Open Access
Add to BibSonomy
Abstract
Electromagnetically induced transparency (EIT) analogs in terahertz regimes have been actively tuned by hybridizing semiconductors into metasurfaces for effective terahertz manipulation. Although several multiple means have been adopted to modulate EIT, the active control of the EIT quality factor is less reported. Multifunctional EIT modulation should also be considered. In this study, we fabricated a terahertz metadevice by embedding Si bridges into a metasurface with optically controllable molecularization of terahertz meta-atoms and experimentally demonstrated dual-functional terahertz EIT modulation, including resonance amplitude modulation and quality-factor modulation. For a low-power pumping of 16 mW, ultrafast EIT resonance amplitude modulation can be achieved with a low-level molecularization, with a normalized amplitude modulation of up to 85%. Then, with a high-power pumping of 360 mW, we performed ultrafast EIT quality-factor switching, where the high Si-bridge conductivity altered the spatial interconnected structure and exhibited high-level molecularization. Both functions have a nanosecond-scale modulation speed. In addition, we numerically and theoretically explained the terahertz modulation and molecularization behaviors, which was consistent with the experimental results. Our proposed all-optical active terahertz metadevice with controllable molecularization provides a novel solution for multifunctional and ultrafast switchable terahertz modulators.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Terahertz waves, as a candidate band for next-generation wireless communications, ranging from 0.1 to 10 THz, play a substantial role in higher data rates [1,2]. Moreover, various applications in the fields of imaging systems, [3,4] nondestructive testing, [5,6] bioscience, [7,8] and security checks [9] also involve the use of THz frequency electromagnetic waves. With abundant spectrum resources, high security, and wide usage, terahertz technology has garnered significant attention from researchers. These practically advanced terahertz applications have inspired the development of terahertz devices, such as terahertz sources, [10–12] modulators, [13–15] and receivers [16,17]. However, the lack of corresponding modulation means and weak terahertz wave interaction with traditional materials still distance us from practical terahertz applications. Reliable methods need to be explored to achieve effective terahertz manipulation. Fortunately, the emergence of metasurfaces provides a unique ability to control light-matter interactions at subwavelength scales for extraordinary phenomena that are unachievable in natural materials [18–21]. The metasurface can offer diverse modulation approaches for manipulating THz waves.
Terahertz metasurfaces, which consist of arrays of metallic structures with micrometer-scale periodicity, have been widely used to design terahertz devices [22–25]. For effectively manipulating terahertz information, the electromagnetically induced transparency-like (EIT-like) effect found in terahertz metasurfaces is widely popular owing to its sensitivity and strong light-matter interaction [26,27]. Several studies have been conducted on terahertz modulation, such as amplitude control, frequency tuning, and slow light effects [28–32]. Active control of the EIT effect in the terahertz regime is crucial for efficiently promoting the development of terahertz modulation, benefiting from the alterable electromagnetic properties and surrounding environment. Various methods for the dynamic tuning of EIT THz metasurfaces using external stimuli, including optical, electrical, and mechanical means, [27,29,33,34] have been widely reported. Compared to other stimulation methods, generally, all-optical control allows for ultrafast speed modulation owing to ultrashort relaxation time in the nanosecond or picosecond scale of semiconductor and two-dimensional materials [34,35]. As an important indicator of the EIT spectrum, the quality (Q) factor evaluates the radiative losses and the interaction strength between terahertz and metasurfaces, which shows the sensitivity of the material response to terahertz [36]. Corresponding works on EIT Q factor modulation are yet to be investigated in terahertz regimes. Thus, it is worth studying the active control of the THz EIT quality factor.
In addition, multiple functions are also one of the most desirable performance indices for terahertz metadevices. In this paper, inspired by the molecularization of meta-atoms, [37,38] we experimentally demonstrated a dual-functional terahertz metasurface for ultrafast all-optical modulation. The molecularization process connects the meta-atoms of a metasurface unit to change the spatial structure. Active control of the THz EIT resonance amplitude and quality factor were performed at different levels of exciting optical pump power. We fabricated a metasurface supported by the EIT effect in which the meta-atoms were connected by Si bridges. With a pump power of 16 mW, the THz device exhibited low-level molecularization for switching the EIT resonance amplitude. With a pump power of 360 mW, high-level molecularization altered the spatial structure and performed an ultrafast modulation of the EIT window width (or quality-factor modulation). Here, Q-factor modulation alters the electric field intensity, but keeps the terahertz transmission nearly unchanged. The modulation speed of both functions was on the order of a few nanoseconds. Our proposed strategy paves the way for multifunctional active controlling terahertz metadevices and provides a novel idea for terahertz EIT modulation.
2. Results
We first demonstrated the design and characterization of a dual-functionally Si-bridged metadevice. The metadevice structure was fabricated on a commercially available silicon-on-sapphire wafer, which was composed of a 485-nm-thick intrinsic epitaxial Si layer on a 500-μm-thick R-plane sapphire substrate. Figure 1(a) describes the working principle of the proposed metadevices supported by the EIT effect. The carriers of the Si-bridge were excited by the excitation of an 800 nm femtosecond laser and then altered the spatial interconnection, accompanied by the efficient modulation of the EIT effect. The incident THz wave with y polarization propagated along the z direction. The spot sizes of the pump light and terahertz waves were ∼5 and 3 mm in diameter, respectively. One unit cell of the metasurface contained a cut wire along the y-direction integrated with two split-ring resonators (SRRs) and four off-state SRRs, as shown in Fig. 1(b). The light blue parts represent the Si bridge. The golden part was the gold metal and the green part was the sapphire substrate. Meanwhile, Si islands were fabricated and bridged the assigned gaps. To achieve a higher degree of molecularization, sufficiently lengthened gold paddles in the off-state SRRs were necessary. The periods of the one-unit cell are px = 100 μm in the x-direction and py = 115 μm in the y-direction. The heights of the gold meta-atoms, Si-bridges, and sapphire substrates were 100 nm, 485 nm, and 500 μm, respectively. The fabricated metadevice was inspected under an optical microscope, as shown in Fig. 1(c), from which the gold metasurface and Si bridges are distinguished. More detailed structure parameters can be found in Fig. 1(d, e). All splitting gaps are g1 = 5 μm, and the width of metal is w = 5 μm. The length of cut-wire is L = 85 μm, the sizes of each SRR are b = 20 μm and c = 36 μm, and the size of each off-state SRR is a = 29 μm. The distance between off-state SRR and SRR is g5 = 3.5 μm. The distance between off-state SRR and cut-wire is g3 = 2 μm and the distance between SRR and cut-wire is g4 = 8 μm. The gap between two paddles is g2 = 1 μm. The whole width of paddles is d = 6 μm. The length of paddles is 20 μm with s1 = 10 μm and s2 = 5 μm.
Fig. 1. Ultrafast optically tunable EIT metadevice with molecularization. The golden part was the gold metal and the green part was the sapphire. The blue layer represents the Si bridge. (a) Artistic illustration of dual-functional metadevice with ultrafast EIT switching under femtosecond optical excitation. The y-polarization terahertz pulse was incident along z direction. (b) Graphical representation of one unit of proposed metadevice above the sapphire substrate. The periods were px = 100 μm in the x direction and py = 115 μm in the y direction. (c) Optical microscopy images of proposed metadevice fabricated on a commercially available SOS wafer. The scale bar in the picture was 50 μm. (d) Planar diagram of one-unit cell. (e) The enlarged image of off-state SRR.
One of the most important characteristics of terahertz modulators is their multifunctionality and ultrafast modulation speed. Two functions of ultrafast modulation were achieved from this metadevice with different levels of pump fluences. Prior to the study of ultrafast EIT modulation, we primarily discussed the EIT effect behaviors as a function of pump power in the experiment, as shown in Fig. 2(a). The frequency-dependent THz transmission was obtained by dividing the sampling transmission spectrum ($|{{E_S}(\omega )} |$) by the reference substrate transmission spectrum ($|{{E_R}(\omega )} |$), described as $T(\omega )= \; |{{E_S}(\omega )/{E_R}(\omega )} |$. Normally, a distinct EIT window occurs in the transmission spectrum of the intrinsic metadevice without a pump light. Afterward, in the case of optical excitation, the EIT effect was generally suppressed as the pump power increased from 0 to 16 mW. Notably, the width of the EIT window defined by two adjacent dips, $\delta f = \; {f_{dip\; 2}} - {f_{dip\; 1}}$, remains nearly unchanged at approximately 0.16 THz, shown in Fig. 3(c) marked by the horizontal dashed line. The important point to note is that a new broader EIT window occurred when the pump power exceeded 40 mW owing to the molecularization of the metasurface. Moreover, in this case, as the pump power increased from 40 to 360 mW, the EIT effect became increasingly apparent. The last transmission spectrum exhibited a broad EIT window of approximately 0.32 THz, at the maximal optically exciting power of 360 mW in our experiment. Figure 2(c) shows the experimental EIT window at various pump powers. This inspired us to realize two functions based on a metadevice as previously mentioned: (I) low-power optical excitation referred to modulating the EIT resonance amplitude; (II) high-power optical excitation suggested the modulation of the EIT window or Q factor. Particularly, we defined the EIT resonance amplitude as $\delta T = {T_{peak}} - {T_{dip1}}$, and the amplitude was strongly modulated with a low pump power of 16 mW. The normalized resonance amplitude as a function of the pump power is shown in Fig. 3(a) to highlight clearer trends towards saturation. The modulation depth was up to 85%. In contrast, the width of the EIT window can be switched from narrow to wide with a high pump power of 360 mW, accompanied by Q from 8 to 2.56.
Fig. 2. Dual-functional EIT modulation behaviors under various optical pump powers. (a) Experimental data of EIT modulation with different level pump powers. The picture at the top displayed behavior of EIT resonance amplitude modulation, whereas the picture at the bottom displayed the behavior of EIT window modulation. (b) Corresponding simulation results of EIT modulation with different Si conductivities. (c) Experimentally extracted width of EIT transparency window with different pump powers from picture a. (d) Simulative extracted width of EIT transparency window with different Si conductivities from picture b. Transition region demonstrated no EIT resonance concluding from the flat transmission spectra in EIT window.
Fig. 3. (a) Normalized EIT resonance amplitude as a function of pump powers extracted from Fig. 2(a) in the main text for Function I, and the maximal modulation depth is up to 85%. The inset defines the EIT resonance amplitude with $\delta T = {T_{peak}} - {T_{dip1}}$. (b) Experimental group delay varied from optical pump powers. The picture at the top showed the modulation of group delay in the case of low-level pump power, whereas the picture at the bottom showed the molecularization degrees for various group delays in the case of high-level pump power.
To better illustrate the experimental results, a numerical analysis was necessary to study the EIT modulation and molecularization approach. In this study, we simulated the transmission amplitude of the terahertz wave as a function of the conductivity of the Si semiconductor using the finite element method (FEM), as shown in Fig. 2(b). Without loss of generality, to merely indicate the tendency between transmission spectra of terahertz waves and Si conductivity, the strict correspondence of Si conductivity in the light of optical pump power was unnecessary (although this relationship can be obtained by THz time-domain system spectra [39]). In the case of low conductivity, the EIT resonance amplitude decreased to vanish as the Si conductivity increased from 1 to 400 S/m. In the case of high conductivity, a broader EIT window took shape, and the EIT resonance amplitude increased with increasing conductivity. In addition, the widths of the EIT window are extracted from Fig. 2(d). All simulation results were consistent with the experimental data, indicating that EIT modulation originated from changes in Si conductivity excited by photogenerated carrier concentration in the experiment. Mainly, we contribute to the differences between experiment and simulation to the fabrication errors and material characteristics. As one of the characteristics of the EIT effect, the slow-light effect has garnered public attention for its intriguing physical phenomena and huge potential applications. In the present study, we characterized the slow-light effect by group delay, $\mathrm{\Delta }{t_g} = \; - d\varphi /d\omega $, where $\omega = 2\pi f$ is the terahertz angular frequency. In addition, the group delay spectra varied with the injected optical powers, as shown in Fig. 3(b), demonstrating efficient EIT effect modulation.
With regard to ultrafast EIT modulation, we utilized optical pump terahertz probe (OPTP) technology to demonstrate the two functions of the proposed metadevice. Function I, the ultrafast modulation of the EIT resonance amplitude, is displayed in Fig. 4(a) with a pump power of 16 mW. The relative time delay in the arrival of the pump light and terahertz pulse at the metasurface was precisely controlled via the pump-probe delay stage. The time-domain terahertz transmission map in Fig. 4(a) shows the ultrafast modulation of the EIT resonance amplitude within 2 ns. At a relative time delay of 0 ns, the EIT effect was performed without a pump for maximal resonance amplitude and thereafter started to weaken. For clarity, part of the temporal evolution of the terahertz transmission spectra in the metadevice under the 16-mW pump light was extracted from Fig. 4(a), as shown in Fig. 4(c). At the time delay of 0.1 ns, the EIT resonance almost entirely disappeared. The inset in Fig. 4(c) shows the embedded Si-brides in meta-atoms with light blue color, indicating that no molecularization took place at a power level of 16 mW. We could conclude that the entire ultrafast EIT switching lasted approximately 1.5 ns owing to the full-recovery EIT resonance at a relative time delay of 1.5 ns. In addition, we demonstrated the ultrafast modulation of group delay behavior mapped in Fig. 4(e), which exhibited the same tendency as the resonance amplitude modulation.
Fig. 4. Temporal evolution dynamics of ultrafast terahertz modulation behaviors under the determinate pump power. (a) Transmission maps for pump power of 16 mW with different time delays. The dotted line showed the frequency of EIT central window and its resonance amplitude. (b) Transmission maps for pump power of 360 mW with different time delays. The dotted line showed the width of EIT transparency window. (c) Extracted transmission spectra as a function of time delay from 0 to 0.3 ns from picture a. The inset revealed the metadevice without molecularization under the low-conductivity Si-bridge. (d) Extracted transmission spectra as a function of time delay from 0 to 0.3 ns from picture b. The inset revealed the metadevice with high-level molecularization under the high-conductivity Si-bridge. (e) Group delay maps for pump power of 16 mW with different time delays. (f) Group delay maps for pump power of 360 mW with different time delays.
We will now discuss Function II in the designed metadevice, ultrafast modulation of the EIT window, with a pump power of 360 mW. Similarly, the terahertz transmission amplitude against the frequency and pump-probe time delay is described by a color map, as shown in Fig. 4(b). At a time delay of 0 ns, the EIT effect exhibited a clear resonance with a narrow transparency window, $\delta f = 0.16\; THz$. To further highlight the ultrafast modulation of Function II, we fit the window spectrum curve using the Fano profile to obtain the quality factor at this moment. The process of quality-factor fitting extraction will be discussed in the next section. As mentioned previously, the quality factor Q ≈ 8 at a time delay of 0 ns with an optical pump of 360 mW, implying that the terahertz pulse arrived at the metadevice before the arrival of the pump beam and no photogenerated carrier was excited about THz EIT modulation. As the delay time increases, a novel broad EIT window is formed after the disappearance of the EIT resonance in the narrow window. For more details, we extracted the portion of temporal evolution of terahertz transmission spectra at 360 mW from Fig. 4(b), as described in Fig. 4(d). A broad EIT window arose at the time delay of 0.1 ns with Q ≈ 2.56, $\delta f = 0.32\; THz$. The inset in Fig. 4(d) shows the brown Si bridge connected to meta-atoms, which indicates giant molecularization for a high pump power of 360 mW. Thus, ultrafast Q-switched modulation was realized in metadevice function II. Notably, the narrow-window EIT resonance was not yet fully recovered at a time delay of 2 ns, because of the longer carrier relaxation for higher pump fluence. In the following discussion, we studied the carrier relaxation time of the Si conductivity for various pump fluences. Limited to the experimental test conditions, only a 2 ns time delay could be measured for the ultrafast dynamic process. Similarly, ultrafast group delay modulation is described by the color map in Fig. 4(f), exhibiting the same behavior as the terahertz transmission spectra for the temporal evolution of ultrafast molecularization. Moreover, performance comparison with the previously reported several typical terahertz metadevices is carried out. It can be seen from Table 1 that, our work demonstrates an all-optically multifunctional terahertz modulator and realizes the active Q-switching function.
Table 1. Performance comparison with several typical terahertz metadevices
To further elucidate the physical mechanism of the modulation based on the EIT resonance amplitude and the molecularization evolution at various pump powers, we performed the simulation of near-field electric magnitude distributions as a function of Si conductivity, as shown in Fig. 5. In this simulation, through FEM, we set the conductivity of gold meta-atoms as $4 \times {10^7}$ S/m with a thickness of 200 nm. The Si conductivities were given as 1 S/m without light excitation and 20000 S/m with 360-mW power pumping. The refractive indices of the sapphire substrate and Si-bridge were set to 3.2 and 3.4, respectively. Periodic boundary conditions were used in the x and y directions, and a pair of perfectly matched layers were used along the propagation direction of the terahertz pulse (z-axis). A plane wave with y-axis polarization was employed to excite the metadevice. The plane of the electric field was monitored above a metasurface of 1 μm. Figure 5(a–c) describe Function I of the metadevice and the modulation of the EIT resonance amplitude with different relatively low Si conductivities. As observed from the simulation maps, the electric magnitude decreased with increasing Si conductivity, implying a waning interaction between bright modes and incident THz pulses. In Function II, molecularization occurred at higher electrical conductivity, and the near-field electric magnitudes are shown in Fig. 5(d–f). Note that the present electric field intensity was weaker than that of Function I. Because tremendous dissipative loss still existed, this could be explained by the extracted Q-factors. As the Si conductivity increased, the electric field magnitude in the cut wire and SRRs decreased while it increased in the off-state SRRs. The red dotted areas demonstrate this evolution, indicating a molecularization process from low to high levels. Overall, the simulation results of the electric field well explained the EIT modulation behaviors in the experiment.
Fig. 5. Numerically extracted electric field amplitudes at EIT window central frequency 0.65 THz for varying Si conductivities ranging from 1 to 20000 S/m. (a–c) Electric field amplitude distributions for low-level conductivity, which indicated modulation of EIT resonance amplitude, Function I in our proposed metadevice. (d–f) Electric field amplitude distributions for high-level conductivity, which indicated modulation of EIT window width, Function II.
3. Discussion
To analyze the physical mechanism of the EIT effect, the EIT modes coupling of various modulation functions is studied. In the case of 0-mW pump, no molecularization of meta-atoms occurred. When we remove the cut-wire and SRRs while retaining the off-state SRRs in the unit, the simulated transmission spectrum shows that the off-state SRRs have no contribution to the EIT formation, shown in Fig. 6(a). Then, a cut-wire is added. The simulated transmission curve displays a dip at 0.65 THz, shown in Fig. 6(b), which means a bright mode resonance. And the electrical field distribution shows a dipole resonance. In addition, we add SRRs instead of cut-wire and obtain its transmission spectrum. The result shows another dip at 0.69 THz, shown in Fig. 6(c), indicating another bright mode resonance. As we can see from Fig. 6(d), the full structure shows a transmission spectrum with an EIT curve. Thus, we can conclude that the EIT feature in the case of no molecularization of meta-atoms is based on bright-bright modes coupling. In the case of 360-mW pump, a new EIT resonance is formed by molecularization of meta-atoms. The SRRs convert into CRRs due to the molecularization, which does not provide terahertz resonance modes as shown in Fig. 6(e). When we remove the off-state SRRs while retaining the cut-wire and SRRs in the unit, the simulated transmission spectrum shows a dip at 0.67 THz, which is from the bright resonance mode of cut-wire, shown in Fig. 6(f). While we remove the cut-wire and retain the off-state SRRs and SRRs in the unit, there are no resonance modes excited, shown in Fig. 6(g), because the dark mode cannot interact directly with terahertz wave. Only when the cut-wire and off-state SRRs exist at the same time, can the EIT effect be triggered, as we can see from Fig. 6(h). Hence, we can conclude that the EIT feature in the case of molecularization of meta-atoms is based on bright-dark modes coupling.
Fig. 6. Explanation of physical origin of the EIT feature. (a-d) In the case of no molecularization, the EIT effect is based on bright-bright modes coupling. The transmission curve and near-field distribution of (a) off-state SRRs, (b) cut-wire and off-state SRRs, (c) SRRs and off-state SRRs, and (d) full structure. (e-h) In the case of molecularization, the EIT effect is based on bright-dark modes coupling. The transmission curve and near-field distribution of (a) SRRs, (b) cut-wire and SRRs, (c) off-state SRRs and SRRs, and (d) full structure.
Benefiting from the short carrier life of Si semiconductors, ultrafast terahertz modulation, including both EIT resonance amplitude and quality factor, can be realized at the nanoscale. For Function I, the normalized resonance amplitude, which is described in Fig. 3(a), exhibited an 85% modulation depth at a pump power of 16 mW. The slow light effect at the transparency window could be reached at 1.9 ps. For Function II, ultrafast Q-factor modulation, we extract the Q-factors of the EIT transparency window by fitting to the following equation: [46,47]
where A denotes a constant, q the asymmetry parameter, $\mathrm{\Omega } = 2({f - {f_{Fano}}} )/\delta f$, ${f_{Fano}}$ the central window resonance frequency, and $\delta f$ the bandwidth of the resonance. Therefore, we can readily calculate the Q-factor as $Q = \; {f_{Fano}}/\delta f$. Without loss of generality, we fit the spectra curves of 0 and 360 mW, as shown in Fig. 7, to illustrate the Q-factor modulation, because the EIT transmission spectrum at 0 mW represents the arrival of the optical pump after the THz pulse. Then, the Q factors before and after modulation were well expressed. The manifestation of different modulation speeds is shown in Fig. 4, and we traced this appearance to the Si carrier lifetimes with various pump powers. To better understand this phenomenon, we characterized the ultrafast relaxation time of the pure Si film with femtosecond pulse laser pumping, as shown in Fig. 8. Particularly, we selected optical powers of 2, 4, 8, and 16 mW to excite pure Si film to study the rules of photogenerated carrier relaxation. On the basis of different carrier concentrations in the time scale, we utilized OPTP to experimentally measure the absolute transmission change, $\mathrm{\Delta }T(t )= T(t )- \; {T_0}$, where $T(t )$ denotes the transmission of the THz pulse through the Si in real time, whereas ${T_0}$ is that of the Si semiconductor in the absence of pump power. The negative differential transmission of the pure Si film, $- \mathrm{\Delta }T/{T_0}$, exhibited a distinct exponential delay. Then, we fit the experimentally measured data by a mono-exponential decay function, determined as follows: [48] where ${A_0}$ denotes a constant amplitude without decay, ${A_1}$ the amplitude determining the weight of the exponential function with the delay time constant τ, and ${t_0}$ refers to the time zero of the exponential fit. The theoretical fitting results are shown in Fig. 8(a), and the delay time constant of the Si semiconductor as a function of pump power can be extracted, as shown in Fig. 8(b). As observed from the figure, the delay time τ increased with an increase in the optical pump fluence. Namely, the longer relaxation time of the Si semiconductor would be for higher pump power excitation, which was an excellent explanation for the distinction of modulation speeds between Functions I and II.Fig. 7. Q factor extraction by Fano fitting at the EIT window. (a) Experimental transmittance spectrum and Fano fit at the pump power of 0 mW, and the extracted Q ≈ 8. (b) Experimental transmittance spectrum and Fano fit at the pump power of 360 mW, and the extracted Q ≈ 2.56.
Fig. 8. Pumping power-dependent relaxation dynamics of the Si film. (a) Negative differential transmission for pump-probe time delays measured at a series of specified optical pump powers. (b) Theoretically fitting time delay constants as a function of optical pump powers.
4. Conclusion
In conclusion, we designed a terahertz metadevice based on an active-tuning metasurface structure supported by the EIT effect and experimentally demonstrated dual-functionality ultrafast modulation. We first displayed Function I at a low pump power, and the EIT resonance amplitude can be efficiently tuned. The ultrafast switching of the EIT resonance amplitude was performed at a pump power of 16 mW. Thereafter, Function II was displayed at a high pump power of 360 mW with high-level molecularization for changing the spatially connected structure, which achieved ultrafast Q-factor/EIT window modulation. To the best of our knowledge, this is the first time that active EIT window modulation is achieved by altering the Q factor, but remains unchanged. We explored this ultrafast process by utilizing OPTP technology, and the modulation speed of both functions was in the order of a few nanoseconds. In addition, we performed a near-field simulation to better understand the EIT modulation and molecularization processes. Overall, our work provides a novel path to attain multifunctional terahertz modulation based on active metasurfaces and enriches the applications of terahertz metadevices.
5. Methods
Sample preparation: We performed the fabrication process of the metadevice structure on a commercially available SOS wafer composed of a 485-nm-thick intrinsic epitaxial Si layer formed on a 500-μm-thick R-plane sapphire substrate. According to the manufacturer, the front surface of the SOS wafer was Epi polished with Ra < 0.3 nm, and the intrinsic resistivity of Si layer was 100 ${\mathbf \Omega }\cdot {\boldsymbol{cm}}$. The fabrication of the metadevice sample started from the pattern aligned to Si photosensitive mesa. Then an E-beam evaporation was utilized to deposit a 10-nm-thick chromium adhesion layer and 200-nm-thick gold material in sequence. A standard UV-lithography technique for patterning the well-designed metasurface structure was performed on the Si layer, followed by a lift-off process. After that, the entire exposed Si layer beyond the metallic regions was subsequently wiped off by reactive ion etching. Next, the unwanted metal on the Si bridge was removed by a second photolithography step. Due to the patterns of photosensitive mesa between the gap of metallic resonators, the assigned Si bridge was formed. Finally, the residual gold film was etched by potassium iodide solution and the chromium adhesion layer was subsequently etched by ammonium ceric nitrate solution. The entirely fabricated periodic Si-bridged metadevice covered an area of $\mathbf{5} \times \mathbf{5}\; {\boldsymbol m}{{\boldsymbol m}^{\mathbf 2}}$.
Terahertz transmission measurement: Optical pump terahertz probe technology from TuoTuo Technology (TTT-02-OPTP) was used to perform our experimental measurements for multifunctional metadevice. The terahertz pulse was generated and detected by nonlinear processes based on the 1-mm-thick ZnTe crystals. A Ti: sapphire regenerative amplifier system (Spectra Physics) was utilized to provide 800 nm (central wavelength) optical laser pulse with 1 kHz repetition and around 100-fs pulse width. The optical laser was divided into two beams. One path of them was used to excite and detect the terahertz pulse and the other was used to pump the Si semiconductor for photogenerated carriers. The spot size of the pump light was 5 mm, which was larger than the terahertz beam spot (∼ 3 mm), ensuing uniform photoexcitation over the metadevice. The terahertz time-domain waveform is recorded by the motion of translational stage of the gated beam. And then these time-domain signals were converted to frequency domain spectra by executing standard Fourier transform. Thus, the frequency-dependent terahertz transmission was obtained by dividing the sampling transmission spectrum ($|{{E_S}(\omega )} |$) by the reference substrate transmission spectrum ($|{{E_R}(\omega )} |$), described as $T(\omega )= |{{E_S}(\omega )/{E_R}(\omega )} |$. The optical pump power was altered through a combination of polarizer and halfwave plate.
Funding
Science Fund for Distinguished Young Scholars of Hunan Province (2020JJ2036); Scientific Researches Foundation of National University of Defense Technology (ZK18-01-03, ZK18-03-22, ZK18-03-36); National Natural Science Foundation of China (11802339, 11804387, 11805276, 11902358, 61801498, 61805282, 62075240).
Disclosures
The authors declare no conflict of interests.
Data availability
The 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. J. F. O’Hara, S. Ekin, W. Choi, and I. Song, “A Perspective on Terahertz Next-Generation Wireless Communications,” Technologies 7(2), 43 (2019). [CrossRef]
2. T. Nagatsuma, G. Ducournau, and C. C. Renaud, “Advances in terahertz communications accelerated by photonics,” Nat. Photonics 10(6), 371–379 (2016). [CrossRef]
3. K. Fan, J. Y. Suen, X. Liu, and W. J. Padilla, “All-dielectric metasurface absorbers for uncooled terahertz imaging,” Optica 4(6), 601–604 (2017). [CrossRef]
4. J. Hawecker, V. Pistore, A. Minasyan, K. Maussang, J. Palomo, I. Sagnes, J. M. Manceau, R. Colombelli, J. Tignon, J. Mangeney, and S. S. Dhillon, “Cavity-based photoconductive sources for real-time terahertz imaging,” Photonics Res. 8(6), 858–863 (2020). [CrossRef]
5. S. Zhong, “Progress in terahertz nondestructive testing: A review,” Front. Mech. Eng. 14(3), 273–281 (2019). [CrossRef]
6. J. Dong, A. Locquet, and D. S. Citrin, “Terahertz Quantitative Nondestructive Evaluation of Failure Modes in Polymer-Coated Steel,” IEEE J. Sel. Top. Quantum Electron. 23(4), 1–7 (2017). [CrossRef]
7. P. H. Siegel, “Terahertz technology in biology and medicine,” IEEE Trans. Microwave Theory Tech. 52(10), 2438–2447 (2004). [CrossRef]
8. W. J. Choi, G. Cheng, Z. Huang, S. Zhang, T. B. Norris, and N. A. Kotov, “Terahertz circular dichroism spectroscopy of biomaterials enabled by kirigami polarization modulators,” Nat. Mater. 18(8), 820–826 (2019). [CrossRef]
9. Y. Cheng, Y. Wang, Y. Niu, and Z. Zhao, “Concealed object enhancement using multi-polarization information for passive millimeter and terahertz wave security screening,” Opt. Express 28(5), 6350–6366 (2020). [CrossRef]
10. M. Tong, Y. Hu, Z. Wang, T. Zhou, X. Xie, X. A. Cheng, and T. Jiang, “Enhanced Terahertz Radiation by Efficient Spin-to-Charge Conversion in Rashba-Mediated Dirac Surface States,” Nano Lett. 21(1), 60–67 (2021). [CrossRef]
11. M. Tong, Y. Hu, X. Xie, X. Zhu, Z. Wang, X. A. Cheng, and T. Jiang, “Helicity-dependent THz emission induced by ultrafast spin photocurrent in nodal-line semimetal candidate Mg3Bi2,” Opto-Electron. Adv. 3(12), 20002301 (2020). [CrossRef]
12. A. V. Balakin, J.-L. Coutaz, V. A. Makarov, I. A. Kotelnikov, Y. Peng, P. M. Solyankin, Y. Zhu, and A. P. Shkurinov, “Terahertz wave generation from liquid nitrogen,” Photonics Res. 7(6), 678–686 (2019). [CrossRef]
13. Z. Chen, X. Chen, L. Tao, K. Chen, M. Long, X. Liu, K. Yan, R. I. Stantchev, E. Pickwell-MacPherson, and J. B. Xu, “Graphene controlled Brewster angle device for ultra broadband terahertz modulation,” Nat. Commun. 9(1), 4909 (2018). [CrossRef]
14. N. Kakenov, M. S. Ergoktas, O. Balci, and C. Kocabas, “Graphene based terahertz phase modulators,” 2D Mater. 5(3), 035018 (2018). [CrossRef]
15. W. He, M. Tong, Z. Xu, Y. Hu, X.A. Cheng, and T. Jiang, “Ultrafast all-optical terahertz modulation based on an inverse-designed metasurface,” Photonics Res. 9(6), 1099–1108 (2021). [CrossRef]
16. N. Wang, S. Cakmakyapan, Y.-J. Lin, H. Javadi, and M. Jarrahi, “Room-temperature heterodyne terahertz detection with quantum-level sensitivity,” Nat Astron 3(11), 977–982 (2019). [CrossRef]
17. B. Paulillo, S. Pirotta, H. Nong, P. Crozat, S. Guilet, G. Xu, S. Dhillon, L. H. Li, A. G. Davies, E. H. Linfield, and R. Colombelli, “Ultrafast terahertz detectors based on three-dimensional meta-atoms,” Optica 4(12), 1451–1456 (2017). [CrossRef]
18. I. Liberal and N. Engheta, “Near-zero refractive index photonics,” Nat. Photonics 11(3), 149–158 (2017). [CrossRef]
19. H. Tang, C. DeVault, S. A. Camayd-Muñoz, Y. Liu, D. Jia, F. Du, O. Mello, D. I. Vulis, Y. Li, and E. Mazur, “Low-Loss Zero-Index Materials,” Nano Lett. 21(2), 914–920 (2021). [CrossRef]
20. M. Liu, E. Plum, H. Li, S. Li, Q. Xu, X. Zhang, C. Zhang, C. Zou, B. Jin, J. Han, and W. Zhang, “Temperature-Controlled Optical Activity and Negative Refractive Index,” Adv. Funct. Mater. 31(14), 2010249 (2021). [CrossRef]
21. S. Kocaman, M. S. Aras, P. Hsieh, J. F. McMillan, C. G. Biris, N. C. Panoiu, M. B. Yu, D. L. Kwong, A. Stein, and C. W. Wong, “Zero phase delay in negative-refractive-index photonic crystal superlattices,” Nat. Photonics 5(8), 499–505 (2011). [CrossRef]
22. L. Cong, V. Savinov, Y. K. Srivastava, S. Han, and R. Singh, “A Metamaterial Analog of the Ising Model,” Adv. Mater. 30(40), 1804210 (2018). [CrossRef]
23. Y. Hu, M. Tong, Z. Xu, X. Cheng, and T. Jiang, “Spatiotemporal Terahertz Metasurfaces for Ultrafast All-Optical Switching with Electric-Triggered Bistability,” Laser Photonics Rev. 15(3), 2000456 (2021). [CrossRef]
24. X. Zhao, C. Chen, K. Kaj, I. Hammock, Y. Huang, R. D. Averitt, and X. Zhang, “Terahertz investigation of bound states in the continuum of metallic metasurfaces,” Optica 7(11), 1548–1554 (2020). [CrossRef]
25. T. Dong, S. Li, M. Manjappa, P. Yang, J. Zhou, D. Kong, B. Quan, X. Chen, C. Ouyang, F. Dai, J. Han, C. Ouyang, X. Zhang, J. Li, Y. Li, J. Miao, Y. Li, L. Wang, R. Singh, W. Zhang, and X. Wu, “Nonlinear THz-Nano Metasurfaces,” Adv. Funct. Mater. 31(24), 2100463 (2021). [CrossRef]
26. L. Cong, Y. K. Srivastava, H. Zhang, X. Zhang, J. Han, and R. Singh, “All-optical active THz metasurfaces for ultrafast polarization switching and dynamic beam splitting,” Light: Sci. Appl. 7(1), 28 (2018). [CrossRef]
27. Y. Hu, J. You, M. Tong, X. Zheng, Z. Xu, X. Cheng, and T. Jiang, “Pump-Color Selective Control of Ultrafast All-Optical Switching Dynamics in Metaphotonic Devices,” Adv. Sci. 7(14), 2000799 (2020). [CrossRef]
28. Z. Fan, Z. Geng, X. Lv, Y. Su, Y. Yang, J. Liu, and H. Chen, “Optical Controlled Terahertz Modulator Based on Tungsten Disulfide Nanosheet,” Sci. Rep.7(1), 14828 (2017). [CrossRef]
29. H. Jung, H. Jo, W. Lee, B. Kim, H. Choi, M. S. Kang, and H. Lee, “Electrical Control of Electromagnetically Induced Transparency by Terahertz Metamaterial Funneling,” Adv. Opt. Mater. 7, 1801205 (2019). [CrossRef]
30. T.-T. Kim, H.-D. Kim, R. Zhao, S. S. Oh, T. Ha, D. S. Chung, Y. H. Lee, B. Min, and S. Zhang, “Electrically Tunable Slow Light Using Graphene Metamaterials,” ACS Photonics 5(5), 1800–1807 (2018). [CrossRef]
31. P. Q. Liu, I. J. Luxmoore, S. A. Mikhailov, N. A. Savostianova, F. Valmorra, J. Faist, and G. R. Nash, “Highly tunable hybrid metamaterials employing split-ring resonators strongly coupled to graphene surface plasmons,” Nat. Commun. 6(1), 8969 (2015). [CrossRef]
32. Y. Hu, T. Jiang, H. Sun, M. Tong, J. You, X. Zheng, Z. Xu, and X. Cheng, “Ultrafast Frequency Shift of Electromagnetically Induced Transparency in Terahertz Metaphotonic Devices,” Laser Photonics Rev. 14(3), 1900338 (2020). [CrossRef]
33. M. Manjappa, Y. K. Srivastava, A. Solanki, A. Kumar, T. C. Sum, and R. Singh, “Hybrid Lead Halide Perovskites for Ultrasensitive Photoactive Switching in Terahertz Metamaterial Devices,” Adv. Mater. 29(32), 1605881 (2017). [CrossRef]
34. P. Pitchappa, M. Manjappa, C. P. Ho, R. Singh, N. Singh, and C. Lee, “Active Control of Electromagnetically Induced Transparency Analog in Terahertz MEMS Metamaterial,” Adv. Opt. Mater. 4(4), 541–547 (2016). [CrossRef]
35. T. Kan, A. Isozaki, N. Kanda, N. Nemoto, K. Konishi, H. Takahashi, M. Kuwata-Gonokami, K. Matsumoto, and I. Shimoyama, “Enantiomeric switching of chiral metamaterial for terahertz polarization modulation employing vertically deformable MEMS spirals,” Nat. Commun. 6(1), 8422 (2015). [CrossRef]
36. Y. Wang, Z. Han, Y. Du, and J. Qin, “Ultrasensitive terahertz sensing with high-Q toroidal dipole resonance governed by bound states in the continuum in all-dielectric metasurface,” Nanophotonics 10(4), 1295–1307 (2021). [CrossRef]
37. H. Jung, J. Koo, E. Heo, B. Cho, C. In, W. Lee, H. Jo, J. H. Cho, H. Choi, M. S. Kang, and H. Lee, “Electrically Controllable Molecularization of Terahertz Meta-Atoms,” Adv. Mater. 30(31), 1802760 (2018). [CrossRef]
38. W. Liang, Y. Jiang, J. Guo, N. Li, W. Qiu, H. Yang, Y. Ji, and S. N. Luo, “Van der Waals heteroepitaxial VO2/mica films with extremely low optical trigger threshold and large THz field modulation depth,” Adv. Opt. Mater. 7(20), 1900647 (2019). [CrossRef]
39. H. J. Joyce, J. L. Boland, C. L. Davies, S. A. Baig, and M. B. Johnston, “A review of the electrical properties of semiconductor nanowires: insights gained from terahertz conductivity spectroscopy,” Semicond. Sci. Technol. 31(10), 103003 (2016). [CrossRef]
40. L. Cong, M. Manjappa, N. Xu, I. Al-Naib, W. Zhang, and R. Singh, “Fano Resonances in Terahertz Metasurfaces: A Figure of Merit Optimization,” Adv. Opt. Mater. 3(11), 1537–1543 (2015). [CrossRef]
41. M. Liu, Z. Tian, X. Zhang, J. Gu, C. Ouyang, J. Han, and W. Zhang, “Tailoring the plasmon-induced transparency resonances in terahertz metamaterials,” Opt. Express 25(17), 19844–19855 (2017). [CrossRef]
42. H.-F. Zhu, L.-H. Du, J. Li, Q.-W. Shi, B. Peng, Z.-R. Li, W.-X. Huang, and L.-G. Zhu, “Near-perfect terahertz wave amplitude modulation enabled by impedance matching in VO2 thin films,” Appl. Phys. Lett. 112(8), 081103 (2018). [CrossRef]
43. Y. Hu, T. Jiang, J. Zhou, H. Hao, H. Sun, H. Ouyang, M. Tong, Y. Tang, H. Li, J. You, X. Zheng, Z. Xu, and X. Cheng, “Ultrafast Terahertz Frequency and Phase Tuning by All-Optical Molecularization of Metasurfaces,” Adv. Opt. Mater. 7(22), 1970084 (2019). [CrossRef]
44. Y. Hu, T. Jiang, J. Zhou, H. Hao, H. Sun, H. Ouyang, M. Tong, Y. Tang, H. Li, J. You, X. Zheng, Z. Xu, and X. Cheng, “Ultrafast terahertz transmission/group delay switching in photoactive WSe2-functionalized metaphotonic devices,” Nano Energy 68(3), 104280 (2020). [CrossRef]
45. Y. Hu, M. Tong, Z. Xu, X. Cheng, and T. Jiang, “Bifunctional Spatiotemporal Metasurfaces for Incident Angle-Tunable and Ultrafast Optically Switchable Electromagnetically Induced Transparency,” Small 17(21), 2006489 (2021). [CrossRef]
46. M. F. Limonov, M. V. Rybin, A. N. Poddubny, and Y. S. Kivshar, “Fano resonances in photonics,” Nat. Photonics 11(9), 543–554 (2017). [CrossRef]
47. J. Niu, Y. Zhai, Q. Han, J. Liu, and B. Yang, “Resonance-trapped bound states in the continuum in metallic THz metasurfaces,” Opt. Lett. 46(2), 162–165 (2021). [CrossRef]
48. M. ElKabbash, A. R. Rashed, B. Kucukoz, Q. Nguyen, A. Karatay, G. Yaglioglu, E. Ozbay, H. Caglayan, and G. Strangi, “Ultrafast transient optical loss dynamics in exciton–plasmon nano-assemblies,” Nanoscale 9(19), 6558–6566 (2017). [CrossRef]
