• Chinese Optics Letters
  • Vol. 23, Issue 7, 071405 (2025)
Jihui Zheng1,2, Youwei Wang1, Xiaoling Jin1, Chen Jiang1..., Bing Sun1, Fumin Zhang2 and Zuxing Zhang1,*|Show fewer author(s)
Author Affiliations
  • 1Advanced Photonic Technology Laboratory, College of Electronic and Optical Engineering and College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications, Nanjing 210023, China
  • 2State Key Laboratory of Precision Measurement Technology and Instruments, Tianjin University, Tianjin 300072, China
  • show less
    DOI: 10.3788/COL202523.071405 Cite this Article Set citation alerts
    Jihui Zheng, Youwei Wang, Xiaoling Jin, Chen Jiang, Bing Sun, Fumin Zhang, Zuxing Zhang, "All-optical tuning of a multi-walled carbon nanotube-coated microcavity," Chin. Opt. Lett. 23, 071405 (2025) Copy Citation Text show less

    Abstract

    In this paper, an all-optical tuning scheme of a multi-walled carbon nanotube (MWCNT)-coated microcavity is introduced, achieving high-speed precise resonance control across the free spectral range (FSR). A modulation laser input through the microcavity tail fiber adjusts the resonance peak position, achieving a tuning efficiency of 107.3 pm/mW below 15 mW, with a maximum range exceeding one FSR and a response time of ∼20 ms. Combined with a fixed-wavelength pump, this scheme can precisely control the microcomb states. The scheme offers high tuning efficiency, simple fabrication, and low cost, making it suitable for applications in microcomb control and optical filters.

    1. Introduction

    The whispering gallery mode (WGM) microcavity[1] utilizes the total reflection effect of light waves at the boundary of the medium, allowing light waves to circulate along the edge of the medium. This confines the optical field within the cavity, forming high-Q resonance modes. This unique optical characteristic causes WGM microcavities to be widely used in important fields such as high-sensitivity sensing[2,3], nonlinear optics[4], microcombs[57], and optical filtering[8]. The common WGM microcavities include microspheres, microdisks, and microbottles[9]. Among these, microsphere cavities are the most widely studied due to their simple and mature fabrication process, high Q-factor, and small mode volume[1012].

    The resonance position is crucial in many applications of microcavities such as the generation of microcavity optical frequency combs (OFCs), in which the resonance position directly determines whether the OFC can be generated[13,14]. Current methods for tuning microcavity resonance include temperature tuning, electrical tuning, mechanical tuning, and optical tuning. Temperature tuning[13,15] primarily involves adding a thermo-electric cooler (TEC) in the microcavity to adjust its internal temperature, thereby changing the refractive index and shifting the resonance peak position. However, the thermal conductivity of microcavities varies depending on the material, making rapid tuning difficult and leading to a certain of response lag. Electrical tuning[16,17] typically involves adding electrodes to the microcavity and utilizing the electro-optic effect to change the refractive index of the material, thereby adjusting the resonance peak position. This method requires a relatively complex process to integrate electrodes onto the microcavity and is widely used in electric tuning materials such as lithium niobate. Mechanical tuning[18,19] involves adding a piezoelectric ceramic actuator to the microcavity, using mechanical actions such as stretching or compressing the microcavity to dynamically modulate the resonance position. However, in mechanical tuning, the coupling position between the microcavity and the fiber taper is prone to interference. The final approach involves the use of an external laser, also known as all-optical tuning[2023]. By adjusting the laser power within the microcavity, the size of the microcavity is altered, thereby tuning the resonance peak position. This method is relatively simple and offers greater versatility. By adjusting the laser power within the microcavity, the size of the microcavity is changed, thereby tuning the position of the resonance peak. This method is relatively simple and offers greater versatility.

    In 2017, Zhu et al. proposed combining a silica microcavity with iron oxide nanoparticles to achieve all-optical control of the microcavity. This approach resulted in a tuning range of 85.9 GHz and a tuning sensitivity of 13.6 GHz/mW, while consistently maintaining a high Q-factor throughout the tuning process[20]. In 2021, Wang et al. proposed an all-optical tunable polymer bottle microcavity fabricated by the droplet dispensing method. The resonant mode inside the cavity was excited by signal light, and the resonant position was adjusted using external and internal pump light methods. The adjustment efficiencies of the two schemes were 5.23 and 68.3 pm/mW, respectively[22]. However, currently these solutions make it difficult to achieve large-scale and rapid adjustment of resonance.

    In this Letter, we demonstrate an all-optical tuning of the multi-walled carbon nanotube (MWCNT)-coated microcavity. By utilizing the excellent adsorption properties and exceptional thermal stability of MWCNTs[24,25], the MWCNTs can be attached to the end face of silica microcavities to improve the efficiency of all-optical modulation and achieve high-speed, resonant modulation across the free spectral range (FSR). Optimizing the optical path of the system further enhances the resonant tuning efficiency. Experimental results show that as the modulation laser power increases, the microcavity resonance gradually redshifts. When the 980 nm modulation laser power is below 15 mW, the all-optical tuning efficiency is 107.3 pm/mW, with the maximum resonance redshift reaching 2.56 nm, exceeding one FSR. Additionally, the system demonstrates a fast response time of 22.2 ms. Finally, the application of the all-optical tuning scheme in microcomb generation is demonstrated, which requires only a fixed-wavelength pump laser, thereby reducing the need for precise wavelength control.

    2. Fabrication of the MWCNT-Coated Microcavity

    The fabrication process for microcavities coated with MWCNT is shown in Fig. 1(a). The fabrication of fiber microsphere cavities mainly adopts methods such as arc discharge or carbon dioxide (CO2) laser ablation. Under the influence of surface tension, the silica material forms a smooth well-defined spherical shape, thereby creating the microcavity. In this work, an arc discharge method is employed. A segment of standard single-mode fiber (SMF-28) is selected, and its protective coating is removed before cleaning the fiber surface with alcohol. The fiber is then placed into one side of a fiber fusion splicer, and the pre-programmed microcavity fabrication process is used to precisely control the electrode discharge intensity, fiber positioning, discharge time, and annealing time. The entire microcavity fabrication process is fully automated, and the standardized procedure ensures that the microcavity size is consistent across each fabrication. Another SMF is then inserted into a test tube containing the MWCNT solution, soaked for a few seconds, and is then quickly pulled out. Due to the excellent adsorption properties of MWCNTs, droplets of varying sizes remain on the fiber surface. The prepared microcavity and the fiber with MWCNT droplets are fixed on a six degree-of-freedom precision translation stage. The translation stages are adjusted to ensure that the microcavity and fiber are aligned perpendicularly, and then the microcavity is slowly moved closer to the MWCNT droplets on the optical fiber. Upon contact, the droplets adhere to the microcavity surface, thereby achieving MWCNT coating on the microcavity surface. The coating size on the microcavity surface can be adjusted by contacting MWCNT droplets of different sizes. The coated microcavity is placed in a drying oven, typically set at 80°C for half an hour, heated, and dried to complete the preparation of MWCNT-coated microcavities. This material transfer method is simple, efficient, and highly successful, ensuring uniform coating.

    Fabrication of a MWCNT-coated microcavity. (a) Fabrication process. (b) Image of a microcavity coated without MWCNT. (c) Image of a microcavity coated with MWCNT. (d) Q-factor measurement of a microcavity without MWCNT coating. (e) Q-factor measurement of a microcavity with MWCNT coating.

    Figure 1.Fabrication of a MWCNT-coated microcavity. (a) Fabrication process. (b) Image of a microcavity coated without MWCNT. (c) Image of a microcavity coated with MWCNT. (d) Q-factor measurement of a microcavity without MWCNT coating. (e) Q-factor measurement of a microcavity with MWCNT coating.

    The concentration of MWCNT used in the experiment is 13%. The comparison of the microcavity before and after MWCNT coating is shown in Figs. 1(b) and 1(c). It can be clearly seen that the end face of the microcavity is uniformly covered with MWCNTs. The shaded area in Fig. 1(c) corresponds to the MWCNT material. Moreover, the MWCNTs do not exceed the equatorial line of the microcavity, resulting in minimal impact on the microcavity coupling. The microcavity has a diameter of 280 µm, corresponding to a FSR of 235 GHz. The Q-factors of the microcavity before and after MWCNT coating are tested using a frequency-modulation (FM) laser, and the laser wavelength is scanned near 1550 nm. The resonance of the microcavity without MWCNT is shown in Fig. 1(d), with a full width at half-maximum (FWHM) of 10.4 MHz, corresponding to a Q-factor of 1.86×107. After coating the microcavity with the MWCNT, its resonance state is shown in Fig. 1(e). The FWHM of resonance is 20.9 MHz, which is broader compared to that of the uncoated microcavity, and the resonance depth is relatively shallower. The Q-factor of the microcavity has decreased to 0.92×107.

    3. Results

    After completing the fabrication of the MWCNT-coated microcavity, precise coupling and packaging are carried out. A single-mode tapered fiber is prepared using a fiber tapering machine and coupled with the microcavity. The coupling distance is finely adjusted to achieve optimal performance. An experimental setup for all-optical tuning of the microcavity is built, as shown in Fig. 2. The 1550 nm pump laser passes through a fiber polarization controller (FPC) before entering the microcavity via one end of the tapered fiber. The other end of the tapered fiber is connected to a photodetector (PD) to measure the resonance signal of the microcavity. After fabricating the fiber microsphere cavity, the tail fiber is connected to a 980 nm modulation laser. By adjusting the output power of the 980 nm modulation laser, the resonance position of the microcavity can be tuned, enabling all-optical control. In the experiment, a 1550 nm pump laser is set to scan continuously from 1550 to 1553 nm at a rate of 6 nm/s, with a pump power of 12 mW. The scanning drive signal for the pump laser is a triangular signal, and an oscilloscope (OSC) is used to simultaneously detect both laser scanning signal and resonance signal of the microcavity.

    Experimental setup for all-optical tuning based on a MWCNT-coated microcavity. FPC, fiber polarization controller; PD, photodetector; OSC, oscilloscope; OSA, optical spectrum analyzer.

    Figure 2.Experimental setup for all-optical tuning based on a MWCNT-coated microcavity. FPC, fiber polarization controller; PD, photodetector; OSC, oscilloscope; OSA, optical spectrum analyzer.

    When the 980 nm laser is not activated (i.e., output power is 0 mW), the resonant peaks generated by the MWCNT-coated microcavity during the 1550 nm laser scanning are shown in Fig. 3(a). The microcavity exhibits multiple resonant modes, with a resonance spacing of 235 GHz within the same mode, corresponding to one FSR. After turning on the 980 nm laser, the output power is gradually increased from 0 to 48.5 mW, and the energy within the cavity changes continuously, leading to variations in the refractive index and the microcavity size. These changes, in turn, affect the resonant position of the microcavity, and the resonant peak of the microcavity gradually redshifts. Taking the first resonant peak on the left side of the microcavity as an example, the variation in the resonant position under different modulation laser powers is observed, and the experimental results are shown in Fig. 3(b). Initially, as the modulation laser power increases, the tuning speed of the resonant peak is relatively fast. When the 980 nm modulation laser power reaches 25.6 mW, the resonant redshift is 2.03 nm, which exceeds one FSR of the microcavity, demonstrating that the microcavity can achieve full-band resonant position tuning under the influence of MWCNT materials. When the 980 nm laser power reaches 48.5 mW, the resonant redshift of the microcavity reaches 2.56 nm. With further increases in modulation laser power, the resonant shift of the microcavity changes little and reaches saturation. Throughout the entire all-optical tuning process, the changes in resonance depth and FWHM are negligible as the modulation laser power increases. This indicates that the coupling state and Q-factor of the device remain stable during tuning, verifying the stability of the all-optical tuning system and avoiding the impact on the coupling state typically associated with mechanical tuning methods.

    Resonance changes of a MWCNT-coated microcavity under all-optical tuning. (a) Microcavity resonant peak at 0 mW modulation laser power. (b) Resonance changes of the microcavity at different modulation laser powers.

    Figure 3.Resonance changes of a MWCNT-coated microcavity under all-optical tuning. (a) Microcavity resonant peak at 0 mW modulation laser power. (b) Resonance changes of the microcavity at different modulation laser powers.

    Similarly, the same experiment is conducted in microcavities that are not coated with MWCNT. As the modulation laser power increases, there is no significant change in the resonance peak of the microcavity. This is because most of the modulation laser is emitted from the surface of the microcavity, preventing efficient photothermal conversion and absorption.

    The wavelength shift of the resonance with varying modulation laser power is also calculated, and the results are shown as red circular markers in Fig. 4. As the 980 nm modulation laser power increases, the wavelength shift initially increases rapidly and then slows down, and it eventually saturates, exhibiting an exponential redshift pattern. An exponential model is fitted to the measurement results, obtaining the function y=2.87e0.05x+2.81, with a root mean square error (RMSE) of 0.058, as shown by the green line in Fig. 4. The measurement results show good linearity for 980 nm modulation laser power less than 15 mW. Linear fitting of the measurement results shows that the wavelength tuning efficiency is 107.3 pm/mW, as indicated by the purple line in Fig. 4. A tuning range exceeding one FSR can be achieved with only a small modulation laser power (<50mW). The blue pentagrams in Fig. 4 indicate the resonance changes as the 980 nm modulation laser power is gradually reduced. At the three measurement points, the repeatability of the resonance wavelength changes during power reduction is better than 8.1 pm compared to when the power is increased, demonstrating superior stability. This confirms the excellent rebound characteristics of the all-optical tuning based on the MWCNT-coated microcavity.

    Resonance wavelength variation with modulation laser power.

    Figure 4.Resonance wavelength variation with modulation laser power.

    The MWCNT-coated microcavity exhibits excellent thermal stability and can quickly respond to different modulation laser powers, enabling real-time adjustment. Response time is an important parameter for the MWCNT-coated microcavity. In the experiment, the wavelength of the 1550 nm laser is fixed, and the output power of the 980 nm modulation laser is set to 14.3 mW. At the moment when the 980 nm laser is turned on or off, the temperature field inside the microcavity undergoes dynamic changes, causing a thermos-optic effect on the resonant wavelength, resulting in a change in the resonant conditions of the 1550 nm laser, manifested as a transient response of the intracavity power. By measuring the rise edge time and fall edge time of power changes inside the microcavity, the dynamic evolution characteristics of the optical field can be extracted, and the response time of all-optical modulation can be calculated. When the modulation laser is turned on, the experimental results, as shown in Fig. 5(a), indicate that the intracavity optical power rapidly increases and tends to stabilize, with a rise time of 22.2 ms. When the 980 nm modulation laser is turned off, as shown in Fig. 5(b), the intracavity optical field decreases with a fall time of 17.1 ms, which is faster than the power rise response time. If a smaller microcavity is used or the thickness of the MWCNT coating is adjusted, the response speed of the all-optical modulation system could potentially be further improved.

    Response time of all-optical tuning based on a MWCNT-coated microcavity. (a) Microcavity response speed when the 980 nm modulation laser is turned on. (b) Microcavity response speed when the 980 nm modulation laser is turned off.

    Figure 5.Response time of all-optical tuning based on a MWCNT-coated microcavity. (a) Microcavity response speed when the 980 nm modulation laser is turned on. (b) Microcavity response speed when the 980 nm modulation laser is turned off.

    By utilizing the characteristics of precise control of the resonant wavelength and a wide adjustment range through all-optical tuning, a 1550 nm laser is used as the pump and combined with an all-optical tuning scheme. The resonance position is precisely tuned to match the pump wavelength, resulting in the generation of a microcomb. The pump wavelength is fixed at 1550.512 nm, and the pump power is set to 40 mW. The experimental setup is shown in Fig. 2, where the output port of the microcavity is connected to an optical spectrum analyzer (OSA) for real-time observation of the intracavity optical field in the MWCNT-coated microcavity.

    As indicated by previous experiments, increasing the all-optical modulation power causes a redshift in the microcavity resonance. However, during the generation of the microcomb, the pump needs to transition gradually from the blue-detuned region to the red-detuned region. To achieve a blue shift of the resonance peak, the 980 nm modulation laser power needs to be reduced. The initial power of the 980 nm modulation laser is set to 2.19 mW. At this point, the pump wavelength does not yet match the resonance, and the microcavity emission spectrum, shown in Fig. 6(a), only displays the pump signal. As the 980 nm modulation laser power is gradually reduced, the resonant blue shift gradually matches the pump wavelength, triggering nonlinear effects within the cavity and generating new spectral components. The intracavity optical field progresses through the primary comb, sub-comb, modulation instability (MI) state, and soliton microcombs, as shown in Figs. 6(b)6(f). Figure 6(f) shows the single-soliton microcomb, with a smooth comb envelope that fits well with the sech2 curve, with a spectral range exceeding 100 nm. Subsequently, the 980 nm modulation laser power is increased to 1.31 mW, and the intracavity optical field returns to the MI state, as shown in Fig. 6(g), which is identical to the spectral state observed at the same power level when the power is previously reduced [Fig. 6(e)]. Further increasing the power results in the extinction of the microcomb, as shown in Fig. 6(h). The experiment demonstrated that the all-optical modulation scheme enabled precise control of the microcomb state. Due to the system being in an open-loop state, the soliton state has a short stability duration. In the future, adding proportional-integral-derivative (PID) control could enhance the stability of the soliton comb. Compared with traditional microcavity comb generation methods such as auxiliary laser heating and laser fast scanning schemes, this system is simpler and does not require precise tuning of the pump wavelength, allowing the use of a single-wavelength laser, which significantly reduces system costs. The all-optical modulation scheme can achieve similar effects using any wavelength modulation laser, and previous related work[21] has been verified.

    Emission spectra of a MWCNT-coated microcavity under different powers of the 980 nm modulation laser, with power values shown in the upper right corner. (a)–(f) Intracavity spectra as the 980 nm modulation laser power is gradually reduced. (f) Single soliton state comb. (g)–(h) Intracavity spectra as the 980 nm modulation laser power is increased.

    Figure 6.Emission spectra of a MWCNT-coated microcavity under different powers of the 980 nm modulation laser, with power values shown in the upper right corner. (a)–(f) Intracavity spectra as the 980 nm modulation laser power is gradually reduced. (f) Single soliton state comb. (g)–(h) Intracavity spectra as the 980 nm modulation laser power is increased.

    4. Discussion

    In summary, we demonstrate an all-optical tunability of a MWCNT-coated microcavity. This scheme transfers MWCNT material to the surface of the microcavity, where it absorbs light and converts it into thermal energy within the microcavity. This process changes the output power of the modulation laser to achieve accurate, fast, and highly repeatable all-optical tuning across the FSR of the microcavity resonance position. Experimental results show that when the modulation laser power is below 15 mW, the all-optical tuning efficiency is 107.3 pm/mW with high linearity. The high tuning efficiency enables this system to achieve a tuning range exceeding one FSR with very low modulation power. Compared to existing reports[21,22,2628], the resonance tuning efficiency has been significantly improved. Turning on and off the modulation laser, the response time of the all-optical tuning is tested, with results showing 20ms. By utilizing its all-optical modulation characteristics, we combine it with a fixed wavelength pump laser to generate a microcomb with a spectral range exceeding 100 nm. The state of the microcomb can be switched by adjusting the modulation laser power, reducing the need for precise control of the pump laser, and resulting in a simpler and more convenient system. In the future, by improving the properties of materials and introducing composite materials, it is expected that the accuracy and response speed of all-optical tuning will be further enhanced, providing new avenues for the development of high-speed optical switches and optical filters.

    References

    [1] X. F. Jiang, C. L. Zou, L. Wang et al. Whispering-gallery microcavities with unidirectional laser emission. Laser Photonics Rev., 10, 40(2016).

    [2] N. Toropov, G. Cabello, M. P. Serrano et al. Review of biosensing with whispering-gallery mode lasers. Light Sci. Appl., 10, 42(2021).

    [3] H. Yu, X. Liu, W. Sun et al. A brief review of whispering gallery mode in sensing. Opt. Laser Technol., 177, 111099(2024).

    [4] Q. T. Cao, R. Liu, H. Wang et al. Reconfigurable symmetry-broken laser in a symmetric microcavity. Nat. Commun., 11, 1(2020).

    [5] P. Del’Haye, A. Schliesser, O. Arcizet et al. Optical frequency comb generation from a monolithic microresonator. Nature, 450, 1214(2007).

    [6] T. Herr, V. Brasch, J. D. Jost et al. Temporal solitons in optical microresonators. Nat. Photonics, 8, 145(2014).

    [7] C. Xiang, J. Liu, J. Guo et al. Laser soliton microcombs heterogeneously integrated on silicon. Science, 373, 99(2021).

    [8] Y. Li, B. Dong, E. Chen et al. Graphene/epoxy composite based broadband all-optical tunable fiber microcavity filter. J. Light. Technol., 40, 1167(2022).

    [9] G. Lin, Q. Song. Kerr frequency comb interaction with Raman, Brillouin, and second order nonlinear effects. Laser Photonics Rev., 16, 2100184(2022).

    [10] S. M. Spillane, T. J. Kippenberg, K. J. Vahala. Ultralow-threshold Raman laser using a spherical dielectric microcavity. Nature, 415, 621(2002).

    [11] S. Zhu, L. Shi, B. Xiao et al. All-optical tunable microlaser based on an ultrahigh- Q erbium-doped hybrid microbottle cavity. ACS Photonics, 5, 3794(2018).

    [12] X. Jin, M. Chen, J. Zheng et al. Magnetic tuning with minimal thermal drift in high-Q microspheres coated with magnetorheological polydimethylsiloxane. Opt. Lett., 49, 5071(2024).

    [13] H. Zhou, Y. Geng, W. Cui et al. Soliton bursts and deterministic dissipative Kerr soliton generation in auxiliary-assisted microcavities. Light Sci. Appl., 8, 50(2019).

    [14] Z. Lu, H. J. Chen, W. Wang et al. Synthesized soliton crystals. Nat. Commun., 12, 1(2021).

    [15] D. Armani, B. Min, A. Martin et al. Electrical thermo-optic tuning of ultrahigh-Q microtoroid resonators. Appl. Phys. Lett., 85, 5439(2004).

    [16] Y. A. Liu, X. S. Yan, J. W. Wu et al. On-chip erbium-doped lithium niobate microcavity laser. Sci. China Phys. Mech. Astron., 64, 234262(2021).

    [17] J. T. Lin, S. Farajollahi, Z. W. Fang et al. Electro-optic tuning of a single-frequency ultranarrow linewidth microdisk laser. Adv. Photonics, 4, 036001(2022).

    [18] H. Qin, Y. Yin, M. Ding. Strain-induced tunable dual-bottle-shaped optical microresonator. Opt. Lett., 44, 6017(2019).

    [19] S. Fujii, K. Wada, S. Kogure et al. Mechanically actuated Kerr soliton microcombs. Laser Photonics Rev., 18, 2301329(2024).

    [20] S. Zhu, L. Shi, S. Yuan et al. All-optical control of ultrahigh-Q silica microcavities with iron oxide nanoparticles. Opt. Lett., 42, 5133(2017).

    [21] H. Lu, J. Tao, L. Chen et al. All-optical tuning of micro-resonator overlaid with MoTe2 nanosheets. J. Light. Technol., 37, 3637(2019).

    [22] H. Wang, M. Liao, H. Xiao et al. All-optical tunable whispering gallery modes in a polymer bottle micro-resonator. IEEE Photonics Technol. Lett., 33, 97(2021).

    [23] S. Zhu, L. Shi, L. Ren et al. Controllable Kerr and Raman-Kerr frequency combs in functionalized microsphere resonators. Nanophotonics, 8, 2321(2019).

    [24] O. B. Mergen, E. Arda, G. A. Evingur. Electrical, optical, and mechanical percolations of multi-walled carbon nanotube and carbon mesoporous-doped polystyrene composites. J. Compos. Mater., 54, 31(2020).

    [25] C. Jiang, Y. Wan, L. Yun et al. Optically tunable mode-locked fiber laser using long-period grating coated with multiwall carbon nanotubes. Appl. Phys. Lett., 124, 111104(2024).

    [26] S. Zhu, L. Shi, S. Yuan et al. All-optical controllable electromagnetically induced transparency in coupled silica microbottle cavities. Nanophotonics, 7, 1669(2018).

    [27] J. Lei, M. Wang, Z. Guo et al. Experimental demonstration of all-optical tunable methods for UCA microbottle cavity. IEEE Photon. Technol. Lett., 36, 1(2024).

    [28] L. Yang, B. Dong, Q. Shao et al. Short carbon nanotube nano-film-based polymer/SiO2 hybrid waveguide micro-ring filter. ACS Appl. Mater. Interfaces, 15, 28555(2023).

    Jihui Zheng, Youwei Wang, Xiaoling Jin, Chen Jiang, Bing Sun, Fumin Zhang, Zuxing Zhang, "All-optical tuning of a multi-walled carbon nanotube-coated microcavity," Chin. Opt. Lett. 23, 071405 (2025)
    Download Citation