm (Figure 5a). The origin of the high-
energy band is likely to be a ligand-centered transition, as the
sodium salts of dithiocarbamate exhibit absorptions in this
region (260 and 291 nm).42 The low-energy band could be
tentatively assigned to the metal to ligand charge transfer
(MLCT) transition (vide infra). The compound 1H exhibits a
yellow emission in both the solid state and solution under UV−
irradiation at 77 K. Direct excitation of the low energy band
resulted in a broad, structureless emission centered at 576 nm
(Figure 5b). In frozen and degassed CH2Cl2, the emission band
is centered at 591 nm (Figure 5c), which is red-shifted about
∼15 nm compared to the emission from the solid state. A large
Stoke shift (>12 600 cm−1) and the long emission lifetimes of
1−3 implied the possibility of a large distortion in the excited
state and emission from a spin-forbidden triplet emissive state.
The broad, featureless nature of the emission band of 1H is
→ [Cu8(H)(S2CR)6]+ + 2(B2H6) + 1.5H2
(3)
The reaction of (NH4)2[Ce(NO3)6] with 1H was monitored
by 1H NMR which reveals that within 5 min the central hydride
peak at 7.05 ppm disappears and after 30 min the ligand related
peaks are also broadened. Reaction of the oxidant was evident
with liberation of hydrogen gas from the hydrido copper(I)
clusters on the formation of empty cubane cluster
[Cu8{S2CNnPr2}6]2+ (eq 1 determined by GC-MS; Supporting
Information Figure S4). We propose [Cu8{S2CNnPr2}6]2+ has
been formed upon oxidation of 1H. In the presence of oxidant,
it has proven extremely difficult to stop the reaction at this step
or fully characterize this sequence. The empty Cu8 cube can be
further oxidized to become [CuII(S2CNnPr2)2] (eq 2). We have
nevertheless managed to theoretically reproduce the UV−vis
absorption spectrum of [Cu8{S2CNR2}6]2+ by DFT modeling
(vide infra). The calculated absorption spectrum shows little
difference from that of compound 1. The reversed process in
which UV−vis spectra are measured to follow the stepwise
titration of (Bu4N)(BH4). This suggests that the dual functions
of BH4 anion as both a reducing agent to reduce CuII to CuI
−
and a hydride donor to form the hydrido CuI8 cluster.
The entrapped hydride is quite stable in [Cr6S8(H)(PEt3)6]
which can be further oxidized by (FeCp2)(BF4) to produce
one-electron oxidized cluster [Cr6S8(H)(PEt3)6]+.17 However,
the oxidizer (CeIV) we used was much stronger than
[FeCp2]+,41 and clearly, this forces the oxidization to proceed
to yield H2 and [CuII(S2CR)2] species.
In order to prove that the inverse UV−vis titration
experiments apply in bulk, we undertook an experiment
combining [CuII(S2CNnPr2)2] and [Cu(CH3CN)4](PF6) in
the ratio of 3:5 in CH3CN followed by the addition of 4 equiv
of (Bu4N)(BH4) (eq 3). The initial deep brown solution turned
yellow over ca. 3 min with the concomitant formation of gas
bubbles in the solution, which suggests that H2 gas was formed.
After 10 min stirring, the species in solution was confirmed as
3
consistent with a MLCT excited state. This is consistent with
our TDDFT calculations (vide infra). To the best of our
knowledge no reports on the emissive properties of any copper
dithiocarbamate complexes have ever been communicated. Of
course the photochemistry of Cu(II) dithiocarbamate in CCl4
has been well established.43
1
the hydride-encapsulated CuI8 cluster by H NMR. On further
addition of 1 equiv of (Bu4N)(BH4) (total 5 equiv), the color
of the mixture turned orange over 5 min, and the solution was
subsequently characterized to contain the hydride-encapsulated
1
CuI7−cluster by H NMR spectroscopy. Thus 1 equiv more of
Theoretical Investigations: [Cu8(H){S2CNH2}6]+. The
major computed data associated with the optimized structure
BH4 reduces one of the copper atoms in 1H to form 4H. The
6583
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
Table 2. Redox Potential (V vs FeCp20/+) of Compounds
a
1H−6H
compd
EpA
EpB
compd
EpA
EpB
1H
2H
3H
0.36
0.34
0.38
0.06
0.07
0.04
4H
5H
6H
0.35
0.38
0.33
0.08
0.10
0.05
a
At a glassy-carbon electrode, scan rate 100 mV s−1
Figure 4. Cyclic voltammograms of 1H recorded in 50% (v/v)
CH3CN−CH2Cl2/0.1 M (nBu4N)(ClO4) using a glassy-carbon
working electrode at different scan rates.
of the octanuclear [Cu8(H){S2CNH2}6]+ model are given in
Table 4. This model is found to adopt the [Cu4(μ4-
H)Cu4{S2CNH2}6]+ tetracapped-tetrahedral (or triakis-tetrahe-
dral)23 geometry of T symmetry, in agreement with the X-ray
structures of 1H and 2H. In particular, there is a satisfying
agreement between the computed Cuv−(μ4-H) distances (1.80
Å) and the corresponding experimental values of 1H and 2H
(Table 1). The tetracapped tetrahedral geometry is found to be
more stable than the perfect cubic one of Th symmetry by 0.28
eV, the latter being not an energy minimum (several imaginary
vibrational frequencies). On the other hand, the empty cluster
[Cu8{S2CNH2}6]2+ is found to adopt the cubic (Th) geometry
in the ground state (Table 4), as generally observed for this
type of noncentered dithiolato octanuclear Cu(I) species,44
including [Cu8{S2CNnPr2}6]2+.24 These results are fully
consistent with previous calculations on related isoelectronic
dichalcogenophosphato octanuclear clusters.21a,b The NBO 1s
population of the encapsulated hydride (1.69) is close to that
found in the related dichalcogenophosphato species (∼1.64),
indicating rather similar covalent interactions resulting from an
electron donation of the occupied 1s(hydride) AO into an in-
phase combination of the vacant CuI spz orbitals. The bonding
energy between H− and the unrelaxed [Cu8{S2CNH2}6]2+
fragment is equal to 11.31 eV and the energy lost by the
empty [Cu8{S2CNH2}6]2+ fragment when distorted from its Th
energy minimum to that which it adopts in [Cu4(μ4-
H)Cu4{S2CNH2}6]+ is 0.69 eV. These values are comparable
with those obtained in the dichalcogenophosphato series.21a,b
The computed NMR chemical shift of the encapsulated hydride
(7.6 ppm) is comparable to those observed for compounds
1H−3H (7.0−7.1 ppm).
The MO diagram of [Cu4(μ4-H)Cu4{S2CNH2}6]+ is shown
in Figure 6, with the MO localizations listed. Most of the
6584
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
Table 3. Photophysical Data for [Cu8(H){S2CR}6](PF6) (R = NnPr2, 1H; NEt2, 2H; aza-15-crown-5, 3H)
Stoke shift (cm−1
12 600
)
λ
max
(nm) (ε/dm3 mol−1 cm−1
)
τ (μs)
ex
em
ab
compd
state (T/K)
λ
max
(nm)
λ
(nm)
max
1H
CH2Cl2 (77)
solid (77)
339−405
329, 445
335−411
330, 433
334−421
325, 432
591
276 (68,000), 345 (37,000)
276 (63,000), 350 (37,000)
275 (66,000), 340 (35,000)
22, 4.6
576
596
580
605
566
2H
3H
CH2Cl2 (77)
solid (77)
13 100
13 400
20, 2.7
20, 4.4
CH2Cl2 (77)
Solid (77)
Figure 5. (a) UV−vis, (b) normalized emission spectra of compound
1 in the solid state at 77K, and (c) emission spectra of 1 in CH2Cl2
glass at 77 K.
highest occupied orbitals are of dominant ligand (mainly sulfur
lone pairs) character, with significant 3d contributions of the
eight copper atoms. The 1a HOMO is an exception with a
metallic polarization (mainly dz2 of the central tetrahedron)
and some hydride admixture. The two lowest unoccupied levels
can be viewed as the six combinations of the π*(dtc) ligands.
Above these two levels, one finds orbitals with dominant Cu
character. The lowest UV−vis optical transitions have been
calculated at the TDDFT level, and the corresponding
simulated absorption spectrum is shown in the middle of
Figure 7 (see Computational Details). With a simulated low
energy band at λmax = 326 nm and a more intense band at λmax
= 264 nm, it fits quite well with the recorded experimental
spectra of 1H, 2H, and 3H (λmax = 325−335 nm and λmax ∼ 270
Figure 6. MO diagram of [Cu4(μ4-H)(μ3-Cu)4{S2CNH2}6]+ (left) and
Cu4(μ4-H)(μ3-Cu)3(S2CNH2)6 (right). The numerical values indicate
the MO localization (%) in the following order: metals/ligands/μ4-H.
nm; see Table 3 and Figure 5a). The low energy band is
associated with MLCT transitions, as exemplified by its major
component computed at 335 nm which corresponds to the 4t
→ 6t transition. Almost all the transitions contributing to the
band at λmax = 264 nm are also of the MLCT type, although its
major component at 282 nm corresponds to the 1a → 7t
transition which is primarily hydride to ligand in character.
Interestingly, the simulated absorption spectrum of the empty
Th cluster [Cu8{S2CNH2}6]2+ (left side of Figure 7) exhibits a
Table 4. Relevant Computed Data for [Cu8(S2CNH2)6]2+, [Cu4(μ4-H)Cu4{S2CNH2}6]+, [Cu7{S2CNH2}6]+, and Cu4(μ4-
H)Cu3{S2CNH2}6 in Their Equilibrium Geometries
[Cu8(S2CNH2)6]2+
S = 0
[Cu4(μ4-H)(μ3-
[Cu4(μ4-H)(μ3-
[Cu7(S2CNH2)6]+
S = 0
Cu4(μ4-H)(μ3-
Cu)3{S2CNH2}6 S = 0
Cu)4{S2CNH2}6]+ S = 0
Cu)4{S2CNH2}6]+ S = 1
symmetry
Th
T
C2
C3
3.88
C3
HOMO−LUMO gap/
3.91
4.15
4.06
eV
Cuv−H/Å
Cuv−Cuv/Å
1.800 × 4
2.939 × 6
(1.812, 1.844) × 2
2.833−3.026 (range)
1.778 × 3, 1.846
(2.829, 2.746, 3.012) (2.763, 3.078) × 3
× 3
2.841 × 12
Cuv−Cucap/Å
Cuv−S/Å
Cucap−S/Å
S···S (bite)/ Å
μ4-H NBO charges
2.634 × 12
2.402 × 12
2.306 × 12
3.081 × 6
−0.69
2.597−2.755 (range)
2.259−2.481 (range)
2.259−2.318 (range)
3.073−3.100 (range)
(2.643, 2.630, 2.612) × 3
2.253−2.306 (range) 2.376−2.593 (range)
2.286−2.327 (range)
3.078−3.088 (range) 3.056−3.071 (range)
2.286 × 4
3.091 × 6
−0.68
NMR chemical shift of
μ4-H
7.6
6.5
6585
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
Figure 7. Simulated absorption spectra of the models [Cu8{S2CNH2}6]2+ (left), [Cu4(μ4-H)(μ3-Cu)3(S2CNH2)6]+ (middle), and Cu4(μ4-H)(μ3-
Cu)3(S2CNH2)6 (right). Solid lines represent the calculated transitions, and their lengths are proportional to their oscillator strengths (os). Only
transitions with os >0.02 are shown. Values in italics indicate the theoretical λmax values on the simulated curves.
very similar peak profile as that of [Cu4(μ4-H)-
Cu4{S2CNH2}6]+, with a low-energy band at λmax = 325 nm
and an intense high-energy band at λmax = 274 nm. Most of the
transitions associated with these bands are also of the MLCT
type. This result matches the interpretation of the spectral
changes shown in Figure 3.
The optimized geometry of the lowest triplet state of
[Cu8(H){S2CNH2}6]+ adopts a distorted tetracapped tetrahe-
dral geometry of C2 symmetry (Table 4). It lies 2.93 eV above
the singlet ground state. The emission wavelength of the cluster
has been calculated as the difference between the energy of the
optimized triplet state and that of the singlet state assuming the
same (unrelaxed) geometry as that of the triplet state. The
corresponding value (506 nm) compares reasonably well with
the corresponding experimental data recorded in the solid state
for 1H, 2H, and 3H (∼570−580 nm; see Table 3). When PCM
solvent (CH2Cl2) corrections are considered,45 the computed
transition shifts to 655 nm, whereas the corresponding
experimental values are close to 600 nm (Table 3).
[Cu7(H){S2CNH2}6]. The optimized geometry of Cu7(H)-
(S2CNH2)6 (Table 4) was found to adopt a Cu4(μ4-
H)Cu3(S2CNH2)6 tricapped-tetrahedral geometry of C3
symmetry. Interestingly, it differs from the X-ray and neutron
structure of 6H in that the central (Cuv)4 tetrahedron is only
moderately elongated along the C3 axis, with an elongated edge
of 3.08 compared to 3.38 Å (av.) as determined in the X-ray
diffraction study of 6H. More importantly, the Cu6 atom is not
bonded to two sulfur atoms as revealed in the crystal structure
of 6H (Figure 2), but to three of them, just like Cu2 and Cu4,
these three metals being symmetry related in C3 symmetry. It
should be noted that calculations commenced from geometry
derived from the experimental structure of 6H ended up at the
unique C3 energy minimum. We attribute this difference
between [Cu7(μ4-H){S2CNH2}6] and 6H to the presence of
bulky aza-crown-5 substituents in 6H, which presumably induce
steric and packing distortions in the solid state. This is
supported by the fact that related isoelectronic heptanuclear
dicalcogenophosphato silver(I) clusters adopt the same
structure as our computed model.46 Thus, the structure of
[Cu4(μ4-H)Cu3{S2CNH2}6] can be described as resulting from
the removal of one capping CuI ion from [Cu4(μ4-H)-
Cu4{S2CNH2}6]+. Apart from this removal, both clusters have
similar connectivities. In particular, the encapsulated hydride of
[Cu7(μ4-H){S2CNH2}6] is found to lie in the middle of the
elongated central tetrahedron with Cu−H distances of 1.78 and
1.85 Å. The Cucap−H distance (2.52 Å) is rather long, thus the
hydride is clearly tetrahedrally coordinated, as demonstrated in
the neutron structure of 6H. The computed NMR chemical
shift of the encapsulated hydride (6.5 ppm) is in excellent
agreement with those observed for compounds 4H−6H (6.5−
6.6 ppm).
Interestingly, the hypothetical empty cluster
[Cu7{S2CNH2}6]+ does not retain the tricapped tetrahedral
arrangement of the copper framework in its equilibrium
geometry, but rather relaxes to a structure of C3 symmetry
which can be approximated to a cube having one vacant vertex,
i.e., deriving from that of [Cu8{S2CNH2}6]2+ by the removal of
one Cu+ ion. Thus, the tetrahedral distortion afforded by the
metal atoms in the hydrido clusters is attributed to a hydride
template effect, both in the hepta- and octanuclear species.
The bonding energy between the encapsulated hydride and
the unrelaxed [Cu7{S2CNH2}6]+ fragment is equal to 8.45 eV,
whereas the energy lost by the empty [Cu7{S2CNH2}6]+
fragment when distorted from its equilibrium geometry to
that it adopts in [Cu4(μ4-H)Cu3{S2CNH2}6] is 0.80 eV. The
bonding energy is lower than that found in [Cu8(μ4-
H){S2CNH2}6]+ (vide supra) whereas the NBO 1s(μ4-H)
population (1.68) is almost equal to that found in the
octanuclear parent. These results indicate similar strengths for
the covalent bonding of H− in [Cu4(μ4-H)Cu3{S2CNH2}6] and
[Cu4(μ4-H)Cu4{S2CNH2}6]+, but weaker ionic bonding in the
former, consistent with the lower charge of the hosting cage.
The MO diagram of [Cu7(μ4-H){S2CNH2}6] is shown in
Figure 6, with the major MO localizations listed. Its frontier
orbitals exhibit very similar characteristics as its octanuclear
relative (vide supra). Its absorption spectrum has also been
simulated from TDDFT calculations (see the right side of
Figure 7). It exhibits a similar shape to its octanuclear relative,
with a low energy MLCT absorption band around 340 nm and
a more intense band around 247 nm, both of them being
largely of the MLCT type.
[CuII(S2CNH2)2]. In order to confirm the interpretation of
the absorption spectral changes mentioned above and shown in
Figure 3, the optical transitions of [CuII(S2CNH2)2] have also
6586
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
spectrophotometer using quartz cells with path length of 1 cm
recording in the 250−750 nm region. The cyclic voltammograms were
recorded on a CH Instruments 611C electrochemical analyzer using a
glassy carbon working electrode, Pt wire auxiliary electrode, Ag/
been computed at the TDDFT level on its equilibrium
geometry which was found to adopt a planar D2h geometry.
A strong optical LMCT transition (oscillator strength = 0.248
nm) is computed at 416 nm, whereas a HOMO-2 → LUMO
symmetry forbidden LMCT transition is found at 647 nm, the
latter being experimentally observable in the [CuII(S2CNR2)2]
(R ≠ H) complexes due to the lower symmetry induced by the
R group. These results fully agree with the indexing to the
[CuII(SCR2)2] species which displays two bands of lowest
energy in Figure 3.
AgNO3 reference electrode (0.33 V
10 mV versus SCE), and
standardized by the redox couple ferricinium/ferrocene. (Bu4N)-
(ClO4) (0.1 M) was used as the supporting electrolyte, and 50% (v/v)
CH3CN−CH2Cl2 solution was used for all of the samples. Electro-
chemical solutions were degassed with N2 and maintained under a
positive pressure for the duration of the experiment. All infrared
spectra were recorded on a Bruker Optics FTIR TENSOR 27
spectrometer at 20 °C using CsI plate. GC/MS analyses were carried
out on a Finnigan trace GC to which a Finnigan trace mass selective
detector was attached. The GC was equipped with an EQUITY-5
fused silica capillary column (30 m × 0.25 mm ×0.25 μm film
thickness). Helium was used as carrier gas and the following
temperature program was employed: initial isothermal period of 1
min at 35 °C, then an increase at 1 °C/min to 50 °C with an
isothermal period of 15 min at 50 °C. Emission spectra were recorded
on a Cary Eclipse B10 fluorescence spectrophotometer. The 77 K
emission lifetimes determined by a scope system at Fu-Jen Catholic
University used a PTI’s Model GL302 Dye Laser pumping by GL-
3300 Nitrogen Laser for excitation and a Hamamatsu P9220 PMT/
E717-63 socket assembly mounted on a Jobin-Yvon H-100
spectrometer for detection. Photomultiplier (PMT) output was
digitized using LeCory WaveRunner 6030A, 350 MHz normal analog
bandwidth, 2.5 GS/s single-shot sample rate/Ch, acquisition Memory
2 MB/4 MB for 4 ch/2 ch. Solid samples were prepared in capillary for
lifetime studies, and the 77 K emission lifetimes determined used
samples in 2 mm cylindrical cells immersed in liquid N2 in a Dewar as
previously described.49
CONCLUSION
■
The synthesis of hydride-centered tetracapped-tetrahedral Cu8
clusters surrounded by six 1,1-dichalcogenolate ligands has
been successfully extended to utilize dithiocarbamates as the
supporting ligands. Exploiting the redox-tolerant nature, which
is not a feature of dichalcogenophosphates, the encapsulated
hydride in compounds 1−3 can be oxidatively released via the
reaction of [Ce(NO3)6]2− to afford [CuII(dtc)2] and CuI salts.
Reactions of [CuII(dtc)2] and CuI in the presence of
borohydride in solution reproduce clusters 1−3 which can be
monitored by the UV−vis titration experiments.
Reactions of 1 equiv of borohydride with Cu8 clusters 1−3
produce neutral, heptanuclear copper clusters, 4−6, with a
copper framework which is a tricapped triangular pyramid in
which the hydride occupies a central tetrahedral site. The four-
coordinate hydride within the Cu7 cluster is unequivocally
authenticated by neutron diffraction analysis performed on a
single crystal of 6. The reactions of CuI ion with hydride-
centered Cu7 clusters in solution reproduce clusters 1−3 in
quantitative yield. In addition, the clusters 4−6 can be directly
synthesized from CuII salts, dithiocarbamates, and borohydrides
in a 7:6:8 molar ratio. Compounds 1−6 are air stable.
DFT calculations on model compounds fully reproduce the
experimental structures of the Cu8H and Cu7H species,
including the position of the hydride in the middle of the
central tetrahedron. The difference in the hydride chemical
shifts between the two species was matched by DFT. The UV−
vis absorption spectra of the studied compounds have been
indexed through TDDFT modeling and the luminescent nature
of 1−3 has been shown to be associated with a 3MLCT excited
state.
Monitoring of the Hydride Abstraction and Cluster
Reformation by UV−vis. Both forward and backtitration experi-
ments were monitored by UV−vis. Compound 1 in 50% (v/v)
CH3CN−CH2Cl2 solution (2 mL, 6.6 × 10−5 M) was added into the
quartz cells (3.5 mL, 1 cm) before determination. The titrants
including both oxidizer (NH4)2[Ce(NO3)6] (4.4 × 10−4 M) and
reductant, (Bu4N)(BH4) (4.4 × 10−4 M) were prepared with 50% (v/
v) CH3CN−CH2Cl2 solution, and 0.33 equiv (0.1 mL) of oxidizer
and/or reductant were stepwise added to the sample solution each
time.
Synthesis: [Cu4(μ4-H)(μ3-Cu)4{S2CNnPr2}6](PF6), 1H. Method A.
[Cu(CH3CN)4](PF6) (0.50 g, 1.338 mmol), Na(S2CNnPr2) (0.20 g,
1.003 mmol), and (Bu4N)(BH4) (0.043 g, 0.167 mmol) in 30 mL of
CH3CN were added to a Schlenk flask (100 mL) and were stirred at
room temperature for 1 h under nitrogen atmosphere. The reaction
mixture was evaporated to dryness under vacuum to obtain a yellow
solid. The yellow solid was added to 30 mL of CHCl3 to form a yellow
solution, and the solution was washed with deionized water (3 × 20
mL). The solution was filtered and the filtrate was evaporated to
dryness under vacuum to yield complex 1H as a yellow powder. Yield:
0.223 g (78%).
EXPERIMENTAL SECTION
■
Reagents and General Procedures. All chemicals were
purchased from commercial sources and used as received. Solvents
were purified following standard protocols. All reactions were
performed in oven-dried Schlenk glassware using standard inert-
atmosphere techniques. All reactions were carried out under N2
atmosphere by using standard Schlenk techniques. The preparation
of [Cu(CH3CN)4](PF6)47 and the Na(S2CR) ligands (R = NnPr2,
NEt2, and aza-15-crown-5) have been reported previously.48 The
syntheses of CuII(S2CR)2 (R = NnPr2, NEt2, and aza-15-crown-5) have
also been described.40 Melting points were measured by using a Fargo
MP-2D melting point apparatus. The elemental analyses were done
using a Perkin-Elmer 2400 CHN analyzer.
Method B. CuII(S2CNnPr2)2 (0.30 g, 0.720 mmol), [Cu-
(CH3CN)4](PF6) (0.447 g, 1.20 mmol), and (Bu4N)(BH4) (0.247
g, 0.960 mmol) in 30 mL CH3CN were added to a 100 mL Schlenk
flask and stirred at room temperature for 10 min under nitrogen
atmosphere. The reaction mixture was evaporated to dryness under
vacuum to obtain a yellow solid. The solid was washed with deionized
water (3 × 20 mL) and dried under vacuum to give complex 1H as a
yellow powder. Yield: 0.292 g (71%). Mp: 249 °C (dec). Anal. Calcd
for Cu8H85C42N6PF6S12·0.5CH3CN: C, 29.80; H, 5.03; N, 5.25%.
Spectroscopic Methods. NMR spectra were recorded on Bruker
Advance DPX300 FT-NMR spectrometer that operates at 300 MHz
1
Found: C, 29.95; H, 5.15; N, 5.19%. H NMR (300 MHz, CDCl3, δ,
3
1
2
while recording H, 121.5 MHz for 31P, and 46.1 MHz for H. The
31P{1H} NMR are referenced externally against 85% H3PO4 (δ = 0
ppm). The chemical shift (δ) and coupling constant (J) are reported in
parts per million and hertz. The NMR spectra were recorded at
ambient temperature. ESI-mass spectra were recorded on a Fison
Quattro Bio-Q (Fisons Instruments, VG Biotech, UK). UV−visible
absorption spectra were measured on a Perkin-Elmer Lambda 750
ppm): 7.05 (bs, 1H, μ4-H), 0.95 (t, JHH = 7.3 Hz, 36H,
3
CH2CH2CH3), 1.83 (m, 24H, CH2CH2CH3), 3.99 (t, JHH = 7.7 Hz,
24H, CH2CH2CH3). 31P NMR (CDCl3): −143.0 (septet, 1JPF = 706.2
Hz, PF6). ESI-MS (m/z)(Cal.): 1568.1 (1567.31) [Cu8(H)-
{S2CNnPr2}6]+. UV−vis [λmax in nm, (ε in M−1 cm−1)], 276(68
000), 345(37 000). IR (CsI, cm−1): ν(Cu−S), 233; ν(S−C), 944;
−
ν(C−N), 1501, ν(PF6 ), 843.
6587
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
[Cu4(μ4-D)(μ3-Cu)4{S2CNnPr2}6](PF6). 1D was prepared in a similar
fashion to 1H by replacing (Bu4N)(BH4) with (Bu4N)(BD4). Yield:
added to it and stirred at room temperature for 10 min. The reaction
mixture was evaporated to dryness under vacuum to obtain a yellow
solid. The solid was washed with ethanol (3 × 5 mL), and then dried
under vacuum to obtain complex 4H as yellow powder. Yield: 0.189 g
( 7 5 % ) . M p : 1 5 4 ° C ( d e c ) . A n a l . C a l c d f o r
Cu7H85C42N6S12·0.5CH2Cl2: C, 33.01; H, 5.61; N, 5.44%. Found: C,
32.99; H, 5.98; N, 5.22%. 1H NMR (300 MHz, CDCl3, δ, ppm): 6.55
0.209
g (73%). Mp: 253 °C (dec). Anal. Calcd for
Cu8H84DC42N6PF6S12 · 1.5CHCl3: C, 27.61; H, 4.66; N, 4.44%.
Found: C, 27.52; H, 4.99; N, 4.39%. H NMR (300 MHz, CDCl3, δ,
ppm): 0.94 (t, JHH = 7.3 Hz, 36H, CH2CH2CH3), 1.82 (m, 24H,
1
3
3
2
CH2CH2CH3), 3.98 (t, JHH = 7.7 Hz, 24H, CH2CH2CH3). H NMR
(46.1 MHz, CDCl3, δ, ppm): 7.07 (bs, 1D). 31P NMR (CDCl3):
(bs, 1H, μ4-H), 0.88 (t, JHH = 7.4 Hz, 36H, CH2CH2CH3), 1.78 (m,
3
1
−143.0 (septet, JPF = 706.1 Hz, PF6). ESI-MS (m/z)(Cal.): 1569.2
24H, CH2CH2CH3), 3.90 (t, 3JHH = 7.8 Hz, 24H, CH2CH2CH3). UV−
vis [λmax in nm, (ε in M−1 cm−1)], 274(65 000), 352(34 000). IR (CsI,
cm−1): ν(Cu−S), 201; ν(S−C), 959; ν(C−N), 1484.
(1568.32) [Cu8(D){S2CNnPr2}6]+. UV−vis [λmax in nm, (ε in
M−1cm−1)], 275(67 000), 345(37 000). IR (CsI, cm−1): ν(Cu−S),
−
233; ν(S−C), 943; ν(C−N), 1501, ν(PF6 ), 843.
[Cu4(μ4-D)(μ3-Cu)3{S2CNnPr2}6]. 4D was prepared as for 4H by
replacing (Bu4N)(BH4) with (Bu4N)(BD4). Yield: 0.128 g (85%). Mp:
159 °C (dec). Anal. Calcd for Cu7H84D1C42N6S12·2CH2Cl2: C, 31.56;
H, 5.42; N, 5.02%. Found: C, 31.49; H, 5.67; N, 4.54%. 1H NMR (300
[Cu4(μ4-H)(μ3-Cu)4{S2CNEt2}6](PF6). 2H was synthesized in a
similar procedure to 1H using Na(S2CNEt2) instead of Na(S2CNnPr2).
Yield: 0.195 g (65%). Mp: 238 °C (dec). Anal. Calcd for
Cu8H61C30N6PF6S12: C, 23.34; H, 3.98; N, 5.44%. Found: C, 23.24;
3
MHz, CDCl3, δ, ppm): 0.88 (t, JHH = 7.3 Hz, 36H, CH2CH2CH3),
1
3
H, 4.14; N, 5.64%. H NMR (300 MHz, CD2Cl2, δ, ppm): 7.02 (bs,
1.77 (m, 24H, CH2CH2CH3), 3.91 (t, JHH = 7.7 Hz, 24H,
1H, μ4-H), 1.31 (t, 3JHH = 7.4 Hz, 24H, CH2CH3), 4.05 (m, 3JHH = 7.4
Hz, 36H, CH2CH3). 31P NMR (CDCl3): −143.0 (septet, 1JPF = 706.1
Hz, PF6). ESI-MS (m/z)(Cal.): 1400.8 (1398.99) [Cu8(H)-
{S2CNEt2}6]+. UV−vis [λmax in nm, (ε in M−1 cm−1)], 276(63 000),
350(37 000). IR (CsI, cm−1): ν(Cu−S), 230; ν(S−C), 970; ν(C−N),
CH2CH2CH3). 2H NMR (46.1 MHz, CDCl3, δ, ppm): 6.40 (bs,
1D). UV−vis [λmax in nm, (ε in M−1 cm−1)], 273(65 000), 352(34
000). IR (CsI, cm−1): ν(Cu−S), 202; ν(S−C), 960; ν(C−N), 1485.
[Cu4(μ4-H)(μ3-Cu)3{S2CNEt2}6]. 5H was synthesized as described
for 4H by using [Cu4(μ4-H)(μ3-Cu)4{S2CNEt2}6](PF6) instead of
[Cu4(μ4-H)(μ3-Cu)4{S2CNnPr2}6](PF6). Yield: 0.106 g (79%). Mp:
137 °C (dec). Anal. Calcd for Cu7H61C30N6S12·2H2O: C, 26.27; H,
−
1505, ν(PF6 ), 843.
[Cu4(μ4-D)(μ3-Cu)4{S2CNEt2}6](PF6). 2D was prepared in a similar
fashion to 2H by replacing (Bu4N)(BH4) with (Bu4N)(BD4). Yield:
1
4.78; N, 6.13%. Found: C, 26.26; H, 4.91; N, 5.90%. H NMR (300
3
0.208
g (69%). Mp: 246 °C (dec). Anal. Calcd for
MHz, CD2Cl2, δ, ppm): 6.55 (bs, 1H, μ4-H), 1.31 (t, JHH = 7.3 Hz,
24H, CH2CH3), 4.04 (m, 3JHH = 8.4 Hz, 36H, CH2CH3). UV−vis [λmax
in nm, (ε in M−1 cm−1)], 275(64 000), 353(37 000). IR (CsI, cm−1):
ν(Cu−S), 210; ν(S−C), 978; ν(C−N), 1489.
Cu8H60DC30N6PF6S12: C, 23.32; H, 4.04; N, 5.44%. Found: C,
23.65; H, 4.04%; N, 5.19%. H NMR (300 MHz, CD2Cl2, δ, ppm):
1
3
3
1.30 (t, JHH = 7.0 Hz, 24H, CH2CH3), 4.05 (m, JHH = 7.0 Hz, 36H,
CH2CH3). 2H NMR (46.1 MHz, CD2Cl2, δ, ppm): 7.02 (bs, 1D). 31
P
[Cu4(μ4-D)(μ3-Cu)3{S2CNEt2}6]. 5D was prepared in a similar
fashion to 5H by replacing (Bu4N)(BH4) with (Bu4N)(BD4). Yield:
1
NMR (CDCl3): −143.0 (septet, JPF = 706.0 Hz, PF6). ESI-MS (m/
z)(Cal.): 1402.5 (1400.00) [Cu8(D){S2CNEt2}6]+. UV−vis [λmax in
nm, (ε in M−1 cm−1)], 276(63 000), 353(37 000). IR (CsI, cm−1):
0.109
g (81%). Mp: 142 °C (dec). Anal. Calcd for
Cu7H60DC30N6S12·1.5CH2Cl2: C, 25.85; H, 4.48; N, 5.74%. Found:
C, 25.78; H, 4.80; N, 5.59%. H NMR (300 MHz, CD2Cl2, δ, ppm):
1.30 (t, JHH = 7.3 Hz, 24H, CH2CH3), 4.04 (m, JHH = 8.4 Hz, 36H,
−
1
ν(Cu−S), 231; ν(S−C), 970; ν(C−N), 1505, ν(PF6 ), 843.
3
3
[Cu4(μ4-H)(μ3-Cu)4{S2C(aza-15-crown-5)}6](PF6). 3H was simi-
larly synthesized as described for 1H using (Na)[S2C(aza-15-crown-5)]
instead of Na(S2CNnPr2). Yield: 0.147 g (77%). Mp: 246 °C (dec).
Anal. Calcd for Cu8H121C66N6PF6O24S12·0.5(CH3)2CO: C, 33.09; H,
2
CH2CH3). H NMR (300 MHz, CD2Cl3, δ, ppm): 6.50 (bs, 1D).
UV−vis [λmax in nm, (ε in M−1 cm−1)], 276(64 000), 351(37 000). IR
(CsI, cm−1): ν(Cu−S), 208; ν(S−C), 978; ν(C−N), 1490.
1
5.10; N, 3.43%. Found: C, 33.17; H, 5.06; N, 3.64%. H NMR (300
[Cu4(μ4-H)(μ3-Cu)3{S2C(aza-15-crown-5)}6]. 6H was synthesized
as described for 4H by using [Cu4(μ4-H)(μ3-Cu)4{S2CN(aza-15-
crown-5)}6](PF6) instead of [Cu4(μ4-H)(μ3-Cu)4{S2CNnPr2}6] (PF6).
Yield: 0.097 g (87%). Mp: 217 °C (dec). Anal. Calcd for
Cu7H121C66N6O24S121: C, 35.83; H, 5.51; N, 3.80%. Found: C, 35.80;
H, 5.39; N, 3.61%. H NMR (300 MHz, CDCl3, δ, ppm): 6.49 (bs,
3
MHz, CDCl3, δ, ppm): 7.03 (bs, 1H, μ4-H), 4.40 (t, JHH = 5.8 Hz,
3
24H, NCH2CH2O), 3.93 (t, JHH = 6.0 Hz, 24H, NCH2CH2O), 3.64
1
(m, 72H, OCH2CH2O). 31P NMR (CDCl3): −143.0 (septet, JPF
=
706.0 Hz, PF6). ESI-MS (m/z)(Cal.): 2275.4 (2275.84) [Cu8(H)-
{S2CNC10H20O4}6]+. UV−vis [λmax in nm, (ε in M−1 cm−1)], 275(66
000), 340(35 000). IR (CsI, cm−1): ν(Cu−S), 233; ν(S−C), 991;
1H, μ4-H), 4.28 (t, 3JHH = 5.4 Hz, 24H, NCH2CH2O), 3.92 (t, 3JHH
=
−
ν(C−N), 1490, ν(PF6 ), 843.
5.7 Hz, 24H, NCH2CH2O), 3.64 (m, 72H, OCH2CH2O). UV−vis
[λmax in nm, (ε in M−1 cm−1)], 275(64 000), 340(36 000). IR (CsI,
cm−1): ν(Cu−S), 201; ν(S−C), 988; ν(C−N), 1483.
[Cu4(μ4-D)(μ3-Cu)4{S2C(aza-15-crown-5)}6](PF6). 3D was pre-
pared in a similar fashion to 3H by replacing (Bu4N)(BH4) with
(Bu4N)(BD4). Yield: 0.137 g (72%). Mp: 247 °C (dec). Anal. Calcd
for Cu8H120D1C66N6PF6O24S12·1(CH3)2CO: C, 33.42; H, 5.20; N,
[Cu4(μ4-D)(μ3-Cu)3{S2C(aza-15-crown-5)}6]. 6D was prepared as
for 6H by replacing (Bu4N)(BH4) with (Bu4N)(BD4). Yield: 0.094 g
(85%). Mp: 220 °C (dec). Anal. Calcd for Cu7H120D1C66N6O24S12: C,
1
3.39%. Found: C, 33.37; H, 5.25; N, 3.45%. H NMR (300 MHz,
3
1
CDCl3, δ, ppm): 4.41 (t, JHH = 5.8 Hz, 24H, NCH2CH2O), 3.93 (t,
35.82; H, 5.56; N, 3.80%. Found: C, 35.78; H, 5.34; N, 3.69%. H
2
3JHH = 6.0 Hz, 24H, NCH2CH2O), 3.66 (m, 72H, OCH2CH2O). H
3
NMR (300 MHz, CDCl3, δ, ppm): 4.30 (t, JHH = 5.8 Hz, 24H,
3
NMR (300 MHz, CDCl3, δ, ppm): 7.02 (bs, 1D). 31P NMR (CDCl3):
NCH2CH2O), 3.93 (t, JHH = 6.0 Hz, 24H, NCH2CH2O), 3.64 (m,
2
1
72H, OCH2CH2O). H NMR (300 MHz, CDCl3, δ, ppm): 6.51 (bs,
−143.0 (septet, JPF = 706.0 Hz, PF6). ESI-MS (m/z)(Cal.): 2276.7
1D). UV−vis [λmax in nm, (ε in M−1 cm−1)], 275(64 000), 340(36
000). IR (CsI, cm−1): ν(Cu−S), 202; ν(S−C), 988; ν(C−N), 1485.
[Cu{S2C(aza-15-crown-5)}2], 7. A synthesis of compound 7 has
been reported,40 but here we present a new synthesis. To a 50% (v/v)
CH3CN−CH2Cl2 solution of compound 3H (0.10 g, 0.041 mmol),
(NH4)2[Ce(NO3)6] (0.091 g, 0.164 mmol) was added. The yellow
solution turned brown rapidly, and the resulting brown solution was
stirred for about 1 h. The solution was filtered to remove any solid,
and the filtrate was evaporated to dryness to give a dark-brown
powder. This was washed with deionized water (5 × 5 mL) and dried
under vacuum to obtain complex 7 as brown powder. Yield: 0.054 g
(67%). Mp: 165 °C. Anal. Calcd for Cu1H40C22N2O8S4·1.5H2O: C,
39.95; H, 6.25; N, 4.24%. Found: C, 39.99; H, 5.95; N, 4.08%. UV−vis
[λmax in nm, (ε in M−1 cm−1)], 271(30 000), 434(14 000), 600(1000).
IR (CsI, cm−1): ν(Cu−S), 381; ν(S−C), 993; ν(C−N), 1499.
(2276.85) [Cu8(D){S2CNC10H20O4}6]+. UV−vis [λmax in nm, (ε in
M−1 cm−1)], 276(66 000), 341(35 000). IR (CsI, cm−1): ν(Cu−S),
−
233; ν(S−C), 990; ν(C−N), 1490, ν(PF6 ), 843.
[Cu4(μ4-H)(μ3-Cu)3{S2CNnPr2}6], 4H. Method A. To a dichloro-
methane solution (20 mL) of [Cu4(μ4-H)(μ3-Cu)4{S2CNnPr2}6](PF6)
(0.171 g, 0.1 mmol) was added (Bu4N)(BH4) (0.026 g, 0.1 mmol).
The solution color changed from yellow to red-brown after 1 h of
stirring, after which the solution was filtered through Al2O3 to remove
copper particles then evaporated to dryness under vacuum. The crude
product was washed with ethanol (3 × 5 mL), and then the solid was
dried under vacuum to obtain complex 4H as yellow powder. Yield:
0.123 g (82%).
Method B. Cu(BF4)2 (0.278 g, 1.171 mmol), Na(S2CNnPr2)2 (0.20
g, 1.003 mmol), and (Bu4N)(BH4) (0.344 g, 1.337 mmol) were
charged in a 100 mL Schlenk flask. A total of 30 mL of CH3CN was
6588
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
X-ray Crystallography. Crystals were mounted on glass fibers
with epoxy resin, and all geometric and intensity data were collected
on a Bruker APEXII CCD diffractometer using graphite monochro-
mated Mo Kα radiation (λ = 0.71073 Å). Data reduction was carried
out with SAINT-Plus software.50 An empirical absorption correction
was applied using the SADABS program.51 Structures were solved by
direct methods and refined by full-matrix least-squares on F2 using the
SHELXTL software package,52 incorporated in SHELXTL/PC version
5.10.53 The data collections for crystals of 1H, 2H, 6H, and 7 were
performed at 223(2), 100(2), 253(2), and 253(2) K, respectively. The
PF6 anion of 1H is disordered over two positions, each in 50%
occupancy. All Cu atoms in 2H are disordered over two positions with
50% occupancy, and the PF6 anion is disordered over two positions
with the major component at 75% occupancy. Several carbon and
oxygen atoms of the aza-15-crown-5 moiety in 6H are found to be
disordered at the temperature studied. O11, C17, and O22 are refined
isotropically the oxygen and carbon atoms (O1, O2, and C18) of the
aza-15-crown-5 in 7 are also disordered and refined isotropically. H-
atoms of the alkyl side chain or aza-15-crown-5 ligands are placed at
idealized positions except for those of the disordered carbon sites.
Pertinent crystallographic data are listed in Supporting Information
Table S1. Metric data of 1H, 2H, 6H, and 6H (neutron data) are in
Tables 1 and 7 of Table S4 in the Supporting Information.
the Def2-TZVP set from EMSL Basis Set Exchange Library.59 All
stationary points were fully characterized as true mimina via analytical
frequency calculations. The geometries obtained from DFT calcu-
lations were used to perform natural orbital population analysis by the
NBO 5.0 program.60 The composition of the molecular orbitals is
calculated using the AOMix program.61 The UV−visible transitions
were calculated by means of time-dependent DFT (TDDFT)
calculations,62 using the same PBE0 functional, but with a smaller
basis set, namely the general double-ξ polarized LANL2DZ basis
set63a−d augmented with Ahlrichs polarization functions on all
atoms,63e in order to reduce the computational demand. It has been
checked before on test calculations that this basis set change does not
modify significantly the energy of the first low energy transitions. Only
singlet−singlet, i.e. spin-allowed, transitions have been computed. The
UV−visible spectra were simulated from the computed TDDFT
transitions and their oscillator strengths by using the SWizard
program,64 each transition being associated with a Gaussian function
of half-height width equal to 3000 cm−1. The gauge including atomic
orbital (GIAO)65 method has been used to compute the 1H chemical
shifts, δ = σTMS − σcluster, where σTMS and σcluster are, respectively, the
isotropic chemical shielding of 1H in tetramethylsilane and of the
encapsulated hydride in the [Cu8(H)(S2CNH2)6]+ cluster.
Neutron Diffraction Details for 6H. Laue data were obtained
using the KOALA54 Laue neutron diffractometer located on a thermal
neutron guide at the OPAL reactor source of the Australian Nuclear
Science and Technology Organization. Images from the stationary
triclinic crystal were collected at 100 K across two series of eleven
frames, interimage 17° rotation of the crystal was perpendicular to the
incident beam. A data set was reduced from these images
corresponding to approximately 80% of the complete unique set to
0.85 < λ < 1.70 Å. The intensities for each set were indexed and
processed via the program LAUE1234,55 and the two sets were
merged Rint = 0.085 at the LAUE4 processing step. Structure
refinement using the CRYSTALS56 program suite was undertaken
commencing from the positions corresponding to the final X-ray
derived structural model. Since only the ratios between unit cell
dimensions can be determined from the Laue technique, the
dimensions found by X-ray diffraction (100 K) were used in the
neutron refinement; the ratios and angles observed were consistent
with the X-ray values. An initial refinement of all positions and
isotropic displacement parameters for the non-H atoms was followed
by deletion from the model of all H atom positions. All hydrogen
atoms presented here were clearly observed in a difference map as
negative density of a magnitude double that of remaining background,
and these H atom sites were reintroduced to the model at their
observed positions and included in the refinement with isotropic
displacement parameters (and in regions of disorder, with fractional
occupancy where consistent with the data). Similarity restraints were
employed in order to better model the disordered regions at the
periphery of the assembly, and light restraints were applied to H-atom
geometry around the aza-crown ether rings. There is no residual
difference density consistent with any additional atomic sites whether
hydrogen or non-hydrogen in crystals of this material except at the
periphery of the molecule where disorder of the crown ether related
groups is modeled and in the solvent containing region, minimum and
maximum residual densities are −1.36 and +1.96.
ASSOCIATED CONTENT
* Supporting Information
■
S
X-ray crystallographic files in CIF format for compounds 1H,
2H, 6H (X-ray structure), 6H (neutron structure), and 7
(CCDC-847804−847808); Figure S1−S5; crystallographic data
(Table S1, S2, and S4); and Cartesian coordinates of all the
DFT computed models (Table S3). This material is available
free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: chenwei@mail.ndhu.edu.tw (C.W.L.), jean-yves.
saillard@univ.rennes1.fr (J.-Y.S.).
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial supports from the National Science Council of
Taiwan (NSC 100-2113-M-259-003) and the Institut Uni-
versitaire de France (IUF) are gratefully acknowledged. Travel
facilities were provided through the French-Taiwanese
ORCHID project no. 19667PA. The award of neutron beam
access at ANSTO to Proposal 1177 is likewise acknowledged.
We thank professor Yuan-Jann Chen of Fu-Jen Catholic
University for the lifetime measurements.
REFERENCES
■
(1) (a) Transition Metal Hydrides; Dedieu, A., Ed.; VCH: New York,
1992. (b) Esteruelas, M. A.; Oro, L. A. Chem. Rev. 1998, 98, 577−588.
(c) Hoskin, A. J.; Stephan, D. W. Coord. Chem. Rev. 2002, 233−234,
107−129. (d) McGrady, G. S.; Guilera, G. Chem. Soc. Rev. 2003, 32,
383−392. (e) Konkol, M.; Okuda, J. Coord. Chem. Rev. 2008, 252,
1577−1591.
Due to the inherent weakness of neutron diffraction data compared
to X-ray derived data, the number of data above 3σ did not permit the
refinement of an anisotropic model and only the Cu and hydride H
atoms were modeled anisotropically. Final refinement cycles of 1017
parameters against 4074 reflections (I > 3σI) with 484 restraints
converged at R = 0.131, Rw = 0.127. A chebychev polynomial
weighting scheme was used and S = 1.00. The hydride H atom
coordinated to four Cu atoms of the cluster is clearly determined. All
geometries reported for this structure are calculated using the full
variance-covariance matrix of the refinement.
(2) (a) Recent Advances in Hydride Chemistry; Peruzzini, M.; Poli, R.,
Eds.; Elsevier: New York, 2001. (b) Maeland, A. J. In Recent Advances
in Hydride Chemistry; Peruzzini, M.; Poli, R., Eds.; Elsevier: New York,
2001; pp 531−556.
(3) (a) Adams, R. D.; Captain, B.; Smith, M. D. Angew. Chem., Int.
Ed. 2006, 45, 1109−1112. (b) Brayshaw, S. K.; Ingleson, M. J.; Green,
J. C.; Raithby, P. R.; Kociok-Kohn, G.; McIndoe, J. S.; Weller, A. S.
̈
Computational Details. Geometry optimizations were carried out
using the Gaussian 09 package,57 employing the PBE1PBE (PBE0)
functional,58 and using the general triple-ξ polarized basis set, namely
Angew. Chem., Int. Ed. 2005, 44, 6875−6878. (c) Brayshaw, S. K.;
Ingleson, M. J.; Green, J. C.; McIndoe, J. S.; Raithby, P. R.; Kociok-
Kohn, G.; Weller, A. S. J. Am. Chem. Soc. 2006, 128, 6247−6263.
̈
6589
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
(d) Shima, T.; Luo, Y.; Stewart, T.; Bau, R.; McIntyre, G. J.; Mason, S.
(b) Holm, R. H.; Ciurli, S.; Weigel, J. A. In Progress in Inorganic
Chemistry: Bioinorganic Chemistry; Lippard, S. J., Ed.; John Wiley &
Sons, Inc.: New York, 1990; Vol 38, p 1.
(29) Liu, P.; Lightstone, J. M.; Patterson, M. J.; Rodrigues, J. A.;
Muckerman, J. T.; White, M. G. J. Phys. Chem. B 2006, 110, 7449−
7455.
(30) (a) Adams, R. D.; Boswell, E. M.; Captain, B. Organometallics
2008, 27, 1169−1173. (b) Stark, J. L.; Harms, B.; Guzmann-Jimenez,
I.; Whitmire, K. H.; Gautier, R.; Halet, J.-F.; Saillard, J.-Y. J. Am. Chem.
Soc. 1999, 121, 4409−4418.
(31) Stephan, H.-O.; Kanatzidis, M. G.; Henkel, G. Angew. Chem., Int.
Ed. Engl. 1996, 35, 2135−2137.
(32) Preut, H.; Haupt, H.-J. Acta Crystallogr., Sect. B 1979, 35, 1205−
1207.
(33) Stevens, R. C.; McLean, M. R.; Bau, R.; Koetzle, T. F. J. Am.
Chem. Soc. 1989, 111, 3472−3473.
(34) (a) Wang, C.; Haushalter, R. C. Inorg. Chim. Acta 1999, 288, 1−
6. (b) Wang, C.; Haushalter, R. C. Chem. Commun. 1997, 1457−1458.
(35) Fenske, D.; Hachgenei, J.; Ohmer, J. Angew. Chem., Int. Ed. Engl.
1985, 24, 706−709.
(36) Bateman, L. W.; Green, M.; Howard, J. A. K.; Mead, K. A.; Mills,
R. M.; Dalter, I. D.; Stone, F. G. A.; Woodward, P. J. Chem. Soc., Chem.
Commun. 1982, 773−775.
(37) Bruce, M. I.; Nicholson, B. K. J. Chem. Soc., Chem. Commun.
1982, 1141−1143.
(38) Whitmire, K. H. J. Am. Chem. Soc. 1985, 107, 1056−1057.
(39) Al-Mandhary, M. R. A.; Buntem, R.; Doherty, C. L.; Edwards, A.
J.; Gallagher, J. F.; Lewis, J.; Li, C.-K.; Raithby, P. R.; Ramire de
Arellano, M. C.; Shields, G. P. J. Clust. Sci. 2005, 16, 127−150.
(40) (a) Granell, J.; Green, M. L. H.; Lowe, V. J.; Marder, S. R.;
Mountford, P.; Saunders, G. C.; Walker, N. M. J. Chem. Soc., Dalton
Trans. 1990, 605−614. (b) Jian, F.-F.; Wang, Z.-X.; Bai, Z.-P.; You, X.-
Z.; Fun, H.-K.; Chinnakali, K.; Razak, I. A. Polyhedron 1999, 18, 3401−
3406. (c) Giovagnini, L.; Sitran, S.; Montopoli, M.; Caparrotta, L.;
Corsini, M.; Rosani, C.; Zanello, P.; Dou, Q. P.; Fregona, D. Inorg.
Chem. 2008, 47, 6336−6343.
(41) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877−910.
(42) Sabin, F.; Ryu, C. K.; Ford, P. C.; Vogler, A. Inorg. Chem. 1992,
31, 1941−1945.
(43) Plyusnin, V. F.; Kolomeets, A. V.; Grivin, V. P.; Larionov, S. V.;
Lemmetyinen, H. J. Phys. Chem. A 2011, 115, 1763−1773 and
references therein.
A.; Hou, Z. Nature Chem. 2011, 3, 814−820.
(4) Kohn, R. D.; Pan, Z.; Mahon, M. F.; Kociok-Kohn, G. Chem.
̈
̈
Commun. 2003, 1272−1273.
(5) Kamiguchi, S.; Imoto, H.; Saito, T.; Chihara, T. Inorg. Chem.
1998, 37, 6852−6857.
(6) Adams, R. D.; Babin, J. E.; Tanner, J. T. Organometallics 1988, 7,
2027−2033.
(7) Jackson, P. F.; Johnson, B. F. G.; Lewis, J.; Raithby, P. R.;
Mcpartlin, M.; Nelson, W. J. H.; Rouse, K. D.; Alibon, J.; Mason, S. A.
J. Chem. Soc., Chem. Commun. 1980, 295−297.
(8) (a) Hart, D. W.; Teller, R. G.; Wei, C.-Y.; Bau, R.; Longoni, G.;
Campanella, S.; Chini, P.; Koetzle, T. F. J. Am. Chem. Soc. 1981, 103,
1458−1466. (b) Hart, D. W.; Teller, R. G.; Wei, C.-Y.; Bau, R.;
Longoni, G.; Campanella, S.; Chini, P.; Koetzle, T. F. Angew. Chem.,
Int. Ed. Engl. 1979, 18, 80−81.
(9) Broach, B. W.; Dahl, L. F.; Longoni, G.; Chini, P.; Schultz, A. J.;
Williams, J. M. In Transition Metal Hydrides, Advances in Chemistry
Series; Bau, R., Ed.; 1978, Vol. 167, Chapter 7.
(10) Yousufuddin, M.; Gutmann, M. J.; Baldamus, J.; Tardif, O.;
Hou, Z.; Mason, S. A.; Mclntyre, G. J.; Bau, R. J. Am. Chem. Soc. 2008,
130, 3888−3891.
(11) (a) Luo, Y.; Baldamus, J.; Tardif, O.; Hou, Z. Organometallics
2005, 24, 4362−4266. (b) Stewart, T.; Nishiura, M.; Konno, Y.; Hou,
Z.; Mcintyre, G. J.; Bau, R. Inorg. Chim. Acta 2010, 363, 562−566.
(12) Cheng, J.; Saliu, K.; Kiel, G. Y.; Ferguson, M. J.; McDonald, R.;
Takats, J. Angew. Chem., Int. Ed. 2008, 47, 4910−4913.
(13) Tanaka, S.; Dubs, C.; Inagaki, A.; Akita, M. Organometallics
2005, 24, 163−184.
(14) Shima, T.; Hou, Z. Dalton Trans. 2010, 39, 6858−6963.
(15) Serrano, C. B.; Less, R. J.; Mcpartlin, M.; Naseri, V.; Wright, D.
S. Organometallics 2010, 29, 5754−5756.
(16) (a) Budzichowski, T. A.; Chisholm, M. H.; Huffman, J. C.;
Eisenstein, O. Angew. Chem., Int. Ed. Engl. 1994, 33, 191−193.
(b) Budzichowski, T. A.; Chisholm, M. H.; Huffman, J. C.; Kramer, K.
S.; Eisenstein, O. J. Chem. Soc., Dalton Trans. 1998, 2563−2568.
(17) Kamiguchi, S.; Saito, T.; Honda, Z. J. Organomet. Chem. 2000,
609, 184−188.
(18) Eady, C. R.; Jackson, P. F.; Johnson, B. F. G.; Lewis, J.; Carlotta,
M.; Mcpartlin, M.; Nelson, W. J. J. Chem. Soc., Dalton Trans. 1980,
383−392.
(19) (a) Eguchi, T.; Heaton, B. T.; Harding, R.; Miyagi, K.; Longoni,
G.; Nahring, J.; Nakamura, N.; Nakayama, H.; Pakkanen, T. A.;
̈
(44) Garland, M. T.; Halet, J.-F.; Saillard, J.-Y. Inorg. Chem. 2001, 40,
Pursiainen, J.; Smith, A. K. J. Chem. Soc., Dalton Trans. 1996, 625−630.
(b) Eguchi, T.; Heaton, B. T. J. Chem. Soc., Dalton Trans. 1999, 3523−
3530.
(20) Li, X.-F.; Baldamus, J.; Nishiura, M.; Tardif, O.; Hou, Z. Angew.
Chem., Int. Ed. 2006, 45, 8184−8188.
3342−3350.
̌
(45) (a) Miertus, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55,
117−129. (b) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J. Comput.
Chem. 2003, 24, 669−681.
(46) Liu, C. W. unpublished results.
(21) (a) Liao, P.-K.; Sarkar, B.; Chang, H.-W.; Wang, J.-C.; Liu, C.-
W. Inorg. Chem. 2009, 48, 4089−4097. (b) Liu, C. W.; Sarkar, B.;
Huang, Y.-J.; Liao, P.-K.; Wang, J.-C.; Saillard, J.-Y.; Kahal, S. J. Am.
Chem. Soc. 2009, 131, 11222−11233. (c) Liu, C. W.; Chang, H.-W.;
Sarkar, B.; Saillard, J.-Y.; Kahlal, S.; Wu, Y.-Y. Inorg. Chem. 2010, 49,
468−475. (d) Liu, C. W.; Chang, H.-W.; Fang, C.-S.; Sarkar, B.; Wang,
J.-C. Chem. Commun. 2010, 46, 4571−4573.
(47) Kubas, G. J. Inorg. Synth. 1979, 19, 90−112.
(48) (a) Kaugars, G.; Rizzo, V. L. J. Heterocycl. Chem. 1981, 18, 411−
412. (b) Green, M. L. H.; Marder, S. R.; Saunders, G. C.; Walker, N.
M. J. Chem. Soc., Dalton Trans. 1988, 1697−1700.
(49) Chen, Y.-J.; Xie, P.; Endicott, J. F. J. Phys. Chem. 2004, 108,
5041−5049.
(50) SAINT V4.043: Software for the CCDC Detector System; Bruker
Analystic X-ray System: Madison, WI, 1995.
(22) Liao, P.-K.; Liu, K.-G.; Fang, C.-S.; Liu, C. W.; Fackler, J. P., Jr.;
Wu, Y.-Y. Inorg. Chem. 2011, 50, 8410−8417.
(51) Sheldrick, G. M. SADABS; University of Gottingen: Gottingen,
Germany, 1996.
(23) Casanova, D.; Llunell, M.; Alemany, P.; Alvarez, S. Chem.Eur.
J. 2005, 11, 1479−1494.
(52) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.
(53) SHELXL 5.10 (PC version): Program Library for Structure
Solution and molecular Graphics; Bruker Analytical X-ray System:
Madison, WI, 1998.
(54) Edwards, A. J. Aust. J. Chem. 2011, 64, 869−872.
(55) Piltz, R. O. Acta Crystallogr. 2011, A67, C155.
(56) Betteridge, P. W.; Carruthers, J. R.; Cooper, R. I.; Prout, K.;
Watkin, D. J. J. Appl. Crystallogr. 2003, 36, 1487.
(57) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A. Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
(24) Cardell, D.; Horgarth, G.; Faulkner, S. Inorg. Chim. Acta 2006,
359, 1321−1324.
(25) Hogarth, G. In Progress in Inorganic Chemistry, Karlin, K. D., Ed;
John Wiley & Sons, Inc.: New York, 2005; Vol 53, pp 71−560.
(26) Bondi, A. J. Phys. Chem. 1964, 68, 441−445.
(27) Cordero, B.; Gomez, V.; Platero-Prates, A. E.; Reves, M.;
Echeverria, J.; Cremades, E.; Barragan, F.; Alvarez, S. Dalton Trans.
2008, 2832−2838.
(28) (a) Berg, J. M.; Holm, R. H. In Metal Ions in Biology, Spiro, T.
G., Ed.; Wiley-Interscience: New York, 1982; Vol 4, Chapter 1.
6590
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591

Inorganic Chemistry
Article
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J.
B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,
O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.;
Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision B.
04; Gaussian, Inc.: Pittsburgh, PA, 2003.
(58) (a) Perdew, J. P.; Ernzerhof, M.; Burke, K. J. Chem. Phys. 1996,
105, 9982−9985. (b) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev.
Lett. 1996, 77, 3865−3868. (c) Perdew, J. P.; Burke, K.; Ernzerhof, M.
Phys. Rev. Lett. 1997, 78, 1396.
(59) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
(60) Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.; Carpenter, J.
E.; Bohmann, J. A.; Morales, C. M.; Weinhold, F. Theoretical
Chemistry Institute, University of Wisconsin, Madison, WI, 2001,
http://www.chem.wisc.edu/∼nbo5.
(61) Gorelsky, S. I. AOMix program, http://www.sg-chem.net/
(62) Burke, K.; Gross, E. K. U. In A guided tour of Time-Dependent
Density Functional Theory, in Density Functionals: Theory and
Applications; Joubert, D., Ed.; Springer: New York, 1998; Vol 500.
(63) (a) Dunning Jr., T. H.; Hay, P. J. In Methods of Electronic
Structure Theory; Schaeffer, H. F., Ed.; Plenum Press: New York, 1977.
(b) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270−283. (c) Hay,
P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 284−298. (d) Hay, P. J.;
Wadt, W. R. J. Chem. Phys. 1985, 82, 299−310. (e) Schafer, A.; Horn,
H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571−2577.
(64) Gorelsky, S. I. SWizard program, revision 4.5, http://www.sg-
chem.net/.
(65) (a) London, F. J. Phys. Radium 1937, 8, 397−409.
(b) McWeeny, R. Phys. Rev. 1962, 126, 1028−1034. (c) Ditchfield,
R. Mol. Phys. 1974, 27, 789−807. (d) Dodds, J. L.; McWeeny, R.;
Sadlej, A. J. Mol. Phys. 1977, 34, 1779−1791. (e) Wolinski, K.; Hinton,
J. F.; Pulay, P. J. Am. Chem. Soc. 1990, 112, 8251−8260.
6591
dx.doi.org/10.1021/ic300135w | Inorg. Chem. 2012, 51, 6577−6591