Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

All-silicon and in-line integration of variable optical attenuators and photodetectors based on submicrometer rib waveguides

Open Access Open Access

Abstract

We demonstrate a monolithic integration of variable optical attenuators (VOAs) and photodetectors (PDs) based on submicrometer silicon (Si) rib waveguide with p-i-n diode structure for near infrared (NIR) light. To make the Si PD absorptive for NIR, we introduced lattice defects at the rib core by means of argon ion implantation. At reverse bias of 5 V, the PD exhibits dark current of ~1 nA, responsivity of ~65 mA/W at 1560-nm wavelength, and a 3-dB cut-off frequency of ~350 MHz, while the VOA shows ~100 MHz. The PD has an absorption coefficient as low as ~0.5 cm−1, which is favorable for an in in-line PD configuration, where the PD absorbs a small portion of the optical power. For DC light, the PD precisely detects the optical power attenuated by the VOA with a detection range of over 40 dB. The 3-dB cut-off frequency of synchronous operation between the VOA and PD is ~50 MHz, which is limited by RF noise originating from the VOA drive current. Putting an isolation groove between the VOA and PD is effective for avoiding performance degradation in DC and RF operation.

©2010 Optical Society of America

1. Introduction

Near infrared (NIR) photodetectors (PDs) integrated with silicon (Si) waveguides are indispensible for photon-electron conversion in Si photonics, where various photonic components are integrated on a Si platform. Two types of waveguide PDs for NIR (1300 - 1550 nm) have been demonstrated: germanium (Ge) PDs [1,2] and Si PDs with lattice defects. Particularly, the latter utilize defect-level-mediated absorption for the Si sub-band-gap by means of ion-implantation of Si [35] or helium [6]. The NIR absorption of irradiated Si has been attributed to the presence of divacancies or V2 [7]. In practice, such PDs offer feedback control when they are synchronously operated with other Si-waveguide-based photonic devices, such as optical modulators or variable optical attenuators (VOAs). The operation principle is as follows: a PD monitors the optical power and feeds the information back to the active photonic devices through electronic circuits, which changes the optical properties of the active devices, such as the refractive index or free carrier absorption, by means of the thermo-optic effect [8] or plasma dispersion effect of Si [9]. Particularly, the Si PD provides simplicity and compatibility for monolithic integration with other Si-based photonic components or electronic circuits. In addition, we can realize an in-line PD configuration for feedback power monitoring, in which a small portion of the optical power is absorbed by the PD and the rest of it propagates out through the waveguide due to the low absorption coefficient of the defective region of the PD (<1 cm−1) for NIR light. In contrast, due to the large absorption coefficient of Ge (~4,000 cm−1) for NIR light, Ge PDs should be implemented with a waveguide tap-coupler to monitor a small fraction of the light power from the bus waveguide [10]. In fact, some researchers have investigated the feasibility of monolithic fabrication between Si defect PDs and Si VOAs [11] or Si Mach-Zehnder modulators [12]. Furthermore, for micro-ring modulators, Geis et al. have proposed athermal feedback control by placing a resistor heater on the modulator, which is connected to an in-line Si p-i-p or n-i-n phototransistor [5]. Liu et al. have theoretically studied signal equalization for bust-mode wavelength division multiplexing (WDM) communications by implementing an in-line integration of a VOA and PD [6,13]. However, there has been no demonstration of synchronous operation of in-line integrated PDs and carrier-injection-based optical modulation devices based on Si waveguides.

In this paper, we report all-Si and in-line integration of VOAs and PDs along submicrometer Si rib waveguides. We investigate the characteristics of the defect PD from the perspective of the in-line PD configuration. We present synchronous operation between the VOA and PD for DC light and radio-frequency (RF) modulated optical signals.

2. Fabrication

Figure 1(a) shows a cross-section of the VOA and PD. The starting substrate was a four-inch silicon-on-insulator (SOI) wafer, which has a boron-doped p-type Si top layer with resistivity of 13.5 - 22.5 Ω cm. The thicknesses of the Si top layer and buried oxide (BOX) are 200 nm and 3 µm, respectively. Si rib waveguides were defined by e-beam lithography and transferred by electron cyclotron resonance (ECR) plasma etching [14]. The rib-core dimensions are 600 nm (width) and 200 nm (thickness), while the slab thickness is 100 nm. To electrically separate adjacent VOA-PD devices, we etched the Si slab down to the BOX to make isolation grooves. However, an isolation groove was not formed between the VOA and PD. Instead, the VOA and PD are separated by a 50-µm-long Si rib waveguide. Boron and phosphorous were ion-implanted for n + and p + contacts to make a lateral p-i-n diode. The peak concentration of ion implantation was intended to be ~1020 cm−3. Note that the intrinsic region (i-region) is lightly p-doped with boron concentration of ~1015 cm−3. The p-i-n diode had an approximately 4-µm-wide intrinsic region, which became ~3 µm after recrystallization annealing at 1000 °C for 60 minutes in nitrogen ambient. To form an absorptive region of the PD, we introduced lattice defects into the Si rib core by implanting argon ions with a dose of 1012 cm−2 at 100 keV. SiO2 was deposited by plasma enhanced chemical vapor deposition (PE-CVD) to a thickness of ~1 µm as an overcladding layer. Then, aluminum was deposited and etched to make electrode pads. Finally, the implanted defects were activated by annealing at 300 °C for 30 minutes. Figure 1(b) shows an optical microscope image of the VOA and PD with 2.5-mm length.

 figure: Fig. 1

Fig. 1 (a) Cross-section of a p-i-n diode for the VOA and PD on a submicrometer rib waveguide. (b) Optical microscope image of the in-line VOA and PD with 2.5-mm length.

Download Full Size | PDF

3. Characterization results

3.1 PD

As a continuous-wave (CW) light source, we used amplified spontaneous emission (ASE) light with 1560-nm peak wavelength. To guide the CW light into the waveguide, we placed adiabatic tapers at the both facets of the chip with a broadened rib-core of 3-µm width. The CW light was coupled in to the waveguide and coupled out to an external power monitor with a lensed fiber. Polarization was fixed to the TE mode. Current-voltage (I-V) curves and the photocurrent of the PD were acquired with a semiconductor parameter analyzer (Agilent 4155C). Figure 2(a) shows I-V curves of the PD under dark and illumination conditions. The dark current is smaller than 1 nA up to the reverse bias of 5 V. The illuminated I-V curve in Fig. 2(a) was obtained when ~0.35 mW of optical power was guided through the PD. The optical power guided into the PD was estimated by considering propagation loss of the rib waveguide and coupling loss at the waveguide facet. Figure 2(b) shows photocurrents versus optical powers at the PD. They have a linear relation in ~40-dB optical power range. From that, we obtained a responsivity of over 65 mA/W, which corresponds to 5.4% of internal quantum efficiency. Considering the dark current and responsivity, the minimum detectable optical power for a CW light is as low as ~10 nW (−50 dBm). To launch a CW light with various wavelengths, we used a tunable fiber mode-lock laser, and the wavelength dependence of the responsivity was obtained as shown in Fig. 2(c). The PD exhibits a uniform responsivity for the C-band wavelength range (1528 - 1560 nm). Figure 2(d) shows responsivity as a function of the PD length. It is linearly proportional, which implies that the defect-level-mediated absorption is proportional to the PD length and thus the propagation loss solely due to the defect-induced absorption corresponds to the absorption coefficient of the PD. Figure 2(e) shows transmission in decibels for different PD lengths. The propagation loss was estimated as ~4.1 dB/cm, which corresponds to ~1 dB of propagation loss for the 2.5-mm-long PD. Note that this value includes the propagation loss from sidewall scattering and p + and n + ion-irradiated damage. We measured the propagation loss of the Si p-i-n rib waveguide to be ~2 dB/cm. By excluding the excess propagation loss from the p-i-n diode structure, the absorption coefficient of the PD is estimated to be ~2 dB/cm (~0.5 cm−1). Further, change of transmission is negligible even when the reverse bias is applied, which is the operation condition of the PD. For instance, the reverse bias of 5 V increases transmission only by ~0.1 dB even though a reverse bias depletes the carriers in the intrinsic region. These properties are favorable for the in-line integration of the PD.

 figure: Fig. 2

Fig. 2 (a) I-V curves of the PD under dark and illumination, (b) photocurrent (left) and responsivity (right) versus optical power at the PD, and (c) responsivity for the C-band wavelength. The PD length is 2.5 mm and a reverse bias of 5 V is applied. (d) Responsivity as a function of the PD length at −5 V. (e) Transmission against the PD length.

Download Full Size | PDF

3.2 Individual bandwidth of the VOA and PD

We measured the 3-dB cut-off bandwidth separately for the VOA and PD. For this measurement, we commonly used the ASE light source and a network analyzer (Agilent E5071C) to launch sinuous signals. First, the frequency response of the VOA was measured by modulating the VOA with the signal from the network analyzer, as shown in Fig. 3(a) . To match 50-Ω impedance between the VOA and transmission cable, a fixed resistor was inserted into a micro-probe near the VOA. The output optical signals of the device chip were guided to an external opto-electro (O-E) converter and then collected by the network analyzer. Second, the frequency response of the PD was measured by launching a modulated light into the PD. A signal from a network analyzer modulated a CW ASE light through a commercial lithium niobate (LN) modulator (Sumitomo Osaka Cement T-MZH1.5-10), as shown in Fig. 3(b). The RF signals from the PD were directly collected by the network analyzer. Figure 3(c) shows frequency responses of the VOA and PD. The 3-dB cut-off frequency of the VOA is ~100 MHz for injection current of 5 and 10 mA. The 3-dB cut-off frequency of the VOA does not vary regardless of injection current up to 50 mA. Above 50 mA, however, we could not obtain a clear frequency response because the output optical power attenuated by the VOA was too low. On the other hand, the bandwidth of the PD depends on the reverse bias. Reverse bias increases the electric field in the intrinsic region of the PD and, consequently, the transit time of photo-generated carriers is shortened. The 3-dB cut-off frequencies of the PD are ~350 and ~800 MHz for reverse bias of 5 and 15 V, respectively.

 figure: Fig. 3

Fig. 3 Block diagram of the frequency-response measurement setups for (a) the VOA and (b) PD. (c) Frequency response of the VOA at injection current of 5 and 10 mA, and that of the PD under reverse bias of 5 and 15 V.

Download Full Size | PDF

3.3 Synchronous operation between the VOA and PD

First, we investigated the synchronous operation between the VOA and PD for the CW light condition. We measured the photocurrent at the PD and simultaneously monitored the optical power attenuated by the VOA with an external optical power meter, as we varied the injection current to the VOA from 0 to 150 mA. The photocurrent measured at the PD accurately detects optical power attenuated by the VOA. From the results in Fig. 2(b), it is clear that the PD has a dynamic range of over 40 dB in optical power, which means the optical attenuation by VOA of ~40 dB is sufficiently detected by the PD. However, we observed a dark-current increase at the PD with increasing injection current. This indicates that there exists an electrical leakage path between the VOA and PD. If the increment of the dark current exceeds the photocurrent, the photocurrent is not able to reliably detect the attenuated power. Therefore, it is important to electrically separate the VOA and PD. A solution to this problem is discussed later.

Second, we investigated the synchronous operation between the VOA and PD for RF signals. The synchronous frequency response was measured with the experimental setup shown in Fig. 5(a) . Note that we used two types of the data-collection methods with a network analyzer. One was a direct collection in the electrical domain from the on-chip Si PD. In the other, the optical signal modulated by the VOA was converted to an electrical one at an external O-E converter and sent back to the network analyzer while a reverse bias was applied to the on-chip PD. Here, the bandwidth of the O-E converter is over 10 GHz. The frequency responses for the direct collection from the on-chip PD are shown in Fig. 5(b). Synchronous 3-dB cut-off frequencies are ~50 and ~60 MHz for the reverse bias of 5 and 15 V at the PD, respectively, when 10 mA of current is injected to the VOA. In contrast, the synchronous 3-dB cut-off frequency collected by the external O-E converter is ~100 MHz regardless of reverse bias of the PD, as shown in Fig. 5(c). We also confirmed that the bandwidth of the PD remains constant regardless of the injection current to the VOA. In theory, the synchronous 3-dB bandwidth should be ~100 MHz, which is the synchronous bandwidth limited by the VOA. However, the measured 3-dB frequency is significantly lower than 100 MHz. Moreover, we observed two intensity dips from 100 to 200 MHz and a noisy frequency band, which becomes prominent over 200 MHz, as shown in Fig. 5(b).

 figure: Fig. 5

Fig. 5 (a) Block diagram of the setup for measuring the synchronous frequency response between the VOA and PD. The frequency response of the synchronous VOA-PD at injection current of 10 mA to the VOA are shown in (b) and (c), respectively, for 1) the direct electrical probing by the PD and 2) electrical conversion at the O-E converter with reverse bias applied to the PD.

Download Full Size | PDF

To study the origin of the degraded synchronous bandwidth and the high-frequency noise when the on-chip Si PD directly collects optical signals modulated by the VOA, we measured the frequency response of the integrated VOA-PD devices in a dark condition i.e., with no light guided through the devices. Before the measurement, we calibrated the measurement setup shown in Fig. 5(a) with an impedance standard substrate (Cascade Microtec Inc. 103-726) instead of the VOA-PD chip. By calibrating with the impedance standard substrate, we can assess the frequency interference purely generated from the VOA and PD. During the measurement, we applied reverse bias of 5 and 15 V to the PD and injected 10 mA of current to the VOA. Figure 6(a) shows two pairs of VOA-PD devices used to measure magnitude and phase as the frequency was being swept. One VOA-PD pair laid along the same waveguide is referred to as “in-line”, while the other VOA-PD pair located along the adjacent waveguide and separated along the isolation groove is referred to as “out-of-line”. Particularly, the out-of-line VOA-PD was measured to see the effect of the isolation groove on the high-frequency interference.

 figure: Fig. 6

Fig. 6 (a) Optical microscope image of frequency response measured in a dark condition between the VOA and PD. In-line and out-of-line refer to a VOA-PD pair placed along the same waveguide (VOA-PD1) and separated along the isolation groove (VOA-PD2), respectively. Magnitudes [(b),(d)] and phases [(c),(e)] as a function of frequency at the reverse bias of 5 and 15 V of the PD and at 10-mA injection current to the VOA after calibration with an impedance standard substrate. (b) and (c) for the in-line VOA-PD1, while (d) and (e) for the out-of-line VOA-PD2.

Download Full Size | PDF

Figure 6(b) shows a significant rise of the noise component from roughly 50 MHz and the noise spectrum at the frequency range over 200 MHz is similar to the high-frequency noise over 200 MHz in Fig. 5(b). This is convincing evidence that the high-frequency noise in Fig. 5(b) is due to the high frequency noise generated along the in-line VOA-PD. The phase shift was also taken into account. As shown in Fig. 6(c), there is as significant phase shift toward 180° from 100 - 125 MHz. In the phase range of 90 - 270 °, the RF leakage from the VOA current cancels out the PD outputs and the 180° phase shift yields the maximum cancellation. We believe that cancellation of optical signals by the phase shift towards 180° is the main cause of the bandwidth degradation in Fig. 5(b). On the other hand, the noise component significantly decreases for the out-of-line VOA-PD devices by ~10 dB [see Fig. 6(d)], although the phase rotation is very similar to the in-line case [see Fig. 6(e)]. This strongly suggests that putting an isolation groove between the VOA and PD would suppress the high-frequency VOA-PD interference, which would result in a clearer synchronous RF response.

4. Discussion

Si defect PDs can be compared to Ge PDs on Si waveguides in terms of a figure of merit (FOM) defined as dark current divided by responsivity. This value represents the minimum detectable optical power in DC illumination condition. Although the quantum efficiency of the Si-defect PD is smaller than that of Ge PDs, the dark current of the Si-defect PD is smaller than that of Ge PDs because defects due to 4% lattice mismatch between Si and Ge act as leakage-current generation centers. Specifically, the FOM of the Si-defect PD in this work is −50 dBm, whereas that of a Ge PD on Si waveguide is typically larger than −40 dBm [10]. This trait of the Si PD is suitable for applications where small optical powers must be detected.

Our in-line integration of the VOA and PD with a short distance (50 µm) caused high-frequency interference in RF synchronous operation as well as a dark-current increase for the PD at DC attenuation by current injection to the VOA. The dark-current increase is most likely due to DC leakage through the lightly p-doped Si slab between p + contact regions of the VOA and PD separated by a distance of with 50 µm, whereas there must be a junction barrier at the lightly p-doped slab between the n + contact regions. Such a dark-current increase reduces the dynamic range of the PD for DC synchronous operation. Furthermore, high-frequency interference due to capacitive coupling between the VOA and PD causes a narrowing of the synchronous bandwidth. Therefore, these problems should be resolved to ensure better synchronous performances in both DC and RF conditions. A solution would be to change the layout of the devices, for instance, by placing two devices far from each other or placing an isolation groove between the devices. We found that an isolation groove is effective for reducing high-frequency noise and also expect that it would eliminate the DC leakage as well.

All-Si and in-line PDs can be implemented in various feedback systems for monolithically integrated carrier-injection-based photonic devices, such as modulators or VOAs. Among various feedback applications, we emphasize the feasibility of signal equalization by the in-line-integrated VOA and PD for wavelength-division multiplexing (WDM) networks. For this purpose, underlying requirements of the PD are sufficient absorption coverage for a large wavelength range, reliable power detection for optical power attenuated by the VOA, and a broad bandwidth exceeding that of the VOA. We showed the full C-band coverage by the PD in Fig. 2(c) and the DC synchronous operation between the VOA and PD with a dynamic range of over 40 dB in Fig. 4 . Further, the synchronous 3-dB bandwidth must be ~100 MHz if the RF noises are to be reduced. In a 10-Gbps-WDM system, 10- to 20-dB of attenuation range and 100 MHz of bandwidth are required for a gain tilt compensation and signal equalization in burst mode operation [13,15]. Therefore, both the DC and RF requirements for the WDM application would be satisfied by our VOA-PD devices.

 figure: Fig. 4

Fig. 4 Photocurrent and increase of dark current at the PD (left) and attenuation (right) at various injection currents to the VOA. Photocurrent was measured with the Si PD under reverse bias of 5 V.

Download Full Size | PDF

5. Conclusion

We demonstrated all-Si and in-line integration of VOA and PD with a p-i-n diode structure fabricated on a submicrometer Si rib waveguide. Argon implantation, followed by post-annealing at 300 °C for 30 min make it possible for the Si PD to absorb sub-band-gap NIR light. The propagation loss of the PD is ~4 dB/cm, ~2 dB/cm (0.5 cm−1) of which corresponds to the absorption coefficient of the PD. Such a low propagation loss and absorption coefficient enables the PD to act as an in-line power monitor. The PD has −50 dBm of minimum detectable optical power, a dynamic range for absorption of over 40 dB, and a 3-dB bandwidth of over 350 MHz. We observed reliable synchronous power monitoring for CW light by the PD in accordance with optical power attenuated by the VOA in spite of a dark current increase with increasing injection current. Furthermore, the synchronous 3-dB bandwidth is narrowed by the RF noise originating from high-frequency interference between the VOA and PD. Those degradations in performance in DC and RF synchronous operation can be overcome by changing the layout of the devices, particularly by placing an isolation groove between them. Such all-Si and in-line integration of a PD and carrier-injection based optical modulation devices would be promising for a feedback control system.

References and links

1. D. Ahn, C.-Y. Hong, J. Liu, W. Giziewicz, M. Beals, L. C. Kimerling, J. Michel, J. Chen, and F. X. Kärtner, “High performance, waveguide integrated Ge photodetectors,” Opt. Express 15(7), 3916–3921 (2007). [CrossRef]   [PubMed]  

2. L. Vivien, M. Rouvière, J.-M. Fédéli, D. Marris-Morini, J. F. Damlencourt, J. Mangeney, P. Crozat, L. El Melhaoui, E. Cassan, X. Le Roux, D. Pascal, and S. Laval, “High speed and high responsivity germanium photodetector integrated in a silicon-on-insulator microwaveguide,” Opt. Express 15(15), 9843–9848 (2007). [CrossRef]   [PubMed]  

3. M. W. Geis, S. J. Spector, M. E. Grein, R. T. Schulein, J. U. Yoon, D. M. Lennon, S. Deneault, F. Gan, F. X. Käertner, and T. M. Lyszczarz, “CMOS-compatible all-Si high-speed waveguide photodiodes with high responsivity in near-infrared communication band,” IEEE Photon. Technol. Lett. 19(3), 152–154 (2007). [CrossRef]  

4. M. W. Geis, S. J. Spector, M. E. Grein, R. J. Schulein, J. U. Yoon, D. M. Lennon, C. M. Wynn, S. T. Palmacci, F. Gan, F. X. Käertner, and T. M. Lyszczarz, “All silicon infrared photodiodes: photo response and effects of processing temperature,” Opt. Express 15(25), 16886–16895 (2007). [CrossRef]   [PubMed]  

5. M. W. Geis, S. J. Spector, M. E. Grein, J. U. Yoon, D. M. Lennon, and T. M. Lyszczarz, “Silicon waveguide infrared photodiodes with >35 GHz bandwidth and phototransistors with 50 AW-1 response,” Opt. Express 17(7), 5193–5204 (2009). [CrossRef]   [PubMed]  

6. Y. Liu, C. W. Chow, W. Y. Cheung, and H. K. Tsang, “In-line channel power monitor based on helium ion implantation in silicon-on-insulator waveguides,” IEEE Photon. Technol. Lett. 18(17), 1882–1884 (2006). [CrossRef]  

7. H. Y. Fan and A. K. Ramdas, “Infrared absorption and photoconductivity in irradiated silicon,” J. Appl. Phys. 30(8), 1127–1134 (1959). [CrossRef]  

8. G. Cocorullo and I. Rendina, “Thermo-optical modulator at 1.5 µm in silicon etalon,” Electron. Lett. 28(1), 83 (1992). [CrossRef]  

9. R. Soref and B. Bennett, “Electrooptical effects in silicon,” IEEE J. Quantum Electron. 23(1), 123–129 (1987). [CrossRef]  

10. T. Tsuchizawa, K. Yamada, T. Watanabe, H. Shinojima, H. Nishi, S. Itabashi, S. Park, Y. Ishikawa, and K. Wada, “Monolithic integration of germanium photodetectors and silicon wire waveguides with carrier injection structures,” in Conference on Lasers and Electro-Optics/International Quantum Electronics Conference, OSA Technical Digest (CD) (Optical Society of America, 2009), paper CTuV2.

11. A. P. Knights, J. D. B. Bradley, S. H. Gou, and P. E. Jessop, “Monolithically integrated photodetectors for optical signal monitoring in silicon waveguides,” Proc. SPIE 6125, 61250J (2006). [CrossRef]  

12. P. E. Jessop, L. K. Rowe, S. M. McFaul, A. P. Knights, N. G. Tarr, and A. Tam, “Study of the monolithic integration of sub-bandgap detection, signal amplification and optical attenuation on a silicon photonic chip,” J. Mater. Sci. Mater. Electron. 20(S1), 456–459 (2009). [CrossRef]  

13. Y. Liu, C. W. Chow, C. H. Kwok, H. K. Tsang, and C. Lin, “Optical burst and transient equalizer for 10Gb/s amplified WDM-PON,” in Proceedings of Optical Fiber Communication and the National Fiber Optic Engineers Conference, (Academic, 2007), OThU7.

14. T. Tsuchizawa, K. Yamada, H. Fukuda, T. Watanabe, J. Takahashi, M. Takahashi, T. Shoji, E. Tamechika, S. Itabashi, and H. Morita, “Microphotonics devices based on silicon microfabrication technology,” IEEE J. Sel. Top. Quantum Electron. 11(1), 232–240 (2005). [CrossRef]  

15. S. Nishihara, M. Nakamura, K. Nishimura, K. Kishine, S. Kimura, and K. Kato, “10.3 Gbit/s burst-mode PIN-TIA module with high sensitivity, wide dynamic range and quick response,” Electron. Lett. 44(3), 222 (2008). [CrossRef]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (6)

Fig. 1
Fig. 1 (a) Cross-section of a p-i-n diode for the VOA and PD on a submicrometer rib waveguide. (b) Optical microscope image of the in-line VOA and PD with 2.5-mm length.
Fig. 2
Fig. 2 (a) I-V curves of the PD under dark and illumination, (b) photocurrent (left) and responsivity (right) versus optical power at the PD, and (c) responsivity for the C-band wavelength. The PD length is 2.5 mm and a reverse bias of 5 V is applied. (d) Responsivity as a function of the PD length at −5 V. (e) Transmission against the PD length.
Fig. 3
Fig. 3 Block diagram of the frequency-response measurement setups for (a) the VOA and (b) PD. (c) Frequency response of the VOA at injection current of 5 and 10 mA, and that of the PD under reverse bias of 5 and 15 V.
Fig. 5
Fig. 5 (a) Block diagram of the setup for measuring the synchronous frequency response between the VOA and PD. The frequency response of the synchronous VOA-PD at injection current of 10 mA to the VOA are shown in (b) and (c), respectively, for 1) the direct electrical probing by the PD and 2) electrical conversion at the O-E converter with reverse bias applied to the PD.
Fig. 6
Fig. 6 (a) Optical microscope image of frequency response measured in a dark condition between the VOA and PD. In-line and out-of-line refer to a VOA-PD pair placed along the same waveguide (VOA-PD1) and separated along the isolation groove (VOA-PD2), respectively. Magnitudes [(b),(d)] and phases [(c),(e)] as a function of frequency at the reverse bias of 5 and 15 V of the PD and at 10-mA injection current to the VOA after calibration with an impedance standard substrate. (b) and (c) for the in-line VOA-PD1, while (d) and (e) for the out-of-line VOA-PD2.
Fig. 4
Fig. 4 Photocurrent and increase of dark current at the PD (left) and attenuation (right) at various injection currents to the VOA. Photocurrent was measured with the Si PD under reverse bias of 5 V.
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.