A single large assembly with dynamically fluctuating swarms of gold

Subscriber access provided by - Access paid by the | UCSB Libraries

Communication

A single large assembly with dynamically fluctuating swarms of gold nanoparticles formed by trapping laser Tetsuhiro Kudo, Shang-Jan Yang, and Hiroshi Masuhara Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.8b02519 • Publication Date (Web): 02 Aug 2018 Downloaded from http://pubs.acs.org on August 3, 2018

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 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 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.

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 14 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

Nano Letters

A single large assembly with dynamically fluctuating swarms of gold nanoparticles formed by trapping laser Tetsuhiro Kudo1,* Shang-Jan Yang1, Hiroshi Masuhara1,2* 1. Department of Applied Chemistry, College of Science, National Chiao Tung University, Hsinchu 30010, Taiwan, 2. Center for Emergent Functional Matter Science, National Chiao Tung University, Hsinchu 30010, Taiwan. Table of Contents

ABSTRACT:

Laser trapping has been utilized as tweezers to three-dimensionally trap nanoscale objects and provided significant impacts in nanoscience and nanotechnology. The objects are immobilized at the position where the tightly focused laser beam is irradiated. Here, we report swarming of gold

ACS Paragon Plus Environment

1

Nano Letters 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 2 of 14

nanoparticles where component nanoparticles are dynamically interacting with each other keeping their long interparticle distance around the trapping laser focus at a glass/solution interface. Two swarms are directionally extended outside the focal spot perpendicular to the linear polarization like a radiation pattern of dipole scattering, while a doughnut shaped swarm is prepared by circularly polarized trapping laser. The light field is expanded as scattered light through trapped nanoparticles, this modified light field further traps the nanoparticles, and scattering and trapping cooperatively develop. Due to these nonlinear dynamic processes, the dynamically fluctuating swarms are evolved up to few tens mirometer. This finding will open the way to create various swarms of nanoscale objects which are interacted and bound through the scattered light depending on the properties of laser beam and nanomaterials.

KEYWORDS: laser trapping, gold nanoparticles, interface, optical binding, light scattering, swarming

TEXT: Laser trapping has been used as tweezers1 to spatiotemporally trap and manipulate nanoscale objects (such as dielectric nanoparticles, metallic nanoparticles, semiconductor quantum dots, proteins, molecular clusters, etc.) in various research fields from fundamental studies to applications2-4. To manipulate the multiple nanoscale objects simultaneously in different way, a spatial light modulator is applied to create multiple trapping sites by controlling the wave front of light5. Also patterning of particles is achieved by spatially-imaged laser patterns6 and repetitive scanning of a focused trapping laser beam with Galvano mirrors7. Evanescent wave generated by total internal reflection with a prism has a steep intensity gradient, which is favorable for the assembling of nanoscale objects at the interface8-10. Laser trapping with locally enhanced electric fields around plasmonic nanostructures11-16 and photonic-crystal waveguides or slabs17-19, have recently drawn much attention with the progress of nanofabrication techniques. In addition, nanoscale objects can be rotated with circularly polarized light and vortex laser beam which carry spin20 and orbital angular momentum21, respectively. In comparison to a continuous wave laser trapping, ultrashort laser pulses have been used for efficient trapping of semiconductor quantum dots22, 23 and dielectric nanoparticles24-26, the latter being accompanied with unique ejection phenomenon depending on the direction of linearly polarized trapping


2

Page 3 of 14 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

Nano Letters

laser25. Thus, designing the light fields is one of the most significant and promising developments in laser trapping studies. It is notable to point out that most of these experiments are focused on how to trap the nanoscale objects by tuning intensity, its spatial gradient, group velocity, angular momentum, wave front, polarization, pulse, time duration, etc. Assembling of various nanoscale objects are also achieved by these trapping lasers. A periodic colloidal structure showing Bragg reflection at certain wavelength of light can be formed for dielectric nanoparticles at the interface27-32. Recently, we found that trapping laser propagates through the periodic colloidal structure prepared by laser trapping, and then the straightly aligned nanoparticles like a horn are stuck out from the assembly along the direction of light propagation, working as a waveguide32. Assembling of metallic and semiconductor nanoparticles are studied with optical spectroscopy upon laser trapping. These assemblies show the energy shift of electronic transition from that of individual nanoparticles33-35. When multiple metallic nanoparticles diffusing in solution are trapped at the irradiated area, the nanoparticles are optically bound and arranged in ordered structure depending on the laser polarization, wavelength, etc.36-39. Apart from nanoparticles, nucleation of molecular systems (amino-acid) and following crystal growth are possible even from unsaturated solution by laser trapping at a solution interface40-42. By controlling the laser polarization, the resultant crystal morphology is biased to certain forms41, 42. Protein growth rate of lysozyme can be controlled by laser trapping43, 44 , and amyloid fibril formation of cytochrome C can be spatiotemporally induced by laser trapping45. In general, the nanoparticles and molecules are trapped, concentrated at the focus, and distributed outside of the focus, giving their clusters, assemblies and even crystals. There the molecules and nanoparticles are stably aligned without large fluctuation, being kept as long as the laser irradiation is continued. Here, we report that a single large assembly with dynamically fluctuating swarms of gold nanoparticles is prepared by laser trapping through light scattering. Initially, several gold nanoparticles are trapped at the focal spot, and scattered light from these nanoparticles modifies the light field which further gathers the nanoparticles around the focus. As a result of these nonlinear dynamic processes, the dynamically fluctuating swarms are evolved up to a few tens micrometer under the laser trapping, where their gold nanoparticles are interacted and bound through the scattered light and its interference. This is the first experimental observation of a dynamic swarm of gold nanoparticles where optical potential is dynamically and nonlinearly evolved over time.


3

Nano Letters 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 14

Figure 1. Optical set up for the laser trapping and assembling experiments of gold nanoparticles at an upper-glass/solution interface. A dashed-lined inset illustrates a sample chamber sandwiched by two glass substrates. The trapping laser is focused at the upper-glass/solution interface of the chamber. NA and CCD are abbreviations for numerical aperture and charge-coupled device, respectively.

The setup for laser trapping system is constructed on an inverted microscope (see Fig. 1), which is similar to that reported previously by our group32. 1064 nm continuous wave laser is introduced as trapping laser and tightly focused at the upper-glass/solution interface by an objective lens (Olympus, 0.95 numerical aperture, 40× magnification). The laser power after the objective lens is 60 mW, and the beam diameter before the objective lens is about 9 mm. The present laser power of 60 mW is about few tens times higher than the minimum laser power that is required to trap a gold nanoparticle at the interface. A half-wave plate is used for turning the direction of linear polarization and a quarter-wave plate is used for generating the circular polarization. Halogen lamp is used as a light source for illuminating the sample through the darkfield condenser lens, and scattering light is recorded by charge-coupled device (CCD) camera (30 frames per second). A short-pass filter (Semrock: FF01-1010/SP-25) is inserted in order to remove the backscattering light of the trapping laser from the images. The diameter of gold nanoparticles we mainly used is 200 nm (BBI Solutions). The extinction spectrum of this nanoparticle is provided in Fig. S1 of Supporting Information. In addition for the comparison, we also use the nanoparticles whose diameter is 150 nm (BBI Solutions), 250 nm (BBI Solutions) and 300 nm (Aldrich). After sonication, the sample solution is sandwiched by glass substrates with a spacer (Electron Microscopy Sciences), and the solution thickness is about 100 µm. We demonstrate the laser trapping (1064 nm in wavelength) of gold nanoparticles (200 nm in diameter) with a single tightly focused laser beam at an upper-glass/solution interface. Initially, the gold nanoparticles are randomly moving near the bottom-glass/solution interface because they are heavier than water. Upon laser irradiation, these nanoparticles are pushed upward and gathered at the upper-glass/solution interface. When the laser is focused inside the solution, the nanoparticles are also accelerated upward by the scattering force while they cannot be trapped at the focus. It is reported that, nanoparticles of 60 nm diffusing below the focal spot are transferred upward upon irradiation of trapping laser46. When they quickly shifted down the observation


4

Page 5 of 14 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

Nano Letters

plane after the laser trapping, a dark circular area around the laser axis were found. The diameter of this area becomes wider when the observation plane is deeper from focus, which indicates this area in each plane is the cross-section of circular-cone shape reflecting the laser beam shape. Thus, we consider that the nanoparticles are transferred from the bottom-glass/solution interface to the upper-glass/solution interface during laser trapping in our studies.

Figure 2. Scattering images of gold nanoparticle assembly formed by linearly polarized laser trapping at the upper-glass/solution interface. (a, b) Single nanoparticle-level images and ensemble images of the gold nanoparticle assembly, respectively. Vertical white dotted lines in (b)-(iii-v) represent the central line where the focal spot is crossed. (c) The scattering images of the assembly of gold nanoparticles whose diameter are 150, 250, 300 nm. The laser is irradiated for 30 min, and arrows denote the direction of trapping laser polarization. The length of the white bars is 1 and 10 µm for (a) single nanoparticle-level and (b) ensemble imaging, respectively. (d) The evolving assembly formation dynamics of gold nanoparticles is illustrated. The circle lines represent the irradiated area by trapping laser. The light-red color represents the light scattering intensity distribution.


5

Nano Letters 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 6 of 14

We start from the results of laser trapping with linearly polarized light as represented in Fig. 2. To observe the early stage of the assembling behavior where the nanoparticles are trapped one by one, a relatively low particle concentration (7×103 particles/µl) is chosen in Fig. 2a, while higher particle concentration (1.4×105 particles/µl) is used to observe the whole assembling behavior in Fig. 2b. We term the former experiment as a single nanoparticle-level imaging and the latter as an ensemble imaging. As shown in Fig. 2a-(i to vi), the nanoparticles are trapped one by one, and the assembly tends to align perpendicualrly to linear polarization with certain interparticle distance. The nanoparticle alignment is sometimes deformed (for example Fig. 2a(v)), but the nanoparticles are again aligned perpendicularly to linear polarization up to about five nanoparticles (Fig. 2a-(vi)). Its center-to-center interparticle distance is approximately 730 nm, which is close to the wavelengh of trapping laser in water (~ 800 nm). Hence, the nanoparticles are not in contact with each other and are separately observed under the optical microscope. The interparticle distance rarely becomes about 1.6 µm which corresponds to twice the wavelength of the trapping laser in water (see Fig. S2 in Supporting Information). We consider that this particle separation is attributed to the optical binding caused by interference of scattered light between the nanoparticles. In fact, it was reported that, the metallic nanoparticles are aligned and optically bound perpendicular to linear polarization with the interval corresonding to the wavelength of trapping laser in medium36-38. Besides, the number of the nanoparticles appearing in the images is sometimes not changed apparently, although we observe that the nanoparticles are trapped from the outside of the assembly. This implies that these nanoparticles are trapped and also bound vertically to the image plane. After about five nanoparticles were perpendicularly aligned, the focal spot is occupied by nanoparticles and the assembly is prepared laterally as shown in Fig. 2a-(vii to xv). Incoming nanoparticles are sequentially trapped next to the nanoparticles that are already perpendicularly aligned and they tend to orient parallel to the linear polarization. The assembling is dynamic, stochastic and its nanoparticles are rearranging frame by frame. For instance, Fig. 2a-(xiii) shows four parallel lines and its interparticle distance in the lines seems to be shorter than that in perpendicular direction to linear polarization. Apart from the perpendicular alignment relative to linear polarization based on light scattering, it was reported that, the metallic nanoparticles are aligned parallel along the linear polarization through near field interaction accompanying the change in scattering spectra of the nanoparticle assembly33. Thus we consider that, near-field interaction optically binds the nanoparticles in parallel direction. It should be emphasized that gold nanoparticles are first aligned perpendicularly due to the dipole scattering and then the additional nanoparticles are aligned parallel to linear polarization due to near-field scattering. Meanwhile it appears that the nanoparticles are ejected out of the focal spot perpendicularly (Fig. 2a-(xiii to xv)) but still being attracted toward the center of the assembly. Interestingly, the distance from the end of this ejected nanoparticle to the focal center is longer (e.g. Fig. 2a-(xv)) compared to the center-to-edge distance of five nanoparticles aligned perpendicularly to linear polarized light (Fig. 2a-(vi)). When the trapping laser is scattered by a central periodic structure (such as several parallel lines aligned at intervals of wavelength), the scattered light creates the constructive interference outside the periodic structure to strongly and directionally scatter the light perpendicular to linear polarization like a Yagi-Uda antenna47. In other words, it is regarded that irradiation of 1064 nm trapping laser prepares the antenna for scattering itself. Thus, we consider that the periodic structure formed by laser trapping enables to strongly scatter the light


6

Page 7 of 14 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

Nano Letters

perpendicular to linear polarization compared to the scattered light by perpendicularly aligned five nanoparticles. It should be emphasized that the nanoparticles constituting the assembly are randomly moving and colliding with each other. Once the central assembly structure becomes asymmetric (Fig. 2d(vi to vii)), the trapping laser is scattered more to the one direction compared to opposite direction. We consider that this light scattering is similar to electromagentic wave scatttering working in Yagi-Uda antenna. Then, the assembly is extended along this direction, and this extended assembly further scatters the light to further expand itself outside the focal spot. When the central structure accidentally takes opposite symmetry, the whole assembly and direction of the light scattering are swiched to opposite side, and vice versa (such as Fig. 2a-(xiii to xv) and Fig. 2d-(viii to x)). Namely, the assembly is alternatively extended outside the focal spot through Yagi-Uda-like scattering.

Figure 3. (a-c) Time evolution of scattering intensity at left (red line) and right (blue line) swarms of the gold nanoparticle assembly. Definition of the left and right areas for respective swarms of the assembly is illustrated in the inset, and the total scattering intensity of left or right area is plotted against irradiation time. (b, c) Their enlarged view at 0 to 2 min and 15 to 17 min, respectively. The black line is the average of the red and blue lines.

Here, we shift our attention to the ensemble imaging of higher particle concentration. Similar to the results of Fig. 2a, a few nanoparticles alignment and the alternative extension are observed as shown in Fig. 2b-(i) and Fig. 2b-(iii to v), respectively. The behavior of alternative extension is represented in Fig. 3b showing the time dependence of scattering intensity at left and right areas of the assembly (see the definition of area in inset of Fig. 3a). The red and blue lines are correlated with each other, so that the scattering intensity at one area is increased and that at


7

Nano Letters 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 8 of 14

another area is decreased. This also suggests that some nanoparticles are moving from one side to the another, and vice versa. As the particle number increases, the length of the assembly extension will become longer until the laser intensity at the edge of the assembly becomes weaker than the intensity required for overcoming thermal fluctuations. Approximately after 1 min, the nanoparticles are trapped at the both directions rather than further extending in one direction (Fig. 2b-(vi) and Fig. 2d-(x to xi)). The assembly extension is gradually observed in both sides accompanying its size increase as shown in Fig. 2b-(vi) to 2b-(viii) and Movie S3. The temporal change in assembly size is shown in Fig. S3 of Supporting Information. The extension length from the focus to the outside edge is about 10 µm which is indeed larger compared to the focal spot. We estimated that about 1000 nanoparticles are included in one swarm by counting the nanoparticles one by one in first one minute and calculating the increase rate of particle number (for detail see Supporting Information). The left and right areas of the assembly are larger and brighter compared with the central area, which indicates that the number of distributed nanoparticles at the side areas are larger than that the central area. Also the nanoparticles in those side areas are dynamically fluctuating (see Movie S3), as well as total scattering intensities at left and right areas are still fluctuating even at 30 min as illustrated in Fig. 3a and 3c. Oscillatory behavior is observed and some amount of intensity is still fluctuating at both areas. From this result, we suggest a certain amount of nanoparticles are still moving to one area to another area over time. It appears that this assembly consists of two swarms of dynamically fluctuating gold nanoparticles. They are connected with each other giving a dumbell-like morphology, while its central part is bright and the surrounding is relatively not shining. This whole and large assembly formed by tightly focused laser beam is suprisingly turned by rotating the linear polarization with half waveplate (Fig. 2b-(viii to x), and Movie S4). We consider that, these two swarms are prepared by directional scattering perpendicular to laser polarization which can be well interpreted in terms of dipole scattering. Namely, an induced dipole oscillated along the linear polarization radiates perpendicular to the dipole moment, and the assembly morphology becomes like the far field radiation pattern of linear dipole. When this radiation pattern is rotated with the half waveplate, the dumbell-like assembly also follows the rotation of linear polarization as we observed in Fig. 2b-(viii to x). The light field of trapping laser is modified by trapped particles, and the modified light field further traps the nanoparticles to further scatter the light for trapping the next particles. Namely, the light field of trapping laser and gold nanoparticle assembly are nonlinearly evolved with time. With increase in the number of the trapped particles at the swarm, the light is multiply scattered by these nanoparticles to further extend the swarm until the multiple light scattering intensity is not enough to further bind the nanoparticles. It is notable that, interference of the scattered light among this assembly holds its dynamic shape by optical binding even outside the focal spot. When the trapping laser is switched off, most of the nanoparticles undergo Brownian motion, the assembly is dispersed to the surrounding area, and the several nanoparticles are sometimes stuck to the glass substrate. When the trapping laser is switched on again, the dispersed nanoparticles are trapped again forming the assembly with two swarms. Since scattering intensity should depend on the nanoparticle size, here we examine the effect of different nanoparticle sizes. The assembly with dynamically fluctuating two swarms of gold nanoparticle is also observed with larger size particles (250 nm and 300 nm) while not with


8

Page 9 of 14 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

Nano Letters

smaller size particle (150 nm) as shown in Fig. 2c. The size of the assembly with two swarms at 30 min becomes larger with increase in nanoparticle size (time evolution of assembly growth is shown in Figs. S5 and S6 of Supporting Information). We consider that these results on size dependence strongly supports our explanation that the assembly with two swarms is formed by scattering and interference of 1064 nm trapping laser. The dipole mode of surface plasmonic resonance for relatively larger sizes of gold nanoparticles are broader and shifted to near infrared region, and light scattering intensity is, of course, proportional to the volume of nanoparticles. Thus, the trapping laser of 1064 nm is much scattered by larger nanoparticles, leading to the single larger assembly with two swarms which reflects the radiation pattern of dipole scattering.

Figure 4. Scattering images of gold nanoparticle assembly formed by circularly polarized laser trapping at the upper-glass/solution interface. (a, b) Single nanoparticle-level scattering images and ensemble scattering images of the gold nanoparticle assembly, respectively. The direction of circularly polarized light is clockwise. The length of the white bars is 1 and 10 µm for (a) single nanoparticle-level and (b) ensemble imaging, respectively. (c) The evolving assembly formation dynamics of gold nanoparticles is illustrated. The circle lines represent the irradiated area by trapping laser. The light-red color represents the light scattering intensity distribution.

It is interesting to demonstrate that the assembly structure and the swarm shape can be controlled by laser polarization. The single nanoparticle-level imaging and the ensemble imaging are given in Fig.4a and Fig. 4b. In the early stage, the nanoparticles are optically bound by light scattering


9

Nano Letters 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 10 of 14

forming a hexagonal structure (Fig. 4a) and its interparticle distance is about 830 nm. It should be emphasized that this structure is created by just focusing the tightly focused laser beam to glass/solution interface instead of using typical nanofabrication technique such as electron beam lithography. Compared to former linear polarization, light is scattered to all direction for circularly polarized light giving a round assembly. Formed structure and the interference pattern of light scattering become different from that of linear polarization, therefore the interparticle distance is different from that of linear one. Besides, the hexagonal structure is rotated in the same direction with circularly polarized light. The assembly gradually becomes dynamic with increase of number of the trapped particles and is evolved, giving a doughnut shaped swarm of dynamically fluctuating nanoparticles as explained in following ensamble imaging. The assembly with doughnut shaped swarm is expanded up to about 13 µm in diameter outside of the focal spot (Fig. 4b-(v), Fig. S3, and Moive S5). In this case, the nanoparticles at the focal volume scatter the light isotropically in all direction. The surrounding gold nanoparticles are gathered along the upper-glass/solution interface and the doughnut shaped swarm is formed. The radius of the swarm formed by circular polarization is shorter than the distance from the assembly center to the edge of the swarm formed by linear one. Namely, the circularly polarized trapping laser gives the shrinked swarm compared to linearly polarized one, because directionally scattered linearly polarized light is isotropically scattered to all directions. Laser heating, thermophoretic force and thermally induced fluid convection may be involved with the formation of dynamically fluctuating gold nanoparticle swarms. Firstly, water may be directly elevated through the overtone absorption of water molecules. However, the temperature elevation of water with 1064 nm laser of the present 60 mW is only estimated to be about 1 degree according to the experimental report48. Secondly, the temperature of the gold nanoparticle will be elevated by absorption of the laser, and the heat is released to the surrounding solvent. In fact, bubbles are generated from the focal spot when the laser power is doubled in the present experiment. This implies that temperature is elevated upto 50 degree which is half value of the boiling point of water. The size of the thermal distribution should be larger than the focal spot size. For example, laser trapping of poly(N-isopropylacrylamide) with tightly focused laser beam provides the several tens micrometer assembly which is much larger than the focal spot size49. They discussed that this is due to the photothermal effect which thermal distribution is extended larger than the focal spot size. Furthermore, there is the experimental report that demonstrated thermophoresis trapping of DNA with laser heating50. In the present work, the nanoparticles recieve the thermal force in all direction outside of the focal spot equivalently following the spatial distribution of the temperature. However the optical force is stronger at focal spot compared to thermal force, the parallel lines and the hexagonal structure are prepared at the assembly center for each polarization. At the outside of the focal spot, the nanoparticles swarm vigorously due to heating and their Brownian motion. However, we believe that the light scattering is the origin of the swarm formation because they are dependent on laser polarization as we shown in above results. We consider that the whole assembly with swarms is realized by the balance between the optical force and the laser heating-induced diffusion. In summary, we have reported large swarms of dynamically fluctuating gold nanoparticles that are nonlinearly evolved at the upper-glass/solution interface under laser trapping. At the initial stage of the assembling experiments, Yagi-Uda like antenna structure for scattering the trapping laser and the hexagonal packing structure aligned at intervals of wavelength are prepared with linearly and circularly polarized lights, respectively. After long-term irradiation, two swarms are


10

Page 11 of 14 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

Nano Letters

directionally extended outside the focus perpedicularly to the linearly polarized light like the radiation pattern of dipole scattering, while a doughnut shaped swarm is formed by circularly polarized light with isotropic scattering in all direction. Scattering light at 1064 nm increases with nanoparticle size, and indeed these patterns are observed for nanoparticles larger than 200 nm in diameter. The trapping laser is scattered by the trapped gold nanoparticles, which further traps and binds the gold nanoparticles. These gold nanoparticles are strongly bound with each other through the light scattering and its intensity is large enough to bind nanoparticles even outside the focal spot. This strong light scattering triggers the formation of the assembly with swarms of dynamically fluctuating gold nanoparticles. The present interpretation on the single large assembly formation can be simulated and now its study is starting from two directions. Single nanoparticle level dynamics is based on self consistent electromagnetic theoretical approach38, while ensemble system is examined from the viewpoint of collective optofluidic dynamics51. Further experimentally various kinds of dynamic swarm could be created with controlling the laser beam (wavelength, polarization, spin and orbital angular momentum, phase, etc.) and nanoparticles (materials, shape, size, electronic resonance, etc.). For example, we may control the dynamic swarm by tuning the wavelength which can change the contribution of the light scattering and thermal heating. The present work based on light scattering will provide a new outlook for laser controllable dynamic swarm whose various nanoparticles are interacted with each other and bound by light scattering. ASSOCIATED CONTENT Supporting Information. Supporting Information, and Movies S1 to S5.

AUTHOR INFORMATION Corresponding Authors *E-mail: (T.K.) [email protected], (H.M.) [email protected]

ACKNOWLEDGMENT The present work is partly supported by the MOE-SPROUT Project (National Chiao Tung University) of the Ministry of Education, Taiwan, to H.M., and by grants from the Ministry of Science and Technology of Taiwan to H.M. (MOST 105-2811-M-009-022) and to T. K. (MOST 106-2113-M-009 -027 -MY2). T.K. deeply acknowledges support from the Japan Society for the Promotion of Science Postdoctoral Fellowships for Research Abroad.


11

Nano Letters 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 12 of 14

REFERENCES 1. Ashkin, A.; Dziedzic, J. M.; Bjorkholm, J. E; Chu, S. Opt. Lett. 1986, 11, 288-290. 2. Li, T.; Kheifet, S.; Medellin, D; Raizen, M. G. Science 2010, 328, 1673-1675. 3. Pollard, M. R.; Botchway, S. W.; Chichkov, B.; Freeman, E.; Halsall, R. N. J.; Jenkins, D. W. K.; Loader, I.; Ovsianikov, A.; Parker, A. W.; Stevens, R. New J. Phys. 2010, 12, 113056. 4. Fazal, F. M.; Block, S. M. Nature Photon. 2011, 5, 318-321. 5. Curtis, J. E., Koss, B. A.; Grier, D. G. Opt. Commun. 2002, 207, 169-175. 6. Misawa, H.; Koshioka, M.; Sasaki, K.; Kitamura, N.; Masuhara, H. Chem. Lett. 1991, 469472. 7. Sasaki, K.; Koshioka, M.; Misawa, H.; Kitamura, N.; Masuhara, H. Opt. Lett. 1991, 16, 1463-1465. 8. Mellor, C. D.; Bain, C. D. ChemPhysChem 2006, 7, 329-332. 9. Mellor, C. D.; Fennerty, T. A.; Bain, C. D. Opt. Express 2006, 14, 10079-10088. 10. Taylor, J. M.; Wong, L. Y.; Bain, C. D.; Love G. D. Opt. Express 2008, 16, 6921-6929. 11. Righini M.; Zelenina A. S.; Girard C.; Quidant R. Parallel and selective trapping in a patterned plasmonic landscape. Nat. Phys. 2007, 3, 477-480. 12. Grigorenko, N. A.; Roberts, W. N.; Dickinson, R. M.; Zhang, Y. Nat. Photon. 2008, 2, 365370. 13. Pang, Y.; Gordon, R. Nano Lett. 2011, 11, 3763-3767. 14. Pang, Y.; Gordon, R. Nano Lett. 2012, 12, 402-406. 15. Tanaka, Y.; Kaneda, S.; Sasaki, K. Nano Lett. 2013, 13, 2146-2150. 16. Shoji, T; Tsuboi, Y. J. Phys. Chem. Lett. 2014, 5, 2957-2967. 17. Yang, A. H. J.; Moore, S. D.; Schmidt, B. S.; Klug, M.; Lipson, M.; Erickson, D. Nature 2009, 457, 71-75. 18. Jaquay, E.; Martínez, L. J.; Mejia, C. A.; Povinelli, M. L. Nano Lett. 2013, 13, 2290-2294. 19. Jaquay, E.; Martínez, L. J.; Huang, N.; Mejia, C. A.; Sarkar, D.; Povinelli M. L. Nano Lett. 2014, 14, 5184-5188. 20. Friese, M. E. J.; Nieminen, T. A.; Heckenberg N. R.; Rubinsztein-Dunlop, H. Nature 1998, 394, 348-350. 21. Curtis, J. E.; Grier, D. G. Phys. Rev. Lett. 2003, 90, 133901. 22. Pan, L.; Ishikawa, A.; Tamai, N. Phys. Rev. B 2007, 75, 161305. 23. Chiang, W.-Y.; Okuhata, T.; Usman, A.; Tamai, N.; Masuhara, H. J. Phys. Chem. B 2014, 118, 14010-14016.


12

Page 13 of 14 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

Nano Letters

24. De, A. K.; Roy, D., Dutta, A.; Goswami, D. Appl. Opt. 2009, 48, G33-G37. 25. Usman, A.; Chiang, W.-Y.; Masuhara, H. J. Photochem. Photobiol., A 2012, 234, 83-90. 26. Liu, T. H.; Chiang, W.-Y.; Usman, A.; Masuhara, H. J. Phys. Chem. C 2016, 120, 2392-2399. 27. Burns, M. M.; Fournier, J. M.; Golovchenko, J. A. Phys. Rev. Lett. 1989, 63, 1233-1236. 28. Burns, M. M.; Fournier, J. M.; Golovchenko, J. A. Science 1990, 249, 749-754. 29. Grzegorczyk, T. M.; Rohner, J.; Fournier, J. M. Phys. Rev. Lett. 2014, 112, 023902. 30. Wang, S.-F.; Yuyama, K.; Sugiyama, T.; Masuhara, H. J. Phys. Chem. C. 2016, 120, 1557815585. 31. Wang, S.-F.; Kudo, T.; Yuyama, K.; Sugiyama, T.; Masuhara, H. Langmuir 2016, 32, 1248812496. 32. Kudo T.; Wang S.-F.; Yuyama K.; Masuhara H. Nano Lett. 2016, 16, 3058-3062. 33. Tanaka, Y.; Yoshikawa, H.; Itoh, T; Ishikawa, M. Opt. Express 2009, 17, 18760-18767. 34. Ohlinger, A.; Nedev, S.; Lutich, A. A.; Feldmann, J. Nano Lett. 2011, 11, 1770-1774. 35. Minowa, Y.; Kawai, R.; Ashida, M. Opt. Lett. 2015, 40, 906-909. 36. Demergis, V.; Florin, E.-L. Nano Lett. 2012, 12, 5756-5760. 37. Yan, Z.; Shah, R. A.; Chado, G.; Gray, S. K.; Pelton M.; Scherer N. F. ACS Nano 2013, 7, 1790-1802. 38. Yan, Z.; Gray, S. K.; Scherer, N. F. Nat. Commun. 2014, 5, 3751. 39. Sule, N.; Yifat, Y.; Gray, S. K.; Scherer, N. F. Nano Lett. 2017, 17, 6548-6556. 40. Sugiyama, T.; Adachi, T.; Masuhara, H. Chem. Lett. 2007, 36, 1480–1481. 41. Yuyama, K.; Rungsimanon, T.; Sugiyama, T.; Masuhara, H. Cryst. Growth Des. 2012, 12, 2427-2434. 42. Sugiyama, T.; Yuyama, K.; Masuhara, H. Acc. Chem. Res. 2012, 45, 1946-1954. 43. Tu, J.-R.; Miura, A.; Yuyama, K.; Masuhara, H.; Sugiyama, T. Cryst. Growth Des. 2014, 14, 15-22. 44. Yuyama, K.; Chang, K.-D.; Tu, J.-R.; Masuhara, H.; Sugiyama, T. Phys. Chem. Chem. Phys. 2018, 20, 6034-6039. 45. Yuyama, K. Ueda, M.; Nagao, S.; Hirota, S.; Sugiyama, T.; Masuhara, H. Angew. Chem. Int. Ed. 2017, 56, 6737-6743. 46. Uwada, T.; Sugiyama, T.; Masuhara, H. J. Photochem. Photobiol., A 2011, 221, 187-193. 47. Curto, A. G.; Volpe, G.; Taminiau, T. H.; Kreuzer, M. P.; Quidant, R.; van Hulst, N. F. Science 2010, 329, 930-932 48. Ito, S.; Sugiyama, T.; Toitani, N.; Katayama, G.; Miyasaka, H. J. Phys. Chem. B 2007, 111, 2365-2371. 49. Hofkens, J.; Hotta, J.; Sasaki, K.; Masuhara, H.; Iwai, K.; Langmuir 1997, 13, 414-419.


13

Nano Letters 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 14 of 14

50. Braun, D.; Libchaber, A. Phys. Rev. Lett. 2002, 89, 188103. 51. Delgado-Buscalioni, R.; Meléndez, M.; Luis-Hita, J.; Marqués, M. I.; Sáenz, J. J. arXiv:1709.04424, 2017.


14

A single large assembly with dynamically fluctuating swarms of gold

Recommend Documents