Relationship between chemical structure and supramolecular effective molarity for formation of intramolecular H-bonds

Article abstract of DOI:10.1021/ja406235d

Effective molarity (EM) is a key parameter that determines the efficiency of a range of supramolecular phenomena from the folding of macromolecules to multivalent ligand binding. Coordination complexes formed between zinc porphyrins equipped H-bond donor sites and pyridine ligands equipped with H-bond acceptor sites have allowed systematic quantification of EM values for the formation of intramolecular H-bonds in 240 different systems. The results provide insights into the relationship of EM to supramolecular architecture, H-bond strength, and solvent. Previous studies on ligands equipped with phosphonate diester and ether H-bond acceptors were inconclusive, but the experiments described here on ligands equipped with phosphine oxide, amide, and ester H-bond acceptors resolve these ambiguities. Chemical double-mutant cycles were used to dissect the thermodynamic contributions of individual H-bond interactions to the overall stabilities of the complexes and hence determine the values of EM, which fall in the range 1-1000 mM. Solvent has little effect on EM, and the values measured in toluene and 1,1,2,2-tetrachloroethane are similar. For H-bond acceptors that have similar geometries but different H-bond strengths (amide and ester), the values of EM are very similar. For H-bond acceptors that have different geometries but similar H-bond strengths (amide and phosphonate diester), there is little correlation between the values of EM. These results imply that supramolecular EMs are independent of solvent and intrinsic H-bond strength but depend on supramolecular architecture and geometric complementarity.

Full text of DOI:10.1021/ja406235d

Article
pubs.acs.org/JACS
Relationship between Chemical Structure and Supramolecular
Effective Molarity for Formation of Intramolecular H‑Bonds
Hongmei Sun, Christopher A. Hunter,* Cristina Navarro, and Simon Turega
Department of Chemistry, University of Sheffield, Sheffield S3 7HF, U.K.
S
* Supporting Information
ABSTRACT: Effective molarity (EM) is a key parameter that
determines the efficiency of a range of supramolecular phenomena
from the folding of macromolecules to multivalent ligand binding.
Coordination complexes formed between zinc porphyrins equipped
H-bond donor sites and pyridine ligands equipped with H-bond
acceptor sites have allowed systematic quantification of EM values for
the formation of intramolecular H-bonds in 240 different systems.
The results provide insights into the relationship of EM to
supramolecular architecture, H-bond strength, and solvent. Previous
studies on ligands equipped with phosphonate diester and ether H-
bond acceptors were inconclusive, but the experiments described here
on ligands equipped with phosphine oxide, amide, and ester H-bond
acceptors resolve these ambiguities. Chemical double-mutant cycles
were used to dissect the thermodynamic contributions of individual H-bond interactions to the overall stabilities of the
complexes and hence determine the values of EM, which fall in the range 1−1000 mM. Solvent has little effect on EM, and the
values measured in toluene and 1,1,2,2-tetrachloroethane are similar. For H-bond acceptors that have similar geometries but
different H-bond strengths (amide and ester), the values of EM are very similar. For H-bond acceptors that have different
geometries but similar H-bond strengths (amide and phosphonate diester), there is little correlation between the values of EM.
These results imply that supramolecular EMs are independent of solvent and intrinsic H-bond strength but depend on
supramolecular architecture and geometric complementarity.
of this phenomenon (geometric complementarity and pre-
organization lead to high affinity), quantification of the
magnitudes of these effects has proved more elusive.16−20
Perfect geometric complementarity and conformational rigidity
are usually difficult to achieve, so it would be useful to have
quantitative guidelines for estimating the penalties for imperfect
design and the potential benefits that would accrue from
synthetically more challenging targets. In this paper, we
describe experiments designed to quantify the relationship
between chemical structure and chelate cooperativity using a
synthetic model system.
The parameter used to quantify chelate cooperativity is
effective molarity (EM), which is defined as the ratio of the
intramolecular association constant for an interaction to the
corresponding intermolecular association constant for the same
interaction in a reference system.21−23 This is a thermodynamic
effective molarity; it is also possible to define a kinetic effective
molarity in the same way by using rate constants in place of
equilibrium constants. A large body of experimental data on
EM for intramolecular covalent bond formation has been
collected, and values as high as 1014 M have been reported.22
Increasing the conformational flexibility of the chain linking the
INTRODUCTION
■
Molecular recognition and self-assembly processes depend on
the interplay of multiple intermolecular interactions. Cooper-
ativity between different non-covalent interaction sites in a
complex interface can have dramatic effects on the overall
binding affinities of the components. This phenomenon has
been extensively exploited in the design of high-affinity
synthetic receptors for ions and in the development of
multivalent ligands for biological receptors.1−7 However,
understanding the relationship between chemical structure
and the magnitude of cooperative effects remains an important
challenge that must be overcome if we are to develop reliable
strategies for the design of new supramolecular systems with
predictable properties.
Cooperativity can take the form of allosteric interactions that
alter the intrinsic binding properties at different interaction
sites, as in hemoglobin, or chelate cooperativity that leads to an
enhanced binding affinity in systems that feature multiple
intermolecular interactions.8,9 Chelate cooperativity leads to
many of the all-or-nothing processes that are characteristic
features of biological systems that switch between different
states over a narrow range of conditions.10−13 Chelate
cooperativity is also central to the design of high-affinity
molecular recognition systems and efficient self-assembly
processes.14,15 Although we have a qualitative understanding
Received: June 22, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja406235d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. (a) Formation of an intermolecular H-bond. (b) Stepwise equilibria in the formation of a porphyrin−pyridine complex containing an
intramolecular H-bond. Kref is the equilibrium constant for formation of the intermolecular H-bond. K0 is the intermolecular association constant for
formation of the zinc−nitrogen interaction. KrefEM is the equilibrium constant for formation of the intramolecular H-bond, and EM is the effective
molarity for the intramolecular interaction.
two reaction sites and decreasing the geometric complemen-
tarity of the linker can reduce the value of EM by many orders
of magnitude.24,25 The experimental data that are available on
EM for intramolecular non-covalent bond formation suggest
that supramolecular EMs are quite different from covalent EMs.
The values of non-covalent EMs do not exceed 103 M and are
subject to less dramatic variation than covalent EMs.26 We have
begun a systematic study using porphyrin−pyridine complexes
to quantify the variation in supramolecular EM with geometric
complementarity, conformational flexibility, intrinsic functional
group interaction strength, and solvent.27−34 The results that
we have reported to date show conflicting trends, so in this
paper, we significantly expand the number of different systems
studied and compare measurements of intramolecular H-
bonding interactions in 240 different complexes.
Approach. We previously developed the porphyrin−ligand
system shown in Figure 1b for the quantification of chelate
cooperativity in complexes that make multiple intermolecular
interactions.27 The pyridine−zinc coordination interaction in
Figure 1b gives a complex in which the formation of an
intramolecular H-bond between the porphyrin phenol groups
and H-bond acceptors on the ligand is possible. Strictly, this H-
bond is a second intermolecular interaction, but in the stepwise
pathway shown in Figure 1b, the H-bond is an intracomplex
interaction, so we will refer to this as an intramolecular
interaction. The double-mutant cycle (DMC) shown in Figure
2 dissects out the thermodynamic contribution of the
intramolecular acceptor·donor (A·D) H-bond to the overall
stability of complex A.35,36 The DMC works by removing the
thermodynamic contributions due to the secondary interactions
that are highlighted in Figure 2. For example, secondary
interactions associated with mutation of the ligand from LA to
L contribute to the difference between the stabilities of
complexes A and B and to the difference between the stabilities
of complexes C and D. Similarly, secondary interactions
associated with mutating the porphyrin from PD to P
contribute to the difference between the stabilities of complexes
A and C and to the difference between the stabilities of
complexes B and D. In eq 1, these differences cancel out,
providing a direct measurement of the free energy contribution
due the A·D H-bond in complex A (ΔΔG°):
°
°
°
°
ΔΔG° = ΔGA − ΔGB − ΔGC + ΔGD
(1)
The EM for the intramolecular interaction can be determined
by comparing the intramolecular interaction with the
corresponding intermolecular interaction measured in a
reference system (Figure 1a). Figure 1b shows that the
equilibrium constant for formation of an intramolecular H-
bond can be expressed as KrefEM, where Kref is the equilibrium
constant for the intermolecular process illustrated in Figure 1a.
This separates the thermodynamic properties of the intra-
molecular interaction into Kref, which is related to the intrinsic
properties of the functional group interaction, and EM, which
measures the influence of the supramolecular architecture on
chelate cooperativity.
In previous work, we investigated chelate cooperativity using
pyridine ligands equipped with either phosphonate diester or
ether H-bond acceptors in two different solvents, toluene and
1,1,2,2-tetrachloroethane (TCE).27,28,31,32 In TCE, the two
different types of H-bond have similar values of EM when they
are mounted on the same supramolecular framework.32 The
values of EM were similar in toluene and TCE for the ether
ligands, but for the phosphonate diester ligands, the values of
EM in toluene were significantly higher than those in TCE.31,32
The phenol−ether H-bond is much weaker than the phenol−
phosphonate diester H-bond, so the number of ether ligands
for which it was possible to measure the value of EM was
relatively small. There are also some differences between the
geometries and steric properties of the ether and phosphonate
diester ligands (Figure 3a), which could account for the
Figure 2. Chemical double-mutant cycle (DMC) for measurement of
the free energy contribution of the A·D H-bond to the stability of
complex A. P is a zinc porphyrin, PD is a zinc porphyrin equipped with
a H-bond donor (D), L is a pyridine ligand, and LA is a pyridine ligand
equipped with a H-bond acceptor (A).
B
dx.doi.org/10.1021/ja406235d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
different positions, and P1b−P4b are the corresponding
methoxy-substituted porphyrins used as the non-H-bonding
mutants in the DMCs. The synthesis of the porphyrins was
reported previously.27 The syntheses of the ligands featuring
phosphonate diester or ether H-bond acceptors (La and Ld)
and the non-H-bonding mutants (Lb and Lc) have been
described in previous publications.27,28
Syntheses. The ligands equipped with tertiary amide, ester,
and phosphine oxide H-bond acceptor sites (Le, Lf, and Lg)
were prepared from the corresponding pyridinecarboxylic acids
via the acid chlorides (Scheme 1). Two of the alcohols used in
the acid chloride couplings are commercially available, N,N-
diethyl-2-hydroxyacetamide and ethyl glycolate. Alcohol 1
required for the phosphine oxide ligands was prepared in one
step from di-tert-butylchlorophosphine and formaldehyde
(Scheme 2a).
Figure 3. Pyridine ligands having an oxygen H-bond acceptor atom
(red) located at the same point on the ligand framework: (a)
previously reported phosphonate diester and ether ligands; (b)
phosphine oxide, amide, and ester ligands reported in this paper.
Phosphine oxide 2 was synthesized in order to measure the
H-bonding properties of this functional group, in particular the
value of Kref for use in determining the EM for intramolecular
phosphine oxide−phenol H-bonds. Scheme 2b shows the
synthetic route, which gave 2 in low yield but in sufficient
difference in behavior. In the present work, we expanded the
range of ligand systems to include phosphine oxide, amide, and
ester H-bond acceptor ligands (Figure 3b) to provide a more
comprehensive data set of supramolecular EMs. The five ligand
systems illustrated in Figure 3 all have a H-bond acceptor
oxygen atom (highlighted in red) located at the same point on
the ligand framework. Comparison of the values of EM
measured for a range of different functional groups mounted on
the same ligand framework provides a good test of whether the
EM for an intramolecular H-bond is influenced by the intrinsic
properties of the functional group interaction, the overall
stability of the complex, or the solvent or is simply a function of
the supramolecular architecture.
37
1
quantities for use in H NMR titrations.
An additional ligand Q equipped with a pyromellitimide
electron acceptor moiety was synthesized for use in
competition titrations to characterize very high affinity
complexes (Scheme 3). Condensation of pyromellitic anhy-
dride with n-hexylamine and 3-(aminomethyl)pyridine gave a
mixture of products, which were separated by column
chromatography to obtain the required unsymmetrical
pyromellitimide ligand Q.38
Measurement of Porphyrin−Ligand Binding Interac-
tions. Binding of the ligands to the porphyrins was studied in
toluene and in TCE using titrations monitored by UV/vis
absorption spectroscopy. There were 144 different porphyrin−
ligand combinations, but for association constants lower than
106 M−1, the titration experiments could be automated using a
UV/vis plate reader.27 Binding of pyridine to a zinc porphyrin
leads to a large change in the Soret absorption band, and a well-
defined isosbestic point indicates the formation of a 1:1
complex. The UV/vis absorption titration data fit well to a 1:1
binding isotherm, and the results are reported in the
Supporting Information.
RESULTS AND DISCUSSION
■
The structures of the porphyrins and ligands used in this study
are shown in Figures 4 and 5, respectively. P1a−P4a are zinc
porphyrins equipped with peripheral H-bond donor sites at
Some porphyrin−ligand combinations gave exceptionally
stable complexes, particularly the phosphine oxide ligands in
toluene, where the H-bonding interactions are strongest. For
these systems, fluorescence titrations were required in order to
measure the association constants. A plate reader equipped with
fluorescence detection allowed automated measurement of
association constants in the range 106 to 107 M−1. A more
sensitive fluorimeter was used to measure association constants
in the range 107 to 108 M−1 by manual titrations. In all cases,
the titration data fit well to a 1:1 binding isotherm. However for
some of the complexes, the association constants were too high
for reliable measurement by these methods, and a competition
experiment was therefore developed. When ligand Q equipped
with a pyromellitimide moiety, which is a good electron
acceptor, was added to any of the zinc porphyrins, the
porphyrin fluorescence was quenched by photoinduced
electron transfer from the porphyrin excited state to the
pyromellitimide (Figure 6a).38 Subsequent addition of a
different ligand that binds more strongly to the porphyrin
displaced Q and restored the porphyrin fluorescence (Figure
6b). This system provided a convenient competition binding
Figure 4. Porphyrin receptors P1a−P4a that can make H-bonds and
P1b−P4b that cannot.
C
dx.doi.org/10.1021/ja406235d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 5. Pyridine ligands.
Scheme 1
assay that could be monitored by fluorescence, allowing
accurate measurement of very high association constants.
The titration data for the ligand displacement experiments
could be fit to a simple 1:1 binding isotherm provided that Q
was present in a large excess relative to the porphyrin and that
the total concentration of Q was constant throughout the
titration. Under these conditions, the apparent value of the
association constant obtained by fitting the data to a 1:1
binding isotherm, Kapp, is related to the true association
constant for formation of the porphyrin−ligand complex, KL,
by eq 2:
KL
Kapp
=
1 + KQ [Q]
(2)
D
dx.doi.org/10.1021/ja406235d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 2
Figure 6. (a) Fluorescence spectra of P3a (5 μM) recorded during a
titration with ligand Q in toluene at 298 K (red corresponds to free
P3a and blue to the end point of the titration). (b) Fluorescence
spectra recorded during a titration of L6g into a solution containing
0.4 μM P3a and 10 mM Q in toluene at 298 K (blue corresponds to
the P3a·Q complex and green to the P3a·L6g complex). The
fluorescence emission intensity (I) is plotted in arbitrary units, and the
excitation wavelength was 427 nm in both cases.
where KQ is the association constant for the formation of the
porphyrin·Q complex and [Q] is the total concentration of Q
(see the Experimental Section for details). Thus, the ligand
solution added to the porphyrin solution during the titration
had to contain the same concentration of Q as present in the
porphyrin solution.
absent. Among the LA·PD complexes, the phosphine oxide
ligand complexes are more stable than the amide ligand
complexes, which in turn are more stable than the ester ligand
complexes. This is consistent with the H-bond acceptor
parameters for these functional groups (β = 10.2 for a
phosphine oxide, β = 8.5 for an amide, and β = 5.3 for an
ester).39 The free energy contributions due to intramolecular
H-bonding interactions, ΔΔG°, were determined using eq 1,
and the results are presented in Tables 1−6.
Figure 7 compares fluorescence titration data collected under
these conditions with the corresponding UV/vis absorption
data for the titration of L6g with P3a. The association constant
for this complex is KL = 8 × 108 M−1, and the UV/vis
absorption data could not be used to determine the association
constant because the tight binding limit was reached under the
conditions of the experiment (Figure 7a). Addition of 10 mM
Q reduced Kapp to 4 × 106 M−1, allowing straightforward
determination of KL by the fluorescence displacement experi-
ment (Figure 7b). Equation 2 shows that varying the
concentration of Q enables Kapp to be adjusted to a
conveniently measurable value, allowing experiments to be
carried out in different concentration regimes for different
complexes, which makes this a particularly versatile technique.
The association constants for these complexes span 8 orders
of magnitude, ranging from 101 to 109 M−1 depending on the
H-bonding functionality, solvent, and porphyrin−ligand com-
plementarity. Figure 8 shows the relationship between the
association constants measured in TCE and the corresponding
values measured in toluene. There is a reasonable correlation
between the results obtained in the two solvents. In all cases,
the association constants are 1−2 orders of magnitude bigger in
toluene than in TCE, which is more polar and competes with
both the zinc−nitrogen coordination and intramolecular H-
bonding interactions. Dissection of the relative contributions of
the coordination bond and H-bonds to the overall association
constant requires the DMC analysis illustrated in Figure 2.
DMC Analysis. The association constants are illustrated
graphically in Figure 9. The complexes are organized and
colored according to their role in the DMC in Figure 2. In
general, the LA·PD complexes, where intramolecular H-
bonding is possible, are more stable than the reference
complexes LA·P, L·PD, and L·P, where these H-bonds are
Intramolecular H-bonds contribute up to 26 kJ mol−1 to the
free energy or 4 orders of magnitude to the stability of the
porphyrin−ligand complexes. In toluene, all 24 of the
phosphine oxide ligand−porphyrin combinations formed
detectable intramolecular H-bonds, but the variation in
geometric complementarity in these systems led to large
variations in the free energy contribution from these
interactions. The values of ΔΔG° for the amide ligand
complexes were somewhat lower than those for the phosphine
oxide ligand complexes, but 22 of the 24 porphyrin−amide
ligand combinations formed detectable intramolecular H-bonds
in toluene (20 out of 24 in TCE). In contrast, values of ΔΔG°
for the ester ligand complexes were much lower, and only three
of the 24 different porphyrin−ester ligand combinations
formed detectable intramolecular H-bonds. These results are
consistent with the H-bond acceptor parameters for the three
functional groups: phosphine oxides (β = 10.2) are more polar
than amides (β = 8.5), which in turn are much more polar than
esters (β = 5.3).39 In general, the H-bonding interactions were
stronger in toluene than in TCE, which reflects the relative
polarity of the two solvents.
One assumption of the DMC approach is that the free
energy contributions that different interactions make to the
overall stability of a complex are additive, and the data in Tables
1−6 provide an opportunity to test the validity of this
assumption.40 The L3 and L6 ligands are symmetrical two-
Scheme 3
E
dx.doi.org/10.1021/ja406235d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 7. (a) Data recorded at two different wavelengths during an automated UV/vis absorption titration of L6g into P3a (5 μM) in toluene at 298
K. The lines correspond to best fits to a 1:1 binding isotherm with a linear correction for nonspecific effects (the curvature during the first phase of
the titration is due to the increase in path length during a plate reader titration). (b) Data recorded at two different wavelengths during a fluorescence
displacement experiment in which a solution containing 40 μM L6g and 10 mM Q was titrated into a solution containing 0.4 μM P3a and 10 mM Q
in toluene at 298 K (excitation at 427 nm). The lines corresponds to best fits to a 1:1 binding isotherm with a linear correction for nonspecific
effects.
Figure 8. Comparison of association constants (log K/M−1) measured
in toluene with the corresponding values measured in TCE. Data for
phosphine oxide (Lg), amide (Le), and ester (Lf) ligands are shown in
black, blue, and red, respectively. The line corresponds to log
K(TCE)/M−1 = log K(toluene)/M−1.
armed versions of the L2 and L5 ligands, respectively. If the
free energy contributions from H-bonding in these complexes
are additive, then the values of ΔΔG° measured for the two-
armed ligands should be double the values measured for the
corresponding one-armed ligands. Figure 10 shows that this is
indeed the case, confirming the additivity in these systems.
Pairwise differences in free energy between mutant
Figure 9. Association constants (log K/M−1) for formation of
complexes where the H-bonding functionality is missing, ΔGB°
porphyrin−ligand complexes at 298 K in (a) toluene and (b) TCE.
− ΔGD° and ΔGC° − ΔG°D, provide information on the
The data are color-coded according to the DMC in Figure 2: LA·PD
(complex A) in blue, L·PD (complex B) in yellow, LA·P (complex C)
in green, and L·P (complex D) in red.
magnitude of secondary interactions in these systems. Figure
11a compares the values of ΔG°B − ΔGD° measured in TCE with
the corresponding values measured in toluene. For most of the
systems, the values are close to zero, indicating that there are no
significant secondary interactions. However, there are six
systems where complex B is between −3 and −8 kJ mol−1
more stable than complex D in both solvents. This is clear
evidence for attractive secondary interactions in complex B. We
previously identified these secondary interactions as ester−
phenol H-bonds involving the ligand linker group in the
complexes of L5b and L6b with porphyrins P1a, P2a, and P3a
(Figure 12).27 The same interactions are present in complex A
of the DMC, so the contribution from these secondary
interactions cancels out in the DMC as shown in Figure 12,
and it is possible to obtain reliable values for intramolecular H-
bonding interactions in the L5 and L6 ligand series.
Figure 11b compares values of ΔG°C − ΔGD° measured in
TCE with the corresponding values measured in toluene. Many
of the systems give values close to zero, but there are two
groups of data points that deviate significantly from zero. The
ester (red) and amide (blue) ligands in the L1, L2, and L3
ligand series give positive values of ΔG°C − ΔGD° in TCE. This
implies that there are secondary interactions with the H-bond
acceptor functionality that destabilize complex C, and we
F
dx.doi.org/10.1021/ja406235d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 1. Free Energy Contributions from Phosphine Oxide−
Phenol H-Bonds (ΔΔG°/kJ mol−1) Determined Using the
Table 5. Free Energy Contributions from Ester−Phenol H-
Bonds (ΔΔG°/kJ mol−1) Determined Using the DMC in
a
a
DMC in Figure 2 at 298 K in Toluene
Figure 2 at 298 K in Toluene
L1g
L2g
L3g
L4g
L5g
L6g
L1f
L2f
L3f
L4f
L5f
L6f
P1a
P2a
P3a
P4a
−7
−8
−6
−2
−13
−7
−9
−26
−13
−20
−15
−8
−9
−13
−10
−10
−7
−7
−21
−12
−18
−20
P1a
P2a
P3a
P4a
0
0
−3
−1
−1
0
−5
−1
−2
0
1
0
0
1
0
1
1
0
−1
0
0
0
−5
−9
−1
0
a
a
Oxygen−zinc coordination interactions in complexes A and C mean
The average error over the data set was 1 kJ mol−1. Values of ΔΔG°
for complexes that do not make detectable H-bonds are set in italic
type.
that these values underestimate the contribution of intramolecular H-
bonds by up to 6 kJ mol−1.
Table 2. Free Energy Contributions from Phosphine Oxide−
Phenol H-Bonds (ΔΔG°/kJ mol−1) Determined Using the
Table 6. Free Energy Contributions from Ester−Phenol H-
Bonds (ΔΔG°/kJ mol−1) Determined Using the DMC in
a
a
DMC in Figure 2 at 298 K in TCE
Figure 2 at 298 K in TCE
L1g
L2g
L3g
L4g
L5g
L6g
L1f
L2f
L3f
L4f
L5f
L6f
P1a
P2a
P3a
P4a
−3
−5
−4
0
−10
−3
−6
−19
−7
−14
−12
−8
−6
−10
−8
−9
−6
−7
−7
−14
−12
−14
−13
P1a
P2a
P3a
P4a
0
0
0
1
−3
−1
−1
1
−5
−1
−2
0
0
0
0
0
1
0
−1
−1
0
−1
−1
−2
0
a
a
Oxygen−zinc coordination interactions in complexes A and C mean
The average error over the data set was 1 kJ mol−1. Values of ΔΔG°
for complexes that do not make detectable H-bonds are set in italic
type.
that these values underestimate the contribution of intramolecular H-
bonds by up to 5 kJ mol−1. Values of ΔΔG° for complexes that do not
make detectable H-bonds are set in italic type.
Table 3. Free Energy Contributions from Amide−Phenol H-
Bonds (ΔΔG°/kJ mol−1) Determined Using the DMC in
a
Figure 2 at 298 K in Toluene
L1e
L2e
L3e
L4e
L5e
L6e
P1a
P2a
P3a
P4a
−5
−5
−3
−2
−10
−4
−6
−19
−8
−11
0
−3
−3
−5
−5
−4
−6
−7
−2
−11
−10
−12
−6
−1
a
The average error over the data set was 1 kJ mol−1. Values of ΔΔG°
for complexes that do not make detectable H-bonds are set in italic
type.
Table 4. Free Energy Contributions from Amide−Phenol H-
Bonds (ΔΔG°/kJ mol−1) Determined Using the DMC in
Figure 10. Total free energy contributions due to intramolecular H-
bonding for ligands with two identical side arms, ΔΔG°(2), compared
with data for the corresponding one-armed ligands, ΔΔG°(1). Data
for phosphine oxide (Lg), amide (Le), and ester (Lf) ligands are
shown in black, blue, and red, respectively. The line corresponds to
ΔΔG°(2) = 2ΔΔG°(1).
a
Figure 2 at 298 K in TCE
L1e
L2e
L3e
L4e
L5e
L6e
P1a
P2a
P3a
P4a
−2
−3
−2
1
−8
−3
−4
1
−14
−4
−8
−2
−1
−3
−3
−4
−4
−3
−2
−7
−7
−7
−4
chemical shift were observed for the signal due to the
phosphine oxide tert-butyl protons (see the Supporting
Information). This implies that the phosphine oxide group
sits over the center of the porphyrin ring current in these
complexes, suggesting that the phosphine oxide oxygen may
compete with the pyridine nitrogen for coordination to the
zinc. To assess the extent of the problem, we titrated P1b into
−1
a
The average error over the data set was 1 kJ mol−1. Values of ΔΔG°
for complexes that do not make detectable H-bonds are set in italic
type.
previously suggested that this could be due to adverse steric
effects associated with solvation of the polar functional groups
by the bulky solvent.30 In contrast, the phosphine oxide ligands
(black) give negative values of ΔG°C − ΔGD° in toluene,
indicating that there are secondary interactions that stabilize
complex C in these systems. The NMR studies discussed below
shed some light on the origin of this behavior.
1
ligand L2g. Figure 13 shows the results of fitting the H NMR
titration data to a 2:1 binding isotherm. The limiting
complexation-induced changes in chemical shift for formation
of the 1:1 complex (K1 = 8 × 103 M−1) were −1.7 ppm for the
tert-butyl signal and −0.5 ppm for the signal due to the pyridine
α proton, and the subsequent changes in chemical shift for
formation of the 2:1 complex (K2 = 7 × 102 M−1) were −0.7
ppm for the tert-butyl signal and −1.5 ppm for the signal due to
the pyridine α proton. This suggests that the predominant
binding mode in the P1b·L2g 1:1 complex is coordination of
1
Structural Information on Binding Modes. H NMR
spectra of 1:1 mixtures of porphyrins and ligands were recorded
in toluene and in TCE. For some of the complexes in toluene,
large upfield complexation-induced changes in 1H NMR
G
dx.doi.org/10.1021/ja406235d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 11. Secondary interactions measured in the DMCs. (a) Comparison of ΔG°B − ΔGD° measured in TCE with corresponding values measured
in toluene. (b) Comparison of ΔGC° − ΔG°D measured in TCE with corresponding values measured in toluene. Data for phosphine oxide (Lg), amide
(Le), and ester (Lf) ligands are shown in black, blue, and red, respectively.
Figure 12. DMC used to measure the magnitude of the phosphine oxide−phenol H-bond in the P3a·L5g complex (complex A). The secondary
ester−phenol H-bonding interactions, which are present in complexes A and B for ligand series L5 and L6, give rise to large negative values of ΔGB°
− ΔGD° but cancel out in the DMC.
zinc by the phosphine oxide oxygen, and the second weaker
binding event in the 2:1 complex is coordination of a second
zinc porphyrin by the pyridine nitrogen. The values of K1 and
K2 indicate that phosphine oxide coordination is about 10 times
stronger than pyridine coordination, and this limits the
accuracy of the DMC analysis for these systems. The values
of ΔΔG° measured by DMCs for the phosphine oxide ligands
underestimate the free energy contributions due to intra-
molecular H-bonds by as much as 6 kJ mol−1 (see the
Supporting Information).
Effective Molarities. The values of ΔΔG° for intra-
molecular H-bonding can be converted into effective molarities
(EM) using the corresponding association constants for
formation of intermolecular H-bonds, Kref. Figure 14 shows
the structures of the compounds used to measure values of Kref
by 1H NMR titrations with p-cresol in TCE and in toluene. The
titration data fit well to a 1:1 binding isotherm, and the results
are shown in Table 7. It is also possible to estimate the value of
Kref using Kcalc from eq 3,
−RT ln Kcalc = −(α − αS)(β − β ) + 6 kJ mol−1
(3)
S
where α and β are the H-bond parameters for the H-bond
donor and acceptor and αS and βS are the H-bond parameters
for the solvent. The agreement with the experimentally
determined values is good (Table 7).39
In a complex held together by multiple non-covalent
interactions, the bound state is actually a mixture of partially
and fully bound states.27,34 For example, the complex shown in
Figure 1a is a mixture the partially bound state, where only the
zinc−nitrogen bond is formed, and the fully bound state, where
both the zinc−nitrogen bond and the intramolecular H-bond
are formed. The association constant that is measured
experimentally, Kobs, is the sum of the association constants
of all bound states (eq 4).
H
dx.doi.org/10.1021/ja406235d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 13. Data from a 1H NMR titration of P1b into L2g in toluene-d8 at 298 K. The changes in chemical shift for the signals due to the tert-butyl
group (gray) and the pyridine α proton (black) are shown along with the lines of best fit to a 2:1 binding isotherm. The predominant binding modes
in the 1:1 and 2:1 complexes are shown.
Table 8. Effective Molarities (EM, in mM) for
Intramolecular Amide−Phenol H-Bonds Measured at 298 K
a
in Toluene
L1e
L2e
L3e
L4e
L5e
L6e
Figure 14. Compounds used to measure K(O) for coordination of H-
bond acceptor groups to a zinc porphyrin and Kref for intermolecular
H-bonding interactions with p-cresol.
P1a
P2a
P3a
P4a
17
16
7
130
15
140
14
7
5
19
38
85
5
51
36
36
8
33
27
22
19
b
b
Table 7. Association Constants (M−1) for the Formation of
Intermolecular H-Bonds Measured by 1H NMR Titrations at
3
−
−
a
b
The average error over the data set was 50%. No interaction
detected.
298 K (Kref) and Estimated Using Equation 3 (Kcalc
)
solvent
complex
α
β
αs
βs
Kref
Kcalc
Table 9. Effective Molarities (EM, in mM) for
TCE
p-cresol·4
p-cresol·5
p-cresol·2
p-cresol·4
p-cresol·5
p-cresol·2
3.8
3.8
3.8
3.8
3.8
3.8
5.4
8.5
2.0
2.0
2.0
1.0
1.0
1.0
1.3
1.3
1.3
2.2
2.2
2.2
2
1
3
2
16
Intramolecular Amide−Phenol H-Bonds Measured at 298 K
a
TCE
22
in TCE
TCE
10.2
5.4
80 10
57
L1e
L2e
L3e
L4e
L5e
L6e
toluene
toluene
toluene
3
1
3
P1a
P2a
P3a
P4a
12
23
10
240
21
220
14
10
59
49
42
15
74
60
64
13
8.5
86 20
110
750
b
−
29
10.2
760 30
50
54
b
b
b
−
−
−
29
Kobs = fK0
(4)
a
b
The average error over the data set was 50%. No interaction
detected.
where K0 is the association constant for formation of only the
zinc−nitrogen coordination bond and f is given by eq 5:
allowing the value of ΔΔG° to be related to the values of Ki and
N
EMi by eq 6,
f = 1 + ∑ σKiEMi + ∑ σ KiEMiKjEMj + ...
i
ij
i
i,j
fA fD
e−ΔΔG°/RT
=
N
fB fC
(6)
+ σij...N ∏ KiEMi
(5)
i
where the values of fA, f B, f C, and fD are given by eq 5.
Solving eqs 5 and 6 for EMi using the values of ΔΔG° in
Tables 1−6 and the values of Ki in Table 7 gave the results
reported in Tables 8−11. Values of EM are not reported for the
phosphine oxide ligands because of contamination of the
DMCs by zinc−oxygen coordination complexes. The values of
EM for formation of intramolecular H-bonds in Tables 8−11
span 2 orders of magnitude, ranging from 3 to 240 mM.
Figure 15 compares the values of EM measured in TCE with
the corresponding values measured in toluene. There is good
agreement between the measurements in these two solvents for
in which Ki is the association constant for formation of the ith
intermolecular H-bond (i.e., Kref,i), EMi is the effective molarity
for formation of the ith intramolecular H-bond, and the
constants σ are statistical factors that account for the
degeneracies of the different bound states (see the Supporting
Information for details of the equations of used for different
complexes).
The zinc−nitrogen coordination bonds are not identical in all
of the complexes, but differences in K0 cancel out in the DMC,
I
dx.doi.org/10.1021/ja406235d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 10. Effective Molarities (EM, in mM) for
Intramolecular Ester−Phenol H-Bonds Measured at 298 K
a
in Toluene
L1f
L2f
L3f
L4f
L5f
L6f
b
b
b
b
P1a
P2a
P3a
P4a
−
170
150
−
−
−
−
−
−
−
−
−
−
−
−
b
b
b
b
b
b
−
−
−
−
b
b
b
b
b
−
38
b
b
b
b
b
b
−
−
−
a
b
The average error over the data set was 50%. No interaction
detected.
Table 11. Effective Molarities (EM, in mM) for
Intramolecular Ester−Phenol H-Bonds Measured at 298 K
a
in TCE
Figure 16. Comparison of effective molarities (EM) for formation of
intramolecular ester−phenol H-bonds (Lf) with the corresponding
values for amide−phenol H-bonds (Le). The line corresponds to log
EM(Lf)/M = log EM(Le)/M.
L1f
L2f
L3f
L4f
L5f
L6f
b
b
b
b
P1a
P2a
P3a
P4a
−
230
230
−
−
−
−
−
−
−
−
−
−
−
−
b
b
b
b
b
b
−
−
−
−
b
b
b
b
b
−
66
striking for the amide ligands (Figure 17b), where there are a
large number of data points available for comparison (r2 = 0.3
for 38 data points). Figure 18 compares the values of EM for
the ether ligands (Ld) with the corresponding values for the
amide and ester ligand families, and here the agreement is much
better. The results in Figures 16−18 suggest that although the
values of EM are similar for different functional groups, there is
sufficient variation in geometry, conformation, and steric bulk
to obscure correlations in the value of EM. The more similar
the geometry of the functional groups (e.g., esters and amides
in Figure 16), the more similar are the values of EM. The EM
values for ethers also show a reasonably good correlation with
the values for esters and amides, but the phosphonate diesters
are sufficiently different in structure to show qualitatively
different behavior.
b
b
b
b
b
b
−
−
−
a
b
The average error over the data set was 50%. No interaction
detected.
CONCLUSIONS
■
We previously used porphyrin−pyridine complexes to measure
values of EM for intramolecular H-bonding of phenols with
phosphonate diesters and ethers in toluene and in
TCE.27,28,31,32 Although some trends were observed, it was
difficult to draw general conclusions from those experiments. In
the present work, we expanded the range of H-bond acceptors
to include phosphine oxides, amides, and esters. Families of
pyridine ligands that place the oxygen H-bond acceptor of these
functional groups at the same point on the ligand framework
were synthesized. The interactions of these ligands with eight
different zinc porphyrins in two solvents (toluene and TCE)
were investigated using UV/vis absorption and fluorescence
titrations and, in the case of very high association constants
(>108 M−1), a competitive ligand displacement experiment.
The 288 association constants were used to construct 144
different DMCs to quantify the contributions of intramolecular
H-bonds to the overall stabilities of the complexes. In
complexes with poor geometric complementarity and weak
H-bond acceptors, no H-bonding was detected. However for
the most polar H-bond acceptor, phosphine oxide, in the least
polar solvent, toluene, intramolecular H-bonding was detected
for all 24 different porphyrin−ligand geometries.
Figure 15. Comparison of effective molarities (EM) for formation of
intramolecular H-bonds measured in TCE with the corresponding
values measured in toluene. Data for amide (Le) and and ester (Lf)
ligands are shown in blue and red, respectively. The line corresponds
to log EM(TCE)/M = log EM(toluene)/M.
both the amide and ester ligands. Figure 16 shows that the
values of EM measured for ester−phenol H-bonds are
practically identical to the corresponding values for amide−
phenol H-bonds if we compare ligands that have the H-bond
acceptor group located at the same point on the ligand
framework. The results for these systems therefore suggest that
EM is independent of the solvent and the polarity of the H-
bond acceptor.
We can now compare the results for these new ligand
systems with the values of EM reported previously for the
corresponding phosphonate diester and ether ligand series (La
and Ld; see Figures 3 and 5).27,28,31,32 Figure 17 compares the
values of EM for the phosphonate diester ligands (La) with the
corresponding values for the other three ligand families. We
previously noted the discrepancy between the values of EM
measured for ether and phosphonate diester ligands (Figure
17a).31 Although the values of EM fall within the same window
(1 mM to 1 M), there is no clear relationship between the
phosphonate diester EM values and the corresponding values
measured for the other ligand systems. This is particularly
1H NMR spectroscopy showed that a phosphine oxide
oxygen binds more strongly than a pyridine nitrogen to a zinc
porphyrin, so the complexes used in DMCs for this ligand
series had different structures. As a result, the DMC
underestimated the contribution of intramolecular phosphine
J
dx.doi.org/10.1021/ja406235d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 18. Comparison of effective molarities (EM) for formation of
intramolecular ether−phenol H-bonds (Ld) with the corresponding
values for (a) amide−phenol H-bonds (Le) and (b) ester−phenol H-
bonds (Lf). The line in each panel corresponds to y = x.
There is a reasonable correlation with the EM values measured
for ether ligands but little correlation with EM values for
phosphonate diester ligands. Although the values of EM are
qualitatively similar for different functional groups with
different H-bonding properties, good quantitative agreement
was obtained only for functional groups that have similar
geometric and conformational properties (e.g., amides and
esters).
In conclusion, we have now collected data on intramolecular
H-bonds between phenol and five different H-bond acceptors
in 24 different supramolecular architectures and two different
solvents. The values of EM fall in the range 1−1000 mM
depending on the geometric complementarity of the system,
but they do not depend on the solvent or the intrinsic strength
of the H-bond (i.e., the polarity of the acceptor group). We
previously noted some variations in EM with the nature of the
H-bond acceptor, but these differences are now ascribed to
differences in conformation. Functional groups with quite
different H-bonding properties and very similar geometrical
properties, such as amides (β = 8.5) and esters (β = 5.3), give
very similar values of EM. In contrast, functional groups with
similar H-bonding properties and different geometrical proper-
ties, such as amides (β = 8.5) and phosphonate diesters (β =
8.9), show little correlation in the values of EM. We previously
noted that EM varies with solvent for the phosphonate diester
ligands but not for the ether ligands. The data reported here
show that EM is independent of solvent for both the amide and
ester ligand series, so it appears that the phosphonate diester
system is anomalous, which may be related to the steric bulk of
Figure 17. Comparison of effective molarities (EM) for formation of
intramolecular phosphonate diester−phenol H-bonds (La) with the
corresponding values for (a) ether−phenol H-bonds (Ld), (b)
amide−phenol H-bonds (Le), and (c) ester−phenol H-bonds (Lf).
The line in each panel corresponds to y = x.
oxide−phenol H-bonds, and it was not possible to determine
accurate values of EM for this ligand series. For the ester and
amide ligand series, it was possible to construct reliable DMCs
and hence to determine values of EM for the intramolecular H-
bonds. The values of EM range from 3 to 240 mM depending
on the supramolecular architecture. There is excellent agree-
ment between the values measured in toluene and in TCE as
well as between the values measured for amide ligands and the
corresponding ester ligands having the oxygen acceptor located
at the same site on the ligand framework. When these results
are compared with the values of EM obtained previously for
different functional groups, the agreement is not so good.
K
dx.doi.org/10.1021/ja406235d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
spectrophotometer at 298 K. A 20 mL solution of Q at known
concentration (about 10 mM) was prepared using spectroscopic-grade
solvent. A 10 mL solution of porphyrin was prepared at known
concentration (about 0.5 μM) by dissolving the porphyrin in the Q
stock solution. A 2 mL solution of ligand was prepared at a known
concentration (about 1 μM) by dissolving the ligand in the
porphyrin−Q stock solution, to ensure that the concentration of the
porphyrin and Q remained constant throughout the titration. A 2 mL
aliquot of the porphyrin−Q stock solution was loaded into a 1 cm path
length fluorescence cuvette, and the fluorescence emission spectrum
was recorded between 500 and 750 nm with excitation at 427 nm.
Aliquots of the ligand solution were added successively to the cuvette,
and the emission spectrum was recorded after each addition. An Excel
spreadsheet was used to fit the fluorescence emission intensity at fixed
wavelengths to a 1:1 binding isotherm with a linear correction to allow
for nonspecific effects (e.g., additional weaker binding interactions at
secondary sites or additional absorbance due to the ligand at high
concentrations). This gave the apparent association constant (Kapp),
which was used to determine the association constant for formation of
the porphyrin−ligand complex as explained below. Each titration was
repeated at least three times, and the experimental error is quoted as
twice the standard deviation at a precision of one significant figure.
Titration data from the displacement experiment could be fit to a
simple 1:1 binding isotherm provided that Q was present in a large
excess relative to the porphyrin and that the total concentration of Q
was constant throughout the titration. This can be understood as
follows. In a simple titration of a porphyrin with a ligand, where Q is
not present, the concentration of the porphyrin−ligand complex, [PL],
is given by eq 7,
this group. The systems for which we have been able to collect
the most data are the phosphonate diester and amide ligands.
Although the values of EM for these two systems fall in the
same window, there is little correlation between the two data
sets. The effects of relatively subtle changes in the geometry of
the H-bond acceptor group therefore appear to be comparable
to the effects of more gross changes in supramolecular
architecture.
EXPERIMENTAL SECTION
■
1H NMR Titrations. NMR titrations were carried out by preparing
a 2 mL sample of host at a known concentration (2−20 mM), after
which 0.6 mL of this solution was removed and an 1H NMR spectrum
was recorded. A 1 mL solution of guest (25−230 mM) was prepared
using the stock host solution to ensure that the concentration of host
remained constant throughout the titration. Aliquots of guest solution
were added successively to the NMR tube containing the host, and the
NMR spectrum was recorded after each addition. Changes in chemical
shift were analyzed in Microsoft Excel using the appropriate binding
isotherms. Each titration was repeated at least three times, and the
experimental error is quoted as twice the standard deviation at a
precision of one significant figure.
Automated UV/vis Absorption Titrations. UV/vis titrations
were carried out using a BMG FLUOstar Omega plate reader
equipped with a UV/vis detector and equilibrated at 298 K. A 5 mL
solution of porphyrin at a known concentration (1−5 μM) and a 10
mL solution of ligand at a known concentration (8−40000 μM) were
prepared using spectroscopic-grade solvent. A 150 μL aliquot of the
porphyrin solution was added to a well of a Hellma quartz microplate,
and the absorbance was recorded at five wavelengths. Aliquots of the
ligand solution (3, 6, or 10 μL) were successively added to the well,
and the absorbance was recorded after each addition. Changes in
absorbance were fit to a 1:1 binding isotherm in Microsoft Excel to
obtain the association constant. Each titration was repeated at least
three times, and the experimental error is quoted as twice the standard
deviation at a precision of one significant figure.
[PL] = KL[P][L]
(7)
where [P] is the concentration of free porphyrin and [L] is the
concentration of free ligand. The total concentrations of porphyrin and
ligand at any point in the titration, [P]T and [L]T, are given by eqs 8
and 9:
[P]T = [P] + [PL]
(8)
Automated Fluorescence Titrations. Fluorescence titrations
were carried out using the BMG FLUOstar Omega plate reader
equilibrated at 298 K. A 10 mL solution of porphyrin at a known
concentration (0.1−1 μM) and a 10 mL solution of ligand at a known
concentration (5−63 μM) were prepared using spectroscopic-grade
solvent. A 150 μL aliquot of the porphyrin solution was added to each
of 12 wells of a Hellma quartz microplate. Different volumes of ligand
solution (0−150 μL) were added to each well, and solvent was added
to give a total volume of 300 μL. The excitation wavelength was set at
420 or 430 nm, and the fluorescence emission was measured at four
wavelengths (590, 600, 620, and 650 nm) for each well. Changes in
fluorescence emission were fit to a 1:1 binding isotherm in Microsoft
Excel to obtain the association constant. Each titration was repeated at
least three times, and the experimental error is quoted as twice the
standard deviation at a precision of one significant figure.
Manual Fluorescence Titrations. Fluorescence titrations were
carried out using a Hitachi F-4500 fluorescence spectrophotometer at
298 K. A 10 mL solution of porphyrin at a known concentration
(0.04−0.05 μM) was prepared in spectroscopic-grade solvent. A 2 mL
aliquot of this host solution was loaded into a 1 cm path length
fluorescence cuvette, and the fluorescence emission spectrum was
recorded from 500 to 750 nm with excitation at 427 nm. A 2 mL
solution of ligand (0.1−1 μM) was prepared using the host stock
solution to ensure that the concentration of the host remained
constant throughout the titration. Aliquots of the ligand solution were
added successively to the cuvette, and the emission spectrum was
recorded after each addition. Changes in fluorescence emission were fit
to a 1:1 binding isotherm in Microsoft Excel to obtain the association
constant. Each titration was repeated at least three times, and the
experimental error is quoted as twice the standard deviation at a
precision of one significant figure.
[L]T = [L] + [PL]
(9)
Substituting [P] from eq 8 and [L] from eq 9 into eq 7 and
rearranging gives eq 10:
[PL] = KL([P]T − [PL])([L]T − [PL])
(10)
This is the equation that is solved in fitting titration data to a 1:1
binding isotherm.
Now consider the effect of the presence of another ligand Q that
competes with L for binding to the porphyrin. Instead of eq 8, the total
porphyrin concentration at any point in the titration is now given by
eq 11:
[P]T = [P] + [PL] + [PQ]
(11)
The concentration of the porphyrin−quencher complex, [PQ], is
given by eq 12:
[PQ] = KQ [P][Q]
(12)
where [Q] is the concentration of free quencher ligand. Substituting
[PQ] from eq 12 into eq 11 and rearranging gives eq 13.
1
[P] =
([P]T − [PL])
1 + KQ [Q]
(13)
Substituting [P] from eq 13 and [L] from eq 9 into eq 7 gives eq 14:
KL
[PL] =
([P]T − [PL])([L]T − [PL])
1 + KQ [Q]
(14)
Comparison of eq 14 with eq 10 shows that the only difference
introduced by the presence of Q is that KL is replaced by KL/(1 +
KQ[Q]). Thus, the association constant obtained by fitting the
displacement titration data to a 1:1 binding isotherm, Kapp, is related to
Fluorescence Displacement Titrations. Fluorescence displace-
ment titrations were carried out using a Hitachi F-4500 fluorescence
L
dx.doi.org/10.1021/ja406235d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
KL by eq 2. The values of KQ used in eq 2 were determined by UV/vis
titrations of Q into the relevant porphyrin solutions (Table 12).
(17) Krishnamurthy, V. M.; Semetey, V.; Bracher, P. J.; Shen, N.;
Whitesides, G. M. J. Am. Chem. Soc. 2007, 129, 1312.
(18) Mulder, A.; Huskens, J.; Reinhoudt, D. N. Org. Biomol. Chem.
2004, 2, 3409.
Table 12. Association Constants for the Formation of Zinc
Porphyrin·Q Complexes (KQ) Measured by UV/vis
Absorption Titrations at 298 K in Toluene
(19) Jiang, W.; Nowosinski, K.; Low, N. L.; Dzyuba, E. V.; Klautzsch,
̈
F.; Schafer, A.; Huuskonen, J.; Rissanen, K.; Schalley, C. A. J. Am.
Chem. Soc. 2012, 134, 1860.
̈
KQ/104 M−1
(20) Mack, E. T.; Snyder, P. W.; Perez-Castillejos, R.; Bilgicer, B.;
Moustakas, D. T.; Butte, M. J.; Whitesides, G. M. J. Am. Chem. Soc.
2012, 134, 333.
porphyrin
P1a
P2a
P3a
P4a
3.3 0.1
3.1 0.1
2.3 0.6
1.6 0.2
(21) Nelson, D. J.; Ashworth, I. W.; Hillier, I. H.; Kyne, S. H.;
Pandian, S.; Parkinson, J. A.; Percy, J. M.; Rinaudo, G.; Vincent, M. A.
Chem.Eur. J. 2011, 17, 13087.
(22) Kirby, A. J. Adv. Phys. Org. Chem. 1981, 17, 183.
(23) Cacciapaglia, R.; Di Stefano, S.; Mandolini, L. Acc. Chem. Res.
2004, 37, 113.
ASSOCIATED CONTENT
* Supporting Information
(24) Illuminati, G.; Mandolini, L. Acc. Chem. Res. 1981, 14, 95.
(25) Lightstone, F. C.; Bruice, T. C. Bioorg. Chem. 1998, 26, 193.
(26) Hogben, H. J.; Sprafke, J. K.; Hoffmann, M.; Pawlicki, M.;
Anderson, H. L. J. Am. Chem. Soc. 2011, 133, 20962.
(27) Hunter, C. A.; Misuraca, M. C.; Turega, S. M. J. Am. Chem. Soc.
2011, 133, 582.
(28) Hunter, C. A.; Misuraca, M. C.; Turega, S. M. J. Am. Chem. Soc.
2011, 133, 20416.
(29) Misuraca, M. C.; Grecu, T.; Freixa, Z.; Garavini, V.; Hunter, C.
A.; van Leeuwen, P.; Segarra-Maset, M. D.; Turegat, S. M. J. Org.
Chem. 2011, 76, 2723.
■
S
Association constants for zinc porphyrin−ligand complexes,
NMR study of phosphine oxide ligand binding modes,
comparison of the phosphine oxide ligand ΔΔG° values with
those for the other ligand series, analysis of the partially bound
states for implementation of eqs 4 and 5, and experimental
details of ligand synthesis and spectroscopic characterization.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(30) Chekmeneva, E.; Hunter, C. A.; Misuraca, M. C.; Turega, S. M.
Org. Biomol. Chem. 2012, 10, 6022.
(31) Hunter, C. A.; Misuraca, M. C.; Turega, S. M. Chem. Sci. 2012,
3, 2462.
(32) Hunter, C. A.; Misuraca, M. C.; Turega, S. M. Chem. Sci. 2012,
3, 589.
(33) Adams, H.; Chekmeneva, E.; Hunter, C. A.; Misuraca, M. C.;
Navarro, C.; Turega, S. M. J. Am. Chem. Soc. 2013, 135, 1853.
(34) Chekmeneva, E.; Hunter, C. A.; Packer, M. J.; Turega, S. M. J.
Am. Chem. Soc. 2008, 130, 17718.
(35) Cockroft, S. L.; Hunter, C. A. Chem. Soc. Rev. 2007, 36, 172.
(36) Camara-Campos, A.; Musumeci, D.; Hunter, C. A.; Turega, S. J.
Am. Chem. Soc. 2009, 131, 18518.
(37) Haynes, R. K.; Au-Yeung, T. L.; Chan, W. K.; Lam, W. L.; Li, Z.
Y.; Yeung, L. L.; Chan, A. S. C.; Li, P.; Koen, M.; Mitchell, C. R.;
Vonwiller, S. C. Eur. J. Org. Chem. 2000, 3205.
AUTHOR INFORMATION
Corresponding Author
c.hunter@sheffield.ac.uk
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank the EPSRC, the China Scholarship Council, and the
University of Sheffield for funding.
■
REFERENCES
■
(1) Dam, T. K.; Roy, R.; Das, S. K.; Oscarson, S.; Brewer, C. F. J. Biol.
Chem. 2000, 275, 14223.
(2) Dam, T. K.; Brewer, C. F. Biochemistry 2008, 47, 8470.
(3) Badjic, J. D.; Nelson, A.; Cantrill, S. J.; Turnbull, W. B.; Stoddart,
J. F. Acc. Chem. Res. 2005, 38, 723.
(38) Hunter, C. A.; Sanders, J. K. M.; Beddard, G. S.; Evans, S. J.
Chem. Soc., Chem. Commun. 1989, 1765.
(39) Hunter, C. A. Angew. Chem., Int. Ed. 2004, 43, 5310.
(40) Mark, A. E.; van Gunsteren, W. F. J. Mol. Biol. 1994, 240, 167.
(4) Seel, C.; Vogtle, F. Angew. Chem., Int. Ed. Engl. 1992, 31, 528.
̈
(5) Ashton, P. R.; Chrystal, E. J. T.; Glink, P. T.; Menzer, S.; Schiavo,
C.; Spencer, N.; Stoddart, J. F.; Tasker, P. A.; White, A. J. P.; Williams,
D. J. Chem.Eur. J. 1996, 2, 709.
(6) Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.; Bruening, R. L. Chem.
Rev. 1991, 91, 1721.
(7) Prins, L. J.; Scrimin, P. Angew. Chem., Int. Ed. 2009, 48, 2288.
(8) Hunter, C. A.; Anderson, H. L. Angew. Chem., Int. Ed. 2009, 48,
7488.
(9) Ercolani, G.; Schiaffino, L. Angew. Chem., Int. Ed. 2011, 50, 1762.
(10) Fasting, C.; Schalley, C. A.; Weber, M.; Seitz, O.; Hecht, S.;
Koksch, B.; Dernedde, J.; Graf, C.; Knapp, E.-W.; Haag, R. Angew.
Chem., Int. Ed. 2012, 51, 10472.
(11) Yonetani, T.; Tsuneshige, A. C. R. Biol. 2003, 326, 523.
(12) Bellelli, A.; Brunori, M.; Miele, A. E.; Panetta, G.; Vallone, B.
Curr. Protein Pept. Sci. 2006, 7, 17.
(13) Luque, I.; Leavitt, S. A.; Freire, E. Annu. Rev. Biophys. Biomol.
Struct. 2002, 31, 235.
(14) Camara-Campos, A.; Hunter, C. A.; Tomas, S. Proc. Natl. Acad.
Sci. U.S.A. 2006, 103, 3034.
(15) Houk, K. N.; Leach, A. G.; Kim, S. P.; Zhang, X. Y. Angew.
Chem., Int. Ed. 2003, 42, 4872.
(16) Mammen, M.; Choi, S. K.; Whitesides, G. M. Angew. Chem., Int.
Ed. 1998, 37, 2755.
M
dx.doi.org/10.1021/ja406235d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX