Non-covalent interactions involving remote substituents influence the topologies of supramolecular chains featuring hydroxyl-O-H?O(hydroxyl) hydrogen bonding in crystals of (HOCH2CH2)2NC(S)N(H)(C6H4Y-

Article abstract of DOI:10.1039/d0ce01810d

Crystallography shows the universal adoption of supramolecular chains featuring hydroxyl-O-H O(hydroxyl) hydrogen bonding in crystals of (HOCH₂CH₂)₂NC(S)N(H)(C₆H₄Y-4) for Y = H (1), Me (2), Cl (3) and

Full text of DOI:10.1039/d0ce01810d

CrystEngComm
View Article Online
View Journal | View Issue
PAPER
Non-covalent interactions involving remote
substituents influence the topologies of
Cite this: CrystEngComm, 2021, 23,
723
1
supramolecular chains featuring hydroxyl-O–
H⋯O(hydroxyl) hydrogen bonding in crystals of
(
HOCH CH ) NC(S)N(H)(C H Y-4) for Y = H, Me,
2
2 2
6 4
Cl and NO2†
Sang Loon Tan
and Edward R. T. Tiekink
*
Crystallography shows the universal adoption of supramolecular chains featuring hydroxyl-O–
H⋯O(hydroxyl) hydrogen bonding in crystals of (HOCH CH NC(S)N(H)(C Y-4) for Y = H (1), Me (2),
Cl (3) and NO (4). However, distinct topologies, i.e. linear (Y = H), helical (Y = Me and Cl) and zig-zag (Y =
NO ) are noted with major differences in the pitch of the polymer. Geometry-optimisation, MEP and NPA
analyses show a distinct electronic structure for the Y = NO derivative, in particular relating to the
2
)
2 2
6 4
H
2
2
2
activation of the aryl ring. An exhaustive analysis of the molecular packing (point-to-point interactions,
crystal structure similarity, Hirshfeld surface analysis, NCI and QTAIM, interaction energies and energy
frameworks) points to the importance of C–H⋯π(aryl) interactions in stabilising the chains but these have a
considerably reduced influence in the crystal with Y = NO2 (4), where π(aryl)⋯π(aryl) interactions are
important. The more open arrangement for the linear chain in 1 facilitates the formation of C–H⋯π(aryl)
interactions and the more compact arrangements enable the formation of stabilising, intra-chain
methylene-C–H⋯S(thione) interactions 2–4. This study highlights the role of second-tier non-covalent
interactions in the arrangement of conventional hydrogen bonding interactions.
Received 15th December 2020,
Accepted 18th January 2021
DOI: 10.1039/d0ce01810d
rsc.li/crystengcomm
interactions provide similar energies of stabilisation in their
crystals and being inherently weak are therefore, flexible,
Introduction
In the organic solid-state, conventional hydrogen bonding
plays privileged role in arranging molecules into
supramolecular assemblies,
being subject to moderation by chemical substitution, steric
effects, etc. The delineation of the role of these different
modes of association is highly desirable in order to
rationalise more fully the assembly of molecules in crystals
as even small changes in molecular packing can influence
macroscopic properties relating to, e.g. optoelectronic
a
1–4
often by design employing
5
the supramolecular synthon approach. This prominent role
notwithstanding, other intermolecular contacts come to the
fore when conventional hydrogen bonding does not occur in
three dimensions or is not present at all. Here, a myriad of
alternative interactions come to the fore, such as π⋯π, C–
H⋯π, lone-pair⋯π, chalcogen bonding, halogen bonding,
7
8
properties, drug discovery and the conformation of
9
molecules. Further, the control of flexible, cooperative
supporting intermolecular interactions will lead to the
strategic design of higher dimensional aggregation patterns
in crystals featuring persistent, structure-directing hydrogen
bonding patterns operating in zero-, one- or two-
dimensions.
6
etc. Along with hydrogen bonding, many of these
Research Centre for Crystalline Materials, School of Science and Technology,
Sunway University, 47500 Bandar Sunway, Selangor Darul Ehsan, Malaysia.
E-mail: edwardt@sunway.edu.my
These
aforementioned
considerations
increasingly
motivate systematic studies of crystals featuring (i) a common
†
Electronic supplementary information (ESI) available: Crystallographic data,
hydrogen bonding aggregation pattern despite the presence
electrostatic potential charge deviations, NPA charges, HOMO–LUMO plots,
10–14
of different substituents,
(ii) multiple hydrogen bonding
PXRD patterns, molecular packing diagram, NCI and QTAIM plots. CCDC
options and their adoption related to small chemical
2047050–2047053 contain the supplementary crystallographic data for this
1
5–23
changes
and (iii) no conventional hydrogen bonding
paper. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/d0ce01810d
present with studies conducted in order to ascertain the
This journal is © The Royal Society of Chemistry 2021
CrystEngComm, 2021, 23, 1723–1743 | 1723

View Article Online
Paper
CrystEngComm
influence of other non-covalent interactions upon molecular Experimental
2
4–27
aggregation.
Herein, an investigation related to scenario
Instrumentation
(i) is presented. As anticipated from the formula of the
All chemicals and solvents were used as purchased without
purification. The melting points (uncorrected) were measured
using a Stuart SMP30 melting point apparatus. The IR spectra
molecules investigated herein, i.e. (HOCH CH ) NC(S)N(H)
2
2 2
(
C H Y-4) for Y = H (1), Me (2), Cl (3) and NO (4), Fig. 1,
6
4
2
hydrogen bonding is prominent and a consistent adoption of
supramolecular chains in the respective crystals is apparent,
in each case mediated by hydroxyl-O–H⋯O(hydroxyl)
interactions. However, the chains display distinct topologies,
i.e. linear (1), helical (2 and 3) and zig-zag (4), and of the
series, only 2 and 3 are isostructural.
were measured on
a
Bruker Vertex 70v FT-IR
−
1
1
13
1
spectrophotometer from 4000 to 80 cm . H and C{ H}
NMR spectra were recorded in DMSO-d6 solutions on a
Bruker Ascend 400 MHz NMR spectrometer with chemical
shifts relative to tetramethylsilane (TMS). The absorption
spectra were measured on 100 μM acetonitrile solutions in
the range of 180–700 nm on a double-beam Shimadzu UV
Compounds 1–4 are examples of tri-substituted derivatives
2
8
of thiourea, a well-known class of compounds. While crystal
structures are known for derivatives conforming to the
3600 Plus UV-vis spectrophotometer. The CHN elemental
2
9–31
analyses were performed on a LECO TruSpec Micro analyser
under a helium atmosphere with glycine being the standard.
The room temperature powder X-ray diffraction (PXRD)
patterns were measured on a Rigaku MiniFlex 600 X-ray
general formula (HOCH
2
CH
2
)N(R)C(S)N(H)R,
for R =
alkyl, aryl, none are known for the di-hydroxyethyl analogues,
i.e. (HOCH CH ) NC(S)N(H)R. Indeed, with the exception
2
2 2
3
2
of 1, which was investigated recently for anti-leishmanial
activity, compounds 2–4 do not appear to have been reported
previously. Herein, the synthesis, spectroscopic and
crystallographic characterisation of 1–4 are described along
with a detailed analysis of the molecular packing in their
crystals with the aim of ascertaining the role of the Y = H,
1
diffractometer with Cu Kα radiation (λ = 1.5418 Å) within a
2
θ range of 5–70° and a step size of 0.02°. The comparisons
between the experimental and calculated (from the respective
CIF) PXRD patterns were performed with Rigaku's PDXL2
software (https://www.rigaku.com/en/products/software/pdxl/
overview).
Me, Cl and NO
association.
2
substituents in the supramolecular
The objectives of the computational studies are three-fold,
i.e. firstly, to validate whether the experimental structures
represent the global minima through a conformational
analysis, an analysis which has significant implications for
their molecular packing. Secondly, to gain insight into the
electronic nature of the molecules through molecular
electrostatic potential (MEP) and natural population analysis
Synthesis
A
common mode of synthesis was adopted for the
preparation of 1,1-bis(2-hydroxyethyl)-3-phenylthiourea (1),
,1-bis(2-hydroxyethyl)-3-(4-tolyl)thiourea (2) and 1,1-bis(2-
1
hydroxyethyl)-3-(4-chlorophenyl)thiourea (3). Thus, 1 mmol of
the corresponding aryl isothiocyanate (phenyl isothiocyanate
(
0.135 g); 4-tolyl isothiocyanate (0.149 g); 4-chlorophenyl
(NPA) studies in order to ascertain any particular features in
isothiocyanate (0.169 g) all from Sigma) was reacted with an
equimolar amount of diethanolamine (Sigma, 0.105 g) in
ethanol (30 ml) followed by stirring for 3 h at room
temperature. White precipitates were formed upon the
addition of dichloromethane (3 ml). The products were
filtered and subsequently washed with cold ethanol (2 ml).
Recrystallisation in hot ethanol resulted in the formation of
colourless blocks after slow evaporation. For 1,1-bis(2-
hydroxyethyl)-3-(4-nitrophenyl)thiourea (4), the product was
obtained by mixing diethanolamine (1 mmol, 0.105 g) in
acetone (5 ml) with an equimolar amount of 4-nitrophenyl
isothiocyanate (Acros, 0.180 g) which was pre-dissolved in
acetone (30 ml). The mixture was then concentrated to half
of the initial volume through slow evaporation with stirring
at room temperature. Upon the formation of a yellow
precipitate, the product was filtered and washed with a
mixture of ethanol and ethyl acetate (4 ml, v/v). Yellow blocks
were obtained through recrystallisation of the crude product
in absolute ethanol under slow evaporation.
the electronic structures of the molecules that may impact
the molecular packing. Finally, to qualitatively and
quantitatively assess the molecular interactions present in
each crystal through Hirshfeld surface analysis, interaction
energy calculations, energy framework simulations, lattice
energy calculation, non-covalent interaction plots and
quantum theory of atoms in molecules (QTAIM). The aim of
these studies is to correlate the molecular conformation and
electronic structure to determine and explain the main
factors that influence the manner in which the thiourea
derivatives pack in their crystals.
Characterisation
1
: colourless crystals, yield: 0.188 g (78%). M.pt.: 362.2–363.8
2 2 2 6 4
Fig. 1 Chemical diagrams for the (HOCH CH ) NC(S)N(H)(C H Y-4)
molecules investigated herein.
K. Calcd. for C H N O S: C 54.98, H 6.71, 11.66%. Found: C
1
1
16 2 2
1724 | CrystEngComm, 2021, 23, 1723–1743
This journal is © The Royal Society of Chemistry 2021

View Article Online
CrystEngComm
Paper
−
1
5
3
1
9
5.01, H 6.79, N 11.38%. IR (ATR, cm ): 3247 (m) ν(O–H),
ppm): δ 9.76 (s, 1H, NH), 7.34 (s, 4H, ortho- and meta-aryl-H),
3
145 (m) ν(N–H), 3093 (m) ν(C–Haro), 3066–2845 (w) ν(C–H),
312 (s) ν(C–N), 1033 (s) ν(CS). H NMR (DMSO-d
.72 (s, 1H, NH), 7.29 (m, 4H, ortho- and meta-phenyl-H), 7.08
m, 1H, para-phenyl-H), 5.31 (br, 2H, OH), 3.84 (t, 4H, J
–N), 3.73 (dt, 4H, JHH = 5.3 Hz, JH–OH = 4.6 Hz,
CH –O). C{ H} NMR (DMSO-d , ppm): δ 182.23 (CS),
5.32 (br, 2H, OH), 3.85 (t, 4H, J = 5.2 Hz, CH –N), 3.73 (dt,
HH
2
1
3
3
13
1
6
, ppm): δ
2
4H, JHH = 5.04 Hz, JH–OH = 4.84 Hz, CH –O). C{ H} NMR
(DMSO-d , ppm): δ 186.84 (CS), 145.06 (ipso-C), 133.14
6
3
(
=
(ortho-C), 133.01 (para-C), 131.07 (meta-C), 64.51 (C–O), 59.57
HH
3
3
−1
−1
5
.3 Hz, CH
2
3
(C–N). UV/vis (acetonitrile, 100 μM, nm, L mol cm ): λmax: 282
(sh), ε = 12 023; 255, ε = 22 909; 226 (sh), 13 490.
1
1
2
6
1
41.34 (ipso-C), 128.52 (ortho-C), 124.71 (meta-C), 124.39
4: yellow crystals, yield: 0.224 g (79%). M.pt.: 449.4–450.8 K.
(
para-C), 59.85 (C–O), 54.78 (C–N). UV/vis (acetonitrile, 100
Calcd. for C11
H
15
N
3
O
4
S: C 46.31, H 5.30, N 14.73%. Found: C
−
1
−1
−1
μM, nm, L mol cm ): λmax: 277 (sh), ε = 10 964; 254, ε =
1
46.34, H 5.36, N 14.44%. IR (ATR, cm ): 3259 (w) ν(O–H), 3224
(w) ν(N–H), 3075 (w) ν(C–Haro), 3015–2832 (w) ν(C–H), 1505 (s)
9 953; 222 (sh), ε = 14 791.
: colourless crystals, yield: 0.207 g (82%). M.pt.: 391.8–
92.5 K. Calcd. for C H N O S: C 56.67, H 7.13, N 11.01%.
2
ν(NOasym), 1474 (s) ν(NOsym), 1291 (s) ν(C–N), 1027 (s)
1
3
ν(CS). H NMR (DMSO-d , ppm): δ 10.37 (s, 1H, NH), 8.18 (d,
12
18
2
2
6
−
1
3
3
Found: C 56.65, H 7.21, N 10.86%. IR (ATR, cm ): 3258 (m)
2H, J
= 9.16 Hz, meta-aryl-H), 7.64 (d, 2H, J
= 9.04 Hz,
HH
HH
ν(O–H), 3192 (m) ν(N–H), 3131 (m) ν(C–Haro), 3057–2885 (w)
ortho-aryl-H), 5.82 (br, 1H, OH), 5.18 (br, 1H, OH), 3.88 (4H, t,
1
3
3
3
ν(C–H), 1291 (s) ν(C–N), 1025 (s) ν(CS). H NMR (DMSO-d ,
JHH = 4.92 Hz, CH –N), 3.76 (4H, dt, J
5.16 Hz, CH –O). C{ H} NMR (DMSO-d , ppm): δ 181.50
2 6
= 5.04 Hz, J
=
6
2
13
HH
H–OH
3
1
ppm): δ 9.60 (s, 1H, NH), 7.18 (d, 2H, JHH = 8.36 Hz,
3
ortho-aryl-H), 7.09 (d, 2H,
J
HH = 8.2 Hz, meta-aryl-H), 5.29
= 5.36 Hz, CH –N), 3.72 (dt,
(CS), 147.87 (para-C), 142.38 (ipso-C), 124.59 (meta-C), 122.37
(ortho-C), 59.61 (C–O), 55.00 (C–N). UV/vis (acetonitrile, 100 μM,
3
(
4
CH
br, 2H, OH), 3.83 (t, 4H, J
HH
2
3
3
−1
−1
H,
J
HH = 5.12 Hz,
J
H–OH = 4.8 Hz, CH
). C{ H} NMR (DMSO-d
ipso-C), 138.33 (para-C), 133.72 (ortho-C3), 129.70 (meta-C),
2
–O), 3.06 (s, 3H,
nm, L mol cm ): λmax: 349, ε = 15 488; 299 (sh), ε = 10 715;
240, ε = 16 982; 224 (sh), ε = 14 454.
1
3
1
3
6
, ppm): δ 187.04 (CS), 143.52
(
6
4.61 (C–O), 59.50 (C–N), 25.72 (methyl-C). UV/vis
−
1
−1
X-ray crystallography
(
acetonitrile, 100 μM, nm, L mol cm ): λmax: 278 (sh), ε =
318; 253, ε = 16 218; 222 (sh), ε = 12 589.
: colourless crystals, yield: 0.215 g (78%). M.pt.: 395.5–396.8
K. Calcd. for C11 15ClN S: C 48.09, H 5.50, N 10.20%. Found:
C 48.01, H 5.58, N 10.02%. IR (ATR, cm ): 3242 (w) ν(O–H),
187 (w) ν(N–H), 3127 (w) ν(C–Haro), 3042–2935 (w) ν(C–H), 1301
8
The crystallographic and refinement data for 1–4 are given in
Table 1. Intensity data were measured at 100 K on an Agilent
Technologies SuperNova Dual diffractometer fitted with an
Atlas detector. Data processing and Gaussian absorption
corrections were accomplished with CrysAlis Pro. Each
structure was solved by direct methods and the refinement
3
H
2 2
O
−1
3
3
3
1
34
(s) ν(C–N), 1062 (s) ν(CS), 691 (m) ν(C–Cl). H NMR (DMSO-d
6
,
Table 1 Crystallographic data and refinement details for 1–4
Compound
1
2
3
4
Formula
11 16 2
C H N O
2
S
C
12 18 2 2
H N O S
11
C H15ClN
2 2
O S
C
11 15 3 4
H N O S
Molecular weight
Crystal size/mm
Colour
Crystal system
240.32
0.30 × 0.30 × 0.30
Colourless
254.34
0.05 × 0.05 × 0.15
Colourless
274.76
0.05 × 0.05 × 0.15
Colourless
Monoclinic
1
P2 /n
285.32
0.30 × 0.35 × 0.40
Yellow
Triclinic
P¯1
3
Monoclinic
Monoclinic
Space group
1
P2 /n
P2
1
/n
a/Å
b/Å
c/Å
α/°
β/°
γ/°
13.4885(1)
11.1767(1)
16.4909(2)
90
98.544(1)
90
2458.53(4)
8
1.299
7.0472(2)
10.7489(2)
16.9533(4)
90
99.109(2)
90
1268.01(5)
4
1.332
7.1366(2)
10.7767(3)
16.6259(4)
90
100.221(2)
90
1258.39(6)
4
1.450
10.8235(5)
11.2124(5)
12.3443(5)
90.050(3)
108.737(4)
114.559(4)
1274.53(11)
4
3
V/Å
Z
−3
D
c
/g cm
1.487
0.269
−1
μ/mm
2.253
2.213
4.184
Measured data
9855
5049
4685
5859
Radiation
Cu Kα
Cu Kα
Cu Kα
Mo Kα
θ range/°
Unique data
4.0–75.0
4995
4.9–75.0
2586
4.9–75.0
2554
2.3–27.5
5859
Observed data (I ≥ 2.0σ(I))
No. parameters
R, obs. data; all data
a; b in weighting scheme
4702
307
0.030; 0.032
0.039; 0.851
0.077; 0.078
−0.30–0.29
2346
164
0.029; 0.033
0.033; 0.456
0.071; 0.074
−0.22–0.21
2243
163
0.031; 0.037
0.040; 0.220
0.075; 0.080
−0.31–0.24
4753
362
0.041; 0.054
0.056; 0.615
0.104; 0.114
−0.30–0.82
R
w
, obs. data; all data
−3
Range of residual electron density peaks/e Å
This journal is © The Royal Society of Chemistry 2021
CrystEngComm, 2021, 23, 1723–1743 | 1725

View Article Online
Paper
CrystEngComm
2
was carried out using full-matrix least squares on F with
charges for the corresponding optimised structures were
54,55
anisotropic displacement parameters for all non-hydrogen
The C-bound hydrogen atoms were placed on
stereochemical grounds and refined with fixed geometries.
The O- and N-bound hydrogen atoms were refined with O–H
0.84 ± 0.01 Å and N–H = 0.88 ± 0.01 Å, respectively. A
weighting scheme of the form w = 1/[σ (F ) + (aP) + bP],
where P = (Fo + 2F )/3, was introduced in each refinement.
Owing to poor agreement, reflections, i.e. (1 3 0) for 1 and
−1 0 7) for 2, were omitted from the final cycles of
refinement. Finally, 4 was refined as a two-component twin
obtained by natural population analysis (NPA)
using
3
5
atoms.
wB97XD/def2-TZVP. The electrostatic potential (ESP) was
mapped onto the electron density iso-surfaces with a constant
3
electronic charge of 0.002 electrons per bohr through the
5
6
=
cubegen utility as available in GaussView6.
Further, a
2
2
o
2
57
molecular packing analysis was performed using Mercury,
with the analysis criteria being set where only molecules within
20% tolerance for both distances and angles were included in
the calculation while molecules with a variation >20% were
discarded. Differences in the molecular structures (i.e. the
substituents in the 4-position) and molecular inversions were
allowed during the calculation.
2
2
c
(
with the fraction due to the minor component = 0.142(3). The
3
6
36
37
programs WinGX, ORTEP-3 for Windows, PLATON and
3
8
DIAMOND were also used in the study.
For the qualitative evaluation of the strength of
interactions, a non-covalent interaction (NCI) visualisation
index was generated for the respective interacting dimers
Computational studies
5
8
using NCIPLOT through the plotting of the reduced density
gradient as a function of the density across the molecules.
The computed density derivatives were mapped as iso-
surfaces which correspond to any favourable or unfavourable
interactions as determined by the sign of the second density
Hessian eigenvalue times the density and visualised using
A conformational search was performed through a Monte Carlo
algorithm using the Merck Molecular Force Field (MMFF), as
39
4
0,41
available in Spartan'16,
9
with the energy cut-off being set to
.6 kcal mol . To increase the accuracy on the Boltzmann
distribution, the generated conformers were subjected to
−1
4
2,43
geometry optimisation using the ab initio HF/3-21G model
5
9
VMD Molecular Graphics Viewer.
followed by energy calculations through the long-range
corrected wB97XD density functional with Grimme's D2
Hirshfeld surface mapping, the corresponding two-
dimensional fingerprint plots and pairwise interaction
calculation were generated using Crystal Explorer 17 (ref. 60)
44
45
dispersion model coupled with Pople's 6-31G(d) basis set.
The long-range corrected hybrid model has been shown to
greatly reduce self-interaction errors and give better accuracy in
6
1
through an established method as reported previously, with
46
the experimental structures being used as the input with X–H
the interaction energies. Upon elimination of redundant
structures with minor conformational changes as well as those
62
bond lengths adjusted to their neutron-derived values. The
interaction energy calculations were performed using the
dispersion corrected CE-B3LYP/6-31G(d,p) model as available
in the program, with the total intermolecular energy being
the sum of energies of the four main components,
comprising electrostatic, polarisation, dispersion and
exchange-repulsion with scale factors of 1.057, 0.740, 0.871
−1
exceeding the 9.6 kcal mol energy window, the remaining
conformers were then submitted for further optimisation at
4
7,48
the wB97XD/6-311+G(d,p) level.
At this stage, a frequency
analysis was performed using the same level of theory and
basis set to ensure the validity of the ground state structures.
Finally, all identified conformers were submitted into
Gaussian16 (ref. 49) for optimisation using wB97XD with
Ahlrichs's valence triple-zeta polarization basis sets (wB97XD/
6
3
and 0.618, respectively. The model was validated against
the B3LYP-D2/6-31G(d,p) counterpoise corrected energy
model as well as the benchmark CCSD(T)/CBS model with
5
0,51
def2-TZVP)
and with the employment of the polarisable
6
4
considerable accuracy. The energy frameworks for 1–4 were
computed for a cluster of 2 × 2 × 2 unit cells with the energy
continuum model by placing the solute in a cavity within an
ethanol solvent reaction field through the integral equation
formalism variant of the polarisable continuum model
−
1
cut-off being set to 1.9 kcal mol . Finally, the total energy
was obtained for a cluster of molecules within a 25 Å radius
from a selected reference molecule through the same level of
theory and basis set model. The lattice energy for the
52
(
IEFPCM). The Gibbs free energies were obtained through
frequency calculations of the optimised structures at the same
level of theory and basis set.
6
5
corresponding crystals was calculated using eqn (2), where
the second term is the cell dipole energy correction, with ρcell
being the vector sum of the molecular dipole moments, Vcell
being the volume and Z being the number of formula units
in the unit cell, respectively. Typically, the cell dipole energy
The relative population of each conformer was determined
53
through a Boltzmann weighting factor using eqn (1), with ΔG
i
being the Gibbs free energy of species i relative to the most
stable conformer, j is the specific conformer (j = 1, 2, 3…), R is
the gas constant and T is absolute temperature set to 298 K.
−
1
correction is negligible (<0.24 kcal mol ) for a unit cell with
6
6
−
ΔGi=RT
a small dipole moment.
e
Boltzmann weighting factor; Pi ¼ P
× 100% (1)
ΔGj=RT
−
e
j¼1
X
2πρcell2
3ZVcell
1
2
AB
total
Elattice ¼
E
−
(2)
RAB<R
Several molecular properties were computed in an attempt
to correlate the molecular packing in 1–4. Briefly, the atomic
1726 | CrystEngComm, 2021, 23, 1723–1743
This journal is © The Royal Society of Chemistry 2021

View Article Online
CrystEngComm
Paper
For 1 and 4, each with Z′ = 2, the lattice energy was
calculated as the average of lattice sums for each molecule in
the asymmetric unit.
LUMO(+2) are located at the delocalised C1N1/C1N2,
C2C3 and C4C5/C6C7 chromophores, respectively,
which indicates that the experimental UV absorption bands
at approximately 280, 250 and 220 nm can be attributed to n
→
π*, π → π* and π → π* transitions, respectively. As for 4,
Results and discussion
Synthesis and characterisation
the delocalised chromophore associated with the nitro group
contributes to LUMO(+1) and hence, the additional
absorption band at approximately 350 nm can be assigned to
π → π*. The PXRD pattern measured for each of 1–4 closely
matches the simulated pattern calculated from their single
crystal data, confirming the phase similarity between the
respective bulk materials (293 K) and experimental structures
(100 K); see ESI† Fig. S2.
The (HOCH CH ) NC(S)N(H)(C H Y-4), Y = H (1), Me (2), Cl
2
2 2
6
4
2
(3) and NO (4), compounds have been prepared in good
yields (78–82%) as colourless (1–3) and yellow (4) crystals. In
the IR spectra, characteristic bands in the regions 1291–1312
−
1
−1
cm
and 1025–1062 cm
are assigned to ν(C–N) and
1
ν(CS), respectively. In the H NMR spectra, measured in
DMSO-d solution, the expected resonances and integration,
6
including those for the N–H and O–H protons, were noted. In
the C{ H} NMR spectra, resonances due to the quaternary-
1
3
1
Experimental molecular structures
C1 atom were seen downfield, in the range of 181.50 (4) to
Crystal structures were established for 1–4; for each of 1 and
4, two independent molecules comprise the asymmetric unit,
henceforth 1a & 1b and 4a & 4b, respectively. The molecular
structures are shown in Fig. 2 and selected geometric data
are collated in Table 2. The first independent molecule of 1,
Fig. 2(a), features a planar C1, N1, N2, S1 chromophore
which exhibits an r.m.s. deviation of 0.0038 Å for the fitted
1
87.04 ppm (2). In order to assign the transitions in the UV
spectra, an analysis on the HOMO–LUMO profile was
performed for the lowest energy conformer at the ground
state (vide infra) for each of 1–4; see ESI† Fig. S1. This
revealed that the HOMO is located at the C1S1
chromophore for 1–3, while the LUMO, LUMO(+1) and
Fig. 2 Molecular structures of (a) 1 (first independent molecule), (b) 1 (second molecule), (c) 2, (d) 3, (e) 4 (first independent molecule) and (f) 4
second independent molecule), all showing atom labelling schemes and displacement ellipsoids at the 50% probability level. Overlap diagrams of
the (g) experimental and (h) geometry optimised structures – the molecules have been overlapped so that the central N S residues are coincident.
(
2
Colour code: 1 (first independent molecule), red; 1 (second independent molecule), green; 2 (inverted), blue; 3, cyan; 4 (first independent
molecule), pink; 4 (inverted second independent molecule), yellow.
This journal is © The Royal Society of Chemistry 2021
CrystEngComm, 2021, 23, 1723–1743 | 1727

View Article Online
Paper
CrystEngComm
Table 2 Selected experimental and calculated (in italics) geometric data (Å, °) characterising 1–4
Parameter
1a
1b
2
3
4a
4b
C1–S1
1.7028(12)
1.6985(12)
1.6986(13)
1.696
1.3573(16)
1.368
1.3539(16)
1.376
127.08(11)
128.1
123.22(10)
122.8
121.49(11)
120.9
115.29(10)
115.8
122.78(10)
123.2
122.69(9)
122.9
114.52(11)
113.9
4.79(19)
13.2
53.74(19)
46.3
1.6968(16)
1.694
1.360(2)
1.372
1.354(2)
1.375
126.00(14)
128.3
123.07(13)
122.9
121.54(13)
120.8
115.39(12)
115.8
122.52(12)
123.2
1.6900(18)
1.6893(18)
1.690
1.374(2)
1.380
1.345(2)
1.372
128.95(15)
129.8
123.44(15)
123.1
121.47(15)
120.8
115.08(14)
115.7
123.14(14)
123.3
1
1
1
1
1
1
1
1
1
1
.695
.370
.376
28.4
22.9
20.9
15.8
23.3
22.9
13.8
C1–N1
1.3518(16)
1.3499(15)
127.40(11)
122.75(10)
122.86(10)
114.35(9)
122.32(9)
122.68(9)
114.99(10)
14.67(18)
51.55(19)
−165.08(9)
17.34(16)
73.33(12)
74.54(13)
1.3511(15)
1.3544(15)
130.19(10)
122.65(10)
122.59(10)
114.70(9)
123.78(9)
121.91(9)
114.29(10)
13.32(18)
31.80(19)
−160.46(9)
16.56(16)
82.80(13)
73.76(13)
1.375(2)
C1–N2
1.344(2)
C1–N1–C2
128.13(15)
123.67(15)
121.06(15)
115.27(14)
122.81(14)
123.09(14)
114.09(15)
3.9(3)
C1–N2–C8
C1–N2–C10
C8–N2–C10
S1–C1–N1
S1–C1–N2
122.83(12)
123.0
114.64(14)
113.8
123.10(13)
123.3
113.74(15)
113.4
N1–C1–N2
S1–C1–N1–C2
C1–N1–C2–C3
S1–C1–N2–C8
S1–C1–N2–C10
N2–C8–C9–O1
N2–C10–C11–O2
r.m.s. deviation CN
−5.3(2)
−5.8(3)
1
3.9
−13.8
−19.0
−55.6(2)
−43.5
42.1(3)
−39.5(3)
−30.8
4
4.0
−166.78(9)
166.73(12)
158.6
−166.45(13)
13.9(2)
165.12(13)
159.8
−
158.3
−158.1
13.32(16)
13.3
−13.6(2)
−16.0(2)
1
7
4
3.1
2.1
9.8
−12.8
−12.0
71.13(14)
72.3
59.47(14)
50.0
−71.15(17)
−72.0
61.7(2)
−63.9(2)
−71.2
−58.80(18)
−49.4
52.5(2)
−55.1(2)
−48.7
2
S
0.0038
0.0065
0.0027
0.0036
0.0048
0.0043
2
CN S/aryl
59.39(4)
39.07(4)
54.3(5)
53.2
56.98(6)
51.0
42.70(7)
41.33(8)
43.4
51.6
atoms. The mono-substituted amine-N1 atom carries a
phenyl ring and the di-substituted amine-N2 atom carries
two hydroxyethyl groups. A significant twist in the molecule
is apparent with the dihedral angle between the central plane
and appended phenyl ring being 59.39(4)°. This observation
plus the presence of two methylene-C atoms bound to the N2
atom suggests that there is no extensive delocalisation of
π-electron density over the molecule; the C1–N bond lengths
are experimentally equivalent. Consistent with the presence
of the C1S1 double bond at the C1 atom, the angles
subtended at the N2 atom involving the C1 atom are the
widest. However, the widest angle at a nitrogen atom is the
C1–N1–C2 angle which reflects the presence of the amine-H
atom. A similar distortion in angles is seen about the C1
atom. Rather than being “dangling”, the hydroxyl groups are
orientated towards the rest of the molecule enabling the
formation of intramolecular hydroxyl-O–H⋯S(thione) and
amine-N–H⋯O(hydroxyl) hydrogen bonds and S(7) loops, as
detailed in Table 3. As seen from Table 2, the key bond
lengths and angles defining the independent molecules of 1,
Fig. 2(b), are generally close with the most significant
difference being a wider angle by about 3° for C1–N1–C2 in
the second molecule. Greater differences are noted in torsion
angles, Table 2. The maximum difference of approximately
20° is noted in the C1–N1–C2–C3 torsion angles followed by
approximately 10° for the N2–C8–C9–O1 torsion angles. A
difference of approximately 20° is also seen in the CN S/aryl
2
torsion angles with reduced splaying between the planes
noted in the second molecule of 1.
Very similar molecular conformations are noted for 2–4,
Fig. 2(c)–(f) and Table 2, including the formation of
intramolecular hydrogen bonds, Table 3. The most notable
differences between 1 and each of 2–4 relate to the more
planar S1–C1–N1–C2 torsion angles and to the reduced N2–
C10–C11–O2 torsion angles, by up to 18°, in 2–4. Two
independent molecules comprise the asymmetric unit of 4
but each presents very similar geometric parameters, Table 2.
However, a distinguishing feature of the two molecules
comprising 4 and each of 1–3 relates to an apparent disparity
in the C1–N1 and C1–N2 bond lengths in 4 not seen in the
latter; this observation is discussed further below. An overlay
diagram of the experimental molecular structures is shown
in Fig. 2(g) from which it is plain that significant
conformational differences with respect to the relative
orientations of both the hydroxyethyl and aryl groups are
evident across the series.
1728 | CrystEngComm, 2021, 23, 1723–1743
This journal is © The Royal Society of Chemistry 2021

View Article Online
CrystEngComm
Paper
Table 3 Summary of intra- and inter-molecular interactions (A–H⋯B; Å, °) operating in the crystals of 1–4
A
H
B
A–H
H⋯B
A⋯B
A–H⋯B
Symmetry operation
1
N1
O2
N21
O22
O1
O21
C9
H1n
H2o
H21n
H22o
H1o
H21o
H9a
H9b
O1
S1
O21
S21
O22
O2
Cg(C22–C27)
Cg(C22–C27)
Cg(C2–C7)
0.873(14)
0.841(14)
0.872(13)
0.838(14)
0.839(11)
0.842(16)
0.99
1.914(14)
2.269(15)
1.900(14)
2.323(15)
1.841(11)
1.842(16)
2.68
2.7514(15)
3.1009(9)
2.7624(14)
3.1470(9)
2.6753(13)
2.6825(14)
3.6030(13)
3.6822(13)
3.8131(14)
160.1(15)
170.1(13)
169.4(13)
167.6(14)
173.0(17)
176.1(15)
155
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
x, y, 1 + z
1 − x, 1 − y, 1 − z
−½ + x, ½ − y, −½ + z
1 − x, 1 − y, 1 − z
C9
C29
0.99
0.99
2.78
2.95
151
147
H29b
2
N1
O2
O1
C8
C10
H1n
H2o
H1o
H8a
H10b
O1
S1
O2
Cg(C2–C7)
0.875(14)
0.839(12)
0.837(15)
0.99
1.906(13)
2.367(15)
1.882(16)
2.89
2.7449(16)
3.1500(10)
2.7137(14)
3.4957(13)
3.6263(19)
160.1(14)
155.4(16)
172.0(19)
120
x, y, z
x, y, z
−½ − x, −½ + y, 1½ − z
−1 + x, y, z
−½ + x, ½ − y, −½ + z
C6
0.99
2.73
150
3
N1
O2
O1
C8
C10
C5
H1n
H2o
H1o
H8a
H10b
Cl1
O1
S1
O2
Cg(C2–C7)
C1l
0.881(18)
0.828(13)
0.837(17)
0.99
0.99
1.7448(17)
1.907(18)
2.388(16)
1.879(17)
2.95
2.85
3.4432(6)
2.7453(19)
3.1591(13)
2.7156(18)
3.5488(17)
3.6738(14)
—
158.3(17)
155.2(18)
178(2)
120
141
x, y, z
x, y, z
−½ − x, ½ + y, ½ − z
−1 + x, y, z
−1½ + x, 1½ − y, −½ + z
2 − x, 1 − y, 1 − z
Cl1
154.00(6)
4
N1
O2
H1n
H2o
O1
S1
O21
S21
O22
O2
O1
O3
0.88(2)
0.84(3)
0.88(2)
0.839(14)
0.838(19)
0.83(2)
0.99
0.99
0.99
0.99
—
2.03(2)
2.40(2)
2.02(2)
2.41(2)
1.890(19)
1.90(2)
2.39
2.39
2.72
2.71
—
2.842(2)
3.1671(16)
2.839(2)
3.1767(16)
2.723(2)
2.732(2)
3.375(3)
3.360(3)
3.475(2)
3.483(2)
3.6105(12)
153(2)
152(3)
154(2)
152(3)
173(2)
172.7(19)
175
165
133
135
0
x, y, z
x, y, z
x, y, z
x, y, z
x, 1 + y, z
x, y, z
−x, 1 − y, 1 − z
x, y, 1 + z
1 − x, 1− y, 2 − z
1 − x, 1 − y, 1 − z
−x, −y, 2 − z
N21
O22
O1
O21
C29
C11
C10
C9
H21n
H22o
H1o
H21o
H29b
H11b
H10b
H9b
Cg(C22–C27)
Cg(C2–C7)
Cg(C22–C27)
Cg(C22–C27)
Molecular packing
differences in the topologies are reflected in the distances
between translationally related pairs of molecules, i.e. 16.49,
10.75, 10.77 and 11.21 Å, respectively, indicating more open
arrangements in the sequence 1 > 4 > 3 and 2.
The geometric parameters characterising the specific
intermolecular contacts operating in the crystals of 1–4 are
collated in Table 3. The common feature of the supramolecular
aggregation is the formation of supramolecular chains
mediated by hydroxyl-O–H⋯O(hydroxyl) hydrogen bonding.
However, the topologies of the resultant chains are distinct. In
In 1, the only other identifiable points of contact between
the supramolecular chains are methylene-C–H⋯π(aryl)
interactions, with the C22–C27 ring accepting two
interactions, one to either side, and these serve to assemble
the chains into a three-dimensional architecture, Fig. 4(a).
In the crystal of 2, the chains are connected into a two-
dimensional array in the ab-plane by methylene-C–H⋯π(C2–
C7) interactions. The layers stack along the c-axis with the
closest interaction between them being a methylene-C–
H⋯C(aryl) contact, Table 3. A view of the unit-cell contents
for isostructural 3 is shown in Fig. 4(c). Here, there is
evidence for weak inter-layer methylene-C–H⋯Cl and Cl⋯Cl
contacts, Table 3 and Fig. 4(b). A distinct molecular packing
1, the two similarly orientated molecules comprising the
asymmetric unit are connected by single hydroxyl-O–
a
H⋯O(hydroxyl) hydrogen bond and the resultant two-molecule
aggregates are connected into a linear supramolecular chain
parallel to the c-axis in the crystal with the monoclinic space
group P2 /n. In isostructural 2 and 3, helical chains are formed,
1
being propagated by 2
monoclinic space group P2
independent molecules are connected into V-shaped aggregates
1
-screw symmetry along the b-axis in the
/n, in their crystals. In 4, the
1
which are connected to translationally related aggregates to
is noted in the crystal of 4 primarily owing to the
form a zig-zag chain along the b-axis in the triclinic (P ¯1 ) crystal.
The side- and end-on views of the supramolecular chains are
illustrated for 1, 2 and 4 in Fig. 3, and for 3 in ESI† Fig. S3. The
participation of hydroxyl- and nitro-O atoms in contacts
along with face-to-face π⋯π contacts, Table 3. Thus, in 4 the
methylene-C–H⋯π(aryl) contacts, present in all crystals,
This journal is © The Royal Society of Chemistry 2021
CrystEngComm, 2021, 23, 1723–1743 | 1729

View Article Online
Paper
CrystEngComm
Fig. 3 Side- and end-on views of the supramolecular chains featuring hydroxyl-O–H⋯O(hydroxyl) hydrogen bonding, shown as orange dashed
lines, in the crystals of (a) 1, (b) 2 and (c) 4.
connect the chains into a centrosymmetric double-layer in
the bc-plane, within which are supporting methylene-C–
H⋯O(nitro) contacts. The double-layers stack along the
a-axis with methylene-C–H⋯O(hydroxyl) and π(aryl)⋯π(aryl)
contacts assembling the layers into a three-dimensional
architecture, Fig. 4(d). The π(aryl)⋯π(aryl) contacts occur
between centrosymmetrically related residues and the rings
are off-set to optimise the attraction: the inter-plane
separation is 3.4128(8) Å and the slippage is 1.18 Å.
chain (4), as highlighted in the overlay diagrams for 1 & 2
and 1 & 4 in Fig. 5(a) and (b), respectively. For comparison,
Fig. 5(c) shows the equivalent image for isostructural 2 and 3
where the r.m.s. deviation is 0.173 Å. An intermediate
situation when 2 (and 3) is compared with 4, Fig. 5(d), where
the r.m.s. deviation is 0.388 Å; the r.m.s. deviation between 3
and 4 is 0.461 Å. A closer inspection of the supramolecular
chains of 2 and 4, Fig. 5(e), shows that every second molecule
1
of 2 has an alternate orientation, reflecting the 2 -screw
symmetry, compared with the molecules having the same
relative orientation reflecting the pseudo-mirror symmetry of
the zig-zag chain of 4.
Molecular packing similarity analysis
A packing similarity analysis was performed between 1–4 to
identify any similarities in the molecular arrangements in
5
7
Conformational analysis
their crystals. The results show that the packing in 1 is
quite distinct from that of 2, 3 and 4, with only one molecule
out of 15 falling within the 20% tolerance in both distance
and angle deviations. The r.m.s. deviation between 1 & 2, 1 &
Owing to the presence of the hydroxyethyl moieties in 1–4,
which may participate in various intra- and inter-molecular
interactions depending on the conformations they adopt, a
3
and 1 & 4 amounts to 0.946, 0.931 and 0.934 Å, respectively.
detailed conformational analysis of
a
representative
The major deviation arises as in the crystal of 1, the
molecules are mainly connected through hydroxyl-O–
H⋯O(hydroxyl) interactions in a linear arrangement, while in
each of 2–4, the molecules are connected by the same
interaction but arranged in a helix (2 and 3) or a zig-zag
molecule, namely 4, was conducted to assess whether the
observed experimental structure represents a conformation at
or close to the global minimum on the potential energy
surface. A striking feature of the molecular structures of 1–4
was the universal formation of intramolecular hydroxyl-O–
1730 | CrystEngComm, 2021, 23, 1723–1743
This journal is © The Royal Society of Chemistry 2021

View Article Online
CrystEngComm
Paper
Fig. 4 Unit-cell contents for (a) 1, viewed down the c-axis, (b) 2, viewed down the b-axis, (c) 3, viewed down the a-axis and (d) 4, viewed down
the b-axis; the views in (a), (b) and (d) are in projection down the axes of propagation of the chains. The hydroxyl-O–H⋯O(hydroxyl), methylene-
C–H⋯π(aryl), methylene-C–H⋯C(aryl), methylene-C–H⋯Cl, Cl⋯Cl, methylene-C–H⋯O(hydroxyl, nitro) and π(aryl)⋯π(aryl) interactions are shown
as orange, purple, pink, dark-red, cyan, blue and dark-green dashed lines, respectively.
H⋯S(thione) and amine-N–H⋯O(hydroxyl) hydrogen bonds,
Fig. 2 and Table 3. The (any atom)N(H)C(S)N(CH CH OH)
true local minima or the global minimum structure on the
potential energy surface. The chemical diagrams for the
identified conformers together with the energy details are
presented in Fig. 6. Clearly, among all possible
conformations identified for 4, conformer 4-1 is the global
minimum structure with the lowest Gibbs free energy and is
also the most dominant conformer with the highest relative
population of about 82%. Two other conformers lie within 2
2
2
2
7
6
fragment is relatively rare in the crystallographic literature,
being restricted to a small number of ArC(O)N(H)C(S)
6
8–71
N(CH CH OH)
2
molecules
and
a
bi-functional
2
2
7
2
analogue. A common feature of the literature precedents is
the formation of the intramolecular amine-N–H⋯O(hydroxyl)
hydrogen bonds but no analogous intramolecular hydroxyl-
O–H⋯S(thione) interactions as in 1–4. Given this
observation, it was thought of interest to ascertain whether
the orientations of the flexible hydroxyethyl residues in 1–4
corresponded to the global potential energy minima.
Accordingly, a conformational analysis through a series of
optimisation steps (see Experimental) was performed on a
representative molecule, namely, 4. In all, nine conformers
with lowest energy were identified upon consecutive
elimination of the redundant conformers and those with
−
1
kcal mol , namely 4-2 and 4-3, with Boltzmann populations
of 10.45 and 3.52%, respectively. Clearly, the intramolecular
hydrogen bonds play a crucial role in stabilising the observed
−
1
molecular conformation, by about 1–5 kcal mol compared
to the conformation without intramolecular hydrogen
bonding, i.e. 4-9.
Additional structural information was revealed through
this analysis. Crucially, the six most stable conformations, 4-
1 to 4-6, have an anti-disposition of the thione-S and amine-
H atoms, with S1–C1–N1–H1n torsion angles in the range of
140.1 to 156.9°, compared syn-dispositions in 4-7 to 4-9 (S1–
C1–N1–H1n: 19.7 to 20.8°) with the difference in energy
−
1
relative energy exceeding 9.6 kcal mol throughout the series
of optimisation steps. As validated through the vibrational
analysis with zero imaginary frequency, the final
optimisation showed that all nine conformers were either
−
1
between 4-6 and 4-7 being 0.2–0.3 kcal mol . The analysis
This journal is © The Royal Society of Chemistry 2021
CrystEngComm, 2021, 23, 1723–1743 | 1731

View Article Online
Paper
CrystEngComm
Fig. 5 Comparisons of the molecular packing leading to the supramolecular chains (only four molecules of each are shown) between (a) 1 (blue)
and 2 (green), (b) 1 and 4 (red), (c) 2 and 3 (magenta), (d) 2 and 4 and (e) end-on view of 2 and 4 with the differently-orientated hydroxyl-O–
H⋯O(hydroxyl) interactions. In (a)–(e), best-fitting molecules are highlighted in ball and stick representation.
Fig. 6 Chemical diagrams for the nine most stable conformers calculated for 4, i.e. 4-1 to 4-9, and their relative energies and Boltzmann
distribution.
1732 | CrystEngComm, 2021, 23, 1723–1743
This journal is © The Royal Society of Chemistry 2021

View Article Online
CrystEngComm
Paper
also highlights the importance of the intramolecular amine-
N–H⋯O(hydroxyl) hydrogen bonds compared with the
variations in geometric parameters notwithstanding, it is
emphasised that the differences are small and are unlikely to
have a significant influence upon the molecular packing.
hydroxyl-O–H⋯S(thione)
and
putative
hydroxyl-O–
H⋯O(hydroxyl) hydrogen bonds, with conformers 4-1 to 4-4
being lower in energy compared with conformers without
amine-N–H⋯O(hydroxyl) hydrogen bonds.
Molecular electrostatic potential (MEP) and natural
population analysis (NPA)
Geometry optimisation calculations for 1–4 were
conducted and an overlay diagram for these are shown in
Fig. 2(h) from which it can be seen that the disparities in the
conformations of the experimentally-determined structures
no longer persist, with only minor differences noted in the
relative orientation conformations of the aryl groups. Selected
geometric parameters for the optimised structures are
collated in Table 2. First and foremost, any differences
apparent between the independent molecules of 1 and of 4
are no longer apparent. For example, the difference in the
C1–N1–C2 angles of approximately 3° in 1 disappears.
Concerning the relative dispositions of the aryl groups,
Fig. 2(h), it was noted above that the S1–C1–N1–C2 angles in
experimental 2–4 were closer to planarity compared with 1
but in the optimised molecules, all approximate the
conformation seen in experimental and theoretical 1, yet a
range of angles, i.e. 13.2° (2) to −19.0° (4) is still apparent.
This and a range of about 12° for the C1–N1–C2–C3 torsion
angles are the exceptional differences with all of the other
angles equal within a degree of each other. With respect to
bond lengths, in the crystals of 1–3 the C1–N1, N2 bond
lengths are equal within experimental error but are distinct
for each independent molecule in 4, Table 2. In the
optimised structures, C1–N2 is marginally longer than C1–N1
in 1–3 but, for 4, there are more significant differences
apparent with the C1–N1 bond length being longer than C1–
N2. This change is related to the influence of the
electronegative nitro substituent in 4. The above systematic
Compounds 1–4 were subjected to molecular electrostatic
potential (MEP) mapping and a natural population analysis
(NPA) in order to better comprehend the distribution of
electron density over the molecules (the relative polarity) with
the view to correlate any systematic trends with the non-
covalent interactions operating in the respective crystals. It is
noted that the NPA approach was chosen for the charge
calculations for its reliability and for being less sensitive to
7
3
the choice of basis set functions.
As shown in Fig. 7, the MEP maps were plotted onto the
iso-density surfaces (0.0004 a.u., the low value being chosen
for the generation of high-quality mapping) for 1–4; a listing
of the electrostatic charges is given in ESI† Table S1. The
most noteworthy features of the MEP plots are the intense
positive (blue) regions centred on the H1o atoms as well as
the negative (orange to red) regions around the S1 and O2
atoms with the electrostatic potential charge (VESP) on the
−
1
surfaces being in the range of +55.73 to +59.49 kcal mol for
−
1
H1o, −30.37 to −36.09 kcal mol for S1, and −39.30 to −42.45
−
1
kcal mol for the O2 atom. The electrostatic potential
charges correspond well with the experimental findings in
that electropositive-H1o interacts with electronegative-O2
through charge-complementary, electrostatic attractions
which results in systematic hydroxy-O–H⋯O(hydroxy)
hydrogen bond formation in 1–4. While there are some
inequivalent distributions of electrostatic potential charge on
the H1o and O1 atoms in 1–4, the net charge (ΔVESP) is
Fig. 7 Electrostatic potentials mapped onto the iso-density surfaces (0.0004 a.u.) for 1–4, in the range of −60.00 to +60.00 kcal mol−1
.
This journal is © The Royal Society of Chemistry 2021
CrystEngComm, 2021, 23, 1723–1743 | 1733

View Article Online
Paper
CrystEngComm
approximately the same across the series with the values
0.50 and 0.48, respectively, in accord with the trends with the
hydroxyl-O atoms. Consistent with the MEP study, there is a
relatively large difference in the natural charges between the
hydroxyl-oxygen and -hydrogen atoms and this is the main
contributing factor for the formation of the common
hydroxy-O–H⋯O(hydroxy) hydrogen bonded chain in 1–4.
It is noted that small but consistent trends in the charges
residing on the S1 and N2 atoms are apparent, with respect
to 1, minor increases in 2 and decreases in 3 and 4 correlated
with the electronegativity of Y. Variations in the natural
charges are also noted in the aryl rings, in particular for the
C5 atoms with respect to 1, observations again related to the
Y-substituents. For 2, the inductive effect of the σ-electron
donating methyl group disperses the charge around the
−
1
being 97.74, 98.44, 98.72 and 98.79 kcal mol , respectively,
indicating similar strengths for these interactions. These net
charge values are relatively greater than the electrostatic
attraction for putative methylene-C–H⋯S(thione) interactions
in the molecular packing for 1 and 4, with energies of 65.12
−
1
and 66.61 kcal mol , respectively, and aryl-C–H⋯S(thione)
−
1
for 1 with an energy = 41.05 kcal mol . However, in 1, these
occur at distances beyond the van der Waals radii so they are
3
7
not indicated in the analysis conducted using PLATON but
are noted in the Hirshfeld surface analysis (vide infra). The
identified methylene-C–H⋯Cl1 and Cl⋯Cl interactions in 3,
−
1
Table 3, have net charge values of 32.40 and −1.88 kcal mol ,
respectively, indicating that the latter is diffusive in nature.
In addition, methylene-C–H⋯O(nitro) and aryl-C–H⋯O(nitro)
7
4
π-system through the resonance effect.
A
similar
−
1
in 4 have energies equal to 49.43 and 44.72 kcal mol ,
respectively; the latter occurs at separations greater than the
van der Waals radii.
observation is found for the Cl substituent in 3 as it able to
donate the lone-pair of electrons to the aryl ring leading to a
similar inductive effect to that for the methyl substituent
despite the chloride atom being known as a weak electron-
The effect of the variation of the 4-Y substituent of the aryl
rings is evidenced through MEP mapping. Thus, the
electrostatic charge on the centroid of the aryl ring becomes
more negative from 1 to 2 due to the electron donating
nature of the Y = Me substituent in 2, while the opposite is
true for 3 and 4, as the aryl rings become less negative owing
to the electron-withdrawing effects of the Y = Cl and NO2
substituents.
7
5
withdrawing group.
The significant differences for all
atoms comprising the aryl ring in 4 relate specifically to the
electronegative nitro substituent. Overall, the net charge
7
6
shift
for the 4-substituted phenylthiourea fragments
P
compared to the parent molecule, i.e. [ q(SCNHC
6
H
4
Y) −
P
q(SCNHC H )], amounts to 0.004, −0.007 and −0.033 e for
2–4, respectively, which correlates to the electron-donating
nature of Me and electron-withdrawing characteristics of Cl
and NO2.
6
5
Similar to the MEP analysis, the NPA study was conducted
to seek trends in the charge distribution on specific atoms
especially those participating in intermolecular interactions
for correlation with molecular packing. A list of selected NPA
values is given in Table 4 and a full listing is given in ESI†
Table S2. The NPA charge analysis shows that the most basic
sites are located on the hydroxyl-O1 and O2 atoms with the
natural charge values in the ranges −0.736 to −0.737 and
Hirshfeld surface analysis
Compounds 1–4 were subjected to Hirshfeld surface analysis
in order to gain further insight on the nature of interactions
present in each crystal, especially those not identified in the
conventional analysis of the molecular packing, as well as
important surface contacts; the analysis includes the
contributions made by the individual components
comprising the asymmetric unit of each of 1 and 4, labelled
henceforth 1a & 1b and 4a & 4b, respectively The mapping of
the normalised contact distances (dnorm) reveals several red
spots on the iso-density surfaces of the molecules ranging
from strong to weak intensity due to the presence of several
close contacts with separations shorter than the sum of van
−
0.743 to −0.749, respectively, indicating that the O2 atom is
marginally more negative. The corresponding H1o and H2o
atoms are the most acidic sites with charge values of about
Table 4 The natural charges for selected atoms in the optimised
molecules of 1–4
Natural charge, |e|
Atom
1
2
3
4
62
der Waals radii. These are categorised into five main types
S1
−0.393
−0.613
−0.396
−0.737
−0.748
0.345
−0.397
−0.611
−0.397
−0.737
−0.749
0.346
−0.387
−0.613
−0.393
−0.737
−0.747
0.344
−0.361
−0.614
−0.384
−0.736
−0.743
0.338
as H⋯O/O⋯H (type I), H⋯C/C⋯H (type II), H⋯S/S⋯H (type
III), C⋯O/O⋯C (type IV) and H⋯Cl/Cl⋯H or Cl⋯Cl (type V),
Fig. 8. A summary of the contacts detected on the Hirshfeld
surfaces is provided in ESI† Table S3 where all the X–H bond
lengths have been adjusted to their neutron values.
The most intense red spots arise from H⋯O/O⋯H
interactions due to the hydroxyl-O–H⋯O(hydroxyl) hydrogen
bonds and these are a common feature of all crystals under
investigation. The differences between the molecules are
observed mainly in the diminutive red spots comprising type
II, III, IV and V contacts. For instance, a type IV contact
appears only for 1b due to thiourea-C1⋯O(hydroxyl)
N1
N2
O1
O2
C1
C2
C3
C4
C5
C6
C7
H1o
H2o
H1n
0.141
0.126
0.141
0.195
−0.215
−0.208
−0.226
−0.205
−0.234
0.496
−0.205
−0.212
−0.024
−0.209
−0.224
0.496
−0.197
−0.224
−0.053
−0.222
−0.216
0.496
−0.222
−0.169
−0.010
−0.164
−0.235
0.498
0.480
0.444
0.480
0.444
0.480
0.445
0.480
0.448
1734 | CrystEngComm, 2021, 23, 1723–1743
This journal is © The Royal Society of Chemistry 2021

View Article Online
CrystEngComm
Paper
Fig. 8 Complementary views of the Hirshfeld surface mapped over dnorm within the range of −0.0788 to 1.0548 arbitrary units, revealing close
contacts shorter than the sum of van der Waals radii through red spots on the surfaces which are categorised into H⋯O/O⋯H (type I), H⋯C/
C⋯H (type II), H⋯S/ S⋯H (type III), C⋯O/O⋯C (type IV) and H⋯Cl/Cl⋯H or Cl⋯Cl (type V) for (a) 1a, (b) 1b, (c) 2, (d) 3, (e) 4a and (f) 4b.
interaction, while type III contacts are observed in 1 and 4
attributed to methylene-/aryl-C–H⋯S(thione) interactions but
no such contact is noted in either 2 or 3. The inclusion of Cl
contacts are also reflected as a pincers-like profile but with d
i
+ d
e
distances in the range of 2.83 to 2.94 Å for 1 & 4 and 2.90
P
to 3.03 Å for 2 & 3 compared with the (vdW)
= 2.89 Å. It
S⋯H
and NO
2
substituents in the 4-positions of 3 and 4 introduces
is for this reason that diminutive red spots were observed in
the relevant plots for 1 and 4 but not in those for 2 & 3; as
discussed below, the H⋯S contacts in 2–4 are intra-chain
contacts. Distinctive characteristics are noted in 3 arising
additional contacts compared with the parent molecule of 1
as evidenced from the presence of red spots, albeit of weak
intensity. For 3, these arise from methylene-C–H⋯Cl and
Cl⋯Cl interactions in 3, while the NO
2
substituent in 4 gives
from H⋯Cl/Cl⋯H as well as Cl⋯Cl contacts with d
i
+ d
e
of
P
rise to methylene-/aryl-C–H⋯O(nitro) interactions. Almost all
of these interactions identified through the Hirshfeld surface
analysis can be considered weak contacts which complement
about 2.78 and 3.42 Å which are shorter than
(vdW)
Cl⋯H
P
and
(vdW)Cl⋯Cl of 2.84 and 3.50 Å, respectively, Fig. 9.
Other contacts co-exist on the Hirshfeld surface but are less
significant owing to long contact separations.
3
7
those interactions detected via PLATON.
The quantification of the close contacts to the Hirshfeld
surface was achieved though the analysis of the two-
dimensional fingerprint plots for the respective molecules in
In terms of contact distributions, crystals 1 and 2 are
dominated by several major contact contributions to the
Hirshfeld surfaces in the order of H⋯H (ca. 57.3–59.5%), H⋯C/
C⋯H (ca. 17.2–18.7%), H⋯S/S⋯H (ca. 10.6–12.0%) and H⋯O/
O⋯H (ca. 8.4–10.8%) followed by other less significant contacts
with each contributing less than 1% as shown in Fig. 10. The
decomposition of the distribution shows that almost all
contacts in 1 and 2 are evenly distributed between the internal
(i.e. the donor or acceptor atoms internal to the surface) and
external (i.e. the donor or acceptor atoms external to the
surface) interactions except for H⋯C/C⋯H and H⋯S/S⋯H
which are slightly inclined toward (internal)-X⋯H-(external)
rather than (internal)-H⋯X-(external) (X = C and S) owing to
their relatively large exposure surfaces that attract the contact
from hydrogen atoms, e.g. for C⋯H, the contact is mainly
concentrated within the aryl ring with a large exposure surface.
Distinctive distributions are noted for each of 3 and 4 owing
1–4. In general, the variation of intermolecular interactions
owing to the differences in molecular packing is reflected in
the fingerprint profiles for the individual molecules, despite
these often being subtle, particularly in the diffuse regions of
the overall profiles as illustrated in ESI† Fig. S4.
The most prominent features of the fingerprint plots are
i e
the pairs of forceps-like spikes tipped at d + d distances
within 1.70–1.76 Å which are much shorter than the sum of
P
van der Waals radii (vdW) for O⋯H [ (vdW)
= 2.61 Å], cf.
O⋯H
the dnorm contact distances listed in ESI† Table S3. These
features arise due to the H⋯O/O⋯H hydrogen bonding
contacts leading to the supramolecular chains. Also
prominent are the pincers-like distributions in the
decomposed fingerprint plots for the H⋯C/C⋯H contacts
with d + d distances in the range of 2.65–2.76 Å which are
to the influence of the Y = Cl and NO substituents, respectively.
i
e
2
P
slightly shorter than
(vdW)C⋯H of 2.79 Å. The H⋯S/ S⋯H
Thus for 3, the contributions are in the order H⋯H (41.3%),
This journal is © The Royal Society of Chemistry 2021
CrystEngComm, 2021, 23, 1723–1743 | 1735

View Article Online
Paper
CrystEngComm
Fig. 9 Decomposed fingerprint plots for 3 delineated into (a) H⋯Cl/Cl⋯H and (b) Cl⋯Cl contacts.
Fig. 10 Relative distribution of different contacts to the Hirshfeld surfaces for individual molecules in 1–4. Other minor but significant contacts
include C⋯Cl/Cl⋯C (2.8%), H⋯N/N⋯H (2.3%) and Cl⋯Cl (1.1%) for 3. For 4a and 4b: H⋯N/N⋯H (2.5%), O⋯C/C⋯O (1.0–2.3%), O⋯S/ S⋯O (1.4–
1
.5%) and O⋯O (0.1–1.3%), and specifically for 4b C⋯C (3.0%).
H⋯C/C⋯H (15.7%), H⋯Cl/Cl⋯H (14.7%), H⋯S/S⋯H (10.8%)
and H⋯O/O⋯H (9.8%) and other minor contacts including the
Cl⋯Cl contact which constitutes only 1.1%. For 4, the order is
H⋯O/O⋯H (ca. 32.2–35.2%), H⋯H (ca. 32.4–33.6%), H⋯C/
C⋯H (ca. 12.2–17.1%), and H⋯S/ S⋯H (ca. 8.5%) followed by
other long contacts (ca. 9.7–10.8%). Similar to 1 and 2,
decomposition of the corresponding contacts exhibits uneven
distributions between the internal and external contacts for
H⋯Cl/Cl⋯H in 3 as well as H⋯O/O⋯H in 4 in addition to the
H⋯C/C⋯H and H⋯S/S⋯H contacts in both molecules, for
which the interactions are inclined toward (internal)-X⋯H-
conducted in an attempt to rank the stabilisation energies
provided by specific contacts in 1–4. The strength of each
interaction as identified from the Hirshfeld surface analysis
was assessed following the approach as detailed in the
Experimental section. As noted from Table 5, among all
pairwise-interactions between molecules, the hydroxyl-O–
H⋯O(hydroxyl)
interactions
provide
the
strongest
interactions with the energy, E , for each pair lying in the
int
−
1
range of −17.81 to −12.09 kcal mol . These strengths are
−
1
comparable to that of ca. −18 and −17 kcal mol as calculated
7
7
for the classical amide-N–H⋯O(amide) and carboxylic-O–
7
8
(external) (X = Cl, O, C and S) indicating the electronegative
H⋯O(carboxylic acid) interactions,
respectively. A close
nature of those acceptor atoms.
inspection of the data shows that Eint(O–H⋯O) is the greatest
in crystal 2, and this is followed by 3, 4 and 1 respectively.
Apart from the O–H⋯O interactions, crystals 1–4 also
feature C–H⋯O and C–H⋯π interactions with the Eint
Interaction energies and energy frameworks
−
1
An analysis of the interaction energies associated with
identified intermolecular contacts was quantitatively
ranging from −2.27 to −10.56 kcal mol and −4.09 to −8.58
−
1
kcal mol , respectively. Additional C–H⋯S and C⋯O
1736 | CrystEngComm, 2021, 23, 1723–1743
This journal is © The Royal Society of Chemistry 2021

View Article Online
CrystEngComm
Paper
−1
Table 5 Corrected interaction energies (kcal mol ) for all intermolecular close contacts present in 1 to 4, with scale factors of 1.057, 0.740, 0.871 and
6
3
0.618 being applied for Eelectrostatic, Epolarization, Edispersion and Erepulsion, respectively, as obtained from the CE-B3LYP/6-31G(d,p) model.
Contact
E
electrostatic
E
polarization
E
dispersion
E
repulsion
E
total
Symmetry operation
1
O1–H1o⋯O22
−16.27
−17.36
−3.11
−3.41
−1.15
−0.35
−1.38
−0.28
−0.94
−2.12
−2.04
−3.73
−9.76
−3.56
−6.08
−1.73
−4.73
−8.97
9.34
9.85
5.29
1.40
4.56
1.46
2.76
4.80
−12.09 x, y, z
O21–H21o⋯O2
−14.65 x, y, 1 + z
−10.56 1 − x, 1 − y, 1 − z
C28–H28b⋯O1/C9–H9a⋯π(C22–C27)/C29–H29a⋯π(C2–C7) −4.93
C5–H5⋯O1
−1.59
−4.85
−1.67
−5.03
−5.48
−4.11
−7.74
−2.20
−7.93
1½ − x, ½ + y, ½ − z
−½ + x, ½ − y, ½ + z
1½ − x, ½ + y, ½ − z
−½ + x, ½ − y, −½ + z
C29–H29b⋯ π(C2–C7)
C4–H4⋯S21
C8–H8a⋯S21/C10–H10a⋯ π(C22–C27)
C21⋯O22
−11.76 1 − x, 1 − y, 1 − z
2
O1–H1o⋯O2
−17.89
−2.75
−2.73
−4.37
−3.71
−0.90
−0.64
−1.49
−6.43
−7.85
−4.91
−7.41
10.22
4.02
3.31
4.70
−17.81 −½ − x, −½ + y, 1½ − z
C6–H6⋯O2/C4–H4⋯C10
C10–H10b⋯π(C2–C7)
C9–H9a⋯π(C2–C7)/C8–H8b⋯π(C2–C7)
−7.48
−4.95
−8.58
½ − x, −½ + y, 1½ − z
−½ + x, ½ − y, −½ + z
−1 + x, y, z
3
O1–H1o⋯O2
−17.13
−4.95
−1.52
−2.85
−1.29
−0.13
−3.91
−0.81
−0.62
−1.22
−0.16
−0.02
−6.45
−7.97
−4.98
−6.85
−1.85
−0.73
10.43
4.67
3.03
3.97
1.29
0.83
−17.09 −½ − x, ½ + y, ½ − z
C6–H6⋯O2/C4–H4⋯C10
C10–H10a⋯π(C2–C7)
C9–H9b⋯π(C2–C7)/C8–H8a⋯π(C2–C7)
C10–H10b⋯Cl1
−9.08
−4.09
−6.69
−2.01
−0.05
½ − x, ½ + y, ½ − z
−½ + x, 1½ − y, −½ + z
−1 + x, y, z
−1½ + x, 1½ − y, −½ + z
2 − x, 1 − y, 1 − z
Cl1⋯Cl1
4
O1–H1o⋯O22
−14.83
−14.45
−0.43
−4.85
−2.58
−2.78
−4.14
−3.79
−3.87
−6.24
−4.00
−4.10
−0.48
−0.78
−0.46
−0.46
−0.88
−0.83
−0.62
−0.87
−7.02
−7.22
−3.98
−2.41
−2.98
−3.02
−7.49
−7.20
−2.17
−10.89
9.70
9.59
2.60
2.42
1.67
1.67
5.29
4.77
1.20
4.46
−16.13 x, 1 + y, z
−16.18 x, y, z
O21–H21o⋯O2
C29–H29b⋯O1
−2.27
−5.62
−4.35
−4.59
−7.24
−7.07
−5.47
−x, 1 − y, 1 − z
C11–H11b⋯O3
x, y, 1 + z
C7–H7⋯O23
x, 1 + y, −1 + z
C27–H27⋯O3
x, y, 1 + z
C30–H30a⋯ π(C2–C7)/C30–H30b⋯S1
C10–H10b⋯ π(C22–C27)/C10–H10a⋯S21
C31–H31a⋯O23
1 − x, 1 − y, 1 − z
1 − x, 1 − y, 2 − z
x, y, −1 + z
π(C22–C27)⋯π(C22–C27)
−13.53 −x, −y, 2 − z
interactions found in 1 have relatively strong interaction
energies with Eint(C–H⋯S) in the range of −2.20 to −7.93 kcal
proportional to the distance of the point charges in accord
with Coulomb's law. Equivalent repulsive forces are not
observed in 1 and 4.
7
9
−
1
mol and Eint(C⋯O) in the range of −4.11 to −11.76 kcal
−
1
mol . The C–H⋯Cl interaction in 3 exhibits a relatively weak
Data in Table 6 indicate that the calculated lattice energies
int of −2.01 kcal mol−1 and the Cl⋯Cl interaction is very
follow the order of 2 (−25.70 kcal mol ) > 1 > 3 > 4 (−23.78
−1
E
−
1
−1
−1
weak with E being close to 0 kcal mol . The only π⋯π
interaction among the series is only present in 4 and gives
rise to a relatively strong Eint of −13.53 kcal mol .
kcal mol ) with the range being about 2 kcal mol . The
order of the lattice energies correlates nicely with the
relatively greater contributions to the energies of stabilisation
provided by the C–H⋯π(aryl) interactions through enhanced
contributions which follow the same order. Further, the
int
−
1
From the data in Table 5, it is evident that the molecular
packing of 1–4 is mainly stabilised by electrostatic forces
owing to the strong O–H⋯O interactions which lead to the
directional topology aligned along the c-axis for 1 and along
b-axes for 2–4, Fig. 11. The crystals also feature dispersive
forces due to the other complementary contacts. The overall
energy framework of 1 has a ladder-like topology in contrast
to the zig-zag, sheet-like energy framework for 4, while
crystals 2 and 3 exhibit a similar, rack-shape topology
consistent with their isostructural relationship. It is
noteworthy that some smaller repulsive forces are observed
within the rack-shaped topology of 2 and 3 owing to the close
proximity of the main electrostatic force resulting from the
O–H⋯O interactions, the magnitude of which is inversely
identified
methylene-C–H⋯Cl(chloride)
and
aryl-C–
H⋯O(nitro) interactions in 3 and 4, respectively, exert little
influence upon the lattice energies due to their weak nature.
By contrast, the energy contribution from the π(aryl)⋯π(aryl)
contact in 4 is significant.
Non-covalent interaction (NCI) plots
5
8
Non-covalent interaction plots were calculated for selected
interactions identified in the crystals of 1–4 to verify the
attractive nature of the interactions through visualisation of a
red-blue-green colour scheme on the iso-surface; red is
This journal is © The Royal Society of Chemistry 2021
CrystEngComm, 2021, 23, 1723–1743 | 1737

View Article Online
Paper
CrystEngComm
Fig. 11 Perspective views of the energy frameworks of 1–4 showing the electrostatic force, dispersion force and total energy diagram. The
cylindrical radius is proportional to the relative strengths of the corresponding energies and was adjusted to the same scale factor of 100 with a
−1
cut-off value of 1.91 kcal mol within 2 × 2 × 2 unit-cells.
indicative of a strong repulsive interaction, blue is indicative
of a strong attractive interaction while green is indicative of a
8
0
weak interaction. The images in Fig. 12 reveal that the
intermolecular hydroxyl-O–H⋯O(hydroxyl) interactions,
−
1
Table 6 The lattice energy (Elattice) in kcal mol and the corresponding
energy components (Eelectrostatic
common in all crystals, along with the intramolecular amide-
N–H⋯O(hydroxyl) and hydroxyl-O–H⋯S(thione) contacts, are
strong and attractive in nature with the low density, low
reduced gradient trough for those interactions lying in the
negative region at about −0.20 to −0.35 a.u. in the respective
two-dimensional NCI plots.
,
polarization repulsion
E , Edispersion and E )
calculated for a cluster of molecules within 25 Å from a reference
molecule through the CE-B3LYP/6-31G(d,p) model
Crystal
E
electrostatic
E
polarization
E
dispersion
E
repulsion
E
lattice
1
2
3
4
−16.19
−14.72
−14.77
−15.01
−4.23
−4.33
−4.29
−3.98
−16.70
−19.30
−18.62
−16.82
12.29
12.65
13.06
12.03
−24.83
−25.70
−24.62
−23.78
Interestingly, for pairs of molecules connected by
hydroxyl-O–H⋯O(hydroxyl) hydrogen bonds in 2–4, there are
1738 | CrystEngComm, 2021, 23, 1723–1743
This journal is © The Royal Society of Chemistry 2021

View Article Online
CrystEngComm
Paper
2
Fig. 12 NCI plots along with the two-dimensional plots of reduced density gradient (RDG) versus sign(λ )ρ(r) for the molecular dimers of 1–4
connected by hydroxyl-O–H⋯O(hydroxyl) hydrogen bonds and highlighting the intramolecular amide-N–H⋯O(hydroxyl) and hydroxyl-O–
H⋯S(thione) hydrogen bonds. The gradient cut-off is set at 0.4 and the colour scale is −0.04 < ρ < 0.04 a.u. Non-essential atoms are truncated
for clarity.
relatively large, diffuse green domains between the
hydroxyethyl-H-and thione-S atoms (see ESI† Fig. S5) that
serve to complement the hydrogen bonds; this is confirmed
in the corresponding QTAIM analysis (see ESI† Fig. S6). The
attractive interactions further strengthen the interactions
leading to the supramolecular chains and are likely
contributors to the reduction in their pitch.
In line with the Hirshfeld surface analysis, additional
contacts are detected in 3, i.e. methylene-C–H⋯Cl(aryl) and
Cl⋯Cl, and 4, i.e. methylene-C–H⋯O(nitro) and aryl-C–
H⋯O(nitro), necessarily absent in 1 and 2. As indicated from
the light-green domains in the respective NCI plots in Fig. 13,
these correspond to weak interactions. Other common
interactions involving the aryl rings and methylene chains are
considered weak and are evidenced through the high-density
localised domains in the NCI plots of ESI† Fig. S5.
8
1
Overview
Crystals of 1 comprise two independent molecules which
associate to form a linear, supramolecular chain sustained,
in part, by hydroxyl-O–H⋯O(hydroxyl) hydrogen bonds. In
Fig. 13 NCI plots for the molecular dimers highlighting weak contacts in 3: (a) methylene-C–H⋯Cl(aryl) and (b) Cl⋯Cl and 4: (c) methylene-C–
H⋯O(nitro) and (d) aryl-C–H⋯O(nitro).
This journal is © The Royal Society of Chemistry 2021
CrystEngComm, 2021, 23, 1723–1743 | 1739

View Article Online
Paper
CrystEngComm
common with isostructural 2 and 3, the space group of 1 is
of systematic studies have been described whereby the
influence of small chemical changes upon supramolecular
monoclinic, P2 /n. There is a simple relationship between the
1
1
0–27
crystals in that the unit-cell edge a in 2 and 3 is about half
that of 1, Table 1. Quite distinct crystal symmetry is noted for
aggregation patterns was investigated.
Except for an
1
0
isostructural series and universal adoption of the carboxylic
1
7
4
, i.e. triclinic, P ¯1 with Z′ = 2. Despite these differences,
dimer synthon in another series, variations in the topology
of the aggregate/chain/layer are the norm as small chemical
changes are made to molecular formulae, even when
comparable hydrogen bonds are evident. Such an observation
confirms that in order to design a crystal let alone an
supramolecular chains featuring hydroxyl-O–H⋯O(hydroxyl)
hydrogen bonds are prominent in the crystals of 1–4.
However, unlike 1, in the isostructural crystals of 2 and 3, the
chains have a helical topology. In 4, zig-zag chains are
apparent. In short, the topology of the supramolecular chains
in 1–4 are syntactic with the crystalline environment in which
8
5
aggregate within a crystal, all supramolecular associations
need to be taken into account.
8
4
they exist.
Finally, it should be mentioned that during the course of
these studies conducted over a period of well over five years
in two different institutions, involving repeated synthesis and
crystallisations, no evidence for polymorphs or solvates was
found. This is not to suggest that different crystalline forms
Isostructural relationships are not uncommon for
molecules differing in a chloro/methyl substituent, which
3
3
have similar volumes, i.e. 19 Å for chloride and 24 Å for
8
2
methyl. Such cases of structural mimicry are commonly
known as the chloro/methyl exchange, but this concept can
8
3
are waiting to be discovered, in accord with the McCrone
1
0
86
be expanded to include other substituents. A chloro/methyl
exchange in isostructural crystals implies similar influences
upon the molecular packing by these substituents. This
suggests that the methylene-C–H⋯Cl(aryl) and Cl⋯Cl
contacts noted in 3, which occur in the inter-layer region, are
not structure-directing even if weakly attractive. The methyl
groups in 2, which are also directed towards the inter-layer
region, do not engage in directional interactions.
axiom.
However, it is stressed that the experimental
conformations in the crystals isolated in this study closely
match the gas-phase optimised geometries suggesting that
the obtained crystals were the thermodynamic forms.
Conclusions
The crystal structure analyses along with detailed analyses of
the supramolecular association in the crystals isolated for
1–4 show the formation of pervasive, hydrogen bonded
chains that differ in topology: linear (1), helical (2 and 3) and
zig-zag (4); no evidence for different crystalline forms was
found in this study. In consideration of the (i) persistence of
the hydroxyl-O–H⋯O(hydroxyl) hydrogen bonds leading to
one-dimensional chains, (ii) the closeness of the calculated
lattice energies, (iii) the relative importance of C–H⋯π(aryl)
contacts operating normal to the chains and (iv) the, on
average, more C–H⋯π(aryl) contacts formed per molecule in
1, compared with 2–4, it is concluded that the different
topologies of the supramolecular chains are related primarily
to the influence of directional C–H⋯π(aryl) contacts. In 1,
where no other directional interactions are apparent, a more
open, linear arrangement facilitates the formation of inter-
chain C–H⋯π(aryl) interactions, by definition occupying a
larger volume of space. In 2 and 3, where C–H⋯π(aryl)
interactions are less dominant, an observation traced to the
Y = Me and Y = Cl substituents, helical chains are apparent.
In 4, where calculations indicate that the aryl rings are
activated, π(aryl)⋯π(aryl) interactions come to the fore and
zig-zag chains are now apparent. Even if remote substituents
do not alter significantly the overall electronic structure of
the molecules, their influence in participating in “second-
tier” supramolecular association can direct the predominant
mode of association between molecules leading to specific
architectures.
In terms of directional interactions, the C–H⋯π(aryl)
contacts featured in each crystal of 1–4, and in 1, lead to a
three-dimensional architecture, in 2 and 3, to a two-
dimensional array and in 4, to a double-layer. Further
directional interactions are largely absent in 2 and 3. By
contrast, methylene-C–H⋯O(hydroxyl) and π(aryl)⋯π(aryl)
interactions assemble the double-layers in 4 (also stabilised
by intra-layer methylene-C–H⋯O(nitro) contacts) into a three-
dimensional arrangement.
At this stage, it is salient to recall the results of the
geometry optimisation calculations which indicate that no
significant influence is exerted by the methyl and chloro
substituents in 2 and 3 but significant activation of the aryl
ring owing to the electronegative nitro group in 4. This latter
observation correlates with the formation of the off-set π(aryl)
⋯
π(aryl) interactions observed in 4 but not in 1–3.
An evaluation of the relative contributions of the different
interaction energies to the overall energies of identified
interactions reveals interesting trends. Thus, the energies
(relative contributions) contributed by the hydroxyl-O–
H⋯O(hydroxyl) hydrogen bonds are approximately −14.72
−
1
−1
and −14.77 kcal mol (46%) for 2 and 3, −15.01 kcal mol
−
1
(
40%) for 4 and −16.19 kcal mol (38%) for 1. This correlates
with the repeat distances of the supramolecular chains with
the greater energy contribution by hydroxyl-O–
H⋯O(hydroxyl) hydrogen bonds to the overall energy of
packing being associated with the chains with the shortest
repeat distances, i.e. 2 (10.75 Å) and 3 (10.77 Å) < 4 (11.21 Å)
In summary, the situation may be envisaged whereby the
molecules precipitate from solution and align to form
supramolecular chains via the predominant hydroxyl-O–
H⋯O(hydroxyl) hydrogen bonds and the adopted topology is
<
1 (16.47 Å).
While an exhaustive literature survey of this phenomenon
is not feasible, as mentioned in the Introduction, a number
1740 | CrystEngComm, 2021, 23, 1723–1743
This journal is © The Royal Society of Chemistry 2021

View Article Online
CrystEngComm
Paper
dictated by the need to optimise the supramolecular
association between chains which, in turn, is moderated by
the specific requirements of the remote substituents.
Bettens, S. K. Seth and E. R. T. Tiekink, CrystEngComm,
2013, 15, 4917–4929.
18 A. P. Voronin, T. V. Volkova, A. B. Ilyukhin, T. P. Trofimova
and G. L. Perlovich, CrystEngComm, 2018, 20, 3476–3489.
19 H. Andleeb, I. Khan, A. Franconetti, M. N. Tahir, J. Simpson,
S. Hameed and A. Frontera, CrystEngComm, 2019, 21,
Author contributions
Sang Loon Tan: data curation; formal analysis; writing –
original draft; writing – review & editing. Edward R. T.
Tiekink: conceptualisation; data curation; formal analysis;
funding acquisition; writing – original draft; writing – review
1
780–1793.
0 P. Mocilac and J. F. Gallagher, CrystEngComm, 2019, 21,
048–4062.
2
2
2
2
2
2
4
1 B. P. A. Gabriele, C. J. Williams, M. E. Lauer, B. Derby and
A. J. Cruz-Cabeza, Cryst. Growth Des., 2020, 20, 7516–7525.
2 C. Weck, E. Nauha and T. Gruber, Cryst. Growth Des.,
&
editing.
Conflicts of interest
2
019, 19, 2899–2911.
3 S. J. Pike, A. Heliot and C. C. Seaton, CrystEngComm,
020, 22, 5040–5048.
There are no conflicts to declare.
2
4 A. Dey, R. K. R. Jetti, R. Boese and G. R. Desiraju,
CrystEngComm, 2003, 5, 248–252.
5 M. D. L. Tonin, S. J. Garden, M. M. Jotani, J. L. Wardell and
E. R. T. Tiekink, Z. Kristallogr. - Cryst. Mater., 2019, 234,
Acknowledgements
Nabihah Al Muna Mohd Nor is thanked for some preliminary
experiments. The authors gratefully acknowledge Sunway
University Sdn Bhd (Grant no. STR-RCTR-RCCM-001-2019) for
support of crystallographic studies.
1
83–200.
2
6 F. Chen, Y. Wang, W. Zhang, T. Tian, B. Bai, H. Wang, F.-Q.
Bai and M. Li, Cryst. Growth Des., 2019, 19, 6100–6113.
7 H. Chung, S. Chen, B. Patel, G. Garbay, Y. H. Geerts and Y.
Diao, Cryst. Growth Des., 2020, 20, 1646–1654.
2
References
1
2
M. C. Etter, Acc. Chem. Res., 1990, 23, 120–126.
G. A. Jeffrey, An Introduction to Hydrogen Bonding; Oxford
University Press: New York, NY, USA, 1997.
T. Steiner, Angew. Chem., Int. Ed., 2002, 41, 48–76.
S. Scheiner, Struct. Chem., 2019, 30, 1119–1128.
G. R. Desiraju, Angew. Chem., Int. Ed. Engl., 1995, 34,
28 D. C. Schroeder, Chem. Rev., 1955, 55, 181–228.
29 J.-P. Griffiths, B. Maliha, M. G. Moloney, A. L. Thompson
and I. Hussain, Langmuir, 2010, 26, 14142–14153.
30 M. Banerjee, R. Karri, A. Chalana, R. Das, R. K. Rai, K. S.
Rawat, B. Pathak and G. Roy, Chem. – Eur. J., 2017, 23,
5696–5707.
3
4
5
2
311–2317.
31 S. M. Lee, A. H. S. Azizan and E. R. T. Tiekink, Molbank,
2018, 2018, M1035.
6
7
8
S. Scheiner, J. Chem. Phys., 2000, 153, 140901.
Q. Li and Z. Li, Acc. Chem. Res., 2020, 53, 962–973.
A. P. Voronin, T. V. Volkova, A. B. Ilyukhin, A. N. Proshin and
G. L. Perlovich, CrystEngComm, 2020, 22, 349–360.
M. Kulułka, M. Srebo-Hooper and M. P. Mitoraj, J. Phys.
Chem. A, 2020, 124, 63–73.
32 G. M. Viana, D. C. Soares, M. V. Santana, L. Henriques do
Amaral, P. W. Meireles, R. P. Nunes, L. C. R. Pereira da Silva,
L. C. de S. Aguiar, C. R. Rodrigues, V. Pereira de Sousa, H. C.
Castro, P. A. Abreu, P. C. Sathler, E. M. Saraiva and L. M.
Cabral, Chem. Pharm. Bull., 2017, 65, 911–919.
33 Agilent Technologies, CrysAlisPro, Santa Clara, CA, USA,
2013.
9
1
1
1
0 A. Deya and G. R. Desiraju, CrystEngComm, 2004, 6, 642–646.
1 D. Das and G. R. Desiraju, CrystEngComm, 2006, 8, 674–679.
2 A. K. Jordaõ, V. F. Ferreira, A. C. Cunha, J. L. Wardell,
S. M. S. V. Wardell and E. R. T. Tiekink, CrystEngComm,
34 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr.,
2008, 64, 112–122.
2
012, 14, 6534–6539.
35 G. M. Sheldrick, Acta Crystallogr., Sect. B: Struct. Sci., Cryst.
Eng. Mater., 2015, 71, 3–8.
36 L. J. Farrugia, J. Appl. Crystallogr., 2012, 45, 849–854.
37 A. L. Spek, Acta Crystallogr., Sect. E: Crystallogr. Commun.,
2020, 76, 1–11.
1
3 S. K. Seth, V. S. Lee, J. Yana, S. M. Zain, A. C. Cunha, V. F.
Ferreira, A. K. Jordão, M. C. B. V. de Souza, S. M. S. V.
Wardell, J. L. Wardell and E. R. T. Tiekink, CrystEngComm,
2
015, 17, 2255–2266.
1
4 L. Chęcińska, A. Jóźwiak, M. Ciechańska, C. Paulmann, J. J.
Holstein, B. Dittrich and M. Małecka, Z. Kristallogr. - Cryst.
Mater., 2018, 233, 675–687.
5 S. Y. Ho, R. P. A. Bettens, D. Dakternieks, A. Duthie and
E. R. T. Tiekink, CrystEngComm, 2005, 7, 682–689.
38 K. Brandenburg, DIAMOND, Crystal Impact GbR, Bonn,
Germany, 2006.
39 T. A. Halgren, J. Comput. Chem., 1996, 17, 490–519.
40 Spartan'16, Wavefunction, Inc., Irvine, CA.
41 Y. Shao, L. F. Molnar, Y. Jung, J. Kussmann, C. Ochsenfeld,
S. T. Brown, A. T. B. Gilbert, L. V. Slipchenko, S. V.
Levchenko, D. P. O'Neill, R. A. DiStasio Jr., R. C. Lochan, T.
Wang, G. J. O. Beran, N. A. Besley, J. M. Herbert, C. Y. Lin, T.
Van Voorhis, S. H. Chien, A. Sodt, R. P. Steele, V. A. Rassolov,
P. E. Maslen, P. P. Korambath, R. D. Adamson, B. Austin, J.
1
1
6 C. R. Kaiser, K. C. Pais, M. V. N. de Souza, J. L. Wardell,
S. M. S. V. Wardell and E. R. T. Tiekink, CrystEngComm,
2
009, 11, 1133–1140.
1
7 A. C. Cunha, V. F. Ferreira, A. K. Jordão, M. C. B. V. de
Souza, S. M. S. V. Wardell, J. L. Wardell, P. A. Tan, R. P. A.
This journal is © The Royal Society of Chemistry 2021
CrystEngComm, 2021, 23, 1723–1743 | 1741

View Article Online
Paper
Baker, E. F. C. Byrd, H. Dachsel, R. J. Doerksen, A. Dreuw,
CrystEngComm
57 C. F. Macrae, I. Sovago, S. J. Cottrell, P. T. A. Galek, P.
McCabe, E. Pidcock, M. Platings, G. P. Shields, J. S. Stevens,
M. Towler and P. A. Wood, J. Appl. Crystallogr., 2020, 53,
226–235.
58 E. R. Johnson, S. Keinan, P. Mori-Sánchez, J. Contreras-
García, A. J. Cohn and W. Yang, J. Am. Chem. Soc., 2010, 132,
6498–6506.
59 W. Humphrey, A. Dalke and K. Schulten, J. Mol. Graphics,
1996, 14, 33–38.
60 S. K. Wolff, D. J. Grimwood, J. J. McKinnon, M. J. Turner, D.
Jayatilaka and M. A. Spackman, Crystal Explorer (Version 17),
University of Western Australia, 2012.
61 S. L. Tan, M. M. Jotani and E. R. T. Tiekink, Acta Crystallogr.,
Sect. E: Crystallogr. Commun., 2019, 75, 308–318.
62 M. A. Spackman and D. Jayatilaka, CrystEngComm, 2009, 11,
19–32.
63 C. F. Mackenzie, P. R. Spackman, D. Jayatilaka and M. A.
Spackman, IUCrJ, 2017, 4, 575–587.
64 M. J. Turner, S. Grabowsky, D. Jayatilaka and M. A.
Spackman, J. Phys. Chem. Lett., 2014, 5, 4249–4255.
65 S. P. Thomas, P. R. Spackman, D. Jayatilaka and M. A.
Spackman, J. Chem. Theory Comput., 2018, 14, 1614–1623.
66 L. Maschio, B. Civalleri, P. Ugliengo and A. Gavezzotti,
J. Phys. Chem. A, 2011, 115, 11179–11186.
67 I. J. Bruno, J. C. Cole, P. R. Edgington, M. Kessler, C. F.
Macrae, P. McCabe, J. Pearson and R. Taylor, Acta
Crystallogr., Sect. B: Struct. Sci., 2002, 58, 389–397.
68 K. R. Koch, C. Sacht and S. Bourne, Inorg. Chim. Acta,
1995, 232, 109–115.
B. D. Dunietz, A. D. Dutoi, T. R. Furlani, S. R. Gwaltney, A.
Heyden, S. Hirata, C.-P. Hsu, G. Kedziora, R. Z. Khalliulin, P.
Klunzinger, A. M. Lee, M. S. Lee, W. Z. Liang, I. Lotan, N.
Nair, B. Peters, E. I. Proynov, P. A. Pieniazek, Y. M. Rhee, J.
Ritchie, E. Rosta, C. D. Sherrill, A. C. Simmonett, J. E.
Subotnik, H. L. Woodcock III, W. Zhang, A. T. Bell, A. K.
Chakraborty, D. M. Chipman, F. J. Keil, A. Warshel, W. J.
Hehre, H. F. Schaefer, J. Kong, A. I. Krylov, P. M. W. Gill and
M. Head-Gordon, Phys. Chem. Chem. Phys., 2006, 8,
3
172–3191.
4
4
2 C. C. J. Roothaan, Rev. Mod. Phys., 1951, 23, 69–89.
3 W. J. Pietro, M. M. Francl, W. J. Hehre, D. J. DeFrees, J. A.
Pople and J. S. Binkley, J. Am. Chem. Soc., 1982, 104,
5
039–5048.
4 J. D. Chai and M. Head-Gordon, Phys. Chem. Chem. Phys.,
008, 10, 6615–6620.
5 G. A. Petersson, A. Bennett, T. G. Tensfeldt, M. A. Al-Laham
and W. A. Shirley, J. Chem. Phys., 1988, 89, 2193–2218.
4
4
4
2
6 C. L. Andersen, C. S. Jensen, K. Mackeprang, L. Du, S.
Jørgensen and H. G. Kjaergaard, J. Phys. Chem. A, 2014, 118,
1
1074–11082.
7 A. D. McLean and G. S. Chandler, J. Chem. Phys., 1980, 72,
639–5648.
8 R. Krishnan, J. S. Binkley, R. Seeger and J. A. Pople, J. Chem.
Phys., 1980, 72, 650–654.
4
4
4
5
9 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci,
G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P.
Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg,
M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M.
Ishida, T. Nakaji-ma, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J.
Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi,
J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R.
Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G.
Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S.
Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J.
Cioslowski and D. J. Fox, Gaussian 16, Gaussian, Inc.,
Wallingford, CT, 2016.
69 L. Hennig, K. Ayala-Leon, J. Angulo-Cornejo, R. Richter and
L. Beyer, J. Fluorine Chem., 2009, 130, 453–460.
70 S. L. Tan, A. H. S. Azizan, M. M. Jotani and E. R. T. Tiekink,
Acta Crystallogr., Sect. E: Crystallogr. Commun., 2019, 75,
1472–1478.
71 S. L. Tan, M. M. Jotani and E. R. T. Tiekink, Acta Crystallogr.,
Sect. E: Crystallogr. Commun., 2020, 76, 155–161.
72 S. A. Bourne, O. Hallale and K. R. Koch, Cryst. Growth Des.,
2005, 5, 307–312.
73 P. Matczak, Computation, 2016, 4, 3.
74 L. Salvatella, Educ. Quim., 2017, 28, 232–237.
75 A. Torrisi, C. Mellot-Draznieks and R. G. Bell, J. Chem. Phys.,
2009, 130, 194703.
76 V. Nikolova, D. Cheshmedzhieva, S. Ilieva and B. Galabov,
J. Org. Chem., 2019, 84, 1908–1915.
5
0 F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys.,
2
005, 7, 3297–3305.
77 S. L. Tan and E. R. T. Tiekink, Acta Crystallogr., Sect. E:
Crystallogr. Commun., 2020, 76, 102–110.
78 S. L. Tan and E. R. T. Tiekink, Acta Crystallogr., Sect. E:
Crystallogr. Commun., 2019, 75, 1–7.
5
5
1 F. Weigend, Phys. Chem. Chem. Phys., 2006, 8, 1057–1065.
2 J. L. Pascual-Ahuir, E. Silla and I. Tuñón, J. Comput. Chem.,
1
994, 15, 1127–1138.
5
5
5
5
3 L. Li, T. Cai, Z. Wang, Z. Zhou, Y. Geng and T. Sun,
Spectrochim. Acta, Part A, 2014, 120, 106–118.
4 E. D. Glendening, A. E. Reed, J. E. Carpenter and F.
Weinhold, NBO Version 3.1.
79 T. Heddle, in Calculations in Fundamental Physics: Electricity
and Magnetism, ed. R. Robinson and N. Hiller, Elsevier,
Amsterdam, 1971, vol. 2, pp. 167–186.
80 J. Contreras-García, E. R. Johnson, S. Keinan, R. Chaudret,
J.-P. Piquemal, D. N. Beratan and W. Yang, J. Chem. Theory
Comput., 2011, 7, 625–632.
5 A. E. Reed, L. A. Curtiss and F. Weinhold, Chem. Rev.,
1
988, 88, 899–926.
6 R. Dennington, T. A. Keith and J. M. Millam, GaussView,
Version 6, Semichem Inc., Shawnee Mission, KS, 2016.
81 T. Lu and F. Chen, J. Comput. Chem., 2011, 33,
580–592.
1742 | CrystEngComm, 2021, 23, 1723–1743
This journal is © The Royal Society of Chemistry 2021

View Article Online
CrystEngComm
Paper
8
2 G. R. Desiraju and J. A. R. P. Sarma, Proc. - Indian Acad. Sci.,
Chem. Sci., 1986, 96, 599–605.
84 D. R. Smyth, B. R. Vincent and E. R. T. Tiekink, Cryst.
Growth Des., 2001, 1, 113–117.
8
3 M. R. Edwards, W. Jones, W. D. S. Motherwell and G. P.
Shields, Mol. Cryst. Liq. Cryst., 2001, 356, 337–353.
85 E. R. T. Tiekink, Chem. Commun., 2014, 50, 11079–11082.
86 J. Haleblian and W. McCrone, J. Pharm. Sci., 1969, 58, 911–929.
This journal is © The Royal Society of Chemistry 2021
CrystEngComm, 2021, 23, 1723–1743 | 1743