Reversible Cu4 ? Cu6 core interconversion and temperature induced single-crystal-to-single-crystal phase transition for copper(I) carboxylate

Article abstract of DOI:10.1021/ic9020367

The first example of a reversible [Cu₄] ? [Cu₆] interconversion for polynuclear copper(I) complexes under controlled experimental settings is reported. It illustrates the key role of specific crystal growth conditions for accessing the target cluster nuclearity that consequently determines physical properties of the resulting solid state products. Thus, when copper(I) benzoate crystallizes from a 1, 2-dichlorobenzene solution at room temperature, it forms [Cu₄]-core based crystalline material, [Cu₄(O₂CC₆H₅)₄] (1). In contrast, crystal growth by deposition from the gas phase at elevated temperatures results in the exclusive formation of [Cu6(O₂CC ₆H₅)₆] (2). Complexes 1 and 2 have been isolated in pure form, fully characterized, and reversibly interconverted into each other. The effect of a core structure on the spectroscopic properties of 1 and 2, such as IR, Raman, and photoluminescence, has been investigated. Additionally, a combination of X-ray powder and single crystal diffraction methods has been used to discover the temperature induced phase transition in the hexanuclear copper(I) system. Two modifications of 2 exhibiting slightly different solid state packing of the [Cu₆(O₂CC ₆H₅)₆] units have been identified at room and low temperature. Moreover, reversible single-crystal-to-singlecrystal transitions between these polymorphic forms have been confirmed. The important role of weak intermolecular interactions between polynuclear copper(I) units in the solid state has also been revealed and discussed.

Full text of DOI:10.1021/ic9020367

1626 Inorg. Chem. 2010, 49, 1626–1633
DOI: 10.1021/ic9020367
Reversible Cu4 T Cu6 Core Interconversion and Temperature Induced
Single-Crystal-to-Single-Crystal Phase Transition for Copper(I) Carboxylate
Alexander S. Filatov,† Oleksandr Hietsoi,† Yulia Sevryugina,† Nikolay N. Gerasimchuk,‡ and
Marina A. Petrukhina*,†
†Department of Chemistry, University at Albany, 1400 Washington Avenue, Albany, New York 12222 and
‡Department of Chemistry, Missouri State University, 901 South National Avenue, Springfield, Missouri, 65897
Received October 14, 2009
The first example of a reversible [Cu4] T [Cu6] interconversion for polynuclear copper(I) complexes under controlled
experimental settings is reported. It illustrates the key role of specific crystal growth conditions for accessing the target
cluster nuclearity that consequently determines physical properties of the resulting solid state products. Thus, when
copper(I) benzoate crystallizes from a 1,2-dichlorobenzene solution at room temperature, it forms [Cu4]-core based
crystalline material, [Cu4(O2CC6H5)4] (1). In contrast, crystal growth by deposition from the gas phase at elevated
temperatures results in the exclusive formation of [Cu6(O2CC6H5)6] (2). Complexes 1 and 2 have been isolated in pure
form, fully characterized, and reversibly interconverted into each other. The effect of a core structure on the spectroscopic
properties of 1 and 2, such as IR, Raman, and photoluminescence, has been investigated. Additionally, a combination of
X-ray powder and single crystal diffraction methods has been used to discover the temperature induced phase transition
in the hexanuclear copper(I) system. Two modifications of 2 exhibiting slightly different solid state packing of the
[Cu6(O2CC6H5)6] units have been identified at room and low temperature. Moreover, reversible single-crystal-to-single-
crystal transitions between these polymorphic forms have been confirmed. The important role of weak intermolecular
interactions between polynuclear copper(I) units in the solid state has also been revealed and discussed.
Introduction
behavior of the resulting systems. However, target preparation
of a desired cluster core with tailored properties remains a
significant synthetic challenge. We have recently turned to
investigation of polynuclear copper(I) complexes supported
by various carboxylate groups since they exhibit a remarkable
structural diversity and that should be the key factor allowing
tuning of their physical properties and chemical reactivity.
Among a rather limited number of the structurally confirmed
examples, there are discrete complexes with tetra-5 and hexa-
nuclear cores,6 as well as with extended structures formed by di-
Polynuclear complexes of copper(I), silver(I), and gold(I) are
subjects of considerable research activity focused on the inter-
play between their geometric and electronic structures,1 as well
as size-reactivity relationships for practical optoelectronic,2
catalytic,3 and biochemical applications.4 Specifically, interac-
tions between closed d10 shells in various clusters attract broad
attention, since the number and arrangement of metal centers in
close proximity affect stability, chemical, and photophysical
and tetranuclear copper units held by additional Cu
O
3 3 3
intermolecular interactions.7,8 Additionally, two new examples
*Corresponding author. E-mail: marina@albany.edu.
(1) (a) Balch, A. L. Struct. Bonding (Berlin) 2007, 123, 1–40. (b) Balch, A.
L. Angew. Chem., Int. Ed. 2009, 48, 2641–2644.
(2) (a) Ford, P. C.; Cariati, E.; Bourassa, J. Chem. Rev. 1999, 99, 3625–3647.
(b) Yam, V. W.-W.; Lo, K. K.-W. Mol. Supramol. Photochem. 1999, 31–112.
(c) Armaroli, N.; Accorsi, G.; Cardinali, F.; Listorti, A. Top. Cur. Chem. 2007, 280,
69–115. (d) Zhang, Q.; Ding, J.; Cheng, Y.; Wang, L.; Xie, Z.; Jing, X.; Wang, F. Adv.
Funct. Mater. 2007, 17, 2983–2990. (e) Osawa, M.; Hoshino, M. Chem. Commun.
2008, 47, 6384–6386. (f) Barbieri, A.; Accorsi, G.; Armaroli, N. Chem. Commun.
of extended helical chains held by cuprophilic Cu Cu inter-
3 3 3
actions in the solid state have been reported for copper(I)
pivalate9 and 3,5-bis(trifluoromethyl)benzoate.10 Analyzing
(5) Sevryugina, Y.; Hietsoi, O.; Petrukhina, M. A. Chem. Commun. 2007,
37, 3853–3855.
€
2008, 19, 2185–2193. (g) Knorr, M.; Guyon, F.; Khatyr, A.; Daschlein, C.;
(6) Sevryugina, Y.; Rogachev, A. Yu.; Petrukhina, M. A. Inorg. Chem.
Strohmann, C.; Aly, S. M.; Abd-El-Aziz, A. S.; Fortin, D.; Harvey, P. D. Dalton
Trans. 2009, 948–955.
(3) See Chem. Rev. 2008, 108 (Coinage Metals In Organic Synthesis),
2793-3442.
2007, 46, 7870–7879.
(7) Cotton, F. A.; Dikarev, E. V.; Petrukhina, M. A. Inorg. Chem. 2000,
39, 6072–6079.
(8) Sevryugina, Y.; Vaughn, D. D., II; Petrukhina, M. A. Inorg. Chim.
(4) (a) Klinman, J. P. Chem. Rev. 1996, 96, 2541–2561. (b) Kim, E.; Chufan,
E. E.; Kamaraj, K.; Karlin, K. D. Chem. Rev. 2004, 104, 1077–1133. (c) Hong, S.;
Huber, S. M.; Gagliardi, L.; Cramer, C. C.; Tolman, W. B. J. Am. Chem. Soc.
2007, 129, 14190–14192. (d) Wang, J.; Schopfer, M. P.; Sarjeant, A. A. N.; Karlin,
K. D. J. Am. Chem. Soc. 2009, 131, 450–451.
Acta 2007, 360, 3103–3107.
(9) Sugiura, T.; Yoshikawa, H.; Awaga, K. Inorg. Chem. 2006, 45, 7584–
7586.
(10) Sevryugina, Y.; Petrukhina, M. A. Eur. J. Inorg. Chem. 2008, 219–
229.
r
pubs.acs.org/IC
Published on Web 01/12/2010
2010 American Chemical Society

Article
Inorganic Chemistry, Vol. 49, No. 4, 2010 1627
these scattered cases, one could assign a certain structural
type to a specific bridging carboxylate, thus creating an
illusion that the former is determined by the ligand nature.
However, our recent investigation on the fragmentation of
polymeric copper(I) 3,5-bis(trifluoromethyl)benzoate has
provided the first system where several polynuclear Cun-core
complexes (n = 2, 4, and 6) were synthesized and structurally
characterized for the same carboxylate ligand.10 Although
crystallized as adducts with different polyaromatic donors,
the isolated discrete CuI complexes of various nuclearities
demonstrated that multiple polynuclear structural arrange-
ments exist for a given carboxylate bridging group. It was
also predicted that variations in the preparation or crystal-
lization conditions should affect the resulting structural out-
come. In this work, we support the above prediction by our
new results obtained for copper(I) benzoate, for which two
complexes of different nuclearity, [Cu4(O2CC6H5)4] ([Cu4])
and [Cu6(O2CC6H5)6] ([Cu6]), have been individually pre-
pared in high yield and fully characterized by X-ray diffrac-
tion and spectroscopic techniques. Importantly, it was
demonstrated that the [Cu4] and [Cu6] units reversibly inter-
convert into each other, providing a remarkable example of
such transformation for discrete polynuclear copper(I) com-
plexes. Moreover, a reversible single-crystal-to-single-crystal
temperature induced phase transition was revealed for the
[Cu6]-product. Thus, this system provided a unique case to
examine the effects of core structures and intermolecular
interactions on the resulting spectroscopic properties, includ-
ing IR, Raman, and photoluminescence.
Figure 1. ORTEP representation of the asymmetric unit in 1 showing
the Cu labeling scheme with the thermal ellipsoids drawn at the 40%
probability level. Cu blue, O red, C gray, H lightgray. This color scheme is
used in all figures.
having fluorinated substituents,5,6,8,10 since the presence
of water (as a result of reaction 1) in highly electrophilic
systems is not desirable. When we applied the latter
synthetic approach culminated with the vapor deposition
to access single crystals of copper(I) benzoate, colorless
blocks (2) having unique unit cell parameters were iso-
lated. Similar to 1, their elemental analysis was consistent
with a 1:1 copper-to-benzoate product composition. The
X-ray structural characterization of 2 performed at 173 K
revealed a different structural type, a planar hexanuclear
[Cu6(O2CC6H5)6] core instead of tetranuclear [Cu4-
(O2CC6H5)4] (1). Thus, copper(I) benzoate provided the
first example when variations in preparation/crystalliza-
tion conditions yielded two different polynuclear carbox-
ylate complexes, [Cu4] and [Cu6]. Therefore, we set out to
carefully investigate this interesting system focusing on
possible structural transformations between [Cu4] and
[Cu6] as well as on their structure-property correlations.
Solid-State Structure of [Cu4(O2CC6H5)4] (1). We have
reproduced the synthesis and solution crystallization sug-
gested by Edwards and co-workers11,12 and have recol-
lected the single-crystal X-ray diffraction experiment for 1
atlow temperature. Importantly, X-ray powder diffraction
confirmed the purity and the [Cu4] identity of the bulk
crystalline product obtained from either reaction 1 or 2, as
well as after its recrystallization from 1,2-dichlorobenzene
(Supporting Information Figures S5 and S9).
Results and Discussion
Syntheses and Structures. In general, CuI compounds
have been less studied than CuII analogues due to their
instability in solutions. This especially applies to the
Lewis acidic complexes, in which electrophilicity at the
CuI centers is enhanced by the presence of fluorinated
carboxylate ligands. Thus, solution chemistry of the
fluorinated benzoate copper(I) complexes was recently
found to be very limited5,6,10 due to facile disproportiona-
tion and oxidation reactions. The nonfluorinated analo-
gues, however, show better stability in solution. For
example, copper(I) benzoate was prepared based on the
reaction 1 and crystallized from solution back in 1973.11
A few years later, its structure was reported to have a
tetranuclear core, [Cu4(O2CC6H5)4] (1).12
Cu2O þ 2C6H5COOH f 2CuðO2CC6H5Þ þ H2O ð1Þ
To access copper(I) carboxylates, we have recently
developed a different preparation scheme (2) based on
the ligand-exchange procedure starting from copper(I)
trifluoroacetate followed by gas phase sublimation-
deposition steps for purification and crystal growth.
Copper(I) benzoate 1 crystallizes in the triclinic P1
space group with two crystallographically independent
tetramers in the asymmetric unit. The latter are similar
but not identical; both have a planar core comprised of
four copper atoms bridged by four benzoate ligands
alternating above and below the [Cu4] plane (Figure 1).
The geometric parameters of these tetramers are very
close to the data reported previously.12 The intramolecular
CuðO2CCF3Þ þ RCOOH f CuðO2CRÞ þ CF3COOH
ð2Þ
This path was shown to be very efficient for the syn-
thesis of several new polynuclear copper(I) benzoates
Cu Cu distances within the [Cu4] plane range from
3 3 3
˚
˚
2.713(2) A for Cu(2) Cu(3) to 2.748(2) A for Cu-
3 3 3
(11) Edwards, D. A.; Richards, R. J. Chem. Soc., Dalton Trans. 1973,
2463–2468.
(12) Drew, M. G. B.; Edwards, D. A.; Richards, R. J. Chem. Soc., Dalton
Trans. 1977, 299–303.
(1) Cu(4). For the second tetramer, these distances span
3 3 3
˚
from 2.718(2) A for Cu(5) Cu(6) to 2.733(2) A for
˚
3 3 3
Cu(5) Cu(8). The average value of Cu Cucarb-bridged
3 3 3
3 3 3

1628 Inorganic Chemistry, Vol. 49, No. 4, 2010
Filatov et al.
˚
Table 1. Selected Bond Distances (A) and Angles (deg) for [Cun(O2CR)n], R = C6H5, (3-F)C6H4, (2,3,4-F)3C6H2, and (3,5-F)2C6H3 (T = 173 K)
b
R
C6H5
(3-F)C6H4
(2,3,4-F)3C6H2
C6H5
(3,5-F)2C6H3
1 [this work]
35
45
2 [this work]
56
n
Cu Cucarb-bridged
4
4
4
2.690(1)
6
6
2.746(2)/2.718(2)
2.713(2)/2.718(2)
2.725(2)/2.723(2)
2.748(2)/2.733(2)
2.904(2)/3.125(2)
4.630(2)/4.460(2)
2.712(1)
2.728(1)
2.679(1)
2.693(1)
2.723(1)
2.706(1)
2.722(1)
2.826(1)
3 3 3
Cu Cunonbridged
3 3 3
2.879(2)
4.614(2)
2.927(1)
4.513(1)
2.981(1)
3.051(1)
4.704(1)
1.863(2)
2.962(1)
2.962(1)
3.189(1)
1.851(3)
a
Cu-Ocarb
1.865(7)/1.858(8)
3.239(2)
3.547(2)
1.857(4)
1.853(2)
Cu Cuintermol
3 3 3
Cu Ointermol
3 3 3
3.098(8)-3.468(8)
64.57(5)/70.19(6)
115.57(7)/109.82(7)
63.83(5)/69.89(6)
116.00(7)/110.10(7)
Cu-Cu-Cu
63.91(3)
116.03(3)
65.94(2)
114.06(2)
67.39(2)
120.55(2)
171.92(2)
66.15(2)
123.48(3)
170.27(3)
a Averaged. b Two crystallographically independent units.
˚
separations in 1 of 2.728(2) A is within the sum of the van
der Waals radii generally accepted for copper (rvdW(Cu) =
1.40 A).13 It is noteworthy that the rvdW value for copper
˚
was recently re-evaluated to be much longer, namely
1.92 A.14 Additional Cu Cu contacts between the non-
˚
3 3 3
˚
bridgedmetalcenterswithinthe[Cu4] cores(2.904(2) A for
˚
Cu(2) Cu(4) and 3.125 (2) A for Cu(5) Cu(7)) can be
3 3 3
3 3 3
mentioned. This tetrameric core is not unique and has
recently been found in two other discrete copper(I) car-
boxylate complexes, namely [Cu4(O2C(3-F)C6H4)4] (3)
and [Cu4(O2C(2,3,4-F)3C6H2)4] (4) (Table 1).5
The above-mentioned [Cu4] complexes bearing fluori-
nated benzoates crystallize in the higher symmetry space
groups than 1 (C2/c and Fddd for 3 and 4, respectively).
Furthermore, all sides of the [Cu4]-core in 4 are equal
(Table 1). Apart from the symmetry considerations, the
tetranuclear complexes 1, 3, and 4 have similar molecular
geometries. However, when checking the packing of
tetramers in 1, we found interesting features that were
overlooked in the past.12 Several types of intermolecular
interactions in the solid-state structure of 1 deserve to be
mentioned. First, there is an alignment of individual tetra-
copper units (Figure 2) with the shortest intermolecular
˚
Figure 2. Schematic representation ofintermolecular contacts(A) in the
solid-state structure of [Cu4(O2CC6H5)4] (1).
thus, may affect the physical properties of the crystalline
products. Importantly, intermolecular interactions of the
Cu Cu and Cu O types are absent in the solid state
structures of [Cu4(O2C(3-F)C6H4)4] and [Cu4(O2C(2,3,4-
F)3C6H2)4]. At the same time, the similarly long (3.23-
3.48 A) Cu π contacts can be identified in both 3 and 4.
Solid-State Structure of [Cu6(O2CC6H5)6] (2, T =173 K).
Copper(I) benzoate (2) crystallizes in the triclinic P1 space
group with one-half of the [Cu6(O2CC6H5)6] moiety being
crystallographically independent. The [Cu6(O2CC6H5)6]
molecule resides on an inversion center positioned in the
middle of its core. A planar hexagon is comprised of six
copper atoms bridged by six benzoate ligands alternating
above and below the plane (Figure 3). The hexanuclear
core structure was first seen in copper(I) 3,5-difluoro-
benzoate (5)6 and was later found in the polyarene adducts
of copper(I) 3,5-bis(trifluoromethyl)benzoate.10,15
˚
Cu(3) Cu(6) separation being 3.239(2) A. The Cu(4)
3 3 3
3 3 3
˚
Cu(5) contact of 3.547(2) A is longer. The corresponding
Cu(7)-Cu(6)-Cu(3) and Cu(8)-Cu(5)-Cu(4) angles are
162.35(7) and 172.30(7)ꢀ. Second, four Cu atoms of the two
tetramers that are not involved in intermolecular cupro-
3 3 3
3 3 3
˚
3 3 3
philic interactions exhibit Cu π(C6H5) contacts ranging
3 3 3
˚
from 3.21 to 3.70 A. Thus, the crystal packing of [Cu4] units
along the crystallographic a axis is based on alternating
weak intermolecular Cu Cu and Cu π interactions.
Third, there are no intermolecular cuprophilic interactions
3 3 3
3 3 3
along the crystallographic b axis. Instead, they are replaced
˚
by Cu O contacts (3.10-3.47 A) with weak Cu
π
3 3 3
3 3 3
˚
interactions (3.17-3.73 A) also being present. All these
intermolecular contacts are within or close to the sums of
the van der Waals radii for copper and oxygen or carbon
(rvdW(Cu, O) = 3.44 A and rvdW(Cu, C) = 3.69 A)14 and,
˚
˚
(13) Bondi, A. J. Phys. Chem. 1964, 68, 441–451.
(14) (a) Batsanov, S. S. Inorg. Mater. 2001, 37, 871–885. (b) Nag, S.;
Banerjee, K.; Datta, D. New J. Chem. 2007, 31, 832–834.
(15) Sevryugina, Y.; Jackson, E. A.; Scott, L. T.; Petrukhina, M. A. Inorg.
Chim. Acta 2008, 361, 3103–3108.

Article
Inorganic Chemistry, Vol. 49, No. 4, 2010 1629
with the simulated pattern based on the unit cell para-
meters obtained for 2 at 173 K (Supporting Information
Figure S6). However, when multiple crystals were re-
checked again by single crystal X-ray diffraction at 173
K, they all conformed to the same unit cell parameters
found for 2173K. Puzzled by these facts, we have recollected
the single crystal diffraction experiment for 2 at room
temperature and found a different set of unit cell para-
meters (2293K, Table 2). The structure solution of 2293K
revealed that its molecular building unit is
a
[Cu6(O2CC6H5)6] complex similar to that in 2173K. In
contrast to the low temperature modification, however,
there are two crystallographically independent hexanu-
clear complexes in the asymmetric unit of 2293K. This
results in the slightly different crystal packing of the
[Cu6] units at low and room temperatures (Figure 4).
While all major bond distances and angles of the
hexacopper(I) complexes in 2293K closely resemble those
in 2173K, some variations in the atom positioning affect
their overall arrangement in the solid state and result in the
observed temperature induced phase transition. The above
changes seem subtle and do not cause the crystal degrada-
tion, thus providing a possibility to study both phases by
single crystal X-ray diffraction. It was found that conver-
sion between the two phases is fully reversible and occurs
between 263 and 253 K. After repeating the cooling-
thawing cycles from 293 to 173 K three times for the same
crystal, the single crystal diffraction data showed only a
minor degradation of the crystal quality in comparison
with the previously collected data at the same temperature
(Table S1 of the Supporting Information). Although the
topochemical reactions in solid state are well-documen-
ted,16 the single-crystal-to-single-crystal phase transitions
are rare and commonly accompanied by problems asso-
ciated with very weak high-angle scattering and/or sub-
stantially increased mosaicity of crystals.17 Moreover,
reversible phase transitions between discrete molecular
structures are exceptional,18 perhaps due to the fact that
cooperative molecular rearrangements in such systems
are rarely accompanied by retention of the overall struc-
tural integrity. In contrast, the temperature- and guest-
induced single-crystal-to-single-crystal transformations
are relatively common in rigid coordination polymers19
and metal-organic frameworks.20 It is worth stressing
Figure 3. ORTEP representation of the [Cu6(O2CC6H5)6] complex
showing the Cu labeling scheme with thermal ellipsoids drawn at the
40% probability level. Symmetry operation: -x þ 1, -y, -z þ 1.
The Cu Cu distances within the [Cu6] core in 2 range
3 3 3
˚
from 2.6797(6) A for Cu(2) Cu(3) to 2.7231(6) A for
˚
3 3 3
Cu(1) Cu(2). The interior angles are 171.92(2)ꢀ for
3 3 3
Cu(1)-Cu(2)-Cu(3), 120.549(19)ꢀ for Cu(2)-Cu(1)-
Cu(3A), and 67.390(16)ꢀ for Cu(2)-Cu(3)-Cu(1A). This
results in additional short Cu Cu contacts between the
˚
nonbridged metal centers, Cu(2) Cu(2A) of 3.0514(8) A,
3 3 3
3 3 3
˚
and Cu(1) Cu(2A) of 2.9812(6) A. The average Cu-
Ocarb distance of 1.863(2) A in 2 is similar to those in 1 and
3 3 3
˚
˚
5 (1.861(7) and 1.851(3) A, respectively).
A comparison of the molecular [Cu6] cores in analo-
gous complexes with benzoate and 3,5-difluorobenzoate
bridges shows a close similarity with small variations in
copper-copper distances and angles (Table 1). In the
solid state, both [Cu6] complexes are not engaged in either
Cu O or Cu Cu intermolecular interactions. The
3 3 3
3 3 3
˚
shortest Cu O distance found in 2 is 4.088(2) A, while
the shortest intermolecular Cu Cu separation is
3 3 3
3 3 3
˚
5.9621(7) A. This contrasts the intermolecular contacts
found in 1 and, importantly, allows a rationalization of
different photophysical properties of 1 and 2, as discussed
(16) (a) Friscic, T.; MacGillivray, L. R. Z. Kristallogr. 2005, 220, 351–363.
(b) Garcia-Garibay, M. A. Angew. Chem., Int. Ed. 2007, 46, 8945–8947. (c)
MacGillivray, L. R.; Papaefstathiou, G. S.; Friscic, T.; Hamilton, T. D.; Bucar, D.-
K.; Chu, Q.; Varshney, D. B.; Georgiev, I. G. Acc. Chem. Res. 2008, 41, 280–291.
(d) Lauher, J. W.; Fowler, F. W.; Goroff, N. S. Acc. Chem. Res. 2008, 41, 1215–
1229.
(17) (a) Ma, J. P.; Dong, Y.-B.; Huang, R.-Q.; Smith, M. D.; Su, C.-Y.
Inorg. Chem. 2005, 44, 6143–6145. (b) Zhang, Y.-J.; Liu, T.; Kanegawa, S.; Sato,
O. J. Am. Chem. Soc. 2009, 131, 7942–7943.
(18) (a) Evers, J.; Klapotke, T. M.; Mayer, P.; Oehlinger, G.; Welch, J.
Inorg. Chem. 2006, 45, 4996–5007. (b) For a nonreversible phase transition
between discrete complexes, see: Mobin, S. M.; Srivastava, A. K.; Mathur, P.;
Lahiri, G. K. Inorg. Chem. 2009, 48, 4652–4654.
(19) (a) Wang, X.-Y.; Scancella, M.; Sevov, S. C. Chem. Mater. 2007, 19,
4506–4513. (b) Vittal, J. J. Coord. Chem. Rev. 2007, 251, 1781–1795. (c)
Kawano, M; Fujita, M. Coord. Chem. Rev. 2007, 251, 2592–2605.
(20) (a) Suh, M. P.; Cheon, Y. E. Aust. J. Chem. 2006, 59, 605–612. (b) Lee,
J. Y.; Lee, S. Y.; Sim, W.; Park, K.-M.; Kim, J.; Lee, S. S. J. Am. Chem. Soc.
2008, 130, 6902–6903. (c) Bernini, M. C.; Gandara, F.; Iglesias, M.; Snejko, N.;
Gutierrez-Puebla, E.; Brusau, E. V.; Narda, G. E.; Monge, M. A. Chem.;Eur. J.
2009, 15, 4896–4905. (d) Montney, M. R.; Supkowski, R. M.; Staples, R. J.;
LaDuca, R. L. J. Solid State Chem. 2009, 182, 8–17.
below. At the same time, weak intermolecular Cu
π
3 3 3
˚
interactions (3.01-3.72 A) can be identified in the solid-
state structures of 2 and 5, similarly to those found in
[Cu4] core complexes with aromatic carboxylate bridges.
Small variations in packing of the [Cu6] units related to a
different engagement of copper(I) atoms in Cu π inter-
3 3 3
actions are found in 2 and 5 (see the Supporting Informa-
tion for details) and that may perturb their electronic
structures and thus affect the photoluminescent proper-
ties of the crystalline materials.
Temperature-Induced Reversible Single-Crystal-to-
Single-Crystal Phase Transition for 2. To confirm the
identity of the bulk crystalline product 2, we relied on
the X-ray powder diffraction method. To our surprise, the
experimental spectrum recorded for single crystals of [Cu6-
(O2CC6H5)6] at room temperature showed no consistency

1630 Inorganic Chemistry, Vol. 49, No. 4, 2010
Table 2. Crystallographic Data and X-ray Experimental Conditions for Cun(O2CC6H5)n (n = 4, 6)
4 [this work]
Filatov et al.
6 [this work]
6 [this work]
n
4 [12]
1
2173K
2293K
empirical formula
formula weight
T (K)
crystal description
crystal color
C28H20Cu4O8
740.08
283-303
C28H20Cu4O8
738.60
173(2)
elongated plate
colorless
triclinic
C42H30Cu6O12
1107.90
173(2)
C42H30Cu6O12
1107.90
293(2)
block
block
colorless
triclinic
P1
colorless
triclinic
P1
crystal system
space group
triclinic
P1
P1
˚
a (A)
15.408(7)
13.784(12)
15.034(9)
92.43(6)
62.34(5)
112.71(6)
2571.981
4
13.4736(12)
14.9872(13)
15.3365(13)
62.6130(10)
67.9140(10)
87.182(2)
2519.0(4)
4
9.2520(7)
9.6674(7)
11.2642(8)
107.5690(10)
90.6550(10)
92.0470(10)
959.63(12)
1
11.5075(17)
13.189(2)
13.933(2)
95.643(2)
102.960(2)
95.054(2)
2037.5(5)
2
1.806(1)
0.0565, 0.1369
0.1090, 0.1660
1.004
˚
b (A)
˚
c (A)
R (deg)
β (deg)
γ (deg)
3
˚
V (A )
Z
Dcalcd (g/cm3)
R1a, wR2b [I > 2σ(I)]
R1a, wR2b (all data)
quality-of-fitc
1.948(1)
1.917(1)
0.064
0.0721, 0.1259
0.2112, 0.2167
1.016
0.0375, 0.0883
0.0561, 0.0961
1.012
3
˚
largest diff. peak/hole, e/A
0.933/1.169
0.581/0.361
1.065/0.532
P
P
P
P
P
aR1 = ||Fo| - |Fc||/ |Fo|. b wR2 = [ [w(Fo2 - Fc2)2]/ [w(Fo2)2]]1/2
.
c Quality-of-fit = [ [w(Fo2 - Fc2)2]/(Nobs - Nparams)]1/2, based on all data.
[Cu4] T [Cu6] Core Transformations. We have repro-
ducibly obtained individual copper(I) benzoate com-
plexes 1 and 2 having different core nuclearities under
controlled experimental conditions. The [Cu4] product
crystallizes from solution (xylene, dichlorobenzene) at
room temperature, while the [Cu6] crystals deposit from
the gas phase at elevated temperatures under reduced
pressure. The X-ray powder diffraction data have con-
firmed the structural identity and purity of both bulk
crystalline materials. For the next step, we applied sub-
limation-deposition procedures to the single crystals of
[Cu4] and completely converted those to [Cu6]. Similarly,
when crystals of [Cu6] are redissolved and then recrystal-
lized from dichlorobenzene, they quantitatively trans-
form back to [Cu4]. These conversion procedures can be
repeated multiple times, confirming a reversible transfor-
mation in the [Cu4] T [Cu6] system, depending on the
crystallization conditions.
Thermogravimetric analysis of [Cu4] and [Cu6] shows
very similar patterns for 1 and 2 (Figures S12 and S15 of
the Supporting Information). The weight loss commences
at ca. 150 ꢀC corresponding to the cleavage of carboxylate
groups and some partial sublimation of compounds
under the dinitrogen flow. After the initial weight loss,
the plateau is observed and the final decomposition
product corresponds to 42% of the mass residue. This is
confirmed by X-ray powder diffraction to be copper(I)
monooxide (Cu2O) in 1 and 2. The thermal behavior of
both complexes was also studied by differential scanning
calorimetry (Figures S13, S14, and S16). The thermo-
graph for [Cu6] is featureless up to ca. 260 ꢀC at which the
sharp endothermic peak is observed and attributed to the
product melting followed by decomposition. In contrast,
the thermograph of [Cu4] shows a broad exothermic peak
at 200-225 ꢀC before the product melting and thermal
decomposition at ca. 260-270 ꢀC. This peak corresponds
to the transformation of [Cu4] to [Cu6], as confirmed by
the X-ray powder diffraction experiments (Figure S8).
When the sample of [Cu4] was heated at 150 ꢀC for 20 days
Figure 4. Fragment of the crystal packing of the [Cu6] units at (a) 173
and (b) 293 K. Benzoate groups are omitted for clarity.
here that temperature induced phase transitions should
be taken into account when examining the solid state
properties of bulk crystalline materials and considering
their structure-property relationships. On many occa-
sions, there is a mismatch between the property measure-
ments (often done at room temperatures) and single
crystal diffraction data (usually collected at low tem-
peratures) used for structural assignments and correla-
tions. In these cases, X-ray powder diffraction should be
used as an indispensable tool to rule out the possibility
of phase transformations in addition to confirming the
full structural match of selected single crystals and bulk
materials.
The reversible single-crystal-to-single-crystal phase
transition found in [Cu6(O2CC6H5)6] allowed us to test
if the observed changes in the solid state packing affect
bulk properties of the same molecular complex (see
Photoluminescent Properties).

Article
Inorganic Chemistry, Vol. 49, No. 4, 2010 1631
in a sealed ampule, no phase changes were observed, while
increasing temperature to 200 ꢀC led to appearance of the
[Cu6] phase. We can only speculate that this transforma-
tion proceeds through local sublimation, gas phase core
rearrangement and ligand redistribution, and crystalliza-
tion sequence based on the fact that this process is
exothermic. It seems rather improbable that such drastic
structural changes may occur in the solid state only, and
in the case of the melt reaction, the endothermic peak
should have been observed.21
Such reversible interconversions between molecular
structures of the same composition are rather unique;
the other examples for the metal carboxylate family
include the triangle-to-square transformations of dirho-
dium(II) oxalate22 and dimolybdenum(II) perfluoroter-
ephthalate.23 Lippard and co-workers have recently re-
ported a reversible interconversion where the use of
different solvents for crystallization led to the isolation
of iron(II) carboxylate complexes with different nuclea-
rities.24 The formation of dinuclear and hexanuclear
copper(II) pivalates that seemed to be dependent on the
basicity and coordinating ability of the base employed in
the reaction can also be mentioned here.25 It should be
emphasized that in contrast to interconversions involving
discrete molecular complexes, the monomer-oligomer-
polymer transformations are very common among the
coordination polymers. For example, solvent-dependent
interconverting systems were reported for coinage,26
d-transition,27 and lanthanide metal ions.28
Spectroscopic Characterization of [Cu4(O2CC6H5)4] (1)
and [Cu6(O2CC6H5)6] (2). The controlled preparation of 1
and 2, having different copper(I) core structures bridged
Figure 5. (a) IR and (b) Raman spectra of [Cu4] (red) and [Cu6] (blue) in
the carboxylate stretching region.
(21) Kaftory, M.; Botoshansky, M.; Kapon, M.; Shteiman, V. Acta
Crystallogr. 2001, B57, 791–799.
(22) Cotton, F. A.; Daniels, L. M.; Lin, C.; Murillo, C. A. J. Am. Chem.
Soc. 1999, 121, 4538–4539.
(23) Cotton, F. A.; Murillo, C. A.; Yu, R. Dalton Trans. 2006, 3900–3905.
(24) Reisner, E.; Telser, J.; Lippard, S. J. Inorg. Chem. 2007, 46, 10754–
10770.
(25) Mikuriya, M.; Azuma, H.; Nukada, R.; Handa, M. Chem. Lett. 1999,
57–58.
by the same benzoate ligands, allowed us to examine the
effect of their molecular geometry and crystal packing on
spectroscopic and photophysical properties.
IR and Raman. Infrared and Raman vibrational spec-
troscopies have been extensively used for structural iden-
tification of metal-ligand complexes29 and proved to be
powerful methods for characterization of small structural
variations in coinage metal complexes.30 Thus, Raman
spectroscopy allowed differentiation of four CuII acetate
structures in aqueous solution, namely protonated, free,
pseudobridging between a water proton and a copper ion,
and bidentate.31
The solid state IR and Raman spectra of 1 and 2 are
nearly identical although most bands resulting from the
same vibrational modes are shifted (Δ = 1-10 cm-1).
Despite the apparent similarities of the vibrational pat-
terns of the [Cu4] and [Cu6] complexes, it is informative to
closely look at the carboxylate stretching region in both
types of spectra (Figure 5). The main difference in the
infrared spectra involves the symmetric CO stretch that
shows up at 1543 cm-1 in 2 but is shifted by 9 cm-1 to the
lower energy in 1. This may be a result of different
intermolecular interactions found in the solid-state struc-
tures of two copper(I) benzoates. As mentioned above,
(26) (a) Matsumoto, N.; Motoda, Y.; Matsuo, T.; Nakashima, T.; Re, N.;
Dahan, F.; Tuchagues, J.-P. Inorg. Chem. 1999, 38, 1165–1173. (b) Chan-
drasekaran, P.; Mague, J. T.; Balakrishna, M. S. Inorg. Chem. 2006, 45, 6678–
6683. (c) Liu, H.-J.; Hung, Y.-H.; Chou, C.-C.; Su, C.-C. Chem. Commun. 2007,
495–497. (d) Heo, J.; Jeon, Y.-M.; Mirkin, C. A. J. Am. Chem. Soc. 2007, 129,
7712–7713. (e) Stadler, A.-M.; Kyritsakas, N.; Vaughan, G.; Lehn, J.-M.
Chem.;Eur. J. 2007, 13, 59–68.
(27) (a) Ramirez, J.; Stadler, A.-M.; Kyritsakas, N.; Lehn, J.-M. Chem.
Commun. 2007, 237–239. (b) Uehara, K.; Kasai, K.; Mizuno, N. Inorg. Chem.
2007, 46, 2563–2570. (c) Kang, H. J.; Noh, T. H.; Jin, J. S.; Jung, O.-S. Inorg.
Chem. 2008, 47, 5528–5530. (d) Weilandt, T.; Troff, R. W.; Saxell, H.; Rissanen,
K.; Schalley, C. A. Inorg. Chem. 2008, 47, 7588–7598.
(28) (a) Mamula, O.; Lama, M.; Stoeckli-Evans, H.; Shova, S. Angew.
Chem., Int. Ed. 2006, 45, 4940–4944. (b) Lama, M.; Mamula, O.; Kottas, G. S.;
De Cola, L.; Stoeckli-Evans, H.; Shova, S. Inorg. Chem. 2008, 47, 8000–8015.
(29) Nakamoto, K. Infrared and Raman Spectra of Inorganic and Co-
ordination Compounds Part. B. Applications in Coordination, Organometallic,
and Bioinorganic Chemistry, 5th Ed.; John Wiley & Sons, Ltd.: New York, 1997.
(30) (a) Bott, R. C.; Bowmaker, G. A.; Davis, C. A.; Hope, G. A.; Jones,
B. E. Inorg. Chem. 1998, 37, 651–657. (b) Zhang, Q.; Cao, R.; Hong, M.; Wu, D.;
Zhang, W.; Zhen, Y.; Liu, H. Inorg. Chim. Acta 1998, 271, 93–98. (c) Alvarez, M.
L.; Ai, J.; Zumft, W.; Sanders-Loehr, J.; Dooley, D. M. J. Am. Chem. Soc. 2001,
123, 576–587. (d) Hagadorn, J. R.; Zahn, T. I.; Que, L., Jr.; Tolman, W. B. Dalton
Trans. 2003, 1790–1794. (e) Vorontsov, I. I.; Kovalevsky, A. Yu.; Chen, Y.-S.;
Graber, T.; Gembicky, M.; Novozhilova, I. V.; Omary, M. A.; Coppens, P. Phys.
Rev. Lett. 2005, 94, 193003-1–193003-4. (f) Phillips, D. L.; Che, C.-M.;
Leung, K. H.; Mao, Z.; Tse, M.-C. Coord. Chem. Rev. 2005, 249, 1476–1490. (g)
Kato, C. N.; Mori, W. C. R. Chimie 2007, 10, 284–294. (h) Dias, H. V. R.; Flores,
J. A.; Wu, J.; Kroll, P. J. Am. Chem. Soc. 2009, 131, 11249–11255.
the additional Cu O interactions are found in 1 but
are absent in 2. In the Raman spectrum of 1, all peaks
3 3 3
are noticeably shifted to higher energy by 3-7 cm-1
(31) Quiles, F.; Burneau, A. Vib. Spectrosc. 1998, 16, 105–117.

1632 Inorganic Chemistry, Vol. 49, No. 4, 2010
Filatov et al.
compared to 2. Thus, the solid [Cu4] and [Cu6] benzoate
complexes can be distinguished by their IR and Raman
spectra, although their X-ray powder diffraction identi-
fication is more straightforward.
Photoluminescent Properties. Although the first reports
on fluorescence of aliphatic copper(I) carboxylates ap-
peared in the literature about 30 years ago,32 there have
been no follow-up studies on the origin of photolumines-
cence(PL) or structure-property relationship forthisclass
of compounds. The emission maxima of copper(I) for-
mate, acetate, propionate, butyrate, valerate, hexanoate,
and heptanoate were found to vary in the broad range of
535-660 nm at room temperature (λex = 305-325 nm),
but those were not correlated with the structures of com-
plexes.32 We have recently found that the emission wave-
lengths for the structurally similar discrete tetranuclear
complexes, [Cu4(O2C(3-F)C6H4)4] (3) and [Cu4(O2C(2,3,
4-F)3C6H2)4] (4), are veryclose.5 Inthe solid state, uponex-
posure to UV radiation (λex = 350 nm), both compounds
exhibited broad emission bands centered at 502 and
507 nm, respectively. In contrast, the PL measurements
for solid tetranuclear copper benzoate, [Cu4(O2CC6H5)4]
(1), revealed an emission being significantly red-shifted to
676 nm (λex = 418 nm). This may result from an essential
difference of the solid-state structure of 1 exhibiting inter-
molecular copper-copper and copper-oxygen interac-
tions between the tetramers. Importantly, the latter
interactions are absent in the structures of 3 and 4. We
have previously observed that the emission wavelength of
the tetranuclear carboxylates shows a dependence on the
general structural type (discrete clusters vs. extended
motifs). For example, in contrast to the above-mentioned
3 and 4 having discrete tetranuclear cores, the polymeric
Figure 6. Variable temperature PL spectra of solid sample 2 taken at 0
(bottom curve), -29, -55, and -83 ꢀC (top curve).
observed with a temperature decrease from 0 to -83 ꢀC,
the overall shift in peak position constituted only 2-3 nm
(λex=260 nm) and can be considered negligible (Figure 6).
The range from 0 to -30 ꢀC, where phase transition for 2
was structurally detected, was carefully examined (mea-
sured at every 2-3 ꢀC from 0 to -40 ꢀC), but subtle
differences in packing of the [Cu6] units seems show no
effect on the electronic communication between Cu(I)
centers in this case.
Conclusions
In summary, two different copper(I) core complexes with
the same bridging ligand, [Cu4(O2CC6H5)4] ([Cu4]) and
[Cu6(O2CC6H5)6] ([Cu6]), have been fully characterized in
this work. While the [Cu4] structure was previously reported,
new features of its solid-state packing have been revealed,
which assisted in rationalizing the photoluminescent trends
for the [Cu4] core based carboxylate series. The structure of
[Cu6] is reported here for the first time and adds the second
example of a hexanuclear core copper(I) carboxylate having
no exogenous ligands. The use of powder X-ray diffraction
allowed us to detect a temperature induced phase transition
for [Cu6(O2CC6H5)6]. Both low and room temperature
phases were studied by single crystal X-ray diffraction to
reveal some minor differences in packing of the hexacopper(I)
units.
The reversible [Cu4] T [Cu6] core interconversion found in
this system represents a unique example for the copper(I)
carboxylate family. It was revealed only after different crystal
growth conditions have been tested and successfully yielded
variable crystalline products for a given metal-ligand com-
bination. These variations in experimental conditions are
very important in controlling the nuclearities and solid-state
structures of copper(I) clusters, which is the key in affecting
their properties, such as photoluminescence.
complexes built on strong axial Cu O interactions be-
3 3 3
tween the tetramers, [Cu4(O2CCF3)2(O2CC6F5)2]¥ and
[Cu4(O2CCF3)4]¥, both display emission red-shifted to
583 nm (λex = 350 nm).5 In this work again, we observe
possible implications of copper-copper and copper-oxy-
gen intermolecular interactions on the photophysical
properties of tetramers.
The emission maxima, λmax, of the crystalline solid
sample of [Cu6(O2CC6H5)6] (2) is 577 nm at room tem-
perature (λex = 350 nm). For comparison, the structu-
rally similar hexanuclear copper(I) 3,5-difluorobenzoate,
[Cu6(O2C(3,5-F)2C6H3)6] (5), exhibits photolumines-
cence at ca. 554 nm (λex = 350 nm) in the solid state.6
Although the molecular [Cu6] core structures of 2 and 5
are quite similar, the observed shift in emission can stem
from electronic structure variations caused by different
bridging carboxylate groups and also from small varia-
tions in their crystal packing patterns (see the Supporting
Information for details).
The revealed phase transition in 2 as well as experi-
mental observations that solid-state photophysical prop-
erties strongly correlate with molecular packing33 and can
be significantly different for different polymorphs1 forced
us to look closely at PL of [Cu6] at variable temperatures.
While a noticeable increase in emission intensity was
Experimental Section
Materials and Methods. All manipulations were performed
under a dinitrogen atmosphere employing standard Schlenk
techniques. Sublimation-deposition procedures were carried
out in small Pyrex glass ampules of 1.1 cm o.d. and of varied
length (5-8 cm). The ampules were evacuated to ca. 10-2 Torr,
sealed, and then placed in electric furnaces having a ca. 5 ꢀC
temperature gradient along the length of the tube. [Cu4-
(O2CCF3)4] was synthesized following a previously reported
procedure.7 Benzoic acid, 99%, was obtained from Acros.
Copper(I) oxide, 97%, and trifluoroacetic anhydride, 99%,
(32) (a) Hardt, H. D.; Weber, P. Z. Anorg. Allg. Chem. 1978, 442, 225–229.
(b) Weber, P.; Hardt, H. D. Inorg. Chim. Acta 1982, 64, L51–L53.
(33) (a) Shin, C. H.; Huh, J. O.; Lee, M. H.; Do, Y. Dalton Trans. 2009,
6476–6479. (b) Kumar, N. S. S.; Varghese, S.; Suresh, C. H.; Rath, N. P.; Das, S. J.
Phys. Chem. C 2009, 113, 11927–11935.

Article
Inorganic Chemistry, Vol. 49, No. 4, 2010 1633
were received from Sigma-Aldrich. All solvents (m-xylene,
anhydrous, 99þ%; benzene, anhydrous, 99.8%; 1,2-dichloro-
benzene, anhydrous, 99%; hexanes, mixture of isomers, anhy-
drous, 99%) were purchased from Sigma-Aldrich and used as
received.
[Cu4(O2CC6H5)4]. The X-ray powder spectrum of the micro-
crystalline bulk material was fully consistent with the calculated
powder pattern based on the single crystal unit cell parameters
of 1 (Figure S5).
Synthesis of [Cu6(O2CC6H5)6] (2). Crystals of 2 were obtained
by sublimation (Tsubl = 230 ꢀC) of the [Cu4(O2CC6H5)4] powder
prepared as described above. Yield of 2 is 35-40% after 12 h
and 70-75% after 2 weeks of sublimation. Anal calc for
C7H5O2Cu: C, 45.53; H, 2.73; O, 17.33. Found: C, 45.25; H,
Physical Measurements. Elemental analyses were performed
by Guelph Chemical Laboratories Ltd., Canada. The room
temperature photoluminescence (PL) spectra of crystalline pow-
ders were collected on a HORIBA Jobin Yvon spectrofluori-
meter using front-face detection. Three full scans were recorded
in each case and then averaged. Default hardware settings (the
slit width of 2 nm and integration time of 0.2 s) were applied. The
solid-state variable temperature PL spectra from fine powder of
[Cu6(O2CC6H5)6] (2) were recorded in the range of 300-900 nm
using the custom-made liquid N2 filled cryostat with 90ꢀ detec-
tion geometry (Supporting Information, Figure S18) that was
designed to work on a Shimadzu RF-5301 PC spectrofluori-
meter. The temperature range was varied from -90 to 0 ꢀC with
the accuracy of temperature monitoring during spectra record-
ing better than 2 ꢀC. Complex 2 was found to be a very bright
emitter regardless of the wavelength in UV and visible region
used for the excitation. Thus, the selected value of λex = 260 nm
was the closest to the visible range wavelength that allowed
recording of an untruncated signal using this particular instru-
ment. Only at λex = 260 nm with the hardware settings of 1 nm
minimal slit width, sampling interval of 1 nm, and a sampling
time of 0.5 s, the recorded PL spectra would fit the vertical scale
of the employed Shimadzu RF-5301 PC instrument. Actual
photographs of 2 under UV-light are shown in Figure S19.
The IR spectra were recorded on a Perkin-Elmer FT-IR Spec-
trometer (Spectrum 100) in the 4000-600 cm-1 range, using a
universal ATR accessory. The Raman spectroscopic measure-
ments were performed on an InVia Raman Microscope
(Renishaw) using 785-nm laser light excitation. The GRAMS/
AI software package (Thermo Galactic) was used for the
spectroscopic data processing.
2.45; O, 17.69%. PL (λex = 350 nm, λmax, nm) 577. IR (νmax
/
cm-1): 3062s, 3025w, 1620w, 1592s, 1543s, 1491s, 1410s, 1313 m,
1279w, 1169m, 1095w, 1066m, 1024m, 1003w, 975w, 930w,
843m, 811w, 704s, 690s, 678s. Raman (νmax/cm-1): 118w,
152w, 182w, 207w, 219w, 616m, 813w, 844s, 1003vs, 1025m,
1143m, 1164m, 1181m, 1418s, 1493m, 1599s, 3070w. The X-ray
powder spectrum of the microcrystalline bulk sample was fully
consistent with the calculated powder pattern based on the
single crystal unit cell parameters for 2293K (Figure S7 of the
Supporting Information).
Transformation of [Cu4(O2CC6H5)4] to [Cu6(O2CC6H5)6].
Crystals of 1 (0.035 g) obtained by recrystallization from 1,2-
dichlorobenzene as described above were sublimed in an evac-
uated glass ampule at 230 ꢀC. The heating was stopped after 12
h, and crystals of [Cu6(O2CC6H5)6] were collected from the walls
of the ampule as colorless blocks. Yield: 0.015 g, 41%.
Transformation of [Cu6(O2CC6H5)6] to [Cu4(O2CC6H5)4].
Crystals of 2 (0.032 g) were suspended in 5 mL of 1,2-dichlor-
obenzene, and the mixture was stirred for 2 h at rt. The saturated
solution was filtered off (leaving 0.021 g of an undissolved solid)
and carefully layered with 5 mL of hexanes. In two days,
colorless crystals of [Cu4(O2CC6H5)4] appeared at the bottom
of the tube. Yield: 0.009 g, 83% (based on the dissolved solid).
X-ray Structural Determinations. The X-ray data sets were
collected for 1 and 2 on a Bruker APEX CCD X-ray diffract-
ometer equipped with a graphite monochromated Mo KR
˚
radiation source (λ = 0.71073 A). Data reduction and integra-
tion were performed with the software package SAINT,34 and
absorption corrections were applied using the program SA-
DABS.35 All structures were solved and refined using the
SHELXTL program package.36 All non-hydrogen atoms were
refined anisotropically. Hydrogen atoms were included at idea-
lized positions as a riding model. Details on crystallographic
data and experimental conditions for copper(I) benzoates are
collected in Table 2. Crystallographic data for the single-crystal-
to-single-crystal phase transition are shown in Table S1
(Supporting Information).
Synthesis of [Cu4(O2CC6H5)4] (1). Benzoic acid (2.5 g, 10.24
mmol) was heated under reflux in 40 mL of m-xylene for 2 h with
a Dean-Stark trap attached to the flask. The resulting solution
was added to copper(I) oxide (0.2 g, 1.40 mmol) and reflux was
continued until all the oxide had reacted (ca. 12 h). The solution
was then slowly cooled to room temperature, and the white
precipitate was filtered off. The polycrystalline powder was
washed three times with m-xylene (20 mL) and dried under
reduced pressure. Yield: 0.387 g, 75%. Anal calc for C7H5O2Cu:
C, 45.53; H, 2.73; O, 17.33. Found: C, 45.83; H, 2.71; O, 16.95%.
PL (λex = 418 nm, λmax, nm) 676. IR (νmax/cm-1): 3065s, 3028w,
1622w, 1594s, 1534s, 1492s, 1410s, 1314m, 1279w, 1173m,
1096w, 1069m, 1024m, 1003w, 976w, 936w, 843m, 811w, 705s,
690s, 678s. Raman (νmax/cm-1): 117w, 146w, 182w, 205w, 217w,
617m, 808w, 845s, 1004vs, 1027m, 1147m, 1161m, 1181m,
1425s, 1497m, 1602s, 3070w.
The X-ray powder pattern for this product was consistent
with the calculated pattern for 1 based on the single crystal unit
cell parameters (Supporting Information, Figure S9). To obtain
single crystals of [Cu4(O2CC6H5)4], 0.040 g of the above powder
was suspended in 10 mL of 1,2-dichlorobenzene and stirred
at room temperature for 2 h. The saturated solution was filtered
off (leaving ca. 0.020 g of an undissolved solid) and layered
with 5 mL of hexanes. Upon hexane diffusion, colorless elon-
gated plates appeared at the bottom of the flask. The solution
was then carefully removed by cannula, and the resulting
crystals were washed two times with m-xylene (5 mL) and dried
under vacuum for 2 days. Yield: 0.011 g, 55% (based on
the dissolved amount of solid). Several crystals from the preci-
pitate were analyzed by single-crystal X-ray diffraction, and all
conformed to those previously reported by Edwards12 for
Acknowledgment. We thank the National Science Founda-
tion for partial financial support (CHE-0546945) and for fund-
ing the X-ray powder diffractometer (CHE-0619422). We are
very grateful to the University at Albany for supporting the
X-ray center at the Department of Chemistry. We also thank
Mr. Vitali Sikirzhytski (Prof. Lednev’s group, University at
Albany) for recording Raman spectra, and Dimitri Vaughn II
and Beth Weaver for assistance in synthesis of copper(I) benzo-
ate complexes during their undergraduate research study at the
University at Albany.
Supporting Information Available: IR and Raman studies, X-
ray powder diffraction details, X-ray data, TGA/DSC details,
and PL data. This material is available free of charge via the
Internet at http://pubs.acs.org.
(34) SAINT, version 6.02; Bruker AXS, Inc.: Madison, WI, 2001.
(35) SADABS; Bruker AXS, Inc.: Madison, WI, 2001.
(36) Sheldrick, G. M. SHELXTL, version 6.14; Bruker AXS, Inc.: Madison,
WI, 2001.