6766
J. Phys. Chem. B 2007, 111, 6766-6771
Origins of Cooperative Noncovalent Host-Guest Chemistry in Mixed Valence Complexes† Benjamin J. Lear and Clifford P. Kubiak* Department of Chemistry and Biochemistry, UniVersity of California, San Diego, La Jolla, California 92093-0358 ReceiVed: December 27, 2006; In Final Form: March 8, 2007
The electronic effects resulting from noncovalent host-guest interactions between calix[6]arene and a ruthenium dimer, [Ru3O(OAc)6(CO)(ppy)]2-µ-pz (ppy ) 4-phenyl pyridine, pz ) pyrazine), are presented. The noncovalent interaction is between the calix[6]arene and the ppy ligands of the dimer. The dimer can bind 2 equiv of calix[6]arene. The complex [Ru3O(OAc)6(CO)(ppy)]2-µ-pz forms a highly stable mixed valence ion with strong electronic coupling between the two Ru3 clusters. The strength of the electronic interaction is found to be moderated by calix[6]arene binding. Addition of calix[6]arene to the mixed valence ion causes the electronic coupling to decrease. The binding of calix[6]arene is found to be cooperative. The origins of cooperative binding are developed in terms of the potential energy surfaces associated with the symmetric and asymmetric mixed valence ion. In particular, it is found that symmetry breaking (through the binding of a single calix[6]arene) destabilizes the mixed valence state. Restoration of symmetry (through the binding of a second calix[6]arene) increases the stability of the mixed valence ion and provides an additional driving force for the binding of the second calix[6]arene.
Introduction With the advent of the fields of supramolecular chemistry and molecular electronics there has been increased interest in the directed self-assembly of functional nanostructures. Noncovalent “host-guest” interactions have been of particular interest. Recently, Stoddart and co-workers discussed the importance of cooperative interactions in guiding the selfassembly of specific supramolecular structures among many possible structures.1 Here, we describe the noncovalent hostguest chemistry of an inorganic mixed valence complex with calixarenes. We show that symmetry breaking of a mixed valence ion caused by 1:1 calixarene binding produces reasonably strong cooperativity in the binding of the second calixarene. The origins of cooperative binding of calixarenes to the mixed valence complex are discussed in terms of the loss and recovery of electronic stabilization energy in asymmetric and symmetric host-guest complexes, respectively. Dimers of trinuclear ruthenium clusters of the type [Ru3O(OAc)6(CO)(L′)]-µ2-BL-[Ru3O(OAc)6(CO)(L′′)] (BL ) bridging ligand) exhibit electron transfer (ET) on the picosecond time scale in their mixed valence states.2,3 The CO ligand provides a strong IR chromophore that, through π-backbonding, is sensitive to the oxidation state of the cluster to which it is attached. Thus, for a neutral cluster, ν(CO) appears at about 1940 cm-1, while for a fully reduced (-2) cluster ν(CO) appears at about 1890 cm-1. Because ET is extremely fast in the mixed valence (-1) states of the “dimers of trimers”, its effects can be seen as dynamic broadening of the ν(CO) band shape in the IR spectra. The overall effect is somewhat analogous to dynamic NMR,4-6 but, as the effect is manifest on the IR time scale, the dynamics probed occur on a time scale approximately one billion times faster than those typically studied by NMR.7 †
Part of the special issue “Norman Sutin Festschrift”. * To whom correspondence should be addressed. E-mail: ckubiak@ ucsd.edu.
The IR spectra that show effects of rapid exchange can be simulated and the rate constants estimated in this way.8 For mixed valence complexes of the type [Ru3O(OAc)6(CO)(L′)]µ2-pz-[Ru3O(OAc)6(CO)(L′)]- (pz ) pyrazine), rate constants, ket, estimated for the rate of intramolecular ET are on the order of 1012 s-1. Such fast ET rates underscore the large degree of electronic coupling between the Ru3 clusters in mixed valence complexes of this type. The coupling is usually expressed as the magnitude of the electronic coupling matrix,9 HAB, which functions to mix the electronic wavefunctions of two sites between which electron exchange occurs.10-12 Historically, determination of the value of HAB has been difficult for complexes which, like ours, reside at the Robin-Day13 Class II/III14 borderline. Recently, we have reported thermodynamic estimates of HAB for asymmetric “dimers of trimers” mixedvalence complexes of the type [Ru3O(OAc)6(CO)(L′)]-µ2-pz[Ru3O(OAc)6(CO)(L′′)]- (pz ) pyrazine).15 These asymmetric dimers form what we have termed, “mixed-valence isomers”.16,17 Mixed-valence isomerism refers to the two possible charge distributions of an asymmetric mixed-valence complex, for example, A--B and A-B-. The equilibrium constant that describes the populations of major and minor mixed-valence isomers pertains directly to the driving force (∆G1) for ET on a strongly coupled adiabatic potential energy surface. The driving force for ET in the uncoupled diabatic limit (∆G0) can be well estimated from electrochemical data. The difference between the driving force in the diabatic and adiabatic limits then is directly related to HAB. These thermodynamic estimates of HAB are quite close to λ/2 (where λ is the total reorganization energy), which is the asymptotic limit for a Robin-Day class III fully delocalized system. However, the unambiguous appearance of both mixed valence isomers in IR spectra clearly indicates the persistence of a double minimum potential energy surface. This is a definitive example of a borderline class II/III system. The large estimated values of HAB also correctly predict many of the properties of this class of mixed-valence complexes, including nearly activationless ET evident in the temperature
10.1021/jp068964l CCC: $37.00 © 2007 American Chemical Society Published on Web 04/18/2007
Cooperative Noncovalent Host-Guest Chemistry independence of rate constants and strong correlation of kET with solvent dipole reorientation times8 that figure prominently in the pre-exponential terms of typical rate constant expressions for ET.18 In the present study, we describe the effects of dynamically induced asymmetry. By “dynamically induced asymmetry” we mean asymmetry that is not permanently manifest via connectivity of the molecule (i.e., ligand substitution) but which can be realized through a noncovalent interaction with some other chemical species. To date, we have only investigated complexes that possess inherent substitutional asymmetry in their molecular structures. We now turn to complexes that are inherently symmetric, but which, through interaction with their environment, can become asymmetric. Specifically, we are interested in the system comprised of [Ru3O(OAc)6(CO)(ppy)]2µ-pz (1) and calix[6]arene. It is known that calixarenes interact with aromatic molecules in a host-guest manner.19-21Thus, it is expected that calix[6]arene will complex with 1 through interaction with the 4-phenyl pyridine ligands. We find that the understanding of potential energy surfaces gained through the study of mixed-valence isomers can be used to explain the effects of dynamic symmetry breaking on symmetric mixedvalence complexes.
Experimental Methods Synthesis. The synthesis of the dimer, [Ru3O(OAc)6(CO)(ppy)]2-µ-pz (1), proceeds by the coupling of two trinuclear clusters. First, 1 equiv of [Ru3O(OAc)6(CO)(H2O)2]22 is stirred with 0.8 equiv of ppy in a 1:1 mixture of methylene chloride and methanol for 2 days to yield [Ru3O(OAc)6(CO)(ppy)(H2O)], which was purified on a silica column using 1% methanol in chloroform. To [Ru3O(OAc)6(CO)(ppy)(H2O)] was added 20 equiv of pyrazine (pz) in methylene chloride and the reaction was stirred for 30 min. The solvent was then removed to a minimal volume and approximately 6 times that volume of hexanes was added. The solution was filtered, and [Ru3O(OAc)6(CO)(ppy)(pz)] was collected as a precipitate. Equal molar quantities of the clusters [Ru3O(OAc)6(CO)(ppy)(H2O)] and [Ru3O(OAc)6(CO)(ppy)(pz)] were then stirred in chloroform for 2 days, yielding the dimer of interest, [Ru3O(OAc)6(CO)(ppy)]2µ-pz (1), which was purified on a Bio-Beads SX-3 size exclusion column packed in chloroform. Anal. Calcd for C52H58N4O28Ru6: C, 34.82; H, 3.26; N, 3.25. Found: C, 34.86; H, 3.53; N, 2.78. 1H NMR (400 MHz, CDCl3): δ 2.00 (s, 12H, acetate), 2.14 (t, 12 H acetate), 2.23 (s, 12 H, acetate), 7.56 (t, 2 H, phenyl), 7.64 (t, 4 H, phenyl), 7.89 (d, 4 H, phenyl), 8.25 (d, 4 H, pyridine), 9.02 (d, 4 H, pyrazine), 9.23 (d-unresolved, 4 H pyridine). Measurements. Cyclic voltammetry and differential pulse voltametry were performed using a Bio-analytical systems
J. Phys. Chem. B, Vol. 111, No. 24, 2007 6767 TABLE 1: Values for E1/2(1), E1/2(2), and ∆E1/2 for Complex 1 as a Function of the Equivalents of Calix[6]arene Presenta equivalents of calix[6]arene
E1/2(1)
E1/2(2)
∆E1/2
0.0 0.2 0.4 0.6 0.8 1.0 2.0 3.0 4.0 5.0 6.0
1.126 1.130 1.131 1.133 1.132 1.133 1.130 1.130 1.130 1.131 1.128
1.500 1.502 1.504 1.505 1.504 1.504 1.489 1.483 1.484 1.484 1.481
0.374 0.372 0.373 0.372 0.372 0.371 0.359 0.353 0.354 0.353 0.353
a Equivalents of calix[6]arene added to a 2 mM solution of 1, together with the position of E1/2 (1), E1/2 (2), and ∆E1/2 vs Fc/Fc+.
CV-50W potentiostat. Electrochemistry was performed in dichloromethane in the presence of 0.1 M tetrabutylammonium hexaflourophosphate (TBAH). The working electrode was a gold disc (3 mm diameter), the counter electrode was a platinum wire, and the reference electrode was the ferrocene/ferrocenium couple. Results and Discussion The electrochemistry of 1 was investigated using both cyclic voltammetry (CV) and differential pulse voltammetry (DPV). In the anodic region, the dimer undergoes two reversible two-electron oxidations at +0.158 V [(RuIIIRuIIIRuII)2 f (RuIIIRuIIIRuIII)2] and at +0.951 V [(RuIIIRuIIIRuIII)2 f (RuIVRuIIIRuIII)2]. In the cathodic region, the two ruthenium clusters are reduced sequentially, giving rise to two reversible one-electron reductions at -1.126 V [(RuIIIRuIIIRuII)2 f (RuIIIRuIIRuII)(RuIIIRuIIIRuII)] and -1.500 V [(RuIIIRuIIRuII)(RuIIIRuIIIRuII) f (RuIIIRuIIRuII)2]. That the reduction of the clusters occurs sequentially is indicative of electronic communication between the clusters. This has been observed in analogous pyrazine bridged mixed-valence complexes.2,3 The degree of splitting (∆E1/2) is a qualitative measurement of the degree of electronic communication or coupling (HAB) between clusters.23 In this case, the value of ∆E1/2 is 374 mV, corresponding to a conproportionation constant, Kc ) 2.18 × 106. These values are comparable to those reported previously for similar mixed-valence dimers.2,3 To study the effects of calixarene binding on the electronic properties of 1 a titration of calix[6]arene into a solution of 1 was performed. Table 1 summarizes the electrochemical data obtained during this titration, while Figure 1 shows the forward (top) and reverse (bottom) DPV of 1 at the beginning (0 equiv of calix[6]arene) and endpoint (6 equiv of calix[6]arene) of this titration. There are three transformations that are immediately obvious from this data. First, the position of the first reduction, E1/2(1), changes very little upon the addition of calix[6]arene, shifting more negative by only 2 mV. Second, the potential of the second reduction, E1/2(2), experiences a comparatively large positive shift upon addition of calix[6]arene, a total of 19 mV. Third, the separation between the two reduction events (∆E1/2) is found to decrease upon the addition of calix[6]arene, changing in magnitude from 374 mV to 352 mV. There are two types of electronic perturbations which give rise to these three observed effects: (1) The change in the intrinsic reduction potentials of the clusters, that is, inductive effects from changes in the pKa of the ppy ligand upon interaction with calix[6]arene,24 and (2) changes in the comproportionation equilibrium constant, Kc, owing to changes in the stability of the mixed-valance ion caused by electronic delocalization.
6768 J. Phys. Chem. B, Vol. 111, No. 24, 2007
Lear and Kubiak
Figure 2. Square scheme for the binding of calix[6]arene to 1 and the reduction of the bound and unbound dimer. The scheme presented is only for the first binding event, though the results are generally applicable for the binding of the second calixarene.
Figure 1. Forward (top) and reverse (bottom) DPV of 2 mM of 1 with 0 equiv of calix[6]arene (solid lines) and 6 equiv of calix[6]arene (dashed lines).
The most straightforward contribution to the change in reduction potentials arises from the decrease in ∆E1/2 (effect 2 above). As ∆E1/2 decreases, the first and second reduction waves are expected to move toward each other by equal amounts. Thus, each wave contributes one-half of the total ∆∆E1/2 and the first wave moves more negative by 11 mV while the second wave moves more positive by 11 mV. While the shifts in potential due to ∆∆E1/2 are easily unraveled, the physical origin of these shifts cannot be understood without first discussing the direct electronic effects of calix[6]arene binding to 1. These effects are a direct result of the changes in cluster orbital energies upon binding of calix[6]arene. The orbital energy levels of the clusters are extremely sensitive to the electronic nature of the attached ancillary ligands.24 The binding of calix[6]arene to the ppy ligand is expected to perturb the electronics of the ppy, which, in turn, brings about a change in the reduction potential of the clusters. Given the total change in the positions of the reduction waves and the contribution arising from ∆∆E1/2, we can extract the degree to which the energy levels of the clusters are adjusted by the binding of calix[6]arene. For E1/2(1), the total potential shift upon exposure of 1 to calix[6]arene was -2 mV. Recalling that the contribution from ∆∆E1/2 was -11 mV, the binding of calix[6]arene must move the potential of E1/2(1) by +9 mV. A similar analysis of E1/2(2) reveals that the binding of calix[6]arene shifts the potential of the reduction positive by +8 mV. Thus, taking into consideration the contribution from ∆∆E1/2, the total shift for E1/2 (2) is observed as +19 mV. From these simple considerations, it is evident that the change in potential due to calix[6]arene binding is essentially the same for both E1/2 (1) (+9 mV) and E1/2 (2) (+8 mV). This is expected as the electronic process associated with these potentials is the addition of an electron into identical orbitals on two different clusters. Since the same orbital is involved in both reductions, the effect of calix[6]arene upon each is expected to be similar.
In view of the overall weak binding of calix[6]arene to 1, it is important to have independent evidence that this interaction is occurring. In a 2D NOESY NMR experiment, the proton of calix[6]arene para to the hydroxyl group was observed to interact with the most distal proton on the ppy ligand. This experiment was performed on a 400 MHz 1H NMR spectrometer with a mixing time, τM ) 1 s and mole ratios of calix[6]arene to 1 of 2:1. This clearly shows that the interior cup of the calix[6]arene is interacting with the 4-phenylpyridine ligand in a host-guest manner. This study confirms calix[6]arene binding, but the low intensities of the 1H-NOESY cross-peaks also highlight that this binding is very weak and reversible. The weak nature of the binding of calix[6]arene to neutral 1, we will show, is a benefit in this study. The weak interaction of calix[6]arene and 1 is found to be required to observe cooperative binding behavior (vide infra). The fact that the electrochemical potential shifts that arise from calixarene binding are positive indicates that the binding of calixarene to the clusters has a stabilizing effect on the cluster to which it binds. This can be seen by examining a simple square scheme for calixarene binding (Figure 2). The shift in the reduction potential for a complex that undergoes a binding event is related to the binding constant by the equation
(Eb - Eu)
( )
Kred nF ) ln RT Kox
(1)
where Eu is the reduction potential of the complex in its unbound state, Eb is the reduction potential of the complex in its bound state, Kox is the binding constant for the complex in its oxidized state, and Kred is the binding constant for the complex in its reduced state. From eq 1, it can be seen that a positive shift in reduction potential means that Kred is larger than Kox. That is, the binding event is more favorable when the complex is in its reduced state. This, in turn, means that the reduced cluster must be more stabilized by the binding of calixarene than the neutral cluster (∆Gred is greater than ∆Gox). Thus, in the reduced state, the binding of calixarene has the effect of lowering the energy (reduction potential) of the cluster to which it is attached. Once the overall effects of the noncovalent binding of calix[6]arene to the mixed-valence complex are understood in terms of the separate changes in intrinsic reduction potential and changes in electron exchange stabilization, the origin of ∆∆E1/2 is easily explained. It has been hypothesized that electronic communication between clusters in the pz-bridged mixedvalence complexes proceeds through the LUMO of the pz.25 The LUMO of the pz is higher in energy than the cluster’s orbitals that are involved in intercluster electronic communication. Since the calix[6]arene has the effect of lowering the orbital energies of the clusters to which it is attached, the energy gap between the clusters and the pz bridge will increase. This, in
Cooperative Noncovalent Host-Guest Chemistry
J. Phys. Chem. B, Vol. 111, No. 24, 2007 6769 the diabatic and adiabatic surfaces for the dimer s requires comment. The energy of the lower potential energy surface, G1, is given by the equation26
1 G1 ) {(Gb + Ga) - [(Gb - Ga)2 + 4HAB2]1/2} 2
(2)
where Ga and Gb are the energies of the reactants and products in the diabatic case (in this case the energy difference between the two clusters in the dimer) and HAB is the effective electronic coupling matrix which mixes the wavefunctions of the reactants and products. In the diabatic case, the wavefunctions are not mixed and HAB is equal to zero. Thus, the stabilization afforded by the electronic coupling, ∆G1, can be obtained by subtracting the lower diabatic potential energy from the lower adiabatic energy. This is given by Figure 3. Plot of ∆E1/2 vs equivalents of calix[6]arene. The sigmiodal curve is fit with an R2 ) 0.9985 and a χ2 ) 1.987 × 10-7. R2 is a relative measure of correlation, the better the correlation between the curve and the data, the closer the value of R2 is to 1. χ2 is a measure of statistical significance, the lower the value of χ2, the more significant the fit.
turn, will lead to smaller effective coupling and less electronic communication between the clusters, reflected by a decrease in ∆E1/2sas is observed by electrochemical methods. There is one important effect that is a result of the modulation of electronic coupling by calix[6]arene binding. This is best illustrated by plotting ∆E1/2 vs equivalents of calix[6]arene (Figure 3). The most striking aspect of this graph is the strong sigmoidal shape of the curve, which is indicative of cooperative binding. In the current system, each dimer contains two separate sites for the binding of calix[6]arene, and the shape of the curve in Figure 3 suggests that binding of one calix[6]arene positively influences the binding of the second. Thus, 1 must be more stable with two calix[6]arenes bound than with just a single calix[6]arene bound. This result is unexpected for two reasons. First, in previous studies which investigated the binding of hosts to multiple sites of the same molecule, the binding events have been observed to be negatiVely cooperative.1 That is, the binding of the first host interferes with the binding of the second. This is often attributed to steric hindrance between the hosts. In our complexes, the hosts (calixarenes) are sufficiently separated spatially that steric interactions are not expected to play a major role in the binding of the second calixarene. However, the elimination of the steric considerations merely removes the reasoning for negative cooperation, it does not explain the observed positive cooperation. The second reason why observation of positive cooperativity is surprising is that there is direct thermodynamic opposition to it. Binding of the calixarene decreases ∆E1/2. Thus, binding of the calixarene lowers the resonance stabilization of the mixed-valence dimer. This is partially compensated by the average decrease in cluster reduction potentials that reflects the stabilizing influence of host-guest complex formation. Overall, however, the changes in (1) intrinsic reduction potentials and (2) ∆E1/2 lead to a shift in reduction potential that is negatiVe. Thus, what we are observing is positive cooperativity for an event that is overall energetically uphill. It is clear that the cooperative binding of calix[6]arene cannot be explained through any direct effects. The cooperative binding of calixarene can be explained however, if one considers the energy of the potential energy surfaces associated symmetric and asymmetric mixed-valence ions of uncomplexed and complexed 1, respectively. In particular, the resonance stabilization s or the difference between
∆G1 ) G1(adiabatic) - G1(diabatic) ) 1 {(Gb - Ga) - [(Gb - Ga)2 + 4HAB2]1/2} (3) 2 In the symmetric dimer, (Gb - Ga) equals zero and ∆G1 is equal to -HAB. In the asymmetric case, the (Gb - Ga) terms remain and it can be seen by inspection that the value of ∆G1 will be more endergonic than -HAB. Specifically, the difference in resonance stabilizations between the asymmetric and symmetric dimers, ∆∆G1, is given by
∆∆G1 ) ∆G1(asymmetric) - ∆G1(symmetric) 1 ∆∆G1 ) {(Gb - Ga) - [(Gb - Ga)2 + 4HAB2]1/2} + HAB 2 (4) Previously, we have reported the properties of asymmetric Ru3 dimers, which gave rise to the observation of mixed-valence isomers.16,17 We have also reported thermodynamic estimates of HAB for these dimers.15 If the assumption is made that the HAB value for the symmetric dimers is similar to those calculated for their closely analogous asymmetric dimers, then we can arrive at a numerical estimation for the value of ∆∆G1. By the use of HAB equal to 4250 cm-1 and Gb - Ga equal to 1850 cm-1 (the values for the dimer in which L ) dmap (4-dimethylaminopyridine) and L′ ) cpy (4-cyanopyridine)), one obtains a value for ∆∆G1 of 825 cm-1 (9.87 kJ/mol). That is, the lower potential energy surface for the asymmetric dimer is expected to be higher in energy than the potential energy surface for the symmetric dimer by 825 cm-1. Thus, it is clear that the introduction of an asymmetry destabilizes the mixed-valence ion by a significant amount. It should be noted that the above calculations are by no means intended to be exhaustive as we have not calculated, ab initio, the potential energy surfaces for these dimers. Rather, the results presented above are meant to demonstrate the effect of the symmetry of a molecule upon potential energy surfaces in general. Because of this, the conclusions reached here are generally applicable to all mixed-valence systems with a reasonably strong degree of electronic coupling. Namely, the property of symmetry can have a strong effect on the energetics of a mixed-valence system undergoing electron exchange. Given that the introduction of asymmetry can have the effect of decreasing resonance stabilization in these complexes, we can now formulate an explanation for the origin of the coopertivity observed for the binding of calix[6]arene to 1 (Scheme 1). In its uncomplexed state, with no calix[6]arene bound, 1 is a symmetric dimer (Scheme 1, stage A). The binding
6770 J. Phys. Chem. B, Vol. 111, No. 24, 2007
Lear and Kubiak
SCHEME 1: Binding of Calix[6]arene to 1a
a In stage A, the dimer, [Ru3O(OAc)6(CO)(ppy)]2-µ-pz, is in its symmetric uncomplexed form. In stage B, a single calix[6]arene has bound to the dimer. This binding is driven by the stabilization of the individual cluster to which it binds. However, the binding also introduces an asymmetry and, as a result, decreases the resonance stabilization of the mixed valence ion. In stage C, a second calix[6]arene has bound to the cluster. This restores the symmetry of the dimer and increases the resonance stabilization. Thus, the binding of the second dimer is driven by both the stabilization of the individual cluster to which it binds and the stabilization of the mixed valence isomer. The additional stabilization of the mixed valence isomer is the “extra” driving force that gives rise to the cooperative binding of calix[6]arene to [Ru3O(OAc)6(CO)(ppy)]2-µ-pz. In this scheme, the reaction coordinate for the potential energy surface is that of all internal nuclear motions as well as solvent modes that must change during the course of the reaction. These coordinates are not quantitatively specified, but are meant as a general indication as to the progress of the reaction (i.e., the products and reactants both have individual minima that can be moved between via a high-energy intermediate, transition state).
of a single calix[6]arene to 1 necessarily introduces an asymmetry to the dimer (Scheme 1, stage B). This has the effect of decreasing the resonance stabilization in the dimer. Binding of a second calix[6]arene restores symmetry to 1 (Scheme 1, stage C). Thus, substantial resonance stabilization is restored as a result of the binding of the second calix[6]arene. It can now be seen why the initial interaction of calix[6]arene with neutral 1 must be small in order for the system to behave in the way that we observe. It is known that the binding constants are related to the change in energy associated with the binding event by the equation ∆G ) -RT ln(Keq). The change in energy associated with electronic delocalization derived in eq 4 is related to a change in binding constant for calix[6]arene; see eq 5.
( )
Keq1 ∆∆G1 ) -RT ln Keq2
arene is reasonably strong. We conclude that if the initial interaction were too strong, we would not observe the sigmoidal curve (Figure 3) at such high molar ratios of calix[6]arane to 1. In summary, the binding of the first calixarene is driven by the stabilization of the cluster to which it binds. This is observed in the electrochemistry as the collective shift of the reduction waves by about +9 mV (vide supra). The binding of the second calixarene is driven by this same energy; however, it is also driven by the increase in resonance energy that is realized upon restoration of symmetry to the cluster. It is this additional driving force accompanying the binding of the second calixarene, which leads to the observation of cooperative binding evidenced by the shape of the curve in Figure 3. Conclusion
(5)
Here, Keq1 is the binding constant associated with the binding of the first calix[6]arene to 1 (introducing an asymmetry) and Keq2 is the binding constant associated with the binding of the second calix[6]arene to 1 (restoring symmetry). In the previous section, we showed that a reasonable estimate of the resonance destabilization, ∆∆G1, caused by symmetry breaking induced by binding of a single calix[6]arene was 825 cm-1. This assumed value of ∆∆G1 and eq 5 show that the binding of the second calix[6]arene is a factor of 54 times stronger than the binding of the first. Thus, even though calix[6]arene interacts weakly with 1, we can expect that the binding of the second calix[6]-
We have demonstrated the ability to control electronic coupling between metal sites through small noncovalent interactions at ancillary positions on the coupled metal clusters. We have shown how relatively weak interactions at distal positions in a complex can give rise to large electronic effects within the complex, confirming that highly coupled borderline class II-III complexes are extremely sensitive to their environment. The implications for the field of supramolecular chemistry are clear: during the self-assembly of a “communicating” system, one must consider the dynamics of the environment and the influence the environment will have on the energetic landscape of the system. Moreover, the translation from a chemical to an electronic signal is a concept that will prove useful in the design and construction of molecular electronic devices. Currently,
Cooperative Noncovalent Host-Guest Chemistry work is underway in our laboratory to identify other modes of noncovalent interaction that are able to give rise to large electronic effects within dimers similar to the one reported here. Acknowledgment. This work was supported by NSF (CHE0315593 and CHE-0616279). References and Notes (1) Badjic, J. D.; Nelson, A.; Cantrill, S. J.; Turnbull, W. B.; Stoddart, J. F. Acc. Chem. Res. 2005, 38, 723. (2) Ito, T.; Hamaguchi, T.; Nagino, H.; Yamaguchi, T.; Washington, J.; Kubiak, C. P. Science 1997, 277, 660. (3) Ito, T.; Hamaguchi, T.; Nagino, H.; Yamaguchi, T.; Kido, H.; Zavarine, I. S.; Richmond, T.; Washington, J.; Kubiak, C. P. J. Am. Chem. Soc. 1999, 121, 4625. (4) Grevels, F. W.; Jacke, J.; Klotzbuecher, W. E.; Krueger, C.; Seevogel, K.; Tsay, Y. H. Angew. Chem. 1987, 99, 960. (5) Turner, J. J.; Gordon, C. M.; Howdle, S. M. J. Phys. Chem. 1995, 99, 17532. (6) Wood, K. A.; Strauss, H. L. J. Phys. Chem. 1990, 94, 5677. (7) Londergan, C. H.; Kubiak, C. P. Chem.-Eur. J. 2003, 9, 5962. (8) Londergan, C. H.; Salsman, J. C.; Ronco, S.; Dolkas, L. M.; Kubiak, C. P. J. Am. Chem. Soc. 2002, 124, 6236. (9) Brunschwig, B. S.; Sutin, N. Coord. Chem. ReV. 1999, 187, 233. (10) Marcus, R. A.; Eyring, H. Annu. ReV. Phys. Chem. 1964, 15, 155.
J. Phys. Chem. B, Vol. 111, No. 24, 2007 6771 (11) Marcus, R. A. J. Chem. Phys. 1956, 24, 966. (12) Allen, G. C.; Hush, N. S. Prog. Inorg. Chem. 1967, 8, 357. (13) Robin, M. B.; Day, P. AdV. Inorg. Chem. Radiochem. 1967, 10, 247. (14) Demadis, K. D.; Hartshorn, C. M.; Meyer, T. J. Chem. ReV. 2001, 101, 2655. (15) Londergan, C. H.; Salsman, J. C.; Lear, B. J.; Kubiak, C. P. Chem. Phys. 2006, 324, 57. (16) Ito, T.; Imai, N.; Yamaguchi, T.; Hamaguchi, T.; Londergan, C. H.; Kubiak, C. P. Angew. Chem., Int. Ed. 2004, 43, 1376. (17) Salsman, J. C.; Kubiak, C. P.; Ito, T. J. Am. Chem. Soc. 2005, 127, 2382. (18) Sutin, N. Prog. Inorg. Chem. 1983, 30, 441. (19) de Namor, A. F. D.; Cleverley, R. M.; Zapata-Ormachea, M. L. Chem. ReV. 1998, 98, 2495. (20) Baudry, R.; Kalchenko, O.; Dumazet-Bonnamour, I.; Vocanson, F.; Lamartine, R. J. Chromatogr. Sci. 2003, 41, 157. (21) Gutsche, C. D. Calixarenes; The Royal Society of Chemistry, 1989. (22) Baumann, J. A.; Wilson, S. T.; Salmon, D. J.; Hood, P. L.; Meyer, T. J. J. Am. Chem. Soc. 1979, 101, 2916. (23) Richardson, D. E.; Taube, H. Coord. Chem. ReV. 1984, 60, 107. (24) Toma, H. E.; Cunha, C. J.; Cipriano, C. Inorg. Chim. Acta 1988, 154, 63. (25) Londergan, C. H.; Kubiak, C. P. J. Phys. Chem. A 2003, 107, 9301. (26) Brunschwig, B. S.; Creutz, C.; Sutin, N. Chem. Soc. ReV. 2002, 31, 168.
