Ion-pair binding: Is binding both binding better?

Article abstract of DOI:10.1002/chem.200900342

It is often tempting to explain chemical phenomena on the basis of intuitive principles, but this practice can frequently lead to biased analysis of data and incorrect conclusions. One such intuitive principle is brought into play in the binding of salts

Full text of DOI:10.1002/chem.200900342

DOI: 10.1002/chem.200900342
Ion-Pair Binding: Is Binding Both Binding Better?
Stefano Roelens,*[a] Alberto Vacca,[b] Oscar Francesconi,[a] and Chiara Venturi[c]
Abstract: It is often tempting to ex-
plain chemical phenomena on the basis
of intuitive principles, but this practice
can frequently lead to biased analysis
of data and incorrect conclusions. One
such intuitive principle is brought into
play in the binding of salts by synthetic
receptors. Following the heuristic con-
cept that “binding both is binding
better”, it is widely believed that ditop-
ic receptors capable of binding both
ionic partners of a salt are more effec-
tive than monotopic receptors because
of a cooperative effect. Using a newly
designed ditopic receptor and a gener-
alized binding descriptor, we show here
that, when the problem is correctly for-
mulated and the appropriate algorithm
is derived, the cooperativity principle is
neither general nor predictable, and
that competition between ion binding
and ion pairing may even lead to inhib-
ition rather than enhancement of the
binding of an ion to a ditopic receptor.
Keywords: affinity descriptors · co-
operative effects · ion binding · ion
pairs · receptors
Introduction
fact, the current literature on binding of ion pairs to ditopic
receptors is essentially based on 1) crystallographic evi-
dence,[3–11] which demonstrates the existence of ion-pair
complexes but not of cooperative effects, 2) transport
through membranes[7,12–15] or extraction from aqueous solu-
tions,[4,16–21] which is not necessarily or solely dependent on
cooperativity, and 3) binding measurements in solu-
tion,[5,6,10,11,22–40] which suffer from a number of inconsisten-
cies that do not allow a convincing and unambiguous assess-
ment of cooperativity. This drawback is caused by the
common practice of comparing the association constant of
an ion pair with a ditopic receptor, usually measured by
1H NMR spectroscopy in organic solvents through a single
signal of a single reagent, to the association constant of the
ditopic receptor with the investigated anion or cation in the
presence of an innocent counterion, but neglecting other
equilibria (ion pairing, higher stoichiometry association,
etc.) that may occur in solution; in addition, in all cases ex-
perimental data are fitted to a 1:1 association model for
both the ion and the ion-pair complexes. Such a simplified
approach is inconsistent for several reasons: 1) systems con-
taining multiple species and/or higher stoichiometry com-
plexes cannot be fitted to a 1:1 binding isotherm; 2) the
comparison is conceptually ill defined, because the ternary
ion pair/receptor complex and the binary ion/receptor com-
plex are incommensurable, having different dimensionality;
3) a three-reagent species like an ion-pair complex cannot
obviously fit into a two-reagent 1:1 model; 4) the counterion
is not inert and cannot be neglected. It is in fact little appre-
ciated that even an innocent counterion, although unbound,
Cooperativity is a concept that is often invoked to explain
phenomena that are not well understood and appear to pro-
duce effects beyond expectation. One such phenomena that
has recently been gaining in popularity is cooperativity in
the binding of an ion pair by synthetic ditopic receptors,
which can simultaneously binding the anion and the cation
of a salt with higher affinity than is observed toward each of
the partners.[1,2] The underlying heuristic principle is that
“binding both is binding better”. This intuitively plausible
concept rests, however, on ambiguous evidence, which does
not allow a clear-cut assessment of true cooperativity. In
[a] Dr. S. Roelens, O. Francesconi
CNR—Istituto di Metodologie Chimiche (IMC)
Dipartimento di Chimica Organica
Universitꢀ di Firenze
Via della Lastruccia, 13
50019 Sesto Fiorentino, Firenze (Italy)
Fax : (+39)055-457-3570
E-mail: stefano.roelens@unifi.it
[b] Prof. A. Vacca
Dipartimento di Chimica
Universitꢀ di Firenze
50019 Sesto Fiorentino, Firenze (Italy)
[c] Dr. C. Venturi
Centro di Risonanze Magnetiche (CERM)
Universitꢀ di Firenze
50019 Sesto Fiorentino, Firenze (Italy)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200900342.
8296
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 8296 – 8302

FULL PAPER
affects binding of the ion under consideration by
taking part in ion-pairing equilibria that determine
the actual availability of the observed ion.
To find an unambiguous way to ascertain whether
cooperativity truly enhances binding of an ion pair,
we re-examined the problem from the basic ques-
tion: “is binding what better than what?”. The first
requirement for a homogeneous comparison is nec-
essarily the definition of an appropriate reference.
Considering that complexes of ions cannot be com-
pared to complexes of ion pairs, the correct refer-
ence should be the affinity of the reactant (the ion)
rather than that of a product (the ion pair), so a di-
mensionally correct approach consists of comparing the af-
finities of individual reacting ions in the presence and ab-
sence of a co-bound counterion. This requires taking into ac-
count all equilibria involving bound species, measuring the
corresponding formation constants, and translating them
into a global affinity of the ion under consideration, since in
multi-equilibrium systems the overall affinity is not ade-
quately described by any of the constants alone. Positive co-
operativity will then emerge when the affinity of the ion for
a ditopic receptor simultaneously binding the counterion ex-
ceeds that found for a corresponding monotopic receptor or
for a ditopic receptor in the presence of an unbound (but
not inert) counterion. We stress that, except for limiting
cases, a simple 1:1 association model is necessarily inade-
quate to describe the system, because a salt undergoes an
ion-pairing equilibrium in solution, especially in organic sol-
vents, and the free ions as well as the ion pair originating
from the equilibrium are both, in general, capable of bind-
ing to the receptor, whether mono- or ditopic, so that a
three-constant model is the minimum required model for
treating the binding of salts. The affinity of an ion will there-
fore be determined by at least three independent, non-negli-
gible association equilibria.
CD3CN (80/20), a medium in which both the salt and the re-
ceptor displayed acceptable solubility. Careful analysis of
the H NMR titration experiments of tetramethylammonium
chloride (QX) with R in the above medium allowed reliable
determination of the formation constants of the complex
species formed in solution.[42] From the results collected in
Table 1 (see the Supporting Information), it can be noted
1
Table 1. Cumulative formation constants b with standard deviations s for
complexes of ureidic receptor R with tetramethylammonium chloride
(QX).[a]
Species
b
lgb
RQX
RX
R2
R2X
QX
(1.32Æ0.02)ꢂ106 mÀ2
(8.83Æ0.02)ꢂ102 mÀ1
(6.06Æ0.06)ꢂ100 mÀ1
(3.25Æ0.11)ꢂ104 mÀ2
(1.23Æ0.01)ꢂ104 mÀ1
6.121Æ0.010
2.946Æ0.008
0.783Æ0.027
4.513Æ0.046
4.090Æ0.005
[a] Measured by 1H NMR (400 MHz) from titration experiments at T=
298 K in CDCl3/CD3CN (80/20) on 0.13/1.06 mmolLÀ1 solutions of QX
using receptor concentrations up to 22 mmolLÀ1. Formation constants
were obtained by simultaneous nonlinear least-squares fit of the shifts of
all available signals from two independent titrations run at different salt
concentrations. Global standard deviation of the fit s=0.00027 ppm
(RMS weighted residual=0.00025). Data are reported from ref. [42].
Results and Discussion
that, besides a small degree of dimerization of the receptor
and a non-negligible ion-pair formation constant, which was
in excellent agreement with the value of logb=4.05Æ0.01
independently measured by a dilution experiment on QX in
the same medium,[42] 1:1 and 2:1 receptor:chloride com-
plexes were detected, together with the complex of the mon-
otopic receptor with the whole ion pair. On the contrary, no
evidence of binding to the tetramethylammonium (Q)
cation could be found by control experiments with the pic-
rate salt. On the other hand, the benzylic receptor B, which
was employed in the course of our studies on the cation–p
interaction, was found to bind to Q in the above medium
with moderate but measurable affinity. The formation con-
stants of the species formed with QX in the above solvent
mixture are collected in Table 2 (see the Supporting Infor-
mation).
Searching for a convenient approach to address the issue of
cooperativity, we devised a ditopic receptor that could be
dissected into the two constituent monotopic binding sites,
one exclusively capable of binding the cation and the other
the anion of a salt. In this way, the association of each ion of
the salt with the ditopic receptor could be investigated and
independently compared under the same conditions to that
with the corresponding monotopic counterpart, avoiding the
presence of “innocent” counterions, while taking into ac-
count the ion-pairing equilibrium of the salt and all com-
plexes of free ions and ion-pairs with each of the three re-
ceptors. Following this approach, we conveniently employed
some tripodal receptors that we had developed for different
purposes in our molecular recognition studies.
The tripodal ureidic receptor R, which shows significant
binding affinities for glycosides of monosaccharides,[41] was
found to effectively bind the chloride anion (X) in CDCl3/
As for R, complexes of the receptor with both the free
and the ion-paired Q cation were detected, together with
ion-pair formation. The corresponding value of the constant
Chem. Eur. J. 2009, 15, 8296 – 8302
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8297

S. Roelens et al.
Table 2. Cumulative formation constants b with standard deviations s for
complexes of benzylic receptor B with tetramethylammonium chloride
(QX).[a]
Table 3. Cumulative formation constants b with standard deviations s for
complexes of ditopic receptor D with tetramethylammonium chloride
(QX).[a]
Species
b
lgb
Species
b
lgb
BQX
BQ
(5.85Æ0.05)ꢂ104 mÀ2
(2.36Æ0.02)ꢂ101 mÀ1
(1.23Æ0.01)ꢂ104 mÀ1
4.767Æ0.021
1.373Æ0.007
4.090Æ0.005
DQX
(2.35Æ0.04)ꢂ105 mÀ2
(1.15Æ0.04)ꢂ1012 mÀ4
(1.55Æ0.02)ꢂ100 mÀ1
(9.61Æ0.05)ꢂ100 mÀ1
(1.36Æ0.05)ꢂ102 mÀ1
(1.23Æ0.01)ꢂ104 mÀ1
5.371Æ0.017
12.061Æ0.018
0.189Æ0.006
0.983Æ0.022
2.132Æ0.020
4.090Æ0.005
DQ2X2
[b]
QX[b]
D2
DQ
[a] Measured by 1H NMR (400 MHz) from titration experiments at T=
298 K in CDCl3/CD3CN (80/20) on 0.185/1.03 mmolLÀ1 solutions of QX
with receptor concentrations up to 59 mmolLÀ1. Binding constants were
obtained by simultaneous nonlinear least-squares fit of the shifts of the
only available Me signal from two independent titrations run at different
salt concentrations. Global standard deviation of the fit s=0.00055 ppm
(RMS weighted residual=0.00051). [b] The ion-pair formation constant
was taken from Table 1 and kept invariant in the refinement.
DX
QX[c]
[a] Measured by 1H NMR (400 MHz) from titration experiments at T=
298 K in CDCl3/CD3CN (80/20) on 0.24/1.08 mmolLÀ1 solutions of QX at
receptor concentrations up to 54 mmolLÀ1. Formation constants were ob-
tained by simultaneous nonlinear least-squares fit of the shifts of all
available signals from two independent titrations run at different salt
concentrations. Global standard deviation of the fit s=0.00016 ppm
(RMS weighted residual=0.00015). [b] The formation constant was kept
invariant in the final refinement. [c] The ion-pair formation constant was
taken from Table 1 and kept invariant in the refinement.
of the latter from Table 1 was used in the fit and kept invari-
ant in the refinement.
In the next step, the two monotopic receptors R and B
were combined into the architecture of ditopic receptor D,
designed for the independent binding of both ionic partners
of QX.
Assessing affinities: Having the formation constants of all
the complex species available for the three receptors, in the
next step we need to assess the affinities of X for receptors
R and D, and of Q for receptors B and D. The question
arises how to evaluate affinities for systems of three re-
agents involving more than one binding constant in which
complex species of different stoichiometry are present. It
may be tempting to compare the formation constants of cor-
responding ternary complexes, such as RQX and DQX. Fol-
lowing this approach, we can easily appreciate that the
cation-binding site does not enhance the affinity of the di-
topic receptor for the anion. However, such an approach is
misleading, because the affinities of the receptors do not
rely solely on ternary complexes; indeed, other species may
compensate the contribution from the ternary complexes.
Rather, what we need is to convert the set of association
constants into a parameter describing the overall affinity.
We have addressed this issue by using the median binding
concentration (BC50) parameter,[41] which we have shown to
be a generalized binding descriptor[45] useful for comparing
heterogeneous binding data and which we have recently ex-
tended to include the binding of salts.[46] The BC50 parame-
ter, which is defined as the total concentration of a reagent
(e.g., a receptor or a ligand) necessary for binding 50% of
an observed species (e.g., a ligand or a receptor), is calculat-
ed from the measured formation constants[47] and is best ex-
pressed as a function of the fraction of bound reagent (i.e.,
the reagent for which we want to evaluate the affinity),
which represents the saturation degree of the reagent itself.
Considering that for 1:1 complexes, when the fraction of
bound reagent is zero, BC50 coincides with the dissociation
constant Kd,[41] the BC50 parameter can be visualized as a
global dissociation constant. Affinities assessed through the
BC50 parameter are thus related to dissociation constants
and, consequently, to binding free energies. The BC50 value
is a convenient and practical affinity descriptor, as 1) it
takes into account the contribution from all the complex
species involved but is independent of the association
model, 2) it can be used for direct comparison of heteroge-
Indeed, due to steric gearing, the alternate topology of
substituents[43] is expected to allow the cation and the anion
to bind to opposite sides of the aromatic ring of D, in such a
way that each side would closely mimic its monotopic coun-
terparts R and B. The synthesis of receptor D was accom-
plished in seven steps from mesitylene with an overall unop-
timized yield of 10% (see the Supporting Information). Fol-
lowing the methodology described for R,[42] a reliable set of
formation constants could be accurately measured for the
complex species detected in the association of ditopic recep-
tor D with QX in CDCl3/CD3CN (80/20) at 298 K (See Sup-
porting Information). The results are reported in Table 3 as
cumulative formation constants.
As expected, besides the complexes of the free ions, com-
plexes with one and two ion pairs were also formed, where-
as dimerization of the receptor was negligible. As for B, the
value of the ion-pair formation constant from Table 1 was
used in the fit and kept invariant in the refinement. The six-
constant model, although quite complex, is the only model
giving an excellent fit with experimental data, far better
than any other model tested, so it can be confidently consid-
ered a reliable description of the system.[44]
8298
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 8296 – 8302

Ion-Pair Binding
FULL PAPER
neous systems because it has the units of a concentration in
all cases, and 3) its usage is straightforward as, in analogy to
the widespread IC50 parameter, the lower the BC50 value,
the higher the affinity. In the present study, the appropriate
approach will consist of calculating the BC50 values of each
ion toward the ditopic receptor from the sets of formation
constants in Tables 1–3 and comparing these values to those
calculated toward the respective monotopic receptors. Con-
sidering that the actual affinity varies with the saturation of
the reagent, the comparison is best accomplished by plotting
the BC50 values against the fraction of bound reagent in the
entire complexation range. Thus, from comparison of the
BC50 curves, the occurrence of binding cooperativity, if any,
exerted by the ditopic receptor can be easily evidenced and
quantitatively assessed.
To define a three-reagent system, BC50 calculation re-
quires fixing, besides the conditions for the two binding
partners, the concentration of the counterion;[42] this can be
done by imposing the condition that the electroneutrality of
the solution must be maintained. Such a condition is satis-
fied by choosing, for each value of the fraction of bound re-
agent, a free concentration of the counterion that will main-
tain the total concentration of the cation equal to that of the
anion. Using the “BC50 calculator” program,[47] BC50 values
were thus calculated for both ions under these conditions
and the results are reported in Figures 1 and 2 (see the Sup-
porting Information). Examination of the affinity profiles of
Q for the monotopic receptor B and the ditopic receptor D
depicted in Figure 1a reveals for the BC50 parameter a hy-
perbolic trend featuring saturation behavior for both recep-
tors, with values asymptotically approaching the infinite for
complete complexation.
Figure 1. a) Plot of the BC50 [molLÀ1] curves calculated for Q toward
monotopic receptor B (*) and the ditopic receptor D (&) at total con-
centration TX =TQ as a function of the fraction of bound cation. BC50
values were calculated from the formation constants of Tables 2 and 3 as
described in ref. [42], by using the BC50 calculator program.[47] Limiting
BC050 values are: 1.95(9)ꢂ10À1 molLÀ1 (B) and 1.19(3)ꢂ10À2 molLÀ1 (D).
b) Plot of the BC50(B)/BC50(D) ratio at TX =TQ as a function of the frac-
tion of bound cation.
At the other end, the curves intersect the BC50 axis for
x
bound cation =0 (BC050), where the unbound cation is forming
the first complex molecule, giving the intrinsic affinities for
the two receptors. The corresponding BC050 values were
1.95(9)ꢂ10À1 molLÀ1 (B) and 1.19(3)ꢂ10À2 molLÀ1(D).
Clearly, the curve of the ditopic receptor lies below the
curve of the monotopic receptor in the whole complexation
range, including the BC500 value, that is, Q indeed binds
more effectively to the ditopic than to the monotopic recep-
tor at any degree of saturation. Considering that the cation
can bind to only one side of the ditopic receptor, which is
identical to the monotopic receptor, it is evident that the
cation “feels” a cooperative contribution from simultaneous
binding of the anion, whatever complex species is involved
in the association. A quantitative assessment of cooperativi-
ty can be obtained from the ratio of the BC50 curves, shown
in Figure 1b, which shows a 16-fold preference of Q for the
ditopic with respect to the corresponding monotopic recep-
tor up to 80% complexation, in terms of cation required for
binding 50% of receptor. The preference decay exhibited
for larger extents of complexation clearly shows that the
BC50 ratio is not constant along the complexation range.
This effect is due to the variable contribution to the BC50
parameter made by complex species involving the counter-
ion, which depend on the (variable) concentration of the
latter imposed by the electroneutrality condition.[42] The in-
trinsic affinity parameter BC050 therefore cannot be used for
direct comparison as in two-reagent systems,[45] since these
contributions vanish when the concentration of the counter-
ion (along with the fraction of bound cation) is zero.
A different situation is apparent when considering the
binding of the anion to the mono- and ditopic receptors
(Figure 2). Up to nearly 85% complexation the curve of the
ditopic receptor lies above that of the monotopic receptor,
that is the chloride anion binds to the monotopic receptor
with a modest but unambiguous preference. Such a prefer-
ence can be quantified through the BC50 ratio (Figure 2b),
which shows an intrinsic twofold advantage in binding to the
monotopic with respect to the ditopic receptor, an advant-
age that decreases with increasing saturation and vanishes at
85% complexation, inverting to a preference for the ditopic
Chem. Eur. J. 2009, 15, 8296 – 8302
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8299

S. Roelens et al.
Figure 2. a) Plot of the BC50 [molLÀ1] curves calculated for X toward
monotopic receptor R (*) and ditopic receptor D (&) at total concen-
tration TQ =TX as a function of the fraction of bound anion. BC50 values
were calculated from the formation constants of Tables 1 and 3 as de-
scribed in ref. [42], by using the BC50 calculator program.[47] BC500 values
are 5.0(1)ꢂ10À3 molLÀ1 (R) and 1.05(2)ꢂ10À2 molLÀ1 (D). b) Plot of the
BC50(D)/BC50(R) ratio at TQ =TX as a function of the fraction of bound
anion.
Figure 3. Plot of the distribution of complex species of a) receptor B and
b) receptor D with QX, expressed as fraction of the total complexed
cation, versus the fraction of bound cation. Distributions were calculated
at TX =TQ from the formation constants of Tables 2 and 3. a) BQX (*),
^
BQ (&); b) DQ2X2 (~), DQX (!), DQ ( ).
tion depicted in Figure 3b for Q binding to D shows a pre-
dominant contribution from the complex of the ditopic re-
ceptor with two ion pairs, with a minor (10%) contribution
from the complex with a single ion pair and a negligible con-
tribution from the complex of the free cation. The strongly
preferred binding of two ion pairs accounts for the observed
cooperativity. Although the global affinity is determined by
a balance of the actual thermodynamic stabilities of the
complexes formed, it is not surprising that, on a purely stoi-
chiometric basis, the ditopic receptor is more effective than
the monotopic in binding the cation. It is noteworthy, how-
ever, that for both receptors the observed affinity is due to
complexes of ion-paired species rather than of free cation,
in contrast to what is generally assumed. In contrast, a dif-
ferent picture is apparent for binding of the anion: while the
species distribution for the ditopic receptor D is very similar
to that of the cation (Figure 4b), with a slightly larger con-
tribution from the complex of the free anion, for the mono-
receptor above this value. In this case, for most of the com-
plexation range an inhibiting effect is evident, rather than
cooperative participation of the simultaneously bound
cation. It must be concluded that, in contrast to the cation,
binding of the anion suffers from an anticooperative effect
from concomitant binding of the counterion, that is, not
only is cooperativity not the rule, but it also may operate in
opposite directions for the two partners of an ion pair.
The observed behavior may be understood by examining
the distribution of the complex species responsible for the
affinity profiles (Figures 3 and 4, Figures S1 and S2 in the
Supporting Information). Concerning the cation, in Fig-
ure 3a it is apparent that the species largely determining the
global affinity of Q for B is the complex of the receptor
with the ion-paired cation, with very limited contribution
from the complex of the free cation. Likewise, the distribu-
8300
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 8296 – 8302

Ion-Pair Binding
FULL PAPER
Conclusions
Is binding both really binding better? We have provided a
quantitative and methodologically correct answer to this
question by conceptually re-examining and experimentally
testing the binding of an ion pair to a ditopic receptor.
When the appropriate algorithm is applied, it is clear that
cooperativity is neither general nor predictable and may
even operate in opposite directions for the two partners of
the ion pair. Several general aspects were focused on in this
analysis: 1) Complexes of ions and of ion pairs are incom-
mensurable; comparison between commensurable entities
(ions) is mandatory. 2) The 1:1 association model is inade-
quate to describe the binding of salts, because ion pairing
and ion-pair binding cannot, in general, be neglected.
3) Binding of the whole ion pair is equally feasible for both
the ditopic and the monotopic receptor. 4) Affinities of ions
cannot be assessed directly by comparing binding constants,
because multiple equilibria are always involved; for this pur-
pose, the BC50 can be used as a general and convenient de-
scriptor of global affinity. Hopefully, these aspects will be
taken into consideration in future studies.
Acknowledgements
The contribution of Dr. P. Gans, University of Leeds and of Prof. A. Sa-
batini, Universitꢀ di Firenze, to the upgrade of the HypNMR program is
gratefully acknowledged. Ente Cassa di Risparmio di Firenze (Italy) is
acknowledged for granting a 400 MHz NMR spectrometer.
[1] J. L. Sessler, P. A. Gale, W.-S. Cho, Anion Receptor Chemistry, RSC,
Cambridge, 2006, Chap. 6, pp. 259–293.
Figure 4. Plot of the distribution of complex species of a) receptor R and
b) receptor D with QX, expressed as fraction of the total complexed
anion, versus the fraction of bound anion. Distributions were calculated
at TQ =TX from the formation constants of Tables 1 and 3. a) RQX (*),
[2] B. D. Smith in Macrocyclic Chemistry: Current Trends and Future
Perspectives (Ed.: K. Gloe), Springer, Dordrecht, 2005, pp. 137–151.
[3] M. T. Reetz, C. M. Niemeyer, K. Harms, Angew. Chem. 1991, 103,
1515–1517; Angew. Chem. Int. Ed. Eng. 1991, 30, 1472–1474.
[4] D. J. White, N. Laing, H. Miller, S. Parsons, S. Coles, P. A. Tasker,
Chem. Commun. 1999, 2077–2078.
^
RX (&), R2X (&); b) DQ2X2 (~), DQX (!), DX ( ).
[5] M. J. Deetz, M. Shang, B. D. Smith, J. Am. Chem. Soc. 2000, 122,
6201–6207.
[6] J. M. Mahoney, A. M. Beatty, B. D. Smith, J. Am. Chem. Soc. 2001,
123, 5847–5848.
[7] M. Barboiu, G. Vaughan, A. van der Lee, Org. Lett. 2003, 5, 3073–
3076.
topic receptor R the contributions from complexes of the
free and the ion-paired anions are equivalent at low com-
plexation (Figure 4a); the complex of the ion-paired anion
is increasingly predominant for higher complexation, where-
as the R2X complex is below 10% in the whole complexa-
tion range. Considering that complexes of the ion pair are
apparently less effective than those of the free anion in con-
tributing to the affinity,[42] in agreement with previous find-
ings showing that electrostatic attraction of the counterion
weakens the interaction of the ion with the receptor,[48,49]
the affinity profile showing a preference of the anion for the
monotopic versus the ditopic receptor up to 85% complexa-
tion may be related to a significant weight of the RX com-
plex in the balance of the thermodynamic stabilities of the
complexes contributing to the overall affinity. Thus, the net
outcome appears to be the loss for the ditopic receptor of
the advantage of binding the free anion when the extent of
complexation increases.
[8] J. L. Atwood, A. Szumna, Chem. Commun. 2003, 940–942.
[9] R. Custelcean, L. H. Delmau, B. A. Moyer, J. L. Sessler, W.-S. Cho,
D. Gross, G. W. Bates, S. J. Brooks, M. E. Light, P. A. Gale, Angew.
Chem. 2005, 117, 2593–2598; Angew. Chem. Int. Ed. 2005, 44, 2537–
2542.
[10] C. Suksai, P. Leeladee, D. Jainuknan, T. Tuntulani, N. Muangsin, O.
Chailapakul, P. Kongsaeree, C. Pakavatchai, Tetrahedron Lett. 2005,
46, 2765–2769.
[11] J. L. Sessler, S. K. Kim, D. E. Gross, C.-H. Lee, J. S. Kim, V. M.
Lynch, J. Am. Chem. Soc. 2008, 130, 13162–13166.
[12] D. M. Rudkevich, J. D. Mercer-Chalmers, W. Verboom, R. Ungaro,
F. de Jong, D. N. Reinhoudt, J. Am. Chem. Soc. 1995, 117, 6124–
6125.
[13] L. A. J. Chrisstoffels, F. de Jong, D. N. Reinhoudt, S. Sivelli, L. Gaz-
zola, A. Casnati, R. Ungaro, J. Am. Chem. Soc. 1999, 121, 10142–
10151.
Chem. Eur. J. 2009, 15, 8296 – 8302
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8301

S. Roelens et al.
[14] A. V. Koulov, J. M. Mahoney, B. D. Smith, Org. Biomol. Chem. 2003,
1, 27–29.
[15] J. M. Mahoney, G. U. Nawaratna, A. M. Beatty, P. J. Duggan, B. D.
Smith, Inorg. Chem. 2004, 43, 5902–5907.
[16] P. D. Beer, P. K. Hopkins, J. D. McKinney, Chem. Commun. 1999,
1253–1254.
[17] H. Miller, N. Laing, S. Parsons, A. Parkin, P. A. Tasker, D. J. White,
J. Chem. Soc. Dalton Trans. 2000, 3773–3782.
[18] F. W. Kotch, V. Sidorov, Y.-F. Lam, K. J. Kayser, H. Li, M. S. Kauch-
er, J. T. Davis, J. Am. Chem. Soc. 2003, 125, 15140–15150.
[19] J. M. Mahoney, A. M. Beatty, B. D. Smith, Inorg. Chem. 2004, 43,
7617–7621.
[20] P. G. Plieger, P. A. Tasker, S. G. Galbraith, Dalton Trans. 2004, 313–
318.
[21] M. P. Wintergerst, T. G. Levitskaia, B. A. Moyer, J. L. Sessler, L. H.
Delmau, J. Am. Chem. Soc. 2008, 130, 4129–4139.
[22] J. Scheerder, J. P. M. van Duynhoven, J. F. J. Engbersen, D. N. Rein-
houdt, Angew. Chem. 1996, 108, 1172–1175; Angew. Chem. Int. Ed.
Engl. 1996, 35, 1090–1093.
[23] N. Pelizzi, A. Casnati, A. Friggeri, R. Ungaro, J. Chem. Soc. Perkin
Trans. 2 1998, 1307–1311.
[24] P. D. Beer, S. W. Dent, Chem. Commun. 1998, 825–826.
[25] S. Nishizawa, V. Shigemori, N. Teramae, Chem. Lett. 1999, 28, 1185–
1186.
[26] S. Kubik, J. Am. Chem. Soc. 1999, 121, 5846–5855.
[27] T. Tozawa, Y. Misawa, S. Tokita, Y. Kubo, Tetrahedron Lett. 2000,
41, 5219–5223.
[28] R. Shukla, T. Kida, B. D. Smith, Org. Lett. 2000, 2, 3099–3102.
[29] A. Arduini, G. Giorgi, A. Pochini, A. Secchi, F. Ugozzoli, J. Org.
Chem. 2001, 66, 8302–8308.
[34] P. T. Gunning, Org. Biomol. Chem. 2005, 3, 3877–3879.
[35] M. K. Chae, J. I. Lee, N. K. Kim, K. S. Jeong, Tetrahedron Lett. 2007,
48, 6624–6627.
[36] M. Cametti, M. Nissinen, A. Dalla Cort, L. Mandolini, K. Rissanen,
J. Am. Chem. Soc. 2007, 129, 3641–3648.
[37] M. D. Lankshear, I. M. Dudley, K. M. Chan, P. D. Beer, New J.
Chem. 2007, 31, 684–690.
[38] H. Miyaji, D.-S. Kim, B.-Y. Chang, E. Park, S.-M. Park, K. H. Ahn,
Chem. Commun. 2008, 753–755.
[39] D. E. Gross, F. P. Schmidtchen, W. Antonius, P. A. Gale, V. M.
Lynch, J. L. Sessler, Chem. Eur. J. 2008, 14, 7822–7827.
[40] M. Hamon, M. Menand, S. Le Gac, M. Luhmer, V. Dalla, I. Jabin, J.
Org. Chem. 2008, 73, 7067–7071.
[41] A. Vacca, C. Nativi, M. Cacciarini, R. Pergoli, S. Roelens, J. Am.
Chem. Soc. 2004, 126, 16456–16465.
[42] S. Roelens, A. Vacca, C. Venturi, Chem. Eur. J. 2009, 15, 2635–2644.
[43] G. Hennrich, E. V. Anslyn, Chem. Eur. J. 2002, 8, 2218–2224.
[44] Since direct experimental detection of all species in solution is nor-
mally unfeasible, rigorous analysis of the error of the fit is the best
evidence available of a chemical model: K. A. Connors, Binding
Constants, Wiley-Interscience, New York, 1987, Chap. 3, pp. 103–
138.
[45] C. Nativi, M. Cacciarini, O. Francesconi, A. Vacca, G. Moneti, A.
Ienco, S. Roelens, J. Am. Chem. Soc. 2007, 129, 4377–4385.
[46] The complete treatment of BC50 for three-reagent systems has been
described in detail.[41]
[47] To expedite the calculation of BC50 from the binding constants, the
computer program BC50 Calculator has been developed and made
available for free upon request. The equations used by the program
to compute BC50 values are described in refs. [44,41] for two- and
three-reagent systems, respectively.
[48] S. Bartoli, S. Roelens, J. Am. Chem. Soc. 2002, 124, 8307–8315.
[49] P. Sarri, F. Venturi, F. Cuda, S. Roelens, J. Org. Chem. 2004, 69,
3654–3661.
[30] J. L. Atwood, A. Szumna, J. Am. Chem. Soc. 2002, 124, 10646–
10647.
[31] J. M. Mahoney, J. P. Davis, A. M. Beatty, B. D. Smith, J. Org. Chem.
2003, 68, 9819–9820.
[32] P. R. A. Webber, P. D. Beer, Dalton Trans. 2003, 2249–2252.
[33] D. Garozzo, G. Gattuso, A. Notti, A. Pappalardo, S. Pappalardo,
M. F. Parisi, M. Perez, I. Pisagatti, Angew. Chem. 2005, 117, 4970–
4974; Angew. Chem. Int. Ed. 2005, 44, 4892–4896.
Received: February 7, 2009
Published online: July 14, 2009
8302
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 8296 – 8302