Interplay between polymorphism and isostructurality in the 2-fur- And 2-thenaldehyde semi-and thiosemicarbazones

Article abstract of DOI:10.3390/molecules25040993

The four compounds, namely: 5-nitro-2-furaldehyde thiosemicarbazone (1); 5-nitro-2-thiophene thiosemicarbazone (2); 5-nitro-2-furaldehyde semicarbazone (3); and 5-nitro-2-thiophene semicarbazone (4) were synthesized and crystallized. The three new crystal structures of 1, 2, and 4 were determined and compared to three already known crystal structures of 3. Additionally, two new polymorphic forms of 1 solvate were synthesized and studied. The influence of the exchange of 2-thiophene to 2-furaldehyde as well as thiosemicarbazone and semicarbazone on the self-assembly of supramolecular nets was elucidated and discussed in terms of the formed synthons and assemblies accompanied by Full Interaction Maps analysis. Changes in the strength of IR oscillators caused by the molecular and crystal packing effects are described and explained in terms of changes of electron density.

Full text of DOI:10.3390/molecules25040993

molecules
Article
Interplay between Polymorphism and Isostructurality
in the 2-Fur- and 2-Thenaldehyde Semi-
and Thiosemicarbazones
Marcin Swiatkowski, Agata Trzesowska-Kruszynska, Agnieszka Danielewicz, Paulina Sobczak
and Rafal Kruszynski *
Institute of General and Ecological Chemistry, Lodz University of Technology, Zeromskiego 116, 90-924 Lodz,
Poland; marcin.swiatkowski@p.lodz.pl (M.S.); agata.trzesowska@p.lodz.pl (A.T.-K.);
aga.danielewicz@wp.pl (A.D.); paulinasobczak97zdw@onet.pl (P.S.)
* Correspondence: rafal.kruszynski@p.lodz.pl; Tel.: +48-42-631-31-37
ꢀꢁꢂꢀꢃꢄꢅꢆꢇ
Academic Editor: Ugo Caruso
Received: 15 January 2020; Accepted: 17 February 2020; Published: 23 February 2020
ꢀꢁꢂꢃꢄꢅꢆ
Abstract:
5-nitro-2-thiophene thiosemicarbazone (
5-nitro-2-thiophene semicarbazone ( ) were synthesized and crystallized. The three new crystal
structures of , and were determined and compared to three already known crystal structures of
Additionally, two new polymorphic forms of solvate were synthesized and studied. The influence
The four compounds, namely: 5-nitro-2-furaldehyde thiosemicarbazone (
1);
2
); 5-nitro-2-furaldehyde semicarbazone ( ); and
3
4
1,
2
4
3.
1
of the exchange of 2-thiophene to 2-furaldehyde as well as thiosemicarbazone and semicarbazone
on the self-assembly of supramolecular nets was elucidated and discussed in terms of the formed
synthons and assemblies accompanied by Full Interaction Maps analysis. Changes in the strength
of IR oscillators caused by the molecular and crystal packing effects are described and explained in
terms of changes of electron density.
Keywords: polymorphism; isostructurality; thiosemicarbazone; semicarbazone
1. Introduction
Knowledge about non-covalent interactions increases with each passing year [1–5], but is still
insufficient to enable the complete design and prediction of the crystal structure of organic compounds,
even with the usage of computational crystal structure prediction. Application of computational
methods allows for the prognostication of crystal packing on the basis of the structural formula, but
these methods still lead to ambiguous results or give an incorrect build of the crystal net in some
cases [
crystallographic data. The hydrogen bond rules [
some functional group [12 16], and molecular tectonics [17
6
,
7
]. Through the years, crystal engineering principles have been developed on the basis of
], exchange rule [10 11], synthon hierarchy for
19] are the most important principles
8,
9
,
–
–
among all of the studied ones. They are very helpful in predicting the results of crystal formation, but
are still insufficient. Better understanding of the nature of intermolecular interactions and the ability to
foresee the molecular structure are the key aspects for the engineering and control of crystal packing
and, hence, the physical and chemical properties.
The molecular geometry (including conformation and specific positions of functional group) as
well as the intermolecular interactions existing between molecules have an influence on the crystal
packing (the mutual arrangement of molecules in the crystal net). Molecular shape prediction is
especially problematic for flexible molecules. It has been shown that similarly shaped molecules,
not necessarily showing the chemical similarity, adopt the same packing arrangement [20]. Different
conformers of the same molecules can have a different spatial arrangement of molecules [21]. Molecules
Molecules 2020, 25, 993; doi:10.3390/molecules25040993
www.mdpi.com/journal/molecules

Molecules 2020, 25, 993
2 of 14
do not always adopt the lowest energy conformer, which is balanced by the possibility of the formation
of stabilizing inter- and intramolecular interactions. Some small changes in the molecular structure
might result in significant changes in crystal packing. Nevertheless, specific functional groups can
be interchanged without affecting the crystal structure. Such groups should not be involved in any
hydrogen bond or electrostatic interactions [11]. Thus, the interplay between these two factors affects
the arrangement of the molecules in the solid state.
The presented study demonstrates the complexity of the problem of the self-assembly of
different isomorphous molecules (i.e., 5-nitro-2-furaldehyde thiosemicarbazone (
thiosemicarbazone ( ), 5-nitro-2-furaldehyde semicarbazone ( ), 5-nitro-2-thiophene semicarbazone (
Scheme S1). Only one compound of this group was previously structurally determined (i.e., the 17 19])
and it created three monoclinic polymorphic forms 3α 22], 3β, and 3γ 23 24]. The introduction of
additional species (different acids) to distinctly affects the present synthons and hydrogen bonding
1), 5-nitro-2-thiophene
2
3
4
),
3
[
–
[
[ ,
3
schemes. The presence of additional hydrogen bond donors and acceptors and, more importantly,
the presence of additional non-isomorphous species in the crystal net causes the impossibility of the
formation of similar crystals [24]. The studied compounds are effectively isomolecular (i.e., have the
same configuration (E) about the imine double bond, possess heteroatoms at the same sites of molecule
and heteroatoms (O and S) [11], which possess the same electron configuration of the valence shell).
As two heteroatoms are present in each molecule, a total of four combinations could be attained and
consequently four compounds were studied. Such an approach allows a comparison of compounds
with gently tuned molecular properties as well as the study of the slight expansion of the atomic core
size on the self-assembly processes, in which very similar hydrogen-bonding groups compete during
synthon formation. Additionally, in all four cases, the molecular space filling is practically the same,
thus it has little influence on the crystal packing, and permits study of the impact of the interactions
themselves (this is not possible in the case of studying compounds that possess other structurally
dissimilar species in the crystal net [24]).
2. Results and Discussion
2.1. Synthesis
The pure compounds
Recrystallization from different solvents always led to the starting form in the case of compounds
and . The use of dmf (pure or in mixture with other solvents) in the crystallization of led to the
formation of two polymorphic forms of dmf solvate of . The 5-nitro-2-furaldehyde thiosemicarbazone
1, 2, and 4 were synthesized in only one polymorphic form (Table 1).
2
3
1
1
dimethylformamide solvate 1·dmfα was created during the slow evaporation of the dmf solution of
1,
while 1·dmfβ was crystallized from the MeOH:dmf (1:1 v/v) solvent mixture (Table 1). Exchanging
MeOH for EtOH or i-PrOH in the mixture does not affect the final form, which suggests that the
presence of the short chain alcohol is a necessary and sufficient condition for the formation of 1·dmfβ
.
The crystals of compounds and are isostructural and these compounds did not create solvates or
2
4
other pseudopolymorphs in the solid state, nevertheless of the solvents used during crystallization. It
must be outlined that the application of very specific conditions such as high pressure can hypothetically
lead to the formation of other polymorphic forms, but any evidence of the formation of other than the
described forms was not registered during the study (including heating up of the compounds to their
decomposition or cooling them to 100 K).

Molecules 2020, 25, 993
3 of 14
Table 1. Crystal data and structure refinement details for the studied compounds.
Compound
1
1·dmfα
1·dmfβ
2
4
CCDC number
Empirical formula
Formula weight
Crystal system
1976784
C6H6N4O3S
214.21
Monoclinic
P21/n
(No. 14)
100.0(1)
0.71073
1976786
C9H13N5O4S
287.30
Monoclinic
P21/n
(No. 14)
100.0(1)
1.54184
1976789
C9H13N5O4S
287.30
Monoclinic
Cc
(No. 9)
100.0(1)
0.71073
λ(MoKα)
1976785
C6H6N4O2S2
230.27
Triclinic
P-1
(No. 2)
100.0(1)
0.71073
λ(MoKα)
1976790
C6H6N4O3S
214.21
Triclinic
P-1
Space group
Temperature (K)
Wavelength (Å)
(No. 2)
100.0(1)
0.71073
λ(MoKα)
λ(MoKα)
λ(CuKα)
Unit cell dimensions
12.262(3)
12.4011(14)
12.670(3)
90.00
113.84(3)
90.00
4.7031(1)
24.9324(5)
11.1468(2)
90.00
98.552(1)
90.00
5.9513(4)
19.6479(15)
11.4607(7)
90.00
104.431(5)
90.00
4.5364(1)
8.7380(2)
12.2261(3)
76.291(2)
87.644(2)
79.401(2)
462.791(19)
2
4.5040(2)
8.5940(3)
11.7247(4)
71.920(3)
86.928(3)
77.680(3)
421.45(3)
2
a (Å)
b (Å)
c (Å)
α (◦)
β (◦)
γ (◦)
Volume (Å3)
1762.3(6)
8
1292.54(3)
4
1297.82(16)
4
Z
Calculated density (Mg/m3)
Absorption coefficient (mm−1
F(000)
1.615
0.354
880
1.476
2.435
600
1.470
0.269
600
1.652
0.553
236
1.688
0.370
220
)
0.098
0.090
0.173
0.131
0.121
0.084
0.107
0.098
0.098
0.093
Crystal size (mm)
0.007
3.416 to
25.049
0.118
4.386 to
78.482
0.024
3.671 to
32.078
0.091
3.430 to
31.793
0.025
3.563 to
31.940
θ Range for data collection (◦)
−14 ≤ h ≤ 14,
−14 ≤ k ≤ 14,
−15 ≤ l ≤ 14
14539/3111
0.0647
−5 ≤ h ≤ 5,
−31 ≤ k ≤ 31,
−14 ≤ l ≤ 13
25641/2646
0.0350
−8 ≤ h ≤ 8,
−28 ≤ k ≤ 28,
−16 ≤ l ≤ 14
12573/3526
0.0464
−6 ≤ h ≤6,
−12 ≤ k ≤ 12,
−17 ≤ l ≤ 17
12959/2743
0.0278
−6 ≤ h ≤ 6,
−12 ≤ k ≤ 12,
−17 ≤ l ≤ 17
12046/2544
0.0338
Index ranges
Reflections collected/unique
Rint
Completeness (%) to θ =
99.8
100.0
99.8
99.8
99.8
25◦(Mo) and θ = 67◦(Cu)
0.48113 and
1.00000
3111/1/271
0.63893 and
1.00000
2646/0/183
0.86402 and
1.00000
3526/2/183
0.94704 and
1.00000
2743/0/136
0.96223 and
1.00000
2544/0/136
Min. and max. transmission
Data/restraints/parameters
Goodness-of-fit on F2
1.044
1.043
1.084
1.078
1.050
R1 = 0.0632,
wR2 = 0.1595
R1 = 0.0834,
wR2 = 0.1705
0.995 and
−0.352
R1 = 0.0271,
wR2 = 0.0698
R1 = 0.0289,
wR2 = 0.0708
0.414 and
−0.205
R1 = 0.0357,
wR2 = 0.0951
R1 = 0.0366,
wR2 = 0.0957
0.582 and
−0.546
R1 = 0.0249,
wR2 = 0.0658
R1 = 0.0290,
wR2 = 0.0674
0.425 and
−0.233
R1 = 0.0282,
wR2 = 0.0725
R1 = 0.0337,
wR2 = 0.0745
0.428 and
−0.229
Final R indices [I > 2σ(I)]
R indices (all data)
Largest diff. peak
and hole (e•Å−3
)
The polymorphic form 3β was created as a result of the condensation reaction between
5-nitro-2-furaldehyde and semicarbazide hydrochloride carried out in water (in the presence of
sulfuric acid) or in methanol (Table S1). The IR spectrum of this form matched the spectrum of the
nitrofural pharmaceutical standard. The recrystallization of form 3β from different solvents and solvent
mixtures (dmf, i-PrOH, water, dmf:acetone (1:1 v/v), MeOH:water (1:1 v/v)) led to the starting form 3β
The usage of the MeOH:MeCN (1:1 v/v) solvent mixture or sole MeCN gave the polymorphic form
3γ (Table S1). According to data presented in [23], the crystallization from MeOH:MeCN (3:1 v/v) or
from A:B (1:3 v/v, where A is H2O, MeOH, EtOH, or butan-2-ol and B is MeCN or acetone) solvent
mixture leads to the formation of 3β and from the water:acetone (7:15 v/v) mixture or pure methanol,
ethanol, propan1-ol, propan-2-ol, butan-2-ol, and acetonitrile leads to the formation of 3γ. These data
are partially in opposition to the current findings and seem to be slightly internally inconsistent as
an excess or shortage of the aprotic solvent in the mixture led to the formation of 3β or 3γ without
any clear tendency, and the usage of pure solvent of any kind gave 3γ. The results of the current
.

Molecules 2020, 25, 993
4 of 14
study suggest that the addition of an equivolume amount or volumetric excess (up to pure solvent) of
MeCN gave the polymorphic form 3γ, while the other used pure solvents or solvent mixtures gave
the polymorphic form 3β. The known polymorphic form 3α [22] could not be obtained regardless
of the crystallization conditions or synthesis method (application of the industrial method, which
was the source of form 3α described in [22] also led to form 3β). It can be suggested that form 3β is
predominant and form 3α can be a “vanishing polymorph” (in subsequent works [23
of form 3α was also not registered).
,24], the presence
2.2. Molecular Conformation
The bond distances in all studied compounds (Figure 1) were similar (Table S2). To ensure that
the temperature of the measurement did not affect the polymorphic form/stability, the structures of 3β
and 3γ were redetermined at 100 K (temperature used in all other measurements of
1–4 structures; in
previous studies the 173 K [24] and room temperature [22 23] were used). The compounds 3β and
,
3γ possessed almost identical structures at all temperatures and slight differences originated from
normal changes of atom motion imposed by temperature. The molecules of thio- and semicarbazones
adopted a nearly planar conformation with the E configuration about the imine C=N bond, but differed
in arrangement of the heterocyclic ring with respect to the thio- and semicarbazone moiety and the
orientation of the amide group. A superimposition of the studied compounds showed two groups of
essentially different conformations, especially visible in the mutual arrangement of the five-membered
ring and amide group (Figure 2). The first most populated arrangement is related to the possibly
longest molecule (i.e., the aliphatic chain expanding along line perpendicular to the two-fold axis
of five-membered ring). This conformation exits in
populated conformation is adopted by isostructural compounds
last case, the deviation from the above-mentioned longest molecule line was greater than for
. The values of O/S(ring)–C–C–N(imine) and N(imine)–N–C–N(amine) torsion angles (Table S2)
confirm that nitrofuran containing compounds ( 3β and 3γ) possess different molecular conformation
than nitrothiophene containing compounds ( and ). The 3α, which additionally shows a different
1
,
1·dmfα
,
1·dmfβ
,
3β, and 3γ. The second, less
as well as by 3α, but in the
and
2
and
4
2
4
1,
2
4
orientation of the amide group, can be considered as belonging to a separate subgroup of the second
general conformation.
The energy barrier of rotation of the thio- and semicarbazone moiety about the C(imine)–C(ring)
bond was much lower than the barrier of rotation of the thio- and amide group about the
C(carbonyl)–N(H) bond (Figure S1). In both cases, the most unfavorable conformation was the
one with an almost perpendicular orientation of the rotated group toward the plane of the ring.
The larger energy barrier of rotation of the thio- and amide group about the C–N bond is related to
the disruption of favorable intramolecular interactions between the amine H-bond donor and imine
N acceptor atoms. This barrier height is larger than the molecule energy difference between the two
conformers of nitrofural, 3α and 3β/3γ. The values of the total molecule energy differences of both
conformers suggest that a more stable molecular conformation is this one adapted by nitrofural in the
forms 3β and 3γ (the difference between the energy of these forms and 3α was 9.6 kcal/mol; value from
the calculation done for optimized molecular geometry with the use of the M06-2X density functional).
In the known polymorph 3α, where the nitrofural exists in another conformation, the hydrogen bond
donors and acceptors point in two different directions, and this is the reason for the distinct hydrogen
bond pattern. The intramolecular repulsions between the electronegative atoms with a lone pair of
electrons can be responsible for the smaller stability of this conformer.

Molecules 2020, 25, 993
5 of 14
Figure 1. Solid state structures of the studied compounds with the atom numbering scheme, plotted
with a 50% probability of displacement ellipsoids of non-hydrogen atoms. Hydrogen atoms are plotted
as spheres of arbitrary radii.
Figure 2. The molecular overlay of the studied compounds. The five-membered rings were arranged
to provide the shortest distances between their constituent atoms.
According to the conformational energy profile, furfural thio- and semicarbazones can adopt
two energetically preferred equivalent conformations, in which the imine N atom and furan O atom
can be syn- or anti-periplanar. These two conformations were observed in the studied compounds,

Molecules 2020, 25, 993
6 of 14
but the syn-periplanar arrangement of heteroatoms was observed only in 3α. The analysis of
structures in the Cambridge Structural Database (CSD) [25] shows that the syn-periplanar isomer
is the predominant isomer for furfural thio- and semicarbazones. For thio- and semicarbazones of
2-thiophenecarboxaldehyde (thenaldehyde), the anti-periplanar isomer was not observed, neither for
the studied compounds nor the CSD structures. According to the conformational energy profile, such
conformation was less preferred in comparison to that of the syn-periplanar arrangement of N(imine)
and S(thiophene) atoms. The conformational preferences of the studied compounds are similar to
those exhibited by aldehyde itself. As in the studied compounds, the pure furfural exists as cis and
trans conformers, whereas the cis form of pure thenaldehyde is favored. The character of electrostatic
interactions between imine nitrogen and ring heteroatom is the major factor determining the more
stable isomer. The stereoelectronic effects cause one conformation of thenaldehyde derived compounds
(2 and 4) to be predominant as a result of the presence of a larger atom of sulfur (in comparison to the
furfural oxygen atom) and subsequently a more favorable and rigid arrangement of a nitrogen one
lone electron pair between two lone electron pairs of sulfur. The blockage of the nitrogen electron pair
is smaller in the case of compounds possessing a furan ring as the distance between respective N and
O atoms is smaller (d[N(imine)•••O(ring)] = 2.70 Å in 3α) than the distance between respective N and
S atoms (d[N(imine)•••S(ring) = 3.07 and 3.02 Å in
2 and 4, respectively). This is also one of important
reasons that make both studied derivatives of thenaldehyde (2 and 4) isostructural.
2.3. Crystal Packing
The Full Interaction Maps (FIMs) software was used to compare the existing non-covalent
interactions formed by the studied compounds with the preferred intermolecular interactions occurring
in similar compounds. The maps were generated using an uncharged NH donor, carbonyl, and oxygen
acceptor probes. For compounds 1 and 2 possessing the thiosemicarbazone group, which showed a
high propensity to form a thioamide dimer synthon, the analysis was expanded in the study of the
thiocarbonyl contact group performed with the IsoStar Intermolecular Contact Database (IsoStar ICD),
as FIMs software are unable to reproduce the respective maps.
The thiosemicarbazones show a preference for the formation of homomeric dimers through both
the primary (in 41%) and secondary (in 40%; observed also in
thiocarbonyl S atom, whereas semicarbazones formed the amide(–C(=O)–NH–)···amide(–C(=O)–NH–)
homosynthon more frequently (this synthon was observed in 49% of compounds; observed also in
1 and 2, Table S3) amine group with the
4,
Table S3) [25]. The heteromeric dimer formed by thioamide(–C(=S)–NH2)···thioamide(–C(=S)–NH-)
and amide(–C(=O)–NH2)···amide(–C(=O)–NH-) groups was rarely observed (only for 10% of crystal
structures; observed also in 3β).
The analysis of the calculated interaction maps and contoured interaction–density plots (Figure S2)
showed that the studied compounds formed interactions consistent with the interaction preferences.
The lack of high-probability areas for the hydrogen bond acceptors and the clearly located areas for
hydrogen bond donors near the thioamide group on FIMs (for
1 and 2, Table S3) is the result of (a)
the limitation in the choice of functional groups to be used as probes and (b) the high propensity
of the thioamide group to form a ring dimer synthon via N–H···S hydrogen bonds. The geometry
of the thioamide group intermolecular interactions for
1 and 2 (Table S3) is also typical, as shown
in the contoured density surface plots obtained from the IsoStar ICD of intermolecular interactions.
The distributions of thiocarbonyl acceptors and NH donors around the thiosemicarbazone group
suggest that both the primary and secondary amine groups are involved in hydrogen bonding with the
C=S group and the hydrogen atoms of the primary amine group are non-equivalent, with one of them
forming hydrogen bonds more frequently. For
1
and
2, the preferred thiocarbonyl group position near
the primary amine is replaced by a nitro group. The molecules of
1
and form only homomeric dimers
2
through secondary amine and thiocarbonyl groups. For both polymorphic forms of 1·dmf solvate, the
dimer motif was not observed. The solvent carbonyl group position is typical for preferred oxygen
atom location, but the thiocarbonyl acceptor forms interactions (of less favored geometry) with the

Molecules 2020, 25, 993
7 of 14
primary amine group. In all cases, the nitro group forms intermolecular interactions of less common
geometries. Typically, the nitro group forms interactions via both oxygen atoms with one donor placed
in the area of O–N–O angle bisector.
The differences between nitrofural polymorphs (3) concern the NH2 donor. The 3γ does not
possess an acceptor near one of the hydrogen atoms of the amine group. The strong interaction map
peaks corresponding to an acceptor near each of the hydrogen atoms of the amine group indicate the
high propensity of this group to form interactions with noticeably preferred geometry. The presence of
this unsatisfied hydrogen-bond donor can explain the reason for the lower possibility of this form being
created. For 3α, the small donor hot spot near the amine group is occupied by a furan oxygen atom,
which suggests that such a position of the acceptor near the amine group is statistically less probable.
To verify the likelihood of the observed hydrogen bond formation for molecules of
solvate, the hydrogen bond propensities were calculated (Table S4). The NH2···O/S=C hydrogen
bonds with the highest propensity values were not observed in the studied compounds ( ).
1–4 and 1
1, 2, 4
The NH2···O=C hydrogen bonds only existed in the crystal structures of two polymorphic forms of
nitrofural (3α and 3β). The hydrogen bonds with high propensity existing in all four compounds
involved the hydrazinic NH group and carbonyl/thiocarbonyl group. The observed interactions
between the amide NH2 group and the nitro O atom also had high propensity, but the formation of
analogous interactions between the hydrazinic NH group and the nitro O atom is highly improbable
(taking into account the low value of propensity), and accordingly, they do not exist in compounds
1
–
4
. The observed NH2(amide)···N(imine) and NH2(amide)···O(furan) interactions in 3α are very rare,
and they are only low propensity interactions existing in the crystal structures of . This might be a
1–4
reason for the problems with the formation of 3α during the synthesis and crystallization.
Both polymorphic forms of 1·dmf solvate showed the most probable combination of hydrogen
bonds. In contrast to 1, the interactions between the amide NH2 group and the thiocarbonyl group
(showing high propensity) are present in the 1·dmf solvate. Moreover, the hydrazinic NH group forms
more possible interactions with the solvent carbonyl group, which is an alternative to the thiocarbonyl
acceptor of 1.
The analysis of tessellation confirmed the above findings. The hexagonal tiles (green, Figure 3)
were created in compounds
in , and exist two additional motif tiles (yellow and orange, Figure 3). These tiles have identical
shapes but different arrangements (i.e., in they are oriented alternatively in each subsequent column,
while in and they have the same alignment). The 3γ also had only one motif tile in the predominant
1, 2, 3β, and 4. It is a sole motif tile in a predominant plane in 3β, whereas
1
,
2
4
1
2
4
plane (similarly to 3β), but had a more irregular shape than the above-mentioned hexagon. 3α possesses
uncommon pentagonal and heptagonal tiling.

Molecules 2020, 25, 993
8 of 14
Figure 3. Tessellations in the studied compounds, created via hydrogen bonds assembling molecules
in an appropriate crystallographic plane: (1 0 1) for and 3α, (2 1 1) for and , (3 0 1) for 3β, (1 0 2) for
1
2
4
3γ (for detailed description see text).
2.4. Polymorph Stability
Structures of the nitrofural (
polymorphs. In three polymorphic forms of
3
) polymorphs possessed different packing than the 1·dmf solvate
3
, hydrogen bonds play a substantial role in controlling
the arrangement of the molecules in solid state (Table S3). The conventional N–H···O hydrogen bonds
existed in all polymorphs, but formed different hydrogen bonding patterns. In general, the number
of hydrogen bonds correlates with the crystal density and packing efficiency and these values are
smaller for smaller numbers of hydrogen bonds. The values of the lattice energy (Table 2) of crystals
3β and 3γ were similar (with difference of 0.3 kcal/mol) and different from 3α, for which the strongest
intermolecular interactions were observed, and the lattice energy was significantly larger in magnitude.
Typically, larger lattice energy differences are observed for conformational polymorphs than for
polymorphs possessing identical conformers [26]. The contribution from Coulombic-polarization
energy to attractive forces was bigger for 3α, due to multiple hydrogen bonds, while the dispersion
energy contribution was smaller for this form. The dispersion interactions had a relevant contribution
to attractive interactions existing in the polymorphic forms of compound
of geometrical parameters, classical π···π pile stacking interactions do not exist in these compounds
(i.e., the interactions between orbitals of the cyclic molecular moieties, in which the centroids of
these moieties are separated at distance shorter than 4 Å) [27 28]. Only very weak interactions at
larger Cg···Cg distances [29 30] were observed in the structures of compound 3. These interactions
3, although, on the basis
π
,
,

Molecules 2020, 25, 993
9 of 14
can be considered as formed between the bonding
π
orbitals of one ring donating electron density
mainly to the antibonding
part to the one-center Rydberg antibonding orbitals of the
co-participation of C
determined on the basis of melting points because only two polymorphs can be currently produced,
and both melt with decomposition.
π
orbitals of the second ring (formal π···π interactions) and at a smaller
-bonded atom of the second ring (formal
cannot be
π
−
H
···π and π···π interactions) [29]. The stability of the polymorphs of
3
Table 2. The packing index and density values derived from structural data. Elatt is the calculated
total lattice energy (as −∆H of sublimation) and its components. EMP2 is the Hartree–Fock energy
with the inclusion of the Møller–Plesset correlation energy correction calculated for the atom positions
taken from x-ray coordinates registered at room temperature after optimization of the hydrogen atoms
positions. EM06-2X is the energy calculated with the M06-2X density functional with optimization of
whole molecule geometry. Energy values are in kJ/mol.
Density
EMP2
EM06-2X
Packing
Index (%)
Epol
Edisp
Erep
Compound
Elatt
ECoul
(g/cm3)
Elatt+EMP2
Elatt+EM06-2X
−1968944
−1973692
3α
3β
3γ
69.9
68.8
1.571
1.547
−
178.6
133.4
−
−
148.4 −60.3
107.1 −38.2
−
117.3
147.3
126.6
−1969123
−1973871
−1968999
−1969132
−1973732
−1973865
−
−
−
−
109.9
−1969005
−1973732
71.2
—
1.592
—
134.7 −99.7
−33.9
116.1
—
114.9
—
−1969140
−1973867
−2815948
−2821619
—
—
—
1
—
—
−3466466
—
—
1·dmfα
1·dmfβ
71.2
70.6
1.476
1.470
−
−
107.6 −81.3
109.1 −84.0
−33.3
−33.5
−
100.2
101.2
107.2
111.8
−3466574
−3466466
−3466575
—
—
−
Taking into account the conformational flexibility of nitrofural compounds and lattice energy
values, three polymorphs represented the two different types of interplay between the molecular and
crystal structure. Two forms (3β and 3γ) adopt lower energy conformation, which is more efficiently
packed in the crystal lattice, but creates a limited number of classical hydrogen bonds (in comparison
to 3α). 3α exists as a higher energy conformer (with different positions of functional groups taking
part in hydrogen bonds formation), which is less efficiently packed in the crystal lattice in comparison
to 3γ, but its stabilization is possible due to intermolecular interactions that allows for stable crystal
packing. The comparison of the energy differences between the two observed conformations of
3
(the geometrically optimized molecule of 3α was 9.6 kcal/mol energetically less favored than 3β and
3γ; Table 2) and the difference between the lattice energy of the polymorphic forms (3β net was 10.8
kcal/mol energetically less stable than the net of 3α, while respective value for 3γ and 3α was 10.5
kcal/mol; see Elatt+EM06-2X values in Table 2) proved that the stabilization energy of the lattice was close
to the energy difference between the favored and unfavored conformation of 3 (at 0.9–1.2 kcal/mol).
Analysis of the differences of the sum of lattice energy and Hartree–Fock energy with the inclusion of the
Møller–Plesset correlation energy correction (calculated for atom positions taken from x-ray coordinates
registered at room temperature after optimization of the hydrogen atoms positions; see Elatt + EMP2
values in Table 2) showed that the solid state forms 3β and 3γ were 3.6 and 4.0 kcal/mol, respectively,
more stable than 3α. This can also explain the problems with the experimental reproduction of the
form 3α, as the formation of 3α in the solid state requires the adoption of energetically less favorable
(in comparison to 3β and 3γ) conformation during crystallization. Nevertheless, all these forms
possessed similar total energy (including the influence of components originating from crystal lattice
and molecular conformation).
For both polymorphic forms of 1·dmf solvate, the calculated density, packing index, and lattice
energy were similar. The differences in crystal packing motifs for these two polymorphs were mainly
reflected in the energetic characteristics. In contrast to 1·dmfβ, for 1·dmfα, whose molecules are

Molecules 2020, 25, 993
10 of 14
stacked in piles along the a axis, the dispersive component is greater for the intermolecular interactions
between the molecules of and between the solvent molecules, at the expense of a smaller dispersive
1
contribution related to the interactions between the molecules of 1 and molecules of the solvent.
2.5. IR Spectra
The IR spectra of the studied compounds were generally similar due to their isomorphicity
(Figure S3), but some clear shifts of bands were observed as a result of the different conformation
of molecules and different intermolecular interactions (Table S5) [31]. All discussed spectra were
registered for the synthesized compounds in transmission mode to the maximal decrease of noise (in
previous works, the spectra were registered in reflectance mode). The formation of intramolecular
hydrogen bonds by the NH2 and N(imine) acceptor further interacted with the thiophene S atom
leads to a blue shift of νas NH2 vibrations present in the spectra of
compounds, Table S5). A similar blue shift was observed for the
in and NH2 vibrations) as a result of different electron
at about 1585 cm−1, together with
2
and
4 (in comparison to other
ν
CC(ring) vibrations (which exists
2
4
σ
withdrawing from the furan and thiophene rings by the NO2 substituent (the more electron rich
thiophene ring allows for the redirection of more electron density to the NO2 group and stabilize
the respective bonds restraining their vibrations). These effects caused a reverse (red) shift of the
ν
C=S vibration (weakening of the bond) in
2
in comparison to
1
and both of its solvates (1·dmf).
Such a red shift (in and ) was also clearly visible for the ρ
2
4
NH2 vibrations due to the reasons given
above. In the far IR spectra, the presence of the oscillations of hydrogen bonds formed by the NH2
group (NH2•••O2N and NH2•••O=C vibrations) caused an appearance of weak bands in the range of
359–376 cm-1. This proves that these bonds are the strongest ones in the studied compounds, and is
subsequently in agreement with the relative strength of the respective hydrogen bondsdonors and
acceptors. The NH•••O=C(dmf) vibrations can be clearly distinguished as they can appear only in
forms of 1·dmf. Respective bands were red shifted in comparison to the bands of hydrogen bonds
formed by the NH2 donor as a result of weaker interactions created by the NH secondary amine donor.
Further oscillators created by sets of hydrogen bonds were observed at smaller wavelengths. However,
it is hard to unambiguously assign specific oscillators to exact bands due to the similar strength of
other (than the above-mentioned) hydrogen bonds. The spectra of compound 3α were not registered
due to the impossibility of reproducing synthesis of this form and the far IR spectra of compounds 3β
and 3γ had no clearly visible bands in this region.
3. Materials and Methods
3.1. Synthesis
Equimolar quantities of thiosemicarbazide (1 mmol) and the appropriate carbonyl compound
(1 mmol of 5-nitro-2-furaldehyde and 5-nitro-2-thiophene, respectively) were dissolved in methanol
(30 mL). The solution mixtures were heated with continuous stirring for 20 min at 65 ◦C. The slow
evaporation of the solvent at room temperature gave the crystalline products: 5-nitro-2-furaldehyde
thiosemicarbazone (
semicarbazone (3β) and 5-nitro-2-thiophene semicarbazone (
only semicarbazide hydrochloride (1 mmnol) was used instead of thiosemicarbazide. Solid samples of
the synthesized compounds ( ) were dissolved in a pure solvent (MeOH, EtOH, i-PrOH, MeCN, or
1) and 5-nitro-2-thiophene thiosemicarbazone (
2). 5-Nitro-2-furaldehyde
4) were synthesized by the same procedure,
1–4
dmf) or solvent mixture (dmf + MeOH, dmf + EtOH, dmf + i-PrOH, dmf + MeCN, or MeOH + MeCN; 1
+ 1 v/v in all cases) and heated for 10 minutes and then left for crystallization under ambient conditions
(in covered, but not sealed, vessels to allow for the decelerated evaporation of solvents). Further
structurally different crystalline products were obtained by the slow evaporation of solvents only
for 1 and 3 (i.e., 5-nitro-2-furaldehyde thiosemicarbazone dimethylformamide dimethylformamide
solvate (1·dmfα, by slow evaporation from dmf solution), 5-nitro-2-furaldehyde thiosemicarbazone
dimethylformamide solvate (1·dmfβ, by slow evaporation from MeOH:dmf (1:1 v/v) solvent mixture),

Molecules 2020, 25, 993
11 of 14
and 5-nitro-2-furaldehyde semicarbazone (3γ, by slow evaporation from MeOH:MeCN (1:1 v/v) solvent
mixture as well as from pure MeCN)).
3.2. Crystal Structure Determination
The crystals of the studied compounds were mounted in turn on a Rigaku Synergy Dualflex
automatic diffractometer equipped with a Pilatus 300K detector and used for data collection. X-ray
intensity data were collected with mirror monochromated MoKα (
λ
= 0.71073 Å) or CuKα (
λ
= 1.54184
Å) radiation generated in a micro-focus sealed PhotonJet x-ray tube. The shutterless
ω
scan mode
was used, and reflections were collected at temperature of 100.0(1) K. The unit cell parameters were
determined from the strongest reflections. Details concerning the crystal data and refinement are
given in Table 1 for the newly determined structures and Table S1 for the redetermined structures.
Examination of the same reference reflections measured before and after measurement showed no loss
of the intensity during measurements. Lorentz, polarization, and empirical absorption corrections
(using spherical harmonics, implemented in SCALE3 ABSPACK scaling algorithm) were applied
during the data reduction. The structure was solved by the dual-space algorithm. All the non-hydrogen
atoms were refined anisotropically using the full-matrix, least-squares technique on F2. All hydrogen
atoms were found from difference Fourier synthesis after four cycles of anisotropic refinement. Carbon
bonded hydrogen atoms were refined as “riding” on the adjacent atom with geometric idealization
after each cycle of refinement. Individual isotropic displacement factors of H atoms were set to be equal
to 1.2 times the value of the equivalent displacement factors of the parent non-methyl carbon atoms,
and 1.5 times that of the parent methyl carbon atoms. Nitrogen bonded hydrogen atom positions
were freely refined (except one case described in the supplementary CIF file) and individual isotropic
displacement factors were fixed to 1.2 of the value of equivalent displacement factors of the parent
atoms. The methyl groups were allowed to rotate about their local three-fold axes. The SHELXT [32],
SHELXL [33], and SHELXTL [34] programs were used for all calculations. Atomic scattering factors
were taken from the International Tables for Crystallography [35]. Selected interatomic bond distances
are listed in Table S2.
CCDC 1976784-1976790 contain the supplementary crystallographic data for this paper. These data
can be obtained free of charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html (or from the CCDC,
12 Union Road, Cambridge CB2 1EZ, UK; Fax: +44-1223-336033; Email: deposit@ccdc.cam.ac.uk)
3.3. IR Spectrometry and Computational Methods
The IR spectra of the coordination compounds were recorded on a Jasco FT/IR 6200
spectrophotometer, in the form of KBr pellets for the spectral range 4000–400 cm−1 and in the
form of CsI pellets for the spectral range of 400–200 cm−1
.
The geometric parameters of the compounds were employed from the crystal structure data.
The calculations of the total molecule energy of polymorphs of and the conformational energy profile
3
were performed in GAUSSIAN09 [36] using M06-2X density functional [37], the MP2 method [38], and
the 6-31+g(d,p) basis set. Calculations using the MP2 and DFT methods showed the same energetic
trend for each conformer, although the MP2 calculations predicted lower energy barriers to rotation.
The experimentally obtained geometry was also used as input for the calculation of the lattice energy
and its components (total lattice energy divided into individual coulombic, polarization, dispersion,
and repulsion contributions to the interaction energies between each two-molecule system) using
the Pixelc module of the Coulomb-London-Pauli (CLP) computer program package [39]. Before the
calculations, the positions of the hydrogen atoms (determined form the x-ray diffraction experiments)
were reset to provide the correct assignment of the requested parameters and values (e.g., atomic
point charges). The required molecular electron densities were calculated using the MP2 [38] method
with a 6-31G(d,p) basis set in the GAUSSIAN09 package [36] in the form of discrete values forming
a three-dimensional grid (the cube). The net energies were calculated on the basis of numerical
integration over the respective number of units of electron densities. The applied numerical integrals

Molecules 2020, 25, 993
12 of 14
corresponded to the equations’ analytical form to provide parameter-free and accurate reproduction of
wavefunction values. The delocalization of electrons (omitted in point-charge methods due to their
nature) were assessed in terms of penetration energy to incorporate all parts of Coulombic energy into
calculations. The polarization terms were computed on the basis of the linear dipole approach [39
]
and dispersion terms on the basis of London’s inverse sixth power calculation (comprising ionization
potentials and polarizabilities) [40].
The Full Interaction Maps tool [41] and Polymorph Assessment tool [42] implemented in Mercury
software [43] were used to evaluate the preferred interactions of molecules, based on pre-extracted
IsoStar [44] interaction data from the CSD [25] and to ranking donors and acceptors based on propensity
scores, respectively.
4. Conclusions
The studied isomorphous molecules derived from 2-furaldehyde/2-thiophene and
semi-/thiosemicarbazone exist in three cases in one polymorphic form (
case in three polymorphic forms ( ). The recrystallization of the studied compounds from different
solvents gave solvates only in one case (of ) and with only one solvent (dmf). This, together with
almost identical conformations of 1·dmfβ 3β, and 3γ implies that the presence of the
1·dmfα
1, 2, and 4) and in one
3
1
1,
,
,
thiocarbonyl donor and 2-thiophene ring is a necessary condition to create a solvate. On the other
hand, the formation of two polymorphic forms of 1·dmf solvate (possessing the same conformation
of the constituent molecules and the same composition of crystal) proves that polymorphism in this
case is caused solely by the intermolecular interactions existing in the crystallization environment
(provided by OH groups of methanol molecules).
The present study outlines that adopting a conformation of the lowest energy is not necessary
to form a stable polymorphic form, as energy gained during the formation of the crystal net from
molecules with higher energy can be greater than the differences between the adopted and the lowest
energy confirmation. Additionally, the study proved that the presence of specific electronic restrains
can be an important factor in diminishing the number of observed polymorphic forms. In the discussed
case, one lone electron pair of nitrogen is located between two lone electron pairs of sulfur (compounds
2
and
2
4
). This reduces the flexibility of molecules and causes the presence of only one conformation
(Figure 2). Subsequently, it led to the creation of only one polymorphic form (for and 4)
for and
4
2
under the studied conditions.
Supplementary Materials: The supplementary materials are available online.
Author Contributions: Conceptualization, A.T.-K.; Methodology, A.T.-K., M.S., and R.K.; Investigation, A.T.-K.,
M.S., R.K., P.S. and A.D.; Writing—Original Draft Preparation, A.T.-K.; Writing—Review & Editing, A.T.-K., M.S.,
and R.K.; Visualization, A.T.-K. and M.S.; Funding Acquisition, A.T.-K. and R.K. All authors have read and agreed
to the published version of the manuscript.
Funding: This work was financed by funds allocated by the Ministry of Science and Higher Education (Poland) to
the Institute of General and Ecological Chemistry, Lodz University of Technology.
Conflicts of Interest: The authors declare no conflicts of interest.
References
1.
2.
3.
Michalczyk, M.; Zierkiewicz, W.; Wysokinski, R.; Scheiner, S. Theoretical studies of IR and NMR spectral
changes induced by sigma-hole hydrogen, halogen, chalcogen, pnicogen, and tetrel bonds in a model protein
environment. Molecules 2019, 24, 3329. [CrossRef] [PubMed]
Jiao, Y.; Weinhold, F. What Is the Nature of Supramolecular Bonding? Comprehensive NBO/NRT Picture of
Halogen and Pnicogen Bonding in RPH2···IF/FI Complexes (R = CH3, OH, CF3, CN, NO2). Molecules 2019
,
24, 2090. [CrossRef] [PubMed]
Trujillo, C.; Alkorta, I.; Elguero, J.; Sanchez-Sanz, G. Cooperative effects in weak interactions: Enhancement
of tetrel bonds by intramolecular hydrogen bonds. Molecules 2019, 24, 308. [CrossRef] [PubMed]

Molecules 2020, 25, 993
13 of 14
4.
5.
6.
7.
Sieranski, T. The intricacies of the stacking interaction in a pyrrole–pyrrole system. Struct. Chem. 2016, 27,
1107–1120. [CrossRef]
Sieranski, T. Discovering the stacking landscape of a pyridine-pyridine system. J. Mol. Model. 2017, 23, 338.
[CrossRef]
Price, S.L. Control and prediction of the organic solid state: A challenge to theory and experiment. Proc. R.
Soc. A 2018, 474, 20180351. [CrossRef]
Iuzzolino, L.; Reilly, A.M.; McCabe, P.; Price, S.L. Use of Crystal Structure Informatics for Defining the
Conformational Space Needed for Predicting Crystal Structures of Pharmaceutical Molecules. J. Chem. Theory
Comput. 2017, 13, 5163–5171. [CrossRef]
8.
9.
Ether, M.C. Encoding and decoding hydrogen-bond patterns of organic compounds. Acc. Chem. Res. 1990
23, 120–126. [CrossRef]
Ether, M.C. Hydrogen bonds as design elements in organic chemistry. J. Phys. Chem. 1991, 95, 4601–4610.
,
[CrossRef]
10. Desiraju, G.R.; Sarma, J.A.R.P. The chloro-methyl exchange rule and its violations in the packing of organic
molecular solids. Proc. Indian Acad. Sci. (Chem. Sci.) 1986, 96, 599–605. [CrossRef]
11. Corpinot, M.K.; Guo, R.; Tocher, D.A.; Buanz, A.B.M.; Gaisford, S.; Price, S.L.; Bucˇar, D.-K. Are Oxygen and
Sulfur Atoms Structurally Equivalent in Organic Crystals. Cryst. Growth Des. 2017, 17, 827–833. [CrossRef]
12. Bis, J.A.; Zaworotko, M.J. The 2-Aminopyridinium-carboxylate Supramolecular Heterosynthon: A Robust
Motif for Generation of Multiple-Component Crystals. Cryst. Growth Des. 2005, 5, 1169–1179. [CrossRef]
13. Shattock, T.R.; Arora, K.K.; Vishweshwar, P.; Zaworotko, M.J. Hierarchy of Supramolecular Synthons:
Persistent Carboxylic Acid···Pyridine Hydrogen Bonds in Cocrystals That also Contain a Hydroxyl Moiety.
Cryst. Growth Des. 2008, 8, 4533–4545. [CrossRef]
14. Kavuru, P.; Aboarayes, D.; Arora, K.K.; Clarke, H.D.; Kennedy, A.; Marshall, L.; Ong, T.T.; Perman, J.;
Pujari, T.; Wojtas, Ł.; et al. Hierarchy of Supramolecular Synthons: Persistent Hydrogen Bonds Between
Carboxylates and Weakly Acidic Hydroxyl Moieties in Cocrystals of Zwitterions. Cryst. Growth Des. 2010
10, 3568–3584. [CrossRef]
,
15. Duggirala, N.K.; Wood, G.P.F.; Fischer, A.; Wojtas, Ł.; Perry, M.L.; Zaworotko, M.J. Hydrogen Bond Hierarchy:
Persistent Phenol···Chloride Hydrogen Bonds in the Presence of Carboxylic Acid Moieties. Cryst. Growth
Des. 2015, 15, 4341–4354. [CrossRef]
16. Budziak, I.; Arczewska, M.; Kamin´ski, D.M. Formation of prenylated chalcone xanthohumol cocrystals:
Single crystal X-Ray diffraction, vibrational spectroscopic study coupled with multivariate analysis. Molecules
2019, 24, 4245. [CrossRef]
17. Wuest, J.D. Engineering crystals by the strategy of molecular tectonics. Chem. Commun. 2005, 5830–5837.
[CrossRef]
18. Hosseini, M.W. Molecular Tectonics: From Simple Tectons to Complex Molecular Networks. Acc. Chem. Res.
2005, 38, 313–323. [CrossRef]
19. Laliberté, D.; Maris, T.; Wuest, J.D. Molecular tectonics. Use of urethanes and ureas derived from
tetraphenylmethane and tetraphenylsilane to build porous chiral hydrogen-bonded networks. Can. J.
Chem. 2004, 82, 386–398. [CrossRef]
20. Cole, J.C.; Groom, C.R.; Read, M.G.; Giangreco, I.; McCabe, P.; Reilly, A.M.; Shields, G.P. Generation of crystal
structures using known crystal structures as analogues. Acta Cryst. Sect. B Struct. Sci. Cryst. Eng. Mater.
2016, 72, 530–541. [CrossRef]
21. Thompson, H.P.G.; Day, G.M. Which conformations make stable crystal structures? Mapping crystalline
molecular geometries to the conformational energy landscape. Chem. Sci. 2014, 5, 3173–3182. [CrossRef]
22. Olszak, T.A.; Peeters, O.M.; Blaton, N.M.; De Ranter, C.J. 5-Nitro-2-furaldehyde semicarbazone. Acta
Crystallogr., Sect. C Cryst. Struct. Commun. 1994, 50, 948–950. [CrossRef]
23. Pogoda, D.; Janczak, J.; Videnova-Adrabinska, V. New polymorphs of an old drug: Conformational and
synthon polymorphism of 5-nitrofurazone. Acta Crystallogr. Sect. B Struct. Sci. Cryst. Eng. Mat. 2016, 72,
263–273. [CrossRef] [PubMed]
24. Sharara, K.N.; Nyamayaro, K.; Wicht, M.M.; Venter, G.A.; Bathori, N.B. Multicomponent crystals of
nitrofurazone – when more is less. CrystEngComm 2019, 21, 1091–1096. [CrossRef]
25. Groom, C.R.; Bruno, I.J.; Lightfoot, M.P.; Ward, S.C. Cambridge Structural Database, CSD v 5.39, August
2018. Acta Crystallogr. Sect. B Struct. Sci. Cryst. Eng. Mater. 2016, 72, 171–179. [CrossRef]

Molecules 2020, 25, 993
14 of 14
26. Cruz-Cabeza, A.J.; Reutzel-Edens, S.M.; Bernstein, J. Facts and fictions about polymorphism. Chem. Soc. Rev.
2015, 44, 8619–8635. [CrossRef]
27. Hunter, C.A.; Sanders, J.K.M. The nature of.pi.-.pi. interactions. J. Am. Chem. Soc. 1990, 112, 5525–5534.
[CrossRef]
28. Huber, R.G.; Margreiter, M.A.; Fuchs, J.E.; von Grafenstein, S.; Tautermann, C.S.; Liedl, K.R.; Fox, T.
Heteroaromatic π-Stacking Energy Landscapes. J. Chem. Inf. Model. 2014, 54, 1371–1379. [CrossRef]
29. Kruszynski, R.; Sieranski, T. Can Stacking Interactions Exist beyond the Commonly Accepted Limits? Cryst.
Growth Des. 2016, 12, 587–595. [CrossRef]
30. Hayashi, N.; Higuchi, H.; Ninomiya, N. X/π Interactions in Aromatic Heterocycles: Basic Principles and
Recent Advances. Top. Heterocycl. Chem. 2009, 18, 1–35.
31. Narayan, V.; Mishra, H.N.; Prasad, O.; Sinha, L. Electronic structure, electric moments and vibrational
analysis of 5-nitro-2-furaldehyde semicarbazone: A D.F.T. study. Comput. Theor. Chem. 2011, 973, 20–27.
[CrossRef]
32. Sheldrick, G.M. SHELXT– Integrated space-group and crystal-structure determination. Acta Crystallogr. Sect.
A Found. Adv. 2015, 71, 3–8. [CrossRef] [PubMed]
33. Sheldrick, G.M. Crystal structure refinement with SHELXL. Acta Crystallogr. Sect. C Struct. Chem. 2015, 71,
3–8. [CrossRef] [PubMed]
34. Sheldrick, G.M. A short history of SHELX. Acta Crystallogr., Sect. A: Found. Adv. 2008, 64, 112–122. [CrossRef]
[PubMed]
35. International Tables for Crystallography, Volume C: Mathematical, Physical and Chemical Tables, 3rd ed.; Prince, E.
(Ed.) Kluwer Academic Publishers: Dordrecht, The Netherlands, 2004.
36. Frisch, M.J.; Trucks, G.W.; Schlegel, H.B.; Scuseria, G.E.; Robb, M.A.; Cheeseman, J.R.; Scalmani, G.; Barone, V.;
Mennucci, B.; Petersson, G.A.; et al. Gaussian 09, Revision, D.01; Gaussian, Inc.: Wallingford, CT, USA, 2009.
37. Zhao, Y.; Truhlar, D.G. The M06 suite of density functionals for main group thermochemistry, thermochemical
kinetics, noncovalent interactions, excited states, and transition elements: Two new functionals and systematic
testing of four M06-class functionals and 12 other functionals. Theor. Chem. Acc. 2008, 120, 215–241.
38. Frisch, M.J.; Head-Gordon, M.; Pople, J.A. A direct MP2 gradient method. Chem. Phys. Lett. 1990, 166,
275–280. [CrossRef]
39. Gavezzotti, A. Efficient computer modeling of organic materials. The atom–atom, Coulomb–London–Pauli
(AA-CLP) model for intermolecular electrostatic-polarization, dispersion and repulsion energies. New J.
Chem. 2011, 35, 1360–1368. [CrossRef]
40. Maschio, L.; Civalleri, B.; Ugliengo, P.; Gavezzotti, A. Intermolecular Interaction Energies in Molecular
Crystals: Comparison and Agreement of Localized Møller–Plesset 2, Dispersion-Corrected Density Functional,
and Classical Empirical Two-Body Calculations. J. Phys. Chem. A 2011, 115, 11179–11186. [CrossRef]
41. Wood, P.A.; Olsson, T.S.G.; Cole, J.C.; Cottrell, S.J.; Feeder, N.; Galek, P.T.A.; Groom, C.R.; Pidcock, E.
Evaluation of molecular crystal structures using Full Interaction Maps. CrystEngComm 2013, 15, 65–72.
[CrossRef]
42. Galek, P.T.A.; Fábián, L.; Motherwell, W.D.S.; Allen, F.H.; Feeder, N. Knowledge-based model of
hydrogen-bonding propensity in organic crystals. Acta Crystallogr. B Struct. Sci. Cryst. Eng. Mater.
2007, 63, 768–782. [CrossRef]
43. Macrae, C.F.; Bruno, I.J.; Chisholm, J.A.; Edgington, P.R.; McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.;
Taylor, R.; van de Streek, J.; Wood, P.A. Mercury CSD 2.0– new features for the visualization and investigation
of crystal structures. J. Appl. Cryst. 2008, 41, 466–470. [CrossRef]
44. Bruno, I.J.; Cole, J.C.; Lommerse, J.P.M.; Rowland, R.S.; Taylor, R.; Verdonk, M.L.J. IsoStar: A library of
information about nonbonded interactions. Comput. Aided Mol. Des. 1997, 11, 525–537. [CrossRef] [PubMed]
Sample Availability: Samples of the compounds
1
,
1·dmfα
,
1·dmfβ, 2, 3β, 3γ and 4 are available from the authors.
©
2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).