- Email: [email protected]
In situ sensing of metal ion adsorption to a thiolated surface using surface plasmon resonance spectroscopy
Journal of Colloid and Interface Science 298 (2006) 543–549 www.elsevier.com/locate/jcis
In situ sensing of metal ion adsorption to a thiolated surface using surface plasmon resonance spectroscopy Jungwoo Moon, Taewook Kang, Seogil Oh, Surin Hong, Jongheop Yi ∗ School of Chemical and Biological Engineering, Institute of Chemical Processes, Seoul National University, Seoul 151-744, South Korea Received 2 November 2005; accepted 30 December 2005 Available online 3 February 2006
Abstract The kinetics of the adsorption of metal ions onto a thiolated surface and the selective and quantitative sensing of metal ions were explored using surface plasmon resonance (SPR) spectroscopy. The target metal ion was an aqueous solution of Pt2+ and a thin-gold-film-coated glass substrate was modified with 1,6-hexanedithiol (HDT) as a selective sensing layer. SPR spectroscopy was used to examine the kinetics of metal ion adsorption by means of the change in SPR angle. The selectivity of the thiolated surface for Pt2+ over other divalent metal ions such as Cu2+ , Ni2+ , and Cd2+ was evident by the time-resolved SPR measurement. SPR angle shift, θSPR , was found to increase logarithmically with increasing concentration of Pt2+ in the range of 1.0 × 10−5 –1.0 mM. The rate of Pt2+ adsorption on HDT observed at both 0.1 and 1 mM Pt2+ accelerates until the surface coverage reaches approximately 17%, after which the adsorption profile follows Langmuirian behavior with the surface coverage. The experimental data indicated that heavy metal ions were adsorbed to the hydrophobic thiolated surface by a cooperative mechanism. A mixed self-assembled monolayer (SAM) composed of HDT and 11-mercaptoundecanoic acid was used to reduce the hydrophobicity of the thiol-functionalized surface. The addition of hydrophilic groups to the surface enhanced the rate of adsorption of Pt2+ onto the surface. The findings show that the adsorption of metal ions is strongly dependent upon the hydrophilicity/hydrophobicity of the surface and that the technique represents an easy method for analyzing the adsorption of metal ions to a functionalized surface by combining SPR spectroscopy with a SAM modification. © 2006 Elsevier Inc. All rights reserved. Keywords: SPR; Metal ions; Adsorption; Kinetics; Mixed SAM
1. Introduction Reactions that occur at a solid/liquid interface include a number of significant chemical reactions, such as charge transfer, ion exchange, and adsorption. However, the in situ monitoring of the surface change is difficult because analytical methods that can be applied to a solid surface are restricted to certain environments. There are many usable surface analysis tools, such as UV-DRS, AES, AT-IR, and SEM; however, almost all of these techniques require high vacuum conditions or the destruction of the sample and cannot be used for the in situ observation of a surface change [1]. Continuous change in a surface can be estimated by the change in concentration of bulk phase molecules, pH, temperatures, and colors. Labeling is an alternative * Corresponding author.
E-mail address: [email protected] (J. Yi). 0021-9797/$ – see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2005.12.066
for visualizing a surface change, but it requires a spectroscopic or radioactive additive and some types of surface changes cannot be observed by these methods [2]. Because of these difficulties, the mechanistic factors that influence interfacial reactions remain poorly understood [3]. Surface plasmon resonance (SPR) spectroscopy is a powerful tool for the in situ real-time characterization of a solid/liquid interface [4–18]. Surface plasmons are collective oscillations of free electrons at an interface between a thin metal film and a dielectric medium. A surface plasmon wave is resonant with the p-polarized beam at a certain incident angle, and the intensity of the reflected light becomes evanescent. This phenomenon is called surface plasmon resonance. The SPR technique typically involves the optical excitation of surface plasmon polariton waves on a gold or silver film employing the Kretschmann configuration. Reflectivity is measured as a function of the incident angle [4]. The change in the surface results in a shift in
544
J. Moon et al. / Journal of Colloid and Interface Science 298 (2006) 543–549
the SPR angle in an attenuated total reflectance (ATR) curve, because SPR is extremely sensitive to the dielectric constant of the medium immediately adjacent to the metal film. Adsorption of molecules onto the metallic film or conformational changes in the adsorbed molecules can be accurately monitored by optical methods [5]. SPR spectroscopy usually detects surface changes in the boundary in the submicrometer range, about 300 nm in this study, because SPR is an evanescent wave technique, possessing maximum sensitivity on the surface and being characterized by the exponential decay of sensitivity with increasing distance from the surface [6]. The power of an optical sensor experimental approach, in which one of the interacting molecules (the analyte) is free in solution and the other (the ligand) is attached to the sensor surface, is that the formation and decomposition of the analyte/ligand complex can be monitored in situ, yielding kinetic data in real time [7,8]. This technique can be used to monitor in situ interactions of an analyte and a ligand on the surface without the need to label the reactant, as is required in spectroscopic and radioactive probes, and is a noninvasive real-time surface analysis method. The availability of real-time SPR kinetic data provides the possibility of understanding the mechanism of the chemical and/or physical binding events for surface reactions [2,4–8]. A self-assembled monolayer (SAM) is a flexible and simple system that permits the interfacial properties of a metal substrate to be tailored for a well-designed functional surface [19– 21]. Alkanethiols can be fabricated on a gold surface by selfassembly, and the bonding of thiol groups to gold is sufficiently strong to be stable. An attempt was made to identify the adsorption mechanism and kinetics, but an approximation was required because the direct observation of surface change was impossible. A better understanding of the adsorption mechanism could permit adsorption reactions to be controlled and the efficiency of heavy metal ion reduction, the catalyst, metal ion sensor, and other techniques using metal ion adsorption might be made more efficient. The adsorption of heavy metal ions onto a thiol-functionalized surface is a meaningful reaction for reducing the level of environmental toxic heavy metal ions such as Hg2+ [22–24], manufacturing metal-doped catalysts [25], and metal ion sensors [4]. In particular, it should be noted that the adsorption of Pt2+ onto the thiolated surface may be applicable to the qualitative and quantitative analyses of aqueous Hg2+ ions in that thiol compounds have a strong affinity to Hg2+ and Pt2+ . We recently investigated in situ SPR spectroscopy as a metal ion sensor in an aqueous phase [4]. A thiolated gold substrate was used as the sensing layer and the electric field near the surface served as a probe. In this study, SPR spectroscopy was used to examine the adsorption of metal ions to a thiolated surface at the molecular level. A thin-gold-film-coated glass substrate was thiol-functionalized with 1,6-hexanedithiol (HDT) by a selfassembly method. The thiol functionality is known to selectively adsorb heavy metal ions in an aqueous solution [4]. SPR spectroscopy was found to be useful in identifying the selective adsorption to a functionalized surface and in quantitatively
analyzing the concentration of Pt2+ . In addition, the specific adsorption kinetics of the adsorption of Pt2+ to an HDT functionalized surface was observed; that is to say, the adsorption of metal ions onto the hydrophobic thiolated surface followed a cooperative adsorption mechanism and was strongly dependent on the hydrophilicity of the surface. 2. Materials and methods 2.1. Chemicals 1,6-Hexanedithiol (HS(CH2 )6 SH, HDT, Aldrich) and 11mercaptoundecanoic acid (HS(CH2 )10 COOH, MUA, Aldrich) were used as purchased. Potassium tetrachloroplatinate (K2 PtCl4 , Riedel de Haen) was used to prepare a solution of Pt2+ . For the preparation of the other metal ion solutions, cadmium nitrate tetrahydrate, copper(II) nitrate trihydrate, and nickel(II) nitrate hexahydrate were purchased from Junsei. The water used was purified using a Milli-QTM Millipore system. Sulfuric acid (H2 SO4 , Fisher Scientific) and hydrogen peroxide (H2 O2 , Fisher Scientific) were used to clean the microscope slides on which the chromium and gold films were evaporated. Hydrogen chloride (HCl, Fisher Scientific) and sodium hydroxide (NaOH, Mallinckrodt Chemicals) were used to control the pH of the metal ion solutions. The SF10 glass with high reflective index (n = 1.78) was used as a prism and a thin gold film was deposited on a glass slide. The microscope slide glasses (1 in. × 1 in., SF10) were immersed in piranha solution (H2 SO4 :H2 O2 = 7:3 v/v) for purification. They were rinsed several times with water and ethanol before deposition of the thin gold film. 2.2. SPR spectroscopy and analysis method The SPR instrument was fabricated in-house and a schematic diagram of the setup is shown in Fig. 1. A 50-nm thin gold film was prepared by thermal evaporation with a 5-nm Cr adhesion layer. A cleaned gold substrate was attached to an SF10 prism with index matching oil (Cargille Laboratories Inc., certified refractive index liquids, n = 1.730 ± 0.0005). A Teflon cell was attached to the gold substrate. The 635-nm He–Ne diode laser (Power Technology Inc.) was p-polarized and focused with a lens through the prism onto the gold substrate. Both the prism and the gold substrate were mounted on a rotating plate to control the angle of the incident light. The reflectance was measured with a photopower meter (Oriel). Quantitative analysis was investigated with angle-resolved analysis. Usually the regulation levels of the toxic heavy metal ions are below than 1 ppm; thus, the initial concentration of solution was fixed at 0.0001, 0.001, 0.01, 0.1, and 1 mM for the quantitative analysis. For the acquisition of kinetic data, time-resolved SPR angle shifts were measured by the fixed-angle method, which assumed the reflectance change, R, to be linearly correlated with the SPR angle shift, θSPR , in a narrow angle range [6]. The fixed incident angle for measuring the reflectance was adjusted before each experiment. Data for the reflectance at a fixed
J. Moon et al. / Journal of Colloid and Interface Science 298 (2006) 543–549
545
Fig. 1. Schematic diagram of the six-phase SPR system and the experimental setup used for SPR angle measurement. The HDT was self-assembled on a thin gold film, and the metal ion solution was added to the sample cell. The metal ion was adsorbed by the HDT SAM on the thin gold film and the surface change was detected by an optical signal.
Fig. 2. SPR angle, at which the intensity is minimal at the curves, shifts after the self-assembly of HDT on a thin gold film and Pt2+ adsorption onto the HDT SAM. The gold-coated substrate was immersed in a 10 mM ethanolic solution for 4 h to form a HDT self-assembled monolayer on the thin gold film. The HDT self-assembled substrate was then allowed to contact a 10 mM aqueous solution of Pt2+ for 48 h because the adsorption needed to be fully saturated.
incident angle were acquired by a computer in real time. There is a very short time lag for acquiring data using a computer and the data can be acquired within a period of 1 ms. The reflectance at the fixed measuring angle is converted to the SPR angle shift in this method. The molecular level adsorption kinetics of Pt2+ onto HDT ligands was obtained by monitoring the change in reflectance. During the monitoring of the reflectance, the prepared metal ion solutions were allowed to flow into a Teflon cell. The concentration of metal ions was controlled at 0.1, 1, and 10 mM in an aqueous solution, the concentration changes of Pt2+ in these experiments are negligible, because the surface area available for adsorption is very small (about 8 × 10−5 mm2 ). All adsorption experiments for quantitative analysis and kinetic data acquisition were established at room temperature. 2.3. Self-assembly of HDT (1,6-hexanedithiol) A self-assembled monolayer on a gold substrate was established by immersing the gold substrate in a 1 mM HDT ethanolic solution for 4 h. HDT was used for the thiol functionalization
Fig. 3. In situ SPR measurements of the adsorption kinetics of Pt2+ on an HDT/MUA mixed SAM or an HDT SAM on a gold substrate in an aqueous 0.1 mM Pt2+ solution. The inset shows a topographical AFM image of a mixed HDT/MUA SAM formed on gold substrates. The bright areas are longer carbon chain molecules, MUA in this study, rich domains.
of the gold thin film and it was found to self-assemble well on the gold layer [9,10]. The HDT self-assembled gold substrate was rinsed with ethanol and dried before each experiment. The atomic composition of the HDT self assembled gold substrate was determined by Auger electron spectroscopy; the atomic ratios were determined to be Au:0.7463, C:0.171, S:0.0826. The HDT SAM on a gold layer was also identified by the SPR angle shift in Fig. 2, and this has been described in a previous report [9]. 2.4. Mixed self-assembled layer of MUA (11-mercaptoundecanoic acid) with HDT (1,6-hexanedithiol) A mixed SAM was prepared by soaking a cleaned gold substrate in a mixed solution of 1 mM of HDT with 1 mM of an ethanolic solution of MUA. The solutions were mixed by volume ratios (MUA:HDT = 1:10), and the time of immersion of the mixed SAM was about 15 h. The atomic composition of the mixed SAM surface was determined by Auger electron spectroscopy; the atomic ratios were Au:0.6831, C:0.2005, S:0.0791, O:0.0373. The surface morphologies of the mixed SAM were also analyzed by AFM, and the image is shown in the inset of Fig. 3.
546
J. Moon et al. / Journal of Colloid and Interface Science 298 (2006) 543–549
3. The interpretation of SPR angle shift A six-phase, denoted as a (012345) SPR system, in the Kretschmann configuration using attenuated total reflection (ATR) was employed in the experiments. The schematic diagram of the six-phase SPR system is shown in the right side of Fig. 1. The different phases labeled as follows: (0) glass slide (SF-10, n = 1.78), optically coupled to a 90◦ ATR prism of the same material; (1) a thin binder layer of Cr (about 5 nm); (2) a thin layer of gold film (about 50 nm); (3) a thiol-terminated SAM of HDT and a mixed SAM; (4) a layer of adsorbed Pt2+ ; (5) an ambient dielectric medium of air (for quantitative analysis) or water (for kinetic analysis). The Maxwell Garnett theory was used to convert θSPR to the coverage of adsorbed Pt2+ for the in situ analysis of Pt2+ adsorption onto a thiolated gold substrate, the possible anisotropy of the adsorbed Pt2+ layer was neglected. SPR angle shift was calculated using the Fresnel equations via SPR software (Winspall version 2.20, Max Planck Institute for Polymer Research, Mainz, Germany) and it was normalized from the following equation: Normalized θSPR =
θSPR (t) . Max θSPR
(1)
The maximum SPR angle shift was obtained from SPR measurement for the fully saturated HDT layer with Pt2+ . For spherical inclusions, assuming the metal ion has a spherical shape on the surface, the effective medium dielectric constant (εeffect ) can be calculated from the following equation [26]: (εeffect − ε4 ) = βf, (εeffect + 2ε4 )
where β =
Fig. 4. Calculated change in the relative SPR angle (normalized θSPR ) at a fixed wavelength of incident light as a function of surface coverage. (A) In the case of refractive index (13) a fixed layer with a variable thickness (0–0.6 Å). (B) In the case of thickness (0.6 Å) a fixed layer with a variable effective refractive index (1.7769–13).
(ε3 − ε4 ) . (ε3 + 2ε4 )
response is nearly identical for both cases until a surface coverage of 30% is reached. Moreover the difference between these two cases for the same surface coverage after 30% is negligible. Therefore, it is reasonable to use the assumption of a constant dielectric constant of the Pt2+ layer with a variable thickness, in probing the in situ kinetics of Pt2+ adsorption. 4. Results and discussion
(2)
Here f is the volume fraction, this quantity is proportional to the surface coverage of adsorbed Pt2+ , and ε3 and ε4 are the dielectric constants of the fully adsorbed Pt2+ and the ambient medium, respectively. We assumed that the value of f reaches 1 when the adsorption is fully saturated. If we assume that the thickness of the adsorbed Pt2+ layer is 0.6 Å (the ionic radius of Pt2+ ), the dielectric constant of fully saturated Pt2+ layer is estimated to be 13. An assumption that a partially adsorbed Pt2+ layer is optically equivalent to a fully saturated Pt2+ layer was used for simulation of the correlation for Pt2+ layer coverage with SPR angle shift (θSPR ), because the optical response is proportional to the average film thickness when the adlayer thickness is very small [27]. On the other hand, an assumption that the dielectric constant is variable with a fixed film thickness of the Pt2+ layer could be considered. Changes in θSPR as a function of surface coverage in these two cases, for a fixed thickness with a variable refractive index and for a variable refractive index with a fixed thickness, were calculated in order to compare the difference [28]. For the latter assumption, the effective dielectric constant of the adsorbed Pt2+ layer was obtained using Eq. (2) with an assumption that the surface coverage is varying with the volume fraction, f . The calculated result is shown in Fig. 4, the optical
4.1. SPR angle shift by Pt2+ adsorption on HDT SAM For the full saturation of Pt2+ adsorption on HDT, the substrate was in contact with the solution for about 48 h. The Pt2+ solution was then removed and the substrate was rinsed with water. The SPR angle of the substrate, was measured and the amount of angle shift was calculated from the results. SPR angle-resolved plots for the bare gold substrate, HDT SAMimmobilized gold substrate, and Pt2+ -adsorbed HDT SAM on gold substrates are shown in Fig. 2. We previously reported in detail on a stepwise self-assembled 1,6-hexanedithiol multilayer by SPR [9]. The thickness and the dielectric constant of the HDT layer in this study were estimated to be 5.9 Å and 2.0, respectively. According to the Chempen 3D software (Hilton), the molecular length of the HDT was estimated to be 10.9 Å. Therefore, the HDT layer was estimated to be tilted by an angle of 57◦ to the surface normal. 4.2. Metal ion complexation with surface atoms The prepared thiolated thin gold film substrate, the SPR angle of which was measured, was immersed in the prepared Pt2+ solutions. The concentrations of Pt2+ in the solutions were fixed at 0.0001, 0.001, 0.01, 0.1, and 1 mM, respectively. After about
J. Moon et al. / Journal of Colloid and Interface Science 298 (2006) 543–549
Fig. 5. Quantitative analysis result by SPR angle shifts after the adsorption of Pt2+ for 48 h in water. The concentrations of Pt2+ were 0.00001, 0.0001, 0.001, 0.01, 0.1, and 1 mM.
48 h for saturated adsorption, the substrate was removed and the SPR angle shift, after the adsorption of Pt2+ onto the substrate reached equilibrium, was measured. All substrates were washed with a copious amount of ethanol and water and then dried in air. The general adsorption equilibrium between Pt2+ in aqueous solution and the HDT–Au surface is established with a ratio of m to n as K
mPt2+ (aq) + nS(s) −→ Ptm Sn , [Ptm Sn ] Kad = K[S]n = 2+ m , [Pt ]
(3) (4)
where S is a thiol-terminated HDT SAM surface, and Ptm Sn the complex (multidentate) formed on the surface. The adsorption equilibrium constant can then be expressed by Eq. (4). The concentration of the complex can be approximated as the surface coverage and be proportional to the angle shift, θSPR . Therefore, Eq. (4) can be expressed as log(θSPR ) = m ln[Pt2+ ] + ln
Kad , k
(5)
where k is the correlation constant between the concentration of the complex and θSPR . The SPR angle shifts are dependent on the number of Pt2+ ions adsorbed to the HDT SAM surface. In other words, the more Pt2+ ions that are adsorbed to the HDT SAM surface, the greater will be the observed angle shift. Fig. 5 shows the θSPR on HDT SAM substrates as a function of Pt2+ concentration in the range of 0.01 µM to 1.0 mM. The slope (m = 0.144) in the linear fit of Eq. (5) indicates that one Pt2+ ion requires approximately six “sulfur” atoms for complexation under the given conditions. Considering that the complexation number of Pt2+ ion is 4 and 6, this result is in good agreement with the theoretical prediction. Fig. 5 also shows that the concentration of Pt2+ in an aqueous solution can be quantified within the concentration range of 1.0 × 10−5 M–1.0 mM by linear interpolation.
547
Fig. 6. Selective adsorption kinetics results for Pt2+ compared with Ni, Cu, and Cd ions was observed by real-time SPR monitoring. The concentration of each metal ion was 1 mM and the solutions were altered at 60-min intervals. The arrows indicate the exchange of each of the metal ion solutions.
4.3. Adsorption selectivity of thiolated surface for a target metal ion The thiol functionality is known to be selective for certain metal ions [25]. To identify the selectivity for Pt2+ , in situ measurements of the adsorption for variable metal ions, Ni2+ , Cu2+ , Cd2+ , and Pt2+ , were made. A 1 mM aqueous solution of each metal ion was prepared. The results for the in situ sensing of metal ion adsorptions are shown in Fig. 6. The metal ion solution in the sample cell was exchanged with each other during the 60 min in which the signal of the SPR equipment was acquired in real time. The reflectance increase was negligible in the case of homo-ionic solutions of Ni2+ , Cu2+ , and Cd2+ ions. However the reflectance in the case of a Pt2+ ion solution increased significantly. The ratio of the signal difference for Ni:Cu:Cd:Pt is about 2:1:1:50. This indicates that the HDT SAM can detect the Pt2+ ion selectively in the presence of Ni2+ , Cu2+ , and Cd2+ ions, because the surface functionality of the HDT SAM, a thiol group, has a selective affinity for Pt2+ ions. 4.4. Metal ion adsorption kinetics to HDT SAM The adsorption kinetics for variable concentration of aqueous Pt2+ to HDT SAM were analyzed as shown in Fig. 7. A aqueous Pt2+ solutions were prepared as the concentration of 0.1, 1, and 10 mM. The reflectance change was accumulated from the time that the solution flowed in the sample cell, and the data were collected by a computer at 1-s intervals. At a relatively higher concentration (10 mM), the adsorption kinetics matched Langmuir adsorption kinetics, but the kinetics had a rather S-shape at lower concentrations (0.1, 1 mM). These phenomena indicate that two steps were involved for the diffusion or contact of Pt2+ to the surface and appeared more definitely in the case of lower Pt2+ concentrations. The kinetics of the adsorption of heavy metal ions, such as Pt2+ and Hg2+ to a thiolated surface have been reported to increase as a function of coverage (cooperative adsorption) [22–29]. From the graph
548
J. Moon et al. / Journal of Colloid and Interface Science 298 (2006) 543–549
Scheme 1. Proposed mechanism for the adsorption of Pt2+ ions onto a HDT SAM on a thin gold film assuming a series of two consecutive surface events.
shown in (b) and (c) in Fig. 7, the adsorption profile can be divided into two steps. In the initial phase of the adsorption, the rate of adsorption increases with the coverage of Pt2+ adsorbed to the surface, after which step, the adsorption behavior follows the Langmuir type. Such a non-Langmuirian cooperative adsorption behavior can be explained by a change in the hydrophilicity of the thiolated surface. The environment of the thiolated surface is initially hydrophobic, resulting in initially slow kinetics, but once complexation occurs, the thiolated surface becomes more hydrophilic as the result of the formation of charged –SM+ moieties. With the increase in hydrophilicity permitting the facile approach of heavy, solvated metal ions to the surface, the adsorption kinetics become faster with coverage, after which, the adsorption follows Langmuirian behavior (Scheme 1). The inflection point for these two cases below appears at a coverage of around 17%. After the surface coverage reaches a certain value, the adsorption of Pt2+ follows Langmuirian behavior. In other words, the acceleration in adsorption rate decreases dramatically, when the coverage exceeds 17%. The kinetics then follow Langmuirian behavior because the affinity of metal ions to the surface was not increased after a certain coverage of Pt2+ ion complex on the surface. Consistent with this explanation, we also find that the coverage with time in Fig. 7 follows first-order Langmuir adsorption kinetics [7]: dΓ Γt (6) = k a C0 1 − . dt Γmax Here, Γ (t) is the time-dependent surface coverage for the adsorbed Pt2+ molecules, Γmax is the maximum possible coverage and is assumed to be 1. From the two lower concentration cases, ka in Langmuir adsorption was determined to be 3.0 × 10−2 s−1 from this fit [22]. The rate constant for Pt2+ adsorption obtained in a kinetic analysis with the different initial concentration are same, the parameter describing the two different cases are selfconsistent. On the other hand, the rate constant for the adsorption of Pt2+ at higher concentrations (10 mM) is smaller than the others. This can be attributed to the rate determining step of adsorption being different in the case of a higher concentration (10 mM) because of a limitation in the diffusion rate of the metal ion in aqueous solution [30]. 4.5. Enhancement of adsorption rate using mixed SAM We showed that cooperative adsorption can be explained by a change in the hydrophilicity of the surface as a result of
the Pt2+ adsorption (via Pt2+ –S− complexation). Therefore, a mixed SAM of HDT and 11-mercaptoundecanoic acid (MUA) was fabricated on a gold substrate in order to create a more hydrophilic surface. SPR measurements of the kinetics of Pt2+ ion adsorption on the mixed SAM immobilized surfaces were established, as shown in Fig. 3. The initial concentration of Pt2+ was 0.1 mM. The time required for saturated adsorption was reduced in the mixed SAM compared with the HDT SAM. The results show that the adsorption of Pt2+ to the mixed SAM immobilized surface also followed an S-type adsorption behavior and the adsorption rate was increased compared to the case of HDT SAM. The reason of faster adsorption rate is considered that the hydrophilicity increase of mixed SAM induce the more facile approach of Pt2+ to the surface than HDT SAM. Moreover, the rate of adsorption accelerates until the coverage reaches ca. 10% in the case of the mixed SAM and ca. 17% in the case of the HDT SAM [22]. The experimental results indicate that the amount of Pt2+ ion complex required for the facile approach of solvated Pt2+ ions to the mixed SAM is less than that for the HDT SAM owing to the presence of hydrophilic carboxyl groups. 5. Summary A method suitable for an in situ optical interfacial analysis that uses an SPR technique after modification of a gold substrate with HDT is described. The HDT-fabricated gold substrate clearly promoted an SPR response to the adsorption of Pt2+ , which was then detected using an optical method. We were able to quantify the concentration of Pt2+ ion in the range of concentration of 0.01 µM to 10 mM in an aqueous solution. In situ SPR experiments on HDT–Au with solutions of different heavy metal ions enabled us to selectively differentiate Pt2+ from Ni2+ , Cu2+ , and Cd2+ . In addition, SPR spectroscopy permitted the direct observation of a cooperative mechanism for the adsorption of Pt2+ to a hydrophobic thiolated surface. The adsorption rate accelerates with the extent of adsorbed Pt2+ coverage, until it reaches ca. 17% in the case of an HDT SAM. This indicates that the adsorption characteristics of metal ions to a functionalized surface are strongly dependent on the hydrophilicity of the surface. Surface modification with hydrophilic molecule was investigated to enhance the adsorption rate, the adsorption characteristics of metal ions onto a bifunctionalized surface could be examined. A kinetic analysis of the adsorption of metal ions to a thiolated surface confirmed that the adsorption kinetics could be controlled by surface modification via the use of a mixed SAM. Thus, the observation of
J. Moon et al. / Journal of Colloid and Interface Science 298 (2006) 543–549
549
metal ion adsorption via SPR measurements is expected to be a useful system for the selective, quantitative detection of metal ion in aqueous solutions and to lead to a better understanding of the mechanism of adsorption. Acknowledgments We are grateful to the Eco-Technopia-21 project of the Ministry of Environment, Korea, for financial support. This research was conducted through the Engineering Research Institute (ERI) at Seoul National University, Korea.
(a)
(b)
(c) Fig. 7. Kinetics data for the adsorption of Pt2+ onto a HDT SAM acquired by SPR. The concentrations of Pt2+ were (a) 10 mM, (b) 1 mM, and (c) 0.1 mM. Comparison of in situ adsorption data with first-order Langmuir kinetics as a function of rate constant ka of (A) 5.5 × 10−3 , (B) 3.0 × 10−2 , and (C) 3.0 × 10−2 , respectively.
References [1] Z. Qi, N. Matsuda, J. Santos, T. Yoshida, A. Takatsu, K. Kato, J. Colloid Interface Sci. 271 (2004) 249. [2] Y.-K. Cho, S. Kim, Y.A. Kim, H.K. Lim, K. Lee, D.S. Yoon, G. Lim, Y.E. Pak, T.H. Ha, K. Kim, J. Colloid Interface Sci. 278 (2004) 44. [3] E.S. Gawalt, M. Mrksich, J. Am. Chem. Soc. 126 (2004) 15613. [4] S. Chah, J. Yi, R.N. Zare, Sens. Actuators B 99 (2004) 216. [5] S. Boussaad, J. Pean, N.J. Tao, Anal. Chem. 72 (2000) 222. [6] T. Zacher, E. Wischerhoff, Langmuir 18 (2002) 1748. [7] R. Georgiadis, K.P. Peterlinz, A.W. Peterson, J. Am. Chem. Soc. 122 (2000) 3166. [8] D. Sarkar, P. Somasundaran, J. Colloid Interface Sci. 261 (2003) 197. [9] S. Chah, J.H. Fendler, J. Yi, Chem. Commun. (2002) 2094. [10] S. Chah, J. Yi, C.M. Pettit, D. Roy, J.H. Fendler, Langmuir 18 (2002) 314. [11] S. Chah, E. Hutter, D. Roy, J.H. Fendler, J. Yi, Chem. Phys. 272 (2001) 127. [12] D. Roy, J.H. Fendler, Adv. Mater. 16 (2004) 479. [13] S. Ekgasit, C. Thammacharoen, W. Knoll, Anal. Chem. 76 (2004) 561. [14] W.L. Barnes, A. Dereux, T.W. Ebbesen, Nature 424 (2003) 824. [15] M. Satjapipat, R. Sanedrin, F. Zhou, Langmuir 17 (2001) 7637. [16] W. Knoll, Annu. Rev. Phys. Chem. 49 (1998) 569. [17] R.J. Whelan, R.N. Zare, Anal. Chem. 75 (2003) 1542. [18] S. Fireman-Shoresh, N. Husing, D. Avnir, Langmuir 17 (2001) 5958. [19] A. Ulman, Chem. Rev. 96 (1996) 1533. [20] J.C. Love, L.A. Estroff, J.K. Kriebel, R.G. Nuzzo, G.M. Whitesides, Chem. Rev. 150 (2005) 1103. [21] T. Baumgart, M. Kreiter, H. Lauer, R. Naumann, G. Jung, A. Jonczyk, A. Offenhäusser, W. Knoll, J. Colloid Interface Sci. 258 (2003) 298. [22] T. Kang, J. Moon, S. Oh, S. Hong, S. Chah, J. Yi, Chem. Commun. (2005) 2360. [23] A. Bibby, L. Mercier, Chem. Mater. 14 (2002) 1591. [24] K.H. Nam, S. Gomez-Salazar, L.L. Tavlarides, Ind. Eng. Chem. Res. 42 (2003) 1955. [25] T. Kang, Y. Park, J. Yi, Ind. Eng. Chem. Res. 43 (2004) 1478. [26] R.J. Gehr, R.W. Boyd, Chem. Mater. 8 (1996) 1807. [27] L.S. Jung, C.T. Campbell, T.M. Chinowsky, M.N. Mar, S.S. Yee, Langmuir 14 (1998) 5636. [28] F.S. Damos, R.C.S. Luz, L.T. Kubota, Langmuir 21 (2005) 602. [29] I.L. Lagadic, M.K. Michell, B.D. Payne, Environ. Sci. Technol. 35 (2001) 984. [30] M. Prasad, S. Saxena, Ind. Eng. Chem. Res. 43 (2004) 1512.
In situ sensing of metal ion adsorption to a thiolated surface using surface plasmon resonance spectroscopy Jungwoo Moon, Taewook Kang, Seogil Oh, Surin Hong, Jongheop Yi ∗ School of Chemical and Biological Engineering, Institute of Chemical Processes, Seoul National University, Seoul 151-744, South Korea Received 2 November 2005; accepted 30 December 2005 Available online 3 February 2006
Abstract The kinetics of the adsorption of metal ions onto a thiolated surface and the selective and quantitative sensing of metal ions were explored using surface plasmon resonance (SPR) spectroscopy. The target metal ion was an aqueous solution of Pt2+ and a thin-gold-film-coated glass substrate was modified with 1,6-hexanedithiol (HDT) as a selective sensing layer. SPR spectroscopy was used to examine the kinetics of metal ion adsorption by means of the change in SPR angle. The selectivity of the thiolated surface for Pt2+ over other divalent metal ions such as Cu2+ , Ni2+ , and Cd2+ was evident by the time-resolved SPR measurement. SPR angle shift, θSPR , was found to increase logarithmically with increasing concentration of Pt2+ in the range of 1.0 × 10−5 –1.0 mM. The rate of Pt2+ adsorption on HDT observed at both 0.1 and 1 mM Pt2+ accelerates until the surface coverage reaches approximately 17%, after which the adsorption profile follows Langmuirian behavior with the surface coverage. The experimental data indicated that heavy metal ions were adsorbed to the hydrophobic thiolated surface by a cooperative mechanism. A mixed self-assembled monolayer (SAM) composed of HDT and 11-mercaptoundecanoic acid was used to reduce the hydrophobicity of the thiol-functionalized surface. The addition of hydrophilic groups to the surface enhanced the rate of adsorption of Pt2+ onto the surface. The findings show that the adsorption of metal ions is strongly dependent upon the hydrophilicity/hydrophobicity of the surface and that the technique represents an easy method for analyzing the adsorption of metal ions to a functionalized surface by combining SPR spectroscopy with a SAM modification. © 2006 Elsevier Inc. All rights reserved. Keywords: SPR; Metal ions; Adsorption; Kinetics; Mixed SAM
1. Introduction Reactions that occur at a solid/liquid interface include a number of significant chemical reactions, such as charge transfer, ion exchange, and adsorption. However, the in situ monitoring of the surface change is difficult because analytical methods that can be applied to a solid surface are restricted to certain environments. There are many usable surface analysis tools, such as UV-DRS, AES, AT-IR, and SEM; however, almost all of these techniques require high vacuum conditions or the destruction of the sample and cannot be used for the in situ observation of a surface change [1]. Continuous change in a surface can be estimated by the change in concentration of bulk phase molecules, pH, temperatures, and colors. Labeling is an alternative * Corresponding author.
E-mail address: [email protected] (J. Yi). 0021-9797/$ – see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2005.12.066
for visualizing a surface change, but it requires a spectroscopic or radioactive additive and some types of surface changes cannot be observed by these methods [2]. Because of these difficulties, the mechanistic factors that influence interfacial reactions remain poorly understood [3]. Surface plasmon resonance (SPR) spectroscopy is a powerful tool for the in situ real-time characterization of a solid/liquid interface [4–18]. Surface plasmons are collective oscillations of free electrons at an interface between a thin metal film and a dielectric medium. A surface plasmon wave is resonant with the p-polarized beam at a certain incident angle, and the intensity of the reflected light becomes evanescent. This phenomenon is called surface plasmon resonance. The SPR technique typically involves the optical excitation of surface plasmon polariton waves on a gold or silver film employing the Kretschmann configuration. Reflectivity is measured as a function of the incident angle [4]. The change in the surface results in a shift in
544
J. Moon et al. / Journal of Colloid and Interface Science 298 (2006) 543–549
the SPR angle in an attenuated total reflectance (ATR) curve, because SPR is extremely sensitive to the dielectric constant of the medium immediately adjacent to the metal film. Adsorption of molecules onto the metallic film or conformational changes in the adsorbed molecules can be accurately monitored by optical methods [5]. SPR spectroscopy usually detects surface changes in the boundary in the submicrometer range, about 300 nm in this study, because SPR is an evanescent wave technique, possessing maximum sensitivity on the surface and being characterized by the exponential decay of sensitivity with increasing distance from the surface [6]. The power of an optical sensor experimental approach, in which one of the interacting molecules (the analyte) is free in solution and the other (the ligand) is attached to the sensor surface, is that the formation and decomposition of the analyte/ligand complex can be monitored in situ, yielding kinetic data in real time [7,8]. This technique can be used to monitor in situ interactions of an analyte and a ligand on the surface without the need to label the reactant, as is required in spectroscopic and radioactive probes, and is a noninvasive real-time surface analysis method. The availability of real-time SPR kinetic data provides the possibility of understanding the mechanism of the chemical and/or physical binding events for surface reactions [2,4–8]. A self-assembled monolayer (SAM) is a flexible and simple system that permits the interfacial properties of a metal substrate to be tailored for a well-designed functional surface [19– 21]. Alkanethiols can be fabricated on a gold surface by selfassembly, and the bonding of thiol groups to gold is sufficiently strong to be stable. An attempt was made to identify the adsorption mechanism and kinetics, but an approximation was required because the direct observation of surface change was impossible. A better understanding of the adsorption mechanism could permit adsorption reactions to be controlled and the efficiency of heavy metal ion reduction, the catalyst, metal ion sensor, and other techniques using metal ion adsorption might be made more efficient. The adsorption of heavy metal ions onto a thiol-functionalized surface is a meaningful reaction for reducing the level of environmental toxic heavy metal ions such as Hg2+ [22–24], manufacturing metal-doped catalysts [25], and metal ion sensors [4]. In particular, it should be noted that the adsorption of Pt2+ onto the thiolated surface may be applicable to the qualitative and quantitative analyses of aqueous Hg2+ ions in that thiol compounds have a strong affinity to Hg2+ and Pt2+ . We recently investigated in situ SPR spectroscopy as a metal ion sensor in an aqueous phase [4]. A thiolated gold substrate was used as the sensing layer and the electric field near the surface served as a probe. In this study, SPR spectroscopy was used to examine the adsorption of metal ions to a thiolated surface at the molecular level. A thin-gold-film-coated glass substrate was thiol-functionalized with 1,6-hexanedithiol (HDT) by a selfassembly method. The thiol functionality is known to selectively adsorb heavy metal ions in an aqueous solution [4]. SPR spectroscopy was found to be useful in identifying the selective adsorption to a functionalized surface and in quantitatively
analyzing the concentration of Pt2+ . In addition, the specific adsorption kinetics of the adsorption of Pt2+ to an HDT functionalized surface was observed; that is to say, the adsorption of metal ions onto the hydrophobic thiolated surface followed a cooperative adsorption mechanism and was strongly dependent on the hydrophilicity of the surface. 2. Materials and methods 2.1. Chemicals 1,6-Hexanedithiol (HS(CH2 )6 SH, HDT, Aldrich) and 11mercaptoundecanoic acid (HS(CH2 )10 COOH, MUA, Aldrich) were used as purchased. Potassium tetrachloroplatinate (K2 PtCl4 , Riedel de Haen) was used to prepare a solution of Pt2+ . For the preparation of the other metal ion solutions, cadmium nitrate tetrahydrate, copper(II) nitrate trihydrate, and nickel(II) nitrate hexahydrate were purchased from Junsei. The water used was purified using a Milli-QTM Millipore system. Sulfuric acid (H2 SO4 , Fisher Scientific) and hydrogen peroxide (H2 O2 , Fisher Scientific) were used to clean the microscope slides on which the chromium and gold films were evaporated. Hydrogen chloride (HCl, Fisher Scientific) and sodium hydroxide (NaOH, Mallinckrodt Chemicals) were used to control the pH of the metal ion solutions. The SF10 glass with high reflective index (n = 1.78) was used as a prism and a thin gold film was deposited on a glass slide. The microscope slide glasses (1 in. × 1 in., SF10) were immersed in piranha solution (H2 SO4 :H2 O2 = 7:3 v/v) for purification. They were rinsed several times with water and ethanol before deposition of the thin gold film. 2.2. SPR spectroscopy and analysis method The SPR instrument was fabricated in-house and a schematic diagram of the setup is shown in Fig. 1. A 50-nm thin gold film was prepared by thermal evaporation with a 5-nm Cr adhesion layer. A cleaned gold substrate was attached to an SF10 prism with index matching oil (Cargille Laboratories Inc., certified refractive index liquids, n = 1.730 ± 0.0005). A Teflon cell was attached to the gold substrate. The 635-nm He–Ne diode laser (Power Technology Inc.) was p-polarized and focused with a lens through the prism onto the gold substrate. Both the prism and the gold substrate were mounted on a rotating plate to control the angle of the incident light. The reflectance was measured with a photopower meter (Oriel). Quantitative analysis was investigated with angle-resolved analysis. Usually the regulation levels of the toxic heavy metal ions are below than 1 ppm; thus, the initial concentration of solution was fixed at 0.0001, 0.001, 0.01, 0.1, and 1 mM for the quantitative analysis. For the acquisition of kinetic data, time-resolved SPR angle shifts were measured by the fixed-angle method, which assumed the reflectance change, R, to be linearly correlated with the SPR angle shift, θSPR , in a narrow angle range [6]. The fixed incident angle for measuring the reflectance was adjusted before each experiment. Data for the reflectance at a fixed
J. Moon et al. / Journal of Colloid and Interface Science 298 (2006) 543–549
545
Fig. 1. Schematic diagram of the six-phase SPR system and the experimental setup used for SPR angle measurement. The HDT was self-assembled on a thin gold film, and the metal ion solution was added to the sample cell. The metal ion was adsorbed by the HDT SAM on the thin gold film and the surface change was detected by an optical signal.
Fig. 2. SPR angle, at which the intensity is minimal at the curves, shifts after the self-assembly of HDT on a thin gold film and Pt2+ adsorption onto the HDT SAM. The gold-coated substrate was immersed in a 10 mM ethanolic solution for 4 h to form a HDT self-assembled monolayer on the thin gold film. The HDT self-assembled substrate was then allowed to contact a 10 mM aqueous solution of Pt2+ for 48 h because the adsorption needed to be fully saturated.
incident angle were acquired by a computer in real time. There is a very short time lag for acquiring data using a computer and the data can be acquired within a period of 1 ms. The reflectance at the fixed measuring angle is converted to the SPR angle shift in this method. The molecular level adsorption kinetics of Pt2+ onto HDT ligands was obtained by monitoring the change in reflectance. During the monitoring of the reflectance, the prepared metal ion solutions were allowed to flow into a Teflon cell. The concentration of metal ions was controlled at 0.1, 1, and 10 mM in an aqueous solution, the concentration changes of Pt2+ in these experiments are negligible, because the surface area available for adsorption is very small (about 8 × 10−5 mm2 ). All adsorption experiments for quantitative analysis and kinetic data acquisition were established at room temperature. 2.3. Self-assembly of HDT (1,6-hexanedithiol) A self-assembled monolayer on a gold substrate was established by immersing the gold substrate in a 1 mM HDT ethanolic solution for 4 h. HDT was used for the thiol functionalization
Fig. 3. In situ SPR measurements of the adsorption kinetics of Pt2+ on an HDT/MUA mixed SAM or an HDT SAM on a gold substrate in an aqueous 0.1 mM Pt2+ solution. The inset shows a topographical AFM image of a mixed HDT/MUA SAM formed on gold substrates. The bright areas are longer carbon chain molecules, MUA in this study, rich domains.
of the gold thin film and it was found to self-assemble well on the gold layer [9,10]. The HDT self-assembled gold substrate was rinsed with ethanol and dried before each experiment. The atomic composition of the HDT self assembled gold substrate was determined by Auger electron spectroscopy; the atomic ratios were determined to be Au:0.7463, C:0.171, S:0.0826. The HDT SAM on a gold layer was also identified by the SPR angle shift in Fig. 2, and this has been described in a previous report [9]. 2.4. Mixed self-assembled layer of MUA (11-mercaptoundecanoic acid) with HDT (1,6-hexanedithiol) A mixed SAM was prepared by soaking a cleaned gold substrate in a mixed solution of 1 mM of HDT with 1 mM of an ethanolic solution of MUA. The solutions were mixed by volume ratios (MUA:HDT = 1:10), and the time of immersion of the mixed SAM was about 15 h. The atomic composition of the mixed SAM surface was determined by Auger electron spectroscopy; the atomic ratios were Au:0.6831, C:0.2005, S:0.0791, O:0.0373. The surface morphologies of the mixed SAM were also analyzed by AFM, and the image is shown in the inset of Fig. 3.
546
J. Moon et al. / Journal of Colloid and Interface Science 298 (2006) 543–549
3. The interpretation of SPR angle shift A six-phase, denoted as a (012345) SPR system, in the Kretschmann configuration using attenuated total reflection (ATR) was employed in the experiments. The schematic diagram of the six-phase SPR system is shown in the right side of Fig. 1. The different phases labeled as follows: (0) glass slide (SF-10, n = 1.78), optically coupled to a 90◦ ATR prism of the same material; (1) a thin binder layer of Cr (about 5 nm); (2) a thin layer of gold film (about 50 nm); (3) a thiol-terminated SAM of HDT and a mixed SAM; (4) a layer of adsorbed Pt2+ ; (5) an ambient dielectric medium of air (for quantitative analysis) or water (for kinetic analysis). The Maxwell Garnett theory was used to convert θSPR to the coverage of adsorbed Pt2+ for the in situ analysis of Pt2+ adsorption onto a thiolated gold substrate, the possible anisotropy of the adsorbed Pt2+ layer was neglected. SPR angle shift was calculated using the Fresnel equations via SPR software (Winspall version 2.20, Max Planck Institute for Polymer Research, Mainz, Germany) and it was normalized from the following equation: Normalized θSPR =
θSPR (t) . Max θSPR
(1)
The maximum SPR angle shift was obtained from SPR measurement for the fully saturated HDT layer with Pt2+ . For spherical inclusions, assuming the metal ion has a spherical shape on the surface, the effective medium dielectric constant (εeffect ) can be calculated from the following equation [26]: (εeffect − ε4 ) = βf, (εeffect + 2ε4 )
where β =
Fig. 4. Calculated change in the relative SPR angle (normalized θSPR ) at a fixed wavelength of incident light as a function of surface coverage. (A) In the case of refractive index (13) a fixed layer with a variable thickness (0–0.6 Å). (B) In the case of thickness (0.6 Å) a fixed layer with a variable effective refractive index (1.7769–13).
(ε3 − ε4 ) . (ε3 + 2ε4 )
response is nearly identical for both cases until a surface coverage of 30% is reached. Moreover the difference between these two cases for the same surface coverage after 30% is negligible. Therefore, it is reasonable to use the assumption of a constant dielectric constant of the Pt2+ layer with a variable thickness, in probing the in situ kinetics of Pt2+ adsorption. 4. Results and discussion
(2)
Here f is the volume fraction, this quantity is proportional to the surface coverage of adsorbed Pt2+ , and ε3 and ε4 are the dielectric constants of the fully adsorbed Pt2+ and the ambient medium, respectively. We assumed that the value of f reaches 1 when the adsorption is fully saturated. If we assume that the thickness of the adsorbed Pt2+ layer is 0.6 Å (the ionic radius of Pt2+ ), the dielectric constant of fully saturated Pt2+ layer is estimated to be 13. An assumption that a partially adsorbed Pt2+ layer is optically equivalent to a fully saturated Pt2+ layer was used for simulation of the correlation for Pt2+ layer coverage with SPR angle shift (θSPR ), because the optical response is proportional to the average film thickness when the adlayer thickness is very small [27]. On the other hand, an assumption that the dielectric constant is variable with a fixed film thickness of the Pt2+ layer could be considered. Changes in θSPR as a function of surface coverage in these two cases, for a fixed thickness with a variable refractive index and for a variable refractive index with a fixed thickness, were calculated in order to compare the difference [28]. For the latter assumption, the effective dielectric constant of the adsorbed Pt2+ layer was obtained using Eq. (2) with an assumption that the surface coverage is varying with the volume fraction, f . The calculated result is shown in Fig. 4, the optical
4.1. SPR angle shift by Pt2+ adsorption on HDT SAM For the full saturation of Pt2+ adsorption on HDT, the substrate was in contact with the solution for about 48 h. The Pt2+ solution was then removed and the substrate was rinsed with water. The SPR angle of the substrate, was measured and the amount of angle shift was calculated from the results. SPR angle-resolved plots for the bare gold substrate, HDT SAMimmobilized gold substrate, and Pt2+ -adsorbed HDT SAM on gold substrates are shown in Fig. 2. We previously reported in detail on a stepwise self-assembled 1,6-hexanedithiol multilayer by SPR [9]. The thickness and the dielectric constant of the HDT layer in this study were estimated to be 5.9 Å and 2.0, respectively. According to the Chempen 3D software (Hilton), the molecular length of the HDT was estimated to be 10.9 Å. Therefore, the HDT layer was estimated to be tilted by an angle of 57◦ to the surface normal. 4.2. Metal ion complexation with surface atoms The prepared thiolated thin gold film substrate, the SPR angle of which was measured, was immersed in the prepared Pt2+ solutions. The concentrations of Pt2+ in the solutions were fixed at 0.0001, 0.001, 0.01, 0.1, and 1 mM, respectively. After about
J. Moon et al. / Journal of Colloid and Interface Science 298 (2006) 543–549
Fig. 5. Quantitative analysis result by SPR angle shifts after the adsorption of Pt2+ for 48 h in water. The concentrations of Pt2+ were 0.00001, 0.0001, 0.001, 0.01, 0.1, and 1 mM.
48 h for saturated adsorption, the substrate was removed and the SPR angle shift, after the adsorption of Pt2+ onto the substrate reached equilibrium, was measured. All substrates were washed with a copious amount of ethanol and water and then dried in air. The general adsorption equilibrium between Pt2+ in aqueous solution and the HDT–Au surface is established with a ratio of m to n as K
mPt2+ (aq) + nS(s) −→ Ptm Sn , [Ptm Sn ] Kad = K[S]n = 2+ m , [Pt ]
(3) (4)
where S is a thiol-terminated HDT SAM surface, and Ptm Sn the complex (multidentate) formed on the surface. The adsorption equilibrium constant can then be expressed by Eq. (4). The concentration of the complex can be approximated as the surface coverage and be proportional to the angle shift, θSPR . Therefore, Eq. (4) can be expressed as log(θSPR ) = m ln[Pt2+ ] + ln
Kad , k
(5)
where k is the correlation constant between the concentration of the complex and θSPR . The SPR angle shifts are dependent on the number of Pt2+ ions adsorbed to the HDT SAM surface. In other words, the more Pt2+ ions that are adsorbed to the HDT SAM surface, the greater will be the observed angle shift. Fig. 5 shows the θSPR on HDT SAM substrates as a function of Pt2+ concentration in the range of 0.01 µM to 1.0 mM. The slope (m = 0.144) in the linear fit of Eq. (5) indicates that one Pt2+ ion requires approximately six “sulfur” atoms for complexation under the given conditions. Considering that the complexation number of Pt2+ ion is 4 and 6, this result is in good agreement with the theoretical prediction. Fig. 5 also shows that the concentration of Pt2+ in an aqueous solution can be quantified within the concentration range of 1.0 × 10−5 M–1.0 mM by linear interpolation.
547
Fig. 6. Selective adsorption kinetics results for Pt2+ compared with Ni, Cu, and Cd ions was observed by real-time SPR monitoring. The concentration of each metal ion was 1 mM and the solutions were altered at 60-min intervals. The arrows indicate the exchange of each of the metal ion solutions.
4.3. Adsorption selectivity of thiolated surface for a target metal ion The thiol functionality is known to be selective for certain metal ions [25]. To identify the selectivity for Pt2+ , in situ measurements of the adsorption for variable metal ions, Ni2+ , Cu2+ , Cd2+ , and Pt2+ , were made. A 1 mM aqueous solution of each metal ion was prepared. The results for the in situ sensing of metal ion adsorptions are shown in Fig. 6. The metal ion solution in the sample cell was exchanged with each other during the 60 min in which the signal of the SPR equipment was acquired in real time. The reflectance increase was negligible in the case of homo-ionic solutions of Ni2+ , Cu2+ , and Cd2+ ions. However the reflectance in the case of a Pt2+ ion solution increased significantly. The ratio of the signal difference for Ni:Cu:Cd:Pt is about 2:1:1:50. This indicates that the HDT SAM can detect the Pt2+ ion selectively in the presence of Ni2+ , Cu2+ , and Cd2+ ions, because the surface functionality of the HDT SAM, a thiol group, has a selective affinity for Pt2+ ions. 4.4. Metal ion adsorption kinetics to HDT SAM The adsorption kinetics for variable concentration of aqueous Pt2+ to HDT SAM were analyzed as shown in Fig. 7. A aqueous Pt2+ solutions were prepared as the concentration of 0.1, 1, and 10 mM. The reflectance change was accumulated from the time that the solution flowed in the sample cell, and the data were collected by a computer at 1-s intervals. At a relatively higher concentration (10 mM), the adsorption kinetics matched Langmuir adsorption kinetics, but the kinetics had a rather S-shape at lower concentrations (0.1, 1 mM). These phenomena indicate that two steps were involved for the diffusion or contact of Pt2+ to the surface and appeared more definitely in the case of lower Pt2+ concentrations. The kinetics of the adsorption of heavy metal ions, such as Pt2+ and Hg2+ to a thiolated surface have been reported to increase as a function of coverage (cooperative adsorption) [22–29]. From the graph
548
J. Moon et al. / Journal of Colloid and Interface Science 298 (2006) 543–549
Scheme 1. Proposed mechanism for the adsorption of Pt2+ ions onto a HDT SAM on a thin gold film assuming a series of two consecutive surface events.
shown in (b) and (c) in Fig. 7, the adsorption profile can be divided into two steps. In the initial phase of the adsorption, the rate of adsorption increases with the coverage of Pt2+ adsorbed to the surface, after which step, the adsorption behavior follows the Langmuir type. Such a non-Langmuirian cooperative adsorption behavior can be explained by a change in the hydrophilicity of the thiolated surface. The environment of the thiolated surface is initially hydrophobic, resulting in initially slow kinetics, but once complexation occurs, the thiolated surface becomes more hydrophilic as the result of the formation of charged –SM+ moieties. With the increase in hydrophilicity permitting the facile approach of heavy, solvated metal ions to the surface, the adsorption kinetics become faster with coverage, after which, the adsorption follows Langmuirian behavior (Scheme 1). The inflection point for these two cases below appears at a coverage of around 17%. After the surface coverage reaches a certain value, the adsorption of Pt2+ follows Langmuirian behavior. In other words, the acceleration in adsorption rate decreases dramatically, when the coverage exceeds 17%. The kinetics then follow Langmuirian behavior because the affinity of metal ions to the surface was not increased after a certain coverage of Pt2+ ion complex on the surface. Consistent with this explanation, we also find that the coverage with time in Fig. 7 follows first-order Langmuir adsorption kinetics [7]: dΓ Γt (6) = k a C0 1 − . dt Γmax Here, Γ (t) is the time-dependent surface coverage for the adsorbed Pt2+ molecules, Γmax is the maximum possible coverage and is assumed to be 1. From the two lower concentration cases, ka in Langmuir adsorption was determined to be 3.0 × 10−2 s−1 from this fit [22]. The rate constant for Pt2+ adsorption obtained in a kinetic analysis with the different initial concentration are same, the parameter describing the two different cases are selfconsistent. On the other hand, the rate constant for the adsorption of Pt2+ at higher concentrations (10 mM) is smaller than the others. This can be attributed to the rate determining step of adsorption being different in the case of a higher concentration (10 mM) because of a limitation in the diffusion rate of the metal ion in aqueous solution [30]. 4.5. Enhancement of adsorption rate using mixed SAM We showed that cooperative adsorption can be explained by a change in the hydrophilicity of the surface as a result of
the Pt2+ adsorption (via Pt2+ –S− complexation). Therefore, a mixed SAM of HDT and 11-mercaptoundecanoic acid (MUA) was fabricated on a gold substrate in order to create a more hydrophilic surface. SPR measurements of the kinetics of Pt2+ ion adsorption on the mixed SAM immobilized surfaces were established, as shown in Fig. 3. The initial concentration of Pt2+ was 0.1 mM. The time required for saturated adsorption was reduced in the mixed SAM compared with the HDT SAM. The results show that the adsorption of Pt2+ to the mixed SAM immobilized surface also followed an S-type adsorption behavior and the adsorption rate was increased compared to the case of HDT SAM. The reason of faster adsorption rate is considered that the hydrophilicity increase of mixed SAM induce the more facile approach of Pt2+ to the surface than HDT SAM. Moreover, the rate of adsorption accelerates until the coverage reaches ca. 10% in the case of the mixed SAM and ca. 17% in the case of the HDT SAM [22]. The experimental results indicate that the amount of Pt2+ ion complex required for the facile approach of solvated Pt2+ ions to the mixed SAM is less than that for the HDT SAM owing to the presence of hydrophilic carboxyl groups. 5. Summary A method suitable for an in situ optical interfacial analysis that uses an SPR technique after modification of a gold substrate with HDT is described. The HDT-fabricated gold substrate clearly promoted an SPR response to the adsorption of Pt2+ , which was then detected using an optical method. We were able to quantify the concentration of Pt2+ ion in the range of concentration of 0.01 µM to 10 mM in an aqueous solution. In situ SPR experiments on HDT–Au with solutions of different heavy metal ions enabled us to selectively differentiate Pt2+ from Ni2+ , Cu2+ , and Cd2+ . In addition, SPR spectroscopy permitted the direct observation of a cooperative mechanism for the adsorption of Pt2+ to a hydrophobic thiolated surface. The adsorption rate accelerates with the extent of adsorbed Pt2+ coverage, until it reaches ca. 17% in the case of an HDT SAM. This indicates that the adsorption characteristics of metal ions to a functionalized surface are strongly dependent on the hydrophilicity of the surface. Surface modification with hydrophilic molecule was investigated to enhance the adsorption rate, the adsorption characteristics of metal ions onto a bifunctionalized surface could be examined. A kinetic analysis of the adsorption of metal ions to a thiolated surface confirmed that the adsorption kinetics could be controlled by surface modification via the use of a mixed SAM. Thus, the observation of
J. Moon et al. / Journal of Colloid and Interface Science 298 (2006) 543–549
549
metal ion adsorption via SPR measurements is expected to be a useful system for the selective, quantitative detection of metal ion in aqueous solutions and to lead to a better understanding of the mechanism of adsorption. Acknowledgments We are grateful to the Eco-Technopia-21 project of the Ministry of Environment, Korea, for financial support. This research was conducted through the Engineering Research Institute (ERI) at Seoul National University, Korea.
(a)
(b)
(c) Fig. 7. Kinetics data for the adsorption of Pt2+ onto a HDT SAM acquired by SPR. The concentrations of Pt2+ were (a) 10 mM, (b) 1 mM, and (c) 0.1 mM. Comparison of in situ adsorption data with first-order Langmuir kinetics as a function of rate constant ka of (A) 5.5 × 10−3 , (B) 3.0 × 10−2 , and (C) 3.0 × 10−2 , respectively.
References [1] Z. Qi, N. Matsuda, J. Santos, T. Yoshida, A. Takatsu, K. Kato, J. Colloid Interface Sci. 271 (2004) 249. [2] Y.-K. Cho, S. Kim, Y.A. Kim, H.K. Lim, K. Lee, D.S. Yoon, G. Lim, Y.E. Pak, T.H. Ha, K. Kim, J. Colloid Interface Sci. 278 (2004) 44. [3] E.S. Gawalt, M. Mrksich, J. Am. Chem. Soc. 126 (2004) 15613. [4] S. Chah, J. Yi, R.N. Zare, Sens. Actuators B 99 (2004) 216. [5] S. Boussaad, J. Pean, N.J. Tao, Anal. Chem. 72 (2000) 222. [6] T. Zacher, E. Wischerhoff, Langmuir 18 (2002) 1748. [7] R. Georgiadis, K.P. Peterlinz, A.W. Peterson, J. Am. Chem. Soc. 122 (2000) 3166. [8] D. Sarkar, P. Somasundaran, J. Colloid Interface Sci. 261 (2003) 197. [9] S. Chah, J.H. Fendler, J. Yi, Chem. Commun. (2002) 2094. [10] S. Chah, J. Yi, C.M. Pettit, D. Roy, J.H. Fendler, Langmuir 18 (2002) 314. [11] S. Chah, E. Hutter, D. Roy, J.H. Fendler, J. Yi, Chem. Phys. 272 (2001) 127. [12] D. Roy, J.H. Fendler, Adv. Mater. 16 (2004) 479. [13] S. Ekgasit, C. Thammacharoen, W. Knoll, Anal. Chem. 76 (2004) 561. [14] W.L. Barnes, A. Dereux, T.W. Ebbesen, Nature 424 (2003) 824. [15] M. Satjapipat, R. Sanedrin, F. Zhou, Langmuir 17 (2001) 7637. [16] W. Knoll, Annu. Rev. Phys. Chem. 49 (1998) 569. [17] R.J. Whelan, R.N. Zare, Anal. Chem. 75 (2003) 1542. [18] S. Fireman-Shoresh, N. Husing, D. Avnir, Langmuir 17 (2001) 5958. [19] A. Ulman, Chem. Rev. 96 (1996) 1533. [20] J.C. Love, L.A. Estroff, J.K. Kriebel, R.G. Nuzzo, G.M. Whitesides, Chem. Rev. 150 (2005) 1103. [21] T. Baumgart, M. Kreiter, H. Lauer, R. Naumann, G. Jung, A. Jonczyk, A. Offenhäusser, W. Knoll, J. Colloid Interface Sci. 258 (2003) 298. [22] T. Kang, J. Moon, S. Oh, S. Hong, S. Chah, J. Yi, Chem. Commun. (2005) 2360. [23] A. Bibby, L. Mercier, Chem. Mater. 14 (2002) 1591. [24] K.H. Nam, S. Gomez-Salazar, L.L. Tavlarides, Ind. Eng. Chem. Res. 42 (2003) 1955. [25] T. Kang, Y. Park, J. Yi, Ind. Eng. Chem. Res. 43 (2004) 1478. [26] R.J. Gehr, R.W. Boyd, Chem. Mater. 8 (1996) 1807. [27] L.S. Jung, C.T. Campbell, T.M. Chinowsky, M.N. Mar, S.S. Yee, Langmuir 14 (1998) 5636. [28] F.S. Damos, R.C.S. Luz, L.T. Kubota, Langmuir 21 (2005) 602. [29] I.L. Lagadic, M.K. Michell, B.D. Payne, Environ. Sci. Technol. 35 (2001) 984. [30] M. Prasad, S. Saxena, Ind. Eng. Chem. Res. 43 (2004) 1512.