Dynamic electrical measurement of biomolecule behavior via plasmonically-excited nanogap fabricated by electromigration

The dynamic motion of DNA oligomers at the nanoscale gap between nanoelectrodes is measured under plasmonic excitation using laser irradiation. The use of a nanogap enables highly sensitive detection of individual molecules using an electrical readout or an optical readout such as Raman spectroscopy. However, the target molecule must reach the nanogap in order to be detected. This study focuses on the use of plasmonic excitation to trap molecules at the nanogap surface. The nanogap electrode is fabricated by electromigration and is, therefore, a much smaller nanogap than the top-down fabrication in the conventional plasmonic trapping studies. To demonstrate the individual molecule detection and to investigate the molecular behavior, the molecules are monitored using an electrical readout under a bias voltage instead of an optical readout used in the conventional studies. The conductance change due to DNA oligomer penetration to the nanogap is observed with the irradiated light intensity of over 1.23 mW. The single-molecule detection is confirmed irradiating the laser to the nanogap. The results suggest that DNA oligomers are spontaneously attracted and concentrated to the nanogap corresponding to the detection point, resulting in high detection probability and sensitivity.


Introduction
A biochemical sensor using a nanogap is a highly sensitive technique by which molecules can be detected at the individual level. An electrical method using a nanogap electrode defined as a pair of nanoelectrodes with a nanometer gap in between them detects biochemical molecules by measuring an electrical conductance change [1][2][3][4]. DNA oligomers are a type of molecule that can be measured using a nanogap electrode for DNA sequencing. This technique is expected for the identification of individual nucleotides, as each nucleotide will produce an electrical conductance in a specific conductance range. To achieve single-molecule detection at the nucleotide level, an electrode gap of counter nanoelectrodes should be comparable to the molecular size, which is around 1 nm or less. We also refer a plasmonic method using a nanogap. Surface-enhanced Raman spectroscopy (SERS) is a promising technique for highly sensitive biochemical sensing [5][6][7][8][9][10]. The method enables single-molecule identification because the Raman spectrum represents a molecular structure. A gold (Au) nanogap is essential for obtaining high sensitivity when performing SERS, as it enhances Raman scattering light from a target molecule due to localization and enhancement of an electromagnetic field at the nanogap by localized plasmonic resonance. The enhancement factor of Raman scattering light is inversely proportional to the nanogap [11]. A reduction in the gap width to 1 nm or less is significantly useful to achieve the sensitivity required for single-molecule identification. Nanoparticle ensembles have frequently been used to form a nanogap of 1 nm or less between particles. It is extremely difficult to fabricate such a small gap by the electron beam (EB) lithography-based top-down process or the focused ion beam (FIB) process [12][13][14][15][16]. These electrical and plasmonic techniques enable us to perform highly sensitive and direct measurements without complicated, costly, and time-consuming preprocessing such as fluorescent labeling. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Although it is often challenging for molecules to reach a small nanogap when the molecules are found in low concentrations, it is possible for nanogaps to detect a single molecule such as DNA oligomers. In most cases of SERS studies using a substrate, measurements are carried out in dry conditions [17,18]. Target molecules are concentrated at nanogaps after drying solutions. Solution-based single-molecule detection presents a challenge of capturing a target molecule at a nanogap working as a sensing spot. Manipulation and trapping of molecules to a nanogap is an effective way to improve the probability that a molecule reaches a nanogap. Optical trapping by laser focusing restricts the minimum molecular size to be manipulated, resulting in the manipulation of micron size objects such as a cell [19][20][21]. Plasmonic trapping of molecules has the potential to allow manipulation of nanometer-scale molecules, such as proteins and DNA [22,23]. This is because the Au plasmonic structure can localize incident light at the nanogap, resulting in much larger optical force due to an enormous electromagnetic field, as mentioned above. This method can detect a single molecule with plasmonic nanogap structures with a nanoaperture or a mechanically-controllable break junction as well as trapping [22][23][24]. Dielectrophoretic and thermophoretic forces due to the gradient of electromagnetic field from the nanogap have been discussed as a trapping principle [25][26][27][28][29][30]. The strong dielectrophoretic force is able to be realized by the large gradient of the electromagnetic field that is generated at the nanogap.
The reported plasmonic nanogaps are around 10 nm or more because they are produced by the EBlithography-based processes or the FIB processes which have a limitation of processing resolution [12][13][14][15][16]. Such a nanogap is insufficiently small in order to realize the detection of individual molecules; the nanogap of around 1 nm or less is required. In addition, to obtain the larger attraction force, the larger electromagnetic enhancement and gradient are required, which is also realized by reduction in nanogap.
In this study, in order to achieve detection of biomolecules such as DNA oligomers at the individual molecular level using the nanogap, we investigated the molecule behavior under light excitation of the nanogap. We focused on a smaller nanogap and an electrical readout in order to trap and detect the individual DNA oligomer. The nanogap was fabricated by electromigration in order for 1 nm or less nanogap [31][32][33][34]. It is expected that a smaller nanogap forms a higher electromagnetic enhancement and gradient at the nanogap resulting in the larger plasmonic trapping force of smaller biomolecules with lower light intensity. The effect of light irradiation on the behavior of single DNA oligomers at the nanogap was investigated using an electrical readout instead of the optical readout. We performed the time-series measurement of an electrical current between counter nanoelectrodes.

Fabrication
The fabrication process for the gold nanoelectrodes starts from a bare silicon wafer with a thickness of 625 μm. The wafer was thermally oxidized so that the thickness of the silicon dioxide film was 0.8 μm. EB lithography was then performed on the SiO 2 film and then a gold film of 50 nm thickness was deposited via the EB evaporation process. We fabricated the gold nanowire using the lift-off process. The SEM image is shown in figure 1(a). The nanowire was then broken by electromigration. The applied voltage was gradually increased until the current abruptly dropped off. The SEM images of the fabricated nanoelectrodes are shown in figures 1(b) and (c).

Characterization
We characterized the fabricated nanoelectrode after breaking by electromigration. The current-voltage (I-V) characteristic was measured as shown in figure 1(d). The interspace distance at the nanogap was estimated by fitting the experimental results to the Fowler-Nordheim equation based on the tunneling current. We observed the exponential increase in current up to the bias voltage of 4.4 V.
Here, A is the counter area of the nanogap, e is the electron charge,  is the reduced Planck constant, f is a work function of gold, m is the mass of electron.
V d is an electric field with the field enhancement factor b, where V and d are the bias voltage and the nanogap, respectively [37][38][39][40]. ( ) = + t y y 1 0.1107 1.33 and ( ) n =y y 1 1.69 are compensation terms considering the mirror images of the electrode surfaces. Here, y is represented as where e 0 is the permittivity of vacuum. The nanogap d was used as a parameter for the fitting curve. In this study, we assume that the electrode tip at the nanogap is a semispherical shape. Some papers have reported that the enhancement factor is represented by the aspect ratio [38][39][40]. In this study, the enhancement factor is regarded as the aspect ratio of 1 and b = 1. The experimental I-V characteristic was consistent with the fitting curve. From the exponential increase and good agreement, we consider that the nanogap and the wiring structures held without melting and breaking against the current at the bias voltage up to 4.4 V. We obtained the gap d of 0.37 nm from the fitting result. We confirmed the nanogap that possessed tunneling current characteristics.

Measurements
We measured the time-series data of the electrical current (I-t) when applying the bias voltage. The polydimethylsiloxane (PDMS) frame was applied to the substrate as shown in figure 1(e). The DNA oligomer solution or ultrapure water was injected into the PDMS frame, and the cover glass was then placed onto the frame. The DNA oligomer concentration and sequence were 10 μM and 5′-AAAAAAAAAAAAAAAA-3′ (Fasmac Co., Ltd), respectively.
We used the He-Ne laser of 632.8 nm wavelength to excite the nanogap. It is because the plasmonic excitation at a nanogap by 632.8 nm laser was confirmed by the SERS experiments using a nanoparticulate system [5][6][7]. The nanogap was located at the laser spot and the focus point. The laser intensity was set to 0, 0.74×10 −3 , 0.11, 1.23, 4.36, 7.05, and 11.4 mW. The diameter of the laser spot was approximately 8.8 μm. The calculated power density of the laser was 18.9 kW cm −2 at the laser intensity of 1.23 mW. The SERS studies on DNA oligomers detection used 6-12 mW laser irradiation [41,42]. Therefore, the laser irradiation used in this study is thought to give no damage to DNA oligomers.
The bias voltage and the sampling rate were set to 0.5 V [1-3] and 1 kHz, respectively. We carried out 9 measurements for each laser intensity. The duration time of measurement was 60 s for each measurement.

Time-series current measurements
We obtained time-series data of measured current from ultrapure water and DNA solution varying a laser intensity, as shown in figure 2, with the measured current ranging from approximately 0 to 4 nA. A constant current was observed in both the cases of ultrapure water and DNA solution under no laser irradiation, as shown in figure 2(a), indicating that no molecule enters the nanogap. We obtained the electrical signal from the DNA solution at the laser intensity of over 1.23 mW, as shown in figures 2(b)-(d), although the ultrapure water exhibited the constant current despite the use of laser irradiation. The current transition indicates that a DNA oligomer has entered the nanogap.
We identified two kinds of typical patterns of the current transition. The first was a step-like pattern, as shown in figures 2(b) and (c). Upon measuring the discrete levels of current, we generated histograms of current acquired from DNA solution as shown on the left side of each graph in figure 2. The current interval in the histogram was set to 0.001 nA. Notably, there were some distinguishable peaks observed in the histogram as shown in figures 2(b) and (c). According to these results, we confirmed that the fabricated nanogap displays single-molecule sensitivity of DNA oligomers and it was realized by the laser irradiation of 1.23 mW or more. An increase of more than two steps from the lowest current peak level, namely more than two peaks, in the timeseries data of currents indicates the step-by-step collection of more than two DNA oligomers at the nanogap. The duration of the constant current suggests that DNA oligomers had stagnated at the nanogap for a long period of time. The second pattern that we observed was a spike-like pattern, as shown in figures 2(b) to (d). Figure 2(d) also shows the magnified graph for the spike-like signals in the inset. These findings suggest that the DNA oligomers passed through a nanogap without stagnation. Figure 3(A) shows the average of current with ultrapure water as a function of laser intensity, indicated by the blue lines in figure 2. The current with ultrapure water, shown as a baseline of time-series data of measured current in figure 2, increased with laser intensity. This is because the temperature of the solution at the nanogap was increased with laser intensity due to the light absorption effect by photothermal conversion of light irradiation, resulting in increased conductivity [28]. The time-series data of the current was integrated at 60 s for the DNA oligomer solution. The ultrapure water data was then subtracted from the DNA oligomer solution data as the background value. The calculation results as a function of laser intensity are shown in figure 3(A). Data points of (a) to (d) in figure 3(B) correspond to the graphs (a) to (d) in figure 2, respectively. In the region including (a) in figure 3(B), we confirmed that the integrated value is small, indicating that no DNA fragment could enter the nanogap. We observed two separate regions of integrated values at a laser intensity of more than 1.23 mW. In the region including (b) and (c) at a laser intensity of more than 1.23 mW, an increase in the current level was observed, indicating laser irradiation concentrates molecules to a nanogap. The region including (d), which shows smaller integrated value even at values higher than 1.23 mW, corresponds to the spike-like signals. However, certain data points of the integrated values remained low even at these higher values. The effect of the heat generated by the laser light is greater than the plasmonic optical force. We assume that the Brownian motion enhanced due to the increase in temperature during laser irradiation, and stagnation times are shorter in the nanogap.
The effect of high-power laser irradiation on the nanogap structure is discussed in this paragraph. After the 11.4 mW laser irradiation, we confirmed the same current levels during the repetitive measurement at the laser intensity from 0 to 11.4 mW. The dependency of the current level of the ultrapure water on the laser intensity as shown in figure 2(a) exhibited no change after 11.4 mW laser irradiation. From these results, the nanogap structure change due to the laser irradiation was thought to be negligible.

Event frequency
Next, we reported the event frequency at which the DNA oligomers enter or exit from the nanogap. The position and the angle at which the DNA oligomers enter the nanogap is not constant. For this reason, the current transition of transporting in the nanogap has a variation. We defined the change of current DI as We defined that a DNA oligomer enters and exits the nanogap when D  I A , t with A t representing the threshold current. Here, we focused on the histogram shown on the left side of the graph in figure 2. Figure 4(a) shows an example of a current histogram, and observed discrete peak currents. We acquired the interval of the current change depending on the laser intensity as shown in figure 4(b). The red dotted line in figure 4(b) indicates a linear fitting line by least squares method. The threshold current A t was defined by the fitting line. When the current change in a measured signal exceeds the threshold current, we determined that a DNA oligomer enters or exits from the nanogap. We calculated the number of events at each laser intensity. Similar to the integration results shown in figure 3(B), we confirmed that many data points exceeded the threshold when the laser intensity was 1.23 mW or more, as shown in figure 5.
The trapping and detection frequency is thought to increase with a concentration of DNA oligomers. The attraction force is a short-range action so that molecules close to the nanogap are attracted. The trapping and the detection are stochastic events. Therefore, the event frequency has some variation as shown in figures 3(b) and 5. The variation is thought to depend on a concentration of DNA oligomers. From the experimental result, we demonstrate that DNA oligomers were attracted to the nanogap by laser light irradiation. It can be suggested that the dielectrophoretic force generated by laser light irradiation promotes the travel of DNA oligomers in the direction of the nanogap. This phenomenon is significantly useful for sensing applications. The results suggest that DNA oligomers are spontaneously attracted and concentrated to the nanogap corresponding to the detection point, resulting in high detection probability and sensitivity.

Stagnation time
In this study, we also measured stagnation time via histogram analysis. The interval time between molecule penetration and exit is defined as the stagnation time, as shown in figure 6(a). The histograms shown in figure 6 have different range settings an interval of the stagnation time. We found the peak at about 22 ms of the stagnation time in figure 6(c). This suggests that the spike-like signals distribute from 0 to around 40 ms, and exhibit peak stagnation time at 22 ms. We considered the short stagnation time to detect DNA oligomers when transporting through the nanogap. According to the histogram shown in figure 6(d), the number of data points in the histogram decreased with increasing stagnation time.
Next, we discuss the stagnation time as a function of laser intensity. Figure 7 shows the histograms of the stagnation time ranging from 0 to 50 ms focusing on the short-time stagnation. The solid lines indicate the normal distribution calculated extracting the stagnation time of 0-40 ms for discussion of spike-like signals. The peak stagnation times were found at 22.5, 21.5, 22.1, and 21.2 ms at the laser intensity of 0.11, 1.23, 4.36, 7.05 mW, and 11.4 mW, respectively. The average stagnation time shows no difference at the laser intensity of 0.11or more. Figure 8 shows the number of events as a function of laser intensity. The range of the stagnation time varies in figure 8. The number of events for each intensity exhibits the same tendency shown in figure 5. At the laser intensity of 1.23 mW or more, the number of events has no large difference.    Notably, adenine molecules adsorb to the gold surface, and we hypothesized that DNA oligomers switch between adsorption and desorption on a gold surface at the nanogap [43]. Moreover, it has been suggested that the adsorption time on the nanogap surface can determine the stagnation time.

Conclusion
In this study, we investigated the effect of nanogap laser irradiation on DNA oligomer behavior using the electrical readout instead of the optical readout used in the conventional studies. We fabricated the gold nanoelectrode with the nanogap using an EB-lithography-based, lift-off technique and subsequent electromigration. The fabricated nanoelectrode was evaluated from the measured I-V curve. By fitting the experimental results by the Fowler-Nordheim equation, we confirmed that the nanogap is small enough to achieve the tunneling current property. The time-series data of the current was measured under the applied bias voltage of 0.5 V. We observed that the current increased by more than 1.23 mW laser intensity when the DNA solution was placed on the substrate. From these results, we confirmed that laser irradiation to the nanogap enables to attract the molecules to the nanogap, and to detect the DNA oligomer with single-molecule sensitivity. This ability can be attributed to the dielectrophoresis force generated from plasmonic excitation of the nanogap. The histogram generated from the time-series experiment measuring the current revealed that the stagnation time peaked at 22 ms. Moreover, the plasmonically-excited nanogap is an extremely useful tool to concentrate and detect individual biochemical molecules, such as a DNA oligomer.