Subscriber access provided by The University of British Columbia Library
Article
Probing the surface properties of gold at low electrolyte concentration Ran Tivony, and Jacob Klein Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.6b01697 • Publication Date (Web): 29 Jun 2016 Downloaded from http://pubs.acs.org on July 1, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 32
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Table of content graphic
_
OH+ H3O+
Gold +
+ +
Pure water
+
+
Mica Adhesive Contact
+
Bulk solution
_
+
+
_
+
Attraction +
+
+
+
+
+ + +
ACS Paragon Plus Environment
+ Bulk
solution
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Probing the surface properties of gold at low electrolyte concentration Ran Tivony and Jacob Klein* Department of Materials and Interfaces, Weizmann Institute of Science, Rehovot, 76100, Israel *Corresponding author:
[email protected] 1 ACS Paragon Plus Environment
Page 2 of 32
Page 3 of 32
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Abstract Using the Surface Force Balance (SFB), we studied the surface properties of gold in aqueous solution with low electrolyte concentration (~10-5M and pH = 5.8), i.e. water with no added salt, by directly measuring the interaction between an ultra-smooth gold surface (ca. 0.2nm RMS roughness) and a mica surface. Under these conditions specific adsorption of ions is minimized and its influence on the surface charge and surface potential of gold is markedly reduced. At open circuit potential, the electrostatic interaction between gold and mica was purely attractive and gold was found to be positively charged. This was further confirmed by force measurements against a positively charged surface, poly-L-lysine coated mica. Successive force measurements unambiguously showed that once gold and mica reach contact all counterions are expelled from the gap, confirming that at contact the surface charge of gold is equal and opposite in charge to that of mica. Further analysis of adhesion energy between the surfaces indicated that adhesion is mostly governed by vdW dispersion force and to a lesser extent by electrostatic interaction. Force measurements under external applied potentials showed that the gold-mica interaction can be regulated as a function of applied potential even at low electrolyte concentration. The gold-mica interaction was described very precisely by the non-linearized Poisson-Boltzmann (PB) equation, where one of the surfaces is at constant charge, i.e. mica, and the other, i.e. gold, is at constant potential. Consequently, the gold surface potential could be determined accurately both at open circuit potential (OCP) and under different applied potentials. Using the obtained surface potentials, we were able to derive fundamental characteristics of the gold surface, e.g. its surface charge density and potential of zero charge (PZC), at very low electrolyte concentration.
2 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 32
Introduction The electrostatic double-layer formed at surfaces immersed in aqueous media affects myriad processes, from colloidal stability1 to supercapacitors2. Measurements of surface forces across such media, which in recent decades have most directly been carried out using scanning probe methods such as atomic force microscopy (AFM)3-6 or surface force balance (SFB)7-9 approaches, can provide detailed information both on the nature of the interactions, and on the properties of the surfaces themselves3, 9. Interactions at metal surfaces are of special interest10, surfaces may be maintained at controlled potentials12,
13.
11,
as such
However, the pH and purity of the
surrounding media, and in particular the detailed structure, local surface roughness13,
14
and
morphology15-17 of the surfaces, which are not always possible to control for, can strongly affect the double-layers and thus the interactions, and may account for discrepancies with theory and between different experimental studies. In an early work the interaction between two gold surfaces in water was measured by Biggs and Mulvaney18 using an AFM. In their study, a long-ranged attraction, which was attributed to a strong van der Waals (vdW) interaction was observed between a gold plate and a gold coated silica sphere across water at open circuit potential (OCP). This result – through the absence of any measured long-ranged repulsion – does not indicate any charge on the gold in water. In later work by Bard and co-workers the effect of applied potential on the interaction between a gold plate and a silica sphere in different alkali-halide electrolyte solutions was studied, again using an AFM12. While this study clearly indicated that the silica-gold interaction was strongly affected by the potential applied to the gold, and by the nature of the electrolyte, their results could not be fitted quantitatively by the classic Gouy-Chapman-Stern (GCS) theory. Moreover, no vdW attraction was observed in salt solution, possibly owing to specific adsorption of halide ions, while in deionized water and at open circuit conditions a long range vdW attraction was observed. In contrast, an SFB study by Chai et al. of the interaction between a smooth gold 3 ACS Paragon Plus Environment
Page 5 of 32
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
surface and mica across purified water with no added salt indicated a long ranged repulsion at open circuit potential, suggesting a negative potential on the gold11,
19.
The effect of applied
potential on the gold-mica interaction was extensively investigated by Vanderlick et al., using an SFB10, 20-22. The effect of potential was examined in 1mM KClO4, but unlike earlier reports3, 12 the interaction did not change from electrostatic repulsion to attraction as a function of potential but merely turned less repulsive. In addition, vdW attraction was not observed and theoretical predictions using the PB equation deviated significantly from the measured forces at small separations, ca. 5nm. Some of these discrepancies may arise from adsorption on the gold of organic molecules23 and anions24, 25, from differences in solution pH24, 26, 27 affecting the surface charge, and, importantly, from differences in surface roughness13, 14, 28 and morphology28, 29. Very recently, we showed that the interaction between a smooth gold surface (ca. 0.2nm rms roughness) and mica in water with no added salt (effective salt concentration ~10-5M, pH = 5.8) is non-monotonic, as it reverts from electrostatic repulsion to electrostatic attraction as the surfaces approach30. This effect, not previously seen directly, could be very well described by the Gouy-Chapman model31, and revealed in particular that when a metal surface at constant potential (gold) interacted with one at constant charge (mica) of the same potential sign, the surface charge on the gold could be reversed as the surfaces approached. In this study the gold surfaces used were made by the template-stripping method of Chai and Klein32. This approach results in large area (~ cm2), exceptionally smooth (to 0.2nm rms roughness) and morphologically well characterized (with respect to the surface crystallographic structure) gold surfaces. Such template-stripped surfaces differ in their reproducible extreme smoothness and morphological uniformity from gold surfaces used in earlier studies; due to their large size they are also particularly well suited for studying in the SFB.
4 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Here, we extend our previous study by using such well-characterized gold surfaces32 in the SFB to examine comprehensively the interaction and surface properties of gold in aqueous solution with low electrolyte concentration, including water with no added salt (pH = 5.8), to minimize any effect of specific adsorption on the gold, especially of hydroxides. We show that in low electrolyte concentration at these pH values, and in the absence of specific adsorption, gold is positively charged. We also demonstrate that its interaction at OCP and under applied potentials can be described very precisely using the nonlinear Poisson-Boltzmann equation where one of the surfaces, mica, is at constant charge while the other, gold, is at constant potential. In consequence, basic characteristics such as the potential of zero charge (PZC) of gold can be determined accurately at low electrolyte concentration. Furthermore, we demonstrate experimentally that once a gold-mica contact is reached, all counterions are squeezed-out to the bulk, so that adhesion is determined by electrostatic force and, mostly, by vdW dispersion forces. Experiment and Materials Materials. Pure water with a total organic content of less than 1 ppb (TOC < 1), a resistivity of 18.2 MΩ cm (so-called conductivity water) and pH = 5.8 was prepared by passing tap water twice through a reverse osmosis system, then passing through an ion exchange column and then through a mechanical filters of mesh size 5 and 2μm before processing in a Barnstead Nanopure Diamond UV/UF system. Gold pellets, 99.999% pure, were purchased from Kurt J. Lesker Co. and evaporated from a graphite crucible. Sodium nitrate, NaNO3, 99.99% pure was purchased from Merck Millipore and used as received. A 0.01wt% solution of Poly-L-lysine (MW = 70,000-150,000Da) in water was purchased from Sigma-Aldrich and used as received. Potassium hydroxide, 99.99% pure (KOH, pKa = 13.5) was purchased from Sigma-Aldrich and used as received. Nitric acid 65% (HNO3, pKa = -1.4) was purchased from Merck Millipore and used as received. 5 ACS Paragon Plus Environment
Page 6 of 32
Page 7 of 32
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Experimental section. Preparation of the gold surface. The preparation process of the ultra-smooth gold surface has been described in detail elsewhere19, 32. Briefly, a thin film (65-70nm) of gold was very slowly evaporated using an e-beam onto a pre-heated ruby muscovite mica (cleaved to sheets 2.5-5μm in thickness and ca. 1 cm x 2 cm in size that were crystallographically smooth on both sides). Following evaporation the sample was annealed for 2hrs and then allowed to slowly cool to room temperature prior to venting the chamber. A sheet of gold-coated mica was then glued onto a cylindrical fused silica lens (for mounting in the SFB), with the gold facing the glue (EPON 1004, Shell Chemicals). Prior to an experiment, the mica was gently peeled-off, using tweezers, to expose the smooth gold surface, and the lens was rapidly mounted and covered with water in the SFB (see below), within 30 – 60 secs of exposure. Preparation of the mica surface. Atomically smooth ruby muscovite mica backing sheet (grade I, S & J Trading, Inc., NY) was freshly cleaved, and mica facets of 2.5-4.5μm in thickness were upstream melt cut (to ensure no Pt contamination) as described elsewhere33 and placed on it with the freshly cleaved side facing downwards. A layer of silver of thickness 45-50nm was then deposited on the backing sheet and mica facets by thermal evaporation. The samples were kept under low humidity conditions to prevent oxidation of the silver layer before an experiment. Adsorption of poly-L-lysine on mica. Prior to the adsorption, a 0.001wt% poly-L-lysine solution was prepared by adding 18ml of water to ~2ml of poly-L-lysine solution (0.01wt%). A piece of silvered mica was glued to a cylindrical lens, with the silvered side facing the lens, and immersed into the poly-L-lysine solution (0.001wt%) for ~24hrs. The lens was taken out and washed under a stream of purified water, for several seconds, and immediately mounted, while still wet, in the SFB. Electrochemical measurements. Cyclic voltammetry measurements were performed with a CHI600C (CH instruments, Inc.) potentiostat and were taken in a custom-designed 6 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
electrochemical glass cell with the gold surface serving as the working electrode and two platinum wires as the reference and the counter electrodes. In general, the gold electrode was prepared by template stripping as mentioned above and all measurements were conducted using a freshly exposed gold surface. NaNO3 solutions were prepared by using pure water with a total organic content of less than 1 ppb (TOC < 1) and resistivity of 18.2 MΩ∙cm. Atomic force microscopy (AFM). Atomic Force Microscopy scans of Poly-L-lysine coated mica and gold were performed in a noncontact tapping mode in water and in air, respectively. In order to measure the roughness of the obtained gold, we glued a piece of gold-coated mica to a glass slide with epoxy resin (EPON 1004), peeled off the mica and scanned the gold surface using AFM (MFP-3D AFM, Asylum research). To eliminate any time effects like humidity or adsorption of contaminants, the scans were carried out as soon as possible after peeling off the mica and exposing the gold. Surface Force Balance (SFB) measurements. The normal and shear forces Fn(D) between surfaces of a gold film and a mica sheet a closest distance D apart were measured using the SFB as previously described in detail11, 34. Figure 1 shows a schematic of the SFB where two surfaces, gold and mica, are glued on opposing fused-silica lenses in a crossed-cylinder configuration, equivalent to the geometry of a sphere over a flat. In this study, normal forces between the surfaces were measured either via a quasi-static step-wise approach or via a dynamic approach19. The separation D between the surfaces (mica and gold) in the SFB is calculated using the multilayer matrix method35, 36, as previously described in detail32. In brief, this method allows to accurately calculate the intensity of interference “fringes of equal chromatic order” (FECO), as a function of wavelength, formed between two reflective surfaces. To determine the separation D between gold and mica, FECO fringes wavelengths were first calculated for a two-layer interferometer Gold/Medium/Mica/Silver, where the thickness of the medium equals zero (i.e. 7 ACS Paragon Plus Environment
Page 8 of 32
Page 9 of 32
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
D=0), to match their measured values between gold and mica at contact. Carrying out the same calculation for different medium layer thicknesses, the position of FECO fringes is then determined for any gold-mica separation D. Finally, a comparison of the calculated FECO fringes wavelengths with those measured in the experiment gives the separation between gold and mica. External potentials were applied on the gold surface by using a three electrode configuration, with the gold surface as a working (W) electrode, platinum as a counter (C) electrode and platinum as a (quasi) reference electrode (second panel of fig. 1, showing also the actual electrode setup). Before each experiment the Platinum electrodes were washed in hot nitric acid solution (30%, ca. 800c) for at least 2 hours and then were passed through a flame to oxidize and remove remains of adsorbed organic molecules. To apply the required potentials on the gold surface, the electrodes were connected to a potentiostat (CHI600C, CH instruments, Inc.) which serves as the control unit. All glassware was cleaned with a 30:70 H2O2/H2SO4 piranha solution and then rinsed with conductivity water and ethanol (Caution: Piranha solution reacts violently with organic compounds and should be handled carefully.). Results and discussion Nature of gold surface: Typical AFM images of a template-stripped polycrystalline gold [111]30, 32
surface used in this study are shown in figure 2. The obtained surface is homogeneous and
ultra-smooth with a RMS roughness of ~0.2nm. X-ray diffraction measurements show that the gold surface lattice is >95% [111] oriented30, 32 Gold-Mica interaction at open circuit potential conditions (OCP). Figure 3 shows the normal force between bare gold and negatively charged bare mica at OCP and in water with no added salt. At these conditions the interaction between mica and gold is purely attractive at all
8 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
separations and onsets at separation D ∼ 150nm, very much larger than the calculated vdW interaction (red dashed line, AH =9×10-20 J), indicating its electrostatic origin. At smaller separations the surfaces jumped-in to adhesive contact, due to the instability of the spring (whenever ∂F/∂X > Ks), from Dj = 25 ± 2nm to a final distance of D0 = 1.0 ± 0.3nm37. When NaNO3 was added to a concentration of 1mM, a purely attractive interaction between gold and mica was again observed, at much shorter range (though still larger than the range of pure vdW attraction), setting on from D ~ 20 nm, as shown in the inset to figure 3. Such shorter range is indeed expected for the attractive electrostatic interactions at the higher salt concentration (κ-1 = 9.6nm). In contrast, the interaction between two mica surfaces across water with no added salt (full black symbols, fig. 3), often measured in previous studies38-40 and repeated here as a control, is monotonically repulsive down to the range of vdW attraction when the surfaces jumpin from Dj ~ 5nm to adhesive contact. The electrostatic nature of the long-ranged interactions in fig. 3 (repulsive for mica-mica or attractive for mica-gold at OCP) is seen from the close fit to the data of the 1-D non-linear Poisson-Boltzmann (PB) ∇ ψx =
ψ
sinh
applicable for such surface interactions with
the appropriate boundary conditions. Here c0 is the concentration of ions in the bulk solution, is the permittivity of free space, is the dielectric constant of the solvent, kB the Boltzmann constant, T the absolute temperature, and e the electronic charge. Mica immersed in aqueous is taken as a constant charge (CC) surface (as in many earlier publications38, 39, 41) and the black solid curve in fig. 3 is the prediction of the PB equation for the mica-mica interaction (black data points). Fit parameters correspond to a Debye length of κ-1 = 100nm, a mica surface potential of Ψ = −110mV and Hamaker constant of A! = 2 × 10$ J, all within the range of literature values for such interactions33, 39, 42. We further use the PB equation to describe the interaction between gold and mica where the surfaces, mica and gold, were set to be at CC and constant potential (CP) boundary conditions, respectively. At these boundary conditions, the black 9 ACS Paragon Plus Environment
Page 10 of 32
Page 11 of 32
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
dashed line in figure 3 is the (best fit) prediction to the gold-mica interaction in water with no added salt. The predicted curve shows a close agreement with the data, yielding a gold surface
Ψ
potential of Ψ%&'( =15mV. Using the Grahame equation, σ = *8, ε kT sinh 0
1 where Ψ is
the surface potential, the corresponding surface charge density is σgold =0.19 mC/m2 , equivalent to σgold =e/837nm2 . This result shows that in water with no added salt (, = 3 × 10$4 M) and at OCP gold is slightly positively charged as also suggested tentatively by Bard et al12. A similarly close fit of the data to the prediction of the PB equation (black dashed line) was also seen for interactions in 1mM NaNO3, as shown in the inset to fig. 3, enabling a surface potential of Ψgold =2mV and surface charge density of σ%&'( (1mM NaNO3 )=0.15 mC/m2 , equivalent to σgold =e/1103nm2 , to be extracted. The origin of positive charges at the surface of gold at OCP can be explained as follows. In cases where specific adsorption of ions occurs on gold, the potential of zero charge (PZC) deviates and turns the gold surface to be negatively charged at OCP24. Consequently, the interaction of gold with a negatively charged surface, like mica or silica, is repulsive25-27. Thus, an attractive goldmica interaction strongly suggests that specific adsorption on gold is negligible. Similarly, in -
1mM NaNO3 solution specific adsorption is also considered minor as nitrate ions (NO3 ) are known to adsorb very weakly and to low extent on gold43, 44. In the absence of specific adsorption the formation of charges at the surface of the gold is mostly determined by the electrochemical potential of charges in the electrolyte phase and by the ordering of water dipole at the surface45. When an uncharged metal comes into contact with a solution the electron distribution at the metal surface is perturbed and electrons are transferred between the phases, with electrons flowing from the phase with the higher electrochemical potential to the phase with the lower electrochemical potential, until equilibrium is reached46, 47. This electron flow changes the surface potential of gold and modifies its charge to positive. 10 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Interaction between gold and a positively charged surface. To confirm that gold is indeed positively charged at OCP in our configuration, as indicated in fig. 3 (but in contrast to some earlier studies, as explained above), we measured its interaction with a positively charged surface. As cationic surfactants are known to form charged patches on mica8, 48, in order to form a smooth and uniform positively charged surface we chose to use for charge reversal the positively charged polyelectrolyte, poly-L-lysine, which adsorbed on mica (mica-PL), when the negative mica surface charge is reversed49. The obtained mica-PL surface was uniform and very smooth with a RMS roughness of 0.145nm as measured under water using the AFM (figure 4). To verify that mica-PL is positively charged, its interaction with mica was measured in water with no added salt (figure 5). The interaction was found to be purely attractive at all distances signifying that mica-PL surface is overall positively charged. By fitting the interaction to the PB equation with constant charge boundary conditions we obtained surface potential of = 105mV and surface charge density of σ $78 = 2.43 mC⁄m e/66nm2 , close to Ψ $78 that of mica but with the opposite charge sign. As shown in figure 5, a very weak long-ranged repulsion was observed between gold and micaPL followed by a jump-in to adhesive contact from a large separation DJ = 40.7 ± 8.9nm to a final distance of D0 = 2.6 ± 0.5nm, corresponding to the thickness of the PL layer (a similar thickness of D0 = 2.8 ± 1nm was also obtained for the mica vs. mica-PL system). The origin of this attraction between a constant charge surface (mica-PL) at a large positive potential and a constant potential surface (gold) at much smaller positive potential arises from the surface charge inversion induced by the approach of the surfaces, as discussed in detail earlier30. This is seen by the fit to the data of the PB equation, solid black curve in fig. 5. Successive approaches at the same contact point, further verifies that poly-L-lysine does not adsorb to the gold surface following jumping-out of contact.
11 ACS Paragon Plus Environment
Page 12 of 32
Page 13 of 32
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
To assure that the attraction is electrostatic and not due to bridging interaction (or both), we measured the shear force Fs between the surfaces at different separations (inset to figure 5). An increase in shear force in response to applied shear motion (i.e. F ≅ 0μN) at separations larger than the thickness L of the PL layer D>3nm is an indication for bridging interaction39. As clearly seen from the shear measurements there is no increase in shear force between the PL chains and the gold surface down to D = 2.5nm. This shows that neither steric repulsion nor bridging interactions act between the surfaces at separations larger than this, signifying that the interaction is governed solely by double layer forces. This confirms that at OCP and in water with no added salt gold is positively charged. The nature of the gold-mica contact. The gold-mica contact is intrinsically different to that between two mica surfaces. For mica-mica contact in low salt concentration solutions their surface charge is balanced by condensation of hydronium ions (H3O+)50. Consequently, the adhesion energy (Wa) is predominantly determined by vdW forces and a typical value of ~10mJ/m2 is obtained51. However, when a gold surface (at constant potential, CP) contacts a mica surface (at constant charge, CC), then, as theoretically predicted by McCormack et al.31 for the approach of a CP surface to one at CC, different behavior ensues. The surface charge on the gold turns more positive so at contact it becomes equal and opposite in sign to that of mica %&'(
%&'(
(|−σEF | = |σ |), where σ and σEF are the (constant) surface charge density of the mica surface, and of gold at contact, respectively. This behavior, which obviously has an effect on adhesion energy between a metal and other surfaces, has not been explicitly demonstrated experimentally to date. As a result, it is not clear, for instance, if once the gold-mica contact is reached, all counterions are excluded from the gap or whether a small portion of H3O+ condenses into the mica surface to decrease its surface charge, as happens for the mica-mica case.
12 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
To address this issue, gold and mica were repeatedly brought into adhesive contact, 5±1 minutes subsequent to separation, and, for each approach, the normal force profile and adhesion energy were measured (Figure 6). It has already been demonstrated by Raviv et al. that the long-range repulsion between two mica surfaces is markedly reduced in a subsequent approach, immediately (~5min) after separation, due to the slow release of condensed counterions to the bulk and the resultant decrease in mica surface charge density52. Thus, if counterions condensation indeed occurs, a reduction of the attractive forces between gold and mica is expected at short times following approach to contact, re-separation and subsequent approach. As clearly shown in figure 6 no such reduction was observed and all force profiles show similar behavior. Furthermore, adhesion energy measurements following each approach gave similar values with an average of W = 52.1 ± 1.5mJ/m . These results directly indicate that once micagold contact is reached all counterions of the electric double layer are excluded from the goldmica gap and no counterion condensation occurs, as schematically shown in the inset to figure 6, indicating that at contact the surface charge of gold is equal and opposite in sign to that of mica as predicted by theory31. Adhesion energy. The adhesion energy (WA) between mica and gold in water with no added salt, as calculated from different experiments, is WA = 64.6 ± 12.6 mJ/m2 (the scatter is attributed to the slight differences in RMS roughness of the various gold surfaces studied in this work). The adhesion energy between the surfaces was calculated using the JKR theory53, WA = −
2Fpull-off 3πR
, where Fpull-off is the measured pull off force and R is the surface radius of
curvature. Using the vdW equation for the energy between two flat surfaces WK(L = −A! ⁄12πD , where AH is the mica-water-gold Hamaker constant and D is the separation between the surfaces, the theoretical vdW interaction between gold and mica can be calculated and its relative contribution to the adhesion energy can be estimated. Taking a “cutoff” contact
13 ACS Paragon Plus Environment
Page 14 of 32
Page 15 of 32
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
distance of D0 = 2.2Å50 and a Hamaker constant of AH = 9×10-20J (obtained from the theoretical fit in figure 3) the predicted vdW interaction across water is WvdW = − 49mJ/m2 . The electrostatic force per unit area (i.e. the pressure) at contact between two surfaces, where one is at constant charge and the other at constant potential, is given by the following equation31 FO ⁄A = − 2πσ − σ%&'( ⁄ε εP, where σ%&'( = 0.19 mC⁄m and σ = −2.7 mC⁄m are the surface charge density of gold and mica at infinity as obtained from theoretical fits in figure 3. Therefore, the electrostatic energy (We) equals the work needed to bring the surfaces from D = 150nm (where electrostatic attraction commences as depicted in figure 3) to E
contact WO = 1⁄A R FO dD. Under the assumption that the electrostatic force is constant at all separations and equal to its contact value (FO ⁄A), the maximal electrostatic energy WO, can be calculated to give WO, = FO ⁄A∆D = −9.7 mJ⁄m . This unambiguously shows that adhesion between gold and mica is mostly governed by vdW interaction as in practice We is even smaller since the electrostatic attraction turns lower with increasing separation. By summing both electrostatic and vdW contributions, the obtained theoretical adhesion energy between gold and mica is W = WO, + WK(L = −58.7mJ/m , corresponding closely to our experimentally measured values noted earlier above. Interaction under external potentials. To determine the potential range at which gold is ideally polarized and no reaction occurs at its surface, a cyclic voltammetry (CV) scan was performed in 1mM NaNO3 and against a platinum quasi-reference electrode (QRE), as shown in figure 7. The obtained voltammogram shows the characteristic anodic (oxidation) and cathodic (reduction) peaks for polycrystalline [111] gold54 and the so-called double layer region (-0.4 to 0.3V), where it is ideally polarized. A CV scan at potential range of -0.3 to 0.2V, shown in the inset to figure 7, further demonstrates that at the double layer region no faradaic current flows and neither oxidation nor reduction occurs at the gold surface.
14 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Typical Fn(D) force profiles are shown in figure 8 for the gold surface held at different applied potentials against a platinum quasi reference electrode (Pt-QRE). As can be clearly seen, the interaction strongly depends on applied potential Ψapp and gradually changes from attraction to repulsion as Ψapp varies from positive to negative values. This also alters the adhesion energy between the surfaces relative to its value at OCP (Figure S3). For all force profiles, at large separations (∼120nm) an electrostatic interaction acts between gold and mica while at short separations (∼10nm) vdW interaction prevails and the surfaces jump-in to adhesive contact. The black solid lines in figure 8 represent the best fit of the PB at CC-CP boundary conditions to the data using a Debye length of κ-1 = 55nm, mica surface potential of Ψsmica = -110mV and Hamaker constant of AH = 9×10-20J. The obtained fits agree very well with the data for all applied potentials and give surface potential values, from top to bottom, of -157, -102, -59, -4 and 24mV. We note, as has been discussed in detail earlier30, that in these conditions electrostatic attraction can occur even between surfaces of similar-sign potential. This is most clearly seen for the blue data in fig. 8, where the gold constant potential surface is at -59mV facing a mica (constant charge surface) at potential -110mV, the attraction arising from charge inversion on the constant potential surface (gold) as the surface approach30. The accurate description of the gold-mica interaction enables us to extract fundamental characteristics of the gold surface, such as its potential of zero charge (PZC). Although such information has been obtained for metals (including gold) using classical electrochemical approaches47,
55,
it has never, to our knowledge, been extracted from direct surface forces
measurements as in this study. The advantage in our approach is that all electrochemical measurements need to be carried out in the presence of sufficient electrolyte concentration (>ca. 1mM) while, our force measurements may be carried out in pure water with no added salt. Typically the PZC of solid metals is obtained by differential double layer capacitance Cd measurements for different applied potentials in dilute solutions (10-4-10-3M)55. The PZC is 15 ACS Paragon Plus Environment
Page 16 of 32
Page 17 of 32
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
determined from the point where the Cd curve has a minimum47. Since the electric double layer (EDL) charge is divided into two layers, immobilized charges in the outer Helmholtz plane and scattered charges in the diffuse layer, the EDL capacity is generally written as 1YC = 1YC + ( ! 1Y , where CD and CH are the diffuse-layer and Helmholtz capacitance, respectively47. At low CE electrolyte concentration, i.e.
