Insulated copper(i) "wires": Structural variations and photoluminescence

Article abstract of DOI:10.1039/c1cc11000d

An unprecedented "molecular wire" type of structure for the copper(i) carboxylate family has been synthesized by utilizing copper-copper interactions and controllable switch of copper-oxygen interactions. Several modifications of the same complex, copper(

Full text of DOI:10.1039/c1cc11000d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 6939–6941
www.rsc.org/chemcomm
COMMUNICATION
Insulated copper(I) ‘‘wires’’: structural variations and photoluminescencew
Oleksandr Hietsoi, Cristina Dubceac, Alexander S. Filatov and Marina A. Petrukhina*
Received 20th February 2011, Accepted 20th April 2011
DOI: 10.1039/c1cc11000d
An unprecedented ‘‘molecular wire’’ type of structure for the
copper(I) carboxylate family has been synthesized by utilizing
copper–copper interactions and controllable switch of
copper–oxygen interactions. Several modifications of the same
complex, copper(I) 2,4,6-triisopropylbenzoate, have been
isolated to allow evaluation of the structural variation effects
on photoluminescent behavior.
dinuclear paddlewheel units.11 Thus, it was selected in this
work to bridge the copper(I) centers and also to favor Cuꢀ ꢀ ꢀCu
instead of Cuꢀ ꢀ ꢀO interactions.
The new copper(I) complex with 2,4,6-triisopropylbenzoate
bridges has been synthesized by refluxing copper(I) acetate
with an excess of the carboxylic acid in m-xylene. The elemental
analysis of the product confirmed a [Cu] : (O2C16H23) = 1 : 1
stoichiometry, suggesting a composition of [Cu(O2C16H23)]
(1).z Sublimation of 1 at 186 1C under reduced pressure
yielded very thin colorless needles. Their X-ray diffraction
study revealed an essentially linear ‘‘copper wire’’ structure
comprised of copper(I) ions bridged by carboxylate groups
and further supported by cuprophilic interactions,
[Cu(O2C16H23)]N,linear (2) (Fig. 1). The Cuꢀ ꢀ ꢀCu distances of
2.9397(5) A are comparable to those in helical copper(I)
structures (ESIw).6,7 The Cu–Cu–Cu angles are 159.43(5)1.
An increase of sublimation temperature to 215 1C resulted
in the formation of large colorless blocks. Their structural
investigation revealed the formation of an extended zigzag
‘‘copper wire’’ [Cu(O2C16H23)]N,zigzag (3), having a different
spatial distribution of the same structural units as in 2 (Fig. 2).
Although the two polymorphic forms crystallize in different
space groups, monoclinic P21/c (2) vs. orthorhombic Pbcn (3),
their cell volumes, normalized by the number of formula units,
are essentially the same (382 and 380 A3 per [Cu(O2C16H23)] in
2 and 3, respectively). A noticeable shortening of ca. 0.06 A is
found for Cuꢀ ꢀ ꢀCu distances along the chains of 3 in
comparison with those of 2. The Cuꢀ ꢀ ꢀCu distances in 2 and
3 are significantly shorter than the sum of their van der Waals
radii (rvdW(Cu) = 1.92 A)12 and fall into the category
of metallophilic interactions, similar to those in Au(I)
compounds.3 The average Cu–Ocarb distances in 2 and 3 are
very similar (1.8445(3) and 1.8483(12) A, respectively).
The zigzag chain 3 is characterized by two Cu–Cu–Cu angles
Studies of extended metallic chains have been expanding over
the past decades due to their potential use as molecular wires.1
These investigations also offer new insights into development
of modern metal–metal bonding theories, understanding of
cooperative reactivity of multiple metal centers, and design of
metal-rich nanomaterials.2 Specifically, interactions between
the closed d10 shells in such networks have attracted
very broad attention.3 Target preparation of linear tri-4 and
tetracopper5 chains has been reported, along with two
extended helical copper(I) polymers held together by cuprophilic
Cuꢀ ꢀ ꢀCu interactions.6,7 The formation of a copper(I) pivalate
helix was explained by a delicate balance between the
cuprophilicity and the steric effect of bulky t-Bu groups.6
The helical structure revealed later for copper(I) 3,5-bis-
(trifluoromethyl)benzoate7 can also fit into the above description.
However, in contrast to copper(I) pivalate that was found to
be nonluminescent, copper(I) 3,5-bis(trifluoromethyl)benzoate
exhibits bright luminescence in the solid state.7 This experi-
mental observation has disproved the explanation provided
earlier6 that the infinite helical arrangement of copper(I) atoms
held by cuprophilicity is responsible for the nonradiative
decay. Overall, the above two helical polymers stand out in
the copper(I) carboxylate family, since all other extended 1D
structures are based on intermolecular copper–oxygen
interactions.8–10 These facts prompted us to investigate how
the controlled on/off switch of axial copper–oxygen inter-
actions can affect the structural outcome and the resulting
photoluminescent behavior.
Among carboxylates, the 2,4,6-triisopropylbenzoate ligand
is known to be sufficiently bulky to prevent self-association of
Department of Chemistry, University at Albany,
1400 Washington Ave., Albany, NY 12222-0100, USA.
E-mail: marina@albany.edu; Fax: +1 518-442-3462;
Tel: +1 518-442-4406
w Electronic supplementary information (ESI) available: Detailed
experimental methods and characterization. CCDC 814791–814794.
For ESI and crystallographic data in CIF or other electronic format
see DOI: 10.1039/c1cc11000d
Fig. 1 A fragment of the molecular structure of [Cu(O2C16H23)]N,linear
(2). Isopropyl groups and hydrogen atoms are omitted for clarity.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 6939–6941 6939

View Article Online
Product 1 is a well-defined chemical compound and cannot
be represented as, for example, a random mixture of species,
such as oligomeric parts of copper(I) wires, because significant
variations in crystallization conditions do not result in any
changes of its XRPD and IR fingerprints. Thus, layering
solutions of 1 in aromatic solvents (benzene, toluene, or
m-xylene) with hexanes at room temperature, using low
temperature-induced crystallization or heating of the m-xylene
solution in a sealed ampoule at 215 1C all led to the recovery of
unchanged 1 (no solvents are co-crystallized within the
crystals). The crystallization attempts are limited to the
utilization of aromatic solvents, because dissolution of 1 in
strong O- or N-donating solvents results in the immediate
disproportionation of the starting material to elemental
copper and Cu(II) carboxylate.
Fig. 2 A fragment of the molecular structure of [Cu(O2C16H23)]N,zigzag
(3). Isopropyl groups and hydrogen atoms are omitted for clarity.
of 96.22(1)1 and 180.00(1)1. The dihedral angles between the
carboxylic groups and phenyl ring planes are ca. 741 and 641 in
2
and 3 in comparison with 691 in the structure of
2,4,6-triisopropylbenzoic acid (ESIw, Fig. S4), showing just a
minor steric repulsion effect of neighboring isopropyl groups
upon formation of ‘‘copper wires’’.
Ultimately, although being unable to produce single crystals
of 1 from solution, we have crystallized the discrete dimeric
product [Cu2(O2C16H23)2ꢀ2C6H4Cl2] (4) from o-dichloro-
benzene at 12 1C. Complex 4 has a Cl-atom of the solvent
axially coordinated at both ends of a dicopper(I) unit with the
Cuꢀ ꢀ ꢀCl contacts of 2.895(1) A (Fig. 3). Moreover, we
were able to obtain the single crystal diffraction data for the
2,4,6-triisopropylbenzoate complex of Ag(I), which exhibited a
Cu(I) acetate-like structure10 based on metal–oxygen inter-
molecular interactions (ESIw, Fig. S12). Importantly,
The individual structural polymorphs of copper(I) 2,4,6-
triisopropylbenzoate, 2 and 3, are reproducibly obtained
under controlled experimental conditions. The X-ray powder
diffraction (XRPD) data confirmed the structural identity and
purity of both bulk crystalline materials (see ESIw for more
details). Interestingly, when crystals of 2 or 3 are dissolved in
aromatic solvents, such as benzene, toluene or xylenes, they
quantitatively transform back to 1, which under gas phase
sublimation conditions re-converts to 2 or 3 (all steps
confirmed by XRPD). These conversion procedures can
be repeated multiple times along a transformation cycle
for the linear wire 2 zigzag wire system, depending on the
crystallization conditions (Fig. 3).
the carboxylate stretches in the IR spectra of
1 and
[Ag2(O2C16H23)2]N are almost identical, providing strong
experimental evidence in favor of the above structural assign-
ment for 1. In addition, thermogravimetric analysis (TGA)
and differential scanning calorimetry (DSC) data show a close
similarity in behavior of 2 and 3 and their noticeable difference
from 1, which is consistent with their overall different solid
state structures based on Cuꢀ ꢀ ꢀCu (2 and 3) vs. Cuꢀ ꢀ ꢀO
interactions (1).
The X-ray powder diffraction data also demonstrated the
structural difference of starting bulk copper(I) 2,4,6-triisopropyl-
benzoate 1 from 2 and 3 (ESIw, Fig. S7). Importantly, careful
consideration of the carboxylate stretches in the IR spectra of
all products showed that 2 and 3 are nearly identical but
distinctly differ from 1 (ESIw, Fig. S9). Numerous attempts to
isolate single crystals by solution crystallization of 1 have
resulted in the formation of very thin colorless tangled fibers
that were not suitable for the single crystal X-ray diffraction
study. Thus, we cannot provide a direct crystallographic proof
and can only speculate that copper(I) 2,4,6-triisopropylbenzoate
(1) exhibits a structure similar to that of copper(I) acetate.10
The latter consists of dicopper units that are further linked
by intermolecular copper–oxygen interactions to form 1D
polymeric chains. This conclusion is supported by various
experimental data and merits further discussion.
The variety of structural forms displayed by copper(I)
2,4,6-triisopropylbenzoate allowed the first analysis of their
structure–photoluminescence correlations. Both polymorphs,
2 and 3, exhibit green-to-orange photoluminescence (PL) upon
exposure to UV radiation in the solid state (Fig. 4).
Specifically, the PL measurements (lex = 350 nm) carried
out at room temperature on crystalline samples revealed an
emission lmax at 563 nm (green region) for 2 and a broad peak
centered at 610 nm (orange region) for 3. Interestingly, the
Fig.
3 Schematic representation of synthetic transformations
between the forms of [Cu(O2C16H23)]. The molecular structure of
Fig. 4 Solid state PL spectra of 1 (black), 2 (blue), and 3 (red).
The emission of 4 is quenched.
[Cu2(O2C16H23)2ꢀ2C6H4Cl2] (4) is also depicted.
c
6940 Chem. Commun., 2011, 47, 6939–6941
This journal is The Royal Society of Chemistry 2011

View Article Online
Crystal data for 2: C16H23CuO2, M = 310.88, monoclinic, P21/c,
a = 17.979(3), b = 5.7850(10), c = 15.146(3) A, b = 104.285(3)1, V =
1526.5(5) A3, Z = 4, T = 100(2) K, m(Mo-Ka) = 1.426 mmꢂ1, 10 313
reflection measured, 2688 unique, Rint = 0.0381, full-matrix least-
squares refinement on F2 converged at R1 = 0.0609 and wR2 = 0.1664
for 218 parameters and 2214 reflections with I 4 2s(I) (R1 = 0.0725,
wR2 = 0.1753 for all data) and a GOF of 1.062.
emission intensity is substantially greater in the case of the
linear ‘‘copper wire’’ compared to the zigzag polymorphic
form. Such a drastic difference in PL properties of 2 and 3
(including both the emission wavelength and intensity) shows
that relatively small variations in positions of Cu atoms held
together by the same bridging ligand may lead to significant
changes in photoluminescence. A substantial difference in PL
emission of two different polynuclear Cu4- and Cu6-core
complexes bridged by the same carboxylate has been reported
by us recently.13 However, this is the first instance when two
structurally close polymorphs, that differ only by a slightly
different arrangement of the same molecular fragments
forming a 1D polymeric chain, show very different PL
properties.
For 3: C16H23CuO2,
M = 310.88, orthorhombic, Pbcn,
a = 34.687(3), b = 10.2092(8), c = 8.5888(7) A, V = 3041.5(4)A3,
Z = 8, T = 100(2) K, m(Mo-Ka) = 1.432 mmꢂ1, 24 294 reflection
measured, 3638 unique, Rint = 0.0396, full-matrix least-squares
refinement on F2 converged at R1 = 0.0375 and wR2 = 0.0966 for
180 parameters and 3228 reflections with I 4 2s(I) (R1 = 0.0425,
wR2 = 0.1003 for all data) and a GOF of 1.086.
For 4: C44H54Cl4Cu2O4,
a = 8.4789(9), b = 11.4630(13), c = 22.026(2) A, b = 92.125(2),
V = 2139.3(4) A3, Z = 2, T = 100(2) K, m(Mo-Ka) = 1.285 mmꢂ1
M = 915.77, monoclinic, P21/c,
,
18 036 reflection measured, 4987 unique, Rint = 0.0371, full-matrix
least-squares refinement on F2 converged at R1 = 0.0341 and wR2 =
0.0879 for 250 parameters and 4341 reflections with I 4 2s(I)
(R1 = 0.0402, wR2 = 0.0919 for all data) and a GOF of 1.047.
In contrast to the extended wire-type structures of 2 and 3,
compounds 1 and 4 are based on dinuclear Cu2(O2C16H23)2
units. The solid sample of 1 brightly emits light with a
maximum wavelength of 535 nm, while crystals of 4 show
no measurable emission. The latter may be due to the heavy
atom effect of the coordinated Cl atoms.14 Noteworthily, 1
emits light in the same region as the only known benzoate9
having the same Cu(I) acetate structural type.
1 J. F. Berry, in Multiple Bonds Between Metal Atoms, ed.
F. A. Cotton, C. A. Murillo and R. A. Walton, Springer, 2005,
pp. 669–706, and references therein; T.-W. Tsai, Q.-R. Huang,
S.-M. Peng and B.-Y. Jin, J. Phys. Chem. C, 2010, 114, 3641;
V. P. Georgiev and J. E. McGrady, Inorg. Chem., 2010, 49, 5591.
2 F. A. Cotton, Chem. Soc. Rev., 1983, 12, 35; M. H. Chisholm and
A. M. Macintosh, Chem. Rev., 2005, 105, 2949; D. L. Phillips,
C.-M. Che, K. H. Leung, Z. Mao and M.-C. Tse, Coord. Chem.
Rev., 2005, 249, 1476; A. Grodzicki, I. Lakomska, P. Piszczek,
I. Szymanska and E. Szlyk, Coord. Chem. Rev., 2005, 249, 2232;
S. Liu, Y.-F. Han and G.-X. Jin, Chem. Soc. Rev., 2007, 36, 1543;
G. Schmid, Chem. Soc. Rev., 2008, 37, 1909; A. S. Filatov and
M. A. Petrukhina, Coord. Chem. Rev., 2010, 254, 2234.
This study has revealed structural variations that can be
found for carboxylate complexes having the same bridging
ligand and the same empirical formula. Changes of experi-
mental conditions allowed us to use isopropyl groups of 2,4,6-
triisopropylbenzoate as a controllable switch for intermolecular
copper–oxygen interactions and to isolate the previously
unknown copper(I) carboxylate wires. The use of gas-phase
crystallization for this system proved to be experimentally
convenient to turn Cuꢀ ꢀ ꢀO intermolecular interactions
off, while solution crystallization put them back on. The
consideration of photoluminescent properties for the structurally
diverse copper(I) 2,4,6-triisopropylbenzoate products revealed
that emission wavelengths and intensities depend not only on
the overall structural type (for example, infinite wires vs.
dimeric structures) but show great sensitivity to subtle
differences in spatial distribution of Cu(I) atoms, even for
the otherwise similar polymeric chains constructed from the
same molecular units (linear vs. zigzag wires).
3 P. Pyykko, Chem. Rev., 1997, 97, 597; P. Pyykko, Angew. Chem.,
¨
¨
Int. Ed., 2004, 43, 4412; A. L. Balch, Gold Bull., 2004, 37, 45;
D. B. Leznoff and J. Lefebvre, Gold Bull., 2005, 38, 47; S. Sculfort,
P. Croizat, A. Messaoudi, M. Benard, M.-M. Rohmer, R. Welter
and P. Braunstein, Angew. Chem., Int. Ed., 2009, 48, 9663.
4 J. Beck and J. Strahle, Angew. Chem., Int. Ed. Engl., 1985, 24, 409;
¨
G. J. Pyrka, M. El-Mekki and A. A. Pinkerton, J. Chem. Soc.,
Chem. Commun., 1991, 84.
5 R. Clerac, F. A. Cotton, L. M. Daniels, J. Gu, C. A. Murillo and
´
H.-C. Zhou, Inorg. Chem., 2000, 39, 4488; Z.-K. Chan, Y.-Y. Wu,
J.-D. Chen, C.-Y. Yeh, C.-C. Wang, Y.-F. Tsai and J.-C. Wang,
Dalton Trans., 2005, 985; Y. Takemura, T. Nakajima and
T. Tanase, Dalton Trans., 2009, 10231.
6 T. Sugiura, H. Yoshikawa and K. Awaga, Inorg. Chem., 2006, 45,
7584.
7 Y. Sevryugina and M. A. Petrukhina, Eur. J. Inorg. Chem., 2008,
219.
We thank the National Science Foundation (NSF
CAREER grant, CHE-0546945) for funding and the University
at Albany for supporting the X-ray center at the Department
of Chemistry.
8 F. A. Cotton, E. V. Dikarev and M. A Petrukhina, Inorg. Chem.,
2000, 39, 6072; Y. Sevryugina, O. Hietsoi and M. A. Petrukhina,
Chem. Commun., 2007, 3853.
9 Y. Sevryugina, D. D. Vaughn and M. A. Petrukhina, Inorg. Chim.
Acta, 2007, 360, 3103.
10 M. G. B. Drew, D. A. Edwards and R. Richards, J. Chem. Soc.,
Chem. Commun., 1973, 124.
11 R. Clerac, F. A. Cotton, K. R. Dunbar, E. A. Hillard,
´
Notes and references
z Synthesis of 1: a mixture of 0.10 g of copper(I) acetate (0.41 mmol),
0.50 g of 2,4,6-triisopropylbenzoic acid (2.02 mmol) (1 : 2.5) and 0.05 g
of metallic copper in 40 mL of m-xylene was refluxed for 43 h with a
Dean-Stark trap. The obtained light brown solution was filtered
through Celite and then concentrated to dryness affording a brown
solid. This solid was washed 4 times with hexanes (4 ꢁ 10 mL) until it
was completely off-white and dried under vacuum at 65 1C for 12 h to
provide 0.25 g (99%) of 1. Anal. Calc. for C16H23CuO2: C, 61.81; H,
7.46. Found: C, 61.72; H, 7.35%. PL (lex = 350 nm, lmax = 535 nm).
M. A. Petrukhina and W. B. Smucker, C. R. Acad. Sci., Ser. IIc:
Chim., 2001, 4, 315; F. A. Cotton, E. A. Hillard and C. A. Murillo,
J. Am. Chem. Soc., 2002, 124, 5658.
12 S. S. Batsanov, Inorg. Mater., 2001, 37, 871; S. Nag, K. Banerjee
and D. Datta, New J. Chem., 2007, 31, 832.
13 A. S. Filatov, O. Hietsoi, Y. Sevryugina, N. N. Gerasimchuk and
M. A. Petrukhina, Inorg. Chem., 2010, 49, 1626.
14 E. F. Healy, S. Manzer, J. Gorman, A. Jones and N. Cristea,
Chem. Phys. Lett., 2010, 485, 258.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 6939–6941 6941