Structural investigations of 5-hydroxy-4,5-dihydroisoxazoles

Article abstract of DOI:10.1016/j.molstruc.2011.09.051

The X-ray diffraction data determined for eight 3-(R³), 4-(R⁴), 4,4-(R⁴/R^4′) and/or 5-(R ⁵) 5-hydroxy-4,5-dihydroisoxazoles [where R³ = Ph, R ⁴/R^4′ = H/H, R⁵ = CCl₃ (1); R³ = 4-Br-C₆H₄, R⁴/R ^4′ = H/H, R⁵ = CCl₃ (2); R³ = thien-2-yl, R⁴/R^4' = H/H, R⁵ = CCl₃ (3); R³ = Ph, R⁴ = Ph, R^4' = OH, R⁵ = Me (4); R³ = Me, R⁴/R^4′ = N-OH, R ⁵ = Me (5); R³ = CF₃, R⁴/R ^4' = H/H, R⁵ = CMe₂CH₂OH (6); R ³ = H, R⁴ = 4-I-C₆H₄, R^4' = H, R⁵ = 4-I-C₆H₄ (7); R³/R ⁴ = -(CH₂)₃-, R^4′ = H, R ⁵ = CF₂CF₂H (8);] are discussed. The crystalline structure of compounds 1-3 is described for the first time and crystalline structure of compounds 4-8 has already been described in literature. It was found that the supramolecular auto-organization of 1-8 is characterized by hydrogen bonds invariably involving the hemiacetal hydroxyl group. Compound 5 is the only exception, where the hydroxyl oxime group is the participant in the hydrogen bond. Compounds 4 and 8 present intermolecular contact between the hydroxyl group of the hemiacetal and the nitrogen atom of the 4,5-dihydroisoxazole ring. Compound 7 presents similar interaction, where the hydroxyl contact is with the oxygen atom of the 4,5-dihydroisoxazole ring. Moreover, the crystal structure of compound 6 was stabilized by O-H?O interaction between the hydroxyl group of hemiacetal and the hydroxyl group of the alcohol function attached at the 5-position of 4,5-dihydroisoxazole. The crystal structure of compounds 1-3, as described here for the first time, was similar to that of compounds 4 and 7, showing a hydrogen bond O(51)-H(51)?N(2) between the hydroxyl group and the nitrogen atom of the isoxazoline ring. This means that the crystal structure of these compounds was governed by hydrogen bonds O-H?N, involving the hydroxyl of the hemiacetal group and the nitrogen atom of the 4,5-dihydroisoxazole ring. This interaction is relatively robust, showing a pattern in the crystal packing. Compounds 1-3 also have their crystal stabilized by more weak interactions of type Cl?Cl, involving the chlorine atom of the trichloromethyl group.

Full text of DOI:10.1016/j.molstruc.2011.09.051

Journal of Molecular Structure 1006 (2011) 462–468
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Structural investigations of 5-hydroxy-4,5-dihydroisoxazoles
Patrick T. Campos a, Pablo Machado a, Clarissa P. Frizzo a, Dayse N. Moreira a, Alexandre R. Meyer a,
Helio G. Bonacorso a, Nilo Zanatta a, Lucas C. Ducati b, Roberto Rittner b, Cláudio F. Tormena b,
Marcos A.P. Martins a,
⇑
a Department of Organic Chemistry, Núcleo de Química de Heterociclos (NUQUIMHE), Federal University of Santa Maria, UFSM, CEP 97105-900, Santa Maria, RS, Brazil
b Department of Organic Chemistry, Chemistry Institute, P.O. Box 6154, State University of Campinas, UNICAMP, CEP 13083-970, Campinas, SP, Brazil
a r t i c l e i n f o
a b s t r a c t
0
The X-ray diffraction data determined for eight 3-(R3), 4-(R4), 4,4-(R4/R4 ) and/or 5-(R5) 5-hydroxy-4,5-
Article history:
0
0
dihydroisoxazoles [where R3 = Ph, R4/R4 = H/H, R5 = CCl3 (1); R3 = 4-Br-C6H4, R4/R4 = H/H, R5 = CCl3 (2);
Received 30 June 2011
Received in revised form 26 September
2011
Accepted 27 September 2011
Available online 20 October 2011
0
R3 = thien-2-yl, R4/R4’ = H/H, R5 = CCl3 (3); R3 = Ph, R4 = Ph, R4’ = OH, R5 = Me (4); R3 = Me, R4/R4 = N–OH,
R
5 = Me (5); R3 = CF3, R4/R4’ = H/H, R5 = CMe2CH2OH (6); R3 = H, R4 = 4-I-C6H4, R4’ = H, R5 = 4-I-C6H4 (7);
0
R3/R4 = –(CH2)3–, R4 = H, R5 = CF2CF2H (8);] are discussed. The crystalline structure of compounds 1–3
is described for the first time and crystalline structure of compounds 4–8 has already been described
in literature. It was found that the supramolecular auto-organization of 1–8 is characterized by hydrogen
bonds invariably involving the hemiacetal hydroxyl group. Compound 5 is the only exception, where the
hydroxyl oxime group is the participant in the hydrogen bond. Compounds 4 and 8 present intermolec-
ular contact between the hydroxyl group of the hemiacetal and the nitrogen atom of the 4,5-dihydrois-
oxazole ring. Compound 7 presents similar interaction, where the hydroxyl contact is with the oxygen
atom of the 4,5-dihydroisoxazole ring. Moreover, the crystal structure of compound 6 was stabilized
by O–Hꢀ ꢀ ꢀO interaction between the hydroxyl group of hemiacetal and the hydroxyl group of the alcohol
function attached at the 5-position of 4,5-dihydroisoxazole. The crystal structure of compounds 1–3, as
described here for the first time, was similar to that of compounds 4 and 7, showing a hydrogen bond
O(51)-H(51)ꢀ ꢀ ꢀN(2) between the hydroxyl group and the nitrogen atom of the isoxazoline ring. This
means that the crystal structure of these compounds was governed by hydrogen bonds O–Hꢀ ꢀ ꢀN, involv-
ing the hydroxyl of the hemiacetal group and the nitrogen atom of the 4,5-dihydroisoxazole ring. This
interaction is relatively robust, showing a pattern in the crystal packing. Compounds 1–3 also have their
crystal stabilized by more weak interactions of type Clꢀ ꢀ ꢀCl, involving the chlorine atom of the trichloro-
methyl group.
Keywords:
4,5-Dihydroisoxazoles
X-ray structure
Hydrogen bonds
Theoretical calculation
Ó 2011 Elsevier B.V. All rights reserved.
1. Introduction
Over the last 20 years, our research group has developed synthe-
ses of new trihalomethyl substituted heterocycles by conventional
Isoxazoles and their derivatives have been of continuous inter-
est because of the numerous applications for such heterocycles in
the pharmaceutical [1–4] and agrochemical industries [1]. Gener-
ally, the pharmacological activity is closely related to the molecu-
lar recognition of the active substance by the active site. Therefore,
a detailed understanding of these molecular interactions is essen-
tial for the development of quantitative approaches to molecular
recognition. Recently, a specific molecular interaction where halo-
gen atoms act by their positive potential surface is under active
investigation [5]. Such interaction is now referred to as halogen
bonding to emphasize the characteristics in parallel with those of
hydrogen bonding in terms of strength and directionality [6].
method [7], using ionic liquid [8], microwave [9] or ultrasonic irra-
diation [10]. Recently, we have also directed our attention to the
understanding of molecular and supramolecular structural proper-
ties of these compounds and their role in biological systems.
Recently, we have revised the synthesis and spectroscopic prop-
erties of 5-halomethyl-5-hydroxy-4,5-dihydroisoxazoles [11].
Moreover, a structural characterization of these compounds from
X-ray data, including intermolecular interactions, has not yet been
performed. Intermolecular contacts which have been described for
similar compounds, such as 4,5-dihydroxy-5-methyl-3,4-diphenyl-4,
5-dihydroisoxazole (4) [12], 5-hydroxy-3,5-dimethylisoxazol-4
(5H)-one oxime (5) [13], 3-trifluoromethyl-5-hydroxy-5-(1,1-
dimethyl-2-ethoxyethyl)-4,5-dihydroisoxazoles (6) [14], 4,5-di
(4-iodophenyl)-5-hydroxy-4,5-dihydroisoxazoles (7) [15] and 3-
hydroxy-3-(1,1,2,2-tetrafluoroethyl)-3a,4,5,6-tetrahydro-3H-cycloi
soxazole (8) [16], are presented in Fig. 1.
⇑
Corresponding author. Tel./fax: +55 55 3220 8756.
E-mail address: mmartins@base.ufsm.br (M.A.P. Martins).
0022-2860/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2011.09.051

P.T. Campos et al. / Journal of Molecular Structure 1006 (2011) 462–468
463
R4'
O
R3
R4
HO
R5
N
Fig. 1. General structure for the studied 4,5-dihydroisoxazoles.
In this context, this work aims to describe the structure of three
new 4,5-dihydroisoxazoles (1–3) and establish a comparison with
the structures of already described 4,5-dihydroisoxazoles (4–8), in
order to identify and characterize the supramolecular syntons
present in these compounds (Fig. 1). Furthermore, the experimen-
tal geometrical data were compared with the results obtained from
semi-empirical molecular orbital methods RM1, AM1, PM3 and
DFT calculations.
2. Experimental
2.1. Synthesis
5-Trichloromethyl-5-hydroxy-4,5-dihydroisoxazoles(1–3)were
synthesized by cyclocondensation reaction of 1,1,1-trichloro-4-
methoxy-4-aryl(heteroaryl)-3-buten-2-ones with hydroxylamine
hydrochloride in the presence of pyridine. This reaction was carried
out in accordance with previously reported procedures [17]. The
crystals used for the data collection were obtained by re-crystalliza-
tion of compounds from methanol followed by slow evaporation at
room temperature.
2.2. Crystallographic data collection and refinement
X-ray data were collected on a Bruker SMART CCD diffractome-
ter [18]. Data were collected using graphite-monochromatized
Mo Ka radiation with k = 0.71073 Å. The structures were solved
with direct methods and refined on F2 by full-matrix least-squares
by SHELXL97 [19]. Hydrogen atoms were included in the refine-
ment in calculated positions with C–H distances of 0.93 Å (aro-
matic CH), 0.97 Å (methylene CH2), 0.98 Å (methine CH) and
0.82 Å (OH) using a riding model. The hydrogen isotropic thermal
parameters were kept equal to Uiso (H) c 1.5 Ueq for Csp3, and
1.2 for Csp2. Molecular graphics were prepared using ORTEP3 for
Windows [20]. The crystal data and details concerning data collec-
tion and structure refinement are given in Supplementary material.
3. Computational details
The geometry of compounds was optimized using semi-empir-
ical RM1, AM1 and PM3 methods implemented in the HyperChem
8.0.6 package [21]. The structures were full optimized without fix-
ing any parameter and, thus, bringing all geometric variables to
their equilibrium values. The energy minimization protocol em-
ployed the Polak-Ribiere conjugated gradient algorithm. Conver-
Fig. 2. A stereoview of part of the crystal structure for 1, 2 and 3, showing the
formation of a hydrogen-bonded chain and Clꢀ ꢀ ꢀCl interactions (For interpretation
of the references to colour in this figure legend, the reader is referred to the web
version of this article.).
gence to
a local minimum was achieved when the energy
gradient was 60.01 kcal molꢁ1
.
DFT calculations were carried out using Gaussian 03 package of
programs [22]. Dihedral angle H51-O51-C5-C51 of compounds 1–3
were changed from 0° to 360° in steps of 10° optimizing the geom-
etry at the HF/cc-pVTZ level. The geometry of the most stable con-
formers was fully re-optimized without any constrain at the
B3LYP/cc-pVTZ level of theory. All geometries were verified as
minima on the potential energy by calculating the Hessian matri-
ces by harmonic frequency calculations.
4. Results and discussion
Intermolecular contacts of 4,5-dihydroisoxazoles 4–7 found in
literature are characterized by hydrogen bonds invariably involv-
ing the hydroxyl group of hemiacetal. There is only one exception,
which is compound 5, where the hydroxyl oxime group is the par-
ticipant in the hydrogen bond. Compounds 4 and 8 present an

464
P.T. Campos et al. / Journal of Molecular Structure 1006 (2011) 462–468
Table 1
Hydrogen bond geometry for structures 1–3 [Å], °.
Comp.
D–Hꢀ ꢀ ꢀA
D–H
Hꢀ ꢀ ꢀA
Dꢀ ꢀ ꢀA
D–Hꢀ ꢀ ꢀA
Symmetry codes
1
2
3
O(51)–H(51) ꢀ ꢀ ꢀN(2)
O(51)–H(51) ꢀ ꢀ ꢀN(2)
O(51)–H(51) ꢀ ꢀ ꢀN(2)
0.78
1.00
0.820
2.05
1.83
2.04
2.827(2)
2.818(5)
2.8569(17)
174.5
172.8
175.5
ꢁx + 1/2, y ꢁ 1/2, ꢁz + 1/2
ꢁx + 1/2, y ꢁ 1/2, ꢁz + 3/2
ꢁx + 1/2, y + 1/2, ꢁz + 1/2
compound 5 had its crystal stabilized by a hydrogen bond between
the hydroxyl group of the oxime function and the nitrogen atom of
the 4,5-dihydroisoxazole ring with an OꢀꢀꢀN distance of 2.762 Å and
O–Hꢀ ꢀ ꢀN angle of 176.2°. Conversely, crystal structure of com-
pound 6 was stabilized by an O–Hꢀ ꢀ ꢀO interaction, between hydro-
xyl groups of alcohol function attached at 5-position of 4,5-
dihydroisoxazole.
Table 2
Halogen Bond Geometry for Structures 1–3 [Å], °.
Comp.
C–Clꢀ ꢀ ꢀCl
Clꢀ ꢀ ꢀCl
C–Clꢀ ꢀ ꢀCl
Symmetry codes
1
2
2
3
3
Cl(2)ꢀ ꢀ ꢀCl(2)
Cl(1)ꢀ ꢀ ꢀCl(1)
C–Cl(3)ꢀ ꢀ ꢀBr
Cl(2)ꢀ ꢀ ꢀCl(2)
Cl(1)ꢀ ꢀ ꢀCl(1)
3.285(1)
3.250(2)
3.584(0)
3.397(9)
3.407(11)
133.5
130.4
120.9
150.9
149.6
ꢁx + 1/2, ꢁy + 5/2, ꢁz + 1
ꢁx + 3/2, ꢁz + 2
ꢁx, ꢁy, 1 ꢁ z
ꢁx, ꢁy + 2, z
ꢁx, y, ꢁz + 1/2
Crystal structures of compounds 1–3, as described here for the
first time, were similar to those of compounds 4 and 7. In other
words, the crystal formation of these compounds was governed
by hydrogen bonds O-Hꢀ ꢀ ꢀN to 1–4 and O–Hꢀ ꢀ ꢀO to 7, involving
the hemiacetal hydroxyl group and the nitrogen or oxygen atom
of the 4,5-dihydroisoxazole ring. This interaction was relatively ro-
bust, showing a pattern in the crystal packing. Compounds 1–3
showed a hydrogen bond O(51)–H(51)ꢀ ꢀ ꢀN(2) between the hydro-
xyl group and the nitrogen atom of the heterocyclic ring, forming
infinite chains along one plane (Fig. 2). The contact distances
O(51)ꢀ ꢀ ꢀN(2) and contact angles O(51)–H(51)ꢀ ꢀ ꢀN(2) are depicted
in Table 1. The interatomic distances observed for these contacts
were shorter than the sum of van der Waals radii of the acceptor
and donor (Dꢀ ꢀ ꢀA) atoms of the proton involved in the interaction
[23]. Compounds 1–3 also had their crystal structures stabilized
by more weak interactions of type Clꢀ ꢀ ꢀCl, involving chlorine atoms
of the trichloromethyl group. Although the interatomic distances
are shorter than the sum of van der Waals radii of the chlorine
atom, this interaction in compounds 1–3 has a weak angular
directionality. Interatomic distances of Clꢀ ꢀ ꢀCl and interaction
angles are shown in Table 2. In addition, compound 2 has a Clꢀ ꢀ ꢀBr
Fig. 3. The two geometries for halogen bonds (R = organic group, X = Cl, Br and I).
intermolecular contact between the hydroxyl group of hemiacetal
and the nitrogen atom of the 4,5-dihydroisoxazole ring, with
O(51)-H(51)ꢀ ꢀ ꢀN(2) angles of 168° and 173.5°, respectively. Com-
pound 7 presents similar interaction; however, the hydroxyl
contact is with the oxygen atom of the 4,5-dihydroisoxazole ring.
Structure 7 was deposited in the CCDC, but without hydrogen
atoms and geometric parameters of this interaction, which were
not found in the publication [15]. As previously mentioned,
4'
R3
R
R4
OH
N
H
O
O
R4'
O
R3
R4
O
H
4'
N
R3
R
R4
HO
OH
N
H
O
O
R4'
R3
R4
H
O
N
O
HO
(a) Compounds 1-4 and 8
(c) Compound 6
(b) Compound 5
(d) Compound 7
Fig. 4. Representation of the supramolecular synthon found in compounds 1–8.

P.T. Campos et al. / Journal of Molecular Structure 1006 (2011) 462–468
465
Fig. 5. View of the asymmetric unit of compounds 1–3, showing the atom labeling scheme. Displacement ellipsoids are drawn at the 50% probability level. H atoms are
represented by circles of arbitrary radii.
intermolecular contact, whose interatomic distance is shorter than
the sum of van der Waals radii of chlorine and bromine atoms.
However, like the Clꢀ ꢀ ꢀCl interaction, this interaction has a weak
angular directionality, with a C(1)–Cl(3)ꢀ ꢀ ꢀBr angle of 120.9° and
C–Brꢀ ꢀ ꢀCl angle of 152.1°.
Table 3
Torsional angle and
DE of conformers determined by HF/cc-pVTZ level.
Comp. Conformer H(51)–O(51)–C(5)–C(51)
D
E
l (D)
Angle(°)
(kcal molꢁ1
)
The presence of these interactions involving halogen atoms in
compounds 1–3 leads us to discuss here their angular require-
ments. Covalently-bonded halogen interacted through positive
electrostatic regions with a negative site, such as a lone pair of a
Lewis base. In particular, in the bonded halogen atom, the aniso-
tropically distributed electron density resulted in a negative charge
concentrated in the equatorial area and a positive charge along the
1
2
3
I
II
I
II
Ia
Ib
IIa
IIb
64.6
305.3
63.3
305.7
64.8
64.6
0.00
4.00
0.00
4.02
0.00
1.11
4.07
5.07
2.66
4.20
1.64
3.33
3.08
2.66
4.68
4.10
305.6
304.9
C-X bond, which is called a
r-hole [6a]. This leads to two preferred
Fig. 6. Energy scan for the dihedral angle H(51)–O(51)–C(5)–C(51) from 0° to 360° for compound 1.

466
P.T. Campos et al. / Journal of Molecular Structure 1006 (2011) 462–468
Compound 1, conformer I
Compound 1, conformer II
Compound 2, conformerI
Compound 2, conformer II
Compound 3, conformer Ia
Compound 3, conformer IIa
Compound 3, conformer Ib
Coampound 3, conformer IIb
Fig. 7. The six optimized structures.
geometries for halogenꢀ ꢀ ꢀLewis base or halogenꢀ ꢀ ꢀhalogen contacts
(Fig. 3). The first arrangement (Type I, Fig. 3) occurs when h1 = h2,
(where h1 and h2 are the R-X1ꢀ ꢀ ꢀX2 and X1ꢀ ꢀ ꢀX2-C angles, respec-
tively). The second geometry (Type II, Fig. 3) arises when
h1 = 180° and h2 = 90°; the perpendicular arrangement. Therefore,
from the angular information selected in Table 2, it is possible to
state that the halogen bonds (Clꢀ ꢀ ꢀCl and Clꢀ ꢀ ꢀBr) observed in com-
pounds 1–3 were similar to those geometries shown in Type I
drawing in Fig. 3. In other words, halogenꢀ ꢀ ꢀhalogen contacts found
in compounds 1–3 have a weak angular directionality, and h1 was
not exactly equal to h2. Crystal stabilization of compound 3, as in
compounds 1 and 2, has an extra interaction, where the sulfur
atoms were responsible for the extra stabilization, rather than
the halogen atoms. The interaction of the type Sꢀ ꢀ ꢀS had an inter-
atomic distance of 3.437(9) Å for S(32)ꢀ ꢀ ꢀS(32) (ꢁx + 1/2, ꢁy + 3/
2, ꢁz + 1) and C(31)–S(32)ꢀ ꢀ ꢀS(32) of 163.9°. This Sꢀ ꢀ ꢀS interaction,
in compound 3, different from the halogenꢀ ꢀ ꢀhalogen contacts
found in compounds 1–3, presents an angle closer to 180°. This
interaction, similar to the halogenꢀ ꢀ ꢀhalogen interaction, can be ex-
plained by the non-spherical distribution of charge in the potential
supramolecular synthon O(51)–H(51)ꢀ ꢀ ꢀN(2) was sensitive to the
addition of a substituent with a hydroxyl group in some position
of the ring (4 and 5), as an oxime (compound 5) and an alcohol
(compound 6). On the other hand, this supramolecular synthon
was not altered by the presence of interactions involving haloge-
nated groups or atoms and sulfur atoms. This can be explained
by the higher ability of oxygen atoms to participate in intermolec-
ular interactions, which is a consequence of the higher dipole in
oxygen atoms, in relation to halogen or sulfur atoms. In this case,
halogenꢀ ꢀ ꢀhalogen (1–3) and sulfur–sulfur (3) interactions perform
an auxiliary role in the extra stabilization of the crystal in com-
pounds 1–3.
We consider it to be worth mentioning some geometric data for
compounds 1–3, which is characterized by the molecular structure
represented by ORTEP in Fig. 5. The five-membered 4,5-dihydrois-
oxazole ring was found as an essentially planar structure for com-
pounds 1–3. These data are in full agreement with compounds 4–8
[12–16]. The C(4)–C(3)–C(31)–C(36) torsion angles made by phe-
nyl and 4-bromophenyl show that these groups are nearly planar
to the 4,5-dihydroisoxazole system. The C(4)–C(3)–C(31)–C(36)
torsion angles were of ꢁ0.4(3)° and 0.8(6)° for compounds 1 and
2, respectively. Similarly, the 4,5-dihydroisoxazole 3 makes C(4)–
C(3)–C(31)–C(35) torsion angle of ꢁ3.0(3)°, demonstrating that
the thien-2-yl group shares the same plane of the 4,5-dihydroisox-
azole ring. This planarity between the 4,5-dihydroisoxazole and its
substituent in the 3-position is associated with the electronic res-
onance between the N(2)@C(3) bond of the heterocycle and its
substituents attached at the 3-position.
surface (
dance, the sulfur atom has positive points in the potential surface
-hole), one for each covalent bond formed. Therefore, from
r-hole) of the covalently bonded sulfur atom. In accor-
(r
structural X-ray analysis of compounds 1–8, it is clear that the
O(51)–H(51)ꢀ ꢀ ꢀN(2) intermolecular hydrogen bond is an important
supramolecular synthon in 4,5-dihydroisoxazoles, due to its
robustness, demonstrated by its presence in five of the eight com-
pounds described here (Fig. 4). In addition, it was found that the

P.T. Campos et al. / Journal of Molecular Structure 1006 (2011) 462–468
467
Table 4
Experimental data and data calculated by RM1, AM1, PM3 and B3LYP/cc-pvtz level for
bond lengths and bond angles.
Bond length
X-ray
RM1
AM1
PM3
B3LYPa
Compound 1
O(1)–N(2)
C(3)–N(2)
C(4)–C(3)
C(4)–C(5)
C(5)–O(1)
C(5)–O(51)
C(51)–C(5)
C(31)–C(3)
1.416(2)
1.289(3)
1.493(3)
1.524(3)
1.457(3)
1.368(3)
1.556(3)
1.460(3)
1.351
1.325
1.496
1.539
1.423
1.373
1.514
1.440
1.314
1.325
1.524
1.549
1.477
1.393
1.547
1.454
1.408
1.309
1.508
1.551
1.445
1.395
1.544
1.460
1.405
1.280
1.505
1.528
1.432
1.389
1.566
1.466
Compound 2
O(1)–N(2)
C(3)–N(2)
C(4)–C(3)
C(4)–C(5)
C(5)–O(1)
C(5)–O(51)
C(51)–C(5)
C(31)–C(3)
1.412(4)
1.277(5)
1.485(5)
1.531(5)
1.460(5)
1.370(5)
1.549(6)
1.470(5)
1.350
1.325
1.496
1.539
1.424
1.372
1.513
1.440
1.313
1.325
1.523
1.550
1.479
1.393
1.547
1.454
1.406
1.309
1.508
1.552
1.446
1.395
1.543
1.461
1.401
1.280
1.505
1.529
1.434
1.388
1.566
1.465
Fig. 8. Plot of experimental and theoretical bond lengths for compounds 1–3. (For
interpretation of the references to colour in this figure legend, the reader is referred
to the web version of this article.)
Compound 3
O(1)–N(2)
C(3)–N(2)
C(4)–C(3)
C(4)–C(5)
C(5)–O(1)
C(5)–O(51)
C(51)–C(5)
C(31)–C(3)
1.416(19)
1.283(2)
1.494(2)
1.537(2)
1.455(2)
1.365(2)
1.546(2)
1.440(2)
1.350
1.325
1.496
1.539
1.424
1.372
1.513
1.440
1.313
1.325
1.523
1.550
1.479
1.393
1.547
1.454
1.406
1.309
1.508
1.552
1.446
1.395
1.543
1.461
1.401
1.280
1.505
1.529
1.434
1.388
1.566
1.465
to the second more stable structure represented by conformer II.
Compound 3, due to the additional presence of the thien-2-yl sub-
stituent, shows four possible structures. Calculated relative ener-
gies show that conformer Ia with the sulfur and oxygen atoms on
the same side was the most stable structure by about 1 kcal molꢁ1
in relation to the other possible structures (Fig. 7). An explanation
of this behavior was given by Lombardo et al. to similar found
[24] in terms of the lack of ability, in the gas phase, to form intramo-
lecular interactions, as the molecule is isolated and not interacting
with other molecules at all. They rationalized that when the mole-
cule is in the bulk, several other molecules surround it and the
intermolecular interactions control the mutual spatial distribution.
In the solid state, where several other interactions contribute to fi-
nal balance of the structure, the intramolecular interaction found in
gas phase is not so relevant. Thus, we believed that this can be
occurring in our study. Although the discrepancies between con-
formers found in gas or solid phase, the geometry of conformers
was so similar. The selected bond lengths and bond angles for the
more stable conformers of compounds 1–3, obtained from semi-
empirical and DFT calculations and from X-ray diffractometry are
presented in Table 4. The simple linear regressions between exper-
imental bond lengths and bond angles from X-ray (average to com-
pounds 1–3) with semi-empirical calculated data (RM1, AM1 and
PM3) and DFT calculations showed good and similar correlation
coefficients. The correlation of semi-empirical and DFT methods
with experimental data showed values of r = 0.925, 0.879, 0.985
and 0.988, for bond lengths, and r = 0.936, 0.942, 0.925, 0.950 for
bond angles, obtained by the methods RM1, AM1 PM3 and DFT,
respectively. In addition, calculated and experimental data were
also analyzed statistically. The sum of bond length errors resulted
in positive values for the RM1 method and negative values for the
others. This indicates that the RM1 method provided longer bond
lengths than did the experimental data, while the other methods
described shorter lengths than did the experimental data. The aver-
age unsigned bond length errors also resulted in positive values for
the RM1 method and negative values for the others, with AM1 and
DFT yielding the lowest errors. In addition, inspection of Fig. 8 re-
veals that DFT was the method that best described the bond length
of compounds 1–3 and the AM1 method was the method that most
failed to describe the O(1)–N(2) bond (Fig. 8).
Bond angle
Compound 1
C(3)–N(2)–O(1)
N(2)–C(3)–C(4)
C(3)–C(4)–C(5)
O(1)–C(5)–C(4)
N(2)–O(1)–C(5)
N(2)–C(3)–C(31)
O(1)–C(5)–C(51)
O(51)–C(5)–O(1)
O(51)–C(5)–C(51)
109.90(18)
112.9(2)
102.85(17)
104.28(17)
109.9(3)
110.5
112.2
100.1
106.1
110.9
123.7
108.4
111.9
107.5
111.9
111.7
100.4
104.1
111.8
126.5
109.6
113.6
103.7
112.4
111.5
101.3
106.5
108.2
124.2
109.8
112.2
104.7
109.8
113.2
100.6
105.1
109.8
121.7
106.7
110.6
110.7
121.7(2)
105.14(17)
109.32(18)
105.96(18)
Compound 2
C(3)–N(2)–O(1)
N(2)–C(3)–C(4)
C(3)–C(4)–C(5)
O(1)–C(5)–C(4)
N(2)–O(1)–C(5)
N(2)–C(3)–C(31)
O(1)–C(5)–C(51)
O(51)–C(5)–O(1)
O(51)–C(5)–C(51)
110.2(3)
113.3(3)
102.8(3)
103.7(3)
109.9(3)
121.0(4)
105.4(3)
109.9(3)
105.8(3)
110.6
112.2
100.0
106.1
110.9
123.6
108.3
112.1
107.5
111.9
111.8
100.4
104.1
111.8
126.4
109.6
113.7
103.6
112.5
111.4
101.3
106.4
108.2
124.2
109.8
112.3
104.7
109.9
113.3
100.6
105.0
109.9
121.5
106.6
110.7
110.6
Compound 3
C(3)–N(2)–O(1)
N(2)–C(3)–C(4)
C(3)–C(4)–C(5)
O(1)–C(5)–C(4)
N(2)–O(1)–C(5)
N(2)–C(3)–C(31)
O(1)–C(5)–C(51)
O(51)–C(5)–O(1)
O(51)–C(5)–C(51)
109.97(13)
113.54(15)
102.11(13)
104.35(12)
110.01(12)
122.62(15)
105.83(13)
105.83(13)
110.12(14)
110.2
112.5
99.9
106.2
111.0
124.6
108.3
112.0
107.5
111.7
112.0
100.2
104.3
111.8
126.6
109.6
113.6
103.6
112.2
111.7
101.1
106.5
108.3
123.8
109.8
112.2
104.6
109.5
113.6
100.3
105.2
109.9
122.2
106.7
110.6
110.7
a
Basis set cc-PVTZ was used.
We have complemented the structural analysis of compounds
1–3 with theoretical calculations. DFT calculated relative energies
of the major possible conformers of 1–3 compounds are given in Ta-
ble 3. The DFT calculated relative energies can be achieved by an en-
ergy scan of 0° to 360° for dihedral angle H(51)–O(51)–C(5)–C(51)
(Fig. 6). The so resulting most stable conformers in the gas phase are
not present in the solid phase. In the gas phase, the energy scan
shows that for compounds 1 and 2, conformer I was the most stable
structure by a difference in energy of about 4 kcal.molꢁ1 in relation
The sum of bond angle errors resulted in negative values for all
methods, showing that the bond angle was described as smaller
than that from experimental data, and DFT was the method that
best described the bond angle for compounds 1–3.

468
P.T. Campos et al. / Journal of Molecular Structure 1006 (2011) 462–468
[3] B. Frølund, L.S. Jensen, S.I. Storustovu, T.B. Stensbøl, B. Ebert, J. Kehler, P.
5. Conclusions
Krogsgaard-Larsen, T. Liljefors, J. Med. Chem. 50 (2007) 1988–1992.
[4] D.A. Patrick, S.A. Bakunov, S.M. Bakunova, E.V.K.S. Kumar, R.J. Lombardy, S.K.
Jones, A.S. Bridges, O. Zhirnov, J.E. Hall, T. Wenzler, R. Brun, R.R. Tidwell, J. Med.
Chem. 50 (2007) 2468–2485.
[5] Y. Lu, T. Shi, Y. Wang, H. Yang, X. Yan, X. Luo, H. Jiang, W. Zhu, J. Med. Chem. 52
(2009) 2854–2862.
[6] (a) P. Politzer, J.S. Murray, M.C. Concha, J. Mol. Model. 14 (2008) 659–665;
(b) Z.P. Shields, J.S. Murray, P. Politzer, Int. J. Quantum Chem. 110 (2010) 2823.
[7] M.A.P. Martins, W. Cunico, C.M.P. Pereira, A.F.C. Flores, H.G. Bonacorso, N.
Zanatta, Curr. Org. Synth. 1 (2004) 391–403.
[8] (a) D.N. Moreira, K. Longhi, C.P. Frizzo, H.G. Bonacorso, N. Zanatta, M.A.P.
Martins, Catal. Commun. 11 (2010) 476–479;
In summary, the crystal structure of 5-trichloromethyl-5-hydro-
xy-4,5-dihydroisoxazoles 1–3 has been characterized and dis-
cussed together with similar structures (4–8) described in the
literature and it was shown that the supramolecular auto-organiza-
tion of 5-hydroxy-4,5-dihydroisoxazoles is characterized by hydro-
gen bonds invariably involving the hydroxyl of the hemiacetal
group. There was only a small substituent effect at the 4-position
of 5-hydroxy-4,5-dihydroisoxazoles when the substituent was an
oxime group. In other words, the crystal of these compounds was
governed by hydrogen bonds O–Hꢀ ꢀ ꢀN (O), involving the hydroxyl
group of hemiacetal and the nitrogen/oxygen atom of the 4,5-
dihydroisoxazole ring. Therefore, this interaction could be consid-
ered a relatively robust supramolecular synthon showing a pattern
in the crystal packing for the eight 5-hydroxy-4,5-dihydroisoxaz-
oles with different substitutions. In addition, the crystal structure
of compounds 1–3 also had its crystal stabilized by more weak
interactions of type Clꢀ ꢀ ꢀCl, involving the chlorine atom of the tri-
chloromethyl group. We believe that the Clꢀ ꢀ ꢀCl interaction, in
these three compounds, could also be considered a supramolecular
synthon, due to its presence in all three compounds substituted by
the chlorinated group. The semi-empirical and DFT calculations
gave similar data in comparison with the experimental data for
bond lengths and angles, where DFT calculations resulted in better
correlations in all cases.
(b) C.P. Frizzo, M.R.B. Marzari, L. Buriol, D.N. Moreira, F.A. Rosa, P.S. Vargas, N.
Zanatta, H.G. Bonacorso, M.A.P. Martins, Catal. Commun. 10 (2009) 1967–
1970.
[9] (a) L. Buriol, C.P. Frizzo, M.R.B. Marzari, D.N. Moreira, L.D.T. Prolla, N. Zanatta,
H.G. Bonacorso, M.A.P. Martins, J. Braz. Chem. Soc. 21 (2010) 1037–1044;
(b) M.A.P. Martins, P. Beck, D.N. Moreira, L. Buriol, C.P. Frizzo, N. Zanatta, H.G.
Bonacorso, J. Heterocycl. Chem. 47 (2010) 301–308.
[10] (a) P. Machado, G.R. Lima, M. Rotta, H.G. Bonacorso, N. Zanatta, M.A.P. Martins,
Ultrason. Sonochem. 18 (2011) 293–299;
(b) G. Sant’anna, P. Machado, P. Sauzem, F. Rosa, M. Rubin, J. Ferreira, H.G.
Bonacorso, N. Zanatta, M.A.P. Martins, Biorg. Med. Chem. Lett. 19 (2009) 546–
549.
[11] M.A.P. Martins, P. Machado, F.A. Rosa, W. Cunico, H.G. Bonacorso, N. Zanatta,
Mini-Rev. Org. Chem. 5 (2008) 53–76.
[12] M. Piloche-Levisalles, Christian Bellec, J. Heterocycl. Chem. 17 (1980) 1537–
1539.
[13] R. Fruttero, R. Calvino, B. Ferrarotti, A. Gasco, S. Aime, R. Gobetto, G. Chiari, G.
Calestani, J. Chem. Soc. Perkin Trans. II (1987) 523–529.
[14] V.Y. Sosnovskikh, M.Y. Mel’nikov, S.A. Pogozhikh, Russ. Chem. Bull. 48 (1999)
1323–1328.
[15] R. Olivera, R. SanMartin, E. Domínguez, X. Solans, M.K. Urtiaga, M.I. Arriortua, J.
Org. Chem. 65 (2000) 6398–6411.
[16] D.V. Sevenard, O.G. Khomutov, K.I. Pashkevich, E. Lork, G. Röschenthalera,
Helv. Chim. Acta 85 (2002) 1960–1972.
Acknowledgements
[17] (a) M.A.P. Martins, G.M. Siqueira, G.P. Bastos, H.G. Bonacorso, N. Zanatta, J.
Heterocycl. Chem. 33 (1996) 1619–1622;
(b) A.F.C. Flores, S. Brondani, L. Pizzuti, M.A.P. Martins, N. Zanatta, H.G.
Bonacorso, D.C. Flores, Synthesis (2005) 2744–2750.
[18] Bruker (2006).APEX2 (Version 2.1), COSMO (Version 1.56), BIS (Version
The authors thank the Conselho Nacional de Desenvolvimento
Científico e Tecnológico – CNPq (Universal/Proc. 485893/2007-0;
Universal/Proc. 471519/2009-0; MAPA/Proc. 578426/2008-0) and
the Fundação de Amparo à Pesquisa do Estado do Rio Grande do
Sul – FAPERGS (PRONEX/Proc. 10/0037-8) for financial support.
The fellowships from CNPq (M.A.P.M., N.Z., H.G.B., D.N.M., P.T.C.,
M.R.B.M.), CAPES (C.P.F., PRODOC/Proc. 2684-32/2010) and FA-
PERGS are also acknowledged.
2.0.1.9), SAINT (Version 7.3A) and SADABS (Version 2004/1)
(Version 2005/4). Bruker AXS Inc., Madison, Wisconsin, USA.
[19] G.M. Sheldrick, Acta Crystallogr A64 (2008) 112–122.
[20] L.J. Farrugia, J. Appl. Cryst. 30 (1997) 565.
& XPREP
[21] HYPERCHEM(TM) Professional 8.0.6 (2008), Hypercube, Inc.,1115 NW 4th
Street, Gainesville, FL 32601, USA.
[22] Gaussian 03, Revision E.01, 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, 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. Al-
Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W.
Chen, M.W. Wong, C. Gonzalez, and J.A. Pople, Gaussian, Inc., Wallingford CT,
2004.
Appendix A. Supplementary material
CCDC numbers of compounds 1–3 is 681049, 681048 and
681050, respectively. These data can be obtained free of charge at
www.ccdc.cam.ac.uk/conts/retrieving.html [or from the Cambridge
Crystallographic Data Centre (CCDC), 12 Union Road, CambridgeCB2
1EZ, UK; fax:+44(0) 1223 336 033; e-mail: deposit@ccdc.cam.ac.uk].
Supplementary data associated with this article can be found, in the
online version, at doi:10.1016/j.molstruc.2011.09.051.
[23] (a) L. Sobczyk, S.J. Grabowski, T.M. Krygowski, Chem. Rev. 105 (2005) 3513–
3560;
References
(b) P. Pyykko, Chem. Rev. 97 (1997) 597–636.
[24] G.M. Lombardo, G. Portalone, M. Colapietro, A. Rescifina, F. Punzo, J. Mol.
Struct. 994 (2011) 87–96.
[1] R.E. Watts, M. Siegel, C. Khosla, J. Med. Chem. 49 (2006) 7493–7501.
[2] P. Conti, M. de Amici, G. Grazioso, G. Roda, A. Pinto, Kbø Hansen, B. Nielsen, U.
Madsen, H. Bräuner-Osborne, J. Egebjerg, V. Vestri, D.E. Pellegrini-Giampietro,
P. Sibille, F.C. Acher, C. de Micheli, J. Med. Chem. 48 (2005) 6315–6325.