Synthesis, characterization, self-assembly and non-ohmic Schottky barrier diode behaviors of two iron(iii) based semiconductors with theoretical insight

Article abstract of DOI:10.1039/d0ce00223b

Two mononuclear iron(iii) complexes, [Fe(L1)(N3)] and [Fe(L2)(N3)], {H2L1 = N,N′-bis(3-methoxysalicylidene)diethylenetriamine and H2L2 = N,N′-bis(3-ethoxysalicylidene)diethylenetriamine} have been synthesized and characterized by elemental, spectral and X

Full text of DOI:10.1039/d0ce00223b

CrystEngComm
View Article Online
View Journal
PAPER
Synthesis, characterization, self-assembly and
non-ohmic Schottky barrier diode behaviors of
two iron(III) based semiconductors with theoretical
insight†
Cite this: DOI: 10.1039/d0ce00223b
b
Tanmoy Basak,a Dhananjoy Das,b Partha Pratim Ray,
c
a
*
Snehasis Banerjee
and Shouvik Chattopadhyay
Two mononuclear iron(III) complexes, [FeIJL1)IJN3)] and [FeIJL2)IJN3)], {H2L1
=
N,N′-bisIJ3-
methoxysalicylidene)diethylenetriamine and H2L2 = N,N′-bis(3-ethoxysalicylidene)diethylenetriamine} have
been synthesized and characterized by elemental, spectral and X-ray crystallographic studies. Structural
features have been examined in detail that reveal the formation of interesting supramolecular networks
generated through weak non-covalent interactions. The current–voltage characteristic curves for an Al/
complex Schottky-barrier diode (SBD) exhibit non-ohmic behavior. Important parameters like ideality
factor, barrier height and series resistance are measured with the help of thermionic emission (TE) theory.
Space-charge-limited current (SCLC) theory is employed to evaluate the charge transport parameters such
as effective carrier mobility and transit time for both complexes. Complex 1 is found to be more
conductive. To obtain insight into the physical nature of weak non-covalent interactions, Bader's quantum
theory of atoms-in-molecules (QTAIM) is used extensively. Additionally, the non-covalent interaction
reduced density gradient (NCI-RDG) methods established nicely the presence of such non-covalent
intermolecular interactions.
Received 13th February 2020,
Accepted 29th June 2020
DOI: 10.1039/d0ce00223b
rsc.li/crystengcomm
Many mononuclear metal complexes have also been used
to fabricate a Schottky diode.5 The electronic charge in these
Introduction
Metal–organic frameworks and coordination polymers have
the ability to be used in optoelectronics.1 A Schottky diode is
one of such optoelectronic devices which may be fabricated
by any semiconducting material showing a rectifying nature
at the metal–semiconductor junction.2 In a Schottky diode, a
junction is formed between metals (such as aluminum, silver
or platinum) and a semiconductor.3 The device performance
and reliability are dependent on the interfacial properties of
metal–semiconductor junctions. Use of semiconducting
inorganic complexes to fabricate a Schottky diode is also
reported in the literature.4
complexes may be transported via their supramolecular
architectures, e.g. H-bonding network, π–π interaction
assembly, etc.6 However, prior to this work, no systematic
analysis was performed to check the ability of these
supramolecular interactions for carrying charges in their
crystalline assembly. In this work, we have synthesized two
more or less similar mononuclear complexes, differing only
in the pendant side arms; one contains methoxy side arms
and the other contains ethoxy side arms. Thus, everything in
these two complexes is practically the same, except their solid
state supramolecular interactions. Our intention is to
correlate electrical properties with their supramolecular
interactions.
a Department of Chemistry, Inorganic Section, Jadavpur University, Kolkata-
700032, India. E-mail: shouvik.chem@gmail.com
b Department of Physics, Jadavpur University, Kolkata-700032, India.
E-mail: parthapray@yahoo.com
c Govt. College of Engineering and Leather Technology, Salt Lake Sector-III, Block-
LB, Kolkata 700106, India
In this work, we concentrate on iron(III), because the
ability of iron(III) Schiff base complexes to exhibit Schottky
barrier diode behavior was not revealed as yet. So, the
behavior
of
iron(III)-Schiff
base
complex
based
semiconductors is completely unknown. Therefore, we have
synthesized two mononuclear iron(III) complexes, [Fe(L1)(N3)]
(1) and [Fe(L2)(N3)] (2), with two similar N3O2 donor Schiff
† Electronic supplementary information (ESI) available: Contains detailed
discussion of supramolecular interactions, IR spectra, device fabrication, etc.
CCDC 1981611 and 1981612 contain the supplementary crystallographic data for
complexes 1 and 2. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/d0ce00223b
base ligands [H2L1
=
N,N′-bisIJ3-methoxysalicylidene)-
diethylenetriamine and H2L2 = N,N′-bisIJ3-ethoxysalicylidene)-
This journal is © The Royal Society of Chemistry 2020
CrystEngComm

View Article Online
Paper
CrystEngComm
diethylenetriamine]. Structures of both complexes were
confirmed by single crystal X-ray diffraction analysis. Non-
covalent interactions in the supramolecular assembly of both
complexes were studied energetically using theoretical (DFT)
calculations. These interactions have also been analyzed
using several computational tools, including Bader's “atoms-
in-molecules” (AIM) and reduced density gradient (NCI-RDG)
methods, which established nicely the presence of such non-
covalent intermolecular interactions. The results show that
complex 2 is much more stabilized by strong H-bonds and C–
H⋯O interactions compared to complex 1. Determination of
electrical parameters indicates that complex 1 is 16 times
more conductive than complex 2. This difference is
correlated with the extent of hydrogen bonding interactions
in these complexes.
separately added to the acetonitrile solutions of the different
Schiff base ligands (H2L1 and H2L2) under refluxing
conditions. Aqueous solutions of sodium azide (∼1 mmol,
0.070 g) were then separately added to each of the solutions
and refluxed further for ca. 30 min. Both solutions were kept
in open atmosphere after filtration. Dark crystalline products
were obtained after one week. X-ray quality single crystals
were collected from these crystalline products.
Complex 1, yield: 0.331 g, 71% (based on iron). Anal. calc.
for C20H23FeN6O4 (467.29): C, 38.26; H, 3.03; N, 14.87%.
Found: C, 38.1; H, 2.9; N, 15.0%. FT-IR (KBr, cm−1): 3297
(νN–H); 2936–2822 (νC–H); 2050 (νN ); 1621 (νCN). λmax (nm)
3
[εmax (L mol−1 cm−1)] (acetonitrile): 518 (2.3 × 102), 330 (5.2 ×
103), 269 (1.3 × 104), 229 (2.4 × 104). Magnetic moment =
5.82μB.
Complex 2, yield: 0.356 g, 72% (based on iron). Anal. calc.
for C22H27FeN6O4 (495.35): C, 51.41; H, 4.96; N, 17.98%.
Found: C, 51.3; H, 4.8; N, 18.1%. FT-IR (KBr, cm−1): 3172
Experimental
Materials
(νN–H); 2980–2830 (νC–H); 2042 (νN ); 1621 (νCN). λmax (nm)
3
[εmax(L mol−1 cm−1)] (acetonitrile): 523 (2.45 × 102), 330 (5.4 ×
103), 268 (1.3 × 104), 229 (2.5 × 104). Magnetic moment =
5.81μB.
All other chemicals were of reagent grade and used as
purchased from Sigma-Aldrich without further purification.
Caution!!! Although no problems were encountered in this
work, organic ligands in the presence of azide are potentially
explosive. Only a small amount of the material should be
prepared and it should be handled with care.
Computational details
All calculations were performed with Gaussian 09. Both
molecules were geometry-optimized in their ground spin
state using spin-unrestricted B3LYP density functional. Here,
the Los Alamos effective core potentials LANL2DZ basis set
was employed for the Fe atom. On the other hand, the split-
valence 6-31G(d) basis set was applied for the other atoms.
The starting structures of the investigated complexes used
were from their X-ray crystallographic data. The geometry
optimization was performed without any constraint, and the
nature of stationary points was confirmed by normal-mode
analysis.
Preparation
Preparation of ligands
Preparation
of
H2L1
{N,N′-bisIJ3-
methoxysalicylidene)diethylenetriamine}. The Schiff base ligand,
H2L1, was prepared by 1 : 2 condensation of diethyltriamine (∼1
mmol, 0.10 mL) and 3-methoxysalicylaldehyde (∼2 mmol, 0.310
g) in acetonitrile (15 mL) for ca. 1 h. The ligand was not isolated
and the yellow coloured acetonitrile solution was directly used
for the synthesis of complex 1.
Preparation
of
H2L2
{N,N′-bisIJ3-
ethoxysalicylidene)diethylenetriamine}. The other Schiff base
ligand, H2L2, was prepared using a similar method to that
used for H2L1, except that 3-ethoxysalicylaldehyde (∼2 mmol,
0.350 g) was used instead of 3-methoxysalicylaldehyde. It was
also not isolated and was used directly for the synthesis of
complex 2.
The topological features derived from Bader's theory of
atoms in molecules (AIM) were applied to understand the
electron-density features like charge density (ρ) and Laplacian of
charge density (∇ ρ) using ADF2014.10. The recently developed
reduced density gradient (RDG) based NCI (non-covalent
interaction) index calculations were applied for real-space
visualization of both attractive (van der Waals and hydrogen-
bonding) and repulsive (steric) interactions based on properties
of the electron density. Herein, single-point calculations were
2
Preparation of the complexes
[Fe(L1)(N3)] (1) and [Fe(L2)(N3)] (2). Two acetonitrile
solutions of iron(III) chloride (∼1 mmol, 0.170 g) were
Scheme 1 Synthetic route for complexes 1 and 2.
CrystEngComm
This journal is © The Royal Society of Chemistry 2020

View Article Online
CrystEngComm
Paper
Table 2 Selected bond lengths (Å) of complexes 1 and 2
based on the structures obtained from X-ray studies and in
these structures, the hydrogen atom positions were normalized
before computation.
In order to calculate the hydrogen bond energy at the very
accurate CCSD(T)/jul-cc-pVTZ level including BSSE correction
and the SAPT2 + (3)δMP2/aug-cc-pVTZ level, the following
latest formula, derived by Emamian et al.,7 was used. This
allows the bond energy to be decomposed into physically
meaningful components to shed light on the electronic
nature of the considered hydrogen bond interactions:
Complex
1
2
Fe(1)–O(1)phenoxo
Fe(1)–O(2)phenoxo
Fe(1)–N(1)imine
1.9243(16)
1.9350(17)
2.1243(4)
2.156(2)
1.939(2)
1.935(3)
2.127(3)
2.167(3)
2.249(3)
2.027(3)
Fe(1)–N(3)imine
Fe(1)–N(2)amine
Fe(1)–N(4)pseudohalide
2.292(2)
2.030(2)
fixed positions. All other hydrogen atoms were placed in their
geometrically idealized positions and constrained to ride on
their parent atoms. Multi-scan empirical absorption corrections
were applied to the data using the program SADABS.9 One
carbon atom is disordered over two positions (C22 and C22A)
and two sets of positions are refined for both with occupancies
of x and 1 − x with x = 0.57. All the H atoms, attached to C(22),
are also disordered and two sets of positions are refined for all
of them with occupancies of x and 1 − x, with x converging to
0.57. Only one (obviously with the higher occupancy) of the
disordered positions is shown in the figure.
Hydrogen bond energy (kcal mol−1) = −223.08 × ρBCP(a. u.)
+ 0.7423
Physical measurements
Elemental analysis (carbon, hydrogen and nitrogen) was
performed using a PerkinElmer 240C elemental analyzer. The
IR spectrum in KBr (4500–500 cm−1) was recorded with a
PerkinElmer Spectrum Two spectrophotometer. Electronic
spectra were recorded in acetonitrile on a Shimadzu UV-1700
spectrophotometer. Magnetic susceptibility measurement was
performed with an EG & PAR vibrating sample magnetometer
(Model 155) at room temperature and diamagnetic
corrections were performed using Pascal's constants.
Hirshfeld surfaces
Hirshfeld surfaces10 and the associated 2D-fingerprint11 plots
were calculated using Crystal Explorer12 which accepted a
structure input file in CIF format. Bond lengths to hydrogen
atoms were set to standard values.13 For each point on the
Hirshfeld isosurface, two distances, de, the distance from the
point to the nearest nucleus external to the surface and di,
the distance to the nearest nucleus internal to the surface,
were defined. The normalized contact distance (dnorm) based
on de and di was given by:
X-ray crystallography
Suitable crystals of the complexes were used for data collection
using a ‘Bruker D8 QUEST area detector’ diffractometer with
graphite-monochromated Mo Kα radiation (λ = 0.71073 Å).
Molecular structures were solved by the direct method and
refined by full-matrix least squares on F2 using the SHELX-18
package.8 Non-hydrogen atoms were refined with anisotropic
thermal parameters. The hydrogen atoms attached to nitrogen
atoms were located in difference Fourier maps and were kept at
À
Á
À
Á
di − rvidw
de − rvedw
dnorm
¼
þ
rvidw
rvedw
where rvidw and revdw are the van der Waals radii of the atoms.
The value of dnorm is negative or positive depending on the
Table 1 Crystal data and refinement details for complexes 1 and 2
Complex
1
2
Formula
Formula weight
Temperature (K)
Crystal system
Space group
a (Å)
C20H23FeN6O4
467.29
273
Monoclinic
P21/n
10.7235(10)
10.3296(9)
19.0512(16)
102.018(3)
4
1.504
0.772
972
28 292
4554
3691
0.039
0.0531, 0.1049
0.0373, 0.0930
C22H27FeN6O4
495.35
273
Monoclinic
P21/c
11.572(12)
14.317(14)
15.366(16)
112.00(3)
4
1.394
0.679
1036
81 077
5192
4694
0.034
0.0484, 0.1361
0.0425, 0.1260
Table 3 Selected bond angles (°) of complexes 1 and 2
Complex
1
2
O(1)–Fe(1)–O(2)
O(1)–Fe(1)–N(1)
O(1)–Fe(1)–N(2)
O(1)–Fe(1)–N(3)
O(1)–Fe(1)–N(4)
O(2)–Fe(1)–N(1)
O(2)–Fe(1)–N(2)
O(2)–Fe(1)–N(3)
O(2)–Fe(1)–N(4)
N(1)–Fe(1)–N(2)
N(1)–Fe(1)–N(3)
N(1)–Fe(1)–N(4)
N(2)–Fe(1)–N(3)
N(2)–Fe(1)–N(4)
N(3)–Fe(1)–N(4)
91.29(7)
83.99(7)
91.20(7)
83.25(7)
b (Å)
c (Å)
β (°)
Z
109.21(7)
174.55(7)
88.49(8)
96.21(7)
155.87(8)
84.47(7)
104.60(8)
74.20(8)
99.85(7)
158.01(8)
75.70(7)
88.97(8)
89.27(8)
108.95(7)
175.09(7)
88.92(8)
95.00(7)
156.17(7)
84.38(7)
103.96(8)
75.51(8)
99.19(7)
159.64(8)
75.86(8)
89.44(8)
90.12(8)
dcal (g cm−3
)
μ (mm−1
F(000)
)
Total reflection
Unique reflections
Observed data[I > 2σ(I)]
R(int)
R1, wR2 (all data)
R1, wR2 [I > 2σ(I)]
This journal is © The Royal Society of Chemistry 2020
CrystEngComm

View Article Online
Paper
CrystEngComm
Fig. 1 Perspective views of complexes 1 and 2 with a selective atom numbering scheme. Hydrogen atoms have been omitted for clarity.
intermolecular contacts, being shorter or longer than the van
der Waals separations. The parameter dnorm displayed a
surface with a red–white–blue colour scheme, where bright
red spots highlighted shorter contacts, white regions
represented contacts around the van der Waals separation,
and blue regions were devoid of close contacts. For a given
crystal structure and set of spherical atomic electron
densities, the Hirshfeld surface was unique14 and it was this
property that suggested the possibility of gaining additional
insight into the intermolecular interaction of molecular
crystals.
diethyltriamine
3-ethoxysalicylaldehyde
with
3-methoxysalicylaldehyde
in acetonitrile,
and
respectively.15
Acetonitrile solutions of both ligands were reacted with
iron(III) chloride under reflux. Aqueous solutions of sodium
azide were used as a co-ligand. Both complexes (1 and 2) are
stable at room temperature. The synthetic route for both
complexes is shown in Scheme 1.
Description of the structures
[Fe(L1)(N3)] (1) and [Fe(L2)(N3)] (2). X-ray crystal structure
determination reveals that both complexes crystallize in the
monoclinic space group P21/n. The details of the
crystallographic data and refinements are given in Table 1.
Important bond lengths and bond angles are gathered in
Tables 2 and 3, respectively. The perspective views of both
complexes are shown in Fig. 1. In each complex, the iron(III)
Results and discussion
Synthesis
The N3O2 donor pentadentate Schiff base ligands H2L1 and
H2L2 were prepared by a 1 : 2 condensation reaction of
Table 4 Comparison of the bond lengths in the coordination sphere of complexes 1 and 2 with other reported complexes
Complexes
[Fe(La)N3]
Fe(III)–Ophenolate
Fe(III) –Nimine
Fe(III) –Namine
Fe(III)–Nazide
Ref
1.8667(9)
1.8859(10)
1.8596(10)
1.8674(9)
1.961(2)
1.922(2)
1.902(2)
1.956(2)
1.971(1)
1.931(1)
1.877(2)
1.931(2)
1.890(2)
1.965(2)
1.9243(16)
1.9350(17)
1.939(2)
1.935(3)
1.9213(11)
1.9532(11)
1.9433(12)
1.9271(11)
2.096(2)
2.072(2)
2.059(3)
2.057(3)
2.104(1)
2.094(1)
2.002(2)
2.013(2)
2.121(2)
2.090(2)
2.1243(4)
2.156(2)
2.127(3)
2.167(3)
2.0012(12)
1.9657(12)
16a
[Fe(La)NCS]
[Fe(Lb)NCSe·C3H6O]
[Fe(Lc)NCS]
[Fe(Lc)(N3)]
[Fe(Ld)(NCS)]
Fe(Le)(N3)]
1
2.0065(12)
2.223(2)
2.167(3)
2.213(1)
2.078(2)
2.159(2)
2.292(2)
2.249(3)
1.9524(12)
2.141(2)
2.128(3)
2.055(1)
2.012(2)
2.080(2)
2.030(2)
2.027(3)
16a
16a
16b
16b
16c
16c
This work
This work
2
La = N,N′-bis({2-hydroxy-3,5-dimethylphenyl}phenyl)-methylidene-1,6-diamino-3-azapentane; Lb = N,N′-bisIJ{2-hydroxy-3-ethoxyphenyl}methylidene)-
1,6-diamino-3-azapentane; Lc = N,N′-bis-(2-aminoethyl)-1,3-propanediamine; Ld = (1-{3-[2-(2-hydroxynaphthalen-1-yl)methylenaminoethylamino]-
propylimino}methyl)naphthalene-2-ol; Le = (2-{3-[2-(1-hydroxynaphthalen-2-yl)methylenaminoethylamino]propylimino}methyl)naphthalene-1-ol.
CrystEngComm
This journal is © The Royal Society of Chemistry 2020

View Article Online
CrystEngComm
Paper
center, Fe(1), adopts
a
six-coordinate pseudo-octahedral
C(10)–N(2) and Fe(1)–N(2)–C(11)–C(12)–N(3), has half-chair
conformations. The puckering parameters18 for the ring
Fe(1)–N(1)–C(9)–C(10)–N(2) are q = 0.467(3) Å and φ =
303.7(3)° in complex 1 and q = 0.457(2) Å and φ = 301.0(2)° in
complex 2. The puckering parameters for the other ring Fe(1)–
N(2)–C(11)–C(12)–N(3) are q = 0.493(3) Å and φ = 130.0(3)° for
complex 1 and q = 0.474(2) Å and φ = 124.3(2)° for complex 2.
geometry. The central iron(III) is coordinated with the
corresponding deprotonated N3O2-donor pentadentate Schiff
base ligand and one terminal N3 ligand. Among the six
coordination sites, the deprotonated Schiff base ligand {(L1)2−
for 1 and (L2)2− for 2} occupies five coordination sites by two
imine nitrogen atoms, N(1) and N(3), one amine nitrogen atom,
N(2), and two phenoxo oxygen atoms, O(1) and O(2). The
remaining sixth coordination site is occupied by the nitrogen
atom, N(4), of an azide ion. The iron(III)–Nimine bond lengths are
shorter than the iron(III)–Namine bond lengths (Table 4), as is
also observed in similar complexes.16 The iron(III)–Ophenoxide
bond lengths (Table 2) are the shortest within the coordination
sphere and are also similar to those observed in other
complexes.16 The values of the iron(III)–Nazide bond lengths vary
in the range of 2.02–2.03 Å, as is also observed in similar
complexes.16a,c The azide moiety is practically linear in each
complex with the N-N-N angles close to 177.9° (in 1) and 178.4°
(in 2). Other Fe(III)-Schiff base complexes with terminal azide
also show similar characteristics.17 Structural parameters of
both the complexes are compared with the reported complexes
and shown in Table 4.
Hirshfeld surface analysis
The Hirshfeld surfaces of both complexes 1 and 2 are
mapped over dnorm (range − 0.1 Å to 1.5 Å), shape index and
curvedness (Fig. S5, ESI†). Red spots on the Hirshfeld
surfaces mapped with dnorm denote the dominant
interactions. Shape index can be used to identify
complementary hollows (red) and bumps (blue) where two
molecular surfaces touch one another. Curvedness is a
function of the root-mean-square curvature of the surface
and maps of curvedness typically show large green regions
(relatively flat) separated by dark blue edges (large positive
curvature).
For both complexes, the main interactions are observed
between the oxygen and hydrogen atoms (O⋯H). Other
visible interactions in the Hirshfeld surfaces correspond to
C⋯H and N⋯H contacts which are shown in Fig. 2.
The deviations of coordinating atoms, O(2), N(1), N(3) and
N(4), in the basal plane from the mean plane passing through
them are 0.2217(17), 0.1620(19), −0.2390(19) and −0.2590(2) Å
for complex 1 and 0.208(14), 0.1522(17), −0.2226(18) and
−0.237(2) Å for complex 2, respectively and also the deviation
of Fe(1) from the same plane is 0.1144(3) Å for complex 1 and
0.0998(3) Å for complex 2. Each of two saturated five-
membered chelate rings for both complexes, Fe(1)–N(1)–C(9)–
Magnetic properties, IR and UV-vis spectra and
thermogravimetric analysis
The magnetic moment of each complex is ∼5.80 BM at room
temperature. This value is closer to the theoretical value (5.92
Fig. 2 Fingerprint plots: different contacts contributed to the total Hirshfeld surface area of complexes 1 and 2.
This journal is © The Royal Society of Chemistry 2020
CrystEngComm

View Article Online
Paper
CrystEngComm
Fig. 3 UV-vis plots of complexes 1 and 2.
BM) for the magnetically isolated high-spin iron(III) ion with
spin forbidden, and therefore they are not observed. The
moderately strong band (at 518 nm for 1 and 523 nm for 2)
in the electronic spectrum of each complex may be assigned
five unpaired electrons (S = 5/2). Thus, both complexes are
2
high spin iron(III) complexes having the t2g3eg electronic
configuration.
to the azide-to-iron(III) charge transfer (CT) transition.17c,22
A
IR and electronic spectra of both complexes are in good
agreement with those of their corresponding crystal
structures. The IR spectra of both complexes exhibit a strong
band at 1618 cm−1, corresponding to the azomethine (CN)
stretching vibration.19 There is a common sharp band that
arises around 2050 cm−1 for both complexes, due to the
presence of terminal azide.17 Characteristic small bands of
the corresponding N–H vibrations are located in the interval
of 3300–3172 cm−1.20 C−H stretching vibrations vary in the
range of 2942–2968 cm−1.21 IR spectra of both complexes are
shown in Fig. S6 (ESI†).
stronger band (at 330 nm for both 1 and 2) may tentatively
be assigned to a superposition of the amine-to-iron(III) and
phenoxy-to-iron(III) CT.23 For both complexes, bands around
230 and 270 nm may be assigned to the intraligand π → π*
and n → π* transitions.24 The band positions and intensities
are
comparable
with
those
found
in
similar
complexes.17c,22,23 UV-vis spectra of both complexes are
shown in Fig. 3.
Thermogravimetric analyses of both complexes were
performed up to 100 °C in nitrogen atmosphere with a rate
of 2 °C min−1 . Both complexes are stable up to this
temperature. However, as both compounds contain metal
bound azide and also some organic ligands, TGA experiment
is not continued above 100 °C. It could be dangerous to raise
The electronic absorption spectra of both complexes are
similar in nature. The d–d transitions in high-spin iron(III) in
an octahedral field should be very weak, as they are strictly
Fig. 4 Tauc plots to estimate the direct bandgap of complexes 1 (a) and 2 (b).
CrystEngComm
This journal is © The Royal Society of Chemistry 2020

View Article Online
CrystEngComm
Paper
resistance (RS) of the fabricated Al/complex/ITO configured
metal–semiconductor diodes were measured (Table 5). The
derivations are provided in the ESI.† The barrier heights of
the fabricated diodes of complexes 1 and 2 were obtained as
0.64 eV and 0.72 eV, respectively.27 It is noticeable that the
rectification ratio of the complex 1 based devices increased
by almost 1.7 times whereas the barrier height decreased
accordingly compared to the complex 2 based devices.28
The conductivity (σ) was calculated in the ohmic region of
the I–V characteristic curve and the values were found to be
16.3 × 10−5 S m−1 and 1.02 × 10−5 S m−1 for complexes 1 and
2, respectively (Table 5).
To understand the charge transport mechanism of the
metal–semiconductor junction, a log I vs. log V curve was
plotted (Fig. 6a). From this figure, it has been observed that
the charge conduction mechanism of the device could be
explained by dividing the current response curve into two
regions. Region-I follows the ohmic (I ∝ V) conduction, which
arises due to the effect of series resistance of the complexes.
In region-II, the current is governed by the power law (I ∝
V2). To explain this region, the space-charge-limited-current
(SCLC) theory has been proven to be effective in the
literature.29
Fig. 5 I–V characteristic curves of the fabricated Schottky diodes of
complexes 1 and 2. Inset is the current response as a function of
voltage in a logarithmic scale.
the temperature above this temperature. However, it could
easily be concluded that both complexes are stable at least
up to 100 °C.
From the SCLC data, important charge transport parameters,
such as effective carrier mobility (μeff), transit time (τ) and
mobility-transit time product (μτ) of the device, have been
evaluated. The effective carrier mobility was estimated from the
slope of the I vs. V2 plot using the Mott–Gurney equation for the
SCLC region (Fig. 6b), which is given by:30
Optical properties
To illustrate the light-absorbing properties, the electronic
absorbance spectra of complexes 1 and 2 were recorded in
the wavelength range of 200–800 nm (Fig. 3). The absorbance
spectra depict that the absorbance edge around 550–650 nm
for complex 1 is shifted towards the higher wavelength region
in comparison to complex 2 which indicates a lower optical
band gap (Eg). Using Tauc's equation (eqn (S1), ESI†),25 the
bandgaps (allowed direct) were estimated to be 1.79 eV and
3.01 eV for complexes 1 and 2, respectively (Fig. 4). So, the
ꢀ
ꢁ
9μeff ε0εrAeff V2
I ¼
(1)
d3
8
where I is the current, Aeff is the effective diode area, ε0 is the
permittivity of free space, εr is the dielectric constant and d is
the thickness of the diode. The values of the dielectric constant
(εr) for both complexes were calculated by utilizing eqn (S4)
(ESI†).31
The effective carrier mobility of complex 1 was found to
be 1.16 × 10−4 m2 V−1 s−1 and for complex 2, it was 6.49 ×
10−7 m2 V−1 s−1 for the ITO/complex/Al configuration. The
transit time (τ), which affects the charge carrier generation
and recombination, is an essential parameter to measure the
suitability of the material for diode application. It is
evaluated using the following equation:32
lower bandgap energy of complex
electronic transition from the valence band to conduction
band.
1 suggests a faster
Electrical properties
The current–voltage (I–V) characterization of the fabricated Al/
complex/ITO device was carried out by applying a dc bias voltage
of 1 V at room temperature (300 K). For both complexes, the
diodes show a non-linear rectifying nature (Fig. 5).
ꢀ ꢁ
The analysis of diode parameters was performed by
employing the thermionic emission theory for the Schottky
diode.26 Using Cheung's equation (eqn (S3) and (S4), ESI†),
the values of ideality factor (η), barrier height (ϕB) and series
9ε0εrAeff
8d
V
I
τ ¼
(2)
Table 5 Schottky diode parameters
RS (Ω)
ΦB (eV)
Ideality
factor
Rectification
ratio
Conductivity
(Sm−1
Complex
)
From G(J) vs. J
From H(J) vs. J
1
2
0.88
0.69
28
17
16.3 × 10−5
1.02 ×10−5
9.87 × 102
1.67 × 104
9.57 × 102
1.48 × 104
0.64
0.72
This journal is © The Royal Society of Chemistry 2020
CrystEngComm

View Article Online
Paper
CrystEngComm
Fig. 6 (a) log I vs. log V and (b) I vs. V2 plots for complexes 1 and 2.
The calculated values of the effective carrier mobility,
transit time and mobility-transit time are tabulated in
Table 6. The carrier mobility (μ) and transit time (τ) of the
charge carriers have a major impact on the charge transport
mechanism of the device.33 Table 6 shows that the carrier
mobility (μ) and transit time (τ) of the carriers of complex 1
are better in comparison with those of complex 2. The lower
Theoretical study on non-covalent interactions
Hydrogen bonding interactions in both complexes are
characterized by AIM analysis, which gives orange spheres as
BCPs (Fig. 7). The existence of hydrogen bonds is also well
supported by the RDG plots and NCI surfaces, as shown in
Fig. 8.
Here, we computed the reduced density gradient (RDG) to
mobility of complex
2 may be related to its (weaker)
represent the deviation from
a homogeneous electron
supramolecular architectures formed by π⋯π and C–H⋯π
distribution. As presented in the following eqn (5), ∇ is the
gradient operator and |∇ρ| is the electronic density gradient
mode. Now, it is a helpful approach to explore and visualize
different kinds of non-covalent interactions (NCIs) in real
space such as both intra and intermolecular weak
interactions like hydrogen bonds and van der Waals forces.
Therefore, the NCI index is used to investigate NCIs in the
investigated complexes.
interactions.6
The excellence of both complexes for device applications
cannot be explained by the carrier mobility only but also by
the density of states (DOS) at the Fermi level. The DOS at the
Fermi level was determined (Table 6) from the I–V
characteristic curves of the SCLC region with the help of the
following equation:34
2εrε0ΔV
N′ðEF Þ ¼
(3)
1
j∇ρj
qed2ΔEF
S ¼
(5)
1
3
4
ρ
2
3
2ð3π Þ
where EF is the quasi-Fermi level, and was estimated by the
equation:
The color mapped isosurfaces and corresponding scatter
diagrams of RDG versus sign(λ2)ρ for the investigated
complexes in the monomers and dimers are shown in Fig. 7.
As discussed, the results characterized the hydrogen bonds
via colored isosurfaces according to the values of sign(λ2)ρ.
The sign of λ2 is used to distinguish bonded (λ2 < 0) and
nonbonded (λ2 > 0) interactions, whereas the electron density
is an indicator of the bonding strength. Large negative values
of sign(λ2)ρ indicate attractive interactions (such as hydrogen
ꢀ
ꢁ
I2V1
I1V2
ΔEF ¼ kBTln
(4)
where, V1 and V2 and I1 and I2 are two different voltages and
currents in the space charge region. So, the lower DOS of the
complex 1 based diode indicates a lower trap density in the
energy bandgap which increases the carrier density through
the Schottky junction.28
Table 6 SCLC parameters of the fabricated diodes
Complex
μeff (m2 V−1 s−1
)
τ (s)
μeff τ (m2 V−1
)
DOS N′ (eV−1 m−3
)
1
2
1.16 × 10−4
6.49 × 10−7
5.89 × 10−9
9.74 × 10−7
6.84 × 10−13
6.32 × 10−13
4.61 × 1039
87.34 × 1039
CrystEngComm
This journal is © The Royal Society of Chemistry 2020

View Article Online
CrystEngComm
Paper
Fig. 7 AIM analyses of the dimers of investigated complexes 1 and 2 showing intermolecular hydrogen bonding. Bond critical points are
represented by orange spheres.
bonding), whereas, if sign(λ2)ρ is large and positive, the
interaction is nonbonding (usually a steric effect). Values
near zero indicate weak vdW interactions. The colors of RDG
(s) vs. sign(λ2)ρ and the isosurfaces have the same meaning;
blue color signifies hydrogen bonds, green, van der Waals
forces and red, steric hindrance. The darker the
corresponding color, the stronger the interaction.
scatter plot. This is also supported by the interaction energy.
The results show that complex 2 is highly stabilized by two
strong H-bonds (ρ(BCP) = 0.0235 a.u.). By applying the
formula as discussed in the computational section, the
calculated hydrogen bond energy is 4.50 kcal mol−1 per
H-bond. As stated, complex 1 has one weak H-bond (ρ(BCP) =
0.0118 a.u.) with 1.89 kcal mol−1 bond energy.
The green region in Fig.
characterizes the hydrogen bonding in the solid-state of both
complexes. Both the complexes show a very similar RDG
8
(NCI picture) clearly
Additionally, there are six C–H⋯O interactions in complex
2 compared to such interactions in complex 1. These
additional interactions also stabilize complex 2.
Fig. 8 Noncovalent interaction (NCI) analysis for the hydrogen bonding and C–H⋯O interaction of the investigated complexes computed at the
B3LYP/Lanl2DZ/6-31G(d) level.
This journal is © The Royal Society of Chemistry 2020
CrystEngComm

View Article Online
Paper
CrystEngComm
Bandgap and density of states of the investigated complexes
The energy difference between the highest occupied and
The result for the theoretical and experimental bandgaps
in complex 1 is quite satisfactory (theoretical 1.76 eV;
experimental 1.79 eV). However, the theoretical bandgap
(3.32 eV) differs a little bit from the experimental bandgap
(3.01 eV) in complex 2. The theoretical band gap was
calculated from the HOMO–LUMO difference of a single
molecule and therefore, the effect of strong H-bonding is not
included in this calculation. The actual bandgap is lowered
due to the presence of strong hydrogen bonding interaction
in complex 2.
lowest unoccupied MOs (ΔE
= ELUMO–EHOMO, eV) is
sometimes assumed as the bandgap. We have performed
DFT calculations to calculate the energy difference between
the HOMO and LUMO, ΔE for both complexes 1 and 2. The
results show that the calculated bandgaps are 1.76 eV (α-
HOMO–α-LUMO) and 1.89 eV (β-HOMO–β-LUMO) for
complex 1 and 3.32 eV (α-HOMO–α-LUMO) and 2.80 eV (β-
HOMO–β-LUMO) for complex 2. The total and partial DOS
calculations shown in Fig. 9 reveal that the valence bands are
mainly dominated by the p-orbitals of the ligands, and to a
lesser extent, by the d-orbitals of the iron atoms. In addition,
the main contributors to the conduction band are the
p-orbitals of the ligand (H2L) and the d-orbitals of the iron
atoms. In both the valence band and conduction band, the
contribution of the s-orbital is poor.
Concluding remarks
In this work, two structurally similar Schiff bases have been
used to synthesize two mononuclear iron(III) complexes of
identical structures (1 and 2). The bandgaps of both
complexes were determined from spectral data and were
corroborated by DFT calculations. The bandgap in 2 (3.01 eV)
Fig. 9 The total and partial DOS calculation results of valence bands.
CrystEngComm
This journal is © The Royal Society of Chemistry 2020

View Article Online
CrystEngComm
Paper
is found to be much higher than that in 1 (1.79 eV). This
indicates that complex 2 is somehow stabilized in its ground
state. The stabilization may come from strong hydrogen
bonds and C–H⋯O interactions in complex 2. NCI-RDG and
QTAIM analyses of supramolecular interactions clearly
indicate that the stabilization energy originating from
hydrogen bonds in complex 1 is much less than that found
in complex 2. This supports the higher bandgap in 2
compared to 1, and in turn, also supports the higher
conductivity (∼16 times) of complex 1 compared to complex
2. The mobility (μeff) of the complex 1 based device is
therefore also enhanced (by ∼175%) compared to that of the
complex 2 based device.
M. G. B. Drew, P. P. Ray and S. Chattopadhyay, New J. Chem.,
2018, 42, 15295–15305.
4 (a) S. Sonmezoglu, S. Senkul, R. Tas, G. Cankaya and M.
Can, Solid State Sci., 2010, 12, 706–711; (b) Y. S. Ocaka, M.
Kulakcib, T. Kılıcoglua, R. Turanb and K. Akkılıc, Synth.
Met., 2009, 159, 1603–1607.
5 S. Konar, A. Dey, S. R. Choudhury, K. Das, S. Chatterjee, P. P.
Ray, J. Ortega-Castro, A. Frontera and S. Mukhopadhyay,
J. Phys. Chem. C, 2018, 122, 8724–8734.
6 S. Roy, A. Dey, M. G. B. Drew, P. P. Ray and S.
Chattopadhyay, New J. Chem., 2019, 43, 5020–5031.
7 S. Emamian, T. Lu, H. Kruse and H. Emamian, J. Comput.
Chem., 2019, 40, 2868–2881.
Mobility of a complex depends on the concentration of
8 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr.,
2008, A 64, 112–122.
impurities,
temperature
defects
and
electron–hole
concentrations. Higher mobility of any complex always leads
to better device performance. Thus, complex 1 is expected to
be a better candidate for the fast switching electronic device
applications than complex 2.
9 G. M. Sheldrick, SADABS, V2014/5, Software for Empirical
Absorption Correction, University of Göttingen, Institute fur
AnorganischeChemiederUniversitat, Gottingen, Germany,
1999–2003.
10 (a) M. A. Spackman and D. Jayatilaka, CrystEngComm,
2009, 11, 19; (b) F. L. Hirshfeld, Theor. Chim. Acta, 1977, 44,
129; (c) H. F. Clausen, M. S. Chevallier, M. A. Spackman and
B. B. Iversen, New J. Chem., 2010, 34, 193.
Conflicts of interest
There are no conflicts of interest.
11 (a) A. L. Rohl, M. Moret, W. Kaminsky, K. Claborn, J. J.
McKinnon and B. Kahr, Cryst. Growth Des., 2008, 8, 4517; (b)
A. Parkin, G. Barr, W. Dong, C. J. Gilmore, D. Jayatilaka, J. J.
McKinnon, M. A. Spackman and C. C. Wilson,
CrystEngComm, 2007, 9, 648; (c) M. A. Spackman and J. J.
McKinnon, CrystEngComm, 2002, 4, 378.
12 S. K. Wolff, D. J. Grimwood, J. J. McKinnon, D. Jayatilaka
and M. A. Spackman, CrystalExplorer 2.0, University of
Western Australia, Perth, Australia, 2007, http://
hirshfeldsurfacenet.blogspot.com.
Acknowledgements
T. B. thanks the CSIR, India, for awarding a Senior Research
Fellowship [Sanction No. 09/096(0861)/2016-EMR-I]. D. D.
acknowledges the University Grants Commission (UGC), India
for providing the NET-Junior Research Fellowship. S. C.
gratefully acknowledges the UGC-CAS II program, Department
of Chemistry, Jadavpur University, for financial support under
the head [Chemicals/Consumables/Glassware]. One of the
authors (PPR) gratefully acknowledges the financial support
for this work from the SERB-DST, Govt. of India (Sanction No.
EMR/2016/005387, Dated - 24.07.2017) and the RUSA 2.0
program of Jadavpur University.
13 F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G. Orpen
and R. J. Taylor, J. Chem. Soc., Perkin Trans. 1, 1987, 2, S1.
14 J. J. Kinnon, M. A. Spackman and A. S. Mitchell, Acta
Crystallogr., Sect. B: Struct. Sci., 2004, 60, 627.
15 T. Basak, K. Ghosh and S. Chattopadhyay, Polyhedron,
2018, 146, 81–92.
References
16 (a) R. Herchel and Z. Trávníček, Dalton Trans., 2013, 42,
16279–16288; (b) P. Masarova, P. Zoufaly, J. Moncol, I.
Nemec, Jan Pavlik, M. Gembicky, Z. Travnıcek, R. Boca and
I. Salitros, New J. Chem., 2015, 39, 508–519; (c) C. Krüger, P.
Augustín, I. Nemec, Z. Trávnícek, H. Oshio, R. Boca and F.
Renz, Eur. J. Inorg. Chem., 2013, 902–915.
17 (a) T. Basak, K. Ghosh, C. J. Gómez-García and S.
Chattopadhyay, Polyhedron, 2018, 146, 42–54; (b) S. Roy, S.
Halder, A. Dey, K. Harms, P. P. Ray and S. Chattopadhyay,
New J. Chem., 2020, 44, 1285–1293; (c) S. Naiya, M. G. B.
Drew, C. Diaz, J. Ribas and A. Ghosh, Eur. J. Inorg. Chem.,
2011, 4993–4999.
18 (a) D. Cremer and J. A. Pople, J. Am. Chem. Soc., 1975, 97,
1354–1358; (b) D. Cremer, Acta Crystallogr., Sect. B: Struct.
Sci., 1984, 40, 498–500.
19 L. Pogany, J. Moncol, J. Pavlik and I. Salitrosm, New J.
Chem., 2017, 41, 5904–5915.
1 (a) Z. Wang, Zi Wang, B. Lin, X. Hu, Y. F. Wei, C. Zhang, B.
An, C. Wang and W. Lin, ACS Appl. Mater. Interfaces, 2017, 9,
35253–35259; (b) B. Bhattacharya, A. Layek, Md. M. Alam,
D. K. Maity, S. Chakrabarti, P. P. Ray and D. Ghoshal, Chem.
Commun., 2014, 50, 7858–7861; (c) S. Roy, A. Dey, P. P. Ray,
J. O. Castro, A. Frontera and S. Chattopadhyay, Chem.
Commun., 2015, 51, 12974–12976.
2 (a) A. Turut and F. Kolleli, J. Appl. Phys., 1992, 72, 818–819;
(b) K. Akkılıç, Y. S. Ocak, T. Kılıçoglu, S. Ilhan and H. Temel,
Curr. Appl. Phys., 2010, 10, 337–341; (c) M. Pattabi, S.
Krishnan, Ganesh and X. Mathew, Sol. Energy, 2007, 81,
111–116.
3 (a) D. Sinha and J. Lee, Nano Lett., 2014, 14, 4660–4664; (b)
G. S. Chunga, K. S. Kima and F. Yakuphanoglu, J. Alloys
Compd., 2010, 507, 508–512; (c) F. Yakuphanoglu, Sol. Energy
Mater. Sol. Cells, 2007, 91, 1182–1186; (d) S. Roy, S. Halder,
This journal is © The Royal Society of Chemistry 2020
CrystEngComm

View Article Online
Paper
CrystEngComm
20 M. Karmakar, T. Basak and S. Chattopadhyay, New J. Chem.,
2019, 43, 4432–4443.
21 T. Basak, A. Bhattacharyya, M. Das, K. Harms, A. Bauza, A.
Frontera and S. Chattopadhyay, ChemistrySelect, 2017, 2,
6286–6295.
27 J. Datta, M. Dasa, S. Sil, S. Kumar, A. Dey, R. Jana, S.
Bandyopadhyay and P. P. Ray, Mater. Sci. Semicond. Process.,
2019, 91, 133–145.
28 M. Das, J. Datta, R. Jana, S. Sil, S. Halder and P. P. Ray, New
J. Chem., 2017, 41, 5476–5486.
22 S. Naiyaa, S. Giri, S. Biswas, M. G. B. Drew and A. Ghosh,
Polyhedron, 2014, 73, 139–145.
29 P. W. M. Blom, M. J. M. de Jong and M. G. van Munster,
Phys. Rev. B: Condens. Matter Mater. Phys., 1997, 55,
R656–R659.
30 B. Gan, J. AhnRusli, Q. Zhang, S. F. Yoon, V. A. Ligatchev, J.
Yu, K. Chew and Q. F. Huang, J. Appl. Phys., 2001, 89,
5747–5753.
31 A. Dey, S. Middya, R. Jana, M. Das, J. Datta, A. Layek and
P. P. Ray, J. Mater. Sci.: Mater. Electron., 2016, 27, 6325–6335.
32 S. Sil, A. Dey, J. Datta, M. Das, R. Jana, S. Halder, J. Dhar, D.
Sanyal and P. P. Ray, Mater. Res. Bull., 2018, 106, 337–345.
33 (a) J. Datta, M. Das, S. Sil, S. Kumar, A. Dey, R. Jana, S.
Bandyopadhyay and P. P. Ray, Mater. Sci. Semicond. Process.,
2019, 91, 133–145; (b) S. Sil, J. Datta, M. Das, R. Jana, S.
Halder, A. Biswas, D. Sanyaland and P. P. Ray, J. Mater. Sci.:
Mater. Electron., 2018, 29, 5014–5024.
23 (a) R. Biswas, C. Diaz, A. Bauzá, A. Frontera and A. Ghosh,
Dalton Trans., 2013, 42, 12274–12283; (b) S. Jana, A.
Bhattacharyya, B. N. Ghosh, K. Rissnen, S. Herrero, R. J.
Aparicio and S. Chattopadhyay, Inorg. Chim. Acta, 2016, 453,
715–723; (c) F. Banse, V. Balland, C. Philouze, E. Rivierea, L.
Tchertanova and J.-J. Girerd, Inorg. Chim. Acta, 2003, 353,
223–230.
24 R. Biswas, M. G. B. Drew, C. Estarellas, A. Frontera and A.
Ghosh, Eur. J. Inorg. Chem., 2011, 2558–2566.
25 J. Tauc, Amorphous and Liquid Semiconductors, Plenum Press,
New York, 1974.
26 (a) S. M. Sze, Physics of Semiconductor Devices, Wiley, New
York, 3rd edn, 2006, p. 198; (b) E. H. Rhoderick and R. H.
Williams, Metal-semiconductor contacts, Clarendon, Oxford,
2nd edn, 1988.
34 M. Şahin, H. Durmuş and R. Kaplan, Appl. Surf. Sci.,
2006, 252, 6269–6274.
CrystEngComm
This journal is © The Royal Society of Chemistry 2020