Sorption processes of ions are often nonlinear (i.e., concentration dependent).
Nonlinear binding isotherms of IOCs have been observed for the sorption to
technical ion exchangers (Liu et al. 2007; Wu et al. 2008) and to soil organic matter
(Droge and Goss 2012b). Nonlinear isotherms must occur, according to the mass
action law, for ion-pair sorption (Escher and Sigg 2004). However, they can also be a
result of the heterogenicity of the sorbent offering sorption sites with different
sorption energies and they will occur close to saturation for any sorbent (homogeneous or heterogeneous) with a limited amount of sorption sites (e.g., ion exchange
sorbent).
In context of the various conceivable sorption processes at charged interfaces it
is also instructive to discuss possible parallels to the sorption of metal ions. The
well-established Nica–Donnan concept (Benedetti et al. 1996; Milne et al. 2003)
describes sorption of metal ions to charged surfaces as the combination of
partitioning to an electrical double layer adjacent to the charged surface and sorption
to specific ion exchange sites. In principle this should also hold for IOCs at charged
surfaces. A quantitative description of these processes will, however, be much more
complex for IOCs than for metal ions. Metal ions are spherical and their charge will
equally distribute about their complete spherical surface. The resulting charge
density which is relevant for the interaction strength can thus be calculated rather
easily and it will have just one distinct value for the complete metal ion. In addition,
the interactions of a metal ion are dominated by the ionic interactions and the cavity
effect. Van der Waals and H-bond interactions are probably negligible in most cases.
This is completely different for IOCs. The charge is located at the functional group
of the molecule and in most cases not evenly distributed about the surface of the
whole molecule. However, the local charge densities on all parts of the molecule
need to be known (which in turn requires knowledge of the 3D structure of the
molecule) if a quantitative understanding of the sorption behavior is required. Next
to the cavity energy and the ionic interactions, van der Waals and H-bond interactions play an important or even dominant role for IOCs. The Nica–Donnan model
may, however, readily be employed to explain differences in the competition of a
given organic cation with various metal cations for ion exchange sites and the
influence of pH (Chen et al. 2013; Iglesias et al. 2009).
3 Bioaccumulation and Biosorption of IOCs
Bioaccumulation of neutral organic chemicals is mainly driven by the total lipid
fraction of an organism and for some neutral chemicals (H-bond donor chemicals)
the protein fraction can also contribute to bioaccumulation (Endo et al. 2013). For
IOC, instead, several sorption phases may be relevant (Figs. 2 and 4).
Figure 4 shows the average composition of the human body (Goss et al. 2018).
Storage lipids (i.e., nonpolar lipids) that are typically the major sink for nonpolar
chemicals are assumed to be negligible as sorbing matrix for IOCs (Schmitt 2008).
Although it appears that the sorption of IOCs to storage lipids was never investigated
by experiment, this assumption is inferred from the weak sorption in octanol for
48
L. Henneberger and K.-U. Goss
Nonlinear binding isotherms of IOCs have been observed for the sorption to
technical ion exchangers (Liu et al. 2007; Wu et al. 2008) and to soil organic matter
(Droge and Goss 2012b). Nonlinear isotherms must occur, according to the mass
action law, for ion-pair sorption (Escher and Sigg 2004). However, they can also be a
result of the heterogenicity of the sorbent offering sorption sites with different
sorption energies and they will occur close to saturation for any sorbent (homogeneous or heterogeneous) with a limited amount of sorption sites (e.g., ion exchange
sorbent).
In context of the various conceivable sorption processes at charged interfaces it
is also instructive to discuss possible parallels to the sorption of metal ions. The
well-established Nica–Donnan concept (Benedetti et al. 1996; Milne et al. 2003)
describes sorption of metal ions to charged surfaces as the combination of
partitioning to an electrical double layer adjacent to the charged surface and sorption
to specific ion exchange sites. In principle this should also hold for IOCs at charged
surfaces. A quantitative description of these processes will, however, be much more
complex for IOCs than for metal ions. Metal ions are spherical and their charge will
equally distribute about their complete spherical surface. The resulting charge
density which is relevant for the interaction strength can thus be calculated rather
easily and it will have just one distinct value for the complete metal ion. In addition,
the interactions of a metal ion are dominated by the ionic interactions and the cavity
effect. Van der Waals and H-bond interactions are probably negligible in most cases.
This is completely different for IOCs. The charge is located at the functional group
of the molecule and in most cases not evenly distributed about the surface of the
whole molecule. However, the local charge densities on all parts of the molecule
need to be known (which in turn requires knowledge of the 3D structure of the
molecule) if a quantitative understanding of the sorption behavior is required. Next
to the cavity energy and the ionic interactions, van der Waals and H-bond interactions play an important or even dominant role for IOCs. The Nica–Donnan model
may, however, readily be employed to explain differences in the competition of a
given organic cation with various metal cations for ion exchange sites and the
influence of pH (Chen et al. 2013; Iglesias et al. 2009).
3 Bioaccumulation and Biosorption of IOCs
Bioaccumulation of neutral organic chemicals is mainly driven by the total lipid
fraction of an organism and for some neutral chemicals (H-bond donor chemicals)
the protein fraction can also contribute to bioaccumulation (Endo et al. 2013). For
IOC, instead, several sorption phases may be relevant (Figs. 2 and 4).
Figure 4 shows the average composition of the human body (Goss et al. 2018).
Storage lipids (i.e., nonpolar lipids) that are typically the major sink for nonpolar
chemicals are assumed to be negligible as sorbing matrix for IOCs (Schmitt 2008).
Although it appears that the sorption of IOCs to storage lipids was never investigated
by experiment, this assumption is inferred from the weak sorption in octanol for
48
L. Henneberger and K.-U. Goss