Vapor-phase deposition-based self-assembled monolayer for an electrochemical sensing platform

Since the first report from Sagiv group in 1980 on the self-assembled monolayer (SAM),¹ it has been a promising surface coating and modification technique due to its good bonding strength, low surface energy, low friction forces, and good thermal stability.^2–5 A SAM has been widely applied to the artificial control of surface chemical functionalities, such as hydrophobicity/hydrophilicity, propensity for hydrogen bonding, and charge polarity/density.⁶ By taking these advantages, a SAM has been utilized for various ranges of research fields including microelectronics, optoelectronics, thin-film technology, protective coatings, bioactive surfaces, cell adhesion, protein adsorption, chemical sensors, and biosensors.^7–11 

The SAMs in most studies are developed in a liquid-phase process by intermolecular interaction between the SAM substrate surfaces and the solution containing adsorbing molecular species.^12–15 Commonly, thiol and alkylsilane groups have served as head groups of absorbing molecules which are covalently bound on gold and glass surfaces, respectively. The liquid-phase approach has an advantage in that it is easy to carry out the experiment in such a manner that the solution comes into contact with the self-assembling reaction surface. However, the liquid-phase deposition process not only requires an additional large amount of solvent potentially but also is difficult to be controlled for the reproducible surfaces.^2,16,17 In addition, it is not sufficient to form a SAM with high quality at the porous surface since the liquid-linked molecular species can hardly penetrate into the narrow area, and it takes a long time to make a saturated SAM, consequently leading to a high production cost.¹⁸ 

In order to overcome these disadvantages in a liquid-phase deposition, several research groups have attempted a vapor-phase deposition process as an alternative.^2,6,16–19 It is based on a solvent-free chemistry by vaporizing the SAM molecules, and it uses the interaction between the SAM surfaces without liquid interaction forces unnecessarily. For the vapor-phase process, it is easier to control the deposition conditions, and the adsorbing molecular species can penetrate deep into the porous surface matrix with efficient mass transfer, resulting in a SAM with a high-aspect-ratio structure. Crooks and his co-workers have demonstrated the chemistry of vapor–solid interfaces by investigating the SAMs constructed onto the Au surface with molecules such as 4-mercaptobenzoic acid (MBA), 3-mercaptopropionic acid (MPA), and 11-mercaptoundecanoic acid (MUA) through FTIR (Fourier-Transform InfraRed spectroscopy)–ERS (External Reflection Spectroscopy) and Thickness-Shear Mode Resonator (TSMR) analyses.²⁰ 

Recently, Oh et al. presented an automated optical analyzer for a simple, reliable, and multiplex detection of allergen-specific immunoglobulin E.²¹ However, such an optical module-incorporated instrument is bulky and expensive, which is a significant limitation to be applied to various situations, e.g., the point-of-care testing (POCT).²² With the aim of overcoming these problems, large amounts of electrochemical sensing strategy-based studies have been reported, in which SAMs prepared in a liquid-phase process serve as a receptor substrate toward various targets.^23–26 However, there is still a significant challenge to overcome a low signal reproducibility and reliability in electrochemical biosensors.²⁷ Herein, we describe a vapor-phase deposition process as an alternative of the SAM preparation method, which results in an enhanced performance of the electrochemical biosensor. In addition, we experimentally compare the electrochemical stability of Au–MPA SAMs prepared in both liquid-phase and vapor-phase processes by measuring using cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS). Furthermore, we optimize the various conditions of the vapor-phase deposition process to verify the reproducibility of the impedance signals from the Au–MPA SAMs and the feasibility of the SAMs as an electrochemical bio-sensing platform.

For fabricating the electrode sample, a glass plate was immersed into a piranha solution (H₂SO₄:H₂O₂ = 4:1) for 20 min, the electrode surface was then thoroughly washed with de-ionized water (DW), and dried under a N₂ gas flow. A stencil mask was produced to make a patterned electrode as shown in Fig. 1(a) and attached onto the cleaned glass surface. The coating of chromium (10 nm) on the glass plate was followed by 100 nm gold (99.999%) thin layer formation by using an electron-beam evaporator. Finally, we attached a Kapton® tape to the electrode to separate the sample solution against the region to be connected to the potentiostat.

The vapor-phase deposition system was constructed as shown in Fig. 1(b). A source vessel containing MPA and a reaction chamber was sufficiently heated over 1 h before the vapor-phase deposition process. The as-prepared Au electrode was loaded into the reaction chamber and heated for 30 min in a vacuum (<10 mTorr). By opening the valve between the source vessel and the reaction chamber, the vaporized MPA is exposed to the reaction chamber to carry out the SAM formation reaction on the Au surface [Fig. 1(c)]. Then, the electrode was immersed in a solution containing 30 mM N-hydroxysuccinimide (NHS) and 150 mM 1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide (EDC) for 1 h, followed by washing with DW, and dried under a N₂ gas flow. 80 µl of 5 µg ml⁻¹ IgE antibody in a phosphate-buffered saline (PBS) buffer (P4417, Sigma-Aldrich) was applied on the gold electrode surface incubated for 1 h at room temperature, followed by incubation in the BSA solution for preventing non-specific binding of target IgE. Finally, 80 µl target IgE in the PBS buffer was allowed to bind with the antibody on the Au electrode surface for 1 h at room temperature.

The left, middle, and right parts in the thin film Au electrode were used as a counter, working, and pseudo reference electrode, respectively. Note that the potential difference between the Au reference and the Ag/AgCl reference and a half sum of the peak potentials is approximately 230 mV and 0 V, respectively.²⁸ Electrochemical impedance spectroscopy (EIS) was performed using Bio-logics VSP (France). Impedance was measured at a DC voltage of 0 V vs reference and an AC voltage of 10 mV in the frequency range from 1 Hz to 100 kHz by using the PBS buffer containing 2.5 mM Fe(CN)₆^3-/4-. The experimental impedance data were analyzed by software, Bio-logic EC-Lab (France), in the form of complex plane diagrams (Nyquist plots). The charge transfer resistance from each impedance spectra was extracted by the Randles circuit model using the software as mentioned above. Cyclic voltammetry was performed with the ideal potentiostat from −0.5 V to 0.5 V with 100 mV/s scan rate.

In this study, we first investigated the electrochemical stability of the Au–MPA SAM deposited in both vapor-phase and liquid-phase processes by measuring the cyclic voltammetry (CV) current signal from the following three substrates: bare Au electrode, Au–MPA SAM prepared in a liquid-phase, and Au–MPA SAM prepared in a vapor-phase. As shown in Fig. 2, reduction and oxidation peak currents of Fe(CN)₆^3-/4- at the bare electrode were measured at ∼−0.09 V and +0.07 V, respectively. In addition, the peak current signals (black dashed line) in the first cycle and the second cycle were almost unchanged. In a liquid-phase deposition, the oxidation peak current at ∼+0.07 V is slightly reduced due to the electrical repulsive phenomenon between the acidic surface of the Au–MPA SAM and Fe(CN)₆^3-/4- contained in a buffer solution. Interestingly, a peak current at ∼+0.45 V was measured, and a subsequently recorded reduction peak current was nearly ideal with that of the bare Au electrode. It seems that the peak current signal at +0.45 V is generated when peeling off the MPA bound on the electrode surface, so the subsequent reduction current is almost the same as that of the bare Au electrode. On the contrary, the current signals at the first and second cycles from the Au–MPA SAM prepared in a vapor-phase process were kept constant, but the peak current signals were significantly reduced than those from the bare Au electrode, which clearly demonstrates that the Au–MPA SAM prepared in a vapor-phase process is sufficiently stable to be utilized as an electrochemical sensing platform.

Next, the cyclic voltammetry current signals from the samples prepared under various conditions such as the temperature of the reaction chambers and the MPA source exposure time were examined. As shown in Fig. 3, the peak currents decrease as the temperature of the reaction chambers and the MPA source exposure time increase. In addition, the peak current deceases as the potential differences between the anodic and the cathodic peaks increase at this time. In addition, no more meaningful decrease in peak currents was observed over 170 °C, which indicates that the surface of the Au electrode was fully covered with the MPA at the chamber temperature of 170 °C. We also measured the time-dependent pressure changes in the reaction chamber from the beginning of the source MPA exposure at various temperatures of the source vessel (Fig. S1). As envisioned, the higher vapor pressure was observed on increasing source temperature. Based on the above results, Au–MPA SAMs were finally prepared at the chamber temperature of 170 °C, the source MPA temperature of 70 °C, and the source exposure time of 30 min, followed by measuring the impedance signals from each electrode sample. However, the EIS signal was observed to be quite poorly reproducible.

With the aim of investigating the cause of reproducibility, AFM experiments were conducted to examine the morphology of the Au electrode surface. As shown in Fig. 4, large amounts of grains were certainly observed in the AFM topography image on the bare Au electrode. After forming the Au–MPA SAM on the Au electrode surface by the vapor-phase deposition process, however, the number of the Au grains was decreased and a couple of grains were seemed to be merged. In addition, Root Mean Square (RMS) surface roughness obtained from the AFM images reduced from 1.305 nm to 1.163 nm, which seems to be due to the excessively deposited MPA on the Au electrode surface during the vapor-phase deposition process.

In order to obtain the reproducible impedance spectra from different electrode samples, the optimal conditions of vapor-phase deposition were examined by calculating a charge transfer resistance from impedance data of each sample. Several experiments were conducted under varying conditions: the chamber temperatures, the source MPA temperature, the source MPA exposure time, and the pumping time after the reaction. The highest reproducibility and precision condition were observed with the lowest coefficient of variation at the chamber temperature of 90 °C, the chamber temperature of 40 °C, the source exposure time of 4 min, and the pumping time of less than 40 min (Fig. 5 and Figs. S3–S6). Finally, we evaluated the feasibility of an electrochemical bio-sensing platform by immobilizing the IgE probe antibody on the Au–MPA SAM prepared in a vapor-phase deposition process and incubating the target human IgE antibody on that platform. As shown in Fig. 6, the R_ct value increases as the target IgE concentration increases because target human IgE bound on the Au electrode surface induces the inhibition of electron transfer of Fe(CN)₆^3-/4- toward the Au electrode. On the other hand, there was no distinguishable change in the Rct value when using electrode samples prepared by the liquid-phase deposition process.

In this study, we investigated the electrochemical Au–MPA SAM stability in a vapor-phase deposition process. The Au–MPA SAM prepared in a vapor-phase deposition process exhibited much higher reproducibility of the impedance signal than in a liquid-phase deposition process since MPA on the Au electrode surface was easily detached when potential was applied. In addition, we optimized the conditions of the vapor-phase deposition process to be used as a reproducible electrochemical sensing platform, and we applied the Au–MPA SAM in the optimal conditions for the electrochemical detection of target human IgE without labeling. We expect that the observations made in this study would accelerate the development of label-free impedimetric biosensors.

See the supplementary material for the experimental and more EIS characterizations.