once a convincing structural model is obtained and validated. The latter can be
achieved by simulating any spectroscopic features, such as UV-Vis, IR, or NMR,
for which experimental data is available. Care must be taken not to overrate the
agreement between experimental and theoretical data, which can be due to error
cancellation or due to correlation between observables. The retinal chromophore of
rhodopsin (Rh) provides an example of a strong correlation between the C¼C
stretching frequency of a charged conjugated chain and its UV-Vis absorption
maximum [1]. Therefore, both properties show the same response to electrostatic
perturbation of the chromophore, they essentially probe the potential difference
between the two ends of the conjugated system and yield redundant information [2].
Further assessment can be achieved by direct comparison between xray and theoretical structure. The validation power of these tests can be amplified by combining
them with site-directed mutation experiments or extended molecular dynamics
(MD) simulations (see Sect. 4.4).
The second benefit of theoretical approaches is to provide a physical explanation
for the observed spectral features. They can be related to individual structural
features or to electrostatic or steric interactions between the chromophore and its
molecular environment. Quantum chemical calculations can reveal the individual
contributions of geometric changes, electrostatic interactions with charged, polar,
or polarisable groups, the effect of dispersive interactions or charge transfer and
delocalisation. These aspects will be discussed in Sect. 4.3.
The great challenge for computational approaches is to capture the essential
physics involved in the observed features of the chromophore and to correctly
describe their response to perturbations from the environment. As the electronically
excited state is involved, this remains one of the critical points in theoretical
predictions and it is important to consider the influence of the employed quantum
mechanical (QM) methods on the results. In particular, it is essential to know
whether or not error cancellation can be assumed for a specific property, which is
often necessary to predict small effects, like frequency shifts due to mutation or
structural changes. This topic will be addressed for individual classes of quantum
chemical methods in Sect. 4.2.
4.2
QM Methods
The demands to be met by the QM method include the qualitative correct response
of the calculated property (e.g., spectroscopic features) to any kind of external
perturbation that may affect this property. These can be grouped as follows: (1)
geometric distortions (in particular bonds lengths, rotation of single and double
bonds), (2) electrostatic polarisation (by H-bonds, counter ions, charged and closeby polar groups), (3) exchange of charge density (mainly across H-bonds), and (4)
dispersion redshift in presence of vicinal highly polarisable groups. (1)–(4) imply
different demands on the employed QM methods and it is essential knowledge to
which extend commonly used approaches meet them, or show systematic errors.
The low-lying electronic valence states of ionic chromophores typically involve
46
M. Wanko and A. Rubio
achieved by simulating any spectroscopic features, such as UV-Vis, IR, or NMR,
for which experimental data is available. Care must be taken not to overrate the
agreement between experimental and theoretical data, which can be due to error
cancellation or due to correlation between observables. The retinal chromophore of
rhodopsin (Rh) provides an example of a strong correlation between the C¼C
stretching frequency of a charged conjugated chain and its UV-Vis absorption
maximum [1]. Therefore, both properties show the same response to electrostatic
perturbation of the chromophore, they essentially probe the potential difference
between the two ends of the conjugated system and yield redundant information [2].
Further assessment can be achieved by direct comparison between xray and theoretical structure. The validation power of these tests can be amplified by combining
them with site-directed mutation experiments or extended molecular dynamics
(MD) simulations (see Sect. 4.4).
The second benefit of theoretical approaches is to provide a physical explanation
for the observed spectral features. They can be related to individual structural
features or to electrostatic or steric interactions between the chromophore and its
molecular environment. Quantum chemical calculations can reveal the individual
contributions of geometric changes, electrostatic interactions with charged, polar,
or polarisable groups, the effect of dispersive interactions or charge transfer and
delocalisation. These aspects will be discussed in Sect. 4.3.
The great challenge for computational approaches is to capture the essential
physics involved in the observed features of the chromophore and to correctly
describe their response to perturbations from the environment. As the electronically
excited state is involved, this remains one of the critical points in theoretical
predictions and it is important to consider the influence of the employed quantum
mechanical (QM) methods on the results. In particular, it is essential to know
whether or not error cancellation can be assumed for a specific property, which is
often necessary to predict small effects, like frequency shifts due to mutation or
structural changes. This topic will be addressed for individual classes of quantum
chemical methods in Sect. 4.2.
4.2
QM Methods
The demands to be met by the QM method include the qualitative correct response
of the calculated property (e.g., spectroscopic features) to any kind of external
perturbation that may affect this property. These can be grouped as follows: (1)
geometric distortions (in particular bonds lengths, rotation of single and double
bonds), (2) electrostatic polarisation (by H-bonds, counter ions, charged and closeby polar groups), (3) exchange of charge density (mainly across H-bonds), and (4)
dispersion redshift in presence of vicinal highly polarisable groups. (1)–(4) imply
different demands on the employed QM methods and it is essential knowledge to
which extend commonly used approaches meet them, or show systematic errors.
The low-lying electronic valence states of ionic chromophores typically involve
46
M. Wanko and A. Rubio