Strong Inclusion of Inorganic Anions into β-Cyclodextrin Immobilized

Jul 5, 2011 - ... of Utah, 383 Colorow Road, Salt Lake City, Utah 84112-0850, United States ... Quantitative analysis of the XPS data suggested a 1:1 ...
0 downloads 0 Views 1MB Size
Download PDF
ARTICLE pubs.acs.org/Langmuir

Strong Inclusion of Inorganic Anions into β-Cyclodextrin Immobilized to Gold Electrode Yasuhiro Domi,† Kentaro Ikeura,† Kazumasa Okamura,† and Katsuaki Shimazu*,†,‡ †

Division of Environmental Materials Science, Graduate School of Environmental Science, and ‡Section of Materials Science, Faculty of Environmental Earth Science, Hokkaido University, Sapporo 060-0810, Japan

Marc D. Porter Departments of Chemistry, Chemical Engineering, and Bioengineering, University of Utah, 383 Colorow Road, Salt Lake City, Utah 84112-0850, United States

bS Supporting Information ABSTRACT: The inclusion of inorganic anions such as SO42, NO3, and HPO42 into the cavity of β-cyclodextrin monolayers on Au was examined by X-ray photoelectron spectroscopy (XPS), a quartz crystal microbalance (QCM), and chronocoulometric measurements of the competitive inclusion with ferrocene. The inclusion amounts of ferrocence in 0.2 M Na2SO4, NaNO3, and Na2HPO4 solutions were less than 6% of the adsorption amount of β-cyclodextrin on Au, resulting in the apparent inhibition of the ferrocene redox reaction. The surface association constants of these anions reached about 10 on a logarithmic scale and were much higher than those for the inclusion of common organic guest compounds. A stronger anion inclusion was also demonstrated by the QCM response corresponding to the replacement of a preincluded organic guest with sulfate upon the injection of the sulfate solution. Quantitative analysis of the XPS data suggested a 1:1 association for each of these anions per surface β-cyclodextrin. There was no detectable inclusion for ClO4, Cl, and Br.

1. INTRODUCTION Cyclodextrins (CDs) are toroidally shaped oligosaccharides with nanometer-size intramolecular cavities.1 Because of their unique capability to include various guests compounds, host guest chemistry of CDs has been widely examined but still receives extensive attention due to growing interest in noncovalent interactions in supramolecular chemistry,2 biomimetic systems,3 and related fields.4 In an aqueous solution, the CDs can include neutral and hydrophobic molecules that are smaller than the cavity due to van der Waals and hydrophobic interactions between a guest and the CD interior. The association constants (log Kguest,sol) in such cases5 typically range from 2 to 5 (e.g., 2.13.3,6,7 2.33.0,7,8 and 3.24.6,9 for benzoic acid, p-nitrophenol, and 1-adamantanecarboxylate, respectively, for the association with β-CD). Based on such a strong molecular recognition ability, it is commonly recognized that the CDs can be used as chemical sensors,10a,11 carriers in drug delivery systems,2b,12 modifiers of stationary phases of chromatographic columns,13 molecular machines,14 removal of pollutants,15 additions (stabilizers) for foods,16 and deliverers of perfumes for cosmetics.17 Recently, thiol-derivatized CDs, immobilized as self-assembled monolayers (SAMs) on solid surfaces, have received increased attention as potential analyte-sensitive sensors and reactantselective catalysts.10 To design and optimize the function of these devices, the knowledge of the association constants at the r 2011 American Chemical Society

surface is essential because surface-immobilized molecules often show different physicochemical properties from those in bulk solutions due to substrate/adsorbate and adsorbate/adsorbate interactions. For the most widely studied acid dissociation properties, for example, the carboxy-terminated alkanethiol monolayers showed low dissociation constants, or high pKa, compared to the corresponding values in a bulk phase due to a lateral repulsive interaction or smaller dielectric environment at the surface than in the bulk solution.18,19 In addition, strong ionic association between the dissociated carboxy and cation that was not observed in a bulk solution took place at the surface.19c For the hostguest association, the surface provides a different environment from the bulk solution, resulting in a lower or higher association constant than the corresponding bulk values.2026 Several studies have been conducted regarding the association between CDs and guests at the surface. For small guest compounds, the size of which is small enough to be included within the CD cavity, the association constant was generally higher at the surface than in the solution. This phenomenon was interpreted in terms of an entropic effect because there is no significant difference in the enthalpy upon the association between the SAM and free CD in Received: December 27, 2010 Revised: June 26, 2011 Published: July 05, 2011 10580

dx.doi.org/10.1021/la1051063 | Langmuir 2011, 27, 10580–10586

Langmuir solution.20 For large guest compounds, chemical interaction with the electrode and steric obstruction affect the association constant at the surface, so that the association constant becomes either higher or lower than that in the solution.20 While the inclusion properties of these surface-bound systems have been examined for several hydrophobic compounds,2026 very little attention has been paid to those for inorganic anions. This is probably due to the belief that these anions have much lower association constants than those for organic compounds even at the surface. In an aqueous solution, small inorganic anions are actually more weakly included (e.g., log Kanion,sol < 1.0 for Cl, Br , NO3 , and SO 4 227 and log Kanion,sol = 1.11.5 for ClO42830). The surface association constant (log Kanion,surf) is known only for the inclusion of NO3 into a per-2,3-dimethyl-6thio-β-CD SAM on a Hg electrode, which exhibited electrode potential dependent values of log Kanion,surf from 0.5 to 1.2.31 In our recent paper,21 mixed per-6-thio-β-cyclodextrin (CD-SH)/ pentanethiol (C5SH) monolayers were constructed using a newly proposed procedure, in which intermolecular vacancies between the surface-immobilized CD-SH molecules are entirely filled with C5SH, while no CD cavity is occupied by C5SH. Therefore, the CD cavities are the only accessible sites for guest compounds.32 This is a key point because, if the intermolecular vacancies exist, it is hard to strictly distinguish at which sites the interaction with the guest compounds and reactants takes place, the CD cavities or intermolecular vacancies, so that accurate evaluation of the phenomena occurring only at the CD cavities becomes difficult. During the study of the redox of metal complexes on our mixed CD-SH/C5SH monolayers, we found for the first time that the redox reaction of cavity-accessible ferrocene is inhibited in the presence of some electrolyte anions (i.e., SO42, NO3, and HPO42) and that these small inorganic anions are strongly included into the CD cavities of the CD-SH/C5SH monolayers, as described in this paper. Evidence in support of the inclusion of these anions, which can rival that of some hydrophobic compounds (e.g., ferrocene), is presented using cyclic voltammetry, chronocoulometry, X-ray photoelectron spectroscopy (XPS), and quartz crystal microbalance (QCM) measurements.

2. EXPERIMENTAL SECTION 2.1. Materials and Substrate. Heptakis(6-deoxy-6-thio)-β-CD (per-6-thio-β-CD, CD-SH) was synthesized according to the previously reported procedure.21 All other chemicals were of reagent or higher grade. Ferrocene, supplied from Tokyo Chemical Industry, was purified by sublimation prior to use. An aqueous ferrocene solution was prepared by injecting a 20 mM ethanolic solution of ferrocene into water. All aqueous solutions were prepared with Milli-Q water. Gold thin film electrodes (thickness: 200 nm) were prepared using either a Ti-coated glass slide or QCM quartz wafer by vacuum evaporation. After flame-annealing, these electrodes showed the typical characteristics of Au(111) for the electrochemical formation/reduction of the surface oxide in 0.1 M H2SO4. 2.2. Preparation of the Mixed CD-SH/C5SH Monolayer. The mixed CD-SH/pentanethiol(C5SH) monolayers were constructed by successively immersing the gold electrode in 1.0 mM CD-SH in DMSO/ H2O (3:2, v/v) solvent for more than 72 h, an aqueous 60 μM ferrocene (Fc) solution over 1 h, and a 2.5 mM Fc + 0.5 mM C5SH solution in an EtOH/H2O (1:1, v/v) solution overnight. The electrode was finally immersed in ethanol to displace ferrocene from the CD cavity. 2.3. Electrochemistry. The cyclic voltammetric and chronocoulometric measurements were performed in a three-electrode cell using a

ARTICLE

computer-controlled polarization system. The working electrode was a mixed CD-SH/C5SH monolayer unless otherwise stated. The reference and counter electrodes were a Ag/AgCl (saturated KCl) electrode and a platinized platinum foil, respectively. All potentials in the text are referred to this electrode. All electrochemical measurements were carried out in deaerated solutions in an Ar filled glovebox.

2.4. Chronocoulometric Measurements of Inclusion Amounts of Ferrocene and Anions. The inclusion amount of ferrocene in the cavities of CD-SH/C5SH monolayers was determined in the presence of ferrocene in the solution. In this case, the electrochemical response becomes the sum of the contributions both from the ferrocene in solution and ferrocene included in the CD cavity. To exclude the contribution of the solution ferrocene, chronocoulometric measurements were conducted at various potentials and concentrations of ferrocene.21 The charge passed during the potential step obeys the following well-known equation. 1=2 

Q ¼

2nFADFc CFc t 1=2 þ Qdl þ nFAΓFc π1=2

ð1Þ

ΓFc represents the amount of ferrocene included in the CD cavities, and the other symbols have their usual meanings. The electrode potential was stepped from 0.10 to 0.30 V and more positive potentials. The intercept 1 at t = 0 in the Q vs t /2 plot corresponds to the sum of the second and third terms of the eq 1. Because only the double layer charge, Qdl, of these two terms is dependent on the potential difference between the initial and stepped potentials, the extrapolation to the initial potential in the intercept versus the stepped potential gives ΓFc. A series of experiments using the same electrode were conducted in solutions of various ferrocene concentrations to obtain the concentration dependence of the inclusion amount (isotherm). To determine the inclusion amount of anions, ΓFc was also measured using the above-described procedures but in solutions containing SO42, NO3, or HPO42 of various concentrations. Prior to the measurements, the maximum ferrocene surface concentration, ΓFc,max, was determined in 0.2 M NaClO4 + 60 μM ferrocence and was used for the following calculation. When the anion is more strongly included than ferrocene, the inclusion amount of ferrocene should decrease in the presence of the anions. Thus, the amount of included anion, Γanion,C, at a given anion solution concentration, C, is given by the equation Γanion,C = ΓFc,max  ΓFc,C, where ΓFc,C is the ferrocene surface concentration at anion concentration C. 2.5. X-ray Photoelectron Spectroscopy (XPS). X-ray photoelectron spectra were obtained using a Rigakudenki model XPS-7000 X-ray photoelectron spectrometer with monochromic Mg KR radiation at 25300 W. The takeoff angle was 90°. The Au 4f7/2 emission was used as the internal reference to determine the binding energies of the elements. 2.6. Quartz Crystal Microbalance. The 5 MHz and AT-cut quartz crystal, one side of which was coated with the mixed CD-SH/ C5SH monolayer, was mounted in a MAXTEK model TPS-550 QCM sensor head. The resonance frequency was measured using an Agilent Technologies model 53131A frequency counter. QCM data were analyzed based on Sauerbrey’s equation, or using the sensitivity of 17.7  109 g cm2 Hz1.

3. RESULTS AND DISCUSSION Mixed CD-SH/C5SH monolayers, prepared according to our past procedures on Au(111) electrodes,21 were used throughout the present study. The average surface concentration for the CD-SH portion of the adlayer was 69.8 ( 4.9 pmol cm2 on a true surface area basis and corresponds to 86% of that expected for a hexagonally close packed monolayer. This packing density is slightly below our earlier reported value of 74.0 ( 6.3 pmol cm2,21 10581

dx.doi.org/10.1021/la1051063 |Langmuir 2011, 27, 10580–10586

Langmuir

ARTICLE

Figure 2. Cyclic voltammograms for the redox of Fc on the mixed CD-SH/C5SH monolayer in 60 μM Fc + 0.1 M NaClO4 + Na2SO4. The sulfate concentrations are shown in the figure. Sweep rate: 0.1 V s1.

Figure 1. Cyclic voltammograms for the redox of Fc on (a) Au and (b) the mixed CD-SH/C5SH monolayer in 60 μM Fc + 0.2 M electrolyte solutions shown in the figure. Sweep rate: 0.1 V s1.

Figure 3. Concentration dependence of the inclusion amount of Fc on the mixed CD-SH/C5SH monolayer in various 0.2 M electrolyte solutions shown in the figure. The inclusion amount was determined by chronocoulometry according to the previously reported procedure.21

but slightly higher than that found by others (4755 pmol cm2).22 Importantly, our earlier preparative procedure effectively fills the intermolecular vacancies between adsorbed CD-SH molecules with C5SH after first protecting the CD cavity with included ferrocene. As such, the interior of the CD cavity is the only accessible site for guest compound inclusion.21 Preliminary evidence for inorganic anion inclusion in our CDSH/C5SH monolayer is shown by the dependence of the cyclic voltammetric response for 60 μM ferrocene in various 0.2 M aqueous electrolytes (i.e., NaClO4, Na2SO4, NaNO3, Na2HPO4, NaCl, and NaBr). Nearly identical voltammograms were obtained for uncoated Au(111) in each of these solutions (Figure 1a) and at Au(111) modified with the CD-SH/C5SH monolayer with NaClO4, NaCl, and NaBr as electrolytes (Figure 1b). Interestingly, the responses for ferrocene in Na2SO4, NaNO3, and Na2HPO4 were notably lower (Figure 1b). Moreover, the current flow for ferrocene decreased with increasing SO42 concentration in 0.1 M NaClO4 (Figure 2). These results show that the SO42, NO3, and HPO42 have inhibiting effects on the redox transformation of ferrocene at Au(111) coated with our CD-SH/C5SH monolayer. Based on these findings, we hypothesized that inhibition is due to the association of SO42, NO3, and HPO42 with the CD-SH/C5SH monolayer. That is, the inclusion of these anions limits access of ferrocene to the CD cavity. As a first test of this hypothesis, the amount of included ferrocene, ΓFc, was estimated using previously described chronocoulometric measurements21 at varied solution concentrations of ferrocene and different values of applied potential. The results of the concentration dependence studies are shown in Figure 3. As evident, the six anions can be categorized into two groups: those that strongly associate with

the CD-SH/C5SH adlayer (i.e., SO42, NO3, and HPO42) and those that weakly interact with the CD-SH/C5SH coating (i.e., ClO4, Cl, and Br). In 0.2 M ClO4, Cl, and Br, ΓFc increases with the solution concentration of ferrocene, approaching a limiting level at ∼60 μM. As summarized in Table 1, the average maximum values for ΓFc, ΓFc,max, determined from several experiments via Langmuirian fits to the observed dependencies, are (1) ∼6070 pmol cm2, (2) anion independent, and (3) similar to the surface concentration of the CD-SH/C5SH adlayer. The results also show that these anions do not measurably inhibit the inclusion of ferrocene; that is, the association of these anions with the CDSH/C5SH adlayer is weak compared to ferrocene. As expected for the anions showing no inhibition, the determined values for log Kguest,surf (Table 1) are similar (4.75.0) but are higher than that (3.6) in solution as already observed in 0.2 M NaClO4.21,24 In contrast to ClO4, Cl, and Br, the values for ΓFc with SO42, NO3, and HPO42 as electrolytes are much smaller across the entire span of ferrocene concentrations. ΓFc,max, which in these cases was determined directly from the plots in Figure 3, are nearly 20 times or more below those for ClO4, Cl, and Br. In fact, ΓFc is immeasurably small for HPO42 at all tested ferrocene concentrations. These data further support the strong inhibition of these anions on the inclusion of ferrocene in the CD cavity. Stated differently, there is a much stronger association of SO42, NO3, and HPO42 than of ClO4, Cl, and Br with the CD-SH/C5SH monolayer. To determine the surface association constants for the anions, Kanion,surf, the anion concentration dependence of ΓFc was measured at a fixed ferrocene concentration (60 μM) in 0.2 M 10582

dx.doi.org/10.1021/la1051063 |Langmuir 2011, 27, 10580–10586

Langmuir

ARTICLE

Table 1. Maximum Adsorption Amount of Fc, Inclusion Efficiency, and Association Constant for the Inclusion of Fc in CD-SH/ C5SH Monolayer in Various Anion Solutions anion ClO4

Cl

ΓFc, maxa/pmol cm2

66.9 ( 4.7

69.2 ( 2.6

inclusion efficiencyb/%

95.8

101.8 ( 3.0

log Kassa

5.0 ( 0.2

4.9 ( 0.3

SO42

NO3

PO43

61.8 ( 4.3

3.8 ( 2.4

2.0 ( 0.7

0.0

93.6 ( 6.1

5.6 ( 3.4

2.7 ( 0.8

0.0

4.7 ( 0.2









Br



For ClO4, Cl, and Br, ΓFc,max and Kguest surf were determined from corresponding Langmuir plots. For SO42, NO3, and HPO42, ΓFc, max was determined directly from the concentration dependence of the inclusion amount; Kguest surf was not determined. b Inclusion efficiency is defined as the ratio of ΓFc, max to the surface concentration of CD-SH. a

Figure 4. Concentration dependence of Γanion,C associated with the CD-SH/C5SH monolayer in 0.2 M NaClO4. The association amount was determined by the procedure described in the text. The inset plots ln[θ/C(1  θ)] vs θ for the determination of Kanion,surf, where θ = Γanion,C/ΓFc,max.

NaClO4. The amount of included anion, Γanion,C, at a given anion solution concentration, C, was found by using the equation Γanion,C = ΓFc,max  ΓFc,C, as described in the Experimental Section. These isotherms follow that of a Frumkin type (Figure 4 and its inset), instead of a Langmuir type, probably due to electrostatic interaction between included anions. The association constants of anions, given as log Kanion,surf and determined from the inset in Figure 4, are 10.7 ( 0.4, 9.7 ( 1.0, and 9.0 ( 0.7 for SO42, NO3, and HPO42, respectively. Therefore, these anions are strongly associated with surface-confined CD-SH. The CD-SH/C5SH monolayer was also characterized by XPS after emersion from a 1 mM solution of each electrolyte that was devoid of ferrocene. The results are shown in Figure 5. As evident from the data in the Cl(2p) and Br(3d) regions, there is no detectable presence for ClO4, Cl, or Br, which is consistent with the weak association of these anions with CD-SH/C5SH adlayer. On the other hand, the stronger association of SO42, NO3 and HPO42 is clearly demonstrated by the presence of S(2p), N(1s), and P(2p) bands, respectively. The observed binding energies of 169 eV for S(2p), 407 eV for N(1s), and 134 eV for P(2p) are in reasonable agreement with reported values of 168.7 ( 0.6 eV for SO42, 405.4 ( 2.6 eV for NO3, and 133.3 ( 0.3 eV for HPO42.33 Note that the strong bands at ∼164 eV for samples immersed in NaClO4 and Na2SO4 reflect the presence of the gold-bound thiolate that results from adlayer formation. This band was observed for all samples. The data in Figure 5 can also be used to estimate the ratio of included anion to surface-confined CD-SH by comparing the

Figure 5. XPS data for CD-SH/C5SH monolayer after 1 h immersions in 1 mM electrolyte: (a) Cl(2p), (b) Br(3d), (c) S(2p), (d) N(1s), and (e) P(2p).

integrated intensities of the corresponding XPS bands after accounting for differences in photoionization cross sections and the spectrometer instrument function (e.g., wavelength dependence of excitation source and detector). The obtained SO42/CD-SH, NO3/CD-SH, and HPO42/CD-SH ratios are 0.15, 0.14, and 0.17, respectively. These values are close to the ratio of the average number of thiols appended to each CD molecule (0.12), pointing to a 1:1 association for each of these anions per immobilized CD-SH. While the isotherm and XPS data confirm the association of SO42, NO3, and HPO42 with the CD-SH/C5SH monolayer, it is not clear if these anions are included in the cavity or simply interact with the outer surface of the adlayer. To clarify this situation, QCM measurements using CD-SH/C5SH monolayers formed on Au(111) coated QCM electrodes were conducted with p-bromophenol (p-BP) and methylene blue (MB) as guest compounds. Upon injection of these solutions, the frequency of the QCM decreases, which corresponds to the mass increase expected from the well-known inclusion of these compounds in the CD cavity (Figure 6a and b). More quantitatively, the total frequency changes (0.49 ( 0.05 and 1.24 ( 0.10 Hz for p-BP and MB, respectively) 10583

dx.doi.org/10.1021/la1051063 |Langmuir 2011, 27, 10580–10586

Langmuir

ARTICLE

Table 2. Frequency Changes Expected for MB, p-BP, and SO42 Inclusion

Figure 6. Time courses of the frequency change of the CD-SH/C5SH monolayer formed on a Au(111)-coated QCM upon injection of SO42 into 0.1 M NaCl. (a) and (b), p-BP or MB was injected prior to SO42 injection, respectively; (c) SO42 injection; and (d) SO42 injection after all CD cavities were filled with C3SH. The concentrations of p-BP, MB, and SO42 were 0.5 mM after the injection.

correspond to those expected for the inclusion of one guest molecule and the exclusion of water molecules already sequestered within the CD cavity. The X-ray crystal structures show that as many as six or seven water molecules can reside within a β-CD cavity.34 The frequency change expected for the inclusion and exclusion of a guest or anion is simply given by Δf ¼ 

ΓCD-SH ðmolcm2 Þ  fMincl ðgmol1 Þ  n  Mexcl ðgmol1 Þg Sðgcm2 Hz1 Þ

ð2Þ where ΓCD-SH is the surface concentration of CD-SH on an apparent surface area basis, which is 121 pmol cm2 (the roughness factor of gold is 1.64), Mincl and Mexcl are the molar masses of the included and excluded species, respectively, n is the number of excluded species per CD cavity, and S is the sensitivity of the ATcut 5 MHz QCM (17.7  109 g cm2 Hz1). When a p-BP or MB solution was injected, the included and excluded species are a guest (p-BP or MB) and water, respectively. If we assume six water molecules are initially included in the cavity, the expected frequency change for the replacement of the included water by p-BP or MB would be 0.42 and 1.14 Hz, respectively; if seven water molecules are displaced, the respective changes would be 0.30 and 1.03 Hz (Table 2). Our data are more consistent with the presence of six and not seven included water molecules. The subsequent injection of a sulfate solution causes the QCM frequency to increase (Figure 6a and b), showing the removal of a heavier guest (i.e., p-BP or MB) and inclusion of a lighter anion (i.e., sulfate). The observed frequency increase of 0.55 ( 0.10 and 1.41 Hz ( 0.18 Hz for p-BP and MB are almost the same as those expected for the simple replacement of preincluded p-BP or MB with sulfate, 0.50 and 1.22 Hz for p-BP and MB, respectively (Table 2). Following the procedures we used for determination of Kanion,surf, but with p-BP and MB, we determined the

included species

excluded species

(molar mass)

(molar mass)

n

Δf /Hz

MB (284.5)

H2O (18.0)

6 7

1.14 1.03

SO42 (96.1)

MB (284.5)

1

1.22

p-BP (173.0)

H2O (18.0)

6

0.42

7

0.30

SO42 (96.1)

p-BP (173.0)

1

0.50

SO42 (96.1)

H2O (18.0)

6

0.08

respective values for log Kguest,surf for p-BP and MB to be 4.38 ( 0.14 and 4.20 ( 0.20, via isotherms of a Langmuir type. Therefore, the replacement with the sulfate, having a higher surface association constant, is thermodynamically favored. It should be mentioned that the values of log Kguest,surf for p-BP and MB are greater than bulk association constants (log Kguest,sol = 2.93 ( 0.07 and 2.323.65 for p-BP and MB, respectively),5,35 as was found for ferrocene. When the sulfate solution was injected without preinclusion of p-BP and MB, the frequency slightly increased (Figure 6c). The observed total frequency change of 0.13 ( 0.01 Hz agrees well with that for the sulfate inclusion with removal of six water molecules (0.08 Hz) (Table 2). The QCM measurements were also conducted after filling cavities with propanethiol (C3SH). In this case, no frequency change upon the sulfate injection was observed (Figure 6d). These results clearly demonstrate that sulfate is included into the cavities of CD-SH immobilized on Au. The findings herein demonstrate that SO42, NO3, and HPO42 are strongly included into the CD cavity of CDSH/C5SH monolayers. The surface association constants (log Kanion, surf) for the inclusion of these anions are 9.010.7, which are much larger than those of common organic guests (e.g., ferrocene, p-bromophenol, and methylene blue) in aqueous solution. As a result, the presence of these anions strongly but unexpectedly inhibits the inclusion of ferrocene into the CD cavity. There are several factors to be considered for the strong inclusion of SO42, NO3, and HPO42 in the CD cavity on the surface. The diameter of these anions is greater than those of halides: 0.33, 0.36, 0.48, 0.53, and 0.54 for Cl, Br, NO3, HPO42, and SO42, respectively. Based on the better size-fit to the CD cavity (0.620.78 nm), a stronger van der Waals interaction is expected.5 The local charge density of the guest is commonly considered for the interaction between the CD and the guest in an aqueous solution as discussed in terms of hydrophobic interaction5 and structure breaking effect in case of the anionic guest.27 One of the surface-specific factors would be the specific adsorption of these anions on Au. However, this cannot be a predominant factor, because halides are usually more strongly adsorbed than sulfate.36 However, it should be mentioned that a possibility remains that halides are also more strongly included in the CD cavity on the surface than in an aqueous solution, because the present data only show that the halide inclusion is weaker than ferrocene. In such case, the specific adsorption of the anion still becomes an important factor. Another possibility would be hydrogen bonding between the anion and hydroxy groups of CD as known for halides.37 Oxoanions would undergo multiple hydrogen bonding and therefore be more stabilized compared to halides. To form hydrogen bonds, the 10584

dx.doi.org/10.1021/la1051063 |Langmuir 2011, 27, 10580–10586

Langmuir

Figure 7. Concentration dependence of the inclusion amount of Fc on the mixed CD-SH/C5SH monolayer prepared by (solid circle) our three-step and (square) Kaifer’s two-step methods in the 0.2 M electrolyte solutions shown in the figure. The inclusion amount was determined by chronocoulometry according to a previously reported procedure.21

secondary hydroxy groups, usually located at the outside of the CD, need to face the inside of the CD cavity. Such a structural change would be assisted by the more hydrophobic nature of the outside environment, namely, the intermolecular vacancies are filled with C5SH. Noninclusion of perchlorate, which has a larger association constant in an aqueous solution compared to SO42 and NO3,2830 is probably due to nonspecific adsorption property of this anion. In addition, its larger size (0.59 nm) compared to the other used anions would be disadvantage for the inclusion into the more rigid surface-confined CD. In previous studies,11b,22,38 it was reported that ferrocene was included in the SAM of CD-SH in an aqueous Na2SO4 solution. For the CD-SH/C5SH monolayers prepared by a two-step procedure, particularly, the inclusion amounts are essentially the same between 0.2 M Na2SO4 (25 pmol cm2 reported by Kaifer’s group22) and 0.2 M NaClO4 (26 ( 3.3 pmol cm2 by us21). This seems to suggest no inhibition of the ferrocene inclusion by the sulfate anions. To examine this, we conducted chronocoulometric measurements for the ferrocene inclusion both in 0.2 M NaClO4 and 0.2 M Na2SO4 using the same CDSH/C5SH monolayers prepared by the two-step procedure. The first determined maximum inclusion amount in 0.2 M NaClO4 was 31 pmol cm2 (the corresponding isotherm is shown in Figure 7, solid square), which is almost within the experimental error of the previous reported value21 and is much lower than that for the CD-SH/C5SH monolayers prepared by the threestep procedure (Figure 7, solid circle) as previously reported.21 When the electrolyte solution was switched from 0.2 M NaClO4 to 0.2 M Na2SO4, the inclusion amount decreased to almost zero (Figure 7, open square more precisely 1.2 pmol cm2 as an average value). Therefore, the inhibition of ferrocene inclusion by sulfate anions occurs independent of how the monolayers were prepared (two-step or three-step procedure). In other words, the appearance/nonappearance of the inhibition effect is not due to the preparation procedure. Because the only difference in the experimental conditions is the substrate materials (i.e., we used Au(111), but the other groups used polycrystalline Au), the anion effect was examined using a polycrystalline Au electrode. Almost no inhibition effect on the redox of ferrocene was observed on the polycrystalline Au electrode (Figure S1 in the Supporting Information). In some other preliminary experiments using different polycrystalline Au electrodes, a weak effect was observed. Although the effect of the structural regularity on the surface

ARTICLE

has not yet quantitatively examined, the magnitude of the inhibition effect seems to be dependent on the degree of regularity on the surface. The origin of such a substrate effect is unclear at this moment. However, among the above-listed factors (size, specific adsorption, and hydrogen bonding), the size-fit is not attributed to the effect, because a smaller sulfate (0.54 nm) can be included in CD cavities that a larger ferrocene (0.66 nm39) can enter, even if the distorted adsorbed CD-SH, and hence smaller CD cavities, is formed on a less-ordered surface. The strong specific adsorption of sulfate on Au(111) reported in the literature40,41 will be a significant reason for the Au(111)-specific inhibition effect, although a weaker adsorption on Au(111) compared to Au(100) and Au(110) was also reported.42 As stated above, we considered that the structural change in the adsorbed CD-SH to form multiple hydrogen bonds between the secondary hydroxy groups and sulfate is assisted by the hydrophobic nature of the outside environment. Based on this idea, the decrease in the hydrophobicity due to the less-ordered structure of the outside C5SH on the polycrystalline Au will have a negative effect on the formation of the hydrogen bonding. Experiments are now underway to more fully explore the scope of present findings, to gain insights into the mechanistic origins of the observed inclusion, the substrate dependence, and the differences between the solution and surface association strengths, and to access this unusual observation on the application of these systems in sensors and catalysis.

’ ASSOCIATED CONTENT

bS

Supporting Information. Cyclic voltammograms for the redox of Fc on the mixed CD-SH/C5SH monolayer on polycrystalline Au are shown. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]; phone: +81-11-706-2276; fax: +81-11-706-4868.

’ REFERENCES (1) Szejtli, J. In Comprehensive Supramolecular Chemistry; Atwood, J. L., Davis, J. E. D., Macnicol, D. D., V€ogtle, F., Szejtli, J., Osa, T., Eds.; Pergamon: Oxford, U.K., 1996; Vol. 5, pp 189203. (2) (a) Ngo, H. T.; Clements, P.; Easton, C. J.; Pham, D.-T.; Lincoln, S. F. Aust. J. Chem. 2010, 63, 687–692. (b) Ma, D.; Tu, K.; Zhang, L.-M. Biomacromolecules 2010, 11, 2204–2212. (c) Dreja, M.; Kim, I. T.; Yin, Y.; Xia, Y. J. Mater. Chem. 2000, 10, 603–605. (3) (a) Kpegba, K.; Murtha, M.; Nesnas, N. Bioorg. Med. Chem. Lett. 2006, 16, 1523–1526. (b) Kataky, R.; Morgan, E. Biosens. Bioelectron. 2003, 18, 1407–1417. (c) Wu, P.; Xiao, R.; Zhang, C.; Zhou, L.; Luo, Q.; Xu, J.; Liu, J. Catal. Lett. 2010, 138, 62–67. (d) Das, S.; Brudvig, G. W.; Crabtree, R. H. Chem. Commun. 2008, 413–424. (e) Milovic, N. M.; Badjic, J. D.; Kostic, N. M. J. Am. Chem. Soc. 2004, 126, 696–697. (f) Singleton, M. L.; Reibenspies, J. H.; Darensbourg, M. Y. J. Am. Chem. Soc. 2010, 132, 8870–8871. (4) (a) Chen, Y.; Liu, Y. Chem. Soc. Rev. 2010, 39, 495–505. (b) Uhlenheuer, D. A.; Petkau, K.; Brunsveld, L. Chem. Soc. Rev. 2010, 39, 2817–2826. (5) Rekharsky, M. V.; Inoue, Y. Chem. Rev. 1998, 98, 1875–1917. (6) (a) Hardee, G. E.; Otagiri, M.; Perrin, J. H. Acta Pharm. Suec. 1978, 15, 188–199. (b) Gelb, R. I.; Schwartz, L. M. J. Inclusion Phenom. Mol. Recognit. Chem. 1989, 7, 465–476. (c) Siimer, E.; K~obu, M.; Kurvits, 10585

dx.doi.org/10.1021/la1051063 |Langmuir 2011, 27, 10580–10586

Langmuir M. Thermochim. Acta 1990, 170, 89–95. (d) Siimer, E.; Kurvits, M.; K€ostner, A. Thermochim. Acta 1987, 116, 249–256. (7) Lewis, E. A.; Hansen, L. D. J. Chem. Soc., Perkin Trans. 2 1973, 2081–2085. (8) (a) Bertrand, G. L.; Faulkner, J. R.; Han, S. M.; Armstrong, D. W. J. Phys. Chem. 1989, 93, 6863–6867. (b) Harata, K. Bioorg. Chem. 1980, 9, 530–539. (9) (a) Briggner, L.-E.; Ni, X.-R.; Tempesti, F.; Wads€o, I. Thermochim. Acta 1986, 109, 139–143. (b) Eftink, M. R.; Andy, M. L.; Bystrom, K.; Perlmutter, H. D.; Kristol, D. S. J. Am. Chem. Soc. 1989, 111, 6765–6772. (c) Weickenmeter, M.; Wenz, G. Macromol. Rapid Commun. 1996, 17, 731–736. (d) Komiyama, M.; Bender, M. L. J. Am. Chem. Soc. 1978, 100, 2259–2260. (e) Godínez, L. A.; Schwartz, L.; Criss, C. M.; Kaifer, A. E. J. Phys. Chem. B 1997, 101, 3376–3380. (10) (a) Lee, J.-Y.; Park, S.-M. J. Phys. Chem. B 1998, 102, 9940–9945. (b) Ghosh-Mukerji, S.; Haick, H.; Schvartzman, M.; Paz, Y. J. Am. Chem. Soc. 2001, 123, 10776–10777. (c) Ghosh-Mukerji, S.; Haick, H.; Paz, Y. J. Photochem. Photobiol. A 2003, 160, 77–85. (11) (a) Choi, S.-J.; Choi, B.-G.; Park, S.-M. Anal. Chem. 2002, 74, 1998–2002. (b) D’Annibale, A.; Regoli, R.; Sangiorgio, P.; Ferri, T. Electroanalysis 1999, 11, 505–510. (c) Ng, S.-C.; Sun, T.; Chan, H. S. O. Macromol. Symp. 2003, 192, 171–181. (d) Ng, S.-C.; Sun, T.; Chan, H. S. O. Tetrahedron Lett. 2002, 43, 2863–2866. (12) (a) Mathews, A. S.; Cho, W.-J.; Kim, I.; Ha, C.-S. J. Appl. Polym. Sci. 2009, 113, 1680–1689. (b) Dahan, A.; Miller, J. M.; Hoffman, A.; Amidon, G. E.; Amidon, G. L. J. Pharm. Sci. 2010, 99, 2739–2749. (c) Zhang, J.; Ellsworth, K.; Ma, P. X. J. Controlled Release 2010, 145, 116–123. (d) Layre, A.-M.; Volet, G.; Wintgens, V.; Amiel, C. Biomacromolecules 2009, 10, 3283–3289. (e) Zugasti, M. S.; Zornoza, A.; Goni, M. D. M.; Isasi, J. R.; Velaz, I.; Martin, C.; Sanchez, M.; MartínezOharriz, M. C. Drug Dev. Ind. Pharm. 2009, 35, 1264–1270. (13) (a) Grisales, J. O.; Lebed, P. J.; Keunchkarian, S.; Gonzalez, F. R.; Castells, C. B. J. Chromatogr., A 2009, 1216, 6844–6851. (b) Varga, 00 G.; Tarkanyi, G.; Nemeth, K.; Ivanyi, R.; Jicsinszky, L.; Toke, O.; Visy, J.; Szente, L.; Szeman, J.; Simonyi, M. J. Pharm. Biomed. Anal. 2010, 51, 84–89. (c) Bogdanski, A.; Wistuba, D.; Larsen, K. L.; Hartnagel, U.; Hirsch, A.; Schurig, V. New J. Chem. 2010, 34, 693–698. (d) Zhou, Z.-M.; Li, X.; Chen, X.-P.; Fang, M.; Dong, X. Talanta 2010, 82, 775–784. (14) (a) Banerjee, I. A.; Yu, L.; Matsui, H. J. Am. Chem. Soc. 2003, 125, 9542–9543. (b) Zhang, Q.; Tu, Y.; Tian, H.; Zhao, Y.-L.; Stoddart, J. F.; Ågren, H. J. Phys. Chem. B 2010, 114, 6561–6566. (c) Willner, I.; Willner, B. J. Mater. Chem. 1998, 8, 2543–2556. (15) (a) Arkas, M.; Allabashi, R.; Tsiourvas, D.; Mattausch, E.; Perfler, R. Environ. Sci. Technol. 2006, 40, 2771–2777. (b) Phan, T. N. T.; Bacquet, M.; Morcellet, M. React. Funct. Polym. 2002, 52, 117–125. (16) (a) Hedges, A. R. Chem. Rev. 1998, 98, 2035–2044. (b) Centini, M.; Maggiore, M.; Casolaro, M.; Andreassi, M.; Facino, R. M.; Anselmi, C. J. Inclusion Phenom. Macrocyclic Chem. 2007, 57, 109–112. (c) Munro, I. C.; Newberne, P. M.; Young, V. R.; B€ar, A. Regul. Toxicol. Pharmacol. 2004, 39, S3–S13. (17) Matsuda, H.; Ito, K.; Fujiwara, Y.; Tanaka, M.; Taki, A.; Uejima, O.; Sumiyoshi, H. Chem. Pharm. Bull. 1991, 39, 827–830. (18) (a) Creager, S. E.; Clarke, J. Langmuir 1994, 10, 3675–3683. (b) Vezenov, D. V.; Noy, A.; Rozsnyai, L. F.; Lieber, C. M. J. Am. Chem. Soc. 1997, 119, 2006–2015. (c) Hu, K.; Bard, A. J. Langmuir 1997, 13, 5114–5119. (19) (a) Shimazu, K.; Teranishi, T.; Sugihara, K.; Uosaki, K. Chem. Lett. 1998, 27, 669–670. (b) Sugihara, K.; Teranishi, T.; Shimazu, K.; Uosaki, K. Electrochemistry 1999, 67, 1172–1174. (c) Sugihara, K.; Shimazu, K.; Uosaki, K. Langmuir 2000, 16, 7101–7105. (20) Kitano, H.; Taira, Y.; Yamamoto, H. Anal. Chem. 2000, 72, 2976–2980. (21) Domi, Y.; Yoshinaga, Y.; Shimazu, K.; Porter, M. D. Langmuir 2009, 25, 8094–8100. (22) Rojas, M. T.; Koeniger, R.; Stoddart, J. F.; Kaifer, A. E. J. Am. Chem. Soc. 1995, 117, 336–343.

ARTICLE

(23) (a) Fukuda, T.; Maeda, Y.; Kitano, H. Langmuir 1999, 15, 1887–1890. (b) Kitano, H.; Taira, Y. Langmuir 2002, 18, 5835–5840. (c) Maeda, Y.; Fukuda, T.; Yamamoto, H.; Kitano, H. Langmuir 1997, 13, 4187–4189. (d) Yamamoto, H.; Maeda, Y.; Kitano, H. J. Phys. Chem. B 1997, 101, 6855–6860. (24) Osella, D.; Carretta, A.; Nervi, C.; Ravera, M.; Gobetto, R. Organometallics 2000, 19, 2791. (25) (a) Beulen, M. W. J.; B€ugler, J.; deJong, M. R.; Lammerink, B.; Huskens, L.; Sch€onherr, H.; Vancso, G. J.; Boukamp, B. A.; Wieder, H.; Offenh€auser, A.; Knoll, W.; van Veggel, F. C. J. M.; Reinhoudt, D. N. Chem.—Eur. J. 2000, 6, 1176–1183. (b) deJong, M. R.; Huskens, J.; Reinhoudt, D. N. Chem.—Eur. J. 2001, 7, 4164–4170. (26) (a) Chmurski, K.; Temeriusz, A.; Bilewicz, R. Anal. Chem. 2003, 75, 5687–5691. (b) Chmurski, K.; Temeriusz, A.; Bilewicz, R. J. Inclusion Phenom. Macrocyclic Chem. 2004, 49, 187–191. (27) Rohrbach, R. P.; Rodriguez, L. J.; Eyring, E. M.; Wojcik, J. F. J. Phys. Chem. 1977, 81, 944–948. (28) Cramer, F.; Saenger, W.; Spatz, H.-Ch. J. Am. Chem. Soc. 1967, 89, 14–20. (29) Matsui, Y.; Ono, M.; Tokunaga, S. Bull. Chem. Soc. Jpn. 1997, 70, 535–541. (30) Dey, J.; Roberts, E. L.; Warner, I. M. J. Phys. Chem. A 1998, 102, 301–305. (31) Chamberlain, R. V., II; Slowinska, K.; Majda, M.; B€uhlmann, P.; Aoki, H.; Umezawa, Y. Langmuir 2000, 16, 1388–1396. (32) To further confirm this conclusion, we examined whether the guest can be adsorbed on the C5SH layer. The experiments were conducted for the adsorption of ferrocene on the self-assembled monolayer of C5SH on Au in 0.2 M NaClO4 using chronocoulometry as will be described in Section 2.4. The results show no adsorption of ferrocene over its concentration range of 060 μM. (33) Wagner, C. D.; Naumlin, A. V.; Kraut-Vass, A.; Alison, J. W.; Powell, C. J.; Rumble, J. R., Jr. NIST X-ray Photoelectron Spectroscopy Database 20 vers. 3.5. http://webbook.nist.gov, 2007. (34) Lindner, K.; Saenger, W. Angew. Chem., Int. Ed. Engl. 1978, 17, 694–695. (35) (a) Liu, Y.; Li, L.; Chen, Y. J. Inclusion Phenom. Macrocyclic Chem. 2002, 42, 151–155. (b) Lee, C.; Sung, Y. W.; Park, J. W. J. Phys. Chem. B 1999, 103, 893–898. (c) Zhao, G.-C.; Zhu, J.-J.; Chen, H.-Y. Spectrochim. Acta Part A 1999, 55, 1109–1117. (36) Lipowski J. Can. J. Chem. 1999, 77, 1163–1176. (37) Fu., Y.; Liu, L.; Guo, Q.-X. J. Inclusion Phenom. 2002, 43, 223–229. (38) Choi, S.-W.; Jang, J.-H.; Kang, Y.-G.; Lee, C.-J.; Kim, J.-H. Colloids Surf., A 2005, 257258, 31–36. (39) Rowe, G. K.; Creager, S. T. Langmuir 1991, 7, 2307–2312. (40) Angerstein-Kozlowska, H.; Conway, B. E.; Hamelin, A.; Stoicoviciu, L. J. Electroanal. Chem. 1987, 228, 429–453. (41) Hamelin, A. J. Electroanal. Chem. 1983, 144, 365–372. (42) Cahan, B. D.; Villullas, H. M.; Yeager, E. B. J. Electroanal. Chem. 1991, 306, 213–238.

10586

dx.doi.org/10.1021/la1051063 |Langmuir 2011, 27, 10580–10586