Open Access
Add to BibSonomy
Abstract
Laser streaming is a phenomenon in which liquid streaming is driven directly from the laser through an in situ fabricated nanostructure. In this study, liquid streaming of a gold nanoparticle suspension driven by a pulsed laser was studied using a high-speed camera. The laser streaming formation time, streaming velocity, and relative energy conversion efficiency of laser streaming was measured for different nanoparticle concentrations, focal lens position, laser powers, and laser repetition rates. In addition to the laser intensity, which played a significant role in the formation process of laser streaming, the optical gradient force was found to be an important approach involved in the transport and provision of nanoparticles during the formation of laser streaming. This finding facilitated a better understanding of the formation mechanism of laser streaming and demonstrated the possibilities of a new potential laser etching technique based on nanosecond lasers and nanoparticle suspensions. This result can also expand the application of laser streaming in microfluids and other fields that require lasers to move macroscopic objects at relatively high speeds.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Lasers are one of the greatest inventions of the 20th century. With key properties such as spatial coherence and low divergence, lasers are employed as ideal energy carriers in various fields where high energy intensity, small size, and long-distance transmission are necessary. Recently, lasers have been used to move and control objects in laser cooling techniques [1,2], control nanoparticles in optical tweezers [3,4], or push spacecraft that have huge solar sails [5]. However, strict conditions and limitations are necessary in these applications. For example, the radiation pressure is too weak to drive macro objects in a non-vacuum environment. In addition to the above, various techniques that transfer the energy of a laser into the momentum of macroscopic objects in an indirect manner have been developed. However, most of these techniques involve photophoresis effect [6] which is owed to a non-uniform temperature distribution of an illuminated particle in a fluid medium, opto-thermocapillary effect [7–9] which arise from the a spatial imbalance of surface tension caused by focused laser, chromocapillary effects [10,11] which is the light-induced droplet motion by a wavelength-dependent liquid/liquid interfacial gradient. These above techniques can only be used in specific cases, such as objects with low velocity. In summary, how to move an object conveniently at a relatively high speed is still a scientific challenge.
A new physical phenomenon, called “laser streaming”, was experimentally discovered by Yanan Wang’s team in 2017 [12]. In their experimental study, a pulsed laser was focused on the inner surface of the cuvette filled with gold nanoparticle suspension. After a few minutes of irradiation, they observed the formation of a nanostructure on the inner surface of the cuvette and the generation of liquid streaming close to the nanostructure and perpendicular to the cuvette surface. Laser streaming was interpreted with two physical effects, namely the photoacoustic [13–19] and acoustic streaming [20–22] effects. Specifically, the laser was first converted into ultrasound by the nanostructure, and the ultrasound then pushed the liquid away. As a single component in laser streaming, a nanostructure with a diameter of approximately 100 μm can be easily fabricated and can drive macroscopic objects at relatively high speeds (several cm/s). These characteristics make laser streaming a potential technique for transferring the energy of a laser into the momentum of an object.
To date, there have been insufficient systematic studies on the laser streaming phenomenon. In this work, we experimentally studied the laser streaming phenomenon using a pulsed laser with a tunable repetition rate and higher laser power. In the following sections, we report the effect of the laser parameters on the formation time, streaming velocity, and energy conversion efficiency of laser streaming and demonstrate some interesting characteristics of this phenomenon.
2. Experimental setup
A schematic of the experimental setup used in this study is shown in Fig. 1. A 1×1×5 cm3 glass cuvette was filled with a gold nanoparticle suspension with a concentration of 4.3×1010 particles/mL. Gold nanoparticles with a diameter of 50 nm were used, considering that the 50 nm particles exhibit a strong localized surface plasmon resonance near 532 nm, thus allowing effective absorption of the incident laser. A 532-nm pulsed laser with a fixed pulse width of 30 ns, adjustable repetition rate of 1–10 kHz, and adjustable laser power of 50–3300 mW was used in this study. The laser was focused on the inner surface of the cuvette with a laser spot of approximately 100 μm using a lens with a 100-mm focal length. The peak laser intensity on the spot reached 3 × 109 W/cm2, which represented the average laser intensity during one pulse. To visualize the laser streaming phenomenon, 3-μm fluorescent polymer microspheres, which are sensitive to a wavelength of 620 nm, were added to the suspension. To excite the fluorescent polymer microspheres, a second 620-nm CW laser was defocused to generate a light sheet on the focal plane of the camera using a cylindrical lens. A high-speed camera with a maximum frame rate of 2000/s was used to capture the motion of the liquid.
Fig. 1. Schematic of the experiment setup. Dfg represents the distance between the focal lens and the inner surface of the glass cuvette.
3. Results and discussion
A laser (with a repetition rate of 10 kHz and laser power of 600 mW) was used to irradiate the cuvette filled with a gold nanoparticle suspension with a concentration of 4.3×1010 particles/mL. The distance between the focal lens and inner surface of the glass cuvette (abbreviated as Dfg in this manuscript) was set to 113 mm. Through repeated experiments, it was observed that laser streaming was mostly formed after approximately 60 s of laser irradiation. Here, the criteria for the formation of laser streaming were as follows. The fluorescent polymer microspheres were moved horizontally for at least 1 mm. Following laser irradiation, the cuvette inner surfaces were observed using an optical microscope. As shown in Fig. 2(a), there were no morphological changes (represented as the nanostructure in this study) on the glass when the laser irradiation time was less than 60 s. Subsequently, obvious nanostructure was found on the glass when the laser irradiation time was longer than 60 s, with heavier nanostructure observed at longer radiation times. Diameter of the nanostructure versus irradiation time is shown in Fig. 2(b). The nanostructure on the glass was found to be truly circular, and the maximum diameter was approximately 55 µm when the radiation time was longer than 120 s. The nanostructure observed with scanning electron microscopy (SEM) is shown in Fig. 2(c). In our opinion, the nanostructure is a cavity on the glass surface and it is decorated with many damaged gold nanostructures. The size of the nanostructure was smaller than that of the laser spot, indicating a laser intensity threshold for the formation of the damaged structure. The radiation time needed to form laser streaming (represented as the formation time in this study) and the diameter of the nanostructure exhibited the same trend. Therefore, this indicated that there was a clear relationship between the formation of laser streaming and the nanostructure on the glass.
Fig. 2. (a) The nanostructure on the inner surface of the cuvette and laser streaming after different laser irradiation times. (b) Diameter of the nanostructure versus irradiation time. (c) SEM image of the nanostructure.
It is evident that the formation of the nanostructure was related to the laser parameters, cuvette surface, and gold nanoparticles. The formation time versus the gold nanoparticle concentration is plotted in Fig. 3. The laser power was 600 mW, the laser repetition rate was 10 kHz, and the Dfg was fixed to 113 mm for Fig. 3. Less radiation time was required when the gold nanoparticle concentration was higher. Laser streaming was not observed for a concentration of 0.5×1010/mL after 60 min of laser irradiation, and it was assumed that the formation of laser streaming was not possible within a limited laser irradiation time when the concentration was lower than 1×1010/mL.
In this study, the formation times under different laser powers and repetition rates were measured, and the results are shown in Fig. 4. The Dfg was set to 107 mm and the gold nanoparticle concentration was set to 4.3×1010/mL. The formation time was shorter when the laser power was higher, and the formation time was not significantly related to the repetition rate but was mainly determined by the laser power. The shortest formation time of less than one second was achieved when the laser power was 3150 mW and the repetition rate was 10 kHz. The experimental data were fitted using the formula, ${T_{formation}} = 5.58 \times {10^6} \times {P_{laser}}^{ - 2}$, where ${T_{formation}}$ is the formation time in seconds and ${P_{laser}}$ is the laser power in mW.
Fig. 4. Formation time versus laser power, the formation time was mainly determined by the laser power rather than the repetition rate
The formation times under different Dfg are shown in Fig. 5. Here, the laser power of 600 mW and laser repetition rate of 10 kHz were used. The formation time curves under different gold nanoparticle concentrations exhibited a similar trend. The curve in Fig. 5 was divided into 4 areas, namely Area (a), Area (b), Area (c), and Area (d). In Area (a), laser streaming could not be formed after 3000 s of laser irradiation. In Area (b), the formation time decreased gradually with increasing Dfg. In Area (c), the formation time was very short and remained almost constant. In Area (d), the formation time increased gradually with Dfg, and subsequently, laser streaming could only be formed occasionally and was very unsteady. Finally, at very large Dfg, laser streaming could not be formed after 3000 s of laser irradiation.
Fig. 5. Formation time versus Dfg. The squares represent the experimental data corresponding to different gold nanoparticle concentrations. The hollow squares indicate that the magnitude of their error was clearly larger than that of the solid squares. The solid squares with hollow circle indicate that the laser streaming could not be formed after 3000 s of laser irradiation.The navy blue circles represent the calculated peak laser intensity on the inner surface of the cuvette. The pink triangles represent the laser intensity gradient close to the inner surface of the cuvette.
When the Dfg was changed, the laser intensity on the inner surface of the cuvette changed. The peak laser intensity on the inner surface of the cuvette was calculated and represented as a navy-blue circle in Fig. 5. Area (a) and Area (d) had almost the same peak laser intensity, as did Area (b) and Area (c). However, the formation time curves in these areas are clearly asymmetrical, which indicates that other factors related to the Dfg might greatly affect the formation time of laser streaming.
Prior to the formation of laser streaming, it was observed that some fluorescent polymer microspheres moved with high speed in a direction opposite to that of the laser. This provided a strong hint that the optical gradient force might play an important role in the formation of laser streaming. The optical gradient force was experienced by particles in the light gradient field, which tended to attract particles from the region with low intensity to that with high intensity [23]. Based on the optical gradient force, the phenomenon shown in Fig. 5 can be well explained and the formation mechanism of laser streaming can be further understood, as described in the next paragraph.
Both the laser intensity and nanoparticle concentration near the inner surface of the cuvette affected the formation time of laser streaming. In Fig. 5, the laser intensity was affected by the Dfg, while the nanoparticle concentration was determined by the liquid flow, particle diffusion, and transportation, which were greatly affected by the optical gradient force. The cases under different Dfg are shown in Fig. 6. Here, cases (a), (b), (c), and (d) correspond to Area (a), Area (b), Area (c), and Area (d) in Fig. 5, respectively. In cases (a) and (d), the laser intensity was lower at the inner surface, rendering it very difficult to form the nanostructure. With respect to the nanoparticle concentration near the inner surface, cases (a) and (d) were different. In case (a), the nanoparticles were transported off the inner surface of the cuvette, which made the formation of the nanostructure even more difficult, as shown in Area (a) of Fig. (5), where laser streaming could not be formed within 3000 s. However, in case (d), the nanoparticles were transported toward the inner surface of the cuvette, which benefitted the formation process, and it can be observed in Area (d) of Fig. (5) that laser streaming could possibly be formed within 3000 s. In both cases (b) and (c), the laser intensity was high and it was easy to form the nanostructure, but the nanoparticles were transported off the inner surface in case (b), while toward the inner surface in case (c). Therefore, the formation time was longer in case (b) than in case (c). In summary, by introducing the nanoparticle transportation effect under the optical gradient force, the phenomenon depicted in Fig. (5) could be explained clearly, and we believe that both the laser intensity and optical gradient force played an important role in the formation of laser streaming.
Fig. 6. Transportation of gold nanoparticles by the optical gradient force before the formation of laser streaming. Here, (a, b, c, d) corresponds to Area (a, b, c, d) in Fig. 5.
The streaming velocity is another key parameter of laser streaming. The streaming velocity was calculated by analyzing the trajectories of the fluorescent polymer microspheres in the video recorded by the high-speed camera. Here, the streaming velocity is the velocity of the polymer microspheres close to the inner surface of the cuvette (within the range of 1 mm), which was measured after the laser streaming reached the steady state. In this experiment, we selected three fastest fluorescent polymer microspheres, and the velocity is the average of the three. The streaming velocity under different Dfg at a laser power of 600 mW, repetition rate of 10 kHz, and gold nanoparticle concentration of 4.3×1010/mL is shown in Fig. 7(a), while the streaming velocity under different nanoparticle concentrations at a laser power of 600 mW, repetition rate of 10 kHz, and Dfg of 113 mm is shown in Fig. 7(b). The velocity reached a maximum value when Dfg was 107 mm, and decreased slightly when Dfg was within the range of 105 to 119 mm. When the Dfg was longer than 120 mm, steady laser streaming could not be formed, and the velocity reduced to zero as shown in Fig. 7(a). As shown in Fig. 7(b), the streaming velocity increased with the gold nanoparticle concentration.
The streaming velocity was measured under different laser powers and repetition rates and is shown in Fig. 8. As the output of our laser device could not maintain long enough to a power value between 1900–3150 mW, there were no data points shown in this range, but the trend in Fig. 8 was not changed. Here, the gold nanoparticle concentration was 4.3×1010/mL and the Dfg was 107 mm. The maximum streaming velocity of 43 cm/s was attained when the maximum available laser power of 3150 mW and repetition rate of 10 kHz were used, which was approximately one order of magnitude faster than the previously reported result of 4 cm/s [12]. No experimental data were available for a laser power of 3150 mW and repetition rate of 5 kHz because the pulse energy exceeded the laser damage threshold of the glass and broke the cuvette. In Fig. 8, it can be observed that the curves corresponding to the repetition rates of 5 kHz and 10 kHz are similar, indicating that the streaming velocity was mainly determined by the laser power and was barely affected by the repetition rate.
As shown in Fig. 8, the streaming velocity curve can be divided into three sections. When the laser power was <500 mW, the streaming velocity increased almost linearly with laser power. When the laser power ranged between 500 mW and 1900 mW, the streaming velocity hardly increased with laser power. When the power was higher than 1900 mW, the streaming velocity increased significantly with laser power. These three clear sections of the streaming velocity curve suggest that different mechanisms were involved in different sections. The snapshots of laser streaming under a laser power of 500 mW, 1200 mW, and 1900 mW are shown in Fig. 9. When the laser power was lower than 500 mW, the trajectories of the microspheres were straight lines and the flow type was laminar, as shown in Fig. 9(a). When the laser power was higher than 1900 mW, the trajectories of the microspheres were all curved lines, as shown in Fig. 9(c), and the flow type was turbulent.
Fig. 9. Snapshots of laser streaming under three laser power conditions, with the laser streaming flowing from right to left. The flow type changed from laminar flow (a) to transition flow (b) and finally to turbulent flow(c).
When the laser power was 1200 mW, the trajectories of the microspheres were a mix of straight and curved lines, as shown in Fig. 9(b), and the flow type should be a combination of laminar and turbulent flow (transition flow). This phenomenon suggests that the relationship between the laser streaming velocity and laser power had a strong correlation with the flow type, which can be summarized as follows: the streaming velocity increased almost linearly with laser power when laser streaming exhibited laminar flow. In addition, the streaming velocity hardly increased with laser power when laser streaming was in the transition range of laminar flow and turbulent flow, while it increased significantly with laser power when laser streaming was under turbulent flow.
Fig. 10. The relative energy conversion efficiency (${v_{streaming}}^2/{P_{laser}}$) curves at different laser powers and repetition rates.
The amount of laser energy that was transported to liquid streaming was too small to measure. Here, the relative energy conversion efficiency was calculated as ${v_{streaming}}^2/{P_{laser}}$. in arbitrary units to represent the amount of energy that was transferred from the laser to streaming. The relative energy conversion efficiency curves at different laser powers with different repetition rates are shown in Fig. 10. Similar to the streaming velocity, the relative energy conversion efficiency was affected by the type of flow: the relative energy conversion efficiency increased with laser power when the laser streaming was laminar or turbulent, but it decreased with laser power for transition flow. These rules were applicable to both laser repetition rates of 5 kHz and 10 kHz.
4. Conclusions
In this study, the formation mechanism of laser streaming was studied. A clear corresponding relationship between the formation of laser streaming and the morphological change on the glass was found. The formation of laser streaming was considerably easier at higher gold nanoparticle concentrations and laser power, and was hardly affected by the laser repetition rate. The shortest formation time of laser streaming was less than even 1 s, which suggested that the laser streaming nanostructure could be formed in-situ and prefabrication was not required in some specific cases. The optical gradient force was found to be an important approach involved in the transport and provision of nanoparticles during the formation of laser streaming. The formation of laser streaming was considerably easier and faster when the optical gradient force was towards the irradiated area. This finding improved our understanding of the formation mechanism of laser streaming.
The streaming velocity and relative energy conversion efficiency were mainly determined by the laser power rather than repetition rate. The streaming velocity and relative energy conversion efficiency both exhibited a strong correlation with the flow type. This correlation can be summarized as follows: both the streaming velocity and relative energy conversion efficiency increased linearly with the laser power for laminar flow and turbulent flow, while the streaming velocity hardly increased and the relative energy conversion efficiency decreased with laser power for transition flow. A maximum laser streaming velocity of 43 cm/s was achieved, which was approximately one order of magnitude faster than the previously reported result of 4 cm/s [12]. This can expand the application of laser streaming in fields where lasers could be used to move macroscopic objects at relatively high speeds. However, the performance of laser streaming under a wider range of laser parameters is still worth exploring.
Funding
National Natural Science Foundation of China (11575121); National Magnetic Confinement Fusion Program of China (2014GB125004); International Visiting Program for Excellent Young Scholars of Sichuan University.
Disclosures
The authors declare that there are no conflicts of interest related to this article.
References
1. J. S. Hangst, J. S. Nielsen, O. Poulsen, P. Shi, and J. P. Schiffer, “Laser cooling of a bunched beam in a synchrotron storage ring,” Phys. Rev. Lett. 74(22), 4432–4435 (1995). [CrossRef]
2. T. W. Hänsch and A. L. Schawlow, “Cooling of gases by laser radiation,” Opt. Commun. 13(1), 68–69 (1975). [CrossRef]
3. R. Kampmann, R. Kleindienst, A. Grewe, E. Burger, A. Oeder, and S. Sinzinger, “Optimized systems for energy efficient optical tweezing,” Proc Spie8637 (2013).
4. P. Kumari, J. A. Dharmadhikari, A. K. Dharmadhikari, H. Basu, S. Sharma, and D. Mathur, “Optical trapping in an absorbing medium: from optical tweezing to thermal tweezing,” Opt. Express 20(4), 4645–4652 (2012). [CrossRef]
5. R. M. Georgevic, “SOLAR RADIATION PRESSURE FORCE AND TORQUES MODEL,” Journal of the Astronautical Sciences 20(5), 257–274 (1973).
6. S. Tehranian, F. Giovane, J. Blum, Y. L. Xu, and B. A. S. Gustafson, “Photophoresis of micrometer-sized particles in the free-molecular regime,” Int. J. Heat Mass Transfer 44(9), 1649–1657 (2001). [CrossRef]
7. C. N. Baroud, M. R. de Saint Vincent, and J. P. Delville, “An optical toolbox for total control of droplet microfluidics,” Lab Chip 7(8), 1029–1033 (2007). [CrossRef]
8. E. Fradet, C. McDougall, P. Abbyad, R. Dangla, D. McGloin, and C. N. Baroud, “Combining rails and anchors with laser forcing for selective manipulation within 2D droplet arrays,” Lab Chip 11(24), 4228–4234 (2011). [CrossRef]
9. K. T. Kotz, K. A. Noble, and G. W. Faris, “Optical microfluidics,” Appl. Phys. Lett. 85(13), 2658–2660 (2004). [CrossRef]
10. A. Diguet, H. Li, N. Queyriaux, Y. Chen, and D. Baigl, “Photoreversible fragmentation of a liquid interface for micro-droplet generation by light actuation,” Lab Chip 11(16), 2666–2669 (2011). [CrossRef]
11. N. Kavokine, M. Anyfantakis, M. Morel, S. Rudiuk, T. Bickel, and D. Baigl, “Light-Driven Transport of a Liquid Marble with and against Surface Flows,” Angew. Chem., Int. Ed. 55(37), 11183–11187 (2016). [CrossRef]
12. Y. N. Wang, Q. H. Zhang, Z. A. Zhu, F. Lin, J. D. Deng, G. Ku, S. C. Dong, S. Song, M. K. Alam, D. Liu, Z. M. Wang, and J. M. Bao, “Laser streaming: Turning a laser beam into a flow of liquid,” Sci. Adv. 3(9), e1700555 (2017). [CrossRef]
13. S. L. Chen, Y. C. Chang, C. Zhang, J. G. Ok, T. Ling, M. T. Mihnev, T. B. Norris, and L. J. Guo, “Efficient real-time detection of terahertz pulse radiation based on photoacoustic conversion by carbon nanotube nanocomposite,” Nat. Photonics 8(7), 537–542 (2014). [CrossRef]
14. Y. Hou, S. Ashkenazi, S. W. Huang, and M. O’Donnell, “Improvements in optical generation of high-frequency ultrasound,” IEEE Trans. Sonics Ultrason. 54(3), 682–686 (2007). [CrossRef]
15. Y. Hou, J. S. Kim, S. Ashkenazi, S. W. Huang, L. J. Guo, and M. O’Donnell, “Broadband all-optical ultrasound transducers,” Appl. Phys. Lett. 91(7), 73507 (2007). [CrossRef]
16. H. Y. Ju, R. A. Roy, and T. W. Murray, “Gold nanoparticle targeted photoacoustic cavitation for potential deep tissue imaging and therapy,” Biomed. Opt. Express 4(1), 66–76 (2013). [CrossRef]
17. F. C. P. Masim, H. L. Liu, M. Porta, T. Yonezawa, A. Balcytis, S. Juodkazis, W. H. Hsu, and K. Hatanaka, “Enhanced photoacoustics from gold nano-colloidal suspensions under femtosecond laser excitation,” Opt. Express 24(13), 14781–14792 (2016). [CrossRef]
18. X. M. Yang, E. W. Stein, S. Ashkenazi, and L. H. V. Wang, “Nanoparticles for photoacoustic imaging,” Wiley Interdiscip. Rev.: Nanomed. Nanobiotechnol. 1(4), 360–368 (2009). [CrossRef]
19. B. L. Zhang, C. Y. Fang, C. C. Chang, R. Peterson, S. Maswadi, R. D. Glickman, H. C. Chang, and J. Y. Ye, “Photoacoustic emission from fluorescent nanodiamonds enhanced with gold nanoparticles,” Biomed. Opt. Express 3(7), 1662–1669 (2012). [CrossRef]
20. J. Lighthill, “Acoustic Streaming,” J Sound Vib 61(3), 391–418 (1978). [CrossRef]
21. B. Moudjed, V. Botton, D. Henry, H. Ben Hadid, and J. P. Garandet, “Scaling and dimensional analysis of acoustic streaming jets,” Phys. Fluids 26(9), 093602 (2014). [CrossRef]
22. B. Moudjed, V. Botton, D. Henry, S. Millet, J. P. Garandet, and H. Ben Hadid, “Near-field acoustic streaming jet,” Phys. Rev. E 91(3), 033011 (2015). [CrossRef]
23. A. Ashkin, J. M. Dziedzic, J. E. Bjorkholm, and S. Chu, “Observation of a Single-Beam Gradient Force Optical Trap for Dielectric Particles,” Opt. Lett. 11(5), 288–290 (1986). [CrossRef]
