310
M. A. Czarnecki et al.
normal coordinate corresponding of ν(OH) mode. The potential energy in each point
was obtained at the B3LYP/6-311++G(3df,3pd) level employing the IPCM solvation
model. Subsequently, to obtain vibrational states, the one-dimensional Schrödinger
equation was solved using Johnson’s reformulation of Numerov’s approach. This
approach yielded the vibrational levels at the accuracy level exceeding 0.001 cm
−1
with respect to the determined vibrational potential, and allowed to avoid approximations such as the fitting of a Morse function [2]. The corresponding transition
intensities were obtained from the integrated absorption coefficient (km mol
−1 , base
e). The accurate calculation of wavenumbers and intensities of the ν(OH), 2ν(OH)
and 3ν(OH) bands enabled a direct interpretation of the observed intensity variations
as well as respective solvent dependency. The calculations reproduced the observed
“parity” in the intensity of the ν(OH) bands ν 01 , ν 02 , ν 03 , and ν 04 in phenols as well
as the respective solvent dependency. The parity effect was notably more prominent
for phenol than for 2,6–dihalogenated phenols. It was concluded that this difference
results from phenol having a stronger intermolecular hydrogen bonding in contrast to
its derivatives that feature a weaker intramolecular hydrogen bond. By this reasoning,
it was suggested that the intermolecular hydrogen bond between the OH group and
the Cl atom is responsible for the observed tendencies [2]. The electrical anharmonicity in the system, manifested as a nonlinear dependence of the transition dipole
moments with respect to the nuclear coordinates [35], may contribute to the parity
effects observed by Gonjo et al. [2].
Most of the combined experimental and computational studies of anharmonic
effects focused on X–H vibrations (e.g. N–H, O–H, and F–H) since the respective
fundamental and overtone bands are relatively strong and well-resolved. Comparatively little knowledge is available about the anharmonicity of other kinds of vibrations and how they are influenced by the molecule’s chemical surrounding. However,
ν(C = O) modes were investigated in the context of anharmonicity and solvent
effects by Chen et al. [9]. The study focused on the IR and NIR spectral regions of
acetone and 2–hexanone. As solvents, n–hexane, CCl 4 , and CHCl 3 , were used and the
study considered vapor phase data for comparison as well. It was confirmed that the
wavenumbers, absorption intensities, and oscillator strengths of the ν(C = O) modes
demonstrate a distinct solvent dependence. In case of the fundamental and the first
overtone bands, the ν(C = O) intensities were found to be significantly stronger than
those of the ν(C–H) vibration. At the same time however, the ν(C = O) and ν(C–H)
bands were found to be comparable in terms of their intensity. Quantum chemical
calculations reproduced the observed trends in integrated intensity upon going from
the fundamental to the first overtone of the ν(C = O), ν(O–H), ν(C–H), and ν(S–H)
vibrations (Fig. 13.8). The combined theoretical and experimental results suggest
that the weak intensity observed for the 2ν(C = O) stretching overtone has a twofold
cause. Low anharmonicity of the vibrational potential and a substantial reduction in
the oscillator strength were suggested to contribute to this spectral effect [9].
As one may conclude from this chapter so far, studies focusing on solvent
effects are of particular importance in physicochemical NIR spectroscopy. However,
advanced computational approaches (e.g. Refs. [2, 9]), explicitly taking into account
solvent effects may often be unsuitable in practice, e.g. for investigations of larger
M. A. Czarnecki et al.
normal coordinate corresponding of ν(OH) mode. The potential energy in each point
was obtained at the B3LYP/6-311++G(3df,3pd) level employing the IPCM solvation
model. Subsequently, to obtain vibrational states, the one-dimensional Schrödinger
equation was solved using Johnson’s reformulation of Numerov’s approach. This
approach yielded the vibrational levels at the accuracy level exceeding 0.001 cm
−1
with respect to the determined vibrational potential, and allowed to avoid approximations such as the fitting of a Morse function [2]. The corresponding transition
intensities were obtained from the integrated absorption coefficient (km mol
−1 , base
e). The accurate calculation of wavenumbers and intensities of the ν(OH), 2ν(OH)
and 3ν(OH) bands enabled a direct interpretation of the observed intensity variations
as well as respective solvent dependency. The calculations reproduced the observed
“parity” in the intensity of the ν(OH) bands ν 01 , ν 02 , ν 03 , and ν 04 in phenols as well
as the respective solvent dependency. The parity effect was notably more prominent
for phenol than for 2,6–dihalogenated phenols. It was concluded that this difference
results from phenol having a stronger intermolecular hydrogen bonding in contrast to
its derivatives that feature a weaker intramolecular hydrogen bond. By this reasoning,
it was suggested that the intermolecular hydrogen bond between the OH group and
the Cl atom is responsible for the observed tendencies [2]. The electrical anharmonicity in the system, manifested as a nonlinear dependence of the transition dipole
moments with respect to the nuclear coordinates [35], may contribute to the parity
effects observed by Gonjo et al. [2].
Most of the combined experimental and computational studies of anharmonic
effects focused on X–H vibrations (e.g. N–H, O–H, and F–H) since the respective
fundamental and overtone bands are relatively strong and well-resolved. Comparatively little knowledge is available about the anharmonicity of other kinds of vibrations and how they are influenced by the molecule’s chemical surrounding. However,
ν(C = O) modes were investigated in the context of anharmonicity and solvent
effects by Chen et al. [9]. The study focused on the IR and NIR spectral regions of
acetone and 2–hexanone. As solvents, n–hexane, CCl 4 , and CHCl 3 , were used and the
study considered vapor phase data for comparison as well. It was confirmed that the
wavenumbers, absorption intensities, and oscillator strengths of the ν(C = O) modes
demonstrate a distinct solvent dependence. In case of the fundamental and the first
overtone bands, the ν(C = O) intensities were found to be significantly stronger than
those of the ν(C–H) vibration. At the same time however, the ν(C = O) and ν(C–H)
bands were found to be comparable in terms of their intensity. Quantum chemical
calculations reproduced the observed trends in integrated intensity upon going from
the fundamental to the first overtone of the ν(C = O), ν(O–H), ν(C–H), and ν(S–H)
vibrations (Fig. 13.8). The combined theoretical and experimental results suggest
that the weak intensity observed for the 2ν(C = O) stretching overtone has a twofold
cause. Low anharmonicity of the vibrational potential and a substantial reduction in
the oscillator strength were suggested to contribute to this spectral effect [9].
As one may conclude from this chapter so far, studies focusing on solvent
effects are of particular importance in physicochemical NIR spectroscopy. However,
advanced computational approaches (e.g. Refs. [2, 9]), explicitly taking into account
solvent effects may often be unsuitable in practice, e.g. for investigations of larger