Open Access
Add to BibSonomy
Abstract
The exploration relevant to the surface changes on optical micro- and nanofibers (MNFs) is still in infancy, and the reported original mechanisms remain long-standing puzzles. Here, by recognizing the combined interactions between fiber heating, mechanically tapering, and high-power pulsed laser guiding processes in MNFs, we establish a general thermal-mechanical-photo-activation mechanism that can explain the surface changes on MNFs. Our proposed activation mechanism can be well supported by the systematical experimental results using high-intensity nanosecond/femtosecond pulsed lasers. Especially we find large bump-like nanoscale cavities on the fracture ends of thin MNFs. Theoretically, on the basis of greatly increased bond energy activated by the fiber heating and mechanically tapering processes, the energy needed to break the silicon-oxygen bond into dangling bonds is significantly reduced from its intrinsic bandgap of ∼9 eV to as low as ∼4.0 eV, thus high-power pulsed lasers with much smaller photon energy can induce obvious surface changes on MNFs via multi-photon absorption. Finally, we demonstrate that using surfactants can repair the MNF surfaces and exploit them in promising applications ranging from sensing and optoelectronics to nonlinear optics. Our results pave the way for future preventing the performances from degradation and enabling the practical MNF-based device applications.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Silica optical fibers are usually composed of amorphous silica and have been widely used in various applications, such as fiber-optic communications and optoelectronics. As a combination of fiber optics and nanotechnology, optical microfiber and nanofibers (MNFs) have been emerging as a novel platform for manipulating light on the wavelength or subwavelength scale, benefitting from the attractive advantages such as high fractional evanescent fields and large tailorable waveguide dispersion [1–4]. MNFs are usually fabricated from standard optical fibers using heating and mechanically stretching methods, and according to the profiles, MNFs can be divided into biconical and single conical MNFs. Biconical MNFs are comprised of a small uniform diameter region (waist region) with each extremity connected to standard fiber by two conical transition regions, and have been widely used in various applications such as microcavity coupling [5], microfiber grating sensing [6,7], and ultrafast laser generation [8,9]. Single conical MNFs can be obtained by breaking the biconical MNFs during the heating and stretching process, and have been usually used for probe sensing [10], and coupling light into functional micro/nanoscale waveguides such as metal and semiconductor nanowires for sensing, nonlinear optics, and micromanipulation, with the attractive advantages of high coupling efficiency and broad spectral regions [11–14].
To date, most of the related research of MNF optics focuses on developing devices with new structures and new functions, and improving certain performance parameters of devices [1–4]. However, less attention has been given to surface changes on MNFs during their guiding of lasers, especially high-density ultrashort pulsed lasers. At high power density, e.g., up to 1013 W cm–2, even conventional bulk silica will suffer from an optically induced breakdown [15–17]. In MNF applications such as nonlinear optics and fiber laser generation [8,9,18–24], high-density pump lasers are usually needed, and due to the small diameters, the confined field intensity in MNFs will be significantly condensed through the tapering region. For reference, for an MNF with a diameter of 1.0 µm and guiding a nanosecond laser (1 ns pulse width) with pulse energy of 1 µJ, the confined density of peak power around the MNF surface can be up to 1012 W cm–2, which nears to the breakdown threshold of fused silica. This may induce surface changes even damages such as surface dangling bonds and degradation, and hinder the practical performances of MNF-based devices such as scattering losses and working life [25]. Therefore, investigating the surface effects on MNFs and finding proper approaches to repair the surface defects is one of the foremost challenges that needs to be overcome to enable their practical performances and commercial applications. Nevertheless, the exploration relevant to pulsed laser-induced surface changes on MNFs, including experimental and theoretical studies, is still in infancy. Even if some results reported strange bump-like structures on MNF surfaces, the original and dynamical mechanisms of this effect remain long-standing puzzles [26,27].
The main obstacle to understanding this issue lies in a long-neglected matter, that is, the effect of the huge diameter deformation of silica fibers (decreased from 125 µm to ∼1 µm) during the fiber heating and tapering processes on the silicon-oxygen (Si−O) bond energy. Herein, for the first time, we establish a thermal-mechanical-photo-activation mechanism to explain the surface changes on MNFs induced by high-intensity pulsed lasers. Experimentally, from systematical investigation by using high-intensity nanosecond/femtosecond pulsed lasers, we revealed that the surface roughness of photoactivated MNFs obviously increased, with a growth dynamic time of about 40 min. Especially we found large bump-like nanoscale cavities on the fracture ends of thin MNFs. Theoretically, on the basis of greatly increased bond energy activated by the fiber heating and mechanically tapering processes, the energy needed to break the silicon-oxygen bond into dangling bonds is significantly reduced from its intrinsic bandgap of ∼9 eV to as low as ∼4.0 eV, thus high-power pulsed lasers with much smaller photon energy (such as at 1064 nm wavelength) can induce obvious surface changes on MNFs via multi-photon absorption. We further discovered that the photoactivated MNFs were easy to adsorb impurities from outside and had larger scattering loss. But we also demonstrated that using surfactants could repair the MNF surfaces and may exploit them in potential applications.
2. Materials and methods
2.1 Fabrication and activation of MNFs
To investigate the MNF surfaces, several nanosecond and femtosecond pulsed laser sources were used in this work. The nanosecond pulsed laser had a wavelength of 1064 nm (MPL-III-1064, Changchun New Industries Optoelectronics Tech. Co., Ltd.), pulse duration of 8.5 ns, and a repetition rate of 4 kHz. By using a frequency-doubling KTP crystal, a 532-nm-wavelength pulsed laser was generated from the 1064-nm-wavelength nanosecond laser. The femtosecond laser had a wavelength of 519 nm (FemtoYL-50, Wuhan Yangtze Soton Laser Co., Ltd.), pulse duration of ∼400 fs, and a repetition rate of 800 kHz.
MNFs were fabricated by using a flame-heated method with uniform waist diameters (Dfiber) ranging from 200 nm to 4.0 µm [13,25]. The excitation pulse lasers were first lens-coupled into standard silica fibers (SMF-28, Corning) and then squeezed into tapered MNFs to activate their surfaces. As shown in Fig. 1(a), the uniform waist diameter region of as-photoactivated MNF was placed on an MgF2 substrate that was coated with a 1.0-µm-thickness polydimethylsiloxane (PDMS) film, and the tapering transition region was fixed on another MgF2 substrate by a low refractive index UV curing adhesive (EFIRON PC-373, Luvantix Co. Ltd.). Here the thin PDMS film had a strong van der Waals force and was used to adsorb the MNFs on the substrate, preventing their movement during the atomic force microscope (AFM) scanning process. The average power of the pulsed lasers launched into the MNFs for photoactivation was on the order of tens of milliwatts. Due to the substrate effect and scattering effect, the transmission loss from the input ends of standard fibers to the tapered tips of MNFs is typically around 10 dB. The whole systems of fabrication and photoactivation, as well as the characterization shown below, were all located in a Class 100,000 cleanroom (ISO 8).
Fig. 1. Experimental illustration of surface roughness measurement on an original tapered MNF. (a) Experimental setup of the surface roughness measurement, which consists of an original MNF placed on an MgF2 substrate with a 1.0-µm-thickness PDMS film. The tapering transition region of the MNF was bonded and fixed by a UV curing adhesive. (b) Long-term aging test of the surface roughness results from a 3.1-µm-diameter MNF with 2-hours repeated measurements (upper) and another 3.4-µm-diameter MNF with 31-hour repeated measurements (lower).
2.2 High-resolution AFM characterization
The surfaces of as-fabricated MNFs were first characterized by a high-resolution AFM (Cypher S, Oxford Instruments Asylum Research, with the noise level less than 15 pm), by using the tapping mode of the AFM system. To obtain a high-resolution root mean square (RMS) value of the surface roughness (Rq), a fast-scanning speed of 10 Hz with a small scanning range of 50 nm × 50 nm was used. Using such fast-scanning speed and small scanning area can greatly minimize the thermal drift and quickly achieve equilibrium. While to obtain the surface morphologies of the MNFs, a slow-scanning speed of 1 Hz with a large scanning range (∼1.0 µm × 1.0 µm) was used.
2.3 Optical characterization
To show the defect types on MNF surfaces, photoluminescence (PL) spectra were used to characterize as-photoactivated MNFs. A 532-nm-wavelength CW laser as the excitation light source (MSL-FN-532, Changchun New Industries Optoelectronics Technology Co., Ltd.) was lens-coupled into the suspended MNFs to excite the surface defects, with average power of 10.0 mW, which was much lower than the normally-used peak power in the photoactivation process (ranging from 200 W to 150 kW) and did not cause detectable surface damage. The PL emissions from the MNF surfaces were collected using a 100× microscope objective and then sent to a spectrometer (IsoPlane SCT-320, Princeton Instruments) for spectral analysis [25]. Optical scattering effect was also used to show the adsorption effect on MNF surfaces. MNFs were placed in an indoor open-air environment or a small sealed plastic chamber, with a glass window for imaging by using a CCD camera (DS-Ri1, Nikon). The used sealed chamber has a volume of ∼1.0 mL and can greatly decrease the adsorption of tiny substances on the MNF surfaces from the open air. The 532-nm-wavelength CW laser with a power of 100 µW was launched into the MNFs as the scattering light source.
3. Surface changes on MNFs
For reference, the surface roughness of the originally fabricated MNFs before photoactivation was first characterized. An original MNF with Dfiber = 3.1 µm was directly placed in the high-resolution AFM and characterized using the fast-tapping mode. The upper of Fig. 1(b) shows that the surface roughness Rq of the MNF maintained stable during the 2-hours test time. The measured average value of Rq is ∼137 pm while the baseline noise of the AFM system is ∼15 pm, indicating that the used AFM characterization method has a small error and high repeatability. The long-term aging test of as-fabricated MNF surfaces was also characterized. The lower of Fig. 1(b) shows that the surface roughness of another MNF (Dfiber = 3.4 µm) only decreases slightly within 31 hours, indicating that the surfaces of original MNFs show a certain degree of stability and the aging-induced fluctuation of Rq during the short time of each AFM measurement (about 10 min) can be neglected.
The original MNFs were first activated by the 532- and l064-nm-wavelength nanosecond pulsed lasers and the surface morphologies of the MNFs were characterized by the AFM with a scanning range of 1.0 µm × 0.5 µm. As shown in Fig. 2(a), after the 532-nm-wavelength nanosecond laser with average power of 8.4 mW was launched into an original MNF (Dfiber = 1.8 µm) for 60 minutes, the surface became obvious uneven. Figure 2(b) illustrates the corresponding surface roughness Rq of the same cross-section segment A–B (red dashed lines) on the MNF surface before (Rq = 135 pm) and after (Rq = 372 pm) photoactivation in Fig. 2(a), in which the measured Rq difference was 237 pm. Figure 2(c) and 2(d) show that an original MNF (Dfiber = 1.8 µm) was activated by the 1064-nm-wavelength nanosecond laser with average power of 22.0 mW for 60 minutes, in which the surface also became obvious uneven with Rq increasing from 247 pm to 353 pm, and the measured Rq difference was 106 pm. In addition, the surface morphologies of the activated MNFs hardly changed, indicating that the surface changes experienced were irreversible.
Fig. 2. Surface variation under nanosecond-laser photoactivation. (a) AFM morphologies of an MNF (Dfiber = 1.8 µm) before and after 532-nm-wavelength nanosecond laser photoactivation, and (b) the corresponding surface roughness Rq denoted in the A‒B cross-section. (c) AFM morphologies of an MNF (Dfiber = 1.8 µm) before and after 1064-nm-wavelength nanosecond laser photoactivation. (d) Corresponding surface roughness Rq denoted in the A‒B cross-section.
A 519-nm-wavelength femtosecond laser with average power of 8.0 mW was also used to activate another MNF (Dfiber = 1.6 µm). Under the same scanning range, the surface roughness Rq increased from 218 pm to 412 pm, with a measured Rq difference of 194 pm after being photoactivated for only 45 minutes, which is roughly consistent with the experimental results by using nanosecond lasers above. Through dozens of experimental samples, it was found that as the MNF diameters increased, the values of Rq became smaller, which may owe to the decreased confined energy density in large-diameter MNFs, and will be discussed in detail later.
To study the dynamic process of surface changes on the activated MNFs, we measured the in-situ morphologies of MNFs after different activation time. A 3.1-µm-diameter MNF was activated by a 1064-nm-wavelength nanosecond pulsed laser with average power of 22.2 mW, and after the pulsed laser was turned off, the surface topography measurements were performed immediately. As shown in Fig. 3(a)‒3(d) at the activation time of 13, 47 and 68 min, respectively, it was found that some bumps, which are also commonly called nanocavities [26–28], grew over time and gradually merged into larger cavities finally. By discriminating the phase difference on the MNF surface (threshold: 131 deg), the area of these nanocavities can be derived. Figure 3(e) summarizes the time evolution process of cavity area on the MNF surface, which exhibits an exponential growth behavior and eventually reaches saturation, with a calculated growth dynamic time of about 42 min. And as far as we know, this is the first time to disclose the growth dynamics of nanoscale cavities on MNF surfaces.
Fig. 3. Dynamic growth process of nanoscale cavities on the photoactivated MNF surface. (Upper) 3D-view AFM morphologies and (lower) the corresponding phase transition diagrams of the 3.1-µm-diameter MNF of (a) before and after (b) 13 min, (c) 47 min, and (d) 68 min photoactivation by a 1064-nm-wavelength nanosecond pulsed laser. (e) Time evolution process of cavity area on the photoactivated MNF surface.
In addition, the surface morphologies and structural reconstruction at the fracture point are different from those at uniform tapering regions because exceptional plastic deformation occurs here. In the heating and stretching process, a standard fiber first underwent plastic deformation under large strain, and then its deformation gradually changed from size-dependent strengthening to brittle-to-ductile transition. After exceptional plastic deformation occurs, the MNF finally broke [29,30]. Experimentally, the highest Rq value was always measured around the fracture point, and even serious surface damage was found. For example, Fig. 4(a) and 4(b) show two AFM images of an MNF (Dfiber = 780 nm) activated by the 519-nm-wavelength femtosecond laser with average power of 8.4 mW for 75 minutes. Here at the fracture region, a necking effect of the diameter reduced from 780 nm to only 550 nm is observed. More obviously, there are many large nanoscale cavities at the fracture region (denoted as End region in Fig. 4(b)), compared to very small protrusions in the uniform diameter region (denoted as Uniform region in Fig. 4(a).
Fig. 4. (Upper) 3D-view AFM morphologies and (lower) the corresponding phase transition diagrams of the MNF. Surface variation of (a) the uniform region (Dfiber = 780 nm) and (b) the fracture region after 75-minutes photoactivated MNF by a 519-nm-wavelength femtosecond laser. The fracture region shows apparent nanoscale cavities on the surface. (c) Microscope image of the used MNF and its PL spectra from the uniform diameter region and the fracture region of the same 1.0-µm-diameter MNF photoactivated by 519-nm-wavelength femtosecond laser, in which a strong peak with a center position of 1.93 eV can be observed.
To further prove the defects on MNF surfaces, the PL spectra were collected in Fig. 4(c). Here, an original MNF (Dfiber = 1.0 µm) was used, and after photoactivation by the 519-nm-wavelength femtosecond laser with average power of 10.0 mW for 30 minutes, the PL spectra from the uniform region and the fracture region were collected respectively. Obviously, compared with the uniform region, the PL signal from the fracture region has a strong peak with a center position of 1.93 eV, which agrees with the reported PL position of oxygen dangling bonds [31], and also further indicates the exceptional surface fracture induced by plastic deformation at the crack tip of photoactivated MNFs.
4. Photoactivation principle
According to our previous work [25], the increased surface roughness originates from the dangling bonds induced by the photoactivation effect on the MNF surface. In amorphous silica, higher-membered siloxane rings (≡Si−O−Si≡, usually 6-membered rings) are dominant surface structures [31,32]. The bond angle of the 6-membered ring is in the range of 140 to 150 degrees, and its bond length is about 1.64 Å, in which each atom in the ring structures is almost relaxed and does not contain strained energy. As illustrated in Fig. 5(a), there are two ways to obtain oxygen dangling bonds from the 6-membered rings on the fiber surface. The first way is directly breaking Si−O bonds in 6-membered rings to generate oxygen dangling bonds. The energy needed is equivalent to the bandgap of silica, ∼9 eV, which is too high for the laser photon energy used here.
Fig. 5. Schematic diagrams of combined effects of fiber heating, mechanically tapering, and the pulsed laser guiding processes on structural changes of siloxane rings. (a) Heating and mechanically stretching processes break the 6-membered rings into highly strained 3-membered rings, and greatly reduce the activation energy of siloxane bonds. Thus, in contrast to the ∼9 eV bandgap value of silica, using a laser with photon energy in the range of ∼4−8eV can break the highly strained 3-membered rings and generate oxygen dangling bonds. (b) Schematic diagram of heating and mechanically stretching process of a standard fiber. (c) Ball and stick model of 6-membered rings, 3-membered rings, and oxygen dangling bonds which are generated by the fiber large-size deformation and photoactivation process.
Another way is producing surface low-membered rings (usually 3-membered rings with a bond angle of ∼132 degrees) as an intermediate transition to generate dangling bonds. During the fiber heating and mechanically stretching process, near the breaking point of the MNF is the place where the fiber tensile deformation is greatest and the density of dangling bonds is also the largest. In our experiment, large-scale deformation can split the 6-membered rings into highly strained 3-membered rings during the fiber heating and mechanically stretching process, and the diameter of optical fiber is decreased from 125 µm to ∼1 µm (Fig. 5(b)). The strained energy in a 3-membered ring is calculated to be 0.26 eV per bond (Fig. 5(c)) [33]. Highly strained 3-membered rings can be easily excited to release high-density oxygen dangling bonds on the MNF surface by pulsed lasers, and greatly lower the activation energy from siloxane bonds. To completely dissociate an oxygen atom (destroy two Si−O bonds at the same time) in the 3-membered rings, ∼8.0 eV energy is needed according to our calculation. But if only one Si−O bond is broken while the other bond retains connected, only 4.0 eV energy is required. So the energy of generating an oxygen dangling bond in a 3-membered ring is in the range of 4.0 eV to 8.0 eV, which is much lower than the energy required in the first way, as illustrated in Fig. 5(c).
In our experiment, for pulsed lasers with wavelengths of 519 (2.39 eV), 532 (2.33 eV), and 1064 (1.17 eV) nm used, the energy of the single-photon absorption is not enough to break the Si−O bond of the 3-membered rings (4‒8 eV), but this process can be compensated by the multi-photon absorption effect. Although the multiphoton absorption coefficient in the standard fiber is relatively weak (e.g., on orders of ∼10−11 m/W for 532 nm light) [34], there have been references showing that the surfaces of conventional bulky amorphous silica can be direct excited by ultra-short pulsed lasers at near-infrared bands (800 or 1064 nm) to generate measurable oxygen dangling bonds, and yet their excitation intensity is very high (usually above 1013 W cm−2) [15–17]. Fortunately, here the small diameters of our tapered MNFs significantly condense the confined light field intensity, and greatly enhance the multiphoton absorption effect, allowing us to use the relatively low-intensity laser to achieve the dangling bonds on the MNF surface. Furthermore, due to the high contrast of refractive indices between the surrounding air and MNFs, the guided light intensity will be greatly enhanced around the MNF surfaces, and the MNFs with smaller diameters will have larger enhancement [35]. Figure 6(a) and 6(b) show the simulated electric field distributions of MNFs with different diameters, in which under the same input laser condition, the calculated energy intensity at the 1.0-µm-diameter MNF surface is larger than that at 3.0-µm-diameter MNF. As the MNF diameters decrease, although the confined energy density in the core decreases, the electric field on the surface is enhanced, which is helpful for inducing more obvious surface effects.
Fig. 6. Simulation of electric field distributions for MNFs with diameters of (a) 3.1 µm and (b) 1.0 µm. Peak field energy on the (left) cross-section and (right) the y-axial direction of the cross-section (the fundamental mode). The inset on the right is an enlarged view at the two MNF surfaces along the y-axial direction.
For the experiments in Fig. 3, the peak power of the 1064-nm-wavelength nanosecond laser input in the standard fiber is 653.7 W, and combined with energy leakage, the peak power density at the MNF surface is calculated to be ∼2.6 × 108 W cm−2. While in Fig. 4, the peak power of the femtosecond laser input in the standard fiber is as high as 26.3 kW, and the peak power density at the MNF surface is up to ∼5.9 × 1010 W cm−2. Since multiphoton absorption is more likely to occur in MNFs with higher surface peak energies, also single photon energy is higher in Fig. 4, the shapes of the photoactivated nanoscale cavities are more obvious in Fig. 4 than in Fig. 3. In addition, it should be noted that in the fraction region of the MNF in Fig. 4(b), the larger deformation makes Si‒O bonds of 3-membered rings store a large amount of tensile strain energy, and thus can be easily excited by pulsed lasers to generate larger nanoscale cavities compared with the uniform region. This makes the fraction point unique.
5. Reparation and utilization
The photoactivation effect induced by the high-intensity pulsed laser may bring unfavorable factors and reduce the performance of the device. The excited dangling bonds are very active and are easy to adsorb tiny substances from outside. Here, scattering characterizations of MNFs were carried out to show the adsorption effect on the MNF surfaces. An original MNF was placed in a small sealed chamber and characterized straightly for the first time; then, it was exposed to open air for 20 minutes and characterized for the second time; subsequently, it was placed in the sealed chamber again and activated by the 519-nm-wavelength femtosecond laser with average power of 50.0 mW for 30 minutes, and characterized for the third time; finally, it was exposed to air for 20 minutes again and characterized for the last time. Figure 7(a) shows the scattering microscope images of an MNF (Dfiber = 890 nm) under different atmosphere conditions and Fig. 7(b) shows the corresponding scattering light intensity distribution along its length direction. Neither the original MNFs nor the photoactivated MNF in the sealed chamber could produce scattered light spots, only the photoactivated MNF after exposed to the open-air environment produced obvious scattered light spots. The calculated loss of photoactivated MNF was about 0.04 dB/mm in the sealed chamber, while it greatly increased to 0.16 dB/mm after exposed to open air for 20 minutes. This can be attributed to the generated high-density active dangling bonds that absorb large numbers of molecular clusters or microparticles from the open-air environment.
Fig. 7. Optical scattering characteristics of an 890-nm-diameter MNF before and after femtosecond laser photoactivation. A 532-nm-wavelength CW laser was used as the scattering denote laser. (a) Optical micrographs and (b) the corresponding scattering intensity curves of the MNF under conditions of original/sealed, original/open air, photoactivated/sealed, and photoactivated/open air states. The characteristic interval between each state was ∼20 minutes.
Since the dangling bonds are so active, it is necessary to passivate them and decrease the adsorption effect for practical applications; on the other hand, we can also exploit them for some useful applications. Blocking dangling bonds is the most direct method. Here, nonionic polyvinylpyrrolidone (PVP) aqueous solution and negatively charged poly (sodium 4-styrenesulfonate) (PSS) aqueous solution which are commonly used in biosensing applications [3,14,36], were used to passivate and modify the as-photoactivated MNF surfaces. Two MNFs with diameters approximately 380 nm were photoactivated by the 519-nm-wavelength femtosecond laser with average power of 50.0 mW for 10 minutes, then immersed in the PVP and PSS aqueous solution (0.01 wt %) for 10 seconds. After that, the MNFs were first placed in a sealed chamber for initial scattering characterization, and then exposed to an open-air environment for 20 minutes before the final scattering characterization.
As shown in Figs. 8(a) and 8(b), compared with the photoactivated sample placed in the sealed chamber, after the photoactivated MNF was modified with PVP and PSS surfactant, the light scattering intensity distribution hardly changed after exposed to the open-air environment for a long time. It can be inferred that the MNF no longer absorbs tiny substances after treated by surfactants and the surfactants have indeed repaired the oxygen dangling bonds on the MNF surfaces. Figures 8(c) and 8(d) illustrate the combination schematic of surfactants with the generated dangling atoms from photoactivated MNFs in an aqueous environment. In aqueous solutions, PVP and PSS molecules can form moderately strong hydrogen bonds with hydroxy groups (‒OH), which come from water or the MNF surface [37]. The oxygen dangling bonds from as-photoactivated MNFs will produce a certain number of hydroxy groups, so the PVP and PSS molecules are easy to be absorbed through hydrogen bonds. Similarly, the silicon dangling bonds from photoactivated MNFs can form silanol groups (≡Si‒OH) [38], which also contain hydroxy groups.
Fig. 8. Reparation and utilization of surface effect on photoactivated MNF surfaces. Scattering intensity distributions on two photoactivated MNFs with the similar diameter around 380 nm, after being modified by (a) a 0.01% PSS aqueous solution and (b) a 0.01% PVP aqueous solution, respectively. A 532-nm-wavelength CW laser was used to denote their scattering effects in sealed chamber or open-air environments. Insets show their optical images of scattering effects on the MNFs. Schematic of generated oxygen dangling atoms binding with (c) PVP or (d) PSS molecules in water. (e) Surface roughness measurement without a photoactivation condition including original samples (black histogram) and PSS-modified original samples (blue histogram), and with a photoactivation condition including PSS-modified samples (green histogram).
To further investigate the photoactivation effect on the adsorption of surfactants on MNF surfaces, the surface roughness of a quartz plane and several MNFs with different diameters were measured. Under the condition without photoactivation, the original surface roughness of a fresh quartz plane and an as-drawn MNF (Dfiber = 1.6 µm) was first measured, and then after the quartz plane and the MNF were immersed in the PSS solution for 10 seconds, the surface roughness was measured for the second time. The results without photoactivation are shown in Fig. 8(e), in which the roughness of the quartz planes and the MNFs exhibits little difference between the original value and the PSS-modified value (around 170 pm level). Under the condition with photoactivation by the 519-nm-wavelength femtosecond laser with average power of 30.0 mW for 10 minutes, all the roughness of the PSS-modified MNFs were above 300 pm, as shown in Fig. 8(e). This set of comparative experiments clearly demonstrates that photoactivation is an essential prerequisite for surfactant adsorption onto the MNFs. It is also worth pointing out that although the surface roughness of surfactant-modified MNFs after photoactivation (∼300 pm) becomes slightly larger, it is still within an acceptable level for the conventional micro/nanowaveguides [2].
6. Conclusion
In summary, we systematically investigated the surface changes of tapered MNFs under the excitation of high-power pulsed lasers. In particular, large bump-like nanoscale cavities or even damaged structures were observed on the fracture ends of thin MNFs, due to the large diameter deformation and the surface enhancement of light confinement. Our proposed thermal-mechanical-photo-activation principle can well explain the observed surface changes on MNFs, by combining the interactions between fiber heating, mechanically tapering, and pulsed laser guiding processes. The generated high-density dangling bonds were so active that can easily adsorb impurities from outside and induce large scattering loss. We demonstrated that using surfactants can passivate the high-density dangling bonds and prevent the performances of MNF-based devices from degradation. On the other hand, we can also exploit photoactivation-induced high-density dangling bonds for some useful applications. Except for the chemical and biosensing, the high-density dangling bonds can be used to tune the optoelectrical properties of atomically thin 2-D semiconductor layers such as the photoluminescence and carrier mobility [25]. Also, the high-density dangling bonds can be used as a graft bridge for functional organic nonlinear molecules [39]. Therefore, our work not only gives in-depth insight of the thermal-mechanical-photo-activation effect on Si‒O bond energy on MNFs, but also inspires new strategies for preventing degradation of MNFs, and designing high-performance of MNF-based devices.
Funding
Natural Science Foundation of Shanghai (21ZR1481100); National Natural Science Foundation of China (62075131, 62122054).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. R. Ismaeel, T. Lee, M. Ding, M. Belal, and G. Brambilla, “Optical microfiber passive components,” Laser Photonics Rev. 7(3), 350–384 (2013). [CrossRef]
2. X. Wu and L. Tong, “Optical microfibers and nanofibers,” Nanophotonics 2(5-6), 407–428 (2013). [CrossRef]
3. Y. Li, Z. Xu, S. Tan, F. Fang, L. Yang, B. Yuan, and Q. Sun, “Recent advances in microfiber sensors for highly sensitive biochemical detection,” J. Phys. D: Appl. Phys. 52(49), 493002 (2019). [CrossRef]
4. W. Luo, Y. Chen, and F. Xu, “Recent progress in microfiber-optic sensors,” Photonic Sens. 11(1), 45–68 (2021). [CrossRef]
5. X. Jiang, A. Qavi, S. H. Huang, and L. Yang, “Whispering-gallery sensors,” Matter 3(2), 371–392 (2020). [CrossRef]
6. Y. Zhang, B. Lin, S. C. Tjin, H. Zhang, G. Wang, P. Shum, and X. Zhang, “Refractive index sensing based on higher-order mode reflection of a microfiber Bragg grating,” Opt. Express 18(25), 26345–26350 (2010). [CrossRef]
7. Y. Wu, B. Yao, A. Zhang, Y. Rao, Z. Wang, Y. Cheng, Y. Gong, W. Zhang, Y. Chen, and K. S. Chiang, “Graphene-coated microfiber Bragg grating for high-sensitivity gas sensing,” Opt. Lett. 39(5), 1235–1237 (2014). [CrossRef]
8. Y. Li, L. Wang, L. Li, and L. Tong, “Optical microfiber-based ultrafast fiber lasers,” Appl. Phys. B 125(10), 192 (2019). [CrossRef]
9. Z. C. Luo, M. Liu, Z. N. Guo, X. F. Jiang, A. P. Luo, C. J. Zhao, X. F. Yu, W. C. Xu, and H. Zhang, “Microfiber-based few-layer black phosphorus saturable absorber for ultra-fast fiber laser,” Opt. Express 23(15), 20030–20039 (2015). [CrossRef]
10. R. Pennetta, S. Xie, and P. S. J. Russell, “Tapered glass-fiber microspike: high-Q flexural wave resonator and optically driven knudsen pump,” Phys. Rev. Lett. 117(27), 273901 (2016). [CrossRef]
11. F. Gu, L. Zhang, G. Wu, Y. Zhu, and H. Zeng, “Sub-bandgap transverse frequency conversion in semiconductor nano-waveguides,” Nanoscale 6(21), 12371–12375 (2014). [CrossRef]
12. S. Kim, N. Yu, X. Ma, Y. Zhu, Q. Liu, M. Liu, and R. Yan, “High external-efficiency nanofocusing for lens-free near-field optical nanoscopy,” Nat. Photonics 13(9), 636–643 (2019). [CrossRef]
13. S. Linghu, Z. Gu, J. Lu, W. Fang, Z. Yang, H. Yu, Z. Li, R. Zhu, J. Peng, Q. Zhan, S. Zhuang, M. Gu, and F. Gu, “Plasmon-driven nanowire actuators for on-chip manipulation,” Nat. Commun. 12(1), 385 (2021). [CrossRef]
14. T. Pan, D. Lu, H. Xin, and B. Li, “Biophotonic probes for bio-detection and imaging,” Light: Sci. Appl. 10(1), 124 (2021). [CrossRef]
15. D. Du, X. Liu, G. Korn, J. Squier, and G. Mourou, “Laser-induced breakdown by impact ionization in SiO2 with pulse widths from 7 ns to 150 fs,” Appl. Phys. Lett. 64(23), 3071–3073 (1994). [CrossRef]
16. B. C. Stuart, M. D. Feit, A. M. Rubenchik, B.W. Shore, and M. D. Perry, “Laser-induced damage in dielectrics with nanosecond to subpicosecond pulses,” Phys. Rev. Lett. 74(12), 2248–2251 (1995). [CrossRef]
17. A. Zoubir, C. Rivero, R. Grodsky, K. Richardson, M. Richardson, T. Cardinal, and M. M. Couzi, “Laser-induced defects in fused silica by femtosecond IR irradiation,” Phys. Rev. B 73(22), 224117 (2006). [CrossRef]
18. V. Grubsky and J. Feinberg, “Phase-matched third-harmonic UV generation using low-order modes in a glass micro-fiber,” Opt. Commun. 274(2), 447–450 (2007). [CrossRef]
19. F. Gu, H. Yu, W. Fang, and L. L. Tong, “Low-threshold supercontinuum generation in semiconductor nanoribbons by continuous-wave pumping,” Opt. Express 20(8), 8667–8674 (2012). [CrossRef]
20. M. A. Gouveia, T. Lee, R. Ismaeel, M. Ding, N. G. R. Broderick, C. M. B. Cordeiro, and G. G. Brambilla, “Second harmonic generation and enhancement in microfibers and loop resonators,” Appl. Phys. Lett. 102(20), 201120 (2013). [CrossRef]
21. X. Zhang, Q. T. Cao, Z. Wang, Y. Liu, C. W. Qiu, L. Yang, Q. Gong, and Y. F. Xiao, “Symmetry-breaking-induced nonlinear optics at a microcavity surface,” Nat. Photonics 13, 21–24 (2019). [CrossRef]
22. Y. Li, Y. Kang, X. Guo, and L. Tong, “Simultaneous generation of ultrabroadband noise-like pulses and intracavity third harmonic at 2 µm,” Opt. Lett. 45(6), 1583–1586 (2020). [CrossRef]
23. L. Huang, Y. Zhang, Y. Cui, J. Qiu, and X. Liu, “Microfiber-assisted GHz harmonic mode-locking in ultrafast fiber laser,” Opt. Lett. 45(17), 4678–4681 (2020). [CrossRef]
24. Z. X. Ding, Z. N. Huang, Y. Chen, C. B. Mou, Y. Q. Lu, and F. Xu, “All-fiber ultrafast laser generating gigahertz-rate pulses based on a hybrid plasmonic microfiber resonator,” Adv. Photonics 2(2), 026002 (2020). [CrossRef]
25. F. Liao, J. Yu, Z. Gu, Z. Yang, T. Hasan, S. Linghu, J. Peng, W. Fang, S. Zhuang, M. Gu, and F. Gu, “Enhancing monolayer photoluminescence on optical micro/nanofibers for low-threshold lasing,” Sci. Adv. 5(11), eaax7398 (2019). [CrossRef]
26. U. Wiedemann, K. Karapetyan, C. Dan, D. Pritzkau, W. Alt, S. Irsen, and D. Meschede, “Measurement of submicrometre diameters of tapered optical fibres using harmonic generation,” Opt. Express 18(8), 7693–7704 (2010). [CrossRef]
27. C. Rodenburg, X. Liu, M. A. Jepson, S. A. Boden, and G. Brambilla, “Surface morphology of silica nanowires at the nanometer scale,” J. Non-Cryst. Solids 357(15), 3042–3045 (2011). [CrossRef]
28. Y. C. Chen, Z. Lu, K. J. Nomura, W. Wang, R. K. Kalia, A. Nakano, and P. Vashishta, “Interaction of voids and nanoductility in silica glass,” Phys. Rev. Lett. 99(15), 155506 (2007). [CrossRef]
29. S. Bonfanti, E. E. Ferrero, A. L. Sellerio, R. Guerra, and S. Zapperi, “Damage accumulation in silica glass nanofibers,” Nano Lett. 18(7), 4100–4106 (2018). [CrossRef]
30. J. Luo, J. Wang, E. Bitzek, J. Y. Huang, H. Zheng, L. Tong, Q. Yang, J. Li, and S. X. Mao, “Size-dependent brittle-to-ductile transition in silica glass nanofibers,” Nano Lett. 16(1), 105–113 (2016). [CrossRef]
31. L. Skuja, M. Hirano, H. Hosono, and K. Kajihara, “Defects in oxide glasses,” Phys. Status Solidi C 2(1), 15–24 (2005). [CrossRef]
32. K. Awazu and H. Kawazoe, “Strained Si-O-Si bonds in amorphous SiO2 materials: A family member of active centers in radio, photo, and chemical responses,” J. Appl. Phys. 94(10), 6243–6262 (2003). [CrossRef]
33. T. Uchino, Y. Kitagawa, and T. Yoko, “Structure, energies, and vibrational properties of silica rings in SiO2 glass,” Phys. Rev. B 61(1), 234–240 (2000). [CrossRef]
34. L. Irimpan, B. Krishnan, V. P. N. Nampoori, and P. Radhakrishnan, “Nonlinear optical characteristics of nanocomposites of ZnO-TiO2-SiO2,” Opt. Mater. 31(2), 361–365 (2008). [CrossRef]
35. G. Fuxing, “Optical waveguiding nanowires and their sensing applications,” Ph. D. dissertation (Zhejiang University, 2012).
36. L. Liu, X. Zhang, Q. Zhu, K. Li, Y. Lu, X. Zhou, and T. Guo, “Ultrasensitive detection of endocrine disruptors via superfine plasmonic spectral combs,” Light: Sci. Appl. 10(1), 181 (2021). [CrossRef]
37. J. Teng, S. Bates, D. A. Engers, K. Leach, P. Schields, and Y. Yang, “Effect of water vapor sorption on local structure of poly(vinylpyrrolidone),” J. Pharm. Sci. 99(9), 3815–3825 (2010). [CrossRef]
38. L. T. Zhuravlev, “The surface chemistry of amorphous silica Zhuravlev model,” Colloids Surf., A 173(1-3), 1–38 (2000). [CrossRef]
39. X. Shen, H. Choi, D. Chen, W. Zhao, and A. M. Armani, “Raman laser from an optical resonator with a grafted single-molecule monolayer,” Nat. Photonics 14(2), 95–101 (2020). [CrossRef]
