Structural, Electronic, and Magnetic Characterization of a Dinuclear Zinc Complex Containing TCNQ- and a μ-[TCNQ-TCNQ]2- Ligand

Article abstract of DOI:10.1021/acs.inorgchem.5b00808

A dinuclear zinc complex containing both a σ-dimerized 7,7,8,8-tetracyanoquinodimethane (TCNQ) ligand ([TCNQ-TCNQ]^2-) and TCNQ^- was synthesized for the first time. This is the first instance of a single molecular complex with a bridging [TCNQ-TCNQ]^2- ligand. Each zinc center is coordinated with two 2,2′-bipyrimidines and one TCNQ^-, and the remaining coordination site is occupied by a [TCNQ-TCNQ]^2- ligand, which bridges the two zinc centers. The complex facilitates π-stacking of TCNQ^- ligands when crystallized, which gives rise to a near-IR charge-transfer transition and strong antiferromagnetic coupling.

Full text of DOI:10.1021/acs.inorgchem.5b00808

Communication
pubs.acs.org/IC
Structural, Electronic, and Magnetic Characterization of a Dinuclear
Zinc Complex Containing TCNQ− and a μ‑[TCNQ−TCNQ]2− Ligand
Juyeong Kim, Alexey Silakov, Hemant P. Yennawar, and Benjamin J. Lear*
Department of Chemistry, The Pennsylvania State University, University Park, Pennsylvania 16802, United States
S
* Supporting Information
properties of [Zn(bpym)2(TCNQ)]2(TCNQ−TCNQ)
ABSTRACT: A dinuclear zinc complex containing both a
(ZnDimer; bpym = 2,2′-bipyrimidine). This is the first example
of a single molecule in which [TCNQ−TCNQ]2− is coordinated
σ-dimerized 7,7,8,8-tetracyanoquinodimethane (TCNQ)
ligand ([TCNQ−TCNQ]2−) and TCNQ− was synthe-
to a metal center. We find that its electronic and magnetic
sized for the first time. This is the first instance of a single
molecular complex with a bridging [TCNQ−TCNQ]2−
ligand. Each zinc center is coordinated with two 2,2′-
bipyrimidines and one TCNQ−, and the remaining
coordination site is occupied by a [TCNQ−TCNQ]2−
ligand, which bridges the two zinc centers. The complex
facilitates π-stacking of TCNQ− ligands when crystallized,
which gives rise to a near-IR charge-transfer transition and
strong antiferromagnetic coupling.
properties are dominated by the nondimerized TCNQ ligands.
ZnDimer is made via the slow diffusion of a methanol solution
of Li(TCNQ) and bpym into a methanol solution of Zn(NO3)2·
6H2O, yielding leaflike purple crystalline plates after 1 day [see
the Supporting Information (SI) for full synthetic details].
Single-crystal X-ray diffraction revealed that crystals of ZnDimer
belong to the monoclinic space group P2(1)/n (Table S1 in the
SI). From the structure, we find that ZnDimer is composed of
two zinc(II) centers, each octahedrally coordinated by two bpym
and one TCNQ− ligand (Figure 1a). The two metal centers are
nterest in the discovery of new functional materials continues
Ito motivate work on the synthesis and characterization of
molecules that could form the basis of materials with a diversity
of electronic and magnetic properties. One common approach to
creating new materials is through tuning noncovalent
interactions between these molecules, as demonstrated in the
rich literature reporting the properties of π-stacks involving
electron-poor 7,7,8,8-tetracyanoquinodimethane (TCNQ) mol-
ecules with electron-rich donors.1 The electronic and magnetic
properties of these simple π-stacked systems are sensitive to the
identity of the electron-rich molecules. However, additional
control over the properties of the TCNQ compounds (e.g.,
electronic conductivity) can be gained through the coordination
of TCNQ to metal centers.2 In addition, the interaction between
the unpaired electron of TCNQ− and a paramagnetic metal
center has shown peculiar magnetism, which results from σ
bonding between the ligand and metal center, π-stacking of
TCNQ, or both of these factors.3
The interest in TCNQ molecules has led, in turn, to the study
of the σ-dimerized form, [TCNQ−TCNQ]2−, which has been
isolated in the solid state as the salt of various cations.4 However,
transition-metal complexes with [TCNQ−TCNQ]2− ligands are
rare compared to the abundance of those incorporating TCNQ−.
Except for a few three-dimensional coordination polymers
reported in the literature,5 we know of only three molecular
[TCNQ−TCNQ]2− transition-metal compounds, and in all
three cases, the metal center exists as a cation stabilizing a
[TCNQ−TCNQ]2− anion rather than forming a coordination
complex.4a,d,f While the properties of the TCNQ σ dimer were
found to be dependent upon the identity of the cation, we
reasoned that coordination of [TCNQ−TCNQ]2− to a metal site
might provide additional control over the properties of the
[TCNQ−TCNQ]2− ligands. Herein we report the synthesis and
Figure 1. (a) Crystal structure of ZnDimer in the bc plane, shown with
anisotropic displacement parameters at 50% probability and hydrogen
atoms not shown. (b) Molecular structures and selected bond lengths of
TCNQ− and [TCNQ−TCNQ]2− in ZnDimer.
then bridged by a [TCNQ−TCNQ]2− ligand. At each zinc
center, the two bpym ligands are cis-disposed to each other,
leaving the TCNQ− and [TCNQ−TCNQ]2− ligands cis-
disposed to each other, and both metal sites exist as the Λ isomer.
The bond lengths from the X-ray crystallographic data also
form the basis for assigning of a charge of 1− to the TCNQ
ligands and 2− for the [TCNQ−TCNQ]2− σ dimer (Figure 1b).
The charge on each TCNQ can be calculated using the
Kistenmacher relationship ρ = A[c/(b + d)] + B, where b is
the C−C bond in the cyano group, c is the adjacent CC bond,
Received: April 15, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.5b00808
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Communication
and d is the C−C bond in the ring.6 Because of coordination of
the TCNQ ligands to the metal centers, the TCNQ and
[TCNQ−TCNQ]2− ligands are inherently asymmetric. Thus,
the bond lengths at the equivalent positions are averaged
which yields lengths similar to those reported for other materials
involving TCNQ− or [TCNQ−TCNQ]2− ligands (Table S2 in
the SI)and used in the equation. With these caveats in mind,
the calculated charges for the TCNQ ligands in ZnDimer are
1.27− and 2.14− for the TCNQ− and [TCNQ−TCNQ]2−
ligands, respectively. These values are close to the anticipated
charges of 1− and 2− for the TCNQ− and [TCNQ−TCNQ]2−
ligands, respectively.
The impact of the ZnDimer packing upon the electronic
properties of the solid are reflected in the UV−vis−near-IR
(NIR) diffuse-reflectance spectrum, which provides insights into
intermolecular charge transfer (CT) between TCNQ ligands
(Figure 3). In order to aid in the interpretation of this spectrum,
The above assignment of the ligand charges is further
supported by IR spectra (Figure S1 and Table S3 in the SI).
From the literature, the δ(C−H) bend mode of a [TCNQ−
TCNQ]2− ligand is shown at 804 cm−1, whereas the TCNQ−
ligand retains its higher energy for the δ(C−H) bend mode at
827 cm−1.5b In ZnDimer, we observe two peaks at 804 and 824
cm−1, which confirms the presence of both the TCNQ− and
[TCNQ−TCNQ]2− ligands. In addition, ZnDimer shows
multiple peaks of its ν(CN) stretch modes that are shifted
to lower energies than those found for free TCNQ−. We ascribe
this bathochromatic shift to back-donation of π electrons from
the zinc center to the TCNQ ligands, supporting the conclusions
drawn from the X-ray structure that the TCNQ− and [TCNQ−
TCNQ]2− ligands carry charges slightly greater than 1− and 2−,
respectively.
While the structure of the individual ZnDimer is informative,
the electronic and magnetic properties of the solid are expected
to depend upon the spatial arrangement of the individual
molecules with respect to one another. On the basis of our crystal
structure, the extended π-stacking of ZnDimer runs along the ab
plane, where two TCNQ− ligands are located between two
[TCNQ−TCNQ]2− ligands (Figure 2). The π-stacking
Figure 3. UV−vis−NIR diffuse-reflectance spectra of ZnDimer (solid
line) and Li(TCNQ) (dotted line) in the solid state.
we also synthesized and acquired the spectrum of Li(TCNQ), an
electrically conductive solid that contains uninterrupted stacks of
TCNQ− anions. The two high-energy bands (α and β) are
associated with transitions present in isolated TCNQ radicals.
The lowest-energy electronic transition of Li(TCNQ) is at 8197
cm−1 and is assigned as an intermolecular CT between face-to-
face-stacked TCNQ radicals (Table S4 in the SI).7 The lowest-
energy band for ZnDimer appears at 9141 cm−1, and by
comparison to the Li(TCNQ) spectrum, we assign this as a CT
transition that promotes an electron between the π-stacked
TCNQ ligands shown in Figure 2. Thus, the CT band of
ZnDimer is shifted ∼1000 cm−1 higher than that of Li(TCNQ),
while the electronic transition energies of the α and β bands in
ZnDimer do not show any significant changes in comparison
with the α and β bands in Li(TCNQ).
Two factors might conspire to produce the blue shift of the CT
band in ZnDimer. First, the staircase-shaped stacking of the
TCNQ ligands in ZnDimer will prohibit the long-range coupling
of adjacent TCNQ‑ ligands (which is observed in the continuous
linear array present in TCNQ salts). Second, the intermolecular
spacing between the TCNQ− and [TCNQ−TCNQ]2− ligands
in ZnDimer (3.72 Å) is longer than that found in TCNQ salts
(3.2−3.5 Å).7b,c This relatively long spacing would prevent
electronic coupling between adjacent TCNQ− and [TCNQ−
TCNQ]2− ligands. The loss of continuous coupling should
localize the electronic ground and excited states, which will
(according to the particle-in-a-box model) result in larger
spacings between the energy levels and a blue shift in the
transition between them.
Both the packing diagram and electronic spectroscopy of
ZnDimer indicate that little electronic coupling is expected to
persist along any one dimension of the solid. This poor electronic
coupling should, in turn, result in poor electrical conductivity.
Indeed, electrical resistance measurements on a single crystal
yielded an estimated specific resistivity on the order of 1012 Ω·cm
(Table S5 in the SI). This value is 7 orders of magnitude greater
than that for Li(TCNQ),8 and confirms ZnDimer as an
insulating solid.
With the electronic properties of ZnDimer thus examined, we
turned to the magnetic properties by using electron para-
magnetic resonance (EPR) on powders. The EPR spectrum of
ZnDimer shows a single isotropic signal with a g value of 2.0059
Figure 2. Extended packing diagram of the TCNQ π-stacks in the ab
plane of ZnDimer. The solid lines indicate the TCNQ− (black) and
[TCNQ−TCNQ]2− (red) ligands, and the inset gives distances
between their stacked layers.
distances are 3.72 Å for the layer between the [TCNQ−
TCNQ]2− and TCNQ− ligands and 3.15 Å for that between two
adjacent TCNQ− ligands. All four ligands (TCNQ− and
[TCNQ−TCNQ]2−) in one stacking unit are coordinated to
different zinc centers. The π-stacking does not form a continuous
chain structure, but it exhibits a staircase-shaped motif.
B
DOI: 10.1021/acs.inorgchem.5b00808
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Communication
(Figure 4a), which is typical for an organic radical. To further
elucidate this paramagnetic property, the spin state was
ACKNOWLEDGMENTS
■
We thank Prof. Hitoshi Miyasaka for help in interpreting the IR
spectra. We thank The Pennsylvania State University and the
NSF (Grant CHE-0131112 for the diffractometer purchase) for
financial support of this work.
REFERENCES
■
(1) (a) Anderson, P. W.; Lee, P. A.; Saitoh, M. Solid State Commun.
1973, 13, 595−598. (b) Cohen, M. J.; Coleman, L. B.; Garito, A. F.;
Heeger, A. J. Phys. Rev. B 1974, 10, 1298−1307. (c) Torrance, J. B. Acc.
Chem. Res. 1979, 12, 79−86. (d) Herbstein, F. H.; Kapon, M. Cryst. Rev.
2008, 14, 3−74.
Figure 4. (a) EPR spectrum of ZnDimer (solid line) and TEMPO
(dotted line) at room temperature and (b) ZnDimer and coal (S = 1/2)
Rabi oscillation at 100 K.
(2) (a) Kaim, W.; Moscherosch, M. Coord. Chem. Rev. 1994, 129, 157−
́ ́
193. (b) Azcondo, M. T.; Ballester, L.; Calderon, L.; Gutierrez, A.;
́
Perpinan, M. F. Polyhedron 1995, 14, 2339−2347. (c) Ballester, L.; Gil,
̃
A. M.; Gutier
Amador, U.; Campo, J.; Palacio, F. Inorg. Chem. 1997, 36, 5291−5298.
(d) Ballester, L.; Gutierrez, A.; Perpinan, M. F.; Azcondo, M. T. Coord.
́
rez, A.; Perpinan, M. F.; Azcondo, M. T.; Sanchez, A. E.;
́
́
̃
́
́
̃
investigated by Rabi oscillation measurements.9 The period of
Chem. Rev. 1999, 190, 447−470. (e) Avendano, C.; Zhang, Z.; Ota, A.;
Rabi oscillation demonstrates that the spin state in ZnDimer
Zhao, H.; Dunbar, K. R. Angew. Chem. 2011, 123, 6673−6677.
1
corresponds to S = /2 spin quantum number (Figure 4b). In
́
(f) Ballesteros-Rivas, M.; Ota, A.; Reinheimer, E.; Prosvirin, A.; Valdes-
Martinez, J.; Dunbar, K. R. Angew. Chem. 2011, 132, 9877−9881.
addition, we quantified the number of unpaired spins in
ZnDimer by comparing the EPR spectrum of (2,2,6,6-
tetramethylpiperidin-1-yl)oxyl (TEMPO) as a radical standard.
Both of the solid materials show isotropic spectra (Figure 4a).
The double integral values allowed us to determine that
ZnDimer contains 0.096 spins per molecule. If each TCNQ−
ligand was the source of this spin, we would expect 2 spins per
molecule. Thus, we conclude that the paramagnetic signal arises
from low-level impurities. We consider it most likely that these
impurities are lattice defects (including the surface of the solids),
where the pairing of TCNQ− ligand is disrupted. Furthermore,
the low number of spins present for ZnDimer supports the
conclusion that (in cases where the pairing between TCNQ−
units is preserved) the TCNQ− species are strongly
antiferromagnetically coupled, resulting in an EPR-silent species.
In total, we find that the electronic and magnetic properties of
ZnDimer are dominated by the properties of the individual
components, especially those of the TCNQ− and [TCNQ−
TCNQ]2− ligands. Specifically, the stacking in the crystal is
arranged to maximize the coupling between TCNQ− molecules
held between [TCNQ−TCNQ]2− ligands. This pairwise
TCNQ− interaction is testified by the CT transition in the
UV−vis−NIR spectrum and the dominance of a ground-state
singlet at room temperature. Nevertheless, the coordination of a
[TCNQ−TCNQ]2− ligand to the zinc center does perturb its
structure, as seen in the IR spectrum. Thus, it is possible that
future changes in the metal will further adjust the electronic and
magnetic properties of the [TCNQ−TCNQ]2− ligand.
(3) (a) Bartley, S. L.; Dunbar, K. R. Angew. Chem., Int. Ed. Engl. 1991,
30, 448−450. (b) Oshio, H.; Ino, E.; Mogi, I.; Ito, T. Inorg. Chem. 1993,
32, 5697−5703. (c) Kunkeler, P. J.; Koningsbruggen, P. J.; Cornelissen,
J. P.; Horst, A. N.; Kraan, A. M.; Spek, A. L.; Haasnoot, J. G.; Reedijk, J. J.
Am. Chem. Soc. 1996, 118, 2190−2197. (d) Miyasaka, H.; Campos-
Fernan
́
́
dez; Clerac, R.; Dunbar, K. R. Angew. Chem. 2000, 112, 3989−
3993. (e) Cler
́
ac, R.; O’Kane, S.; Cowen, J.; Ouyang, X.; Heintz, R.;
Zhao, H.; Bazile, M. J.; Dunbar, K. R. Chem. Mater. 2003, 15, 1840−
1850. (f) Miyasaka, H.; Izawa, T.; Takahashi, N.; Yamashita, M.;
Dunbar, K. R. J. Am. Chem. Soc. 2006, 128, 11358−11359.
(4) (a) Dong, V.; Endres, H.; Keller, H. J.; Moroni, W.; Nothe, D. Acta
̈
Crystallogr., Sect. B 1977, 33, 2428−2431. (b) Morosin, B.; Plastas, H. J.;
Coleman, L. B.; Stewart, J. M. Acta Crystallogr., Sect. B 1978, 34, 540−
543. (c) Harms, R. H.; Keller, H. J.; Nothe, D.; Werner, M.; Gundel, D.;
̈
Sixl, H.; Soos, Z. G.; Metzger, R. M. Mol. Cryst. Liq. Cryst. 1981, 65,
179−196. (d) Hoffmann, S. K.; Corvan, P. J.; Singh, P.; Sethulekshmi, C.
N.; Metzger, R. M.; Hatfield, W. E. J. Am. Chem. Soc. 1983, 105, 4608−
4617. (e) Radhakrishnan, T. P.; Van Engen, D.; Soos, Z. G. Mol. Cryst.
Liq. Cryst. 1987, 150, 473−492. (f) Alonso, C.; Ballester, L.; Gutier
́
rez,
A.; Perpinan, M. F.; Sanchez, A. E.; Azcondo, M. T. Eur. J. Inorg. Chem.
2005, 486−495.
́
́
̃
(5) (a) Zhao, H.; Heintz, R. A.; Dunbar, K. R. J. Am. Chem. Soc. 1996,
118, 12844−12845. (b) Zhao, H.; Heintz, R. A.; Ouyang, X.; Dunbar, K.
R. Chem. Mater. 1999, 11, 736−746. (c) Shimomura, S.; Horike, S.;
Matsuda, R.; Kitagawa, S. J. Am. Chem. Soc. 2007, 129, 10990−10991.
(d) Shimomura, S.; Matsuda, R.; Kitagawa, S. Chem. Mater. 2010, 22,
4129−4131.
(6) Kistenmacher, T. J.; Emge, T. J.; Bloch, A. N.; Cowan, D. O. Acta
Crystallogr., Sect. B 1982, 38, 1193−1199.
(7) (a) Boyd, R. H.; Phillips, W. D. J. Chem. Phys. 1965, 43, 2927−
2929. (b) Iida, Y. Bull. Chem. Soc. Jpn. 1969, 42, 71−75. (c) Hoekstra, A.;
Spoelder, T.; Vos, A. Acta Crystallogr., Sect. B 1972, 28, 14−25.
(8) (a) Kepler, R. G.; Bierstedt, P. E.; Merrifield, R. E. Phys. Rev. Lett.
1960, 5, 503−504. (b) Melby, L. R.; Harder, R. J.; Hertler, W. R.;
Mahler, W.; Benson, R. E.; Mochel, W. E. J. Am. Chem. Soc. 1962, 84,
3374−3387.
ASSOCIATED CONTENT
■
S
* Supporting Information
Crystallographic data in CIF format, materials and methods,
crystal data, IR spectra, supplementary tables, and specific
resistivity. The Supporting Information is available free of charge
on the ACS Publications website at DOI: 10.1021/acs.inorg-
chem.5b00808.
(9) (a) Isoya, J.; Kanda, H.; Norris, J. R.; Tang, J.; Bowman, M. K. Phys.
Rev. B: Condens. Matter Mater. Phys. 1990, 41, 3905−3913. (b) Moro, F.;
Kaminski, D.; Tuna, F.; Whitehead, G. F. S.; Timco, G. A.; Collison, D.;
Winpenny, R. E. P.; Ardavan, A.; Mclnnes, E. J. L. Chem. Commun. 2014,
50, 91−93.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: bul14@psu.edu.
Notes
The authors declare no competing financial interest.
C
DOI: 10.1021/acs.inorgchem.5b00808
Inorg. Chem. XXXX, XXX, XXX−XXX