Internal rotation and intramolecular hydrogen bonding in thiosalicylamide: gas phase electron diffraction study supported by quantum chemical calculations

Article abstract of DOI:10.1007/s11224-019-01369-y

The molecular structure of thiosalicylamide (2-hydroxythiobenzamide) was investigated in the gas phase at 401?K by means of gas electron diffraction (GED) combined with quantum chemical (QC) calculations. Special attention was paid to the internal rotation of the thioamide group. Structural refinement was performed taking into account rovibrational corrections to the thermal-average internuclear distances calculated with harmonic and anharmonic (cubic) MP2/cc-pVTZ force constants in terms of static and dynamic models. It was shown that both models fitted the GED data equally well. The results of the GED refinement revealed that in the equilibrium structure, the thioamide group is twisted by about 30° with respect to the phenol ring plane. This is the result of an interatomic repulsion of hydrogen atom in the amide group from the closest hydrogen atom of the benzene ring, which overcomes the energy gain from the π?π conjugation of the thioamide group and the aromatic system of thiosalicylamide. Natural bond orbital (NBO) analysis and comparison of the thiosalicylamide molecular structure with those of related compounds revealed hydrogen-bonded fragment between the hydroxyl and thiocarbonyl groups. The structure of thiosalicylamide in the gas phase was found to be markedly different from that in the solid phase due to the effect of intermolecular hydrogen bonding in the crystal.

Full text of DOI:10.1007/s11224-019-01369-y

Structural Chemistry
https://doi.org/10.1007/s11224-019-01369-y
ORIGINAL RESEARCH
Internal rotation and intramolecular hydrogen bonding
in thiosalicylamide: gas phase electron diffraction study supported
by quantum chemical calculations
Inna N. Kolesnikova1 & Anatolii N. Rykov1 & Maxim V. Shuvalov1 & Igor F. Shishkov1
Received: 19 March 2019 /Accepted: 30 May 2019
#
Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract
The molecular structure of thiosalicylamide (2-hydroxythiobenzamide) was investigated in the gas phase at 401 K by means of
gas electron diffraction (GED) combined with quantum chemical (QC) calculations. Special attention was paid to the internal
rotation of the thioamide group. Structural refinement was performed taking into account rovibrational corrections to the thermal-
average internuclear distances calculated with harmonic and anharmonic (cubic) MP2/cc-pVTZ force constants in terms of static
and dynamic models. It was shown that both models fitted the GED data equally well. The results of the GED refinement revealed
that in the equilibrium structure, the thioamide group is twisted by about 30° with respect to the phenol ring plane. This is the
result of an interatomic repulsion of hydrogen atom in the amide group from the closest hydrogen atom of the benzene ring,
which overcomes the energy gain from the π−π conjugation of the thioamide group and the aromatic system of thiosalicylamide.
Natural bond orbital (NBO) analysis and comparison of the thiosalicylamide molecular structure with those of related compounds
revealed hydrogen-bonded fragment between the hydroxyl and thiocarbonyl groups. The structure of thiosalicylamide in the gas
phase was found to be markedly different from that in the solid phase due to the effect of intermolecular hydrogen bonding in the
crystal.
.
.
.
.
Keywords Thiosalicylamide 2-Hydroxythiobenzamide Molecular structure Gas electron diffraction Quantum chemistry
Internal rotation Intramolecular hydrogen bonding
.
Introduction
(XRD) [7]. According to XRD, the asymmetric unit of
thiosalicylamide contains two crystallographically indepen-
dent molecules with similar molecular parameters. Two types
of hydrogen bonds are present in the crystal phase: intramo-
lecular O–H...S bonds (rH...S ~ 2.0 Å) and intermolecular N–
H...S and N–H...O bonds (rH...S ~ 2.6 Å and rH...O ~ 2.1 Å).
The investigation of the structural properties of
thiosalicylamide and its metal complexes made by means of
IR spectroscopy confirmed that the OH group is bonded in the
solid state [12]. Unlike salicylamide which is almost flat in the
crystal phase, the thioamide group in thiosalicylamide is twist-
ed by approximately 20° relative to the plane of the ring [7].
While theoretical investigations of free molecules were per-
formed for a few compounds similar to thiosalicylamide, such
as 4-hydroxythiobenzamide [13] and 2-hydroxy-N-
metylthiobenzamide [14], neither computational nor experi-
mental data could be found for the gaseous structure of
thiosalicylamide. Thus, steric effect of the large sulfur atom
on the conformation of gaseous thiosalicylamide molecule in
comparison with its oxo-analog has remained unknown.
Salicylamides and thiosalicylamides are compounds of special
interest in medicinal chemistry [1–5]. While the molecule
structure of salicylamide, a well-known non-prescription an-
tiseptic drug, has been extensively studied [6–11], data are
scarce concerning the structure and conformation of its
bioisosteric sulfur-containing analog, thiosalicylamide
(Fig. 1). The molecular structure of thiosalicylamide in the
solid phase was determined by means of X-ray diffraction
Electronic supplementary material The online version of this article
(https://doi.org/10.1007/s11224-019-01369-y) contains supplementary
material, which is available to authorized users.
* Inna N. Kolesnikova
i_n_kolesnikova@mail.ru
1
Chemistry Department, Lomonosov Moscow State University,
Leninsky Gory 1, Moscow, Russian Federation 119991

Struct Chem
MACO EM-FILM EMS photo films. The details of the GED
experiments are given in Table 1.
An Epson Perfection 4870 photo scanner was used to pro-
cess the films with recorded diffraction patterns in order to
obtain digital images of GED data in 16-bit/600-dpi gray
scale. The resulting data for LD and SD were transformed into
mean optical density functions of s, with an increment of
0.2 Å−1 by applying the PLATE program [16]. The experi-
mental intensity curves are presented in Fig. 2, which also
displays the background lines B(s) obtained for a dynamic
model. The background curves for a static model are given
in Fig. S1 of the Electronic Supplementary Material (ESM).
Fig. 1 Structural formula of thiosalicylamide and the numbering of atoms
This report is aimed to fill this gap and presents the first
study of thiosalicylamide molecular structure in the gas phase
by means of gas electron diffraction (GED) supported by
quantum chemical (QC) calculations. The obtained results
are used to discuss the structural changes caused by the re-
placement of oxygen atom in salicylamide by the larger and
less electronegative sulfur and the effect of intermolecular
hydrogen bonding on the structure of thiosalicylamide in the
crystal phase.
QC calculations
Computational programs and methods
All QC calculations presented in this work were made using
Gaussian03 software [17]. The methods were B3LYP [18, 19]
and MP2 [20] including basis sets 6-31G(d,p) [21] and cc-
pVTZ [22], with appropriate scale factors (hhtp://cccbsb.nist.
gov/vibscalejust.asp). The natural bond orbitals (NBO) anal-
ysis [23] was performed with the NBO program integrated in
Gaussian03. The program ChemCraft [24] was used for visu-
alization of molecular orbitals.
Experimental section
Syntheses of thiosalicylamide
Scans of potential energy
Salicylamide (0.5 g, 4.4 mmol) and Lawesson’s reagent (1 g,
2.5 mmol) were stirred in tetrahydrofuran (4 mL) under Ar
atmosphere at 55 °C within for 4 h. After aluminum oxide
(10 g) was added to the mixture, the solvent was evaporated
in vacuum. Thiosalicylamide was eluted from the resulting
solid with acetonitril–ethanol (10:1) using column chromatog-
raphy. The yield was 50%, m. p. 118–119 °C (cf. Ref. [15]: m.
p. 119–120 °C). The purity of the substance was confirmed by
NMR spectra recorded in a Bruker Avance 400 NMR spec-
trometer in CDCl3 at frequencies 400.13 MHz (1H) and
100.61 MHz (13C) using SiMe4 as internal standard. The fol-
lowing chemical shifts (δ, ppm) were observed in 1H NMR:
6.88 (t, 1H, C6H5, J = 7.69 Hz), 7.05 (d, 1H, C6H5, J =
8.47 Hz), 7.41 (m, 1H, C6H5), 7.46 (broad s, 2H, NH2), 7.61
(m, 1H, C6H5), 11.7 (broad s, 1H, OH), and 13C NMR:
119.2(С3), 119.6(С1), 119.9(С5), 124.2 (С6), 134.8 (C4),
160.2 (C2), 198.4 (C=S).
In order to find possible conformers of thiosalicylamide, two-
dimensional potential energy surface (PES) scan for simulta-
neous internal rotation of the thioamide and the hydroxyl
groups was performed at the B3LYP/6-31 G(d,p) level of the-
ory (Fig. 3). The internal coordinate describing internal rota-
tion of the thioamide group was the dihedral angle φ1(C2−C1
−C7−S) and the scanned coordinate for the hydroxyl group
was the dihedral φ2 (C1−C2−O−H). Both dihedral angles φ1
and φ2 were changed in steps of 10°.
Table 1 Summary of the GED experiments
LD
SD
T, K
398
403
Ielectron beam, μA
λelectronsa, Å
s rangeb, Å−1
Presidual gas, Pa
Number of photo films
2.2
2.8
0.049708
3.0–18.4
4.0·10−3
3
0.049850
7.6–31.6
3.0·10−3
3
GED experiment
a The electron wavelength was calibrated against gaseous CCl4
The GED patterns of thiosalicylamide were recorded in an
electron diffraction apparatus EG-100 M at long (36 cm) and
short (19 cm) camera distances (LD an SD, respectively) on
4πsin2θ
b s ¼
with scattering angle θ ¼ arctanDr , where r is the distance
λelectrons
from the center of the plate and D is the nozzle-to-plate distance

Struct Chem
Fig. 4 Relaxed scans for rotation of the thioamide group around the C1
−C7 bond in thiosalicylamide using QC calculations
Fig. 2 Experimental intensity curves and background lines for
thiosalicylamide
According to the results B3LYP/6-31G(d,p), four con-
formers of thiosalicylamide with different orientations of the
thioamide and the hydroxyl groups are possible in the gas phase
(Fig. 5). Note that conformers 1, 2, and 4 have enantiomeric
isomers 1′, 2′, and 4′, respectively, since they possess C1 sym-
metry, while conformer 3 possessing Cs symmetry is single.
However, B3LYP/cc-pVTZ and MP2/cc-pVTZ computa-
tions do not confirm that conformer 4 is possible the gas
phase, and MP2/cc-pVTZ method predicts C1 symmetry for
conformer 3. The energy differences between conformers and
their distribution at 401 K (the average temperature of the
GED experiment) at various levels of theory are given in
Table 2. All computations show that conformer 1 is predom-
inant in the gas phase (more than 99%).
A low energy barrier to rotation about the C1−C7 bond
separates two optical isomers 1 and 1′ (see Fig. 4). It is esti-
mated to be ~ 1 kcal mol−1 according to QC calculations and is
comparable with the average thermal energy of the internal
rotation mode RT = 0.80 kcal mol−1. The thioamide group
torsion vibration is estimated to be less than 100 cm−1 (ca.
61 cm−1, 56 cm−1, and 57 cm−1 according to B3LYP/6-
31G(d,p), B3LYP/cc-pVTZ, and MP2/cc-pVTZ calculations,
respectively). Thus, the results of QC calculations indicate that
only the lowest energy form (conformer 1) should be consid-
ered in the GED analysis. Moreover, the flexibility of con-
former 1 makes it reasonable to interpret the GED data in
terms of a dynamic model. To consider the thioamide torsion
group vibration as a large-amplitude motion, the additional
QC calculations were performed for pseudo-conformers with
the dihedral angle φ1 (C2−C1−C7−S) ranging from 0 to 60°.
Additional one-dimensional scans for internal rotation of
the thioamide group were performed for the lowest energy
form at B3LYP/cc-pVTZ and MP2/cc-pVTZ levels of theory;
the scanned angle φ1 was varied in steps of 10° (Fig. 4). It is
noteworthy that the E(φ1) = E(−φ1) symmetry exists.
Optimizations of conformers
For all the minima detected on the two-dimensional PES, the
geometries were fully optimized at B3LYP/6-31G(d,p),
B3LYP/cc-pVTZ, and MP2/cc-pVTZ, using the opt = tight
and int = ultrafine parameters. The harmonic vibration fre-
quencies were calculated at the same levels of theory to char-
acterize the optimized stationary points and to confirm that
negative eigenvalues are absent in the Hessian matrix. The
optimized Cartesian coordinates of all thiosalicylamide con-
formers obtained at the different levels of theory are available
in the ESM (Tables S1−S10).
Pseudo-conformers optimizations
The geometries of thiosalicylamide pseudo-conformers were
optimized at MP2/cc-pVTZ level of theory using the opt =
Fig. 3 Potential energy surface for internal rotation of the thioamide (φ1)
and the hydroxyl groups (φ2) in thiosalicylamide

Struct Chem
Fig. 5 Conformers of
thiosalicylamide (optical isomers
are not shown)
tight and int = ultrafine parameters and option frozen for the
dihedral angle φ1. Subsequent frequency calculations were
performed at the same level of theory. The optimized
Cartesian coordinates of thiosalicylamide pseudo-conformers
are available in Section 1 of the ESM (Tables S11−S16). Note
that pseudo-conformer with φ1 = 0° possesses Cs symmetry
and has one imaginary vibrational frequency (ν(A")43 = −
158 cm−1) as it corresponds to a transitional state for internal
rotation.
the same level of theory applying VibModule program [28].
During the refinement procedure, some of the independent
equilibrium structural parameters (bond lengths, bond, and
dihedral angles) were combined in groups as indicated in
Table 3. The differences between the parameters in groups
were fixed at quantum-chemical values. Estimated standard
deviations for the refined parameters were multiplied by a
factor of three in order to include possible uncertainties due
to data correlation and scale uncertainty [26]. Note that geo-
metrical parameters involving hydrogen atoms were kept
equal to theoretical values due to a weak sensitivity of GED
to the positions of hydrogens [29]. The mean vibrational am-
plitudes were fixed at theoretical values.
Structure refinements
General considerations
Static model
The structure of thiosalicylamide was refined in terms of static
(rigid) and dynamic models using UNEX program [25]. The
function that was minimized had a standard form and a crite-
rion of minimum of the functional served R-factor [26]. The
initial geometrical parameters of thiosalicylamide were taken
from the results of QC computations at MP2/cc-pVTZ level of
theory and the vibrational parameters (the mean square ampli-
tudes of atom pairs l and shrinkage corrections re − ra [27])
were derived from quadratic and cubic force fields obtained at
The refinement was performed with an assumption that only
the lowest energy form of thiosalicylamide (conformer 1) is
present in the gas phase as it had been predicted by QC com-
putations. The conventional algorithm of structural analysis
for a molecule undergoing small-amplitude vibrations was
applied [26]. The contribution of the all vibration frequencies
including the lowest energy vibration were used to derive the
amplitudes of vibrations and the vibration corrections. The
Table 2 Relative total electronic
energiesa, relative standard Gibbs
free energiesb, and percentage
abundancesc for thiosalicylamide
conformers at the average
temperature of the GED
experiment
Method
Conformer
Symmetry
ΔE (kcal mol−1
)
ΔG°401 (kcal mol−1
)
P401 (%)
B3LYP/6-31G(d,p)
1
2
3
4
1
2
3
1
2
3
C1
C1
Cs
C1
C1
C1
Cs
C1
C1
C1
0.00
5.75
4.81
8.99
0.00
6.22
5.35
0.00
4.67
4.98
0.00
5.72
4.42
7.95
0.00
6.15
3.89
0.00
4.58
4.35
99.5
0.1
0.4
5 · 10−3
99.2
4 · 10−2
0.8
B3LYP/cc-pVTZ
MP2/cc-pVTZ
99.3
0.3
0.4
a The corresponding values contain ZPVE correction
b For conformers with C1 symmetry, which are present as an equimolar mixture of the two optical isomers ΔG°
includes the entropy of mixing (Rln2)
401
c The corresponding values are obtained using Boltzmann equation

Struct Chem
Table 3 Bond distances (r, Å),
bond, and dihedral angles (∠, °) of
thiosalicylamide obtained by
various methods
Parameters
GEDa
XRD[7]b
B3LYP/
MP2/
Static
model
Dynamic
model
I
II
6-
cc-
pVTZ
cc-
pVTZ
31G(d,p)
r(C1 C2)
r(C2 C3)
r(C3 C4)
r(C4 C5)
r(C5 C6)
r(C6 C1)
r(C–C)
1.409(2)1
1.395(2)1
1.379(2)1
1.391(2)1
1.379(2)1
1.403(2)1
1.463(7)2
1.347(12)3
1.670(3)4
1.081c
1.405(2)1
1.391(2)1
1.376(2)1
1.387(2)1
1.376(2)1
1.399(2)1
1.460(7)2
1.351(12)3
1.668(4)4
1.081c
1.401(5) 1.389(5)
1.417(5) 1.394(5)
1.350(5) 1.352(7)
1.379(7) 1.375(7)
1.384(5) 1.384(3)
1.388(5) 1.404(5)
1.490(5) 1.489(2)
1.311(4) 1.308(4)
1.687(4) 1.684(4)
1.427
1.408
1.383
1.403
1.383
1.417
1.473
1.351
1.692
1.086
1.009
1.337
0.993
119.4
121.2
120.3
119.1
122.4
117.5
122.7
125.1
116.6
118.3
124.7
118.4
2.5
1.421
1.402
1.377
1.397
1.378
1.411
1.470
1.346
1.685
1.081
1.005
1.335
0.991
119.4
121.2
120.2
119.1
122.5
117.4
122.8
125.0
116.7
118.2
124.5
118.5
2.4
1.415
1.400
1.385
1.396
1.385
1.408
1.473
1.347
1.664
1.081
1.006
1.343
0.985
119.1
121.3
120.0
119.3
121.8
118.4
122.3
124.7
115.2
120.1
124.4
119.3
2.5
r(C–N)
r(C=S)
r(C–H)av
r(N–H)av
r(C–O)
0.95(3)
1.00(4)
0.96(4)
0.91(4)
1.006c
1.335(9)5
0.985c
1.006c
1.332(7)5
0.985c
1.347(4) 1.370(4)
1.04(4) 0.91(3)
r(O–H)
∠C1C2C3
∠C2C3C4
∠C3C4C5
∠C4C5C6
∠C5C6C1
∠C6C1C2
∠C2C1C7
∠C1C7S8
∠C1C7N7
∠S8C7N7
∠C1C2O10
∠H16N7H17
∠C1C2C3C4
∠C2C3C4C5
∠C2C1C7S8
∠C6C1C7N7
119.0(2)6
121.2(2)6
120.4(8)d
118.7(9)d
122.1(8)d
118.3(2)6
122.3(2)6
125.6(5)7
116.3(8)8
118.1(9)d
123.6(6)9
119.3c
119.0(2)6
121.3(2)6
120.4(7)d
118.7(9)d
122.2(8)d
118.3(2)6
122.2(2)6
125.2(5)7
116.3(8)8
118.4(9)d
123.0(5)9
119.3c
2.5c
0.2c
28.3c
25.9c
177.9c
14.0c
119.3(3) 120.3(3)
120.6(4) 120.3(4)
121.1(4) 120.8(4)
118.8(4) 119.6(4)
122.2(4) 121.9(4)
117.9(3) 117.0(3)
122.7(3) 122.5(3)
123.4(3) 123.4(3)
117.1(3) 117.3(3)
119.4(3) 119.3(3)
124.8(3) 123.8(3)
122(3)
1.4(6)
110(3)
2.5c
0.2c
31(3)10
28(3)10
− 2.4(6)
− 0.6(6)
− 19.4(5) 21.2
− 18.9(5) 19.3
− 0.8(6)
19.4(3)
19.0(5)
0.2
0.1
0.2
20.3
28.3
18.7
25.9
∠C7S8N7H16 177.9c
∠C7S8N7H17 14.0c
− 177(2) 177(3)
− 2(3) − 13(3)
177.9
15.3
178.3
13.3
177.9
14.0
RLD
RSD
Rtot
3.9
5.1
4.4
3.4
5.6
4.4
a For the GED study, equilibrium parameters (re, ∠e) are presented, R-factors are in %
b For the XRD study, geometrical parameters are given for the both independent molecules (I and II) of the unit
cell
c Assumed at the values of MP2/cc-pVTZ calculations
d Dependent parameters
1-10 The same numerical superscript indicates the refinement in one group
corresponding values of l and Δ(ra − re) used in the refinement
for static model can be found in the ESM (Table S17). As a
result of the refinement, the equilibrium structural parameters
of thiosalicylamide were obtained, see Table 3; the corre-
sponding Cartesian coordinates can be found in the
Table S18 of the ESM. The disagreement factor appeared to
be 4.4%. The matrix of correlations for a static model is de-
posited as Table S19 in the ESM. The plots of the experimen-
tal and theoretical molecular intensities sM(s) along with their
difference curve as well as the plots of the experimental and
theoretical radial distributions f(r) with their difference curve
can be found in the ESM as Figs. S2 and S3, respectively.

Struct Chem
The possibility of the GED method to distinguish different
conformers of thiosalicylamide was demonstrated by consid-
ering the other forms (conformer 2 and conformer 3) in the
refinements. The values of R-factor exceeded 9.0% in the both
cases. The corresponding sM(s) functions are presented in the
ESM as Figs. S4 and S5, respectively. A model of a mixture of
conformers 1 and 2 was also taken into account. However,
after the refinement procedure independent of the starting val-
ue of percent abundance of conformer 2, the amount of con-
former 1 always reached 100%.
a Boltzmann equation can be found in Table S21 of the ESM.
The refined equilibrium parameters of thiosalicylamide (φ1 =
28.3°) from a dynamic model are presented in Table 3; the
corresponding Cartesian coordinates can be found in
Table S22 of the ESM. The matrix of correlations is given in
Table S23 of the ESM. The reduced scattering intensity curves
and the radial distribution curve are shown in Figs. 6 and 7,
respectively.
Results and discussion
Dynamic model
As it can be seen from Table 3, a dynamic model does not
significantly improve the quality of the experimental analysis
comparing with a static model: structural parameters, experi-
mental uncertainties, and disagreement factors are quite similar
in both cases (see Table 3). The usage of a dynamic model for
the description of a large-amplitude motion of the thioamide
group allowed to refine the potential function parameters.
In general, the geometry of thiosalicylamide molecule is
better predicted by means of MP2/cc-pVTZ than by means
of B3LYP/6-31G(d,p) and B3LYP/cc-pVTZ methods (see
Table 3). For example, B3LYP/6-31G(d,p) overestimates the
value of the C=S bond by approximately 0.02 Å and under-
estimates the angular twist of the thioamide group relative to
the benzene ring plane by approximately 10°; however, this
method better predicts the value of the C–O bond length and
the values of the SCN and the C1C7N valence angles than
MP2/cc-pVTZ (see Table 3).
As it has been predicted by QC calculations, enantiomers 1
and 1′ are separated by a relatively low energy barrier (see Fig.
4) and a significant fraction of the molecules in the molecular
beam would have sufficient energy to pass over the barrier at
the temperature of the GED experiment. Therefore, an algo-
rithm derived for molecules undergoing large-amplitude vi-
brations should be applied.
To consider a large-amplitude motion of the thioamide
group, the structure of thiosalicylamide was modeled
by seven pseudo-conformers, which include conformer 1
(φ1 = 28.3°), three pseudo-conformers with φ1 < 28.3°
(φ1 = 0°, 10°, 20°), and three pseudo-conformers with φ1 >
28.3° (φ1 = 40°, 50°, and 60°). Since E(φ1) = E(−φ1), statis-
tical weight of two was used for all pseudo-conformers, except
Cs pseudo-conformer (φ1 = 0°) for which a statistical weight
of one was applied. The vibrational amplitudes and
anharmonic corrections for each pseudo-conformer were cal-
culated by including the contributions from all normal modes
except the lowest vibrational mode. The obtained data for
conformer 1 calculated without the contribution of the lowest
energy vibration can be found in the ESM (Table S20).
Structural parameters of each of the pseudo-conformers were
refined simultaneously with the corresponding parameters of
equilibrium configuration (conformer 1), assuming differ-
ences between them at the values from the MP2/cc-pVTZ
calculations.
The joint GED and QC analysis established that
thiosalicylamide possesses a non-planar structure with the
thioamide group being twisted by angle ~ 30° with respect
The potential function for torsion vibration of thioamide
group was refined in the form
2
V2k
Vðφ1Þ ¼ V0 þ ∑
ð1−cosð2kφ1ÞÞ
2
k¼1
The parameters V2 and V4 were varied independently; the
V0 constant was adjusted from the condition V(φ1) ≥ 0.
Refinement of this model converged to give an R-factor of
4.4%. After the refinement procedure, the resulting PES pa-
rameters for torsion vibration of the thioamide group were
estimated to have the following values: V0 = 1.2 kcal/mol,
V2 = − 5.8(8) kcal/mol, and V4 = 3.6(4) kcal/mol. The relative
abundance of each pseudo-conformer calculated according to
Fig. 6 Reduced molecular scattering intensity curves sM(s)
(experimental—white circles, theoretical—solid lines) for the long and
short camera distances (LD and SD) of thiosalicylamide (dynamic mod-
el). The difference (experimental − theoretical) curve is shown at the
bottom

Struct Chem
The non-planarity of thiosalicylamide and salicylamide
molecules may be caused by the intramolecular repulsion of
the two nearby hydrogen atoms: one of the amine group and
the other one of the benzene ring (the corresponding H...H
distance in the both molecules is about 2.1 Å, whereas van
der Waals radius of hydrogen atom is ~ 1.2 Å [32]). The dif-
ference in the degree of non-planarity of thiosalicylamide
molecule and its oxo-analog is probably connected with dif-
ferent sizes of sulfur and oxygen atoms. The larger radius of
sulfur atom causes the angle C1C7S in thiosalicylamide to be
~ 5° bigger than the C1C7O angle in salicylamide (Table 4)
that consequently leads to a higher non-planarity
thiosalicylamide skeleton in comparison with its oxo-analog.
Comparison of structural parameters of thiosalicylamide
with those of related compounds (Table 4) shows that the
value of the C=S bond in thiosalicylamide is 0.03 Å longer
and the value of the C1−C7 bond is ~ 0.02 Å shorter than in
thiobenzamide; the value of the C−O bond in thiosalicylamide
is ~ 0.03 Å shorter than the corresponding value in phenol
(Table 4). These differences in the bond lengths may be ex-
plained by the formation of an intramolecular hydrogen bond
O−H...S, which consequently leads to a higher contribution of
the resonance form B into the overall structure of
thiosalicylamide (see Scheme 1). Thus, the hydrogen bond
formation decreases the double bond character of the C=S
bond and increases the double bond character of the C1−C7
bond comparing to thiobenzamide, and it also increases the
double bond character of the C−O bond comparing to phenol
(the corresponding values of the bond orders calculated ac-
cording Wiberg scheme [33] are presented in Table 4).
Fig. 7 Radial distribution curves f(r) (experimental—black circles,
theoretical—solid line) of thiosalicylamide (dynamic model). The differ-
ence (experimental − theoretical) curve is shown at the bottom. Vertical
bars indicate relatively important interatomic terms
to the phenol plane. However, the corresponding angular twist
of the amide group in salicylamide is ~ 15° (Table 4). These
data and the results of NBO analyses reflect the fact that the
thioamide moiety is substantially less conjugated with the
electron system of the phenol ring than the amide group in
salicylamide. Thus, according to NBO analyses, the following
weak interactions in thiosalicylamide molecule were revealed:
σ(C1 C6) → σ*(C=S) (stabilization energy E = 3.5 kcal/
mol), σ(C=S) → σ*(C1 C6) (E = 3.2 kcal/mol), σ(C1 C2)
→ σ*(C−N) (E = 2.5 kcal/mol), σ(C−N) → σ*(C1 C2) (E =
2.3 kcal/mol). On the contrary, in salicylamide molecule, the
considerable conjugation is kept even with the distortion from
planarity: σ(C1 C6) → σ*(C=O) (E = 2.1 kcal/mol),
π(C1 C2) → π*(C=O) (E = 27.5 kcal/mol), π (C=O) →
π*(C1 C2) (E = 2.6 kcal/mol), σ(C=O) → σ*(C1 C6) (E =
1.5 kcal/mol), σ(C1 C2) → σ*(C−N) (E = 2.7 kcal/mol), σ(C
−N) → σ*(C1 C2) (E = 1.5 kcal/mol).
The comparison of the numerical values of the S...H dis-
tance in S...H−O fragment (2.07(3) Å) and the sum of the van
der Waals radii of sulfur atom (1.85 Å [32]) and hydrogen
atom (1.2 Å [32]) supports the idea of intramolecular hydro-
gen bond formation. The S...H distance in thiosalicylamide is
by about 0.5 Å longer than the O...H distance in the O...H−O
fragment of salicylamide (see Table 4), that fact is in
Table 4 Selected structural
Parametersa
Thiosalicylamideb
Thiobenzamidec
Salicylamided
Phenole
parameters of thiosalicylamide,
salicylamide, thiobenzamide, and
phenol molecules (atom-atom
overlap-weighted NAO bond or-
ders are given after the slash
symbol)
r(C=X)
r(C–N)
r(C–C)
r(C–O)
X...H
1.670(3)/1.18
1.347(12)/1.05
1.463(7)/0.95
1.335(9)/0.94
2.07(3)/0.12
125.6(5)
1.641(4)/1.24
1.352(2)/1.04
1.478(9)/0.90
1.232(6)/1.22
1.357(5)/1.04
1.496(4)/0.93
1.337(5)/0.94
1.61(8)/0.10
120.1(9)
1.381(3)/0.88
∠C1C7X
∠C2C1C7X
123.4(5)
31(4)
31(3)
15(5)
a Bond lengths are in Å and bond angles are in degrees
b Present work, re structure
c Ref. [30], re structure
d Ref. [11], rh1 structure
e Ref. [31], rg structure

Struct Chem
Scheme 1 Resonance forms of
thiosalicylamide.
agreement with the smaller size of oxygen atom (the corre-
sponding van der Waals radius is of oxygen atom is 1.4 Å
[32]).
The angle C2C1C7S turns out to be by about 10° less in the
crystal than in the gas phase (Table 3) probably due to the
effect of crystal packing as the nearby molecules compensate
an interatomic repulsion in a single molecule. The similar
difference in the angular twist of the amide group relatively
the phenol ring plane is observed for the gaseous and the solid
salicylamide (15 5° in the gas phase [11] versus 2.1 0.1° in
the crystal phase [7]).
NBO analysis confirms the presence of the intramolecular
hydrogen bridge as the following stabilization interactions
were found: LP1(S) → σ*(О−Н) (E = 2.3 kcal/mol), LP2(S)
→ σ*(О−Н) (E = 25.2 kcal/mol), σ(C=S) → σ*(О−Н) (E =
0.7 kcal/mol), π(C=S) → σ*(О−Н) (E = 1.1 kcal/mol). The
NBO LP2(S) → σ*(О−Н) orbital overlap diagram
representing the O−H...S hydrogen bond is shown in Fig. 8.
The geometry of the six-membered pseudo-ring containing
intramolecular hydrogen bond of gaseous thiosalicylamide
was found to be quite different from that in the solid phase
(see Table 3). In the average, in the gas phase, the C=S bond is
~ 0.02 Å shorter, the C−N bond is ~ 0.04 Å longer and the C
−O bond is ~ 0.03 Å longer than corresponding parameters in
the crystal. That fact is probably caused by the intermolecular
hydrogen bonding in the crystal (see Fig. 9) and as a conse-
quence with higher contribution of resonance form C into the
structure of crystalline thiosalicylamide (see Scheme 1).
The intermolecular hydrogen bond formation seems to de-
crease the double bond character of the C=S bond, causing its
elongation, and to increase the double bond character of the C
−N bond, causing its shrinkage comparing to the gas phase.
The weakening of the intramolecular hydrogen bond caused
by the formations intermolecular hydrogen bonds also leads to
the elongation of the C−O bond in the solid phase.
Concluding remarks
The molecular structure of thiosalicylamide and the internal
rotation of the thioamide group were investigated in the gas
phase by the joint application of the GED method and QC
calculations. According to the obtained results,
thiosalicylamide molecule, being free from the intermolecular
interactions in the gas phase, exists as a mixture of two non-
planar enantiomers with a barrier of their mutual interconver-
sion ~ 1 kcal/mol. The application of GED augmented by QC
computations afforded to obtain equilibrium structural param-
eters of thiosalicylamide as well as parameters of the potential
function for the internal rotation of the thioamide group.
According to GED, the equilibrium structure of
thiosalicylamide in the gas phase is not planar as an interatom-
ic repulsion of nearby hydrogens tries to tilt the thioamide
group with respect to the phenol ring. The structural features
Fig. 8 NBO overlap interaction surface plot (LP2(S) → σ*(О−Н))
reflecting the formation of an intramolecular hydrogen bond in
thiosalicylamide
Fig. 9 Hydrogen-bonding motifs observed for thiosalicylamide in the
crystal phase

Struct Chem
15. Briel D (2005). Heterocycles 65(6):1295–1309
16. Kochikov IV, Kovtun DM, Tarasov YI (2008). Num Meth Program
9:12–18 http://num-meth.srcc.msu.ru/zhurnal/tom_2008/pdf/
v9r202.pdf. Accessed 13.09.2010
of thiosalicylamide molecule as well as NBO analysis indicate
that the conjugation between thioamide group and aromatic
ring is almost absent. The existence of intramolecular hydro-
gen bond was revealed by NBO calculations and the compar-
ison of structural parameters of thiosalicylamide with those of
the related molecules. The geometry of gaseous
thiosalicylamide was found to be different from that in the
solid phase due to the absence of intermolecular hydrogen
bonding and the effects of crystal packing in the gas phase.
17. Frish MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
Cheeseman JR, Zakrzewski VG, Montgomery JA, Stratmann JR,
Burant JC, DapprichS, Millam JM, Daniels AD, Kudin KN, Strain
MC, Farkas O, Tomasi J, Barone V, Cossi M, Cammi R, Mennucci
B, Pomelli C, Adamo C, Clifford S, Ochterski J, Petersson GA,
Ayala PY, Cui Q, Morokuma K, Malick DK, Rabuck AD,
Raghavachari K, Foresman JB, Cioslowski J, Ortiz JV, Baboul
AG, Stefanov BB, Liu G, Liashenko A, Piskorz P, Komazomi I,
Gomperts R, Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng
CY, Nanayakkara A, Challacombe M, Gill PM, Johnson B, Chen
W, Wong MW, Andres JL, Gonzales C, Head-Gordon M, Replogle
ES, Pople JA (2003) Gaussian 03 (Revision D01). Gaussian Inc.,
Pittsburgh
18. Becke AD (1988). Phys Rev A 38(6):3098–3000. https://doi.org/
10.1103/PhysRevA.38.3098
19. Lee C, Yang W, Parr RG (1988). Phys Rev B 37:785–789. https://
doi.org/10.1103/PhysRevB.37.785
20. Møller C, Plesset MS (1934). Phys Rev 46:618–622. https://doi.
org/10.1103/PhysRev.46.618
21. Petersson GA, Bennett A, Tensfeldt TG, Al-Laham MA, Shirley
WA, Mantzaris J (1988). J Chem Phys 89:2193–2218. https://doi.
org/10.1063/1.455064
Acknowledgments The authors express their gratitude to Dr. Ilya I.
Marochkin from Lomonosov Moscow State University, Arseniy A.
Otlyotov, Dr. Yury A. Zabanov, and Prof. Nina I. Giricheva from
Ivanovo State University of Chemistry and Technology for valuable con-
sultations which were very useful for preparing this manuscript.
Funding information This project was made with financial support of the
Russian Foundation for Basic Research (Grant Number 18-33-00546
mol_a).
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflicts of
interest.
22. Dunning TH (1989). J Chem Phys 90:1007–1023. https://doi.org/
10.1063/1.456153
23. Weinhold F, Landis CR (2001). Chem Educ Res Pract 2:91–104.
https://doi.org/10.1039/B1RP90011K
24. G. A. Zhurko, D. A. Zhurko, ChemCraft 1.6 build vol. 332 https://
www.chemcraftprog.com/index.html. Accessed 20.06.2010
25. Vishnevskiy YV, UNEX, 2007, version 1.5. http://unexprog.org.
Accessed 2.09.2013
26. Ischenko AA, Girichev GV, Tarasov YI (2013) Electron diffraction:
structure and dynamics of free molecules and condensed state of
substance. Fizmatlit, Moscow
27. Sipachev VA (2000). Struct Chem 11:167–172. https://doi.org/10.
1023/A:1009217826943
References
1. Pospisilova S, Michnova H, Kauerova T, Pauk K, Kollar P, Vinsova
J, Imramovsky A, Cizek A, Jampilek J (2018). Bioorg Med Chem
Lett 28(12):2184–2188. https://doi.org/10.1016/j.bmcl.2018.05.
011
2. Ueda J, Khan ST, Takagi M, Shin-ya K (2010). J Antibiot 63(5):
267–269. https://doi.org/10.1038/ja.2010.26
3. Hardie DG (2013). Diabetes 62(7):2164–2172. https://doi.org/10.
2337/db13-0368
28. Vishnevskiy YV, Zhabanov YA (2015). J Phys Conf Ser 633:
012076. https://doi.org/10.1088/1742-6596/633/1/012076
29. Tikhonov DS, Vishnevskiy YV, Rykov AN, Grikina OE, Khaikin
LS (2017). J Mol Struct 1132:20–27. https://doi.org/10.1016/j.
molstruc.2016.05.090
4. Mehanna AS, Kim JY (2005). Bioorg Med Chem 13(13):4323–
4331. https://doi.org/10.1016/j.bmc.2005.04.012
5. Zhang Y, Mantravadi PK, Jobbagy S, Bao W, Koh JT (2016). ACS
Chem Biol 11(10):2797–2802. https://doi.org/10.1021/
acschembio.6b00659
6. Palomar J, De Paz JLG, Catalán J (1999). Chem Phys 246(1–3):
167–208. https://doi.org/10.1016/s0301-0104(99)00159-7
7. Pertlik F (1990). Monatsh Chem 121:129–139. https://doi.org/10.
1007/BF00809525
8. Velcheva EA, Stamboliyska BA (2008). J Mol Struct 875(1–3):
264–271. https://doi.org/10.1016/j.molstruc.2007.04.038
9. Anandan K, Kolandaivel P, Kumaresan R (2005). Int J Quantum
Chem 104(3):286–298. https://doi.org/10.1002/qua.20559
10. Manin AN, Voronin AP, Perlovich GL (2013). Thermochim Acta
551:57–61. https://doi.org/10.1016/j.tca.2012.10.013
11. Aarset K, Page EM, Rice DA (2013). J Phys Chem A 117(14):
3034–3040. https://doi.org/10.1021/jp311003d
30. Kolesnikova IN, Putkov AE, Rykov AN, Shishkov IF (2018). J Mol
Struct 1161:76–82. https://doi.org/10.1016/j.molstruc.2018.01.084
31. Portalone G, Schultz G, Domenicano A, Hargittai I (1992). Chem
Phys Lett 197:482–488. https://doi.org/10.1016/0009-2614(92)
85804-J
32. Pauling L (1960) The nature of the chemical bond3rd edn. Cornell
University Press, Ithaca
33. Wiberg KB (1968). Tetrahedron 24:1083–1196. https://doi.org/10.
1016/0040-4020(68)88057-3
12. Banerjee K, Raychaudhury S (1982). Bull Chem Soc Jpn 55(11):
3621–3624. https://doi.org/10.1246/bcsj.55.3621
13. Sambathkumar K (2015). Spectrochim Acta A 147:51–66. https://
doi.org/10.1016/j.saa.2015.03.052
Publisher’s note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
14. Jezierska A, Panek JJ, Mazzarello R (2009). Theor Chem Accounts
124(5–6):319–330. https://doi.org/10.1007/s00214-009-0612-2