S.I. Vdovenko et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 103 (2013) 368–377
371
2
D spectrum was achieve when two overlapped bands changing
in position and intensity in opposite directions (similar to those
in Fig. 7 of Supplementary Part) were used. The
2.4%
m
~ðC@CÞ vibrations
4.9%
5.1%
Me
ꢀ1
were represented as intensive band at ca. 1555 cm , nevertheless,
Me Me
N
similarly to 2, analysis of 2D correlation spectrum of 3 revealed
existence of two bands at 1562 and 1555 cm , respectively. Com-
H
1
Me
COCF3
N
Me
COCF3
4
parison of spectra of the 3 in CCl and acetonitrile showed that
H
ꢀ1
Me
the intensities of pair of the bands at ca. 1664 and 1558 cm in-
creased when solvent polarity increased, whereas the intensities
0
.5%
X
ꢀ1
of the pair of the bands at ca. 1656 and 1565 cm decreased under
these circumstances. Moreover, the intensities of the former pair of
the bands decreased when temperature increased with simulta-
neous intensity decrease of the latter pair of bands. Starting from
results of quantum chemical calculations (comparison of relative
energies of conformers in Table 1), Taylor and Smith criteria
[15,17] and temperature investigations we ascribed the bands at
NOE-data of the enaminone 2
C-Me irradiated
NOE-data of the enaminone 3
=CH irradiated
=
Fig. 1. NMR NOE investigations of the enaminoketones 2 and 3.
the corresponding 1D spectrum [9]. The intensity changes, band
shifts, and changes in band shapes are typical spectral variations
observed under dilution and temperature changes for fluorinated
enaminoketones [5,6]. The effects of complex spectral feature vari-
ations, such as band shift and line intensity changing, are best visu-
alized with simulated model data emphasizing the specific aspect
of spectral changes. Fig. 2b shows results of simulation studies of
the case, where position shift of two bands is coupled with simul-
ꢀ1
1664 and 1558 cm to the (E–s–Z) isomer and the bands at ca.
ꢀ1
1656 and 1565 cm to the (Z–s–Z) isomer of the 3.
Results of quantum chemical calculations
As it can be clearly seen from Table 1, that relative energies (the
difference between the energies of the (E–s–E) and (E–s–Z) con-
former) calculated with DFT method including different basis sets
ꢀ
1
taneous intensity decrease of the band at 1692 cm and intensity
increase of the band at 1660 cm1 (similarly to those depicted in
Fig. 5a and b of Supplementary Part). The same simulation was
⁄⁄
⁄⁄
(viz. B3LYP/6-31G and B3LYP/6-311G set) are quite close. The
energy of the (E–s–E) conformer to be higher than the energy of
the (E–s–Z) conformer for 1, whereas energies of the (E–s–Z) and
(E–s–E) conformers of 2 are practically the same, although energy
of the (E–s–E) conformer is somewhat higher. Moreover, the (E–s–
Z) and (E–s–E) conformers of 1 and 2 are practically planar,
whereas the (Z–s–Z) and (Z–s–E) conformers are considerably dis-
torted, especially the conformers of 2. Energy of the (E–s–Z) isomer
of enaminoketone 3 is 13 kJ/mol lower than energy of the (Z–s–Z)
isomer, whereas relative energies of the (E–s–E) and (Z–s–E) isomer
are much higher than RE of the (E–s–Z) and (Z–s–Z) isomer. The (E–
made for
b of Supplementary Part). It is apparent from the Figs. 2 and 3 that
both
~ðC@OÞ and ~ðC@CÞ band consists of two overlapped bands
which position and intensity varies during dilution, therefore we
fitted the profile of
~ðC@OÞ and ~ðC@CÞ vibrations with two Lorentz
bands (see Fig. 1 of Supplementary). As contrasted to bands at
m
~ðC@CÞ vibrations (using bands like those in Fig. 6a and
m
m
m
m
ꢀ1
ꢀ1
1
672, 1660, and 1605 cm the band at 1564 cm did not change
in intensity under temperature or concentration variation there-
fore, similarly to the 1, we ascribed this band to the
½
m
~ðC ꢀ NÞ þ
bers in view of mass increase of CH@C(CH
rise resulted in intensity increase of the pair of bands at 1660
m
~ðC@CÞꢂ vibration which is shifted to lower wavenum-
s–E) isomer of
3
is significantly distorted (dihedral angle
) of the
3
) moiety. Temperature
u
= 154.1°). Although evaluation of a dihedral angle (u
⁄⁄
⁄⁄
(Z–s–E) isomer of 3 with basis set 6-311G and 6-31G gives dif-
ferent results (176.5° and 150.9°, respectively), nevertheless it is
clear that this isomer is also distorted.
ꢀ1
and 1605 cm with simultaneous intensity decrease of the pair
ꢀ1
of bands at 1672 and 1598 cm The opposite trend was observed
when passing from non polar carbon tetrachloride to polar aceto-
nitrile: intensities of the bands at 1672 and 1598 cm increased,
According to experimental data for the enaminoketone 1 the
(E–s–E) conformer is more stable in comparison with the (E–s–Z)
conformer in contrast with results of DFT calculations though. In
previous work [18] we showed that this discrepancy between re-
sults of spectroscopic measurements and quantum chemical calcu-
lations is a consequence of two reasons: first – a hydrogen bond
formation (including intra- and intermolecular H-bonds in self-
association) and dispersion interactions; second – an additional
influence of the solvent polarity (particularly, dramatic variation
of er when total enaminoketone concentration rises). The same is
true for enaminoketone 3, where (Z–s–Z) isomer is more stable
than (E–s–Z) isomer (spectroscopic measurements) although as
per DFT calculations this stability is reversed (vide supra). More de-
tailed discussion of this problem is done in Discussion part of this
work. In contrast to 1 and 3 the most stable conformer of 2 is (E–s–
E). Despite the fact that energy of the (E–s–Z) conformer is very
close to that of the (E–s–E) conformer (see above) this result of
DFT calculations conforms to experimental data. As it follows from
Table 1 the (E–s–E) and (E–s–Z) conformers of 2 are the least dis-
torted structures with dihedral angle of the C@CAC@O moiety clo-
sely approximated to 180°.
ꢀ1
ꢀ1
whereas intensities of the bands at 1660 and 1605 cm decreased
see e.g. Figs. 2 and 3 of Supplementary). Basing on concentration
and temperature investigations we attributed the bands at 1672
(
ꢀ
1
and 1598 cm to the
m
~ðC@OÞ and ~ðC@CÞ vibrations of the (E–s–
m
ꢀ1
Z) and the bands at 1660 and 1605 cm
to the
m
~ðC@OÞ and
m
~ðC@CÞ vibrations of the (E–s–E) conformer, respectively (Table 2).
It is worth to note that according to NMR data and results of quan-
tum chemical calculations there is possibility for existence of a (Z)
conformer, namely, the (Z–s–Z), to which we attributed the bands
ꢀ
1
ꢀ1
with very low intensity at 1698 cm and 1628 cm (the
and
m
~ðC@OÞ
m
~ðC@CÞ vibrations, respectively). To this attribution we came
from the results of our calculations according to which the system
C@CAC@O of the (Z–s–Z) isomer was significantly distorted (dihe-
dral angle C@CAC@O is equal 21°, Table 1), hence the conjugation
between C@O and C@C double bonds is weakened and the
m
~ðC@OÞ
and
m
~ðC@CÞ vibrations are more isolated. Since the quantity of this
isomer was negligible and changed insignificantly under tempera-
ture investigations, we excluded it from our considerations.
In polar solvents IR-spectra of the enaminone 3 also comprised
two distinct
m
~ðC@OÞ bands (see, e.g. Fig. 5), although in non polar
carbon tetrachloride those bands are strongly overlapped. Fig. 4
Solvent influence on IR spectra of the enaminoketones
shows 2D correlation IR-spectrum of (3) under dilution in the re-
gion of
m
(C@O) vibrations. Accordance between 2D correlation
Solvent effects on the
gated in twelve various solvents and wavenumbers obtained are
m
~ðC@OÞ and ~ðC@CÞ bands were investi-
m
spectrum in the region of the
m
~ðC@OÞ vibrations and simulated
Vdovenko, Sergey I.
