Femtosecond Laser Trapping Dynamics of Nanoparticles: A Single

Subscriber access provided by University of Florida | Smathers Libraries

Article

Femtosecond Laser Trapping Dynamics of Nanoparticles: A Single Transient Assembly Formation Leading to Their Directional Ejection Wei-Yi Chiang, Anwar Usman, Teruki Sugiyama, Johan Hofkens, and Hiroshi Masuhara J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.7b08127 • Publication Date (Web): 23 Oct 2017 Downloaded from http://pubs.acs.org on October 28, 2017

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 free 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 accessible to all readers and 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.

The Journal of Physical Chemistry C 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 28

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

The Journal of Physical Chemistry

Femtosecond Laser Trapping Dynamics of Nanoparticles: A Single Transient Assembly Formation Leading to Their Directional Ejection

Wei-Yi Chiang,1,3 Anwar Usman,2,* Teruki Sugiyama,1,4* Johan Hofkens,3,* Hiroshi Masuhara1,* 1

Department of Applied Chemistry, College of Science, National Chiao Tung University, 1001 Ta Hsueh Rd., Hsinchu 30010, Taiwan, 2

Department of Chemistry, Faculty of Science, Universiti Brunei Darussalam, Jalan Tungku Link, Gadong BE1410, Negara Brunei Darussalam 3

Single Molecule Unit, Department of Chemistry, Katholieke Universiteit Leuven, Celestijnenlaan 200F, 3001 Heverlee, België 4

Graduate School of Materials Science, Nara Institute of Science and Technology, 8916-5 Takaya-ma-cho, Ikoma, Nara 630-0192, Japan

Page | 1

ACS Paragon Plus Environment


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 28

ABSTRACT:

We investigated femtosecond laser trapping dynamics of silica nanoparticles with different hydrophobic surface properties. We demonstrated that hydrophobic surface on the silica nanoparticles facilitates mutual association of the nanoparticles in the optical trapping site. Such association of optically trapped nanoparticles is a prerequisite to induce their directional ejection away from the trapping site. The directional ejection of the optically trapped NPs is most probably due to asymmetric three-dimensional ejecting forces generated by the electromagnetic interaction between transient assembly in the focal spot and the incident pulses. These findings provide important insights into the directional ejection of nanoparticles from the tapping site in the femtosecond laser trapping, and this physico-chemical phenomenon is controlled by both the trapping laser and material properties.

Page | 2


Page 3 of 28

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


1.

INTRODUCTION Optical trapping which utilizes a tightly focused single continuous-wave (CW) laser

beam has emerged as a powerful tool for optical manipulation of nano- and micro-meter sized particles.1-6 In particular, the radiation pressure of the CW lasers induces optical trapping and molecular assembly in the focal spot, leading to the phase transition of the polymer,7 crystallization,8 and photopolymerization.9 In recent years, applying high-repetition-rate ultrashort-pulsed laser instead of the conventional CW lasers as the trapping beam has been proven to induce high trapping efficiency of Rayleigh particles.10,11 Moreover, high impulsive electromagnetic field of femtosecond (fs) laser-pulse train has been demonstrated to be able to split potential trapping minimum of gold nanoparticles (NPs) into two minima along the laser polarization,12 to enable reversible trapping and release of λ-DNA,13 and to induce directional ejection of polystyrene NPs from the trapping site.14 These interesting optical trapping phenomena have been evaluated in detail and they have been related to the optical properties of target particles and laser pulses, including nonlinear optical properties,15,16 particle number density and interparticle interactions,17 the impulsive peak power of a fs pulse which depends on the pulse width,18,19,20 longitudinal and transversal light-field distribution in the focal spot,18,19 and relaxation time between pulses.21-23 Complexity of the fundamental repetitive optical gradient forces, impulsive peak power, intrinsic optical properties, and dynamics of the particles in the trapping site prevents a detailed understanding on the optical trapping mechanism of the ultrashort laser-pulse train,18,23 although the role of intrinsic optical properties, for instance, the enhanced trapping efficiency of quantum dots or dye-doped NPs due to two-photon absorption,10,11 nonlinear susceptibility,12,15 and nonlinear refractive index16 have been experimentally and theoretically evaluated in detail. The optical trapping under the ultrashort laser-pulse train has attracted great interests due to its fascinating characteristics, particularly its ability to control optical manipulation of target NPs in the Rayleigh regime.10,11,23 We believed that by adjusting optical properties of the NPs, Page | 3



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 28

impulsive peak power, repetition rate or relaxation time between pulses, and the numerical aperture (NA) of objective lens in the trapping experiments, one can achieve high efficiency and full control of their optical traps, offering the next generation of single-beam laser trapping for new opportunities of optical nanomanipulation and technology. Recent advances in optical trapping under the ultrashort laser-pulse train demonstrated that pulse width of the fs laser, related to impulsive peak power, can control the number of optically trapped polystyrene NPs and their migration distances from the trapping site.17 On the other hand, by varying the time interval between the pulses we further demonstrated the role of relaxation time in drag and release of the NPs into and out from the trapping site.23 We have also shown that the NPs are always non-simultaneously ejected in the two directions perpendicular to the polarization vector of the incident laser,14,17 suggesting that the NPs should experience asymmetrical optical forces which are generated by the interaction between a large group NPs and the electromagnetic field in the focal spot.18 In this sense, even for spherical NPs, we have shown that the magnitude of gradient force along the polarization vector is always larger than that perpendicular to it. Moreover, temporal force which is the intrinsic character of optical trapping with ultrafast laser pulses is anisotropic, where it tends to push the NPs along the beam propagation. Since individual NPs in the Rayleigh regime ejected from the trapping site is not detected by an optical microscope, the observable directional ejection should be a large group of optically trapped NPs, which are associated into a kind of transient assembly and ejected from the focal spot. With this consideration, the formation of such transient assembly in the trapping site is essential for the directional ejection. The transient assembly, however, can dissociate within the ns time scales and, thus, directional ejection occurs only when the NPs are trapped under high-repetition-rate laser pulses. Under such trapping condition, the NPs are gathered, trapped, and ejected rapidly with the dynamics much faster than 33-ms time window. The Page | 4


Page 5 of 28

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


filling up of the trapping site with the NPs and their ejection occur within a few tens of ms.23 Therefore, we have demonstrated that the ejection in the two directions perpendicular to the linear polarization of the incident laser is always detected by charge coupled device (CCD) camera, as if the ejection is a continuous process.14,17,23 In order to understand fully this issue, the effects of chemical property of NPs on optical trapping and directional ejection of NPs under fs laser trapping is pursued.17,23 In this work, we evaluated optical trapping dynamics of chemically-modified silica NPs under high-repetition-rate fs laser-pulse train. Here, we present that through silanization to generate hydrophobic surface on the silica NPs we can modify their interparticle interactions leading to the formation of their transient assembly. We then examine the effect of such chemical modification, in addition to concentration and size effects, on the directional ejection of the silica NPs. By comparing their optical trapping behavior, we demonstrate that the optical trapping and directional ejection of the NPs are governed by the interparticle interactions of the NPs, implying the role of the formation of transient assembly on the directional ejection of the NPs.

2.

EXPERIMENTAL SECTION 2.1. Sample Preparation. Polystyrene NPs (diameter = 50 nm, refractive index = 1.59

at 800 nm)24 and nonporous silica spheres (diameter = 50 nm and 100 nm, refractive index = 1.43-1.46) suspended in distilled water (refractive index = 1.33 at room temperature) were purchased from PolyScience. These silica spheres are abbreviated as silica NPs-(50) and NPs-(100), respectively. Hydrophobic activation on the silica NPs was prepared by adding dried silica NPs into triethoxy(octyl)silane (OTES; refractive index = 1.42; ≥97.5%; Sigma-Aldrich) solution in toluene. The mixture was stirred, followed by sonication at room temperature for 12 hr, and the silica NPs then were washed with toluene for four times. OTES Page | 5



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 28

in the solution would be chemically bonded or self-assembled on the surface of silica NPs. In order to evaluate the effect of silane layer on the trapping dynamics of silica NPs, we have prepared OTES-silica NPs using four different concentrations of OTES (0.32, 0.16, 0.038, and 0.019×10-3 M). The sizes of silica NPs-(50) and -(100) before and after silanization coverages were evaluated by dynamic light scattering (DLS) measurement by employing a digital correlator DLS (BI 9000, Brookhaven Instruments Corp.) with the laser source being a diode-pumped laser (532 nm; 10 mW). Briefly, in prior to DLS measurement, the chemically modified silica NPs were filtered off using a membrane with the pore size of 0.5 µm to avoid large aggregates. The solution of the silica NPs was irradiated and the scattering light signal perpendicular to the laser source was collected. The data were fitted with an autocorrelation function and the particle sizes were computed with a non-negative least-squares method. In this measurement, the sample chamber was kept at 20℃. Scanning electron microscopy (SEM) (JEOL JSM67005) has also been employed to evaluate the morphology and size of the silica NPs before and after silanization coverages. SEM images and size distributions of the silica NPs-(50) and (OTES)-silica NPs-(50) are shown in Fig. 1. Accordingly, the size distribution of silica NPs-(100) and (OTES)-silica NPs-(100) are shown in Fig. S1 in Supporting Information. The SEM images directly indicate that the hydrophobic silica NPs become larger than the bare NPs; the mean size of silica NPs-(50) were enlarged to 130 nm. Similarly, upon silanization, silica NPs-(100) were enlarged to around 130, 150 and 450 nm depending on the OTES concentration. Hereafter, these modified silica NPs are abbreviated as (OTES)-silica NPs-(50→130), NPs-(100→130), NPs-(100→150), and NPs-(100→450), respectively. For optical trapping experiments, the colloidal solutions of polystyrene, silica NPs-(50), or (OTES)-silica NPs-(50→130) with the same number density of 3.8×1014 particles/mL Page | 6


Page 7 of 28

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


were put into sample cells consisting of two cover glasses (32×24×0.17 mm; Matsunami) and a silicon chamber (1 mm thickness, 10 mm diameter of inner well). In one set of experiments, the number density of 3.9×1013 particles/mL is used for all of silica NPs-(100) or (OTES)-silica NPs-(100). 2.2. Optical Setup. We employed an optical trapping system on inverted microscope (Eclipse Ti) integrated with dark-field microscopy, adopting the experimental setup reported earlier17 with a slight modification. In brief, fundamental mode of laser light from a Ti:sapphire laser (Mai Tai; 800 nm, 80 MHz) acting as a trapping beam was collimated by a pair of positive lenses with focal length being 100 and 200 mm, respectively. In the current study, the laser pulse was passed through a coupled prism to adjust its moderate pulse-width (100 fs), so that the distance of NPs ejected from the trapping site is still within the image frame of the CCD camera. The beam then was passed through a half-wave plate to control the linear polarization state. Finally, the beam was focused by an objective lens (UPLanApo; 60×, NA 0.90) into the sample cell mounted on the stage of an inverted microscope. In all experiments, the laser power at the sample was constantly controlled to be at 350 mW, equivalent to a peak intensity of 4.38 nJ per pulse. 2.3. Dark-Field Microscopy. The dark-field microscopy was set by passing white-light illumination from halogen lamp (λ=400−750 nm) through a dark-field condenser lens (NA 1.2−1.4) into the sample cell. The probe light was collected by the same objective, and it was sent into a charge coupled device (CCD) camera (JAI; CV-A55IR E) running at 30 interlaced frames per second, allowing the imaging of the trapped and ejected NPs at 33 ms per frame. In all measurements, the elastic light scattering originating from the laser trapping beam was completely cut by a shortwave-pass filter with transmission at λ = 380−720 nm (Semrock; Brightline 750/SP) in the front of the CCD camera. With such an optical setup and detection system, Rayleigh scattering of probe light by the NPs was detected and the intensity of the Page | 7



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 28

elastic scattering light was linearly correlated with the number of NPs at the trapping site or ejected to the surrounding area. Optical trapping and directional ejection dynamics of the NPs were analyzed from successive images which were continuously recorded by the CCD camera for ∼120 s.

3.

RESULTS AND DISCUSSION 3.1. Trapping and Directional Ejection of Silica NPs and Silane-Coated Silica NPs.

We have demonstrated that polystyrene NPs shows a unique trapping behavior upon fs laser irradiation, so that we study fs laser trapping dynamics of silica NPs as another representative dielectric NPs in order to understand this behavior more generally. In Fig. 2, we show a series of Rayleigh scattering images of silica NPs-(50) and (OTES)-silica NPs-(50) optically trapped by fs laser pulse train. We found silica NPs-(50) are just trapped and no directional ejection is clearly induced. The scattering intensity profile along the axis horizontally crossing the focus indicates that only the center is bright. Consequently, it is concluded that the silica NPs are stably trapped at the focal spot. On the other hand, under the same experimental conditions, (OTES)-silica NPs-(50) exhibit directional ejection (Fig. 2b), similarly to the case of polystyrene NPs.14 This means that the hydrophobic surface completely changed the fs trapping behavior of the silica NPs. Upon switching on the trapping laser, (OTES)-silica NP-(50→130) streamed out from the focus along the direction perpendicular to the laser polarization vector. The ejection is always induced only to one direction or the opposite one, and the ejection to both side is never observed, namely the ejection switches from right to left and from left to right arbitrarily. However, the probability for both sides is likely similar when we observed the trapping behavior for a certain period. It is noteworthy that when the polarization vector of the trapping laser was rotated, ejection direction was also rotated keeping its perpendicular relation to the polarization vector. The Page | 8


Page 9 of 28

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


images shown in Fig. 2(b) were prepared by combining two halves of two different image frames separated by the dashed line. Thus the combined images give a pair of partially folded fan-like scattering of NPs. As shown in Fig. 2(b), the scattering light intensity at the focal spot is higher for (OTES)-silica NPs as compared with that of silica NPs. This should be ascribed to higher trapping efficiency or higher number of (OTES)-silica NPs in the trapping site. This suggests that hydrophobic surface treatment enhances trapping efficiency of the silica NPs, and that their interfacial and interparticle interactions suppress the dynamics of individual NPs in the trapping site. Considering that the refractive index of OTES is 1.42, which is almost the same with that of silica NPs, the refractive index of the silica NPs remains unchanged upon silanization. Thus, the change in the refractive index that can induce huge difference of trapping behavior of (OTES)-silica NPs compared with that of silica NPs can be ignored. Furthermore, the directional ejection of the (OTES)-silica NPs is also clearly observed perpendicular to the polarization vector of the trapping laser. This behavior is quite similar to that of polystyrene NPs. It is considered that hydrophobic surface of silica NPs enables mutual association and the resultant volume of the association receives a strong scattering force. This finding further indicates that the directional ejection depends essentially on the formation of the transient assembly, and it supports our previous notion that polystyrene NPs are actually interacting with each other due to their hydrophobicity to form the transient assembly in the trapping site.14,17 For the silica NPs, the low scattering intensity at the focal spot and the absence of directional ejection could be explained by the following two viewpoints. Firstly, the relative refractive index of silica NPs with respect to water medium is lower than that of polystyrene NPs, resulting in lower optical cross-sections, optical forces, and scattering light intensity of silica NPs as compared with those of polystyrene NPs. In this case, generated optical forces Page | 9



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 28

exerted on silica NPs would not be able to gather high number of silica NPs in the trapping site, thus there is no directional ejection observed.27 This notion has been clearly demonstrated by the dependence of directional ejection of polystyrene NPs on its particle number density.17 From this viewpoint, we have increased the particle number density of silica NPs solution by a factor of four, but we were unable to observe the directional ejection of the silica NPs. Thus the refractive index difference of silica NP to water medium is not a possible reason giving no ejection. Secondly, the silica NPs is much less hydrophobic compared with polystyrene NPs. In water medium, unlike polystyrene NPs, the hydrophilic surface of the silica NPs prevents them to interact with each other. Therefore, even large number of silica NPs are gathered in trapping site, they may not associate to form an assembly to generate stronger gradient, scattering, and temporal forces.7 This means that individual NPs in the trap moves vigorously, never forming any transient assembly. An ideal approach to unambiguously demonstrate whether it is the former or the latter mechanism which governs the directional ejection is to compare the trapping behavior of silica NPs and (OTES)-silica NPs. By modifying the surface of silica NPs to be hydrophobic, we expect that they can associate in the trapping site. 25,26 We note that (OTES)-silica NP-(50→130) has a larger volume compared to silica NPs-(50). One may point out that volume expansion upon the chemical modification of the surface contributes to efficient trapping and surface property is not so critical. In this sense, scattering and temporal forces exerted on a Rayleigh particle is super-linearly proportional to its volume. To examine this possible size effect on the trapping and ejection phenomena, we have carried out laser trapping of silica NPs with different sizes; silica NPs-(100), NPs-(300), and NPs-(500). Under the same laser trapping conditions, all these silica NPs scarcely showed directional ejection, as summarized in Fig. 2(c). This result suggests that, although silica NPs receive the stronger gradient, scattering, and temporal forces with increasing their Page | 10


Page 11 of 28

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


size, they did not show the directional ejection of large number of NPs. Thus, we consider that the change in surface property from hydrophilic to hydrophobic is responsible for realizing the directional ejection behavior for (OTES)-silica NP-(50→130). In this sense, chemically modified hydrophobic surface of silica NPs enables mutual association of the NPs in water under high concentration forming an assembly.25,26 We can therefore rationalize these observations by considering that hydrophobic surface treatment enhances the interfacial and interparticle interactions between NPs leading to a transient assembly formation in the trapping site, and that the strong scattering and temporal forces then ejects them from the focal spot. As we have previously reported, the ejection occurs rapidly after the trapping laser was switched on and the two ejection directions perpendicular to the polarization vector occur stochastically and alternatively in average.14,17 Thus further quantitative analyses on the trapping and ejection dynamics can only be provided by ultrafast spatial and temporal spectroscopy. 3.2. Trapping and Directional Ejection Dynamics of Silane-Coated Silica NPs with Different Thickness. In order to elucidate the correlation between the trapping and ejection dynamics of hydrophobic silica NPs, we studied silica NP-(100) and (OTES)-silica NPs-(100→130, 150, and 450) under the same experimental conditions. We should note that the central core of silica NP was 100 nm and common to all, while the thickness of silane layer varied from 30 to 350 nm. As shown in Fig. 3a, the directional ejection is also absent in the optical trapping of silica NPs-(100), similarly to silica NPs-(50).27 In contrary, (OTES)-silica NPs show the alternative directional ejection perpendicular to the laser polarization vector, as shown Fig. 3b-d. Interestingly, the directional ejection is observed even for those with low silanization coverage, (OTES)-silica NPs-(100→130). Notably, as shown by their dark-field microscopy images in Fig. 3b-d, the scattering spatial profiles of (OTES)-silica NPs are different depending on the thickness of their silane Page | 11



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 28

layer. While (OTES)-silica NPs-(50→130) shows a folding fan-shape alternating in the two opposite directions with similar shape of scattering images to polystyrene NPs (Fig. 3b),14,17 those with thicker silane layer, (OTES)-silica NPs-(100→150) and NPs-(100→450), give two sphere-like profiles in both sides of the polarization directions as shown in Fig. 3c-d. Moreover, radius of the sphere-like swarm tends to be larger with the thickness of silane layer. The sphere-like structure indicates that interparticle interactions in an assembly formed in the trap are enough strong to keep its shape even after being ejected from the focal spot. Directional flow of the assembly may distort its shape and NPs at the edge may be removed partly, which should be overcome by associative force among the NPs in the assembly. We can consider that once the surface becomes hydrophobic the strength of interparticle attractive force does not depend on the thickness of the silane layer. However, the surface area of (OTES)-silica NPs must increase with the thickness of the silane layer, leading to wider contact and eventually resulting in an assembly of larger number of NPs. Upon interacting with the trapping laser the larger size of assembly generates stronger gradient, scattering, and temporal forces. This in turn supports our proposal that the larger number of NPs were associated in the optical potential, forming an assembly which can result in stronger optical forces in the trapping site, leading to their directional ejection.14,17 Indeed the size of the ejected scattering patterns was larger with the increase in the thickness of the silane layer, as clearly demonstrated in Fig. 3, indicating that the size of the assembly in the trapping site is determined by the thickness of the silane layer on the silica NPs. To exemplify further the aforementioned issue, we observed trapping and ejection behavior of (OTES)-silica NP-(100→450), the biggest one among the present samples, in more detail. In this case, we observed that two spherical scattering patterns were produced sequentially from the focal spot, as shown in Fig. 4(a), where back scattering images are presented without any combination of different frames. The intensity profiles along the Page | 12


Page 13 of 28

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


horizontal line including the focus is also given above the images. They directly show how scattering patterns underwent dispersion, namely the one dimensional distribution became wide with time scale of a few tens ms. These observations are well interpreted by the following two processes. Firstly, the assembly is pushed out from the focal volume as a whole because of strong association of OTES-silica NPs. The assembly lost directional motional energy due to diffusional friction with solvent, and then dissolution of the ejected assembly is started. The ejection is completed in 33 ms (one frame) and the following dispersion takes place in a few tens ms as observed in the pictures of Fig. 4(a). If we consider that the sphere-like images of the ejected (OTES)-silica NPs-(100) with different thickness can be attributed to a large size of assembly of NPs flowing away from the trapping site, these results indicate the relationship between the silanization coverages and interfacial and interparticle interactions. One can expect that higher silanization coverages assemble larger number of NPs, as mentioned above. In this case, when this transient assembly is ejected, the trapping site is quickly refilled with NPs, and the repetitive optical trapping and directional ejection of the transient assemblies occur rapidly. The successive assemblies of NPs flowing away from the trapping site were observed as discrete sphere-like swarm images, as seen in Figs. 3c-d. As the image of the spheres is much larger than the focal spot, we may also anticipate that the assembly of NPs in the trapping site should be expanded from the focal volume accompanying NP density lowering. When the assembly of NPs flows away from the trapping site its morphology has been considerably changed and interparticle interactions were weakened, giving apparently larger aggregate. Indeed, eventually the flowing assembly of NPs dissociates into the surroundings area away from the focal spot, and such dissociation was clearly observed after switching off the trapping laser beam, as discussed below. 3.3. Direct Observation of a Transient Assembly of (OTES)-Silica NPs. Mutual Page | 13



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 28

reactions on the surface of silica NPs result in an increase in their aggregation, as demonstrated by the sizes of (OTES)-silica NPs-(100→130, 150, and 450). However, we recall that the particle size is not responsible for the ejection behaviors for silica NPs (see section 3.1). Therefore, we have to consider that chemical modification using OTES at the surface of the silica NPs provide interfacial and interparticle interactions, leading to the formation of a transient assembly of NPs in the focal spot. Along this idea, one can anticipate that the NPs ejected from the trapping site should be in a large size of assembly. Considering that the energy of hydrophobic interactions in only 3–7 kcal/mol, the energies required for an assembly of NPs would be very low. Thus, when the hydrophobic surface is formed, even with low silanization coverage, (OTES)-silica NPs can easily form an assembly in the trapping site, leading to the directional ejection. The directional ejection of the assembly from the trapping site is governed by the asymmetric gradient, scattering, and temporal forces exerting on the assembly, rather than on individual NPs. To exemplify further this issue, additionally we examined the trapping and ejection behavior of (OTES)-silica NPs-(100→450) by optical transmission microscopy, by which we observed spherical dark areas emerging and diffusing out from the focal spot. As summarized in Fig. 4(b), several images were observed during the laser irradiation (upper pictures) and after switching off the laser (lower pictures). All these images were observed at one frame. The dark area in transmission images indicates that the probing light is much reduced through transmission, namely the density of NPs of the area is high. Although Fig. 4(a) and (b) show the same trapping and diffusion behavior of NPs, the size of the dark area is always smaller than that of scattering pattern. We consider that the spherical light scattering pattern represents distribution of ejected NPs, as described above, and the dark spherical area corresponds to an assembly where NPs are tightly packed. Examining various combination of dark-field and transmission images Page | 14


Page 15 of 28

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


measured simultaneously, it is concluded that the dark area is located at the center of the scattering pattern. This single assembly, represented by a single dark area, was ejected sequentially while keeping its size and density of NPs, and its size was reduced during the diffusion. The direct observation of the single assembly of (OTES)-silica NP-(100→450) is thus achieved in the present work. In line with the sphere-like swarm shape in the dark-field images (see Fig. 3(b)), the bright field images show a large size of assembly of NPs at the trapping site, seen as a black spot in Figs. 4b(1)-(3). The assembly flows away from the trapping site when the laser was switched off (see Figs. 4b(4)-(6)). The size of the black spot changes with time due to diffusion and floating of the NPs out from the focal plane. This finding indicates that the assembly of NPs migrates three dimensionally away from the trapping site. The directional ejection suggests that the NPs are trapped and ejected by asymmetric three-dimensional optical forces, which are generated by the electromagnetic interaction between the transient assembly of NPs optically trapped in the focal spot with successive incident laser pulses.28,29 Thus, this finding suggests that the transient assembly of NPs has either an asymmetric structure or its center of mass shifted from the beam center even when the assembly is symmetric, which may cooperatively generate the asymmetric optical force leading the directional ejection, as described in section 3.4. The ejection of assembly of NPs is fast, thus we could not follow its diffusion from the focal spot with the frame resolution of 33 ms. In Fig. 4(b)(3), we found the second dark area near the focus, while the first dark area was observed at the left side of the focus. The intensity profiles did not correspond to the assembly, as they were obtained along the axis crossing the focus horizontally and the assemblies diffused out from the axis. The assembly just ejected from the focus is monitored on the focal plane, but three-dimensional diffusion of the dark area results in de-focusing image more or less. Fortunately, we can still guess its Page | 15



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 16 of 28

shape and size from the scattering and transmission images, as shown in Fig. 4(b)(4)-(6). The trapping laser was switched off after 9.2 s, so new assembly was not produced anymore and we could follow the size and shape change. This observation indeed represents the dynamics of a single assembly of (OTES)-silica NP-(100→450), and its shrinking in the dark area took place in about 2 s. The close observation at 11.3 which is given in enlarged picture of Fig. 4(b)(6) told us that the central shape is non-spherical, namely, it is asymmetric as we propose in the previous papers.17,23 3.4. Formation and Dissociation Dynamics of a Transient Assembly and Its Asymmetric Structure. Repetitive fs pulses gather (OTES)-silica NPs in the optical potential and form their assembly. The relatively larger assembly receives stronger gradient force and grows more rapidly, as the force is linearly proportional to their size. A bigger one grows faster compared to other smaller ones, and then a resultant single assembly of NPs may have an important role in the trapping dynamics of (OTES)-silica NPs. The assembly is dynamically moving in the potential. At this stage the gradient force exerted on the assembly overcomes the scattering and temporal forces, while the gradient force becomes relatively smaller than the scattering and temporal forces for the assembly with a certain size, due to that the former and the latter forces is linearly and super-linearly proportional to the size of the assembly, respectively, pushing the assembly out front the trapping site. Now we have two possible explanations on alternatively switched ejection of (OTES)-silica NPs from the focal spot. One is based on the asymmetric transient assembly formation, as we proposed. An asymmetric assembly grows rapidly, not reflecting the symmetrical potential and reaches closely the size of the trap. Asymmetric geometrical arrangement of the assembly in the focus may lead to asymmetric ejection when the ejecting force becomes stronger than the gradient force. This possibility is schematically illustrated in Fig. 5. In this work, 80 MHz repetitive pulses are employed to trap the NPs, so that attraction Page | 16


Page 17 of 28

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


force toward the focal center (~100 fs) and release (12.5 ns) are repetitively applied on individual NPs. Association and aggregation of minor numbers of NPs evolves nonlinearly with size and time, giving a single asymmetric assembly in the trap. When the interparticle hydrophobic interaction between NPs is moderate, the ejected assembly may be decomposed into its component NPs giving a partially folded-fan scattering pattern (Fig. 5 a-f). This is demonstrated for (OTES)-silica NP-(50→130). Upon increasing the thickness of silane coated on the silica NPs, the spherical pattern is obtained. This can be indicative that, out from the trapping site, the assembly with larger number of NPs is not deformed or dissociated, keeping its structure. This understanding on an asymmetrical assembly interacting with the symmetrical optical field is illustrated in Fig. 5 a-f’(the lower part). The higher thickness of silane layer indicates more flexible surface structure, giving wider contact areas. Furthermore, the polymers and oligomers at the surface of (OTES)-silica NP-(100→450) may undergo mutual interpenetration of contacting NPs, making the assembly more stable and tighter. This is the reason why we have succeeded in detecting the assembly directly and elucidating its temporal behavior. The other plausible explanation is related to the position of the assembly with respect to the beam center when the scattering force becomes enough strong to kick it out from the trapping site into the surrounding area. If the assembly is accidentally located on the axis of the propagating trapping pulses, it is shifted toward upper, as the fs pulses are introduced from the bottom, and then pulled back to center of the focus by the gradient force. If the assembly is located at the right or left half, the scattering force pushes the assembly outside of the potential trap leading to right or left side ejection, respectively. At present we have no data supporting the second possibility on accidental location, while we have succeeded in observing directly the asymmetric transient assembly. Thus, we conclude that the asymmetric transient assembly formation is a main reason for giving alternatively switched ejection of NPs. Page | 17



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

4.

Page 18 of 28

CONCLUSIONS In conclusion, we have shown that trapping and ejection dynamics characteristic of fs

laser pulses is realized for hydrophobic NPs and directional ejection behavior depends on the hydrophobicity and/or the thickness of hydrophobic layer. When the surface is hydrophilic, directional ejection does not follow the trapping, showing conventional trapping at the focus. As hydrophobicity is increased, directional ejection is observed first as a partially folded fan-like scattering pattern and then as a spherical pattern. In the latter case the directional ejection consists of two processes; rapid ejection of an assembly with larger number of NPs from the focal spot and then it is dissociated slowly to component NPs. This spherical pattern is an assembly of NPs formed in the trapping potential and it is observed directly in transmission images as a dark area. By comparing both dark-field and transmission optical microscopic images, the ejection and dissociation dynamics were analyzed for silane-coated silica NPs-(100→450). We found that ejection and dissociation takes place in less than 33 ms and a few s, respectively. We discussed possible mechanisms of the alternative ejection behavior in terms of symmetrical optical field, asymmetrically formed assembly, and/or location of the assembly in the trap. The results reported here provide important insights into the directional ejection of NPs from the tapping site in the fs laser trapping, and this optical phenomenon is controlled by both parameters of the trapping laser and material properties. These insights would allow further development of fully controlled optical trapping behaviors of nanoparticles in Rayleigh regime with fs laser pulses, such as high trapping efficiency, directional ejection, and optofluidic flow of the nanoparticles around the focal spot as the next generation of single-beam optical traps for new opportunities of optical nanomanipulation and technology.

Page | 18


Page 19 of 28

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


ASSOCIATED CONTENT Supporting Information. The size distribution of silica NPs-(100), (OTES)-silica NPs-(100→130), (OTES)-silica NPs-(100→150), and (OTES)-silica NPs-(100→450). The Supporting Information is available free of charge on the ACS Publications website.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected] (A.U.); [email protected] (T.S.); [email protected] (J.H.); [email protected] (H.M.) Notes The authors declare no competing financial interest.

ACKNOWLEDGMENTS The present work is partly supported by MOE-ATU Project (National Chiao Tung University) of the Ministry of Education, Taiwan to H.M., and by grants from Ministry of Science and Technology to H.M. (MOST 105-2811-M-009-022 and MOST 106-2113-M-009-026) and to T.S. (MOST 104-2113-M-009-021), and by JSPS KAKENHI Grant Number JP16H06507 in Scientific Research on Innovative Areas “Nano-Material Optical-Manipulation” to T.S. J.H. acknowledges financial support from the Research Foundation-Flanders (FWO, grant ZW15_09 GOH6316N) and the Flemish government through long term structural funding Methusalem (CASAS2, Meth/15/04), W.-Y.C. acknowledges the financial support from NCTU through the scholarship of 2016 Global Networking Talent Program and the KULeuven research council through the IDO program (interdisciplinary research IDO/13/008) for a personal grant. We sincerely thank Prof. Chia-Ching Chang of College of Bioscience and Technology of NCTU who supported us to carry out DLS measurement and Mr. Jim Jui-Kai Chen who operated optical microscope. Page | 19



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 20 of 28

Page | 20


Page 21 of 28

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


REFERENCES (1) Ashkin, A. Acceleration and Trapping of Particle by Radiation Pressure. Phys. Rev. Lett. (2)

1970, 24, 156−159. Ashkin, A.; Dziedzic, J. M.; Bjorkholm, J. E.; Chu, S. Observation of a Single-Beam

(3)

Gradient Force Optical Trap for Dielectric Particles. Opt. Lett. 1986, 11, 288−290. Lehmuskero, A., Johansson, P., Rubinsztein-Dunlop, H., Tong, L.; Käll, M. Laser

(4)

Trapping of Colloidal Metal Nanoparticles. ACS Nano 2015, 9, 3453–3469. Grier, D. G. A Revolution in Optical Manipulation. Nature 2003, 424, 810–816.

(5)

Dholakia, K., Reece, P.; Gu, M. Optical Micromanipulation. Chem. Soc. Rev. 2008, 37, 42–55.

(6)

Neuman, K. C.; Block, S. M. Optical Trapping. Rev. Sci. Instrum. 2004, 75, 2787−2809.

(7)

Hofkens, J.; Hotta, J.; Sasaki, K.; Masuhara, H.; Iwai, K. Molecular Assembling by the Radiation Pressure of a Focused Laser Beam: Poly(N-isopropylacrylamide) in Aqueous

(8) (9)

Solution. Langmuir 1997, 13, 414−419. Sugiyama, T.; Yuyama, K.; Masuhara, H. Laser Trapping Chemistry: From Polymer Assembly to Amino Acid Crystallization. Acc. Chem. Res. 2012, 45, 1946−1954. Ito, S.; Tanaka, Y.; Yoshikawa, H.; Ishibashi, Y.; Miyasaka, H.; Masuhara, H. Confinement of Photopolymerization and Solidification with Radiation Pressure. J. Am.

Chem. Soc. 2011, 113, 14472−14475. (10) Pan, L., Ishikawa, A.; Tamai, N. Detection of Optical Trapping of CdTe Quantum Dots by Two-Photon-Induced Luminescence. Phys. Rev. B 2007, 75, 161305. (11) Chiang, W.-Y., Okuhata, T., Usman, A., Tamai, N.; Masuhara, H. Efficient Optical Trapping of CdTe Quantum Dots by Femtosecond Laser Pulses. J. Phys. Chem. B 2014, 118, 14010−14016. (12) Jiang, Y.; Narushima, T.; Okamoto, H. Nonlinear Optical Effects in Trapping Nanoparticles with Femtosecond Pulses. Nat. Phys. 2010, 6, 1005–1009. (13) Shoji, T.; Saitoh, J.; Kitamura, N.; Nagasawa, F.; Murakoshi, K.; Yamauchi, H.; Ito, S.; Miyasaka, H.; Ishihara, H.; Tsuboi, Y. Permanent Fixing or Reversible Trapping and Release of DNA Micropatterns on a Gold Nanostructure Using Continuous-Wave or Femtosecond-Pulsed Near-Infrared Laser Light. J. Am. Chem. Soc. 2013, 135, 6643– 6648. (14) Usman, A., Chiang, W.-Y.; Masuhara, H. Optical Trapping and Polarization-Controlled Scattering of Dielectric Spherical Nanoparticles by Femtosecond Laser Pulses. J. Photochem. Photobiol. A Chem. 2012, 234, 83–90. (15) Kudo, T.; Ishihara, H. Proposed Nonlinear Resonance Laser Technique for Manipulating Nanoparticles. Phys. Rev. Lett. 2012, 109, 087402. (16) Devi, T.; De, A. K. Theoretical Investigation on Optical Kerr Effect in Femtosecond Page | 21



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 22 of 28

Laser Trapping of Dielectric Microspheres. J. Opt. 2017, 19, J. Opt. 2017, 19, 065504. (17) Chiang, W.-Y., Usman, A.; Masuhara, H. Femtosecond Pulse-Width Dependent Trapping and Directional Ejection Dynamics of Dielectric Nanoparticles. J. Phys. Chem. C 2013, 117, 19182–19188. (18) Usman, A., Chiang, W.-Y.; Masuhara, H. Optical Trapping of Nanoparticles by Ultrashort Laser Pulses. Sci. Prog. 2013, 96, 1–18. (19) Wang, L.-G.; Chai, H.-S. Revisit on Dynamic Radiation Forces Induced by Pulsed Gaussian Beams. Opt. Express 2011, 19, 14389–14402. (20) Shane, J. C., Mazilu, M., Lee, W. M.; Dholakia, K. Effect of Pulse Temporal Shape on Optical Trapping and Impulse Transfer Using Ultrashort Pulsed Lasers. Opt. Express 2010, 18, 7554–7668 (21) Agate, B., Brown, C., Sibbett, W.; Dholakia, K. Femtosecond Optical Tweezers for In-Situ Control of Two-Photon Fluorescence. Opt. Express 2004, 12, 3011−3017. (22) Liu, T.-H., Chiang, W.-Y., Usman, A.; Masuhara, H. Optical Trapping Dynamics of a Single Polystyrene Sphere: Continuous Wave versus Femtosecond Lasers. J. Phys. Chem. C 2016, 120, 2392–2399. (23) Muramatsu, M., Shen, T.-F., Chiang, W.-Y., Usman, A.; Masuhara, H. Picosecond Motional Relaxation of Nanoparticles in Femtosecond Laser Trapping. J. Phys. Chem. C 2016, 120, 5251–5256. (24) Miles, R. E. H., Rudić, S., Orr-Ewing, A. J.; Reid, J. P. Influence of Uncertainties in the Diameter and Refractive Index of Calibration Polystyrene Beads on the Retrieval of Aerosol Optical Properties Using Cavity Ring Down Spectroscopy. J. Phys. Chem. A 2010, 114, 7077–7084. (25) Lee, Y.-L., Du, Z.-C., Lin, W.-X.; Yang, Y.-M. Monolayer Behavior of Silica Particles at Air/Water Interface: A Comparison Between Chemical and Physical Modifications of Surface. J. Colloid Interface Sci. 2006, 296, 233–24. (26) Bagwe, R. P., Hilliard, L. R.; Tan, W. Surface Modification of Silica Nanoparticles to Reduce Aggregation and Nonspecific Binding. Langmuir 2006, 22, 4357–4362. (27) Usman, A.; Chiang, W.-Y.; Masuhara, H. Femtosecond Trapping Efficiency Enhanced for Nanosized Silica Spheres. Proc. SPIE 2012, 8458, 845833. (28) Harada, Y.; Asakura, T. Radiation Forces on a Dielectric Sphere in the Rayleigh Scattering Regime, Opt. Commun. 1996, 124, 529−541. (29) Wang, L.-G.; Zhao, C.-L. Dynamic Radiation Force of a Pulsed Gaussian Beam Acting on a Rayleigh Dielectric Sphere. Opt. Express 2007, 15, 10615–10621.

Page | 22


Page 23 of 28

Figure 1 (a)

(b)

(c)

(d) 100

Intensity / %

100

Intensity / %

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


50

0 56

58

60

62

Diameter / nm

50

0 30

40

50

120

150

180

Diameter / nm

Fig. 1. Representative SEM images and size distribution determined by DLS measurement. (a) and (c), silica nanoparticles of 50 nm diameter (silica NP-(50)). (b) and (d), modified silica nanoparticles of 50 nm diameter showing maximum peak around 130 nm ((OTES)-Silica NP-(50→130)).

Page | 23



Figure 2 (a)

(b)

100

(d)

0

-40

-20

0 µm

20

40

0 -40

-20

0 µm

20

40

20

(e)

0 -40

-20

0 µm

20

Is / a.u.

(c)

Is / a.u.

100

Is / a.u.

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 24 of 28

40

Fig. 2. Dark-field Rayleigh scattering images of optical trapping and directional ejection of (a) silica NPs-(50), (b) (OTES)-silica NPs-(50130), (c) silica NPs-(100), (d) silica NPs-(300) and (e) silica NPs-(500) in water. The images in (b) were prepared by combining two halves of two different image frames separated by the dashed line, showing the directional ejection of the NPs perpendicular to the polarization vector as indicated by the solid arrows. The graphs above images (c), (d) and (e) represent line profiles crossing the focal point. The laser polarization here is the same with that of (b).

Page | 24


Page 25 of 28

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


Figure 3 (a)

(b)

(c)

(d)

Fig. 3. Dark-field Rayleigh scattering images of optical trapping and directional ejection of silica NPs-(100) with different silane coverages (a) silica NPs, (b) (OTES)-silica NPs-(100→130), (c) (OTES)-silica NPs-(100→150), and (d) (OTES)-silica NPs-(100→450). These images were prepared by combining two halves of two different image frames, showing the alternative directions of the ejection perpendicular to the polarization vector. In this case, the polarization vector is on the vertical direction.

Page | 25



Figure 4 (a)

(b) µm

µm -40

Is/a.u.

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 26 of 28

-20

0

20

40

-40

-20

0

µm 20

40

-40

-20

0

20

40

20 0

1. (9.13 s)

2. (9.16 s)

3. (9.2 s)

Laser on

×2 4. (9.3 s)

5. (9.7 s)

6. (11.3 s)

Laser off

Fig. 4. (a) Rayleigh scattering images of sequential ejection of (OTES)-silica NPs-(100→450) from the focus to the surrounding. The red arrow indicates the focal spot of the trapping beam, the white bar corresponds to 10 µm, and the double headed arrows denotes the polarization direction. The graph on the top row is line profile on each image passing through the focal spot perpendicular to the polarization direction. The bright spots are the probed light scattered by the (OTES)-silica-NPs ejected from the trapping site. (b) Transmission image of modified silica nanoparticles of (OTES)-Silica NP-(100→450). A sphere-like swarms which are pushed out from the focal point were observed with optical transmission microscopy (see text). Red arrows represent the focal point when the trapping laser was switched on, while white arrows represent the focal point after the trapping laser was switched off. The polarization direction of the trapping fs laser is shown as a double headed arrow and the white bar corresponds to 10 µm. The graph on the top row is line profile passing on images 1-3 through the focal spot perpendicular to the polarization direction. Page | 26


Page 27 of 28

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


Figure 5

Fig. 5. The mechanism of fs trapping of silica NPs: (a) a dashed box representing dragging and releasing NPs in the trapping site, (a→c) the high repetitive dragging and releasing of the NPs indicated by the arrows, and (d) eventually a large number of NPs optically trapped in the trapping site, forming a kind of transient assembly. The transient assembly is not necessarily symmetry and the relative position of its center of mass may be shifted from to the focus center. As the results, the optical forces exerted on the assembly is asymmetric. (e) the assembly ejected out from the trapping site perpendicular to the polarization vector by the scattering and temporal forces, (f) for (OTES)-silica NPs-(50→130), NPs-(100→130) and NPs-(100→150), the assembly rapidly dissociated, and observed as a partial folded-fan in the dark-field scattering imaging, and (f’) for thick silane layer of (OTES)-silica NPs-(100→450), forming more stable and tighter assembly, and observed as a sphere-like swarm in the dark-field scattering imaging.

Page | 27



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 28 of 28

TOC

Page | 28


Femtosecond Laser Trapping Dynamics of Nanoparticles: A Single

Recommend Documents