Molecular structure of clonidine: Gas-phase electron diffraction, single-crystal X-ray diffraction and quantum chemical studies

Article abstract of DOI:10.1039/c6cp08401j

This study presents the first determination of the molecular structure of the antihypertensive drug clonidine in the gas phase using gas electron diffraction (GED). The refinement was supported by quantum chemical calculations (QCs). The tautomeric and conformational distribution was investigated theoretically, providing an explanation for the presence of the single conformer in the gas phase. The molecular conformation of clonidine has been shown to have a nearly perpendicular arrangement of the phenyl and imidazolidine rings as described by the torsion angle C2-N6-C7-C8 = -72(6)°. The following structural parameters were obtained (bond lengths in Angstroms and bond angles in degrees with 3σ in parentheses): r(C_HH-C_HH) = 1.549(7), r(C_HH-N_H)_av = 1.470(7), r(N_H-C)_av = 1.388(2), r(CN) = 1.286(7), r(C-N) = 1.388(2), r(CzC)_av = 1.403(2), r(C-Cl)_av = 1.737(2); ∠(N_H-C-N_H) = 108.1(11), ∠(C_HH-N_H-C)_av = 109.7(12), ∠(C_HH-C_HH-N_H)_av = 100.9(12), ∠(C-NC) = 122.5(12), ∠(C_ClzCzC_Cl) = 114.9(2), and ∠(C_HzC_ClzC)_av = 123.1(2). The standard enthalpy of formation of clonidine in the gas phase was calculated using G4 theory with both atomisation and isodesmic reaction approaches, yielding the corresponding value of. The molecular structure of clonidine in the solid phase was determined using X-ray diffraction (XRD). Clonidine crystallizes in the monoclinic space group P2₁/c as a twinned crystal. The imino-tautomer, as an equimolar mixture of the two conformers with geometries close to the enantiomeric pair, is present in the solid phase. The identical conformers are linked into centrosymmetric dimers by paired N-H?N hydrogen bonds. The geometries of gaseous and solid clonidine differ especially in the immediate vicinity of the intermolecular hydrogen bonds formed in the crystal.

Full text of DOI:10.1039/c6cp08401j

View Article Online
View Journal
PCCP
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: I. N. Kolesnikova,
A. N. Rykov, I. F. Shishkov, V. Tafeenko and L. A. Aslanov, Phys. Chem. Chem. Phys., 2017, DOI:
10.1039/C6CP08401J.
This is an Accepted Manuscript, which has been through the
Volume 18 Number
1 7 January 2016 Pages 1–636
Royal Society of Chemistry peer review process and has been
accepted for publication.
PCCP
Accepted Manuscripts are published online shortly after
Physical Chemistry Chemical Physics
www.rsc.org/pccp
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9076
PERSPECTIVE
Darya Radziuk and Helmuth Möhwald
Ultrasonically treated liquid interfaces for progress in cleaning and
separation processes
rsc.li/pccp

Page 1 of 19
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP08401J
Molecular structure of clonidine: gas-phase electron diffraction,
single-crystal X-ray diffraction and quantum chemical studies
Inna N. Kolesnikova*, Anatolii N. Rykov, Igor F. Shishkov, Victor A. Tafeenko, Leonid A.
Aslanov
Department of Chemistry, M.V. Lomonosov Moscow State University, 119991 Moscow, Russia
*eꢀmail: kolesnikova@phys.chem.msu.ru
Abstract
This study presents the first determination of molecular structure of the antihypertensive drug
clonidine in the gas phase using gas electron diffraction (GED). The refinement was supported by
quantum chemical calculations (QC). The tautomeric and conformational distribution was
investigated theoretically, providing an explanation for the presence of the single conformer in the
gas phase. The molecular conformation of clonidine has been shown to have a nearly
perpendicular arrangement of the phenyl and imidazolidine rings as described by the torsion angle
C2ꢀN6ꢀC7ꢀC8 = − 72(6)o. The following structural parameters were obtained (bond lengths in
Angstroms and bond angles in degrees with 3σ in parentheses): r(CHH−CHH) = 1.549(7),
r(CHH−NH)av = 1.470(7), r(NH−C)av = 1.388(2), r(C=N) = 1.286(7), r(C−N) = 1.388(2), r(C C)av
= 1.403(2), r(C−Cl)av = 1.737(2); ∠(NH−C−NH) = 108.1(12), ∠(CHH−NH−C)av = 109.7(12),
∠(CHHꢀCHHꢀNH)av = 100.9(12), ∠(C−N=C) = 122.5(12), ∠(CCl C CCl) = 114.9(2),
∠(CH CCl CH)av = 123.1(2). The standard enthalpy of formation of clonidine in the gas phase
was calculated using G4 theory with both an atomisation and isodesmic reactions approach,
yielding the corresponding value of ꢀ f H2o98 (C9H9N3Cl2, g) = 129.2 ± 4.0 kJ·molꢀ1. The molecular
structure of clonidine in the solid phase was determined using Xꢀray diffraction (XRD). Clonidine
crystallizes in the monoclinic space group P21/c as a twinned crystal. The iminoꢀtautomer, as an
equimolar mixture of the two conformers with geometries close to the enantiomeric pair, is
present in the solid phase. The identical conformers are linked into centrosymmetric dimers by
paired NꢀH…N hydrogen bonds. The geometries of gaseous and solid clonidine differ especially
in the immediate vicinity of the intermolecular hydrogen bonds formed in the crystal.
Keywords: clonidine, molecular structure, gas electron diffraction, Xꢀray diffraction, quantum
chemical calculations, standard enthalpy of formation in the gas phase, tautomer, imino, amino,
conformer, enantiomer.
1

Physical Chemistry Chemical Physics
Page 2 of 19
View Article Online
DOI: 10.1039/C6CP08401J
1 Introduction
Clonidine is an antihypertensive drug1, which has been in clinical use for over 40 years2.
Many commercial drugs exist in different tautomeric forms3, thus the identification of the correct
tautomer is an important task for drug discovery programs. Clonidine molecules contain a
guanidine fragment with possible aminoꢀimino tautomerizm, see Fig. 1. In the iminoꢀtautomer (I)
the C=N bond is situated outside the imidazolidine ring (exocyclic), whereas in the iminoꢀ
tautomer (A) it is inside the fiveꢀmembered cycle (endocyclic).
Fig. 1 Tautomeric equilibrium of clonidine.
1
The tautomer distribution of clonidine was previously investigated experimentally by H
and 13C NMR spectroscopy in CDCl3 solution4. According to spectral studies, the imino form is
the predominant tautomer. The conformationalꢀtautomeric composition in the gas phase was
investigated with the use of QC5ꢀ7. The imino form was predicted to be more stable than amino
form by about 1.0 ꢀ 6.6 kcal/mol depending on method and basis applied. However, the fact that
the distribution of conformers and structural parameters of clonidine molecules obtained with the
use of QC cannot be compared to those obtained experimentally in the gas phase due to the
absence of the latter was mentioned7.
In this work GED data was obtained and analyzed for the first experimental determination
of the gaseous molecular structure of clonidine. Conclusions regarding the tautomericꢀ
conformational composition in the GED experiment were drawn using theoretical estimations
based on QC computations. The commonly used, geometrically consistent, rh1 structure8ꢀ9 of
clonidine has been determined on the basis of GED and QC data. Molecular structure of clonidine
in the solid phase has been determined by the means of XRD. To satisfy the need for reliable
thermochemical data, the standard gasꢀphase enthalpy of formation of clonidine was predicted by
QC.
2 Experimental details
2.1 Synthesis and crystal growth of 2-[2,6-dichlorophenylimino]imidazolidine (clonidine)
2

Page 3 of 19
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP08401J
A solution of clonidine hydrochloride (500 mg, 1.88 mmol) in H2O (8 ml) was treated
⁰
dropwise with NaOH (83 mg, 2.1 mmol) in H2O (0.5 ml) at 50 C. The mixture was cooled to
room temperature and the crystals that formed were filtered off, washed with H2O and dried in
o
10
⁰
vacuo. The yield of clonidine was 94 % (400 mg, 1.8 mmol), mp 137ꢀ139 C (lit., 137ꢀ139 C).
A colorless block crystal of clonidine suitable for study by Xꢀray diffraction was
obtained after 2 days of slow evaporation at room temperature of the solution of clonidine (40 mg)
in CHCl3 (0.5 ml) in a preꢀconditioned glass Pyrex ampoule.
A sample of clonidine hydrochloride with purity greater than 99.8% was generously
supplied by the Joint Stock Company Federal Research and Production Center ALTAI and used
for clonidine synthesis with no further purification.
2.2 GED measurements
The GED experiment was carried out using the electron diffraction apparatus EGꢀ100 M at
the M. V. Lomonosov Moscow State University. The diffraction patterns were recorded on photo
films (MACO EMꢀFILM EMS) at short (SD) and long (LD) nozzleꢀtoꢀfilm distances at 420 K.
The electron wavelength was calibrated against gaseous CCl4. The differences between
wavelengths of electrons at the beginning and at the end of each of two experiments (SD and LD)
were less than 0.05%. The diffraction patterns were scanned using an Epson Perfection 4870
photo scanner. Transformation of scanned images into total intensity curves I(s) and determination
of wavelengths were done using the UNEX11 program. A summary of the experimental conditions
is given in Table 1.
Table 1 Conditions of the GED experiment.
SD
LD
Nozzleꢀtoꢀplate distance, [mm]
Nozzle temperature, [K]
Acceleration voltage, [kV]
Electron beam current, [ꢁA]
Electrons wavelength, [Å]
s rangea, [Åꢀ1]
193.9
419
362.3
421
60
60
4.5
4.5
0.050388
6.8–32.8
3.3·10ꢀ3
135
0.049988
3.4–17.8
4.0·10ꢀ3
95
Residual gas pressure, [Pa]
Exposure time, [s]
a s =4πλꢀ1sinθ/2, where θ is the scattering angle and λ is the electron wavelength
The total intensity curves with the final backgrounds are shown in Fig. 2. Numerical values
of the intensities and backgrounds are available in section 1 of the ESI.
3

Physical Chemistry Chemical Physics
Page 4 of 19
View Article Online
DOI: 10.1039/C6CP08401J
I(s)
LD
I(s)
SD
0
5
10
15
20
25
30
35
s, Åꢀ1
Fig. 2 Full experimental intensity (solid line) and background curves (dashed line) for clonidine.
2.3 XRD measurements
The Xꢀray diffraction experiment was carried out using the STOE STADI VARI diffractometer
with Pilatus100K detector at the M. V. Lomonosov Moscow State University. The intensity data
was collected by rotation method mode with multilayer focusing mirror collimation at room
temperature. Data collection, image processing, cell refinement and data reduction were carried
out using the STOE XꢀArea 1.67 software12. Integration and scaling of the intensity data were
accomplished using the program LANA12. The crystal data that was automatically measured on
the diffractometer, the refined cell constants and the conditions of the Xꢀray diffraction
experiment are listed in Table 2.
Table 2 Data collection statistics and conditions of the Xꢀray diffraction experiment.
Molecular formula
C9H9Cl2N3
Molecular weight [g/mol]
Crystal size [mm]
230.09
1.0×0.25×0.12
Crystal system
Monoclinic
Space group
P21/c
Cell dimensions: a, b, c [Å]
α, β, γ [o]
Unit cell volume [Å3]
Electrons wavelength [Å]
No. of formula units per unit cell, Z
Density (calculated) [g/cm3]
17.8433(1), 8.2767(3), 15.1223(6)
90, 115.072(1), 90
2022.88(11)
1.54186
8
1.511
4

Page 5 of 19
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP08401J
F(000)
944
Radiation type
CuꢀKα
Absorption coefficient, ꢀ [mmꢀ1]
θ range [o]
5.463
5.040 – 72.765
67.686 – 100.000
ꢀ18 – 22, ꢀ7 – 10, ꢀ18 – 14
15511
Completeness to θ [%]
Limiting indices: h, k, l
No. of reflections measured
No. of independent reflections
Temperature [K]
3994
295(2)
Rint
0.0490
3 Theoretical calculations
3.1 General computational considerations
All the QC were performed using the Gaussian03 program package13. Computational
methods B3LYP14,15 and MP216 as well as basis sets 6ꢀ31G(d,p)17 and ccꢀpVTZ18 were used as
implemented in this program. The geometries of the conformers were optimized using the opt =
tight and int = ultrafine parameters. Quadratic force fields and vibrational frequencies of the
optimized species were calculated at B3LYP/6ꢀ31G(d,p), B3LYP/ccꢀpVTZ, and MP2/ccꢀpVTZ
levels of theory, using the scale factors γ = 0.961, γ = 0.965, and γ = 0.95019, respectively, since γ
was designed to the satisfy the equation vexp = γvcalc, where v is the experimental and commutated
harmonic frequencies. To correct for systematic error, the force elements of the quadratic force
field were modified as fij = γ2fij20. The interatomic distance corrections (ꢂrh1ꢀa = rh1ꢀra) and mean
vibrational amplitudes (l) were obtained using the VibModule program21.
The gasꢀphase enthalpy of formation of clonidine was calculated using the Gaussianꢀ4
(G4) method22 applied to both the atomisation reaction23 and isodesmic reactions24. Although the
G4 method applied to atomisation reactions was found to be accurate in calculating the enthalpies
of formations of most classes of organic compounds22, in some cases it underestimates the
corresponding values by up to 20 kJ·molꢀ1 25, 26. To check the accuracy of ꢀ f H2o98 (g) values
obtained from atomisation energies, the method of isodesmic reactions was also used to calculate
the enthalpy of formation of clonidine.
3.2 Computation of clonidine conformers
Ab initio computations identified the existence of one conformer for the iminoꢀtautomer
(I) and three conformers for aminoꢀtautomer (A1, A2 and A3), see Fig. 3. After the geometry
5

Physical Chemistry Chemical Physics
Page 6 of 19
View Article Online
DOI: 10.1039/C6CP08401J
optimizations, subsequent frequency calculations confirmed that the structures obtained are local
minima as no negative eigenvalues were found in the Hessian matrix, thus demonstrating that no
imaginary frequencies were present. The optimized Cartesian coordinates of all clonidine
conformers at B3LYP/6ꢀ31G(d,p), B3LYP/ccꢀpVTZ, and MP2/ccꢀpVTZ levels of theory are
available in section 2 of the ESI.
Fig. 3 Tautomers of clonidine – I is the imino tautomer, A1, A2, and A3 are conformers of the
amino tautomer.
The thermochemial analysis of clonidine conformers was carried out 420 K (the average
temperature of the GED experiment). Results of the MP2/ccꢀpVTZ calculations are summarized in
Table 3. The corresponding data for B3LYP/6ꢀ31G(d,p) and B3LYP/ccꢀpVTZ can be found in
section 2 of the ESI. The iminoꢀtautomer (I) was found to be the lowest in energy. The energy
difference between conformer I and the next lowest conformer (A1 according to MP2/ccꢀpVTZ)
was found to be 8.1 kcal·molꢀ1, thus it can be confidently assumed that only the iminoꢀtautomer is
present in the gas phase. The same conclusion may be drawn from the B3LYP/6ꢀ31G(d,p) and
B3LYP/ccꢀpVTZ results (see section 2 of the ESI).
6

Page 7 of 19
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP08401J
Table 3 Equilibrium distribution of clonidine conformers at 420 K according to the MP2/ccꢀpVTZ
calculations.
Conformer
ꢂEe
ꢂ
H
S420
p420
I
0.0
8.1
8.8
8.7
0.0
7.9
9.0
8.4
135.6
137.1
137.5
136.8
100
A1
A2
A3
0
0
0
ꢂ
Ee (kcal·molꢀ1) is the relative electronic energy;
H
=
Ee
+
ZPE (kcal·molꢀ1), where ZPE is the
zero point energy correction calculated from scaled vibrational frequencies; S420 (cal/(K·mol)) is
the entropy value at 420 K obtained using scaled frequencies including the entropy of mixing
(
ꢂ
/
R
H
ln2), since all conformers present as an equimolar mixture of the two optical isomers; ꢂGT
) + ST ) = [exp(–ꢂGT
) is the Gibbs free energy difference (kcal·molꢀ1); pT
RT)]/Σ[exp(–GT )/RT)] is the mole fraction of conformers (%).
(i) =
(i)
(i
T
ꢂ
(
(i
i
(i
3.3 Computation of clonidine standard enthalpy formation in the gas phase
The clonidine gasꢀphase enthalpy of formation was estimated, taking into account the
computed enthalpies of reactions described by equations (1) – (3) (Table 4) and the experimental
enthalpies of formation in the gas phase of the other atoms and molecules involved.
For the atomisation reaction procedure [Equation (1)] the experimental enthalpies of
formation of the gaseous atoms at
T = 0 K and the corresponding thermal corrections for the
elements in their standard states were taken from the reference book27. The isodesmic reactions
[Equations (2) and (3)] were chosen depending on the availability of reliable experimental
thermochemical data of the relevant compounds: 2,5ꢀdichloroaniline28, imidazole29, H2O30 and
imidazolidinꢀ2ꢀone31.
The G4 energies of the molecules involved in Equations (1) ꢀ (3) were calculated for the
lowestꢀenergy conformers. Geometry optimizations and conformational analyses for the flexible
molecules were performed at the B3LYP/6ꢀ31G(d,p) level of theory. To characterize the
optimized stationary points, the harmonic vibrational frequencies were calculated at the same
level. The resulting geometries were used as inputs in the G4 calculations.
The standard enthalpy of formation of gaseous clonidine, ꢀ f H2o98 (C9H9N3Cl2, g), was
obtained through both the atomisation and isodesmic reaction procedures and is given in Table 4.
Table 4 Enthalpy of formation of clonidine calculated from atomisation reаction and isodesmic
reactions using G4 energies (in kJ·molꢀ1)
reaction
ꢀ f H2o98 (g)
atomisation reaction
7

Physical Chemistry Chemical Physics
Page 8 of 19
View Article Online
DOI: 10.1039/C6CP08401J
C9H9N3Cl2(g)(clonidine) → 9C(cr)+4.5H2(g)+1.5N2(g)+Cl2(g)
(1)
(2)
(3)
125.9
isodesmic reactions
Cl
Cl
NH
NH
NH
N
NH2
+
N
Cl
Cl
129.8
clonidine
2,5ꢀdichloroaniline imidazole
Cl
Cl
O
NH
NH
N
H2O
NH2
HN
NH
+
+
Cl
Cl
131.9
clonidine
2,5ꢀdichloroaniline imidazolidineꢀ2ꢀone
average of all reactions :
129.2
4 Structural analysis
4.1 GED structural analysis
The GED data set was analyzed by applying the leastꢀsquares method to the molecular
intensities using the UNEX program. The structure was refined in terms of rh1 parameters. The
functional that was minimized in the structural refinement had a standard form32.
sM(s)
LD
SD
ꢂsM(s)
0
5
10
15
20
25
30
35
s, Å ꢀ1
Fig. 4 Experimental (open circles) and theoretical (solid line) reduced molecular scattering
intensity sM ) for clonidine; corresponding difference curves ꢂsM ).
(s
(s
Starting values for the geometric parameters for the GED analyses were taken from the
B3LYP/6ꢀ31G(d,p) calculations. The parameters that were used as independent variables are listed
8

Page 9 of 19
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP08401J
in Table 5. To reduce the correlation between very similar parameters, their values were refined in
groups as indicated in Table 5, thus keeping their differences within the same group constant. The
estimated error limits were taken as three times the standard deviations (3
calculations.
σ) from the leastꢀsquares
Starting values of the vibrational amplitudes, together with harmonic vibrational
corrections to the internuclear distances, were calculated with the VibModule program using
structural parameters and the quadratic field obtained at the B3LYP/6ꢀ31G(d,p) level of theory.
The vibrational amplitudes were refined in groups whereas the amplitudes assigned to closelyꢀ
spaced distances were linked to maintain constant differences. The constant differences were
selected on the basis of the calculated values. The amplitudes for the hydrogenꢀcontaining terms
were not refined and were kept constant at the calculated values. The estimated error limits were
taken as three times the standard deviations (3σ) from the leastꢀsquares calculations. The values of
the predicted and refined vibrational amplitudes from the harmonic approximation are listed in
section 1 of the ESI.
The final geometrical parameters obtained from the GED analysis can be found in Table 5.
A final
Rꢀfactor of 4.6 % was obtained. The overall fit of the model to the data can be seen from
Fig. 4. The full results of the refinement (Cartesian coordinates and the correlation matrix) can be
found in section 1 of the ESI. The radial distribution curves
presented in Fig. 5.
f(r) with difference curves are
Table 5 Geometrical parameters of clonidine obtained by gasꢀphase electron diffraction, Xꢀray
diffraction and quantum chemical calculations.
Parametersa
GED
XRD
B3LYP/ B3LYP
MP2
I
I
IA
6ꢀ31G(d,p) /ccꢀpVTZ /ccꢀpVTZ
Independent parameters
r(N1–C2)
r(C2–N3)
r(N3–C4)
r(N1–C5)
r(C4–C5)
1.390(2)b
1.385(2)b
1.468(7)c
1.472(7)c
1.549(7)c
1.379(6)
1.354(5)
1.435(9)
1.476(6)
1.386(6)
1.335(6)
1.489(8)
1.465(6)
1.387
1.382
1.459
1.463
1.540
1.383
1.378
1.458
1.462
1.535
1.383
1.379
1.456
1.459
1.526
1.531(10) 1.509(10)
r(C2=N6)
r(N6–C7)
1.286(7)d
1.388(2)e
1.291(6)
1.396(7)
1.303(6)
1.427(7)
1.286
1.387
1.285
1.385
1.278
1.381
r
r
r
r
r
r
(C7 C8)
(C8 C9)
(C9 C10)
(C10 C11)
(C11 C12)
(C12 C7)
1.417(2)e
1.396(2)e
1.395(2)e
1.397(2)e
1.394(2)e
1.418(2)e
1.397(9)
1.381(8)
1.356(13) 1.375(14)
1.374(11) 1.374(12)
1.393(10) 1.399(10)
1.412(10) 1.398(10)
1.398(9)
1.383(9)
1.414
1.393
1.392
1.394
1.391
1.415
1.409
1.392
1.391
1.392
1.391
1.407
1.408
1.387
1.387
1.389
1.385
1.410
r(N–H)av
1.03(3)f
0.91(5)
1.07(7)
1.010
1.009
1.006
9

Physical Chemistry Chemical Physics
Page 10 of 19
View Article Online
DOI: 10.1039/C6CP08401J
r(C–H)av
1.102(10)g
0.950n
0.950n
1.096
1.092
1.090
r
(C8–Cl19)
1.741(2)h
1.748(5)
1.726(8)
1.741(2)
1.719(7)
1.765
1.757
1.759
1.750
1.737
1.729
r(C12–Cl23)
1.733(2)h
∠
∠
(N1ꢀC2ꢀN3)
(C2ꢀN3ꢀC4)
108.1(11)i
110.7(11)i
108.0(4)
111.1(5)
108.6(4)
110.8(4)
107.5
110.1
107.5
110.2
107.7
109.0
∠
∠
∠
(N3ꢀC2ꢀN6)
(C2ꢀN6ꢀC7)
(N6ꢀC7ꢀC8)
122.9(12)j
122.5(12)j
124.4(12)j
122.6(4)
120.4(5)
126.4(5)
121.4(4)
118.8(5)
124.9(5)
122.5
122.1
124.0
122.7
122.9
123.8
123.1
117.3
123.1
∠
∠
∠
(C7ꢀC8ꢀC9)
(C8ꢀC9ꢀC10)
(C9ꢀC10ꢀC11)
123.0(2)k
119.6(2)k
119.8(2)k
123.9(6)
119.6(5)
119.6(5)
122.4(5)
119.0(5)
121.7(6)
123.0
119.6
119.8
122.9
119.6
119.7
122.7
119.3
120.1
∠
∠
(C7ꢀC8ꢀCl19)
(C7ꢀС12ꢀCl23)
118.5(22)l
118.0(22)l
119.0(5)
118.5(4)
120.8(5)
118.9(4)
119.0
118.6
119.1
118.5
119.1
118.7
τ
τ
(C7ꢀC8ꢀC9ꢀC10)
ꢀ0.1m
0.7(9)
0.5(7)
ꢀ2.2(9)
ꢀ0.9(8)
ꢀ0.1
0.0
0.0
ꢀ0.1
0.0
ꢀ0.2
(C8ꢀC9ꢀC10ꢀC11) 0.0m
τ(N1ꢀC2ꢀN3ꢀC4)
τ(C4ꢀN3ꢀC2ꢀN6)
τ(C2ꢀN6ꢀC7ꢀC8)
τ(N6ꢀC7ꢀC8ꢀC9)
ꢀ11.8m
ꢀ4.5(6)
5.8(6)
ꢀ11.8
166.6
ꢀ73.8
ꢀ174.0
ꢀ11.6
166.7
ꢀ73.3
ꢀ174.1
ꢀ12.9
165.6
ꢀ73.5
ꢀ175.0
166.6m
171.5(5)
ꢀ58.0(9)
ꢀ170.7(5)
60.4(8)
ꢀ72(6)
ꢀ174.0m
ꢀ178.3(6) 175.2(6)
Dependent parameters
∠
∠
∠
(N3ꢀC4ꢀC5)
(N1ꢀC5ꢀC4)
(C2ꢀN1ꢀC5)
99.9(12)
101.9(12)
108.6(12)
102.6(5)
99.8(5)
109.1(4)
100.3(5)
102.0(4)
108.3(4)
100.7
101.0
109.6
100.9
101.1
109.8
100.2
100.4
108.6
∠
(N1ꢀC2ꢀN6)
129.0(11)
129.1(4)
130.0(4)
129.9
129.7
129.2
∠
∠
∠
(C10ꢀС11ꢀC12)
(C11ꢀC12ꢀC7)
(C8ꢀC7ꢀC12)
119.5(2)
123.2(2)
114.9(2)
121.0(7)
121.1(6)
114.7(4)
118.0(7)
122.7(6)
116.1(4)
119.7
122.9
115.1
119.8
122.8
115.1
119.5
122.6
115.8
τ(C2ꢀN3ꢀC4ꢀC5)
27.3(12)
22.0(6)
ꢀ23.2(6)
27.8
27.1
31.2
RL
RS
Rtot
3.9
5.3
4.6
a Bond lengths are in Å and bond angles are in degrees.
The factors (in %) are given for the two data sets combined (Rtot) and for the two data sets
R
separately as well, for the long (RL) and short (RS) camera distances.
b
Refined in one group. Differences between parameters in the group were assumed at the values
of B3LYP/6ꢀ31G(d,p) calculations.
c, d, e, f, g, h, i ,j, k ,l See Footnote b.
10

Page 11 of 19
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP08401J
m Assumed at the values of B3LYP/6ꢀ31G(d,p) calculations
n Assumed at the values of SHELX calculations.
f(r)
C12–Cl23
C8–Cl19
С2
C6
–
–
N3
N7
N3
N1
–
C4
C5
–
С2=N6
ꢂf(r)
C
C C4–C5
0
1
2
3
4
5
6
7
8
r, Å
Fig. 5 Experimental (dots) and theoretical (solid line) radial distribution curves
f
(r
) with difference
curve ꢂf(r) for clonidine. The distribution of nonꢀhydrogen distances is indicated by vertical bars.
4.2 XRD structural analysis
The crystal structure was solved by direct methods applying the SHELX33 program.
Hydrogen atoms attached to carbon were placed in calculated positions and allowed to ride on
their parent atoms [CꢀH 0.93 – 0.98; Uiso = 1.2 Ueq (parent atom)], whereas those attached to
nitrogen were located in difference maps and refined freely. All nonꢀhydrogen atoms were
modeled with anisotropic displacement parameters. The refinement proceeded by fully matrix
leastꢀsquares against
6).
F
2. Two molecules of clonidine were found in the asymmetric unit (see Fig.
11

Physical Chemistry Chemical Physics
Page 12 of 19
View Article Online
DOI: 10.1039/C6CP08401J
Fig. 6 The asymmetric unit of clonidine molecule. Displacement ellipsoids are drawn in
50% probability level and H atoms are shown as small spheres of arbitrary radii.
However, the refinement converged to a rather high R1ꢀfactor of 18 %, and in the list of the
'all most disagreeable reflections' FO2 was much greater than FC2 . Thus, it seemed reasonable that
this crystal structure was twinned. The twin law was found using the program PLATON34. The
refinement applying the twin law (TWIN ꢀ10ꢀ1 0ꢀ10 001) and an initial fraction (BASF) of 0.50
lead to an R1ꢀfactor of 5.2 % and to a BASF of 0.45. However, the difference Fourier analysis
indicated the existence of high residual (0.9 Åꢀ3) electron density localization at a distance of 1.8
Å from carbon atom C11 (see Fig. 6). Further refinement with an assumption that this position
could be attributed to a chlorine atom lead to a final conventional R1ꢀfactor of 4.0 %. Thus, it was
found that the clonidine sample obtained in this work contains its structural isomer 2ꢀ[2,5ꢀ
dichlorophenylimino]imidazolidine in the amount of 4 – 5 %. Final refinement statistics are given
in Table 6. CCDCꢀ1512005 contains the supplementary crystallographic data. These data can be
obtained free of charge from the Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Table 6 Refinement statistics.
Data, restrains, parameters
3994, 4, 278
0.0718
0.1089
0.0402
0.0937
0.894
Final R1 values (all data)
Final w
Final R1 values (Io > 2
Final wR
2) values (Io > 2
Goodness of fit against F2
R
(F
2) values (all data)
Io))
σ(
(F
σ(Io))
Largest diff. peak and hole [e/Å ꢀ3]
Largest diff. hole [e/Å ꢀ3]
0.300
ꢀ0.243
5
Results and discussion
It has been established by means of both GED and QC that gaseous clonidine prefers to
adopt a structure with the exocyclic double bond (imino form). The enhanced stability of the
imino tautomer is attributed to the delocalization of the exocyclic N lone pair and C=N pꢀorbital
into the phenyl ring πꢀsystem35,36. This delocalization is additionally stabilized by the ortho
disubstitution of the phenyl ring with σꢀaccepting chlorine atoms. The imidazolidine and aromatic
rings adopt an approximately perpendicular arrangement [τ
(C2ꢀN6ꢀC7ꢀC8) = ꢀ 72(6)o]. It appears
that the bulky chlorine atoms force steric repulsion and make the imidazolidine ring stay out of the
12

Page 13 of 19
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP08401J
plane defined by the diclorophenyl ring. However, it is not only the steric nature of the
substituents on the phenyl ring but also their electronic nature and the electronic nature of nitrogen
atoms of imidazolidine moiety which influence the overall ring conformation. According to
MP2/ccꢀpVTZ calculations, the Mulliken effective electron charges of both the chlorine and
nitrogen atoms are negative [N1 (ꢀ0.23 e), N3 (ꢀ0.22 e), Cl19 (ꢀ0.16 e), Cl23 (ꢀ0.15 e)]. Thus,
electronic repulsion of the chlorine and endocyclic nitrogen atoms also forces the phenyl and
imidazolidine rings to adopt an almost orthogonal orientation. One of the chlorine atoms of the
dichlorobenzene ring forms an intramolecular hydrogen bond with the nearest amine hydrogen of
imidazolidine ring (Cl19…H distance is 2.872 Å).
According to the GED data (Table 5) the dichlorobenzene ring of clonidine is planar and
the imidazoline ring is slightly bent. The bond lengths and angles within the imidazoline ring are
similar to those observed in ureido rings37, with the short N1–C2 and C2–N3 distances indicative
of doubleꢀbond character. The endocyclic N lone pairs are involved in conjugation with the C=N
pꢀorbital, and the sp2 hybridized carbon atom of the C=N bond is situated in the plane N1ꢀN6ꢀN3
within the GED limit of error.
The average value of the С С bond [1.403(2) Å] is typical for aromatic compounds. The
benzene ring angles differ substantially from 120°, which is a common feature of substituted
benzenes38. Thus, the angles with the central atom bonded to chlorine atoms have an average
value of 123.1(2)°, in agreement with the chlorine atoms being σꢀacceptors39, while the angle of
the imidazolidine ring is reduced to 114.9(2)° due to the πꢀaccepting C=N moiety39. The
deformation of the benzene ring in the clonidine molecule is bigger than that of the relative
monosubstituted benzenes (e.g. αipso = 121.7(2)° in chlorobenzene40 and αipso = 118.9(4)° in
Nꢀ
benzilidene aniline41) as the effects of different substituents are superimposed. The average length
of the C–Cl bond in clonidine [1.737(2) Å] is almost the same as in chlorobenzene [1.739(2) Å40].
The structural parameters obtained by GED are in a good agreement with the QC (see
Table 5). It is worth noting that values of the bond lengths are better predicted by the B3LYP/6ꢀ
31G(d,p) method especially for C4–C5 (see Table 5) than by MP2/ccꢀpVTZ. However, the lengths
of the C–Cl bonds are quite accurately predicted by MP2/ccꢀpVTZ, while the B3LYP/6ꢀ31G(d,p)
method overestimates the distance by 0.02 Å. The values obtained with the use of B3LYP/ccꢀ
pVTZ are somewhere in between those obtained with the B3LYP/6ꢀ31G(d,p) and MP2/ccꢀpVTZ
calculations. The angles predicted by QC agree with the GED results in the GED limit of error.
The XRD study of single crystal of clonidine at 295 K revealed that clonidine crystallizes
in the monoclinic space group
P21/c in a twinned crystal as the iminoꢀtautomer with two
molecules in the asymmetric unit (see Fig. 6). The geometry of two molecules is close to the
13

Physical Chemistry Chemical Physics
Page 14 of 19
View Article Online
DOI: 10.1039/C6CP08401J
geometry of the enantiomeric pair (see Table 5). The imidazolidine ring exists in a halfꢀchair
conformation. The dihedral angle between the planes of imidazolidine and the aromatic rings is ꢀ
58.0(9)° (conformer I) in one of the conformers and 60.4(8)° in the other one (conformer IA).
Two types of centrosymmetric dimers linked by paired N–H…N hydrogen bonds are
observed in the crystal: the first one is formed by conformer I and its optical isomer, the second
one is formed by conformer IA and its optical isomer (see Fig. 7).
Fig 7 Part of the crystal structure showing the formation of centrosymmetric dimers.
Table 7 Hydrogen bonding in clonidine.
D–H...A a
D–H
H...A
D...A
∠D–H...A
N3–H3...N6i
N3A–H3A...N6Aii
0.91(5)
1.07(7)
2.09(5)
1.93(7)
2.977(7)
3.000(6)
166(4)
175(5)
a Bond lengths are in Å and angles are in degrees.
i Symmetry transformations used to generate equivalent atoms: ꢁx+2
ii Symmetry transformations used to generate equivalent atoms: ꢁx+1
,
ꢁy+1,ꢁz+1
ꢁy+1 ꢁz+1
,
,
As can be seen from Table 7 the intermolecular hydrogen bonds in the first type of dimers
seem to be weaker than in the second type with the H...N distances being about 0.26 Å longer and
N–H Å bond lengths being about 0.16 Å shorter in the first type of dimer than in the second one.
The distances between the N3 atom of one molecule and the N6 atom of the other molecule in the
first type of dimer is smaller than the corresponding distances in the second type of dimer by only
about 0.02 Å.
Stronger intermolecular hydrogen bonds in the second type of dimer seem to be the reason
why the structure of conformer IA is more distorted in comparison with the molecular structure of
gaseous clonidine than the structure of conformer I (see Table 5). Thus, the C2–N3 bond of
conformer IA in the solid phase is shorter by about 0.05 Å than in the gas phase, whereas for
conformer I it is shorter only by 0.03 Å. The C2–N6 and N6–C7 bonds of conformer IA are
14

Page 15 of 19
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP08401J
longer by about 0.02 Å and 0.04 Å in the solid than in the gas phase, whereas the corresponding
bonds of conformer I are almost equal in the solid and gas phases. In the crystal, as in the gas
phase, one of the chlorine atoms of the dichlorobenzene ring forms an intramolecular hydrogen
bond with the nearest amine hydrogen of imidazolidine ring (the Cl19…H distance is 2.58 Å in
conformer I and 2.63 Å in conformer IA).
The dihedral angle between ring planes in clonidine (τ(C2ꢀN6ꢀC7ꢀC8)) closes in
appreciably from ꢀ72(6)° in the gaseous molecule to ꢀ58.0(9)° for conformer I and to 60.4(8)° for
the conformer IA in the crystal. For the crystal structure the relative decrease of the corresponding
dihedral angle in comparison with the gas phase may be caused by the effects of molecular
packing and due to formation of dimers via hydrogen bonding.
Though structural comparisons of solid clonidine with the gaseous molecule are
complicated by difference of temperature and technique, with Xꢀray diffraction, unlike electron
diffraction, measuring distances between centers of maximum electron density, in general it can
be summarized that the others parts of in structures of gaseous and solid are very similar (see
Table 5).
It is worth noticing that the XRD study discovered that the clonidine sample obtained in
this work also contains 2ꢀ[2,5ꢀdichlorophenylimino]imidazolidine, a structural isomer of
clonidine, in the amount of approximately 4 – 5 %, which is too small to be detected by the GED
study. This compound could be formed either by migration of one of chlorine atoms to the
neighbouring carbon atom during the process of hydrolisation of clonidine hydrochloride or 2ꢀ
[2,5ꢀdichlorophenylimino]imidazolidine hydrochloride might be present in the parent sample of
clonidine hydrochloride.
To reveal how the protonation of parent drug molecules changes their geometry the
molecular structure of clonidine base is compared with its crystalline salt40. As can be seen from
Table 8, the geometry of the free base and its salt differ in the vicinity of the bridging nitrogen
atom which is protonated in clonidine hydrochloride. Thus, the exocyclic Cim–N bond of the
protonated clonidine is longer by about 0.03 Å, whereas the endocyclic C–NH are shorter by 0.04
Å, respectively, indicating the loss of double bond character in the former and gain of double bond
character in the latter. The deformation of the benzene ring is smaller in clonidine hydrochloride
due to the loss of double bond character of the Cim–N bond. The pyramidal endocyclic N atoms in
the clonidine base adopt an equivalent planar geometry in clonidine hydrochloride. Both structures
contain a perpendicular orientation of aryl and imidazolidine rings, though the torsional angle in
clonidine base is about 14° less than in the solid phase.
15

Physical Chemistry Chemical Physics
Page 16 of 19
View Article Online
DOI: 10.1039/C6CP08401J
Table 8 Selected structural parameters of clonidine base and clonidine hydrochloride in
the solid phase.
c
Parametera
Crystal b
(clonidine)
Crystal rα(∠ )
α
(clonidine·HCl)
r
r
r
r
r
r
r
(CHH–CHH)
(NH–CHH)av
(C–NH)av
(Cim–N)
(N–CAr)
(C C)av
(C–Cl)av
1.509(10)
1.477(8)
1.361(6)
1.303(6)
1.427(7)
1.386(13)
1.730(2)
1.540(3)
1.450(3)
1.321(3)
1.332(3)
1.419(3)
1.389(3)
1.732(3)
∠
∠
∠
(CHH–CHH–NH)av
(CHH–NH–C)av
NH–C–NH
102.0(4)
108.3(4)
108.6(4)
102.8(2)
111.5(2)
111.5(2)
∠
∠
∠
CAr–N–Cim
(CCl CN CCl)
(CH CCl CH)av
118.8(5)
116.1(4)
122.6(5)
122.6(2)
117.3(2)
121.5(2)
τ
(imidazolidineꢀdiclorphenyl)
60.4(8)
74.5(14)
a Bond lengths are in Å and bond angles are in degrees.
b Present work, molecular parameters are given for conformer IA
c Xꢀray crystallography42.
6. Conclusions
This theoretical study set out to determine tautomeric equilibria, stable conformations and
gasꢀphase enthalpy of formation of clonidine for which a relatively small amount of experimental
physicochemical data previously existed, considering its pharmacological importance. Using the
results of GED and XRD structural analysis the following conclusions can be drawn.
(1) In has been shown that clonidine is present only as iminoꢀtautomer both in the gaseous
and solid phases.
(2) In the gas phase clonidine exists as a mixture of two optical isomers. Surprisingly, the
enantiomers of gaseous clonidine noticeably changed their molecular structures in the crystal, the
conformers were found to be linked into the centrosymmetric dimers by paired NꢀH…N hydrogen
bonds. The hydrogen bonds in the one type of dimers seem to be stronger than in the other one and
thus have a stronger influence on the corresponding molecular structures.
(3) A structural comparison of the clonidine base with its salt structure is established.
Protonation occurs at the imino Nꢀatom of 2ꢀiminoꢀimidazolidine group and results in clear
changes to the neighbouring distances.
16

Page 17 of 19
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP08401J
(4) The standard enthalpy of formation of the clonidine molecule in the gas phase was
calculated by the application of G4 theory to both atomisation and isodesmic reactions.
Acknowledgements
We wish to thank Chairman and Director General of the Joint Stock Company Federal
Research and Production Center ALTAI, Corr.ꢀmember of RAS, Prof. Aleksander S. Zharkov for
a generous supply of clonidine hydrochloride. The authors express their gratitude to Prof.
Vladimir V. Kuznetsov and to Dr. Maxim V. Shuvalov for their assistance during the development
of the experiment, to Dr. Yury V. Vishnevsky for initial processing of the experimental data and
to Dr. Yury A. Zabanov for consultations on the VibModule program. We would like to thank Dr.
Sarah Masters for the great help in preparing this manuscript. This study was supported by
Lomonosov Moscow State University Program of Development.
References
1 D. T. Lowenthal, K .M. Matzek and T. R. MacGregor, Clinical Pharmacokinetics, 1988, 14,
287–310.
2 M. J. E. Neil, Current Clinical Pharmacology, 2011, 6, 280–287.
3 P. Pospisil, P. Ballmer, L. Scapozza and G. Folkers, Journal of Receptors and Signal
transduction, 2003, 23, 361–371.
4 L. M. Jackman and T. Jen, Journal of the American Chemical Society, 1975, 97, 2811–2818.
5 A. P. Mazurek and B. KukawskaꢀTarnawska, Journal of Molecular Structure (Theochem), 1991,
236, 23–28.
6 M. Remko, O. A. Walsh and W. G. Richards, Physical Chemistry Chemical Physics, 2001, 3,
901–907.
7 M. Remko, M. Swart and F. M. Bickelhaupt, Bioorganic and Medicinal Chemistry, 2006, 14,
1715–1728.
8 S. Gundersen, V. P. Novikov, S. Samdal, R. Seip, D. J. Shorokhov and V. A. Sipachev, Journal
of Molecular Structure, 1999, 485–486, 97–114.
9 L. S. Khaikin, O. E. Grikina, V. A. Sipachev, A. V. Belyakov, E. T. Bogoradovskii and M.
Kolonits, Journal of Molecular Structure, 2000, 523, 23–37.
10. T. Jen., H. V. Hoeven, W. Groves, R. A. McLean, B. Love, Journal of Medicinal Chemistry
,
1975, 18, №1, 90–99.
11. Y. V. Vishnevskiy, UNEX version 1.5, 2007, http://unexprog.org.
12 STOE & Cie GmbH, Darmstadt, Germany, 2013, https://www.stoe.com/
17

Physical Chemistry Chemical Physics
Page 18 of 19
View Article Online
DOI: 10.1039/C6CP08401J
13. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J.
A. Montgomery, T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V.
Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada,
M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, M. Klene, X. Li, J. E Knox, H. P Hratchian, 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, P.
Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S.
Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G.
Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. AlꢀLaham,
C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W Gill, B. Johnson, W Chen, M. W. Wong,
C. Gonzalez and J. A. Pople, 2004, Gaussian Inc Wallingford CT.
14 A. D. Becke, Physical Review A, 1988, 38, 3098–3000.
15 C. Lee, W. Yang and R. G. Parr, Physical Review B, 1988, 37, 785–789.
16 C Møller and M. S. Plesset, Physical Review,1934, 46, 618–622.
17 G. A. Petersson, A. Bennett, T. G. Tensfeldt, M. A. AlꢀLaham, W. A. Shirley and J. Mantzaris,
The Journal of Chemical Physics, 1988, 89, 2193ꢀ218.
18 T. H. Dunning, The Journal of Chemical Physics, 1989, 90, 1007–1023.
19 hhtp://cccbsb.nist.gov/vibscalejust.asp
20 A.V. Sipachev, Journal of Molecular Structure, 2004, 693, 2350ꢀ240.
21 Y. V. Vishnevskiy and Y. A. Zhabanov, Journal of Physics: Conference Series, 2015, 633,
012076.
22 L. A. Curtiss, P. C. Redfern and K. Raghavachari, The Journal of Chemical Physics, 2007, 126,
84108.
23 L. A. Curtiss, K. Raghavachari, P. C. Redfern and J. A. Pople, The Journal of Chemical
Physics, 1997, 106, 1063–1079.
24 K. Raghavachari and B. B. Stefanov, Molecular Physics, 1997, 91, 555–559.
25 O. V. Dorofeeva, I. N. Kolesnikova, I. I. Marochkin and O.N. Ryzhova, Structural Chemistry
2011, 22, 1303–1314.
,
26 M. A. Suntsova and O. V Dorofeeva, Journal of Chemical and Engineering Data, 2014, 59,
2813–2826.
27 L. V. Gurvich, I. V. Veytz and C. B. Alcock, Eds. Thermodynamic Properties of Individual
Substances, Hemisphere, New York, 1990.
28 M. A. V. Ribeiro da Silva, L. M. P. F. Amaral and J. R. B. Gomes, The Journal of Physical.
Chemistry A, 2006, 110, 9301–9306.
18

Page 19 of 19
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP08401J
29 M. Zaheeruddin and Z. N. Lodhi, Phys. Chem. (Peshawar Pak.), 1991, 10, 111–118.
30 B. Ruscic, The Journal of Physical Chemistry A, 2015, 119, 7810−7837.
31 M. D. M.C. Ribeiro da Silva, M. A. V. Ribeiro da Silva, V. L. S. Freitas, M. V. Roux, P.
Jimenez, M. Temprado, J. Z. Davalos, P. Cabildo, R. M. Claramunt and J. Elguero, The Journal of
Chemical Thermodynamics, 2008, 40, 386–393.
32 N. W. Mitzel and D. W. H. Rankin, Dalton Transactions, 2003, 3650–3662.
33 G. M. Sheldrick, Acta Crystallographica A, 2008, 64, 112–122.
34 A. L. Spek, Acta Crystallographica D, 2009, 65, 148–155.
35 L. M. Jackman and T. Jen, Journal of the American Chemical Society, 1975, 97, 2811.
36 A. P. de Jong, J. Med. Chem., 1980, 23, 889.
37 J. B. Nanubolu, B. Sridhar, K. Ravikumar, CrystEngComm, 2014, 1–36.
38 A. R. Campanelli, A. Domenicano, F. Ramondo and I. Hargittai, The Journal of Physical
Chemistry A, 2004, 108, 4940–4948.
39 Attila Kovacs and Istvan Hargittai, Structural Chemistry, 2000, 11, 193–201.
40 G. Schultz, I. Hargittai, A. Domenicano, Journal of Molecular Structure, 1980, 68, 281.
41 H. B. Bürgi and J. D. Dunitz, Helvetica Chimica Acta, 1970, 53, 1747–1764
42 V. Cody and G. T. de Titta, Journal of Crystal and Molecular Structure, 1979, 9, 33–43.
19