Open Access
Add to BibSonomy
Abstract
We propose and experimentally demonstrate an all-optical microwave filter with tunable central frequency and bandwidth based on two cascaded silicon opto-mechanical microring resonators (MRRs). Due to the Vernier effect, transmission spectrum of the cascaded MRRs is a series of notch bimodal distribution. In the case of intensity modulation with optical double-sideband (ODSB) signals, the optical carrier is fixed between the two resonant peaks of one notch bimodal distribution. By injecting two pump powers to control the above two resonance red-shifts based on the nonlinear effects in opto-mechanical MRRs, the frequency intervals between the optical carrier and the two resonances could be flexibly manipulated for tunable microwave processing. In the experiment, with the highest required pump powers of 1.65 mW and 0.96 mW, the central frequency and bandwidth of the notch microwave photonic filter (MPF) could be tuned from 5 GHz to 36 GHz and 6.7 GHz to 10.3 GHz, respectively. The proposed opto-mechanical device is competent to process microwave signals with dominant advantages of all-optical control, compact footprint, wide tuning range and low-power consumption, which has significant applications in on-chip microwave systems.
© 2017 Optical Society of America
1. Introduction
With significant advantages of large bandwidth, superior tunability and reconfigurability, microwave photonic filters (MPFs) have attracted great interest in the past decades [1–5]. In radar and wireless communication systems, MPFs with tunable central frequency and bandwidth are highly desired to process random and unpredictable microwave signals [6–8]. Different implementations of tunable MPFs have been demonstrated by fiber optical technology [9, 10], such as fiber Bragg grating [11], optical frequency comb [12], Raman fiber laser [13] and high-birefringence linear chirped grating [14]. To pursue better integration and reliability, nanotechnology has been widely used in microwave systems [15–17]. Due to the dominant advantages of silicon-on-insulator (SOI) technology, such as high refractive index difference and complementary metal-oxide semiconductor (CMOS) compatibility, integrated circuits based on silicon nano-photonic technology has become one of the most promising photonic integration platforms in the last decade [18–20]. Passive silicon waveguide structures have shown an unprecedented reduction in device footprint [21]. All-optical microwave chips based on the SOI technology could largely reduce the system complexity and power consumption [22–26]. To date, only several on-chip schemes have demonstrated all-optical microwave filters with tunable central frequency and bandwidth, such as by utilizing stimulated Brillouin scattering (SBS) effect [27]. However, the relatively long waveguide length and high pump power for SBS effect limit their practical applications in optical systems [28–30].
Recently, silicon opto-mechanical structures have attracted widespread attentions [31–34]. Especially in free-hanging microring resonators (MRRs), the nonlinear effects (including the thermo-optic effect and opto-mechanical effect) could be activated by lower pump powers. The thermo-optic effect plays an important role to induce resonance red-shifts of free-hanging MRRs for reducing the pump power. As the removal of the oxide substrate could significantly decrease the device heatsink, the temperature rise of opto-mechanical microring would be much higher. On the other hand, the opto-mechanical effect could be excited in free-hanging MRRs by low pump powers, because the gradient of optical field is significantly enhanced in microring resonators (MRRs) [35]. The generated optical gradient force between the free-hanging arc and underneath substrate could cause nanometer or even micrometer mechanical deformations of the microring, which induces resonance red-shifts of the MRRs [36, 37]. Therefore, the opto-mechanical MRRs provide an all-optical control and low-power solution to process microwave signals in pure silicon platform [38]. In our previous work, we have realized MPFs with tunable central frequency and bandwidth through adjusting the laser wavelengths [39] or electrodes [40]. Subsequently, in order to improve the tunable method, we have experimentally demonstrated all-optical microwave filters with continuously tunable central frequency based on an opto-mechanical MRR [41]. However, as the spectrum shape of the passive silicon MRR is hard to be changed, the MPF bandwidth is not tunable. Moreover, the scheme requires external assistance of an optical filter to realize single sideband modulation which increases the system complexity and power consumption. Therefore, an effective solution with all-optical control, low-power consumption and compact size to realize MPFs with tunable central frequency and bandwidth is highly desired.
In this paper, we experimentally demonstrate an all-optical microwave filter based on cascaded opto-mechanical MRRs. Due to the Vernier effect, transmission spectrum of the cascaded MRRs is a series of notch bimodal distribution. By injecting two pump powers to control the corresponding MRR, the frequency intervals of the bimodal distribution could be flexibly manipulated. In the case of optical double-sideband (ODSB) modulation, the central frequency and bandwidth of the MPF could be tuned from 5 GHz to 36 GHz and 6.7 GHz to 10.3 GHz, respectively. The highest required pump power for the tunable central frequency and bandwidth are 1.65 mW and 0.96 mW, respectively. The proposed scheme based on silicon opto-mechanical devices provides an all-optical control approach to realizing MPFs with low power consumption, wide tuning range and compact footprint, which is significant for low-power all-optical microwave chips.
2. Operation principle
The tuning mechanism of the all-optical microwave filter is based on the nonlinear effects in opto-mechanical MRRs to manipulate the device spectrum. As shown in Fig. 1(a), half arc of the MRR is free-hanging by removing the underneath oxide substrate. When the pump light at the resonance wavelength λ0 is injected into the silicon device, the nonlinear effects including the opto-mechanical effect and thermo-optic effect would result red-shifts of the MRR spectrum. For the opto-mechanical effect, the evanescent fields between the MRR free-hanging arc and the substrate would induce optical mechanical nonlinearity [42], as shown in Fig. 1(b). In this case, the free-hanging ring is bent downwards to the oxide substrate driven by the generated optical gradient force. Due to the mechanical deformations of the MRR, the ring effective length becomes larger which causes the red-shifts of the MRR resonances [43].
Fig. 1 (a) Schematic diagram of the free-hanging MRR. (b) Cross-sectional illustration of the deflected MRR influenced by the optical gradient force.
The resonant wavelength red-shift δλ1 induced by the opto-mechanical effect can be expressed as [38]
where is the opto-mechanical tuning efficiency, neff is the effective index of the free-hanging MRR, g represents the waveguide separation between the free-hanging arc and the substrate, Pm is the circulating pump power on the ring for opto-mechanical effect, and k is the beam stiffness.On the other hand, the resonance red-shift δλ2 of the MRR owing to the thermo-optic effect can be described by [44]
where λ0 is the resonant wavelength, ng is the group index, Γth is the effective confinement factor corresponding to the thermo-optic effect, kth is silicon thermo-optic coefficient, Rth is the thermal resistance of the silicon ring resonator, and Pt is the optical pump power for the thermo-optic effect.Therefore, the total resonance red-shift δλ can be determined by
As the MRR resonance red-shifts δλ are proportional to the input pump power Ppump (mainly including Pm and Pt), the MRR spectrum red-shifts could be manipulated by adjusting the pump powers. Because the coupling efficiencies of the MRR resonance wavelengths are higher, the pump light aligned at the resonances could induce stronger opto-mechanical interaction and thermal effect which would cause the larger resonance red-shifts.
Figure 2 illustrates that the key device to realize tunable MPF is the two cascaded MRRs (R1 and R2) with different free spectral ranges (FSRs, FSR1 and FSR2). Due to the Vernier effect, transmission spectrum of the cascaded MRRs is a series of notch bimodal distribution, as shown in Fig. 2(a). The resonance wavelengths of R1 and R2 are λ1, λ3 (blue lines) and λ2, λ4 (red lines), respectively. The first and second bimodal regions are utilized as the working region (λ1, λ2) and pump region (λ3, λ4), respectively. Namely, the pump powers of λ3 and λ4 are used to manipulate the transmission spectrum of the working region (λ1, λ2). It should be noted that the wavelength of optical carrier λ0 is fixed at the left edge of resonance λ2, shown as the green arrow. In this case, the frequency intervals between the optical carrier λ0 and λ1, λ2 are written as f3 and f1, respectively. As the wavelength of the optical carrier λ0 is far away from the MRR resonances λ1 and λ2, its influence on the MRR resonance red-shifts could be negligible.
Fig. 2 Operation principle of the tunable MPF central frequency. (a) The transmission spectrum of the cascaded MRRs. The frequency intervals between the optical carrier and two resonances of the working region are equally tuned as (b) f1 (λ3: Pump on, λ4: Pump off), (d) f2 (λ3: Pump on, λ4: Pump on) and (f) f3 (λ3: Pump off, λ4: Pump on), respectively. The central frequencies of the MPF response are (c) f1, (e) f2 and (g) f3, respectively.
The tuning principle of the MPF central frequency is described as follows. When a random radio frequency (RF) is modulated onto the optical carrier by a Mach–Zehnder modulator (MZM) under small signal modulation, optical double-sideband (ODSB) signals could be generated. Firstly, as shown in Fig. 2(b), the pump power of λ3 is turned on while the pump power of λ4 is turned off. By adjusting the pump power of λ3, the initial resonance λ1 shifts to the location with a frequency interval f1 from λ0, shown as the blue dashed line. Thus a notch MPF with central frequency of f1 could be obtained, as shown in Fig. 2(c). Secondly, as shown in Fig. 2(d), the two pump powers of λ3 and λ4 are both turned on, in order to tune the two resonances of the working region with a same frequency interval f2 from λ0. In this case, Fig. 2(e) shows that the central frequency of the notch MPF could be tuned from f1 to f2. Thirdly, the pump power of λ3 is turned off. By adjusting the pump power of λ4, the initial resonance λ2 shifts to the location with a frequency interval f3 from λ0, shown as the red dashed line in Fig. 2(f). In this case, Fig. 2(g) shows that a notch MPF with central frequency of f3 could be realized. Therefore, by adjusting the pump powers of λ3 and λ4, the central frequency of the all-optical microwave filter could be tuned from f1 to f3.
The 3dB bandwidth of the MPF could be tuned at any central frequency ranging from f1 to f3. Figure 3 shows that f3 is chosen as the MPF central frequency to illustrate the bandwidth tuning process. As discussed in Fig. 2(f), by adjusting the pump power of λ4, the notch MPF with a narrow bandwidth could be obtained, shown as the red line in Fig. 2(g). Subsequently, the two pump powers are both turned on to slightly shift the two resonances of λ1 and λ2 with a frequency offset f4-f3. As shown in Fig. 3(a), the frequency intervals between the optical carrier λ0 and the two resonances are tuned as 2f3-f4 and f4, respectively. In this case, the notch MPF with a larger bandwidth could be realized with maintaining the MPF central frequency of f3, shown as the blue line in Fig. 3(b). Therefore, by adjusting the frequency offset, the 3dB bandwidth of the notch MPF could be tuned.
Fig. 3 Operation principle of the tunable MPF bandwidth. (a) The frequency intervals between the optical carrier and two resonances of the working region are tuned as 2f3-f4 and f4. λ3: Pump on, λ4: Pump on. (b) The MPF response with a larger bandwidth, shown as the blue line.
The frequency response of the notch MPF is deduced as follows. In order to realize ODSB modulation, the bias of the MZM is set at the linear region. The voltage offset is expressed by
where Vπ is the half-wave voltage of the MZM, α is the modulation depth and ωRF is the angular frequency of the RF signal.The output optical field of the MZM can be described as
where E0 is the input optical field and ω0 is the angular frequency of the optical carrier.In the case of small signal modulation and neglecting the high order sidebands, the output optical field of the MZM can be expressed by
where Jn is the nth-order Bessel function of the first kind and m is the intensity modulation index.Then, the generated ODSB signal is sent into the silicon chip. As each frequency component would multiply a different amplitude weight H(ω) of the cascade MRRs, the output optical field can be written as
With neglecting the J12 term, the alternative current (AC) term in the square-law photodetector (PD) can be described as
As the wavelength of optical carrier λ0 is fixed at the edge of the resonance λ2, H(ω0) is a constant. Equation (8) reveals that the MPF frequency response is mainly determined by the two sideband responses H(ω0-ωRF) and H(ω0 + ωRF), which are controlled by the two resonances of the working region. Therefore, by adjusting the frequency intervals (or the frequency offsets) of the two resonances, the MPF central frequency (or 3dB bandwidth) could be accordingly tuned.
In order to demonstrate the all-optical tunable MPF, we chose the commercial SOI wafer with a 220 nm thick silicon layer to fabricate the cascaded opto-mechanical MRRs. The device is fabricated by selective etching process [35], including two E-beam lithography (EBL), two inductively coupled plasma (ICP) etching and one hydrofluoric (HF) wet etching. Firstly, the cascaded MRRs and straight waveguides were transferred to photoresist by the first EBL and etched downwards for a set value (written as h1) through the first ICP etching. Secondly, only half of each MRR away from the straight waveguides were patterned and etched downwards for another depth (written as h2, h1 + h2 = 220 nm) through the second EBL and ICP etching [41]. In this case, half of each MRR has formed a corrosion window and the oxide substrates around them are exposed to the air. In contrast, the structure of the other half MRR is ridge waveguide with silicon slab layer to protect these fixed waveguides from later HF wet etching. Finally, HF acid wet etching was utilized to selectively undercut the oxide layer of the corrosion windows [45]. In this case, the free-hanging arcs of the cascaded MRRs could be released.
After designing the device structure and fabrication steps, we utilize the waveguide theory and simulations to optimize the device parameters, such as the MRR radius, ridge height (h1) and slab height (h2). Firstly, according to the beam propagation method [46], the bending loss of the silicon MRR could be calculated, shown as the blue dashed line (width = 400 nm), red dotted line (width = 450 nm) and green solid line (width = 500 nm) in Fig. 4(a). Considering the device footprint and waveguide single-mode condition, the radius and width of the MRR are designed as 20 μm and 450 nm, respectively. In this case, the MRR bending loss could be negligible. Then, we use the finite-element mode solver (COMSOL Multiphysics) to design the other parameters. When the ridge height h1 and slab height h2 are set as 190 nm and 30 nm respectively, the effective indexes of the ridge MRR, free-hanging MRR and straight waveguide are shown as the blue dashed line, red dotted line and green solid line in Fig. 4(b). In this case, the effective index differences of the above three structures could be negligible, which benefits the signal coupling and transmission between different waveguides. Moreover, the energy profiles of the fundamental modes in the straight waveguide, the ridge MRR and free-hanging MRR are shown as the three insets respectively.
Fig. 4 (a) Calculated bending loss under different MRR waveguide widths. (b) Effective indexes and energy profiles of the fundamental modes in different waveguide structures.
The resonance wavelengths of the MRR could be described as
where R are the microring radius and m is the order of the resonant mode. Utilizing Eq. (9) and the effective index in Fig. 4(b), we could realize the required MRR resonance wavelengths around the fiber communication wavelength (i.e. 1550 nm).Utilizing the above device parameters and fabrication method, we have fabricated the cascaded opto-mechanical MRRs, whose scanning electron microscope (SEM) image is shown in Fig. 5(a). The radii of the two MRRs (i.e. R1 and R2) are 20 μm and 20.17 μm, respectively. The coupling gaps between the straight waveguides and R1, R2 are 200 nm and 195 nm, respectively. Figure 5(b) shows the SEM image of R1. The width of the whole waveguides is 450 nm. The vertical grating couplers are employed to couple the optical signals from the fibers to the silicon chip. The period and duty cycle of the grating coupler are 610 nm and 50% respectively, as shown in Fig. 5(c).
Fig. 5 SEM images of (a) the cascaded opto-mechanical MRRs, (b) R1 and (c) the grating coupler, respectively.
The transmission response of the cascaded MRRs has been measured, as shown in Fig. 6(a). The device optical spectrum is a series of notch bimodal distributions with two different FSRs of 4.26 nm (R1) and 4.22 nm (R2), respectively. As the middle region of the second bimodal distribution is flat and extinction ratios of the two resonances λ1 and λ2 are almost equal as 18 dB, the second bimodal distribution is chosen as the working region in order to achieve better beat performances. On the other hand, the third bimodal distribution (resonances λ3 and λ4) with the lowest transmission loss is selected as the pump region to realize the optimal coupling of pump light. The wavelengths of the four resonances are 1551.42 nm (λ1), 1551.75 nm (λ2), 1555.68 nm (λ3) and 1555.97 nm (λ4), respectively. The zoom in image of the working region is shown as Fig. 6(b). The frequency interval between λ1 and λ2 is almost 41 GHz. The wavelength of the optical carrier is fixed at 1551.71 nm, which is 36 GHz [corresponding to f3 in Fig. 2(a)] away from λ1 and 5 GHz [corresponding to f1 in Fig. 2(a)] away from λ2.
Fig. 6 (a) Measured transmission spectrum of the cascaded MRRs. (b) Zoom in image of the working region.
In order to investigate the tuning property of the cascaded MRRs, different pump powers of λ3 and λ4 are injected into the device. The initial transmission spectrum of the working region is shown as the green line in Fig. 7(a). Then, with injecting two pump powers of 0.96 mW [Pin(λ3) = Pin(λ4) = 0.96 mW] and 1.63 mW [Pin(λ3) = Pin(λ4) = 1.63 mW], the spectrum responses of the device are shown as the blue dashed line and red dash-dotted line, respectively. Finally, to more clearly illustrate the relationships between the input pump powers of λ3 (or λ4) and red-shifts of λ1 (or λ2), pump powers ranging from 0 mW to 2 mW in steps of 0.25 mW are injected into the silicon MRRs. As shown in Fig. 7(b), the red-shifts of resonances λ1 (blue solid line) and λ2 (red dashed line) both increase linearly with the input pump powers.
Fig. 7 (a) Measured transmission spectra of the cascaded MRRs under different pump powers. (b) The relationships between the red-shifts of λ1, λ2 and input pump powers.
3. Experimental results and discussions
The schematic illustration of the experimental setup is shown in Fig. 8. The electrical path and optical path are represented by the red dotted lines and blue solid lines, respectively. The wavelength of optical carrier emitted from laser diode 1 (LD1) is fixed at 1551.71 nm. The RF signals emitted from the vector network analyzer (VNA) are amplified by an electrical amplifier (EA) and then modulated on the optical carrier by the MZM. Under small signal modulation, ODSB signals could be generated and then launched into the cascaded MRRs. The pump path including two narrow linewidth lasers (i.e. LD2 and LD3) whose linewidths are several kilohertz (KHz), and two variable optical attenuators (VOAs) are utilized to provide different pump powers. The wavelengths of pump light emitted from LD2 and LD3 are fixed at 1555.68 nm (λ3) and 1555.97 nm (λ4), respectively. Subsequently, the pump powers are coupled together by the optical coupler (OC) and injected into the silicon device through the optical circulator. Finally, the output optical signal of the silicon chip is converted to alternative current by the PD and analyzed in the VNA.
Fig. 8 Schematic illustration of the experimental setup. The red dotted lines: electrical path, the blue solid lines: optical path. LD: laser diode, PC: polarization controller, MZM: Mach–Zehnder modulator, EDFA: erbium-doped fiber amplifier, VOA: variable optical attenuator, OC: optical coupler, PD: photodetector, VNA: vector network analyzer, EA: electrical amplifier.
The experimental results of the MPF tunable central frequency is shown in Fig. 9(a). By finely adjusting the input powers of λ3 and λ4, the central frequency of the notch MPF could be tuned from 5 GHz to 36 GHz. The MPF central frequencies of 5 GHz (the lowest frequency) and 36 GHz (the highest frequency) are achieved in the case of Pin(λ3) = 1.48 mW and Pin(λ4) = 0 mW, Pin(λ3) = 0 mW and Pin(λ4) = 1.65 mW, respectively. The power combinations of λ3 and λ4 to realize different MPF central frequencies are shown in Table 1. The highest required powers of λ3 and λ4 are 1.48 mW and 1.65 mW, respectively. As shown in Fig. 9(b), the rejection ratios and 3dB bandwidths of the MPF are around 40 dB (the blue line) and 7 GHz (the green line), respectively.
Fig. 9 (a) Measured notch MPFs with tunable central frequency. (b) Features of MPF rejection ratio and 3dB bandwidth.
Table 1. Input pump powers of λ3 and λ4 corresponding to different MPF central frequencies
According to Fig. 3, the bandwidth of the MPF could be tuned by inducing different frequency offsets. In the experiment, central frequency of 20 GHz is chosen to illustrate the tunability of MPF bandwidth. Firstly, the pump powers of λ3 and λ4 are set as 0.76 mW and 0.83 mW, respectively. In this case, the frequency intervals (written as fa and fb) between the optical carrier and two resonances of the working region are both 20 GHz (i.e. frequency offset Δf = 0 GHz). Thus the MPF response with a narrow bandwidth of 6.7 GHz could be obtained, shown as the blue line in Fig. 10(a). Secondly, the pump powers of λ3 and λ4 are tuned to 0.81 mW and 0.89 mW respectively, in order to induce a frequency offset Δf of 1 GHz. Namely, the two frequency intervals fa and fb are changed to 19 GHz and 21 GHz, respectively. The green line shows that the MPF response with a larger bandwidth of 7.6 GHz has been realized. Thirdly, a frequency offset Δf of 1.5 GHz is generated to change fa and fb as 18.5 GHz and 21.5 GHz, respectively. The required pump powers are 0.83 mW (λ3) and 0.92 mW (λ4). In this case, the MPF response with a much larger bandwidth of 9.2 GHz is shown as the pink line in Fig. 10(b). Finally, fa and fb are adjusted to 18 GHz and 22 GHz (i.e. frequency offset Δf = 2 GHz), when the two pump powers are tuned to 0.86 mW (λ3) and 0.96 mW (λ4) respectively. Thus the MPF response with the largest bandwidth of 10.3 GHz could be realized, shown as the red line in Fig. 10(c). Features of the MPF rejection ratios and 3dB bandwidths under different frequency offsets are shown in Fig. 10(d). It is clear to see that the MPF bandwidth could be tuned from 6.7 GHz to 10.3 GHz by adjusting the frequency offsets from 0 GHz to 2 GHz.
Fig. 10 Variations of MPF bandwidths under different frequency offsets. (a) Frequency offset of 0 GHz (blue line): Pin(λ3) = 0.76 mW and Pin(λ4) = 0.83 mW, frequency offset of 1 GHz (green line): Pin(λ3) = 0.81 mW and Pin(λ4) = 0.89 mW. (b) Frequency offset of 1.5 GHz (pink line): Pin(λ3) = 0.83 mW and Pin(λ4) = 0.92 mW. (c) Frequency offset of 2 GHz (red line): Pin(λ3) = 0.86 mW and Pin(λ4) = 0.96 mW. (d) Features of the MPF rejection ratio and 3dB bandwidth under different frequency offsets.
Table 2 illustrates the operation methods and experimental performances of recent on-chip tunable MPFs using different nonlinear effects. Firstly, the waveguide lengths to motivate SBS effect in chalcogenide waveguide [27, 28] and silicon devices [29] are both centimeter magnitude and the pump powers are relatively high, which limit their practical applications in large-scale integration. Secondly, the modulation method of optical single sideband modulation by phase modulator (PM) requires external assistance of an optical filter [41, 47], and the simultaneous use of PM and dual-parallel Mach-Zehnder modulator (DPMZM) would increase the complexity and power consumption of the optical systems [28]. Thirdly, only several MPF bandwidths could be tunable but the tuning range is not large enough due to the intrinsic characteristic of SBS [27, 28]. In contrast, the modulation method of this work is intensity modulation (IM) which does not require assistances of optical filters. Moreover, the highest pump power low as 1.65 mW is strong enough to tune the MPF central frequency from 5 GHz to 36 GHz by utilizing the nonlinear effects in compact opto-mechanical MRRs. Meanwhile, with highest pump power low as 0.96 mW, the MPF bandwidth could be tuned from 6.7 GHz to 10.3 GHz. Therefore, the proposed scheme to achieve all-optical tunable microwave filter is significant in on-chip microwave systems with low-power consumption and large tuning ranges.
Table 2. Performance comparisons of recent on-chip tunable MPFs using nonlinear effects
The tunability of the MPF central frequency and bandwidth are mainly determined by the performances of the silicon cascaded MRRs, including the quality (Q) factor, extinction ratios and waveguide separation between the free-hanging MRR and substrate. Firstly, the tuning range of the MPF central frequency could be extended by improving the Q factor of MRRs with post-processing techniques, such as thermal oxidation [48]. In this case, the optical carrier λ0 could be located more closely to λ2 in Fig. 2(a). Namely, the lower MPF central frequency f1 could be realized. Therefore, the tuning range of the MPF central frequency would be larger. Secondly, the tunability of the MPF bandwidth is limited by the MPF rejection ratios, which are related to the MRR extinction ratios. By designing the MRRs at the critical coupling, the extinction ratios could be optimized. In this case, the frequency offset f4-f3 could be set at a larger value in Fig. 3. Namely, the higher MPF bandwidth could be realized. Therefore, the larger tuning range of MPF bandwidth could be realized. Finally, the required pump powers to tune the MPF central frequency and bandwidth could be further improved. By optimizing the waveguide separation between the free-hanging MRR and substrate, the nonlinear effects could be enhanced to decrease the required pump powers.
4. Conclusion
We have experimentally demonstrated an all-optical microwave filter with tunable central frequency and bandwidth based on the nonlinear effects in silicon opto-mechanical MRRs. The MPF central frequency and 3dB bandwidth could be tuned from 5 GHz to 36 GHz and 6.7 GHz to 10.3 GHz, respectively. The highest required pump powers to tune the MPF central frequency and bandwidth are 1.65 mW and 0.96 mW, respectively. The integrated device provides a tunable notch MPF scheme with dominant advantages of all-optical control, low power consumption, compact footprint and wide tuning range, which has significant applications in on-chip microwave photonic systems.
Funding
The National Key Scientific Instrument & Equipment Development Program of China (2012YQ09016701), the National Natural Science Foundation of China (NSFC) (61503350 and 61604135), and the Fundamental Research Funds for the Central Universities, China University of Geosciences (Wuhan) (CUG170637).
References and links
1. J. Capmany and D. Novak, “Microwave photonics combines two worlds,” Nat. Photonics 1(6), 319–330 (2007). [CrossRef]
2. J. Yao, “Microwave Photonics,” J. Lightwave Technol. 27(3), 314–335 (2009). [CrossRef]
3. A. J. Seeds and K. J. Williams, “Microwave photonics,” J. Lightwave Technol. 24(12), 4628–4641 (2006). [CrossRef]
4. Y. Liu, Y. Yu, S. Yuan, X. Xu, and X. Zhang, “Tunable megahertz bandwidth microwave photonic notch filter based on a silica microsphere cavity,” Opt. Lett. 41(21), 5078–5081 (2016). [CrossRef] [PubMed]
5. J. Capmany, B. Ortega, and D. Pastor, “A tutorial on microwave photonic filters,” J. Lightwave Technol. 24(1), 201–229 (2006). [CrossRef]
6. D. Zhang, X. Feng, and Y. Huang, “Tunable and reconfigurable bandpass microwave photonic filters utilizing integrated optical processor on silicon-on-insulator substrate,” IEEE Photonics Technol. Lett. 24(17), 1502–1505 (2012). [CrossRef]
7. B. Vidal, M. A. Piqueras, and J. Martí, “Tunable and reconfigurable photonic microwave filter based on stimulated Brillouin scattering,” Opt. Lett. 32(1), 23–25 (2007). [CrossRef] [PubMed]
8. A. Byrnes, R. Pant, E. Li, D. Y. Choi, C. G. Poulton, S. Fan, S. Madden, B. Luther-Davies, and B. J. Eggleton, “Photonic chip based tunable and reconfigurable narrowband microwave photonic filter using stimulated Brillouin scattering,” Opt. Express 20(17), 18836–18845 (2012). [CrossRef] [PubMed]
9. J. Sancho, N. Primerov, S. Chin, Y. Antman, A. Zadok, S. Sales, and L. Thévenaz, “Tunable and reconfigurable multi-tap microwave photonic filter based on dynamic Brillouin gratings in fibers,” Opt. Express 20(6), 6157–6162 (2012). [CrossRef] [PubMed]
10. X. Feng, C. Lu, H. Y. Tam, and P. K. A. Wai, “Reconfigurable microwave photonic filter using multiwavelength erbium-doped fiber laser,” IEEE Photonics Technol. Lett. 19(17), 1334–1336 (2007). [CrossRef]
11. J. Mora, B. Ortega, J. Capmany, J. Cruz, M. Andres, D. Pastor, and S. Sales, “Automatic tunable and reconfigurable fiberoptic microwave filters based on a broadband optical source sliced by uniform fiber Bragg gratings,” Opt. Express 10(22), 1291–1298 (2002). [CrossRef] [PubMed]
12. M. Song, C. M. Long, R. Wu, D. Seo, D. E. Leaird, and A. M. Weiner, “Reconfigurable and tunable flat-top microwave photonic filters utilizing optical frequency combs,” IEEE Photonics Technol. Lett. 23(21), 1618–1620 (2011). [CrossRef]
13. H. Ou, H. Fu, D. Chen, and S. He, “A tunable and reconfigurable microwave photonic filter based on a Raman fiber laser,” Opt. Commun. 278(1), 48–51 (2007). [CrossRef]
14. X. Yi, C. Lu, X. Yang, W. D. Zhong, F. Wei, L. Ding, and Y. Wang, “Continuously tunable microwave-photonic filter design using high-birefringence linear chirped grating,” IEEE Photonics Technol. Lett. 15(5), 754–756 (2003). [CrossRef]
15. R. Soref, “The past, present, and future of silicon photonics,” IEEE J. Sel. Top. Quantum Electron. 12(6), 1678–1687 (2006). [CrossRef]
16. B. Jalali and S. Fathpour, “Silicon photonics,” J. Lightwave Technol. 24(12), 4600–4615 (2006). [CrossRef]
17. L. Liu, J. Dong, and X. Zhang, “Chip-integrated all-optical 4-bit Gray code generation based on silicon microring resonators,” Opt. Express 23(16), 21414–21423 (2015). [CrossRef] [PubMed]
18. M. Chamanzar, M. Soltani, B. Momeni, S. Yegnanarayanan, and A. Adibi, “Hybrid photonic surface plasmon-polariton ring resonators for sensing applications,” Appl. Phys. B 101(1-2), 263271 (2010). [CrossRef]
19. L. Liu, J. Dong, D. Gao, A. Zheng, and X. Zhang, “On-chip passive three-port circuit of all-optical ordered-route transmission,” Sci. Rep. 5(1), 10190 (2015). [CrossRef] [PubMed]
20. M. Chamanzar and A. Adibi, “Hybrid nanoplasmonic-photonic resonators for efficient coupling of light to single plasmonic nanoresonators,” Opt. Express 19(22), 22292–22304 (2011). [CrossRef] [PubMed]
21. W. Bogaerts, P. De Heyn, T. Van Vaerenbergh, K. De Vos, S. Kumar Selvaraja, T. Claes, P. Dumon, P. Bienstman, D. Van Thourhout, and R. Baets, “Silicon microring resonators,” Laser Photonics Rev. 6(1), 47–73 (2012). [CrossRef]
22. D. Marpaung, C. Roeloffzen, R. Heideman, A. Leinse, S. Sales, and J. Capmany, “Integrated microwave photonics,” Laser Photonics Rev. 7(4), 506–538 (2013). [CrossRef]
23. V. R. Almeida, C. A. Barrios, R. R. Panepucci, and M. Lipson, “All-optical control of light on a silicon chip,” Nature 431(7012), 1081–1084 (2004). [CrossRef] [PubMed]
24. Q. Xu and M. Lipson, “All-optical logic based on silicon micro-ring resonators,” Opt. Express 15(3), 924–929 (2007). [CrossRef] [PubMed]
25. L. Liu, R. Kumar, K. Huybrechts, T. Spuesens, G. Roelkens, E. J. Geluk, and G. Morthier, “An ultra-small, low-power, all-optical flip-flop memory on a silicon chip,” Nat. Photonics 4(3), 182–187 (2010).
26. C. Koos, P. Vorreau, T. Vallaitis, P. Dumon, W. Bogaerts, R. Baets, B. Esembeson, I. Biaggio, T. Michinobu, F. Diederich, W. Freude, and J. Leuthold, “All-optical high-speed signal processing with silicon–organic hybrid slot waveguides,” Nat. Photonics 3(4), 216–219 (2009). [CrossRef]
27. D. Marpaung, B. Morrison, M. Pagani, R. Pant, D. Y. Choi, B. Luther-Davies, S. J. Madden, and B. J. Eggleton, “Low-power, chip-based stimulated Brillouin scattering microwave photonic filter with ultrahigh selectivity,” Optica 2(2), 76–83 (2015). [CrossRef]
28. A. Choudhary, I. Aryanfar, S. Shahnia, B. Morrison, K. Vu, S. Madden, B. Luther-Davies, D. Marpaung, and B. J. Eggleton, “Tailoring of the Brillouin gain for on-chip widely tunable and reconfigurable broadband microwave photonic filters,” Opt. Lett. 41(3), 436–439 (2016). [CrossRef] [PubMed]
29. A. Casas-Bedoya, B. Morrison, M. Pagani, D. Marpaung, and B. J. Eggleton, “Tunable narrowband microwave photonic filter created by stimulated Brillouin scattering from a silicon nanowire,” Opt. Lett. 40(17), 4154–4157 (2015). [CrossRef] [PubMed]
30. R. Pant, D. Marpaung, I. V. Kabakova, B. Morrison, C. G. Poulton, and B. J. Eggleton, “On-chip stimulated Brillouin Scattering for microwave signal processing and generation,” Laser Photonics Rev. 8(5), 653–666 (2014). [CrossRef]
31. M. Li, W. H. P. Pernice, C. Xiong, T. Baehr-Jones, M. Hochberg, and H. X. Tang, “Harnessing optical forces in integrated photonic circuits,” Nature 456(7221), 480–484 (2008). [CrossRef] [PubMed]
32. M. L. Povinelli, M. Lončar, M. Ibanescu, E. J. Smythe, S. G. Johnson, F. Capasso, and J. D. Joannopoulos, “Evanescent-wave bonding between optical waveguides,” Opt. Lett. 30(22), 3042–3044 (2005). [CrossRef] [PubMed]
33. S. Weis, R. Rivière, S. Deléglise, E. Gavartin, O. Arcizet, A. Schliesser, and T. J. Kippenberg, “Optomechanically induced transparency,” Science 330(6010), 1520–1523 (2010). [CrossRef] [PubMed]
34. D. Van Thourhout and J. Roels, “Optomechanical device actuation through the optical gradient force,” Nat. Photonics 4(4), 211–217 (2010). [CrossRef]
35. B. Dong, H. Cai, G. I. Ng, P. Kropelnicki, J. M. Tsai, A. B. Randles, M. Tang, Y. D. Gu, Z. G. Suo, and A. Q. Liu, “A nanoelectromechanical systems actuator driven and controlled by Q-factor attenuation of ring resonator,” Appl. Phys. Lett. 103(18), 181105 (2013). [CrossRef]
36. J. Tao, J. Wu, H. Cai, Q. Zhang, J. M. Tsai, J. Lin, and A. Liu, “A nanomachined optical logic gate driven by gradient optical force,” Appl. Phys. Lett. 100(11), 113104 (2012). [CrossRef]
37. Y. F. Yu, M. Ren, J. B. Zhang, T. Bourouina, C. S. Tan, J. M. Tsai, and A. Q. Liu, “Force-induced optical nonlinearity and Kerr-like coefficient in opto-mechanical ring resonators,” Opt. Express 20(16), 18005–18015 (2012). [CrossRef] [PubMed]
38. H. Cai, B. Dong, J. F. Tao, L. Ding, J. M. Tsai, G. Q. Lo, A. Q. Liu, and D. L. Kwong, “A nanoelectromechanical systems optical switch driven by optical gradient force,” Appl. Phys. Lett. 102(2), 023103 (2013). [CrossRef]
39. J. Dong, L. Liu, D. Gao, Y. Yu, A. Zheng, T. Yang, and X. Zhang, “Compact notch microwave photonic filters using on-chip integrated microring resonators,” IEEE Photonics J. 5(2), 5500307 (2013). [CrossRef]
40. L. Liu, M. He, and J. Dong, “Compact continuously tunable microwave photonic filters based on cascaded silicon microring resonators,” Opt. Commun. 363, 128–133 (2016). [CrossRef]
41. L. Liu, Y. Yang, Z. Li, X. Jin, W. Mo, and X. Liu, “Low power consumption and continuously tunable all-optical microwave filter based on an opto-mechanical microring resonator,” Opt. Express 25(2), 960–971 (2017). [CrossRef] [PubMed]
42. M. Ren, J. Huang, H. Cai, J. M. Tsai, J. Zhou, Z. Liu, Z. Suo, and A. Q. Liu, “Nano-optomechanical actuator and pull-back instability,” ACS Nano 7(2), 1676–1681 (2013). [CrossRef] [PubMed]
43. J. Rosenberg, Q. Lin, and O. Painter, “Static and dynamic wavelength routing via the gradient optical force,” Nat. Photonics 3(8), 478–483 (2009). [CrossRef]
44. M. Xu, J. Wu, T. Wang, X. Hu, X. Jiang, and Y. Su, “Push-pull optical nonreciprocal transmission in cascaded silicon microring resonators,” IEEE Photonics J. 5(1), 2200307 (2013). [CrossRef]
45. L. Martinez and M. Lipson, “High confinement suspended micro-ring resonators in silicon-on-insulator,” Opt. Express 14(13), 6259–6263 (2006). [CrossRef] [PubMed]
46. V. Subramaniam, G. N. De Brabander, D. H. Naghski, and J. T. Boyd, “Measurement of mode field profiles and bending and transition losses in curved optical channel waveguides,” J. Lightwave Technol. 15(6), 990–997 (1997). [CrossRef]
47. Y. Long and J. Wang, “All-optical tuning of a nonlinear silicon microring assisted microwave photonic filter: theory and experiment,” Opt. Express 23(14), 17758–17771 (2015). [CrossRef] [PubMed]
48. A. Gondarenko, J. S. Levy, and M. Lipson, “High confinement micron-scale silicon nitride high Q ring resonator,” Opt. Express 17(14), 11366–11370 (2009). [CrossRef] [PubMed]
