2-Substituted-1H-perimidines: Synthesis, crystal structure and DFT calculations

Article abstract of DOI:10.1016/j.molstruc.2013.03.020

Perimidines or 1H-perimidines represent an important class of heterocyclic compounds built up of a dihydropyrimidine ring peri-fused to a naphthalene moiety. They exhibit 1,3-annular tautomerism due to two possible hydrogen positions at one or the other endocyclic nitrogen atoms or both resulting in the imine or enamine forms, respectively. The imine forms of 2-substituted-1H- perimidines can exhibit cis or trans orientation of 2-substituent in respect of the double CN bond. Here we report synthesis of four 2-substituted perimidine derivatives: 2-hydroxymethyl-perimidine 2, ethyl-1H-perimidine-2-carboxylate 3, perimidine-2-carboxylic acid 4 and 2-cyanomethyl-perimidine 5. The compounds are characterized by MS, IR spectroscopy, single-crystal X-ray diffraction (SCXRD) (2), and powder X-ray diffraction (PXRD) (5), as well as DFT calculations (3, 4 and 5) and by the NMR spectroscopy in the solution (2-5). Single-crystal X-ray diffraction proved that the compound 2 exists in the solid state as imine tautomer. The molecular structure that compound 5 adopt in its crystalline state could not be determined reliably. The results from the PXRD structural analysis cannot univocally indicate which of the forms is more likely to be present in the solid state. On the contrary, DFT calculation shows that the enamine form of 5 is more stable than imine form.

Full text of DOI:10.1016/j.molstruc.2013.03.020

Journal of Molecular Structure 1041 (2013) 156–163
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
2-Substituted-1H-perimidines: Synthesis, crystal structure and DFT calculations
a
b,
c
d
⇑
´
´
Irena Sovic , Gordana Pavlovic , Anastasios G. Papadopoulos , Dubravka Šišak ,
Grace Karminski-Zamola a
a
´
Department of Organic Chemistry, Faculty of Chemical Engineering and Technology, University of Zagreb, Marulicev trg 20, P.O. Box 177, HR-10000 Zagreb, Croatia
b Faculty of Textile Technology, Laboratory of General Chemistry, Department of Applied Chemistry, University of Zagreb, Prilaz baruna Filipovi´ca 28a, HR-10000 Zagreb, Croatia
c Aristotle University of Thessaloniki, Department of Chemistry, P.O. Box 135, GR-541 24 Thessaloniki, Greece
d Laboratory of Crystallography, ETH Zürich, Wolfgang-Pauli-Strasse 10, 8093 Zürich, Switzerland
h i g h l i g h t s
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
1H-perimidines exhibit 1,3-annular tautomerism resulting in the imine or enamine forms.
The imine forms of 2-substituted-1H-perimidines can exhibit cis or trans orientation.
Synthesis of four 2-substituted perimidine derivatives are reported.
The compounds are characterized by MS, IR, NMR spectroscopy.
Single-crystal X-ray diffraction, powder X-ray diffraction and DFT calculations are applied.
a r t i c l e i n f o
a b s t r a c t
Article history:
Perimidines or 1H-perimidines represent an important class of heterocyclic compounds built up of a
dihydropyrimidine ring peri-fused to a naphthalene moiety. They exhibit 1,3-annular tautomerism due
to two possible hydrogen positions at one or the other endocyclic nitrogen atoms or both resulting in
the imine or enamine forms, respectively. The imine forms of 2-substituted-1H-perimidines can exhibit
cis or trans orientation of 2-substituent in respect of the double C@N bond.
Received 16 January 2013
Received in revised form 11 March 2013
Accepted 11 March 2013
Available online 18 March 2013
Here we report synthesis of four 2-substituted perimidine derivatives: 2-hydroxymethyl-perimidine 2,
ethyl-1H-perimidine-2-carboxylate 3, perimidine-2-carboxylic acid 4 and 2-cyanomethyl-perimidine 5.
The compounds are characterized by MS, IR spectroscopy, single-crystal X-ray diffraction (SCXRD) (2),
and powder X-ray diffraction (PXRD) (5), as well as DFT calculations (3, 4 and 5) and by the NMR spec-
troscopy in the solution (2–5).
Keywords:
2-Substituted-1H-perimidines
Single-crystal and powder X-ray diffraction
DFT calculations
Single-crystal X-ray diffraction proved that the compound 2 exists in the solid state as imine tautomer.
The molecular structure that compound 5 adopt in its crystalline state could not be determined reli-
ably. The results from the PXRD structural analysis cannot univocally indicate which of the forms is more
likely to be present in the solid state.
On the contrary, DFT calculation shows that the enamine form of 5 is more stable than imine form.
Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction
ponent systems with interesting pharmaceutical properties: anti-
fungal, antiulcer, antitumor or antimicrobial [7–10]. They are also
Perimidines or 1H-perimidines, firstly discovered by de Aguiar
[1] in 1874, represent an important class of heterocyclic com-
pounds built up of a dihydropyrimidine ring peri-fused to a naph-
thalene moiety. They exhibit 1,3-annular tautomerism due to two
possible hydrogen positions at one or the both endocyclic nitrogen
atoms [2]. In the past decades efficient synthetic methods for the
preparation [3–6] of perimidines are in focus of research as well
as their targeted functionalizations to give rise to new multicom-
used as coloring materials for polyester fibers [7] and as fluores-
cent materials [4]. Recently, Li and Deng reported the introduction
of the dihydroperimidine group as a very effective electron donor
group to the fluorene resulting in the formation of the spiro struc-
ture [11]. Such compounds have promising applications in organic
light-emitting diodes (OLEDs).
Here we report synthesis of four 2-substituted perimidine
derivatives (Scheme 1), IR (solid state) and NMR spectroscopic re-
sults (solution), single-crystal X-ray structure analysis (2) and
powder X-ray structure analysis (5) as well as DFT calculations
for compounds 3–5.
⇑
Corresponding author. Tel.: +385 1 3712 590; fax: +385 1 3712 599.
E-mail address: gpavlov2@ttf.hr (G. Pavlovic´).
0022-2860/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molstruc.2013.03.020

´
157
I. Sovic et al. / Journal of Molecular Structure 1041 (2013) 156–163
Scheme 1. Reaction scheme for preparation of 2-substituted-1H-perimidines and numeraton of protons for tautomers 5a and 5b.
and the resulting product was filtered off. Crude product was
recrystallized from methanol to obtain 0.66 g (48%) of green pow-
der. After repeated recrystallization from methanol dark green
2. Experimental section
2.1. Synthesis and characterization
crystals were obtained; mp = 192–195 °C; IR (diamond) (m
/cmꢁ1):
3060, 1639, 1583; 1H NMR (DMSO-d6) (d/ppm): 10.31 (bs, 1H,
NH), 7.09 ((t, 2H, J = 7.81 Hz, Hperim), 6.98 (d, 2H, J = 7.83 Hz,
Instrumental methods of detection for compounds 2–5
Melting points were determined on a Koffler hot stage micro-
scope and are uncorrected. IR spectra were recorded on a Bruker
Vertex 70 spectrophotometer with diamond crystal. Mass spectra
were recorded on an Agilent 1200 series LC/6410 QQQ instrument.
Elemental analysis for carbon, hydrogen and nitrogen were per-
formed on a Perkin–Elmer, Series II, CHNS Analyzer 2400. Where
analyses are indicated only as symbols of elements, analytical re-
sults obtained are within 0.4% of the theoretical value. All com-
pounds were routinely checked by thin layer chromatography
(TLC) using precoated Merck silica gel 60F-254 plates and the spots
were detected under UV light (254 nm).
H
perim), 6.51 (d, 2H, J = 6.92 Hz, Hperim), 5.63 (bs, 1H, OH), 4.10 (s,
2H, CH2); 13C NMR (DMSO-d6) (d/ppm): 157.83 (s, 2C), 135.60 (s,
2C), 128.82 (s, 4C), 122.39 (s), 118.65 (s, 2C), 61.80 (s); MS (m/z):
199.1 ([M+1]+); Anal. Calcd. for C12H10N2O: C, 72.71; H, 5.08; N,
14.13. Found: C, 72.90; H, 5.06; N, 14.10.
2.1.2. Ethyl-1H-perimidine-2-carboxylate 3
The solution of 1,8-diaminonaphthalene 1 (3.2 g, 0.02 mol) and
triethylamine (8.1 g, 0.08 mol) in dichloromethane (80 ml) was
cooled to –5 °C and a mixture of ethyl oxalyl chloride (2.7 g,
0.02 mol) in dichloromethane (10 ml) was added dropwise. The
reaction mixture was stirred for 40 min and then warmed up to
a room temperature and stirred overnight. Then, reaction mixture
was washed successively with water (3 ꢂ 50 ml) and dried over
anhydrous magnesium sulfate. The solvent was removed under
2.1.1. 2-Hydroxymethyl-perimidine 2
The mixture of 1,8-diaminonaphthalene 1 (1.1 g, 0.007 mol) and
glycolic acid (0.76 g, 0.01 mol) was heated at 120 °C for 2 h. Then,
ammonium solution was added in the warm reaction mixture

´
158
I. Sovic et al. / Journal of Molecular Structure 1041 (2013) 156–163
Table 1
reduced pressure and the crude residue was recrystallised from
hexane–ethyl acetate (1:2) to obtain 2.8 g of red powder (59%);
mp = 214–216 °C (mplit = 212–215 °C) [3].
General and crystal data and summary of intensity data collection and structure
refinement for the compounds 2 (1.1) and 5 (1.2).
Compound
2
1.1.
Formula
Mr
Crystal system, color and habit
Space group
Crystal dimensions (mm)
Unit cell parameters:
2.1.3. Perimidine-2-carboxylic acid 4
C
12H10N2
Compound 3 (0.72 g, 0.003 mol) was dissolved in ethanol and
diluted aqueous hidrochloride acid was added. The reaction mix-
ture was refluxed for half an hour. After cooling precipitate was fil-
tered off and recrystallized from ethanol to obtain 0.48 g of green
198.22
Monoclinic, green, plate
P 21/c
0.58 ꢂ 0.49 ꢂ 0.19
powder (75%); mp = 186–188 °C; IR (diamond) (m
/cmꢁ1): 3067–
a (Å)
b (Å)
c (Å)
b (°)
V (Å3)
Z
15.3746(2)
4.6415(1)
13.6041(2)
104.7066(11)
939.00(3)
4
1.402
0.739
2654, 1680, 1624, 1578; 1H NMR (DMSO-d6) (d/ppm): 7.33 (s,
1H), 7.09 (t, 2H, J = 7.80 Hz, Hperim), 6.99 (d, 2H, J = 8.18 Hz, Hperim),
6.39 (d, 2H, J = 7.22 Hz, Hperim); 13C NMR (DMSO-d6) (d/ppm):
146.89 (s), 141.33 (s), 135.75 (s), 128.85 (s), 123.39 (s), 118.99
(s), 108.09 (s); MS (m/z): 213.1 ([M+1]+); Anal. Calcd. for
Dc (g cmꢁ3
(mmꢁ1
F(000)
)
C12H8N2O2: C, 67.92; H, 3.80; N, 13.20. Found: C, 68.12; H, 3.81;
l
)
N, 13.16.
416
h Range for data collection (°)
h,k,l range
Scan type
3.0, 72.5
ꢁ18 to 18; ꢁ5 to 5; ꢁ16 to 16
2.1.4. 2-Cyanomethyl-perimidine 5
x
The mixture of 1,8-diaminonaphthalene 1 (0.5 g, 0.003 mol) and
2-cyanoacetamide (0.54 g, 0.006 mol) was heated at 200 °C for
15 min. Then, mixture of water and ethanol was added to the
warm mixture, until the crude product was dissolved and decolor-
ized with charcoal. After cooling, precipitate was filtered off and
0.32 g (52%) of green crystalline substance was obtained;
No. measured reflections
No. independent reflections (Rint
No. refined parameters
4740
1840, 0.019
142
)
No. observed reflections, I P 2
g1, g2 in w
r(I)
1560
0.0764, 0.0805
0.0383, 0.1140
0.0442, 0.1193
1.05
R, wR [I P 2r(I)]
R, wR [all data]
mp = 116–118 °C; IR (diamond)
(
m
/cmꢁ1): 3311, 3207, 3045,
Goodness of fit on F2, S
Max., min. electron density (e Åꢁ3
)
0.15, ꢁ0.15
2942, 2258, 2173, 1638, 1610, 1588; 13C NMR (DMSO-d6) (d/
ppm): 151.58, 148.66, 134.98, 134.65, 134.51, 133.66, 128.86,
128.00, 127.91, 127.86, 121.36, 121.22, 119.73, 118.09, 118.00,
117.66, 115.98, 113.73, 112.99, 105.27, 103.68, 102.27, 41.39,
23.97; MS (m/z): 208.1 ([M+1]+); Anal. Calcd. for C13H9N3: C,
75.35; H, 4.38; N, 20.28. Found: C, 75.62; H, 4.36; N, 20.35.
Maximum
D/r
<0.001
5
1.2.
Chemical composition
C
13N3H9ꢃH2O
Data collection
Instrument
Instrument geometry
Capillary
STOE STADI P
Transmission
0.3 mm
2.1.4.1. 2-Cyanomethyl-perimidine 5a, imine tautomeric form. 1H
NMR (DMSO-d6) (d/ppm): 10.73 (s, 1H, NH), 7.00–7.24 (m, 4H,
Wavelength
2h range
Step size
Cu Ka1 1.5406 Å
6.80°–60.00°
0.01°2h
H
perim), 6.61 (d, 1H, J = 7.11 Hz, Hperim), 6.30 (d, 1H, J = 7.01 Hz,
Hperim), 3.79 (s, 2H, CH2).
Measurement type
Exposure time
Number of points
Temperature
Continuous
Variable
5321
296(2) K
2.1.4.2. 2-Cyanoethenyl-perimidine 5b, enamine tautomeric
form. 10.19 (s, 1H, NH), 9.91 (s, 1H, NH), 7.00–7.24 (m, 4H, Hperim),
6.90 (d, 1H, J = 7.15 Hz, Hperim), 6.47 (d, 1H J = 7.10 Hz, Hperim), 3.43
(s, 1H, CHalif).
Unit cell parameters
a
b
c
19.64(2) Å
15.0676 (2) Å
7.336(8) Å
100.8 (6)°
P 21/n
2.2. X-ray diffraction experiment
b
Space group
Z
Z0
2.2.1. Single-crystal diffraction experiment for compound 2
The general and crystallographic data for the compounds 2 is
listed in Table 1. The bond distance table (Table 2) contains rele-
vant geometrical features for structure 2. Table 3 lists geometrical
parameters of hydrogen bonds and weak interactions for com-
pound 2.
4
2
Structure solution
Method program
Direct space method FOX
Structure refinement
Rietveld refinement program
dmin
Number of contributing reflections
Number of geometric restraints
CAC/CAN
XRS
1.54 Å
611
80
The data collection for structure 2 was carried out on Oxford
Xcalibur Nova four-circle kappa geometry single-crystal diffrac-
tometer with CCD detector and k(Cu K ) = 1.54184 Å at ambient
a
37
temperature 296(2) K. Data collection has been performed by
applying the CrysAlis Software system, Version 1.171.32.29. [12].
The Lorentz-polarization effect was corrected and the intensity
data reduced by the CrysAlis RED application of the CrysAlis Soft-
ware system, Version 1.171.32.29. [12]. The diffraction data have
been scaled for absorption effects by multiscan method. Coordi-
nates of the non-hydrogen atoms were determined by direct meth-
ods with the SHELXS program [13]. The coordinates and
anisotropic thermal displacement parameters for all non-hydrogen
atoms were refined by the least-squares methods based on F2 using
the SHELXL97 program [13].
CACAC/CANAC/CACAN
Number of structural parameters refined
Hydrogen treatment
Number of profile parameters
Common weight factor for the restraints
45
132
Calculated
11
3.6
The hydrogen atoms, which belong to Csp2 and Csp3 carbon
atoms were generated at ideal geometrical positions [Csp2-H
0.93 Å, Uiso(H) = 1.2 Ueq(C) and Csp3-H 0.97 Å, Uiso(H) = 1.2 Ueq(C)]
and constrained to ride on the parent carbon atom. The hydrogen

´
159
I. Sovic et al. / Journal of Molecular Structure 1041 (2013) 156–163
Table 2
programs [24]. The hybrid method was applied (B3LYP) with
Becke’s three-parameter functional [25] and the nonlocal correla-
tion is provided by the LYP expression [26]. The basis set which
used was the augmented 6-311++G(d,p) [27,28]. Harmonic fre-
quency calculations were performed for all the optimized struc-
tures to establish that the stationary points found are minimal.
In calculating the interaction energy in the hydrogen-bonded
complexes studied, we accounted for the basis set superposition
error (BSSE) by recalculating the monomer energies using the full
dimer basis at the optimized geometry of the dimer using the
counterpoise method. In this context, the BSSE corrected interac-
tion energy of the Aꢃ ꢃ ꢃB dimer is calculated as:
Selected interatomic distances (Å) and valence angles
(°) for the compound 2.
2
O1AC12
N1AC1
N1AC2
N2AC10
N2AC1
C1AC12
1.410(2)
1.295(2)
1.413(2)
1.399(2)
1.349(2)
1.509(2)
C1AN1AC2
117.55(10)
121.60(10)
119.6(10)
118.7(11)
125.01(10)
C1AN2AC10
C1AN2AH2N
C10AN2AH2N
N1AC1AN2
D
EðBSSEÞ ¼ E þ BSSE
D
D
E ¼ EAfABBgðABÞ ꢁ EAfAgðAÞ ꢁ EBfBgðBÞ
BSSE ¼ EAfABgðAÞ þ EfABBgðBÞ ꢁ EfAABBgðAÞ ꢁ EAfABBgðBÞ
atoms which belong to ANH and AOH groups were found as a
small electron-density at 0.90(2) Å apart from the N2 nitrogen
atom and equally from the O1 oxygen atom and they refine freely.
The molecular geometries and graphics were done using PovRay
[14] and PLATON programmes [15].
D
EðBSSEÞ ¼ EAfABBgðABÞ ꢁ EAfAgðAÞ ꢁ EBfBgðBÞ þ EAfABgðAÞ þ EAfBBgðBÞ
ꢁ EfAABBgðAÞ ꢁ EAfABBgðBÞ
where EAfABBgðABÞ is the energy of the optimized dimer, EfAAgðAÞ and
EfBBgðBÞ the energies of the monomers A and B respectively at their
equilibrium geometries calculated in their own bases {A} and {B},
EfAABgðAÞ and EfABBgðBÞ the energies of the monomers A and B respec-
tively at geometries found in the dimer calculated in their own
bases {A} and {B}, and EfAABBgðAÞ and EAfABBgðBÞ the energies of the mono-
mers A and B respectively at geometries found in the dimer calcu-
lated with the full basis of the dimer {AB} [29–31].
2.2.2. Powder diffraction experiment for compound 5
The data collection for compound 5 was carried out on a STOE
STADI P diffractometer with a linear position sensitive detector
(PSD) and Cu Ka1 (1.5406 Å) radiation. The sample was filled into
a 0.3 mm glass capillary and measured in transmission geometry,
using variable exposure times. Indexing of the data was per-
formed using the software CMPR [16] and revealed a monoclinic
cell. Subsequent examination of the systematic absences resulted
in the recognition of P 21/n symmetry (Table 1.2). The unit cell
volume and symmetry implied that there would be two mole-
cules in the asymmetric unit. As the basic parameters (molecular
model, unit cell parameters, symmetry) were known, direct-space
optimization was used for the structure determination [17]. A
molecular model was constructed using a similar structure found
in the CSD database [18], the imine tautomer of 2-tBu-perimidine
(reference code DELKUB, [19]). Using the program in the program
Mercury [20], -tBu group from the DELKUB structure was replaced
by a ACH2CN group and molecular model for compound 5 was
obtained. Two molecular models were input to the direct-space
program FOX [21], and optimized using the parallel tempering op-
tion. Soft restraints were imposed on bond lengths and angles of
the molecules but torsion angles were allowed to change freely.
The resulting structure was then subjected to Rietveld refinement
[22,23]. After just a few cycles, the difference Fourier maps re-
vealed one significant electron density maximum that was inter-
preted to be a water molecule.
3. Results and discussion
3.1. General considerations of synthetic procedure
Starting material for the synthesis of perimidine ring was 1,8-
diaminonaphtalene. Depending on the reagent, different deriva-
tives were prepared (Schemes 1–4.). 2-hydroxymethyl-perimidine
2 was prepared using glycolic acid, ethyl-1H-perimidine-2-carbox-
ylate 3 was prepared using ethyl oxalyl chloride and 2-cyanometh-
yl-perimidine 5 was prepared using 2-cyanoacetamide. Attempt to
prepare perimidine-2-carboxylic acid 4 by oxidating 2-hydroxy-
methyl-perimidine using KMnO4 as oxidazing agent was not
successful so another approach was used, acidic hydrolysis of
ethyl-1H-perimidine-2-carboxylate 3.
The prepared perimidine derivatives were confirmed by NMR
spectroscopy in solution based on the analysis of chemical shifts
and HAH coupling constants. In the 1H NMR spectra of compounds
2 and 4 two doublets and one triplet for six perimidine protons can
be observed. The different substituents at the 2 position of perimi-
dine nuclei disturb only protons 4 and 9 (numbering of the H-
atoms is shown in Scheme 1), while the chemical shifts of other
four protons remain the same. This is also observed in the spectra
of compound 5. For the compound 2, the signal of the NH proton of
perimidine nuclei is broad and shifted to the lower magnetic field
2.3. Quantum chemical calculations
Density functional theory (DFT) full geometry optimizations of
the compounds (3), (4) and (5) and the H-bonded of the compound
(5) have been carried out using the Gaussian 03W suite of
Table 3
Hydrogen bond and interaction geometry (Å, °) for the compound 2.
DAHꢃ ꢃ ꢃA
DAH
Hꢃ ꢃ ꢃA
Dꢃ ꢃ ꢃA
\DAHꢃ ꢃ ꢃA
Symmetry code
2
N2AH2Nꢃ ꢃ ꢃO1
O1AH1Oꢃ ꢃ ꢃN1
C3AH3ꢃ ꢃ ꢃO1
0.90(2)
0.90(2)
0.93
0.97
0.97
2.09(2)
1.89(2)
2.96
2.78
2.97
2.933(1)
2.790(1)
3.592(2)
3.565(2)
3.722(2)
162
179
127
138
136
1ꢁx,ꢁy,1ꢁz
1ꢁx,ꢁ1/2+y,3/2ꢁz
ꢁx+1,+y+1/2, ꢁz+3/2
ꢁx+1,+y+1/2, ꢁz+3/2
x,+y+1,+z
C12AH12Aꢃ ꢃ ꢃN1
C12AH12Bꢃ ꢃ ꢃO1

´
160
I. Sovic et al. / Journal of Molecular Structure 1041 (2013) 156–163
2
1
NH
1
N
N
N1H
1
2
1
2
1
2
1
1
1
N
N
NH
NH
2
3
3
NC
CN
2
2
1
a
a'
O
O
O
O
a
b
2
1
NH
HN
1
NH
HN
Scheme 2. Schematic overview of trans (a) and cis (b) isomers of imine tautomeric
form of compound (3).
C
2
C
3
NC
H
H
CN
b
b'
Scheme 4. Schematic overview of trans (a) and cis (a0) isomers of imine tautomeric
1
form of compound (5) and its enamine tautomeric form (b).
1
2
2
2
1
1
NH
N
NH
N
2
2
1
1
2
H
O
O
O
O
H
a
b
Scheme 3. Schematic overview of trans (a) and cis (b) isomers of imine tautomeric
form of compound (4).
(10.31 ppm) while for the compound 4 that signal is a sharp singlet
and shifted to the higher magnetic field (7.33 ppm). This difference
is due to an intramolecular hydrogen bond(s) between NH group of
perimidine nuclei and hydroxyl and carboxyl group, respectively.
Analysis of 1H and 13C NMR spectra in solution of compound 5
has shown the existence of two tautomers in 2:1 ratio in the favor
of 5a (see Sections 2.1.4.1 and 2.1.4.2. and Scheme 1). This is sup-
ported by observed signals at 3.97 ppm indicating two methyl pro-
tons and at 3.43 ppm indicating one ethenyl proton. Also, the NH
protons of perimidine nuclei are observed, one stronger signal at
10.73 ppm and two weaker signals at 9.91 and 10.19 ppm,
respectively.
Fig. 1. ORTEP drawing of 2-hydroxymethyl-1H-perimidine,
numbering scheme. The thermal ellipsoids are drawn at the 50% probability level at
296 K.
2 with the atomic
N1AC2 and N2AC10 bond distances are longer having more pro-
nounced
r character. Therefore, the geometry of the perimidine
3.2. Crystal structure description of compounds 2 and 5
structural fragment including N1, C1, N2 and C12 atoms is strongly
affected by the type of 2-substituent (aryl vs. aliphatic), proton-
ation or alkylation of the N2 atom and by the possibilities of hydro-
gen bonding arrangements.
The C1AN2 bond value of 1.349(2) Å is comparable with the
structures of 2-(2-hydroxyphenyl)perimidine (1.361(4) Å) and 2-
(9-anthryl)perimidine (1.352(2) Å), on the contrary to longer
bonds in 1-methyl perimidine derivatives such as: 1-methyl-2-
phenylperimidine (1.375(2) Å) and 1-methyl -2-(3-methoxy-
phenyl)perimidine (1.377(2) Å) [3].
The perimidine moiety of (1H-perimidine-2-yl)-methanol mol-
ecule, 2, is essentially planar with the largest r.m.s. deviation of
0.065(1) Å for the protonated N2 atom (Fig. 1). The methylene
C12 atom and the hydroxyl O1 atom deviate from the mean per-
imidine moiety plane by 0.088(1) Å and –0.583(1) Å, respectively.
The protonation of the N2 site is strongly supported by the bond
distance arrangement around N2. The C1AN2 bond is longer than
N1AC1 bond, although they all have a great degree of charge delo-
calization (Table 2; 1.349(2) vs. 1.295(2) Å). On the contrary, the

´
161
I. Sovic et al. / Journal of Molecular Structure 1041 (2013) 156–163
The similar trend is not observed with the N1AC1 bond value of
1.295(2) Å which is comparable with the analogous bond in 1-
methylperimidine derivatives rather than with 1H-perimidine
derivatives. The reason for that is probably that all mentioned 2-
substituted perimidines are 2-aryl derivatives, while (1H-perimi-
dine-2-yl)-methanol contains C12 atom which is sp3 hybridized
representing aliphatic discontinuance.
Crystal structure of 2 (Fig. 2 and Table 3) is built up on the basis
of two intermolecular hydrogen bonds formed between the alcho-
col hydroxyl group and the ring nitrogen atom (O1AH10ꢃ ꢃ ꢃN1) and
between ANH perimidine group and hydroxyl oxygen atom
(N2AH2Nꢃ ꢃ ꢃO1) thus joining molecules into infinite 2D chains.
The gross crystal structure of compound (5) is characterized by
two perimidine moieties and one water molecule in the asymmet-
ric unit. The obvious difference between these perimidine mole-
cules arises from the torsion angle of the cyano group with
respect to the perimidine backbone, affecting the gross planarity
of the molecule (Fig. 3). Rietveld refinement of this ‘‘gross struc-
ture’’ was hindered by the fact that the data resolution was poor
(dmin = 1.54 Å), and that the exact molecular structures of the two
perimidines in the asymmetric unit were not known. As this mol-
ecule shows tautomerism, it can adopt either the imine (5a) or en-
amine (5b) form. Moreover, the imine tautomer can adopt either
the trans (5a) or cis (5a0) conformation of the cyano group with re-
spect to the double C@N bond (as presented in Scheme 4). Given
that there are two perimidine molecules in the asymmetric unit
and each can adopt any of the three possible forms (a,a0,b), there
are seven theoretically possible combinations. Some of these pos-
sible structures, labeled XX, where X = {a, a0, b} were constructed
and subjected to the restrained Rietveld refinement. It was rea-
soned that although the correct structure would not be obvious,
a careful examination of the resulting molecular geometries and
their corresponding R values might indicate which structure is
more likely to be correct. For each structure, the corresponding re-
straints on bond distances and angles were assigned. In the cases of
the imine tautomer (a and a0), these geometrical restraints were
obtained from closely correlated structures (2) and [18] that were
solved from single crystal data. Geometrical restraints for the
refinement of tautomer b were obtained from an analogous struc-
ture found in the CSD database, reference code GAMNAK [29]. In all
cases, Rietveld refinement included refinement of the molecular
geometry by using comparably high weights on the restraints. This
ensured that the molecular geometry was optimized with respect
to the expected values. Then, as the refinement progressed, the
Fig. 3. Crystal structure of compound 5. Hydrogen atoms are omitted for clarity.
weights were reduced systematically, i.e. giving less weight to
the geometric restraints and more weight to the diffraction data.
The positions of the hydrogen atoms were calculated using the
program Mercury and their coordinates were not refined. An at-
tempt to assign the hydrogen atoms to the water molecule was
not undertaken. In order to compensate for these two electrons,
the water was modeled as an oxygen atom with an increased pop-
ulation parameter. The final Rietveld refinement resulted in R val-
ues that showed no clear preference for any of the tested structures
(Table 4). However, the restraint R-value proved to be noticeably
higher for the enamine tautomer (bb), indicating that its molecular
structure does not follow the expected values as well as the others.
Next, the molecular geometries of all models were examined.
For those with the imine tautomer, selected bond lengths and an-
gles are presented in Table S1. In all cases, refinement of the atomic
coordinates resulted in stable values for the CAN bonds, showing
no extreme discrepancies from the expected values. The CAN
bonds are all at about 1.37 Å, giving a comparably fair agreement
with these reported in the structures (2) and [19] (1.35 and
1.36 Å, respectively). The C@N double bond for the imine tautomer
range between 1.31 and 1.32 Å, while the analogous bond in struc-
tures (2) and [19] is 1.30 Å. In the (bb) model, both CAN and C@N
bonds are slightly longer than their analogues in the GAMNAK
structure. The (C1AC12AC13) angles on sp2 hybridized carbons
also showed a larger discrepancy than for an sp3 hybridized car-
bon. A detailed comparison of the molecular geometries is pro-
vided in Table S2.
Possible inter- and intramolecular interactions were calculated
using the program PLATON. All constructed structures showed ex-
tended H-bond networks (Table S3). However, no clear or detailed
description of these networks is possible, because of the ambiguity
regarding the hydrogen atoms in the water molecule. Although
Table S3 may suggest that the number of possible interactions
would favor structures (a0a0) and (bb), aforementioned ambiguities
prevent us from drawing definite conclusions.
Although NMR results show that two tautomers (2:1) are pres-
ent in the liquid state, the possibility of a cocrystalization of these
two tautomers was not investigated here. That is, no attempt to
construct an asymmetric unit with one imine tautomer (a or a0)
and one enamine (b) was undertaken.
In conclusion, PXRD analysis could not provide exact results in
this specific case, that is, none of the structural models discussed
can be explicitly favored or excluded. Moreover, physical mixtures
Table 4
Results of the Rietveld refinement for five constructed structures.
(aa’)
(a’a’)
(aa)
(a’a)
bb
RF
Rwp
Rexp
Rrestraints
0.076
0.108
0.057
0.006
0.075
0.107
0.057
0.0075
0.076
0.109
0.057
0.006
0.076
0.109
0.057
0.007
0.075
0.108
0.057
0.011
Fig. 2. Crystal structure of 2-hydroxymethyl-1H-perimidine, 2 viewed along y axis.
Hydrogen bonds of the OAHꢃ ꢃ ꢃN and the NAHꢃ ꢃ ꢃO type showing as dashed lines are
realized via alcohol hydroxyl group, endocyclic thiazolyl nitrogen atom and
thiazolyl ANH group.

´
I. Sovic et al. / Journal of Molecular Structure 1041 (2013) 156–163
162
of these structures, or cocrystalization of two tautomers is also
possible.
level of theory. Approach of the water molecule was assumed in
the pathway that would contribute more to the stabilization of
the final product through the most hydrogen bonds formed in each
case. The optimized structures are given at Fig. S4. The relative sta-
bility of the entities studied is analogous to the one computed for
the neat conformers, however the energy differences are more
widespread and favor 5b H2O by 1.4 kcal molꢁ1 over 5a ꢂ H2O
and that by 6.8 kcal molꢁ1 over 5a0 ꢂ H2O. The C1AN1 and C1AC2
bond distances in hydrogen bonded tautomers remain the same,
comparing to the non-hydrogen bonded tautomeric forms
(Table S5). However, the dihedral angle N1AC1AC2AN3 of the s-
trans imine tautomer (5a) appears distortion by 39.9°. The same
is observed in the s-cis imine tautomer (5a0) (24.4°). A slightly dis-
tortion of the dihedral angle N1AC1AC2AN3 in enamine tautomer
(5b) is observed (1.5°). The calculated hydrogen bonds N1AH1ꢃ ꢃ ꢃO1
and O1AH2ꢃ ꢃ ꢃN3 in the s-trans imine tautomer (5a) are 1.95 and
2.27 Å, respectively. In enamine tautomer (5b) the calculated
hydrogen bonds N1AH1ꢃ ꢃ ꢃO1 and O1AH2ꢃ ꢃ ꢃN3 are 1.95 and 2.05 Å,
respectively. The calculated hydrogen bonds O1AH2ꢃ ꢃ ꢃN3 and O1-
AH3ꢃ ꢃ ꢃN2 in the s-trans imine (5a0) tautomer are 2.82 and 2.05 Å,
respectively. These hydrogen bond distances are indicative of a
stronger binding of water to enamine (5b) tautomer. This is further
confirmed by the higher interaction energy values in water-enam-
ine tautomer complex (29.7 kJ molꢁ1) compared to that of water-
imine complexes (24.7 kJ molꢁ1 and 15.4 kJ molꢁ1 for s-trans and
s-cis imine tautomers, respectively), listed before and after inclu-
sion of the BSSE correction calculated by the counterpoise method,
as well as the ZPE correction in Table S8.
3.3. Theoretical calculations for compounds 3, 4 and 5
Relative energies and geometrical parameters of calculated con-
formers are depicted at Tables S4–S8, optimized geometries are gi-
ven at Figs. S1–S4 and the scan diagrams around the dihedral angle
N1AC1AC2AO1 as calculated at the B3LYP/6-311++G(d,p) level of
theory are enclosed at Figs. S5–S7.
Geometry optimization computations were carried out on the
possible tautomeric forms of compounds 3 and 4 assuming a single
molecule with no interactions to a surrounding medium as de-
picted in Schemes 2 and 3. The two stable conformers for each
compound 3 and 4, respectively, are presented at Figs. S1 and S3.
Conformer 3a, forms s-cis configuration between the carbonyl
bond of the ester group (AC@O) and the pyrimidine ring double
bond C1AN2 while conformer 3b form s-trans configuration be-
tween the carbonyl bond of the ester group (AC@O) and the pyrim-
idine ring double bond C1AN2. The computation reveals
a
preference for conformer 3b by 2.5 kcal molꢁ1, as shown in
Table S4. Its stability being verified by a scan calculation around
the N1AC1AC2AO1 dihedral angle as shown in Fig. S5, as well as
by QST3 calculations and reinforced by the intramolecular hydro-
gen bond N1AH1ꢃ ꢃ ꢃO2 (2.24 Å). For compound 4, conformer 4a
forms s-cis configuration between the carbonyl bond of the carbox-
ylic group (AC@O) and the pyrimidine ring double bond C1AN2,
while conformer 4b forms s-trans configuration between the car-
bonyl bond of the carboxylic group (AC@O) and the pyrimidine
ring double bond C1AN2 as depicted in Fig. S2. Conformer 4b is
computed to be more stable by 15.1 kcal molꢁ1, as shown in
Table S5. The stability being reinforced by the intramolecular
hydrogen bonds N1AH1ꢃ ꢃ ꢃO2 and O1AHꢃ ꢃ ꢃN2 (2.36 and 2.03 Å
respectively). The stability of 4b was also verified by a scan calcu-
lation around the dihedral angle of N1AC1AC2AO1 as shown in
Fig. S6, as well as by QST3 calculations.
4. Conclusion
Here we report synthesis of four 2-substituted perimidine
derivatives, IR (solid-state) and NMR (solution) spectroscopic
results, single-crystal X-ray diffraction (SCXRD) (1H-perimidine-
2-yl)-methanol, 2) and powder X-ray diffraction (PXRD)
(2-cyanomethyl-perimidine, 5) as well as DFT calculations for
compounds 3, 4 and 5.
These compounds can exist in the two possible tautomeric
forms: the imine form and the enamine form.
The compound 2 exists in the solid state as imine tautomer as it
is established by SCXRD method.
Geometry optimization computations carried out on the possi-
ble tautomeric forms of compounds 3 and 4 assuming a single mol-
ecule with no interactions to a surrounding medium revealed the
enamine forms of 3 and 4 to be more stable.
Information obtained from the PXRD data was not conclusive
about what tautomer of the compound 5 is present in the solid
state. Moreover, possibility of cocrystalization of two tautomers,
or even physical mixtures of two tautomers (or cocrystal) cannot
be excluded.
The subsequent step of study the structures of the hydrated
conformers of 5 were optimized at the B3LYP/6-311++G(d,p) level
of theory. The relative stability of the entities studied is analogous
to the one computed for the neat conformers, however the energy
differences are more widespread and favor the enamine form
5b ꢂ H2O by 1.4 kcal molꢁ1 over 5a ꢂ H2O and that by
6.8 kcal molꢁ1 over 5a0 ꢂ H2O conformer.
The investigation of compound 5 is more demanding and inter-
esting in a view of the various but not decisive experimental re-
sults concerning the tautomeric form in which it is present. In
view of the synthesis procedure applied which facilitates produc-
tion of the thermodynamically more stable conformers and of
the follow-up treatment with aqueous medium from which crys-
talline hydrate is precipitated it was assumed suitable to study at
first the possible tautomers as single molecules in a traditional
computational scheme. The two possible forms of the imine and
the enamine tautomer computed are presented at Scheme 4.
Conformer 5a form s-trans configuration between the cyano
group (ACN) and the pyrimidine ring double bond C1AN2 while
conformer 5a0 form s-cis configuration between the cyano group
(ACN) and the pyrimidine ring double bond C1AN2. The conformer
5a0 has higher energy by 3.0 kcal molꢁ1 as shown in Table S6. The
stability of 5a was also verified by a scan calculation around the
dihedral angle of N1AC1AC2AN3 as shown at Fig. S7., as well as by
QST3 calculations. The enamine 5b is marginally more stable rela-
tive to 5a by 0.1 kcal molꢁ1. Based on the above results the predic-
tion can be made that under the synthetic procedure applied
conformers 5a and 5b are expected to be present in the final prod-
uct in almost equal quantities while 5a0 should not exceed 0.1% in
abundance. One should not overlook however the synthetic proce-
dure where the crude product is treated with aqueous methanol
especially in view of the computed dipole moments of the conform-
ers which are 1.27, 5.8 and 5.5 D for 5a, 5a0 and 5b respectively, giv-
ing a substantial preference for 5a0 to be present in the final product
even if an equilibrium should be assumed between this and 5a.
Therefore as a subsequent step of study the structures of the
hydrated conformers were optimized at the B3LYP/6-311++G(d,p)
Acknowledgement
The Ministry of Science and Technology of the Republic of Croa-
tia, Zagreb (Grant Nos. 125-0982464-1356 and 098-1191344-
2943) supported this work.

´
163
I. Sovic et al. / Journal of Molecular Structure 1041 (2013) 156–163
[14] <www.povray.org>.
[15] A. Spek, J. Appl. Crystallogr. 36 (2003) 7.
Appendix A. Supplementary material
[16] B.H. Toby, Appl. Cryst 38 (2005) 1040.
[17] R. Cerny, V. Favre-Nicolin, Z. Kristallogr. 222 (2007) 105.
[18] F.H. Allen, Acta Cryst. B58 (2002) 380–388.
[19] T. Murata, Y. Morita, K. Fukui, K. Tamaki, H. Yamochi, G. Saito, K. Nakasuji, Bull.
Chem. Soc. Jpn. 79 (2006) 894.
ˇ
CCDC numbers CCDC 919277 (compound 2) and 919278 (com-
pound 5) contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge at www.ccdc.ca-
m.ac.uk/conts/retrieving.html [or from the Cambridge Crystallo-
graphic Data Centre (CCDC), 12 Union Road, Cambridge CB2 1EZ,
UK; fax: +44(0)1223-336033; email: deposit@ccdc.cam.ac.uk].
Structure factors table is available from the authors. Supplemen-
tary data associated with this article can be found, in the online
version, at http://dx.doi.org/10.1016/j.molstruc.2013.03.020.
[20] C.F. Macrae, I.J. Bruno, J.A. Chisholm, P.R. Edgington, P. McCabe, E. Pidcock, L.
Rodriguez-Monge, R. Taylor, J. van de Streek, P.A. Wood, J. Appl. Crystallogr. 41
(2008) 466.
ˇ
[21] V. Favre-Nicolin, R. Cerny, J. Appl. Cryst. 35 (2002) 734.
[22] Baerlocher, Ch. (1990) EXTRACT, ETH, Zürich, Switzerland.
[23] Baerlocher, Ch., Hepp, A. (1982) XRS-82, ETH, Zürich, Switzerland.
[24] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
J.A. Montgomery, Jr., 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, 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. Al-Laham, C.Y. Peng, A. Nanayakkara, M.
Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A.
Pople, Gaussian 03, Revision E.01-SMP, Gaussian, Inc., Pittsburgh PA, 2003.
[25] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
[26] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B37 (1988) 785.
[27] K. Raghavachari, J.A. Pople, E.S. Replogle, M. Head-Gordon, J. Phys. Chem. 94
(1990) 5579.
[28] M.J. Frisch, J.A. Pople, J.S. Binkley, J. Chem. Phys. 80 (1984) 3265.
[29] R. Davis, N. Tamaoki, Org. Lett. 7 (2005) 1461.
[30] (a) S.F. Boys, F. Bernardi, Mol. Phys. 19 (1970) 553;
(b) E.R. Davidson, S.J. Chakravorty, Chem. Phys. Lett. 217 (1994) 48.
[31] (a) F.B. Van Duijneveldt, J.G.C.M. Van Duijneveldt-van de Rijdt, J.H. Van Lenthe,
Chem. Rev. 94 (1994) 1873;
References
[1] A. De Aguiar, Ber. Dtsch. Chem. Ges. 7 (1874) 309.
[2] R.M. Claramunt, J. Dotor, D. Sanz, C. Foces-Foces, A.L. Llamas-Saiz, J. Elguero, R.
Flammang, J.P. Morizur, E. Chapon, J. Tortajada, Helv. Chim. Acta 77 (1994) 121.
[3] I. Yavari, M. Adib, F. Jahani-Moghaddam, H.R. Bijanzadeh, Tetrahedron 58
(2002) 6901.
[4] G. Varsha, V. Arun, P.P. Robinson, M. Sebastian, D. Varghese, P. Leeju, V.P.
Jayachandran, K.K.M. Yusuff, Tetrahedron Lett. 51 (2010) 2174.
[5] M.M. Belmonte, E.C. Escudero-Adán, J. Benet-Buchholz, R.M. Haak, A.W. Kleij,
Eur. J. Org. Chem. (2010) 4823.
[6] I.A.S. Smellie, A. Fromm, R.M. Paton, Tetrahedron Lett. 50 (2009) 4104.
[7] J.M. Herbert, P.D. Woodgate, W.A. Denny, J. Med. Chem. 30 (1987) 2081.
[8] R.M. Claramunt, J. Dotor, J. Elguero, Ann. Quim. 91 (1995) 151.
[9] D.R. Luthin, A.K. Rabinovich, D.R. Bhumralkar, K.L. Youngblood, R.A. Bychowski,
D.S. Dhanoa, J.M. May, Bioorg. Med. Chem. Lett. 9 (1999) 765.
[10] X. Bu, L.W. Deady, G.J. Finlay, B.C. Baguley, W.A. Denny, J. Med. Chem. 44
(2001) 2004.
[11] Z. Li, W. Deng, Spectrochimica Acta Part A 82 (2011) 56.
[12] Oxford Diffraction Ltd., Xcalibur CCD system, CrysAlis Software system,
Version 1.171.32.29 Abingdon, Oxfordshire, England, 2010.
[13] G.M. Sheldrick, Acta Crystallogr. A64 (2008) 112.
(b) B. Jeziorski, R. Moszynski, K. Szalewicz, Chem. Rev. 94 (1994) 1887.