A unique copper(ii)-assisted transformation of acetylacetone dioxime in acetone that leads to one-dimensional, quinoxaline-bridged coordination polymers

Article abstract of DOI:10.1039/C6DT03595G

The reactions of copper(ii) carboxylate sources with acetylacetone dioxime (acacdoH₂) in Me₂CO have been studied and a novel, metal ion-assisted ligand transformation has been discovered. The reaction of [Cu₂(diba)₄(dibaH)₂] and acacdoH₂ (1 : 1.5) in Me₂CO has provided access to the complex {[Cu₂(diba)₄(qunx)]}_n (1) in low yield (25-30%), where dibaH is 3,3-dimethylbutyric acid and qunx is quinoxaline. The [Cu₂(piv)₄(pivH)₂]/acacdoH₂ (1 : 1.5) reaction system in warm Me₂CO, where pivH is pivalic acid, gave the analogous complex {[Cu₂(piv)₄(qunx)]}_n (2) in moderate yield (～50%). Complexes 1 and 2 can be easily prepared by the direct 1 : 1 reactions between the corresponding copper(ii) carboxylate starting materials and qunx in Me₂CO and MeOH, respectively. The formation of coordinated qunx in 1 and 2 is Cu^II-promoted (assisted) as suggested by the failure to synthesize the free qunx by a variety of reactions of acacdoH₂ and Me₂CO under aerobic conditions in the absence or even the presence of dibaH and pivH, respectively. The observed acacdoH₂ → qunx transformation is catalytic and new in the chemistry of the dioximes of β-diketones, and a mechanism has been proposed based on well-established reactions of organic chemistry. The mechanism is based on a double Beckmann rearrangement-type transformation and the overall scheme is represented by the 1 : 1 : 1 reaction between acacdoH₂, Me₂CO and O₂. Complexes 1 and 2 have similar molecular structures consisting of paddle-wheel {Cu₂(η¹:η¹:μ-O₂CR)₄} units bridged by qunx ligands in a zigzag 1D chain arrangement. The geometry of the Cu^II ions is square pyramidal with a quinoxaline nitrogen atom occupying the apical position at each metal ion. Weak H bonds are present within the chains, the donors being qunx carbon atoms and the acceptors being coordinated carboxylate oxygen atoms. Neighbouring chains interact through C-H ...π interactions between diba^-/piv^- methyl groups and the “pyrazine” part of qunx forming layers which are stacked along the b (1) or a (2) axis through weak van der Waals interactions. The packing of the layers is different in the two structures, due to the different nature of the carboxylate ligands. Hirshfeld surface analysis of the two structures reveals the similarity of the interchain (intralayer) interactions. The IR and Raman data of 1 and 2 are discussed in terms of the coordination mode of the carboxylate groups and permit assignments of some characteristic bands/peaks of coordinated qunx. Dc magnetic susceptibility studies in the 1.8-310 K range reveal very strong antiferromagnetic Cu^II ...Cu^II exchange interactions within the carboxylate-bridged Cu₂ units (J = -479 K for 1 and -532 K for 2 using the H = - J∑S₁·S₂ spin Hamiltonian) and weaker antiferromagnetic interactions between the Cu₂ units via the qunx superexchange pathways, with the latter being ～10% in strength compared to the former. A critical discussion of the acacdoH₂ → qunx transformation in 1 and 2 is provided in the light of other impressive, recently discovered Cu^II-assisted transformations of acacdoH₂, pointing out the key role of the solvent in the processes known to date.

Full text of DOI:10.1039/C6DT03595G

Dalton
Transactions
View Article Online
View Journal
PAPER
A unique copper(II)-assisted transformation of
acetylacetone dioxime in acetone that leads to
one-dimensional, quinoxaline-bridged
coordination polymers†‡
Cite this: DOI: 10.1039/c6dt03595g
Zoi G. Lada,a,b Amaia Soto Beobide,a Aikaterini Savvidou,c
Catherine P. Raptopoulou,c Vassilis Psycharis,*c George A. Voyiatzis,*a
Mark M. Turnbull*d and Spyros P. Perlepes*a,b
The reactions of copper(II) carboxylate sources with acetylacetone dioxime (acacdoH2) in Me2CO have
been studied and a novel, metal ion-assisted ligand transformation has been discovered. The reaction of
[Cu2(diba)4(dibaH)2] and acacdoH2 (1 : 1.5) in Me2CO has provided access to the complex
{[Cu2(diba)4(qunx)]}n (1) in low yield (25–30%), where dibaH is 3,3-dimethylbutyric acid and qunx is
quinoxaline. The [Cu2(piv)4(pivH)2]/acacdoH2 (1 : 1.5) reaction system in warm Me2CO, where pivH is pivalic
acid, gave the analogous complex {[Cu2(piv)4(qunx)]}n (2) in moderate yield (∼50%). Complexes 1 and 2
can be easily prepared by the direct 1 : 1 reactions between the corresponding copper(II) carboxylate start-
ing materials and qunx in Me2CO and MeOH, respectively. The formation of coordinated qunx in 1 and 2
is CuII-promoted (assisted) as suggested by the failure to synthesize the free qunx by a variety of reactions
of acacdoH2 and Me2CO under aerobic conditions in the absence or even the presence of dibaH and
pivH, respectively. The observed acacdoH2 → qunx transformation is catalytic and new in the chemistry of
the dioximes of β-diketones, and a mechanism has been proposed based on well-established reactions
of organic chemistry. The mechanism is based on a double Beckmann rearrangement-type trans-
formation and the overall scheme is represented by the 1 : 1 : 1 reaction between acacdoH2, Me2CO and
O2. Complexes 1 and 2 have similar molecular structures consisting of paddle-wheel {Cu2(η1:η1:μ-O2CR)4}
units bridged by qunx ligands in a zigzag 1D chain arrangement. The geometry of the CuII ions is square
pyramidal with a quinoxaline nitrogen atom occupying the apical position at each metal ion. Weak
H bonds are present within the chains, the donors being qunx carbon atoms and the acceptors being co-
ordinated carboxylate oxygen atoms. Neighbouring chains interact through C–H⋯π interactions between
diba−/piv− methyl groups and the “pyrazine” part of qunx forming layers which are stacked along the b (1)
or a (2) axis through weak van der Waals interactions. The packing of the layers is different in the two
structures, due to the different nature of the carboxylate ligands. Hirshfeld surface analysis of the two
structures reveals the similarity of the interchain (intralayer) interactions. The IR and Raman data of 1 and 2
are discussed in terms of the coordination mode of the carboxylate groups and permit assignments of
some characteristic bands/peaks of coordinated qunx. Dc magnetic susceptibility studies in the 1.8–310 K
range reveal very strong antiferromagnetic CuII⋯CuII exchange interactions within the carboxylate-
bridged Cu2 units (J = −479 K for 1 and −532 K for 2 using the H = − J∑S1·S2 spin Hamiltonian) and
weaker antiferromagnetic interactions between the Cu2 units via the qunx superexchange pathways, with
Received 15th September 2016,
Accepted 28th November 2016
DOI: 10.1039/c6dt03595g
www.rsc.org/dalton
aFoundation for Research and Technology- Hellas (FORTH), Institute of Chemical
Engineering Sciences (ICE-HT), Platani, P.O. Box 1414, 26504 Patras, Greece.
E-mail: gvog@iceht.forth.gr; Tel: +30 2610 965253
bDepartment of Chemistry, University of Patras, 26504 Patras, Greece.
E-mail: perlepes@patreas.upatras.gr; Tel: +30 2610 996730
†This work is dedicated to Professor Jan Reedijk for his great contributions
to inorganic chemistry over the last 45 years and on the occasion of his
retirement.
‡Electronic
supplementary
information
(ESI)
available:
Various
structural plots (Fig. S1–S10), tables with selected interatomic
cInstitute of Nanoscience and Nanotechnology, NCSR “Demokritos”, 153 10 Aghia
Paraskevi Attikis, Greece. E-mail: v.psycharis@inn.demokritos.gr; Tel: +30 210 6503346
dCarlson School of Chemistry and Biochemistry, Clark University, Worcester, MA,
USA. E-mail: MTurnbull@clarku.edu; Tel: +1-508-7937167
distances and angles (Tables S1 and S2) and IR spectra (Fig. S11) of
complexes
1
and 2. CCDC 1503220 and 1503221. For ESI and
crystallographic data in CIF or other electronic format see DOI: 10.1039/
c6dt03595g
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
the latter being ∼10% in strength compared to the former. A critical discussion of the acacdoH2 → qunx
transformation in 1 and 2 is provided in the light of other impressive, recently discovered CuII-assisted
transformations of acacdoH2, pointing out the key role of the solvent in the processes known to date.
bearing carbonyl (ketones and aldehydes),6,7 or oxime8,9
groups. We have observed that the reactivity at various sites in
Introduction
The term “ligand” (Latin ligare, to bind) was first proposed by an organic ligand may be enhanced or diminished upon
Alfred Stock when lecturing in Berlin on borane and silane in coordination, and that very subtle control over the organic
1916,1 but it came into extensive use among English-speaking chemistry of the ligand may be exerted.
chemists in the 1950s, largely through the popularity of Jannik
Restricting further discussion to the reactivity of co-
Bjerrum’s PhD thesis.2 The proper use of old ligands and the ordinated oximes,8–10 oxime and oximato groups can bind a
design of new ones underlie much of modern molecular in- metal ion in various coordination modes (Scheme 1) exhibit-
organic chemistry. Classical coordination chemistry employed ing versatile reactivity. In contrast to many functional groups
simple inorganic and organic ligands, e.g. H2O, Cl−, NH3, en, (e.g. amine, thiol, nitrile, carbonyl, nitrosyl, etc.), a coordinated
dien, py, etc., in the 1950–1970 period. However, the use oxime/oximate contains three active sites (C, N and O atoms).
of polydentate ligands with a variety of functional groups, Many types of metal-mediated transformations are possible for
e.g. macrocyclic ligands, crown ethers, cryptands, sepulchrates the coordinated oximes/oximates, e.g. nucleophilic reactions,
and ligands of unusual reach, which are designed to achieve electrophilic reactions, template synthesis, Beckmann
specific purposes, continues to be an important research rearrangements, intramolecular redox reactions, etc. Retention
frontier in contemporary coordination chemistry.3 Theoretical or loss of integrity of the CvNO(H) fragment during coordi-
concepts related to such and other specialized organic ligands nation is the general characteristic used for the classification
are the chelate effect, the macrocyclic effect, the conformation of the metal-mediated reactions of oximes.10c,d
of chelating rings, the isoelectronic and isolobal relationships
and some aspects of the HSAB model.
Most studies on the reactivity of coordinated oximes have
been performed with ligands containing a single oxime
While metal complexes are most often synthesized by the group,10 e.g. 2-pyridyl oximes.8,9 With our previous experience
reaction of a metal precursor and a presynthesized organic as a background, we have embarked on a new project concern-
ligand which is incorporated intact in the product, it is also ing the in-depth study of the metal ion-assisted/mediated reac-
possible to generate the ligand in situ from an easily available tions of dioximes and we started our efforts with 1,3-di-
organic compound. This synthetic approach allows the reactiv- oximes11 which present a rather rich reactivity in organic
ity of the metal ion to activate a proligand, transforming it chemistry. We selected acetylacetone dioxime (pentane-2,4-
through an in situ reaction, sometimes providing coordination dione dioxime or N,N′-dihydroxypentane-2,4-diimine), abbre-
complexes with ligands that cannot be easily prepared by con- viated hereafter as acacdoH2 (Scheme 2), as the first ligand for
ventional organic synthesis.4 The change from a lone pair to a our studies. The coordination chemistry of acacdoH2 was
bonding pair, consequent upon the formation of the coordi- practically unknown before 2014 and the only complexes
nate bond, has effects on the other atoms or groups, which are that had been crystallographically characterized were the
bonded to the ligand donor atom(s). These effects are the tetranuclear titanium(IV) clusters [Ti4(OiPr)8(acacdo)4] and
origin of the modification of ligand reactivity that is observed [Ti4O2(OiPr)8(acacdo)2].12 During our first reactions of
in coordination complexes. Nowadays, the altered reactivity of acacdoH2 and copper(II) sources, we came across an exciting
organic ligands as a consequence of their coordination to a report by Papaefstathiou’s group which described the struc-
metal centre attracts the intense interest of inorganic tures of [Cu2Cl4(acacdoH2)2] (we had also obtained this di-
II
16
chemists, because it forms the basis for the use of metal com- nuclear complex) and an aesthetically beautiful {Cu } wheel
plexes as stoichiometric reagents and also as homogeneous
catalysts in organic synthesis. Polarization of a substrate co-
ordinated to a metal ion is also one of the most obvious, and
best investigated, means by which the reactivity of coordinated
ligands may be modified in biochemistry. Biological systems
have made extensive use of these processes, and Zn2+ plays an
important role in this function. Typical examples include the
polarization of a coordinated biological ligand and its acti-
vation towards attack by a nucleophile such as H2O or a H−
donor (such as NADH), or the polarization of a H2O ligand
and subsequent stabilization of a coordinated OH− ion.5
In the last 20 years or so, we have developed an understand-
Scheme 1 Crystallographically established coordination modes of the
ing of the many ways in which coordination to a metal ion
may modify and control the reactivity of organic molecules
oxime and oximate groups; M is a metal ion and the dashed line
represents a strong hydrogen bond.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
noxaline (qunx, Scheme 2) in acetone. Details of this transform-
ation and the resulting 1D, carboxylate and qunx-containing
copper(II) coordination polymers are described in this work.
Experimental section
Materials and physical measurements
All manipulations were performed under aerobic conditions
using materials (reagent grade) and solvents as received. The
ligands acetylacetone dioxime (acacdoH2) and quinoxaline
(qunx) are commercially available. The copper(II) starting
materials [Cu2(piv)4(pivH)2]16 and [Cu2(diba)4(dibaH)2],17
where pivH is pivalic acid and dibaH is 3,3-dimethylbutyric
acid (tert-butylacetic acid) were synthesized as described in the
literature. Elemental analyses (C, H, N) were performed by the
University of Patras microanalytical service. FT-IR spectra
(4000–400 cm−1) were recorded using a Perkin Elmer 16PC
FT-IR spectrometer with samples prepared as KBr pellets. The
backscattering Raman spectra were obtained using the
T-64000 model of Jobin Yvon (Horiba group) excited with a
Cobolt Fandango™ ISO, operating at 515 nm. The laser inten-
sity on the samples was 0.4 mW. Dispersion and detection of
the Raman phonons are done by using a 600-groove per nm
grating and a 2D CCD detector (liquid nitrogen-cooled at
140 K), respectively. The spectrometer was calibrated using the
standard 1001 cm−1 Raman peak position of polystyrene.
Electronic absorption spectra (both in solution and in the
solid state) were recorded with a Perkin Elmer Lambda
950 UV/VIS spectrophotometer in the 250–800 nm range.
Variable-temperature, solid state magnetic susceptibility data
were collected using a Quantum Design MPMS-XL SQUID mag-
netometer on samples of finely ground crystals of the com-
plexes, which were loaded into gelatin capsules, in the
1.8–310 K range. Magnetization data were collected as a func-
tion of field from 0 to 50 kOe at 1.8 K. The moment for the
pivalate compound 2 (vide infra) approached saturation at 50
kOe with a moment of 55 emu mol−1 suggesting that the only
contributions to the moment at that temperature arise from a
trace paramagnetic impurity. The tert-butylacetate copper(II)
complex 1 (vide infra) behaved similarly with a moment of
61 emu mol−1 at 50 kOe (at 1.8 K). Magnetic susceptibility data
were then collected from 1.8 to 310 K in a 1 kOe applied field.
All data were corrected for the background signal of the
sample mounting materials (measured independently), for the
diamagnetic contributions of the constituent atoms as
estimated via Pascal’s constants and for the temperature-inde-
pendent paramagnetism of the CuII ions. The data were fit
using the Hamiltonian H = −J∑S1·S2.
Scheme 2 Ligands discussed in the text.
with the formula [Cu16(L1)4(L2)8(L3)8]·8H2O.13 The latter com-
prises three different ligands, namely (L1)2−, (L2)2− and (L3)−,
Scheme 2, all derived from acacdoH2; it is remarkable that the
original ligand is completely absent from the complex. The
II
{CuI2I} and {Cu } complexes were prepared using EtOH and
16
EtOH/H2O as solvents, respectively. Although the transform-
ation of acacdoH2 to (L1)2−, (L2)2− and (L3)− was not anti-
cipated, the authors were able to fully rationalize the for-
mation of these anions. Methylene moieties (–CH2–) attached
to electron withdrawing groups can be aerially oxidized to the
corresponding ketones with14 or without the presence of metal
ions.14b The dianion of the oxidized form of acacdoH2, (L2)2−
,
is probably metal ion-assisted since the 1H NMR spectra of
pure acacdoH2 in D2O or CD3OD remain unchanged for
several months. Ketones, like (L2)2−, may undergo nucleophilic
addition to the carbonyl C atom to form the respective hydrate
(the dianion of the gem-diol form), (L1)2−. The electrophilic
character of the carbonyl C atom is increased by coordination
of the carbonyl O atom (direct polarization) or by coordination
of the remote
N or O atoms (induced polarization).
Alternatively, acacdoH2 may be first oxidized to H2L1 which
upon dehydration forms the ketone form H2L2. Pyrazole
N-oxides, like (L3)−, have been previously reported to form by
metal ion-assisted transformations of β-diketone dioximes.11
On studying the old literature more, we found another
example of
a metal ion-assisted reaction of acacdoH2.
The CuCl2·2H2O/acacdoH2 reaction system in MeOH/H2O
(1 : 20 v/v) under heating leads to CuII → CuI reduction and
gives the polymeric complex {[CuI2Cl2(L4)]}n, where L4 is the
3,5-dimethylisoxazole ligand.15 The authors proposed that the
formation of L4 occurs through H2NOH elimination and CuII
reduction, associated with, or followed by, ring closure. Since
the presence of LnIII ions (Ln = Nd, Gd, Er) in the reaction
mixtures has been essential for the isolation of the hexadeca-
nuclear cluster (although the LnIII ion does not appear in the
Synthetic details
{[Cu2(diba)4(qunx)]}n (1). Method A. A colourless slurry of
final product) and given the interesting acacdoH2 → L4 trans- acacdoH2 (0.039 g, 0.30 mmol) in Me2CO (2 mL) was slowly
formation promoted only by copper, we started our efforts added to a pale greenish blue solution of [Cu2(diba)4(dibaH)2]
using exclusively copper(II) and were pleased to discover a new (0.164 g, 0.20 mmol) in the same solvent (9 mL). The resulting
CuII-assisted transformation of acacdoH2 to coordinated qui- green-blue solution was stirred for a further 20 min at room
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
temperature and filtered. Upon vapour diffusion of Et2O, the
Method B. A green solution containing [Cu2(piv)4(pivH)2]
colour of the reaction solution turned dark green. X-ray (0.086 g, 0.10 mmol) and qunx (0.013 g, 0.10 mmol) in MeOH
quality, green crystals of the product were formed in a period (20 mL) was refluxed for 2 h. No noticeable colour change
of 5–6 d. The crystals were collected by filtration, washed with occurred, but a very small quantity of a green powder was
cold Me2CO (2 mL) and Et2O (5 × 2 mL), and dried in vacuo precipitated during the reflux. The green solid was removed by
over CaCl2. Typical yields were in the 25–30% range (based on filtration and the filtrate was left undisturbed in a closed flask
the CuII available). Analytical data, calcd for C32H50Cu2N2O8 at room temperature. Dark green crystals of the product were
(found values are in parentheses): C 53.54 (53.85), H 7.04 formed over a period of 4 d. The crystals were collected by
(7.18), N 3.90 (3.77)%. IR bands (KBr, cm−1): 2962m, 2940sh, filtration, washed with cold MeOH (1 mL) and Et2O (2 × 5 mL),
2910w, 1614s, 1504w, 1463sh, 1410m, 1362m, 1304w, 1280w, and dried in vacuo over CaCl2. The yield was 48% (based on
1238w, 1218m, 1140w, 1044m, 964m, 866m, 802w, 758m, the CuII available). Analytical data, calcd for C28H42Cu2N2O8
734m, 655m, 460w, 422w. Raman peaks (cm−1): 3076w, 2957s, (found values are in parentheses): C 50.82 (50.33), H 6.41
2928w, 2908m, 2866sh, 1589m, 1536w, 1510w, 1446w, 1433m, (6.64), N 4.23 (4.31)%. The finely ground crystals had IR,
1379m, 1362s, 1296m, 1239w, 1213w, 1148w, 1048w, 1021w, Raman and solid-state UV/VIS spectra identical to those of the
966w, 937w, 914w, 771s, 750m, 540m, 420m, 369m, 234m, authentic material 2 prepared by method A.
199m. Solid-state UV/VIS bands (nm): 645, 399, 335, 260.
UV/VIS bands in MeOH (nm): 635wb, 321, 310, 255.
Reaction of [Cu2(diba)4(dibaH)2] with an excess of acacdoH2.
A colourless slurry of acacdoH2 (0.156 g, 1.20 mmol) in Me2CO
Method B.
A
pale greenish blue solution of (5 mL) was slowly added to a pale greenish-blue solution of
[Cu2(diba)4(dibaH)2] (0.164 g, 0.20 mmol) in Me2CO (9 mL) [Cu2(diba)4(dibaH)2] (0.164 g, 0.20 mmol) in the same solvent
was added to a stirred yellow solution of qunx (0.026 g, (6 mL). The resulting green-blue solution was stirred for a
0.20 mmol) in the same solvent (2 mL). The resulting dark further 20 min at room temperature. Upon vapour diffusion of
green solution was stirred for a further 20 min, filtered to Et2O, the colour of the reaction solution became dark green.
remove a very small quantity of a pale green solid and the Green crystals were formed over a period of 4–5 d. The crystals
filtrate was stored in a closed flask at room temperature. X-ray were collected by filtration, washed with cold Me2CO (1 mL)
quality, rod-like green crystals of the product were precipitated and Et2O (1 mL), and dried in air. The green material proved
after 2 d. The crystals were collected by filtration, washed with to be compound 1 by IR spectroscopic comparison with the
cold Me2CO (1 mL) and Et2O (2 × 4 mL), and dried in vacuo authentic material (vide supra). The yield was 35% (based on
over CaCl2. The yield was 45% (based on the CuII available). CuII). Analytical data, calcd for C32H50Cu2N2O8 (found values
Analytical data calcd for C32H50Cu2N2O8 (found values are in are in parentheses): C 53.54 (53.03), H 7.04 (6.87), N 3.90
parentheses): C 53.54 (53.65), H 7.04 (7.40), N 3.90 (3.93)%. (4.11)%. The green filtrate from the above procedure was
The finely ground crystals had IR, Raman and solid-state UV/ stirred overnight at room temperature. The resulting green
VIS spectra identical to those of the authentic sample prepared solution was slowly filtered through a plug of silica gel
by method A.
(Davisil, Grade 634, 1.0 cm of silica gel into a porous glass
{[Cu2(piv)4(qunx)]}n (2). Method A. A colourless slurry of funnel); after the addition of the filtrate on the funnel, the
acacdoH2 (0.039 g, 0.30 mmol) in Me2CO (3 mL) was slowly solution was allowed to stand undisturbed for 2 min and then
added to a green solution of [Cu2(piv)4(pivH)2] (0.147 g, the vacuum was turned on. This procedure was repeated and
0.20 mmol) in the same solvent (8 mL). Upon slight warming the colour of the final filtrate was yellow. Evaporation of the
(35 °C) and vigorous stirring a bluish green solution was solvents (Me2CO, Et2O) in a rotary evaporator at room tempera-
obtained, which was stirred for a further 10 min, filtered and ture gave a pale brown oily solid; the vacuum had to be turned
layered with Et2O (10 mL). Slow mixing gave a green solution off shortly after the removal of the solvents because the solid
from which were subsequently isolated dark green crystals of is rather volatile. The residue was treated with petroleum
the product. The crystals were isolated by filtration, washed spirits, the resulting mixture was filtered, and the collected
with cold Me2CO (1 mL) and Et2O (2 × 2 mL), and dried in air. solid was washed with cold Et2O (2 × 1 mL) and dried in air to
Typical yields were in the 40–45% range (based on the CuII give ∼0.14 g (95% based on the acacdoH2 used) of a pale
available). Analytical data, calcd for C28H42Cu2N2O8 (found yellow solid. Careful IR and Raman studies of the product
values are in parentheses): C 50.82 (52.95), H 6.41 (6.59), proved that the solid was free qunx with slight contamination
N 4.23 (4.09)%. IR bands (KBr, cm−1): 2966sh, 2954m, by acacdoH2. Raman peaks (cm−1): 1369s, 760s, 604m, 405m.
2940sh, 2918w, 1602s, 1505w, 1488m, 1454w, 1422s, 1359m,
Reaction of CuBr2 with an excess of acacdoH2. A brown solu-
1354m, 1224m, 1142w, 1040w, 964m, 894w, 866m, 790m, tion of CuBr2 (0.045 g, 0.20 mmol) in Me2CO (10 mL) was
754m, 618m, 442m, 414w. Raman peaks (cm−1): 3102w, slowly added to a colourless slurry of acacdoH2 (0.130 g,
3064w, 2985w, 2960w, 2924m, 2874w, 1589m, 1532w, 1487w, 1.00 mmol) in the same solvent (10 mL). The resulting green-
1455w, 1422m, 1379s, 1364s, 1296m, 1213m, 1146w, 1048w, ish brown solution was stirred in a closed flask at room temp-
1023w, 941w, 895w, 868w, 808s, 775s, 637w, 540m, 441m, erature for 10 min. The resulting pale green-brown solution
415w, 325w, 214s, 192s, 157w. Solid-state UV/VIS bands (nm): was left undisturbed for 4–5 d, during which time a colour
652, 410, 330, 290. UV/VIS bands in Me2CO (nm): 645wb, change to greenish yellow and precipitation of a yellow solid
315, 310, 305.
was observed. The microcrystalline solid was collected by
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Table 1 Crystallographic data for compounds 1 and 2
filtration, washed with cold Et2O (2 × 1 mL) and dried in air.
The yellow material proved to be the known compound
{[CuBr2(qunx)]}n by microanalyses and IR spectroscopic
comparison with the authentic material prepared as
described in the literature18 by the direct reaction of CuBr2
and qunx. The yield was 78% (based on the CuII available).
Analytical data, calcd for C8H6CuN2Br2 (found values are in
parentheses): C 27.18 (27.79), H 1.71 (1.60), N 7.93 (7.45)%.
The greenish yellow filtrate from the above mentioned
procedure was stirred overnight at room temperature and
slowly filtered through a plug of silica gel (Davisil, Grade 634,
1.0 cm of silica gel into a porous glass funnel); after the
addition of the filtrate on the funnel, the solution was
allowed to stand undisturbed for 2 min and then the vacuum
was turned on. Evaporation of the solvent (mainly Me2CO) in
a rotary evaporator at room temperature gave a yellow solid;
the vacuum had to be turned off shortly after the removal of
the solvent because the solid is rather volatile. The residue
was treated with petroleum spirits, the resulting mixture was
filtered, and the collected solid was washed with cold Et2O
(3 × 1 mL) and dried in air to give ∼0.10 g (92% based on
the acacdoH2 used) of a pale yellow solid. Careful IR and
Raman studies of the product proved this to be free qunx with
negligible contamination by acacdoH2. Raman peaks (cm−1):
1370s, 759s, 604m, 406m.
Parameter
1
2
Formula
Fw
C32H50Cu2N2O8
717.82
C28H42Cu2N2O8
661.71
Orthorhombic
Pna21
21.2259(4)
17.2370(3)
9.3165(2)
90.0
90.0
90.0
3408.64(11)
4
1.289
Crystal system
Space group
Triclinic
P1
ˉ
a/Å
b/Å
c/Å
a/°
9.9649(2)
11.0985(2)
17.6722(3)
71.044(1)
82.983(1)
73.748(1)
1773.56(6)
2
1.344
160(2)
Cu Kα/1.875
18 264/5724
(0.0456)
4789
β/°
γ/°
V/Å3
Z
ρcalcd/g cm−3
T/K
293(2)
Radiation/μ (mm−1
)
Cu Kα/1.908
16 260/4009
(0.0629)
2559
374
0.0594/0.1744
1.133
Reflns collected/unique (Rint
)
Reflns with I > 2σ(I)
No. of parameters
R1[I > 2σ(I)], wR2 (all data)
GOF(F2)
520
0.0406/0.113
1.096
0.630/−0.573
1503220
Δρmax/Δρmin (e Å−3
)
0.544/−0.861
1503221
CCDC number
Results and discussion
Synthetic comments
Our approach to activate the proligand acetylacetone dioxime
for further reactivity was to treat a copper(II) starting material
Single-crystal X-ray crystallography
A suitable single-crystal of 1 (0.14 × 0.33 × 0.43 mm) was taken with acacdoH2. Our initial efforts were with CuX2 (X = Cl, Br),
from the mother liquor and immediately cooled to −113 °C; a Cu(ClO4)2·6H2O and Cu(NO3)2·6H2O using alcohols or/and
crystal of 2 (0.04 × 0.37 × 0.50 mm) was mounted in air. All the MeCN as solvents and various crystallization techniques.
crystallisation efforts for 2 gave poorly diffracting crystals Powders were mostly obtained which could not be character-
which were very thin plates. Finally, data collection was ized, the only crystalline compound obtained being
performed at room temperature. X-ray diffraction data for [Cu2Cl4(acacdoH2)]. Then we came across the very interesting
1
and
2
were collected on
a
Rigaku R-AXIS SPIDER communication by Papaefstathiou’s group which reported the
II
Image Plate diffractometer using graphite-monochromated chloro-bridged copper(II) dimer and the exciting {Cu } wheel
16
Cu Kα radiation. Data collection (ω-scans) and processing mentioned in Introduction; the latter contains the trans-
(cell refinement, data reduction and empirical absorption formed ligands (L1)2−, (L2)2− and (L3)− shown in Scheme 2.
correction) were performed using the CrystalClear program This report convinced us that there was a rich reactivity of the
package.19 The structures were solved by direct methods coordinated acacdoH2. In order to explore further reaction
using SHELXS-9720a and refined by full-matrix least-squares pathways for this ligand, we changed the nature of the
techniques on F2 with the latest version of SHELXL (2014/7).20b metal starting materials and the solvent. Treatment of
For 1 the H atoms were refined isotropically at positions [Cu2(diba)4(dibaH)2] with acacdoH2 (1 : 1.5) in Me2CO at
located either from difference Fourier maps or at the room temperature gave a green solution from which were
calculated ones using a riding model. The structure of 2 is subsequently isolated green crystals of {[Cu2(diba)4(qunx)]}n (1)
not centrosymmetric and it was refined as an inversion in rather low yields (25–30%) upon vapour diffusion with Et2O.
twin (the Flack parameter takes the value of 0.44). For 2 all The crystals were of X-ray quality and single-crystal X-ray crys-
H atoms were introduced at the calculated positions as tallography revealed the identity of the product and the novel
riding on their respective bonded atoms. All non-H atoms acacdoH2 → qunx transformation. We do believe that the for-
for both complexes were refined anisotropically. Plots of the mation of qunx in 1 (and in 2, vide infra) is CuII-promoted
structures were drawn using the Diamond
3 program (or assisted). To date, we have failed to synthesize the free
package.21 The packing of the complexes in the crystal qunx by reactions of acacdoH2 and Me2CO (stoichiometric
structures was further investigated using the CrystalExplorer amount or excess) under aerobic conditions in the absence or
1
package.22
even in the presence of dibaH. In addition, the H NMR spec-
Important crystallographic data are listed in Table 1. Full trum of pure acacdoH2 in (CD3)2CO remains unchanged for
details can be found in the CIF files.
several weeks. Since the acacdoH2 transformations in the
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
II
16
{Cu } wheel had been observed in the presence of several in which the carbocation is intercepted by an internal O atom.
lanthanide(III) ions,13 we repeated the procedure described in It is well known that the carbocation may combine with a
the Experimental section (method A) by also employing species possessing an electron pair (a Lewis acid–base reac-
Dy(NO3)3·5H2O in a [Cu2(diba)4(dibaH)2] : DyIII reaction ratio of tion). This species may be OH−, halide ion, or any other nega-
2 : 1. Somewhat to our surprise, crystals of 1 (unit cell determi- tive atom (as in the present case), or it may be a neutral
nation) were again precipitated in a somewhat lower yield species with an electron pair to donate, in which case, of
(∼20%). Thus, it seems that the acacdoH2 → qunx transform- course, the immediate product must bear a positive charge.23
ation is assisted by either CuII or a CuII–DyIII combination. The intermediate VI, formed after two cycles of Beckmann
Dy(NO3)3·5H2O/acacdoH2 and Dy(NO3)3·5H2O/acacdoH2/dibaH rearrangement and carbocation interception, gives the new
reaction systems in Me2CO have failed to date to give any intermediate XI, most probably through the formation of X.
characterisable solid.
The X → XI process is a [1,3] sigmatropic migration of hydro-
Having structurally characterized compound 1, we gen.23,24 Several examples of thermal or photochemical
wondered whether this 1D coordination polymer could be rearrangements in which a hydrogen atom migrates from one
prepared by the direct reaction between [Cu2(diba)4(dibaH)2] end of a system of a π bond to the other have been reported,
and qunx in Me2CO and this turned out to be the case. The though the reaction is subject to geometrical conditions.
1 : 1 reaction (CuII : qunx = 2 : 1) gave crystals of 1 in a ca. 50% Pericyclic mechanisms are involved, and the hydrogen atom
yield, eqn (1).
must, in the transition state, be in contact with both ends of
the chain at the same time. Furthermore, there are two geo-
metrical pathways by which any sigmatropic rearrangement
can take place. In one of the two pathways, the hydrogen
moves along the top or bottom of the π system (suprafacial
migration). In the other pathway, the hydrogen moves across
Â
Ã
n Cu2ðdibaÞ4ðdibaHÞ2 þ n qunx
Me2CO
ð1Þ
ꢀꢀꢀ! f½Cu2ðdibaÞ4ðqunxÞꢁgn þ 2n dibaH
1
In a next step we tried to investigate the influence of the the π system from top to bottom or vice versa (antarafacial
carboxylate on the transformation and product identity. migration). The suprafacial [1,3]-hydrogen sigmatropic
The [Cu2(piv)4(pivH)2] starting material led to completely rearrangement is forbidden by orbital symmetry. The antarafa-
analogous results with [Cu2(diba)4(dibaH)2]. Thus, the cial [1,3]-hydrogen sigmatropic rearrangement is allowed by
[Cu2(piv)4(pivH)2]/acacdoH2 (1 : 1.5) reaction mixture in orbital symmetry; however, this type of rearrangement is geo-
Me2CO (the solvent had to be warmed to facilitate the reaction) metrically impossible due to the inability to maintain the
gave a blue-green solution from which were subsequently iso- overlap of the 1s atomic orbital of the migrating hydrogen
lated green crystals of {[Cu2(piv)4(qunx)]}n (2) in moderate atom with the necessary lobes of the terminal 2p atomic orbi-
yields (∼50%) upon liquid diffusion with Et2O. As in the case tals at the midpoint of the migration.24a In the present case
of 1, compound 2 can be easily prepared by the direct 1 : 1 the migration is facilitated because the double bonded carbon
reaction between [Cu2(piv)4(pivH)2] and qunx in a ca. 50% atom to which the hydrogen atom migrates also participates in
yield. The product obtained in Me2CO repeatedly gave analyti- a second double bond (cumulative double bonds).23 XI is
cal results slightly outside the acceptable range. The preferable transformed into the new intermediate XII through an electro-
solvent for its preparation is MeOH. The reaction solution had cyclic rearrangement (such rearrangements take place by peri-
to be refluxed to precipitate a minor quantity of a byproduct cyclic mechanisms), which involves redistribution of π bonds
whose IR spectrum is similar (but not identical) to that of 2. and the formation of a new C–C σ bond.23 The intermediate
The reactions of acacdoH2 and other copper(II) carboxylate XII undergoes a conjugate addition by the coordinated enolate
sources, e.g. [Cu2(O2CMe)4(H2O)2] and Cu(O2CPh)2·xH2O, in species (or simply by the free enolate ion) to give XIII; the con-
Me2CO gave green powders that we could not crystallize; jugate addition (also called Michael addition) is a typical 1,4-
however their IR spectra show the typical bands of the co- addition to the carbon–carbon double bond.23 Step viii in
ordinated qunx.
Scheme 3 is an oxidative aromatization of the heterocyclic ring
The observed acacdoH2 → qunx transformation is new in which leads to XIV. Generally, six-membered alicyclic rings can
the chemistry of 1,3-dioximes.11,13,15 A proposed mechanism, be aromatized in a number of ways.25 The reaction can also be
based on well-established reactions in organic chemistry,23 is applied to heterocyclic 5- and 6- (as in the present case) rings.
shown in Scheme 3. The overall reaction is illustrated in Aromatization is accomplished most easily if there are already
Scheme 4. The mechanism is based on a double Beckmann one or two (as in the case of XIII) double bonds in the ring.23
rearrangement-type reaction. The CuII ion ensures the correct Among other reagents that have been used are atmospheric
configuration of the groups during the double Beckmann oxygen (the case here), H2O2, MnO2, SeO2, various strong
rearrangement which demands anti conformation of the bases and chromic acid. The intermediate XIV undergoes a
leaving (H2O) and migrating alkyl groups (CH2R at the begin- nucleophilic aromatic substitution and, after dehydration, the
ning and CH3 later on).
dihydroquinoxaline XV is formed. The nucleophilic aromatic
Steps i and ii in Scheme 3 represent Beckmann rearrange- substitution most probably proceeds through an SNAr mechan-
ments. The group that migrates is the one anti to the hydroxyl ism and is facilitated by the presence of the strongly activated
group.23 Each Beckmann rearrangement is followed by a step hetero nitrogen atoms at the α and γ positions relative to the
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Scheme 3 A proposed mechanism that leads to the acacdoH2 → qunx transformation. Important steps are: (i) 1st Beckmann rearrangement;
(ii) interception of the carbocation by an internal O atom; (iii) 2nd Beckmann rearrangement; (iv) interception of the carbocation by an internal
O atom; (v) 1,3-hydride shift (sigmatropic rearrangement); (vi) electrocyclic reaction; (vii) conjugate addition; (viii) oxidative aromatization;
(ix) nucleophilic aromatic substitution; (x) dehydration; (xi) oxidative aromatization. A coordination number of four (4) is arbitrarily assigned to the
copper ion. The vacant coordination sites might be filled by solvent molecules. The dashed line represents a H bond. XH = dibaH, pivH.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
In the case of acacdoH2, the “normal” double Beckmann
rearrangement would be expected to lead to the
N-acetylglycine methylamide product XXII (Scheme 5).
The presence of this product as an intermediate in the
acacdoH2 → qunx transformation could not be ruled out.
Work is in progress in our laboratories to synthesize XXII and
to investigate whether its reaction with [Cu2(diba)4(dibaH)2]
and [Cu2(piv)4(pivH)2] in Me2CO would give compounds 1 and 2.
Another question that arises from our results is the follow-
ing: “Is the acacdoH2 → qunx transformation catalytic or stoi-
chiometric?” We have strong evidence that the reaction is cata-
lytic.27 As detailed in the Experimental section, the reaction of
[Cu2(diba)4(dibaH)2] with a large excess of acacdoH2 in Me2CO
at room temperature gives complex 1 and free qunx with
almost no signs of contamination by acacdoH2. After removal
of the precipitated amount of 1 by filtration, the coloured fil-
trate was passed through silica gel in which the CuII content
(from the soluble quantity of the complex) remained bound.
Evaporation of the filtrate gave free qunx. The catalytic role of
CuII in the transformation was also shown27 by the reaction of
CuBr2, i.e. a non-carboxylate copper(II) source, with a large
excess of acacdoH2 in Me2CO at room temperature which gave
Scheme 4 The overall reaction that takes place during the preparation
of the 1D coordination polymers 1 and 2. RCOOH = dibaH, pivH.
leaving group.23 Finally, qunx arises from XV through oxidative
aromatization (vide supra) by H2O2 which has been produced
in the previous step of the sequence. We want to emphasize at
this point that we have not investigated the mechanism of the
acacdoH2 → qunx transformation by solution methods (kine-
tics, UV/VIS, EPR, etc.). Due to the complicated nature of the
transformation, it is doubtful if such studies will provide any
reliable mechanistic information. In addition, the paramag-
netic nature of copper(II) precludes the use of NMR spec-
troscopy which might be more useful. We do believe that the
proposed mechanism is not entirely speculative and the
sequences shown in Scheme 3 represent a logical mechanistic
approach in which the role of CuII is to activate groups by
coordination, facilitating their reactions.
18
the known complex {[CuBr2(qunx)]}n and almost pure qunx;
full details are given in the Experimental section.
The rearrangement of ketoximes to the corresponding
amides, known as the Beckmann reaction or rearrangement, is
a common method in organic chemistry and a topic of current
interest. It accomplishes both the cleavage of a carbon–carbon
bond and the formation of a carbon–nitrogen bond.26 It rep-
resents a powerful method particularly for manufacturing
ε-caprolactam in the chemical industry. The reaction generally
requires high reaction temperatures and dehydrating media.
Thus, the reaction can lead to large amounts of byproducts
and precludes its application to sensitive substrates. Several
catalysts, including metal ions, have been employed for this
reaction.
Description of structures
The aspects of the molecular and crystal structures of 1 and 2
are shown in Fig. 1–4 and S1–S7.‡ Selected interatomic dis-
tances and angles are listed in Tables S1 and S2.‡
Compound 1 contains two crystallographically independent
paddle-wheel {Cu2(diba)4} units (secondary building units,
SBUs), each at an inversion centre, and these dimers are linked
into one-dimensional (1D) chains parallel to the [−1, 0, 1]
direction through the use of qunx as a bidentate bridging
ligand (Fig. S1‡). The asymmetric unit of the cell consists of
two CuII atoms (Cu1, Cu2), four syn,syn-η1:η1:μ diba− groups
Scheme 5 The anticipated “normal” product from the CuII-assisted double Beckmann rearrangement of acacdoH2. Steps (i) and (ii) represent
Beckmann rearrangements. A coordination number of four is arbitrarily assigned to the copper ion. The vacant coordination sites might be filled by
solvent molecules. The dashed line represents a H bond. The intermediate XVII is the same as I in Scheme 3. XH = dibaH, pivH.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Fig. 4 The stacking of two successive layers along the a axis in the
crystal structure of 2. Thick red and thin blue lines are used for the rep-
resentation of the chains in the bottom and top layers, respectively; the
successive bottom and top layers are extended parallel to the [0, −1, 1]
and [0, 1, 1] crystallographic directions, respectively.
Fig. 1
A portion of a chain that is present in the structure of
{[Cu2(diba)4(qunx)]}n (1); only two Cu2 units are shown. The dashed
green lines represent intrachain H bonds. Symmetry operations used to
generate equivalent atoms: (’) −x + 1, −y + 2, −z; (’’) −x, −y + 2, −z + 1.
and one qunx ligand and thus the complex is best formulated
as {[Cu2(diba)4(qunx)]}n. The CuII⋯CuII vectors are not co-
linear but form a zigzag line, with Cu1⋯Cu2⋯Cu2″ and Cu1′
⋯Cu1–Cu2 angles of 160.3(1) and 158.2(1)°, respectively. There
are three distinct CuII⋯CuII separations along the chain which
alternate in an ⋯ABCBABCB⋯ pattern where (using the view-
point of Fig. 1) A = Cu1⋯Cu1′ = 2.609(2) Å, B = Cu1⋯Cu2 =
7.159(2) Å and C = Cu2⋯Cu2′ = 2.611(1) Å. Note that the four
syn,syn bridging tert-butylacetates allow the closest approach
of metal centres with the Cu1⋯Cu1′ and Cu2⋯Cu2′ distances
being typical of those found with a number of [Cu2(O2CR)4L2]
complexes (2.6–2.7 Å), where L is a monodentate ligand.28
The geometry of Cu1 and Cu2 can be described as square
pyramidal with a quinoxaline nitrogen atom occupying the
apical position at each metal ion. Analysis of the shape-deter-
mining angles using the approach of Addison, Reedijk et al.29
yields values for the trigonality index, τ, of ∼0 (Cu1/Cu1′) and
0.005 (Cu2/Cu2′) (τ = 0 and 1 for square pyramidal and trigonal
bipyramidal geometry, respectively). Weak H bonds are present
within the chains (intrachain H bonds). The donors are
carbon atoms from the phenyl and diazine parts of qunx (C25,
C28, and C31) and acceptors are the carboxylate oxygen atoms
O4, O8 and O2.
Fig. 2 The formation of a layer parallel to the (010) plane through
C(methyl)–H⋯π(diazine) interactions in 1; the interactions are shown
with dashed orange and dashed magenta lines.
Neighbouring chains communicate through C–H⋯π inter-
actions between the diba− methyl groups bearing C11 and
C17, and the “pyrazine” part of qunx (H11C⋯Cg1 = 2.634 Å
and H17C⋯Cg1 = 3.233 Å, where Cg1 is the centroid of the
diazine ring) forming layers parallel to the (010) plane (Fig. 2,
the stronger C11–H11C⋯Cg1 and the weaker C17–H17C⋯Cg1
interactions are indicated with dashed orange and dashed
magenta lines, respectively). The distance between the neigh-
bouring chains within these layers is 9.104 Å. The layers are
stacked along the b axis, probably interacting through weak
van der Waals interactions (Fig. S2‡). Other views of the supra-
molecular structure of 1 are shown in Fig. S3 and S4.‡
Fig. 3
A portion of a chain that is present in the structure of
{[Cu2(piv)4(qunx)]}n (2); only two Cu2 units are shown. The dashed green
lines represent intrachain H bonds. Symmetry operations used to gene-
rate equivalent atoms: (’) −x + 3/2, y + 1/2, z + 1/2; (’’) −x + 3/2, y − 1/2,
z − 1/2.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
The molecular structure of 2 is similar to that of 1 but the
symmetry elements differ. Complex 2 crystallizes in the non-
centrosymmetric space group Pna21. The asymmetric unit of
the cell consists of two CuII atoms, four piv− groups and one
qunx ligand and thus the complex is best formulated as
{[Cu2(piv)4(qunx)]}n. There is one crystallographically indepen-
dent dinuclear unit consisting of Cu1 and Cu2; these metal
ions are bridged by four, crystallographically independent syn,
syn-η1:η1:μ piv− groups forming a paddle-wheel unit, the
Cu1⋯Cu2 distance being 2.606(2) Å. A nitrogen atom of the
bridging qunx ligand completes five-coordination at each
metal ion. All this part of the structure lies close to the n glide
plane which is normal to the a axis and the result of this sym-
metry element is the formation of polymeric chains parallel to
the [0, 1, −1] direction (Fig. S5‡). As in the case of 1, the
CuII⋯CuII vectors in 2 are not co-linear forming a zigzag line
with Cu1⋯Cu2⋯Cu1″ and Cu2⋯Cu1″⋯Cu2″ of 166.5(4) and
164.2(4)°, respectively. There are two distinct CuII⋯CuII separ-
ations along the chain which alternate according to the
sequence ⋯ABAB⋯ where (using the viewpoint of Fig. 3) A =
Cu1⋯Cu2 = 2.606(2) Å and B = Cu2⋯Cu1″ = 7.274(2) Å. The
coordination geometries of Cu1 and Cu2 are square pyramidal
(τ = 0.023 and 0.047, respectively). The apical site of each
copper(II) square pyramid is occupied by a nitrogen atom of
the bridging qunx ligand. Weak H bonds are present within
the chains. The donor atoms are carbon atoms from the
phenyl and azine parts of qunx (C21, C22, C25, and C28) and
acceptors are the carboxylate oxygen atoms O3, O4, O7 and O8.
Neighbouring chains interact through C–H⋯π interactions
involving the piv− methyl groups bearing C5 and C14 as
donors and the “pyrazine” part of the coordinated qunx as
acceptors forming layers parallel to the (100) plane (Fig. S6,‡
the C–H⋯π interactions are indicated with dashed orange and
magenta lines). The H5A⋯Cg1 and H14A–Cg1 distances are
2.884 and 2.867 Å, respectively, where Cg1 is the centroid of
the diazine ring of qunx. The distance between the neighbour-
ing chains within these layers is 8.196 Å. These layers are
stacked along the a axis as shown in Fig. 4. Fig. S7‡ shows the
portions of the chains of 1 and 2 in an overlapped mode; it is
clear that the chains are similar. Also similar are the layers
formed in the two compounds. However, careful inspection of
Fig. S2‡ and 4 reveals that the packing of the layers, i.e. the
overall 3D arrangement, is different in the two complexes,
most probably due to the different nature of the carboxylate
ligands involved (diba− vs. piv−).
Fig. 5 Hirshfeld surface dnorm maps (upper plots) and parts of the
molecular structures (bottom plots) for (a) compound 1 and (b) com-
pound 2.
Fig. 6 S map for 1. Circled with white lines (a) and black lines (b) are
the contact areas of HS which indicate the C11–H11C⋯Cg1 and C17–
H17C⋯Cg1 interactions, respectively.
Fig. 7 Histograms summarizing the percentages of the types of atomic
contacts, based on 2D fingerprint plots, for 1 and 2.
figures are the S property maps for the compounds and the
circled regions identify complementary hollows (red) and
bumps (blue) where neighbouring molecular HSs touch one
another. Each C–H⋯π interaction is clearly demonstrated on
the S maps as a broad depression (of red colour and concave
curvature) in the surface above the diazine ring and as a
corresponding blue region of the convex curvature above the
C–H donor region.
A simple way to examine the HS is by using the “fingerprint
plot”. The two dimensional fingerprint plot derived from the
HS shows the frequency of occurrence of each combination of
de, di (each such combination corresponds to some kind of
interaction) in the molecular surface; de is the distance from
Hirshfeld surface (HS) analysis30 was applied to evaluate
the magnitude of interchain interactions in the crystals of 1
and 2; data are shown in Fig. 5–8 and S8–S10.‡
A visual inspection of the dnorm surfaces in both 1 and 2
does not indicate any significant intermolecular interactions
(Fig. 5). It should be mentioned at this point that in such an
analysis, the colour scheme corresponds to the magnitude of
intermolecular interactions ranging from strong (red colour) to
moderate (white) to weak (blue). On the contrary, the HSs
decorated with the shape index, S, reveal the importance of the
C–H⋯π interactions in both complexes (Fig. 6 and S8‡). These
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
the neighbouring chains within the layers of compound 1. The
overall conclusion from the HS analysis is the similarity of the
interchain (intralayer) interactions in the two structures.
Compounds 1 and 2 join a handful of quinoxaline-bridged
polymeric copper(II) complexes;18,31 there are no quinoxaline-
bridged carboxylate copper(II) coordination polymers and thus
1 and 2 become the first such complexes. The dinuclear
complex [Cu2(O2CPh)4(qunx)2]32a and the polymeric com-
pound {[Cu(HCO2)2(qunx)]}n 32b contain both qunx and carbox-
ylate ligands, but the qunx group is terminal (i.e. mono-
dentate). Compounds 1 and 2 are also members of a rather
large family of 1D coordination polymers of the general
formula {[Cu2(O2CR)4(L–L)]}n (R various, L–L = N,N′-bidentate
bridging ligand).30d,33–37
Fig. 8 O⋯H/H⋯O (a) and C⋯H/H⋯C (b) contributions (blue areas),
along with the other kinds of contacts (gray area), into the 2D fingerprint
plot for compound 1. For the meaning of the green line, see the text.
Vibrational spectra
the surface to the nearest-neighbouring atom exterior and di is
the distance from the surface to the nearest-neighbouring
atom interior. In a fingerprint plot, the observation and identi-
fication of the pairs of atoms that contribute to a specific de, di
combination reveal the importance of each possible inter-
action type. The percentages of each type of interaction for 1
and 2 are given in Fig. 7. More than 95% of the interactions
are of the H⋯H, O⋯H and C⋯H types in both cases. The van
der Waals-type H⋯H interactions, which contribute ∼80% in
both structures, are weak and provide little stabilization. The
distributions of the types of interactions are slightly different
for the two complexes.
In the analysis of the 2D fingerprint plot, it is important to
discuss the di + de distances of the neighbouring atoms with
values less than the sum of their van der Waals radii. When
these values are small (i.e. the non-bonded interatomic dis-
tances are short), the intermolecular interactions should be
strong.30d Fig. 8 and S9‡ summarise the 2D fingerprint plots
for O⋯H/H⋯O and C⋯H/H⋯C interatomic contacts in the
crystals of 1 and 2, respectively. The green lines correspond to
the di + de distances that are equivalent to the sums of the van
der Waals radii of oxygen (1.52 Å) and hydrogen (1.20 Å),
Fig. 8a and S9a,‡ and of carbon (1.70 Å) and hydrogen (1.20 Å),
Fig. 8b and S9b.‡ It is clear from Fig. 8a and S9a‡ that a small
portion of O⋯H interactions (the part of the fingerprint plot
which overlaps with the triangular area), corresponding to
∼10% of the total interactions, are effective contacts, i.e. the
distances are shorter than the sum of the van der Waals radii;
these H⋯O interactions are of the intrachain type (vide supra).
The other important types of interactions, also contributing
∼10% to the total interactions, consist of the interchain
C–H⋯π interactions; these interactions generate the character-
The Raman and IR spectra of compounds 1 and 2 are shown
in Fig. 9 and S11,‡ respectively. In the IR spectra, the strong
bands at 1614 (1) and 1602 (2) cm−1 and the strong to medium
intensity bands at 1410 (1) and 1422 (2) cm−1 are assigned to
the νas(CO2) and νs(CO2) carboxylate stretching modes, respect-
ively, the former most probably overlapping with an aromatic
stretch.38 The value of the spectral parameter Δ, where Δ =
νas(CO2) − νs(CO2), is less (204 cm−1 in 1, 180 cm−1 in 2) than
the corresponding values in the “ionic” sodium salts of dibaH
and pivH (∼250 cm−1), in accordance with the bidentate brid-
ging mode of the carboxylate ligation38 in the two coordination
polymers. The νas(CO2) and νs(CO2) peaks in the Raman
spectra appear at 1589 (1 and 2), 1433 (1), and 1422 (2) cm−1
,
respectively. The two qunx IR bands at 604 cm−1 (in plane
deformation) and 405 cm−1 (out-of-plane deformation) are
shifted to higher frequencies due to coordination, the former
to 655 (1) and 618 (2) cm−1, and the latter is split into 460,
422 (1) and 442, 414 (2) cm−1 39
The Raman spectra of both
.
compounds are dominated by the spectral contribution of the
qunx moiety. The characteristic qunx peaks at 3076 (1) and
3102, 3064 (2) cm−1 are due to ν(C–H)qunx, the ∼1590 (1 and 2)
istic distinct pattern of the pairs of “wings” for de, di
≈
1.6–1.8 Å (Fig. 8b and S9b‡ for compounds 1 and 2, respect-
ively). Thus, the overlap of the fingerprint plot with the green
area corresponds to the section that contains C⋯H inter-
actions with interatomic distances shorter than the sum of the
van der Waals radii of carbon and hydrogen and therefore
more effective for the crystal packing. Fig. S10‡ illustrates a
combined HS dnorm map and the molecular representation of
Fig. 9 Raman spectra of compounds {[Cu2(diba)4(qunx)]}n (1, bottom
plot) and {[Cu2(piv)4(qunx)]}n (2, upper plot).
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
and 1379, ∼1360 (1 and 2) cm−1 peaks are assigned to the ring
stretching vibrations of B1 and A1 types, respectively, while the
strong peak at 808 cm−1 for 2 mostly arises from a qunx
breathing mode. The Raman peaks at 771 (1) and 775 (2) are
tentatively attributed to a skeletal bending vibration.40 The
different nature of the diba− and piv− ligands is reflected in
Table 2 Fitted magnetic exchange parameters for compounds 1 and 2
Compound
Ca (emu K mol−1 Oe−1
)
J/Kb (K)
θc (K)
ρd (%)
1
2
0.78(1)
0.78(2)
−479(4)
−532(4)
−48(9)
−67(13)
9.5(5)
10.9(6)
a Curie constant. b Magnetic exchange constant. c Weiss constant.
the 2990–2860 cm−1 IR and Raman bands/peaks, accurate d Paramagnetic impurity.
assignments of the νas(CH3), νs(CH3), νas(CH2) and νs(CH2)
modes [the latter two only in the case of 1] being difficult.41
defects in the structure (broken dimers in the crystal). Given
Magnetic susceptibility studies
the low values of χ as a result of the strong exchange, the
effects of impurities dominate the magnetic data and behavior
below 100 K for both complexes.
Magnetic susceptibility data were collected as a function of
temperature from 1.8 to 310 K (Fig. 10). As anticipated from
the presence of paddle-wheel {Cu2(O2CR)4} units, the two com-
pounds exhibit very strong antiferromagnetic exchange inter-
actions with the molar magnetic susceptibility (χ) maxima
occurring near or above room temperature. Both complexes
show the presence of a paramagnetic impurity; this is the
result of both potential chemical impurities in the sample and
The data were fit to the Bleaney–Bowers dimer model42 with
a Curie–Weiss correction for interdimer interactions and a
paramagnetic impurity term.43 The resulting parameters are
given in Table 2. The magnetic data reveal the expected strong
antiferromagnetic exchange within the carboxylate dimer por-
tions of the chains with J values near −500 K for both com-
pounds. These values compare well with other reported tetra
(η1:η1:μ-carboxylato)dicopper(II) dimers.44–46
The large values for the Weiss constant (θ) suggest that
there are significant antiferromagnetic interactions between
the carboxylate-bridged dinuclear units via the quinoxaline
superexchange pathways, but these values are still only ∼10%
of the values for the interaction within the dinuclear units. We
recommend that these values should be taken as
qualitative measures of the parameters. The −48(9) (1) and
−67(13) (2) K values for the interdimer exchange are in good
agreement with those reported in most qunx-bridged copper(II)
complexes.18,31a
Conclusions and perspectives
In this report, we have shown that the use of acacdoH2 in reac-
tions with CuII carboxylate sources in Me2CO has led to a novel
transformation of this ligand and the preparation of interest-
ing quinoxaline- and carboxylate-bridged 1D coordination
polymers which have been fully characterized. It is important
to state that the originally employed ligand is not present in
the reaction products. The process is catalytic and this opens
the door for the catalytic synthesis of qunx and substituted
qunx compounds using acacdoH2-based starting materials.
Traditionally, the synthesis of quinoxalines is accom-
plished47,48 by: (a) cyclization of benzene substrates already
bearing appropriate substituents, (b) cyclocondensation of
benzene substrates with acyclic synthons to provide one or
more of the ring atoms required to complete the pyrazine ring,
(c) analogous processing of the performed pyrazine substrates,
or (d) rearrangement, ring expansion/contraction, degradation,
or modification of the appropriate derivatives of other hetero-
cyclic systems. By far the most common synthetic method
for quinoxalines involves the cyclocondensation of an
o-phenylenediamine (or a closely related substrate) with a
synthon containing an oxalyl [–C(vO)–C(vO)–] or equivalent,
Fig. 10 The temperature dependence of the magnetic susceptibility of
compounds 1 (upper plot) and 2 (bottom plot). The solid lines are the
best fits to the dimer model with a Curie–Weiss correction term to
account for interdimer interactions.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
e.g. HC(vO)–CuN, group.48 As far as the free qunx is
concerned, its synthesis is accomplished by the 1 : 1 reaction
of o-phenylenediamine (1,2-benzenediamine) and the glyoxal-
sodium bisulfite adduct, OHCCHO·2NaHSO4·H2O (the
synthon), in CH3COOH 2 M/Na2OCMe 4 M/H2O (pH = 5) at
60 °C in 85% yield.47a This work shows that metal-mediated
reactions of 1,3-dioximes are still in their infancy. The present
results also demonstrate the importance of the solvent in
3 (a) C. E. Housecroft and A. G. Sharpe, Inorganic Chemistry,
Pearson, Harlow, UK, 4th edn, 2012, pp. 225–236, 339–343;
(b) F. A. Cotton, G. Wilkinson, C. A. Murillo and
M. Bochmann, Advanced Inorganic Chemistry, Wiley,
New York, 6th edn, 1999, pp. 27–37.
4 (a) E. C. Constable, Metals and Ligand Reactivity, VCH,
Weinheim, Germany, new, revised and expanded edn,
1996, pp. 22–62; (b) A. J. L. Pombeiro and
V. Yu. Kukushkin, in Comprehensive Coordination Chemistry
II, ed. J. A. McCleverty and T. J. Meyer, Elsevier,
Amsterdam, 2004, vol. 1, pp. 585–594.
5 S. J. Lippard and J. M. Berg, Principles of Bioinorganic
Chemistry, University Science Books, Mill Valley, CA, USA,
1994, pp. 35–37, 257–278.
6 For review-type articles, see: (a) G. S. Papaefstathiou and
S. P. Perlepes, Comments Inorg. Chem., 2002, 23, 249;
(b) A. J. Tasiopoulos and S. P. Perlepes, Dalton Trans., 2208,
5537 (perspective); (c) Th. C. Stamatatos, C. G. Efthymiou,
C. C. Stoumpos and S. P. Perlepes, Eur. J. Inorg. Chem.,
2009, 3361 (microreview).
such reactions. The novel CuII-assisted acacdoH2 → (L1)2−
(L2)2− (L3)− transformation (Scheme 2), discovered by
,
,
Papaefstathiou’s group,13 took place in EtOH/H2O, the H2O
being responsible for the hydration of the ketone (L2)2− to the
double deprotonated gem-diol (L1)2−. On the contrary, the
acacdoH2 → qunx transformation takes place in Me2CO which
seems to participate actively in the (proposed) mechanism
(Scheme 3), providing the extra three carbon atoms required
for the transformation (Scheme 4). Investigations using other
transition metal ions under similar reaction conditions are in
progress; preliminary results with NiII show different and unre-
lated products.
We have no reasons to believe that this research area is
exhausted of new results. Indeed, ongoing studies in our
groups reveal that we have scratched only the surface of the
area of the metal ion-assisted reactivity of dioximes, including
benzil dioxime and 1,3-di(pyridine-2-yl)propane. As far as
future perspectives are concerned, variation of the metal ion
(e.g. use of NiII, ZnII, etc.), and the R,R′ groups in the RC(NOH)
CH2C(NOH)R′ (e.g. R = R′ = Ph and R = CH3, R′ = Ph) family of
dioximes may provide a systematic modification of products
(both transformed ligands and the resulting complexes); we
are currently working on these ideas and results will be
reported soon. Meanwhile, we have been developing the use of
solvents other than Me2CO, e.g. MeCN and MeNO2, in the reac-
tions; these solvents can give strongly nucleophilic carbanions
under strongly alkaline conditions, favouring the formation of
unusual ligands when reacting with oxime-containing mole-
cules in the presence of metal ions. We believe that inorganic
chemists will more actively participate in the future in projects
related to the many varied reactions of oximes that can take
place in the presence of metal ions.
7 Selected studies from our group: (a) A. A. Kitos,
C. G. Efthymiou, M. J. Manos, A. J. Tasiopoulos,
V. Nastopoulos, A. Escuer and S. P. Perlepes, Dalton Trans.,
2016, 45, 1063; (b) A. A. Kitos, D. P. Giannopoulos,
C.
Papatriantafyllopoulou,
L.
Cunha-Silva
and
S. P. Perlepes, Inorg. Chem. Commun., 2016, 64, 53;
(c)
C.
C.
C.
P.
Papatriantafyllopoulou
G.
Efthymiou,
I.
Psycharis,
and S.
Mylonas-Margaritis,
Raptopoulou,
V.
A.
P.
Escuer,
Perlepes,
Magnetochemistry, 2016, 2, 30; (d) C. G. Efthymiou,
C. P. Raptopoulou, V. Psycharis, A. J. Tasiopoulos,
A. Escuer, S. P. Perlepes and C. Papatriantafyllopoulou,
Polyhedron, 2013, 64, 30; (e) H. Sartzi, C. C. Stoumpos,
M. Giouli, I. Verginadis, S. C. Karkabounas, L. Cunha-Silva,
A. Escuer and S. P. Perlepes, Dalton Trans., 2012, 41, 11984;
(f)
C.
G.
Efthymiou,
A.
N.
Georgopoulou,
C. Papatriantafyllopoulou, A. Terzis, C. P. Raptopoulou,
A. Escuer and S. P. Perlepes, Dalton Trans., 2010, 39, 8603;
(g) C. C. Stoumpos, I. A. Gass, C. J. Milios, N. Lalioti,
A. Terzis, G. Aromi, S. J. Teat, E. K. Brechin and
S.
P.
Perlepes,
Dalton
Trans.,
2009,
307;
(h) Th. C. Stamatatos, V. Tangoulis, C. P. Raptopoulou,
A. Terzis, G. S. Papaefstathiou and S. P. Perlepes, Inorg.
Chem., 2008, 47, 7969; (i) C. Papatriantafyllopoulou,
C. G. Efthymiou, C. P. Raptopoulou, R. Vicente, E. Manessi-
Zoupa, V. Psycharis, A. Escuer and S. P. Perlepes, J. Mol.
Struct., 2007, 829, 176; ( j) G. S. Papaefstathiou, A. Escuer,
F. A. Mautner, C. Raptopoulou, A. Terzis, S. P. Perlepes and
R. Vicente, Eur. J. Inorg. Chem., 2005, 879;
(k) A. K. Boudalis, F. Dahan, A. Bousseksou, J.-P. Tuchagues
and S. P. Perlepes, Dalton Trans., 2003, 3411;
(l) G. S. Papaefstathiou, A. Escuer, M. Font-Bardia,
S. P. Perlepes, X. Solans and R. Vicente, Polyhedron, 2002,
21, 2027; (m) G. S. Papaefstathiou, A. Escuer,
C. P. Raptopoulou, A. Terzis, S. P. Perlepes and R. Vicente,
Eur. J. Inorg. Chem., 2001, 1567; (n) A. Tsohos,
Acknowledgements
We thank Professor Dionyssios Papaioannou for extremely
helpful discussions concerning the acacdoH2
transformation.
→ qunx
Notes and references
1 J. Ribas Gispert, Coordination Chemistry, Wiley-VCH,
Weinheim, Germany, 2008, pp. XXXIX, XL.
2 J. Bjerrum, Metal Ammine Formation in Aqueous Solution,
P. Haase and Son, Copenhagen, 1957.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
S. Dionyssopoulou, C. P. Raptopoulou, A. Terzis, 17 M. Goto, Y. Kani, M. Tsuchimoto, S. Ohba, H. Matsuhima
E. G. Bakalbassis and S. P. Perlepes, Angew. Chem., Int. Ed.,
1999, 38, 983; (o) V. Tangoulis, C. P. Raptopoulou,
and T. Tokii, Acta Crystallogr., Sect. C: Cryst. Struct.
Commun., 2000, 56, 7.
S. Paschalidou, E. G. Bakalbassis, S. P. Perlepes and 18 C. P. Landee, A. Delcheva, C. Galeriu, G. Pena,
A. Terzis, Angew. Chem., Int. Ed. Engl., 1997, 36, 1083.
8 For comprehensive review, see: C. J. Milios,
M. M. Turnbull and R. D. Willett, Polyhedron, 2003, 22,
2325.
a
Th. C. Stamatatos and S. P. Perlepes, Polyhedron, 2006, 25, 19 CrystalClear, Rigaku/ MSC Inc., The Woodlands, TX, USA,
134 (report). 2005.
9 Selected studies from our group: (a) V. Douros, H. Sartzi, 20 (a) G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam.
S. J. Teat, Y. Sanakis, O. Roubeau and S. P. Perlepes, Inorg.
Chem. Commun., 2014, 50, 117; (b) K. F. Konidaris,
Crystallogr., 2008, 64, 112; (b) G. M. Sheldrick, Acta
Crystallogr., Sect. C: Cryst. Struct. Commun., 2015, 71, 3.
C. D. Polyzou, G. E. Kostakis, A. J. Tasiopoulos, 21 DIAMOND: Crystal and Molecular Structure Visualization, ver.
O. Roubeau, S. J. Teat, E. Manessi-Zoupa, A. K. Powell and 3.1, Crystal Impact GbR, Bonn, Germany, 2011.
S. P. Perlepes, Dalton Trans., 2012, 41, 2862; 22 S. K. Wolff, D. J. Grimwood, J. J. McKinnon, M. J. Turner,
(c) C. C. Stoumpos, R. Inglis, O. Roubeau, H. Sartzi,
D. Jayatilaka and M. A. Spackman, CrystalExplorer, The
A. A. Kitos, C. J. Milios, G. Aromi, A. J. Tasiopoulos,
University of Western Australia, Crawley, Australia, 2012.
V. Nastopoulos, E. K. Brechin and S. P. Perlepes, Inorg. 23 J. March, Advanced Organic Chemistry, Wiley, New York, 4th
Chem., 2010, 49, 4388; (d) C. J. Milios, P. Kyritsis,
C. P. Raptopoulou, A. Terzis, R. Vicente, A. Escuer and
edn, 1992, pp. 174, 641–676, 741–743, 1062, 1095–1097,
1110–1112, 1121–1125, 1162–1164.
S. P. Perlepes, Dalton Trans., 2005, 501; (e) C. J. Milios, 24 For example, see: (a) D. J. Pasto and J. E. Brophy, J. Org.
E. Kefalloniti, C. P. Raptopoulou, A. Terzis, A. Escuer,
R. Vicente and S. P. Perlepes, Polyhedron, 2004, 23, 83.
10 For reviews, see: (a) V. Yu. Kukushkin, D. Tudela and
Chem., 1991, 56, 4554; (b) M.-H. Yeh, L. Linder,
D. K. Hoffman and Th. J. Barton, J. Am. Chem. Soc., 1986,
108, 7849.
A. J. L. Pombeiro, Coord. Chem. Rev., 1996, 156, 333; 25 P. P. Fu and R. G. Harvey, Chem. Rev., 1978, 78, 317.
(b) V. Yu. Kukushkin and A. J. L. Pombeiro, Coord. Chem. 26 (a) L. De Luca, G. Giacomelli and A. Porcheddu, J. Org.
Rev., 1999, 181, 147; (c) A. J. L. Pombeiro and
V. Yu. Kukushkin, in Comprehensive Coordination
Chem., 2002, 67, 6272; (b) A. Mishra, A. Ali, S. Upreti and
R. Gripta, Inorg. Chem., 2008, 47, 154.
Chemistry II, ed. J. A. McCleverty and T. J. Meyer, Elsevier, 27 We thank a referee for her/his suggestions.
Amsterdam, 2004, vol. 1, pp. 631–637; (d) D. A. Garnovskii 28 (a) S. P. Perlepes, E. Libby, W. E. Streib, K. Folting and
and V. Yu. Kukushkin, Russ. Chem. Rev., 2006, 75, 111;
(e) D. S. Bolotin, N. A. Bokach and V. Yu. Kukushkin,
Coord. Chem. Rev., 2016, 313, 62; (f) C.-I. Yang,
Z.-Z. Zhang and S.-B. Lin, Coord. Chem. Rev., 2015,
289–290, 289; (g) N. Gerasimchuk, Eur. J. Inorg. Chem.,
2014, 4518 (microreview); (h) A. G. Beirakhov, I. M. Orlova
and Yu. N. Mikhailov, Russ. J. Inorg. Chem., 2014, 59,
1679.
G. Christou, Polyhedron, 1992, 11, 923; (b) R. J. Doedens,
Prog. Inorg. Chem., 1976, 21, 209; (c) J. Catterick and
P. Thornton, Adv. Inorg. Chem. Radiochem., 1977, 20, 291;
(d) M. Melnik, Coord. Chem. Rev., 1982, 42, 259; (e) M. Kato
and Y. Muto, Coord. Chem. Rev., 1988, 92, 45;
(f) R. D. Harcourt, F. L. Skrezenek and R. G. A.
R. Maclagan, J. Am. Chem. Soc., 1986, 108, 5403.
29 A. W. Addison, T. N. Rao, J. Reedijk, J. V. Rijn and
G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349.
11 For a review concerning the organic chemistry of 1,3-diox-
imes, see: A. Kotali and V. P. Papageorgiou, Org. Prep. 30 (a) M. A. Spackman and D. Jayatilaka, CrystEngComm, 2009,
Proced. Int., 1991, 23, 593.
11, 19; (b) J. J. McKinnon, F. P. A. Fabbiani and
M. A. Spackman, Cryst. Growth Des., 2007, 7, 755;
(c) J. J. McKinnon, M. A. Spackman and A. S. Mitchell, Acta
Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater., 2004, 60,
627; (d) K. Takahashi, N. Hoshino, T. Takeda, S. Noro,
T. Nakamura, S. Takeda and T. Akutagawa, Inorg. Chem.,
2015, 54, 9423; (e) M. A. Spackman and J. J. McKinnon,
CrystEngComm, 2002, 4, 378.
12 S. O. Baumann, M. Bendova, H. Fric, M. Puchberger,
C. Vismescu and U. Schubert, Eur. J. Inorg. Chem., 2009,
3333.
13 A. K. Kostopoulos, A. D. Katsenis, J. M. Frost, V. G. Kessler,
E. K. Brechin and G. S. Papaefstathiou, Chem. Commun.,
2014, 50, 15002.
14 (a) H.-C. Yao, M.-M. Li, G.-S. Yang, Z.-J. Li and Y. Zhu,
Inorg. Chim. Acta, 2007, 360, 3959; (b) F. D. S. Miranda, 31 (a) J. Jornet-Somoza, N. Codina-Castillo, M. Deumal,
F. G. Menezes, J. Vicente, A. J. Bortoluzzi, C. Zucco,
A. Neves and N. S. Goncalves, J. Mol. Struct., 2009, 938, 1;
(c) J. Zhang, Z. Zhang, Z. Chen and X. Zhou, Dalton Trans.,
2012, 41, 357.
F. Mota, J. J. Novoa, R. T. Butcher, M. M. Turnbull,
B. Keith, C. P. Landee and J. L. Wikaira, Inorg. Chem., 2012,
51, 6315; (b) T. Kogane, N. Koyama, T. Ishida and
T. Nogami, Polyhedron, 2007, 26, 1811; (c) B. C. Keith,
F. Xiao, C. P. Landee, M. M. Turnbull and A. Zheludev,
Polyhedron, 2011, 30, 3006; (d) P. Lumme, S. Lindroos and
E. Lindell, Acta Crystallogr., Sect. C: Cryst. Struct. Commun.,
1987, 43, 2053; (e) A. A. Massoud, V. Langer, L. Ohrstrom
15 K. J. Oliver, T. N. Waters, D. F. Cook and C. E. F. Rickard,
Inorg. Chim. Acta, 1977, 24, 85.
16 S. I. Troyanov, E. G. Il’ina and K. M. Dunaeva, Koord. Khim.,
1991, 17, 1692.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
and M. A. M. Abu-Youssef, J. Coord. Chem., 2009, 62, 519;
(f) S. Lindroos and P. Lumme, Acta Crystallogr., Sect. C:
1986, pp. 231–233; (b) G. B. Deacon and R. J. Phillips,
Coord. Chem. Rev., 1980, 33, 227.
Cryst. Struct. Commun., 1990, 46, 2039; (g) S. E. H. Etaiw, 39 P. Lumme, S. Lindroos and E. Lindell, Inorg. Chim. Acta,
M. E. El-Zaria, T. A. Fayed and S. N. Abdou, J. Inorg.
Organomet. Polym. Mater., 2011, 21, 465.
32 (a) E. Y. Lee, B. K. Park, C. Kim, S.-J. Kim and Y. Kim, Acta
Crystallogr., Sect. E: Struct. Rep. Online, 2008, 64, m286;
1988, 144, 75.
40 (a) H. F. Shurvell and J. W. Rusell, J. Raman Spectrosc.,
1975, 3, 407; (b) H. M. Badawi, A. A. Al-Saadi and S. A. Ali,
J. Mol. Struct., 2014, 1058, 284.
(b) L. Sieron, Acta Crystallogr., Sect. E: Struct. Rep. Online, 41 (a) F. R. Dollish, W. G. Fateley and F. F. Bentley,
2008, 64, m53.
Characteristic Raman Frequencies of Organic Compounds,
Wiley, New York, 1974, pp. 1–11; (b) C. N. R. Rao, Chemical
Applications of Infrared Spectroscopy, Academic Press,
New York, 1963, pp. 125–136; (c) R. M. Silverstein,
G. C. Bassler and T. C. Morrill, Spectrometric Identification
of Organic Compounds, Wiley, New York, 4th edn, 1981, pp.
106–108.
33 M. Barqin, M. J. González Garmendia, L. Larrínaga,
E. Pinilla and M. R. Torres, Z. Anorg. Allg. Chem., 2005, 631,
2210.
34 A. D. Burrows, M. F. Mahon, P. R. Raithby, A. J. Warren,
S. J. Teat and J. E. Warren, CrystEngComm, 2012, 14, 3658.
35 K. A. H. Alzahrani and R. J. Deeth, Dalton Trans., 2016, 45,
11944.
42 B. Bleaney and K. D. Bowers, Philos. Mag., 1952, 214, 451.
36 (a) S. Takamizawa, E. Nakata, T. Saito and H. Yokoyama, 43 C. P. Landee and M. M. Turnbull, Eur. J. Inorg. Chem.,
Inorg. Chem. Commun., 2003, 6, 1326; (b) S. Takamizawa, 2013, 2266.
E. Nakata and H. Yokoyama, Inorg. Chem. Commun., 2003, 44 A. Mookherji and S. C. Mathur, Nature, 1962, 196, 370.
6, 763; (c) S. Takamizawa, E. Nakata and T. Saito, Inorg. 45 Y. Muto, T. Tokii, K. Chijiwa and M. Kato, Bull. Chem. Soc.
Chem. Commun., 2004, 7, 1; (d) S. Takamizawa, E. Nakata,
Jpn., 1984, 57, 1008.
T. Saito, T. Akatsuka and K. Kojima, CrystEngComm, 2004, 46 (a) Md. E. Ali and S. N. Datta, J. Mol. Struct.: THEOCHEM,
6, 197; (e) S. Takamizawa, E. Nakata, T. Akatsuka,
R. Miyake, Y. Kakizaki, H. Takeuchi, G. Maruta and
S. Takeda, J. Am. Chem. Soc., 2010, 132, 3783;
(f) S. Takamizawa, T. Akatsuka and R. Miyake,
CrystEngComm, 2010, 12, 82; (g) K. Takahaski, N. Hoshino,
2006, 775, 19; (b) A. G. Bellesis, C. P. Landee, M. Polson,
M. M. Turnbull and J. L. Wikaira, Eur. Chem. Bull., 2015, 4,
110; (c) M. Barquin, N. Cocera, M. J. Garmendia,
L. Larrinaga, E. Pinilla and M. R. Torres, J. Coord. Chem.,
2010, 63, 2247.
T. Takeda, S. Noro, T. Nakamura, S. Takeda and 47 (a) J. C. Cavagnol and F. Y. Wiselogle, J. Am. Chem. Soc.,
T. Akutagawa, Dalton Trans., 2014, 43, 9081; (h) Y. Takasaki
and S. Takamizawa, Nat. Commun., 2015, 6, 1.
37 I. H. Hwang, Y. D. Jo, H. Kim, K. B. Kim, K.-D. Jung, C. Kim,
Y. Kim and S.-J. Kim, Inorg. Chim. Acta, 2013, 402, 39.
38 (a) K. Nakamoto, Infrared and Ramon Spectra of Inorganic
and Coordination Compounds, Wiley, New York, 4th edn,
1947, 69, 795; (b) H. Yuan, K. Li, Y. Chen, Y. Wang, J. Cui
and B. Chen, Synlett, 2013, 2315; (c) C. Qi, H. Jiang,
L. Huang, Z. Chen and H. Chen, Synthesis, 2011, 387.
48 Quinoxalines: Supplement II, Chemistry of Heterocyclic
Compounds, ed. D. J. Brown, Wiley, New York, 2004, vol. 61,
pp. 1–44.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.