organic compounds
ꢀ-electron delocalization in chemical compounds. It is based
on the geometric criterion of aromaticity, which stipulates that
bond lengths in aromatic systems lie between values that are
typical for single and double bonds (Kruszewski & Krygowski,
1973; Krygowski, 1993). Therefore, HOMA = 0 for a model
´
non-aromatic system (e.g. the Kekule structure of benzene)
and HOMA = 1 for a system with all bonds equal to the
optimal value, assumed to be realised for fully aromatic
systems. The HOMA value (based on B3LYP/6-311++G**
optimized geometries) calculated for ring A (0.960) is lower
than that of ring C (0.979). This loss of aromaticity of ring A is
caused by both the electron-withdrawing properties of the
neighbouring carbonyl group and the above-mentioned
interactions with the quasi-rings formed by intramolecular
hydrogen bonds, in particular by C13—H13ꢀ ꢀ ꢀO7. Breaking
the intramolecular hydrogen bonds by a twist of 90ꢁ around
the C5—C8 bond results in an increase in energy of
5.65 kcal molꢂ1 (1 kcal molꢂ1 = 4.184 kJ molꢂ1), based on the
B3LYP/6–311++G** calculations, as well as an increase in the
aromaticity of ring A (HOMA = 0.989). The aromaticity of
ring C remains almost unchanged (HOMA = 0.978).
Figure 2
Similar sequences of values were obtained for a twist about
the C3—C14 bond. In this case, the increase in energy is
5.22 kcal molꢂ1, and the HOMA value rises to 0.988 for ring C
and decreases to 0.958 for ring A. Finally, in a hypothetical
conformation without intramolecular hydrogen bonds, viz.
with both phenyl groups perpendicular to the triazine ring, the
energy is higher by 11.75 kcal molꢂ1 and the HOMA index is
the same for rings A and C (0.989).
A packing diagram for (I), showing the N1—H1ꢀ ꢀ ꢀO7i hydrogen bonds as
dashed lines. [Symmetry code: (i) ꢂx + 1, ꢂy + 2, ꢂz + 1.]
atoms N1/N2/C3/N4/C5/C6) and C (the phenyl ring containing
atoms from C14–C19). These are nearly coplanar in the
crystalline state, with the angles between the ring planes for
A/B and B/C being similar [8.6 (2) and 8.4 (2)ꢁ, respectively].
The A and C planes are arranged in a mutually cis position.
The twists (described by torsion angles) around the C3—C14
and C5—C8 bonds are less than 10ꢁ (Table 1). These small
deformations from planarity probably result from the inter-
molecular interactions present in the crystal lattice. A density
functional theory (DFT) study predicts a completely planar
conformation (Cs point-group symmetry) as the preferred one
for the isolated molecule of (I). See Supplementary materials
for further details of the DFT calculations.
Note that there are no significant differences between the
values of the bond lengths and angles of (I) in the solid state
and those found for the calculated planar structure;ꢁ the
˚
differences do not exceed 0.02 A for bond distances and 2 for
bond angles. All bond distances and angles are normal
(Table 1) and are in good agreement with the geometry of
other 3,5-disubstituted 1,2,4-triazin-6(1H)-one derivatives
(Buscemi et al., 2006; Garg & Stoltz, 2005; Sanudo et al., 2006;
´ ´ˇ
Travnıcek et al., 1995).
The near planarity of the system favours the formation of
intramolecular hydrogen bonds and ꢀ-electron delocalization.
The molecular structure of (I) contains four weak intra-
molecular hydrogen bonds (Fig. 1 and Table 2); one C—Hꢀ ꢀ ꢀO
interaction, which forms a six-membered ring, and three C—
Hꢀ ꢀ ꢀN interactions that form five-membered rings, denoted
quasi-rings. The resulting rings can be investigated as mol-
ecular patterns of intramolecular resonance assisted by
hydrogen bonds. The position of the extra ring formed by the
substituent interacting through the hydrogen bond is found to
influence both the strength of that hydrogen bond and the
local aromaticity of the polycyclic aromatic hydrocarbon
skeleton. Relatively speaking, a greater loss of aromaticity of
the ipso-ring (phenyl ring) can be observed for these kinked-
like structures because of the greater participation of ꢀ-elec-
trons from the ipso-ring in the formation of the quasi-ring
(Krygowski et al., 2010; Palusiak et al., 2009).
Experimental
d-(ꢂ)-ꢁ-Phenylglycine hydrazide (3.30 g, 20 mmol) was added to a
mixture of triethyl orthobenzoate (4.57 g, 20 mmol) and p-toluene-
sulfonic acid (0.1 g) in xylene (20 ml) and the resulting solution kept
under reflux for 3 h (monitored by thin-layer chromatography). After
cooling, the mixture was washed with water (30 ml), dried over
MgSO4 and then concentrated under reduced pressure. The oily
residue was subjected to column chromatography (silica gel; eluent:
hexane–AcOEt, 1:2 v/v), yielding 3,5-diphenyl-4,5-dihydro-1,2,4-tri-
azin-6(1H)-one (2.60 g). This crude product was dissolved in ethanol
(50 ml) and left in solution for 10 d at room temperature. Yellow
needles of (I) were filtered off and dried in air [yield 0.35 g, 14%; m.p.
494–495 K, reference 491–493 K (Camparini et al., 1978)]. Due to the
fact that the oxidation of 3,5-diphenyl-4,5-dihydro-1,2,4-triazin-
6(1H)-one occurred simultaneously with the crystallization of (I), the
crystals of (I) obtained were of poor quality (cracked) and weakly
diffracting.
The harmonic oscillator model of aromaticity (HOMA) is a
leading method for the quantitative determination of cyclic
ꢃ
o150 Ejsmont et al. C15H11N3O
Acta Cryst. (2012). C68, o149–o151
Ejsmont, Krzysztof
