Influence of the central metal ion in controlling the self-assembly and magnetic properties of 2D coordination polymers derived from [(NiL)2M]2+ nodes (M = Ni, Zn and Cd) (H2L = salen-type di-Schiff base) and dicyanamide spacers

Article abstract of DOI:10.1039/c4dt02823f

Three new 2D coordination polymers (CPs) ²_∞[(NiL)₂Ni(μ_1,5-N(CN)₂)₂]_n (1), ²_∞[(NiL)₂Cd(μ_1,5-N(CN)₂)₂]_n (2) and ²_∞[(NiL)₂Zn(μ_1,5-N(CN)₂)₂]_n (3) have been synthesized by reacting a [NiL] "metalloligand" (where H₂L = N,N′-bis(salicylidene)-1,3-propanediamine) with three different metal(ii) (Ni, Cd and Zn) perchlorates and sodium dicyanamide, with identical molar ratios of the reactants. All three products have been characterized by IR and UV-Vis spectroscopies, elemental analyses, powder and single crystal X-ray diffraction and variable temperature magnetic measurements. The isomorphous compounds 1 and 2 consist of similar [(NiL)₂M(μ_1,5-N(CN)₂)] (M = Ni for 1 and Cd for 2) angular trinuclear units in which two terminal "metalloligands" [NiL] coordinate to the central nickel(ii) (in 1) or cadmium(ii) (in 2) ion through phenoxido oxygen atoms. The μ_1,5-bridging dicyanamido spacers connect the central Ni(ii) or Cd(ii) of one node to terminal Ni(ii) of two different nodes giving rise to 2D CPs. Compound 3 also contains trinuclear units with the same formula as those of 1 and 2: [(NiL)₂M(μ_1,5-N(CN)₂)] (M = Zn in 3). The main differences are that these units are linear in 3 and the dicyanamide spacers link only the nickel atoms of neighbouring nodes. As in 1 and 2, these trinuclear units are connected to four other units via four μ_1,5-bridging dicyanamido ligands, giving rise to 2D CP with a similar topology: a uninodal 4-connected underlying net with the sql (Shubnikov tetragonal plane net) topology and (4⁴·6²) point symbol. The magnetic properties show the presence of moderate intra-trimer antiferromagnetic interactions in 1 (J = -12.9 cm^-1) and weak antiferromagnetic interactions between the terminal Ni(ii) ions in 2 (J = -2.4 cm^-1). In 3 the Ni(ii) ions are well isolated by the central Zn(ii) ion and accordingly, only a very weak antiferromagnetic interaction through the single μ_1,5-bridging dicyanamido ligands is observed (J = -0.44 cm^-1, D = -3.9 cm^-1). This journal is

Full text of DOI:10.1039/c4dt02823f

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Influence of the central metal ion in controlling
the self-assembly and magnetic properties of 2D
coordination polymers derived from [(NiL)2M]2+
nodes (M = Ni, Zn and Cd) (H2L = salen-type
di-Schiff base) and dicyanamide spacers†
Cite this: Dalton Trans., 2015, 44,
1292
Lakshmi Kanta Das,a Carlos J. Gómez-Garcíab and Ashutosh Ghosh*a
2
2
Three new 2D coordination polymers (CPs) ∞[(NiL)2Ni(µ1,5-N(CN)2)2]n (1), ∞[(NiL)2Cd(µ1,5-N(CN)2)2]n (2)
and 2∞[(NiL)2Zn(µ1,5-N(CN)2)2]n (3) have been synthesized by reacting a [NiL] “metalloligand” (where H2L =
N,N’-bis(salicylidene)-1,3-propanediamine) with three different metal(II) (Ni, Cd and Zn) perchlorates and
sodium dicyanamide, with identical molar ratios of the reactants. All three products have been character-
ized by IR and UV-Vis spectroscopies, elemental analyses, powder and single crystal X-ray diffraction and
variable temperature magnetic measurements. The isomorphous compounds 1 and 2 consist of similar
[(NiL)2M(µ1,5-N(CN)2)] (M = Ni for 1 and Cd for 2) angular trinuclear units in which two terminal “metalloli-
gands” [NiL] coordinate to the central nickel(II) (in 1) or cadmium(II) (in 2) ion through phenoxido oxygen
atoms. The μ1,5-bridging dicyanamido spacers connect the central Ni(II) or Cd(II) of one node to terminal
Ni(II) of two different nodes giving rise to 2D CPs. Compound 3 also contains trinuclear units with the
same formula as those of 1 and 2: [(NiL)2M(µ1,5-N(CN)2)] (M = Zn in 3). The main differences are that these
units are linear in 3 and the dicyanamide spacers link only the nickel atoms of neighbouring nodes. As in
1 and 2, these trinuclear units are connected to four other units via four μ1,5-bridging dicyanamido
ligands, giving rise to 2D CP with a similar topology: a uninodal 4-connected underlying net with the sql
(Shubnikov tetragonal plane net) topology and (44·62) point symbol. The magnetic properties show the
presence of moderate intra-trimer antiferromagnetic interactions in 1 (J = −12.9 cm−1) and weak anti-
ferromagnetic interactions between the terminal Ni(II) ions in 2 (J = −2.4 cm−1). In 3 the Ni(II) ions are well
isolated by the central Zn(II) ion and accordingly, only a very weak antiferromagnetic interaction through
the single μ1,5-bridging dicyanamido ligands is observed (J = −0.44 cm−1, D = −3.9 cm−1).
Received 15th September 2014,
Accepted 5th November 2014
DOI: 10.1039/c4dt02823f
www.rsc.org/dalton
CPs are an intriguing class of hybrid crystalline materials
that are constructed by the spontaneous self-assembly of metal
Introduction
In the last decade, the design and construction of coordination ions or clusters (node or connector) and organic ligands
polymers (CPs) have been emerging areas of research due to (linker or spacer) driven by metal–ligand coordination inter-
their structural diversity and vast potential applications in action that extend infinitely to one, two or three dimensions.2
areas such as catalysis, conductivity, porosity, chirality, lumi- In the year 1990, Robson established and illustrated “the node-
nescence, magnetism, spin-transition and non-linear optics.1 and-spacer approach”3 which has been remarkably successful
in producing various CPs with predictable network architec-
tures and desired topologies. This can be easily achieved by
choosing the metal ions according to their appropriate coordi-
aDepartment of Chemistry, University College of Science and Technology, University
of Calcutta, 92, A.P.C. Road, Kolkata-700 009, India. E-mail: ghosh_59@yahoo.com
bInstituto de Ciencia Molecular (ICMol), Universidad de Valencia, C/Catedrático José
nation number and geometry, charge and HSAB behaviour
(the concept of Hard and Soft Acids and Bases), as well as bridg-
ing spacers with suitable denticity, shape, size and flexibility.4
Recently, we and others have shown that oligonuclear nodes
are excellent building blocks in designing novel CPs due to
their higher geometrical flexibility because of the presence of
two or more metal ions.5–8 In this regard, hetero-trinuclear
Beltrán, 2, 46980 Paterna, Valencia, Spain
†Electronic supplementary information (ESI) available: Experimental and simu-
lated X-ray powder diffractograms of the three compounds (Fig. S1), IR spectra
of the three compounds (Fig. S2–S4), 2D coordination network for compound 2
(Fig. S5) and isothermal magnetization of the three compounds (Fig. S6–S8).
CCDC 1022779–1022781 for 1–3. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c4dt02823f
1292 | Dalton Trans., 2015, 44, 1292–1302
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
(3d–3d′–3d, 3d–3d–4f, 3d–3d′–4f, etc.) cationic or neutral com- metalloligands are nearly perpendicular to each other, making
pounds of cyclic as well as acyclic Schiff base ligands5,6,9 its shape unique. The variable-temperature magnetic suscepti-
deserve special mention as they can be used to construct poly- bility measurement of all three polymers is presented in detail.
mers of various dimensionalities and topologies. Moreover,
several important properties like magnetic, optical, redox or
catalytic properties which depend upon metal–metal intra-
Experimental section
Starting materials
and inter-node interactions10 in the resulting CPs can be
modified by the introduction of the hetero-metal ions into the
trinuclear nodes.6,7
All chemicals including salicylaldehyde and 1,3-propanedi-
Recently, we synthesized some hetero-metallic trinuclear amine were purchased from Lancaster and were of reagent
compounds of the general formula {[(ML)2M′(N(CN)2)a]Xb} grade. They were used without further purification.
(where M = Cu(II) or Ni(II), M′ = Co(II), Zn(II) or Cd(II), L = salen-
Caution! Perchlorate salts are potentially explosive. Only a
type di-Schiff base ligand, X = ClO4−, a = 1 or 2 and b = 1 or 0) small amount of the material should be prepared and handled
and used them as nodes to construct CPs.5,6 These trinuclear with care.
compounds are conformationally and coordinatively flexible.
Synthesis of the Schiff base ligand N,N′-bis(salicylidene)-1,3-
Depending upon the coordination geometry of the central propanediamine (H2L) and the “metalloligand” [NiL]. The
metal ions and the coordination mode of the anionic coli- Schiff base ligand was synthesized by standard methods:
gands, the trinuclear compounds can vary from linear to bent 5 mmol of 1,3-propanediamine (0.42 mL) was mixed with
shapes leading to different spatial orientations of the “metallo- 10 mmol of salicylaldehyde (1.04 mL) in methanol (20 mL).
ligand” in the trinuclear species. As a consequence, when such The resulting solution was refluxed for ca. 2 h, and allowed to
species are used as nodes the resulting CPs can be of various cool. The yellow methanolic solution was used directly for
dimensions and topologies. For example, using {[ML]2M′}2+, compound formation. An aqueous solution (20 mL) of
(M = Cu(II) or Ni(II), L = salen-type di-Schiff base ligands and Ni(ClO4)2·6H2O (1.820 g, 5 mmol) and 10 mL of ammonia
M′ = Co(II), Zn(II) or Cd(II)) as nodes and dicyanamide as solution (20%) were added to a methanolic solution of H2L
spacers we succeeded in obtaining species ranging from hexa- (10 mL, 5 mmol) to prepare the “metalloligand”, [NiL] as
nuclear cluster to 1D to 2D to 3D polymers, some of them reported earlier.11
2
representing very rare examples of genuine supramolecular
Synthesis of ∞[(NiL)2Ni(µ1,5-N(CN)2)2]n (1). The precursor
isomers.5 We observed that besides the nature of the central “metalloligand” [NiL] (0.642 g, 2 mmol) was dissolved in
hetero-metal atoms, the dimension and topology of the coordi- methanol (20 mL) and then an aqueous solution (1 mL) of
nation network depend also upon the coordination numbers Ni(ClO4)2·6H2O (0.364 g, 1 mmol) followed by an aqueous
of the terminal “metalloligands” and the different bridging solution (1 mL) of sodium dicyanamide (0.180 g, 2 mmol)
modes and spatial orientation of the dicyanamide spacer.
were added to this solution. The mixture was stirred for 1 h at
Herein, we report the synthesis and structural features of room temperature, producing a suspension containing a green
2
2
three new 2D CPs ∞[(NiL)2Ni(µ1,5-N(CN)2)2]n (1), ∞[(NiL)2Cd- solid. This was filtered off, washed with a methanol–water
2
(µ1,5-N(CN)2)2]n (2) and ∞[(NiL)2Zn(µ1,5-N(CN)2)2]n (3) (where mixture and dried to give 1. The filtrate was allowed to stand
H2L = N,N′-bis(salicylidene)-1,3-propanediamine) assembled overnight under an open atmosphere resulting in the for-
from the [NiL] “metalloligand”, the dicyanamide spacer and mation of green prismatic shaped X-ray quality single crystals
metal salts of Ni(II), Cd(II) and Zn(II). Interestingly, the trinuc- of 1. These crystals were washed with a methanol–water
lear nodes [(NiL)2M]2+ (M = Ni(II) for 1 and Cd(II) for 2) of 1 mixture and dried in a desiccator containing anhydrous CaCl2
and 2 adopt a bent molecular shape whereas the trinuclear to give the second crop of 1, and then characterized by elemen-
node, [(NiL)2Zn]2+ in 3 is linear (Scheme 1), in which two tal analysis, spectroscopic methods and X-ray diffraction.
Compound 1: Yield: 0.773 g, 89% (including the green pre-
cipitate and the crystalline compound). Anal. calc. for
C38H32Ni3N10O4: C 52.53, H 3.71, N 16.12; found: C 52.69,
H 3.54, N 16.40%. UV/Vis: λmax(MeOH) = 581, 406 and 356 nm
and λmax (solid, reflectance) = 860, 593 and 378 nm. IR (KBr
pellet, cm−1): ν(CvN) 1623 and ν(N(CN)2−) 2180, 2242, 2305.
2
Synthesis of ∞[(NiL)2Cd(µ1,5-N(CN)2)2]n (2). Compound 2
was synthesized in a similar way to 1 but using anhydrous
Cd(ClO4)2 instead of Ni(ClO4)2·6H2O. The precursor “metalloli-
Scheme 1 Bent and linear trinuclear nodes are formed depending gand” [NiL] (0.642 g, 2 mmol) was dissolved in methanol
on the coordination geometry of the central hetero-metal ions (M).
(a) Octahedral environment of M (M = Ni(II) and Cd(II)) results in a bent
node in which two “metalloligands” are almost parallel to each other. (b)
Tetrahedral environment of M (M = Zn(II)) results in a linear node where
(20 mL) and then an aqueous solution (1 mL) of anhydrous
Cd-(ClO4)2 (0.311 g, 1 mmol) followed by an aqueous solution
(1 mL) of sodium dicyanamide (0.180 g, 2 mmol) were added
to this solution. The solution was stirred for 1 h at room
temperature. Here also, a green product separated out during
two “metalloligands” are nearly perpendicular to each other. Here, H2L
is N,N’-bis(salicylidene)-1,3-propanediamine and X is dicyanamide.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 1292–1302 | 1293

View Article Online
Paper
Dalton Transactions
stirring. It was collected by filtration, washed with a metha- Crystallographic data collection and refinement
nol–water mixture and dried to give 2. The second crop of 2 as
green rhombic shaped X-ray quality single crystals was
obtained by the slow evaporation of the filtrate in air. The crys-
tals were washed with a methanol–water mixture and dried in
a desiccator containing anhydrous CaCl2.
Compound 2: Yield: 0.785 g, 85% (including the green pre-
cipitate and the crystalline compound). Anal. calc. for
Suitable single crystals of each compound were mounted on a
Bruker-AXS SMART APEX II diffractometer equipped with a
graphite monochromator and Mo-Kα (λ = 0.71073 Å) radiation.
The crystals were positioned at 60 mm from the CCD. Frames
(360) were measured with a counting time of 5 s. The struc-
tures were solved by the Patterson method using the SHELXS
97 program. The non-hydrogen atoms were refined with aniso-
C38H32Ni2N10O4Cd: C 49.47, H 3.50, N 15.18; found: C 49.49,
tropic thermal parameters. The hydrogen atoms bonded to
H 3.69, N 15.09%. UV/Vis: λmax(MeOH) = 581, 407 and 360 nm
carbon were included in geometric positions and given
thermal parameters equivalent to 1.2 times those of the atom
to which they were attached. Successful convergence was indi-
cated by the maximum shift/error of 0.001 for the last cycle of
and λmax(solid, reflectance) = 854, 600 and 372 nm. IR
(KBr pellet, cm−1): ν(CvN) 1624 and ν(N(CN)2−) 2178, 2240,
2310.
2
Synthesis of ∞[(NiL)2Zn(µ1,5-N(CN)2)2]n (3). Compound 3
the least squares refinement. Absorption corrections were
was prepared in a similar way to 1 and 2 but using Zn
carried out using the SADABS program.13 All the calculations
(ClO4)2·6H2O. The precursor “metalloligand” [NiL] (0.642 g,
were carried out using SHELXS 97,14 SHELXL 97,15 PLATON
2 mmol) was dissolved in methanol (20 mL) and then an
99,16 ORTEP-3217 and WINGX systems ver-1.64.18 Data collec-
aqueous solution (1 mL) of Zn(ClO4)2·6H2O (0.372 g, 1 mmol),
tion, structure refinement parameters and crystallographic
followed by an aqueous solution (1 mL) of sodium dicyana-
data for all three compounds are given in Table 1.
mide (0.180 g, 2 mmol), were added to this solution. The solu-
tion was stirred for 1 h at room temperature, producing a light
blue microcrystalline product that was collected by filtration,
washed with a methanol–water mixture and dried to furnish 3.
The second crop of 3 as blue rhombic shaped X-ray quality
Results and discussion
Synthesis, IR and UV-Vis spectra of the compounds
single crystals was obtained by the slow evaporation of the fil- The Schiff-base ligand N,N′-bis(salicylidene)-1,3-propanedi-
trate in air. The crystals were washed with a methanol–water amine (H2L) and its corresponding Ni(II) compounds [NiL] were
mixture and dried in a desiccator containing anhydrous CaCl2. synthesized according to the reported procedure.11 The [NiL]
Compound 3: Yield: 0.683 g, 78% (including the light blue “metalloligand” on reaction with three different metal(II)
precipitate and the crystalline compound). Anal. calc. for (Ni, Cd and Zn) perchlorate salts and sodium dicyanamide, in
C38H32Ni2N10O4Zn: C 52.13, H 3.68, N 16.00; found: C 51.96, a 2 : 1 : 2 molar ratio in methanol–water medium (10 : 1, v/v),
2
H 3.91, N 16.16%. UV/Vis: λmax(MeOH) = 583, 406 and 362 nm yielded three 2D CPs
and λmax(solid, reflectance) = 1026, 566, 420 and 357 nm. IR
∞[(NiL)2Ni(µ1,5-N(CN)2)2]n (1),
2
2
∞[(NiL)2Cd(µ1,5-N(CN)2)2]n (2) and ∞[(NiL)2Zn(µ1,5-N(CN)2)2]n
(KBr pellet, cm−1): ν(CvN) 1624 and ν(N(CN)2−) 2180, 2248, (3) (Scheme 2). All three polymers are constructed by joining
2323.
the [(NiL)2M]2+ nodes with the help of dicyanamide spacers.
However, the trinuclear nodes are angular in CPs 1 and 2 and
linear in 3, resulting in interesting differences in the networks
of both polymers (Scheme 2). The phase purity of these three
Physical measurements
Elemental analyses (C, H and N) were performed using a compounds (1–3) was confirmed by their powder XRD pattern
Perkin-Elmer 2400 series II CHN analyzer. IR spectra in KBr (Fig. S1†).
pellets (4000–500 cm−1) were recorded using a Perkin-Elmer
All three compounds were initially characterized by IR spec-
RXI FT-IR spectrophotometer. Electronic spectra in methanol troscopy. The precursor “metalloligand” [NiL] does not
and in the solid state (1200–300 nm) were recorded on a contain any dicyanamide ion and it is included in the three
−
Hitachi U-3501 spectrophotometer. Powder X-ray diffraction compounds. The presence of dicyanamido ligands N(CN)2 in
patterns were recorded on a Bruker D-8 advance diffractometer the compounds is confirmed by the presence of three
operated at 40 kV voltage and at 40 mA current and calibrated sharp and strong characteristic stretching frequencies in the
with a standard silicon sample, using Ni-filtered Cu-Kα (λ = 2323–2178 cm−1 region. These bands are attributed to νas
+
0.15406 nm) radiation. Magnetic susceptibility measurements νs(C–N) combination modes, νas(CuN) and νs(CuN)
were carried out in the temperature range 2–300 K with an vibrations,19 and are observed at 2305, 2242 and 2180 cm−1 in
applied magnetic field of 0.1 T on polycrystalline samples of 1, 2310, 2240 and 2178 cm−1 in 2 and 2323, 2248 and
1–3 (with masses of 71.32, 31.33 and 46.76 mg) with a 2180 cm−1 in 3. These bands are shifted towards higher fre-
Quantum Design MPMS-XL-5 SQUID susceptometer. The quencies with respect to the free dicyanamide ion that shows
susceptibility data were corrected for the sample holders pre- the same pattern of bands at 2291, 2231 and 2173 cm−1, indi-
viously measured using the same conditions and for the cating the bridging coordination mode of the dicyanamide ion
diamagnetic contributions of the salt as deduced by using in these three compounds.20 In addition, a strong and sharp
Pascal’s constant tables12 (χdia = −440.98 × 10−6, −452.98 × band due to the azomethine ν(CvN) group of the Schiff base
10−6 and −443 × 10−6 emu mol−1 for 1–3, respectively).
appears at 1623, 1624 and 1624 cm−1 for compounds 1–3,
1294 | Dalton Trans., 2015, 44, 1292–1302
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Table 1 Crystal data and structure refinement of compounds 1–3
Compounds
1
2
3
Formula
M
Crystal system
Space group
C
38H32Ni3N10O4
C38H32Ni2N10O4Cd
922.53
Monoclinic
P21/c
C38H32Ni2N10O4Zn
875.51
Orthorhombic
Pbcn
868.81
Monoclinic
P21/c
a/Å
b/Å
c/Å
α/°
16.236(5)
10.705(5)
21.821(5)
90
16.713(5)
10.748(5)
21.986(5)
90
14.9315(7)
19.8367(9)
12.5335(6)
90
β/°
107.244(5)
90
3622(2)
4
1.593
1.601
109.293(5)
90
3728(2)
4
1.644
1.620
90
90
3712.3(3)
4
1.566
1.701
1792
γ/°
V/Å3
Z
Dc/g cm−3
µ/mm−1
F (000)
1784
1864
R(int)
0.039
52 678
8362
6531
0.0300, 0.0778
293
0.040
0.038
40 281
3382
2757
0.0268, 0.0696
293
Total reflections
Unique reflections
I > 2σ(I)
33 780
33 780
5899
0.0311, 0.0855
293
R1 a, wR2
b
Temp. (K)
GOFc on F2
1.02
1.06
1.07
4 1/2
a R1 = ∑||Fo| − |Fc||/∑|Fo|. b wR2 (Fo2) = [∑[w(Fo2 − Fc2)2/∑w Fo
]
.
c GOF = [∑[w(Fo2 − Fc2)2/(Nobs–Nparams)]1/2
.
Scheme 2 Construction of 2D coordination polymers.
respectively (Fig. S2–S4†). The rest of the spectral pattern and
band positions of the respective compounds and “metallo-
ligand” are very similar.
The UV-Vis spectra of the compounds in methanolic solu-
tion and their solid state reflectance spectra are shown in
Fig. 1. The electronic spectra of all the compounds in metha-
nol are almost identical, but they differ appreciably in the
solid state, especially in the visible region. Thus, they show a
sharp single absorption band near 356, 360 and 362 nm in
methanol and 378, 372 and 357 nm in the solid state for 1–3
respectively, attributed to ligand-to-metal charge transfer tran-
sitions. Besides this band, a broad absorption band (ν1) is
observed in the visible region at 581, 581 and 583 nm along
with a less intense shoulder (ν2) at 406, 407 and 406 nm in
Fig. 1 UV-Vis spectra of compound 1–3 and [NiL] (a) in methanolic
solution and (b) in the solid state.
methanol for 1–3 respectively, while the “metalloligand” [NiL]
shows band maxima (ν1) at 592 nm along with a less intense
shoulder (ν2) at 406 nm. This band is typical of d–d transitions
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 1292–1302 | 1295

View Article Online
Paper
Dalton Transactions
of Ni(II) ions with a square planar environment. The electronic
spectrum for a four coordinate nickel(II) compound with a
square planar geometry is expected to exhibit absorption
bands near 610 (ν1) and 500 nm (ν2) corresponding to the spin
Table 2 Bond distances (Å) and angles (°) for compounds 1 and 2
Compound 1
Compound 2
Ni(1)–O(11)
Ni(1)–O(31)
Ni(1)–N(19)
1.9813(18)
2.0036(17)
2.039(2)
2.014(2)
2.087(2)
2.086(2)
2.0532(16)
2.0192(16)
2.1966(18)
2.032(2)
2.026(2)
1.9778(16)
2.0199(17)
2.014(2)
2.028(2)
2.036(2)
80.92(6)
89.04(8)
166.96(8)
95.01(8)
169.80(7)
91.74(7)
95.86(7)
97.78(9)
86.71(8)
96.47(8)
77.31(6)
84.93(6)
73.62(6)
97.13(7)
169.97(7)
156.98(5)
81.52(6)
98.27(7)
98.63(7)
79.47(5)
98.38(7)
96.42(7)
170.61(7)
96.81(7)
92.51(8)
84.90(5)
89.52(6)
170.76(7)
94.76(7)
163.64(7)
87.46(7)
96.47(7)
96.43(7)
99.31(8)
91.25(8)
2.033(2)
1.982(2)
2.043(2)
2.031(3)
2.038(3)
2.531(2)
2.190(2)
2.215(2)
2.362(3)
2.246(3)
2.242(3)
2.023(2)
2.001(3)
2.033(3)
2.056(2)
2.090(3)
88.70(8)
86.57(9)
166.29(10)
95.83(10)
174.73(9)
88.34(8)
92.04(9)
95.72(10)
90.76(10)
97.65(11)
72.56(7)
74.08(7)
63.92(9)
98.27(9)
164.89(9)
142.48(7)
77.79(9)
99.19(10)
101.65(10)
72.45(8)
102.13(10)
106.19(10)
162.11(11)
101.46(11)
96.43(11)
84.60(11)
90.66(11)
172.23(9)
91.33(11)
168.78(12)
88.20(12)
93.46(12)
96.94(12)
96.81(12)
86.20(12)
allowed d–d transitions Blg←lAg and B3g←lAg respectively.21
The observation of the ν1 and ν2 bands confirms the square
planar environment around Ni(II) in methanol solutions.
However, in the solid state, the electronic spectra of 1 and 2 in
the visible regions show absorption bands at 593 and 600 nm,
respectively, which are associated to weaker ones centred at
860 and 854 nm. The electronic spectrum for a five coordinate
nickel(II) compound with a square-pyramidal geometry is
expected to exhibit absorption bands near 1150 (ν1), 950 (ν2),
and 600 nm (ν3), corresponding to the spin allowed d–d tran-
l
l
Ni(1)–N(23)
Ni (1)–N(5)a
Ni(2)/Cd(2)–O(11)
Ni(2)/Cd(2)–O(31)
Ni(2)/Cd(2)–O(41)
Ni(2)/Cd(2)–O(61)
Cd(2)/Ni(2)–N(1)
Ni(2)/Cd(2)–N(6)
Ni(3)–O(41)
Ni(3)–O(61)
Ni(3)–N(49)
Ni(3)–N(53)
3
3
3
sitions T2g(F)←3A2g(ν1), T1g(F)←3A2g(ν2) and T1g(P)←3A2g(ν3),
respectively.22 In the present case, the ν1 band cannot be
located. The observation of the ν2 and ν3 bands suggest that
the Ni(II) present a square-pyramidal geometry in the solid
state. On the other hand, 3 exhibits three distinct bands
at 420, 566 and 1026 nm, which can be assigned to the spin-
allowed d–d transitions 3T1g(P)←3A2g,3T1g(F)←3A2g and
3T2g(F)←3A2g respectively. These values agree with the litera-
ture values for octahedral Ni(II) compounds.23 Thus the
spectral data in the solid state agree with the X-ray structural
data (see below).
Ni(3)–N(10)b
O(11)–Ni(1)–O(31)
O(11)–Ni(1)–N(19)
O(11)–Ni(1)–N(23)
O(11)–Ni(1)–N(5)a
O(31)–Ni(1)–N(19)
O(31)–Ni(1)-N(23)
O(31)–Ni(1)–N(5)a
N(19)–Ni(1)–N(23)
N(19)–Ni(1)–N(5)a
N(23)–Ni(1)–N(5)a
O(11)–Ni(2)/Cd(2)–O(31)
O(11)–Ni(2)/Cd(2)–O(41)
O(11)–Ni(2)/Cd(2)–O(61)
O(11)–Ni(2)/Cd(2)–N(1)
O(11)–Ni(2)/Cd(2)–N(6)
O(31)–Ni(2)/Cd(2)–O(41)
O(31)–Ni(2)/Cd(2)–O(61)
O(31)–Ni(2)/Cd(2)–N(1)
O(31)–Ni(2)/Cd(2)–N(6)
O(41)–Ni(2)/Cd(2)–O(61)
O(41)–Ni(2)/Cd(2)–N(1)
O(41)–Ni(2)/Cd(2)–N(6)
O(61)–Ni(2)/Cd(2)–N(1)
O(61)–Ni(2)/Cd(2)–N(6)
N(1)–Ni(2)/Cd(2)–N(6)
O(41)–Ni(3)–O(61)
O(41)–Ni(3)–N(49)
O(41)–Ni(3)–N(53)
O(41)–Ni(3)–N(10)b
O(61)–Ni(3)–N(49)
O(61)–Ni(3)–N(53)
O(61)–Ni(3)–N(10)b
N(49)–Ni(3)–N(53)
N(49)–Ni(3)–N(10)b
N(53)–Ni(3)–N(10)b
Structure description of the compounds
The structure of 1 is shown in Fig. 2 together with the atomic
numbering scheme. Bond parameters within the metal coordi-
nation spheres are given in Table 2. The neutral trinuclear
entity of this compound [(NiL)2Ni(µ1,5-N(CN)2)2] contains two
[NiL] “metalloligands” (where H2L is N,N′-bis(salicylidene)-1,3-
propanediamine), one Ni(II) ion and two µ1,5-dicyanamido
(dca−) moieties. The Ni(II) ions of the [NiL] units (Ni1 and Ni3)
present
a penta-coordinated square pyramidal geometry
a
b
Symmetry transformation = −x, −1/2 + y, 1/2 − z and
= 1−x,
a
b
−1/2 + y, 1/2 − z for 1 and = 2 − x, −1/2 + y, 3/2 − z and = 1 − x,
−1/2 + y, 3/2 − z for 2.
(Ni1 could also be viewed as an elongated octahedron if we
consider the weak Ni1–O61 bond distance of 2.441(6) Å). The
basal planes around these nickel atoms are constituted by the
two imine nitrogen atoms and two phenoxido oxygen atoms
from the tetradentate Schiff base ligand. The axial positions of
both nickel atoms are occupied by the terminal nitrogen
atoms of dca− ligands. The basal and axial Ni–O and Ni–N
bond distances in both nickel atoms are very similar (Table 2).
The four donor atoms in the equatorial plane show r.m.s. (root
Fig. 2 The coordination environment of metal ions in the trinuclear
unit of compound of 1 with ellipsoids at 30% probability (symmetry
transformation = −x, −1/2 + y, 1/2 − z and = 1 − x, −1/2 + y, 1/2 − z).
The weak Ni1–O61 bond is shown by an open bond.
a
b
1296 | Dalton Trans., 2015, 44, 1292–1302
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
mean squared) deviations of 0.071 and 0.083 Å for Ni1 and to each other as indicated by the dihedral angle (13.9(5)°)
Ni3, respectively. The metal atoms are shifted by 0.119(1) and between the two N2–Ni–O2 planes. The cadmium atom, Cd(2),
0.191(1) Å, respectively, from their mean plane towards the has a similar distorted octahedral environment to Ni(2)
axially coordinated nitrogen atoms. The Addison parameters in compound 1. The cis [63.92(9)–106.19(10)°] and trans
(τ = 0.047 for Ni1 and 0.119 for Ni3) indicate that distortion [142.48(7)–164.89(9)°] angles also indicate significant distor-
towards trigonal bipyramid is negligible for both metal atoms tions from the ideal octahedral geometry around the cadmium
(τ = 0 for the ideal square pyramid and τ = 1 for the trigonal atom. The Cd⋯Ni distances are 3.269(2) and 3.162(3) Å
bipyramid).24 The dihedral angle between the two N2–Ni–O2 whereas the Ni⋯Ni distance is 3.668(3) Å. As in 1, the Ni1–
planes is 15.78(9)°, indicating that the two “metalloligands” Cd2–Ni3 angle (69.53(3)°) indicates an extremely bent arrange-
are almost parallel to each other.
ment of the three metal atoms in the Ni2Cd unit.
The trinuclear nodes in both compounds
The central Ni2 atom has an octahedral environment
1 and 2
formed by four oxygen atoms from the two chelating “metallo- are angular with Ni–Ni–Ni or Ni–Cd–Ni angles of 77.72(1) or
ligands” and by two terminal cis-dca− ligands with very similar 69.53(3)°, respectively. In both compounds, the two dicyan-
bond lengths (Table 2). The cis [73.62(6)–98.63(7)°] and the amido bridges present a V-shaped geometry with NC–N–CN
trans [156.98(5)–170.61(7)°] angles indicate significant distor- angles of 122.1(2) and 123.6(2)° in 1 and 123.30(1) and
tions from an ideal octahedral geometry. The Ni1⋯Ni2, 121.83(1)° in 2, and occupy two cis positions of the octahedral
Ni2⋯Ni3 and Ni3⋯Ni1 distances are 2.938(2), 3.033(2) and coordination sphere of the central Ni2 or Cd2 atom. These two
3.746(2) Å, respectively. The Ni1–Ni2–Ni3 angle is 77.72(1)° dca− bridges connect each trinuclear unit with the terminal
indicating a bent arrangement of the metal atoms in the Ni3 Ni centres of two different neighbouring trinuclear units. As a
unit.
result, each trinuclear Ni3 or Ni2Cd is connected with four
Compound 2 is isostructural to 1 but with a central Cd(II) neighbouring trinuclear units: with two of them acting as the
ion instead of a Ni(II) (Fig. 3), thus forming a neutral trinuclear dca-donor and with the other two as the dca-acceptor. This
unit of formula [(NiL)2Cd(µ1,5-N(CN)2)2]. As expected, a com- connectivity results in square 2D coordination networks for
parison of the bond lengths and angles between 1 and 2 both compounds (Fig. 4a for 1 and Fig. S5 in ESI† for 2).
shows very small differences (Table 2). The r.m.s. deviations of The topological analysis of this network can be simplified by
the coordinated atoms in the basal planes are 0.103 and considering the centroid of the trinuclear [(NiL)2M]2+ units
0.102 Å around Ni1 and Ni3, respectively. The metal atoms are (M = Ni for 1 and Cd for 2) as 4-connected cluster nodes. This
0.142(2) and 0.077(2) Å from their mean planes. The Addison uninodal 4-connected net features the sql (Shubnikov tetra-
parameters (0.140 for Ni1 and 0.058 for Ni3) also confirm a gonal plane net) topology with the point symbol of (44·62).
slightly distorted square pyramidal geometry around the metal
centers.
Compound 3 contains neutral trinuclear units of formula
[(NiL)2Zn(µ1,5-N(CN)2)2] presenting a crystallographic 2-fold
As observed in 1, the geometry around Ni3 may be viewed axis passing through the central Zn atom (Fig. 5). In this tri-
as a distorted elongated octahedra with an elongated Ni3–O11 nuclear unit, the three metal atoms (two terminal Ni atoms
bond distance of 2.363(2) Å and an axial trans angle of and the central Zn atom) are almost co-linear, in clear contrast
164.54(10)°. The two “metalloligands” are also nearly parallel to the observed bent geometry in the trinuclear units 1 and 2.
The two equivalent terminal Ni(II) ions present an elongated
octahedral geometry where the basal plane is formed by the
two imine nitrogen atoms and two phenoxido oxygen atoms
from one “metalloligand”. These four donors in the equatorial
plane show r.m.s. deviation from their mean plane around the
Ni center of 0.081 Å while the metal atom deviates 0.004(1) Å
from this plane in the direction of the N1 atom. The basal
Ni–O and Ni–N bond distances are very similar (Table 3). The
apical positions are occupied by the terminal nitrogen atoms
from two dicyanamido ligands. The apical Ni–N bond lengths
are slightly longer than the basal Ni–N ones. The axial trans
angle (177.62(8)°) is close to 180°. In contrast to 1 and 2, the
dihedral angle between the two N2–Ni–O2 planes is 67.65(18)°
indicating that the two “metalloligands” are almost perpen-
dicular (Fig. 5).
Another remarkable difference between 1–2 and 3 is the tetra-
hedral environment of the Zn(II) ion which is bonded to four
bridging phenoxido oxygen atoms from two different [NiL]
units. The Zn–O bond lengths are very similar (1.9462(15) and
1.9538(15) Å) forming a distorted-tetrahedron with O–Zn–O
bond angles in the range 84.05(6)–131.45(6)° (Table 3). The
Fig. 3 The coordination environment of metal ions in the trinuclear
unit of compound 2 with ellipsoids at 30% probability (symmetry trans-
a
b
formations = 2 − x, −1/2 + y, 3/2 − z and = 1 − x, −1/2 + y, 3/2 − z).
The weak Ni3–O11 bond is shown by an open bond.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 1292–1302 | 1297

View Article Online
Paper
Dalton Transactions
Table 3 Bond distances (Å) and angles (°) for compound 3
Ni(1)–O(11)
Ni(1)–O(31)
Ni(1)–N(19)
Ni(1)–N(23)
2.0306(15)
2.0566(15)
2.0198(19)
2.020(2)
2.139(2)
2.120(2)
1.9462(15)
1.9538(15)
79.40(6)
94.33(7)
90.65(7)
O(31)–Ni(1)–N(19)
O(31)–Ni(1)–N(23)
O(31)–Ni(1)–N(5)c
N(1)–Ni(1)–N(19)
N(1)–Ni(1)–N(23)
N(1)–Ni(1)–N(5)c
N(19)–Ni(1)–N(23)
N(19)–Ni(1)–N(5)c
N(23)–Ni(1)–N(5)c
O(11)–Zn(1)–O(31)
O(11)–Zn(1)–O(11)d
O(11)–Zn(1)–O(31)d
O(31)–Zn(1)–O(31)d
169.00(7)
90.75(7)
92.75(7)
88.30(8)
90.41(8)
177.62(8)
99.56(8)
91.59(8)
87.27(8)
84.05(6)
112.26(7)
131.45(6)
119.89(6)
Ni(1)–N(1)
Ni(1)–N(5)c
Zn(1)–O(11)
Zn(1)–O(31)
O(11)–Ni(1)–O(31)
O(11)–Ni(1)–N(1)
O(11)–Ni(1)–N(19)
O(11)–Ni(1)–N(23)
O(11)–Ni(1)–N(5)c
O(31)–Ni(1)–N(1)
168.87(8)
88.05(7)
87.79(7)
c
d
Symmetry transformation = 1/2 + x, 1/2 − y, 2 − z and = −x, y,
3/2 − z for 3.
distorted tetrahedral geometry around the Zn(II) ion is
suggested by the dihedral angle of 75.37(14)° between the two
O–Zn–O planes (the dihedral angle is 0° for a perfectly square
planar arrangement and 90° for a perfect tetrahedral arrange-
ment) and confirmed by its τ4 index of 0.77. The τ4 index is
defined as τ4 = [360° − (α + β)]/141°, with α and β (in °) being
the two largest angles around the central metal in the com-
pound with τ4 = 0 for a perfect square planar and τ4 = 1 for a
perfect tetrahedron.25 The Ni1⋯Zn2 and Ni1⋯Ni1 distances
are 3.016(1) and 6.032 Å respectively. The Ni1–Zn2–Ni3 angle
is 179.22(1)°.
The connectivity of the trinuclear units in 3 is established
through V-shaped dicyanamido bridges with NC–N–CN angles
of 119.9(2)° (close to those observed in 1 and 2). However,
these dca− bridges occupy trans positions in the two terminal
Ni atoms, giving rise to a corrugated topology where each
Ni2Zn is connected to four different Ni2Zn trinuclear units
through single µ1,5-dicyanamido bridges between only termi-
nal nickel centers (Fig. 6a). These layers stack in an eclipsed
way along the b-axis (Fig. 6b). The topological analysis of this
structure in 3 assuming the centroid atom of the trinuclear
unit is the same uninodal 4-connected net observed in 1 and
2: Shubnikov tetragonal plane net topology with the point
symbol of (44·62) (Fig. 4b).
Fig. 4 (a) The 2D coordination network in 1 constructed by assembling
the trinuclear [(NiL)2Ni]2+ units through the central and terminal Ni
centres with the dca− bridges. All H atoms are omitted for clarity, Ni =
green, N = blue, O = red, C = grey. (b) Simplified uninodal 4-connected
net with the sql topology and the point symbol of (44·62). Centroids of
the 4-connected trinuclear units are shown as violet balls.
Magnetic properties
Compound 1 shows at room temperature a χmT value of
ca. 3.5 cm3 K mol−1 (χm is the magnetic susceptibility per Ni3
trinuclear unit), which is the expected value for three isolated
Ni(II) S = 1 ions (Fig. 7). When the sample is cooled, χmT shows
a progressive decrease with a smoothening of the slope
between 10 and 5 K where the χmT value is ca. 1.2 cm3 K mol−1
(inset in Fig. 7). This value is the expected one for a S = 1 spin
state. Below ca. 5 K χmT shows a more abrupt decrease reach-
ing a value of ca. 0.6 cm3 K mol−1 at 2 K (inset in Fig. 7). This
result indicates that the three Ni(II) ions in 1 present an anti-
ferromagnetic coupling that results in a S = 1 spin ground
state for the trinuclear unit. At very low temperatures this spin
state presents a zero field splitting (ZFS) and/or inter-trinuclear
Fig. 5 The coordination environment of the metal ions in the structure
of compound 3 with ellipsoids at 30% probability (symmetry = 1/2 + x,
1/2 − y, 2 − z and d = −x,y, 3/2 − z).
c
1298 | Dalton Trans., 2015, 44, 1292–1302
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
antiferromagnetic interactions, through the single dca−
bridges, responsible for the abrupt decrease below ca. 5 K.
Although the structure of the Ni3 compound show that it is a
bent trinuclear unit, in fact, it can be considered as a linear tri-
nuclear unit from the magnetic point of view since there are
no direct bridges connecting the two terminal Ni atoms (if we
neglect the long Ni1–O61 bond distance of 2.441(6) Å). Since
the central Ni2 atom is connected with the two terminal Ni1
and Ni3 atoms through similar double phenoxido bridges, we
can consider, in a first approach, that compound 1 is a linear
symmetric trinuclear unit with only one intra-trinuclear coup-
ling constant (J). Finally, to account for the possible inter-
trinuclear interactions through the single dca− bridges, we
have included an inter-trinuclear term using the molecular
field approximation. Accordingly, we have fit the magnetic pro-
perties of 1 to a linear centrosymmetric S = 1 trinuclear model
(J) with inter-trinuclear interactions (zj ) and a paramagnetic
monomeric S = 1 impurity (c) to account for possible mono-
meric impurities and vacant compounds. This model repro-
duces very satisfactorily the magnetic properties of compound
1 in the 5–300 K temperature range with the following para-
meters: g = 2.127, J = −12.9 cm−1, zj = −0.4 cm−1 and c = 5.9%
(the Hamiltonian is written as H = −JSiSi+1). At very low temp-
eratures (below 5 K) the fit is not good because we have not
included a ZFS of the resulting S = 1 spin state since both para-
meters (ZFS and inter-trinuclear coupling) are similar in mag-
nitude and are strongly correlated, precluding their precise
determination. Note also that, although the two J values are
not equivalent by symmetry, they must be very similar, as
demonstrated by the good agreement between the experi-
mental and theoretical values. The antiferromagnetic coupling
in the linear trinuclear unit leads to a S = 1 ground spin state
as confirmed by the isothermal magnetization at 2 K that
shows a saturation value close to 2.0μB, the expected value for
a S = 1 spin state with g ≈ 2 (Fig. S6†).
Fig. 6 (a) The 2D coordination network in 3 constructed by assembling
the trinuclear [(NiL)2Zn]2+ units through only terminal Ni centres with
the help of µ1,5-dicyanamido linkers. All H atoms are omitted for clarity.
(b) Simplified view of the eclipsed layers stacking in the⋯AA⋯fashion
(only the bridging ligands and the coordinating sphere of the metals are
drawn). Ni = green, Zn = pink, N = blue, O = red and C = grey.
Compound 2 shows at room temperature a χmT value of ca.
2.5 cm3 K mol−1 per Ni2Cd trinuclear unit, the expected value
for two isolated Ni(II) S = 1 ions (Fig. 8). When the sample is
Fig. 8 Thermal variation of χmT for compound 2. Solid and dashed lines
are the best fit to the dimer and monomer S = 1 models, respectively.
The inset shows the low temperature region of the thermal variation
Fig. 7 Thermal variation of χmT for compound 1. Solid line is the best fit
to the model (see text). Inset shows the low temperature region.
of χm.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 1292–1302 | 1299

View Article Online
Paper
Dalton Transactions
cooled, χmT remains constant down to ca. 50 K and below this reach a value of ca. 1.0 cm3 K mol−1 at 2 K. Since the Zn(II)
temperature it shows a progressive decrease to reach a value of ions are diamagnetic, we can consider that the only possible
ca. 0.2 cm3 K mol−1 at 2 K. This behaviour indicates the pres- exchange pathway is the one taking place through the dca−
ence of very weak antiferromagnetic interactions between the single bridges connecting the Ni2Zn trinuclear units. This
two Ni ions in 2. Since the central Cd(II) ion is diamagnetic, exchange pathway gives rise to regular Ni(II) chains as clearly
the Ni(II) ions are magnetically quite well isolated, except for shown in Fig. 6b. Accordingly, we have fit the magnetic data of
the presence of weak Ni–O bonds, connecting both terminal 3 to a regular S = 1 chain model with ZFS.26 Since, a priori
Ni(II) ions in the trinuclear unit. Therefore, we can assume we do not know the sign of D, we have used both models (for
that the decrease observed at low temperature may arise from D positive and D negative). As can be clearly observed in Fig. 9,
a weak antiferromagnetic coupling between the two Ni the model with positive D (g = 2.172, J = −0.47 cm−1 and D =
ions through weak Ni–O–Ni bridges and/or from the ZFS of the 0.8 cm−1, dashed line in Fig. 9, the Hamiltonian is written as
Ni(II) ions. Accordingly, we have used two different models to H = −JSiSi+1) gives a poorer agreement than the model with
fit the magnetic properties: (1) a S = 1 monomer with ZFS and negative D (g = 2.148, J = −0.44 cm−1 and D = −3.9 cm−1, solid
(2) a S = 1 dimer with ZFS. The first model reproduces quite line in Fig. 9) and, accordingly, we can assume that compound
satisfactorily the magnetic data with the following parameters: 3 presents a negative D value. The very weak antiferromagnetic
g = 2.254 and |D| = 29.5 cm−1 (dashed line in Fig. 8). The coupling is also confirmed by the isothermal magnetization at
second model reproduces even better the magnetic data, 2 K that shows a saturation value close to 2.3μB, the expected
especially at low temperatures, with
g
=
2.187,
J
=
one for a S = 1 ion with g = 2.15 (Fig. S8†).
−2.4 cm−1and |D| = 14.8 cm−1 (solid line in Fig. 8, the Hamil-
The weak antiferromagnetic coupling found in compound 1
tonian is written as H = −JS1S2). Although it could be argued is surprising since previous magneto-structural correlations in
that the better agreement is due to the use of an additional similar double oxido bridged Ni(II) clusters indicate that the
fitting parameter (J), the presence of a weak intra-trinuclear coupling is expected to be ferromagnetic when the Ni–O–Ni
antiferromagnetic Ni⋯Ni interaction is confirmed by the pres- bond angle is in the range 90–98°.27 In compound 1 the
ence of a maximum in the χm plot at ca. 10 K (inset in Fig. 8). Ni–O–Ni bond angles are in the range 91.88–98.70° with
As expected, only the dimer model is able to reproduce satis- average values of 92.60° and 95.29° for the Ni1⋯Ni2 and
factorily this maximum. A limitation of the used model is the Ni2⋯Ni3 couplings, respectively. However, the dihedral Ni–O–
fact that the D and J values are strongly correlated, precluding O–Ni angle also plays an important role.28 In compound 1
their reliable determination. In fact, the D value obtained is these dihedral angles are far from 180° (these are 139.35° for
quite high. Note also that we have not considered any inter- Ni1–O11–O31–Ni2 and 157.38° for Ni2–O41–O61–Ni3) and,
trinuclear interaction because the single dca− bridges connect- therefore, the magnetic coupling is expected to be weak and
ing the trinuclear units always link a Ni(II) with a Cd(II) ion. antiferromagnetic, as observed experimentally. In compound 2
Finally, the isothermal magnetization confirms the presence the weak antiferromagnetic coupling found (J = −2.4 cm−1) is
of the antiferromagnetic coupling and shows an almost linear not surprising since the two Ni(II) ions are bridged by a double
behaviour without reaching saturation even at 5 T (Fig. S7†).
asymmetric phenoxido bridge. One of them, Ni3–O11, has a
Compound 3 shows at room temperature a χmT value of long distance of 2.363(2) Å and a Ni1–O11–Ni3 large bond
ca. 2.3 cm3 K mol−1 per Ni2Zn trinuclear unit, the expected angle of 112.89(1)°, which is expected to give rise to a weak
value for two isolated Ni(II) S = 1 ions (Fig. 9). On cooling antiferromagnetic coupling. The second bridge can be neg-
down the sample, χmT remains constant down to ca. 10 K. lected since it presents a very long Ni1–O61 bond distance of
Below this temperature χmT shows a progressive decrease to 2.81(1) Å and a Ni1–O61–Ni3 bond angle close to the
crossing point between ferro- and antiferromagnetic coupling
(97.96(2)°). In compound 3 the very weak antiferromagnetic
coupling found (J = −0.44 cm−1) has to be attributed to
the long NC–N–CN− inter-trinuclear bridge since there is no
Ni⋯Ni intra-trinuclear interactions. This ligand is very well
known to give rise to weak antiferromagnetic interactions,
in agreement29 with the result found in compound 3.
Conclusions
In this paper, using trinuclear {[NiL]2M}2+ nodes derived from
an acyclic tetradentate N2O2 type Schiff base and dicyanamide
spacers, we synthesized three 2D CPs having different net-
works. Among them, the isomorphous compounds 1 and 2 are
Fig. 9 Thermal variation of χmT for compound 3. Solid and dashed lines
are the best fit to the S = 1 regular chain model with D < 0 or D > 0,
constructed by linking bent trinuclear nodes {[NiL]2M}2+ (M =
respectively. The inset shows the low temperature region.
Ni for 1 and Cd for 2) via µ1,5-bridging dicyanamido spacers
1300 | Dalton Trans., 2015, 44, 1292–1302
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
through the central Ni or Cd of one node to terminal Ni
centres of two different nodes. In contrast, the unique 2D CP
of 3 results from linear trinuclear nodes {[NiL]2Zn}2+ which are
connected by µ1,5-bridging dicyanamide spacers through termi-
nal Ni centres of neighbouring nodes. The different shapes of
the trinuclear nodes originate from the different coordination
environment of the central metal ions. The octahedral environ-
ment of Ni or Cd in 1 or 2 results in bent nodes in which the
two [NiL] “metalloligands” are almost parallel to each other.
On the other hand, in 3, Zn is tetrahedrally coordinated,
making the two [NiL] “metalloligands” nearly perpendicular in
linear disposition. The presence of different central hetero-
metal ions in the trinuclear nodes brings about significant
variation in the magnetic properties of the CPs. In 1 the intra-
trinuclear interactions between the three phenoxido bridged
Ni(II) ions are moderately antiferromagnetic, whereas in 2 the
weak antiferromagnetic intra-trinuclear interaction is only
between the two terminals Ni(II). The very weak inter-trinuclear
antiferromagnetic interactions between Ni(II) is through
the single μ1,5-bridging dicyanamide spacer in 3. The present
system thus reveals that the central metal ion in this kind of
trinuclear nodes is important in determining the final
network of the CP and their magnetic interactions. Therefore,
different trinuclear nodes can potentially be explored in future
studies of CPs, namely towards the design of novel metal–
organic materials with diverse topologies and functional
properties by changing metal ions.
2011, 13, 6118–6129; ( j) J.-P. Zhang, X.-C. Huang and
X.-M. Chen, Chem. Soc. Rev., 2009, 38, 2385–2396;
(k) A. Y. Robin and K. M. Fromm, Coord. Chem. Rev., 2006,
250, 2127–2157; (l) A. Erxleben, Coord. Chem. Rev., 2003,
246, 203–228; (m) G. Givaja, P. Amo-Ochoa, C. J. Gómez-
García and F. Zamora, Chem. Soc. Rev., 2012, 41, 115–147.
2 (a) C. Janiak and J. K. Vieth, New J. Chem., 2010, 34, 2366–
2388; (b) N. N. Adarsh and P. Dastidar, Chem. Soc. Rev.,
2012, 41, 3039–3060.
3 (a) R. W. Gable, B. F. Hoskins and R. Robson, J. Chem. Soc.,
Chem. Commun., 1990, 1677–1678; (b) B. F. Hoskins and
R. Robson, J. Am. Chem. Soc., 1990, 112, 1546–1554.
4 M. Andruh, Chem. Commun., 2007, 2565–2577.
5 (a) L. K. Das and A. Ghosh, CrystEngComm, 2013, 15, 9444–
9456; (b) L. K. Das, A. M. Kirillov and A. Ghosh, CrystEng-
Comm, 2014, 16, 3029–3039.
6 (a) S. Ghosh, S. Mukherjee, P. Seth, P. S. Mukherjee and
A. Ghosh, Dalton Trans., 2013, 42, 13554–13564;
(b) S. Biswas, C. J. Gómez-García, J. M. Clemente-Juan,
S. Benmansour and A. Ghosh, Inorg. Chem., 2014, 53, 2441–
2449.
7 S. Ghosh, Y. Ida, T. Ishida and A. Ghosh, Cryst. Growth
Des., 2014, 14, 2588–2598.
8 (a) M. Andruh, Pure Appl. Chem., 2005, 77, 1685–1706;
(b) M. Andruh, D. G. Branzea, R. Gheorghe and
A. M. Madalan, CrystEngComm, 2009, 11, 2571–2584.
9 (a) A. M. Madalan, H. W. Roesky, M. Andruh,
M. Noltemeyer and N. Stanica, Chem. Commun., 2002,
1638–1639; (b) J.-P. Costes, R. Gheorghe, F. Dahan,
M. Andruh, S. Shova and J.-M. Clemente Juan, New
J. Chem., 2006, 30, 572–576; (c) D. G. Branzea, A. Guerri,
O. Fabelo, C. Ruiz-Pérez, L.-M. Chamoreau, C. Sangregorio,
A. Caneschi and M. Andruh, Cryst. Growth Des., 2008, 8,
941–949; (d) D. Visinescu, A. M. Madalan, M. Andruh,
C. Duhayon, J.-P. Sutter, L. Ungur, W. V. Heuvel and
L. F. Chibotaru, Chem. – Eur. J., 2009, 15, 11808–11814;
(e) R. Gheorghe, A. M. Madalan, J.-P. Costes,
W. Wernsdorferd and M. Andruh, Dalton Trans., 2010, 39,
4734–4736.
Acknowledgements
L.K.D. is thankful to CSIR, India for awarding a Senior
Research Fellowship [Sanction no. 09/028 (0805)/2010-EMR-I].
The authors also thank the Department of Science and Tech-
nology (DST), New Delhi, India, for financial support (SR/S1/
IC/0034/2012). Crystallography was performed at the DST-FIST;
India funded Single Crystal Diffractometer Facility at the
Department of Chemistry, University of Calcutta. We also
thank the Spanish MINECO (CTQ-2011-26507) and the Gener-
alitat Valenciana (Prometeo and ISIC-Nano programs) for 10 (a) M. Andruh, Chem. Commun., 2011, 47, 3025–3042;
financial support.
(b) D. G. Branzea, A. M. Madalan, S. Ciattini, N. Avarvari,
A. Caneschi and M. Andruh, New J. Chem., 2010, 34, 2479–
2490.
11 O. Atakol, H. Nazir, C. Arici, S. Durmus, I. Svoboda and
H. Fuess, Inorg. Chim. Acta, 2003, 342, 295–300.
Notes and references
1 (a) T. Yamada, K. Otsubo, R. Makiura and H. Kitagawa, 12 G. A. Bain and J. F. Berry, J. Chem. Educ., 2008, 85, 532–536.
Chem. Soc. Rev., 2013, 42, 6655–7044; (b) O. Kahn, 13 SAINT, version 6.02, SADABS, version 2.03, Bruker AXS, Inc.,
J. Larionova and L. Ouahab, Chem. Commun., 1999, 945–
952; (c) M. J. Zaworotko, Chem. Commun., 2001, 1–9; 14 G. M. Sheldrick, SHELXS 97, Program for Structure Solution,
(d) B. Moulton and M. J. Zaworotko, Chem. Rev., 2001, 101, University of Göttingen, Germany, 1997.
1629–1658; (e) M. Du, C.-P. Li, C.-S. Liu and S.-M. Fang, 15 G. M. Sheldrick, SHELXL 97, Program for Crystal
Madison, WI, 2002.
Coord. Chem. Rev., 2013, 257, 1282–1305; (f) C. Janiak,
Dalton Trans., 2003, 2781–2804; (g) S. L. James, Chem. Soc.
Structure Refinement, University of Göttingen, Germany,
1997.
Rev., 2003, 32, 276–288; (h) K. Biradha, M. Sarkar and 16 A. L. Spek, PLATON. Molecular Geometry Program, J. Appl.
L. Rajput, Chem. Commun., 2006, 4169–4179; (i) Y.-Y. Liu, Crystallogr., 2003, 36, 7–13.
Y.-Y. Jiang, J. Yang, Y.-Y. Liu and J.-F. Ma, CrystEngComm, 17 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 1292–1302 | 1301

View Article Online
Paper
Dalton Transactions
18 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
19 (a) A. Biswas, M. G. B. Drew, J. Ribas, C. Diaz and
A. Ghosh, Eur. J. Inorg. Chem., 2011, 2405–2408;
24 A. W. Addison, T. N. Rao, J. Reedjik, J. van Rijn and
C. G. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349–
1356.
(b) J. Carranza, C. Brennan, J. Sletten, F. Lloret and 25 L. Yang, D. R. Powell and R. P. Houser, Dalton Trans., 2007,
M. Julve, J. Chem. Soc., Dalton Trans., 2002, 3164–3170.
20 S. Biswas and A. Ghosh, Polyhedron, 2012, 39, 31–37.
21 (a) J. Ferguson, J. Chem. Phys., 1961, 34, 611–615;
955–964.
26 J. J. Borrás-Almenar, E. Coronado, J. Curely and R. Georges,
Inorg. Chem., 1995, 34, 2699–2704.
(b) M. P. Sigalas and C. A. Tsipis, Inorg. Chem., 1986, 25, 27 (a) M. A. Halcrow, J.-S. Sun, J. C. Huffman and G. Christou,
1875–1880.
Inorg. Chem., 1995, 34, 4167–4177; (b) J. M. Clemente-Juan,
B. Chansou, B. Donnadieu and J. P. Tuchagues, Inorg.
Chem., 2000, 39, 5515–5519.
22 E. M. Boge, D. P. Freyberg, E. Kokot, G. M. Mockler and
E. Sinn, Inorg. Chem., 1977, 16, 1655–1660.
23 (a) M. Dey, C. P. Rao, P. K. Saarenketo and K. Rissanen, 28 S. Shit, M. Nandy, G. Rosair, C. J. Gómez-García, J. J. Borras
Inorg. Chem. Commun., 2002, 5, 924–928; (b) S. Banerjee, Almenar and S. Mitra, Polyhedron, 2013, 61, 73–79.
M. G. B. Drew, C.-Z. Lu, J. Tercero, C. Diaz and A. Ghosh, 29 C. Paraschiv, J.-P. Sutter, M. Schmidtmann, A. Müller and
Eur. J. Inorg. Chem., 2005, 2376–2383. M. Andruh, Polyhedron, 2003, 22, 1611–1615.
1302 | Dalton Trans., 2015, 44, 1292–1302
This journal is © The Royal Society of Chemistry 2015