Polymorphism in hetero-metallic tri-nuclear CuII 2CdII complexes of salicylaldimine ligand: Structural analysis and theoretical study

Article abstract of DOI:10.1016/j.poly.2012.04.029

Three new trinuclear hetero-metallic copper(II)-cadmium(II) complexes [(CuL)₂Cd (NCS)₂] (1A and 1B), [(CuL)₂Cd(NCO) ₂] (2) have been synthesized using [CuL] as so-called "ligand complex" (where H₂L = N,N′-bis(salicylidene)-1,3- propanediamine) and structurally characterized. Crystal structure analyses reveal that all three complexes have trinuclear entity in which two square planar [CuL] units are bonded to a cadmium ion through double phenoxido bridges. The Cd(II) is in a six-coordinate distorted octahedral environment being bonded additionally to two mutually cis nitrogen atoms of terminal thiocyanate (in 1A and 1B) and cyanate (in 2). Complexes 1A and 1B have the same molecular formula but crystallize in different monoclinic unit cells (P2₁/c for 1A and P2₁/n for 1B) and can be considered as polymorphs. On the other hand, complex 2 is crystallized only in one form and its conformation is similar to that observed in 1A. A theoretical investigation shows that 1A is slightly more (2.25 kcal/mol) stable than 1B. However, in the crystal lattice, 1A establishes a C-H/π interaction while 1B interacts through a C-H···S hydrogen bond to form dimer and the more favorable dimerization energy in the later (ca. 3 kcal/mol) compensates the difference in the relative energy of the polymorphs. Complex 2 crystallizes only in one form and conformation of the molecule is similar to 1A. Analyses of the noncovalent interactions of 2 show a weakening of the dimerization energy compared to 1A or 1B. Moreover, the H-bonding interaction that facilitates the formation of 1B is unlikely to be established in 2. Therefore this polymorph is not isolated.

Full text of DOI:10.1016/j.poly.2012.04.029

Polyhedron 52 (2013) 1416–1424
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Polymorphism in hetero-metallic tri-nuclear CuII CdII complexes of salicylaldimine
2
ligand: Structural analysis and theoretical study
a,
Lakshmi Kanta Das a, Apurba Biswas a, Antonio Frontera b, , Ashutosh Ghosh
⇑
⇑
a Department of Chemistry, University College of Science, University of Calcutta, 92, A.P.C. Road, Kolkata 700 009, India
b Department of Chemistry, Universitat de les Illes Balears, Crta. de Valldemossa km 7.5, 07122 Palma de Mallorca (Baleares), Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 3 May 2012
Three new trinuclear hetero-metallic copper(II)–cadmium(II) complexes [(CuL)2Cd (NCS)2] (1A and 1B),
[(CuL)2Cd(NCO)2] (2) have been synthesized using [CuL] as so-called ‘‘ligand complex’’ (where H2L =
N,N0-bis(salicylidene)-1,3-propanediamine) and structurally characterized. Crystal structure analyses
reveal that all three complexes have trinuclear entity in which two square planar [CuL] units are bonded
to a cadmium ion through double phenoxido bridges. The Cd(II) is in a six-coordinate distorted octahedral
environment being bonded additionally to two mutually cis nitrogen atoms of terminal thiocyanate (in
1A and 1B) and cyanate (in 2). Complexes 1A and 1B have the same molecular formula but crystallize
in different monoclinic unit cells (P21/c for 1A and P21/n for 1B) and can be considered as polymorphs.
On the other hand, complex 2 is crystallized only in one form and its conformation is similar to that
observed in 1A. A theoretical investigation shows that 1A is slightly more (2.25 kcal/mol) stable than
This paper is dedicated to Professor Alfred
Werner on the occasion of the 100th
anniversary of the award of the Nobel Prize
in Chemistry to him.
Keywords:
Copper(II)
Cadmium(II)
Heterometallic complexes
Polymorphism
1B. However, in the crystal lattice, 1A establishes a C–H/
p interaction while 1B interacts through a C–
HꢀꢀꢀS hydrogen bond to form dimer and the more favorable dimerization energy in the later (ca. 3 kcal/
mol) compensates the difference in the relative energy of the polymorphs. Complex 2 crystallizes only
in one form and conformation of the molecule is similar to 1A. Analyses of the noncovalent interactions
of 2 show a weakening of the dimerization energy compared to 1A or 1B. Moreover, the H-bonding inter-
action that facilitates the formation of 1B is unlikely to be established in 2. Therefore this polymorph is
not isolated.
Crystal structure
Density functional calculations
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
due to the different intermolecular interactions. It is an intriguing
phenomenon which has gained much current interest in academic
Isomers, that played a very important role in the development
of the chemistry of coordination compounds [1], are substances
that have the same number and kinds of atoms arranged differ-
ently in solid state, gas phase or in solution. Some well-known
isomers namely ionization, hydrate, linkage, polymerization,
geometric (cis–trans, fac–mer), optical, etc. were recognized from
the time of Werner. In fact, by isolating the previously unknown
and industrial purposes [6–9]. Different polymorphs with compa-
rable energies appear because of different arrangements and dif-
ferent conformation of molecules in the crystal lattice [10]. If the
ligand system contains p-electron clouds and quite a few electro-
negative atoms then different polymorphs of comparable energies
can be stabilized in solid state through various non-covalent forces
like H-bonds,
p–p, and C–H/p interactions [11–15]. The different
cis isomer [2] of [Co(NH3)4Cl2]Cl and by resolving [3] the
D
- and
forms exhibit different solid-state properties, like thermal stability,
crystal shape, compressibility and density which leads to differ-
ences in handling and processing properties of the compound
[9,16].
L-isomers of cis-[Co(en)2(NH3)Cl]Cl2 Warner proved the octahedral
coordination beyond any doubt and firmly established the modern
coordination theory proposed by him. On the other hand, poly-
morphs are the compound with identical chemical composition
and the same molecular connectivity, but allowing for different
conformations by rotations about single bonds resulting from the
possibility of at least two different arrangements of the molecules
of that compound in the solid state [4,5]. Polymorphism provides a
useful tool for investigating the structure–property relationship
Recently, we have focused our attention on the synthesis of
some hetero-metallic polynuclear complexes based on neutral
copper(II) complexes of salen type di-Schiff base ligands [17–21].
In the present investigation, we use [CuL] where H2L = N,N0-bis
(salicylidene)-1,3-propanediamine (Scheme 1) as ‘‘ligand complex’’
to form three hetero-metallic coordination compounds (1A, 1B and
2) with Cd(II) along with the pseudohalides, SCNꢁ and OCNꢁ as an-
ionic coligands. All of them are angular trinuclear species having
general formula [(CuL)2Cd(X)2] (X = SCNꢁ or OCNꢁ). Interestingly,
the thiocyanate ion results in the isolation of the crystals of
⇑
Corresponding authors. Fax: +91 33 2351 9755 (A. Ghosh).
E-mail addresses: toni.frontera@uib.es (A. Frontera), ghosh_59@yahoo.com
(A. Ghosh).
0277-5387/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2012.04.029

L.K. Das et al. / Polyhedron 52 (2013) 1416–1424
1417
the brown compound, 1A is isolated. The nearly colorless filtrate of
this refluxed mixture gives rhombic shaped deep brown (Fig. S1)
X-ray quality single crystals of 1A on slow evaporation at room
temperature.
Complex 1A: Yield: 0.700 g (75%, brown microcrystalline com-
pound), Anal. Calc. for C36H32CdCu2N6O4S2: C, 47.19; H, 3.52;
N, 9.17. Found: C, 47.63; H, 3.22; N, 9.11%. IR (KBr):
m(C@N),
1614 cmꢁ1 (SCN), 2068 and 2030 cmꢁ1, kmax (DMF): 603 and
;
m
365 nm, kmax (solid, reflectance): 590 and 368 nm.
2.4. Synthesis of the complex [(CuL)2Cd(NCS)2] (1B)
Complex 1B was prepared by mixing the same components like
1A at room temperature. A brown product (mixture of 1A and 1B)
separated immediately from the reaction mixture. The filtrate was
allowed to stand at open atmosphere when a rhombic shaped deep
brown (1A) and a rectangular shaped light brown (1B) X-ray qual-
ity single crystals appeared at the bottom of the vessel. The crystals
were isolated by hand under the microscope. The rectangular
shaped light brown crystals were used for X-ray diffraction and
other analyses.
Scheme 1. Formation of the complexes (1A, 1B and 2).
polymorphs 1A and 1B in which conformation of the trinuclear
complexes are different in two forms. Theoretical study has been
performed using DFT calculations at the M06/SVP level of theory,
to analyze (i) the energy difference between the two conforma-
tions, (ii) how the Cd–N–C angle between the central metal of
the metal–ligand complex and the thiocyanate ligand influences
the stability of the complexes, (iii) some non-covalent interactions
observed in the crystal packing of both the polymorphs, and (iv)
probable reasons for the isolation of only one form for complex
2. It is to be noted that different conformations were observed pre-
viously in this type of trinuclear adducts containing different ‘‘li-
gand complex’’ and consequently they were not polymorphs [21].
Therefore, to our knowledge, complexes 1A and 1B are the first re-
port of polymorphism in trinuclear hetero-metallic species base on
‘‘ligand complex’’ of salen type Schiff base.
Complex 1B: Anal. Calc. for C36H32CdCu2N6O4S2: C, 47.19; H,
3.52; N, 9.17. Found: C, 46.99; H, 3.37; N, 9.18%. IR (KBr):
m(C@N),
1617 cmꢁ1 (SCN), 2055 cmꢁ1, kmax (DMF): 603 and 364 nm, kmax
;
m
(solid, reflectance): 592 and 370 nm.
2.5. Synthesis of the complex [(CuL)2Cd(NCO)2] (2)
The ‘‘ligand complex’’ [CuL] (0.718 g, 2 mmol) was dissolved in
methanol (20 mL) and then water solution (1 mL) of Cd(ClO4)2
(0.311 g, 1 mmol) followed by the aqueous solution (1 mL) of so-
dium cyanate (0.130 g, 2 mmol) was added to this solution. A
brown product separated immediately like 1. The filtrate was al-
lowed to stand overnight in open atmosphere when rectangular
shaped red X-ray quality single crystals appeared. On keeping the
filtrate no other crystalline form was found to appear as was exam-
ined under microscope.
2. Experimental
2.1. Starting materials
Salicylaldehyde, and 1,3-propanediamine were purchased from
Lancaster and were of reagent grade. They were used without fur-
ther purification.
Complex 2: Yield: 0.480 g (54%), Anal. Calc. for C36H32CdCu2-
N6O6: C, 48.90; H, 3.65; N, 9.50. Found: C, 48.63; H, 3.62;
N, 9.51%. IR (KBr): m ; m
(C@N), 1615 cmꢁ1 (OCN), 2185 cmꢁ1, kmax
Caution! Perchlorate and azide salts of metal complexes with
organic ligands are potentially explosive. Only a small amount of
material should be prepared and it should be handled with care.
(DMF): 605 and 365 nm, kmax (solid, reflectance): 593 and 370 nm.
Table 1
Crystal data and structure refinement of complexes 1A, 1B and 2.
2.2. Synthesis of the Schiff base ligand (H2L) N,N0-bis(salicylidene)-1,
3-propanediamine
1A
1B
2
Formula
M
Crystal system
Space group
a (Å)
C
36H32CdCu2N6O4S2 C36H32CdCu2N6O4S2 C36H32CdCu2N6O6
916.33
monoclinic
P21/c
13.706(5)
11.616(5)
21.834(5)
90
92.207(5)
90
3474(2)
4
916.33
monoclinic
P21/n
884.19
The Schiff base ligand was synthesized by standard method.
Briefly, 5 mmol of 1,3-propanediamine (0.42 mL) was mixed with
10 mmol of the salicylaldehyde (1.04 mL) in methanol. The result-
ing solution was refluxed for ca. 2 h, and allowed to cool. The yel-
low colored methanolic solution was used directly for complex
formation.
triclinic
ꢀ
P1
11.157(5)
16.544(5)
19.541(5)
90
91.055(5)
90
10.058(5)
11.913(5)
14.080(5)
91.332(5)
100.645(5)
91.590(5)
1656.7(12)
2
b (Å)
c (Å)
a
(°)
b (°)
c
(°)
V (Å3)
3606(2)
4
2.3. Synthesis of the complex [(CuL)2Cd(NCS)2] (1A)
Z
Dcalc (g cmꢁ3
)
1.752
1.990
1840
1.688
1.773
To a methanolic solution (20 mL) of Cu(ClO4)2ꢀ6H2O (1.852 g,
l
(mmꢁ1
)
1.917
1.966
5 mmol) was added
a methanolic solution of H2L (5 mmol,
F(000)
1840
888
10 mL) to prepare the ‘‘ligand complex’’ [CuL] as reported earlier
[22]. The resulted complex [CuL] (0.718 g, 2 mmol) was dissolved
in methanol (20 mL) and then a water solution (1 mL) of Cd(ClO4)2
(0.311 g, 1 mmol) followed by an aqueous solution (1 mL) of so-
dium thiocyanate (0.162 g, 2 mmol) were added to this solution.
The resulted mixture is put under reflux for about 4–5 h and then
Rint
0.029
14476
0.085
35782
6321
4468
0.0396, 0.1108
293
0.022
18004
6690
5850
0.0321, 0.1142
293
Total reflections
Unique reflections 5086
I > 2
R1, wR2
r(I)
4072
0.0341, 0.0879
293
T (K)

1418
L.K. Das et al. / Polyhedron 52 (2013) 1416–1424
2.6. Physical measurements
crystals were positioned at 60 mm from the CCD. Frames (360)
were measured with a counting time of 5 s. The structures were
solved using Patterson method by using the SHELXS 97. Subsequent
difference Fourier synthesis and least-square refinement revealed
the positions of the remaining non-hydrogen atoms. Non-
hydrogen atoms were refined with independent anisotropic dis-
placement parameters. Hydrogen atoms were placed in idealized
positions and their displacement parameters were fixed to be 1.2
times larger than those of the attached non-hydrogen atom. Suc-
cessful convergence was indicated by the maximum shift/error of
0.001 for the last cycle of the least squares refinement. Absorption
corrections were carried out using the SADABS program [23]. All cal-
culations were carried out using SHELXS 97, [24] SHELXL 97, [25] PLATON
99, [26] ORTEP-32 [27] and WINGX system ver-1.64 [28]. Data collec-
tion and structure refinement parameters and crystallographic
data for the three complexes are given in Table 1. The selected
bond lengths and bond angles are summarized in Table 2.
Elemental analyses (C, H and N) were performed using a Perkin-
Elmer 240C elemental analyzer. IR spectra in KBr pellets (4000–
500 cmꢁ1) were recorded using a Perkin-Elmer RXI FT-IR spectro-
photometer. Electronic spectra (800–300 nm) were recorded in
DMF as well as in solid state in a Hitachi U-3501 spectrophotome-
ter. Thermal analysis (TG-DTA) were carried out on a Mettler Tole-
do TGA/SDTA 851 thermal analyzer in a dynamic atmosphere of
dinitrogen (flow rate = 30 cm3 minꢁ1). The samples were heated
in an alumina crucible at a rate of 10 °C minꢁ1. Powder X-ray dif-
fraction patterns are recorded on a Bruker D-8 Advance diffractom-
eter operated at 40 kV voltage and 40 mA current and calibrated
with a standard silicon sample, using Ni-filtered Cu Ka (a =
0.15406 nm) radiation.
2.7. Crystallographic data collection and refinement
2.8. Theoretical methods
Suitable single crystals of each complexes were mounted on a
Bruker-AXS SMART APEX II diffractometer equipped with a graph-
The present theoretical study has been carried out using DFT
calculations by means of the GAUSSIAN-09 package [29]. The level
of theory is M06/SVP for geometries and M06/TZVP//M06/SVP for
energies, which is an adequate compromise between the accuracy
of the results and the size of the systems studied herein. The M06
functional is one of the most successful functional for general
applications and in particular for non-covalent interactions in
chemical systems [30]. The binding energies were calculated with
correction for the basis set superposition error (BSSE) by using the
Boys–Bernardi counterpoise technique [31].
For the study of the relevant noncovalent forces observed in the
crystal lattice of the compounds, we have used the crystallographic
coordinates for the calculations in order to estimate the contribu-
tions of the interactions. Other possible conformations of the crys-
tal fragments have not been considered because the ultimate aim
of this particular study is to analyze the binding properties of the
non-covalent interaction in the geometry that they have in the so-
lid state. This approximation has been successfully used by some of
us [18,32–38] and others [39–41] to evaluate non-covalent interac-
tions in the solid state.
ite monochromator and Mo K
a (k = 0.71073 Å) radiation. The
Table 2
Bond distances (Å) and angles (°) for complexes 1A, 1B and 2.
1A
1B
2
Cd(1)–O(1)
Cd(1)–O(2)
Cd(1)–O(3)
Cd(1)–O(4)
Cd(1)–N(5)
Cd(1)–N(6)
Cu(1)–O(1)
Cu(1)–O(2)
Cu(1)–N(1)
Cu(1)–N(2)
Cu(2)–O(3)
Cu(2)–O(4)
Cu(2)–N(3)
Cu(2)–N(4)
2.386(3)
2.271(3)
2.393(3)
2.338(3)
2.251(5)
2.203(5)
1.928(3)
1.920(3)
1.953(4)
1.952(4)
1.929(3)
1.924(3)
1.962(4)
1.978(4)
2.303(4)
2.363(4)
2.402(4)
2.290(4)
2.234(6)
2.219(5)
1.933(4)
1.921(3)
1.944(6)
1.979(5)
1.927(3)
1.925(4)
1.949(5)
1.983(4)
2.389(3)
2.375(2)
2.432(3)
2.354(3)
2.171(4)
2.193(4)
1.924(2)
1.911(3)
1.958(3)
1.960(3)
1.936(3)
1.918(2)
1.965(3)
1.983(3)
O(1)–Cd(1)–O(2)
O(1)–Cd(1)–O(3)
O(1)–Cd(1)–O(4)
O(1)–Cd(1)–N(5)
O(1)–Cd(1)–N(6)
O(2)–Cd(1)–O(3)
O(2)–Cd(1)–O(4)
O(2)–Cd(1)–N(5)
O(2)–Cd(1)–N(6)
O(3)–Cd(1)–O(4)
O(3)–Cd(1)–N(5)
O(3)–Cd(1)–N(6)
O(4)–Cd(1)–N(5)
O(4)–Cd(1)–N(6)
N(5)–Cd(1)–N(6)
O(1)–Cu(1)–O(2)
O(1)–Cu(1)–N(1)
O(1)–Cu(1)–N(2)
O(2)–Cu(1)–N(1)
O(2)–Cu(1)–N(2)
N(1)–Cu(1)–N(2)
O(3)–Cu(2)–O(4)
O(3)–Cu(2)–N(3)
O(3)–Cu(2)–N(4)
O(4)–Cu(2)–N(3)
O(4)–Cu(2)–N(4)
N(3)–Cu(2)–N(4)
Cd(1)–O(1)–Cu(1)
Cd(1)–O(2)–Cu(1)
Cd(1)–O(3)–Cu(2)
Cd(1)–O(4)–Cu(2)
Cd(1)–N(5)–C(35)
Cd(1)–N(6)–C(36)
65.84(10)
104.64(11)
164.37(11)
95.92(13)
91.87(13)
76.43(11)
101.01(10)
95.24(14)
151.46(14)
62.51(11)
151.83(13)
93.49(14)
93.63(13)
97.57(13)
105.03(16)
82.32(12)
92.01(14)
164.80(15)
163.20(15)
92.49(15)
96.71(17)
79.18(13)
92.03(15)
170.15(15)
170.96(15)
91.67(15)
97.24(17)
100.75(12)
105.14(13)
102.09(14)
104.24(13)
130.4(4)
65.59(12)
89.00(12)
144.06(12)
100.52(16)
106.60(16)
76.36(11)
84.90(12)
97.03(17)
164.61(16)
63.31(12)
165.10(16)
90.64(15)
103.16(16)
96.70(16)
97.5(2)
81.97(14)
92.8(2)
168.60(18)
169.2(2)
90.86(18)
95.7(2)
79.55(14)
91.96(16)
169.80(17)
170.81(17)
90.74(17)
97.91(18)
104.41(15)
102.61(14)
98.30(14)
102.24(15)
155.2(5)
63.94(8)
99.36(8)
157.29(7)
99.12(12)
92.80(12)
74.03(9)
3. Results and discussion
3.1. Synthesis of the complexes
96.89(8)
94.15(12)
152.24(12)
61.84(8)
150.69(12)
96.70(11)
94.10(12)
101.59(12)
104.91(15)
82.27(9)
90.65(14)
167.06(13)
165.61(13)
92.46(14)
96.92(17)
79.34(9)
91.97(13)
169.73(11)
169.94(14)
91.66(11)
97.41(15)
102.30(9)
103.22(9)
100.55(10)
103.86(9)
160.0(4)
The Schiff-base ligand (H2L) and its Cu(II) complex, [CuL] were
synthesized using the reported procedure [22]. The Cu(II) complex
on reaction with cadmium perchlorate and the sodium thiocyanate
in MeOH–H2O medium (10:1, v/v) resulted in polymorphs,
[(CuL)2Cd(NCS)2] (1A) and [(CuL)2Cd(NCS)2] (1B) depending upon
the temperature. The X-ray powder diffraction patterns (Fig. S2)
of the brown microcrystalline product that separated immediately
on mixing the Cu(II) complex, cadmium perchlorate and sodium
thiocyanate at room temperature show that it is indeed a mixture
of 1A and 1B. The brownish filtrate on keeping at open atmosphere
for several days yields two distinctly different rhombic and rectan-
gular shaped crystals which are identified by single crystal X-ray
analysis as the polymorphs 1A and 1B, respectively (Fig. S1). How-
ever, if the mixture is stirred, the amount of the brown product in-
creases gradually and in ꢂ3h the supernatant liquid becomes
almost colorless. The brown product is a mixture of 1A and 1B as
before and the resulting filtrate on slow evaporation in open atmo-
sphere produces only few crystals but both of 1A and 1B. However,
if the mixture is put under reflux for about 4–5 h and then the
brown compound is isolated, the product is pure 1A as is evident
from the powder XRD. The nearly colorless filtrate of this refluxed
173.6(4)
150.5(5)
149.6(4)

L.K. Das et al. / Polyhedron 52 (2013) 1416–1424
1419
mixture gives the crystals of only 1A on slow evaporation. On the
other hand, if the mixing of the components is done at low temper-
ature (around 0 °C), the immediately separated brown product is
found to be the mixture of 1A and 1B as in the case of room tem-
perature but the filtrate on keeping in a refrigerator (ꢂ5 °C) pro-
duced mostly the crystals of 1B along with a few crystal of 1A.
The crystals were identified by their characteristic shape and IR
spectra. Therefore it may be concluded that both 1A and 1B are
formed when the constituent are mixed but at higher temperature
(when the mixture is put under reflux) 1B transforms to 1A. Com-
plex [(CuL)2Cd(NCO)2] (2) was synthesized by following a similar
procedure to that of 1, by using sodium cyanate instead of sodium
thiocyanate. However, the brown microcrystalline product which
was isolated immediately is the single species of complex 2 as its
powder XRD pattern is identical to that of the simulated one
(Fig. S3). Moreover, unlike 1 on refluxing the mixture no change
in the powder XRD pattern of this brown product is observed.
The crystals which were isolated on slow evaporation of the filtrate
were examined under the microscope, and found to be of only one
type. The determination of the cell dimensions for several crystals
by diffractometer as well as the powder XRD pattern of these crys-
tals confirm that only one product is formed for 2.
that of 1B. Again, X-ray crystallography data revels that the density
of 1A (1.752 g cmꢁ3
) is slightly greater than that of 1B
(1.688 g cmꢁ3) indicating that packing of the molecules is more
efficient in 1A and thus explain its higher thermal stability.
3.4. Crystal structures of the complexes
Each of structures consists of trinuclear neutral molecules of
formula [(CuL)2CdX2], where X is a monodentate anion, SCNꢁ in
1A and 1B and OCNꢁ in 2. The structure of 1A is shown in Fig. 1
(left) together with the atomic numbering scheme. Dimensions
in the metal coordination sphere are given in Table 2. The copper
atoms are four coordinate with a square planar environment being
bonded to two oxygen atoms and two nitrogen atoms of the Schiff-
base ligand. The four donor atoms in the equatorial plane show
r.m.s. deviation of 0.240 and 0.045 Å for Cu(1) and Cu(2), respec-
tively, with the copper atoms at 0.013(1), 0.014(1) Å from the
plane. The Cu–O bond lengths in both the copper atoms are similar
within the range from 1.920(3) to 1.929(3) Å but the Cu–N bond
lengths in Cu(1) [1.953(4) and 1.952(4) Å] are distinctly shorter
than those in Cu(2) [1.962(4) and 1.978(4) Å]. The trans angles
around Cu(1) and Cu(2), O(1)–Cu(1)–N(2) 164.80(15)°, O(2)–
Cu(1)–N(1) 163.20(15)°, O(3)–Cu(2)–N(4) 170.15(15)° and O(4)–
Cu(2)–N(3) 170.96(15)° indicate that the geometry of Cu(1) is
more distorted than that of Cu(2) from the ideal square planar
geometry. The greater tetrahedral distortion around Cu(1) is also
apparent in the dihedral angles between the two N–Cu–O planes,
which are 19.80, 3.85° for Cu(1) and Cu(2) respectively. This angle
is 0° for a perfectly square planar arrangement and 90° for a perfect
tetrahedral arrangement. A notable difference is found in the con-
formations of the two ‘‘ligand complex’’. Taking the angle between
the two aromatic rings as a measure, the angle between planes
around Cu(1) is 51.44°, while that around Cu(2) is 11.91°.
The cadmium ion has a six-coordinate distorted cis-octahedral
environment: four oxygen atoms from the two different ‘‘ligand
complex’’ and two terminal nitrogen atoms of two SCNꢁ ligands
are bonded to Cd(1) ion to complete its coordination sphere.
The bonds to four oxygen atoms are 2.386(3) Å for Cd(1)–O(1),
2.271(3) Å for Cd(1)–O(2), 2.393(3) Å for Cd(1)–O(3) and 2.338(3) Å
for Cd(1)–O(4) while that to two nitrogen atoms are 2.251(5) Å
for Cd(1)–N(5) and 2.203(5) Å for Cd(1)–N(6). Both the cis
[62.51(11)–105.03(16)°] and trans [151.46(14)–164.37(11)°] an-
gles indicate considerable distortions from ideal octahedral geom-
etry. Interestingly, the orientation of two SCNꢁ ligands is
considerably different which is reflected in their Cd–N–C bond an-
gles [Cd(1)–N(5)–C(35) 130.4(4)° and Cd(1)–N(6)–C(36) 173.6(4)°].
The two CdꢀꢀꢀCu distances are 3.335(2) and 3.374(2) Å while the
CuꢀꢀꢀCu separation is 5.095 Å.
3.2. IR and electronic spectra of the complexes
Besides elemental analysis, all the complexes were initially
characterized by IR spectra. The precursor ‘‘ligand complex’’
[CuL] is neutral and obviously not have any counter anion. All three
synthesized hetero-nuclear complexes contain anions (SCNꢁ in 1A
and 1B and OCNꢁ in 2). These anions show their characteristic
bands in their IR spectra. The complexes 1A and 1B were distin-
guished not only by the peak positions of m(SCN), but the patterns
of peak was also different. In case of 1A (Fig. S4) a strong and broad
peak at 2068 cmꢁ1 along with a shoulder at 2030 cmꢁ1 appeared,
while there was only
a single strong and sharp peak at
2055 cmꢁ1 in the spectrum of 1B (Fig. S5). The splitting of the thio-
cyanate peak may be attributed to the presence of two different
types of SCNꢁ anion in 1A. The characteristic intense peaks for cya-
nate at 2185 cmꢁ1 was also clearly detected in the IR spectra of 2.
The rest of the spectral patterns and band positions of the all three
complexes are very similar to the precursor, [CuL]. In all com-
plexes, a strong and sharp band due to the azomethine
t(C@N)
group of the Schiff base appears at 1614, 1617 and 1615 cmꢁ1 for
complexes 1A, 1B and 2, respectively.
The electronic spectra were measured in DMF as well as in solid
state. All the complexes show a broad single absorption band near
at 365, 364 and 365 nm in DMF and at 368, 370 and 370 nm in
solid state for 1A, 1B and 2 respectively (Figs. S6 and S8) attributed
to ligand-to-metal charge transfer transitions. Besides this band, a
broad absorption band is observed in the visible region at 603, 603
and 605 nm in DMF and at 590, 592 and 593 nm in solid state for
1A, 1B and 2, respectively (Figs. S7 and S8). This band is typical of
d–d transitions of Cu(II) ions with a square planar environment
and, as expected, is also observed in the mononuclear CuL precur-
sor complex. The electronic spectral data for the two polymorphs
were found almost identical in both solution as well as solid state.
This indicates that almost identical complexes had formed in both
the cases.
The structure of 1B is shown in Fig. 1 (middle). The connectivity
is the same as in 1A but the dimensions in the metal coordination
spheres differ considerably. Around the copper atoms the donor
atoms in the equatorial planes show r.m.s. deviation of 0.147 and
0.052 Å for Cu(1) and Cu(2), respectively, with the copper atoms
0.007(1), 0.004(1) Å from the plane. Like 1A the Cu–O bond lengths
of the two copper atoms vary within a short range from 1.921(3) to
1.933(4) Å but Cu–N bond distances [Cu(1)–N(1) 1.944(6) Å,
Cu(1)–N(2) 1.979(5) Å, Cu(2)–N(3) 1.949(5) Å and Cu(2)–N(4)
1.983(4) Å] are significantly different from each other in the indi-
vidual copper atom although the average Cu–N bond distances
[1.962 Å for Cu(1) and 1.966 Å for Cu(2)] are similar in both the
copper atoms. Similarly, the tetrahedral distortion around Cu(1)
is larger than that of Cu(2) as is evident from the dihedral angles
between the two N–Cu–O planes, which are 12.10, 4.32° for
Cu(1) and Cu(2), respectively. By contrast, the trans angles around
the Cu(1) [168.60(18)° and 169.2(2)°] and Cu(2) [169.80(17)° and
170.81(17)°] are almost equivalent. Like 1A the angles between
3.3. Thermogravimetric analysis of the complexes 1A and 1B
Thermogravimetric analysis (TGA) of complexes 1A and 1B
shows similar type of thermogram (Fig. S9). From the thermogram
it is found that complex 1B starts to decompose (Ti = 297 °C) at
lower temperature than that of complex 1A (Ti = 325 °C). Therefore,
between the two polymorphs 1A is thermally more stable than

1420
L.K. Das et al. / Polyhedron 52 (2013) 1416–1424
Fig. 1. ORTEP representation of 1A, 1B and 2 with ellipsoids at 30% probability.
the two phenyl rings around Cu(2) is 9.75° but this angle around
Cu(1) [35.54°] is significantly smaller than that in 1A.
O(4)] and the trans angle [O(1)–Cd(1) –O(4)] around the cadmium
atom, which are originated from the O atoms of two different ‘‘li-
gand complex’’ are drastically different in two structures. In case
of complex 1A both the cis angles [104.64(11)° and 101.01(10)°]
are significantly greater than 90° whereas the trans angle
[164.37(11)°] is less by about 15.63° from the ideal value. On the
other hand, in complex 1B, both the cis angles [89.00(12)° and
84.90(12)°] are less than 90° while the corresponding trans angle
[144.06(12)°] is smaller by 35.94° from 180°. The relative disposi-
tion of the two ‘‘ligand complexes’’ around Cd atom is different:
they slide away from each other more in 1A than in 1B resulting
a greater CuꢀꢀꢀCu separation in 1A.
The Cd–O and Cd–N bond lengths are also similar to 1A in the
ranges from 2.290(4) to 2.402(4) Å and 2.219(5) to 2.234(6) Å,
respectively. The cis [63.31(12)°–106.60(16)°] and trans [144.06
(12)°–165.10(16)°] angles around the metal deviate considerably
from octahedral geometry but in a different manner than those
in 1A as further discussed below. Unlike 1A, the orientation of
two SCNꢁ ligands are similar with bond angles 155.2(5)° and
150.5(5)°. The two CdꢀꢀꢀCu distances are 3.355(2) and 3.290(2) Å
while the CuꢀꢀꢀCu separation at 4.264 Å is significantly less than
that in 1A.
The two structures (1A and 1B) can be considered as poly-
morphs. The different degree of distortion particularly in the bond
angles around the cadmium atom is also observed between the
two polymorphs. A close examination of the two crystal structures
reveals that the two cis angles [O(1)–Cd(1)–O(3) and O(2)–Cd(1)–
We have studied theoretically some interesting non-covalent
interactions observed in the crystal packing of both polymorphs
that influence the Cd–NCS angle, which is one of the differences
between both polymorphs. One of the most important assemblies
observed in both crystal structures is the formation of a dimer that
Fig. 2. Dimers of 1A and 1B observed in the solid state and their location in the crystal packing.

L.K. Das et al. / Polyhedron 52 (2013) 1416–1424
1421
has a crucial influence in the solid state architecture of the com-
plexes. The dimeric assemblies of 1A and 1B and their location in
the crystal packing are represented in Fig. 2. It is interesting to note
that the Cu atom that is bonded to the aromatic ligand that partic-
ipates in the stacking presents an almost perfect square planar
arrangement, thus facilitating the planarity of the system and the
stacking. The distortion of the other Cu atom with respect to the
ideal square planar disposition is considerably higher.
A close examination of the non-covalent interactions present in
the self-assembled dimers reveals that the aliphatic chain that con-
nects the nitrogen atoms of the ligand plays a crucial role in the
formation of the dimer. Two hydrogen atoms of one unit establish
two hydrogen bonds with two oxygen atoms of the other unit and
vice versa, giving rise to four C–HꢀꢀꢀO hydrogen bonds. This hydro-
gen bond network is observed in the dimers of both polymorphs
(Fig. 3). An additional and very interesting interaction that differ-
entiates both polymorphs involves the SCNꢁ ligand and particu-
larly the Cd–N–C angle. In 1A the angle is 173.50° and the ligand
served that the formation energy of the dimer of 1B is slightly more
favorable than the dimer of 1A, as expected taking into account
that hydrogen bonding interactions (C–HꢀꢀꢀS) are more favorable
than C–H/p interactions. The energetic difference in the formation
of the dimers shown in Fig. 4 (ca. 3 kcal/mol) should compensate
the difference in the relative energy of the polymorphs, as it is
demonstrated below.
The structure of 2 is shown in Fig. 1 (right). Dimensions are
compared in Table 2 and show only minor differences from 1A.
Thus, the ranges of Cu–O distances are from 1.911(3) to
1.936(3) Å, Cu–N from 1.958(3) to 1.983(3) Å, Cd–O at 2.354(3)–
2.432(3) Å and Cd–N at 2.171(4)–2.193(4) Å. Here also the trans
angles around Cu(1) [167.06(13)° and 165.61(13)°] and Cu(2)
[169.73(13)° and 169.94(14)°] indicate that Cu(1) is suffered great-
er tetrahedral distortion than Cu(2). The higher tetrahedral distor-
tion around Cu(1) is also apparent in r.m.s. deviation of the
coordinating atoms from their mean planes [0.193 and 0.079 Å
for Cu(1) and Cu(2), respectively] and in the dihedral angles be-
tween the two N–Cu–O planes which are 16.02, 6.44° for Cu(1)
and Cu(2), respectively. Here, the angles between the two aromatic
rings [50.19° for Cu(1) and 14.58° for Cu(2)] are comparable with
1A. The distortion from octahedral geometry around cadmium is
indicated by the smallest trans O(3)–Cd(1)–N(5) angle of 150.69
(12)°. The two Cd–N–C bond angles, [Cd(1)–N(5)–C(35) 160.0(4)°
and Cd(1)–N(6)–C(36) 149.6(4)°] differ about 10° indicating that
the orientation of two OCNꢁ ligands are not alike. The two CdꢀꢀꢀCu
establishes a C–H/p interaction with the central carbon atom of
the aliphatic chain. In contrast, the angle in 1B is 150.50° and the
type of interaction is a C–HꢀꢀꢀS hydrogen bond. We will demon-
strate below that the energy cost of bending the Cd–N–CS angle
from 180 to 150° is almost negligible, therefore the binding energy
associated to both C–H/
p
and C–HꢀꢀꢀS should be similar (vide infra).
We have further analyzed and studied the binding energy of the
dimers. The dimerization energies are shown in Fig. 4. It can be ob-
Fig. 3. Noncovalent interactions observed in the dimers of 1A and 1B, distances in Å.
Fig. 4. Formation energy of the dimers of polymorphs 1A and 1B.

1422
L.K. Das et al. / Polyhedron 52 (2013) 1416–1424
distances are 3.372(2) and 3.374(2) Å while the CuꢀꢀꢀCu separation
is 4.844 Å. The degree of distortion around cadmium atom in 2 is
close to that of complex 1A as the two cis angles [O(1)–Cd(1)–
O(3) 99.36(8)° and O(2)–Cd(1)–O(4) 96.89(8)°] around cadmium
atom are greater than 90° whereas the trans angles [O(1)–Cd(1)–
O(4) 157.29(7)°] is considerably less than 180°. The CuꢀꢀꢀCu separa-
tion (4.844 Å) also indicates that the structure of 2 is close to 1A.
In addition, there is a geometric feature in complex 2 that is dif-
ferent from 1A that strongly influences the dimerization. That is
the conformation of the aliphatic chain that connects the nitrogen
atoms of the ligand. In the solid state structure compound 2 the
dimer that we have described above for polymorphs 1A and 1B
is also formed. In Fig. 5 we show the partial view of the dimer of
2 and it can be appreciated that the central carbon atom of the
chain is pointing to the OCNꢁ ligand, the opposite side with respect
to 1A (see Fig. 2, top). This change causes the formation of a differ-
ent hydrogen bonding network. In 1A two hydrogen atoms belong-
ing to two different carbon atoms interact with two phenolic
oxygen atoms of the other molecule. In 2A one hydrogen atom of
the central carbon atom forms a bifurcated C–HꢀꢀꢀO2 hydrogen
energetic data have been computed at the M06/TVZP//M06/SVP le-
vel of theory in order to increase the accurateness of the results.
We have divided the study into two parts. First, we have analyzed
the energetic difference between both polymorphs observed
experimentally. The experimental and optimized geometries are
included in Fig. 6. It can be observed that the optimized structures
are similar to the experimental ones, apart from the orientation of
the SCNꢁ ligands, which have been reoriented during the optimiza-
tion because the absence of additional interactions between the
SCNꢁ ligands and neighboring molecules present in the crystal. It
is interesting to note that both polymorphs have been located on
the potential hypersurface, having different formation energies.
bond and the other hydrogen atom p-interacts with the triatomic
ligand. This change provokes a weakening of the dimerization en-
ergy, which is approximately ꢁ46 kcal/mol (see Fig. 5), compared
to the ones computed for the dimers of 1A and 1B. The different
conformation of the central atom of the aliphatic chain likely pre-
vents the formation of the other polymorph, since the hydrogen
bond that characterizes the dimer 1B cannot be established in this
conformation (too small C–HꢀꢀꢀO angle).
3.5. Theoretical study
This theoretical study has been performed using DFT calcula-
tions at the M06/SVP level of theory, which is a good compromise
between the size of the system and the accuracy of the results. The
Fig. 7. The polymorph 1A is shown with indication of the angle used for
constructing the energetic profile.
Fig. 5. Noncovalent interactions observed in the stacked dimers of 2A and its location in the crystal packing. Distances in Å. The dimerization energy is indicated. The
aliphatic chain is highlighted in light green. (Color online.)
Fig. 6. Left: X-ray structures 1A and 1B. Right: DFT-optimized structures and relative energies.

L.K. Das et al. / Polyhedron 52 (2013) 1416–1424
1423
Fig. 8. Left: Histogram for the TR–N–C angle generated from X-ray structures of transition metal complexes with NCS ligand. Right: Histogram for the Cd–N–C angle
generated from X-ray structures of Cd complexes with NCS ligand.
Fig. 9. Left: X-ray structure of 2. Right: DFT-optimized structures and relative energies of polymorphs 2A and 2B.
Polymorph 1A is 2.25 kcal/mol more stable than 1B. This difference
is compensated by the noncovalent interactions established in the
crystal packing (vide supra) since the formation of the dimer of 1A
is 3 kcal/mol more favorable than the formation of the dimer of 1B.
In the second part of the present theoretical study we have ana-
lyzed how the Cd–N–C angle between the central metal of the
metal–ligand complex and the thiocyanate ligand influences the
stability of the complex. For this study we have used the 1A. In
Fig. 7 we highlight the SCNꢁ ligand that presents the lower Cd–
N–C angle, which is 130.40°. We have progressively increased
and decreased this angle from the experimental value to 180°
and 90°, respectively, in order to know the preference angle in
the isolated metal–ligand adduct and the energetic behavior. We
have computed the energy of the complex changing the angle
every 10° covering the 90°–180° range. The plot of the relative en-
transition metal bounded to thiocyanate applying a unique restric-
tion, which is the SCNꢁ cannot act as a bridging ligand. The histo-
gram plot is shown at the left part of Fig. 8. It can be clearly
observed that a large number of structures present angles between
160° and 180° with a slight preference for linearity. When the
search is restricted to cadmium complexes the histogram plot
(right part of Fig. 8) drastically changes. Curiously and in contrast
to the general trend the Cd–NCS linearity is not preferential in
the cadmium complexes. A likely explanation is that the d10 config-
uration of the CdII ion does not favors a preferential direction due
to orbital effects. From the energetic and CSD results, it is clear that
the Cd–N–C angle can vary from 130° to 180° and that the angle
depends upon the non-covalent interactions present in the crystal
packing.
We have also optimized the two possible polymorphs of 2 (2A
and 2B) to study their difference in energy. The results are shown
in the right part of Fig. 9 and it can be observed that the most stable
polymorph is 2A. Experimentally, polymorph 2B is not formed
probably because the size of the ligand is too short compared to
SCNꢁ, and the hydrogen bond that characterizes the dimer forma-
tion in 1B (see Fig. 3) cannot be established.
ergy of the system versus
a is shown at the bottom of Fig. 7. It can
be observed that in the 180°–130° range the different points are al-
most isoenergetic, indicating that there is not any preference for a
given angle, although the global minimum is found at 180°. The
energetic difference between the 180° and the 130° is lesser than
0.1 kcal/mol, indicating a lack of directionality. A further reduction
of the angle causes a destabilization of the system, being very mod-
est at 120° (0.4 kcal/mol) and increasing rapidly for smaller angles.
Curiously a local minimum is found at 130°, which is the value for
the Cd–N–C angle in the crystal for one ligand (this point is repre-
sented in blue in the plot of Fig. 7).
4. Conclusions
We have synthesized two polymorphs 1A and 1B of the trinu-
clear complex [(CuL)2Cd (NCS)2] that the same molecular formula
but crystallize in different monoclinic unit cells (P21/c for 1A and
P21/n for 1B). By means of the theoretical study we have shown
that both polymorphs 1A and 1B are stationary points. In spite of
To further investigate this lack of directionality for the SCNꢁ li-
gand, we have explored the CSD, which is a big storehouse of geo-
metric information, to learn the general behavior of this ligand. We
have performed two searches: in the first one we have used any

1424
L.K. Das et al. / Polyhedron 52 (2013) 1416–1424
[8] G.P. Stahly, Cryst. Growth Des. 7 (2007) 1007.
the energetic difference computed for the polymorph (2.5 kcal/
mol), both are obtained experimentally. A likely explanation is that
this difference is compensated by non-covalent interactions pres-
ent in the solid state. As deduced by the computational study,
the energy difference seems to be counterbalanced in the forma-
tion of the self-assembled dimers of polymorphs 1A and 1B. We
have also studied the influence of the Cd–N–CS angle on the energy
of the complex and we have shown that in a wide range of values it
is negligible. We have compared this result with experimental data
retrieved from the CSD and shown that the preference angle in Cd
complexes (150°–160°) is different compared to the general trend
observed in the rest of transition metal complexes (170°–180°).
In an attempt to obtain additional polymorph, we have synthe-
sized similar complex using OCNꢁ instead of SCNꢁ. The X-ray char-
[9] G.R. Desiraju, Angew. Chem., Int. Ed. 46 (2007) 8342.
[10] N. Blagden, R.J. Davey, Cryst. Growth Des. 3 (2003) 873.
[11] J. Miao, Y. Nie, C. Hub, Z. Zhang, G. Li, M. Xu, Guoxin Sun, J. Mol. Struct. 1014
(2012) 97.
[12] C.A. Bessel, R.F. See, D.L. Jameson, M.R. Churchill, K.J. Takeuchi, J. Chem. Soc.,
Dalton Trans. (1992) 3223.
[13] K.F. Bowes, I.P. Clark, J.M. Cole, M. Gourlay, A.M.E. Griffin, M.F. Mahon, L. Ooi,
A.W. Parker, P.R. Raithby, H.A. Sparkes, M. Towrie, CrystEngComm 7 (2005)
269.
[14] J.E. Beves, P. Chwalisz, E.C. Constable, C.E. Housecroft, M. Neuburger, S.
Schaffner, J.A. Zampese, Inorg. Chem. Commun. 11 (2008) 1009.
[15] S.K. Dey, G. Das, Cryst. Growth Des. 10 (2010) 754.
[16] S. Gracin, A.C. Rasmuson, Cryst. Growth Des. 4 (2004) 1013.
[17] P. Mukherjee, M.G.B. Drew, A. Figuerola, A. Ghosh, Polyhedron 27 (2008)
3343.
[18] S. Biswas, S. Naiya, M.G.B. Drew, C. Estarellas, A. Frontera, A. Ghosh, Inorg.
Chim. Acta 366 (2011) 219.
[19] S. Biswas, A. Ghosh, Polyhedron 30 (2011) 676.
[20] S. Biswas, S. Naiya, C.J. Gómez-García, A. Ghosh, Dalton Trans. 41 (2012) 462.
[21] S. Biswas, A. Ghosh, Indian J. Chem. 50A (2011) 1356.
[22] M.G.B. Drew, R.N. Prasad, R.P. Sharma, Acta Crystallogr., Sect. C 41 (1985)
1755.
[23] SAINT, Version 6.02; SADABS, Version 2.03, Bruker AXS, Inc., Madison, WI, 2002.
[24] G.M. Sheldrick, SHELXS 97, Program for Structure Solution, University of
Göttingen, Germany, 1997.
ꢀ
acterization (it crystallizes in the space group P1) shows that only
form A is obtained. A detailed analysis of the solid state structure
reveals that the aliphatic chain that connects the two nitrogen
atoms of the ligand is oriented differently compared to the SCNꢁ
structure and it probably causes the formation of only one form.
[25] G.M. Sheldrick, SHELXL 97, Program for Crystal Structure Refinement, University
of Göttingen, Germany, 1997.
Acknowledgements
[26] A.L. Spek, J. Appl. Crystallogr. 36 (2003) 7.
[27] L.J.J. Farrugia, Appl. Crystallogr. 30 (1997) 565.
[28] L.J.J. Farrugia, Appl. Crystallogr. 32 (1999) 837.
L.K.D. and A.B. are thankful to CSIR, India for awarding Junior
Research Fellowship [Sanction No. 09/028 (0805)/2010-EMR-I]
and Senior Research Fellowship [Sanction No. 09/028 (0717)/
2008-EMR-I], respectively. Crystallography was performed at the
DST-FIST, India funded Single Crystal Diffractometer Facility at
the Department of Chemistry, University of Calcutta. A.F. thanks
CONSOLIDER–Ingenio 2010 (CSD2010-0065) and the MICINN of
Spain (project CTQ2011-27512/BQU, FEDER funds) for financial
support and the CESCA for computational facilities.
[29] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro,
M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M.
Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth,
P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B. Foresman,
J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision B.01, Gaussian, Inc.,
Wallingford, CT, 2009.
Appendix A. Supplementary data
[30] D.G. Truhlar, Y. Zhao, Theor. Chem. Acc. 120 (2006) 215.
[31] S.F. Boys, F. Bernardi, Mol. Phys. 19 (1970) 553.
[32] A. Das, S.R. Choudhury, C. Estarellas, B. Dey, A. Frontera, J. Hemming, M.
Helliwell, P. Gamez, S. Mukhopadhyay, CrystEngComm 13 (2011) 4519.
[33] S.K. Seth, I. Saha, C. Estarellas, A. Frontera, T. Kar, S. Mukhopadhyay, Cryst.
Growth Des. 11 (2011) 3250.
[34] R. Biswas, M.G.B. Drew, C. Estarellas, A. Frontera, A. Ghosh, Eur. J. Inorg. Chem.
(2011) 2558.
[35] M. Barceló-Oliver, C. Estarellas, A. Terrón, A. García-Raso, A. Frontera, Chem.
Commun. 47 (2011) 4646.
CCDC 859826–859828 contains the supplementary crystallo-
graphic data for 1A, 1B and 2, respectively. These data can be ob-
tained free of charge via http://www.ccdc.cam.ac.uk/conts/
retrieving.html, or from the Cambridge Crystallographic Data
Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223-
336-033; or e-mail: deposit@ccdc.cam.ac.uk. Supplementary data
associated with this article can be found, in the online version, at
http://dx.doi.org/10.1016/j.poly.2012.04.029.
[36] M. Barceló-Oliver, C. Estarellas, A. Garcia-Raso, A. Terrón, A. Frontera, D.
Quinonero, I. Mata, E. Molins, P.M. Deyà, CrystEngComm 12 (2010) 3758.
[37] S. Naiya, M.G.B. Drew, C. Estarellas, A. Frontera, A. Ghosh, Inorg. Chim. Acta 363
(2010) 3904.
References
[38] A. Garcia-Raso, F.M. Alberti, J.J. Fiol, Y. Lagos, M. Torres, E. Molins, I. Mata, C.
Estarellas, A. Frontera, D. Quiñonero, P.M. Deyà, Eur. J. Org. Chem. (2010) 5171.
[39] S.R. Choudhury, B. Dey, S. Das, A. Robertazzi, A.D. Jana, C.-Y. Chen, H.M. Lee, P.
Gamez, S. Mukhopadhyay, Dalton Trans. 38 (2009) 7617.
[40] P. de Hoog, A. Robertazzi, I. Mutikainen, U. Turpeinen, P. Gamez, J. Reedijk, Eur.
J. Inorg. Chem. (2009) 2684.
[1] S.F.A. Kittle, Physical Inorganic Chemistry, Oxford University Press, 2002. p. 43.
[2] G.B. Kauffman, J. Chem. Educ. 36 (1959) 521.
[3] A. Werner, V.L. King, Ber. Dtsch. Chem. Ges. 44 (1911) 1887.
[4] A.J. Gavezzotti, Pharm. Sci. 96 (2007) 2232.
[5] J. Elguero, Cryst. Growth Des. 11 (2011) 4731.
[6] J. Bernstein, Polymorphism in Molecular Crystals, Clarendon Press, Oxford,
2002.
[41] R.J. Goetz, A. Robertazzi, I. Mutikainen, U. Turpeinen, P. Gamez, J. Reedijk,
Chem. Commun. (2008) 3384.
[7] R. Hilfiker, Polymorphism in the Pharmaceutical Industry, Wiley-VCH,
Weinheim, 2006.