acronyms (‘ch’, which stands for ‘change’, followed by a number) to enable easy
referencing to them within this text.
The water molecule appears to prefer to bind to an sp
2 O rather than to an sp
3 O.
Thus, inputs in which the water molecule is attached to O19, or to both O19 and
O23, optimize to adducts with the water molecule attached to O23 (ch-1). Inputs in
which the water molecule is attached to the donor of the first IHB (O8 or O12, for d
and s conformers respectively), in conformers in which O14 is the only sp
2 O in the
vicinity, often optimize to adducts in which water binds to O14 (ch-2, ch-3). If O23
is also available in the vicinity (e.g., in inputs involving 3-d conformers), the water
molecule often shifts from O8 to O23 (ch-4) or, sometimes, to O26 (ch-5). These
changes may be related to the hydrophobic character of IHB regions for hydroxybenzenes in general [33], and of the first IHB of ACPLs in particular [17].
A clear preference appears for the water molecule to bind to two sites simultaneously, when two sites are geometrically suitable; this may lead to adducts in
which it binds to an sp
2 O and an sp
3 O simultaneously (e.g., O8 and O23; ch-6,
ch-7). The water molecule also appears to prefer to be acceptor to an OH group
(consistently also with the known tendency of phenol OHs to be donors in intermolecular H-bonds [16, 31]). This may involve rotation of H15, H16, H17 or H27,
to enable the formation of the H-bond with the water molecule (it is interesting to
recall that the phenol OHs in ACPLs do not usually rotate to form IHBs [16],
whereas the adducts calculated here show that the OHs in ARZ may rotate to form
an intermolecular H-bond with the solvent). The water molecule may change
binding site completely in order to be acceptor to the H of an OH (ch-8, ch-9,
ch-10). On the other hand, its tendency to bind to two geometrically suitable sites
simultaneously may lead to adducts in which it binds simultaneously to the H of an
OH and to a suitably close O atom (ch-11 to ch-19). The tendency to bind to two
sites appears also for inputs in which the water molecule is initially placed as
acceptor to an OH (ch-20 to ch-25).
The water molecule does not break O-H⋯O IHBs. It appears, however, to be
able to act on the O-H⋯π interaction, by breaking it (ch-26) or prompting it
(ch-27), or changing its pattern (ch-28, in which it changes from η to ξ). The water
molecule itself may interact with the C29=C30 π bond; the interaction usually
appears during optimisation (ch-29, ch-30), whereas inputs having the interaction
do not often optimise to adducts in which it is maintained (ch-31 to ch-34).
The changes just outlined appear both with HF and with DFT optimization. In
most cases, the same change occurs with the same input, i.e., HF and DFT optimisations lead to the same changed output. In some cases, the changes are different
with the two methods, leading to different outputs. Figure 6 shows a case in which
HF and DFT lead to outputs that are different from the input and different from each
other. In such cases, it is not easy to estimate a priori which of the two outcomes
might be closer to reality; for the specific case shown in Fig. 6, the HF result seems
more probable, as it shows preference for an sp
2 O, which can form a stronger
H-bond than an sp
3 O like O8 or O26.
Table 3 reports the relative energy and the ARZ-water interaction energy for
representative adducts, selected in such a way as to comprise nearly all the
292
L. Mammino
referencing to them within this text.
The water molecule appears to prefer to bind to an sp
2 O rather than to an sp
3 O.
Thus, inputs in which the water molecule is attached to O19, or to both O19 and
O23, optimize to adducts with the water molecule attached to O23 (ch-1). Inputs in
which the water molecule is attached to the donor of the first IHB (O8 or O12, for d
and s conformers respectively), in conformers in which O14 is the only sp
2 O in the
vicinity, often optimize to adducts in which water binds to O14 (ch-2, ch-3). If O23
is also available in the vicinity (e.g., in inputs involving 3-d conformers), the water
molecule often shifts from O8 to O23 (ch-4) or, sometimes, to O26 (ch-5). These
changes may be related to the hydrophobic character of IHB regions for hydroxybenzenes in general [33], and of the first IHB of ACPLs in particular [17].
A clear preference appears for the water molecule to bind to two sites simultaneously, when two sites are geometrically suitable; this may lead to adducts in
which it binds to an sp
2 O and an sp
3 O simultaneously (e.g., O8 and O23; ch-6,
ch-7). The water molecule also appears to prefer to be acceptor to an OH group
(consistently also with the known tendency of phenol OHs to be donors in intermolecular H-bonds [16, 31]). This may involve rotation of H15, H16, H17 or H27,
to enable the formation of the H-bond with the water molecule (it is interesting to
recall that the phenol OHs in ACPLs do not usually rotate to form IHBs [16],
whereas the adducts calculated here show that the OHs in ARZ may rotate to form
an intermolecular H-bond with the solvent). The water molecule may change
binding site completely in order to be acceptor to the H of an OH (ch-8, ch-9,
ch-10). On the other hand, its tendency to bind to two geometrically suitable sites
simultaneously may lead to adducts in which it binds simultaneously to the H of an
OH and to a suitably close O atom (ch-11 to ch-19). The tendency to bind to two
sites appears also for inputs in which the water molecule is initially placed as
acceptor to an OH (ch-20 to ch-25).
The water molecule does not break O-H⋯O IHBs. It appears, however, to be
able to act on the O-H⋯π interaction, by breaking it (ch-26) or prompting it
(ch-27), or changing its pattern (ch-28, in which it changes from η to ξ). The water
molecule itself may interact with the C29=C30 π bond; the interaction usually
appears during optimisation (ch-29, ch-30), whereas inputs having the interaction
do not often optimise to adducts in which it is maintained (ch-31 to ch-34).
The changes just outlined appear both with HF and with DFT optimization. In
most cases, the same change occurs with the same input, i.e., HF and DFT optimisations lead to the same changed output. In some cases, the changes are different
with the two methods, leading to different outputs. Figure 6 shows a case in which
HF and DFT lead to outputs that are different from the input and different from each
other. In such cases, it is not easy to estimate a priori which of the two outcomes
might be closer to reality; for the specific case shown in Fig. 6, the HF result seems
more probable, as it shows preference for an sp
2 O, which can form a stronger
H-bond than an sp
3 O like O8 or O26.
Table 3 reports the relative energy and the ARZ-water interaction energy for
representative adducts, selected in such a way as to comprise nearly all the
292
L. Mammino