Evidence for partially bound states in cooperative molecular recognition interfaces

Article abstract of DOI:10.1021/ja803434z

A zinc porphyrin equipped with four amide H-bonding sites provides a rigid molecular receptor for the study of cooperative multipoint binding interactions. The interaction of this receptor with a variety of pyridine ligands bearing zero, one, and two H-bonding sites has been studied using UV/vis absorption, ¹H and ³¹P NMR spectroscopy, and isothermal titration calorimetry in five different solvents. The results are analyzed in terms of a bound state that populates an ensemble of different complexes in which zero, one, or two of the potential H-bond interactions are formed. The key parameter that determines the behavior of the system is the product of the association constant for the H-bond interaction, K_H, and the effective molarity for the intramolecular interaction, EM. In the system reported here, EM is 0.1-1 M for all of the intramolecular interactions. For strong H-bonds (large K _H in nonpolar solvents), all of the interactions are formed in the complex and the fully bound state dominates. In this case, additional binding interactions produce incremental increases in complex stability. However, for weaker H-bonds (small K_H in polar solvents), the formation of additional interactions does not lead to an increase in the overall stability of the complex, due to the population of partially bound states.

Full text of DOI:10.1021/ja803434z

Published on Web 12/08/2008
Evidence for Partially Bound States in Cooperative Molecular
Recognition Interfaces
Elena Chekmeneva,† Christopher A. Hunter,*,† Martin J. Packer,‡ and
Simon M. Turega†
Department of Chemistry, UniVersity of Sheffield, Sheffield S3 7HF, and AstraZeneca,
Alderley Park, Cheshire SK10 4TG, United Kingdom
Received May 8, 2008; E-mail: C.Hunter@sheffield.ac.uk
Abstract: A zinc porphyrin equipped with four amide H-bonding sites provides a rigid molecular receptor
for the study of cooperative multipoint binding interactions. The interaction of this receptor with a variety of
pyridine ligands bearing zero, one, and two H-bonding sites has been studied using UV/vis absorption, 1H
and 31P NMR spectroscopy, and isothermal titration calorimetry in five different solvents. The results are
analyzed in terms of a bound state that populates an ensemble of different complexes in which zero, one,
or two of the potential H-bond interactions are formed. The key parameter that determines the behavior of
the system is the product of the association constant for the H-bond interaction, KH, and the effective molarity
for the intramolecular interaction, EM. In the system reported here, EM is 0.1-1 M for all of the intramolecular
interactions. For strong H-bonds (large KH in nonpolar solvents), all of the interactions are formed in the
complex and the fully bound state dominates. In this case, additional binding interactions produce incremental
increases in complex stability. However, for weaker H-bonds (small KH in polar solvents), the formation of
additional interactions does not lead to an increase in the overall stability of the complex, due to the
population of partially bound states.
Introduction
Multiple intermolecular interactions between two molecules
in a supramolecular system usually lead to cooperative stabiliza-
tion of the resulting noncovalent complex.1-3 The magnitude
of the cooperative effect can vary significantly from one system
to another, but the relationship among the noncovalent interac-
tions, the supramolecular architecture, and cooperativity is a
relatively unexplored area. The field of supramolecular chem-
istry has been directed by the preorganization principle to rigid
scaffolds and strong binding interactions that give rise to very
stable complexes and readily predictable behavior4-7 (Figure
1a). In contrast, biology works with flexible molecular frame-
works and relatively weak binding interactions, relying on
cooperativity between many weak interactions to organize
relatively complex systems and resulting in behavior that is
Figure 1. Cooperative binding interactions at a molecular recognition
interface. (a) Systems that are highly preorganized (high EM) and/or have
very favorable binding site interactions (high K) form stable well-defined
complexes, where all interactions are fully made. (b) Systems that are
flexible (low EM) and have weak binding site interactions (low K) have
the potential to populate a large number of equilibrium states and can show
dramatic cooperative responses to changes in conditions.
much more difficult to predict (Figure 1b).8-11 All or nothing
processes such as protein folding are a direct consequence of
the delicate balance between the formation of a large number
of weakly favorable intermolecular interactions and the restric-
tion of a large number of conformational degrees of freedom.
However, the complexity of biomolecules makes any attempt
to dissect the molecular details of these processes a real
challenge. In this paper, we outline an approach to the
characterization of structure-cooperativity relationships in a
synthetic supramolecular system.
† University of Sheffield.
‡ AstraZeneca.
(1) (a) Mulder, A.; Huskens, J.; Reinhoudt, D. N. Org. Biomol. Chem.
2004, 2, 3409–3424. (b) Philp, D.; Stoddart, J. F. Angew. Chem., Int.
Ed. 1996, 35, 1155–1196. (c) Lehn, J. M. Angew. Chem., Int. Ed.
1990, 29, 1304–1319.
(2) (a) Tobey, S. L.; Anslyn, E. V. J. Am. Chem. Soc. 2003, 125, 10963–
10970. (b) Hughes, A. D.; Anslyn, E. V. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 6538–6543.
(3) Jadhav, V. D.; Schmidtchen, F. P. Org. Lett. 2006, 8, 2329–2332.
(4) Mammen, M.; Choi, S. K.; Whitesides, G. M. Angew. Chem., Int. Ed.
1998, 37, 2755–2794.
(5) Hilser, V. J.; Garcia-Moreno, B.; Oas, T. G.; Kapp, G.; Whitten, S. T.
Chem. ReV. 2006, 106, 1545–1558.
(8) Cockroft, S. L.; Hunter, C. A. Chem. Soc. ReV. 2007, 36, 172–188.
(9) Hunter, C. A.; Tomas, S. Chem. Biol. 2003, 10, 1023–1032.
(10) Shiozawa, H.; Chia, B. C. S.; Davies, N. L.; Zerella, R.; Williams,
D. H. J. Am. Chem. Soc. 2002, 124, 3914–3919.
(6) Williams, D. H.; Bardsley, B. Angew. Chem., Int. Ed. 1999, 38, 1173–
1193.
(7) Munoz, V.; Ghirlando, R.; Blanco, F. J.; Jas, G. S.; Hofrichter, J.;
Eaton, W. A. Biochemistry 2006, 45, 7023–7035.
(11) Williams, D. H.; Stephens, E.; O’Brien, D. P.; Zhou, M. Angew. Chem.,
Int. Ed. 2004, 43, 6596–6616.
9
17718 J. AM. CHEM. SOC. 2008, 130, 17718–17725
10.1021/ja803434z CCC: $40.75
2008 American Chemical Society

Cooperative Molecular Recognition Interfaces
A R T I C L E S
Figure 2. Supramolecular recognition interface with one metal-ligand
interaction and two peripheral H-bonds.
The key parameter that quantifies the magnitude of cooper-
ativity in a supramolecular complex is the effective molarity
for the intramolecular interaction (EM). The magnitude of the
EM depends on the flexibility of the linker between two binding
sites and the degree of conformational strain introduced on
cyclization. This relationship has been thoroughly characterized
for the formation of intramolecular covalent bonds using both
kinetic and thermodynamic measurements of EM.12-14 We
might assume that noncovalent interactions will show similar
behavior; i.e., the EM will decrease with increasing conforma-
tional flexibility and with increasing conformational strain.15,16
However, noncovalent interactions are much looser and con-
formationally less demanding than covalent bonds, which could
lead to important differences in behavior.
Figure 3. Synthesis of porphyrin 3.
The system illustrated in Figure 2 was designed to quantify
cooperativity at a relatively complicated but tractable molecular
recognition interface. The porphyrin receptor provides a rigid
scaffold on which a set of binding sites can be displayed: a
metal-ligand binding site and four H-bonding sites.17,18 Zinc
porphyrins are generally five-coordinate in solution and so have
a single coordination site available for complexation of one
pyridine ligand. The zinc porphyrin-pyridine coordination bond
provides a strong noncovalent interaction site that can be
used to guarantee the formation of a complex with a well-defined
topology. Formation of this interaction can be easily monitored
using 1H NMR, UV/vis absorption, or fluorescence spectroscopy
across a wide range of concentrations and solvents. The weaker
H-bonding sites provide up to four additional interactions. In
this study, we have varied the number of H-bonds and tuned
the strength of the individual binding interactions by changing
the polarity of the solvent.19,20 The degree of flexibility in the
linker that connects the coordination site and H-bond sites is
another key parameter, and the conformational flexibility of the
linkers used in this study leads to rather more complicated
Figure 4. Synthesis of ligands 4a-d.
complexation equilibria than usually observed in conventional
supramolecular assemblies.
Results
The tetraamidoporphyrin used in this study was synthesized
using 5,10,15,20-tetrakis(3-aminophenyl)-21H,23H-porphyrin
(1) as a starting material (Figure 3).21 Porphyrin 1 was coupled
with isononyl chloride to provide porphyrin 2 in 50% yield.
Porphyrin 2 was then metalated with zinc acetate in 50% yield
to give porphyrin 3. A set of complementary pyridine ligands
containing peripheral H-bond acceptors were synthesized from
the corresponding pyridine carboxylic acids (Figure 4). The
pyridine carboxylic acids were first converted to the acid
chloride and then coupled with the appropriate alcohol to
generate ligands 4a-d. Two additional control compounds,
amide 5 and phosphatodiester 6, were used to study the
properties of the amide-phosphatodiester H-bond. Compound
6 is commercially available, and 5 was prepared as shown in
Figure 5.
(12) Bruice, T. C.; Lightstone, F. C. Acc. Chem. Res. 1999, 32, 127–136.
(13) Cacciapaglia, R.; Di Stefano, S.; Mandolini, L. Acc. Chem. Res. 2004,
37, 113–122.
(14) Kirby, A. AdV. Phys. Org. Chem. 1980, 17, 183.
(15) Camara-Campos, A.; Hunter, C. A.; Tomas, S. Proc. Natl. Acad. Sci.
U.S.A. 2006, 103, 3034–3038.
Binding of ligands 4a-d to porphyrin 3 was investigated in
five solvents, toluene, 1,1,2,2-tetrachloroethane (TCE), dichlo-
romethane (DCM), chloroform, and acetone, using UV/vis
absorption spectroscopy and where possible 1H NMR spectros-
copy, 31P NMR spectroscopy, and isothermal titration calorim-
(16) Chi, X. L.; Guerin, A. J.; Haycock, R. A.; Hunter, C. A.; Sarson,
L. D. Chem. Commun. 1995, 2563–2565.
(17) Aoyama, Y.; Asakawa, M.; Yamagishi, A.; Toi, H.; Ogoshi, H. J. Am.
Chem. Soc. 1990, 112, 3145–3151.
(18) Aoyama, Y.; Asakawa, M.; Matsui, Y.; Ogoshi, H. J. Am. Chem. Soc.
1991, 113, 6233–6240.
(19) Cook, J. L.; Hunter, C. A.; Low, C. M. R.; Perez-Velasco, A.; Vinter,
J. G. Angew. Chem., Int. Ed. 2007, 46, 3706–3709.
(20) Hunter, C. A. Angew. Chem., Int. Ed. 2004, 43, 5310–5324.
(21) Semeikin, A. S.; Koifman, O. I.; Berezin, B. D. IzV. Vyssh. Uchebn.
ZaVed., Khim. Khim. Tekhnol. 1985, 28, 47–51.
9
J. AM. CHEM. SOC. VOL. 130, NO. 52, 2008 17719

A R T I C L E S
Chekmeneva et al.
Table 2. Association Constants (M-1) for the Formation of the 5·6
Complex at 298 K
a
solvent
Rs
ꢀs
Keq 1
Kexptl
toluene
TCE
DCM
CHCl3
acetone
1.0
1.9
1.9
2.2
1.5
2.2
1.1
1.1
0.8
5.8
16
2.1
2.1
0.87
0.51
24
2.9
1.7
0.46
1.3
a Measured using 1H NMR or 31P NMR spectroscopy. The average
percentage error over the data set is (40%.
Figure 5. Synthesis of control compound 5. Control compound 6 is
commercially available.
Table 1. Association Constants (M-1) for the Formation of 1:1
theoretical estimates based on eq 1 using R ) 2.9 for the amide
and ꢀ ) 8.9 for the phosphatodiester. The agreement between
the two sets of values gives us some confidence that the values
are reasonable, despite our inability to access the full binding
isotherm.
Complexes at 298 Ka
solvent
3·4a
3 ·4b
3·4c
3 ·4d
method
toluene
TCE
TCE
1900
1600
940
2000
1000
790
28000
6000
2000
11000
2300
2400
1400
1400
850
400
310
1200
800
970
120
120
240000
2500
1700
10000
3200
3200
360
UV/vis
UV/vis
ITCb
UV/vis
UV/vis
ITCb
The possibility of oxygen metal coordination makes it
necessary to consider an alternative mode of cooperative binding
between porphyrin 3 and ligands 4b and 4d. This alternative
mode of binding would involve the oxygen of the phosphato-
diester coordinating to the zinc of 3 and the pyridine nitrogen
forming a peripheral H-bond with the amide of the porphyrin.
Phosphine oxide coordination to zinc porphyrins has been
observed, but with association constants of 24-32 M-1 in
chloroform,25 which is more than an order of magnitude lower
than the corresponding pyridine-zinc porphyrin association
constant. Nevertheless, it is possible to directly test the
magnitude of the oxygen-zinc interaction in the system
described here by titrating porphyrin 3 with control compound
6 using UV/vis absorption spectroscopy to monitor binding. The
data fit well to a 1:1 binding isotherm with an association
constant of 94 ( 10 M-1 in DCM, which is an order of
magnitude lower than the pyridine-zinc association constant
in this solvent (2000 M-1). We can therefore rule out any
significant oxygen-zinc interaction, and the coordination bonds
and the H-bonds are effectively orthogonal in this system.
DCM
CHCl3
CHCl3
acetone
acetone
250
240
UV/vis
180
1H NMR
a The average percentage error over the data set is (30%. b The 1H
NMR spectra are broad in CDCl3 and TCE, which is indicative of
aggregation, so these data are considered less reliable than the UV/vis
results.
etry. UV/vis absorption spectroscopy dilution experiments were
first used to investigate self-association of porphyrin receptor
3.22-24 In the micromolar concentration range in all five
solvents, the absorbance is linear with concentration, and there
is no change in the wavelength of the Soret band. Thus, at these
concentrations, there are no complications associated with self-
interaction of the receptor. Receptor 3 is soluble at the millimolar
concentrations required for NMR and isothermal titration
calorimetry (ITC) experiments in TCE, chloroform, and acetone.
1
However, the H NMR spectrum is very broad in chloroform,
indicating significant aggregation in this solvent at these
concentrations (see the Supporting Information). Binding of
ligands 4a-d to porphyrin 3 was investigated using titrations
in five solvents, toluene, TCE, DCM, chloroform, and acetone,
Spectroscopic evidence for the formation of H-bonds in the
complexes can be obtained from limiting complexation-induced
1
changes in chemical shift in the H NMR titrations. At the
1
1
using UV/vis absorption spectroscopy and where possible H
concentrations required for NMR experiments (mM), the H
NMR spectroscopy, 31P NMR spectroscopy, and ITC. The
association constants were determined by fitting the titration
data to a 1:1 binding isotherm, and there were no indications
of the formation of any higher stoichiometry complexes (Table
1).
NMR spectrum of 3 is very broad in chloroform, which is
1
indicative of aggregation. However, the H NMR spectrum of
3 in acetone is much sharper, suggesting that the H-bonding
solvent is able to break up amide-amide H-bonding interactions
that lead to aggregation in chloroform. The limiting complex-
ation-induced changes in chemical shift obtained from titration
experiments in acetone show some interesting behavior. The
The association constants for the H-bonded 5·6 control
1
complex were determined using H and 31P NMR titrations
1
(Table 2), but the values are close to the limit of the range
accessible in NMR titration experiments, because saturation
cannot be achieved. The magnitude of association constants for
simple H-bonded complexes of this type can be estimated using
eq 1 as described in ref 20, where R and ꢀ are the H-bond donor
and acceptor parameters of the solutes, and Rs and ꢀs are the
corresponding parameters for the solvent.
amide region of the H NMR spectra of mixtures of 3 and
ligands 4a-d is shown in Figure 7. Addition of 4a or 4c to 3
does not affect the amide signal, because these ligands do not
form H-bonds. In contrast, large changes are observed on
addition of 4b or 4d. There are large downfield shifts that are
characteristic of the formation of H-bonding interactions with
the phosphatodiester groups of the ligands. In addition, the
singlet observed for 3 splits into several broad nonequivalent
signals in the presence of the H-bonding ligands. Rotation
around the porphyrin meso-phenyl group bonds is slow on the
1H NMR time scale at room temperature, and receptor 3
therefore exists as a mixture of four different atropisomers
(Figure 6). In the absence of a H-bonding ligand, the chemical
∆G ) -(R - Rs)(ꢀ - ꢀs) + 6 kJ mol-1
(1)
Table 2 compares the experimental estimates of the 5·6
association constant obtained from the NMR titrations with the
(22) Kobuke, Y.; Miyaji, H. J. Am. Chem. Soc. 1994, 116, 4111–4112.
(23) Maeda, C.; Yamaguchi, S.; Ikeda, C.; Shinokubo, H.; Osuka, A. Org.
Lett. 2008, 10, 549–552.
(24) Stibrany, R. T.; Vasudevan, J.; Knapp, S.; Potenza, J. A.; Emge, T.;
Schugar, H. J. J. Am. Chem. Soc. 1996, 118, 3980–3981.
(25) Officer, D. L.; Lodato, F.; Jolley, K. W. Inorg. Chem. 2007, 46, 4781–
4783.
9
17720 J. AM. CHEM. SOC. VOL. 130, NO. 52, 2008

Cooperative Molecular Recognition Interfaces
A R T I C L E S
Figure 6. Atropoisomers of 3. For a statistical equilibrium distribution, the relative populations are expected to be RRRR, RRꢀꢀ, RꢀRꢀ, and RRRꢀ in a ratio
of 1:2:1:4, which gives the relative intensities for the six amide signals A-F shown above (cf. Figure 7).
1
Table 3. Average Limiting Complexation-Induced Changes in H
NMR Chemical Shift (ppm) for the Signal due to the 3 Amide
Proton on Formation of 1:1 Complexes in Acetone-d6 at 298 K
3·4a
3·4b
3·4c
3·4d
5 ·6
∆∂
-0.03
0.27
-0.02
0.22
1.11
Table 4. Thermodynamic Data (kJ mol-1) Obtained from ITC of
Porphyrin 3 and Ligands 4a-4d in TCE at 298 Ka
ligand
∆G
∆H
(-)T∆S
∆∆H
4a
4b
4c
4d
-17.0
-18.8
-14.2
-18.4
-20.6
-25.6
-19.5
-30.8
3.6
6.8
5.3
-5.0
12.4
-11.3
a Average errors (kJ mol-1) are (1 in ∆G, (1 in ∆H, and (2 in
T∆S.
complexation-induced changes in chemical shift observed for
complexes of 3·4b and 3·4d are about a quarter of this value,
which implies that on average in both porphyrin ligand
complexes only one of the four amide groups makes a H-bond
with the ligand.
Figure 7. Amide region of the 400 MHz 1H NMR spectra of porphyrin 3
in acetone-d6: (a) 3·4d complex, (b) 3·4c complex, (c) 3·4b complex, (d)
3·4a complex, (e) unbound 3. In all cases, the porphyrin is greater than
60% bound. There is a minor porphyrin impurity at 9.52 ppm (*) that is
unaffected by the presence of ligands.
Calorimetry was used to obtain further evidence for the extent
of H-bonding in the complexes. ITC experiments were carried
out for the complexes of porphyrin 3 and ligands 4a-d in TCE.
The solubility of 3 is too low to carry out the corresponding
experiments in toluene, there is significant aggregation in
chloroform, and the volatility of DCM and acetone makes
calorimetry in these solvents unreliable. The thermodynamic
parameters obtained from the ITC experiments are shown in
Table 4. The magnitude of the observed enthalpy changes
provides evidence for the extent to which potential binding
interactions are realized in the complexes. Specifically, com-
parison of the values of ∆H for the control ligands 4a and 4c
with the corresponding functional ligands 4b and 4d provides
a measure of the extent of H-bond formation with the phos-
phatodiester ligands (eq 2). The results suggest that H-bonds
are formed in all of the complexes that contain phosphatodiester
groups, and the very large enthalpy change observed for the
3·4d complex in TCE indicates the formation of two H-bonds
in this complex.
shifts of all of the amide protons in all of these isomers are
similar, so a simple singlet is observed. The H-bond interactions
with the phosphatodiester groups change this, so that all of the
nonequivalent signals in each atropisomer are resolved in the
1H NMR spectrum of the 3·4d complex (Figure 7).
The magnitude of the average limiting complexation-induced
change in chemical shift of the amide protons provides a
measure of the extent of H-bonding in the porphyrin-ligand
complexes (Table 3).26,27 The 1.1 ppm increase in chemical shift
observed for the 5·6 control complex corresponds to the
formation of one phosphatodiester-amide H-bond. The limiting
(26) Hunter, C. A.; Packer, M. J. Chem.sEur. J. 1999, 5, 1891–1897.
(27) Packer, M. J.; Zonta, C.; Hunter, C. A. J. Magn. Reson. 2003, 162,
102–112.
9
J. AM. CHEM. SOC. VOL. 130, NO. 52, 2008 17721

A R T I C L E S
Chekmeneva et al.
∆∆HH-bond ) ∆H4b⁄d - ∆H4a⁄c
(2)
Discussion
The spectroscopic data indicate that, in all cases, the pyridine
ligands coordinate to the zinc center of the porphyrin, and the
ligands equipped with phosphatodiester groups form H-bonds
with the amide groups around the porphyrin periphery. However,
the thermodynamic properties of these interactions vary sig-
nificantly with solvent. The ITC experiments suggest that the
3·4d complex forms two H-bonds in TCE, but the NMR
chemical shift data in acetone suggest that on average only one
of these H-bonds is formed. Thus, both the number and the
strength of the binding interactions vary with the solvent. To
interpret the experimental data, it will therefore be important
to consider the properties of partially bound states; i.e., the
bound complex in these systems must be treated as an ensemble
of complexes, where only subsets of all potential binding
interactions that are available are simultaneously populated.
These species are illustrated in Figure 8. The overall association
constant that is observed is given by the Boltzmann-weighted
average of all partially bound states present in the complex.
We ignore the possibility of complexes in which the metal-
ligand coordination bond is broken, as this interaction is an order
of magnitude more stable than the H-bonds. In addition, any
degeneracy associated with complexation to one of two faces
of the porphyrin is included in K0 and is identical for all of the
complexes.
Figure 8. Sequential binding scheme for the formation of singly and doubly
H-bonded complexes.
Table 5. Effective Molarities, EMs (M), and Sequential Equilibrium
Constants for H-Bond Interactions, K1 and K2 (M-1), in the
Complexes Formed between Porphyrin 3 and Ligands 4b and 4d
at 298 Ka
3·4b
3·4d
solvent
K1
EM1
K2
EM2
toluene
TCE
DCM
CHCl3
acetone
16
4
6
3
6
0.2
0.4
0.8
1
10
0.5
0.5
0.2
1
We consider a stepwise binding model28-30 with formation
of a metal-ligand interaction, K0, followed by the formation
of one H-bond, K1, and then the second, K2. The association
constant for the 3·4b complex is given by
a Errors: EM1, (60%; K1, (40%; EM2, (80%; K2, (60%.
5·6, K0 is the association constant for the control complex 3·4a,
and the statistical factor reflects the fact that there are four
degenerate amide binding sites where H-bonds can be formed
on the porphyrin.
2
1 + K1
Kobsd ) K
(3)
0 1 + K1
However, the observed association constant is perturbed by
population of the non-H-bonded state and so is not a true
reflection of the stability of the complex in which both the
H-bond and the coordination interaction are made simulta-
neously. If we take the partially bound state into account using
eq 3, then a more accurate value can be obtained for the EM in
this system:
where the control complex 3·4a is used to estimate K0.
The pyridine ligand in 4d has an additional substituent, so
the value of K0 is different for the 3·4d complex. Thus, using
the control complex 3·4c to estimate K0, the association constant
for the 3·4d complex is given by
2
2
1 + K12 + K1 K2
Kobsd ) K0
(4)
1 + K1 + K1K2
K1
Kobsd 1 + 1 + 4(K /Kobsd)(1 - K0/Kobsd
)
√
0
EM1 )
)
(
)
The influence of solvent on the individual binding interactions
can be estimated from the properties of the control complexes
3·4a, 3·4c, and 5·6 that feature single-point interactions (K0
and KH). These data can in turn be used to estimate effective
molarities (EMs) for the intramolecular binding interactions in
the complexes that contain cooperative multipoint contacts
(Figure 8). However, if the bound state is an ensemble of
different complexes, analysis of the EM values is not straight-
forward. For example, the value of EM for the intramolecular
H-bond in the 3·4b complex would conventionally be given
by
4KH 4K0KH
2
(6)
where Kobsd is the observed association constant for the 3·4b
complex, KH is the association constant for the control complex
5·6, and K0 is the association constant for the control complex
3·4a.
The results obtained for K1 and EM1 for the 3·4b complex
are presented in Table 5. The EM values are in the range 0.1-1
M, and in all cases, K1 > 1, so the H-bonded complex is the
dominant species in the bound state. The population of
the H-bonded complex is highest in toluene (95%), where the
H-bonding interactions are strongest. The maximum population
of the partially bound state where the H-bond is not made is
about 30% in chloroform and TCE. It is important to note that
the population of the partially bound state depends on both the
strength of the binding interaction, KH, and the effective
molarity, EM. In chloroform, there is a significant population
of partially bound state due to solvent competition that weakens
the H-bond interaction, but in TCE, a low EM leads to a similar
population of partially bound state despite a stronger H-bond.
Kobsd
EM1 )
(5)
4K0KH
where Kobsd is the observed association constant for the 3·4b
complex, KH is the association constant for the control complex
(28) Heitmann, L. M.; Taylor, A. B.; Hart, P. J.; Urbach, A. R. J. Am.
Chem. Soc. 2006, 128, 12574–12581.
(29) Rodriguez-Docampo, Z.; Pascu, S. I.; Kubik, S.; Otto, S. J. Am. Chem.
Soc. 2006, 128, 11206–11210.
(30) Jencks, W. P. Proc. Natl. Acad. Sci. U.S.A. 1981, 78, 4046–4050.
9
17722 J. AM. CHEM. SOC. VOL. 130, NO. 52, 2008

Cooperative Molecular Recognition Interfaces
A R T I C L E S
Table 6. Estimated Overall Association Constants for Complexes
Formed between Porphyrin 3 and Ligand 4d That Make Either
One or Two H-Bonds (M-1) and Populations (%) of Partially Bound
States Assuming EM1 ) EM2 ) 0.5 M
singly H-bonded complex
doubly H-bonded complex
zero one
zero one
H-bonds H-bond
two
H-bonds H-bond H-bonds
solvent
Kest
Kest
toluene 8.2 × 104
1
8
13
35
16
99
92
87
65
84
6.8 × 105
3.6 × 103
6.4 × 103
1.1 × 103
4.9 × 102
0
4
8
32
11
10
46
56
58
59
90
50
35
10
29
TCE
3.3 × 103
7.1 × 103
Figure 9. Two trans and four cis modes of binding in the doubly H-bonded
3·4d complex.
DCM
CHCl3 1.2 × 103
acetone 5.6 × 102
If we assume that the EM1 values measured for the 3·4b
complex can be used to estimate the stability of the singly
H-bonded state in the 3·4d complex, then the partially bound
state analysis can be extended to the 3·4d complex. The
equilibrium constant between the bound state in which no
H-bonds are made and the bound state in which one H-bond is
made in the 3·4d complex is given by
bonded complexes. If we assume a fixed value of EM in all
systems (0.5 M), we can estimate the expected values of the
observed association constants for formation of complexes in
which both or only one H-bond can be made using eqs 3 and
4. Table 6 shows the results along with the relevant populations
of partially bound states. For toluene and TCE, the doubly
H-bonded complex is more stable than the singly H-bonded
complex, but the opposite is true in the other three solvents. In
other words, in DCM, chloroform, and acetone, there is nothing
to be gained by forming the second H-bond, and the population
of the doubly H-bonded state is correspondingly low in these
solvents. The H-bonding interactions in toluene and TCE are
stronger than in the other more polar solvents, so it is only in
these two solvents that the gain in interaction energy is sufficient
to offset the entropic organization energy required to generate
the fully bound, doubly H-bonded complex. The structural
evidence that supports this analysis is the ITC results that
indicate the formation of two H-bonds in TCE and the NMR
data that indicate the formation of only one of the two H-bonds
in acetone.
K1 ) 8(EM1)KH
(7)
because ligand 4d has twice as many H-bonding sites as ligand
4b.
Thus, combining eqs 4, 6, and 7 with the value of EM1
determined from the 3·4b complex, it is possible to determine
EM2 for the formation of the second H-bond in the 3·4d
complex:
8K2 8Kobsd
EM2 )
)
×
6KH 6K0KH
2
1 + 1 + 4(K /Kobsd){1 + K1 - (K0/Kobsd)(1 + K1 )}
√
0
(8)
(
)
2
where Kobsd is the observed association constant for the 3·4d
complex, KH is the association constant for the control complex
5·6, K0 is the association constant for the control complex 3·4c,
and the statistical factor reflects the degeneracies of the free
(8) and bound (6) states.
Conclusions
In complexes that make multiple intermolecular interactions,
partially bound states, where not all of the individual binding
sites are simultaneously populated, can make a significant
contribution to the behavior of the system.31-33 There is a
balance between the favorable free energy change associated
with formation of an individual interaction and the free energy
penalty associated with formation of a more organized complex.
The organization energy is quantified by the effective molarity
(EM) for the intramolecular binding interaction, and the free
energy contribution associated with formation of an individual
binding interaction can be estimated from the association
constants of model reference complexes where only one binding
interaction is made (K). Thus, the product K(EM) is the key
parameter that dictates the extent to which cooperative multi-
point binding interactions are made in a particular system. We
assume that the free energy contributions of individual binding
interactions are approximately constant, additive, and indepen-
dent of their context, provided that the interaction is fully made
and there are no significant secondary interactions.34
There are eight degenerate states for the singly H-bonded
complex, because there are four amide H-bond donors on the
porphyrin and two phosphatodiester H-bond acceptors on the
ligand. Assumptions about the structure of the complex are
required to estimate the number of degenerate states in
the doubly H-bonded complex. If we assume that the two
H-bonds can be formed using any pair of meso-amide substit-
uents on the porphyrin, the statistical factor is 6 (Figure 9).
However, if there were a strong preference for interaction with
meso substituents that are trans across the face of the porphyrin,
the statistical factor would be 2 (Figure 9). Similarly, if there
were a strong preference for interaction with meso substituents
that are cis across the face of the porphyrin, the statistical factor
would be 4. This would affect the numerical results but does
not significantly affect the substance of the analysis presented
below. It is also important to reiterate that the control complex
used to estimate K0 in this system is different from that used
for the 3·4b complex, because the ligands have different
substituents.
The results obtained for K2 and EM2 for the 3·4d complex
are presented in Table 5. The EM values are similar to those
found for the 3·4b complexes, between 0.1 and 1 M. However,
eq 8 only has a real solution for the results obtained in toluene
and TCE. This implies that two H-bonds are not formed to any
significant extent in the other solvents and that the 3·4d
complexes formed in DCM, chloroform, and acetone exist as a
mixture of the non-H-bonded and predominantly singly H-
In the experiments presented here, we have used a porphyrin-
ligand system that can form up to three noncovalent intermo-
lecular binding interactions: a metal-ligand coordination in-
(31) Williams, D. H.; Davies, N. L.; Koivisto, J. J. J. Am. Chem. Soc. 2004,
126, 14267–14272.
(32) Calderone, C. T.; Williams, D. H. J. Am. Chem. Soc. 2001, 123, 6262–
6267.
(33) Williams, D. H.; Davies, N. L.; Zerella, R.; Bardsley, B. J. Am. Chem.
Soc. 2004, 126, 2042–2049.
9
J. AM. CHEM. SOC. VOL. 130, NO. 52, 2008 17723

A R T I C L E S
Chekmeneva et al.
teraction that is so favorable that it can be assumed to be fully
to the bound state in which no H-bonds are made, and this
provides an additional entropic stabilization of the H-bond in
the partially bound state.
35,36
bound in all situations (K ) 100-2000 M-1
)
and two
weaker H-bonding interactions (K ) 0.5-25 M-1). The H-
bonding sites are connected via moderately flexible linkers
giving effective molarities in the 0.1-1 M range; i.e., there is
a small unfavorable organization energy associated with the
formation of an intramolecular H-bond that reduces the effective
molarity from 10 M, which is the value expected on the basis
of estimates of the cost of restricting intermolecular motion.20
Although there is some variation in the measured effective
molarities reported here, the results are rather similar for all of
the complexes and do not differ significantly, given the
experimental error.
The system described here represents a new tool for probing
the properties of complicated recognition interfaces and delin-
eating the rules that govern cooperativity in molecular interac-
tions. The expression of individual binding site interactions in
the overall stability of a complex is determined by the
relationship between the binding site interaction strength, K,
and the effective molarity, EM, and the relationship between
these parameters, the chemical structure, and the supramolecular
architecture should be a fruitful area for further investigation.
Systems that are poised at the balance point of K(EM) ≈ 1 are
a hallmark of biological organization and recognition, and not
only will synthetic counterparts prove useful in understanding
the molecular basis for the properties of such systems, they may
also exhibit functionally related behavior.
The free energy contribution associated with formation of
intramolecular H-bonds was varied by tuning the polarity of
the solvent. In toluene, there is very little solvent competition,
so there is a large driving force for the formation of a H-bond
(K ) 24 M-1). In this solvent, K(EM) ≈ 10, so that the fully
bound complex is formed in which all three binding sites are
fully populated (>90% occupancy of bound states). In this case,
a simple increase in stability is observed for each interaction
that is added to the system: when the ligand cannot make any
H-bonds, K ≈ 103 M-1, for a ligand that can make one H-bond,
K ≈ 104 M-1, and for a ligand that can make two H-bonds, K
Experimental Section
Synthesis. 5,10,15,20-Tetrakis(3,5,5-trimethyl-N-phen-3-ylhex-
anamide)-21H,23H-porphine, 2. To a mixture of 1 (0.325 g, 0.482
mmol) and triethylamine (1.63 mL, 11.6 mmol) in THF (20 mL)
under nitrogen at room temperature was added isononanoyl chloride
(0.366 mL, 1.93 mmol), and the solution was stirred for 18 h. The
solution was concentrated under reduced pressure and the resulting
blue gum dissolved in DCM (250 mL), washed with 10%
NaHCO3(aq) (2 × 200 mL) and brine (200 mL), and condensed
under reduced pressure. The crude product was purified on silica
(30 g) eluting with DCM/5% methanol. The product 2 was isolated
≈ 105 M-1
.
The behavior in the more polar solvents is more complicated.
Here K(EM) ≈ 1, so that the system is delicately balanced and
the probability of formation of an intramolecular H-bond is 50:
50. In these solvents, complexes that could potentially form zero,
one, or two H-bonds all have rather similar overall stability
constants, and a detailed dissection of all of the small contribu-
tions that affect the association constants is required to interpret
the results. The individual H-bonding interactions are most
favorable in TCE (K ) 3 M-1), and in this solvent, there is
clear evidence from ITC experiments for the formation of two
intramolecular H-bonds in the bound state. In other words, the
K(EM) balance is tipped in favor of the fully bound, doubly
H-bonded state by K in this case. However, detailed analysis
of the thermodynamic properties of this complex suggests a
significant population of partially bound states in which only
one H-bond is made (40%), and no increase in the overall
stability of the complex is observed. For this system, although
the spectroscopic data clearly show that H-bonds are made, this
does not lead to any significant increase in the association
constant as the number of H-bonding groups increases. In the
other three solvents, DCM, chloroform, and acetone, K(EM) <
1, due to increased solvent competition with the H-bonding
interactions, and there is no evidence for significant population
1
as a purple solid: yield 0.297 g (50%); mp 250 °C dec; H NMR
(250 MHz, CD3SOCD3) δH ) 10.25 (1H, s), 8.89 (2H, s), 8.50
(1H, d, J ) 10 Hz), 8.08 (1H, s), 7.92 (1H, d, J ) 8 Hz), 7.74 (1H,
dd, J ) 8 Hz, J ) 8 Hz), 2.39-2.2 (2H, m), 2.19 (1H, s), 1.36-1.06
(2H, m), 0.98 (3H, d, J ) 5 Hz), 0.88 (9H, s), -2.96 (0.5H, s);
MS (EI+) m/z (rel intens) ) 675 (100), 1236 (10) [M + H+];
HRMS (EI+) m/z calcd for C80H99N8O4 1235.7789, found 1235.7825;
UV/vis (DMSO) λmax/nm (ε/mol-1 cm-1) 423 (5.1 × 105), 516 (1.8
× 104), 551 (7.9 × 103), 590 (5.3 × 103), 645 (4.2 × 103).
5,10,15,20-Tetrakis(3,5,5-trimethyl-N-phen-3-ylhexanamide)por-
phine zinc(II), 3. A mixture of 2 (0.300 g, 0.243 mmol) and zinc
acetate (0.890 g, 4.85 mmol) in chloroform (20 mL)/MeOH (30
mL) was stirred under nitrogen at room temperature for 1.5 h. The
solution was concentrated under reduced pressure, and the resulting
blue gum was purified on deactivated alumina (80 g) eluting with
DCM. The product 3 was isolated as a purple solid: yield 0.159 g
(50%); mp 195 °C dec; 1H NMR (250 MHz, CDCl3, CD3OD 2%)
δH ) 8.89 (2H, s), 8.76 (1H, s), 8.19-8.14 (1H, m), 8.00-7.86
(2H, m), 7.57 (1H, s), 2.38-2.28 (2H, m), 2.14-2.04 (1H, m),
1.28-1.02 (2H, m), 0.96 (3H, d, J ) 5 Hz), 0.83 (9H, s); MS (EI+)
m/z (rel intens) ) 1299 (100) [M+]; HRMS (EI+) m/z calcd for
C80H96N8O464Zn 1296.6846, found 1296.6835; UV/vis (DMSO)
1
of the doubly H-bonded complex. However, H NMR spec-
λ
max/nm (ε/mol-1 cm-1) 421 (4.6 × 105), 549 (1.6 × 104), 595 (4.7
troscopy shows that on average one of the porphyrin amides
forms a H-bond with the ligand in acetone, and detailed analysis
of the thermodynamic properties of these complexes indicates
that the major species present in the bound state is the singly
H-bonded complex (60%). This complex in which only one of
the two H-bonds is formed enjoys an 8-fold degeneracy relative
× 103).
(Diethoxyphosphoryl)methyl Nicotinate, 4b. A mixture of nico-
tinic acid (1 g, 8.12 mmol), toluene (10 mL), DMF (10 µL), and
SOCl2 (15 mL) was refluxed for 1 h under protection by a DryRite
drying tube. The solution was condensed under reduced pressure,
and DCM (120 mL) was added. Then diethyl hydroxymethyl
phosphonate (1.20 mL, 8.12 mmol) in DCM (20 mL) and
triethylamine (3.39 mL, 24.4 mmol) were added in small portions.
The solution was allowed to stir for 18 h, then washed with
NaHCO3(aq) (2 × 100 mL) and brine (100 mL), dried with NaSO4,
and condensed under reduced pressure. The crude product was
purified on silica (25 g) eluting with EtOAc containing 0.25%
triethylamine. The product was isolated as a clear oil: yield 2.71 g
(34) We assume that there is no cooperativity (positive or negative)
associated with changes in the intrinsic affinity of the metal-ligand
or H-bond interactions (K0 and KH) on formation of multiple
intermolecular contacts in these systems: Ercolani, G. J. Am. Chem.
Soc. 2003, 125, 16097–16103.
(35) Dattagupta, N.; Malakar, D.; Ramcharan, R. G. J. Inorg. Nucl. Chem.
1981, 43, 2079–2080.
1
(98%); H NMR (250 MHz, CDCl3) δH 9.19 (d, 1H, J ) 2 Hz),
(36) Zielenkiewicz, W.; Lebedeva, N. S.; Antina, E. V.; Vyugin, A. I.;
Kaminski, M. J. Solution Chem. 1998, 27, 879–886.
8.75 (dd, 1H, J ) 2, 5 Hz), 8.26 (dt, 1H, J ) 2, 8 Hz), 7.35 (m,
9
17724 J. AM. CHEM. SOC. VOL. 130, NO. 52, 2008

Cooperative Molecular Recognition Interfaces
A R T I C L E S
Table 7. λmax of Porphyrin 3 in the Solvents Used in This Study
1H), 4.59 (d, 2H, J ) 9 Hz), 4.16 (dq, 4H, J ) 7, 14 Hz), 1.30 (t,
6H, J ) 7 Hz); 13C NMR (62.9 MHz, CDCl3) δC ) 164.42 (d, J )
8 Hz), 153.97, 151.08, 137.33, 125.24, 123.48, 62.97 (d, J ) 6
Hz), 57.58 (d, J ) 170 Hz), 16.51 (d, J ) 6); MS (EI+) m/z (rel
intens) ) 274 (40) [M + H+], 547 (100) [2M + H+], 569 (60)
[2M + Na+]; HRMS (EI+) m/z calcd for C11H17NO5P 274.0844,
found 274.0857; FT-IR (thin film) νmax/cm-1 2982, 2934, 1736,
1593, 1423, 1329, 1275, 1249, 1119, 1052, 1025, 971.
toluene
DCM
CHCl3
acetone
λmax/nm
425
421
420
425
that the concentration of porphyrin remained constant throughout the
titration. Aliquots of pyridine solution were added successively to the
cell containing the porphyrin solution, and the UV/vis spectrum was
recorded after each addition. Changes in absorbance for the Soret band
of the porphyrin were fit to a 1:1 binding isotherm in Microsoft Excel
to obtain the association constant. Each titration was repeated at least
twice, and the experimental error is quoted as 2 times the standard
deviation at a precision of one significant figure.
UV/vis dilutions of 3 were carried out by preparing a 0.03 mM
solution of porphyrin 3 in DCM. Aliquots of porphyrin 3 solution
were added successively to a cell containing 2 mL of DCM, and
the UV/vis spectrum was recorded after each addition. Changes in
the absorbance of the Soret band fit well to a Lambert-Beer law
showing no signs of aggregation.
Isothermal Titration Calorimetry. ITC experiments were per-
formed at 25 °C on a VP-ITC MicroCal titration calorimeter (MicroCal,
Inc., Northampton, MA). In a typical calorimetric measurement,
porphyrin was dissolved in spectroscopic grade solvent (CHCl3,
C2H2Cl4, toluene, DCM) at a concentration on the order of 0.5-1 mM,
and the solution was loaded into the sample cell of the microcalorim-
eter. The guest solution (15-20 times more concentrated than the host
solution) was loaded into the injection syringe. The number of
injections was between 50 and 60, and the volumes of the injections
were between 5 and 7 µL with 30-40 s of duration and 300 s of
spacing between the injections. Dilution experiments were performed
for each titration by loading the guest solution (at the same concentra-
tion as the titration experiment) into the injection syringe and adding
the guest to pure solvent in the cell. The dilution data were subtracted
from each host titration thermogram. The data fitting was performed
by using ORIGIN (version 7.0, Microcal, LLC ITC) and a 1:1 binding
isotherm (One Set of Sites model), fixing the stoichiometry number
to 1 and allowing the stability constant K and binding enthalpy ∆H
values to float. At least two independent measurements were performed
for each complex, and the average results are shown.
Diethyl Pyridine-3,5-dicarboxylate, 4c. A mixture of 3,5-py-
ridinedicarboxylic acid (1 g, 5.98 mmol), toluene (10 mL), DMF (15
µL), and SOCl2 (30 mL) was refluxed for 1 h under protection by a
DryRite drying tube. The solution was condensed under reduced
pressure, and DCM (120 mL) was added. Then ethanol (3.50 mL, 60
mmol) in DCM (20 mL) and triethylamine (6.78 mL, 48.8 mmol) were
added in small portions. The solution was allowed to stir for 18 h,
then washed with NaHCO3(aq) (2 × 100 mL) and brine (100 mL),
dried with NaSO4, and condensed under reduced pressure. The crude
product was purified on silica (25 g) eluting with EtOAc/hexane. The
product was isolated as a clear oil: yield 1.18 g (91%); 1H NMR (250
MHz, CDCl3) δH ) 9.30 (s, 1H), 8.79 (s, 0.5H), 4.39 (q, 2H, J ) 7.1
Hz), 1.37 (t, 3H, J ) 7.1 Hz); 13C NMR (62.9 MHz, CDCl3) δC )
164.45, 154.09, 137.88, 126.20, 61.80, 14.24; MS (EI+) m/z (rel intens)
) 196 (20), 224 (100) [M + H+]; HRMS (EI+) m/z calcd for
C11H14NO4 244.0923, found 224.0931; FT-IR (thin film) νmax/cm-1
2987, 1730, 1601, 1450, 1369, 1314, 1259, 1243, 11.01, 1027.
Bis(diethoxyphosphoryl) Pyridine-3,5-dicarboxylate, 4d. A mix-
ture of 3,5-pyridinedicarboxylic acid (1 g, 5.98 mmol), toluene (10
mL), DMF (15 µL), and SOCl2 (30 mL) was refluxed for 1 h under
protection by a DryRite drying tube. The solution was condensed under
reduced pressure, and DCM (120 mL) was added. Then diethyl
hydroxymethyl phosphonate (1.20 mL, 8.12 mmol) in DCM (20 mL)
and triethylamine (6.78 mL, 48.8 mmol) were added in small portions.
The solution was allowed to stir for 18 h. DCM (200 mL) was added,
and the resulting solution was then washed with NaHCO3(aq) (2 ×
100 mL) and brine (100 mL), dried with NaSO4, and condensed under
reduced pressure. The crude product was purified on silica (60 g)
eluting with DCM/0-5% methanol gradient. The product was isolated
as a clear oil: yield 2.20 g (80%); 1H NMR (250 MHz) δH ) 9.36 (d,
2H, J ) 2 Hz), 8.85 (t, 1H, J ) 2 Hz), 4.62 (d, 4H, J ) 9 Hz), 4.18
(dq, 8H, J ) 7, 8 Hz), 1.31 (t, 12H, J ) 7 Hz); 13C NMR (62.9 MHz,
CDCl3) δC ) 163.68 (d, J ) 8 Hz), 154.98, 138.78, 125.54, 63.28 (d,
J ) 6 Hz), 58.18 (d, J ) 170 Hz), 16.74 (d, J ) 6 Hz); MS (EI+) m/z
(rel intens) ) 468 (100) [M + H+], 490 (40) [M + Na+]; HRMS
(EI+) m/z calcd for C17H28NO10P2 468.1188, found 468.1170; FT-IR
(thin film) νmax/cm-1 2985, 2931, 1737, 1600, 1331, 1260, 1227, 1103,
1054, 1025, 971.
Dilution experiments were used to check for the effects of
aggregation of the porphyrin host. The porphyrin solution was loaded
into the sample cell, and pure solvent was added from the syringe. In
all cases, the dimerization constant and the dimerization enthalpy were
too low to be measured, so aggregation can safely be ignored in these
experiments.
1H NMR Titrations. NMR titrations were carried out by preparing
a 3 mL sample of host at known concentration (1 mM for porphyrin
hosts and 50 mM for 6). Then 0.6 mL of this solution was removed,
3,5,5-Trimethyl-N-phenylhexanamide, 5. A mixture of aniline
(1.51 g, 16.1 mmol) and triethylamine (6.79 mL, 48.3 mmol) in DCM
(125 mL) protected by a CaCl2 drying tube was cooled to 0 °C. To
this was added isononyl chloride (2.84 g, 16.1 mmol), and the resulting
mixture was stirred for 18 h. The solution was then washed with
NaHCO3(aq) (2 × 100 mL) and brine (100 mL), dried with NaSO4,
and condensed under reduced pressure. The crude product was purified
on silica (80 g) eluting with hexane/ethyl acetate. The product was
1
and a H NMR spectrum was recorded. A 2 mL solution of guest
(10-1400 mM) was prepared using the host solution, so that the
concentration of porphyrin 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 for the porphyrin aromatic signals
were analyzed by using the appropriate binding isotherms in Microsoft
Excel. Each titration was repeated at least twice, and the experimental
error is quoted as twice the standard deviation at a precision of one
significant figure.
1
isolated as a waxy solid: yield 2.79 g (74%); H NMR (250 MHz,
CDCl3) δH ) 8.02 (s, 1H), 7.70 (d, 2H, J ) 8 Hz), δ 7.44 (t, 2H, J )
8 Hz), 7.24 (t, 1H, J ) 7 Hz), 2.41 (m, 3H), 1.36 (m, 2H)1.18 (d, 3H,
J ) 6 Hz), 1.60 (s, 9H); 13C NMR (62.9 MHz, CDCl3) δC ) 171.45,
138.21, 129.03, 124.29, 120.25, 50.76, 47.63, 31.22, 30.16, 27.65,
22.81; MS (EI+) m/z (rel intens) ) 234 (100) [M + H+], 256 (60)
[M + Na+]; HRMS (EI+) m/z calcd for C15H24NO 234.1858, found
234.1866; FT-IR (thin film) νmax/cm-1 3295, 3058, 2956, 2905, 2862,
1654, 1603 1543 1499, 1444, 1364, 1302, 1251, 1204.
UV/Vis Absorption Titrations. UV/vis titrations were carried out
by preparing a 10 mL sample of the porphyrin 3 at known concentra-
tion (1-10 µM) in spectroscopic grade solvent. A 2 mL portion of
this solution was removed, and a UV/vis spectrum was recorded using
a Peltier thermostat set at 298 K (Table 7). A 2 mL solution of pyridine
ligand (5-2000 µM) was prepared using the porphyrin solution, so
Acknowledgment. We thank the EPSRC (C.A.H. and S.M.T.)
for funding.
Supporting Information Available: UV/vis absorption dilution
1
data for 3 in all five solvents, H NMR spectra of 3 in acetone,
TCE, and chloroform, and NMR spectroscopic characterization of
2, 3, 4b, 4c, 4d, and 5. This material is available free of charge
via the Internet at http://pubs.acs.org.
JA803434Z
9
J. AM. CHEM. SOC. VOL. 130, NO. 52, 2008 17725