Open Access
Add to BibSonomy
Abstract
Reconfigurable optical mode multiplexers/demultiplexers have attracted increasing attention in the academic community, because they enable convenient construction of flexible and complex on-chip optical networks. Here, we propose and demonstrate a scheme of reconfigurable and scalable optical mode multiplexer/demultiplexer with large operation bandwidth, based on three-waveguide-coupling structures. As proof of concept, a reconfigurable device that can multiplex input signals to the fundamental and first-order quasi-transverse electric mode is fully fabricated and demonstrated successfully. Static response spectra show that the optical crosstalk at the output ports of the device are less than −14.3 dB and −13.7 dB over the entire C band (> 40 nm), respectively. The dynamic performance with data transmission speeds of 40 Gbps for each multiplexing channel are also demonstrated successfully. The presented device is believed to be a potential candidate for future on-chip optical network with large-scale integration, flexible functionality, and low cost.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Optical communication plays a crucially important role in long-distance, short-distance, and on-chip networking because of its unique features such as high speed, parallel processing, and low latency [1–8]. Currently, an ever-growing demand for high capacity and flexible data transmission has become an increasingly serious challenge [4–6]. To further increase the communication capacity carried in a physical channel, researchers have developed many effective optical multiplexing technologies, such as wavelength-division multiplexing (WDM) [9,10], polarization-division multiplexing (PDM) [11], and mode-division multiplexing (MDM) [12–15]. Among them, WDM has been successfully deployed in optical communications systems owing to its ability in greatly increasing the channel capacity. However, WDM technology can be expensive and complex requiring many lasers and complex wavelength management [9,10]. For on-chip PDM, only two polarizations can be adopted to increase the capacity, which will quickly become a limitation for future applications [11]. Recently, the MDM technology has attracted widespread attention to further increase capacity as it allows multiple channels of data to be transmitted in parallel through a multimode waveguide by multiplexing several orthogonal spatial modes using one wavelength [12–15]. MDM can also be used in combination with WDM and PDM, enabling communication capacity to be significantly increased [14,16]. For MDM, multiplexer/demultiplexer is the most basic component. Various multiplexers/demultiplexers have been proposed and demonstrated based on different structures such as asymmetrical directional couplers (ADC) [17,18], adiabatic couplers [19], asymmetrical Y-junctions [20,21], micro-ring resonators (MRR) [14,22], multimode interference (MMI) couplers [23], among others. However, most of the proposed mode multiplexers/demultiplexers cannot be reconfigured, which means that each input fundamental mode can only be converted to one specific high-order mode at the mode multiplexer [12–23]. Obviously, this would limit the application of multiplexer/demultiplexer in the increasingly complex and flexible MDM networks. Although some reconfigurable multiplexers/demultiplexers are also demonstrated successfully based on MMI couplers and Y junctions [24–30], it is difficult for them to be scaled to implement higher-order mode (de)multiplexing. More recently, some mode selective switches and add-drop multiplexers have been demonstrated based on MRR, Mach-Zehnder interference coupler and micro-electromechanical system [31–35]. These devices can be converted into reconfigurable mode multiplexers/demultiplexers by introducing an extra passive mode multiplexer, but they are complicated in structure. Therefore, there still exists a demand to develop a reconfigurable multiplexer/demultiplexer characterized with simple and scalable structure.
A new mode conversion scheme based on cascading three-waveguide coupling structures has been briefly proposed recently [36]. This type of mode conversion scheme holds great promise for scalable reconfigurable mode multiplex/demultiplex functionality, however, detailed analysis of the working principle and experimental demonstration of this mode conversion scheme have not yet been reported. In this paper, we propose and experimentally demonstrate an on-chip reconfigurable broadband optical mode multiplexer/demultiplexer using the novel three-waveguide-coupling structure. The proposed device can implement the function of MDM to increase channel capacity and importantly can convert the signal from any of the two input channels to an arbitrary mode of the output channel. For proof of concept, a reconfigurable mode multiplexer/demultiplexer which can multiplex the input signals to fundamental and first-order quasi transverse electric mode (TE0 and TE1) is fabricated and demonstrated. The static response spectra show that the optical crosstalk at the output ports of the device are less than −14.3 dB and −13.7 dB over the entire C band, respectively. To demonstrate the utility of this structure, a simple communications demonstration is presented and clear and open eye diagrams with data transmission speed of 40 Gbps are achieved.
2. Device principle
As shown in Fig. 1(a), the proposed reconfigurable wideband optical mode multiplexer/demultiplexer is composed of one 2 × 2 MMI coupler and two three-waveguide-coupling structures. Each three-waveguide-coupling structure consists of two single mode waveguides and a multimode waveguide. A linear adiabatic taper is used to connect the two multimode waveguides and form a bus waveguide. The two input ports and one output port are labeled X, Y and Z, respectively. According to the principle of MMI [37,38], the light input from port X or Y would be divided to two parts with equal amplitudes and phase difference of π/2 if the length of multimode region LMMI satisfies LMMI = 3Lπ/2, where Lπ is the beat length of two lowest-order modes [Figs. 1(b) and 1(c)]. Two thermal phase shifters A and B are adopted to tune the phases of the two paths [Fig. 1(a)]. Once the phase difference of the two divided light beams is 0 or 2π, the two divided light beams would couple into the bus waveguide in the form of even-order mode (TE0 or TE2 or TE4 or any higher order even mode where the field distribution is symmetrical about the waveguide axis) at the first three-waveguide coupling region (TWCR1) [39], and the coupled even-order mode would finally be directed to the output port Z [Fig. 1(d)]. The multimode waveguide of TWCR1 is designed to support several modes, and the supported highest-order mode is the desired even-order mode. The effective refractive index of the even-order mode in the multimode waveguide of TWCR1, which is determined by the width of the multimode waveguide, must match the effective refractive index of the TE0 mode in the single mode waveguides to achieve efficient conversion (phase matching condition) [14,39]. Once the phase difference of the two divided light beams is π, the two divided light beams would pass through the TWCR1 and continue to propagate in two single mode waveguides with no disturbance. At the second TWCR (TWCR2), the two divided light beams with phase difference of π would couple into the bus waveguide and be converted to an odd-order mode (TE1 or TE3 or TE5 or any higher order odd mode where the field distribution is anti-symmetrical about the waveguide axis) [Fig. 1(e)]. The multimode waveguide of TWCR2 is designed to support several modes, and the supported highest-order mode is the desired odd-order mode. The effective refractive index of the odd-order mode in the multimode waveguide of TWCR2, which is determined by the width of the multimode waveguide, must match the effective refractive index of the TE0 mode in the single mode waveguides. In this case, the width of TWCR2’s multimode waveguide should be larger than TWCR1’s to ensure the multiplexed even-order mode from TWCR1 can propagate in TWCR2’s multimode waveguide and implement the function of mode multiplexing. Since the two divided light beams have an original phase difference of π/2, there are two ways to accomplish a total phase difference of 0 or 2π. The first approach is to apply an appropriate voltage on the micro-heater above one of the Arm waveguides (Arm1 or Arm2) which has a small phase to compensate the phase difference, the other approach is to apply an appropriate voltage on the micro-heater above one of the Arm waveguides (Arm1 or Arm2) which has large phase to tune an extra 3π/2. The same methods can be adopted to realize a phase difference of π.
Fig. 1 Schematic of the proposed reconfigurable mode multiplexer/demultiplexer (MMI: multimode interference coupler, TWCR: three-waveguide-coupling region). (a) schematic of the device, (b)-(c) schematic of the 2 × 2 MMI when only X input and Y input respectively, (d)-(e) schematic of TWCRs when two light beams propagate in the two arms with same phases and π phase difference respectively, Φ1 and Φ2 are the phases of light beam in Arm1 and Arm2 respectively, (f)-(k) schematic of the proposed device at different working states, ΔΦA and ΔΦB are the tuned phases in phase shifter A and B respectively.
Without applied voltage to the heaters, the phase in the two Arm waveguides will be different due to the nature of the coupler. When light is coupled from port X, the phase of divided light in Arm 1 is π/2 larger than the one of light in Arm 2 [Fig. 1(b)]; when light is coupled from port Y, the phase of divided light in Arm 2 is π/2 larger than the divided light in Arm 1 [Fig. 1(c)]. For one scenario, if only phase shifter B on Arm 2 is tuned to π/2 (ΔΦA = 0, ΔΦB = π/2), the phase difference of the two optical paths from input port X would be 0 before arriving at the TWCR1. In this case, the even-order mode will be excited at TWCR1 and will be finally directed to the Output Z [Figs. 1(b), 1(d) and 1(f)]. Meanwhile, the phase difference of the two optical paths from input port Y would be π before arriving at the TWCR1, resulting in no excitation at this point due to orthogonality of the odd symmetry of the excitation from the Arms and the even symmetry of the available phase matched bus mode. The light would thus continue to propagate in the two Arm waveguides until they coupled to an odd-order mode in the multimode waveguide of TWCR2 [Figs. 1(c), 1(e) and 1(g)]. Hence, when light is simultaneously input from both X and Y ports, both even-order mode and odd-order mode will be obtained in Output Z at the state of ΔΦA = 0, ΔΦB = π/2 [Fig. 1(h)].
For another scenario, if only phase shifter A above Arm 1 is tuned π/2 (ΔΦA = π/2, ΔΦB = 0), the phase difference of the two paths from input port X would be π before arriving at the TWCR1, the light would not couple to the bus waveguide and would continue to propagate in two Arm waveguides until it coupled into the bus waveguide at TWCR2, and finally will be converted to an odd-order mode [Figs. 1(b), 1(e) and 1(i)]. Meanwhile, the phase difference of the two divided light beams from input port Y would be 0 before arriving at the TWCR1, these two divided light beams would couple into the bus waveguide at TWCR1 in the form of even-order mode and will be finally directed to the Output Z [Figs. 1(c), 1(d) and 1(j)]. When light is simultaneously input from both X and Y ports, both even-order mode and odd-order mode will be obtained in Output Z at the state of ΔΦA = π/2, ΔΦB = 0 [Fig. 1(k)]. Similar results can be obtained if only one of the phase shifters is tuned 3π/2.
According to the principle discussed above, the performance of the proposed mode multiplexer/demultiplexer is simulated using 3-dimension finite difference beam propagation method (3D-FDBPM) [40,41]. The refractive indices for silicon and silicon dioxide are 3.45 and 1.45 respectively. The wavelength used in simulation is 1550 nm. For illustration purposes, the simulated results of a device which can multiplex TE0 & TE1 modes and a device which can multiplex TE2 & TE3 modes are shown in Fig. 2. The differences of the two devices are the widths of multimode waveguides and the coupling lengths of TWCRs since the two devices are adopted to multiplex different modes. At the tuning state of ΔΦA = 0 and ΔΦB = π/2 (or ΔΦA = 3π/2 and ΔΦB = 0), when light is input from port X, a TE0 mode is obtained in Fig. 2(a); when light is input from port Y, a TE1 mode is obtained in Fig. 2(b); when light is input from ports X and Y simultaneously, they are multiplexed to two modes (TE1 + TE0) in the bus waveguide, as shown in Figure 2(c).At the tuning state of ΔΦA = π/2 and ΔΦB = 0 (or ΔΦA = 0 and ΔΦB = 3π/2), when light is input from port X, a TE1 mode is obtained in Fig. 2(d); when light is input from port Y, a TE0 mode is obtained in Fig. 2(e); when light is input from ports X and Y simultaneously, they are multiplexed to two modes (TE0 + TE1) in the bus waveguide in Fig. 2(f). By increasing the width of the multimode waveguide in TWCR1 to support TE2, and the width of the multimode waveguide in TWCR2 to support TE3, the results of multiplexing TE2 and TE3 modes are obtained as shown in Figs. 2(g)-2(l). Note that the proposed device can implement higher-order mode multiplexing by choosing different widths of the bus waveguide to satisfy effective refractive index matching condition. Therefore, we can conclude that when light is input from both X and Y ports simultaneously, they would be multiplexed to two modes in the bus waveguide, and the relation between input ports and multiplexed modes is depending on the phases difference introduced by the phase shifters. The fact that two phase shifters can be adopted to tune the optical phase independently to implement the function greatly increase the flexibility of the device. Besides, one outstanding advantage for the proposed device is that the reliability of device can be improved.
Fig. 2 Simulated performance of the mode multiplexers based on three-waveguide-coupling structure. ΔΦA and ΔΦB are the phases that phase shifters A and B are tuned respectively. (a)-(f) the mode distributions of the device which can multiplex fundamental and first-order quasi transverse electric mode. (g)-(l) the mode distributions of the device which can multiplex second-order mode and third-order mode. (a)(d)(g)(j) are the mode distributions when light only input from X port, (b)(e)(h)(k) are the mode distributions when light only input from Y port, (c)(f)(i)(l) are the mode distributions when light is input from both X and Y ports, respectively.
3. Fabrication and experimental setup
As a proof of concept, the basic reconfigurable mode multiplexer/demultiplexer, which can implement the reconfigurable multiplexing of TE0 mode and TE1 mode was fabricated and demonstrated. The device was fabricated on an 8-inch silicon-on-insulator wafer with 220 nm top silicon layer and 3 μm buried silicon dioxide layer. To reduce the scattering loss induced by sidewalls, ridge waveguides with 220 nm in height, 70 nm in slab thickness, 400 nm in width for single-mode waveguide and 880 nm in width for multimode-mode waveguide were adopted to construct the proposed device. To implement efficient coupling between TE0 mode and TE1 mode, the effective refractive indices for TE0 mode in 400 nm waveguide and TE1 mode waveguide should be equal. The width of TE1 mode waveguide is chosen 880 nm, of which the effective refractive index is calculated about 2.42 at the wavelength of λ = 1550 nm. The adiabatic linear taper adopted to connect the two multimode waveguides of TWCRs was 150 μm in length. 248 nm deep ultraviolet photolithography was employed to define the pattern of the device, and inductively coupled plasma (ICP) was adopted to etch the waveguides. Plasma enhanced chemical vapor deposition (PECVD) was used to deposit a 1500 nm thick silicon dioxide layer. After that, the titanium nitride (TiN) with thickness of 120 nm and width of 2 μm were deposited above the two arms of the device to form the thermal heater for phase tuning. Then a silicon dioxide layer with thickness of 500 nm was deposited by PECVD. Via holes are etched till TiN layer to get electrical access, and finally, aluminum wires along with pads are fabricated to form the metal trace. At the input and output ports of the device, spot size converters are fabricated to couple light into/out from the device. The fabricated device is shown in Fig. 3, of which the partial enlarged graphs show the MMI and two TWCRs. The effective footprint of the device including a reconfigurable mode multiplexer/demultiplexer and an ADC-based normal demultiplexer used to demultiplex the TE0 and TE1 modes from the reconfigurable mode multiplexer/demultiplexer is about 50 μm × 1000 μm. The input and output ports of the device are named X, Y, P and Q, respectively. The left ADC of the mode demultiplexer is adopted to convert TE1 mode obtained from the reconfigurable mode multiplexer/demultiplexer to TE0 mode and to direct it to output P port, while the right ADC is adopted to demultiplex the TE0 mode obtained from the reconfigurable mode multiplexer/demultiplexer and to direct it to output Q port. The lengths of the left and right ADCs are 13.7 μm and 9.2 μm respectively.
Fig. 3 Micrograph of the fabricated device. (a) micrograph of the device which consists of a reconfigurable mode multiplexer/demultiplexer and a normal mode multiplexer, (b)-(d) partial enlarged graphs of the multimode interference coupler, the first three-waveguide-coupling region, and the second three-waveguide-coupling region, respectively.
The experimental setup utilized to characterize the proposed device is shown in Fig. 4. For static spectra measurement, an amplified spontaneous emission (ASE) source, a tunable voltage source (TVS), and an optical spectrum analyzer (OSA) are employed. The ASE is adopted to generate broad band light over the entire C band. Inverse tapered single-mode fiber is adopted to coupled light into the input port of the device under test. Voltages generated by the TVS are applied on the pads to tune the phase shifters. At the output port of the device, another inverse tapered single-mode fiber is utilized to couple the light out and fed it into the OSA for spectrum measurement. According to the working principle described in section 2, each tuned phase corresponds to a TE0 or TE1 mode multiplexed by the reconfigurable mode multiplexer, and thus results in high and low level of power in each output port. For instance, if light is input from port X, when the applied voltage on phase shifter B is increased from 0, the phase of the light propagate in the arm under this phase shifter will increase. Once the increased phase reaches to π/2, a TE0 mode will be obtained in the reconfigurable mode multiplexer. Thus, the light at P port will be in its lowest level while the power at Q port will be in its highest level. This voltage is then chosen as the voltage to implement phase tuning of ΔΦB = π/2. Further increasing of this voltage can implement ΔΦB = 3π/2. The voltages in other tuning states can also be determined by using the same method.
Fig. 4 Experimental setup utilized to characterize the fabricated device (ASE: amplified spontaneous emission; TVS: tunable voltage source; DUT: device under test; OSA: optical spectrum analyzer; TL: tunable laser; OM: optical modulator; PC: polarization controller; BPG: bit pattern generator; EDFA: erbium-doped fiber amplifier; OTF: optical tunable filter; DCA: digital communication analyzer).
For the fabricated device, there are two methods to implement reconfigurable mode multiplexing by applying appropriate voltages on the phase shifters as mentioned above. Figure 5 shows the measured static response spectra of the device. When the broadband light is input from port X, by applying voltages VA = 0 V and VB = 3.3 V to the phase shifters A and B respectively, the light divided between two Arm waveguides would couple into the bus waveguide in the form of TE0 mode at TWCR1, and will be finally directed to the Output Q [Figs. 2(a) and 5(c)]. Meanwhile the optical power at Output P port is at a low level [Fig. 5(a)].
Fig. 5 Measured static response spectra of the device, (a)-(b) the transmission spectra from input ports X and Y to output port P when different voltages are applied to the corresponding phase shifters, (c)-(d) the transmission spectra from input ports X and Y to output port Q when different voltages are applied to the corresponding phase shifters. (CT: crosstalk).
By applying voltages VA = 0 V and VB = 6.1 V to phase shifters A and B respectively, the light divided between the two Arm waveguides would couple into the bus waveguide in the form of TE1 mode at TWCR2, and will be finally directed to the Output P [Figs. 2(d) and 5(b)]. Meanwhile the optical power at Output Q port is at a low level [Fig. 5(d)], which are opposite to the previous case with applied voltage VA = 0 V and VB = 3.3 V.
By applying voltages VA = 3.9 V and VB = 0 V to the phase shifters A and B respectively, the light divided between the two Arm waveguides would couple into the bus waveguide in the form of TE1 mode at TWCR2, and will be finally directed to Output P [Figs. 2(d) and 5(b)]. Meanwhile the optical power at Output Q port would be at a low level [Fig. 5(d)].
By applying voltages VA = 6.7 V and VB = 0 V to the phase shifters A and B respectively, the light divided between the two Arm waveguides would couple into the bus waveguide in the form of TE0 mode at TWCR1, and will be finally directed to Output Q [Figs. 2(a) and 5(c)]. Meanwhile the optical power at Output P port would be at a low level [Fig. 5(a)].
Similarly, when the light is input from port Y, we would expect the opposite results compared to light input from port X [Figs. 5(a)-5(d)]. The optical crosstalk (CT) at Output P and Q are all less than −14.3 dB and −13.7 dB respectively. The insertion losses for multiplexing channels are about 10 dB, which include about 6.0 dB coupling loss between the device and the external single-mode fibers when light is coupled into and out from the device, about 0.5 dB coupling loss for the 2 × 2 MMI, 0.3 dB transmission loss, 1.1-2.0 dB for TWCR1 and 1.2-2.8 dB for TWCR2 over the wavelength range of 1525-1565 nm, 0.9-2.0 dB and 0.7-1.8 dB propagation losses of mode demultiplexer for TE0 and TE1 modes over the wavelength range of 1525-1625 nm, respectively. Further optimization could be made on the demultiplexers and the TWCR structures to reduce the insertion loss of each multiplexing channel. Besides, the proposed device is fabricated with 180 nm CMOS process, therefore, more advanced fabrication process such as 90 nm fabrication process, even Electron Beam Lithography (EBL) technology can be considered for device fabrication in order to further reduce the loss and improve the performance of the device, since the structural parameters of the fabricated device can be more in agreement with the design targets in these ways. All these are left as further works for us in future.
The eye diagrams of the device are also measured. For dynamic measurement, a tunable laser (TL), a LiNbO3 optical modulator (OM), a polarization controller (PC), a TVS, a bit pattern generator (BPG), an erbium-doped fiber amplifier (EDFA), an optical tunable filter (OTF) and a digital communication analyzer (DCA) are adopted. A monochromatic continuous wave generated by the TL is fed into a LiNbO3 OM and PC before it is coupled into the input ports of the fabricated device. The LiNbO3 modulator is driven by a 40-Gbps pseudo-random binary sequence (PRBS) with a length of 231-1 generated by the BPG. When the light is output from the device, the light is first amplified by the EDFA and then filtered by the OTF before it is fed into the DCA for eye diagrams observation. The BPG and DCA are triggered with same speed clock signal at the rate of 40 Gbps. For the DCA, the sensitivity is >-5 dBm and its bandwidth is 65 GHz. The fibers employed to direct light outside the device are all normal single-mode fibers. The cables adopted to transmit the direct-current, the PRBS data and the corresponding clock are radio-frequency cables. Figure 6 shows the measured clear and open eye diagrams for data transmission speed of 40 Gbps. Note that, the working wavelength of the device can cover the entire C band. For clarity, only the results measured at 1545 nm and 1555 nm are shown here. If dynamic signals with data rate of 40 Gbps are input from both X and Y ports, the total throughput for the reconfigurable MDM device can reach 80 Gbps. By cascading more 2 × 2 MMIs and more TWCRs, the total throughput for the device can be dramatically increased.
4. Discussion
We have shown proof of concept functionality with a reconfigurable low order multimode multiplexer/demultiplexer. However, we believe the proposed reconfigurable mode multiplexer can be used as a basic component to realize advanced functions. Figure 7 shows two kinds of advanced reconfigurable mode multiplexer/demultiplexer. The first one is an advanced reconfigurable two-mode multiplexer scaled by cascading several TWCRs to the bus waveguide [Fig. 7(a)] [36]. The widths of the multimode waveguides in TWCRs such as TWCR1, TWCR2 and TWCR3 are designed to support TE0, TE1, and TE2 modes, respectively, of which the effective refractive indices should be equal to the TE0 mode in the single-mode waveguide. The two input signals can be dynamically multiplexed to two arbitrary modes in this device by tuning the phase shifters, and thus the flexibility of the device can be greatly increased. For instance, at the tuning state of ΔΦA = 0, ΔΦB = π/2, ΔΦC = 0, ΔΦD = 0, ΔΦE = 0 and ΔΦF = 0, when light is input from Input1 port, a TE0 mode will be obtained at the TWCR1, when light is input from Input2 port, a TE1 mode will be obtained at the TWCR2; at the tuning state of ΔΦA = π/2, ΔΦB = 0, ΔΦC = 0, ΔΦD = 0, ΔΦE = 0 and ΔΦF = 0, when light is input from Input1 port, a TE1 mode will be obtained at the TWCR2, when light is input from Input2 port, a TE0 mode will be obtained at the TWCR1; at the tuning state of ΔΦA = π/2, ΔΦB = 0, ΔΦC = π, ΔΦD = 0, ΔΦE = 0 and ΔΦF = 0, when light is input from Input1 port, a TE2 mode will be obtained at the TWCR3, when light is input from Input2 port, a TE0 mode will be obtained at the TWCR1; other modes can be obtained by changing the tuning state of the phase shifters. Another potential application is a scaled reconfigurable multimode multiplexer which can multiplex many input signals by cascading several basic reconfigurable mode multiplexers/demultiplexers described in section 2 [Fig. 7(b)]. The relationship between the input signals and the multiplexed modes in this device can be dynamically changed according to the principle discussed in section 2. Phase shifters A, B and TWCR1, TWCR2 are utilized to reconfigurably multiplex TE0 and TE1 modes, phased shifters C, D and TWCR3, TWCR4 are utilized to reconfigurably multiplex TE2 and TE3 modes. Higher-order modes can also be reconfigurably multiplexed by cascading more basic reconfigurable mode multiplexers/demultiplexers. Moreover, further researches can be made on the proposed basic reconfigurable mode multiplexer/demultiplexer to implement other advanced functions.
Fig. 7 Two kinds of advanced reconfigurable mode multiplexer/demultiplexer scaled by the proposed basic reconfigurable mode multiplexer/demultiplexer, (a) reconfigurable two arbitrary modes multiplexer, (b) reconfigurable multimode multiplexer.
5. Conclusion
In summary, we propose and demonstrate a basic reconfigurable wideband optical multiplexer/demultiplexer using a 2 × 2 MMI and two TWCRs. The basic component consists of a reconfigurable mode multiplexer/demultiplexer which can multiplex two input fundamental quasi transverse electric mode to two arbitrary modes. Two kinds of advanced reconfigurable mode multiplexer/demultiplexer are proposed based on the basic reconfigurable multiplexer/demultiplexer. As proof of concept, a basic reconfigurable optical multiplexer/demultiplexer which can multiplex the fundamental and first-order quasi transverse electric mode is fabricated and demonstrated. The static response spectra show that the optical crosstalk at the output ports of the device are less than −14.3 dB and −13.7 dB over the entire C band (over 40 nm), respectively. Clear and open eye diagrams for data transmission at the rate of 40 Gbps in all multiplexing channels are successfully demonstrated. The demonstrated device characterized with reconfigurability and scalability would have applications in deploying MDM to complex optical networks.
Funding
Open Project of Key Laboratory for Magnetism and Magnetic Materials of the Ministry of Education (LZUMMM2017005); Opened Fund of the State Key Laboratory on Integrated Optoelectronics (IOSKL2016KF14).
Acknowledgement
The authors would like to thank Dr. Hao Jia and Professor Lin Yang for their support in the device’s dynamic measurement.
References
1. M. A. Taubenblatt, “Optical interconnects for high-performance computing,” J. Lightwave Technol. 30(4), 448–457 (2012). [CrossRef]
2. L. Yang, T. Zhou, H. Jia, S. L. Yang, J. F. Ding, X. Fu, and L. Zhang, “General architectures for on-chip optical space and mode switching,” Optica 5(2), 180–187 (2018). [CrossRef]
3. C. Zhang, S. J. Zhang, J. D. Peters, and J. E. Bowers, “8 × 8 × 40 Gbps fully integrated silicon photonic network on chip,” Optica 3(7), 785–786 (2016). [CrossRef]
4. R. W. Tkach, “Scaling optical communications for the next decade and beyond,” Bell Labs Tech. J. 14(4), 3–9 (2010). [CrossRef]
5. R. J. Essiambre and R. W. Tkach, “Capacity trends and limits of optical communication networks,” Proc. IEEE 100(5), 1035–1055 (2012). [CrossRef]
6. E. Agrell, M. Karlsson, A. R. Chraplyvy, D. J. Richardson, P. M. Krummrich, P. Winzer, K. Roberts, J. K. Fischer, S. J. Savory, B. J. Eggleton, M. Secondini, F. R. Kschischang, A. Lord, J. Prat, I. Tomkos, J. E. Bowers, S. Srinivasan, M. Brandt-Pearce, and N. Gisin, “Roadmap of optical communications,” J. Opt. 18(6), 063002 (2016). [CrossRef]
7. R. G. Beausoleil, P. J. Kuekes, G. S. Snider, S.-Y. Wang, and R. S. Williams, “Nanoelectronic and nanophotonic interconnect,” Proc. IEEE 96(2), 230–247 (2008). [CrossRef]
8. D. A. B. Miller, “Device requirements for optical interconnects to silicon chips,” Proc. IEEE 97(7), 1166–1185 (2009). [CrossRef]
9. A. Liu, L. Liao, Y. Chetrit, J. Basak, H. Nguyen, D. Rubin, and M. Paniccia, “Wavelength division multiplexing based photonic integrated circuits on silicon-on-insulator platform,” IEEE J. Sel. Top. Quantum Electron. 16(1), 23–32 (2010). [CrossRef]
10. W. Bogaerts, P. Dumon, D. Van Thourhout, D. Taillaert, P. Jaenen, J. Wouters, S. Beckx, V. Wiaux, and R. G. Baets, “Compact wavelength-selective functions in silicon-on-insulator photonic wires,” IEEE J. Sel. Top. Quantum Electron. 12(6), 1394–1401 (2006). [CrossRef]
11. B. Shen, P. Wang, R. Polson, and R. Menon, “An integrated-nanophotonics polarization beam splitter with 2.4× 2.4 μm2 footprint,” Nat. Photonics 9(6), 378–382 (2015). [CrossRef]
12. D. J. Richardson, J. M. Fini, and L. E. Nelson, “Space-division multiplexing in optical fibres,” Nat. Photonics 7(5), 354–362 (2013). [CrossRef]
13. R. G. H. van Uden, R. A. Correa, E. A. Lopez, F. M. Huijskens, C. Xia, G. Li, A. Schülzgen, H. de Waardt, A. M. J. Koonen, and C. M. Okonkwo, “Ultra-high-density spatial division multiplexing with a few-mode multicore fibre,” Nat. Photonics 8(11), 865–870 (2014). [CrossRef]
14. L. W. Luo, N. Ophir, C. P. Chen, L. H. Gabrielli, C. B. Poitras, K. Bergmen, and M. Lipson, “WDM-compatible mode-division multiplexing on a silicon chip,” Nat. Commun. 5(1), 3069 (2014). [CrossRef] [PubMed]
15. X. R. Wu, C. R. Huang, K. Xu, W. Zhou, C. Shu, and H. K. Tsang, “3 × 104 gb/s single-λ interconnect of mode-division multiplexed network with a multicore fiber,” J. Lightwave Technol. 36(2), 318–324 (2018). [CrossRef]
16. D. X. Dai, C. L. Li, S. P. Wang, H. Wu, Y. C. Shi, Z. H. Wu, S. M. Gao, T. G. Dai, H. Yu, and H.-K. Tsang, “10-channel mode (de)multiplexer with dual polarizations,” Laser Photonics Rev. 12(1), 1700109 (2018). [CrossRef]
17. M. Greenberg and M. Orenstein, “Multimode add-drop multiplexing by adiabatic linearly tapered coupling,” Opt. Express 13(23), 9381–9387 (2005). [CrossRef] [PubMed]
18. Y. Ding, J. Xu, F. Da Ros, B. Huang, H. Ou, and C. Peucheret, “On-chip two-mode division multiplexing using tapered directional coupler-based mode multiplexer and demultiplexer,” Opt. Express 21(8), 10376–10382 (2013). [CrossRef] [PubMed]
19. J. Xing, Z. Li, X. Xiao, J. Yu, and Y. Yu, “Two-mode multiplexer and demultiplexer based on adiabatic couplers,” Opt. Lett. 38(17), 3468–3470 (2013). [CrossRef] [PubMed]
20. J. B. Driscoll, R. R. Grote, B. Souhan, J. I. Dadap, M. Lu, and R. M. Osgood, “Asymmetric Y junctions in silicon waveguides for on-chip mode-division multiplexing,” Opt. Lett. 38(11), 1854–1856 (2013). [CrossRef] [PubMed]
21. W. Chen, P. Wang, T. Yang, G. Wang, T. Dai, Y. Zhang, L. Zhou, X. Jiang, and J. Yang, “Silicon three-mode (de)multiplexer based on cascaded asymmetric Y junctions,” Opt. Lett. 41(12), 2851–2854 (2016). [CrossRef] [PubMed]
22. Y. D. Yang, Y. Li, Y. Z. Huang, and A. W. Poon, “Silicon nitride three-mode division multiplexing and wavelength-division multiplexing using asymmetrical directional couplers and microring resonators,” Opt. Express 22(18), 22172–22183 (2014). [CrossRef] [PubMed]
23. T. Uematsu, Y. Ishizaka, Y. Kawaguchi, K. Saitoh, and M. Koshiba, “Design of a compact two-mode multi/demultiplexer consisting of multimode interference waveguides and a wavelength-insensitive phase shifter for mode-division multiplexing transmission,” J. Lightwave Technol. 30(15), 2421–2426 (2012). [CrossRef]
24. D. Melati, A. Alippi, and A. Melloni, “Reconfigurable photonic integrated mode (de)multiplexer for SDM fiber transmission,” Opt. Express 24(12), 12625–12634 (2016). [CrossRef] [PubMed]
25. W. K. Zhao, J. Feng, K. X. Chen, and K. S. Chiang, “Reconfigurable broadband mode (de)multiplexer based on an integrated thermally induced long-period grating and asymmetric Y-junction,” Opt. Lett. 43(9), 2082–2085 (2018). [CrossRef] [PubMed]
26. Y. L. Xiong, R. B. Priti, and O. Liboiron-Ladouceur, “High-speed two-mode switch for mode-division multiplexing optical networks,” Optica 4(9), 1098–1102 (2017). [CrossRef]
27. F. Guo, D. Lu, R. K. Zhang, H. T. Wang, and C. Ji, “A two-mode (de) multiplexer based on multimode interferometer coupler and y-junction on inp substrate,” IEEE Photonics J. 8(1), 1–8 (2016). [CrossRef]
28. W. Y. Chan and H. P. Chan, “Reconfigurable two-mode mux/demux device,” Opt. Express 22(8), 9282–9290 (2014). [CrossRef] [PubMed]
29. C. Sun, Y. Yu, G. Chen, and X. Zhang, “Integrated switchable mode exchange for reconfigurable mode-multiplexing optical networks,” Opt. Lett. 41(14), 3257–3260 (2016). [CrossRef] [PubMed]
30. R. B. Priti and O. Liboiron-Ladouceur, “A reconfigurable multimode demultiplexer/switch for mode-multiplexed silicon photonics interconnects,” IEEE J. Sel. Top. Quantum Electron. 24(6), 1–10 (2018). [CrossRef]
31. B. Stern, X. L. Zhu, C. P. Chen, L. D. Tzuang, J. Cardenas, K. Bergman, and M. Lipson, “On-chip mode-division multiplexing switch,” Optica 2(6), 530–535 (2015). [CrossRef]
32. S. Wang, H. Wu, H. K. Tsang, and D. Dai, “Monolithically integrated reconfigurable add-drop multiplexer for mode-division-multiplexing systems,” Opt. Lett. 41(22), 5298–5301 (2016). [CrossRef] [PubMed]
33. H. Jia, T. Zhou, L. Zhang, J. Ding, X. Fu, and L. Yang, “Optical switch compatible with wavelength division multiplexing and mode division multiplexing for photonic networks-on-chip,” Opt. Express 25(17), 20698–20707 (2017). [CrossRef] [PubMed]
34. Y. Zhang, Q. Zhu, Y. He, C. Qiu, Y. Su, and R. Soref, “Silicon 1× 2 mode and polarization-selective switch,” in Optical Fiber Communications Conference and Exhibition (OFC) (IEEE, 2017), pp. 1–3.
35. N. P. Diamantopoulos, M. Hayashi, Y. Yoshida, A. Maruta, R. Maruyama, N. Kuwaki, K. Takenaga, H. Uemura, S. Matsuo, and K. Kitayama, “Mode-selective optical packet switching in mode-division multiplexing networks,” Opt. Express 23(18), 23660–23666 (2015). [CrossRef] [PubMed]
36. W. Y. Liu, “Flexible on-chip mode-division switching with a new mode converter design,” https://arxiv.org/abs/1712.01658.
37. H. F. Xiao, L. Deng, G. L. Zhao, Z. L. Liu, Y. H. Meng, X. N. Guo, G. P. Liu, S. Liu, J. F. Ding, and Y. H. Tian, “Optical mode switch based on multimode interference couplers,” J. Opt. 19(2), 025802 (2017). [CrossRef]
38. L. B. Soldano and E. Pennings, “Optical multi-mode interference devices based on self-imaging: principles and applications,” J. Lightwave Technol. 13(4), 615–627 (1995). [CrossRef]
39. A. Yariv, Optical Electronics in Modern Communications, 6th ed. (Oxford University, 2007).
40. R. Scarmozzino, A. Gopinath, R. Pregla, and S. Helfert, “Numerical Techniques for Modeling Guided-Wave Photonic Devices,” IEEE J. Sel. Top. Quantum Electron. 6(1), 150–162 (2000). [CrossRef]
41. R. Scarmozzino and R. M. Osgood, “Comparison of finite-difference and Fourier-transform solutions of the parabolic wave equation with emphasis on integrated-optics applications,” J. Opt. Soc. Am. A 8(5), 724–731 (1991). [CrossRef]
