Synthesis, characterization and theoretical study in gaseous and solid phases of the imine 4-Acetyl-N-(4-methoxybenzylidene)aniline

Article abstract of DOI:10.1016/j.molstruc.2017.06.062

In the present paper we describe the synthesis and characterization of the Schiff's base or imine 4-Acetyl-N-(4-methoxybenzylidene)aniline (1), which provided experimental support for the theoretical calculations. The imine was characterized by infrared s

Full text of DOI:10.1016/j.molstruc.2017.06.062

Accepted Manuscript
Synthesis, characterization and theoretical study in gaseous and solid phases of the
imine 4-Acetyl-N-(4-methoxybenzylidene)aniline
J.F.N. Batista, J.W. Cruz, Jr., A.C. Doriguetto, C. Torres, E.T. de Almeida, I. Camps
PII:
S0022-2860(17)30843-8
10.1016/j.molstruc.2017.06.062
MOLSTR 23948
DOI:
Reference:
To appear in: Journal of Molecular Structure
Received Date: 6 June 2016
Revised Date: 6 June 2017
Accepted Date: 14 June 2017
Please cite this article as: J.F.N. Batista, J.W. Cruz Jr., A.C. Doriguetto, C. Torres, E.T. de Almeida,
I. Camps, Synthesis, characterization and theoretical study in gaseous and solid phases of the imine
4-Acetyl-N-(4-methoxybenzylidene)aniline, Journal of Molecular Structure (2017), doi: 10.1016/
j.molstruc.2017.06.062.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Synthesis, characterization and theoretical study in gaseous and solid
phases of the imine 4-Acetyl-N-(4-methoxybenzylidene)aniline
J.F.N. Batista, J.W. Cruz Jr., A.C. Doriguetto, C. Torres, E.T. de Almeida
Instituto de Qu´ımica. Universidade Federal de Alfenas - Unifal-MG, Rua Gabriel Monteiro da Silva, 700,
CEP 37130-000, Alfenas, MG, Brazil
I. Camps
Laborato´rio de Modelagem Computacional - LaModel, Instituto de Ciˆencias Exatas. Universidade Federal
de Alfenas - Unifal-MG, Rua Jovino Fernades Sales, 2600, CEP 37130-000, Alfenas, MG, Brazil
Abstract
In the present paper we describe the synthesis and characterization of the Schiff’s base
or imine 4-Acetyl-N-(4-methoxybenzylidene)aniline (1), which provided experimental sup-
port for the theoretical calculations. The imine was characterized by infrared spectroscopy
and single crystal XRD techniques. The computational studies were performed using the
density functional theory (DFT) for the gaseous and solid phases. As similar compounds
already shown biological activity, the pharmacokinetic properties of (1) were evaluated.
Our results shown that (1), in its gaseous form, it is electronically stable and has phar-
macological drug like properties. Due to its structural similarity with commercial drugs, it
is a promise candidate to act as a nonsteroidal anti-inflammatory and to treat dementia,
sleep disorders, alcohol dependence, and psychosis. From the solid state calculations we
obtain that (1) is a low gap semiconductor and can act as an absorber for electromagnetic
radiations with energy greater that ∼ 0.9 eV .
Keywords: Schiff’s base, ab initio calculations, electronic structure, ADME properties
Preprint submitted to J. Molecular Structure
June 16, 2017

ACCEPTED MANUSCRIPT
1. Introduction
1
2
3
4
5
6
7
8
9
Nitrogenated compounds obtained by condensation reaction of primary amines and
aldehydes are called Schiff’s base or imines and have a characteristic C=N double bond,
where the nitrogen atom is bonded to an aryl or alkyl group. Imines play an important
role in Coordination Chemistry due to the formation of stable complexes with transition
metals [1]. In our recent papers, several nitrogenated ligands, mainly imines, are employed
coordinating to the metal center, such as palladium(II), in order to obtain new complex
species with antitumour [2], tuberculostatic [3; 4] and leishmanicidal activities [5].
Computational methods are widely used in material science [6] and in drug devel-
opment [7] to simulate and predict physical and chemical relevant properties. In material
10 science, properties like electron density, electrostatic potential, electronic orbitals and lower
11 energy conformer (for isolated molecules) and electronic bands and crystal energy (for pe-
12 riodical systems) are among the successfully predicted properties [6]. In drug development
13 process, the quantitative structure activity relationship and the absorption, distribution,
14 metabolism and excretion (ADME) properties are some predicted properties [7].
15
Due to its wide variety of applications [8; 9], Schiff’s base form an important class of
16 organic compounds. In the literature, several computational studies of imines that could aid
17 the design and development of new drugs, the understanding of the coordination geometry
18 and electronic properties of isolated and complexes Schiff’s base have been performed [10–
19 14].
20
The aim of this work is to present an experimental and theoretical study of the title
21 compound (see Fig. 1 for a 2D diagram). The paper is organized as follows. The ex-
22 perimental details are presented in section 2. The theoretical methodology and tools are
23 described in section 3. In section 4, the discussion of the results for the gaseous and solid
24 phase of (1) are presented. Our conclusions are summarized in section 5.
2

ACCEPTED MANUSCRIPT
Figure 1: 2D structure diagram. Chemical Abstracts Service (CAS) registry number: 23596-02-3.
25 2. Experimental methodology
26 2.1. Synthesis
27
The materials commercially available were used without purification. The synthesis of
28 the title imine (1) was carried out at room temperature. This compound was prepared as
29 described: to a solution containing p-acetophenoneamine 1.2014 g (0.01 mol) and 30.0 mL
30 of ethanol, 1.3614 g (0.01 mol) of p-anisaldehyde was added. The resulting solution was
31 stirred for 8 h and then concentrated under reduced pressure. Distilled water was added
32 and a white solid was formed. The solid was filtered off, washed thoroughly with diethyl
o
33 ether and water and dried in vacuum. Yield: 90%, mp: 125 C.
34 2.2. Infrared Spectroscopy
35
Infrared (IR) spectrum was recorded on a SHIMADZU Prestige 21 FTIR spectropho-
36 tometer. The spectrum of (1), in the form of KBr pellets, was recorded in the region
37 4000 − 400 cm−1
.
38 2.3. Single crystal X-ray diffraction
Suitable crystals of (1) were grown by slow lower the temperature in the ethanol.
39
40 Intensity data were measured with the crystal at room temperature (298 K) using the
41 Enraf-Nonius Kappa-CCD diffractometer monochromated by graphite, with MoKα radia-
˚
42 tion (λ = 0.71073 A). The cell refinements were performed using the software Collect [15]
43 and Scalepack [16] and the final cell parameters were obtained on all reflections. Data re-
44 duction was carried out using the software Denzo-SMN and Scalepack [16]. The structure
3

ACCEPTED MANUSCRIPT
45 was solved and refined using the software SHELXS97 and SHELXL97, respectively [17].
46 Non-hydrogen atoms of the molecules were unambiguously solved and full-matrix least-
47 squares refinement of these atoms with anisotropic thermal parameters was carried on.
48 The H atoms were located from the difference electron density synthesis and allowed to
˚
˚
49 ride on their parent atoms, with CH(aromatic)= 0.95 A and CH(aliphatic)= 0.97 A and
50 Uiso(H) = 1.5Ueq for methyl H atoms or 1.2Ueq for the remaining H atoms. Tables were
51 generated by WinGX [18] and the structure representations by ORTEP-3 [18] and Mer-
52 cury [19]. The intramolecular parameters were analyzed using Mogul [20], a knowledge
53 database of molecular geometry derived from Cambridge Structural Database CSD [21],
54 which provides access to information on the ideal values of bond lengths, valence angles
55 and acyclic torsion angles of similar molecular fragments.
56 3. Theoretical methodology
57
To study the structural and electronic properties of the isolated molecule (gaseous
58 phase) we performed quantum mechanics calculations. These calculations were carried out
59 with the ab initio density functional theory (DFT) methodology [22–24] together with the
60 6-311++G** basis-set considering both polarization (**) and diffuse (++) functions and
61 the B3LYP exchange-correlation functional [25] as implemented in the Jaguar package [26].
62 As input, the molecular geometry determined by the X-ray diffraction analysis (see section
63 2.3 for details) was used and then optimized. As this conformation is not necessary in the
64 lower energy conformation, all properties presented here were calculated over the optimized
65 structure.
66
Using the Janak’ s theorem [27] it is possible to estimate the chemical potential (µ),
67 the molecular hardness (η) [28] and the electrophilicity index (ω) [29] from the HOMO and
68 LUMO energies ꢀH and ꢀL, respectively as following:
ꢀL + ꢀH
∼
µ
(1)
=
2
4

ACCEPTED MANUSCRIPT
ꢀL − ꢀH
∼
η
(2)
(3)
=
2
µ2
ω =
2η
69
To determine the ADME properties the simulations were done with the QikProp soft-
70 ware [30]. When performing an evaluation, QikProp quickly analyzes the type of atoms
71 and its charges, and angles, volume and surface area of the molecule. QikProp then uses
72 these information together with the calculated physical descriptors, using Monte Carlo sta-
73 tistical mechanics in the regression equations and semi-empirical calculations. The result
74 is a precise prediction of the pharmacologically relevant properties of a molecule.
75
To study the structural and electronic properties of the solid phase, the calculations
76 were done considering the DFT methodology for periodical systems [24; 31]. In the case
77 of organic molecular crystals, one of the main interaction that maintain the cohesion of
78 the crystal is the hydrogen interaction [32; 33]. This type of non-chemically bonded,
79 nonlocal and long-ranged interaction is often named as “van der Waals” (vdW) interaction.
80 Until a few years ago, the mostly used DFT approximations implemented in the solid
81 phase simulations software were the local density approximation (LDA) and the generalized
82 gradient approximation (GGA). Since the beginning of the last decade, the first works
83 appear with new approximations including van der Waals dispersion forces [34–39].
84
We used the software SIESTA [40] together with the van der Waals density functional
85 as proposed by Klimeˇs et al. [36]. Norm conserving Troullier–Martins pseudopotentials [41]
86 were generated with the ATOM software (part of the SIESTA package) to avoid the explicit
87 treatment of the electrons. For the valence electrons, we used a split-valence double-zeta
88 basis set with polarization functions (DZP) [42]. The structure optimizations were done
89 until the Hellman-Feynman forces were below 0.04 eV A−1. To obtain results sufficiently
˚
90 accurate, convergence studies were done for the mesh cutoff energy and the number of
91 k-points. The total energy convergence for the system was obtained for a mesh cutoff of
92 300 Ry and for a k-point set of 8 × 8 × 8. The Brillouin zone was sampled folowing the
5

ACCEPTED MANUSCRIPT
93 Monkhorst and Pack scheme [43].
To better understand the intermolecular interactions, the use of tables with the inter-
94
95 atomic distances is widely used. This methodology is difficult to interpret, suffering of a
96 natural lack of information. Here we decided to use the Hirshfeld surface and fingerprint
97 tools [44–48] to easily visualize and analyze the intermolecular interactions and the packing
98 behavior.
99
The purpose of the Hirshfeld surface is to define the molecule occupied space in a crystal
100 from the partition of electron density into molecular fragments. Using this idea, a weight
101 function (wA (r)) for a molecule A in a crystal is defined as:
wA (r) = ρpromolecule (r) /ρprocrystal (r) ,
(4)
102 where ρpromolecule (r) is the sum of the electron density over the atoms in the molecule A
103 (promolecule) and ρprocrystal (r) is the sum of the electron density over the crystal (pro-
104 crystal). The Hirshfeld surface is obtained setting wA (r) = 0.5.
105 4. Results and discussion
106 4.1. Structural properties
107
The crystal data and refinement parameters are presented in Table 1. Crystallographic
108 data for (1) has been deposited in the Cambridge Crystallographic Data Centre as a
109 supplementary publication, CCDC789276.
110
Figure 2 shows an ORTEP representation of the molecular structure of (1). An overlay
111 of the molecular backbones in both the experimental and the theoretical intramolecular
112 structures of (1) clearly shows the conformational similarity between them. The r.m.s.
˚
113 deviation between analogous non-H atoms is 0.015 A.
114
The Mogul [20] analysis revealed that all bond lengths correspond with the expected
115 values for a good X-ray diffraction structure refinement. There are two molecular moieties,
116 which are individually almost flat: the moiety 1 containing the atoms O2, C15 and N1
6

ACCEPTED MANUSCRIPT
Table 1: Crystal data and structure refinement for (1).
Empirical formula
C16H15N1O2
Formula weight (gmol−1
Temperature (K)
)
253.29
298(2)
˚
Wavelength (A)
0.71069
monoclinic
P21/c
Crystal system
Space group
Unit cell dimensions:
˚
a(A)
16.553(5)
˚
b(A)
7.928(5)
˚
c(A)
10.581(5)
β (o)
107.727(5)
3
˚
Volume (A )
1322.6(11)
Z (dimer/cell)
D(calc) (Mg/m3)
4
1.272
Absorption coefficient (mm−1
F(000)
)
0.084
536
Crystal size
0.04 × 0.21 × 0.24 mm3
Theta range for data collection (o) 3.27 to 25.00
Index ranges
−19 <= h <= 18, −9 <= k <= 9, −12 <= l <= 11
7904
Reflections collected
Independent reflections
2325 [R(int) = 0.0443]
Completeness to theta = 0.50o (%) 0.0
Refinement method
Full-matrix least-squares on F2
Data / restraints / parameters
Goodness-of-fit on F2
2325 / 0 / 172
1.024
Final R indices [I>2sigma(I)]
R indices (all data)
R1 = 0.0485, wR2 = 0.1246
R1 = 0.0732, wR2 = 0.1419
7
3
˚
Largest diff. peak and hole (A )
0.212 and −0.159

ACCEPTED MANUSCRIPT
Figure 2: ORTEP representation of (1).
117 and those ones present in rings A and the moiety 2 containing ring B and N1, O1, C7
118 and C16. The largest deviations from the least squares plane through the moieties 1 and
˚
119 2 are −0.124(2) and −0.054(2) A for N1 and C7 atoms, respectively. The least squares
120 planes of the moieties 1 and 2 form an angle of 50.52(6)o. This geometric feature is also
121 highlighted by the torsional angle considering C7-N1-C8-C13 (torsion= 45.08(1)o). The N1
˚
122 atom, which belongs to both planar moieties, is slightly closer to the moiety 2 (0.039(1)A)
˚
123 than the moiety 1 (−0.124(2)A).
Figure 3: Non-classical hydrogen bonds of (1) forming chains along [101] direction.
124
The supramolecular analysis of (1) shows that there are two non-classical hydrogen
125 bonds contributing to the stabilization of the crystal packing (figure 3). Either the C5H5
126 or C16H16A groups act as intermolecular hydrogen bond donors to the O2, forming a
127 bifurcated hydrogen bond that gives rise to chains along the [101] direction. Parallel
128 chains are linked together by van der Waals interactions forming a infinite two-dimensional
129 network parallel to the plane (010). No significant π − π interactions is observed and the
130 parallel planar 2D network on the (010) planes are linked along the [010] direction by van
8

ACCEPTED MANUSCRIPT
131 der Waals interactions completing the 3D packing.
Starting from the X-ray structure, it were performed two types of structure optimiza-
132
133 tions. The first one was done setting the cell parameters fixed and letting the atoms free to
134 move (hereafter called Type-I structure). The second optimization was considering the cell
135 parameters and the atoms free to move (hereafter called Type-II structure). The crystal
136 structure parameters are summarized in Table 2.
Table 2: Crystal parameters as obtained from X-ray analysis and for the optimized crystal.
Parameter
Type-I
16.553(5)
7.928(5)
10.581(5)
90.0
Type-II
15.9987
7.5556
9.6178
90.0
˚
a (A)
˚
b (A)
˚
c (A)
α
β
γ
107.727(5) 103.7432
90.0 90.0
3
˚
V olume (A ) 1322.6(11) 1129.314
137
From Table 2 we can see that Type-II (full optimized) structure is a more dense struc-
138 ture. The Type-II structure has lower values for the cell parameters and the angles did
139 not present appreciable variations. As a result, the full optimized structure maintain its
140 monoclinic symmetry and the decreased in volume after the optimization gave place to a
141 more compact structure.
142
The calculated and experimental infrared spectra are shown in Fig. 4. All calculated
143 oscillation frequencies were positive (red ticks in Fig. 4), indicating that a minimum energy
144 optimized structure was reached.
145
The solid state experimental IR spectrum (upper solid blue line in Fig. 4), shows high
146 intensity bands at 1670 cm−1, assigned to C=N stretch characteristics of the imines or
147 Schiff’s base and the bands at 1585 cm−1 and 1510 cm−1 are assigned to C=C stretches of
9

ACCEPTED MANUSCRIPT
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
IR Experimental
IR Calculated
3000
2500
2000
1500
1000
500
Frequency (cm-1
)
Figure 4: Infrared spectra. Experimental spectrum (upper solid blue line), theoretical spectrum (down
solid red line) and calculated frequency positions (red ticks).
148 the aromatic rings. Two strong bands appears in 831 cm−1 and 846 cm−1. These bands
149 are seen in the typical spectrum of para-disubstituted rings present in (1). In the 2800 to
150 3500 cm−1 region, several bands assigned to C-H stretching of the methyl groups appear.
151
The bands observed at 2330 cm−1 and 1700 cm−1 are assigned to ν(CO) from CO2
1
152 and ketonic group, respectively. Also the larger band at 3520 cm− was assigned to ν(OH)
153 of water, contaminants hard to eliminate by the KBr pellets technique.
154
The theoretical spectrum was generated from the calculated peak positions/intensities
155 using Gaussian functions with 100 cm−1 width. It is well known that ab initio vibrational
156 frequencies are typically larger than the experimental ones [49; 50]. Taking this shift into
157 account, Fig. 4 shows a good correspondence between experimental and calculated IR
158 spectra in the peaks positions. This means that the theoretical methodology used here is
159 suitable to predict the molecular properties.
160 4.2. ADME properties
161
In silico predictions of the ADME properties had become a very useful tool decreasing
162 costs in the discovery and development of new drugs. Consequently, eliminating compounds
10

ACCEPTED MANUSCRIPT
163 with high-risk or bad ADME properties earlier in the drug production pathway has become
164 an attractive approach [51]. Alternatively, the ADME properties could guide scientists to
165 suggest modifications based on the structure-activity relationship.
166
To know the drugability of (1), the ADME properties were simulated using the QikProp
167 software [30]. Some of the most used descriptors are shown in Table 3. As it can be seen,
168 all the parameters are in agreement with the statistically favorable values. It is noteworthy
169 that no violation occurs to the Lipinski’s rule of five [52] and Jorgensen’s rule of three [53],
170 widely used descriptors in drugability predictions.
Table 3: Some of the calculated ADME properties of compound (1).
Property
V alue
253.300
11.553
6.020
3.259
-3.993
0
Recommended
130.0 − 725.0
8.0 − 35.0
4.0 − 45.0
2.0 − 6.5
MW1
QPlogPoct2
QPlogPw3
QPlogPo/w4
QPlogS5
−6.5 − 0.5
maximum is 4
RuleOfFive6
RuleOfThree7
0
maximum is 3
1
3
2
Molecular weight of the molecule. Predicted octanol/gas partition coefficient.
4
Predicted water/gas partition coefficient. Predicted octanol/water partition
5
coefficient. Predicted aqueous solubility, logS. S in mol · dm3 is the concentration of the
6
solute in a saturated solution that is in equilibrium with the crystalline solid. Number
7
of violations of Lipinski’s rule of five. Number of violations of Jorgensen’s rule of three.
171
Together with the calculation of ADME descriptors, the QikProp package is able to
172 identify similar drug molecules. In this case, the top 4 compounds with similarity greater
R
173 that 90% were Melperoneꢀ (used to treat dementia, sleep disorders, alcohol dependence
11

ACCEPTED MANUSCRIPT
R
174 and psychosis), Medifoxamineꢀ (an antidepressant which acts as a dopamine reuptake
R
R
175 inhibitor), Fenbufenꢀ and Ketoprofenꢀ both used as nonsteroidal anti-inflammatory drug
176 (NSAID). These results point out (1) as a promise candidate for treating these diseases.
177 4.3. Electronic properties
178
The highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular
179 orbital (LUMO) are shown in Fig. 5. From the diagram we can see that these orbitals are
180 asymmetrically spread over the entire molecular structure. This is a desirable characteristic
181 for molecular electronic since the device can behave differently with different bias.
(a)
(b)
Figure 5: Molecular orbitals (a) HOMO and (b) LUMO (diagrams rendered with the Maestro software [54]).
182
The calculated values of chemical potential (µ), the molecular hardness (η) and the
183 electrophilicity index are shown in Table 4. From them, we can infer that as η has a posi-
184 tive value, the redistribution of the electrons in the molecule is energetically unfavorable.
185 Also, the higher η value is, the more stable the molecule is and, therefore, the harder the
186 rearrangement of its electrons [55]. The electrophilicity index, ω, can be used as a measure
187 of the molecule energy lowering due to the maximal electron flow between the environ-
188 ment and the molecule [29]. This is a very useful tool when studying the ligand-binding
189 phenomena in drug design: higher values of ω implies lower energies for the ligand (more
190 stability).
191
In Fig. 6 the Hirshfeld surface together with the hydrogen interaction for Type-II struc-
192 ture is shown. The figure stands for a mapping of dnorm over the Hirshfeld surface. The
12

ACCEPTED MANUSCRIPT
Table 4: Calculated parameters ꢀL, ꢀH , µ, η and ω.
Property V alue(eV)
ꢀL
ꢀH
µ
-2.183
-6.221
-4.202
2.019
4.372
η
ω
193 function dnorm represents a symmetric function of distances from the inside and outside
194 of the Hirshfeld surface nuclei (di and de, respectively), relative to their respective van
195 der Waals radii. The red regions represent distances shorter than vdW separations (high-
196 lighting both donor and acceptor equally) whereas white regions represent distance equal
197 to vdW and blue regions for distances greater than vdW. The red spot in Fig. 6 indicate
198 that the van der Waals interactions are the main responsible for the crystal stability and
199 cohesion, therefore a more compact structure.
Figure 6: Hirshfeld surface and hydrogen interactions (dashed green lines). Diagram rendered with the
CrystalExplorer software [56].
200
Another useful tool to visualize the distribution and type of interactions inside the
201 crystal is the fingerprint plot. The fingerprint plots correspond to 2D histograms for the
202 distribution of (di, de) pairs from the Hirshfeld surface. In Fig. 7 the calculated fingerprint
13

ACCEPTED MANUSCRIPT
203 plots for Type-II structure are shown.
In Fig. 7a the contributions of all atoms from the molecule are represented. At first
204
˚
205 sight, it can be seen that the intermolecular contact distribution is bellow 2.2A. This is
206 in accordance to the fact that Type-II structure is more compact than Type-I were the
˚
207 contact distances extend to more than 2.5A. The sharp spikes in the plots characterize the
208 hydrogen interactions: longer spikes represent stronger hydrogen bonds, the upper spark
209 corresponds to the hydrogen interaction donor (de > di) and the other one to the hydrogen
210 interaction acceptor (di > de).
de
de
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
di
di
(Å) 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
(Å) 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
(a) All atoms
(b) C
de
de
de
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
2.4
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.6
di
di
di
(Å) 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
(Å) 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
(Å) 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
(c) H
(d) O
(e) N
Figure 7: Fingerprint plots for Type-II structure. Maps for atoms in the promolecule (a) all atoms; (b)
carbon atoms; (c) hydrogen atoms; (d) oxygen atoms and (e) nitrogen atoms. Diagrams rendered with the
CrystalExplorer software [56].
211
To get a better understanding about what atoms contribute more to the crystal cohesion
212 and how they do that, it is possible to plot the fingerprints for a given type of atom from
213 the promolecule, representing in this way, the interaction of it with all other atoms from
14

ACCEPTED MANUSCRIPT
214 the procrystal. In both structures, the carbon and hydrogen atoms are responsible for
215 more than 80% of the interactions. Their contributions are shown in figures 7b (carbon
216 atoms) and figures 7c (hydrogen atoms). Figure 7d and Fig. 7e shown the contributions
217 of oxygen and nitrogen atoms, respectively. Both atoms behave like hydrogen interaction
218 acceptors.
219
One of the properties that can be calculated within the quantum mechanical ap-
220 proach is the partial atomic charge distribution. To accomplish this task, there are several
221 methodologies like (i) Mulliken population analysis [57–60], (ii) natural population analysis
222 (NPA) [61], (iii) the Breneman-Wiberg (BW) model [62], (iv) Merz-Kollman-Singh (MKS)
223 electrostatic potential derived charges [63; 64], (v) the Bader partition scheme [65–68],
224 (vi) the Voronoi partition scheme [69; 70] and (vii) the Hirshfeld partition scheme [71].
225 However, none of them is universally accepted as the “best” for computing partial atomic
226 charges. Here, we decided to use the Hirshfeld partition scheme as it proves to be highly
227 insensitive to the choice of the basis set [71; 72]. In Fig. 8 the calculated charges for Type-II
228 structure are shown.
0.180
0.135
0.090
0.045
0.000
-0.045
-0.090
-0.135
Figure 8: Charge distribution using the Hirshfeld [71] partition scheme for Type-II structure.
229
In Fig. 9 the calculated total density of states (DOS) together with each atom contri-
230 bution are shown. The systems did not show spin polarization. The states over the Fermi
231 energy (EF , dashed green line) are the states available in the conduction band and the
15

ACCEPTED MANUSCRIPT
232 states below the Fermi energy represents the states from the valence bands. As it can be
233 seen, the conduction states are formed mainly with carbon orbitals whereas the valence
234 band states are formed with the contribution of carbon and oxygen orbitals.
H
N
O
EF = -4.572 eV
Εg = 0.861 eV (1440.0 nm)
C
Total
EF
20
10
0
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
Energy (eV)
Figure 9: Total density of state (DOS, solid black line) for Type-II structure and individual atoms con-
tributions: carbon (solid gray line), oxygen (solid red line), nitrogen (solid blue line) and hydrogen (solid
orange line).
235
A visible energy gap (Eg) equal to equal to 0.861 eV appears. From an optical point
236 of view, this energy gap implies that (1) is transparent for electromagnetic radiation with
237 energies below 0.861 eV and have available states for the absorption of radiation with ener-
238 gies greater that 0.861 eV including part of the infrared region, the visible and ultraviolet
239 regions. These properties indicate that (1) can be used as a low gap organic semiconductor
240 very useful in optoelectronic applications.
241 4.4. Gaseous vs solid
242
The structural and electronic properties of an isolated molecule (gaseous form) are
243 very different when the same molecule is in its crystal form. This is due to the collective
244 contribution of each molecule and the geometrical restriction imposed by crystal periodicity.
16

ACCEPTED MANUSCRIPT
245
The title compound (1), in its gaseous phase, has the lower possible symmetry (C1) for
246 a molecule. In its crystal form, the system belongs to the P21/c space group, which is also
247 a low–symmetry space group. A low–symmetry system has low geometrical restrictions
248 which are translated in fewer limitations for the physical properties [73].
249
In table 5, the calculated root–mean–square deviation (RMSD) and maximum distance
250 between two equivalent atoms (MaxD) using the molecule overlay feature of Mercury [19]
software is shown for the three structures.
˚
Table 5: Calculated RMSD/MaxD (in A)
Gaseous
0
Type-I
Type-II
Gaseous
0.8645/2.2184 0.2919/0.5968
Type-I 0.8645/2.2184
0
0.8460/2.4884
Type-II 0.2919/0.5968 0.8460/2.4884
0
251
252
The lower values for the RMSD and MaxD is for the pair gaseous/Type-II structures.
253 This is expected as the Type-II was optimized without any constrains, permitting the full
254 variations of atoms position together with cell parameters.
255
In Fig. 10 the calculated infrared spectra for the gaseous and solid phases (Type-II
256 structure) are shown.
257
The infrared spectra of molecules in the gaseous phase are characteristic of a free
258 molecule without any kind of interaction. In the solid phase, the crystalline order has
259 two effects. First, the symmetry of each molecule can be changed due to distortions in
260 its structure owing to the interactions with its nearest neighbors. The second effect is
261 the appearance of new oscillation modes due to the coupling (in-phase or out-of-phase)
262 between neighboring molecules in the crystal unit cell [74].
263
The oscillation modes of (1) in its gaseous phase are concentrated in the regions
264 1750 cm−1 to 250 cm−1 and over 3000 cm−1. The first region of oscillations is also visi-
265 ble in the spectrum for the solid phase whereas the oscillations in the second region were
17

ACCEPTED MANUSCRIPT
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Solid
Gaseous
3000
2500
2000
1500
1000
500
Frequency (cm-1)
Figure 10: Infrared spectra. Gaseous phase (upper blue lines), solid phase (Type-II structure, down red
lines) and calculated frequency positions (ticks).
266 not reproduced at the solid phase. The appearance of the other bands in the solid phase
267 spectrum were associated with the coupling of the 4 molecules present in the unit cell.
268
The molecular orbitals of a single molecule are formed from the superposition of atomic
269 orbitals and can be calculated using the linear combinations of atomic orbitals (LCAO)
270 approximation [6]. The results are discrete energy levels. In the solid phase (organic
271 crystals), the overlap of atomic and molecular orbitals produce bands instead discrete
272 energy levels. In this case, the translational symmetry of the crystal gives rise to another
273 quantum number k that represents the momentum of the electrons.
274
The results from the Natural Bond Orbital (NBO) analyzes [75; 76] (using the Canonical
275 Molecular Orbital Composition) for the HOMO and LUMO orbitals are summarized in
276 table 6. The HOMO has significant contributions from at least 9 NBOs whereas the
277 LUMO has a contribution from 6 NBOs. All the participant NBOs has a p character. The
278 results obtained from NBO analysis are in agreement with the orbital distribution (see
279 inset of Fig. 11).
18

ACCEPTED MANUSCRIPT
Table 6: Natural bond orbital analysis for HOMO and LUMO
HOMO
LUMO
0.395*[33]: BD(2) C3-C7
0.356*[29]: BD(2) N2-C5
0.337*[42]: BD(2) C6-C11
0.298*[21]: LP(2) O1(lp)
-0.290*[37]: BD(2) C4-C13
0.272*[23]: LP (1) C10(lp)
-0.269*[100]: BD*(2) C12-C15*
0.248*[68]: LV(1) C14(lv)
0.236*[93]: BD*(2) C9-C16*
0.536*[72]: BD*(2) N2-C5*
0.373*[90]: BD*(2) C8-O19*
0.316*[80]: BD*(2) C4-C13*
-0.311*[76]: BD*(2) C3-C7*
0.273*[68]: LV(1) C14(lv)
-0.247*[50]: BD(2) C9-C16
BD: bond, BD*: antibond, LP: lone pair (unfilled valence-shell), LV: lone vacancy (unfilled valence nonbonding).
O2s
N2s
C2s
O2p
N2p
C2p
H1s
10
0
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
Energy (eV)
Figure 11: Partial density of states for Type-II system and HOMO/LUMO molecular orbitals.
19

ACCEPTED MANUSCRIPT
280
For the solid state phase, the partial density of states is shown in Fig. 11. As can be
281 seen, the main contributions to the conduction and valence bands are from the carbon
282 atoms. For the valence band, there is also a contribution from the 2p orbitals from oxygen
283 and nitrogen atoms.
284
Comparing both results (table 6 and Fig. 11) it is possible to affirm that the same type
285 of atoms contributes for the HOMO/LUMO and valence/conduction bands, respectively.
286 5. Conclusions
287
The structural and electronic properties of compound (1) were determined in gaseous
288 and solid phases from experimental and theoretical approaches.
In this case, we obtain a good correspondence between experimental and calculated IR
289
290 spectra indicating that the theoretical methodology is adequate to simulate the properties
291 of these compound. The calculated molecular hardness (η) and the electrophilicity index
292 (ω) show an electronically stable molecule which is suitable to bind or interact with other
293 chemical species without appreciable electronic modifications. The calculation of ADME
294 properties indicate that (1) has drug like properties and from similarity with other com-
295 mercial drugs, it is a promise candidate to act as a nonsteroidal anti-inflammatory and to
296 treat dementia, sleep disorders, alcohol dependence and psychosis.
297
For the solid phase, the crystal was grown and its structure was experimentally de-
298 termined using single crystal X-ray diffraction. Starting with the experimental crystal
299 structure, two structure optimizations were done considering van der Waals dispersion
300 forces, very important in organic crystals. The type of interactions between atoms and
301 their contribution to the crystal cohesion were graphically analyzed using the Hirshfeld
302 surface and fingerprints. The calculated density of states show that (1) is a low gap semi-
303 conductor and can act as an absorber for electromagnetic radiations with energy greater
304 that ∼ 0.9 eV .
20

ACCEPTED MANUSCRIPT
305 6. Acknowledgments
We would like to thank CAPES, CNPq, FINEP, and FAPEMIG (Grant CEX-APQ-
306
307 01984-14) for financial support. This work is also a collaboration research project of a
308 member of the Rede Mineira de Qu´ımica (RQ-MG) supported by FAPEMIG (Grant REDE-
309 113/10). Part of the results presented here were developed with the help of CENAPAD-SP
310 (Centro Nacional de Processamento de Alto Desempenho em S˜ao Paulo) grant UNICAMP-
311 FINEP-MCT.
312
The authors express sincere thanks to Prof. Javier Ellena at IFSC (Instituto de F´ısica
313 da USP, S˜ao Carlos-SP) for the measurements and support of the X-ray facilities.
21

ACCEPTED MANUSCRIPT
314
315
316
[1] X. Wu, M. Tamm, Transition metal complexes supported by highly basic imidazolin-
2-iminato and imidazolin-2-imine N-donor ligands, Coord. Chem. Rev. 260 (2014)
116–138.
317
318
319
[2] N. C. CampanellaI, M. da Silva Demartini, C. Torres, E. T. de Almeida, C. M. C. ao
Paiva Gouvˆea, The cytotoxic and growth inhibitory effects of palladium(II) complexes
on MDA-MB-435 cells, Genet. Mol. Biol. 35 (2012) 159–163.
320
321
322
323
324
[3] A. C. Moro, A. C. Urbaczek, E. T. de Almeida, F. R. Pavan, C. Q. Leite, A. V.
Netto, A. E. Mauro, Binuclear cyclopalladated compounds with antitubercular ac-
tivity: synthesis and characterization of [Pd(C2, N − dmba)(X) 2(µ − bpp)](X =
Cl, Br, NCO, N 3; bpp=1,3-bis(4-pyridyl)propane), J. Coord. Chem. 65 (2012) 1434–
1442.
325
326
327
[4] C. G. Oliveira, P. I. da S. Maia, M. Miyata, F. R. Pavan, C. Q. F. Leite, E. T.
de Almeida, V. M. Deflon, Cobalt(III) complexes with thiosemicarbazones as potential
anti-Mycobacterium tuberculosis agents, J. Braz. Chem. Soc. 25 (2014) 1848–1856.
328
329
330
331
[5] L. P. Franco, E. P. de G´ois, B. S. Codonho, A. L. R. Pavan, I. de Oliveira Pereira, M. J.
Marques, E. T. de Almeida, Palladium(II) imine ligands cyclometallated complexes
with a potential leishmanicidal activity on Leishmania (L.) amazonensis, Med. Chem.
Res. 22 (2013) 1049–1056.
332
333
[6] F. Giustino, Materials Modelling using Density Functional Theory. Properties and
Predictions, Oxford University Press, 2014.
334
335
336
[7] S. Ekins, B. Wang (Eds.), Computer Applications in Pharmaceutical Research and
Development, Wiley series in drug discovery and development, Wiley Interscience,
2006.
337
338
[8] S. Kumar, D. N. Dhar, P. N. Saxena, Applications of metal complexes of Schiff base-A
review, J. Sci. Ind. Res. 68 (2009) 181–187.
22

ACCEPTED MANUSCRIPT
339
340
[9] A. Kajal, S. Bala, S. Kamboj, N. Sharma, V. Saini, Schiff Bases: A versatile pharma-
cophore, J. Catal. 2013 (2013) 893512.
341 [10] R. Joseph, J. P. Chinta, C. P. Rao, Lower Rim 1,3-Diderivative of Calix[4]arene-
342
343
344
Appended Salicylidene Imine (H2L): Experimental and Computational Studies of the
Selective Recognition of H2L toward Zn2+ and Sensing Phosphate and Amino Acid
by [ZnL], J. Org. Chem. 75 (2010) 3387–3395.
345 [11] A. K. Rochani, B. Suma, S. Kumar, J. Jays, V. Madhavan, QSAR, ADME and QSTR
346
studies of some synthesized anti-cancer 2-indolinone derivatives, Int. J. Pharm. Bio.
Sci. 1 (2010) 208–218.
347
348 [12] P. Anwar, S. Bobby, Design, docking study and ADME prediction of Chalcone deriva-
tives as potent Tubulin inhibitors, J. Comput. Methods Mol. Des. 4 (2014) 1–5.
349
350 [13] M. E. Castro, M. J. Percino, M. Cer´on, G. Soriano, V. M. Chapela, Theoretical inhi-
351
bition efficiency study of Schiff base (E)-2-(2-hydroxybenzylideneamino)phenylarsonic
acid and its isomers, Int. J. Electrochem. Sci. 9 (2014) 7890–7903.
352
353 [14] N. S. H. N. Moorthy, U. B. Vittal, C. Karthikeyan, V. Thangapandian, A. P.
354
355
356
Venkadachallam, P. Trived, Synthesis, antifungal evaluation and in silico study of
novel Schiff bases derived from 4-amino-5 (3,5-dimethoxy-phenyl)-4H-1,2,4-triazol-3-
thiol, Arabian J. Chem. in press.
357 [15] Nonius [or Hooft, R. W. W.] (1998). COLLECT. Nonius BV, Delft, The Netherlands.
358 [16] Z. Otwinowski, W. Minor, [20] Processing of X-ray diffraction data collected in oscil-
359
lation mode , in: J. Charles W. Carter (Ed.), Macromolecular Crystallography Part
A, Vol. 276 of Methods in Enzymology, Academic Press, 1997, pp. 307 – 326.
360
361 [17] G. M. Sheldrick, A short history of shelx, Acta Crystallogr. A 64 (2008) 112–122.
23

ACCEPTED MANUSCRIPT
362 [18] L. J. Farrugia, WinGX and ortep for windows: an update, J. Appl. Cryst. 45 (2012)
363
849–854.
364 [19] C. F. Macrae, P. R. Edgington, P. McCabe, E. Pidcock, G. P. Shields, R. Taylor,
365
366
M. Towler, J. van de Streek, Mercury: visualization and analysis of crystal structures,
J. Appl. Cryst. 39 (2006) 453–457.
367 [20] I. J. Bruno, J. C. Cole, M. Kessler, J. Luo, W. D. S. Motherwell, L. H. Purkis,
368
369
370
B. R. Smith, R. Taylor, R. I. Cooper, S. E. Harris, A. G. Orpen, Retrieval of
Crystallographically-Derived Molecular Geometry Information, J. Chem. Inf. Com-
put. Sci. 44 (2004) 2133–2144.
371 [21] F. H. Allen, The Cambridge Structural Database: a quarter of a million crystal struc-
tures and rising, Acta Crystallogr. B 58 (2002) 380–388.
372
373 [22] W. Koch, M. C. Holthausen, A Chemist’s Guide to Density Functional Theory, Wiley-
VCH, 2001.
374
375 [23] D. S. Sholl, J. A. Steckel, Density Functional Theory. A Practical Introduction, John
Wiley & Sons, Inc., 2009.
376
377 [24] J. P. Perdew, A. Ruzsinszky, L. A. Constantin, J. Sun, G. I. Csonka, Some fundamental
378
issues in ground-state density functional theory: A guide for the perplexed, J. Chem.
Theory Comput. 5 (2009) 902–908.
379
380 [25] C. Lee, W. Yang, R. G. Parr, Development of the Colle-Salvetti correlation-energy
formula into a functional of the electron density, Phys. Rev. B 37 (1988) 785–789.
381
382 [26] Jaguar, version 7.6, Schr¨odinger, LLC, New York, NY, 2009.
383 [27] J. Janak, Proof that ∂E/∂ni = εi in density-functional theory, Phys. Rev. 1978 (18)
384
7165–7168.
24

ACCEPTED MANUSCRIPT
385 [28] C.-G. Zhan, J. A. Nichols, D. A. Dixon, Ionization potential, electron affinity, elec-
386
tronegativity, hardness, and electron excitation energy: Molecular properties from
density functional theory orbital energies, J. Phys. Chem. A 107 (2003) 4184–4195.
387
388 [29] R. G. Parr, L. V. Szentp´aly, S. Liu, Electrophilicity index, J. Am. Chem. Soc. 121
(1999) 1922–1924.
389
390 [30] QikProp, version 3.2, Schr¨odinger, LLC, New York, NY (2009).
391 [31] R. Martin, Electronic Structure: Basic Theory and Practical Methods, Cambridge
392
University Press, 2004.
393 [32] J. D. Dunitz, X-Ray Analysis and the Structure of Organic Molecules, Wiley-VCH,
1992.
394
395 [33] E. Kaxiras, Atomic and Electronic Structure of Solids, Cambridge University Press,
2003.
396
397 [34] M. Dion, H. Rydberg, E. Schr¨oder, D. Langreth, B. Lundqvist, Van der Waals density
functional for general geometries, Phys. Rev. Lett. 92 (2004) 246401.
398
399 [35] G. Rom´an-P´erez, J. M. Soler, Efficient implementation of a van der Waals density
400
functional: Application to double-wall carbon nanotubes, Phys. Rev. Lett. 103 (2009)
096102.
401
402 [36] J. Klimeˇs, D. R. Bowler, A. Michaelides, Chemical accuracy for the van der Waals
403
density functional, J. Phys. Condens. Matter 22 (2010) 022201.
´
404 [37] K. Lee, Eamonn D. Murray, L. Kong, B. I. Lundqvist, D. C. Langreth, Higher-accuracy
405
van der Waals density functional, Phys. Rev. B 82 (2010) 081101.
406 [38] J. F. Dobson, T. Gould, Calculation of dispersion energies, J. Phys.: Condens. Matter
24 (2012) 073201.
407
25

ACCEPTED MANUSCRIPT
408 [39] D.-L. Chen, W. A. Al-Saidi, J. K. Johnson, The role of van der Waals interactions in
409
the adsorption of noble gases on metal surfaces, J. Phys.: Condens. Matter 24 (2012)
424211.
410
411 [40] J. Soler, E. Artacho, J. Gale, A. Garc´ıa, J. Junquera, P. Ordej´on, D. S´anchez-Portal,
412
The SIESTA method for ab-initio order-N materials simulation, J. Phys. Condens.
Matter 14 (2002) 2745–2779.
413
414 [41] N. Troullier, J. L. Martins, Efficient pseudopotencials for plane-wave calculation, Phys.
Rev. B 43 (1991) 1993–2006.
415
416 [42] E. Artacho, D. S´anchez-Portal, P. Ordej´on, A. Garc´ıa, J. Soler, Linear-scaling ab-initio
calculations for large and complex systems, Phys. Stat. Sol. (b) 215 (1999) 809–817.
417
418 [43] H. Monkhorst, J. Pack, Special points for Brillouin-zone integrations, Phys. Rev. B
13 (12) (1976) 5188–5192.
419
420 [44] M. A. Spackman, D. Jayatilaka, Hirshfeld surface analysis, Cryst. Eng. Comm. 11
(2009) 19–32.
421
422 [45] M. A. Spackman, J. J. McKinnon, Fingerprinting intermolecular interactions in molec-
ular crystals, Cryst. Eng. Comm. 4 (2002) 378–392.
423
424 [46] J. J. McKinnon, M. A. Spackman, A. S. Mitchell, Novel tools for visualizing and
425
exploring intermolecular interactions in molecular crystals, Acta Crystallogr. B 60
(2004) 627–668.
426
427 [47] M. A. Spackman, Molecules in crystals, Phys. Scr. 87 (2013) 048103.
428 [48] J. J. McKinnon, D. Jayatilaka, M. A. Spackman, Towards quantitative analysis of
429
intermolecular interactions with Hirshfeld surfaces, Chem. Commun. 37 (2007) 3814–
3816.
430
26

ACCEPTED MANUSCRIPT
431 [49] A. P. Scott, L. Radom, Harmonic vibrational frequencies: An evaluation of Hartree-
432
Fock, Moller-Plesset, quadratic configuration interaction, density functional theory,
and semiempirical scale factors, J. Phys. Chem. 100 (1996) 16502–16513.
433
434 [50] M. L. Laury, M. J. Carlson, A. K. Wilson, Vibrational frequency scale factors for
435
density functional theory and the polarization consistent basis sets, J. Comp. Chem.
30 (2012) 2380–2387.
436
437 [51] Torsten Schwede, Manuel C. Peitsch (Eds.), Computational Structural Biology. Meth-
ods and Applications, World Scientific, 2008.
438
439 [52] C. A. Lipinski, F. Lombardo, B. W. Dominy, P. J. Feeney, Experimental and com-
440
putational approaches to estimate solubility and permeability in drug discovery and
development settings, Adv. Drug Delivery Rev. 46 (2001) 3–26.
441
442 [53] W. L. Jorgensen, E. M. Duffy, Prediction of drug solubility from structure, Adv. Drug
Delivery Rev. 54 (2002) 355–366.
443
444 [54] Maestro, version 11.1.011, Schr¨odinger, LLC, New York, NY (2017).
445 [55] R. G. Parr, R. G. Pearson, Absolute hardness: companion parameter to absolute
446
electronegativity, J. Am. Chem. Soc. 105 (1983) 7512–7516.
447 [56] S. Wolff, D. Grimwood, J. McKinnon, M. Turner, D. Jayatilaka, M. Spackman, Crys-
talExplorer (Version 3.1), University of Western Australia (2012).
448
449 [57] R. Mulliken, Electronic population analysis on LCAO–MO molecular wave functions.
I, J. Chem. Phys. 23 (1955) 1833–1840.
450
451 [58] R. Mulliken, Electronic population analysis on LCAO–MO molecular wave functions.
452
II. Overlap populations, bond orders, and covalent bond energies, J. Chem. Phys. 23
(1955) 1841–1846.
453
27

ACCEPTED MANUSCRIPT
454 [59] R. Mulliken, Electronic population analysis on LCAO-MO molecular wave functions.
455
III. Effects of hybridization on overlap and gross AO populations, J. Chem. Phys. 23
(1955) 2338–2342.
456
457 [60] R. Mulliken, Electronic population analysis on LCAO-MO molecular wave functions.
458
IV. Bonding and antibonding in LCAO and valence-bond theories, J. Chem. Phys. 23
(1955) 2343–2346.
459
460 [61] A. E. Reed, R. B. Weinstock, F. Weinhold, Natural population analysis, J. Chem.
Phys. 83 (1985) 735–746.
461
462 [62] C. M. Breneman, K. B. Wiberg, Determining atom-centered monopoles from molecular
463
electrostatic potentials. The need for high sampling density in formamide conforma-
tional analysis, J. Comp. Chem. 11 (1990) 361–373.
464
465 [63] U. C. Singh, P. A. Kollman, An approach to computing electrostatic charges for
molecules, J. Comp. Chem. 5 (1984) 129–145.
466
467 [64] B. H. Besler, K. M. Merz, P. A. Kollman, Atomic charges derived from semiempirical
methods, J. Comp. Chem. 11 (1990) 431–439.
468
469 [65] R. F. W. Bader, Atoms in Molecules: A Quantum Theory, Clarendon Press, 1994.
470 [66] G. Henkelman, A. Arnaldsson, H. J´onsson, A fast and robust algorithm for Bader
471
decomposition of charge density, Comput. Mater. Sci. 36 (2006) 354–360.
472 [67] E. Sanville, S. Kenny, R. Smith, G. Henkelman, An improved grid-based algorithm
for Bader charge allocation, J. Comp. Chem. 28 (2007) 899–908.
473
474 [68] W. Tang, E. Sanville, G. Henkelman, A grid-based Bader analysis algorithm without
lattice bias, J. Phys.: Condens. Matter 21 (2009) 084204.
475
28

ACCEPTED MANUSCRIPT
476 [69] F. M. Bickelhaup, N. J. R. van Eikema Hommes, C. F. Guerra, E. J. Baerends, The
477
carbon–lithium electron pair bond in (CH 3Li) n (n = 1, 2, 4), Organometallics 15
478
(1996) 2923–2931.
479 [70] C. F. Guerra, J. W. Handgraaf, E. J. Baerends, F. M. Bickelhaupt, Voronoi deforma-
480
tion density (VDD) charges: Assessment of the Mulliken, Bader, Hirshfeld, Weinhold,
and VDD methods for charge analysis, J. Comp. Chem. 25 (2004) 189–210.
481
482 [71] F. L. Hirshfeld, Bonded-atom fragments for describing molecular charge densities,
Theor. Chim. Acta 44 (1977) 129–138.
483
484 [72] F. Martin, H. Zipse, Charge distribution in the water molecule. A comparison of
methods, J. Comput. Chem. 26 (2005) 97–105.
485
486 [73] C. Giacovazzo, H. L. Monaco, G. Artioli, D. Viterbo, M. Milanesio, G. Gilli, P. Gilli,
487
G. Zanotti, G. Ferraris, Fundamentals of crystallography, Oxford University Press,
1992.
488
489 [74] M. Donahue, E. Botonjic-Sehic, D. Wells, C. W. Brown, Understanding infrared and
raman spectra of pharmaceutical polymorphs, Am. Pharmaceut. Rev. 14 (2011) 1.
490
491 [75] F. Weinhold, C. R. Landis, Discovering chemistry with natural bond orbitals, John
Wiley & Sons, Inc., Hoboken, New Jersey, 2012.
492
493 [76] F. Weinhold, E. D. Glendening, NBO 6.0 Program Manual. Natural Bond Orbital
Analysis Programs (2013).
494
29