Electrochemistry of the Inclusion Complexes Formed Between the

Publication Date (Web): June 22, 2009 ... Binding Interactions between Cucurbit[n]uril Hosts and Tritopic, Dicationic Guests ... Wei Li and Angel E. K...
0 downloads 3 Views 3MB Size
Download PDF
pubs.acs.org/Langmuir © 2009 American Chemical Society

Electrochemistry of the Inclusion Complexes Formed Between the Cucurbit [7]uril Host and Several Cationic and Neutral Ferrocene Derivatives† Lu Cui, Suresh Gadde, Wei Li, and Angel E. Kaifer* Center for Supramolecular Science and Department of Chemistry, University of Miami, Coral Gables, Florida 33124-0431. Received April 28, 2009. Revised Manuscript Received June 2, 2009 We have investigated the formation of inclusion complexes between the host cucurbit[7]uril (CB7) and three cationic and four neutral ferrocene-containing guests: (ferrocenylmethyl)trimethylammonium (2þ), butyl(ferrocenylmethyl)dimethylammonium (3þ), (ferrocenylmethyl)heptyldimethylammonium (4þ), hydroxymethylferrocene (5), (((methoxy)-ethoxy)ethoxy)methylferrocene (6), 1,10 -di(hydroxymethyl)ferrocene (7), and 1,1’-di((((methoxy)ethoxy)ethoxy) methyl)ferrocene (8). The formation of highly stable inclusion complexes (K > 107 M-1) was verified in all cases using NMR spectroscopic techniques. From cyclic voltammetric experiments, we observed that CB7 complexation of the cationic guests (2þ-4þ) leads to significant anodic shifts on the ferrocene oxidation half-wave potentials, while the measured potential shifts were smaller in the case of the neutral guests (5-8). Encapsulation of all guests resulted in a substantial decrease of the standard rate constant for heterogeneous electron transfer. However, inclusion complexation of the neutral guests led to quasi-reversible voltammetric behavior, in which the anodic peak potential is more sensitive to scan rate than the corresponding cathodic peak potential, suggesting a minor degree of strutural rearrangement in the neutral inclusion complex after oxidation.

Introduction 1-3

The family of cucurbit[n]uril (CBn) hosts has been receiving considerable attention in the past few years, as these molecular container receptors can reach high binding affinities with suitable guests in aqueous solution.4 The formation of inclusion complexes between ferrocene and cucurbit[7]uril (CB7) was first reported by our group5 in 2003. Two years later, in collaboration with other groups, we reported that a cationic ferrocene derivative (guest 2þ in Figure 2) is bound by CB7 with an equilibrium association constant (K) of 4 1012 M-1 in pure water at 25 °C.6 This finding led to the investigation of the inclusion complex formed between a dicationic ferrocene derivative (12þ in Figure 1) and CB7,7 which was shown to exhibit an extremely high K value (1015 M-1), fully comparable to that of the avidin-biotin host-guest pair.8 Our previous work on these complexes has focused on the determination of their structures and thermodynamic stabilities.5-7 We have also described the voltammetric behavior of some water-soluble ferrocene derivatives in the presence of the CB7 host.6 For instance, the half-wave potential (E1/2) for electrochemical oxidation of the ferrocene residue in 2þ shifts ca. 110 mV to more positive values in the presence of 1.0 equiv of the host. † Part of the “Langmuir 25th Year: Molecular and macromolecular selfassemblies” special issue. *[email protected].

(1) Lee, J. W.; Samal, S.; Selvapalam, N.; Kim, H.-J.; Kim, K. Acc. Chem. Res. 2003, 36, 621. (2) Lagona, J.; Mukhopadhyay, P.; Chakrabarti, S.; Isaacs, L. Angew. Chem., Int. Ed. 2005, 44, 4844. (3) Ko, Y. H.; Kim, E.; Hwang, I.; Kim, K. Chem. Commun. 2007, 1305. (4) Liu, S.; Ruspic, C.; Mukhopadhyay, P.; Chakrabarti, S.; Zavalij, P. Y.; Isaacs, L. J. Am. Chem. Soc. 2005, 127, 15959. (5) Ong, W.; Kaifer, A. E. Organometallics 2003, 22, 4181. (6) Jeon, W. S.; Moon, K.; Park, S. H.; Chun, H.; Ko, Y. H.; Lee, J. Y.; Lee, E. S.; Samal, S.; Selvapalam, N.; Rekharsky, M. V.; Sindelar, V.; Sobransingh, D.; Inoue, Y.; Kaifer, A. E.; Kim, K. J. Am. Chem. Soc. 2005, 127, 12984. (7) Rekharsky, M. V.; Mori, T.; Yang, C.; Ko, Y. H.; Selvapalam, N.; Kim, H.; Sobransingh, D.; Kaifer, A. E.; Liu, S.; Isaacs, L.; Chen, W.; Moghaddam, S.; Gilson, M. K.; Kim, K.; Inoue, Y. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 20737. (8) Green, N. M. Biochem. J. 1963, 89, 585.

Langmuir 2009, 25(24), 13763–13769

This means that the one-electron oxidation of ferrocene is thermodynamically hindered by encapsulation inside the CB7 cavity. It also means that the equilibrium association constant of the CB7 complex decreases by 1-2 orders of magnitude upon oxidation of the ferrocene residue.9 In contrast to this, similar voltammetric experiments with the neutral ferrocene derivative, hydroxymethylferrocene (guest 5 in Figure 2), revealed that the E1/2 value experiences a modest anodic shift of only ca. 10 mV upon complexation by CB7.6 The contrast between the CB7-induced, half-wave potential shift in these two cases is striking and suggests that a more detailed investigation is needed. Yuan and Macartney10 have also studied the homogeneous self-exchange electron transfer between the CB7 3 2þ inclusion complex and its one-electron, oxidized form (CB7 3 22þ) and found a slight acceleration in the corresponding rate constant as compared to that between the unbound ferrocene compounds (2þ and 22þ). We are thus quite interested in determining the rate constants for heterogeneous electron transfer for the CB7 inclusion complexes of cationic and neutral ferrocene guests. Finally, a detailed investigation of the electrochemical behavior of CB7ferrocene complexes may be of interest as it may disclose accessible methods to weaken the stability of these complexes. Anticipated improvements in the existing synthetic methods for functionalization of CB hosts11-13 may eventually lead to the preparation of compounds containing multiple CB7 host units, which can serve as avidin analogues for binding to several ferrocene derivatives (acting as biotinylated compounds). Unlike the avidin-biotin host-guest pair, which usually requires harsh (9) Kaifer, A. E.; Gomez-Kaifer, M. Supramolecular Electrochemistry; Wiley-VCH: Weinheim, 1999; Chapter 9. (10) Yuan, L.; Macartney, D. H. J. Phys. Chem. B 2007, 111, 6949. (11) Jon, S. Y.; Selvapalam, N.; Oh, D. H.; Kang, J.-K.; Kim, S.-Y.; Jeon, Y. J.; Lee, J. W.; Kim, K. J. Am. Chem. Soc. 2003, 125, 10186. (12) Jeon, Y. J.; Kim, H.; Jon, S.; Selvapalam, N.; Oh, D. H.; Seo, I.; Park, C.-S.; Jung, S. R.; Koh, D.-S.; Kim, K. J. Am. Chem. Soc. 2004, 126, 15944. (13) Hwang, I.; Baek, K.; Jung, M.; Kim, Y.; Park, K. M.; Lee, D.-W.; Selvapalam, N.; Kim, K. J. Am. Chem. Soc. 2007, 129, 4170.

Published on Web 06/22/2009

DOI: 10.1021/la9015096

13763

Article

Cui et al.

Figure 1. Structures of the dicationic ferrocene guest (12þ) and CB7 host that partner to form an inclusion complex with similar stability to the biotin-avidin complex.

nitrogen inlet Teflon tubing. Nitrogen gas was purified before use and bubbled through the solution to remove dissolved oxygen before the measurements. During the voltammetric measurements, nitrogen flow was maintained above the solution to minimize redissolution of oxygen. The working electrode was polished on a soft, felt surface using an aqueous slurry of alumina (0.05 μm). Before use, the working electrode surface was rinsed extensively with water and sonicated to remove any particulate left from polishing. In some voltammetric experiments, we used a carbon fiber ultramicroelectrode (BAS, 5 μm radius) as the working electrode. NMR spectroscopic experiments were recorded in Bruker Avance instruments (400 or 500 MHz). In some cases, sodium acetate was added to the sample solution and the methyl acetate protons used for calibration in order to measure the concentration of other species in the D2O (Cambridge Isotopes) solution.

Results and Discussion

Figure 2. Structures of the cationic and neutral ferrocene-containing guests surveyed in this work.

conditions for dissociation, the CB7-ferrocene host-guest system may perhaps be weakened under milder conditions, which should facilitate dissociation on demand. For all these reasons, we decided to carry out a detailed investigation of the electrochemical behavior of the CB7 inclusion complexes formed by a series of cationic and neutral ferrocene derivatives (see Figure 2 for structures). Ferrocene-containing guests 2þ-4þ share the same (ferrocenylmethyl)dimethylammonium binding site and differ in the length of the fourth alkyl substituent attached to the positively charged nitrogen. Compounds 5-8 are all neutral and have either one or two identical substituents directly attached to the cyclopentadienyl rings.

Experimental Section Materials. CB7 was prepared as reported by Day and coworkers.14 This host hydrates upon standing, and we found that it is necessary to determine its effective molecular weight before use. This was done by UV-vis spectroscopic titration with a standard aqueous solution of cobaltocenium.5 Guests 2þ-4þ were prepared (as their bromide salts) following procedures previously reported by our group.15 Guest 5 is commercially available, 7 was prepared by reduction of ferrocenedialdehyde,16 and 6 and 8 were prepared according to a literature report.17 Procedures. The electrochemical experiments were recorded with a BAS 100W system. A single-compartment glass cell was used for all voltammetric experiments. Typically, the cell was fitted with a glassy carbon working electrode (0.071 cm2), a Pt or W auxiliary electrode, a Ag/AgCl reference electrode, and (14) Day, A.; Arnold, A. P.; Blanch, R. J.; Snushall, B. J. Org. Chem. 2001, 66, 8094. (15) Isnin, R.; Salam, C.; Kaifer, A. E. J. Org. Chem. 1991, 51, 35. (16) Faux, N.; Razafimahefa, D.; Picart-Goetgheluck, S.; Brocard, J. Tetrahedron: Asymmetry 2005, 16, 1189. (17) Balzani, V.; Becher, J.; Credi, A.; Nielsen, M. B.; Raymo, F. M.; Stoddart, J. F.; Talarico, A. M.; Venturi, M. J. Org. Chem. 2000, 65, 1947.

13764 DOI: 10.1021/la9015096

NMR Spectroscopic Binding Studies with Guests 2þ-4þ. The binding interactions of the cationic ferrocene derivative 2þ with the CB7 host in aqueous solution have already been described by our group.6 The resulting inclusion complex is highly stable (K=41012 M-1 in pure water), and CB7 is believed to include the ferrocenyl unit in its cavity, while the positively charged nitrogen is located close to the center of one of the cavity portals, generating favorable ion-dipole interactions with the carbonyl groups at the cavity opening. We investigated the binding interactions of the other two cationic ferrocene derivatives (3þ and 4þ) with CB7 and found very similar results. For instance, Figure 3 shows 1H NMR spectroscopic data on the interaction of CB7 and butyl ferrocene derivative 3þ. The presence of the host leads to a pronounced upfield shift of the ferrocene protons (1, 10 , and 2), while the signals corresponding to the protons on the butyl chain (a-d) shift slightly downfield. This pattern of complexation-induced shifts reveals that the ferrocenyl unit is included by CB7, while the butyl chain remains outside the cavity. The N-methyl protons (e) shift slightly to higher field, which indicates that they may be closer to the cavity than the a-d protons. The methylene protons (f) that bridge the ferrocenyl unit to the nitrogen shift upfield along with the ferrocenyl protons, providing clear evidence that they are all fully included by CB7. The simultaneous observation of proton signals for the free and CB7-bound guest (Figure 3B) indicates that the chemical exchange between both species is slow in the NMR time scale, in excellent agreement with previous observations6 made with guest 2þ. In fact, the NMR data set shown in Figure 3 suggests that the binding interactions between the butyl derivative 3þ and CB7 are essentially identical to those previously observed between 2þ and CB7. Further lengthening of the aliphatic substituent (guest 4þ) has no significant effect on the NMR spectroscopic data (see Supporting Information, Figure S1). We conclude that the binding interactions of all three cationic derivatives with host CB7 are basically identical, giving rise to highly stable complexes in which the ferrocenylmethyl unit is engulfed inside the CB7 cavity. The extremely high association equilibrium constant of the complex formed between guest 2þ and CB7 in pure water and the strong similarities of the NMR spectroscopic data obtained in binding studies between the host and all the ferrocene-containing cationic guests clearly imply that the K values in 0.1 M NaCl for guests 2þ-4þ are more than high enough to guarantee quantitative complexation of the guest upon mixing with 1.0 equiv of the host at millimolar concentration levels. Voltammetric Investigation of the CB7 Complexes of Guests 2þ-4þ. As anticipated, the cyclic voltammetric (CV) Langmuir 2009, 25(24), 13763–13769

Cui et al.

Figure 3. 1H NMR spectra (400 MHz, 0.1 M NaCl/D2O) for 1.0 mM 3þ (A) in the absence and in the presence of (B) 0.5 equiv and (C) 1.0 equiv CB7. The symbol *denotes the signal for the acetate protons.

behavior of guest 3þ and 4þ is dominated by the one-electron oxidation of the ferrocene residue. This electrochemical process is known to be extremely fast, leading to reversible behavior in the CV experiments (Figure 4). Addition of 1.1 equiv of CB7 to the 3þ solution fully converts the guest into its CB7 inclusion complex, which exhibits a more positive half-wave potential for ferrocene oxidation and decreased current levels. This behavior is identical to our reported observations with guest 2þ in the presence of CB7.6 The CB7 complex is obviously bulkier than the free guest, and its diffusion to the electrode surface is slower, thus yielding smaller currents. The anodic shift in the E1/2 value reflects the differential stabilization by CB7 of the reduced (ferrocene) form of the guest as compared to its oxidized (ferrocenium) form. It also means that the stability of the inclusion complex is somewhat decreased by oxidation of the ferrocene residue. While the CV behavior of the free guests is clearly reversible (Figure 4), as evidenced by the potential difference between the anodic and cathodic peaks (ΔEp ∼60 mV), the CV response of the CB7 3 3þ and CB7 3 4þ complexes is clearly quasi-reversible (ΔEp ∼90 mV) at the relatively slow scan rate used (100 mV s-1). As a result of this observation, we decided to measure the standard rate constants (ko) for the heterogeneous electron transfer reactions of the CB7 complexes. In order to address this matter, we started with the simplest cationic guest, 2þ, and carried out a scan rate study of its cyclic voltammetric response. In agreement with previous findings, the free guest shows perfectly reversible behavior as evidenced by the invariance of the anodic and cathodic peak potentials as the scan rate increases from 50 mV s-1 to 1.0 V s-1 (Figure 5A). In clear contrast with this behavior, the CB7 3 2þ complex shows quasi-reversible behavior, as the ΔEp peak-to-peak splitting increases with the scan rate. Clearly, these data indicate that encapsulation of the ferrocenyl unit in the CB7 cavity leads to a measurable attenuation of the electrochemical kinetics, that is, a decrease in the ko value. Nicholson’s method18 can be utilized to measure the ko values using data sets of the type shown in Figure 5. However, the diffusion coefficients (Do) for the electroactive species involved must be independently estimated. We determined the Do values using two separate techniques: (1) voltammetric measurements with ultramicroelectrodes (UMEs) and (2) pulse gradient stimulated (18) Nicholson, R. S. Anal. Chem. 1965, 37, 1351.

Langmuir 2009, 25(24), 13763–13769

Article

echo (PGSE) NMR measurements.19 We have shown before that both techniques correlate well with one another, provided that viscosity corrections (to take into account viscosity differences in the two types of solutions used in these measurements) are properly applied.20 Representative UME voltammograms are shown in the Supporting Information (Figures S2 and S3) and the measured diffusion coefficient values are given in Table 1. The values in Table 1 show all the expected trends. For instance, the Do values decrease as the length of the aliphatic N-substituent increases, although this effect is more pronounced in going from methyl to butyl (2þ to 3þ) than from butyl to heptyl (3þ to 4þ). The CB7 complexes exhibit values considerably smaller than the free guests, reflecting the larger molecular volume and hydrodynamic radii of the former. The partial encapsulation of the guest by the CB7 host is also evidenced by the smaller dispersion of Do values observed in the CB7 complexes as compared to the values measured with the free guests. Finally, the relative proximity of the values in the fourth column of the table to the experimentally determined ratio of viscosities [η(NMR)/η(UME) = 1.18] for the solutions used in both types of experiments provides further validation to the internal consistency of the data set. Nicholson evaluated numerically18 the relationship between the dimensionless function Ψ and the potential difference between the anodic and cathodic peak potentials (ΔEp). Therefore, from the experimentally determined ΔEp value at a given scan rate (v), one can obtain the corresponding Ψ value and compute the standard rate constant ko using eq 1 Ψ ¼

pffiffiffiffiffiffiffiffi ðDo =DR ÞR=2 k° RT ðπnFDo vÞ1=2

ð1Þ

where Do and DR stand for the diffusion coefficients of the oxidized and reduced species, R is the charge transfer coefficient, and the remaining terms have their usual meanings. As is rather customary, we took Do =DR and R=0.5 and used the diffusion coefficients determined in the steady-state voltammetric experiments to determine the ko values. The resulting values are listed in Table 2. The ko values in Table 2 confirm that encapsulation of the ferrocene centers leads to a pronounced attenuation of electrochemical kinetics in this system. Clearly, inclusion inside CB7 decreases the measured value of ko by a factor of 20 or more. This finding can be rationalized as the result of inclusion complexation keeping the ferrocene center further away from the electrode surface, due to the presence of a physical barrier between the two. In other words, the electroactive species diffuses toward the electrode surface and typically undergoes electron transfer from or near the so-called Outer Helmholtz Plane (OHP). The increase in the hydrodynamic radius of the ferrocene center upon encapsulation is thus equivalent to a displacement of the OHP away from the electrode surface, which is thought to be the key factor leading to the decreased rate for the heterogeneous electron transfer process. NMR Spectroscopic Binding Studies with Neutral Guests 5-8. The 1H NMR spectrum of hydroxymethylferrocene, guest 5, in 0.1 M NaCl-D2O solution exhibits three resonances in the chemical shift region 4.2-4.4 ppm (see Figure 6A). In the presence of 1.0 equiv CB7, all the resonances broaden and shift upfield to ca. 3.5 ppm (Figure 6). The considerable CB7-induced upfield shift is consistent with inclusion binding of this guest by the CB7 host. (19) Cohen, Y.; Avram, L.; Frish, L. Angew. Chem., Int. Ed. 2005, 44, 520. (20) Sun, H.; Chen, W.; Kaifer, A. E. Organometallics 2006, 25, 1828.

DOI: 10.1021/la9015096

13765

Article

Cui et al.

Figure 4. CB7 effect on the CV response on glassy carbon (0.071 cm2) of 1.0 mM ferrocenyl guest in 0.1 M NaCl. Scan rate: 0.1 V s-1. (A) 1.0 mM guest 3þ in the absence (black) and in the presence of 1.1 equiv CB7 (magenta). (B) 1.0 mM guest 4þ in the absence (black) and in the presence of 1.1 equiv CB7 (blue).

Figure 5. Scan rate dependence of the CV response on glassy carbon (0.071 cm2) of (A) free 2þ and (B) CB7 3 2þ complex in 0.1 M NaCl. Scan

rates: 0.05 (black), 0.1 (red), 0.2 (green), 0.5 (blue), and 1.0 (purple) V s-1. Table 1. Diffusion Coefficients (cm2 s-1  106) Measured Using UME Steady-State Voltammetry and PGSE NMR Spectroscopic Techniques at 25 °C sample

Do (UME)a

Do (NMR)b

Do (UME)/ Do (NMR)c

2þ 7.99 ( 0.51 6.76 ( 0.10 1.17 6.82 ( 0.43 5.60 ( 0.20 1.21 3þ 6.35 ( 0.19 5.27 ( 0.12 1.19 4þ þ 4.01 ( 0.02 3.29 ( 0.04 1.21 CB7 3 2 þ 3.56 ( 0.04 2.86 ( 0.01 1.23 CB7 3 3 3.42 ( 0.12 2.80 ( 0.12 1.21 CB7 3 4þ a Measured in 0.1 M NaCl from the voltammetric steady state current i=4nFDoCr, where n=1 and r=5 μm. b Measured in 0.1 M NaCl/D2O. c Assuming identical hydrodynamic radii in both media, Do(UME)/ Dο(NMR)=η(NMR)/η(UME). Experimentally, the solution viscosity ratio was found to be 1.18 at 25 °C.

13766 DOI: 10.1021/la9015096

Compound 6, with a longer chain covalently attached to the cyclopentadienyl ring of the ferrocene unit, behaves similarly (see Supporting Information, Figure S4) although the resonances are less broadened than in the case of guest 5. Similar NMR studies with biarmed guests 7 and 8 (Supporting Information Figures S5 and S6) allow us to conclude that the main binding site is the ferrocenyl residue in all cases. However, the known K value for the CB7 3 5 complex in pure water is 3 orders of magnitude lower than that reported for CB7 3 2þ, and further substitution of the ferrocene center;along with the presence of 0.1 M NaCl21 in the solution;may lower the K values for CB7 complexation of guests 6-8 substantially. Therefore, we felt that it was necessary to determine the K values with the neutral guests in order to assess (21) Ong, W.; Kaifer, A. E. J. Org. Chem. 2004, 69, 1383.

Langmuir 2009, 25(24), 13763–13769

Cui et al.

Article

Table 2. Standard Rate Constants for Heterogeneous Electron Transfer (ko in cm s-1) and Half-Wave Potentials (E1/2 in V vs Ag/ AgCl) Measured in 0.1 M NaCl at 25 °C ko

redox couple 2þ/22þ 3þ/32þ 4þ/42þ CB7 3 2þ/CB7 3 22þ CB7 3 3þ/CB7 3 32þ CB7 3 4þ/CB7 3 42þ a Reversible behavior was 57-62 mV.

E1/2

>0.8a 0.432 ( 0.002 >0.8a 0.437 ( 0.008 a >0.8 0.440 ( 0.002 0.041 ( 0.005 0.542 ( 0.006 0.012 ( 0.002 0.527 ( 0.006 0.022 ( 0.003 0.523 ( 0.002 observed, with ΔEp values in the range

Figure 6. 1H NMR spectra (500 MHz, 0.1 M NaCl/D2O) of 2.0 mM 5 (a) in the absence and in the presence of (b) 0.5 equiv and (c) 1.0 equiv CB7.

whether simple mixing of 1.0 equiv of guest and 1.0 equiv of host at millimolar concentration levels would guarantee quantitative complex formation in the electrochemical experiments. To determine the K values, we resorted to competition experiments with the guest cobaltocenium5 (Cobþ). The single proton signal of this guest resonates at chemical shifts between 5.58 ppm (free) to 4.97 ppm (CB7-bound) and, since this complex typically exhibits fast exchange in the NMR time scale,5 the observed chemical shift, when both free and CB7-bound Cobþ coexist, is given by the following equation δobs ¼ 5:58Xf þ4:91Xc

ð2Þ

where Xf and Xc are the molar fractions of free and complexed guest, respectively. Therefore, if we mix known concentrations of Cobþ and one of the neutral ferrocene guests 5-8 with a concentration of CB7 insufficient to fully bind both guests, the observed chemical shift for the Cobþ protons allows us to determine Xf and Xc, since Xf þ Xc=1, and thus the respective concentrations. Assuming that the concentration of free CB7 in equilibrium is negligibly small, we can calculate the concentrations of free and bound ferrocenyl guest, which we can use to express the ratio of the two relevant equilibrium association constants. For instance, for guest 5 we can write ½CB7 3 5½Cobþ  K5 ¼ KCobþ ½5½CB7 3 Cobþ 

ð3Þ

where all the concentrations on the right side are at equilibrium. Using this straightforward procedure, we measured the relative K values for CB7 binding of guests 5-8 using KCobþ as a common Langmuir 2009, 25(24), 13763–13769

reference. We obtained 0.32, 0.24, 0.13, and 0.063 for guests 5, 6, 7, and 8, respectively. We have previously measured the equilibrium association constant between Cobþ and CB7 as 5.7  109 M-1 in 50 mM sodium acetate at 25 °C.22 In the medium used in this work (0.1 M NaCl), the corresponding K value should be smaller by a factor of 2-3.21 This means that the K values for guests 5-8 must all clearly exceed 107 M-1, which guarantees quantitative complexation when mixing equal host and guest concentrations at millimolar levels. Computational Investigation of the CB7 3 7 Complex. The CB7 inclusion complexation of biarmed guests 7 and 8 is interesting, as it offers two distinct structural possibilities for the resulting host-guest complex. Basically, the two ferrocenyl side arms can protrude out of the host cavity through the same portal (syn conformation) or through different portals (anti conformation). In principle, one may anticipate that these two structures can be differentiated by NMR spectroscopy, as the doublet corresponding to the inner methylene bridge protons of CB7, may be sensitive to the two different chemical environments in the anti conformation. However, our own NMR spectroscopic data with the CB7 3 2þ, CB7 3 3þ, and CB7 3 4þ complexes show that this is not always the case. In all of these complexes, the equatorial plane of symmetry of the CB7 host is lost, and this lack of symmetry would be expected to yield nonequivalent inner methylene protons on the two sides of the cavity. While this is observed in the 1H NMR spectrum of the first complex (CB7 3 2þ), it is not observed with the other two cationic complexes, for reasons that are unclear at this point. However, these experimental NMR data prompted us to look for other methods to assess the relative stabilities of the two forms (syn/anti) of the inclusion complexes of biarmed guests 7 and 8. We decided to assess the stability of the syn and anti forms of the CB7 3 7 complex using computational density functional theoretical (DFT) methods. The flexible side arms of guest 8 would create unnecessary complexity in these calculations, which justifies our exclusive computational focus on the CB7 3 7 complex. First, we optimized the syn and anti forms of CB7 3 7 complexes at the B3LYP/STO-3G* level in the gas phase. We manually varied the starting positions of the host-guest complexes and run full energy minimizations from each starting position in order to locate the most stable complexes in both forms. After reaching the global minima of the complexes, we performed single point energy calculations, utilizing the same basic conditions, but now in water to more accurately determine the binding energies. The completely optimized geometries of the CB7 3 7 complexes (syn and anti) are shown in Figure 7. In the gas phase, as well as in water, compound 7 exists in the syn form rather than in the anti form, the former being more stable than the latter by ∼3 kcal mol-1. These results can be explained by possible interactions between the methanolic OH groups in free 7. Furthermore, the high polarity of water probably favors the more polar syn conformer. In contrast, the anti form of the CB7 3 7 complex is more stable than the corresponding syn form. This is mainly due to the possible steric problems arising between the CB7 carbonyl groups and the ferrocenyl methanolic OH groups, as both side arms are projecting out of the host cavity through the same portal in the syn form. The distortion of CB7 is more pronounced in the case of the syn form, as the host must accommodate both side arms on the same portal, leading to a complex with higher energy content. Also, the nonpolar character of the host cavity also favors the less polar anti conformer. This is evidenced in the binding energies, as the formation of the CB7 inclusion complex from the anti form of the guest (22) Sobransingh, D.; Kaifer, A. E. Langmuir 2006, 22, 10540.

DOI: 10.1021/la9015096

13767

Article

Cui et al.

Figure 8. CB7 effect on the CV response on glassy carbon (0.071 cm2) of 1.0 mM 6 in 0.1 M NaCl solution (black) in the absence and in the presence of (red) 0.5 and (green) 1.0 equiv of CB7. Scan rate= 0.1 V s-1.

Figure 7. Side and top views of the anti (top) and syn (bottom) conformer minimized structures for the CB7 3 7 complex (B3LYP/ STO-3G*). Table 3. Electrochemical Data for Guests 5-8 and Their CB7 Complexes in 0.1 M NaCl at 25 °C compound 5 6 7 8 CB7 3 5 CB7 3 6 CB7 3 7 CB7 3 8

E1/2 (V vs Ag/AgCl)

Do (cm2 s-1)

ko (cm s-1)

0.290 ( 0.004 0.315 ( 0.006 0.329 ( 0.008 0.376 ( 0.004 0.308 ( 0.002 0.311 ( 0.008 0.361 ( 0.006 0.401 ( 0.002

(7.81 ( 0.29)  10-6 (5.41 ( 0.19)  10-6 (6.65 ( 0.04)  10-6 (3.70 ( 0.28)  10-6 (3.65 ( 0.17)  10-6 (3.47 ( 0.14)  10-6 (3.61 ( 0.14)  10-6 (1.90 ( 0.16)  10-6

>0.8 >0.8 >0.8 >0.8 0.033 ( 0.002 0.012 ( 0.002 0.012 ( 0.002 0.008 ( 0.002

releases -2.0 kcal mol-1, whereas that from the syn form would require the absorption of 4.1 kcal mol-1. A full set of computational results is given in the Supporting Information (Table S1). Voltammetric investigation of the CB7 Complexes of Guests 5-8. As anticipated, the voltammetric behavior of all the neutral ferrocene compounds is fully reversible in the range of scan rates investigated. The corresponding half-wave potentials for the oxidation of the ferrocene residues are given in Table 3. The presence of CB7 in the solution depresses the current levels observed for the oxidation of the ferrocene derivative, which is a clear indication of inclusion binding by the host, since the inclusion complex is larger in molecular weight and hydrodynamic radius than the free guest, leading to decreased diffusivity. The half-wave potential for the oxidation of the ferrocene residue is only slightly affected by the presence of CB7, in analogy to what we already reported with compound 5.6 This is in strong contrast to the cationic ferrocene guests, in which CB7 binding leads to a substantial increase of the E1/2 value. As an illustrative example, Figure 8 shows voltammetric data for guest 6. We also investigated the electrochemical kinetics of the CB7 complexes of guests 5-8 following the same method already described for the complexes of the cationic guests. The corresponding data are collected in Table 3. One clear trend evident from the E1/2 values is that the formation of CB7 inclusion complexes has only a relatively minor effect on the half-wave potential for oxidation of the ferrocene residue. The maximum CB7-induced E1/2 shift is 32 mV (observed with guest 7). In contrast to this small effect on the half-wave potentials, CB7 binding exerts a more pronounced effect on the electrochemical 13768 DOI: 10.1021/la9015096

Figure 9. Scan rate dependence of the anodic and cathodic peak potentials measured for (A) 1.0 mM CB7 3 2þ and (B) 1.0 mM CB7 3 6 in 0.1 M NaCl.

kinetics for the ferrocene oxidation process. Compared to the free guests, all ko values are substantially lower in the CB7 complexes, which exhibit voltammetric behavior in the quasi-reversible regime, while the free guests are fully reversible within the range of scan rates surveyed. The range of ko values exhibited by the neutral complexes is similar to that measured with the cationic complexes. In spite of the similarities in the recorded ko values for the cationic and neutral CB7 complexes, we must note an important difference in their quasi-reversible voltammetric behavior. Figure 9 shows the dependence of the peak potentials on the voltammetric scan rate for the CB7 complexes of guest 2þ and 6. In both cases, the peak-to-peak separation increases with scan rate. However, for CB7 3 2þ both peaks depart from their “reversible” positions (measured at slow scan rate) at similar rates. In contrast to this, for CB7 3 6 the anodic peak clearly moves faster than the cathodic peak. The reasons behind these differences are not entirely clear, but it seems reasonable to conclude that the oxidation of the ferrocene nucleus in CB7 3 6 may result in some minor degree of structural reorganization of the inclusion complex. The scan rate dependence of the peak potentials for all four neutral complexes of CB7 is similar, suggesting that the postulated structural change may be related to the ferrocene/ ferrocenium core, not to any of the substituents. It is interesting to reiterate here that, upon complexation with CB7, the E1/2 values for oxidation of guests 2þ-4þ are shifted anodically by ca. 100 mV, while those for guests 5-8 are shifted only by 32 mV or less. Langmuir 2009, 25(24), 13763–13769

Cui et al.

The cationic guests are anchored relative to the CB7 cavity by the presence of the positively charged nitrogen on the side arm, which is tightly bound to one of the carbonyl-laced cavity portals. Thus, the generation of a positive charge in the ferrocene residue, located in the middle of the host cavity, is thermodynamically unfavorable and leads to a decrease in the stability of the inclusion complex. The situation is quite different in the complexes between CB7 and uncharged guests 5-8. Here, the main driving force for complex formation results from the good fit between the ferrocene residue and the host cavity coupled to the strong hydrophobic interactions between them.7 Electrochemical oxidation of the guest creates a positive charge in the ferrocene center, which is more ‘mobile’ than that generated by oxidation of the cationic guests, since there are no additional locking ion-dipole interactions that may force the ferrocene to stay in the exact cavity center. We propose that oxidation of the neutral guest may give rise to a small sliding movement of the ferrocene toward one of the two carbonyl-laced openings of the cavity, which may explain the differences in the relative displacement of the anodic and cathodic peak potentials as a function of scan rate. A final issue that we must address here is the attenuation of the electrochemical kinetic rates upon complexation by CB7, which are observed in this work, compared to the slight increase observed by Macartney and co-workers on the kinetic rates for the homogeneous self-exchange electron transfer reaction10 CB7 3 2þ þ CB7 3 22þ h CB7 3 22þ þ CB7 3 2þ. While a detailed assessment of reorganization energies still needs to be carried out, the self-exchange reaction requires bringing together two positively charged reactants and the corresponding work function is substantially decreased by CB7 complexation, as the distance of maximum approach between the two charged reactants increases

Langmuir 2009, 25(24), 13763–13769

Article

due to the presence of the encapsulating CB7 hosts. This appears to be the overriding factor, which predominates over the intrinsic distance effect on the electron transfer. In the electrochemical reactions, the work function may not be so strongly affected by CB7 complexation, but its effect on the distance over which the electron transfer must take place is still important (displacement of the OHP further away from the electrode surface). Thus, qualitatively, it may be possible that complexation by CB7 may slightly increase the rate of homogeneous self-exchange electron transfer, while decreasing the rate of the electrochemical processes. In conclusion, we have shown that the cationic guests 2þ-4þ and the neutral guests 5-8 all form very stable inclusion complexes with the host CB7. The half-wave potential for ferrocene oxidation in the complexes is shifted to more positive values in all cases, but this effect is much more pronounced in the cationic complexes. Finally, the standard rate constants for heterogeneous electron transfer decrease considerably upon encapsulation inside CB7, in such a way that the voltammetric behavior of all the ferrocene guests goes from perfectly reversible in their free, unbound state to the quasi-reversible regime in the CB7-bound state. Acknowledgment. The authors are grateful to the National Science Foundation for the generous support of this work (to AEK, CHE-0600795 and CHE-0848637). AEK gratefully acknowledges an excellent suggestion from Prof. Christian Amatore. Supporting Information Available: Additional NMR spectroscopic, computational and cyclic voltammetric data, as mentioned in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

DOI: 10.1021/la9015096

13769