Substituent regulated photoluminescent thermochromism in a rare type of octahedral Cu4I4 clusters

Article abstract of DOI:10.1039/c8nj00505b

A series of Cu₄I₄ clusters (1-5) supported by two pyrazolate-type zwitterionic ligands N-methyl-[4-(5-R-pyrazolyl)]pyridinium (1: R = tert-butyl; 2: R = trifluoromethyl; 3, 4, 5: R = phenyl) have been prepared and fully characterized

Full text of DOI:10.1039/c8nj00505b

NJC
View Article Online
View Journal
PAPER
Substituent regulated photoluminescent
thermochromism in a rare type of octahedral
Cite this:DOI: 10.1039/c8nj00505b
Cu4I4 clusters†
Ya-Dong Yu, Ling-Bin Meng, Qiu-Cheng Chen, Guang-Hui Chen
Xiao-Chun Huang
and
*
A series of Cu4I4 clusters (1–5) supported by two pyrazolate-type zwitterionic ligands N-methyl-[4-(5-R-
pyrazolyl)]pyridinium (1: R = tert-butyl; 2: R = trifluoromethyl; 3, 4, 5: R = phenyl) have been prepared
and fully characterized. These compounds all exhibit a distorted ‘‘octahedral’’ geometry and their
luminescence behavior is quite distinct from those of previously reported tetranuclear copper halide
clusters. The crystalline powders of all five clusters only show a single broad low-energy (LE) emission
band from 300 to 80 K. Cluster 1 shows temperature-independent orange phosphorescence despite a
slight blue-shift of the low-energy flank upon cooling. Intriguingly, for clusters 2–5, a hypsochromic
phenomenon is clearly observed by decreasing the temperature. Using DFT/TDDFT calculations,
we find that the nature of the electronic transitions responsible for the LE emission is dominated by the
steric demand and/or electronic characteristics of the substituents on the pyrazolyl moiety. Specially,
contribution of cuprophilic interactions in the lowest triplet excited states is ruled out by the rigid
ligands. The single LE emission of 1 and 2 is attributed to pure 3ILCT and 3(M + X)LCT transitions,
respectively, while the emission of 3 and 4 is assigned to a mixed 3ILCT and 3(M + X)LCT transition with
different proportions.
Received 29th January 2018,
Accepted 19th March 2018
DOI: 10.1039/c8nj00505b
rsc.li/njc
1. Introduction
Multinuclear copper(I) halide clusters possessing diverse structural
and rich photophysical properties1–5 have been of great interest
for decades due to their wide range of potential applications,
including stimuli-responsive luminescent materials,6–14 white-
light emitters,15 and organic light-emitting diodes (OLEDs).16–20
In particular, tetranuclear Cu4X4Ln (X = Cl, Br, and I) clusters
have been extensively investigated because of their original
luminescence behavior. The Cu4X4 cores often adopt cubane,
octahedral, or stair-step structures with four copper atoms
forming a Cu4 tetrahedron, a square plane, or a zigzag chain,
respectively (Fig. 1a). The cubane-like clusters are the most
widely studied structural motifs, which are well-known for their
thermochromic luminescence properties.6–9,21,22 They usually
display two distinct emission bands from two different triplet
Department of Chemistry and Key Laboratory for Preparation and Application of
Ordered Structural Materials of Guangdong Province, Shantou University,
Guangdong 515063, China. E-mail: xchuang@stu.edu.cn; Fax: (+86)-0754-82902767
† Electronic supplementary information (ESI) available: The X-ray crystallo-
graphic file (CIF), tabulated structural parameters, supplemental theoretical
and experimental spectra, and other characterization data are provided. CCDC
1525210–1525215 and 1525981–1525983. For ESI and crystallographic data in CIF
or other electronic format see DOI: 10.1039/c8nj00505b
Fig. 1 Illustrations of (a) structural motifs of the Cu4X4 cores (redrawn
from the structural data, Cu, yellow; X, purple), (b) bidentate ligands used
in the ‘‘octahedral’’ clusters, (c) the zwitterionic ligands used in this work,
(d) the chemical structure of one of the title compound.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
NJC
excited states with relative intensities that vary sharply with
To better comprehend the optical properties of the octa-
temperature. The low-energy band (around 550–600 nm) origi- hedral copper halide clusters and explore the impact of the
nating from triplet cluster-centered (3CC) charge transfer ligands on the origin of the electronic transitions, we have both
transitions dominates at room temperature (RT), in which the experimentally and theoretically investigated several copper
wavelength is closely related to cuprophilic interactions, the iodide clusters bearing pyrazolate-type zwitterionic ligands.
shorter the Cu–Cu distances, the longer is the wavelength. Pyrazolate ligands have been widely investigated in cyclic
While at low temperatures, the high-energy band (around trinuclear copper(I) complexes.41–45 Pyridinium derivatives are
400–450 nm) originating from halide-to-ligand charge transfer strongly electron deficient and have been reported to exhibit
excited states (3XLCT) (sometimes mixed with metal-to-ligand thermochromic phenomena in the solid state upon hydration
charge transfer, 3MLCT) becomes more prominent. Distinct and dehydration.46 The combination of pyrazole and pyridi-
from the cubane-type copper halide clusters, the stair-step nium with different substituents on the pyrazolyl moiety gives
3
clusters exhibit phosphorescence only from the XLCT/3MLCT rise to three proligands with iodide as the counterion (HL1ÁI,
excited states due to very weak cuprophilic interactions.8,23
HL2ÁI, and HL3ÁI), which convert into neutral but charge-
Octahedral Cu4X4L2 motifs are less-studied compared to cubane separated zwitterionic ligands after losing a molecule of HI
analogues presumably due to the difficulty in synthesis.24–40 (Fig. 1c). The chemical structure of one of the title compound is
According to the theoretical calculations by Thompson et al. presented in Fig. 1d. These ligands without the bridging atoms
the bare octahedral Cu4I4 cores are energetically unfavorable between the two N-donors are expected to be more rigid than
relative to the cubane Cu4 array.29 In the limited reports of previously reported phosphine-based chelating ligands and
octahedral Cu4X4L2 compounds, two copper atoms are bridged possess a charged organic surface which is likely to exhibit
by phosphine-based bidentate ligands bearing small bite angles, intriguing luminescence properties. tert-Butyl, trifluoromethyl,
such as P^N donors (I–IV), P^P ligands (V–VIII), and P^C carbene and phenyl are expected to finely tune and modulate the
IX (Fig. 1b). Among these reports, the synthesis and structural emission properties of the clusters.
characterization are the majority and only scarce photophysical
studies are available.25–31
Herein, we report the synthesis, structural characterization,
photophysical properties, and theoretical study of five octa-
On the basis of these studies on octahedral Cu4X4L2 complexes, hedral copper iodide clusters [Cu4I4(L1)2]ÁCH3CN (1), [Cu4I4(L2)2]
the rigidity and steric demands of the ligands play a significant (2), [Cu4I4(L3)2]ÁCH3CN (3), [Cu4I4(L3)2] (4), and [Cu4I4(L3)2]Á
role on the optical properties of the clusters. In the example with 2C13H10O (5, C13H10O denotes benzophenone). Thermochromic
ligand I, Thompson and co-workers synthesized a series of behavior was observed for these clusters in the solid state in
octahedral copper iodide complexes formulated as Cu4I4PR2, which the LE emission bands are blue-shifted upon lowering the
which exhibit dual temperature-dependent 3XLCT and 3CC temperature. This unique hypsochromic phenomenon differs
emission similar to copper cubanes when the ligand contains from previous reports on traditional cubane-type clusters and
phenyl, cyclohexyl, or ethyl groups. However, the bulky tert-butyl other octahedral analogues. To rationalize these interesting
or iso-propyl substituent on ligand I imparts the clusters with optical properties in the solid state, the nature of the excited
3
only one high energy band from the XLCT excited states. The state involved in the absorption and emission spectra for
authors reasoned that the structural relaxation of the 3CC excited clusters 1–4 has been investigated using density functional
state is limited by the bulky ligands, impeding the development theory (DFT) and time dependent DFT (TDDFT) calculations.
of the LE emission band.29 With the more rigid ligand II in There is no convincing evidence which shows that these
which the CH2 spacer linking the organophosphine moiety and clusters maintain integrity in solution. Therefore, the photo-
the pyridine ring is missing compared to ligand I, Catalano and physical properties in acetonitrile solution of clusters 1–4 were
co-workers have also prepared several octahedral Cu4X4L2 (X = Cl not investigated in detail.
or Br) clusters showing only one single temperature-independent
LE emission band at around 520 nm assigned to (M + X)LCT.
Despite the appreciable cuprophilic interactions in these
complexes, such an optical phenomenon is in stark contrast to
copper cubanes or other octahedral Cu4I4L2 clusters, however,
2. Results and discussion
2.1 Synthesis and structural characterization of the clusters
resembling the stair-step copper halide clusters.28 Employing The pyrazolate-type zwitterionic ligands were synthesized by a
ligand III (R = H or CH3) which introduces a phospholane unit to Claisen condensation reaction giving rise to diketones, followed
replace the PPh2 moiety in ligand II, Musina obtained two by treatment with hydrazine hydrate and iodomethane
octahedral Cu4I4L2 compounds with photophysical properties (Scheme S1, ESI†). The IR and 1H NMR spectra of the three
in analogy with the cubane structures through both experi- ligands are shown in Fig. S1–S4 (ESI†). Clusters 1–3 and 5 were
mental and theoretical methods.26 From these studies, the synthesized by slowly vaporizing the solutions while complex 4
emissions observed seem to have different origins defined was prepared by solvothermal reactions. The five clusters are
by the ligands. Therefore, the influence of the properties of merely moderately soluble in acetonitrile. All of these com-
the ligands, such as rigidity, steric hindrance and electronic pounds were characterized by elemental analyses, IR (Fig. S1,
characteristics, on the luminescence behavior of the octahedral ESI†), 1H NMR (Fig. S5–S9, ESI†), ESI-MS (Fig. S10–S17, ESI†)
Cu4I4L2 clusters has yet not been addressed.
spectroscopies and PXRD analysis (Fig. S18, ESI†). More details
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
about the synthesis and characterization are found in the ESI.† complexes 3–5 are composed of the same Cu4I4 cores and
In PXRD analysis, the peaks of the powdered sample correspond ligands, showing significantly different emissions in the solid
to those calculated from the data of the single crystal diffraction state (vide infra). Compounds 3 and 4 could be treated as cis–trans
measurements, which confirms its purity.
isomers, while compound 5 additionally contains cocrystallized
These five copper iodide clusters are also characterized by benzophenone molecules compared to 4.
single crystal X-ray analysis at RT. For compounds 2–5, the data
The Cu–N bond lengths of the five clusters fall between
were also collected at 100 K. The X-ray crystallographic data of 1.896 and 2.028 Å (average = 1.967(8) Å) similar to the values of
the clusters are provided in Table S1 (ESI†). As shown in Fig. 2, the reported octahedral clusters with P^N donors.29 The average
these clusters possess a rare ‘‘octahedral’’ geometry that differs distance between Cu and m2-bound iodide (2.568(1) Å) is greatly
significantly from the cubane or the stair-step configurations. shorter than that between Cu and m3-bridging iodide (2.936(2) Å).
Except for cluster 3, the four copper atoms are arranged in The Cu–Cu bond distances at the short sides of the Cu4 array
a parallelogram with obtuse angles ranging from 93.141 to range from 2.466(4) to 2.637(5) Å (average = 2.547(5) Å) and vary
110.801. In particular, the Cu4 array forms a trapezoid with an from 2.834(4) to 2.918(9) Å (average = 2.882(9) Å) at the long sides
obtuse angle of 91.721 for cluster 3. The planar Cu4 array is of the array, which are comparable with those in previously
axially capped by two m3-iodides forming a distorted octahedron reported octahedral clusters. The Cu–Cu distances for 2–5
and two additional m2-iodides bound on the each short side of between 293 K and 100 K are almost identical within the
the Cu4 array. There is a long nonbonding interaction between experimental error (Tables 1–5). This is in stark contrast to the
the iodide ions and the copper atoms. The zwitterionic ligands cubane-like clusters in which the Cu–Cu distances always
are coordinated to the copper atoms on the opposing long sides become shorter when lowing the temperature.21
of the Cu4 array in a ‘‘head-to-tail’’ arrangement for 1, 2, 4 and
2.2 Luminescence properties in the solid state
5, while in a ‘‘head-to-head’’ manner for 3. For the first time,
to our knowledge, two dissymmetric chelating ligands are Clusters 1–5 are yellow or brown crystalline powers under
arranged in a ‘‘head-to-head’’ manner in octahedral clusters. ambient light. At RT, 2, 3 and 5 display a bright red emission,
Complex 1 contains two independent clusters in the asymmetric while 1 and 4 emit orange and yellow light, respectively. The
unit (1a and 1b). The tetranuclear copper core adopts a more thermochromic luminescence properties of the clusters are
skewed parallelogram geometry in 1a than in 1b with obtuse revealed by dipping the samples into liquid nitrogen. Under
angles of 110.801 and 100.811, respectively. The three analogous the same UV irradiation, all the five clusters exhibit intense
Fig. 2 Molecular structure of clusters 1–5 (H atoms are omitted for clarity. 1a and 1b are the two independent clusters in the asymmetric unit for 1).
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
NJC
Table 1 Selected bond distances (Å) and angles (deg) of 1 at 293 K
Table 4 Selected bond distances (Å) and angles (deg) of 4 at 293 K and
100 K
1a
1b
4 (293 K)
4 (100 K)
Cu1–Cu2
Cu1–Cu2a
Cu1–I1
Cu1–I2a
Cu1–I1a
Cu2–I2
Cu2–I1a
Cu2–I1
Cu1–N2
2.834(4) Cu3–Cu4a
2.549(4) Cu3–Cu4
2.696(4) Cu3–I3
2.700(3) Cu3–I4a
2.964(4) Cu3–I4
2.473(4) Cu4–I3
2.684(4) Cu4–I4
3.739(4) Cu4–I4a
2.028(15) Cu3–N5
1.942(16) Cu4–N6a
110.80(14) Cu4–Cu3–Cu4a
69.20(14) Cu3–Cu4–Cu3a
2.891(4)
2.544(4)
2.607(3)
2.665(4)
2.993(4)
2.499(3)
3.028(4)
3.107(4)
1.974(15)
1.896(16)
100.81(13)
79.19(13)
Cu1–Cu2a
Cu1–Cu2
Cu1–I2
Cu1–I1a
Cu1–I1
Cu2–I2
Cu2–I1
Cu2–I1a
Cu1–N1
2.906(11) Cu1–Cu2a
2.572(9) Cu1–Cu2
2.597(8) Cu1–I2
2.795(8) Cu1–I1a
2.907(9) Cu1–I1
2.528(8) Cu2–I2
2.732(9) Cu2–I1
3.207(9) Cu2–I1a
1.978(4) Cu1–N1
1.949(3) Cu2–N2a
2.915(11)
2.575(9)
2.612(8)
2.747(8)
2.885(9)
2.527(8)
2.696(8)
3.288(8)
1.994(4)
1.952(4)
98.49(3)
81.51(3)
À61.6(4)
0.0
Cu2–N3
Cu2a–Cu1–Cu2
Cu1a–Cu2–Cu1
Cu2a–Cu1–Cu2–Cu1a
N2–Cu1–Cu2–N3
Cu2–N2a
Cu2–Cu1–Cu2a
Cu1–Cu2–Cu1a
N1–Cu1–Cu2–N2a
Cu2a–Cu1–Cu2–Cu1a
95.70(3)
84.30(3)
À50.1(4)
0.0
Cu2–Cu1–Cu2a
Cu1–Cu2–Cu1a
N1–Cu1–Cu2–N2a
Cu2a–Cu1–Cu2–Cu1a
0.0
À7.2(8)
Cu4a–Cu3–Cu4–Cu3a 0.0
N5–Cu3–Cu4–N6a
À30(3)
Table 2 Selected bond distances (Å) and angles (deg) of 2 at 293 K and
100 K
Table 5 Selected bond distances (Å) and angles (deg) of 5 at 293 K and
100 K
2 (293 K)
2 (100 K)
5 (293 K)
5 (100 K)
Cu1–Cu2
Cu1–Cu2a
Cu1–I2
Cu1–I1a
Cu1–I1
Cu2–I2a
Cu2–I1
Cu2–I1a
Cu1–N1
2.918(9) Cu1–Cu2
2.520(8) Cu1–Cu2a
2.572(7) Cu1–I2
2.782(8) Cu1–I1a
2.935(9) Cu1–I1
2.514(7) Cu2–I2a
2.727(9) Cu2–I1
3.262(9) Cu2–I1a
1.972(3) Cu1–N1
1.947(3) Cu2–N2
2.920(15)
2.515(13)
2.589(11)
2.765(11)
2.854(13)
2.506(11)
2.679(13)
3.331(13)
1.980(6)
1.951(6)
99.74(4)
80.26(4)
0.0
Cu1–Cu2a
Cu1–Cu2
Cu1–I2
Cu1–I1
Cu1–I1a
Cu2–I2
Cu2–I1
Cu2–I1a
Cu1–N2
2.895(16) Cu1–Cu2a
2.544(14) Cu1–Cu2
2.594(11) Cu1–I2
2.807(14) Cu1–I1
3.036(14) Cu1–I1a
2.576(11) Cu2–I2
2.732(12) Cu2–I1
2.971(13) Cu2–I1a
1.986(5) Cu1–N2
1.993(5) Cu2–N1a
2.892(10)
2.545(10)
2.591(8)
2.791(8)
2.997(8)
2.575(7)
2.707(7)
2.947(7)
1.979(4)
1.990(4)
87.04(3)
92.96(3)
À8.6(4)
0.0
Cu2–N2
Cu2a–Cu1–Cu2
Cu1a–Cu2–Cu1
Cu2a–Cu1–Cu2–Cu1a
N1–Cu1–Cu2–N2
96.71(3)
83.29(3)
0.0
Cu2a–Cu1–Cu2
Cu1a–Cu2–Cu1
Cu2a–Cu1–Cu2–Cu1a
Cu2–N1a
Cu2–Cu1–Cu2a
Cu1–Cu2–Cu1a
N2–Cu1–Cu2–N1a
Cu2a–Cu1–Cu2–Cu1a
86.86(5)
93.14(5)
À6.1(7)
0.0
Cu2–Cu1–Cu2a
Cu1–Cu2–Cu1a
N2–Cu1–Cu2–N1a
Cu2a–Cu1–Cu2–Cu1a
2.81(18) N1–Cu1–Cu2–N2
3.0(3)
Table 3 Selected bond distances (Å) and angles (deg) of 3 at 293 K and
100 K
of 1–5), and the corresponding photophysical data are provided
in Table S2 (ESI†). At 300 K, the emission spectra all display a
single featureless broad low-energy emission band centered at
640 nm for 1, at 650 nm for 2, at 655 nm for 3, at 615 nm for 4,
and at 695 nm for 5 in agreement with the corresponding
emission color observed at RT. The different excitation wave-
lengths have no influence on the emission maxima (see Fig. S20,
ESI†). The broad excitation and emission spectra indicate the
charge transfer (CT) nature of the involved excited states.47
By lowering the temperature down to 80 K, a significant
3 (293 K)
3 (100 K)
Cu1–Cu2
2.852(4)
2.637(5)
2.466(4)
2.741(3)
2.540(3)
3.242(3)
2.615(3)
2.703(3)
3.010(3)
1.993(12) Cu1–N1
1.947(13) Cu2–N2
88.28(6)
91.72(6)
2.4(6)
0.0
Cu1–Cu2
Cu1–Cu1a
Cu2–Cu2a
Cu1–I1
Cu1–I4
Cu1–I3
Cu2–I2
Cu2–I3
Cu2–I1
2.841(14)
2.651(19)
2.474(17)
2.725(11)
2.547(11)
3.210(11)
2.618(10)
2.706(10)
2.958(10)
1.981(5)
1.972(5)
88.22(2)
91.78(2)
1.9(2)
Cu1–Cu1a
Cu2–Cu2a
Cu1–I1
Cu1–I4
Cu1–I3
Cu2–I2
Cu2–I3
Cu2–I1
Cu1–N1
blue-shift of the emission from 650 to 597 nm (E1366 cmÀ1
)
Cu2–N2
for 2, from 655 to 607 nm (E1207 cmÀ1) for 3, from 615
to 577 nm (E1071 cmÀ1) for 4, and from 695 to 644 nm
(E1139 cmÀ1) for 5 were observed with a concomitant increase
of the intensity. The luminescent behavior of 1 is very different
from those of 2–5. Indeed, despite a slight blue-shift of the
low energy flank observed on cooling, the emission position is
Cu1a–Cu1–Cu2
Cu2a–Cu2–Cu1
N1–Cu1–Cu2–N2
Cu1a–Cu1–Cu2–Cu2a
Cu1a–Cu1–Cu2
Cu2a–Cu2–Cu1
N1–Cu1–Cu2–N2
Cu1a–Cu1–Cu2–Cu2a
0.0
yellow emission (Fig. 3 inset). The original emission colors are almost unchanged with maxima at 640 nm at 300 K and 633 nm
recovered when the samples gradually warm up to RT again, at 80 K. In addition, the emission lifetimes of the five clusters
indicating a reversible thermochromism for the clusters.
are around 5 ms at 300 K and become slightly longer at 80 K,
Solid-state emission spectra were recorded for clusters 1–5 suggesting that they are all phosphorescence stemming from
from 300 to 80 K (Fig. 3 for normalized emission and excitation triplet states. A striking feature is that the three analogues
spectra of 1–5 and Fig. S19 (ESI†) for original emission spectra 3–5 exhibit distinct luminescence, in particular, the cis–trans
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
Fig. 3 Normalized temperature dependence of solid-state luminescence spectra of cluster 1–5 from 300 to 80 K (lex = 330 nm for 1 and lex = 340 nm
for 2–5) and corresponding excitation spectra (yellow lines), corresponding lem 640 nm for 1, 650 nm for 2, 640 nm for 3, 600 nm for 4, 685 nm for 5.
The photos of 1–5 under UV irradiation at 365 nm (UV lamp) at RT and in the liquid nitrogen (77 K) are presented in the inset.
isomers 3 and 4 display red and yellow emission at RT, 2.4 Exploration of the photophysical properties for complexes
respectively.
1–4 in the solid state
2.4.1 Computed bond distances. To gain insight into
the photophysical properties, combined DFT and TDDFT
2.3 Investigation of the luminescence properties in solution
Because the octahedral clusters Cu4X4(PPh2py)2 (X = Br, Cl, calculations were performed for clusters 1–4 using coordinates
PPh2py = 2-(diphenylphosphino)pyridine) were reported to be obtained from the X-ray data. The results reported below merely
dynamic in solution,28 the stability of clusters 1–4 in acetonitrile concern optimized geometries of isolated clusters unless
solution was investigated. Fig. S21 (ESI†) shows the concentration otherwise stated. The geometries of the singlet ground state
dependence of the UV-vis absorption spectra of clusters 1–4 in (S0) and of the lowest triplet state (T1) under spin-unrestricted
CH3CN at RT. As exemplified in the inset of Fig. S21 (ESI†), a calculations (UDFT) were fully optimized for 1–4. Geometry
linear Beer’s law plot of concentration-dependent absorbance for optimization of the T1 with the TDDFT approach was only
1 and 2 was observed, suggesting a single species or a collection of successful for 2 and 3. We found that the PBE0 and M06-2X
species of identical composition existing in CH3CN solution. functional will severely underestimate the emission energy
However, only fragments of the parent clusters like [Cu3I2(L1)2]+ T1 - S0 for 2, thus the more suitable range-separated oB97X-D
or [Cu3I2(L2)2]+ were found in the MALDI-TOF-MS spectra functional was used throughout except the geometry optimization
(Fig. S22, ESI†), which implies that 1 and 2 probably do not calculations (Table S3, ESI†).
maintain integrity in acetonitrile and undergo fragmentation to
species of lower nuclearities. The obvious non-linear relationships TT1DDFT) are compared to the experimental metrical values (exp)
of 3 and 4 indicate that they dissociate in CH3CN solution.
in Table S4 (ESI†). The computed values for the ground state
Relevant computed bond distances for 1–4 (S0, TU1 DFT, and
The emission spectra of the proligands in acetonitrile are in fair agreement with the X-ray experimental data, despite
solution at RT are provided in Fig. S23 (ESI†). The lumines- a small overestimation of the bond distances that can be
cence spectra of clusters 1–4 in dry acetonitrile solution at RT probably traced back to the neglect of the solid-state packing
and at 77 K are shown in Fig. S24 (ESI†). Distinct from the low effects. For 2–4, it is worth noting that the Cu–Cu separations
energy (LE) phosphorescence in the solid state, these clusters from T1 optimized geometries are longer and the Cu–I bond
display high energy (HE) emission in acetonitrile solution at RT. lengths become shorter than those at S0 optimized structures,
When the temperature is decreased to 77 K, an intense LE emission whereas it is reversed in 1.
band is observed in the glassy CH3CN matrix for all 1–4 from 655 to
2.4.2 Molecular orbitals.
A schematic representation
670 nm which is in agreement with the red light observed.
and isodensity surface plots of the frontier molecular orbitals
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
NJC
Fig. 4 Schematic representation and isosurface plots (isovalue 0.0005) of the frontier molecular orbitals for the singlet ground state calculated at the S0
optimized geometries. The orbital compositions (in brackets), HOMO and LUMO energies (in green), and the HOMO–LUMO energy gaps (in red) are also
listed.
calculated at the S0 geometries for 1–4 are shown in Fig. 4. the energy gaps of 3 bearing two ‘‘head-to-head’’ ligands and of
Isosurface plots of other selected molecular orbitals (HOMOÀ1/ 4 possessing two ‘‘head-to-tail’’ ligands are 5.28 eV and 5.72 eV,
HOMOÀ4, LUMO+1/LUMO+8) are shown in Fig. S25–S28 respectively.
(ESI†). The four clusters have similar frontier orbitals. As shown
2.4.3 Optical absorption spectra. The solid-state absorp-
in the figure, the highest occupied molecular orbitals (HOMOs) tion spectra of clusters 1–4 are shown in Fig. 5. The computed
of these clusters are combinations of copper d and iodine p data in vacuo are also shown in Fig. 5, their height being
orbitals (495%), while the unoccupied molecular orbitals proportional to their oscillator strength ( f ). The calculated
(LUMOs) are mainly located on the pyrazolyl groups and TDDFT singlet–singlet of the first 10 and most intense
N-methyl pyridinium moieties of the ligands (497%). The ( f 4 0.03) transitions and of the 10 lowest singlet–triplet
LUMO+4 for 1–2 and LUMO+6 for 3–4 represent a combination transitions with transition assignments in terms of electron
of the copper and iodine s–p state, which have a bonding density difference (EDD) maps at the S0 optimized geometry in
character among the Cu4 array and the antibonding character- the gas state for all 1–4 are listed in Tables S5–S12 (ESI†).
istic between all the Cu and I atoms. Such molecular orbital
As depicted in Fig. 5, the solid-state absorption spectra
compositions are similar to a previous study for the cubane- contain three absorption regions: (1) the first region, a high
type and other reported octahedral clusters.21,29 Although the energy band originating from 200–260 nm for all 1–4; (2) the
compositions of the frontier orbitals are not impacted by the second region, a broad absorption band from 260–325 nm
substituent groups, the energy levels of the orbitals and centered at 280 nm for 1, 3, 4 and from 250–300 centered at
HOMO–LUMO energy gaps of these clusters are significantly 270 nm for 2; (3) the third region, a low energy band from
affected by them. For instance, the electron-withdrawing CF3 325 nm tailing to 550 nm centered at 370 nm for 1, 3, 4, and
group in 2 will stabilize the frontier orbitals relative to the bulky from 300–550 centered at ca. 360 nm for 2. The calculated
butyl group in 1 or the p-conjugated phenyl group in 3 and 4. wavelength and oscillator strength nicely correlate with the
Notably, the symmetry of the molecules has a significant presence of mainly three absorption regions in the experi-
influence on the HOMO–LUMO gaps. As described in Fig. 6, mental spectra. The intense transitions of sizable intensity
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
compound is in the solid state, it seems that the characteristics
of the ligands, such as rigidity, steric demands, and electronic
properties, are operative in modifying the photophysical
character of the octahedral copper iodide clusters. On one
hand, only a single emission band is observed upon cooling
in all five compounds similar to the octahedral clusters with
rigid ligand II in ref. 28. On the other hand, the evident blue-
shift of the LE phosphorescence in clusters 2–5 by lowering the
temperature had not been observed in copper cubanes or other
octahedral Cu4X4L2 clusters in which the LE emission band is
apt to show the bathochromic shift due to the shortening of
Cu–Cu distances upon cooling. Notably, the blue-shift behavior
of the luminescence stemming from the 3XLCT excited state of
tetranuclear clusters [Cu4I4(PPh3)4] with a stair-step configu-
ration or a simple dimeric compound [Cu2I2(PPh3)3] has been
reported.12,23 These observations indicate that the emission
properties of these clusters are not controlled by the Cu–Cu
interactions in spite of the short Cu–Cu distances falling within
Fig. 5 Experimental solid-state absorption spectra of 1–4 (black lines)
and calculated transitions in vacuo at the S0 optimized geometries with bar
heights proportional to the oscillator strengths (vertical red lines).
3
the range found for CC transitions.
The electron density difference maps presented in Fig. 6
provide insights into the origin of the excited states responsible
( f 4 0.1) correspond well to the three absorption regions for LE phosphorescence. Intriguingly, the emission of complex
respectively (Table S13, ESI†). Generally, it is shown in vacuo 1 is dominated by pure 3ILCT (intraligand charge transfer)
that the low energy absorption (ca. 4370 nm) is dominated by transitions while the emission of complex 2 is dominated by
pure 1(X + M)LCT transitions with a low oscillator strength, the 3(X + M)LCT transitions. The luminescence of complexes 3 and
1ILCT (intraligand charge transfer) starts to manifest itself in 4 is assigned to the emission from the mixed 3(X + M)LCT/3ILCT
the region of ca. 250–370 nm showing a high oscillator strength excited states with different proportions. In complex 3, the
with the 1(X + M)LCT character still being the majority, and the 3ILCT transitions are the major contribution while in complex 4
cluster-centered transition (1CC) is only observed at the high the 3(X + M)LCT transitions are the major contribution. The
energy absorption zone (ca. o250 nm). Another interesting considerable influence of the substituents on the origin of the
finding is that transitions with a high oscillator strength electronic transitions involved in the LE emission arise from
bearing the prominent ILCT character are not found in the the zwitterionic character of the ligands. The pyridinium part of
spin-allowed singlet–singlet transitions, however, observed in the ligand is highly electron-deficient, which could be affected
the first two lowest spin-restricted singlet–triplet transitions at by the characteristics of the substituent on the pyrazolyl moiety.
ca. 525 nm for 1, at 470 nm for 2, at 566 nm for 3, and at 520 nm For instance, the electron-withdrawing CF3 group in 2 presum-
for 4. These spin-restricted transitions may be the origin of the ably inhibits the charge transfer from the pyrazolyl part to the
very weak tails displayed in the solid-state absorption spectra. pyridinium part, resulting in the absence of 3ILCT in 2 which is
2.4.4 Investigation into the emission properties in the observed for 1 and 3–4 in the lowest triplet excited states.
solid state. As far as the luminescence behavior of the title No 3CC transition in the lowest triplet excited state which is
Fig. 6 The density difference plots (isovalue 0.0005) for the lowest triplet excitation of complex 1–4 in vacuo, calculated at the UDFT optimized
T1 geometry. Blue and purple represent zones of depletion and augmentation of electron density in the T1 excited state versus the S0 ground state.
The calculated T1 - S0 emission wavelengths together with the experimental ones at 300 K are also listed for the sake of comparison.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
NJC
typical for the classic cubane-like copper clusters and some 3. Conclusions
previously reported octahedral clusters is found in our calcula-
A rare family of tetranuclear copper(I) iodide clusters with a
rare octahedral configuration supported by charge-separated
zwitterionic ligands have been synthesized and fully characterized.
The photophysical properties of these complexes in the solid state
at varying temperatures have been systematically investigated.
The results of the DFT/TDDFT calculations have provided a
fundamental understanding and deep insights into the nature
and origin of the excited states involved in the absorption spectra
and those responsible for the single LE emission bands for the
clusters described here.
tions. The assignments of the emission for 2 and 3 calculated at
the UDFT optimized T1 geometry are consistent with those
calculated at the TDDFT optimized T1 geometry (see Fig. S29,
ESI†). The calculated emission energies at the UDFT optimized
T1 geometry are overestimated relative to the experimental
values at 300 K; however, at the TDDFT optimized T1 geometry
the energy is underestimated for 2 and is in fair agreement with
the experimental one for 3. Generally, the calculated T1 - S0
emission wavelengths are in rather good agreement with the
corresponding experimental ones, especially the trend observed
in the emission energies of 2–4 follows the experimental trend
(4 4 2 4 3).
Upon cooling in the solid state, the position of the single LE
emission band is considerably blue-shifted for clusters 2–5
in contrast to that of cubane or other octahedral clusters.
However, for complex 1 with bulky tert-butyl groups, the single
LE emission band hardly moves when lowering the temperature.
According to TDDFT calculations, the emission of complex 1
originates from 3ILCT and complex 2 is dominated by pure
3(X + M)LCT transitions. The luminescence of complexes 3 and
From these theoretical results, the emission origin of copper
iodide clusters is closely related to the properties of the ligands
as we supposed. Both experimental and theoretical studies of
the cubane or some octahedral clusters reveal that structural
relaxation is required to shorten the Cu–Cu distances to produce
the LE emission from the 3CC excited state.21,29 For cubane
clusters, concerns about how the rigidity/steric demands of
the ligands would influence the contraction of the Cu–Cu
separations in the excited state can almost be neglected because
monodentate ligands, such as the pyridine-derivatives and
phosphine-based molecules, are employed which show less
influence on the LE emission band. However, when it comes
to octahedral clusters, the rigidity/steric hindrance of the ligands
must be considered since bidentate ligands are generally
required to obtain the planar Cu4 arrangement. If the bidentate
ligands are too rigid or bulky, the contraction of the Cu–Cu
distances in the excited state may be severely restricted. Such an
assumption is consistent with the reported studies on the
octahedral clusters in which emission from the 3CC excited state
will be blocked by very rigid bidentate ligands or ligands with
very bulky substituents in spite of the short Cu–Cu bond
lengths.29 The crystallographic data that the Cu–Cu distances
of 2–5 at 293 K and 100 K are basically the same supported that
the rigid pyrazolate-type ligand employed in this work hinders the
shortening of Cu–Cu separations upon cooling. Therefore,
the studies described here confirm that the emission from
3
3
4 arises from the mixed (X + M)LCT/3ILCT excited states. CC
states are not involved in the luminescence. Although these
complexes contain short Cu–Cu contacts less than or close to
2.8 Å (the van der Waals radius of Cu is 1.4 Å), our results suggest
that the rigid pyrazolate-type ligands will block the 3CC emission
probably by restricting the structural relaxation in the excited
state. Furthermore, via modifying the characteristic of the sub-
stituents, the origin of the electronic transitions involved in the
emission can be tuned. In addition, a pair of cis–trans isomers
(3 and 4) displaying distinct luminescence behavior have been
reported for the first time in the family of octahedral copper
halide clusters.
To conclude, the unique photophysical properties of this
class of clusters lead to the development of new luminophores.
The detailed DFT and TDDFT calculations reported here
provide a significantly enhanced understanding of the role of
ligands in the luminescence of the copper cluster family.
4. Experimental
4.1 Materials
3
the CC excited state in octahedral copper iodide clusters will
Reagents and solvents employed were commercially available
and used as received. Spectroscopically pure solvents for photo-
physical measurements were additionally dried over molecular
sieves or sodium.
be blocked by the rigid bidentate pyrazolate-type though strong
cuprophilic interactions existed.
It is commonly known that the energy of emission con-
trolled by 3ILCT transitions is not liable to be affected by
varying the temperature. Therefore, the slight blue-shift of the
low-energy flank observed upon cooling for cluster 1 may be
explained by freezing energetically higher lying emissions from
4.2 Synthesis of the ligands
4-(5-(tert-Butyl)-1H-pyrazol-3-yl)-1-methylpyridin-1-ium iodide
an inhomogeneously broadened distribution in the powder (HL1ÁI). HL1ÁI was prepared using a literature method with a
sample.5 On the basis of the current experimental results or slight modification.48 Firstly, 3,3-dimethyl-2-butanone (11.2 mL,
computational data, we cannot elucidate the molecular details 90 mmol) was added to a suspension of potassium tert-butoxide
of the hypsochromic shift for 2–5. Considering the great (11.2 g, 100 mmol) in dry THF (100 mL). The mixture was stirred
difference of the emission energies of the two pseudo- at RT for about 30 min. Methyl isonicotinate (10 mL, 85 mmol)
polymorphs 4 and 5, one possible explanation is that the slight was slowly added to the mixture. After stirring for 12 h, dilute
distortion of the structures upon cooling results in such a acetic acid (10 mL acetic acid in 30 mL H2O) was added to the
considerable blue shift.
solution and further stirred for 10 min. The aqueous solution
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
was extracted with diethyl ether (3 Â 20 mL). The combined 1332 (w), 1265 (w), 1217 (m), 1186 (s), 1056 (w), 964 (s), 859 (w), 841
organic phase was washed with brine, and then the diethyl ether (w), 807 (m), 771 (vs), 724 (w), 691 (m), 663 (w), 527 (w), 507 (w), 490
solvent was removed under vacuum to yield the crude 1,3-dione. (w), 438 (w). 1H NMR (400 MHz, CD3CN) d 12.20 (s, 1H), 8.56 (d, J =
Secondly, this 1,3-dione was dissolved in ethanol (200 mL) and 6.7 Hz, 2H), 8.35 (d, J = 6.4 Hz, 2H), 7.77 (d, J = 7.6 Hz, 2H), 7.54
the resulting ethanol solution was treated with an excess of (t, J = 7.5 Hz, 2H), 7.47 (t, J = 7.3 Hz, 1H), 7.39 (s, 1H), 4.24 (s, 3H).
hydrazine (80%, 15 mL) and then refluxed for 12 h. The ethanol
4.3 Synthesis of the complexes
solvent was removed under reduced pressure and the crude
product was recrystallized from methanol and H2O (v/v 1 : 1).
Synthesis of [Cu4I4(L1)2]ÁCH3CN (1). First, HL1ÁI (17.15 mg,
A white solid was obtained (8.04 g, yield: 44%). Thirdly, the 0.05 mmol) in 1 mL CH3CN and CuI (28.5 mg, 0.15 mmol) in
product (2.01 g, 10 mmol) from the last step was dissolved 1 mL CH3CN were mixed together. Then, 3 mL of benzene was
in acetonitrile (200 mL) and treated with excess iodomethane added to the solution. The solution was allowed to stand at
(15 mL). The mixture was stirred for 24 hours at RT. Then the RT for about a week to afford yellow crystals. Yield: 60% (based
solvent was removed under reduced pressure and the crude on CuI). IR (KBr pellet, cmÀ1): 3457 (b), 3046 (w), 2955 (m),
product was recrystallized from ethanol and H2O (v/v 1 : 1). 1637 (vs), 1560 (w), 1540 (s), 1499 (w), 1476 (w), 1381 (w), 1333
A pale yellow solid was obtained (3.0 g, yield: 90%). IR (KBr (w), 1280 (w), 1249 (w), 1199 (m), 1181 (m), 1032 (w), 980 (w),
1
pellet, cmÀ1): 3480 (b), 3184 (s), 3184 (s), 3134 (m), 3101 (m), 848 (w), 831 (w), 789 (m), 524 (w), 410 (w). H NMR (400 MHz,
3033 (w), 2964 (s), 2867 (w), 1643 (vs), 1579 (w), 1538 (s), 1479 CD3CN) d 8.47 (d, J = 6.2 Hz, 8H), 8.29 (d, J = 3.9 Hz, 8H),
(m), 1401 (m), 1364 (w), 1332 (w), 1285 (w), 1249 (w), 1209 (s), 6.83 (s, 2H), 4.17 (s, 6H), 1.38 (s, 18H). Anal. calcd for
1137 (m), 1174 (w), 1000 (m), 963 (m), 853 (s), 811 (m), 710 (w),
684 (w), 647 (w), 525 (w), 511 (w), 442 (m). H NMR (400 MHz, + 1 (CH3CN): C, 27.27; H, 3.02; N, 7.95. Found: C, 26.57; H, 2.71;
C26H34Cu4I4N6: C, 26.19; H, 2.87; N, 7.05 and for C26H34Cu4I4N6
1
CD3CN) d 11.70 (s, 1H), 8.51 (d, J = 6.7 Hz, 2H), 8.26 (d, J = 6.7 Hz, N, 7.31. ESI-MS: the parent ion was not observed in either
2H), 6.85 (s, 1H), 4.22 (s, 3H), 1.37 (s, 9H).
positive or negative ion mode. In the positive mode, the
1-Methyl-4-(5-(trifluoromethyl)-1H-pyrazol-3-yl) pyridin-1-ium fragment corresponding to the ligand was found: m/z 216.15
iodide (HL2ÁI). The synthesis procedure of HL2ÁI is similar to that (HL1)+. In the negative mode, fragments corresponding to
of HL1ÁI. Firstly, 4-acetylpyridine (5.6 mL, 50 mmol) was added to various inorganic anions were found: m/z 1269.91 (Cu6I7)À,
a suspension of potassium tert-butoxide (6.72 g, 60 mmol) in dry m/z 1078.08 (Cu5I6)À, m/z 888.24 (Cu4I5)À, m/z 698.41 (Cu3I4)À,
THF (100 mL). The mixture was stirred at RT for about 30 min. m/z 506.58 (Cu2I3)À, m/z 380.72 (I3)À, m/z 316.74 (CuI2)À. MALDI-
Ethyl trifluoroacetate (6.5 mL, 55 mmol) was slowly added to the TOF-MS (trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylidene]-
mixture. After stirring for 24 h, dilute acetic acid (10 mL acetic malononitrile as the matrix, positive mode): m/z +874.9
acid in 30 mL H2O) was added to the solution and further stirred {[Cu4I4(L1)2] À CuI2}+.
for 10 min. The aqueous solution was extracted with diethyl
Synthesis of [Cu4I4(L2)2] (2). First, HL2ÁI (35.5 mg, 0.1 mmol)
ether (3 Â 20 mL). The combined organic phase was washed in 0.5 mL CH3CN and CuI (28.5 mg, 0.15 mmol) in 1 mL CH3CN
with brine; the diethyl ether solvent was removed under vacuum were mixed together. Then, 3 mL of 1,3-dimethoxybenzene was
to yield the crude 1,3-dione. Secondly, this 1,3-dione was dis- added to the solution. The solution was allowed to stand at RT
solved in ethanol (60 mL) and the resulting ethanol solution was for about a week to afford brown crystals. Yield: 65% (based on
treated with an excess of hydrazine (80%, 15 mL) and then CuI). IR (KBr pellet, cmÀ1): 3441 (b), 3127 (w), 3051 (w), 1637
refluxed for 12 h. After then, the ethanol solvent was removed in (vs), 1566 (w), 1542 (s), 1524 (m), 1477 (w), 1402 (w), 1349 (w),
a vacuum and the residue was re-dissolved in 50 mL of ethanol 1281 (s), 1255 (vs), 1220 (w), 1197 (s), 1152 (vs), 1125 (vs), 1008
along with 1 mL of conc. HCl. The solution was refluxed for an (s), 976 (w), 840 (w), 829 (m), 808 (s), 744 (w), 512 (w), 517 (w),
1
additional 5 h. The ethanol was evaporated again under vacuum 440 (w). H NMR (400 MHz, CD3CN) d 8.63 (d, J = 6.5 Hz, 4H),
and the resulting solid was recrystallized from CH2Cl2 and 8.24 (s, 7H), 7.48 (s, 2H), 4.25 (s, 3H). Anal. calcd for
n-hexane (v/v 1 : 1). A white solid was obtained (4.26 g, yield:
C20H16Cu4I4F6N6: C, 19.75; H, 1.33; N, 6.91. Found: C, 20.04;
40%). The next step is similar to that of HL1ÁI, and a brown solid H, 1.36; N, 6.94. ESI-MS: the parent ion was not observed in
was obtained (3.2 g, yield: 90%). IR (KBr pellet, cmÀ1): 3440 (b), either the positive or negative ion mode. In the positive mode,
3105 (m), 3052 (m), 2824 (b), 1639 (s), 1615 (s), 1587 (w), 1575 (w), the fragment corresponding to the ligand was found: m/z
1522 (m), 1490 (m), 1452 (w), 1417 (s), 1341 (w), 1284 (s), 1260 (vs), 228.08 (HL2)+. In the negative mode, fragments corresponding
1222 (s), 1159 (vs), 1126 (vs), 1063 (w), 1005 (m), 973 (vs), 897 (w), to various inorganic anions were found: m/z 1269.91 (Cu6I7)À,
820 (s), 746 (s), 717 (w), 694 (m), 667 (w), 618 (w), 560 (w), 517 (w), m/z 1078.08 (Cu5I6)À, m/z 888.25 (Cu4I5)À, m/z 698.41 (Cu3I4)À,
431 (w). 1H NMR (400 MHz, CD3CN) d 8.65 (d, J = 6.8 Hz, 2H), 8.32 m/z 506.58 (Cu2I3)À, m/z 380.72 (I3)À, m/z 316.74 (CuI2)À. MALDI-
(d, J = 6.7 Hz, 2H), 7.48 (s, 1H), 4.24 (s, 3H).
TOF-MS (trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylidene]
1-Methyl-4-(3-phenyl-1H-pyrazol-5-yl)pyridin-1-ium iodide malononitrile as the matrix, positive mode): m/z +898.7
(HL3ÁI). The synthesis process of HL3ÁI is almost the same as {[Cu4I4(L2)2] À CuI2}+, +706.9 {[Cu4I4(L2)2] À Cu2I3}+.
that of HL1ÁI except that 3,3-dimethyl-2-butanone is replaced by
Synthesis of [Cu4I4(L3)2]ÁCH3CN (3). The HL3ÁI (18.1 mg,
acetophenone in the first step. Finally, a white solid was 0.05 mmol) in mixed DMF/EtOH (v/v = 1 : 1, 2 mL) and CuI
obtained (3.26 g, yield: 90%). IR (KBr pellet, cmÀ1): 3456 (b), (19.0 mg, 0.1 mmol) in 1 mL CH3CN were mixed together.
3127 (b), 3041 (w), 1639 (vs), 1571 (w), 1539 (s), 1480 (s), 1395 (m), The solution was allowed to stand at RT for about a week to
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
NJC
afford brown crystals. Yield: 65% (based on CuI). IR (KBr pellet, mounted on the top of fiberglass with glue. Data collection for
cmÀ1): 3441 (b), 3113 (w), 3054 (w), 1672 (s), 1637 (vs), 1560 (w), 1–5 at 293 K and 2 at 100 K was performed using an Oxford
1542 (s), 1521 (w), 1478 (m), 1431 (w), 1385 (m), 1215 (w), 1196 Diffraction Gemini E (Enhance Mo X-ray source, Ka, l = 0.71073 Å)
(m), 1152 (w), 1094 (m), 990 (m), 840 (m), 798 (w), 769 (m), 723 equipped with a graphite monochromator and an ATLAS CCD
(w), 694 (w), 667 (w), 530 (w), 507 (w), 493 (w), 444 (w). 1H NMR detector (CrysAlis CCD, Oxford Diffraction Ltd). Data collection for
(400 MHz, CD3CN) d 8.54 (d, J = 5.3 Hz, 4H), 8.34 (d, J = 6.8 Hz, 3–5 at 100 K was performed using a Bruker APEX2 (Cu X-ray
4H), 7.92 (s, 1.5H), 7.77 (d, J = 7.6 Hz, 4H), 7.57–7.48 (m, 4.5H), source, Ka, l = 1.54178 Å). Structures were solved by direct
7.46 (d, J = 6.9 Hz, 3H), 7.37 (s, 2H), 4.19 (s, 3H). Anal. calcd for methods (SHELXTL-97) and refined on F2 using full-matrix
C
32H29Cu4I4N7: C, 30.18; H, 2.30; N, 7.70. Found: C, 30.53; H, least-squares (SHELXTL-97).49 Restraints on bond lengths for the
2.48; N, 7.35. ESI-MS: the parent ion was not observed in either disordered t-butyl groups were employed for cluster 1 to obtain
the positive or negative ion mode. In the positive mode, the reasonable thermal displacement parameters. All non-hydrogen
fragment corresponding to the ligand was found: m/z 236.12 atoms were refined with anisotropic thermal parameters, and all
(HL3)+. In the negative mode, fragments corresponding to hydrogen atoms were placed in geometrically calculated positions.
various inorganic anions were found: m/z 1269.93 (Cu6I7)À, m/z PLATON/SQUEEZE50 was used to correct the diffraction data for
1078.09 (Cu5I6)À, m/z 888.25 (Cu4I5)À, m/z 698.42 (Cu3I4)À, m/z the contribution from disordered acetonitrile solvent molecules in
506.58 (Cu2I3)À, m/z 380.72 (I3)À, m/z 316.74 (CuI2)À.
complex 3.
Synthesis of [Cu4I4(L3)2] (4). The HL3ÁI (18.1 mg, 0.05 mmol)
4.5 Characterization
in mixed DMF/EtOH (v/v = 1 : 1, 2 mL) and CuI (19.0 mg,
0.1 mmol) in 1 mL CH3CN were mixed together and loaded 1H NMR spectra were recorded on a Bruker Biospin Avance
into a 15 mL Teflon-lined reactor, and then heated to 120 1C in a 400 MHz spectrometer and were referenced to the residual
programmable oven for 72 h, followed by slow cooling (3 1C hÀ1) to solvent signal (CH3CN at 1.94 ppm). Elemental analyses (C, H,
RT. Yellow crystals were obtained. Yield: 55% (based on CuI). IR and N) were carried out using an Elementar Vario EL Cube
(KBr pellet, cmÀ1): 3441 (b), 3113 (w), 3054 (w), 1637 (vs), 1560 (w), instrument. Powder X-ray diffraction (PXRD) experiments were
1542 (s), 1521 (w), 1478 (m), 1431 (w), 1385 (m), 1215 (w), 1196 (m), performed using a D8 Advance X-ray diffractometer with CuKa
1152 (w), 1094 (m), 990 (m), 840 (m), 798 (w), 769 (m), 723 (w), 694 radiation. ESI-MS spectra were recorded on a Thermo Finnigan
(w), 667 (w), 530 (w), 507 (w), 493 (w), 444 (w). 1H NMR (400 MHz, LCQ DECA XP quadrupole ion trap mass spectrometer using an
CD3CN) d 8.55 (d, J = 6.7 Hz, 4H), 8.34 (d, J = 6.8 Hz, 4H), 7.76 (d, J = electrospray ionization source with acetonitrile as the mobile
7.4 Hz, 4H), 7.57–7.51 (m, 5H), 7.50–7.44 (m, 4H), 7.38 (s, 2H), 4.24 phase and were acquired in both positive and negative ion
(s, 3H). Anal. calcd for C30H26Cu4I4 N6: C, 29.24; H, 2.13; N, 6.82. mode. MALDI-TOF mass spectra were obtained on a 4800 Plus
Found: C, 29.10; H, 2.09; N, 6.75. ESI-MS: the parent ion was not MALDI TOF Analyzer (ABI) spectrometer using trans-2-[3-(4-
observed in either the positive or negative ion mode. In the positive tert-butylphenyl)-2-methyl-2-propenylidene]malononitrile (DCTB)
mode, the fragment corresponding to the ligand was found: m/z as the matrix. Solid UV-vis spectra were recorded on a Bio-Logic
236.12 (HL3)+. In the negative mode, fragments corresponding MOS-450/AF-CD spectrometer. Infrared spectra (KBr pellet) were
to various inorganic anions were found: m/z 1269.93 (Cu6I7)À, m/z recorded on a Nicolet Avatar 360 FTIR spectrometer in the range
1078.09 (Cu5I6)À, m/z 888.25 (Cu4I5)À, m/z 698.41 (Cu3I4)À, m/z of 4000–400 cmÀ1 (abbreviations: w = weak, m = medium,
506.58 (Cu2I3)À, m/z 316.74 (CuI2)À.
b = broad, s = strong, and vs = very strong).
Synthesis of [Cu4I4(L3)2]Á2C13H10O (5, C13H10O denotes benzo-
Steady-state luminescence spectra were recorded on a PTI
phenone). First, HL3ÁI (9.0 mg, 0.025 mmol) in 1 mL CH3CN and QuantaMaster Model QM-TM scanning spectrofluorometer.
CuI (9.5 mg, 0.05 mmol) in 1 mL of CH3CN were mixed together. The emission spectra were corrected for the detector response.
Then, 0.45 g of benzophenone in 3 mL of CH3CN was added to the Lifetime data were acquired using phosphorescence subsystem
solution. The solution was allowed to stand at RT for about a week add-ons to the PTI instrument. A pulsed excitation source of
to afford brown crystals. Yield: 45% (based on CuI). IR (KBr pellet, 337 nm was generated from a nitrogen laser. A liquid-nitrogen
cmÀ1): 3429 (b), 3113 (w), 3054 (w), 1656 (s), 1638 (vs), 1597 (w), cryostat (Janis Research Model VPF-100 System) was used to
1542 (s), 1521 (w), 1476 (w), 1445 (w), 1380 (w), 1316 (m), 1278 (s), control the sample temperature. Microcrystalline samples were
1216 (w), 1196 (m), 1152 (w), 1096 (w), 1074 (w), 991 (w), 942 (w), used for the photoluminescence measurements in the solid state.
918 (w), 841 (m), 813 (w), 798 (w), 766 (s), 72w (w), 704 (s), 661 (w),
1
4.6 Computational details
638 (m), 533 (w), 507 (w), 493 (w), 444 (w). H NMR (400 MHz,
CD3CN) d 8.55 (d, J = 6.7 Hz, 4H), 8.34 (d, J = 6.7 Hz, 4H), 7.77 Density functional theory (DFT) and time-dependent DFT
(d, J = 7.1 Hz, 25H), 7.65 (t, J = 7.4 Hz, 14H), 7.54 (t, J = 7.6 Hz, 26H), (TDDFT) have been carried out with the Gaussian09 software
7.49–7.44 (m, 4H), 7.38 (s, 1H), 4.21 (s, 4H). Anal. calcd for without symmetry constraints.51 The Lanl2dz52 effective core
C56H46Cu4I4N6O2: C, 42.12; H, 2.90; N, 5.26. Found: C, 42.26; potential (ECP) was applied for Cu and I while the 6-31G** basis
H, 3.20; N, 5.03.
set53 was used for C, N, O, F, and H atoms throughout. The
hybrid PBE1PBE (PBE0)54 functional was exerted for geometry
optimizations while oB97X-D was used for molecular orbitals
4.4 Crystal structure determination
Single crystals of clusters 1–5 suitable for X-ray analysis were (MOs) and vertical excitation energies calculations. Spin-
obtained as described in the synthesis section. Crystals were unrestricted calculations were also performed in the case of
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
12 Q. Benito, B. Baptiste, A. Polian, L. Delbes, L. Martinelli,
T. Gacoin, J. P. Boilot and S. Perruchas, Inorg. Chem., 2015,
54, 9821–9825.
13 Q. Benito, X. F. Le Goff, S. Maron, A. Fargues, A. Garcia,
C. Martineau, F. Taulelle, S. Kahlal, T. Gacoin, J. P. Boilot
and S. Perruchas, J. Am. Chem. Soc., 2014, 136, 11311–11320.
14 S. Perruchas, X. F. Le Goff, S. Maron, I. Maurin, F. Guillen,
A. Garcia, T. Gacoin and J.-P. Boilot, J. Am. Chem. Soc., 2010,
132, 10967–10969.
triplet states. Vibrational frequency calculations were only
carried out on the optimized ground state (S0) structures to
ensure they are minima on the potential energy surface. In the
calculation of UV-vis transitions, the 150 singlet–singlet or
singlet–triplet transitions of lowest energy were computed.
Note that we performed TDDFT calculations without including
spin–orbit coupling, so that singlet–triplet excitations have zero
oscillator strengths ( f ). The molecular orbital compositions
were obtained by the Hirshfeld method55 and using Multiwfn
packages.56 The electron density difference (EDD) maps were
also generated from Multiwfn packages. More detailed compu-
tational results including coordinates of geometry optimized
structures (Table S14) are given in the ESI.†
15 W. Liu, Y. Fang, G. Z. Wei, S. J. Teat, K. Xiong, Z. Hu, W. P.
Lustig and J. Li, J. Am. Chem. Soc., 2015, 137, 9400–9408.
16 D. Volz, D. M. Zink, T. Bocksrocker, J. Friedrichs, M. Nieger,
¨
T. Baumann, U. Lemmer and S. Brase, Chem. Mater., 2013,
25, 3414–3426.
17 D. M. Zink, D. Volz, T. Baumann, M. Mydlak, H. Flugge,
J. Friedrichs, M. Nieger and S. Brase, Chem. Mater., 2013, 25,
4471–4486.
Conflicts of interest
The authors declare no competing financial interests.
18 H. Yersin, A. F. Rausch, R. Czerwieniec, T. Hofbeck and
T. Fischer, Coord. Chem. Rev., 2011, 255, 2622–2652.
19 M. J. Leitl, F. R. Kuchle, H. A. Mayer, L. Wesemann and
H. Yersin, J. Phys. Chem. A, 2013, 117, 11823–11836.
20 M. J. Leitl, D. M. Zink, A. Schinabeck, T. Baumann, D. Volz
and H. Yersin, Top. Curr. Chem., 2016, 374, 25.
21 F. De Angelis, S. Fantacci, A. Sgamellotti, E. Cariati, R. Ugo
and P. C. Ford, Inorg. Chem., 2006, 45, 10576–10584.
22 A. Vega and J.-Y. Saillard, Inorg. Chem., 2004, 43, 4012–4018.
23 L. Maini, D. Braga, P. P. Mazzeo and B. Ventura, Dalton
Trans., 2012, 41, 531–539.
Acknowledgements
We gratefully acknowledge financial support from the NSFC
(No. 21571122), the National Basic Research Program of China
(973 Program, 2013CB834803), and the Department of Educa-
tion in Guangdong Province (No. 2014KCXTD012). We thank Dr
Ji Zheng for his kind help with the theoretical calculations and
Prof. Ming-Hua Zeng for the measurements of ESI-MS.
24 G. Nardin and L. Randaccio, Acta Crystallogr., 1974, 30,
1377–1379.
References
25 W. F. Fu, X. Gan, C. M. Che, Q. Y. Cao, Z. Y. Zhou and
N. N. Zhu, Chem. – Eur. J., 2004, 10, 2228–2236.
26 E. I. Musina, A. V. Shamsieva, I. D. Strelnik, T. P.
Gerasimova, D. B. Krivolapov, I. E. Kolesnikov, E. V.
Grachova, S. P. Tunik, C. Bannwarth, S. Grimme, S. A.
Katsyuba, A. A. Karasik and O. G. Sinyashin, Dalton Trans.,
2016, 45, 2250–2260.
27 L. Maini, P. P. Mazzeo, F. Farinella, V. Fattori and D. Braga,
Faraday Discuss., 2014, 170, 93–107.
28 K. Chen, J. Shearer and V. J. Catalano, Inorg. Chem., 2015,
54, 6245–6256.
1 P. C. Ford, E. Cariati and J. Bourassa, Chem. Rev., 1999, 99,
3625–3648.
2 M. Wallesch, D. Volz, D. M. Zink, U. Schepers, M. Nieger,
T. Baumann and S. Brase, Chem. – Eur. J., 2014, 20,
6578–6590.
3 R. Peng, M. Li and D. Li, Coord. Chem. Rev., 2010, 254, 1–18.
4 T. Hofbeck, U. Monkowius and H. Yersin, J. Am. Chem. Soc.,
2015, 137, 399–404.
5 M. J. Leitl, V. A. Krylova, P. I. Djurovich, M. E. Thompson
and H. Yersin, J. Am. Chem. Soc., 2014, 136, 16032–16038.
6 T. H. Kim, Y. W. Shin, J. H. Jung, J. S. Kim and J. Kim,
Angew. Chem., Int. Ed., 2008, 47, 685–688.
29 Z. Liu, P. I. Djurovich, M. T. Whited and M. E. Thompson,
Inorg. Chem., 2012, 51, 230–236.
30 S. Naik, J. T. Mague and M. S. Balakrishna, Inorg. Chem.,
2014, 53, 3864–3873.
7 H. Kitagawa, Y. Ozawa and K. Toriumi, Chem. Commun.,
2010, 46, 6302–6304.
8 Q. Benito, X. F. Goff, G. Nocton, A. Fargues, A. Garcia,
A. Berhault, S. Kahlal, J. Y. Saillard, C. Martineau, 31 E. W. Emerson, M. F. Cain, M. D. Sanderson, C. B. Knarr,
J. Trebosc, T. Gacoin, J. P. Boilot and S. Perruchas, Inorg.
Chem., 2015, 54, 4483–4494.
D. S. Glueck, J. C. Ahern, H. E. Patterson and A. L.
Rheingold, Inorg. Chim. Acta, 2015, 427, 168–172.
9 S. Perruchas, C. Tard, X. F. Le Goff, A. Fargues, A. Garcia, 32 R. Ahuja, M. Nethaji and A. G. Samuelson, Inorg. Chim. Acta,
S. Kahlal, J. Y. Saillard, T. Gacoin and J. P. Boilot, Inorg.
Chem., 2011, 50, 10682–10692.
10 Q. Benito, I. Maurin, T. Cheisson, G. Nocton, A. Fargues,
2011, 372, 220–226.
33 A. Camus, G. Nardin and L. Randaccio, Inorg. Chim. Acta,
1975, 12, 23–32.
´
A. Garcia, C. Martineau, T. Gacoin, J. P. Boilot and 34 N. Mezailles, P. Le Floch, K. Waschbu¨sch, L. Ricard, F. Mathey
S. Perruchas, Chem. – Eur. J., 2015, 21, 5892–5897. and C. P. Kubiak, J. Organomet. Chem., 1997, 541, 277–283.
11 X.-C. Shan, H.-B. Zhang, L. Chen, M.-Y. Wu, F.-L. Jiang and 35 N. Marsich, G. Nardin and L. Randaccio, J. Am. Chem. Soc.,
M.-C. Hong, Cryst. Growth Des., 2013, 13, 1377–1381. 1973, 95, 4053–4054.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
NJC
36 S. Ramaprabhu, N. Amstutz, E. A. C. Lucken and G. r. 51 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
Bernardinelli, J. Chem. Soc., Dalton Trans., 1993, 871.
37 S. Daly, M. F. Haddow, A. G. Orpen, G. T. A. Rolls, D. F. Wass
and R. L. Wingad, Organometallics, 2008, 27, 3196–3202.
38 E. Ku¨hnel, I. V. Shishkov, F. Rominger, T. Oeser and
P. Hofmann, Organometallics, 2012, 31, 8000–8011.
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin,
V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand,
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox,
Gaussian 09, Revision A.02, Gaussian, Inc., Wallingford CT,
2009.
39 D. M. Zink, T. Grab, T. Baumann, M. Nieger, E. C. Barnes,
¨
W. Klopper and S. Brase, Organometallics, 2011, 30, 3275–3283.
¨
40 M. Scherer, D. Stein, F. Breher, J. Geier, H. Schonberg and
H. Gru¨tzmacher, Z. Anorg. Allg. Chem., 2005, 631, 2770–2774.
41 J. H. Wang, M. Li, J. Zheng, X. C. Huang and D. Li, Chem.
Commun., 2014, 50, 9115–9118.
42 S. Z. Zhan, M. Li, X. P. Zhou, J. H. Wang, J. R. Yang and
D. Li, Chem. Commun., 2011, 47, 12441–12443.
43 Q. Xiao, J. Zheng, M. Li, S. Z. Zhan, J. H. Wang and D. Li,
Inorg. Chem., 2014, 53, 11604–11615.
44 H. V. Dias, H. V. Diyabalanage, M. G. Eldabaja,
O. Elbjeirami, M. A. Rawashdeh-Omary and M. A. Omary,
J. Am. Chem. Soc., 2005, 127, 7489–7501.
45 H. V. Dias, H. V. Diyabalanage, M. A. Rawashdeh-Omary, 52 (a) P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82,
M. A. Franzman and M. A. Omary, J. Am. Chem. Soc., 2003,
125, 12072–12073.
46 J. S. Moore and S. I. Stupp, Macromolecules, 1986, 19,
1815–1824.
270–283; (b) W. R. Wadt and P. J. Hay, J. Chem. Phys., 1985,
82, 284–298; (c) P. J. Hay and W. R. Wadt, J. Chem. Phys.,
1985, 82, 299–310.
53 P. C. Hariharan and J. A. Pople, Mol. Phys., 1974, 27, 209–214.
47 J. Nitsch, F. Lacemon, A. Lorbach, A. Eichhorn, F. Cisnetti 54 (a) J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett.,
and A. Steffen, Chem. Commun., 2016, 52, 2932–2935.
48 C.-C. Cheng, W.-S. Yu, P.-T. Chou, S.-M. Peng, G.-H. Lee,
1996, 77, 3865–3868; (b) J. P. Perdew, K. Burke and
M. Ernzerhof, Phys. Rev. Lett., 1997, 78, 1396.
P.-C. Wu, Y.-H. Song and Y. Chi, Chem. Commun., 2003, 2628. 55 F. L. Hirshfeld, Theor. Chim. Acta, 1977, 44, 129–138.
¨
¨
49 G. M. Sheldrick, SHELX-97, University of Gottingen, Gottingen, 56 (a) T. Lu, Multiwfn, Revison 3.3.3, University of Science and
Germany, 1997.
Technology Beijing, Beijing, People’s Republic of China,
2014; (b) T. Lu and F. W. Chen, J. Comput. Chem., 2012, 33,
580–592.
50 A. L. Spek, Acta Crystallogr., Sect. A: Found. Crystallogr., 1990,
46, C34.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018