- Email: [email protected]
Potential driven deposition of polyelectrolytes onto the surface of cysteine monolayers assembled on gold
Journal of Colloid and Interface Science 342 (2010) 499–504
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Potential driven deposition of polyelectrolytes onto the surface of cysteine monolayers assembled on gold Wesley Sanders a, Mark R. Anderson b,* a b
Department of Chemistry, Virginia Polytechnic Institute and State University, Blacksburg, VA 24060-0212, United States University of Colorado Denver, Department of Chemistry, Campus Box 194, PO Box 173364, Denver, CO 80217-3364, United States
a r t i c l e
i n f o
Article history: Received 10 September 2009 Accepted 15 October 2009 Available online 21 October 2009 Keywords: Self-assembled monolayers Electrostatic deposition Impedance spectroscopy
a b s t r a c t Electrochemical impedance spectroscopy and the quartz crystal microbalance measurements are used to examine the ability of potential applied to a substrate to create, in situ, conditions favorable for the electrostatic deposition of polyelectrolytes onto a gold substrate modified by the self-assembly of cysteine. Cysteine is a zwitterionic compound that, when confined to a substrate, has the ability to establish either a net positive or a net negative interfacial charge, depending on the conditions. As such, cysteine modified interfaces could possibly be used as a versatile substrate for deposition of either cationic or anionic polyelectrolytes. The potential of zero charge of a gold electrode modified by self-assembly with cysteine in the presence of 0.10 mol L1 KCl and buffered at pH 5 is found by differential capacitance measurement to be 0.12(±0.02) V vs. Ag/AgCl. When 0.05 V vs. Ag/AgCl is applied to the substrate (a potential positive of the PZC) in the presence of different polyelectrolytes, both impedance spectroscopy and quartz crystal microbalance data suggest the accumulation of anionic poly(sodium styrenesulfonate) along the cysteine modified interface. Conversely, when 0.40 V vs. Ag/AgCl is applied to the substrate (a potential negative of the PZC), experimental results suggest the accumulation of cationic poly(diallydimethylammonium chloride). Ó 2009 Elsevier Inc. All rights reserved.
1. Introduction Electrostatic and molecular self-assembly are relatively simple methods that have been widely applied to create chemically modified interfaces. Each of these methods take advantage of the through-space interactions that exist between molecules to drive their assembly onto solid substrates. This suggests that experimental manipulation of these interactions may provide some measure of control over the assembly process, and potentially can be used to control the structure and/or properties of the modified interface. Several research groups have looked at the role that different intermolecular interactions play in establishing the ensemble structure and properties of self-assembled monolayers. Bain and Whitesides generated modified interfaces by mixing two different mercaptans in the adsorption solution [1–3]. They found that the surface composition did not linearly track the composition of the adsorption solution; rather, differences in the extent of the through-space interactions among the molecules played a significant role in determining the composition of the interfacial layer. Subsequent studies by Hobara et al. [4–6] show that often with these mixed monolayers, the different components of the adsorption solution will phase segregate when they adsorb to the sur* Corresponding author. Fax: +1 303 556 4776. E-mail address: [email protected] (M.R. Anderson). 0021-9797/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2009.10.033
face. They propose that the phase segregation can be used to create two-dimensional molecular patterns along the interface. These results show that the strength of lateral through-space interactions help to establish the structure of self-assembling interfacial systems. Others have effectively used through-space interactions to create three-dimensional interfacial structures by depositing polymeric materials on top of the initial monolayer. Decher et al. take advantage of electrostatic interactions to deposit alternating layers of polyelectrolytes onto surfaces [7–10]. Electrostatic deposition is a versatile, robust method that has been applied for the deposition of a variety of ionic materials, including proteins and enzymes [11–13]. Hammond et al. leverage both lateral interactions and layer-by-layer deposition by combining contact printing with electrostatic deposition to create complex three-dimensional structures at interfaces [14–16]. Others show that experimentally modulating the through-space interactions along an interface can change the structure and alter the properties of the modified interface. For example, Willner et al. created a monolayer in which access to the substrate is modulated by experimentally inducing an isomerization of the molecules that comprise the monolayer [17–20]. Upon isomerization, different steric interactions exist at the interface that alter the physical properties (e.g. the permeability of electroactive species to the substrate surface) of the modified interface.
500
W. Sanders, M.R. Anderson / Journal of Colloid and Interface Science 342 (2010) 499–504
Our interest is in creating a modified interface and then manipulating the properties of the interface so that the structure can be subsequently modified in a controlled, reproducible fashion [21,22]. We previously demonstrated that potential applied to a substrate modified with a 3-mercaptopropionic acid monolayer could be used to modulate the properties of the modified interface to drive the electrostatic assembly of polycationic poly(diallydimethyl ammonium chloride), PDDA, onto the modified substrate [21]. In this application, potential applied to the substrate alters the interfacial interactions from conditions that do not favor polyelectrolyte deposition to conditions where polycation deposition is favored. We found that when potentials positive of the potential of zero charge (PZC) were applied to the substrate, the cationic polymer PDDA deposits onto the modified substrate, while PDDA does not adsorb if potentials negative of the PZC are applied. This result was interpreted as being due to the pH local to the interface becoming more basic when potentials more positive than the PZC are applied to the substrate, and the pH change causing the deprotonation of the terminal acid groups. Once deprotonated, the anionic charge of the interface creates conditions favorable for the deposition of a cationic polymer. A cysteine monolayer has the ability to have either a net positive or a net negative charge depending on the solution pH. We recently demonstrated that this property could be leveraged to electrostatically deposit either polycationic or polyanionic polymers onto a cysteine monolayer by adjusting the bulk solution pH [22]. In this manuscript, we explore using applied potential to drive the adsorption of either polycationic PDDA or polyanionic poly(sodium styrenesulfonate), PSS, onto a substrate modified with cysteine. 2. Experimental 2.1. Chemicals Cysteine, poly(diallyldimethylammonium chloride) and poly (sodium styrene sulfonate) were purchased from the Aldrich Chemical Company (Milwaukee, WI). Potassium hexacyanoferrate (II) trihydrate, potassium hexacyanoferrate (III), potassium hydrogen phosphate monohydrate, potassium dihydrogen phosphate monohydrate, phosphoric acid and potassium hydroxide were purchased from Fischer Scientific company. Poly(diallyldimethylammonium chloride), PDDA, was low molecular weight, 20% in water. Poly (sodium styrene sulfonate), PSS, was 20 wt.% in water. All chemicals were analytical grade and were used without further purification. Unless given otherwise, all solutions were prepared with water deionized with an 18 MX Milli-Q ion exchange filter from Millipore Incorporated.
2.3. Measurements Electrochemical impedance measurements were conducted using a CH Instruments (Austin, TX) model 604B electrochemical analyzer. The impedance measurements were performed in a standard three electrode cell with the modified gold electrode serving as the working electrode, a platinum wire as the auxiliary electrode, and a Ag/AgCl reference electrode. The supporting electrodissolved in 0.10 lyte consisted of 0.005 mol L1 FeðCNÞ3=4 6 mol L1 KCl and a phosphate buffer system containing a total phosphate concentration of 0.050 mol L1 and adjusted to pH 5. Impedance data was obtained at frequencies ranging from 100,000 to 0.1 Hz using a 5 mV amplitude sinusoidal potential modulation that is centered about the formal potential of the redox couple. Quantitative estimates of the charge-transfer resistance are obtained by fitting the experimental data to the Randles equivalent circuit using the nonlinear least squares fitting routines of the software package LEVM 7.0 (available from Solartron, www.solartronanalytical.com). For the potential induced polyelectrolyte deposition experiments, cysteine modified electrodes were placed in an aqueous phosphate buffer solution (pH 5) that contains 0.10 mol L1 KCl and either 1.8 104 mol L1 PDDA or 1.8 104 mol L1 PSS (the concentration is determined with respect to the monomer formula weight) and subject to applied potentials that are either positive or negative of the potential of zero charge for 60 s. The potential was then returned to the open circuit value and the substrate was then removed from this solution and placed into an electrolyte solution composed of 0.005 mol L1 FeðCNÞ63=4 , 0.10 mol L1 KCl, and pH 5 phosphate buffer for the impedance measurements. These experimental conditions were chosen to be consistent with similar experiments conducted previously with a gold substrate modified with a 3-mercaptopropionic acid monolayer [21]. Quartz crystal microbalance measurements were performed with an in-lab constructed QCM oscillator connected to an HP model 5334B frequency analyzer [23]. Quartz crystals from International Crystal Manufacturing (Oklahoma City, OK) having a 1.3 cm diameter gold electrode and 5 MHz resonant frequency were used as the substrate. Prior to cysteine immobilization onto the crystal’s gold electrodes, the quartz crystals were immersed in piranha solution (3:1 concentrated H2SO4:30% H2O2) for less than one minute to clean the surface. They were then rinsed with water, dried in a stream of N2, and immersed in a hexane solution containing 0.005 mol L1 cysteine. The quartz crystal frequencies were measured in air for 15 min before and after the monolayer modified crystals were exposed to the different experimental conditions (described above). During the measurement, the crystal frequency was sampled at 1 min intervals. For each trial, the measured frequency is the average of these 15 samples. The frequency changes reported are an average of measurements conducted for three different modified crystals.
2.2. Monolayer preparation Two millimeter gold disk working electrodes (CH Instruments, Austin, TX) were polished with 0.05 lm alumina followed by sonication in deionized water. The electrodes were then electrochemically cleaned in 0.5 mol L1 sulfuric acid by cycling the potential between 2.0 and 0.8 V vs. Ag–AgCl at 0.050 V/s for 25 complete cycles. Following the electrochemical cleaning, the electrodes were rinsed with deionized water, dried in a stream of nitrogen, and then immersed in a solution of hexane containing 0.005 mol L1 cysteine for 15 min. After cysteine immobilization, the electrode was rinsed with ethanol and deionized water. Reductive desorption measurements were conducted with electrodes modified by this method, and the surface coverage was found to be 4.8(±0.9) 1010 mol cm2.
3. Results For all measurements, the solution is buffered at pH 5, a value close to the isoelectric point for cysteine [24]. At this pH, the monolayer should be net neutral and provide no driving force for the electrostatic deposition of polyelectrolyte. Differential capacitance measurements are used to determine the potential of zero charge for the cysteine monolayer modified Au substrate [25,26]. From these measurements, the PZC is found to be 0.12(±0.05) V vs. Ag/AgCl. This value is consistent with experimental PZC’s measured with other monolayer modified Au electrodes [25,26]. Analogous to the observed behavior of 3-mercaptopropionic acid monolayers [21], by applying potential positive or negative
W. Sanders, M.R. Anderson / Journal of Colloid and Interface Science 342 (2010) 499–504
501
of the PZC we anticipated creating conditions that favor either a net negative or a net positive charge, respectively, of the cysteine monolayer due to the potential induced changes to the interfacial pH. We previously demonstrated that altering the net charge of the cysteine monolayer by adjusting the bulk solution pH in the absence of applied potential, the electrostatic deposition of either PDDA (when using basic pH) or PSS (when using acidic pH) could be accomplished [22]. Others have also shown that changes to bulk pH create electrostatic interactions at an ionizable interface capable of driving the deposition of polyelectrolytes [8,16]. By analogy, we anticipate that interfacial pH changes brought about by changes to the substrate potential should allow the electrostatic deposition of an appropriately charged polyelectrolyte. Changes to the interfacial properties brought about by the deposition of a polyelectrolyte are monitored by impedance spectroscopy. Fig. 1 shows the impedance response of a cysteine monolayer modified interface before (open squares) and after (closed squares) exposure of the interface to a solution containing 1.8 104 mol L1 PDDA and a potential positive of the PZC (Eapplied = 0.05 V vs. Ag–AgCl) applied to the substrate. Several research groups show that electrostatic interactions between a Redox probe and the monolayer modified interface influence the measured impedance [27,28]. Under these experimental conditions, the impedance decreases slightly after exposure of the interface to PDDA and positive potential applied to the substrate. This result suggests that there is a small change in the properties of the interface after the application of potential; however, the observed impedance change is not nearly as large as that found when conducting similar experiments with a 3-mercaptopropionic acid monolayer [29], nor is it as large as the impedance change found before and after exposure of the cysteine monolayer modified interface to a 0.10 mol L1 NaOH solution containing 1.8 104 mol L1 PDDA [22]. The magnitude of the impedance change found with the cysteine monolayer when subjected to positive potentials in the presence of PDDA suggests that only a small amount of the PDDA is deposited. Potentials positive of the PZC were also applied to cysteinemodified substrates in the presence of the anionic polyelectrolyte PSS, and the impedance subsequently measured (Fig. 1). Under these conditions, the impedance is found to have a larger change after the application of potential than the impedance decrease seen with exposure to PDDA. The increased impedance is characteristic of electrostatic repulsion between the FeðCNÞ64=3 redox probe and the modified interface, and suggests that PSS deposition occurred when the positive potential was applied to the substrate. This
was an unexpected outcome as we anticipated that the positive potential applied to the substrate would establish a pH local to the interface that is basic of the bulk pH (buffered at pH 5) and create conditions that favor a net negative charge of the cysteine monolayer. If the interface had a net negative charge, those conditions would inhibit deposition of the polyanionic PSS, the opposite behavior of what is observed experimentally. The magnitude of the impedance change when measured with these modified substrates is characteristic of either the amount of material deposited onto the monolayer, the charge density of the interface, or to some combination of these two parameters [27–29]. Quartz crystal microbalance measurements were conducted using these same experimental conditions to estimate the amount of PDDA and PSS deposition under this experimental condition. After a potential positive of the PZC is applied to the cysteine-modified substrate in the presence of 1.8 104 mol L1 PDDA, a frequency decrease of 1.4(±0.6) Hz is measured. This frequency change corresponds to deposition of 32(±14) ng of PDDA polymer deposited. In contrast, when the QCM experiment is conducted in the presence of 1.8 104 mol L1 PSS, the frequency decreases by 5.3(±0.4) Hz. This frequency decrease corresponds to a deposition of 124(±9) ng of PSS polymer. The results of the impedance and QCM experiments show that, when potentials positive of the PZC are applied to the substrate, deposition of the anionic PSS is favored relative to that of cationic PDDA. This behavior, combined with the impedance data, suggests that when potentials positive of the PZC are applied to the cysteine monolayer modifed substrate, a positive charge along the monolayer/solution interface develops. The experimental evidence that PSS deposition is favored when potentials positive of the PZC are applied to the substrate suggests that a mechanism other than the applied potential influencing the solution pH local to the interface (which in turn influences the net charge of the monolayer) is at work. If this is the case, then the behavior when potentials negative of the PZC are applied should also be different from that expected if the local pH were controlling the net charge of the cysteine monolayer. Fig. 2 shows the impedance of the cysteine modified interface before and after a potential negative of the PZC is applied to the substrate in the presence of 1.8 104 mol L1 PDDA. Under these conditions, the impedance measured with an anionic redox probe decreases after the application of the negative potential, consistent with an electrostatic attraction between the FeðCNÞ64=3 redox probe and the monolayer modified substrate. This EIS result is characteristic of an increase in
Fig. 1. Representative complex impedance plots obtained by electrochemical impedance spectroscopy measured with a gold electrode modified with a layer of cysteine exposed to a 0.10 mol L1 KCl solution buffered at pH 5 containing 1 FeðCNÞ3 0.005 mol L1 FeðCNÞ4 6 =0:005 mol L 6 (i) before (open squares) and after (solid squares) application of a potential positive of the potential of zero charge in the presence of 1.8 104 mol L1 PDDA and (ii) before (open circles) and after (solid circles) application of a potential positive of the potential of zero charge in the presence of 1.8 104 mol L1 PSS.
Fig. 2. Representative complex impedance plots obtained by electrochemical impedance spectroscopy measured with a gold electrode modified with a layer of cysteine exposed to a 0.10 mol L1 KCl solution buffered at pH 5 containing 1 FeðCNÞ3 0.005 mol L1 FeðCNÞ4 6 =0:005 mol L 6 (i) before (open squares) and after (solid squares) application of a potential negative of the potential of zero charge in the presence of 1.8 104 mol L1 PDDA and (ii) before (open circles) and after (solid circles) application of a potential positive of the potential of zero charge in the presence of 1.8 104 mol L1 PSS.
502
W. Sanders, M.R. Anderson / Journal of Colloid and Interface Science 342 (2010) 499–504
the net positive charge of the interface, and suggests that the PDDA adsorbs to the substrate under these conditions. In contrast, when potential negative of the PZC is applied to the cysteine-modified substrate in the presence of anionic PSS (Fig. 2, open, before the potential is applied, and closed circles, after the potential is applied) the impedance does not significantly change. The impedance results suggest that, when a potential negative of the PZC is applied to the substrate, conditions that favor selective adsorption of cationic polyelectrolyte are generated; and, when potentials positive of the PZC are applied, conditions that favor selective adsorption of anionic polyelectrolytes are established. These impedance results are opposite of that expected if the potential were only creating local pH conditions that change the net charge of the interface by changing the relative ionization of the amine or carboxylic acid group of the confined cysteine molecules. QCM experiments were again conducted to confirm the impedance measurements. When the modified quartz crystal is exposed to potentials negative of the PZC in the presence of PDDA, the frequency decreases by 4.7(±0.6) Hz, corresponding to deposition of 110(±4) ng of PDDA. Under the same experimental conditions, except that the cysteine interface is exposed to anionic PSS, a frequency decrease of only 1.0(±0.5) Hz is measured, corresponding to deposition of 23(±12) ng. The impedance and QCM results obtained when applying potential to the substrate of a cysteine modified interface in the presence of cationic PDDA and anionic PSS are opposite of that expected by analogy to the behavior of 3-mercaptopropionic acid monolayers when subjected to applied potential, and cannot be attributed to changes in the pH local to the interface altering the net charge of the monolayer. Several reports show that potential applied to a substrate can be used to change the orientation of molecules containing charged groups that are part of a modified interface [30–33]. Kong et al. propose that potential applied to the substrate changes the conformation of the 16-mercaptohexadecanoate molecules that make-up a low-density monolayer [30]. In this study, they show that at positive applied potentials, the negative carboxylate group is attracted to the substrate, and this reorientation of the molecules that comprise the monolayer influence the adsorption of proteins to the monolayer modified interface. Somorjai et al. also show by surface energy and second harmonic generation measurements that potential applied to the substrate reorients the 16-mercaptohexadecanoate acid molecules of the monolayer [31]. In this application, they use potential to control the surface energy of the modified interface. Reorientation of cysteine and homocysteine molecules that are confined to an interface has also been reported by Brolo et al. [32] and Zhang et al. [33] In these reports, potential applied to the substrate created conditions that attracted either the cationic ammonium group or the anionic carboxylate group. Because cysteine is zwitterionic over a wide pH range, small changes in the solution pH adjacent to the interface will only have a small influence on the net charge of the monolayer. Potential applied to the substrate, however, may create electrostatic conditions that can attract either the cationic ammonium group or the anionic carboxylate group toward the substrate as suggested by
previous studies. If this reorientation were to occur, the oppositely charged group of the confined amino acid will be oriented toward the solution side of the interface (Scheme 1). For example, when a potential positive of the PZC is applied to the substrate, the carboxylate group is attracted to the substrate and the cationic ammonium group repelled by the interface. This reorientation causes the cationic ammonium group to be exposed to the adjacent solution. The presence of the exposed ammonium group at the interface creates conditions that potentially favor deposition of the anionic PSS, and inhibits deposition of the cationic PDDA. Conversely, when potentials negative of the PZC are applied to the substrate, conditions that favor attraction of the positively charged ammonium group are created, and the cysteine molecules reorient and expose the negatively charged carboxyl group to the solution. These conditions then favor deposition of the cationic PDDA and inhibit deposition of the anionic PSS. This description is consistent with our experimental data and with reports in the literature. In order for the molecules present at the interface to reorient with applied potential, the individual molecules need sufficient space to change conformation. Reductive desorption measurements for the cysteine monolayer, shown in Fig. 3, yield a coverage of 4.8(±0.9) 1010 mol cm2. This coverage is lower than the value found for saturation coverage of an n-alkanethiol monolayer (7.8 1010 mol cm2), or the coverage found by reductive desorption for monolayers prepared with 3mercaptopropionic acid (7.1(±0.6) 1010 mol cm2) [29,34]. At this coverage, each cysteine molecule occupies 0.34 nm2. This footprint is smaller than the 0.56 nm2 required for the reorientation of 16-mercaptohexadecanoic acid molecules to create a measurable surface energy change, as determined by Somorjai et al. [31]. Cysteine reorientation, however, does not require as much structural change as 16-mercaptohexadecanoic acid molecules to create a different net charge exposed at the monolayer–solution interface. Comparison of the QCM results obtained with potential induced deposition of polyelectrolyte onto the cysteine monolayer modified interface to results obtained by adjusting the interfacial charge with solution pH changes (in the absence of an applied potential) show that the mass of polyelectrolyte deposited by potential induced deposition is much smaller than the mass deposited when adjusting the bulk solution pH (Table 1). The differences in the mass of polyelectrolyte deposited under the two experimental conditions suggest that the cysteine monolayer interfaces are not the same. This is contrasted with the behavior of a 3-mercaptopropionic acid monolayer evaluated under similar experimental conditions in which the applied potential and the bulk solution pH had nearly identical impact on the interface [21]. The electrostatic driving force for polyelectrolyte deposition created when potential is applied to a substrate modified with a cysteine monolayer is considerably smaller than that caused by changes to the solution pH. This result is consistent with the applied potential inducing the confined cysteine molecules to reorient at the interface, rather than the applied potential significantly altering the net charge of the monolayer by influencing the solution pH local to the interface. Potential induced monolayer reorientation will not change the net
Scheme 1. Schematic representation of the proposed influence of substrate applied potential on the orientation of the confined cysteine molecules.
503
W. Sanders, M.R. Anderson / Journal of Colloid and Interface Science 342 (2010) 499–504
Fig. 3. Representative current vs. voltage curve for the reductive desorption of a cysteine monolayer self-assembled on a gold substrate. The reductive desorption experiment was conducted in an 0.5 mol L1 KOH solution.
Table 1 Comparison of the mass deposited onto a cysteine monolayer modified gold substrate under different experimental conditions. Experimental conditions
Mass deposited (ng)
E = 0.40 V vs. Ag–AgCl, 1.8 10-4 mol L1 PDDA E = 0.05 V vs. Ag–AgCl, 1.8 104 mol L1 PSS Open circuit potential, 0.10 mol L1 HCl, 1.8 104 mol L1 PSS [21] Open circuit potential, 0.10 mol L1 NaOH, 1.8 104 mol L1 PDDA [21]
110(±14) 124(±9) 700(±20) 770(±40)
charge of the interface; rather, the structure change will reorient the dipole of the cysteine molecules that comprise the monolayer and that will create conditions favorable for selective deposition of either polycationic or polyanionic species. Examination of the impedance and QCM data also show that small changes in the QCM frequency and the EIS impedance occur even for the conditions that do not favor the polyelectrolyte adsorption. This suggests that some amount of polyelectrolyte adsorbs to the interface even if the potential applied to the substrate does not favor the electrostatic adsorption of that particular polyelectrolyte. This result also suggests that the orientation of the charged groups on the cysteine modified interface does not provide absolute selectivity for the deposition of either polycationic PDDA or polyanionic PSS. Rather, potential applied to the cysteine-modified substrate establishes conditions for the preferential adsorption of different polyelectrolytes.
4. Summary In this research, we demonstrated that potential applied to the substrate of a cysteine modified interface can create conditions that favor the deposition of either polycations or polyanions, dependent on the value of the applied potential relative to the PZC of the modified substrate. The mechanism that creates these conditions, however, is different from that proposed when applying potential to monolayers of 3-mercaptoproprionic acid. With the cysteine monolayers, the electrostatic attraction between the charged ammonium or carboxylate functional groups of the confined zwitterionic cysteine molecules and the substrate excess charge creates interfacial conditions that favor the reorientation
of the cysteine molecules at the interface. The reorientation arranges the molecules at the interface so that one of these functional groups is attracted to the substrate and the other is oriented away from the substrate, toward the adjacent solution. This electrostatic ordering of the interface then apparently favors the deposition of a polyelectrolyte whose charge is opposite that of the group extending away from the surface. Although the amount of polyelectrolyte deposited is less than found when the interfacial charge is adjusted by bulk solution pH changes, potential applied to the cysteine-modified substrate clearly shows a selectivity for deposition of one charged polyelectrolyte over the oppositely charged polyelectrolyte. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]
C.D. Bain, G.M. Whitesides, J. Am. Chem. Soc. 111 (1989) 7164–7175. C.D. Bain, G.M. Whitesides, Angew. Chem., Int. Ed. Engl. 28 (1989) 506–512. C.D. Bain, G.M. Whitesides, Science 240 (1988) 62–63. S. Imabayashi, N. Gon, T. Sasaki, D. Hobara, T. Kakiuchi, Langmuir 14 (1998) 2348–2351. S.I. Imabayashi, D. Hobara, T. Kakiuchi, W. Knoll, Langmuir 13 (1997) 4502– 4504. T. Kakiuchi, K. Sato, M. Iida, D. Hobara, S. Imabayashi, K. Niki, Langmuir 16 (2000) 7238–7244. G. Decher, J.D. Hong, J. Schmitt, Thin Solid Films 210 (1992) 831–835. G. Decher, Y.M. Lvov, J. Schmitt, Thin Solid Films 244 (1994) 772–777. G. Decher, Science 277 (1997) 1232–1237. G. Decher, Photon. Optoelectron. Polym. 672 (1997) 445–459. T. Cassier, K. Lowack, G. Decher, Supermol. Sci. 5 (1998) 309–315. A.C. Harper, M.R. Anderson, Electroanalysis 18 (2006) 2397–2404. E.J. Calvo, A. Wolosiuk, ChemPhysChem 5 (2004) 235–239. S.L. Clark, M. Montague, P.T. Hammond, Supramol. Sci. 4 (1997) 141–146. P.T. Hammond, G.M. Whitesides, Macromolecules 28 (1995) 7569–7571. P.T. Hammond, Curr. Opin. Colloid Interface Sci. 4 (1999) 430–442. I. Willner, R. Blonder, A. Dagan, J. Am. Chem. Soc. 116 (1994) 9365–9366.
504 [18] [19] [20] [21] [22] [23] [24]
W. Sanders, M.R. Anderson / Journal of Colloid and Interface Science 342 (2010) 499–504
M. Lion-Dagan, E. Katz, I. Willner, J. Am. Chem. Soc. 116 (1994) 7913–7914. E. Katz, M. Lion-Dagan, I. Willner, J. Electroanal. Chem. 382 (1995) 25–31. E. Katz, I. Willner, Electroanalysis 7 (1995) 417–419. W. Sanders, M.R. Anderson, Langmuir 24 (2008) 12766–12770. W. Sanders, M.R. Anderson, J. Colloid Interface Sci. 331 (2009) 318–321. O. Melroy, K. Kanazawa, J.G. Gordon, D. Buttry, Langmuir 2 (1986) 697–700. L. Quigwen, G. Hong, W. Yiming, L. Guoan, M. Jie, Electroanalysis 13 (2001) 1342–1346. [25] M.A. Bryant, R.M. Crooks, Langmuir 9 (1993) 385–387. [26] A.M. Becka, C.J. Miller, J. Phys. Chem. 97 (1993) 6233–6239. [27] T. Komura, T. Yamaguchi, H. Shimatani, R. Okushio, Electrochim. Acta 49 (2004) 597–606.
[28] K. Kim, J. Kwak, J. Electroanal. Chem. 512 (2001) 83–91. [29] W. Sanders, R. Vargas, M.R. Anderson, Langmuir 24 (2008) 6133–6139. [30] L. Mu, Y. Liu, S. Zhang, B.H. Liu, J. Kong, New J. Chem. 29 (2005) 847– 852. [31] J. Lahann, S. Mitragotri, T.N. Tran, H. Kaido, J. Sundaram, I.S. Choi, S. Hoffer, G.A. Somorjai, R. Langer, Science 299 (2003) 371–374. [32] A.G. Brolo, P. Germain, G. Hager, J. Phys. Chem. B 106 (2002) 5982–5987. [33] J.D. Zhang, A. Demetriou, A.C. Welinder, T. Albrecht, R.J. Nichols, J. Ulstrup, Chem. Phys. 319 (2005) 210–221. [34] M.M. Walczak, D.D. Popenoe, R.S. Deinhammer, B.D. Lamp, C.K. Chung, M.D. Porter, Langmuir 7 (1991) 2687–2693.
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Potential driven deposition of polyelectrolytes onto the surface of cysteine monolayers assembled on gold Wesley Sanders a, Mark R. Anderson b,* a b
Department of Chemistry, Virginia Polytechnic Institute and State University, Blacksburg, VA 24060-0212, United States University of Colorado Denver, Department of Chemistry, Campus Box 194, PO Box 173364, Denver, CO 80217-3364, United States
a r t i c l e
i n f o
Article history: Received 10 September 2009 Accepted 15 October 2009 Available online 21 October 2009 Keywords: Self-assembled monolayers Electrostatic deposition Impedance spectroscopy
a b s t r a c t Electrochemical impedance spectroscopy and the quartz crystal microbalance measurements are used to examine the ability of potential applied to a substrate to create, in situ, conditions favorable for the electrostatic deposition of polyelectrolytes onto a gold substrate modified by the self-assembly of cysteine. Cysteine is a zwitterionic compound that, when confined to a substrate, has the ability to establish either a net positive or a net negative interfacial charge, depending on the conditions. As such, cysteine modified interfaces could possibly be used as a versatile substrate for deposition of either cationic or anionic polyelectrolytes. The potential of zero charge of a gold electrode modified by self-assembly with cysteine in the presence of 0.10 mol L1 KCl and buffered at pH 5 is found by differential capacitance measurement to be 0.12(±0.02) V vs. Ag/AgCl. When 0.05 V vs. Ag/AgCl is applied to the substrate (a potential positive of the PZC) in the presence of different polyelectrolytes, both impedance spectroscopy and quartz crystal microbalance data suggest the accumulation of anionic poly(sodium styrenesulfonate) along the cysteine modified interface. Conversely, when 0.40 V vs. Ag/AgCl is applied to the substrate (a potential negative of the PZC), experimental results suggest the accumulation of cationic poly(diallydimethylammonium chloride). Ó 2009 Elsevier Inc. All rights reserved.
1. Introduction Electrostatic and molecular self-assembly are relatively simple methods that have been widely applied to create chemically modified interfaces. Each of these methods take advantage of the through-space interactions that exist between molecules to drive their assembly onto solid substrates. This suggests that experimental manipulation of these interactions may provide some measure of control over the assembly process, and potentially can be used to control the structure and/or properties of the modified interface. Several research groups have looked at the role that different intermolecular interactions play in establishing the ensemble structure and properties of self-assembled monolayers. Bain and Whitesides generated modified interfaces by mixing two different mercaptans in the adsorption solution [1–3]. They found that the surface composition did not linearly track the composition of the adsorption solution; rather, differences in the extent of the through-space interactions among the molecules played a significant role in determining the composition of the interfacial layer. Subsequent studies by Hobara et al. [4–6] show that often with these mixed monolayers, the different components of the adsorption solution will phase segregate when they adsorb to the sur* Corresponding author. Fax: +1 303 556 4776. E-mail address: [email protected] (M.R. Anderson). 0021-9797/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2009.10.033
face. They propose that the phase segregation can be used to create two-dimensional molecular patterns along the interface. These results show that the strength of lateral through-space interactions help to establish the structure of self-assembling interfacial systems. Others have effectively used through-space interactions to create three-dimensional interfacial structures by depositing polymeric materials on top of the initial monolayer. Decher et al. take advantage of electrostatic interactions to deposit alternating layers of polyelectrolytes onto surfaces [7–10]. Electrostatic deposition is a versatile, robust method that has been applied for the deposition of a variety of ionic materials, including proteins and enzymes [11–13]. Hammond et al. leverage both lateral interactions and layer-by-layer deposition by combining contact printing with electrostatic deposition to create complex three-dimensional structures at interfaces [14–16]. Others show that experimentally modulating the through-space interactions along an interface can change the structure and alter the properties of the modified interface. For example, Willner et al. created a monolayer in which access to the substrate is modulated by experimentally inducing an isomerization of the molecules that comprise the monolayer [17–20]. Upon isomerization, different steric interactions exist at the interface that alter the physical properties (e.g. the permeability of electroactive species to the substrate surface) of the modified interface.
500
W. Sanders, M.R. Anderson / Journal of Colloid and Interface Science 342 (2010) 499–504
Our interest is in creating a modified interface and then manipulating the properties of the interface so that the structure can be subsequently modified in a controlled, reproducible fashion [21,22]. We previously demonstrated that potential applied to a substrate modified with a 3-mercaptopropionic acid monolayer could be used to modulate the properties of the modified interface to drive the electrostatic assembly of polycationic poly(diallydimethyl ammonium chloride), PDDA, onto the modified substrate [21]. In this application, potential applied to the substrate alters the interfacial interactions from conditions that do not favor polyelectrolyte deposition to conditions where polycation deposition is favored. We found that when potentials positive of the potential of zero charge (PZC) were applied to the substrate, the cationic polymer PDDA deposits onto the modified substrate, while PDDA does not adsorb if potentials negative of the PZC are applied. This result was interpreted as being due to the pH local to the interface becoming more basic when potentials more positive than the PZC are applied to the substrate, and the pH change causing the deprotonation of the terminal acid groups. Once deprotonated, the anionic charge of the interface creates conditions favorable for the deposition of a cationic polymer. A cysteine monolayer has the ability to have either a net positive or a net negative charge depending on the solution pH. We recently demonstrated that this property could be leveraged to electrostatically deposit either polycationic or polyanionic polymers onto a cysteine monolayer by adjusting the bulk solution pH [22]. In this manuscript, we explore using applied potential to drive the adsorption of either polycationic PDDA or polyanionic poly(sodium styrenesulfonate), PSS, onto a substrate modified with cysteine. 2. Experimental 2.1. Chemicals Cysteine, poly(diallyldimethylammonium chloride) and poly (sodium styrene sulfonate) were purchased from the Aldrich Chemical Company (Milwaukee, WI). Potassium hexacyanoferrate (II) trihydrate, potassium hexacyanoferrate (III), potassium hydrogen phosphate monohydrate, potassium dihydrogen phosphate monohydrate, phosphoric acid and potassium hydroxide were purchased from Fischer Scientific company. Poly(diallyldimethylammonium chloride), PDDA, was low molecular weight, 20% in water. Poly (sodium styrene sulfonate), PSS, was 20 wt.% in water. All chemicals were analytical grade and were used without further purification. Unless given otherwise, all solutions were prepared with water deionized with an 18 MX Milli-Q ion exchange filter from Millipore Incorporated.
2.3. Measurements Electrochemical impedance measurements were conducted using a CH Instruments (Austin, TX) model 604B electrochemical analyzer. The impedance measurements were performed in a standard three electrode cell with the modified gold electrode serving as the working electrode, a platinum wire as the auxiliary electrode, and a Ag/AgCl reference electrode. The supporting electrodissolved in 0.10 lyte consisted of 0.005 mol L1 FeðCNÞ3=4 6 mol L1 KCl and a phosphate buffer system containing a total phosphate concentration of 0.050 mol L1 and adjusted to pH 5. Impedance data was obtained at frequencies ranging from 100,000 to 0.1 Hz using a 5 mV amplitude sinusoidal potential modulation that is centered about the formal potential of the redox couple. Quantitative estimates of the charge-transfer resistance are obtained by fitting the experimental data to the Randles equivalent circuit using the nonlinear least squares fitting routines of the software package LEVM 7.0 (available from Solartron, www.solartronanalytical.com). For the potential induced polyelectrolyte deposition experiments, cysteine modified electrodes were placed in an aqueous phosphate buffer solution (pH 5) that contains 0.10 mol L1 KCl and either 1.8 104 mol L1 PDDA or 1.8 104 mol L1 PSS (the concentration is determined with respect to the monomer formula weight) and subject to applied potentials that are either positive or negative of the potential of zero charge for 60 s. The potential was then returned to the open circuit value and the substrate was then removed from this solution and placed into an electrolyte solution composed of 0.005 mol L1 FeðCNÞ63=4 , 0.10 mol L1 KCl, and pH 5 phosphate buffer for the impedance measurements. These experimental conditions were chosen to be consistent with similar experiments conducted previously with a gold substrate modified with a 3-mercaptopropionic acid monolayer [21]. Quartz crystal microbalance measurements were performed with an in-lab constructed QCM oscillator connected to an HP model 5334B frequency analyzer [23]. Quartz crystals from International Crystal Manufacturing (Oklahoma City, OK) having a 1.3 cm diameter gold electrode and 5 MHz resonant frequency were used as the substrate. Prior to cysteine immobilization onto the crystal’s gold electrodes, the quartz crystals were immersed in piranha solution (3:1 concentrated H2SO4:30% H2O2) for less than one minute to clean the surface. They were then rinsed with water, dried in a stream of N2, and immersed in a hexane solution containing 0.005 mol L1 cysteine. The quartz crystal frequencies were measured in air for 15 min before and after the monolayer modified crystals were exposed to the different experimental conditions (described above). During the measurement, the crystal frequency was sampled at 1 min intervals. For each trial, the measured frequency is the average of these 15 samples. The frequency changes reported are an average of measurements conducted for three different modified crystals.
2.2. Monolayer preparation Two millimeter gold disk working electrodes (CH Instruments, Austin, TX) were polished with 0.05 lm alumina followed by sonication in deionized water. The electrodes were then electrochemically cleaned in 0.5 mol L1 sulfuric acid by cycling the potential between 2.0 and 0.8 V vs. Ag–AgCl at 0.050 V/s for 25 complete cycles. Following the electrochemical cleaning, the electrodes were rinsed with deionized water, dried in a stream of nitrogen, and then immersed in a solution of hexane containing 0.005 mol L1 cysteine for 15 min. After cysteine immobilization, the electrode was rinsed with ethanol and deionized water. Reductive desorption measurements were conducted with electrodes modified by this method, and the surface coverage was found to be 4.8(±0.9) 1010 mol cm2.
3. Results For all measurements, the solution is buffered at pH 5, a value close to the isoelectric point for cysteine [24]. At this pH, the monolayer should be net neutral and provide no driving force for the electrostatic deposition of polyelectrolyte. Differential capacitance measurements are used to determine the potential of zero charge for the cysteine monolayer modified Au substrate [25,26]. From these measurements, the PZC is found to be 0.12(±0.05) V vs. Ag/AgCl. This value is consistent with experimental PZC’s measured with other monolayer modified Au electrodes [25,26]. Analogous to the observed behavior of 3-mercaptopropionic acid monolayers [21], by applying potential positive or negative
W. Sanders, M.R. Anderson / Journal of Colloid and Interface Science 342 (2010) 499–504
501
of the PZC we anticipated creating conditions that favor either a net negative or a net positive charge, respectively, of the cysteine monolayer due to the potential induced changes to the interfacial pH. We previously demonstrated that altering the net charge of the cysteine monolayer by adjusting the bulk solution pH in the absence of applied potential, the electrostatic deposition of either PDDA (when using basic pH) or PSS (when using acidic pH) could be accomplished [22]. Others have also shown that changes to bulk pH create electrostatic interactions at an ionizable interface capable of driving the deposition of polyelectrolytes [8,16]. By analogy, we anticipate that interfacial pH changes brought about by changes to the substrate potential should allow the electrostatic deposition of an appropriately charged polyelectrolyte. Changes to the interfacial properties brought about by the deposition of a polyelectrolyte are monitored by impedance spectroscopy. Fig. 1 shows the impedance response of a cysteine monolayer modified interface before (open squares) and after (closed squares) exposure of the interface to a solution containing 1.8 104 mol L1 PDDA and a potential positive of the PZC (Eapplied = 0.05 V vs. Ag–AgCl) applied to the substrate. Several research groups show that electrostatic interactions between a Redox probe and the monolayer modified interface influence the measured impedance [27,28]. Under these experimental conditions, the impedance decreases slightly after exposure of the interface to PDDA and positive potential applied to the substrate. This result suggests that there is a small change in the properties of the interface after the application of potential; however, the observed impedance change is not nearly as large as that found when conducting similar experiments with a 3-mercaptopropionic acid monolayer [29], nor is it as large as the impedance change found before and after exposure of the cysteine monolayer modified interface to a 0.10 mol L1 NaOH solution containing 1.8 104 mol L1 PDDA [22]. The magnitude of the impedance change found with the cysteine monolayer when subjected to positive potentials in the presence of PDDA suggests that only a small amount of the PDDA is deposited. Potentials positive of the PZC were also applied to cysteinemodified substrates in the presence of the anionic polyelectrolyte PSS, and the impedance subsequently measured (Fig. 1). Under these conditions, the impedance is found to have a larger change after the application of potential than the impedance decrease seen with exposure to PDDA. The increased impedance is characteristic of electrostatic repulsion between the FeðCNÞ64=3 redox probe and the modified interface, and suggests that PSS deposition occurred when the positive potential was applied to the substrate. This
was an unexpected outcome as we anticipated that the positive potential applied to the substrate would establish a pH local to the interface that is basic of the bulk pH (buffered at pH 5) and create conditions that favor a net negative charge of the cysteine monolayer. If the interface had a net negative charge, those conditions would inhibit deposition of the polyanionic PSS, the opposite behavior of what is observed experimentally. The magnitude of the impedance change when measured with these modified substrates is characteristic of either the amount of material deposited onto the monolayer, the charge density of the interface, or to some combination of these two parameters [27–29]. Quartz crystal microbalance measurements were conducted using these same experimental conditions to estimate the amount of PDDA and PSS deposition under this experimental condition. After a potential positive of the PZC is applied to the cysteine-modified substrate in the presence of 1.8 104 mol L1 PDDA, a frequency decrease of 1.4(±0.6) Hz is measured. This frequency change corresponds to deposition of 32(±14) ng of PDDA polymer deposited. In contrast, when the QCM experiment is conducted in the presence of 1.8 104 mol L1 PSS, the frequency decreases by 5.3(±0.4) Hz. This frequency decrease corresponds to a deposition of 124(±9) ng of PSS polymer. The results of the impedance and QCM experiments show that, when potentials positive of the PZC are applied to the substrate, deposition of the anionic PSS is favored relative to that of cationic PDDA. This behavior, combined with the impedance data, suggests that when potentials positive of the PZC are applied to the cysteine monolayer modifed substrate, a positive charge along the monolayer/solution interface develops. The experimental evidence that PSS deposition is favored when potentials positive of the PZC are applied to the substrate suggests that a mechanism other than the applied potential influencing the solution pH local to the interface (which in turn influences the net charge of the monolayer) is at work. If this is the case, then the behavior when potentials negative of the PZC are applied should also be different from that expected if the local pH were controlling the net charge of the cysteine monolayer. Fig. 2 shows the impedance of the cysteine modified interface before and after a potential negative of the PZC is applied to the substrate in the presence of 1.8 104 mol L1 PDDA. Under these conditions, the impedance measured with an anionic redox probe decreases after the application of the negative potential, consistent with an electrostatic attraction between the FeðCNÞ64=3 redox probe and the monolayer modified substrate. This EIS result is characteristic of an increase in
Fig. 1. Representative complex impedance plots obtained by electrochemical impedance spectroscopy measured with a gold electrode modified with a layer of cysteine exposed to a 0.10 mol L1 KCl solution buffered at pH 5 containing 1 FeðCNÞ3 0.005 mol L1 FeðCNÞ4 6 =0:005 mol L 6 (i) before (open squares) and after (solid squares) application of a potential positive of the potential of zero charge in the presence of 1.8 104 mol L1 PDDA and (ii) before (open circles) and after (solid circles) application of a potential positive of the potential of zero charge in the presence of 1.8 104 mol L1 PSS.
Fig. 2. Representative complex impedance plots obtained by electrochemical impedance spectroscopy measured with a gold electrode modified with a layer of cysteine exposed to a 0.10 mol L1 KCl solution buffered at pH 5 containing 1 FeðCNÞ3 0.005 mol L1 FeðCNÞ4 6 =0:005 mol L 6 (i) before (open squares) and after (solid squares) application of a potential negative of the potential of zero charge in the presence of 1.8 104 mol L1 PDDA and (ii) before (open circles) and after (solid circles) application of a potential positive of the potential of zero charge in the presence of 1.8 104 mol L1 PSS.
502
W. Sanders, M.R. Anderson / Journal of Colloid and Interface Science 342 (2010) 499–504
the net positive charge of the interface, and suggests that the PDDA adsorbs to the substrate under these conditions. In contrast, when potential negative of the PZC is applied to the cysteine-modified substrate in the presence of anionic PSS (Fig. 2, open, before the potential is applied, and closed circles, after the potential is applied) the impedance does not significantly change. The impedance results suggest that, when a potential negative of the PZC is applied to the substrate, conditions that favor selective adsorption of cationic polyelectrolyte are generated; and, when potentials positive of the PZC are applied, conditions that favor selective adsorption of anionic polyelectrolytes are established. These impedance results are opposite of that expected if the potential were only creating local pH conditions that change the net charge of the interface by changing the relative ionization of the amine or carboxylic acid group of the confined cysteine molecules. QCM experiments were again conducted to confirm the impedance measurements. When the modified quartz crystal is exposed to potentials negative of the PZC in the presence of PDDA, the frequency decreases by 4.7(±0.6) Hz, corresponding to deposition of 110(±4) ng of PDDA. Under the same experimental conditions, except that the cysteine interface is exposed to anionic PSS, a frequency decrease of only 1.0(±0.5) Hz is measured, corresponding to deposition of 23(±12) ng. The impedance and QCM results obtained when applying potential to the substrate of a cysteine modified interface in the presence of cationic PDDA and anionic PSS are opposite of that expected by analogy to the behavior of 3-mercaptopropionic acid monolayers when subjected to applied potential, and cannot be attributed to changes in the pH local to the interface altering the net charge of the monolayer. Several reports show that potential applied to a substrate can be used to change the orientation of molecules containing charged groups that are part of a modified interface [30–33]. Kong et al. propose that potential applied to the substrate changes the conformation of the 16-mercaptohexadecanoate molecules that make-up a low-density monolayer [30]. In this study, they show that at positive applied potentials, the negative carboxylate group is attracted to the substrate, and this reorientation of the molecules that comprise the monolayer influence the adsorption of proteins to the monolayer modified interface. Somorjai et al. also show by surface energy and second harmonic generation measurements that potential applied to the substrate reorients the 16-mercaptohexadecanoate acid molecules of the monolayer [31]. In this application, they use potential to control the surface energy of the modified interface. Reorientation of cysteine and homocysteine molecules that are confined to an interface has also been reported by Brolo et al. [32] and Zhang et al. [33] In these reports, potential applied to the substrate created conditions that attracted either the cationic ammonium group or the anionic carboxylate group. Because cysteine is zwitterionic over a wide pH range, small changes in the solution pH adjacent to the interface will only have a small influence on the net charge of the monolayer. Potential applied to the substrate, however, may create electrostatic conditions that can attract either the cationic ammonium group or the anionic carboxylate group toward the substrate as suggested by
previous studies. If this reorientation were to occur, the oppositely charged group of the confined amino acid will be oriented toward the solution side of the interface (Scheme 1). For example, when a potential positive of the PZC is applied to the substrate, the carboxylate group is attracted to the substrate and the cationic ammonium group repelled by the interface. This reorientation causes the cationic ammonium group to be exposed to the adjacent solution. The presence of the exposed ammonium group at the interface creates conditions that potentially favor deposition of the anionic PSS, and inhibits deposition of the cationic PDDA. Conversely, when potentials negative of the PZC are applied to the substrate, conditions that favor attraction of the positively charged ammonium group are created, and the cysteine molecules reorient and expose the negatively charged carboxyl group to the solution. These conditions then favor deposition of the cationic PDDA and inhibit deposition of the anionic PSS. This description is consistent with our experimental data and with reports in the literature. In order for the molecules present at the interface to reorient with applied potential, the individual molecules need sufficient space to change conformation. Reductive desorption measurements for the cysteine monolayer, shown in Fig. 3, yield a coverage of 4.8(±0.9) 1010 mol cm2. This coverage is lower than the value found for saturation coverage of an n-alkanethiol monolayer (7.8 1010 mol cm2), or the coverage found by reductive desorption for monolayers prepared with 3mercaptopropionic acid (7.1(±0.6) 1010 mol cm2) [29,34]. At this coverage, each cysteine molecule occupies 0.34 nm2. This footprint is smaller than the 0.56 nm2 required for the reorientation of 16-mercaptohexadecanoic acid molecules to create a measurable surface energy change, as determined by Somorjai et al. [31]. Cysteine reorientation, however, does not require as much structural change as 16-mercaptohexadecanoic acid molecules to create a different net charge exposed at the monolayer–solution interface. Comparison of the QCM results obtained with potential induced deposition of polyelectrolyte onto the cysteine monolayer modified interface to results obtained by adjusting the interfacial charge with solution pH changes (in the absence of an applied potential) show that the mass of polyelectrolyte deposited by potential induced deposition is much smaller than the mass deposited when adjusting the bulk solution pH (Table 1). The differences in the mass of polyelectrolyte deposited under the two experimental conditions suggest that the cysteine monolayer interfaces are not the same. This is contrasted with the behavior of a 3-mercaptopropionic acid monolayer evaluated under similar experimental conditions in which the applied potential and the bulk solution pH had nearly identical impact on the interface [21]. The electrostatic driving force for polyelectrolyte deposition created when potential is applied to a substrate modified with a cysteine monolayer is considerably smaller than that caused by changes to the solution pH. This result is consistent with the applied potential inducing the confined cysteine molecules to reorient at the interface, rather than the applied potential significantly altering the net charge of the monolayer by influencing the solution pH local to the interface. Potential induced monolayer reorientation will not change the net
Scheme 1. Schematic representation of the proposed influence of substrate applied potential on the orientation of the confined cysteine molecules.
503
W. Sanders, M.R. Anderson / Journal of Colloid and Interface Science 342 (2010) 499–504
Fig. 3. Representative current vs. voltage curve for the reductive desorption of a cysteine monolayer self-assembled on a gold substrate. The reductive desorption experiment was conducted in an 0.5 mol L1 KOH solution.
Table 1 Comparison of the mass deposited onto a cysteine monolayer modified gold substrate under different experimental conditions. Experimental conditions
Mass deposited (ng)
E = 0.40 V vs. Ag–AgCl, 1.8 10-4 mol L1 PDDA E = 0.05 V vs. Ag–AgCl, 1.8 104 mol L1 PSS Open circuit potential, 0.10 mol L1 HCl, 1.8 104 mol L1 PSS [21] Open circuit potential, 0.10 mol L1 NaOH, 1.8 104 mol L1 PDDA [21]
110(±14) 124(±9) 700(±20) 770(±40)
charge of the interface; rather, the structure change will reorient the dipole of the cysteine molecules that comprise the monolayer and that will create conditions favorable for selective deposition of either polycationic or polyanionic species. Examination of the impedance and QCM data also show that small changes in the QCM frequency and the EIS impedance occur even for the conditions that do not favor the polyelectrolyte adsorption. This suggests that some amount of polyelectrolyte adsorbs to the interface even if the potential applied to the substrate does not favor the electrostatic adsorption of that particular polyelectrolyte. This result also suggests that the orientation of the charged groups on the cysteine modified interface does not provide absolute selectivity for the deposition of either polycationic PDDA or polyanionic PSS. Rather, potential applied to the cysteine-modified substrate establishes conditions for the preferential adsorption of different polyelectrolytes.
4. Summary In this research, we demonstrated that potential applied to the substrate of a cysteine modified interface can create conditions that favor the deposition of either polycations or polyanions, dependent on the value of the applied potential relative to the PZC of the modified substrate. The mechanism that creates these conditions, however, is different from that proposed when applying potential to monolayers of 3-mercaptoproprionic acid. With the cysteine monolayers, the electrostatic attraction between the charged ammonium or carboxylate functional groups of the confined zwitterionic cysteine molecules and the substrate excess charge creates interfacial conditions that favor the reorientation
of the cysteine molecules at the interface. The reorientation arranges the molecules at the interface so that one of these functional groups is attracted to the substrate and the other is oriented away from the substrate, toward the adjacent solution. This electrostatic ordering of the interface then apparently favors the deposition of a polyelectrolyte whose charge is opposite that of the group extending away from the surface. Although the amount of polyelectrolyte deposited is less than found when the interfacial charge is adjusted by bulk solution pH changes, potential applied to the cysteine-modified substrate clearly shows a selectivity for deposition of one charged polyelectrolyte over the oppositely charged polyelectrolyte. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]
C.D. Bain, G.M. Whitesides, J. Am. Chem. Soc. 111 (1989) 7164–7175. C.D. Bain, G.M. Whitesides, Angew. Chem., Int. Ed. Engl. 28 (1989) 506–512. C.D. Bain, G.M. Whitesides, Science 240 (1988) 62–63. S. Imabayashi, N. Gon, T. Sasaki, D. Hobara, T. Kakiuchi, Langmuir 14 (1998) 2348–2351. S.I. Imabayashi, D. Hobara, T. Kakiuchi, W. Knoll, Langmuir 13 (1997) 4502– 4504. T. Kakiuchi, K. Sato, M. Iida, D. Hobara, S. Imabayashi, K. Niki, Langmuir 16 (2000) 7238–7244. G. Decher, J.D. Hong, J. Schmitt, Thin Solid Films 210 (1992) 831–835. G. Decher, Y.M. Lvov, J. Schmitt, Thin Solid Films 244 (1994) 772–777. G. Decher, Science 277 (1997) 1232–1237. G. Decher, Photon. Optoelectron. Polym. 672 (1997) 445–459. T. Cassier, K. Lowack, G. Decher, Supermol. Sci. 5 (1998) 309–315. A.C. Harper, M.R. Anderson, Electroanalysis 18 (2006) 2397–2404. E.J. Calvo, A. Wolosiuk, ChemPhysChem 5 (2004) 235–239. S.L. Clark, M. Montague, P.T. Hammond, Supramol. Sci. 4 (1997) 141–146. P.T. Hammond, G.M. Whitesides, Macromolecules 28 (1995) 7569–7571. P.T. Hammond, Curr. Opin. Colloid Interface Sci. 4 (1999) 430–442. I. Willner, R. Blonder, A. Dagan, J. Am. Chem. Soc. 116 (1994) 9365–9366.
504 [18] [19] [20] [21] [22] [23] [24]
W. Sanders, M.R. Anderson / Journal of Colloid and Interface Science 342 (2010) 499–504
M. Lion-Dagan, E. Katz, I. Willner, J. Am. Chem. Soc. 116 (1994) 7913–7914. E. Katz, M. Lion-Dagan, I. Willner, J. Electroanal. Chem. 382 (1995) 25–31. E. Katz, I. Willner, Electroanalysis 7 (1995) 417–419. W. Sanders, M.R. Anderson, Langmuir 24 (2008) 12766–12770. W. Sanders, M.R. Anderson, J. Colloid Interface Sci. 331 (2009) 318–321. O. Melroy, K. Kanazawa, J.G. Gordon, D. Buttry, Langmuir 2 (1986) 697–700. L. Quigwen, G. Hong, W. Yiming, L. Guoan, M. Jie, Electroanalysis 13 (2001) 1342–1346. [25] M.A. Bryant, R.M. Crooks, Langmuir 9 (1993) 385–387. [26] A.M. Becka, C.J. Miller, J. Phys. Chem. 97 (1993) 6233–6239. [27] T. Komura, T. Yamaguchi, H. Shimatani, R. Okushio, Electrochim. Acta 49 (2004) 597–606.
[28] K. Kim, J. Kwak, J. Electroanal. Chem. 512 (2001) 83–91. [29] W. Sanders, R. Vargas, M.R. Anderson, Langmuir 24 (2008) 6133–6139. [30] L. Mu, Y. Liu, S. Zhang, B.H. Liu, J. Kong, New J. Chem. 29 (2005) 847– 852. [31] J. Lahann, S. Mitragotri, T.N. Tran, H. Kaido, J. Sundaram, I.S. Choi, S. Hoffer, G.A. Somorjai, R. Langer, Science 299 (2003) 371–374. [32] A.G. Brolo, P. Germain, G. Hager, J. Phys. Chem. B 106 (2002) 5982–5987. [33] J.D. Zhang, A. Demetriou, A.C. Welinder, T. Albrecht, R.J. Nichols, J. Ulstrup, Chem. Phys. 319 (2005) 210–221. [34] M.M. Walczak, D.D. Popenoe, R.S. Deinhammer, B.D. Lamp, C.K. Chung, M.D. Porter, Langmuir 7 (1991) 2687–2693.