256
as seen in Sect. 7.2. They may also form complex with proteins, sometimes highly
insoluble (e.g., tannins). Polar polyphenols have been traditionally extracted using
aqueous mixtures of methanol, ethanol, and acetone in conventional solid/liquid
extraction (SLE) (Ruesgas-Ramón et al. 2017). Ethanol and methanol are also used
as cosolvent in supercritical carbon dioxide (Sc-CO 2 ) extraction of polyphenol from
plant materials such as grape pomaces and seeds, rosemary and olive leaves, and
pistachio hulls. The Sc-CO 2 extracts mainly contain gallic acid, catechin, and epicatechin, whereas the SLE extracts are richer in proanthocyanidin. This underlines
the importance of the choice of the solvent and extraction method for extracting
selectively the compounds of interest, which logically depend on their structures.
Thanks to their ability to form hydrogen bonds, polyphenols are potentially good
targets for extraction using deep eutectic solvent. As mentioned previously, hydroxyl
groups on polyphenols are good hydrogen donors but poor acceptors and preferably
form bonds with good hydrogen bond acceptors (Burghoff et al. 2008; Martins et al.
2019). Thus, amines and phosphine oxides are excellent phenol extractants. It might
be also interesting to introduce aromatic rings in the deep eutectic solvent to take
benefit of π-π interactions. Depending on their constituents, deep eutectic solvents
and especially natural deep eutectic solvents could be tuned with a wide range of
polarities to accommodate from the less to the most polar polyphenols. In addition,
they could interact via numerous hydrogen bonds with the hydroxyl groups of polyphenols, thus promoting their solubilization (Dai et al. 2013c). For these reasons,
natural deep eutectic solvents have been considered as promising solvents for the
selective solubilization and extraction of polyphenols (Faggian et al. 2016). Even if
(natural) deep eutectic solvents do not overtake the extraction yields obtained with
benchmark solvents, an advantage is expected in terms of selectivity in addition to
environmental acceptability.
Interestingly, it is also suggested that natural deep eutectic solvents also enhance
the biological activity of polyphenols (Faggian et al. 2016) and, thanks to the biocompatibility of the natural deep eutectic solvents, extracts may be used directly in
food, cosmetic, and pharmaceutical formulations.
In front of the abundance of possible (natural) deep eutectic solvent formulations, the in silico methods could provide a valuable guidance to select the most
promising solvents for the selective extraction of a desired compound or class of
compounds (Jeliński and Cysewski 2018; Silva et al. 2018). Particularly, the
COSMO-RS (quantum conductor-like screening model for real solvent) has been
successfully applied to ionic liquids and deep eutectic solvents to rank solvent candidates for the extraction of polyphenols (Burghoff et al. 2008; Jeliński and
Cysewski 2018) and other compounds of interest. The COSMO-RS model enables
to compute the chemical potential of molecules in liquid solutions, from which
other parameters such as activity and solubility can be derived. One limitation of the
use of COSMO-RS for (natural) deep eutectic solvents is that the degree of dissociation of weak acids and bases in the mixture, which are the main components of
natural deep eutectic solvent, is generally unknown; therefore, rough assumptions
have to be made. Their accuracy can be validated by comparison with experimental
solubility data sets.
L. Percevault et al.
as seen in Sect. 7.2. They may also form complex with proteins, sometimes highly
insoluble (e.g., tannins). Polar polyphenols have been traditionally extracted using
aqueous mixtures of methanol, ethanol, and acetone in conventional solid/liquid
extraction (SLE) (Ruesgas-Ramón et al. 2017). Ethanol and methanol are also used
as cosolvent in supercritical carbon dioxide (Sc-CO 2 ) extraction of polyphenol from
plant materials such as grape pomaces and seeds, rosemary and olive leaves, and
pistachio hulls. The Sc-CO 2 extracts mainly contain gallic acid, catechin, and epicatechin, whereas the SLE extracts are richer in proanthocyanidin. This underlines
the importance of the choice of the solvent and extraction method for extracting
selectively the compounds of interest, which logically depend on their structures.
Thanks to their ability to form hydrogen bonds, polyphenols are potentially good
targets for extraction using deep eutectic solvent. As mentioned previously, hydroxyl
groups on polyphenols are good hydrogen donors but poor acceptors and preferably
form bonds with good hydrogen bond acceptors (Burghoff et al. 2008; Martins et al.
2019). Thus, amines and phosphine oxides are excellent phenol extractants. It might
be also interesting to introduce aromatic rings in the deep eutectic solvent to take
benefit of π-π interactions. Depending on their constituents, deep eutectic solvents
and especially natural deep eutectic solvents could be tuned with a wide range of
polarities to accommodate from the less to the most polar polyphenols. In addition,
they could interact via numerous hydrogen bonds with the hydroxyl groups of polyphenols, thus promoting their solubilization (Dai et al. 2013c). For these reasons,
natural deep eutectic solvents have been considered as promising solvents for the
selective solubilization and extraction of polyphenols (Faggian et al. 2016). Even if
(natural) deep eutectic solvents do not overtake the extraction yields obtained with
benchmark solvents, an advantage is expected in terms of selectivity in addition to
environmental acceptability.
Interestingly, it is also suggested that natural deep eutectic solvents also enhance
the biological activity of polyphenols (Faggian et al. 2016) and, thanks to the biocompatibility of the natural deep eutectic solvents, extracts may be used directly in
food, cosmetic, and pharmaceutical formulations.
In front of the abundance of possible (natural) deep eutectic solvent formulations, the in silico methods could provide a valuable guidance to select the most
promising solvents for the selective extraction of a desired compound or class of
compounds (Jeliński and Cysewski 2018; Silva et al. 2018). Particularly, the
COSMO-RS (quantum conductor-like screening model for real solvent) has been
successfully applied to ionic liquids and deep eutectic solvents to rank solvent candidates for the extraction of polyphenols (Burghoff et al. 2008; Jeliński and
Cysewski 2018) and other compounds of interest. The COSMO-RS model enables
to compute the chemical potential of molecules in liquid solutions, from which
other parameters such as activity and solubility can be derived. One limitation of the
use of COSMO-RS for (natural) deep eutectic solvents is that the degree of dissociation of weak acids and bases in the mixture, which are the main components of
natural deep eutectic solvent, is generally unknown; therefore, rough assumptions
have to be made. Their accuracy can be validated by comparison with experimental
solubility data sets.
L. Percevault et al.