Fast Kinetics of Thiolic Self-Assembled Monolayer Adsorption on Gold

Oct 29, 2014 - However, it is well-known that SAMs have a fast growth rate accompanied by a rearrangement step, from the beginning of the self-assembl...
0 downloads 7 Views 3MB Size
Article pubs.acs.org/JPCB

Fast Kinetics of Thiolic Self-Assembled Monolayer Adsorption on Gold: Modeling and Confirmation by Protein Binding Sasan Asiaei,†,§ Patricia Nieva,*,† and Mathilakath M. Vijayan‡,∥ †

Department of Mechanical and Mechatronics Engineering and ‡Department of Biology, University of Waterloo, Waterloo, ON, Canada N2L 3G1 § School of Mechanical Engineering, Iran University of Science and Technology, Tehran, Iran 1684613114 ∥ Department of Biological Sciences, University of Calgary, Calgary, AB, Canada T2N 1N4 ABSTRACT: This study presents an improved kinetics for the formation of selfassembled monolayers (SAMs) of thiols on gold substrates. Based on predictions of a computational model developed to study the SAM growth kinetics, SAMs of 11-mercaptoinic acid and 1-octanethiol were successfully formed for the first time within 15 min by incubation of planar gold chips in a 10 mM solution of thiols in pure ethanol. The performance of this new rapid SAM formation protocol is compared to the conventional 24 h incubation protocol by evaluating the binding capacity of a fluorescent-labeled antibody to the SAM samples prepared using both protocols. Tetramethylrhodamine conjugated polyclonal goat γ-globulin (IgG) was bound to all SAMs previously modified with 1-ethyl-3-(3-(dimethylamino)propyl)carbodiimide (EDC) to improve antibody immobilization. Resulting binding density of the fast SAM was evaluated using epifluorescence and atomic force microscopy (AFM) and found to be comparable with reported values in the literature using conventional 24 h protocols.

I. INTRODUCTION

few days, depending on the SAM type and its concentration.1,4,9−12 Currently, the majority of the work performed on SAM kinetics employs the Langmuir kinetics model and assumes that the rate of adsorption is proportional to the free space on the surface.4 Although the Langmuir model provides consistent description of SAM growth kinetics,13−16 its implementation is limited to very low concentrations (∼1 μM) of adsorbates.12 In addition, mutual movement of molecules within the monolayer, called rearrangement step, is excluded. However, it is wellknown that SAMs have a fast growth rate accompanied by a rearrangement step, from the beginning of the self-assembly process.11 In this study, a numerical model, in which the mass transport of thiol is coupled to its surface reaction, is presented for the prediction of SAM growth. The model accounts for the initially fast growth rate of SAMs, as well as the rearrangement step, by assigning higher kinetics constants to the fast step and considerably lower kinetic constants to the rearrangement step.11 The model shows that increasing the concentration nearly exponentially enhances the SAM growth kinetics. Based on the model predictions of the time required for the selfassembly, gold surfaces were incubated for 15 min in a 10 mM self-assembly solution. Additional samples were also prepared by using a conventional 24 h incubation process performed in a 1 mM solution.9,10 The quality of the SAM formed in each set

Self-assembled monolayers (SAMs) of thiols have numerous applications in different areas, including biology, electrochemistry, electronics, and micro/nano fabrication. This is due to the unique physical, chemical, and photonic properties of SAMs.1,2 SAMs can easily form a packed, robust, and thin interface in the order of a few nanometers, and their surface properties allow the mimicking of natural environments, providing a suitable substrate for molecular recognition in biomedical and analytical devices.2 In these applications, SAMs are often formed by chemisorption of thiols from an ethanolic solution to the surface of noble metals such as gold and silver.1,3,4 Typically, SAMs are formed with carboxylic end groups to facilitate the binding of the amine group of a probe protein to the SAM.5,6 The probe protein is selective to a given biomolecule of interest, called the target. After immobilization of the probe protein (which in most cases is an antibody) to the SAM, the interface consisting of SAM and probe protein is ready to bind to the target molecule. An interface formed from a high quality SAM usually results in a higher binding density of probe proteins.7,8 One of the most influential factors in successful self-assembly of thiolic monolayers is the processing time.1,4 SAMs are conventionally formed by incubation of the metal surfaces in the thiolic solution for times ranging from 24 h to a few days.1,4,9,10 SAM growth from a low concentration of thiols, such as 1 μM, is known to follow a slow reaction-limited kinetics with a diffusive thiol transport mode. Hence, monolayer formation is expected to take several hours to a © XXXX American Chemical Society

Received: October 2, 2014 Revised: October 28, 2014

A

dx.doi.org/10.1021/jp509986s | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

where θ0 = 7.3 × 10−10 mol m−2. The corresponding coverage rates are described by the following equations:11

was assessed by binding a fluorescently tagged antibody to the SAMs. The binding efficiency was also examined using atomic force microscopy (AFM). Experimental results for both incubation periods are comparable to previously reported surface binding density of antibodies,6,10 confirming our prediction of the formation of a monolayer within a 15 min self-assembly process. The fast SAM growth achieved is therefore promising for biomedical applications as it considerably reduces the usually long conventional 24 h processing times involved.

dθ1 = kac − (kac + kd + k t)θ1 − kacθ2 dt

(4)

dθ2 = k tθ1 dt

(5)

where ka = 11.5 × 10−3 m3 mol−1 s−1, kd = 6 × 10−4 s−1, and kt = 4.5 × 10−4 s−1 respectively denote the association, dissociation, and rearrangement rate constants.11 As mentioned before, Rs is proportional to the coverage rate. Using eq 3, the reaction rate (Rs) is described as11

II. MATHEMATICAL MODEL The typical configuration of a self-assembly chamber is shown in Figure 1. The solution volume for all samples is chosen in

R s = θ0(dθ1 + dθ2)/dt

(6)

Substitution of eqs 4−6 into eq 2 yields the following governing equation for the thiols’ rate of surface concentration change on the reaction surface: ∂cs = ∇·(Ds∇cs) + kac(θ0 − cs) − kdθ0θ1 ∂t

The boundary conditions of eq 1 provide the necessary additional equations required to solve eqs 1−7 for cs. The surface concentration of chemisorbed thiols (cs) on the reaction surface, which is contained in the Rs term (eq 2), is coupled to the bulk concentration of thiols (c) by the boundary condition of eq 1, as a mass flux equation:

Figure 1. Typical configuration of a self-assembly chamber (not to scale).

n ⃗ ·( −D∇c) = −R s

such a way that the chamber contains enough thiols, and solution concentration remains almost constant during the experiments.17 Gold surfaces were assumed to be flat. In this configuration, the SAM growth kinetics can be reasonably considered two-dimensional, as the variation of the properties across the depth is negligible. The purpose of the simulations is finding the instantaneous concentration of the chemisorbed thiols (cs) on the reaction surface. In this regard, the concentration of the thiols in the bulk fluid (c) should be determined first. Diffusion transports the thiol molecules to the self-assembly sites and the mass transport equation can be written as follows: ∂c = ∇·(D∇c) ∂t

n ⃗ ·( −D∇c) = 0

(9)

Moreover, the concentration of bulk fluid is c0 at the beginning of the reaction: c = c0

for t0 = 0

(10)

COMSOL Multiphysics was used to simulate the physics, which was introduced using a triangular mesh. The independency of the simulation results from the mesh size was assessed with a less than 1% numerical results variation margin. The simulations completed when the surface concentration of thiol species (cs) exceeded 99.6% of its maximum possible value (θ0 = 7.3 × 10−10 mol m−2)9,10 for the first time.

(1)

where D = 5.7 × 10−10 m2 s−1 is the diffusion coefficient. The concentration of the thiols in the bulk fluid is related to the surface concentration of the chemisorbed thiols (cs) via the reaction rate term (Rs) and by taking into account the surface diffusion of the thiols:

III. EXPERIMENTAL VERIFICATION For the formation of the SAMs, gold-coated glass substrates (Platypus Technologies, LLC, Madison, WI) were diced into 1.27 cm × 1.27 cm (0.5 in. × 0.5 in.) chips and rinsed for 2 min with hexane to remove hydrophobic contaminants. The rinsed chips were then quickly transferred to ethanol for another 2 min wash to remove hydrophilic contaminants, followed by 10 min acetone, 2 min isopropyl alcohol (IPA), and 2 min deionized water (DI) washes. To prepare them for incubation, the chips were thoroughly rinsed with the SAM solvent (ethanol) and completely dried under a stream of nitrogen. A consistent 1.12 nm root-mean-square (rq-rms) surface roughness of gold surfaces was measured by the atomic force microscope (AFM) for chip surfaces. After the self-assembly

(2)

In eq 2, Ds = 2 × 10−18 m2 s−1 is the SAMs’ surface (lateral) diffusion constant18 and Rs is the chemisorption rate of thiol molecules in mol m−2 s−1. Rs is proportional to the coverage rate (dθ/dt). The coverage (θ) is defined as the ratio between the surface concentration of the chemisorbed thiols (cs) on the reaction surface and the maximum possible surface concentration of thiols (θ0); in other words: θ = cs/θ0.11 The coverage has two components, θ1 and θ2, defined as the SAM coverage in the first stage of the adsorption and rearranged assembled monolayer, respectively: θ = cs/θ0 = θ1 + θ2

(8)

where n⃗ is the unit normal vector to the surface. The material balance for the thiol molecules at the nonreacting surfaces is “insulating”,19 which indicates that there is a zero normal flux of thiols into these surfaces:

14

∂cs + ∇·( −Ds∇cs) = R s ∂t

(7)

(3) B

dx.doi.org/10.1021/jp509986s | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

the roughness increased to rq = 1.21 nm. The topographic changes observed here follow the same trend as those reported in the literature.20 It is worth noting that the trends observed are mainly concerned here rather than the global values. The difference in the values reported here and those stated elsewhere is due to the difference in the initial roughness of the gold substrate used.20,21 The ratio of thiols in the solution was always kept at 1:3 (11MUA:1-OT), irrespective of whether the chips were incubated using the fast SAM or the 24 h conventional protocols, to avoid steric hindrance.9 Thiols employed in this research are 11mecaptoundecanoic acid (11-MUA) with carboxylic end groups, and 1-octanethiol (1-OT) which serves as a spacer and support for 11-MUA (both purchased from Sigma-Aldrich, Oakville, Canada). After incubation, the chips were rinsed thoroughly with pure ethanol to remove any unbound SAM molecules from the gold surface. Immediately after the SAM formation, the chips were loaded with fluorescent antibodies using a standard zero-length cross-linking protocol to evaluate the binding capacity of the monolayer formed.6 The antibody bound to the SAM was tetramethylrhodamine conjugated goat immuno-γ-globulin (IgG) (Invitrogen, Burlington, Canada). A 10 mM phosphate buffered saline (PBS) buffer of pH 7.4 containing a maximum of 0.05 mg/mL of antibodies and 100 mM of carbodiimide EDC (also called EDAC or EDCI, acronyms for 1-ethyl-3-(3-(dimethylamino)propyl; Sigma) was used as the antibody/SAM cross-linker.6 In preparation for the characterization of the binding efficiency, the rate of evaporation from the chip surface was controlled by placing the chips in a humidified chamber during the cross-linking process. Control of evaporation is necessary to avoid precipitation of the fluorescent antibodies and crystallization of the buffer salts on the chip surface. A glass slide coverslip was laid on the chip surface to further reduce the surface evaporation. The chips were kept at 4 °C overnight and rinsed the following day with large volumes of 10 mM PBS. The moisture was removed from the surface before fluorescence and AFM characterization, since surface moisture prevents accurate fluorescence and AFM imaging. Fluorescence images were captured using a Zeiss Axiovert inverted epifluorescence microscope set at 550 nm (Zeiss, Toronto, Canada). The AFM images were captured in air with a NanoScope III (Digital Instruments, Santa Barbara, CA) in tapping mode. Images were obtained at a scan rate of 2 Hz with a silicon cantilever (Nanoprobe, cantilever length = 125 μm and resonant frequency = 307−375 kHz). A drive voltage between 150 and 200 mV was used. Unless stated otherwise, all the experiments were performed at an ambient temperature of 22 °C.

Figure 2. Calculated self-assembly periods predicted for different initial concentrations of thiols in ethanol, plotted in log scale. Selfassembly periods decrease significantly by increasing the thiol concentration in the bulk fluid. Data points are connected by straight lines to aid visualization.

reactions were monitored in real time using localized surface plasmon resonance (LSPR) spectroscopy, and the resulting LSPR absorbance peak shift was comparable to the experimental results reported in the literature.23 The key observation from the simulation results is the fast kinetics associated with higher concentrations of thiols in the bulk. The results of the simulations presented in Figure 2 show a nearly exponential reduction of the self-assembly period as a function of thiol concentration, and they indicate complete SAM formation after 15 min when 10 mM thiol solution is used. Hereafter, this 15 min assembly process using a 10 mM SAM solution is called a “fast SAM”. Hence, by adjusting the SAM concentration, we can considerably increase the rate of self-assembly and, thus, reduce the process time. These findings are very promising for biosensing applications, since SAM of 11-MUA and 1-OT is conventionally formed by incubating the chip in the self-assembly solution for 24 h.9,10 A uniform surface distribution of bound antibodies usually suggests a successful immobilization protocol and, therefore, a uniform spread of thiol molecules over the chip surface area.7,8 In general, conventional fluorescence imaging helps confirm the uniformity of antibody binding over a reasonably large surface area (∼1 mm2), while AFM characterization helps verify the binding density of the antibodies. Nevertheless, AFM only provides local data about the antibody binding density and cannot be used to image large surface areas, without sacrificing resolution.24 In this paper, the quality of the SAMs resulting from both assembly processes was first determined using epifluorescence imaging. Figure 3 displays pictures of two chips, one prepared with the fast SAM process (A) and the other using the conventional 24 h SAM process (B), showing similar spreading binding patterns of antibodies. Fluorescence imaging of these samples was performed for more than six times for each chip and in at least two sets of chips for each protocol. To determine the strength of each one of the used immobilization protocols, the number of antibodies bound to the chip surface is counted. Two representative areas in Figure 3 were chosen, and the number of fluorescent spots was counted as depicted by the insets in Figure 4. The regions chosen were those areas with either minimum or maximum density of antibodies observable, along with minimum

IV. RESULTS AND DISCUSSION The self-assembly periods calculated for a number of initial concentrations of thiols in ethanol using the numerical model described in section III are shown in Figure 2. The figure depicts the period required for the completion of the selfassembly process as a function of the initial concentration of thiol molecules in the solution. Simulated self-assembly completion time agreed very well with the literature results for the 1 mM concentration of 11-MUA, with less than 5% variation,11 and Prato et al.’s results for octadecanethiol (C18) SAMon gold surface with the same experimental conditions.22 In another study, we have examined the binding capacity of the 10 mM alkanethiol SAM using biotin−streptavidin. Binding C

dx.doi.org/10.1021/jp509986s | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

Figure 3. Fluorescent antibodies uniformly bound to the SAMs and distributed on the chip using (A) fast SAM and (B) conventional 24 h protocol.

Figure 4. Method used to quantify the immobilization strength of the SAM formation protocols. The fluorescent antibodies are colored with white dots and counted in two randomly chosen representative areas which had no precipitation. On average, 5−60 spots were resolved in the fast SAM (A) and 2−15 spots in the conventional protocol (B) within a 100 μm × 100 μm area.

inferred because the upper resolution limit of the microscope used was 1.1 μm and therefore cannot resolve a single antibody that requires a precision of about 14 nm.26 The fluorescent spots are thus attributed to a dense accumulation of antibodies (cluster) attached to the surface. The clustering occurs because the conjugation chemistry used can attach antibodies together.6 The resulting surface loading density of antibodies is comparable to the corresponding protein loading density reported in the literature.6 The EDC conjugation chemistry is reported to have a yield of 1−5% in the present reaction environment.6 Therefore, the total surface concentration of the available sites for antibody immobilization is between 7.3 × 10−13 and 3.65 × 10−12 mol m−2, corresponding to 4.4 × 103−

precipitation. Precipitation arises due to uneven evaporation of the buffer fluid and crystallization of buffer salts over the chip surface during the antibody conjugation process. Areas with precipitation can be easily resolved under the fluorescence microscope, as they resemble reasonably large crystallized salts over the surface. The number of spots with maximum detection resolution of 1.1 μm25 was counted in a 100 μm × 100 μm square. An average of 5−60 fluorescent spots were detectable in the fast SAM protocol, while the corresponding average for the conventional protocol was 2−15 spots. While the existence of fluorescent spots corresponds directly to the immobilization of antibodies on the surface, they are not each related to an individual antibody molecule. This can be D

dx.doi.org/10.1021/jp509986s | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

Figure 5. AFM image of the fluorescent antibodies bound to a sample prepared using the fast SAM, in the left (A′, B′, C′), and conventional selfassembly protocol, in the right (A, B, C); (A) and (A′) are 10 μm2 surface area of each chip which are selected randomly and imaged by AFM; 1 μm2 region of the imaged area (inside the initial 10 μm2 surface area) is selected randomly and depicted in (B) and (B′) showing clusters of antibodies; (C, C′) show side view of antibody clusters with smooth height transitions.

2.2 × 104 binding events in an 100 μm × 100 μm area.9,10 The fluorescent spots detected here (Figure 4) correspond to antibody clusters that have been cross-linked to each other and also to the surface. The detected clusters are at least 1.1 μm in width, and since each antibody is around 14 nm in width,27,28 it is estimated that each cluster contains at least around 80 immobilized antibodies. Therefore, a rough estimate of the number of cross-linked antibodies within each cluster is approximately 160−1200 for the conventional protocol and 400−4800 antibodies in the fast SAM protocol, in each 100 μm × 100 μm area. These values are below the corresponding literature predictions (4400−22 000).6 However, since the antibodies are roughly 14 nm in width,27 they are, in principle, only detectable by AFM or spatially modulated illumination (SMI) microscopes.29,30 Therefore, in those regions of the chip where no fluorescent spot is detectable, no conclusions can be made regarding the strength of the immobilization protocol. Nevertheless, as shown in Figure 3, both protocols seem to effectively produce a strong bond between the carboxylic group of SAM and the amine group of antibody. On the other hand, visual counting of the binding events only provides a qualitative assessment of the immobilization protocol strength. It should

be noted that this work did not attempt to quantify antibody binding capacity and compare the strength of the two protocols, but to confirm it. Moreover, a mathematically uniform distribution of fluorescent spots is not expected here. The distribution range reported in the literature (4.4 × 103−2.2 × 104 binding events in an 100 μm × 100 μm area9,10) is an statistical average and there might be regions on surface with lower or higher antibody loading densities, compared to the expected range, irrespective of the SAM formation and/or conjugation protocol used. This has been experimentally confirmed using a SMI microscope,29,30 in which the spread of antibodies over the surface was not observed to follow a mathematically uniform distribution. To confirm antibody binding, AFM characterization was performed for two sets of chips using an apparatus with a maximum resolution of 2 nm for solid objects with a sampling number of 512 per μm. The surface area imaged during all the tests performed in this study could not be increased above 1 μm2 to keep the resolution around 2 nm. The images generated by the AFM for the fast and conventional 24 h SAM protocols are shown in Figure 5. E

dx.doi.org/10.1021/jp509986s | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

The right-hand AFM images in Figure 5 correspond to the 24 h conventional protocol, and the left-hand images correspond to the fast SAM. Figures 5A and 5A′ show the results obtained for antibody-binding density measurement on two of the two chip pairs tested. Figures 5B and 5B′ present the zoomed-in images of the 1 μm2 bounded regions in Figures 5A and 5A′, showing two and three objects, respectively. Figures 5C and 5C′ show their respective side view. In Figures 5C and 5C′ it can be observed that the objects have a smooth height transition at their edges. Moreover, their relative height is around 7 nm, which is in accordance with the shape and values reported for these antibodies’ height in the literature.20 Nevertheless, their width is variable and different from the typical 14 nm width of an antibody,27,28 which can be justified by the conjugation chemistry used.5 In other words, the protocol used to cross-link the antibody to the SAM allows clustering because antibodies have carboxylic groups as well as amine groups. Therefore, the cross-linker can potentially attach antibodies to each other, forming antibody clusters with variable widths.5 Because of their shape and height, the objects seen in the AFM image are considered to be antibody clusters bound to the surface. Assuming that each antibody is around 14 nm in width,27,28 the left object in Figure 5B can be considered a cluster of around 7 antibodies and the one in the right is considered to be a cluster of 10 antibodies. Similarly, for the fast SAM process the left object in Figure 5B′ can be considered a cluster of around 6 antibodies, the one in the center is considered to be a cluster of 3 antibodies, and the middle right corner cluster encloses 6 antibodies. In general, the number of antibodies in clusters varied randomly over all the chips independently of the SAM protocol used. Imaging using the AFM was repeated for 6 times for each chip due to the antibody instability at room temperature.31 This allows the resolving 0−2 clusters per μm2 for the conventional protocol and 0−3 clusters per μm2 for the fast SAM protocol. Both of these ranges are comparable to the value of 0.44−2.2 antibodies/μm2, calculated from the reported range of antibody binding densities in the literature.6,10 The cases in which the binding density observed is closer to zero might be to denaturation of the antibodies over time during the AFM experiments.31 Moreover, the reported distribution range of 0.44−2.2 antibodies/μm2 was calculated from statistical data provided on larger than 1 μm2 areas. Since a fraction of an individual antibody is hardly resolved, a more realistic distribution range might be any digits between 0 and 2 clusters/μm2. Slightly higher antibody binding densities were observed for the fast SAM protocol when using AFM. Although similar values were observed using fluorescence microscopy, due to the lower resolution of this method, results on the antibody binding density were not conclusive. The lower binding density observed for the conventional 24 h protocol might be attributed to the loss of activity of the carboxyl-ended thiols (11-MUA) in the prolonged conventional SAM formation process, due to the oxidation, or desorption from the surface to be replaced by the other spacer thiols (1-OT).17 The protein characterization results from the AFM experiments have been summarized in Table 1 and compared with relevant literature values for both protocols. The results from fluorescence imaging confirm a quite uniform antibody binding, while the AFM experiments verify the repeatability of the distribution range of antibody binding density. However, AFM imaging was conducted only up to 6 times for each chip with considerably large overall dimensions

Table 1. Summary of SAM/Antibody Binding Characterization of Samples Prepared in This Study Using the Fast SAM and the 24 h Conventional Protocols Compared to the Literature SAM formation protocol 15 min fast SAM 24 h process (this work) 24 h process (literature)

average height of the antibody/type 7 nm/rhodamine goat IgG (this work)

antibody binding density 0−3/μm2 0−2/μm2

7 nm/bovine IgG20

0.44−2.2/μm2 6,10

of 1.27 × 1.27 cm, due to the instability of the antibodies at room temperature.31 Therefore, it did not allow for statistical testing on the efficiency of antibody surface density distribution with the two techniques. We believe that the low range of antibody distribution observed does not correspond to an improper conjugation or SAM formation protocol. In fact, the fluorescence imaging performed confirms a similar spread of antibodies for the chips loaded with fast SAMs and the ones loaded with the 24 h conventional protocol. In addition, the protein characterization results from the AFM experiments (Table 1) are comparable to the literature data for both protocols.6,10 Therefore, we would like to conclude that for high quality chips to be used in practice the fast SAM protocol entails similar protein binding capabilities as the conventional 24 h protocol.

V. CONCLUSIONS In this paper, the effect of higher concentration of thiols on the process time of self-assembled monolayers on gold surfaces was studied by developing a numerical model incorporating the mass transport of thiol coupled to its surface reaction. The model indicated that a fast 15 min self-assembly process is feasible using a 10 mM thiol solution. Experiments were conducted to examine the quality of the fast SAM in comparison to the SAMs formed using a conventional 24 h protocol by binding fluorescently tagged antibodies to both monolayers. Fluorescence microscopy and AFM were used to assess the antibody/SAM binding density on the surface. Observations during this study indicate that the fast SAM has almost similar antibody binding capabilities when compared to the conventional 24 h protocol. This new fast SAM protocol has particular benefits for the development of biosensors that use SAM based interfaces. Preparation of high quality chips from planar gold surfaces using a 15 min SAM formation protocol would greatly cut down the time and cost of current immobilization procedures that use conventional 24 h SAM incubations.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (P.N.). Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors acknowledge the funding for this work provided by the Natural Sciences and Engineering Research Council (NSERC) of Canada, the Ministry of Research and Innovation (MRI), Canada, the Foothills Research Institute, and Early Warning Inc. The authors also acknowledge the products and F

dx.doi.org/10.1021/jp509986s | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

(21) Vertova, A.; Forlini, A.; Rondinini, S. Probing the Electron Transfer Process of Cytochrome C Embedded in Mixed Thiol SAM on Electrodeposited Gold. J. Electrochem. Soc. 2012, 159, 181. (22) Prato, M.; Moroni, R.; Bisio, F.; Rolandi, R.; Mattera, L.; Cavalleri, O. Optical Characterization of Thiolate Self-Assembled Monolayers on Au(111). J. Phys. Chem. C 2008, 112 (10), 3899−3906. (23) Asiaei, S.; Denomme, R. C.; Marr, C.; Nieva, P. M.; Vijayan, M. M. Fast Self-Assembly Kinetics of Alkanethiols on Gold Nanoparticles: Simulation and Characterization by Localized Surface Plasmon Resonance Spectroscopy; SPIE: Microfluidics, BioMEMS, and Medical Microsystems: San Francisco, CA, 2012. (24) Eaton, P. J. Atomic Force Microscopy; West, P., Ed.; Oxford University Press: London, 2010. (25) Axio Observer Inverted Research Microscope. Home page http://www.zeiss.com/microscopy/en_de/products/lightmicroscopes/cell-biology-axio-observer.html (accessed Oct 23, 2014). (26) Lemmer, P. G. Using Conventional Fluorescent Markers for Far-Field Fluorescence Localization Nanoscopy Allows Resolution in the 10-nm Range. J. Microsc. 2009, 235 (2), 163−171. (27) Nobbmann, U.; Connah, M.; Fish, B.; Varley, P.; Gee, C.; Mulot, S. Dynamic Light Scattering As A Relative Tool For Assessing The Molecular Integrity And Stability of Monoclonal Antibodies. Biotechnol. Genet. Eng. Rev. 2007, 24, 117−128. (28) Gunkel, M.; Erdel, F.; Rippe, K.; Lemmer, P.; Kaufmann, R.; Hörmann, C. Dual Color Localization Microscopy of Cellular Nanostructures. Biotechnol. J. 2009, 4, 927−938. (29) Baddeley, D.; Batram, C.; Weiland, Y.; Cremer, C.; Birk, U. J. Nanostructure Analysis Using Spatially Modulated Illumination Microscopy. Nat. Protoc. 2007, 2 (10), 2640−2646. (30) Reymann, J.; Baddeley, D.; Gunkel, M.; Lemmer, P.; Stadter, W.; Jegou, T. High-Precision Structural Analysis of Subnuclear Complexes in Fixed and Live Cells via Spatially Modulated Illumination (SMI) Microscopy. Chromosome Res. 2008, 16 (3), 367−382. (31) Zhengjian, L.; Jianhua, W.; Guoping, C. The Wettability and Topography of Self-Assembled Protein Monolayer Linked by Alkanethiols, 3rd international conference on bioinformatics and biomedical engineering, iCBBE, 2009; IEEE Computer Society: Beijing, China, 2009.

services provided by Canadian Mycrosystems Corporation (CMC) that facilitated this research.



REFERENCES

(1) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Self-Assembled Monolayers of Thiolates on Metals as a Form of Nanotechnology. Chem. Rev. 2005, 105 (4), 1103−70. (2) Ferretti, S.; Paynter, S.; Russell, D. A.; Sapsford, K. E.; Richardson, D. J. Self-Assembled Monolayers: A Versatile Tool for the Formulation of Bio-Surfaces. TrAC, Trends Anal. Chem. 2000, 19 (9), 530−540. (3) Nuzzo, R. G.; Allara, D. L. Adsorption of Bifunctional Organic Disulfides on Gold Surfaces. J. Am. Chem. Soc. 1983, 105 (13), 4481− 4483. (4) Schwartz, D. K. Mechanisms and Kinetics of Self-Assembled Monolayer Formation. Annu. Rev. Phys. Chem. 2001, 52, 107−37. (5) Gizeli, E.; Lowe, C. R. Biomolecular Sensors; Taylor & Francis: London, 2002. (6) Hermanson, G. T. Bioconjugate Techniques; Academic Press: New York, 2008. (7) Park, T.; Choo, J.; Lee, M.; Kim, Y. S.; Lee, E. K.; Lee, H. S. Enhancement of the Protein Loading Density by a Pre-Cleaning Process of a Gold Substrate: Confocal Laser Scanning Microscopic Study. Anal. Sci. 2004, 20 (9), 1255−1258. (8) Sam, S.; Touahir, L.; Salvador, A. J.; Allongue, P.; Chazalviel, J.; Gouget, L. A. Semiquantitative Study of the EDC/NHS Activation of Acid Terminal Groups at Modified Porous Silicon Surfaces. Langmuir 2010, 26 (2), 809−814. (9) Malinsky, M. D.; Kelly, K. L.; Schatz, G. C. Chain Length Dependence and Sensing Capabilities of the Localized Surface Plasmon Resonance of Silver Nanoparticles Chemically Modified with Alkanethiol Self-Assembled Monolayers. J. Am. Chem. Soc. 2001, 123 (7), 1471−1482. (10) Haes, A. J.; Van Duyne, R. P. A Nanoscale Optical Biosensor: Sensitivity and Selectivity of an Approach Based on the Localized Surface Plasmon Resonance Spectroscopy of Triangular Silver Nanoparticles. J. Am. Chem. Soc. 2002, 124 (35), 10596−10604. (11) Damos, F. S.; Luz, R. C. S.; Kubota, L. T. Determination of Thickness, Dielectric Constant of Thiol Films, and Kinetics of Adsorption Using Surface Plasmon Resonance. Langmuir 2005, 21 (2), 602−609. (12) Peterlinz, K. A.; Georgiadis, R. In Situ Kinetics of Self-Assembly by Surface Plasmon Resonance Spectroscopy. Langmuir 1996, 12 (20), 4731−4740. (13) Himmelhaus, M.; Eisert, F.; Buck, M. Self-Assembly of NAlkanethiol Monolayers. A Study by IR-Visible Sum Frequency Spectroscopy (SFG). J. Phys. Chem. B 2000, 104 (3), 576−584. (14) Jung, L. S.; Campbell, C. T. Sticking Probabilities in Adsorption of Alkanethiols from Liquid Ethanol Solution onto Gold. J. Phys. Chem. B 2000, 104 (47), 11168−11178. (15) Schessler, H. M.; Karpovich, D. S.; Blanchard, G. J. Quantitating the Balance between Enthalpic and Entropic Forces in Alkanethiol/ Gold Monolayer Self-Assembly. J. Am. Chem. Soc. 1996, 118 (40), 9645−9651. (16) Xu, S.; Cruchon-Dupeyrat, S.; Garno, J.; Liu, G.; Jennings, G.; Yong, T. In Situ Studies of Thiol Self-Assembly on Gold from Solution Using Atomic Force Microscopy. J. Chem. Phys. 1998, 108 (12), 5002−5012. (17) Graham, D.; Dingman, S. A Step-by-Step Guide for Solution Based Self-Assembly. Mater. Matters 2006, 1 (3), 18−19. (18) Ionita, P.; Volkov, A.; Jeschke, G.; Chechik, V. Lateral Diffusion of Thiol Ligands on the Surface of Au Nanoparticles: An Electron Paramagnetic Resonance Study. Anal. Chem. 2008, 80 (1), 95−106. (19) Jomeh, S.; Hoorfar, M. Numerical Modeling of Mass Transport in Microfluidic Biomolecule-Capturing Devices Equipped with Reactive Surfaces. Chem. Eng. J. 2010, 165 (2), 668−677. (20) Veiseh, M.; Zareie, M. H.; Zhang, M. Highly Selective Protein Patterning on Gold-Silicon Substrates for Biosensor Applications. Langmuir 2002, 18 (17), 6671−6678. G

dx.doi.org/10.1021/jp509986s | J. Phys. Chem. B XXXX, XXX, XXX−XXX