Open Access
Add to BibSonomy
Abstract
We demonstrate an InP deep-ridge type multi-quantum-wells (MQWs) waveguide-based active polarization control scheme by using two positive-intrinsic-negative diode phase shifters. The polarization state of horizontal/vertical linearly polarized input light is rotated along the ± 45°-linear polarization axis on the Poincaré sphere by 45°-eigenmode-rotated first phase shifter 1 and subsequently rotated along the vertical/horizontal axis by second normal phase shifter 2. The rotation of the eigenmode axes is obtained by using the surface plasmonic effect. The effective index of the waveguide is changed via quantum-confined Stark effect in the MQW core. The length of each phase shifter is 240 µm. The applied reverse bias voltages are −2.7 V and −1.95 V for the phase shifter 1 and 2, respectively.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Polarization control and adjustment in traditional optical fiber communications is a prerequisite process to achieve the optimal performance in optical devices. The majority of conventional optical devices are polarization-sensitive unless they have been carefully designed to be polarization-insensitive [1]. Recent advances in photonic integrated circuits (PICs) such as silicon and InP photonics are based on high-index contrast core/cladding systems [2,3]. In such high-index contrast conditions, birefringence between the fundamental propagation modes in the optical waveguides is huge and strongly affected even by small fabrication dimension errors. This makes the design of polarization-insensitive optical devices practically difficult. It is common to choose one polarization state in the high-index contrast optical device design. For silicon photonics, horizontally linear polarized (also known as transverse-electric, TE) light is chosen to utilize the higher effective index than the effective index for vertically linear polarized (transverse-magnetic, TM) light. In case of InP photonics as well, TE-polarized light is utilized due to the dominant TE-polarized light gain in the heavy-hole transitions of multi-quantum-wells (MQWs) structures.
Today’s optical communication technology uses advanced modulation formats such as quadrature phase-shift keying (QPSK), quadrature amplitude modulation (QAM), and polarization-division multiplexing (PDM). These coherent optical communication technologies increase spectral efficiency sending more than one bit per symbol to satisfy the requirements of exploding data capacity in optical links [4]. As a cost of the higher efficiency, coherent optical systems are complicated and require more than several tens of optical components. High-index contrast PICs are well suited to the requirements of modern optical communication technology [3,5]. Nowadays, more than 1,700 photonic functions have been integrated in a single chip [6]. Despite advances in integration, the polarization control of light is still performed by off-chip bulk optics. To achieve integration of polarization control, integrated passive TE/TM mode converters have been extensively studied [7–11]. In active polarization control, conventional optical fiber-based controllers use fiber squeezers to induce birefringence at 0° and 45° to the horizontal plane [12]. Polarization manipulation by the fiber squeezing birefringence effect can be mathematically expressed as the product of Jones matrices [13]. It can be represented by the cascaded 3-dB couplers, phase shifters, and polarization rotators as a PICs version [14]. Such integration has been realized in silicon photonics [15,16]. Another active control scheme is to use the cascade of passive polarization rotator/phase shifter pairs. This scheme has been demonstrated in thin-film inserted polymer waveguide, silicon photonics, and InP photonics [17–20].
For the design of the integrated polarization manipulation, the polarization rotation process is necessary. In InP photonics, the InP waveguide is basically an epitaxially grown ridge type structure with a strong birefringence geometry between the lateral and vertical axes. This strong birefringence hinders the rotation process of the eigenmode axes in the InP waveguides. There are a number of InP based polarization rotator results that have successfully overcome this constraint [21–28]. However, these results may be difficult to implement in active device configurations as their performance strongly depends on the shape of the waveguide. In our previous work, we proposed an InP based polarization rotator using the surface plasmonic effect [29]. The surface plasmonic effect shows the strong polarzation interaction between the metal/dielectric interface and light. It can be used in TE/TM mode converters [30]. By using this strong light-metal interaction phenomenon, it is possible to overcome the birefringence in the InP waveguide geometry and rotate the eigenmode axes with respect to the position of the Au in the partially dry-etched topcladding layer. The simple structure of our polarization rotator is suitable for active operation in which the waveguide operates as a phase shifter.
In this paper, we propose an InP based active polarization control scheme by using two positive-intrinsic-negative (PIN) diode phase shifters. The eigenmode axes of the first phase shifter is rotated at 45° to the horizontal axis by using the surface plasmonic effect. The second phase shifter is a normal phase shifter (0° to the horizontal axis). Its operation principle follows that of the conventional fiber squeezing phase retarders [13]. In general, three phase shifters (0°, 45°, 0°) are needed to convert an arbitrary input polarization state to an arbitrary desired output polarization state. Here, we assume that horizontal/vertical linearly polarized light is launched into the waveguide because TE-polarized light is predominantly used in InP photonics. By using these phase shifters, we demonstrate active polarization control in InP photonics.
2. Device structure and simulation results
2.1 InP polarization controller concept
Figure 1(a) shows a schematic diagram of the device. The waveguide is a deep-ridge type structure and has an InGaAsP MQWs core. It operates as singlemode condition. Two propagation modes of the singlemode waveguide are aligned to the horizontal (TE) and vertical (TM) directions. The polarization controller consists of two cascaded PIN diode phase shifters. Both have lengths of 240 µm each. The effective refractive index in the waveguide is changed by using the quantum-confined Stark effect (QCSE) in the MQWs core. In the first phase shifter (phase shifter 1), the eigenmode axes of the waveguide are rotated by 45° to the horizontal axis as shown in Fig. 1(b). Launched TE/TM modes in phase shifter 1 are split evenly into two 45°-rotated propagation modes. The effective index change in PIN diode 1 by reverse voltage biasing induces birefringence change between two 45°-rotated propagation modes. After passing through PIN diode 1, the split light is combined with a polarization state change resulting from the relative phase variation which is the product of the birefringence change and the length of the phase shifter. This polarization change by phase shifter 1 is visualized on the Poincaré sphere as a rotation of the polarization state along the ± 45°-linear polarization axis as shown in Fig. 1(d). The second phase shifter (phase shifter 2) is based on a normal singlemode waveguide. The eigenmode axes of the waveguide are along the horizontal and vertical directions as shown in Fig. 1(c). The state of polarization (SoP) change by the effective index change in PIN diode 2 is represented on the Poincaré sphere as a polarization state rotation along the horizontal/vertical linear polarization axis as shown in Fig. 1(d). The changed polarization state by two phase shifters is coupled to a normal InP waveguide. It is highly birefringent. The state of polarization is continuously changing during propagation along the waveguide. Therefore, we cannot get the exact controlled polarization state at the end of the phase shifters. Instead, we can change the polarization state of the output of the InP waveguide which incorporates the polarization change by InP waveguide after phase shifters. By combining the two phase shifters, input horizontal (TE)/vertical (TM) polarized light can be controlled into any polarization state.
Fig. 1. (a) Schematic of the integrated InP polarization controller with deep-ridge waveguide and InGaAsP MQWs core. The polarization controller consists of two 240 µm long PIN diode phase shifters. (b) The eigenmode axes of phase shifter 1 is rotated by 45° to the horizontal axis by using the surface plasmonic effect. The rotation angle is controlled by the position of the Au in the topcladding layer. (c) Phase shifter 2 is a normal waveguide. The eigenmode axes are along the horizontal (TE) and vertical (TM) directions. (d) When TE/TM modes are launched into the device, the effective index change in phase shifter 1 rotates the SoP of light along the ± 45° linear polarization axis. The effective index change in phase shifter 2 rotates the SoP of light along the horizontal/vertical linear polarization axis.
2.2 InP polarization controller waveguide structure
The InP polarization controller has four types of waveguide structures. Figure 2 shows the cross-sectional schematics of the waveguide structures. Figure 2(a) shows a cross-sectional view of the normal phase shifter 2. The PIN diode is reverse biased through a 40-nm thick p+-InGaAs layer to induce the effective index variation in the MQWs core via QCSE. The MQWs consist of 22 pairs of 1.48Q InGaAsP well (10 nm) and 1.15Q InGaAsP barrier (8 nm). Its photoluminescence peak is in the range of 1.41–1.42 µm. It has a 100 nm thickness of lower-separate confinement heterostructure (SCH) and a 130 nm thickness of upper-SCH. The total InGaAsP core dimension is 634 nm in height and 1.1 µm in width. It has the InP topcladding layer with a thickness of 475 nm. The InP topcladding layer is p-doped (Zn) with three doping levels: undoped, 5 × 1017 /cm3, and 1 × 1018 /cm3. A p+-InGaAs ohmic contact layer (2 × 1019 /cm3) is placed above the InP topcladding layer. This p+-InGaAs layer delivers an electric potential from a metal electrode to the entire area of the phase shifter. The waveguide is covered with a 500 nm thick SiN layer. Above SiN layer Ti/Au layer is covered. The Au layer can be used as a heater electrode. Figure 2(b) shows a cross-sectional view of the phase shifter 1 with its eigenmode axes rotated by 45°. The dimension and structure of the waveguide is the same as the normal phase shifter except for a Au via structure which rotates the eigenmode axes of the waveguide. The bottom corner of the Au via is shifted 290 nm from the waveguide center and penetrates the InP topcladding layer by 150 nm. Figure 2(c) shows a cross-sectional view of a phase shifter with the metal electrode interconnection. The Ti/Au layer is connected to the topcladding p+-InGaAs layer. The length of the interconnection is 8 µm. Figure 2(d) shows a cross-sectional view of a passive deep-ridge waveguide. The p+-InGaAs and p-doped layers in the InP topcladding are removed to reduce propagation loss.
Fig. 2. Cross-sectional view of (a) normal phase shifter 2, (b) phase shifter 1 with eigenmode axes rotated by 45°, (c) metal electrode interconnected phase shifter, and (d) passive waveguide.
2.3 Eigenmode axes rotation principle and simulation
The rotation of the eigenmode axes in the waveguide is accomplished by using the surface plasmonic effect. Surface plasmon polaritons (SPPs) are electromagnetic waves perpendicular to planar metal-dielectric interface. As shown in Fig. 3(a), the Au via is placed above the InGaAsP core and displaced from the center of the waveguide core. This causes the mode profile of the InGaAsP core to be affected by the SPPs on both the side and bottom Au surfaces. The fundamental eigenmodes of the waveguide become a hybrid form of the InGaAsP core dielectric and SPP modes. As a result, the eigenmode axes are rotated with respect to the position of the bottom corner of the Au via. To calculate the rotation angle (θ) of the eigenmode axes, the rotation angle parameter R is calculated as follows [31]:
Fig. 3. Eigenmode axes rotation simulation. (a) Simulation condition. The bottom corner of the Au via is shifted from the waveguide center by X and etched down from the waveguide top by Y. (b) Contour map of eigenmode 1 axis rotation angle with respect to the metal lateral shift (X) and the partial etch depth (Y). (c) Contour map of eigenmode 2 axis rotation angle with respect to the metal lateral shift (X) and the partial etch depth (Y). (d) Contour map of eigenmode 1 propagation loss (dB/100 µm) with respect to the metal lateral shift (X) and the partial etch depth (Y). (e) Contour map of eigenmode 2 propagation loss (dB/100 µm) with respect to the metal lateral shift (X) and the partial etch depth (Y).
As in Eq. (1), a rotation of an eigenmode in the rectangular waveguide can be expressed as an intensity ratio change between horizontal and longitudinal field intensity distributions which constitute the eigenmode. Therefore, the birefringence of the rotated eigenmodes can be modulated by horizontal/longitudinal birefringence change in MQWs by QCSE.
The mode profile characteristics of the waveguide with respect to the position of the Au via are calculated using the full-vector finite difference method (FDM). The refractive indices of the materials used in the simulation at 1.55 µm are set as follows: SiN 2.05, n-InP 3.1624, p- InP 3.1652, SCH 3.309, QW barrier 3.309, QW well 3.478, p-InGaAs 3.55 + 0.0617i, Au 0.559 + 9.81i, respectively. We used a mesh size of 2 nm/1 nm in the MQWs X/Y region and 0.5 nm within ± 40 nm of the Au surface. Figure 3(b) shows a contour map of the eigenmode 1 axis rotation angle distribution with respect to the metal lateral shift (X) and the partial etch depth (Y). The green region represents a deviation within ± 3° from 45°. Figure 3(c) shows a contour map of the eigenmode 2 axis rotation angle distribution. Figure 3(d) shows a contour map of eigenmode 1 propagation loss. Figure 3(e) shows a contour map of eigenmode 2 propagation loss. The scale bar for the propagation loss contour map is dB/100 µm. In consideration of the propagation loss, we chose a lateral shift of 290 nm and etch depth of 150 nm, as denoted by the X mark in Fig. 3.
3. Fabrication process
The fabrication process of the polarization controller is shown in Fig. 4. The InGaAs layer is first patterned (Fig. 4(a)). A 150 nm-thick SiN layer is deposited by plasma-enhanced chemical vapor deposition (PECVD) followed by photolithography, and then etched by reactive ion etching (RIE). The patterned SiN layer is used as a hard mask pattern in the wet chemical etching of the InGaAS layer with a ratio of 10H2O:1H3PO4:1H2O2. After wet chemical etching, the InP layer is etched by RIE to remove the heavily p-doped topcladding layer except at the phase shifter region to reduce propagation loss. Then, the SiN layer is removed with 6:1 buffered oxide etch (BOE). The InGaAs layer is placed only on the phase shifter area. Figure 4(b) shows the deep-ridge waveguide structure patterned by RIE. 300 nm thick SiN is deposited by PECVD as the hard mask layer. The waveguide pattern is transferred onto the SiN layer by photolithography and RIE. Next, the InP and MQWs layers are etched by RIE, followed by the deposition of a 500 nm thick SiN layer by PECVD (Fig. 4(c)). The topcladding layer is then partially etched at the 45°-eigenmode rotated phase shifter area (Fig. 4(d)). The covered SiN layer is patterned by electron beam lithography (EBL, nB4-Nanobeam ltd.) and RIE. Then, opened InGaAs-InP layers are etched by RIE at an etching depth of 150 nm. The InGaAs contact area for the metal electrode interconnection is opened (Fig. 4(e)) The contact area in the SiN layer is patterned by EBL and RIE. The p-metal electrode is patterned by lift-off process (Fig. 4(f)). Photolithography is performed to form the electrode photoresist pattern. Ti/Au layers of 30 nm/300 nm thickness are deposited by electron beam evaporation, after which the photoresist is removed by acetone immersion lift-off. Rapid-thermal annealing (RTA) is then performed to form the p-ohmic contact. 200 nm thick Au layer is deposited by sputtering (Fig. 4(g)). The deposited Au is filled into the partially etched topcladding layer in phase shifter 1. In addition, the Au layer becomes a seed layer for subsequent electroplating. Photoresist is patterned by photolithography, and electroplating is performed to thicken the electrode (Fig. 4(h)). After electroplating, the photoresist is removed by acetone immersion. The Au seed layer is etched with Au etchant. The two phase shifter electrodes are then isolated from each other (Fig. 4(i)). Lapping is performed to thin down the InP wafer to 120 µm for easy scribing and chip breaking (Fig. 4(j)). After lapping, 50 nm/350 nm thick Cr/Au layers are deposited as n-metal electrodes onto the backside of the wafer by evaporation. Finally, RTA is performed to form n-ohmic contacts. The inset of Fig. 4(k) shows a bird’s eye view scanning electron microscope (SEM) image of the fabricated normal phase shifter cross-section. Pt was deposited on Au during the focused-ion beam (FIB) milling process to prevent milling damage. The height of the deep-ridge waveguide is about 1.8 µm but appears distorted in the figure due to the SEM imaging angle. The actual height of the waveguide is about 2.3 µm. Figure 4(l) shows a SEM image of the cross section of the 45°-eigenmode rotated phase shifter 1. The Au via is shifted by about 240 nm from the waveguide center and penetrates into the InP topcladding layer by about 150 nm. Figure 4(m) is a microscope image of the phase shifters.
Fig. 4. Fabrication process for the InP polarization controller. (a) InGaAs wet etching for phase shifter area definition and p-doped InP layer etching. (b) Deep-ridge waveguide etching. (c) SiN deposition. (d) SiN and InP top cladding partial etching. (e) SiN etching to open InGaAs contact. (f) Ti/Au deposition and lift-off for p-electrode formation. (g) Au seed layer sputtering for electroplating. (h) Electroplating. (i) Au seed layer etching. (j) Lapping and backside Cr/Au deposition for n-electrode formation. (k) SEM image of the normal phase shifter 2. (l) SEM image of the 45°-rotated eigenmode phase shifter 1. (m) Microscope image of the phase shifters.
4. Device characterization
Figure 5 shows the experimental setup for characterizing the fabricated device. A continuous-wave (CW) laser light of 1.55-µm wavelength is launched from an external-cavity laser (ECL, Keysight 81989A). The launched TM mode light from the ECL goes through a polarization maintaining fiber (PMF) and is converted into light propagating in free space via a fiberport collimator (FC, Thorlabs). A rotating quarter-wave (λ/4) retarder is used to control the light to be linear-elliptical-circular polarization. The free space light is coupled into a PMF through FC again. The light is focused into the InP polarization controller chip through a lensed PMF. The InP chip is placed on a thermoelectric cooler (TEC)-controlled sample holder. The temperature is maintained at 25 degree. The polarization of the incident light in the InP chip is adjusted to TE mode with a fiber rotator (FR). The SoP of the light launched into the chip is manipulated by the fabricated polarization controller. A reverse DC voltage is applied to phase shifter 1 and a reverse sweep voltage is applied to phase shifter 2. The required relative phase shift between the two propagation modes in the phase shifters are π and 2π in phase shifters 1 and 2, respectively, for full control of the SoP. These phase shifts are plotted on the Poincaré sphere as half circular and full circular sweeps, respectively. The polarization-manipulated output light from the chip is coupled into a lensed singlemode fiber and analyzed with a polarimeter (PAX1000IR1, Thorlabs). Figure 6(a) shows the measured locus of the polarization controller plotted on the Poincaré sphere. Phase shifter 1 was reverse biased at −2.7 V for the half- circular trajectory and phase shifter 2 at −1.95 V for the complete circular trajectory. The full sweep voltage of −1.95 V in phase shifter 2 is much lower than the previous result of −14 V [20]. The voltage reduction is due to the tighter light confinement and higher electric field intensity in the MQWs enhancing QCSE in the 1.1 µm wide and 22-pair MQWs deep-ridge waveguide here than the wider 7-pair MQWs shallow-ridge type waveguide in the previous work. Compared to the −1.95 V reverse bias at phase shifter 2 for the complete circular trajectory, phase shifter 1 requires higher reverse voltage of −2.7 V even for the shorter half circular trajectory. This indicates that the eigenmode axes of phase shifter 2 (0°-rotation) coincide with the QCSE birefringence orientation and the orthogonal QCSE birefringence is used fully. The TE mode is more strongly affected than the TM mode. In phase shifter 1, the eigenmode axes are rotated by 45° and, therefore, both propagation modes are affected by the superposition of QCSE birefringence, resulting in weaker relative phase shift. Figure 6(b) shows the trajectories by the polarization controller when right circular elliptical mode is launched into the controller. The input TE mode is changed to right circular elliptical polarization by rotating the λ/4 retarder. Then, the polarization rotation axis on the Poincaré sphere of the phase shifter 1 (45°-rotation) is changed from ± 45°-linear polarization axis to right/left polarization axis. The SoP rotation along right/left polarization axis in phase shifter 1 becomes a linear polarization rotation input to phase shifter 2. Then, polarization rotation axis of phase shifter 2 (0°-rotation) is changed from horizontal/vertical polarization axis. It is described by the Stokes vector, [cos(2T), sin(2T), 0], where T is the input fast-axis angle with respect to the horizontal axis. T is varied by phase shifter 1. As a result, the circle trajectory by the phase shifter 2 voltage sweep rotates along right/left circular polarization axis with respect to the voltage biasing of phase shifter 1. In this case, the reverse bias voltage in phase shifter 1 required for the half-circular trajectory is −2.3 V, which indicates a better phase shift result than the TE mode input in Fig. 6(a). The measured device loss is about −5.5 dB and the operation loss is below −2 dB.
Fig. 5. Schematic of the characterization setup. ECL: external-cavity laser. PMF: polarization-maintaining fiber. FC: fiberport collimator. λ/4: rotating quarter-wave retarder. FR: fiber rotator. SMF: singlemode fiber.
Fig. 6. SoP trajectories on the Poincaré sphere for (a) TE-polarized input and (b) right-circular elliptically polarized input. In (a), reverse biasing of phase shifter 2 sweeps out a circular trajectory around the horizontal/vertical polarization axis, and the position of the circle along the ± 45° polarization axis is controlled by phase shifter 1. The applied reverse voltage is −2.7 V for phase shifter 1 and −1.95 V for phase shifter 2. In (b), the polarization rotation axis of phase shifter 1 is changed to the right/left circular polarization axis. For phase shifter 2, the rotation angle is varied according to the Stokes vector [cos(2T), sin(2T), 0], where T is the input fast-axis angle with respect to the horizontal axis. T is varied by phase shifter 1. The required reverse voltage in phase shifter 1 is −2.3 V.
5. Conclusions
We have proposed an InP based polarization controller scheme using two phase shifters. Besides a normal phase shifter, for which the eigenmode axes are along the horizontal and vertical directions, it is necessary to have a phase shifter in which the eigenmode axes are rotated by 45° to the horizontal axis. We achieve this rotation by using the surface plasmonic effect. The topcladding layer of the deep-ridge waveguide is partially etched and filled in with Au. The propagating light through the MQWs core is affected by the surface plasmon mode in Au, and the resulting hybrid modes are rotated with respect to the position of the bottom corner of the Au layer. The fabrication process of the polarization controller is totally compatible with conventional InP based Mach-Zehnder modulators and lasers. The birefringence effect in the phase shifters is obtained by using the QCSE in the MQWs. By exploiting the strong QCSE birefringence, tight light confinement, and high electric field intensity of the deep-ridge waveguide structure, the operating reverse voltage for the normal phase shifter is −1.95 V, and for the 45°-axes rotated phase shifter the reverse voltages are −2.7 V and −2.3 V for TE-polarized input and right-circular elliptically polarized input, respectively. Each phase shifter has a length of 240 µm. The total device length is 510 µm. This is the smallest integrated polarization controller to the best of the authors’ knowledge. We expect that our polarization control scheme can be incorporated with polarization-division multiplexing, high-speed coherent polarization modulations, quantum key distribution in quantum communications, and polarization modulating lasers.
Funding
Ministry of Science, ICT and Future Planning (GK19N0400).
Disclosures
The authors declare no conflicts of interest.
References
1. H. Takahashi, Y. Hibino, and I. Nishi, “Polarization-insensitive arrayed-waveguide grating Wavelength Multiplexer on Silicon,” Opt. Lett. 17(7), 499–501 (1992). [CrossRef]
2. E. Lee, S. Lee, B. O, S. Park, M. Chung, K. Kim, and S. Song, “Fabrication and integration of VLSI micro/nano-photonic circuit board,” Microelectron. Eng. 83(4-9), 1767–1772 (2006). [CrossRef]
3. S. Arafin and L. A. Coldren, “Advanced InP Photonic Integrated Circuits for Communication and Sensing,” IEEE J. Sel. Top. Quantum Electron. 24(1), 1–12 (2018). [CrossRef]
4. K. Roberts, D. Beckett, D. Boertjes, J. Berthold, and C. Laperle, “100G and beyond with digital coherent signal processing,” IEEE Commun. Mag. 48(7), 62–69 (2010). [CrossRef]
5. R. Nagarajan, M. Kato, J. Pleumeekers, P. Evans, S. Corzine, S. Hurtt, A. Dentai, S. Murthy, M. Missey, R. Muthiah, R. A. Salvatore, C. Joyner, R. Schneider, M. Ziari, F. Kish, S. Member, and D. Welch, “InP Photonic Integrated Circuits,” IEEE J. Sel. Top. Quantum Electron. 16(5), 1113–1125 (2010). [CrossRef]
6. W. Williams, P. Evans, D. Gold, M. Lai, T. Vallaitis, A. James, D. Pavinski, P. Studenkov, M. Ziari, D. Welch, F. Kish, M. Missey, J. Summers, J. Sena, J. Ou-Yang, S. Corzine, and M. Fisher, “Monolithic InP-based coherent transmitter photonic integrated circuit with 2.25 Tbit/s capacity,” Electron. Lett. 50(16), 1150–1152 (2014). [CrossRef]
7. M. R. Watts and H. A. Haus, “Integrated mode-evolution-based polarization rotators,” Opt. Lett. 30(2), 138 (2005). [CrossRef]
8. H. Fukuda, K. Yamada, T. Tsuchizawa, T. Watanabe, H. Shinojima, and S. Itabashi, “Polarization rotator based on silicon wire waveguides,” Opt. Express 16(4), 2628 (2008). [CrossRef]
9. J. Zhang, M. Yu, G. Q. Lo, and D. L. Kwong, “Silicon-waveguide-based mode evolution polarization rotator,” IEEE J. Sel. Top. Quantum Electron. 16(1), 53–60 (2010). [CrossRef]
10. M. Aamer, A. M. Gutierrez, A. Brimont, D. Vermeulen, G. Roelkens, J. M. Fedeli, A. Hakansson, and P. Sanchis, “CMOS compatible silicon-on-insulator polarization rotator based on symmetry breaking of the waveguide cross section,” IEEE Photonics Technol. Lett. 24(22), 2031–2034 (2012). [CrossRef]
11. J. N. Caspers, M. Z. Alam, and M. Mojahedi, “Compact hybrid plasmonic polarization rotator,” Opt. Lett. 37(22), 4615–4617 (2012). [CrossRef]
12. S. Yao, Polarization in Fiber Systems: Squeezing out More Bandwidth (Laurin Publishing, 2003).
13. R. Noe, H. Heidrich, and D. Hoffmann, “Endless Polarization Control Systems for Coherent Optics,” J. Lightwave Technol. 6(7), 1199–1208 (1988). [CrossRef]
14. L. Möller, “WDM polarization controller in PLC technology,” IEEE Photonics Technol. Lett. 13(6), 585–587 (2001). [CrossRef]
15. W. D. Sacher, T. Barwicz, B. J. F. Taylor, and J. K. S. Poon, “Polarization rotator-splitters in standard active silicon photonics platforms,” Opt. Express 22(4), 3777 (2014). [CrossRef]
16. W. D. Sacher, Y. Huang, L. Ding, T. Barwicz, J. C. Mikkelsen, B. J. F. Taylor, G.-Q. Lo, and J. K. S. Poon, “Polarization rotator-splitters and controllers in a Si3N4-on-SOI integrated photonics platform,” Opt. Express 22(9), 11167 (2014). [CrossRef]
17. J. W. Kim, S. H. Park, W.-S. Chu, and M. C. Oh, “Integrated-optic polarization controllers incorporating polymer waveguide birefringence modulator,” Opt. Express 20(11), 12443 (2012). [CrossRef]
18. J. D. Sarmiento-Merenguel, R. Halir, X. Le Roux, C. Alonso-Ramos, L. Vivien, P. Cheben, E. Durán-Valdeiglesias, I. Molina-Fernández, D. Marris-Morini, D.-X. Xu, J. H. Schmid, S. Janz, and A. Ortega-Moñux, “Demonstration of integrated polarization control with a 40 dB range in extinction ratio,” Optica 2(12), 1019 (2015). [CrossRef]
19. T. Tanemura and Y. Nakano, “Compact InP stokes-vector modulator and receiver circuits for short-reach direct-detection optical links,” IEICE Trans. Electron. 101(7), 594–601 (2018). [CrossRef]
20. M. Kazi, S. Ghosh, M. Sugiyama, and Y. Nakano, “Offset-Quantum-Well-Based Integrated Stokes Vector Modulator on InP,” IEEE Photonics Technol. Lett. 31(15), 1233–1236 (2019). [CrossRef]
21. M. V Kotlyar, L. Bolla, M. Midrio, L. O. Faolain, and F. Thomas, “Compact polarization converter in InP-based material,” Opt. Express 13(13), 5040–5045 (2005). [CrossRef]
22. B. M. Holmes and D. C. Hutchin, “Realization of novel low-loss monolithically integrated passive waveguide mode converters,” IEEE Photonics Technol. Lett. 18(1), 43–45 (2006). [CrossRef]
23. L. M. Augustin, S. Member, J. J. G. M. Van Der Tol, E. J. Geluk, and M. K. Smit, “Short Polarization Converter Optimized for Active – Passive Integration in InGaAsP – InP,” IEEE Photonics Technol. Lett. 19(20), 1673–1675 (2007). [CrossRef]
24. D. M. Beggs, M. Midrio, and T. F. Krauss, “Compact polarization rotators for integrated polarization diversity in InP-based waveguides,” Opt. Lett. 32(15), 2176–2178 (2007). [CrossRef]
25. R. Zhang, H. G. Bach, and M. Schell, “Polarization rotator for InP rib waveguide,” Opt. Lett. 37(3), 335–337 (2012). [CrossRef]
26. J. J. G. M. Van Der Tol, M. Felicetti, and M. K. Smit, “Increasing tolerance in passive integrated optical polarization converters,” J. Lightwave Technol. 30(17), 2884–2889 (2012). [CrossRef]
27. M. Zaitsu, T. Tanemura, A. Higo, Y. Nakano, and I. Ingaasp, “Experimental demonstration of self-aligned InP / InGaAsP polarization converter for polarization multiplexed photonic integrated circuits,” Opt. Express 21(6), 6910 (2013). [CrossRef]
28. D. C. Hutchings and B. M. Holmes, “A waveguide polarization toolset design based on mode beating,” IEEE Photonics J. 3(3), 450–461 (2011). [CrossRef]
29. S. An and O.-K. Kwon, “Integrated InP polarization rotator using the plasmonic effect,” Opt. Express 26(2), 1305 (2018). [CrossRef]
30. S. An, H. Lee, Y. Jeong, Y. C. Jun, S. G. Lee, S. Park, and E. Lee, “Nanofocusing of light using three-dimensional plasmonic mode conversion,” Opt. Express 21(23), 27816–27825 (2013). [CrossRef]
31. H. Deng, D. O. Yevick, C. Brooks, and P. E. Jessop, “Design rules for slanted-angle polarization rotators,” J. Lightwave Technol. 23(1), 432–445 (2005). [CrossRef]
