Solvothermal synthesis of oxamate-based helicate: Temperature dependence of the hydrogen bond structuring in the solid

Article abstract of DOI:10.1021/cg300877r

Solvothermal synthesis has been successfully tested as a new synthetic method toward the formation of oxamate-based coordination compounds. The reaction of 1,3-benzenedioxamic acid (H₄mpba) with divalent metal ions has afforded the dinuclear co

Full text of DOI:10.1021/cg300877r

Article
pubs.acs.org/crystal
Solvothermal Synthesis of Oxamate-Based Helicate: Temperature
Dependence of the Hydrogen Bond Structuring in the Solid
Laurent Lisnard,§,† Lise-Marie Chamoreau,§,‡ Yanling Li,§,† and Yves Journaux*,§,†
§Institut Parisien de Chimie Molec
†CNRS, Institut Parisien de Chimie Molec
‡Centre de res
olution structurale, Institut Parisien de Chimie Molec
* Supporting Information
́
ulaire, UPMC Univ Paris 06, Paris 75005, France
ulaire, UMR 7201, Paris 75005, France
ulaire, Paris 75005, France
́
́
́
S
ABSTRACT: Solvothermal synthesis has been successfully tested as a new
synthetic method toward the formation of oxamate-based coordination
compounds. The reaction of 1,3-benzenedioxamic acid (H4mpba) with
divalent metal ions has afforded the dinuclear compounds TBA2[Co2-
(H2mpba)3]·2DMF·5H2O (1a), (HNEt3)2[Co2(H2mpba)3]·6DMF·5H2O
(1b), TBA2[Ni2(H2mpba)3]·2DMF·2.5H2O (2), and (HNEt3)2[Co2-
(H2mpba)3] (3). Although the 3:2 ligand to metal ratio is known for the
(mpba)4− ligand under bench conditions, these complexes are the first
examples of oxamate-based helicate. Furthermore, crystallographic studies
show a temperature-dependent hydrogen bond structuring that leads to
racemic or chiral hexagonal 3D networks.
growing metal−metal distances13,14 inevitably leading to a
INTRODUCTION
■
dramatic drop in the ordering temperature value when the
system is scaled up with longer organic linkers. To tackle this
problem, we have designed over the past few years new
polytopic oxamate-based ligands able to efficiently transmit
electronic interaction across extended aromatic bridges via a
spin polarization mechanism. Starting from these ligands, we
were able to obtain homometallic polynuclear complexes that
effectively exhibit relatively strong ferromagnetic interactions
despite large metal−metal separation.15−18 When reacted with
additional metal ions, some of these complexes give 2D or 3D
magnetic MOFs19−21 with magnetic ordering in the 6.5 to 22.5
K temperature range. The metal−metal distance through the
bridging oxamate ligand is not tunable but with the help of
serendipity large octogonal pores have been obtained.21
Furthermore these networks are resizable by changing the
aromatic parts while retaining a significant interaction.
Nevertheless the final geometry of the nets depends on subtle
chemical equilibria, and to fully exploit this family of ligands, we
have decided to explore their reactivity in solvothermal
conditions. This method imposes strong constraints both in
pressure and in temperature to the reaction media and allows
its user to work beyond the solvent’s boiling point.22
Consequently the solubility and the mobility of the reactants
are increased, and reactions that could not be accomplished or
would not yield crystals under bench conditions become
The synthesis of molecular building blocks that self-assemble or
are capable of connexion with another component to form
molecular objects or networks of defined topology or geometry
is one of the major subjects in crystal engineering1,2 and in
supramolecular chemistry.3−5 The assembly process relies on
weak chemical interactions such as H-bonds, electrostatic
interactions, or coordination bonding. However, due to subtle
equilibria between these weak interactions, there are still many
surprises in the outcome and the final product can be quite
different from the expected one. These unexpected results often
open up new opportunities and contribute greatly to the beauty
of synthetic chemistry. Among all the research work on the
crystal engineering of nets, the synthesis of metal−organic
frameworks has increased exponentially.6 The reason for this
success is related to the chemical flexibility of these materials
owing to the presence of tunable organic and inorganic parts,
which allows one to scale up the system or to introduce new
properties. Moreover MOFs have many potential applications
in catalysis but also in hydrogen7 and CO2 storage, which are
8
two of the most pressing technological social needs. MOFs are
also highly interesting for their physical properties such as
luminescence9,10 or ferroelectricity11 and magnetism.12 Regard-
ing magnetic MOFs, the goal is the obtainment of magnets and
more precisely porous magnets, so in addition to the control
over atom arrangement, the ability of organic ligands to
propagate the electronic interaction between the magnetic ions
is of paramount importance. At first sight, large porosity seems
incompatible with magnetic ordering due to the exponential
decrease of the interaction between the magnetic ions with
Received: June 29, 2012
Revised: August 20, 2012
Published: September 4, 2012
© 2012 American Chemical Society
4955
dx.doi.org/10.1021/cg300877r | Cryst. Growth Des. 2012, 12, 4955−4962

Crystal Growth & Design
Article
feasible.23 This method has convincingly proven its efficiency
for the synthesis of multidimensional materials24,25 as well as
the synthesis of coordination clusters26 and represents therefore
a convenient tool to investigate polyoxamate’s reactivity.
To start our study, we have been focusing on the 1,3-
benzenedioxamate ligand (Figure 1a) since it is known to
helicate geometry, which has never been observed to date in
oxamate-based complexes. X-ray diffraction studies also show
that the supramolecular hydrogen-bond arrangement of the
complexes in the solid is strongly dependent on the reaction
temperature: reactions performed at 100 °C have afforded a
racemic hexagonal 3D structure in contrast to a chiral phase
that forms at 150 °C.
EXPERIMENTAL SECTION
■
Materials and Methods. All reagents were used as purchased with
no further purification. Diethyl N,N′-1,3-benzenedioxamate
(H2Et2mpba) was prepared according to the literature procedure.27
1H and 13C NMR spectra were collected on a 400 MHz Bruker Avance
spectrometer at 298 K. ATR/FT-IR spectra were collected on a Bruker
TENSOR 27 equipped with a simple reflection ATR diamond plate of
the Harrick MPV2 series. Magnetic measurements in dc mode were
performed on a Quantum Design MPMS SQUID on polycrystalline
samples restrained in a plastic film. Data were corrected for the
diamagnetism contributions of the samples using Pascal constants. The
sample holder diamagnetism was measured and subtracted from the
raw data.
Synthesis. Synthesis of H4mpba·3H2O. Hydrate N,N′-1,3-
benzenedioxamic acid was prepared as follows: aqueous NaOH
(6.22 g, 1.55 mol, 100 mL) was added dropwise to a suspension of
H2Et2mpba (12 g, 38.9 mmol) in water (800 mL). The solution was
stirred for 30 min and filtered. Addition of 4 M HCl (38.9 mL, 1.55
mol) to the resulting solution leads to the formation of a white
precipitate. After 15 min, the precipitate is collected on a sintered glass
filter, washed with water and then with 50% v. EtOH, and dried for
several hours in air and overnight at 45 °C in an oven. Yield: 10.9 g
1
(91.5%). H NMR (Bruker Avance 300 MHz, DMSO-d6, 298 K) δ
10.74 (s, 2H, NH), 8.27 (s, 1H), 7.50 (d, 2H, J = 7.95 Hz), 7.34 (t,
1H, J = 8.1 Hz). 13C NMR (Bruker Avance 400 MHz, DMSO-d6,
298K) δ 112.68, 116.83, 128.86, 137.83, 157.02, 162.11. IR (ATR,
cm−1): 3427 s; 3311 m; 3092 s; 2523 s; 1998 s; 1675 S; 1612 m; 1550
m; 1479 s; 1457 s; 1286 m; 1258 m; 1225 S; 1209 m; 1179 s; 1147 s;
1097 s; 993 s; 926 s; 875 s; 822 s; 787 m; 729 S; 678 s; 534 m; 500 m;
408 S; 262 S. Elemental analysis (%) calculated for C10H14N2O9 (Mr =
306.22 g mol−1): C 39.22, H 4.60, N 9.14. Found: C 39.21, H 4.44, N
9.07.
Synthesis of TBA2[Co2(H2mpba)3]·2DMF·5H2O, 1a. CoCl2·6H2O
(0.119 g, 0.5 mmol), H2Et2mpba (0.153 g, 0.5 mmol), and TBAOH (1
M in MeOH, 1 mL, 1 mmol) were placed in a 23 mL Teflon-lined
autoclave filled with DMF (8 mL) and heated at 100 °C for 12 h. After
slow cooling to room temperature, pink crystals of 1a were collected,
washed by sonication in DMF, and dried in air. Yield: 0.152 g (39%,
based on Co). IR (ATR, cm−1): 3470 s; 3197 s; 3083 s; 2963 s; 2875 s,
1661 S; 1624 S; 1595 S; 1543 S; 1474 S; 1456 m; 1409 s; 1364 S; 1294
m; 1255 s; 1228 s; 1171 s; 1093 s; 1062 s; 1031 s; 988 s; 934 s; 893 S;
837 s; 798 S; 760 s; 739 s; 690 s; 660 s; 598 s; 539 s; 515 s; 473 s; 445
m; 404 s; 357 s; 305 m; 254 S. Elemental analysis (%) calculated for
C68H114Co2N10O25 (Mr = 1589.55 g mol−1): Co 7.41, C 51.38, H 7.22,
N 8.81. Found: Co 7.65, C 51.10, H 6.70, N 9.10.
Synthesis of (HNEt3)2[Co2(H2mpba)3]·6DMF·5H2O, 1b.
CoCl2·6H2O (0.119 g, 0.5 mmol), H4mpba·3H2O (0.154 g, 0.5
mmol), and NEt3 (99%, 0.28 mL, 2 mmol) were placed in a 23 mL
Teflon-lined autoclave filled with DMF (9 mL) and heated at 100 °C
for 12 h. After slow cooling to room temperature, red-brown crystals
of 1b were collected, washed by sonication in DMF, and dried in air.
Yield: 0.043 g (11%, based on Co). IR (ATR, cm−1): 3435 s; 3197 s;
3084 s; 1621 S; 1595 S; 1545 S; 1475 s; 1456 m; 1364 S; 1295 m;
1255 s; 1229 s; 1174 s; 1094 s; 1062 s; 1034 s; 988 s; 934 s; 892 s; 837
s; 799 S; 760 s; 689 s; 660 s; 598 s; 538 m; 517 s; 444 m; 405 s; 357
m; 307 S; 251 S; 214 m. Elemental analysis (%) calculated for
C60H102Co2N14O29 (Mr = 1601.39 g mol−1): C 45.0, H 6.42, N 12.24.
Found: C 44.65, H 6.06, N 12.46.
Figure 1. (a) Representation of the 1,3-benzenedioxamic acid; (b)
structure of the anionic complex in 1b, with legend as an inset, H
atoms are omitted for clarity; (c) representation of the N(H)···O
bond (orange dotted lines) between adjacent molecules in 1b; H
atoms are omitted for clarity; the dinuclear complex is schemed with
an intramolecular Co−Co bound pictured in purple.
promote ferromagnetic exchange between paramagnetic ions
through a spin polarization mechanism.15 We have thus reacted
this ligand with cobalt(II) or nickel(II) salts in DMF varying
the temperature of the solvothermal treatment. Reactions
performed at 100 °C in the presence of TBAOH or NEt3 have
yielded the bimetallic compounds TBA2[Co2(H2mpba)3]·
2DMF·5H2O (1a), (HNEt3)2[Co2(H2mpba)3]·6DMF·5H2O
(1b), and TBA2[Ni2(H2mpba)3]·2DMF·2.5H2O (2), whereas
the compound (HNEt3)2[Co2(H2mpba)3] (3) has been
obtained at 150 °C in the presence of triethylamine.
Crystallographic studies reveal that all the complexes adopt a
Synthesis of TBA2[Ni2(H2mpba)3]·2DMF·2.5H2O, 2. NiCl2·6H2O
(0.119 g, 0.5 mmol), H2Et2mpba (0.154 g, 0.5 mmol), and TBAOH (1
M in MeOH, 1 mL, 1 mmol) were placed in a 23 mL Teflon-lined
4956
dx.doi.org/10.1021/cg300877r | Cryst. Growth Des. 2012, 12, 4955−4962

Crystal Growth & Design
Article
autoclave filled with DMF (8 mL) and heated at 100 °C for 12 h. After
slow cooling to room temperature, green crystals of 2 were collected,
washed by sonication in DMF, and dried in air. Yield: 0.111 g (29%,
based on Ni). IR (ATR, cm−1): 3463 s; 3229 s; 3084 s; 2963 s; 2875 s;
1662 m; 1625 S; 1595 S; 1545 S; 1474 s; 1457 s; 1409 s; 1366 S; 1295
s; 1255 s; 1229 s; 1171 s; 1094 s; 1062 s; 1031 s; 988 s; 935 s; 894 s;
837 s; 802 m; 762 s; 739 s; 690 s; 660 s; 599 s; 540 s; 517 s; 499 s; 483
s; 462 s, 446 s; 406 s; 360 s, 318 m; 274 m. Elemental analysis (%)
calculated for C68H109N10Ni2O22.5 (Mr = 1544 g mol−1): Ni 7.60, C
52.89, H 7.11, N 9.07. Found: Ni 8.10, C 50.74, H 6.75, N 9.6.
Synthesis of (HNEt3)2[Co2(H2mpba)3], 3. CoCl2·6H2O (0.119 g, 0.5
mmol), H4mpba·3H20 (0.153 g, 0.5 mmol), and NEt3 (99%, 0.280
mL, 2 mmol) were placed in a 23 mL Teflon-lined autoclave filled with
DMF (9 mL) and heated at 150 °C for 12 h. After slow cooling to
room temperature, purple crystals of 3 were collected, washed by
sonication in DMF, and dried in air. Yield: 0.135 g (50.3%, based on
Co). IR (ATR, cm−1): 3228 s; 1631 S; 1598 S; 1540 m; 1475 m; 1456
m; 1361 S; 1290 m; 1233 s; 1161 s; 1017 s; 979 s; 934 s; 885 s; 800 S;
707 s; 684 m; 630 s; 604 s; 545 m; 509 m; 439 m; 407 s; 358 s; 296 m;
273 S; 243 S. Elemental analysis (%) calculated for C42H50Co2N8O18
(Mr = 1072.75 g mol−1): Co 10.98, C 47.02, H 4.69, N 10.44. Found:
Co 10.66, C 47.56, H 4.64, N 10.66.
Crystallographic Studies. Data were collected on a Bruker
Kappa-APEX II CCD diffractometer for 1b and 3 (Mo Kα, λ =
0.71073 Å). Crystals were mounted on a Hamilton cryoloop using
Paratone-N oil and placed in the cold flow produced with an Oxford
Cryocooling device. Partial hemispheres of data, predefined with the
APEX II software,28a were collected using ϕ and ω scans (110 s/frame
for 1b, 100 s/frame for 3). Integrated intensities were obtained with
SAINT+28a and were corrected for absorption with SADABS;28b,c
structure solution and refinement was performed with the SHELXTL-
package.28a The structures were solved by direct methods and
completed by iterative cycles of ΔF syntheses and full-matrix least-
squares refinement against F2. Compounds 1a, 1b, and 2 are poorly
diffracting, and despite numerous efforts, the best data set obtained, for
1b, still gives some large reliability factors, and restraints have been
used to model counterions and solvent molecules. Crystallographic
data and refinement parameters for 1b and 3 are given in Table 1.
X-ray powder diffraction patterns for 1a, 1b, and 2 are presented in
the Figure S1 of the Supporting Information. Diagrams were collected
on a Philips X’pert Pro diffractometer using Cu Kα1 monochromatized
radiation (λ = 1.54060 Å) and equipped with an X’celerator linear
detector.
Table 1. Crystallographic Data for Compounds 1b and 3
1b
3
a
formula
C51H91Co2N11O31
1472.21
C42H50Co2N8O18
1072.76
monoclinic
P21
FW [g mol−1
cryst syst
space group
a [Å]
]
trigonal
R3c
̅
17.527(7)
17.527(7)
46.452(7)
90
10.3375(6)
15.2574(15)
15.0870(15)
90
b [Å]
c [Å]
α [deg]
β [deg]
γ [deg]
V [Å3]
Z
90
91.745(7)
90
120
12359(3)
6
2378,5(4)
2
T [K]
200(2)
0.71073
1.187
200(2)
0.71073
1.498
λ [Å]
ρcalc [g cm−3
]
μ(Mo Kα) [mm−1
]
0.480
0.780
F(000)
4656
1112
cryst size [mm3]
θ limits [deg]
meas reflns
unique reflns
Rint
0.32 × 0.19 × 0.17
2.21−24.99
19834
2416
0.31 × 0.15 × 0.15
3.01−30.01
28005
12872
0.0960
1595
0.0345
reflections I > 2σ(I)
params
9744
124
633
restraints
9
1
b
R1 [I > 2σ(I)]
0.0849
0.2981
1.135
0.0422
c
wR2
0.0955
GOF
1.000
largest residuals [e Å3]
−0.543; 1.823
−0.319; 0.429
a
b
c
Including solvate molecules. R1 = ∑||Fo| − |Fc||/∑|Fo|. wR2
=
[∑(ω(Fo2 − F2c)2)/∑(ω(F2o)2)]1/2
.
established that a 2D net made of octahedra sharing edges is
only obtained for a perfect alternation of Λ and Δ
isomers.31−33(On the other hand, a 3D (10,3) chiral framemork
is obtained for enanantiopure octahedra sharing edges; see, for
instance, ref 33.) Therefore both enantiomers of the helicate
are found in the structure and are alternatively linked to one
another through the double H-bonds to yield the heterochiral
honeycomb-like framework A4[Δ-MΔ-M(H2mpba)3][Λ-MΛ-
RESULTS
■
Compounds 1a, 1b, and 2 are isostructural and crystallize in the
R3c centrosymmetric space group with an asymmetric unit that
̅
+
consists of one-sixth of the bimetallic complex. Each complex is
made of two divalent ions in octahedral geometry bridged by
three partially deprotonated (H2mpba)2− ligands. The metal
ions bind two oxygen atoms from the oxamate groups leaving
one carbonyl and one amido function uncoordinated. Such a
binding mode arises from the nondeprotonation of the amido
groups independently of the strength of the base used in the
reaction and results in the formation of a triple-stranded chiral
helicate complex where the oxamate groups lie in the plane of
the benzyl rings (Figure 1b).
M(H2mpba)3] (A = TBA+, HNEt3 ; M = Ni, Co) (Figure 2). It
is worth noticing that the H-bonds seem to solely drive the
three-dimensional arrangement of the complexes since no
obvious structural differences could be observed between the
+
TBA+ and the HNEt3 salts. The formation of this relatively
strong supramolecular network could potentially tend to
stabilize the helicate complex and limit the deprotonation of
the ligands.
This binding mode of the (Hxmbpa)(4−x)− ligand is
encountered for the first time. Indeed, under bench conditions
and in aqueous media, the fully deprotonated form of the
ligand, (mbpa)4−, also binds octahedral ions in a 3:2 fashion but
affords a heterochiral triple-stranded mesocate bimetallic
complex [Co2(mpba)3]8− where the oxamate groups are
perpendicular to the benzyl rings (Figure 3a).20 This difference
in the geometrical arrangement in 1b and in [Co2(mpba)3]8−
changes the relative orientation of the coordination polyhedra
between the two halves of the dinuclear complexes. In 1b, the
two polyhedra have a staggered conformation when viewed
along the Co−Co axis, whereas in [Co2(mpba)3]8− the two
The formation of multiple hydrogen bonds between the
amide (proton donor) and carbonyl (proton acceptor)
functions of oxamic acids from adjacent complexes leads to a
3D framework composed of 2D hexagonal layers pillared by the
helicoidal complexes (Figure 1c and 2). The found N(
H)···O distance of 2.81(1) Å is relatively short in comparison
to the intermolecular H-bond distances reported for the
H2Et2mpba ligand.29,30 The 2D (6,3) hexagonal networks are
composed of cobalt octahedra linked via their edges by double
H-bonds through the oxamate groups, and it is now well
4957
dx.doi.org/10.1021/cg300877r | Cryst. Growth Des. 2012, 12, 4955−4962

Crystal Growth & Design
Article
Figure 4. Staking of hexagonal networks in (a) 1b and (b)
Figure 2. Crystal packing of 1b showing the H-bond structuring
(orange dotted lines) into a heterochiral hexagonal 3D network; H
atoms are omitted for clarity; dinuclear complexes are schemed with
an intramolecular Co−Co bound pictured in purple in the bc plane;
benzyl rings are omitted for clarity in the representation of the ab
plane.
[Li3Co2(mpba)3(H2O)6]5−; benzyl rings are omitted for clarity.
ions leads to a 3D honeycomb framework with hexagonal
channels and is topologically like the lonsdaleite or wurtzite
structures, that is, “lon” in RCSR database. By contrast, the
staggered conformation of the helicate does not lead to large
pores in 1b.
Compound 3 crystallizes in the P21 chiral space group, and a
whole complex is found in the asymmetric unit. The bimetallic
complex is identical to the previously described compounds;
however, there are drastic differences in the 3D layout since the
H-bond-driven arrangement of the bimetallic species leads to a
homochiral framework (Figure 5). The H-bond network is far
more complex than in the previous cases with each ligand of the
polyhedra have an eclipsed conformation (Figure 3b,c).
Consequently these conformations lead to different networks.
In 1b, the double H-bonds between adjacent complexes lead to
a stacking of staggered hexagonal networks (Figure 4a) with the
same topology as diamond or sphalerite structures, that is, “dia”
n e t i n R C S R d a t a b a s e , 3 4 w h e r e a s i n
Li5[Li3Co2(mpba)3(H2O)6]·31H2O, the coordination of the
two cis carbonyl-oxygen atoms of the oxamato groups to Li+
Figure 3. (a) Structure of the anionic complex [Co2(mpba)3]8−;20 (b) stick model of 1b along the Co−Co axis; (c) Stick model of [Co2(mpba)3]8−
along the Co−Co axis.
4958
dx.doi.org/10.1021/cg300877r | Cryst. Growth Des. 2012, 12, 4955−4962

Crystal Growth & Design
Article
2.8708(2) to 3.0886(3) Å with a mean value of 2.97 Å. One can
suppose that performing the reaction at 150 instead of 100 °C
leads to a higher autogenous pressure during the solvothermal
treatment, which in turn constrains the system to stabilize into
a denser phase and hence this more complex H-bond
networking. The consequence of this densification is the lack
of one of the two enantiomers of the bis-cobalt(II) helicate in
the hexagonal lattice. Although distorted, the hexagonal lattice
is still visible in Figure 5 with vertices alternatively defined by
cobalt ocathedra and the entanglement of hydrogen bonds
leading then to an enantiopure network.
No signals were found in circular dichroism measurements
performed on the bulk material indicating that both
enantiomers (HNEt3)2[Δ-CoΔ-Co(H2mpba)3] and
(HNEt3)2[Λ-CoΛ-Co(H2mpba)3] crystallize out from the
solvothermal reaction. Attempts to measure circular dichroism
data on a single crystal diluted in KBr proved unsuccessful
whereas measurements in solution were made impossible
because compound 3 is insoluble in most common solvents,
including DMSO or DMF.
Magnetic Properties. Magnetic measurements in DC
mode were performed on all compounds (Figure 7). At room
temperature, the χMT values, approximately 6.5 cm3·K·mol−1
for 1a, 1b, and 3, are close to the expected one for two isolated
Co(II) ions. Upon cooling, the χMT products steadily decrease
and reach 3.7 cm3·K·mol−1 at 2 K. In the 300−2 K temperature
range, the experimental curves can be modeled using the
theoretical behavior of two isolated distorted octahedral
cobalt(II) ions.35 The best fit of the data obtained by full-
matrix diagonalization of the appropriate spin Hamiltonian gave
λ = −128 cm−1, D = −391 cm−1, and α = 0.99 for 1a; λ = −179
cm−1, D = −565 cm−1, and α = 0.76 for 1b; and λ = −168 cm−1,
D = −494 cm−1, and α = 0.79 for 3, where λ, D, and α are the
spin orbit coupling constant, the distortion parameter, and the
orbital reduction factor, respectively. The theoretical curves
reproduce quite well the experimental data over the whole
range of temperature showing that there is no interaction
between the cobalt ions within the helicates. For the bis-nickel
helicate 2, the χMT value remains constant in the 300−100 K
temperature range and is equal to 2.2 cm3·K·mol−1, which is
close to the expected one for two magnetically isolated Ni(II)
ions. Below 100 K, the value of χMT slowly decreases to reach a
value of 1.9 cm3·K·mol−1 at 2 K. This behavior could be
interpreted either by a very weak interaction between the two
Figure 5. Crystal packing of 3 showing the H-bond structuring
(orange dotted lines) into a homochiral 3D network; H atoms are
omitted for clarity; dinuclear complexes are schemed with an
intramolecular Co−Co bound pictured in purple in the bc plane;
benzyl rings are omitted for clarity in the representation of the ab
plane.
same molecular unit displaying a different H-bonding mode to
two or three adjacent molecules (Figure 6). Triethylammonium
cations also form hydrogen bonds with the carbonyl groups of
the ligand and are located in the cavities of the distorted
hexagonal architecture defined by the metallic complexes
(Figure S2 in the Supporting Information). The intermolecular
N(H)···O distances found in the structure range from
Figure 6. Schematic view of the N(H)···O bonds (orange dotted lines) between adjacent molecules in 3; H atoms are omitted for clarity; the
dinuclear complex is schemed with an intramolecular Co−Co bond pictured in purple.
4959
dx.doi.org/10.1021/cg300877r | Cryst. Growth Des. 2012, 12, 4955−4962

Crystal Growth & Design
Article
deprotonation of the mpba ligands completely changes the
magnetic properties of the dinuclear compounds from
ferromagnetic coupling to an uncoupled system, and this is a
perfect illustration that the important parameter in the
exchange interaction is not primarily the metal−metal distance
but the existence of molecular orbitals able to transmit the
interaction.
DISCUSSION
■
Unexpectedly, solvothermal synthesis using polyoxalamide
ligands has led to new coordination compounds with ligands
bearing double H-bonds sites. These complexes are built from
metal ions and partially protonated ligands comprising oxamide
anions. Oxamide anions are known to be flexible building
blocks for hydrogen-bonded architectures.38 Indeed, the
compounds presented in this paper are interesting examples
of networks built simultaneously from coordination complexes
and complementary hydrogen bonding sites.39−41 In inorganic
crystal engineering, this strategy is, in theory, very effective. The
geometrical preferences of the coordination sphere of the metal
ions (square planar, octahedral, etc.) induce a precise
orientation of hydrogen-bonding groups that can also
propagate geometrical information, which would lead, in
principle, to predictable results. In practice, the situation is
less clear. For instance, with 2,2-bisimidazolato tris-chelate
complex building blocks, the directions of the H-bonds are
coplanar and separated by an angle of 120°, so the obvious
construction is a honeycomb network, and some results
confirm this expectation.42 However, there are several examples
of more complicated structures that include interpenetrating
nets42 and achiral,43 racemate,44 and chiral 3D structures.45
Even, 0D, 1D zigzag, and 1D helical structures were
obtained.43,46 Switching from one structure to another depends
on other subtle intermolecular interactions, such as the role of
the counterions or the control of the chirality of the metal-ion
coordination sphere. Therefore, whatever the type of
connection used to form a network either by hydrogen bonds
or by coordination bonds, if tris-chelate complex building
blocks appear as a reliable way to obtain self-organized 2D or
3D nets, the final dimensionality remains difficult to predict.
This is especially true for 3D nets because they seem difficult to
achieve although it is not completely unexpected since a basic
unit bearing six connecting points is necessary to build a 3D net
by translation.47,48 Consequently, a tris-chelate building block
with only three connecting points does not meet this
requirement. Since the basic unit must be connected to six
others, it is then compulsory to have at least four tris-chelates in
the basic unit in order to build a three-connected 3D net, that
is, a tetranuclear complex as building block (Scheme 1). In the
case of monomeric metal complexes, only an octahedral
complex with six bis-monodentate ligands like hexacyanome-
tallates is able to build a 3D network.49,50 For dinuclear
building blocks, it is necessary to have a connectivity of four at
each node to obtain 3D networks, and the helicates described
in this paper meet this requirement.
Figure 7. (a) Plots of χMT vs T measured from 300 to 2 K at 1 kOe for
◆
●
▲
▼
1a ( ), 1b ( ), 2 ( ), and 3 ( ) and respective fits in solid lines:
1a, blue; 1b, green; 2, red; 3, orange. (b) Plots of M vs H measured
◆
●
▲
▼
from 0 to 7 T at 2 K for 1a ( ), 1b ( ), 2 ( ), and 3 ( ).
Ni(II) ions within the helicate or by the effect of the zero field
splitting of the individual Ni(II) ions. The best data fits gave g =
2.1, J = −0.6 cm−1 for the first model and gNi = 2.1 and DNi
=
−6.5 cm−1 for the second one. The agreement factor R =
[(χMT)exp − (χMT)calcd]2/[(χMT)exp]2 is slightly better for the
first model with values equal to R = 10−4 and 2 × 10−4,
respectively. However, the lack of interaction in the cobalt-
based helicates leads us to favor the second model.
The lack of interaction in these helicates contrasts with the
results obtained for the cobalt and nickel mesocates, in which
the (mpba)4− ligand is deprotonated, where a weak but
noticeable ferromagnetic interaction is observed with J = 1
cm−1 and J = 3.6 cm−1, respectively.20,36 This difference in the
magnetic behavior is explained by the relative orientation of
oxamate groups with the benzene rings. In the helicate family,
the oxamate groups are almost in the same plane as the
benzene cycles. As a consequence, the interaction between the
magnetic orbitals of the cobalt or nickel ions and the π orbitals
of the benzene rings is very weak, and there is no efficient
pathway with the σ bonds of the benzene rings. In contrast, in
the mesocates the oxamate groups are perpendicular to the
benzene ring; this conformation allows a σ overlap between the
eg magnetic orbitals and the lone pair of the deprotonated
amide nitrogen atom, which are in strong interaction with the
extended π conjugated bond system of the phenylene spacers.37
This situation leads to a quite large spin polarization effect
through the π orbitals of the benzene rings, and despite a
metal−metal distance that goes up 6.85 Å, which is actually
longer than in the case of the helicates, there is a magnetic
interaction in the mesocate family. The simple protonation or
Actually, these new helicate complexes bearing six double H-
bonds sites illustrate nicely the “recipe” A.F. Wells proposed in
the 1950s to build a 3D four-connected net,51,52 “The simplest
planar 3-connected net is the hexagonal net with 2 points in the
repeat unit. By connecting alternate points to points in layers
above and below there arises the simplest three- dimensional 4-
connected net.” This description corresponds to the exact
structure of compound 1b. Furthermore, owing to the structure
4960
dx.doi.org/10.1021/cg300877r | Cryst. Growth Des. 2012, 12, 4955−4962

Crystal Growth & Design
Article
study is thus promising. Indeed the use of solvothermal
conditions combined with the rich family of ligands that
oxamate derivatives constitute clearly opens a new synthetic
route toward multidimensional molecular magnetic materials.
Scheme 1. Simplest Building Blocks Providing the Six
Necessary Connecting Points To Build 3D Nets: (a)
Mononuclear, (b) Dinuclear, and (c, d) Tetranuclear
ASSOCIATED CONTENT
* Supporting Information
■
S
X-ray crystallographic files for 1b and 3 in CIF format and
figures showing experimental X-ray powder diffraction patterns
for 1a, 1b, and 2 and calculated patterns for 1a and 1b, and
intermolecular hydrogen bonds between the bimetallic complex
+
and its HNEt3 counterions in 3. This information is available
free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*Phone: +33 1 44 27 55 62. Fax: +33 1 44 27 38 41. E-mail:
yves.journaux@upmc.fr.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Minister
̀
e de la Recherche et
de l’Enseignement Super
de la Recherche Scientifique (CNRS).
́
ieur (MRES) and the Centre National
REFERENCES
■
(1) Aakeroy, C. B.; Champness, N. R.; Janiak, C. CrystEngComm
2010, 12, 22−43.
(2) Robson, R. J. Chem. Soc., Dalton Trans. 2000, 3735−3744.
(3) Fujita, M. Chem. Soc. Rev. 1998, 27, 417−425.
(4) Swiegers, G. F.; Malefetse, T. J. Chem. Rev. 2000, 100, 3539−
3539.
(5) Chakrabarty, R.; Mukherjee, P. S.; Stang, P. J. Chem. Rev. 2011,
111, 6810−6918.
of the helicate, the resulting net in 1b was perfectly predictable.
However, it should be noted that the structure of 3 with the
entanglement of hydrogen bonds that results from a temper-
ature effect was not anticipated even though the same initial
synthon is present. To summarize, the dinuclear triple-stranded
helicates or mesocates are perfect building blocks to obtain 3D
networks because they offer the six required connecting points.
It is worth noting that the polynuclear complexes {Zn4O-
(RCO2)6} and {Cr3O(RCO2)6} which are used as SBUs
(secondary building units) in 3D MOF 553 and MIL 10154 also
possess six connecting points.
(6) Long, J. R.; Yaghi, O. M. Chem. Soc. Rev. 2009, 38, 1213−1214.
(7) Murray, L. J.; Dinca, M.; Long, J. R. Chem. Soc. Rev. 2009, 38,
1294−1314.
(8) Sumida, K.; Rogow, D. L.; Mason, J. A.; McDonald, T. M.; Bloch,
E. D.; Herm, Z. R.; Bae, T.-H.; Long, J. R. Chem. Rev. 2012, 112, 724−
781.
(9) Allendorf, M. D.; Bauer, C. A.; Bhakta, R. K.; Houk, R. J. T.
Chem. Soc. Rev. 2009, 38, 1330−1352.
(10) Cui, Y.; Yue, Y.; Qian, G.; Chen, B. Chem. Rev. 2012, 112,
1126−1162.
(11) Zhang, W.; Xiong, R.-G. Chem. Rev. 2012, 112, 1163−1195.
(12) Kurmoo, M. Chem. Soc. Rev. 2009, 38, 1353−1379.
(13) Coffman, R. E.; Buettner, G. R. J. Phys. Chem. 1979, 83, 2387−
2392.
CONCLUSION
■
Solvothermal synthesis has successfully led to novel chiral
helicate oxamate-based dinuclear complexes where supra-
molecular H-bond networking in the solid influenced by the
synthesis temperature leads to 3D diamond-like or 3D chiral
structures. From these results obtained fortuitously, it is
possible to conclude that, as in the synthesis of MOF, the
use of polynuclear metal complexes as building blocks is a good
strategy to build 3D networks in a predictable and controlled
way. To the best of our knowledge, the strategic use of discrete
polymetallic coordination compounds bearing hydrogen bond
sites specifically designed to construct networks rationally has
not been explored so far in crystal engineering, and the present
results open a synthetic route that is worth following. There
had been no example of oxamate-based coordination
compounds obtained solvothermally, and this preliminary
(14) Pardo, E.; Carrasco, R.; Ruiz-Garcia, R.; Julve, M.; Lloret, F.;
Munoz, M. C.; Journaux, Y.; Ruiz, E.; Cano, J. J. Am. Chem. Soc. 2008,
130, 576−585.
(15) Fernandez, I.; Ruiz, R.; Faus, J.; Julve, M.; Lloret, F.; Cano, J.;
Ottenwaelder, X.; Journaux, Y.; Munoz, M. C. Angew. Chem., Int. Ed.
2001, 40, 3039−3042.
(16) Dul, M. C.; Pardo, E.; Lescouezec, R.; Chamoreau, L. M.;
Villain, F.; Journaux, Y.; Ruiz-Garcia, R.; Cano, J.; Julve, M.; Lloret, F.;
Pasan, J.; Ruiz-Perez, C. J. Am. Chem. Soc. 2009, 131, 14614−14615.
(17) Pardo, E.; Ferrando-Soria, J.; Dul, M.-C.; Lescouezec, R.;
̈
Journaux, Y.; Ruiz-García, R.; Cano, J.; Julve, M.; Lloret, F.; Canadillas-
̃
Delgado, L.; Pasan
12851.
́
, J.; Ruiz-Per
́
ez, C. Chem.Eur. J. 2010, 16, 12838−
(18) Dul, M. C.; Ottenwaelder, X.; Pardo, E.; Lescouezec, R.;
Journaux, Y.; Chamoreau, L. M.; Ruiz-Garcia, R.; Cano, J.; Julve, M.;
Lloret, F. Inorg. Chem. 2009, 48, 5244−5249.
4961
dx.doi.org/10.1021/cg300877r | Cryst. Growth Des. 2012, 12, 4955−4962

Crystal Growth & Design
Article
(19) Pereira, C. L. M.; Pedroso, E. F.; Stumpf, H. O.; Novak, M. A.;
Ricard, L.; Ruiz Garcia, R.; Riviere, E.; Journaux, Y. Angew. Chem., Int.
Ed. 2004, 43, 955−958.
(53) Li, H.; Eddaoudi, M.; O’Keeffe, M.; Yaghi, O. M. Nature 1999,
402, 276.
(54) Fer
́
ey, G.; Mellot-Draznieks, C.; Serre, C.; Millange, F.; Dutour,
J.; Surble,
́
S.; Margiolaki, I. Science 2005, 309, 2040−2042.
(20) Pardo, E.; Cangussu, D.; Dul, M.-C.; Lescouezec, R.; Herson, P.;
Journaux, Y.; Pedroso, E. F.; Pereira, C. L. M.; Munoz, M. C.; Ruiz-
̃
Garcia, R.; Cano, J.; Amoros
Ed. 2008, 47, 4211−4216.
́
, P.; Julve, M.; Lloret, F. Angew. Chem., Int.
(21) Ferrando-Soria, J.; Ruiz-Garcia, R.; Cano, J.; Stiriba, S. E.;
Vallejo, J.; Castro, I.; Julve, M.; Lloret, F.; Amoros, P.; Pasan, J.; Ruiz-
Perez, C.; Journaux, Y.; Pardo, E. Chem.Eur. J. 2012, 18, 1608−
1617.
(22) Laudise, R. A. Chem. Eng. News 1987, 65, 30−43.
(23) Rabenau, A. Angew. Chem., Int. Ed. 1985, 24, 1026−1040.
(24) Ferey, G. Chem. Mater. 2001, 13, 3084−3098.
(25) Rosi, N. L.; Eddaoudi, M.; Kim, J.; O’Keeffe, M.; Yaghi, O. M.
CrystEngComm 2002, 401−404.
(26) Laye, R. H.; McInnes, E. J. L. Eur. J. Inorg. Chem. 2004, 2004,
2811−2818.
(27) Cervera, B.; Sanz, J. L.; Ibanez, M. J.; Vila, G.; Lloret, F.; Julve,
M.; Ruiz, R.; Ottenwaelder, X.; Aukauloo, A.; Poussereau, S.; Journaux,
Y.; Munoz, M. C. J. Chem. Soc., Dalton Trans. 1998, 781−790.
(28) (a) BrukerAXS Inc, Madison, Wisconsin, USA, 1998.
(b) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112. (c) Blessing,
R. H. Acta Crystallogr. 1995, A51, 33.
(29) Blay, G.; Fernandez, I.; Pedro, Jose, R.; Ruiz-García, R.; Munoz,
́
́
̃
M. C.; Cano, J.; Carrasco, R. Eur. J. Org. Chem. 2003, 1627−1630.
(30) Padilla-Martinez, I. I.; Chaparro-Huerta, M.; Martinez-Martinez,
F. J.; Hopfl, H.; Garcia-Baez, E. V. Acta Crystallogr. 2003, E59, O825−
O827.
(31) Tamaki, H.; Zhong, Z. J.; Matsumoto, N.; Kida, S.; Koikawa, M.;
Achiwa, N.; Okawa, H. J. Am. Chem. Soc. 1992, 114, 6974−6979.
(32) Decurtins, S. Inorg. Chim. Acta 1994, 216, 65.
(33) Gruselle, M.; Train, C.; Boubekeur, K.; Gredin, P.; Ovanesyan,
N. Coord. Chem. Rev. 2006, 250, 2491−2500.
(34) O’Keeffe, M.; Peskov, M. A.; Ramsden, S. J.; Yaghi, O. M. Acc.
Chem. Res. 2008, 41, 1782−1789.
̀
(35) Lloret, F.; Julve, M.; Cano, J.; Ruiz-GarcIa, R.; Pardo, E. Inorg.
Chim. Acta 2008, 361, 3432−3445.
(36) Pardo, E.; Morales Osorio, I.; Julve, M.; Lloret, F.; Cano, J.; Ruiz
Garcia, R.; Pasan, J.; Ruiz Perez, C.; Ottenwaelder, X.; Journaux, Y.
Inorg. Chem. 2004, 43, 7594−7596.
(37) Pardo, E.; Ruiz-Garcia, R.; Cano, J.; Ottenwaelder, X.;
Lescouezec, R.; Journaux, Y.; Lloret, F.; Julve, M. Dalton Trans.
2008, 2780−2805.
(38) Aakeroy, C. B.; Hughes, D. P.; Nieuwenhuyzen, M. J. Am. Chem.
Soc. 1996, 118, 10134−10140.
(39) Burrows, A. D.; Chan, C.-W.; Chowdhry, M. M.; McGrady, J. E.;
Mingos, D. M. P. Chem. Soc. Rev. 1995, 24.
(40) Tadokoro, M.; Nakasuji, K. Coord. Chem. Rev. 2000, 198, 205−
218.
(41) Brammer, L. Chem. Soc. Rev. 2004, 33, 476−489.
(42) Tadokoro, M.; Isobe, K.; Uekusa, H.; Ohashi, Y.; Toyoda, J.;
Tashiro, K.; Nakasuji, K. Angew. Chem., Int. Ed. 1999, 38, 95−98.
̈
(43) Ohrstrom, L.; Larsson, K.; Borg, S.; Norberg, S. T. Chem.Eur.
̈
J. 2001, 7, 4805−4810.
(44) Tadokoro, M.; Shiomi, T.; Isobe, K.; Nakasuji, K. Inorg. Chem.
2001, 40, 5476−5478.
(45) Larsson, K.; Ohrstrom, L. CrystEngComm 2003, 5, 222−225.
(46) Tadokoro, M.; Kanno, H.; Kitajima, T.; Shimada-Umemoto, H.;
Nakanishi, N.; Isobe, K.; Nakasuji, K. Proc. Natl. Acad. Sci. U.S.A. 2002,
99, 4950−4955.
(47) Wells, A. F. Acta Crystallogr. 1954, 7, 535−544.
(48) Ohrstrom, L.; Larsson, K. Dalton Trans. 2004, 347−353.
(49) Buser, H. J.; Schwarzenbach, D.; Petter, W.; Ludi, A. Inorg.
Chem. 1977, 16, 2704−2710.
(50) Witzel, M.; Babel, D. Acta Crystallogr. 1984, A40, C216−C216.
(51) Wells, A. F. Acta Crystallogr. 1954, 7, 545−554.
(52) Wells, A. F. Structural Inorganic Chemistry; Clarendon Press:
Oxford, 1984.
4962
dx.doi.org/10.1021/cg300877r | Cryst. Growth Des. 2012, 12, 4955−4962