Novel applications of functionalized 2,1,3-benzothiadiazoles for coordination chemistry and crystal engineering

Article abstract of DOI:10.1039/c4ra03342f

Two novel applications of functionalized 2,1,3-benzothiadiazoles for metal coordination chemistry and crystal engineering of organic solids are presented. 4-Amino-2,1,3-benzothiadiazole (1) forms a 2:1 complex with ZnCl₂ (complex 2) and a 1:1 complex with 4-nitro-2,1,3-benzothiadiazole (3) (complex 4). The structures of compounds 1-4 were confirmed by single-crystal X-ray diffraction and studied by UV-Vis and IR spectroscopy, and DFT and QTAIM calculations. Complex 2 is the first structurally defined Zn complex with 2,1,3-benzothiadiazole ligands. In this complex, both molecules 1 are coordinated to Zn only by amino groups, thus revealing a novel type of coordination of this ligand to the metal center. According to ¹H NMR data, complex 2 dissociates in CHCl₃, THF or DMSO solutions. There are only a few examples of similar complexes, which are also considered to dissociate in solutions. In crystalline complex 4, molecules 1 and 3 form infinite alternating π-stacks connected by lateral S...N interactions between the neighboring stacks. Intermolecular S...N interactions are also observed in the crystals of individual 1 and 3 but the packing motifs are different from those in 4. DFT calculations predict a small charge transfer (CT, ～0.02e at B97-D3 level) from 1 to 3 upon the formation of 4, which therefore is an unprecedented CT complex where both donor and acceptor moieties are the derivatives of the 2,1,3-benzothiadiazole ring system. This finding creates some new prospects for the crystal engineering of organic solids. Crystalline complex 4 is characterized by an intense CT absorption band with a maximum at ～550 nm. However, according to DFT and QTAIM calculations the complex is weakly bonded and its formation is not observed in CH ₂Cl₂ solution with ¹H NMR and UV-Vis techniques.

Full text of DOI:10.1039/c4ra03342f

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Novel applications of functionalized 2,1,3-
benzothiadiazoles for coordination chemistry and
Cite this: RSC Adv., 2014, 4, 28309
crystal engineering†
ab
a
ab
bc
Denis A. Bashirov, Taisiya S. Sukhikh, Natalia V. Kuratieva, Elena A. Chulanova,
a
ce
ab
def
Irina V. Yushina, Nina P. Gritsan, Sergey N. Konchenko* and Andrey V. Zibarev*
Two novel applications of functionalized 2,1,3-benzothiadiazoles for metal coordination chemistry and
crystal engineering of organic solids are presented. 4-Amino-2,1,3-benzothiadiazole (1) forms a 2 : 1
2
complex with ZnCl (complex 2) and a 1 : 1 complex with 4-nitro-2,1,3-benzothiadiazole (3) (complex 4).
The structures of compounds 1–4 were confirmed by single-crystal X-ray diffraction and studied by UV-
Vis and IR spectroscopy, and DFT and QTAIM calculations. Complex 2 is the first structurally defined Zn
complex with 2,1,3-benzothiadiazole ligands. In this complex, both molecules 1 are coordinated to Zn
only by amino groups, thus revealing a novel type of coordination of this ligand to the metal center.
1
According to H NMR data, complex 2 dissociates in CHCl
3
, THF or DMSO solutions. There are only a
few examples of similar complexes, which are also considered to dissociate in solutions. In crystalline
complex 4, molecules 1 and 3 form infinite alternating p-stacks connected by lateral S/N interactions
between the neighboring stacks. Intermolecular S/N interactions are also observed in the crystals of
individual 1 and 3 but the packing motifs are different from those in 4. DFT calculations predict a small
charge transfer (CT, ꢀ0.02e at B97-D3 level) from 1 to 3 upon the formation of 4, which therefore is an
unprecedented CT complex where both donor and acceptor moieties are the derivatives of the 2,1,3-
benzothiadiazole ring system. This finding creates some new prospects for the crystal engineering of
organic solids. Crystalline complex 4 is characterized by an intense CT absorption band with a maximum
Received 13th April 2014
Accepted 3rd June 2014
DOI: 10.1039/c4ra03342f
at ꢀ550 nm. However, according to DFT and QTAIM calculations the complex is weakly bonded and its
1
www.rsc.org/advances
formation is not observed in CH
2
Cl
2
solution with H NMR and UV-Vis techniques.
functional materials (relevant literature is too abundant to be
1
. Introduction
1–10
cited completely; for selected references see
and the refs
Currently, 2,1,3-benzothiadiazole and its derivatives attract therein). For crystal engineering, these compounds are prom-
increasing attention because of their redox, optical and coor- ising due to unlike charges on their S and N atoms, causing
11
dination properties, which can be used in materials science for intermolecular S/N contacts in the solid state together with
the design and synthesis of various new molecule-based extended p-systems capable of p–p interactions in the same
12
state. As a result, multidimensional networks can be formed,
which are believed to cause non-trivial chemical and physical
1
3–17
a
properties of this type of compounds.
At the same time, the coordination chemistry of compounds
Nikolaev Institute of Inorganic Chemistry, Siberian Branch, Russian Academy of
Sciences, 630090 Novosibirsk, Russia. E-mail: konch@niic.nsc.ru
b
Department of Natural Sciences, National Research University – Novosibirsk State under discussion is not well-developed, and only a limited
University, 630090 Novosibirsk, Russia
number of structurally characterized transition metal
complexes, including those of Ag, Cd, Co, Cu, Pd, Ru and
c
Voevodsky Institute of Chemical Kinetics and Combustion, Siberian Branch, Russian
Academy of Sciences, 630090 Novosibirsk, Russia
d
trimeric
peruoro-ortho-phenylenemercury
(anti-crown
Vorozhtsov Institute of Organic Chemistry, Siberian Branch, Russian Academy of
14–24
compound), are known.
Sciences, 630090 Novosibirsk, Russia
e
Recently, we reported the preparation and single-crystal X-
Department of Physics, National Research University – Novosibirsk State University,
6
30090 Novosibirsk, Russia
ray diffraction (XRD) characterization of Ir complexes with the
f
Department of Chemistry, National Research University – Tomsk State University, parent 2,1,3-benzothiadiazole, its 4-amino- (1) and 4-hydroxy-
6
34050 Tomsk, Russia
derivatives, and related acenaphtho[1,2-c][1,2,5]thiadiazole as
†
Electronic supporting information (ESI) available: Crystallographic data,
25
ligands. In continuation of our research, in this paper, we
additional gures and theoretical calculations. CCDC 994141 (1), 994142 (2),
94143 (3), 994144 (4). For ESI and crystallographic data in CIF or other
report two novel applications of functionalized 2,1,3-benzo-
thiadiazoles in coordination chemistry and crystal engineering,
9
electronic format see DOI: 10.1039/c4ra03342f
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 28309–28316 | 28309

View Article Online
RSC Advances
Paper
which covers the synthesis and characterization of a novel 2 : 1
It is known that compound 1 can coordinate metal centers by
complex between compound 1 and ZnCl2 (complex 2; previ- N atoms of both the heterocyclic moiety and amino group (Fig. 2,
1
4
14,18,25
25
ously, Zn complexes with any 2,1,3-benzothiadiazoles were atoms N and N , respectively).
Both the [Ir(1)Cp*Cl2]
26
18
unknown), and unprecedented 1 : 1 co-crystals of compound 1 and [Cd(1) Br ] complexes contain ligands 1 coordinated by the
2
2
1
and its synthetic precursor, 4-nitro-2,1,3-benzothiadiazole (3)
N
atom (Fig. 2, coordination type 1). Binuclear complex
CN) ](PF (ref. 14) is the only example of a
(
complex 4). All the compounds 1–4 were characterized by [{Cu(1)(CH
3
2
}
2
6 2
)
1
4
single-crystal XRD, and complexes 2 and 4 were also charac- bridging 1 ligand coordinated by both N and N atoms (Fig. 2,
terized by UV-Vis and IR spectroscopy in combination with DFT coordination type 2). In complex 2, the ligand 1 is coordinated to
4
and QTAIM calculations to clarify their bonding situations.
Zn by the N atom only, and this is the rst example of a novel
In crystalline complex 2, both molecules 1 are coordinated to mode of coordination (Fig. 2, coordination type 3).
Zn only by amino groups, thus revealing a novel type of coor- Noticeably, an attempt to prepare the complex [Zn(1)Cl2]
dination of this ligand to the metal center. However, the analogous to [{Cu(1)(CH CN) } ](PF ) (ref. 14) was unsuccess-
3
2 2
6 2
complex is weakly bonded and dissociates in CHCl
DMSO solutions.
3
, THF or ful. From the reaction mixture with 1 : 1 molar ratio of 1 and
ZnCl , only 2 was isolated, and the formation of other
2
In crystalline complex 4, molecules of 1 and 3 form innite complexes was not detected.
alternating p-stacks connected by the lateral S/N interactions.
In complex 2, the geometry of ligand 1 differs only slightly
This packing motif is different from those observed with XRD in from that of free molecule with the only exception of the elon-
4
˚
˚
the homo-crystals of 1 and 3. The bonding situation in complex gation of bond C–N by about 0.05 A (1.3869(18) A for 1 and
4
˚
4
characterized by a small charge transfer (CT) from molecule 1 1.4368(19) A for 2). It should be noted that the C–N bond of 1 in
18
to 3 is equal to ꢀ0.02e at the B97-D3 level of theory. In the the complex [Cd(1) Br ] revealing coordination type 1 is close
2
2
˚
crystalline state, complex 4 features intense CT absorption band to that in complex 2 (1.417 A). At the same time, the corre-
2
5
with a maximum at ꢀ550 nm. Therefore, 4 can be considered to sponding bond length in the complex [Ir(1)Cp*Cl
is 1.373 A˚ ,
be the rst example of CT complex, where both donor and which is even shorter than the C–N bond in individual 1.
acceptor moieties are the derivatives of 2,1,3-benzothiadiazole The crystal packing of complex 2 reveals stacks along the b
]
2
4
ring system. Similar to complex 2, complex 4 is weakly bonded, axis (Fig. 3A). Asymmetric form of the molecules results in a
and it was impossible to detect 4 in CH Cl solution due to the primitive rectangular net in the ac plane, which is slightly gof-
2
2
very low equilibrium constant of its formation.
fered (Fig. 3B). The p-planes of the neighboring ligands 1 in the
stacks are parallel with an interplanar separation of 3.08 A˚
(
Fig. S1, ESI†). It should be noted that there are pairs of short-
2
. Results and discussion
˚
ened S/N intermolecular contacts of 3.112(1) A between
neighboring almost coplanar ligands 1 (a maximum least-
squares deviation of the atoms from the mean plane is 0.15 A˚ ,
Fig. 3A).
2.1. Synthesis and characterization of 2
Complex 2 was prepared by the reaction of anhydrous ZnCl2
with two equivalents of compound 1 in Et O followed by the
crystallization of the reaction precipitate from THF–Et O
2
mixture. Performing the reaction directly in THF gave the same
results.
2
It is known that the coordination of 1 affects the NH
27
stretching modes of its NH group. For individual 1 in KBr
pellet, the n(NH ) are 3354 and 3190 cm , whereas in complex 2
they are 3283 and 3226 cm . The band shis independently
indicate that the ligand 1 coordinates with the metal center by
the N atom of the amino group. In agreement with this, the N–H
2
ꢁ
1
2
ꢁ1
According to the XRD data (Table S1, ESI†), compound 2 is a
2
molecular complex with two NH groups coordinating with the
Zn atom (Fig. 1). Two Cl atoms ll the Zn coordination envi-
ronment to a distorted tetrahedron.
ꢁ
1
stretching band of individual 1 at 3296 cm related to the
intermolecular hydrogen bond (N –H/N , for atom numbering,
4
4
see Fig. 2) is absent in the IR spectrum of complex 2.
1
In H NMR spectra, the solutions of complex 2 in CHCl , THF
3
or DMSO reveal signals of free ligand 1, indicating the
Fig. 1 ORTEP plot of 2 showing 50% probability ellipsoids (H atoms
are shown as circles). Selected bond distances ( A˚ ) and angles ( ): Zn1– Fig. 2 Coordination types of 1: by N atom (type 1), by N and N
Cl1 2.2400(3), Zn1–N4 2.0666(12), N4–Zn1–Cl1 106.19(4).
ꢂ
1
1
4
4
atoms (type 2), and by N atom (type 3).
28310 | RSC Adv., 2014, 4, 28309–28316
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
The dissociation of complex 2 in THF solution agrees with the
DFT calculations predicting that a complex of ZnCl with two
2
THF molecules is thermodynamically more favorable than
0
ꢁ1
complex 2 (DG ¼ 2.4 kcal mol for reaction Zn(THF) Cl + 21
2
2
¼
2 2
Zn(1) Cl + 2THF).
2.2. Synthesis and characterization of 4
2
,1,3-Benzothiadiazoles are moderate electron acceptors
2
involved in the synthesis of numerous CT complexes. To the
best of our knowledge, authentic (i.e. not highly polarized
30
hybrid molecules ) 2,1,3-benzothiadiazoles have never been
used as donor components of CT complexes.
Nevertheless, the co-crystallization of 1 with its synthetic
precursor 3 from hexane gives their 1 : 1 complex 4. Excess of
any component in the reaction solution does not affect the
stoichiometry of the crystalline product. Interestingly, complex
4
can be prepared under the conditions of incomplete reduction
Fig. 3 Packing diagram of 2 viewed along b axis (A) and c axis (B).
of compound 3 when bulky Fe chips are used instead of Fe
powder or the reaction time is shortened.
According to the XRD data (Table S1, ESI†), complex 4 is
composed of planar molecules 1 and 3, which are parallel to
each other (Fig. 4). The corresponding average bond lengths of
the thiadiazole units in 4 are similar and fall within 1.6058(16)–
dissociation of this complex. There are only a few examples of
similar complexes, which are also believed to dissociate in
solutions.
To rationalize the experimental results, DFT calculations
were performed. Structural optimization for the 2 : 1 complexes
with different types of coordination (type 3 for both ligands and
type 3 only for one ligand) revealed that all the structures
correspond to the minima on the potential energy surface
27
˚
˚
1
1
.6144(16) A for N–S, 1.343(2)–1.351(2) A for N–C and 1.436(2)–
˚
.437(2) A for C–C bonds. These distances are nearly equal to
that of the parent 2,1,3-benzothiadiazole and that of the
31
complex 2 (Fig. 1) and individual 1 and 3 (see below). It should
4
˚
be noted that the C–N bond in 4 (1.3602(24) A) is slightly
(Fig. S2, ESI†). Among them, one structure (Fig. S2A†) was
˚
shortened by ꢀ0.03 A as compared with that of individual 1. The
similar to that observed in the crystal (Fig. 1), and it features
crystal packing of 4 reveals tilted stacks of alternating 1 and 3
molecules. The stacks are parallel to the crystallographic axis a
and form hexagonal rod-packing in the perpendicular direction
˚
identical Zn–Cl bond distances (2.24 A) but noticeably elon-
˚
gated Zn–N distances (by ꢀ0.1 A) compared to single-crystal
XRD geometry. In contrast to the experiment, however, calcu-
lations predict that this structure is less favorable thermody-
namically (Fig. S2†).
(
Fig. 5). The neighboring stacks are connected by the shortened
˚
S/N contacts of 3.029(2) A between coplanar (maximum devi-
˚
ation of atoms from the mean plane is less that 0.13 A) 1 and 3
QTAIM analysis of the electron density distribution in
complex 2 with single-crystal XRD geometry was also per-
formed. The bond critical points (BCPs) lying between Zn and Cl
and Zn and N atoms were localized, and their topological
descriptors were calculated. The values of topological descrip-
tors, namely, electron density (r ¼ 0.075 au for Zn–Cl bond and
molecules (Fig. S3A, ESI†). The S/N distances are noticeably
shorter than the sum of the corresponding van der Waals radii
0
0
.061 au for Zn–N bond) and Laplacian of electron density (Dr ¼
.238 for Zn–Cl bond and 0.230 for Zn–N bond) were found to be
similar and typical of a closed-shell (i.e. predominantly elec-
28
trostatic) interaction. Recently, the ratio of potential and
kinetic energy densities (|V |/G ) was proposed to be an
b
b
adequate indicator for classifying bonding interactions: those
with |V |/G > 2 are covalent, with |V |/G < 1 are non-covalent,
and a partially covalent bonding was proposed in the interme-
b
b
b
b
29
b b
diate case (1 < |V |/G < 2). For the BCPs belonging to the Zn–
Cl and Zn–N bonds, this ratio was found to be 1.19 and 1.11,
respectively, and these bonds in complex 2 can be classied as
partially covalent bonds. Note that the values of all the
descriptors under consideration are slightly higher for the BCP
of the Zn–Cl bond. This is consistent with the weaker Zn–N
Fig. 4 ORTEP plot of complex 4 showing 50% probability ellipsoids (H
atoms are shown as circles). Selected distances ( A˚ ): N24/S12 3.745(2),
bonding, which manifests in the dissociation of 2 into ZnCl
2
1
and free 1 in CHCl
3
, THF or DMSO solutions ( H NMR data). N24/N13 3.746(2), N14/S22 3.795(2).
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 28309–28316 | 28311

View Article Online
RSC Advances
Paper
transitions for compounds 1 and 3 in CH
2
Cl
2
solution. It is
observed that the calculations reproduce experimental spectra
reasonably well. Fig. 7B shows the UV-Vis spectra of 1 and 3
recorded in the solid state. These spectra are recalculated from
the diffused reectance spectra of the samples, containing a
small fraction of compound (1 or 3) dispersed in the BaSO
4
matrix. Comparison of Fig. 7A and B demonstrates the quali-
tative similarity of the spectra in solid state and solution.
It is clear from Fig. 7B that the UV-Vis spectrum of 4 has an
intense absorption band in the range of 500–650 nm, which is
absent in the spectra of individual 1 and 3. Fig. 7B also repre-
sents calculated positions and oscillator strengths of the elec-
tronic transitions for the stacked tetramer of alternating 1 and 3
molecules with tilted geometry obtained from the XRD struc-
ture of 4 (Fig. 5). It is seen that the calculations well reproduce
the spectrum of the solid 4. Note that the M06-HF functional
used for calculations is known to well reproduce the CT tran-
33,34
sitions.
According to calculations, electronic transition at
about 550 nm is composed mainly of electron promotion from
the HOMO localized on 1 to the LUMO localized on 3 (Fig. 7C).
Thus, as expected, the long-wavelength band in the UV-Vis
Fig. 5 Packing pattern of complex 4 viewed along crystallographic spectrum of 4 is a CT band. B97-D3/def2-TZVP calculations
axis a.
estimate CT from 1 to 3 upon the formation of 4 to be ꢀ0.02e.
Some changes were also observed in the IR spectrum of 4 as
compared with the superposition of the spectra of individual
of 3.35 A. The arrangement of the pairs of stacks reveals that the compounds 1 and 3. The most signicant differences are
˚
˚
planes of the pairs are stepwise shied by 0.98 A along the associated with the NH
bands because they are shied up to
2
ꢁ
1
stacking axis (Fig. S3B and C, ESI†). The interplanar separation 100 cm to higher wave numbers in 4, as compared with
˚
of 1 and 3 molecules in the stacks is 3.39 A, suggesting their p–p individual 1. Analogous shis were observed earlier for the CT
3
5,36
interaction. This spacing is signicantly larger than the corre- complexes of ArNH
2
derivatives.
Furthermore, the band at
ꢁ1
sponding trans-annular distance in dimeric syn- and anti- 3296 cm in the spectrum of 1 associated with the intermo-
2.2](4,7)benzothiadiazolophanes (2.98 A and 2.94 A, respec- lecular N –H/N hydrogen bond (for atom numbering, see
tively). This fact could be explained by the appearance of Fig. 2) disappears upon the formation of 4. Indeed, the structure
4
4
˚
˚
[
5
additional chemical bonds, forming bridges between two of 4 (Fig. 4 and 5) does not show any hydrogen bonds.
1
benzothiadiazole units in the phanes.
According to H NMR data, complex 4 dissociates in solu-
It follows from XRD data (Table S1, ESI†) that the packing tion. Furthermore, the UV-Vis spectrum of 4, as well as the
pattern of 4 (Fig. 5) differs from those of individual 1 and 3 spectra of mixtures of 1 and 3 with concentrations up to 2.5 ꢃ
ꢁ
3
ꢁ2
(
Fig. 6; for XRD molecular structures of 1 and 3 see Fig. S4, 10 M (1) and 2.5 ꢃ 10 M (3) have no appreciable absorption
ESI†), although both 1 and 3 also form coplanar pairs featuring at l > 500 nm. Thus, the upper limit of the equilibrium constant
˚
˚
2 2
shortened S/N contacts (3.122(1) A for 1 and 3.088(1) A for 3; for the formation of 4 from 1 and 3 in CH Cl solution was
ꢁ1
maximum deviation of the atoms from the mean plane is 0.19 A˚ estimated to be K # 0.1 M
.
˚
for 1 and 0.28 A for 3). However, there are two S/N contacts
Very low value of equilibrium constant is in agreement with
within a pair in the homo-crystals of 1 and 3 (Fig. 6). Similar to the results of DFT calculations. We were able to optimize the
, molecules in 1 and 3 are packed as stacks with an interplanar structures of the four types of 1 : 1 complexes between 1 and 3
4
˚
˚
separation of 3.52 A for 1 and 3.22 A for 3; the angles between (complexes 4a–d; Fig. S5, ESI†); all are p-stacked. Among them,
molecules in different stacks are 85.4 in 1 and 68.4 in 3. It structure 4a is similar to that observed by XRD (Fig. 4), although
ꢂ
ꢂ
should be noted that there are intermolecular hydrogen bonds the distances between the N atoms of amino or nitro groups and
4
4
˚
N –H/N in 1 with a N/N distance of 3.206(2) A and N–H/N S atom of heterocyclic partner in optimized structures (3.54 and
ꢂ
˚
˚
angle of 169.2 . The bonds are rather long but they t to weak 3.56 A, respectively) are shorter by about 0.2 A than those in the
˚
hydrogen bonding according to the statistical research of R. S. XRD structure (3.80 and 3.75 A, respectively). In optimized
32
Rowland and R. Taylor. Overall, the crystal structure of 1 can structures 4b and 4c, amino and nitro groups are in close
˚
be described as shbone packing (Fig. 6B).
proximity (N/N, 3.30 A). In structures 4c and 4d, the hetero-
The intense dark-red color of crystalline 4 and yellow color of rings are under one another. For all the complexes 4a–d, the
the mechanical mixture of its components testify that 4 is a Gibbs free energy of formation is predicted to be positive, and
complex characterized by a CT band in the UV-Vis spectrum. therefore the equilibrium constants are expected to be as low as
ꢁ
3
ꢁ2
ꢁ1
Fig. 7A displays the UV-Vis spectra as well as the M06-HF- ꢀ10 to 10
M , which is in agreement with the experiment
calculated positions and oscillator strengths of the electronic result. For all the complexes (4a–d), calculations predict a small
28312 | RSC Adv., 2014, 4, 28309–28316
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Fig. 6 Above: packing diagram of 1 viewed along the c axis, (A) showing relative arrangement of stacks (B). Below: packing diagram of 3 viewed
along the a axis, (C) showing relative arrangement of stacks (D).
CT (ꢀ0.02–0.05e) from 1 to 3 moiety and a CT band in the UV- responsible for the non-covalent bonding interactions between
Vis spectra with a maxima at 550–600 nm (Fig. S5, ESI†).
1 and 3 were localized. Two of the BCPs belong to S22/N14 and
QTAIM analysis of the electron density distribution for the S12/C25 bond paths, and the other three belong to the N13/
XRD structure of complex 4 was also performed. Five BCPs C24, C14/C27 and C17/C26 bond paths (Fig. 4). As expected,
2 2
Fig. 7 (A) UV-Vis spectra of compounds 1 (black spectrum 1) and 3 (gray spectrum 2) in the CH Cl solution. Vertical bars indicate the positions
and oscillator strengths (f, right axis) of the electronic transitions calculated for 1 (black bars) and 3 (gray bars) at the M06-HF/def2-TZVP level of
theory. (B) Solid-state UV-Vis spectra (Kubelka–Munk functions) for complex 4 (red spectrum) and its constituents 1 (black solid curve) and 3
(
black dotted curve). Vertical red bars indicate the positions and oscillator strengths (f, right axis) of the electronic transitions calculated at the
M06-HF/def2-TZVP level for stacked tetramer of alternating 1 and 3 molecules with geometry of XRD structure of 4. (C) HOMO and LUMO of
complex 4 calculated at the same level of theory.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 28309–28316 | 28313

View Article Online
RSC Advances
Paper
both the sulfur atoms (S12 and S22) and N14 atom of the amino
substituent are involved in the bonding interactions. EuroEA3000 analyzer. H NMR spectra were recorded with a
These BCPs are characterized by very low electron density Bruker DRX-500 spectrometer at the frequency of 500.13 MHz,
Elemental analysis data were obtained with a Eurovector
1
(
r ꢀ 0.005 au) and its Laplacian (Dr ꢀ 0.015–0.019 au) and the solvent signals were used as internal standards (dH ¼ 2.50
positive total electronic energy density (HBCP ꢀ 0.001 au). These for DMSO-d
6
, 1.73 and 3.58 for THF-d , and 7.24 for chloroform-
8
37–39
values agree with the very low energy of bonding interaction.
d). IR spectra were recorded with a Bruker IFS28 instrument.
The UV-Vis spectra in solutions were measured with an Ultro-
spec 3300 pro UV/Visible spectrophotometer. The diffuse
reectance UV-Vis spectra were obtained with a Shimadzu UV-
3
. Conclusions
3
101 PL spectrophotometer. Samples for the diffuse reectance
In this work, two novel applications of functionalized 2,1,3-
benzothiadiazoles for metal coordination chemistry and crystal
engineering of organic solids are presented.
measurements were prepared by a thorough grinding of a
mixture of the compounds under study with BaSO , which was
used also as a standard. Spectral dependence of the diffuse
reectance was converted into a spectrum of a Kubelka–Munk
function.
4
2
4-Amino-2,1,3-benzothiadiazole (1) and ZnCl formed 2 : 1
complex (2) characterized by XRD. Complex 2 is the rst struc-
turally dened Zn complex with 2,1,3-benzothiadiazole ligands.
In this complex, both molecules 1 are coordinated to Zn only by
amino groups, thus revealing a novel type of coordination of
this ligand to the metal center. According to the DFT and
40
4
.2. X-ray structure determination
Single-crystal XRD data for 1–4 (Table S1, ESI†) were collected
using the graphite-monochromated MoKa-radiation (l ¼
QTAIM calculations, complex 2 is weakly bonded. As evidenced
1
by H NMR data, it dissociates in CHCl
solutions.
3
, THF or DMSO
˚
0
.71073 A) on a Bruker APEX DUO diffractometer (3) and Bruker
Nonius X8 APEX diffractometer (1, 2, 4) equipped with a 4k CCD
area detector at 150(2) K (2, 3) and at 293(2) K (1, 4). The 4-scan
technique was employed to measure intensities. Absorption
corrections were applied using the SADABS program. The
crystal structures were solved by direct methods, and rened by
the full-matrix least squares techniques with the use of the
SHELXTL package. Atomic thermal parameters for non-
hydrogen atoms were rened anisotropically. For all
compounds and amino-group in 1 and 4, the positions of
hydrogen atoms were calculated corresponding to their
geometrical conditions and rened using the riding model. For
Compound 1 and 4-nitro-2,1,3-benzothiadiazole (3) formed a
: 1 complex (4). According to XRD, in the co-crystals of 1 and 3
1
their molecules are arranged as innite alternating p-stacks
connected by lateral S/N interactions. This packing motif
differs from those observed by XRD in the homo-crystals of 1
and 3. In the crystalline state, complex 4 is characterized by the
intense CT absorption band with a maximum at ꢀ550 nm,
which was well-reproduced by M06-HF calculations. B97-D3
calculations predict a CT of ꢀ0.02e from 1 to 3 upon the
formation of 4. Overall, 4 can be considered to be a novel CT
complex, where both donor and acceptor moieties are the
derivatives of 2,1,3-benzothiadiazole ring system. Previously,
41
42
2, the hydrogen atoms of amino-groups coordinated to Zn were
localized from the electron difference map and rened in
2,1,3-benzothiadiazole derivatives were used only as the
isotropic approximation.
acceptor components of CT complexes, including those which
are promising for materials science. However, according to the
DFT and QTAIM calculations and H NMR and UV-Vis
measurements, complex 4 is weakly bonded and does not exist
in solution.
These ndings motivate further work towards the synthesis
of new metal complexes of 2,1,3-benzothiadiazole ligands also
covering their Se (and hopefully Te) congeners, as well as of new
CT complexes with these heterocycles as both donor and
acceptor components. Importantly, the crystal structure of such
complexes can be expected to be p-stacked, i.e. crystals can be
engineered.
4
.3. Syntheses
1
43
4-Amino-2,1,3-benzothiadiazole (1). (modied method ). To
a suspension of Fe powder (15 g) in boiling water (125 ml),
concentrated acetic acid (2.5 ml) and compound 3 (5.0 g, 28
mmol) were simultaneously added. The reaction mixture was
reuxed for 10 min, and the hot mixture was ltered into an ice-
cooled bath. The lter cake was washed by several portions of
hot water. The crystalline yellow-orange precipitate formed in
the cold bath was ltered and dried in air. At this stage,
compound 1 was obtained in the yield of 1.7 g (40%). Additional
1
was obtained by acidifying the ltrate with concentrated HCl
(5 ml), evaporating its volume to 2/3 of its initial volume, and
4
. Experimental and computational
extraction with several portions of CHCl
of 1 was 2.0 g (50%), m. p. 69–70 C. Found (calculated) for
C H N S (%): C 47.6 (47.7), H 3.3 (3.3), N 27.6 (27.8), S 20.6
6 5 3
3
. The combined yield
ꢂ
details
4.1. General
(
21.2).
1
Solvents and starting materials were used as received from
H NMR spectra of 1, especially signal of amino group, are
suppliers. Preparations for synthesis of 2 were performed under solvent-dependent, d: DMSO-d : 7.40 (t, 1H, Ar), 7.14 (d, 1H, Ar),
6
an argon atmosphere using standard Schlenk technique. Et O 6.57 (d, 1H, Ar), 6.19 (s, 2H, NH ); THF-d : 7.33 (t, 1H, Ar), 7.14
2
2
8
and THF used for the synthesis of 2 were dried by common (d, 1H, Ar), 6.51 (d, 1H, Ar), 5.62 (s, 2H, NH
drying agents and redistilled under argon. (M, 2H, Ar), 6.60 (d, 1H, Ar), 4.65 (s, 2H, NH
2
); CDCl
).
3
: 7.31–7.39
2
28314 | RSC Adv., 2014, 4, 28309–28316
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
IR (KBr), cm : 3354 m, 3296 m, 3190 m, 1635 m, 1605 m,
RSC Advances
ꢁ
1
To estimate the thermodynamics of the formation of the
1
9
548 s, 1490 s, 1428 m, 1345 m, 1293 m, 1147 w, 1076 w, 1019 w, complex in solution, the following calculations were performed.
00 m, 875 m, 836 m, 800 s, 741 s, 607 w, 513 m, 474 m, 458 m. Thermochemical corrections were obtained from the gas phase
Single crystals of 1 suitable for XRD were obtained by the optimization/frequency calculations at the B97-D3/def2-TZVP
slow evaporation of the toluene solution of 1.
ZnCl (1) ] (2). At ambient temperature Et
added to a mixture of anhydrous ZnCl (0.051 g, 0.374 mmol) and energies aer outlying charge correction were used to
level of theory without gCP. Single point calculations were
50
[
2
2
2
O (10 ml) was performed for THF solution using he COSMO solvation model,
2
and 1 (0.115 g, 0.760 mmol). The reaction mixture was stirred calculate enthalpies and Gibbs free energies. Note that the
for 1 day, and the precipitate was then ltered off and washed standard state for the calculations of Gibbs free energy was
with several portions of Et O and dried in air. Complex 2 was chosen to be 1 M concentration solution.
2
obtained in the form of yellow microcrystalline solid, yield 0.107
To understand the nature of the bonding interactions
37–39
g (65%). Found (calculated for C H Cl N S Zn) (%): C 32.6 between 1 and 3, QTAIM analysis was performed.
B97-D3/
1
2
10
2 6 2
(
32.8), H 2.4 (2.3), N 18.7 (19.1), S 14.7 (14.6).
def2-TZVP densities were used in QTAIM calculations. QTAIM
1
H NMR, d (DMSO-d
d, 1H, Ar), 6.19 (s, 2H, NH
6
): 7.40 (t, 1H, Ar), 7.13 (d, 1H, Ar), 6.57 calculations and graphics were produced using the Multiwfn
53
(
2
); d
8
-THF: 7.33 (T, 1H, Ar), 7.15 (d, program (version 3.2).
). UV-Vis spectra were calculated by time-dependent (TD) DFT
IR (KBr), cm : 3283 s, 3226 s, 3114 w, 1618 w, 1570 s, 1544 s, with meta-GGA M06-HF functional, especially proposed for this
54
1
H, Ar), 6.52 (d, 1H, Ar), 5.63 (s, 2H, NH
2
ꢁ1
3
3,34
1
8
491 m, 1341 w, 1258 w, 1074 w, 1056 w, 1015 m, 899 m, 848 w, type of calculations.
14 m, 808 m, 756 w, 680 m, 610 w, 500 w, 454 w.
These calculations were also performed
51
with the ORCA suit of programs.
Single crystals of 2 suitable for XRD were obtained by slow
diffusion of Et O into its solution in THF at ambient
2
temperature.
Acknowledgements
4-Nitro-2,1,3-benzothiadiazole (3). Compound
3
4
was
4
prepared according to the known literature method. Single The authors are grateful to the Russian Foundation for Basic
crystals suitable for XRD were obtained by the slow evaporation Research (projects no. 13-03-01088, 14-03-31268, 14-03-31466)
of an ethanol solution of 3.
and the Ministry of Education and Science of the Russian
1$3 (4). (a) A mixture of compounds 1 (0.105 g, 0.695 mmol) Federation (federal entitlement program Kadry, contract no.
and 3 (0.126 g, 0.696 mmol) was dissolved in boiling hexane (10 8631, and project of Joint Laboratories of Siberian Branch of the
ꢂ
ml), and the solution formed was cooled to 2 C. The precipitate Russian Academy of Sciences and National Research Universi-
was ltered and dried in air. Complex 4 was obtained in the ties) for nancial support.
form of dark-red crystals suitable for XRD, yield 0.193 g (85%),
ꢂ
m. p. 77–78 C. Found (calculated for C
6
H
5
N
3
S$C
6
H
3
N
3
SO
2
) (%)
C 43.1 (43.4), H 2.5 (2.4), N 25.3 (25.3), S 19.7 (19.3).
Notes and references
1
H NMR, d (CDCl ): 8.56 (d, 1H, Ar-NO ), 8.38 (d, 1H, Ar-NO ),
3
2
2
7
.77 (t, 1H, Ar-NO ), 7.31–7.39 (M, 2H, Ar-NH ), 6.60 (d, 1H, Ar-
1 B. A. D. Neto, A. A. M. Lapis, E. N. da Silva J u´ nior and
J. Dupont, Eur. J. Org. Chem., 2013, 2013, 228–255.
2 A. Lonchakov, O. Rakitin, N. Gritsan and A. Zibarev,
Molecules, 2013, 18, 9850–9900.
3 G. K. Dutta and S. Patil, Org. Electron., 2012, 13, 1266–1276.
4 G. Berionni, A.-M. Goncalves, C. Mathieu, T. Devic,
A. Etcheberry and R. Goumont, Phys. Chem. Chem. Phys.,
2011, 13, 2857–2869.
2
2
NH ), 4.66 (s, 2H, NH ).
2
2
ꢁ1
IR (KBr), cm : 3452 m, 3552 m, 3080 w, 1631 m, 1591 w,
547 m, 1514 s, 1488 m, 1421 m, 1371 m, 1330 s, 1312 s, 1368 m,
128 w, 996 w, 900 m, 837 m, 818 s, 735 s, 618 w, 566 w, 537 w,
56 m.
1
1
4
(b) Complex 4 can be prepared directly from compound 3
under the conditions of its reduction to compound 1 (ref. 44) by
making the reduction incomplete via the replacement of Fe
powder by Fe chips or reducing the reaction time to 5 min.
5 M. Watanabe, K. Goto, M. Fujitsuka, S. Tojo, T. Majima and
T. Shinmyozu, Bull. Chem. Soc. Jpn., 2010, 83, 1155–1161.
6
M. Watanabe, K. Goto, M. Shibahara and T. Shinmyozu,
J. Org. Chem., 2010, 75, 6104–6114.
4
.4. Quantum chemical calculations
7
T. Linder, E. Badiola, T. Baumgartner and T. C. Sutherland,
Org. Lett., 2010, 12, 4520–4523.
The geometries of 1–4 were optimized at the B97-D3 level of
theory
damping function was used in all the dispersion corrected
calculations.
45,46
47
with the def2-TZVP basis set. The Becke–Johnson
8 J.-Y. Lee, W.-S. Shin, J.-R. Haw and D.-K. Moon, J. Mater.
Chem., 2009, 19, 4938–4945.
9 Y. Shuku, R. Suizu, K. Awaga and O. Sato, CrystEngComm,
2009, 11, 2065–2068.
48
Geometrical counterpoise (gCP) correction
49
scheme was used to semi-empirically treat the BSSE effects.
50
The COSMO solvation model was employed to take into 10 R. T. Boer ´e and T. L. Roemmele, Coord. Chem. Rev., 2000,
account the solvent inuence on the geometry. The resolution- 210, 369–445.
of-the-identity approximation (RI) was used to speed up the 11 A. F. Cozzolino, I. Vargas-Baca, S. Mansour and
GGA computations. All quantum chemistry calculations were
performed using the ORCA suit of programs (version 3.0.0).
51
A. H. Mahmoudkhani, J. Am. Chem. Soc., 2005, 127, 3184–
3190.
52
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 28309–28316 | 28315

View Article Online
RSC Advances
Paper
12 V. V. Zhivonitko, A. Y. Makarov, I. Y. Bagryanskaya, 27 E. A. Bezzubets, E. K. Diachenko, I. I. Fadeeva and
Y. V. Gatilov, M. M. Shakirov and A. V. Zibarev, Eur. J.
Inorg. Chem., 2005, 2005, 4099–4108.
3 Y. Suzuki, M. Fujimori, H. Yoshikawa and K. Awaga, Chem.–
Eur. J., 2004, 10, 5158–5164.
4 M. Munakata, H. He, T. Kuroda-Sowa, M. Maekawa and
Y. Suenaga, J. Chem. Soc., Dalton Trans., 1998, 1499–1502.
5 M. Munakata, T. Kuroda-Sowa, M. Maekawa, M. Nakamura,
N. A. Ostapkevich, Zh. Obshch. Khim., 1983, 53, 612–619.
28 R. F. W. Bader and H. Ess ´e n, J. Chem. Phys., 1984, 80, 1943–
1960.
29 E. Espinosa, I. Alkorta, J. Elguero and E. Molins, J. Chem.
Phys., 2002, 117, 5529–5542.
1
1
1
30 Y. Yamashita and M. Tomura, J. Mater. Chem., 1998, 8, 1933–
1944.
S. Akiyama and S. Kitagawa, Inorg. Chem., 1994, 33, 1284– 31 V. Luzzati, Acta Crystallogr., 1951, 4, 193–200.
1291.
32 R. S. Rowland and R. Taylor, J. Phys. Chem., 1996, 100, 7384–
16 G. S. Papaefstathiou, S. P. Perlepes, A. Escuer, R. Vicente,
7391.
A. Gantis, C. P. Raptopoulou, A. Tsohos, V. Psycharis, 33 Y. Zhao and D. G. Truhlar, J. Phys. Chem. A, 2006, 110, 13126–
A. Terzis and E. G. Bakalbassis, J. Solid State Chem., 2001,
59, 371–378.
13130.
1
34 Y. Zhao and D. G. Truhlar, Acc. Chem. Res., 2008, 41, 157–167.
17 G. S. Papaefstathiou, A. Tsohos, C. P. Raptopoulou, A. Terzis, 35 A. Hindawey, Y. Issa, Y. Marghalani and R. Issa, Monatsh.
V. Psycharis, D. Gatteschi and S. P. Perlepes, Cryst. Growth
Des., 2001, 1, 191–194.
8 L. G. Kuzmina, L. P. Grigorieva, Y. T. Struchkov,
Chem., 1980, 111, 1143–1150.
36 G. Smith, U. D. Wermuth, R. C. Bott, J. M. White and
A. C. Willis, Aust. J. Chem., 2001, 54, 165–170.
1
Z. I. Ezhkova, B. E. Zaitsev, V. V. Davidov and 37 The Quantum Theory of Atoms in Molecules, ed. C. F. Motta
A. K. Molodkin, Zh. Neorg. Khim., 1980, 25, 2931–2938. and R. J. Boid, Wiley, Weinheim, 2007.
9 N. W. Alcock, A. F. Hill and M. S. Roe, J. Chem. Soc., Dalton 38 R. F. W. Bader, Monatsh. Chem., 2005, 136, 819–854.
1
2
Trans., 1990, 1737–1740.
39 R. F. W. Bader, Atoms in Molecules: A Quantum Theory, Oxford
0 K. Skorda, G. S. Papaefstathiou, A. Vaadis, A. Lithoxoidou,
University Press, Oxford, 1990.
C. P. Raptopoulou, A. Terzis, V. Psycharis, E. Bakalbassis, 40 P. Kubelka, J. Opt. Soc. Am., 1948, 38, 448–457.
V. Tangoulis and S. P. Perlepes, Inorg. Chim. Acta, 2001, 41 Bruker AXS Inc., APEX2 (Version 2.0), SAINT (Version 8.18c),
3
26, 53–64.
and SADABS (Version 2.11), Bruker Advanced X-ray Solutions,
Madison, Wisconsin, USA, 2000–2012.
42 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr.,
2008, 64, 112–122.
2
1 M. W. Renner, K. M. Barkigia, D. Melamed, K. M. Smith and
J. Fajer, Inorg. Chem., 1996, 35, 5120–5121.
2 A. S. Antsyshkina, O. A. Tokarskaya, V. S. Tsveniashvili,
2
V. N. Ostrikova, L. Y. Ukhin, M. A. Porai-Koshits and 43 L. S. Efros and R. M. Levit, Zh. Obshch. Khim., 1955, 25, 199–
A. D. Garnovskii, Koord. Khim., 1989, 15, 214–220. 209.
3 F. S. Mancilha, L. Barloy, F. S. Rodembusch, J. Dupont and 44 L. S. Efros and R. M. Levit, Zh. Obshch. Khim., 1953, 23, 1552–
M. Pfeffer, Dalton Trans., 2011, 40, 10535–10544. 1553.
4 N. A. Semenov, I. Y. Bagryanskaya, A. V. Alekseev, 45 S. Grimme, J. Comput. Chem., 2006, 27, 1787–1799.
Y. V. Gatilov, E. Lork, R. Mews, G.-V. Roeschenthaler and 46 S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys.,
2
2
A. V. Zibarev, Zh. Strukt. Khim., 2010, 51, 569–574.
2010, 132, 15404.
2
2
5 D. A. Bashirov, T. S. Sukhikh, N. V. Kuratieva, D. Y. Naumov, 47 F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005,
S. N. Konchenko, N. A. Semenov and A. V. Zibarev,
Polyhedron, 2012, 42, 168–174.
7, 3297–3305.
48 S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem.,
2011, 32, 1456–1465.
6 Historically, co-crystals have been referred to as molecular
complexes or, for clarity, multi-component crystals 49 H. Kruse and S. Grimme, J. Chem. Phys., 2012, 136, 154101.
containing charge neutral species, as opposed to salts. Yet 50 S. Sinnecker, A. Rajendran, A. Klamt, M. Diedenhofen and
the prediction of whether co-crystallization will occur is
F. Neese, J. Phys. Chem. A, 2006, 110, 2235–2245.
not yet possible and must, at. present, be answered 51 F. Neese, J. Comput. Chem., 2003, 24, 1740–1747.
empirically, see N. Blagden, D. J. Berry, A. Parkin, H. Javed, 52 F. Neese, WIREs Comput. Mol. Sci., 2012, 2, 73–78.
A. Ibrahim, P. T. Gavan, L. L. De Matos and C. C. Seaton, 53 T. Lu and F. Chen, J. Comput. Chem., 2012, 33, 580–592.
New J. Chem., 2008, 32, 1659–1672.
54 A. Dreuw and M. Head-Gordon, Chem. Rev., 2005, 105, 4009–
037.
4
28316 | RSC Adv., 2014, 4, 28309–28316
This journal is © The Royal Society of Chemistry 2014