Open Access
Add to BibSonomy
Abstract
A Fe-InP-based planar array antenna-coupled InGaAs/InAlAs multiple quantum well (MQW) optical phase modulator is proposed and demonstrated for radio over fiber (RoF) applications with 60 GHz-band millimeter-wave wireless signals. The modulator comprises five types of five-layer asymmetric coupled quantum wells (FACQWs) and a two-element array antenna. The FACQWs are designed to have a significant electric-field-induced refractive index change with small electric fields induced in the antenna. In the fabricated modulator, a carrier-to-sideband ratio (CSR) of up to 45.9 dB was successfully obtained at a power density of 11 W/m2, corresponding to a phase shift of 10.1 mrad. Furthermore, data transmission of a 2 GHz modulated wave with a 60 GHz wireless carrier wave was demonstrated.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Millimeter-wave signals have attracted much attention for the realization of next-generation high-speed wireless communication systems. Among them, 60 GHz-band millimeter-wave signals realize high-speed and large-capacity wireless communications. However, their range is limited because millimeter waves are easily attenuated in air owing to oxygen’s lack of diffraction and absorption [1,2]. Millimeter-wave radio-over-fiber (MMW RoF) is a technology for high-speed and low-loss transmission of large-capacity wireless signal information over optical fibers in mobile front-haul/back-haul. It contributes to the realization of high-speed wireless communication systems using 60 GHz-band millimeter waves [3,4]. Efficient conversion from wireless signals to optical signals and from optical signals to wireless signals is essential for RoF technology.
A typical RoF system mainly consists of an antenna unit that receives wireless signals, a coaxial cable that transmits electric signals, and an optical modulator that converts electric signals to optical signals. Since electric signals flowing through coaxial cables are high-frequencies in the MMW RoF system, the influence of the skin effect cannot be ignored, and signal quality degradation, such as distortion and delay, becomes a problem. In addition, an external power supply is also required to operate the optical modulator, which can also be a noise source. Therefore, antenna-coupled optical modulators, which can directly modulate optical signals with wireless signals, are attracting attention. The modulator part of this type of device is composed of an electrooptic material (EO material), and electric fields derived from high-frequency electromagnetic waves propagating in the space received by the antenna are directly applied to the modulator to be operated. The devices have been proposed and demonstrated on the basis of EO polymer/plasmonic waveguides [5,6], EO polymer waveguides [7,8], lithium niobate (LN) [9,10], and compound semiconductors [11,12]. Since these devices do not require cables or external power supplies, they have many advantages, such as low power consumption and high signal quality in high-frequency wireless communication systems. However, they have various problems, such as large device size owing to low modulation efficiency and increased insertion loss when integrated with a semiconductor laser as a light source.
InGaAs/InAAs multiple quantum wells (MQWs) have a relatively large electric field refractive index change due to the quantum confinement Stark effect (QCSE). In particular, using a five-layer asymmetric coupled quantum well (FACQW) can achieve large electric-field-induced refractive index change characteristics with less absorption loss [13]. We expect that the use of the FACQWs in the waveguide core layer of antenna-coupled optical modulators and the use of p-InP and n-InP as cladding layers in a PIN structure will enable efficient wireless-to-optical signal conversion and thus contribute to the miniaturization of the devices. In addition, since the device is composed of III-V compound semiconductors (InP), the same as C-band laser diodes, it can be easily integrated with semiconductor lasers, which are the light sources.
We previously reported an n-InP/glass-based MQW optical phase modulator [11]. This device is based on n-InP, which has high conductivity and is unsuitable for antenna substrates. A glass substrate on the top surface of the antenna reduces the effective dielectric constant of the antenna substrate and increases the aperture area of the antenna, thereby improving modulation efficiency. In this device, a carrier sideband ratio (CSR) of 62.7 dB, corresponding to a modulation depth of 1.46 mrad, was observed when irradiated with 57.5 GHz radio signals, which is the first experimental demonstration of a compound semiconductor-based antenna-coupled optical modulator. However, this modulation efficiency was insufficient for practical use. We also reported a device using semi-insulating InP (Fe-InP) as a substrate, which can suppress leakage current [14]. In this device, 59.9 dB CSR was observed at 60.0 GHz radio signal irradiation, which corresponds to a modulation depth of 2.06 mrad.
In this paper, we propose and demonstrate a newly designed MQW optical modulator with arrayed antennas for 60 GHz radio signals to improve modulation efficiency. Five types of FACQWs are redesigned, and a waveguide core layer with a wide electric field operating region and large electric field refractive index change characteristics is achieved. An array antenna with two elements is also designed to improve the reception efficiency for wireless signals. The modulation performance of the fabricated is characterized and discussed.
2. Device structure and simulation of optical and MMW characteristics
2.1 Device structure
Schematic overall and cross-sectional views of a Fe-InP-based two-electrode-array antenna-coupled multiple quantum well optical phase modulator are shown in Figs. 1(a) and (b). This device consists of an optical modulator and rectangular planar antenna electrodes with a small gap on the modulator. The modulator has a single transverse electric (TE) mode high mesa with a height of 2.1 µm and a width of 1.45 µm. The propagation loss of this waveguide is estimated to be approximately 1.4 dB/mm. The waveguides are embedded with benzocyclobutene (BCB), a low-k resin. The substrate of this modulator is 200 µm-thick semi-insulating Fe-doped InP (Fe-InP). The semi-insulating InP substrate suppresses the leakage current when the planar antenna receives radio signals. As a result, the electric field derived from the wireless signal can be efficiently induced in the waveguide core layer. The waveguide consists of a core layer of multiple quantum wells and cladding layers of p-type InP (p-InP) and n-type InP (n-InP) above and below the core layer. This PIN structure allows the applied electric field from the antenna to be concentrated in the core layer of the waveguide [15]. The thickness of the semi-insulating substrate affects the electric field enhancement factor and the resonance frequency. The enhancement factor is maximum at the substrate thickness of 150 µm. In this study, a 200 µm-thick substrate was used considering easy handling of the fragile InP chip.
This device has a small gap of 3.0 µm in the center of the planar antenna and ground electrode because the electric field derived from the received radio signals can be concentrated in the gap [16]. In other words, larger electric fields than the received ones are generated at the antenna gap. Therefore, waveguides are placed below the antenna gap to apply this efficiently amplified electric field to the waveguide core layer in the z-direction [14]. Here, we define the electric field enhancement factor, |Ez/E0|, as the ratio of the electric field applied in the z direction at the waveguide core layer, Ez, to the electric field of the radio signal received by the antenna, E0.
Figure 2 shows the calculated distribution of the amplified electric fields applied to the waveguides when receiving 60-GHz-band radio signals. We use ANSYS HFSS as a finite element method (FEM) simulator to calculate the MMW characteristics of this device. The material properties used in the simulations are listed in Table 1. As shown in Fig. 2(b), the electric field is concentrated in the non-doped layer (I-layer) with a high electric field enhancement factor.
Fig. 2. Simulated distribution of z-component electric field Ez under irradiation of 60 GHz band wireless signals. (a) Top view, (b) Cross-sectional view, where Antenna width Wa = 650 µm, Antenna length La = 840 µm, Antenna gap Ga = 3 µm, Antenna height ta = 0.78 µm, Waveguide height twav = 2.1 µm, Distance between waveguides dwav = 2.8 µm, Substrate height tsub = 200 µm.
Table 1. Material properties used for simulations.
2.2 Waveguide structure
The waveguide core layer of the modulator comprises In0.53Ga0.47As/In0.52Al0.48As FACQWs, which exhibit giant refractive index change characteristics due to the unique QCSE [13,17]. The FACQW in the modulator needs to be operated with the small change in the electric field induced by an antenna around the electric field caused by a built-in potential in the PIN structure. In addition, the characteristics of the electric-field refractive index change in the FACQW are highly sensitive to the electric field in the core layer. The electric field is non-uniform in the non-doped intrinsic layer owing to the space charge caused by residual impurities. Therefore, we designed and optimized the FACQW core layer considering the non-uniformity of the electric field and its operation region.
Figure 3(a) shows the schematic view of the designed waveguide structure. The core layer is composed of five types of FACQW structures, which makes it possible to exhibit a large refractive index change even with the non-uniform electric field strongly depending on the residual impurity density.
Fig. 3. (a) Schematic cross-sectional view of waveguide. (b) Conduction potential profile of 6 and 7th FACQW.
The conduction band potential profile of one of the FACQW structures (the 6 and 7th FACQWs) is shown in Fig. 3(b). Unlike the previous report [11], a p + -contact layer was not inserted on top of the p-InP upper cladding layer. The doping level of the p-InP layer may not be sufficient for low-resistance ohmic contact. However, unlike standard optical modulators, the electric field is applied to the space around the antenna gap to the waveguide core layer, and it is confirmed with a simulation that the conductivity in the contact does not significantly influence the characteristics of the electric field enhancement. It is confirmed with a simulation that the conductivity in the contact does not have a significant influence on the characteristics of the electric field enhancement. Therefore, it does not necessarily have an ohmic contact between the metal electrode and p-InP. In addition, when a highly doped p + -layer is placed directly under the antenna, the effective phase shifter length (electric field application region) is increased, which will deteriorate radio-optical phase matching. We did not adopt a highly doped contact layer for the above reasons.
Figure 4 shows the calculated built-in electric field distribution of the core layer with various residual impurity densities. This distribution shows that the built-in electric field in the core layer significantly differs depending on both position and residual impurity density. In particular, the electric field markedly varies in the region close to the n-InP. The residual impurity could be difficult to control. For these reasons, the FACQW structure was optimized for each position to exhibit a large electric field-induced refractive index change.
Fig. 4. Built-in electric field distribution in FACQW core layer. (p = 2.0 × 1018 cm−3, n = 3.0 × 1018 cm−3)
Figure 5 shows the refractive index change characteristics of each newly designed FACQW and the operation electric field region. The 1−7th FACQWs in layers are designed to exhibit a large refractive index change under each applied electric field region because the variation in the electric field caused by the residual impurities is relatively small, while the 8−12th FACQWs have the thinner central InAlAs barrier layer and are designed to have a wide operating range for high tolerance to the non-uniformity of the applied electric field.
Here, we discuss the design of the FACQW for the proposed modulator. The change in absorption coefficient in a QW is caused by the change in the overlap integral of electron and hole wavefunctions, leading to the change in the refractive index according to the Kramers–Kronig relation [18]. Therefore, the characteristics of the electric field-induced refractive index change can be designed by altering the thicknesses of the well and barrier layers and the ratio of the left and right side well layers. The basic designing principle of the InGaAs FACQW is discussed in Ref. [15]. The FACQW is regarded as a coupled QW composed of a single QW and a small coupled QW. For example, the FACQW shown in Fig. 3(b) comprises QW1 and QW2, where QW1 is a 22-monolayer (ML) single QW, and QW2 is an (8 + 4 + 17)-ML coupled QW. In the following discussion, an FACQW structure is denoted as t1_ t2_t3_ t4_ t5, where ti is the thickness of the i-th layer in an FACQW in ML, as shown in Fig. 3(b). For instance, the FACQW in Fig. 3(b) is expressed as 22_8_8_4_17. Figure 6(a) shows the wavefunction distributions in 19_8_8_4_17 and 22_8_8_4_17 under various electric fields. In both FACQWs, the electrons in the ground state (E1) and in the first excited state (E2) are respectively distributed in QW1 and QW2 under no electric field. Although E1 and E2 start to move to the opposite QW with the increase in the electric field, E1 and E2 in 19_8_8_4_17 move to the opposite QW respectively at a lower electric field than those in 22_8_8_4_17. This is because the electron energy levels in QW1 and QW2 in 19_8_8_4_17 are closer than those in 22_8_8_4_17. E1 and E2 can, therefore, tunnel through the center barrier layer at a lower electric field. On the other hand, the wavefunction distributions of the holes do not change. Figure 6(b) shows the refractive index change characteristics of the FACQW with various t1. The operation electric field region, where the wavefunction distributions of E1 and E2 alter, can be changed by adjusting t1.
Fig. 6. (a) Wavefunction distributions in FACQW structures 19_8_8_4_17 and 22_8_8_4_17 under various electric fields. (b) Refractive index change characteristics of FACQW with various well layer thickness t1.
Figure 7(a) shows the wavefunction distributions in 22_7_8_4_17 and 22_10_8_4_17 under various electric fields. In both FACQWs, E1 and E2 are respectively distributed in QW1 and QW2 under no electric field. Although E1 and E2 start to move to the opposite QW with the increase in the electric field, E1 and E2 in 22_7_8_4_17 move to the opposite QW more gradually than those in 22_10_8_4_17. This is because the thickness of the center barrier layer t2 in 22_7_8_4_17 is smaller, and E1 and E2 start tunneling through the center barrier layer at a lower electric field. Figure 7(b) shows the refractive index change characteristics of the FACQW with various t2. The gradient of the refractive index change Δn versus the applied electric field F characteristic in the operation electric field region can be changed by adjusting t2. With the decrease in t2, the electric field-induced change in the refractive index of the FACQW becomes more gradual, which is suitable for widening the operation electric field region.
Fig. 7. (a) Wavefunction distributions in FACQW structures 22_7_8_4_17 and 22_10_8_4_17 under various electric fields. (b) Refractive index change characteristics of FACQW with various barrier layer thickness t2.
In the waveguide structure, the doping concentration and the thickness of the InP cladding layers were also optimized. Since the electric field is concentrated in the waveguide core layer owing to the PIN structure, the doping concentrations of the p-InP and n-InP layers greatly influence the enhancement of the induced electric field in the core layer.
Figure 8(a) shows the calculated relationships between the doping concentration in the n-InP upper layer and the electric field enhancement factor. The doping concentration (carrier density) in the p-InP layer is set to 2.0 × 1018 cm−3, and the thicknesses of the p-InP and n-InP are assumed to be 800 and 600 nm, respectively. The doping concentrations are converted to conductivities of the layer [19]. As shown in Fig. 8(a), the highest electric field enhancement factor is obtained at n = 3.0 × 1017 cm−3. Figure 6(b) shows the calculated relationships between the thickness of the p-InP layer and the electric field enhancement factor. The highest electric field enhancement factor is obtained at tp-InP = 900 nm. We determined the waveguide structure shown in Fig. 3(a) from these results.
Fig. 8. Calculated relationship between electric field enhancement factor and (a) doping density of n-InP, n, (b) thickness of p-InP, tp-InP.
Figure 9 shows the calculated electric field refractive index change characteristics of the waveguide, as shown in Fig. 3(a). The designed waveguide has characteristics that are highly robust to the variation in residual impurity density. The electrooptic coefficients of FACQW rFACQW are expected to be 15.2 × 10−10– 25.1 × 10−10 m/V, which are several tens of times higher than those of LN 31 × 10−12 m/V [20]. Figure 10 shows the refractive index change characteristics of the designed core layer with an impurity density of 3.0 × 1015 cm−3 for various wavelengths. It shows that a large change in refractive index is expected even in the longer wavelength region.
Fig. 9. Refractive index change characteristics of designed core layer with various impurity densities.
Fig. 10. Refractive index change characteristics of designed core layer with impurity density of 3.0 × 1015 cm−3 for various wavelengths.
2.3 Antenna structure
When x-polarized MMW signals are incident on a planar rectangular antenna, surface currents are generated on the antenna. If a small gap is placed in the planar antenna, displacement currents are induced at the antenna gap owing to the continuity of the currents. As a result, electric fields derived from the MMW signals are generated around the gap. The proposed modulator uses these electric fields for phase modulation of light. These theories are discussed in detail in [21]. The electric field in the z-direction at the antenna gap is given by
The geometry of the rectangular antenna in this device has already been discussed in [14]. The dependence of the accumulated phase shift on Da for the array device with 2 antenna elements illustrated in Fig. 1(a) is shown in Fig. 11. On the basis of this result, Da is set to 1450 µm in the device in order to maximize the accumulated phase shift.
3. Device fabrication
The FACQW waveguide layer structure shown in Fig. 2(a) was grown on a 2-inch Fe-InP wafer by molecular beam epitaxy (MBE) and was cut to a chip with a size of $1 \times 1\; \textrm{c}{\textrm{m}^2}$. Several devices are simultaneously fabricated on this chip. The fabrication process is summarized in Fig. 12.
First, a SiO2 film and negative electron beam (EB) resist (OEBR-CAN) were deposited on the chip. Then, waveguide patterns were formed on the EB resist by EB lithography. The waveguide pattern formed on the resist was transferred to the SiO2 film by CHF3-based inductive coupled plasma reactive ion etching (ICP-RIE). After removing the resist, we further transferred the waveguide patterns from the SiO2 film to the chip by Cl2-based ICP-RIE. After etching, the unneeded SiO2 film is completely removed by wet etching with buffered hydrofluoric acid (BHF). Then, benzocyclobutene (BCB) was deposited using a spin-coating method to embed the ridge waveguides and thermally solidified by baking. After that, BCB on the top of the waveguide was removed by capacitive coupled plasma-RIE (CCP-RIE). Positive EB resist (OEBR-CAP) was deposited, and the antenna pattern was formed by EB lithography. AlSi is then deposited, and the antenna is formed by a lift-off method. Finally, the substrate was polished, and a Au electrode was deposited on the backside of the substrate.
Figure 13 shows optical microscope image and scanning microscope (SEM) images of the fabricated device. The size of the device is 4.5 × 4.5 cm2, which is approximately the designed value. The antenna gap width is 3.6 µm, which is larger than the designed value of 3.0 µm.
Fig. 13. Optical microscope image and scanning microscope images of fabricated device. (a) Overall view. (b) Top view.
Figure 14 is an SEM image of a cross-section of the fabricated device. The BCB between the antenna and the waveguide was completely removed by etching. The height of the waveguide ridge is 2.8 µm, which is higher than the designed height of 2.1 µm, resulting in the formation of an unintentional step in the antenna.
4. Measurement of modulation characteristics
4.1 Carrier-to-sideband ratio (CSR)
The phase-modulated optical signal fPM (t) is assumed to be [23]
The second and third terms in Eq. (10) are called the first-order (n = 1) sideband signals. The peak difference between the amplitude of the carrier signal and the first-order sideband signal is defined as the carrier-to-sideband ratio (CSR), which is an important factor for the modulation depth m. According to Eq. (10), the relationship between CSR and m is given by,
4.2 Modulation characteristics under irradiation of 60 GHz band signal
Figure 15 shows the experimental setup for the modulation characteristics of the fabricated device. The 60 GHz-band radio signals are generated by the optical two-tone technique, and the optical signals are converted into RF signals by a uni-traveling carrier photodiode (UTC-PD). The RF signals are incident on a horn antenna (Antenna efficiency is approximately 10 dB) through an amplifier, and the 60 GHz-band radio signals are irradiated to the fabricated device on a metal pedestal. For transmission measurements, a tunable laser with an output power of up to 14.55 dBm. This output light is set to the TE mode by the polarization controller and then incident into the input port of the fabricated device through the polarization-maintaining fiber. The light output from the output port of the fabricated device is input to the optical spectrum analyzer through an optical fiber. When the device receives 60 GHz wireless signals from the horn antenna, and this wireless signal directly modulates the optical signals, the output signals are observed by the optical spectrum analyzer.
Fig. 15. Experimental setup of modulation measurement under irradiation of 60 GHz band wireless signals.
Figure 16 shows the measured modulation characteristics of the fabricated device. Theoretical characteristics are also shown. Residual impurities in the core layer were taken into account in the calculation because the residual impurities tend to decrease not only the EO coefficient but also the electric field enhancement factor because of its conductivity. Therefore, it is assumed that rFACQW is 15.2 × 10−10 m/V for the impurity density of 7.0 × 1015 cm−3 and that |Ez/E0| is 16% of the value expected when the core layer is a perfect insulator. The effect of the metal device holder is also considered. The metal device holder is expected to decrease |Ez/E0| to approximately 40% of the value obtained when a device holder made of insulator is used. Figure 16(a) shows one of the optical spectra of the output from the antenna-coupled optical modulator under the irradiation of 60 GHz band radio signals. A sideband signal with a CSR of 45.9 dB (λop = 1540 nm, fRF = 57.4 GHz) at minimum is observed under the condition of the distance between the horn antenna and the fabricated device of 25 mm and the RF power density input to the device of P ∼ 11 W/m2. The CSR of 45.9 dB corresponds to a phase shift $\Delta \phi $ of 10.1 mrad. Figure 16(b) shows the CSR dependence on 60 GHz band radio frequency. The CSRs of approximately 45−55 dB were observed in the 60 GHz band. Two CSR peaks exist around 57.5 GHz and 60.0 GHz. However, the calculated CSR curve indicates that there should be a single CSR peak.
Fig. 16. Measured modulation characteristics of fabricated device (blue line and dot). Theoretical characteristics are also shown (dotted yellow line). (a) One of measured optical spectra. (b) RF frequency dependence of CSR. (c) Wavelength dependence of CSR.
Figure 16(c) shows the CSR dependence on the optical wavelength (s = 2.5 mm, fRF = 57.4 GHz). The CSR decreases with the increase in wavelength. This result is consistent with the theoretical characteristics of the refractive index change of the FACQW shown in Fig. 11.
In spite of the deterioration in the characteristics, the CSR obtained with this device with two antenna elements is larger than those with other similar modulators compared to other devices that use a rectangular antenna with a small gap (for example, 50 dB for a LiNbO3-based device with 9 elements [16], 42 dB for an EO polymer-based device with 10 elements [7]). Therefore, we believe that the experimental results show the superiority of the proposed modulator. The EO coefficient was estimated to be 36 × 10−11 m/V from the estimated wireless signal irradiation power and electric field enhancement factor. This EO coefficient is approximately 2.5 times higher than that obtained in our previous report [11].
4.3 Demodulation characteristics after modulation with 60 GHz band wireless signal
The experimental setup of data transmission of the fabricated device is shown in Fig. 17. In addition to the experimental setup shown in Fig. 15, a 1 km optical fiber is connected from the output port of the fabricated device to convert the phase-modulated signals into the intensity-modulated signals by using the wavelength dispersion of this optical fiber [24]. After the signal conversion, the signals are amplified using an erbium-doped fiber amplifier (EDFA). The signals are demodulated into 60 GHz-band electrical signals by the UTC-PD and then observed by the spectrum analyzer. The data modulation of the 60 GHz wireless carrier wave is modulated by a 2 GHz binary phase shift keying (BPSK) digital signal (symbol rate: 1 Mbps) generated by a digital RF signal generator (Agilent E4432B). The modulated radio signals are irradiated to the device, and the modulator is operated.
Figure 18 shows the demodulation results of simple 60 GHz-band radio carrier signals without data signals. The demodulated spectra of the 57.4 GHz and 65.0 GHz radio carrier signals are shown in Figs. 18(a) and (b), respectively. A maximum of −53 dBm electric signal with a signal-to-noise ratio (SNR) of 21 dB was observed. Figure 19(a) shows the transmission data signal irradiated to the fabricated device, and Fig. 19(b) shows the received data signal demodulated after passing through the device and the 1 km-long fiber. The received signals are recognizable as BPSK signals, although there are some data dispersions in the received signal in comparison with the transmission signals. The SNR of the demodulated signals under the data modulation was approximately 12 dB, and the bit error rate (BER) was estimated to be 2.6 × 10−4.
Fig. 18. Demodulation signals of non-modulated RF signal, (a) 57.4 GHz, (b) 65.0 GHz, after propagating in fabricated device.
Fig. 19. Data-transfer characteristics of fabricated device. (a) Transmission signal (b) Received signals after propagating in fabricated device.
5. Conclusion
The two-element planar array antenna-coupled InGaAs/InAlAs MQW optical phase modulator has been proposed and demonstrated for 60 GHz-band millimeter-wave wireless signals. The modulator is composed of five types of FACQWs and the two-electrode array antenna. The FACQWs are designed to have a significant electric-field-induced refractive index change with small electric fields induced in the antenna. In the fabricated modulator, the CSR of as low as 45.9 dB was successfully obtained at a power density of 11 W/m2, which corresponds to the phase shift of 10.1 mrad and a significantly large value for a modulator with an antenna of two elements. The EO coefficient of the FACQW was estimated to be 36 × 10−11 m/V, which is 10 times higher than that of LiNbO3. In addition, data transmission of 2 GHz BPSK digital data signals with a 60 GHz wireless carrier wave was demonstrated. The proposed optical modulator is expected to make a significant contribution to highly efficient wireless-optical signal conversion in RoF systems.
Funding
Support Center for Advanced Telecommunications Technology Research Foundation; Japan Society for the Promotion of Science (21H01841, 21K18169).
Acknowledgments
The authors express sincere thanks to Professor Joo-Hyong Noh of Kanto Gakuin University for supporting the electromagnetic analysis using the FEM simulator. The authors also thank the Advanced ICT Laboratory of National Institute of Information and Communications Technology (NICT) for support in device fabrication. This work was partly supported by a Grant-in-Aid for Scientific Research B (No. 21H01841 and No.21K18169) from the Ministry of Education, Culture, Sports, Science and Technology. A part of this work was conducted at Takeda Sentanchi Supercleanroom, the University of Tokyo, supported by ARIM Japan (JPMXP1222UT1049).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. T. S. Rappaport, S. Sun, R. Mayzus, et al., “Millimeter Wave Mobile Communications for 5 G Cellular: It Will Work!” IEEE Access 1, 335–349 (2013). [CrossRef]
2. T. Yilmaz and O. B. Akan, “State-of-the-Art and Research Challenges for Consumer Wireless Communications at 60 GHz,” IEEE Trans. Consum. Electron. 62(3), 216–225 (2016). [CrossRef]
3. J. Capmany and D. Novak, “Microwave photonics combines two worlds,” Nat. Photonics 1(6), 319–330 (2007). [CrossRef]
4. C. Lim, A. Nirmalathas, M. Bakaul, et al., “Fiber-wireless networks and subsystem technologies,” J. Lightwave Technol. 28(4), 390–405 (2010). [CrossRef]
5. Y. Salamin, B. Baeuerle, W. Heni, et al., “Microwave plasmonic mixer in a transparent fibre-wireless link,” Nat. Photonics 12(12), 749–753 (2018). [CrossRef]
6. Y. Horst, T. Blatter, L. Kulmer, et al., “Transparent optical-THz-optical link at 240/192 Gbit/s over 5/115 m enabled by plasmonics,” J. Lightwave Technol. 40(6), 1690–1697 (2022). [CrossRef]
7. T. Kaji, I. Morohashi, Y. Tomonari, et al., “D-band optical modulators using electrooptic polymer waveguides and non-coplanar patch antennas,” Opt. Express 31(11), 17112–17121 (2023). [CrossRef]
8. T. Kaji, I. Morohashi, Y. Tomonari, et al., “W-band optical modulators using electrooptic polymer waveguides and patch antenna arrays,” Opt. Express 29(19), 29604–29614 (2021). [CrossRef]
9. H. Murata, “Millimeter-wave-band electrooptic modulators using antenna-coupled electrodes for microwave photonic applications,” J. Lightwave Technol. 38(19), 5485–5491 (2020). [CrossRef]
10. Y. N. Wijayanto, A. Kanno, H. Murata, et al., “W-band millimeter-wave patch antennas on optical modulator for runway security systems,” 2017 IEEE Conference on Antenna Measurements & Applications, 79–82 (2017).
11. Y. Miyazeki, H. Yokohashi, S. Kodama, et al., “InGaAs/InAlAs multiple-quantum-well optical modulator integrated with a planar antenna for a millimeter-wave radio-over-fiber system,” Opt. Express 28(8), 11583–11596 (2020). [CrossRef]
12. H. Kamada and T. Arakawa, “Proposal of Highly Efficient Quantum Well Microring Resonator-Loaded Optical Phase Modulator Integrated with Antenna-Coupled Electrodes for Radio-over-Fiber,” Photonics 8(2), 37 (2021). [CrossRef]
13. T. Arakawa, T. Toya, M. Ushigome, et al., “InGaAs/InAlAs Five-Layer Asymmetric Coupled Quantum Well Exhibiting Giant Electrorefractive Index Change,” Jpn. J. Appl. Phys. 50(3R), 032204 (2011). [CrossRef]
14. G. Sekiguchi, R. Nakazawa, S. Nakamori, et al., “Development of planar antenna integrated quantum well phase modulator for 60 GHz band signal,” 35th URSI General Assembly and Scientific Symposium (GASS), 2023 (2023).
15. Y. Miyazeki and T. Arakawa, “Proposal of InGaAs/InAlAs multiple quantum well Mach–Zehnder modulator integrated with array of planar antennas,” Jpn. J. Appl. Phys. 58(SJ), SJJE05 (2019). [CrossRef]
16. Y. N. Wijayanto, H. Murata, and Y. Okamura, “Electrooptic Millimeter-Wave–Lightwave Signal Converters Suspended to Gap-Embedded Patch Antennas on Low-k Dielectric Materials,” IEEE J. Sel. Top. Quantum Electron. 19(6), 33–41 (2013). [CrossRef]
17. T. Arakawa, T. Hariki, Y. Amma, et al., “Low-Voltage Mach–Zehnder Modulator with InGaAs/InAlAs Five-Layer Asymmetric Coupled Quantum Well,” Jpn. J. Appl. Phys. 51(4R), 042203 (2012). [CrossRef]
18. J. Weiner, “Quadratic electrooptic effect due to the quantum-confined Stark effect in quantum wells,” Appl. Phys. Lett. 50(13), 842–844 (1987). [CrossRef]
19. D. Pasquariello, E. S. Björlin, D. Lasaosa, et al., “Selective undercut etching of InGaAs and InGaAsP quantum wells of for improved performance of long-wavelength optoelectronic devices,” J. Lightwave Technol. 24(3), 1470–1477 (2006). [CrossRef]
20. E. H. Turner, “High-frequency electrooptic coefficients of lithium niobate,” Appl. Phys. Lett. 8(11), 303–304 (1966). [CrossRef]
21. D. H. Park, V. R. Pagán, T. E. Murphy, et al., “Free space millimeter wave-coupled electrooptic high speed nonlinear polymer phase modulator with in-plane slotted patch antennas,” Opt. Express 23(7), 9464–9476 (2015). [CrossRef]
22. M. Born and E. Wolf, Principles of Optics, 7th ed. (Cambridge University, 1999).
23. J. D. Gibson, Principles of digital and analog communications, 2nd ed. (Macmillan, 1993).
24. K. Takaba and H. Murata, “Microwave/millimeter-wave signal control using high-speed optical modulator and optical fiber dispersion effect,” IEICE Tech. Rep.121(352), 91–94 (2022).
