Open Access
Add to BibSonomy
Abstract
We report the demonstration of a normal-incidence p-i-n germanium-tin () photodetector on silicon-on-insulator substrate for 2 μm wavelength application. The DC and RF characteristics of the devices have been characterized. A dark current density under bias of approximately is achieved at room temperature, and the optical responsivity of 14 mA/W is realized for illumination wavelength of 2 μm under reverse bias. In addition, a 3 dB bandwidth () of around 30 GHz is achieved at , which is the highest reported value among all group III–V and group IV photodetectors working in the 2 μm wavelength range. This work illustrates that a GeSn photodetector has great prospects in 2 μm wavelength optical communication.
© 2021 Chinese Laser Press
1. INTRODUCTION
At present, the industry standard wavelength range for telecommunication lies between 1.3 and 1.6 μm. However, the fiber-optic telecommunication system is gradually approaching its capacity limit since the exponential growth of internet data transmission, and the phenomenon of “capacity crunch” of the optical communication system may happen in the future [1,2]. This crisis can be temporarily alleviated by compressing the transmission data or adopting multiple parallel links communication. But the recent studies in hollow-core photonic bandgap fibers (HC-PBGFs) provide a more elegant solution; HC-PBGFs have demonstrated that the loss window can be extended from 1.55 to 2 μm, and the theoretical minimum loss is below 0.1 dB/km [3,4]. This loss is lower than the best conventional single-mode fiber (SMF) (0.1484 dB/km), and the proposed HC-PBGF–based communication system is compatible with silicon photonics, as the loss of silicon dioxide is low at 2 μm, and some strip and ridge waveguides with low transmission loss on silicon-on-insulator (SOI) platforms have been reported [5]. In addition, some optical components that can be used for 2 μm waveband optical communication have also made significant progress. For example, the optical gain window of thulium-doped fiber amplifiers (TDFAs) resides at around 1810–2050 nm, and it can be used as the equivalent to erbium-doped fiber amplifiers (EDFAs) in a 2 μm communication system [6,7]. The eight-channel wavelength division multiplexing system has a data transmission rate of 100 Gb/sat 2 μm, which consists of four internal direct modulation channels and four externally modulated channels [8]. Also, on the germanium-on-silicon (Ge-on-Si) platform, the all-optical modulator based on free carrier absorption working at around 55 MHz across the wavelength range of 2–3.2 μm [9], and a silicon-based high-speed Mach–Zehnder interferometer (MZI) modulator with a data transmission rate of 20 Gb/s at 2 μm have also been demonstrated [10]. Significant progress has also been made in 2 μm optical attenuators (VOAs) based on free carrier injection on the SOI platform and the Ge-on-Si platform [11]. These studies indicate that the 2 μm waveband has great potential as a candidate spectral window for future telecommunication systems.
The high-speed photodetector as an important optical component is also indispensable in optical communication systems. Because the bandgap of germanium-tin () alloys is adjustable and compatible with the existing complementary metal-oxide–semiconductor (CMOS) platform, some high-speed photodetectors based on GeSn binary alloys for 1.55 and 2 μm have been reported [12–15]. However, due to the large parasitic parameters, the bandwidth of the photodetectors working in the 2 μm band only achieves or less. Moreover, the differences in lattice constants between the Ge element and the Sn element and between Si and Sn are 14.7% and 19.5%, respectively. As the composition of Sn in the GeSn alloy increases, the compressive strain of the GeSn alloy will gradually increase. When the stress is unbearable, the strain will eventually be released in the form of dislocations. It will cause deterioration of material quality, and the electrical performance and high-speed performance of the device will also be seriously affected.
In this work, high-speed photodetectors on SOI substrates were designed and fabricated. The high-quality GeSn binary alloy was created using solid-source molecular beam epitaxy (MBE). A room temperature dark current density of is achieved when biased at . The parasitic capacitance and junction capacitance of the device are also extracted by the C–V curves. A remarkable optical responsivity of 14 mA/W is achieved at 2 μm. In addition, the frequency response characteristics of these photodetectors are studied. A 3 dB bandwidth up to 30 GHz is achieved at reverse bias, which is among the highest values reported in the literature for all group III–V and group IV photodetectors working in the 2 μm wavelength range. These results pave the way toward more promising feasibility of a new spectral window at the 2 μm wavelength.
2. MATERIAL GROWTH AND CHARACTERISTICS
The materials were grown using solid-source MBE on an SOI substrate, which contains a 220 nm thick Si top layer and a 2 μm thick buried oxide layer (BOX). Before the materials were created, the superficial Si layer went through phosphorus diffusion to form the -type contact layer. Afterward, a standard Radio Corporation of America (RCA) wet-chemical cleaning recipe was used to clean the substrate. Then, the wafer was loaded into the chamber for degassing at 300°C. Subsequently, the substrate was deoxidized at 850°C for 10 min. The complete layer sequence of nominally designed epitaxial layer includes a 200 nm thick Ge-buffer layer grown using a two-step growth process, composed of a 70 nm thick low temperature (LT-Ge) deposited at 300°C and a layer of 130 nm thick high temperature (HT-Ge) grown at 600°C. In order to reduce the lattice mismatch dislocations, the Ge-buffer underwent cyclic annealing from 600°C to 750°C after growth. A 350 nm thick GeSn alloy layer was deposited at 200°C, and a 100 nm thick -type boron-doped GeSn layer was grown at the same temperature to avoid segregation of Sn atoms in the GeSn layer.
The cross-sectional transmission electron microscopy (TEM) image of the materials is shown in Fig. 1(a), and distinct interfaces of the Ge/Si and GeSn/Ge heterojunction are observed. The inset above in Fig. 1(a) is the selected-area diffraction pattern of the GeSn layer, which indicates this layer has a single-crystalline nature with a diamond cubic structure. The red square areas of the Ge/Si and GeSn/Ge interface are zoomed into the HR-TEM image. In Fig. 1(b), the HR-TEM image, a smooth and abrupt interface between GeSn and Ge-buffer, is observed. No threading dislocations can be seen, which shows that the GeSn layer has high quality. In Fig. 1(c), although the interface has a large number of misfit dislocations, it is restricted to the LT-Ge layer. In addition, it can clearly be seen that there are almost no threading dislocations extending upward, which ensures the high quality of the upper GeSn layer. The secondary ion mass spectrometry (SIMS) results in Fig. 1(d) show the distribution and concentration of various elements in the as-grown sample. It can be seen that the composition of Sn in the GeSn layer is about , and the higher doping concentration of boron and phosphorus is beneficial to form ohmic contacts. To verify the concentration of Sn in the GeSn layer, the X-ray diffraction reciprocal space map (XRD-RSM) around the asymmetric () reflection was obtained, as shown in Fig. 1(e). It shows two distinct areas, the upper region can be assigned to the Si substrate, and the bottom region corresponds to the Ge-buffer and GeSn film. It can be seen that there are two diffraction peaks of Si, which correspond to top-Si and bottom-Si, respectively. Because the SOI substrate is obtained by a bonding process, it is difficult to completely match the crystal orientation of top-Si and bottom-Si. Moreover, since the thickness of the top-Si layer is much smaller than that of bottom-Si layer, the weaker diffraction peak originates from the top-Si layer and the stronger one from the Si handle substrate. The RSM reveals that the Ge-buffer contains a tensile strain of roughly 0.09%, which is due to cyclic annealing and the difference between the thermal expansion coefficient of the Ge and Si substrates [16,17]. The diffraction peaks of the GeSn layer are positioned well on the pseudomorphic line; it means that the GeSn layer has the same in-plane lattice constant as the Ge-buffer. The shoulder peak of the GeSn layer corresponds to -GeSn, which is because the doping of boron reduces the average lattice constant of the crystal [18]. Also, according to the Bragg formula, the composition of the GeSn binary alloy can be calculated as , which is in good agreement with the result of the SIMS. Material characterization results of this material are summarized in Table 1.
Fig. 1. (a) Cross-sectional transmission electron microscopy (TEM) image of the epitaxial material grown on the SOI substrate; the inset above is the selected-area diffraction pattern of the GeSn layer. (b) High-resolution TEM (HR-TEM) image of interface between epitaxial GeSn and Ge-buffer. (c) HR-TEM image of interface between epitaxial Ge-buffer and top-Si substrate. (d) The SIMS depth profile analysis of various elements in the as-grown sample. (e) X-ray diffraction reciprocal space map (XRD-RSM) around the asymmetric () reflection of the material.
Table 1. Summary of Lattice Constant, In-Plane Strain and Sn Concentration
3. DEVICE FABRICATION
Normally illuminated p-i-n GeSn photodetectors were realized with a -type GeSn layer, an absorption layer (including Ge-buffer layer and GeSn layer), and an -type contact layer. The inductively coupled plasma (ICP) dry-etching process was adopted to form a double-mesa structure. Subsequently, a 500 nm thick passivation was deposited by plasma-enhanced chemical vapor deposition (PECVD) on the surface and sidewalls to reduce the surface leakage current. Ni\Al\Ti\Au for the pad contact was deposited by e-beam evaporation and patterned by lift-off. Then, a 380 nm thick layer was deposited by PECVD to reduce the surface reflection and protect the metal electrodes. The three-dimensional (3D) schematic of the normally illuminated p-i-n photodetector is shown in Fig. 2(a), and Fig. 2(b) shows the top-view scanning electron microscopy (SEM) image of this device with a 10 μm diameter mesa.
Fig. 2. (a) 3D structure schematic of the normally illuminated p-i-n photodetector. (b) Top-view SEM image of the device with a 10 μm diameter mesa.
4. ELECTRICAL CHARACTERISTICS
The typical current-voltage (I–V) and capacitance-voltage (C–V) characteristics of the devices were measured using an Agilent B1500A semiconductor parameter analyzer at room temperature. Figure 3 shows the I–V curves from to 1 V of the photodetectors with various diameters. The devices exhibit a well-defined rectifying behavior with a high on/off current ratio near between 1 and , which indicates the excellent quality of the epitaxial materials and p-i-n junction. Also, the ideal factor can be obtained by the current-voltage formula [19]: , where is reverse saturated current, is elementary charge, is the Boltzmann constant, and is the absolute temperature. The calculated value of 1.36 indicates that the current is diffusion-current dominant rather than a carrier recombination current and also implies the low threading-dislocation density in the layer [20,21]. In addition, the series resistances of the devices can also be extracted from the I–V curves in Fig. 3. The series resistance of the devices with diameters of 30–10 μm is as low as , indicating that superior-quality ohmic contact is formed. It is generally known that dark current is one of the most critical characteristics of high-performance photodetectors. Low dark current can reduce the noise of photodetectors, which is very important for the high signal-to-noise ratio (SNR) for optical receivers. The dark current of the photodetectors with a diameter of 10 μm is 230 nA at , and this value is tolerable for high-speed optical receivers [22]. The total dark current density () can be divided into the bulk leakage current density () and surface leakage current density () using the following equation:
where is the diameter of the devices, and the relationship between the total dark current of the multiple devices and is shown in the inset in Fig. 3. The deviation of some data points is mainly due to the unevenness of the device manufacturing process. The and extracted from the linear fitting are and 52.7 μA/cm, respectively. Since is proportional to the threading-dislocation density, the low indicates the high quality of the GeSn sample [23], and it is also a low value among the GeSn photodetectors with a similar Sn concentration [24–26].Fig. 3. Typical I-V characteristics of the photodetectors (, 12, 15, 18, 20, 25, and 30 μm); the inset is the dark current densities of devices at versus .
Figure 4(a) shows the C–V curves of the photodetectors at bias voltage from 0 to . It can be seen that the device gradually becomes depleted as the bias voltage increases, and the devices are fully depleted at . At bias voltages of , the capacitances of the photodetectors with diameters of 10, 12, 15, 18, 20, 25, and 30 μm are 48, 65, 94, 123, 149, 230, and 312 fF, respectively. In addition, the capacitance () can be divided into the junction capacitance () and parasitic capacitance () using the following equation:
where , is the area of the devices, , is the thickness of the intrinsic layers, and and are the dielectric constants. The total capacitances of different photodetectors at versus are shown in Fig. 4(b). The extracted parasitic capacitance and junction capacitance per unit area are 17.5 fF and , respectively.Fig. 4. (a) C-V characteristics of the photodetectors with various diameters. (b) The capacitances of different devices at versus area.
5. PHOTOCURRENT AND SPECTRUM RESPONSIVITY
The photocurrents of the devices were measured using an Agilent B1500A semiconductor parameter analyzer, a probe station, and a 2000 nm laser at room temperature. The incident light was introduced by a single-mode tapered lensed fiber to the top surface of the photodetectors, and the output optical power was measured to be 1 mW by a calibrated commercial reference detector. The I–V characteristics of a 10 μm diameter device with/without light incidence are shown in Fig. 5(a). The optical responsivity of 14 mA/W is achieved for an illumination wavelength of 2000 nm at . In addition, the optical responsivity could be further improved by increasing the Sn composition in GeSn alloy or photon-trapping microstructures [13,27–29].
Fig. 5. (a) I-V characteristics of the GeSn photodetector with a diameter of 10 μm with/without light incidence. (b) Spectrum response and optical responsivity of the photodetector as functions of wavelength under zero-bias. Devices were measured by an FTIR optical spectrometer and lasers (two tunable lasers, one is located between 1260 nm and 1360 nm, and the other is located between 1500 nm and 1630 nm, and a laser at 2000 nm). The data are shown as a blue curve and a scatter plot, respectively.
The spectral response of the photodetector was measured using a Nicolet 6700 Fourier transform infrared (FTIR) spectrometer, and a commercial InGaAs photodetector was used to calibrate spectrum responsivity. Figure 5(b) shows the spectral response of the photodetector under zero-bias. The scatter plot, which was measured by a 1260–1360 nm tunable laser, a 1500–1630 nm tunable laser, and a 2 μm laser at 0 V reverse-bias voltage, corresponds well to the spectrum response indicated by the blue curve. A resonance responsivity peak around is clearly observed, and the responsivities are 0.25 and 0.31 A/W at 1550 and 1625 nm wavelengths, respectively. Obviously, compared with Ge-on-Si photodetectors, the device in this work can significantly extend the detection wavelength. The optical response covers the O, E, S, C, L, and U telecommunication bands completely, and a new 2 μm communication window can also be implied.
6. FREQUENCY RESPONSES
The frequency response at 2 μm of the photodetector was measured using an optical heterodyne beat frequency measurement system [30], as shown in Fig. 6. The two lasers are used to emit two beams of similar frequencies, which are coupled to produce high-speed signals. Then, the power is amplified by the TDFA, 95% of the light is coupled into the photodetector by the beam splitter, the microwave signal is generated by the photoelectric conversion photodetector, and the microwave signal is input to the RF power meter (test range is between 0 and 30 GHz) through the microwave probe and bias-T, so as to characterize the frequency response performance of the detector. The other 5% of the light generated by the beam splitter is input to a reference photodetector (PD) to detect the generated signal. The DC path of bias-T is connected to an ammeter and a bias voltage source. The ammeter is used to test the current of the photodetector, and the bias voltage source is used to add bias voltage to the photodetector.
Figure 7(a) shows the normalized frequency responses of GeSn photodetectors with various diameters at 2 μm, and the noise of the signal is mainly due to the low responsivity of the device at 2 μm. The 3 dB bandwidth remarkably increases with a decrease in diameter of the photodetector. At bias, the 3 dB bandwidths of the devices with diameters of 10, 15, 20, 25, and 30 μm are 30, 28, 17, 11, and 8 GHz, respectively.
Fig. 7. (a) Normalized frequency responses of the photodetectors with various diameters at 2 μm (, 15, 20, 25, and 30 μm). (b) The theory RC-limited bandwidth, transit-time-limited bandwidth, and combined bandwidth of devices with different diameters.
The bandwidth of photodetector is mainly dominated by the carrier transit-time-limited bandwidth () and resistor capacitor (RC)-limited bandwidth () in the active region. and can be written as [31]
where is the saturation drift velocity, and is the thickness of the intrinsic layer. is the load resistance ( in this case). , , and are the series resistance, parasitic capacitance, and junction capacitance, respectively, which can be extracted from the I-V curves and C-V curves in Figs. 3 and 4. The theoretical RC-limited bandwidth, transit-time-limited bandwidth, and combined bandwidth of the devices are shown in Fig. 7(b). The experimental bandwidths of the photodetectors with various diameters at are also shown in Fig. 7(b) for comparison. The experimental values are consistent with the theoretical values. It can be seen that the bandwidth of the detector is mainly limited by the RC constant, and if the diameter of the device continues to decrease, the bandwidth can be further improved.In order to understand the electrical characterization of the device more intuitively, the dark current density, capacitance, and 3 dB bandwidth of the devices with different diameters are summarized, as shown in Table 2. In addition, it is worth emphasizing that high-speed photodetectors for 2 μm wavelength optical communications have been reported in recent years, including group III–V and group IV systems. The 3 dB bandwidth achieved in this work is the highest value among the reported high-speed photodetectors operating at 2 μm, as shown in Fig. 8 [14,15,30,32–35].
Table 2. Summary of Dark Current Density, Capacitance, and 3 dB Bandwidth of Devices with Different Diameters
Fig. 8. Comparison of 3 dB bandwidth of high-speed photodetectors for 2 μm-wavelength light detection in different groups.
7. CONCLUSIONS
In summary, a high-speed silicon-based photodetector is fabricated on SOI substrates for 2 μm wavelength light detection. A low dark current density of is achieved at . The parasitic capacitance and junction capacitance are extracted as 17.5 fF and , respectively. At bias voltages of , the optical responsivity of the photodetector is 14 mA/W at 2 μm. In addition, this GeSn photodetector achieved a 3 dB bandwidth as high as 30 GHz, which is among the highest values for 2 μm wavelength optical communications. And, the experimental values of the 3 dB bandwidth under different diameters are consistent with simulated results. The excellent high-speed performance of this device proves that Si-based GeSn photodetectors have great potential in the new 2 μm communication band, which can effectively increase the communication capacity in the future.
Funding
National Natural Science Foundation of China (61774143, 61874109, 61975121, 61975196); National Key Research and Development Program of China (2018YFB2200501, 2019YFB2203400).
Disclosures
The authors declare no conflicts of interest.
REFERENCES
1. D. J. Richardson, “Filling the light pipe,” Science 330, 327–328 (2010). [CrossRef]
2. A. D. Ellis, Z. Jian, and D. Cotter, “Approaching the non-linear Shannon limit,” J. Lightwave Technol. 28, 423–433 (2010). [CrossRef]
3. P. Roberts, F. Couny, H. Sabert, B. Mangan, D. Williams, L. Farr, M. Mason, A. Tomlinson, T. Birks, J. Knight, and P. St. J. Russell, “Ultimate low loss of hollow-core photonic crystal fibres,” Opt. Express 13, 236–244 (2005). [CrossRef]
4. E. Desurvire, C. Kazmierski, F. Lelarge, X. Marcadet, A. Scavennec, F. A. Kish, D. F. Welch, R. Nagarajan, C. H. Joyner, R. P. Schneider, S. W. Corzine, M. Kato, P. W. Evans, M. Ziari, A. G. Dentai, J. L. Pleumeekers, R. Muthiah, S. Bigo, M. Nakazawa, D. J. Richardson, F. Poletti, M. N. Petrovich, S. U. Alam, W. H. Loh, and D. N. Payne, “Science and technology challenges in XXIst century optical communications,” C. R. Phys. 12, 387–416 (2011). [CrossRef]
5. D. E. Hagan and A. P. Knights, “Mechanisms for optical loss in SOI waveguides for mid-infrared wavelengths around 2 μm,” J. Opt. 19, 025801 (2017). [CrossRef]
6. Z. Li, A. M. Heidt, N. Simakov, Y. Jung, J. M. Daniel, S. U. Alam, and D. J. Richardson, “Diode-pumped wideband thulium-doped fiber amplifiers for optical communications in the 1800–2050 nm window,” Opt. Express 21, 26450–26455 (2013). [CrossRef]
7. Z. Li, A. M. Heidt, J. M. Daniel, Y. Jung, S. U. Alam, and D. J. Richardson, “Thulium-doped fiber amplifier for optical communications at 2 µm,” Opt. Express 21, 9289–9297 (2013). [CrossRef]
8. N. Ye, M. R. Gleeson, M. U. Sadiq, B. Roycroft, C. Robert, H. Yang, H. Zhang, P. E. Morrissey, N. Mac Suibhne, K. Thomas, A. Gocalinska, E. Pelucchi, R. Phelan, B. Kelly, J. Ocarroll, F. H. Peters, F. C. Garcia Gunning, and B. Corbett, “InP-based active and passive components for communication systems at 2 μm,” J. Lightwave Technol. 33, 971–975 (2015). [CrossRef]
9. L. Shen, N. Healy, C. J. Mitchell, J. S. Penades, M. Nedeljkovic, G. Z. Mashanovich, and A. C. Peacock, “Mid-infrared all-optical modulation in low-loss germanium-on-silicon waveguides,” Opt. Lett. 40, 268–271 (2015). [CrossRef]
10. W. Cao, D. Hagan, D. J. Thomson, M. Nedeljkovic, C. G. Littlejohns, A. Knights, S.-U. Alam, J. Wang, F. Gardes, W. Zhang, S. Liu, K. Li, M. S. Rouifed, G. Xin, W. Wang, H. Wang, G. T. Reed, and G. Z. Mashanovich, “High-speed silicon modulators for the 2 μm wavelength band,” Optica 5, 1055–1062 (2018). [CrossRef]
11. J. Kang, M. Takenaka, and S. Takagi, “Novel Ge waveguide platform on Ge-on-insulator wafer for mid-infrared photonic integrated circuits,” Opt. Express 24, 11855–11864 (2016). [CrossRef]
12. M. Oehme, K. Kostecki, K. Ye, S. Bechler, K. Ulbricht, M. Schmid, M. Kaschel, M. Gollhofer, R. Korner, W. Zhang, E. Kasper, and J. Schulze, “GeSn-on-Si normal incidence photodetectors with bandwidths more than 40 GHz,” Opt. Express 22, 839–846 (2014). [CrossRef]
13. H. Zhou, S. Xu, Y. Lin, Y. C. Huang, B. Son, Q. Chen, X. Guo, K. H. Lee, S. C. Goh, X. Gong, and C. S. Tan, “High-efficiency GeSn/Ge multiple-quantum-well photodetectors with photon-trapping microstructures operating at 2 µm,” Opt. Express 28, 10280–10293 (2020). [CrossRef]
14. Y. Dong, W. Wang, S. Xu, D. Lei, X. Gong, X. Guo, H. Wang, S. Y. Lee, W. K. Loke, S. F. Yoon, and Y. C. Yeo, “Two-micron-wavelength germanium-tin photodiodes with low dark current and gigahertz bandwidth,” Opt. Express 25, 15818–15827 (2017). [CrossRef]
15. S. Xu, W. Wang, Y. C. Huang, Y. Dong, S. Masudy-Panah, H. Wang, X. Gong, and Y. C. Yeo, “High-speed photo detection at two-micron-wavelength: technology enablement by GeSn/Ge multiple-quantum-well photodiode on 300 mm Si substrate,” Opt. Express 27, 5798–5813 (2019). [CrossRef]
16. L. Peng, X. Li, Z. Liu, X. Liu, J. Zheng, C. Xue, Y. Zuo, and B. Cheng, “Horizontal GeSn/Ge multi-quantum-well ridge waveguide LEDs on silicon substrates,” Photon. Res. 8, 899–903 (2020). [CrossRef]
17. Z. Liu, B.-W. Cheng, Y.-M. Li, C.-B. Li, C.-L. Xue, and Q.-M. Wang, “Effects of high temperature rapid thermal annealing on Ge films grown on Si(001) substrate,” Chin. Phys. B 22, 116804 (2013). [CrossRef]
18. D. Schwarz, H. S. Funk, M. Oehme, and J. Schulze, “Alloy stability of Ge1−xSnx with Sn concentrations up to 17% utilizing low-temperature molecular beam epitaxy,” J. Electron. Mater. 49, 5154–5160 (2020). [CrossRef]
19. C. Popescu, J. C. Manifacier, and R. Ardebili, “I-V curve shape factor for thin p-n junctions at high injection levels,” Phys. Status Solidi A 158, 611–621 (1996). [CrossRef]
20. C. Sah, R. N. Noyce, and W. Shockley, “Carrier generation and recombination in P-N junctions and P-N junction characteristics,” Proc. IRE 45, 1228–1243 (1957). [CrossRef]
21. Z. Liu, F. Yang, W. Wu, H. Cong, J. Zheng, C. Li, C. Xue, B. Cheng, and Q. Wang, “48 GHz high-performance Ge-on-SOI photodetector with zero-bias 40 Gbps grown by selective epitaxial growth,” J. Lightwave Technol. 35, 5306–5310 (2017). [CrossRef]
22. D. Ahn, C. Y. Hong, J. Liu, W. Giziewicz, M. Beals, L. C. Kimerling, J. Michel, J. Chen, and F. X. Kartner, “High performance, waveguide integrated Ge photodetectors,” Opt. Express 15, 3916–3921 (2007). [CrossRef]
23. Y. Zhao, N. Wang, K. Yu, X. Zhang, X. Li, J. Zheng, C. Xue, B. Cheng, and C. Li, “High performance silicon-based GeSn p–i–n photodetectors for short-wave infrared application,” Chin. Phys. B 28, 128501 (2019). [CrossRef]
24. K. Ye, W. Zhang, M. Oehme, M. Schmid, M. Gollhofer, K. Kostecki, D. Widmann, R. Körner, E. Kasper, and J. Schulze, “Absorption coefficients of GeSn extracted from PIN photodetector response,” Solid-State Electron. 110, 71–75 (2015). [CrossRef]
25. C.-H. Tsai, B.-J. Huang, R. A. Soref, G. Sun, H. H. Cheng, and G.-E. Chang, “GeSn resonant-cavity-enhanced photodetectors for efficient photodetection at the 2 μm wavelength band,” Opt. Lett. 45, 1463–1466 (2020). [CrossRef]
26. W.-T. Hung, D. Barshilia, R. Basu, H. H. Cheng, and G.-E. Chang, “Silicon-based high-responsivity GeSn short-wave infrared heterojunction phototransistors with a floating base,” Opt. Lett. 45, 1088–1091 (2020). [CrossRef]
27. H. Cong, C. Xue, J. Zheng, F. Yang, K. Yu, Z. Liu, X. Zhang, B. Cheng, and Q. Wang, “Silicon based GeSn p-i-n photodetector for SWIR detection,” IEEE Photon. J. 8, 6804706 (2016). [CrossRef]
28. H. Cansizoglu, C. Bartolo-Perez, Y. Gao, E. Ponizovskaya Devine, S. Ghandiparsi, K. G. Polat, H. H. Mamtaz, T. Yamada, A. F. Elrefaie, S.-Y. Wang, and M. S. Islam, “Surface-illuminated photon-trapping high-speed Ge-on-Si photodiodes with improved efficiency up to 1700 nm,” Photon. Res. 6, 734–742 (2018). [CrossRef]
29. Y. Gao, H. Cansizoglu, K. G. Polat, S. Ghandiparsi, A. Kaya, H. H. Mamtaz, A. S. Mayet, Y. Wang, X. Zhang, T. Yamada, E. P. Devine, A. F. Elrefaie, S.-Y. Wang, and M. S. Islam, “Photon-trapping microstructures enable high-speed high-efficiency silicon photodiodes,” Nat. Photonics 11, 301–308 (2017). [CrossRef]
30. Y. Chen, Z. Xie, J. Huang, Z. Deng, and B. Chen, “High-speed uni-traveling carrier photodiode for 2 μm wavelength application,” Optica 6, 884–889 (2019). [CrossRef]
31. M. Jutzi, M. Berroth, G. Wohl, M. Oehme, and E. Kasper, “Ge-on-Si vertical incidence photodiodes with 39-GHz bandwidth,” IEEE Photon. Technol. Lett. 17, 1510–1512 (2005). [CrossRef]
32. A. Joshi and D. Becker, “High-speed low-noise p-i-n InGaAs photoreceiver at 2-μm wavelength,” IEEE Photon. Technol. Lett. 20, 551–553 (2008). [CrossRef]
33. B. F. Andresen, A. Joshi, S. Datta, G. F. Fulop, and P. R. Norton, “High-speed, large-area, p-i-n InGaAs photodiode linear array at 2-micron wavelength,” Proc. SPIE 8353, 83533D (2012). [CrossRef]
34. Y. Chen, X. Zhao, J. Huang, Z. Deng, C. Cao, Q. Gong, and B. Chen, “Dynamic model and bandwidth characterization of InGaAs/GaAsSb type-II quantum wells PIN photodiodes,” Opt. Express 26, 35034–35045 (2018). [CrossRef]
35. B. Tossoun, J. Zang, S. J. Addamane, G. Balakrishnan, A. L. H. Holmes, and A. Beling, “InP-based waveguide-integrated photodiodes with InGaAs/GaAsSb type-II quantum wells and 10-GHz bandwidth at 2 μm wavelength,” J. Lightwave Technol. 36, 4981–4987 (2018). [CrossRef]
