it has been suggested that organometallic benzene derivatives may even serve as
potential superconducting materials [42].
Organometallic catalysis has allowed the development of an impressive number
of chemical transformations that could not be achieved using classical methodologies. Most of these reactions have been accomplished in organic solvents, and in
many cases in the absence of water, and under air-free conditions. The increasing
pressure to develop more sustainable transformations has stimulated the discovery of
metal-catalyzed reactions that can take place in water. A particularly attractive
extension of this chemistry consists of the use of biologically relevant aqueous
solvents, as this might set the basis to transform catalytic metal complexes into
biological settings [8, 11, 43–45]. These trends go in line with the recently invoked
interest of pharmaceutical research and development for organometallics [46], in
particular their use as potential anticancer drugs [47–49].
There is a continuous scientific endeavor being aimed at optimizing currently
applied organometallic compounds and/or finding new ones with advanced properties and a broader scope of applications. In particular, modifying and fine-tuning
homogeneous organometallic catalysts has been at the focus. However, there is a
huge number of possible combinations between one of the 28 transition metals of the
first, second, and third transition metal period (excluding Tc) or one of the 12 metals
of periods 2–6 and an even larger amount of possible ligands (L). Attempts to find
suitable organometallic complexes for catalysis reach from trial-and-error procedures to educated guesses and model-based strategies [50–58], which is nowadays
strongly supported and guided by quantum chemical catalyst design exploring the
catalytic reaction mechanism [59] or the physical properties of the catalyst
[60]. Physical property-based approaches, which recently started to involve data
science and machine learning techniques, provide the ability to examine large swaths
of chemical space [61, 62], but the translation of this information into practical
catalyst design may not be straightforward, in particular if the proposed properties do
not directly relate to measurable quantities.
In this connection, one has searched since decades for measurable parameters that
can be used as suitable descriptors to assess the catalytic activity of a metal complex,
mainly focusing on identifying possible metal–ligand (ML) bond descriptors. The
ML interaction, often highly covalent in nature in the case of organometallics, plays
a key role in determining their diverse properties and rich chemistry, combining
aspects of traditional inorganic and organic chemistry. Therefore, the detailed
understanding of the ML bond is a necessary prerequisite for the fine-tuning of
existing and the design of the next generation of organometallic catalysts. Two
popular strategies to describe the catalytic activity of a transition metal complex in
homogeneous catalysis as a function of the ML bond are based (1) on the ML bond
dissociation energies (BDEs) [63–68] and (2) on molecular geometries to predict via
BDE values and/or bond lengths the ease replacement of a given ligand or the
possibility of enlarging the coordination sphere of a transition metal during catalysis.
While these attempts have certainly contributed to the chemical understanding of
metal and transition metal complexes, one has to realize that BDE values or bond
lengths provide little insight into the intrinsic strength of the ML bond. The BDE is a
reaction parameter that includes all changes, which take place during the dissociation
230
E. Kraka and M. Freindorf
potential superconducting materials [42].
Organometallic catalysis has allowed the development of an impressive number
of chemical transformations that could not be achieved using classical methodologies. Most of these reactions have been accomplished in organic solvents, and in
many cases in the absence of water, and under air-free conditions. The increasing
pressure to develop more sustainable transformations has stimulated the discovery of
metal-catalyzed reactions that can take place in water. A particularly attractive
extension of this chemistry consists of the use of biologically relevant aqueous
solvents, as this might set the basis to transform catalytic metal complexes into
biological settings [8, 11, 43–45]. These trends go in line with the recently invoked
interest of pharmaceutical research and development for organometallics [46], in
particular their use as potential anticancer drugs [47–49].
There is a continuous scientific endeavor being aimed at optimizing currently
applied organometallic compounds and/or finding new ones with advanced properties and a broader scope of applications. In particular, modifying and fine-tuning
homogeneous organometallic catalysts has been at the focus. However, there is a
huge number of possible combinations between one of the 28 transition metals of the
first, second, and third transition metal period (excluding Tc) or one of the 12 metals
of periods 2–6 and an even larger amount of possible ligands (L). Attempts to find
suitable organometallic complexes for catalysis reach from trial-and-error procedures to educated guesses and model-based strategies [50–58], which is nowadays
strongly supported and guided by quantum chemical catalyst design exploring the
catalytic reaction mechanism [59] or the physical properties of the catalyst
[60]. Physical property-based approaches, which recently started to involve data
science and machine learning techniques, provide the ability to examine large swaths
of chemical space [61, 62], but the translation of this information into practical
catalyst design may not be straightforward, in particular if the proposed properties do
not directly relate to measurable quantities.
In this connection, one has searched since decades for measurable parameters that
can be used as suitable descriptors to assess the catalytic activity of a metal complex,
mainly focusing on identifying possible metal–ligand (ML) bond descriptors. The
ML interaction, often highly covalent in nature in the case of organometallics, plays
a key role in determining their diverse properties and rich chemistry, combining
aspects of traditional inorganic and organic chemistry. Therefore, the detailed
understanding of the ML bond is a necessary prerequisite for the fine-tuning of
existing and the design of the next generation of organometallic catalysts. Two
popular strategies to describe the catalytic activity of a transition metal complex in
homogeneous catalysis as a function of the ML bond are based (1) on the ML bond
dissociation energies (BDEs) [63–68] and (2) on molecular geometries to predict via
BDE values and/or bond lengths the ease replacement of a given ligand or the
possibility of enlarging the coordination sphere of a transition metal during catalysis.
While these attempts have certainly contributed to the chemical understanding of
metal and transition metal complexes, one has to realize that BDE values or bond
lengths provide little insight into the intrinsic strength of the ML bond. The BDE is a
reaction parameter that includes all changes, which take place during the dissociation
230
E. Kraka and M. Freindorf