A microchip fabrication process that incorporates plasmonic tweezers is presented here. The microchip enables the imaging of a trapped particle to measure maximal trapping forces.
Plasmonic tweezers use surface plasmon polaritons to confine polarizable nanoscale objects. Among the various designs of plasmonic tweezers, only a few can observe immobilized particles. Moreover, a limited number of studies have experimentally measured the exertable forces on the particles. The designs can be classified as the protruding nanodisk type or the suppressed nanohole type. For the latter, microscopic observation is extremely challenging. In this paper, a new plasmonic tweezer system is introduced to monitor particles, both in directions parallel and orthogonal to the symmetric axis of a plasmonic nanohole structure. This feature enables us to observe the movement of each particle near the rim of the nanohole. Furthermore, we can quantitatively estimate the maximal trapping forces using a new fluidic channel.
The ability to manipulate microscale objects is an indispensable feature for many micro/nano experiments. Direct contact manipulations can damage the manipulated objects. Releasing the previously held objects is also challenging because of stiction problems. To overcome these issues, several indirect methods using fluidic1, electric2, magnetic3, or photonic forces4,5,6,7,8 have been proposed. Plasmonic tweezers that use photonic forces are based on the physics of extraordinary field enhancement several orders larger than the incident intensity9. This extremely strong field enhancement enables the trapping of extremely small nanoparticles. For example, it has been shown to immobilize and manipulate nanoscale objects, such as polystyrene particles7,10,11,12,13,14, polymer chains15, proteins16, quantum dots17, and DNA molecules8,18. Without plasmonic tweezers, it is difficult to trap nanoparticles because they quickly disappear before they are effectively examined or because they are damaged due to the high intensity of the laser.
Many plasmonic studies have used various nanoscale gold structures. We can categorize the gold structures as protruding nanodisk types12,13,14,15,19,20,21 or suppressed nanohole types7,8,10,11,22,23. In terms of imaging convenience, the nanodisk types are more suitable than the nanohole types because, for the latter, the gold substrates can obstruct the observation view. Moreover, the plasmonic trapping occurs near the plasmonic structure and makes observation even more challenging. To the best of our knowledge, plasmonic trapping on nanohole types was only verified using indirect scattering signals. However, no successful direct observations, such as microscopic images, have been reported. Few studies have described the position of trapped particles. One such result was presented by Wang et al. They created a gold pillar on a gold substrate and observed the particle motion using a fluorescent microscope24. However, this is only effective for monitoring lateral movements not in the direction parallel to the beam axis.
In this paper, we introduce new fluidic microchip design and fabrication procedures. Using this chip, we demonstrate the monitoring of plasmonically trapped particles, both in directions parallel and orthogonal to the plasmonic nanostructure. Furthermore, we measure the maximal force of the immobilized particle by increasing the fluid velocity to find the tipping velocity in the microchip. This study is unique because most studies on plasmonic tweezers cannot quantitatively show the maximal trapping forces used in their experimental setups.
Caution: Please refer to all relevant material safety regulations before use. Several of the chemicals used in microchip fabrication are acutely toxic and carcinogenic. Please use all appropriate safety practices when performing the photolithography and etching processes, including the use of engineering controls (fume hood, hot plate, and aligner) and personal protective equipment (safety glasses, gloves, lab coat, full-length pants, and closed-toe shoes).
1. Fabrication of the PDMS Microchannel
2. Etching Process of the Gold Plate
3. Assembly of the Microchip
4. Improvement of the Microchip Side Surface Roughness by PDMS Coating
NOTE: The gold plate with fixed dimensions of 400 x 150 µm2 is relatively more difficult to cut out than the PDMS material. Therefore, to detach the PDMS microchannel from the wafer, a razor blade is used to cut out a larger piece than the gold plate. After combining the two parts, the excess parts of the PDMS relative to the gold plate must then be cut so that the inside of the channel can be observed from the side using a microscope (Figure 4a). However, the cut surface, which is used as a window, has a high surface roughness and consequently produces cloudy images of the particles that flow in the channel (Figure 4b). Coating with the PDMS solution is performed again to resolve this problem.
5. Laser Coupling to Insert the SMF Cable to the Microchip
NOTE: For the plasmonic tweezer system, an optical fiber incident laser with a 1,064-nm wavelength is used. The SMF cable is used because the diameter of the incident laser (5 mm) is too immense to emit the laser beam at the nanohole milled on the gold block (400 x 150 µm2) in the microchip. The cladding diameter of the SMF cable is 125 µm. Thus, the incident laser and SMF cable must be coupled.
6. Plasmonic Trapping of Single Fluorescent Polystyrene Particle in the Microchip
The fabrication process of the PDMS microchannel and nanohole gold plate is shown in Figures 1 and 2. The method to combine the two parts and the actual microchip is shown in Figure 3. The PDMS was cut to reveal the inside of the channel from the side of the microchip. However, it was difficult to observe the particles flowing in the channel because of the surface roughness of the cutting plane. Therefore, we introduced the PDMS coating method to solve this problem, as shown in Figure 4.
We observed 5-µm, flowing polystyrene particles in the microchip to confirm the effect of the PDMS coating. Figure 5 shows the actual fabricated microchip and particles observed in the microchip using the microscope. Figure 5a and c are the before and after appearances of the microchip. Figure 5b and d are the magnified surfaces of each. Figure 5e shows blurred particles flowing, whereas Figure 5f shows that the edges of the particles are notably clear and that movements can be monitored. As above, the PDMS coating of the microchip surface is essential to the monitoring of trapped particles.
Figure 6 shows the 100-nm polystyrene particle undergoing plasmonic optical trapping by the plasmonic tweezer system. An SMF cable with a 0.14 numerical aperture (NA) was used. A tube was inserted at the inlet/outlet holes of the microchip channel. A micropump was used to insert and collect the 100-nm fluorescent polystyrene particle solution. To emphasize the interior appearance of the trapped particle by the plasmonic phenomenon, the dotted parts of Figure 6a have been enlarged as an inset, Figure 6b. The laser delivered by the SMF cable inserted in the microchip was emitted to the nanohole of the gold block, and the particle flowed from left to right. Movement was detected at the nanohole rim, where the field enhancement was the strongest.
Figure 7 shows consecutive images where a 100-nm fluorescent polystyrene particle that flowed in the microchannel was trapped and released at the nanohole at the intensity of 0.42 mW/µm2. The particles flowed at a constant speed of 3.4 µm/s in the fluid direction, as shown in Figure 7a. After the laser was turned on, one of the particles was trapped at the nanohole, as shown in Figure 7b. On the contrary, another particle flowed into the stream, as shown in Figure 7c. Then, the flow speed was increased until the trapped particle escaped. Figure 7d shows the particle escaping from the trap. At this moment, we can estimate the trapping force with direct observation by measuring the fluid velocity when the particle escaped. We also worked in the opposite direction. Instead of increasing the fluid velocity, we gradually decreased the laser power in decrements of 1 mW and recorded the intensity when the particle escaped. This laser intensity is defined as the minimum trapping laser intensity and was measured to be 0.24 mW/µm2.
Figure 1. Fabrication of the PDMS microchannel. (a) Preparation of the Si wafer. (b) Photoresist spin coating of the wafer. (c) Fabricated microchannel mold by the photolithography process. (d) PDMS solidification using an oven after pouring the PDMS solution on the wafer. (e) PDMS microchannel cutting. (f) PDMS microchannel detachment from the wafer. (g) Inlet/outlet and SMF cable holes punctured on the PDMS microchannel. (h) Actual solidified PDMS on the wafer. (i) Actual detached PDMS microchannel. Please click here to view a larger version of this figure.
Figure 2. Fabrication of the nanohole on the gold plate after the etching process. (a) Deposition of Au and Ti on the glass. (b) Photoresist spin coating of the gold plate. (c) Dissolved photoresist removal after UV light exposure. (d) Au etching. (e) Ti etching. (f) Remaining photoresist removal. (g) Nanohole milling by a focused ion beam on the gold block. (h) Actual fabricated gold block. (i) Actual milled nanohole on the gold block. Please click here to view a larger version of this figure.
Figure 3. Assembly process of the microchip. (a) Fix the PDMS microchannel and gold plate on the mask holder and substrate stage, respectively, equipped on the aligner. (b) Combination of the PDMS microchannel part and the gold plate after surface treatment with O2 plasma. (c, d) Assembled microchip after combination. (e) Removal of excess amount of the PDMS microchannel. Please click here to view a larger version of this figure.
Figure 4. Process of surface roughness improvement by PDMS coating. (a) Remove excess amount using a razor blade after combining the two parts. (b) High surface roughness of the microchip after cutting. (c) PDMS solution spin coating in a Petri dish. (d) Dipping the window surface of the microchip into the spin-coated PDMS solution. (e) PDMS-coated microchip detachment from the Petri dish. (f) Improvement of surface roughness by PDMS coating. Please click here to view a larger version of this figure.
Figure 5. Assembled microchip and observation of 5-µm polystyrene particles in the microchannel before and after PDMS coating. (a, b) Microchip before PDMS coating and magnified view. (c, d) Microchip after PDMS coating and magnified view. (e, f) Observation of particles in the microchannel before and after PDMS coating. Please click here to view a larger version of this figure.
Figure 6. Designed plasmonic tweezer system. (a) Schematic of the plasmonic tweezer system. (b) Trapping of a 100-nm fluorescent polystyrene particle at the rim of the nanohole in the microchip. Please click here to view a larger version of this figure.
Figure 7. Trapping and releasing of a 100-nm fluorescent polystyrene particle in the microchannel. (a) The microchannel with a particle flowing into the stream. (b, c) Trapped particle at the nanohole compared to another particle. (d) Particle that escaped from the trap due to the increased fluid force. Please click here to view a larger version of this figure.
The SMF cable was inserted in the SMF cable hole on the microchip, as shown in the rectangular dot of Figure 6a. Because the SMF cable hole is larger than the cable diameter, epoxy glue was used to seal the gap to block the leakage of the flowing particle solution. Before the application of epoxy glue, the gold block and cable edge should be coaxially aligned by hand using a microscope. Although it is ideal for the inserted cable edge and the nanohole to be coaxially aligned, a slight misalignment can be tolerated because the laser beam diverges once it is emitted from the end of the 0.14 NA SMF cable edge, and the beam affects a much larger region. Because the microchip was configured to be perpendicular to the optical axis of the microscope, we could not directly observe the location of the nanohole. The location of the nanohole can only be indirectly determined by observing the location of the plasmonically trapped particle at the nanohole. A solution can be provided by installing a camera at the fiber cable and using it to monitor the gold block.
The distinctive feature of the microchip is its ability to monitor particle motion near the plasmonic nanohole in real time. The motion of the particle follows the scenario described below. When the fluid streams the particles forward, some particles move toward the gold block. In some cases, a particle gets notably close to the rim of the nanohole due to attraction to the nanohole and eventually becomes immobilized. At this moment, the optical force exerted on the particle exceeds the fluid force. Subsequently, the immobilized particle escapes from the nanohole rim when the fluid velocity increases; thus, the fluid force becomes stronger than the optical force. The maximal trapping force can be measured from this terminal fluid velocity. However, the conventional drag force equation cannot be used because the particle is in physical contact with the gold wall at the nanohole. To consider the surface effect of the gold wall, we used the finite-element method, which considers the fluid motion near the surface, and obtained the fluid force.
We have introduced a new plasmonic tweezer setup that enables the monitoring of particle dynamics along the laser beam axis. In contrast, previous studies have only introduced particle movement in the plane perpendicular to the laser beam axis, such as with the nanoblock12, nanodisk13,14,19,21, nanostick20, and nanopyramid18. Furthermore, in the case of nanohole types, trapping can only be witnessed by monitoring the scattering signal, and not by visual monitoring10,11,23. However, we could not precisely measure the particle position because of the limited capabilities of current imaging techniques. The imaging quality should be further improved to confirm the exact dislocation measurements. This technique can be applied in the characterization and biosensing of a single molecule.
The authors have nothing to disclose.
This work was supported by the ICT R&D program of MSIP/IITP (R0190-15-2040, Development of a contents configuration management system and a simulator for 3D printing using smart materials).
Negative photoresist | MicroChem | SU-8 2075 |
Developer | MicroChem | SU-8 Developer |
Positive photoresist | Merck Ltd. | AZ GXR-601 |
AZ Photoresist Developers | Merck Ltd. | AZ 300 MIF |
HMDS | Merck Ltd. | AZ Adhesion Promoter |
Aligner | Midas System | MDA 400M |
Atmospheric plasma machine | Atmospheric Process Plasma Co. |
IDP-1000 |
Polydimethylsiloxane (PDMS) | Dow Corning | Sylgard 184 A/B |
Gold coated test slides | EMF Co. | TA124(Ti/Au) |
Au etchant | Transene Inc. | TFA |
Ti etchant | Transene Inc. | TFT |
40X objective lens | Edmund Optics | 40X DIN |
60X water immersion objective lens |
Olympus | LUMPLFLN 60XW |
Optical fiber incident laser | IPG Photonic | YLR 10 |
SMF coupler | Thorlabs | MBT612D/M |
Syringe micropump | Harvard | PC2 70-4501 |
Fluorescent microscope | Olympus | IX-51 |
Plasma system | Femto Science Inc | CUTE-MPR |