Open Access
Add to BibSonomy
Abstract
We report the demonstration of a germanium-tin (Ge0.9Sn0.1) multiple-quantum-well p-i-n photodiode on silicon (Si) substrate for 2 μm-wavelength light detection. Characterization of the photodetector in both direct current (DC) and radio frequency (RF) regimes was performed. At the bias voltage of −1 V, a dark current density of 0.031 A/cm2 is realized at room-temperature, which is among the lowest reported values for Ge1−xSnx-on-Si p-i-n photodiodes. In addition, for the first time, a 3 dB bandwidth (f3dB) of around 1.2 GHz is achieved in Ge1−xSnx photodetectors operating at 2 μm. It is anticipated that further device optimization would extend the f3dB to above 10 GHz.
© 2017 Optical Society of America
1. Introduction
Two-micron-wavelength photodetectors are critical for a variety of emerging applications, including fiber-optic communication [1,2], biomedical sensing [3], and 3D light detection and ranging (LiDAR) for atmospheric remote sensing [4]. All these applications benefit significantly from having photodetectors with a low dark current (Idark) and a high response speed. Germanium-tin (Ge1−xSnx) is a versatile group IV alloy that is very promising for 2 μm light-detection due to its large absorption coefficient at this wavelength [5]. Photodetectors comprising Ge1−xSnx can be monolithically-integrated with other silicon (Si) photonic and electronic devices to realize compact, low-cost, and multi-functional opto-electronic integrated circuits (OEICs). Remarkable progress has been made in the development of Ge1−xSnx photodetectors on Si substrate over the last decade [6–25], and photodetectors with low-Idark have been achieved [20,23,25]. However, to the best of our knowledge, 2 μm-wavelength Ge1−xSnx photodetector with high response speed has not been reported yet.
In this work, we demonstrate a Ge0.9Sn0.1 multiple-quantum-well (MQW) p-i-n photodiode on Si substrate for 2 μm-wavelength applications, featuring a device with not only a low Idark but also a 3dB bandwidth (f3dB) in the gigahertz (GHz) range. A room-temperature dark current density (Jdark) of 0.031 A/cm2 is achieved when biased at −1 V, which is among the lowest values reported in the literature for Si-based Ge1−xSnx p-i-n diodes. In addition, f3dB of the photodetector is measured to be around 1.2 GHz and an f3dB of above 10 GHz is expected with further device optimization.
2. Device design and fabrication
A 3D schematic of the vertically-illuminated Ge0.9Sn0.1 MQW p-i-n photodiode is shown in Fig. 1(a). The MQW structure brings about two advantages as compared to a bulk-Ge0.9Sn0.1 layer. First, the Ge0.9Sn0.1 MQW has a larger critical thickness (for strain-relaxation) than the bulk one. Second, MQW structure is favorable for other photonic devices, e.g. electro-optical modulators based on the Quantum Confined Stark Effect (QCSE) [26]. Therefore, the implementation of the MQW structure allows monolithic integration of the photodetector with other photonic devices using a single MQW stack [27]. In this work, ten periods of Ge0.9Sn0.1 QWs were grown by molecular beam epitaxy (MBE) at 150 °C on a commercially available Ge-on-Si(100) virtual substrate, which comprised a 350 nm-thick Ge buffer, a 1 μm-thick n+-Si contact layer, and a high-resistance Si(100) substrate. This was followed by the epitaxy of a 50 nm-thick in situ Ga-doped p+-Ge layer. The doping concentrations of the n+-Si and p+-Ge layers are 2 × 1019 cm−3 and 5 × 1019 cm−3, respectively.
Fig. 1 (a) 3D schematic of the fabricated Ge0.9Sn0.1 MQW p-i-n photodiode that is vertically illuminated by a single-mode (SM) optical fiber. (b) Cross-sectional schematic of the Ge0.9Sn0.1 MQW. The thicknesses of the Ge0.9Sn0.1 well and the Ge barrier are 15 and 20 nm, respectively. There are 10 Ge0.9Sn0.1 quantum wells in this design.
In each period of the MQW, the thicknesses of Ge0.9Sn0.1 well (dwell) and Ge barrier (dbarrier) are 15 and 20 nm, respectively, as shown in Fig. 1(b). The energy bandgap of a 15 nm-thick Ge0.9Sn0.1 QW (sandwiched by Ge barriers) at Γ-valley (Eg,Γ) is calculated to be around 0.56 eV using the method described in Ref. 9. Therefore, the Ge0.9Sn0.1 MQW has a cutoff wavelength λc (defined as λc = 1.24/Eg,Γ in μm) of around 2.2 μm. In addition, the Ge0.9Sn0.1 MQW is fully-strained to the underlying Ge buffer as confirmed by the reciprocal space X-ray diffraction mapping, avoiding the generation of threading dislocations (TDs) that may originate from the Ge0.9Sn0.1/Ge interfaces. A lower density of TD is preferred as it generally results in a smaller bulk leakage current density for a p-i-n photodiode [28].
After growing the MQW and the p+-Ge layer, circular mesa structures were formed by chlorine-based inductively coupled plasma (ICP) etch. A 250 nm-thick silicon dioxide (SiO2) capping layer was then deposited by plasma enhanced chemical vapor deposition (PECVD) at 250 °C. After that, the contact windows were opened by fluorine-based ICP etch and the subsequent wet etch using diluted hydrogen fluoride (DHF). Finally, aluminum (Al) top and bottom electrodes were formed by sputtering and chlorine-based ICP etch. The electrodes were patterned in a ground-signal-ground (GSG) configuration for RF measurement. The processing temperatures of all the device fabrication steps were kept below 250 °C in order to prevent Sn diffusion, segregation, or clustering.
Transmission electron microscopy (TEM) analysis was performed to examine the crystalline quality of each semiconductor layer after device fabrication. The cross-sectional scanning-TEM (STEM) image of the photodiode near mesa sidewall region is shown in Fig. 2(a). The mesa etching process leads to an over-etching of around 350 nm into the n+-Si layer. It can be observed that ten Ge0.9Sn0.1 QWs are uniformly grown on the Ge buffer. Figure 2(b) shows the cross-sectional TEM image at the Ge0.9Sn0.1 MQW region. The Ge and Ge0.9Sn0.1 thicknesses are consistent with the design illustrated in Fig. 1(b). In addition, the high-crystalline-quality of the Ge0.9Sn0.1 QW is confirmed by the high-resolution TEM (HRTEM) image shown in Fig. 2(c).
Fig. 2 (a) Cross-sectional STEM image of the Ge0.9Sn0.1 MQW p-i-n photodiode showing the edge of the mesa or the mesa sidewall. (b) Cross-sectional TEM image at the Ge0.9Sn0.1 MQW region. (c) HRTEM image at one Ge0.9Sn0.1 QW layer.
3. Results and discussion
3.1 DC measurements
Dark current is one of the key figures-of-merit when evaluating the performance of an infrared photodetector. In most of the applications, e.g. fiber-optic communication, a p-i-n photodiode is normally reverse biased during operation to achieve higher response speed. Thus, the leakage current of the photodetector will contribute to standby power consumption. In addition, a high Idark increases the noise of a detector, resulting in degradation of the signal-to-noise ratio (SNR) of the optical receiver.
The dark current of the Ge0.9Sn0.1 MQW p-i-n photodiode was characterized at various measurement temperatures (T) ranging from 4 to 320 K, as plotted in Fig. 3(a). The diameter (D) of the diode mesa is 20 μm. The arrow indicates the direction of increasing T. In general, an Idark value of less than 1 μA is considered as an acceptable value for a high speed optical receiver, below which the trans-impedance amplifier (TIA) noise becomes the main noise source [29]. It can be seen that the photodiode in this work exhibits sub-1 μA Idark at the bias voltage (Vbias) between −2 to 0 V. In addition, an excellent forward/reverse current ratio of around 5 orders of magnitude is achieved at Vbias = ± 1 V and room-temperature. Figure 3(b) shows the Idark-T characteristics of the photodiode at different Vbias. In the high temperature regime (T > 240 K), it is observed that Idark decreases exponentially with reducing T. This indicates that Idark is dominated by a temperature-dependent current component. In the low temperature regime (T < 30 K), the thermionic process is significantly suppressed. Thus, the current component which is less dependent on temperature becomes more prominent. This can also be seen from Fig. 3(a), where the two Idark-Vbias curves at 4 and 30 K are almost superimposed.
Fig. 3 (a) Idark-Vbias characteristics of the Ge0.9Sn0.1 MQW p-i-n photodiode at various T ranging from 4 to 320 K. The arrow points in the direction of increasing T. The diameter of the diode is 20 μm. (b) Idark-T characteristics of the diode with different Vbias. Idark becomes less temperature-dependent at lower temperatures. The arrow points in the direction of more negative Vbias. (c) Plot of ln(Idark/T3/2) vs. 1/kT for the photodiode with different Vbias. Activation energy (EA) of the photodiode is extracted by linear fitting.
More insights into the mechanism of Idark at reverse bias voltage (Vbias < 0 V) can be obtained from the Arrhenius plot shown in Fig. 3(c). T ranges from 250 to 320 K. The data points are linearly fitted using a least-square-regression method. The gradient of the fitting line yields the thermal activation energy (EA) of Idark [30], according to
where B is a constant, and k is the Boltzmann constant. At Vbias of −0.5, −1, −1.5, and −2 V, EA is extracted to be 0.333, 0.297, 0.261, and 0.206 eV, respectively. It is noted that the EA values are around half of the indirect bandgap of the Ge0.9Sn0.1 QW (Eg,L ≈0.515 eV according to Ref. 9), indicating that Idark at the high T regime is dominated by the Shockley-Reed-Hall (SRH) generation-recombination process via deep-level traps [31,32]. The traps are most likely associated with the core of TDs resulting from the mismatched epitaxy [33,34]. This is not unexpected as a substantial amount of TDs could be generated from the Ge/Si interface and may propagate through the Ge0.9Sn0.1 MQW region. One also sees that EA decreases with increasing the reverse bias, which could be explained by the electric-filed-enhanced emission mechanism, essentially the Poole-Frenkel [35] and the phonon-assisted-tunneling [36] effects. The former states that the thermionic emission energy barrier of a trapped carrier decreases due to the bending of energy band caused by electric field. As for the latter, if the emission process is coupled with suitable phonon(s), the energy barrier will be even lower and the carrier will tunnel through the barrier into the respective conduction or valence bands. Therefore, increasing Vbias enlarges the internal electric filed within the i-layer, leading to the reduction in EA.Next, the dark current of the photodiode in this work is compared with those of the other Ge1−xSnx-on-Si p-i-n photodiodes from the literature [10–25]. Idark is normalized by its diode junction area and Vbias is fixed at −1 V for fair comparison, as shown in Fig. 4. It should be noted that a higher Sn composition generally leads to a larger Jdark due to the smaller bandgap. Despite the highest Sn composition of 10%, Jdark of the Ge0.9Sn0.1 MQW p-i-n photodiode is lower than most of the reported values. It should be noted that Ge1−xSnx with x > 0.06 is normally needed for photodetection beyond 2 μm. Thus, the photodiode in this work has a lowest Jdark among all the 2 μm-wavelength Ge1−xSnx-on-Si p-i-n photodiodes. In addition, the achieved Jdark of 0.031 A/cm2 at −1 V is even comparable to those of the Ge-on-Si p-i-n photodiodes which have typical values ranging from 0.01 to 0.10 A/cm2 [37]. Further suppression of Jdark could be possibly achieved by reducing the threading dislocation density (TDD) of the Ge buffer. This could be realized by increasing the thickness of Ge buffer (dbuffer) [38,39], optimizing the growth technique [40], or implementing a cyclic annealing process [41].
Fig. 4 Benchmarking of dark current density Jdark of Ge1−xSnx-on-Si p-i-n photodiodes at Vbias = −1 V. L is the side length of a square mesa. The photodiode demonstrated in this work has a record high Sn composition and the Jdark is lower than most of the reported values.
Another critical metric of a photodetector is the optical responsivity (Rop), which is defined as the ratio of photocurrent to the incident light power (Pin). The room-temperature direct current (DC) photoresponse of a 20 μm-diameter Ge0.9Sn0.1 MQW p-i-n photodiode was characterized, as plotted in Fig. 5(a). The arrow indicates the direction of increasing light wavelength (λ) from 1530 to 2003 nm. Pin is fixed at 0.24 mW. Based on the measurement results, the wavelength-dependent Rop of the detector at −1 V was obtained, as shown in Fig. 5(b). Rop of the photodiode are 0.216 and 0.023 A/W at the wavelengths of 1530 and 2003 nm, respectively. As compared to Ge-on-Si photodetectors, the device in this work exhibits a significantly extended detection wavelength, covering not only the traditional O to U telecommunication bands but also the novel fiber-optic communication band near 2 μm. The Rop of the photodiode can be further improved by increasing the number of QWs or by implementing the lateral-illuminated (waveguide) structure, which are subjects for future work.
Fig. 5 (a) I-Vbias characteristics of the Ge0.9Sn0.1 MQW p-i-n photodiode (D = 20 μm) illuminated at different light wavelengths (λ) ranging from 1530 to 2003 nm. The arrow indicates the direction of increasing λ. (b) Wavelength-dependent responsivity of the photodiode at Vbias = −1 V.
3.2 RF characterization
Besides DC performance, the radio frequency (RF) response of the Ge0.9Sn0.1 MQW p-i-n photodiode was also investigated. An illustration of the RF measurement system is shown in Fig. 6. The light source was a fiber-coupled Fabry-Pérot laser diode (Thorlabs FPL2000S) with λ = 2 μm and a maximum output power of 15 mW. A polarization controller was used before the lithium-niobate electro-optical (EO) modulator (Photline MX2000-LN-10) rated for 10 GHz. The modulator was driven by signals from an analog signal generator (Keysight N5173B). A single-mode (SM) fiber connected to the modulator output was used to couple light onto the top surface of the photodiode. The spot size of the SM fiber is less than 10 μm, which is smaller than the size of the photodiode mesa. The electrical contact to the photodiode was provided by a microwave probe with a GSG configuration, as shown in the inset of Fig. 6. A bias-tee was used to apply Vbias to the diode. The small signal response of the photodiode was finally measured by a signal analyzer (Agilent N9030A). Calibration measurements were done to remove the effects of modulator, GSG probe, bias-tee, and cables from the final RF response results.
Fig. 6 Illustration of the photodiode RF measurement system at 2 μm-wavelength. The inset is a top-view image of the photodiode during measurement.
One imperative figure-of-merit of a high-speed photodetector is the 3 dB bandwidth, which is defined as the frequency at which the electrical output power is reduced by a factor of two. In general, f3dB of a p-i-n photodiode is mainly limited by 3 factors: (i) diffusion current, (ii) carrier transit time, and (iii) resistance-capacitance (RC) delay. In this work, the first factor can be ignored since negligible photocurrent is generated in the p+-Ge and the n+-Si layers at 2 μm-wavelength. Thus, f3dB of the photodiode can be calculated as
where fT and fRC are the carrier transit frequency and the RC frequency, respectively [42]. The minimum time for carriers to transit the i-region is given by the thickness of the i-region (di) and the saturation drift velocity (vsat). Thus, fT can be calculated byThe RC delay frequency fRC is determined by the series resistance (Rs), load resistance (Rload), junction capacitance (Cj), and parasitic capacitance (Cpar).The RF response of the Ge0.9Sn0.1 MQW p-i-n photodiode with D = 20 μm is plotted in Fig. 7. Assuming an identical vsat value for Ge0.9Sn0.1 and Ge (vsat,GeSn = vsat,Ge = 6 × 106 cm/s), fT of the detector is calculated to be around 39.7 GHz, which is much larger than the as-measured f3dB of around 1.2 GHz. It should also be noted that increasing the reverse bias from −1 to −4 V does not result in a noticeable change in the f3dB value. Therefore, f3dB of the detector should be limited by the RC delay.
Fig. 7 Normalized frequency response of the 20-μm-diameter photodiode at 2 μm-wavelength. Vbias ranges from −1 to −4 V. The 3dB bandwidth f3dB of the detector is around 1.2 GHz.
A quantitatively analysis by SPICE simulation was performed to provide more insights into the fRC of the detector. Figure 8(a) shows the equivalent circuit of the p-i-n photodiode [43]. Rs and Rload are 34 and 50 Ω, respectively. Rs is estimated by fitting the room-temperature Idark-Vbias curve of the detector at a forward bias of around 2 V. The diode junction capacitance Cj is around 6.61 × 10−14 F, as calculated by
where ε0 is the vacuum permittivity and Ad is the diode junction area. The relative permittivity of Ge0.9Sn0.1 (εGeSn) is estimated by linear interpolation between the values of Ge (εGe) and Sn [44]. The parasitic capacitance Cpar is mainly the overlap capacitance between the central signal electrode and the n+-Si layer, separated by a SiO2 film. Cpar can be calculated bywhere εSiO2 is the relative permittivity of SiO2, Aoverlap is the overlapping area, and dSiO2 is the thickness of the SiO2 film. A Cpar value of around 2.07 × 10−12 F is obtained, which is much larger than that of the diode capacitance Cj. Using the resistance and capacitance values calculated above, fRC is calculated to be 1.5 GHz by SPICE simulation. Thus, the theoretical f3dB value of the photodiode is around 1.5 GHz according to Eq. (2), which is slightly larger than the experimental result of 1.2 GHz. One possible reason for the deviation of the experimental data from the calculated one is the carrier trapping effect, which results from the energy band discontinuity of the MQW structure [45].Fig. 8 (a) Equivalent circuit of a p-i-n photodiode. (b) f3dB of the photodiode as a function of Cpar. An f3dB of larger than 10 GHz is expected by implementing a double-mesa structure, which can significantly reduce Cpar of the detector.
It should be noted that the f3dB result in this work is mainly limited by the large parasitic capacitance of the electrode rather than the Ge0.9Sn0.1 p-i-n diode itself. In fact, Cpar can be significantly reduced by implementing a double-mesa structure [46]. This will result in a much higher f3dB, as shown in Fig. 8(b). With a Cpar value in the range of 10−14 F (which is typical for a vertically-illuminated double-mesa p-i-n photodiode as reported in Ref. 42), f3dB is expected to be higher than 10 GHz for the 20 μm-diameter Ge0.9Sn0.1 MQW p-i-n photodiode.
4. Conclusion
A Ge0.9Sn0.1 MQW p-i-n photodiode is demonstrated for 2 μm-wavelength light detection. Temperature-dependent Idark of the detector is measured at T = 4 to 320 K. A Jdark of 0.031 A/cm2 is achieved at Vbias = −1 V and room-temperature, which is among the lowest values for Si-based Ge1−xSnx p-i-n diodes reported in the literature. In addition, the spectral response of the photodetector covers wavelengths ranging from 1530 to 2003 nm. Rop of the photodiode are 0.216 and 0.023 A/W at λ = 1530 and 2003 nm, respectively. Finally, response speed of the photodiode is characterized using a 2 μm-wavelength RF measurement system. The detector exhibits an f3dB of around 1.2 GHz, which is limited by the Cpar of the electrode rather than the Ge0.9Sn0.1 p-i-n diode itself. Further device optimization would extend f3dB to above 10 GHz.
Funding
National University of Singapore (Trailblazer Grant: R-263-000-B43-733).
Acknowledgments
The authors acknowledge Prof. Gengchiau Liang, Sachin Yadav at NUS and Dr. Chongyang Liu at NTU for technical discussions.
References and links
1. 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. O’Carroll, 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(5), 971–975 (2015). [CrossRef]
2. K. Xu, Q. Wu, Y. Xie, M. Tang, S. Fu, and D. Liu, “High speed single-wavelength modulation and transmission at 2 μm under bandwidth-constrained condition,” Opt. Express 25(4), 4528–4534 (2017). [CrossRef] [PubMed]
3. P. S. Jensen, J. Bak, and S. Andersson-Engels, “Influence of temperature on water and aqueous glucose absorption spectra in the near- and mid-infrared regions at physiologically relevant temperatures,” Appl. Spectrosc. 57(1), 28–36 (2003). [CrossRef] [PubMed]
4. P. F. Ambrico, A. Amodeo, P. Di Girolamo, and N. Spinelli, “Sensitivity analysis of differential absorption lidar measurements in the mid-infrared region,” Appl. Opt. 39(36), 6847–6865 (2000). [CrossRef] [PubMed]
5. G. He and H. A. Atwater, “Interband transitions in SnxGe1−x alloys,” Phys. Rev. Lett. 79(10), 1937–1940 (1997). [CrossRef]
6. A. Gassenq, F. Gencarelli, J. Van Campenhout, Y. Shimura, R. Loo, G. Narcy, B. Vincent, and G. Roelkens, “GeSn/Ge heterostructure short-wave infrared photodetectors on silicon,” Opt. Express 20(25), 27297–27303 (2012). [CrossRef] [PubMed]
7. B. R. Conley, J. Margetis, W. Du, H. Tran, A. Mosleh, S. A. Ghetmiri, J. Tolle, G. Sun, R. Soref, B. Li, H. A. Naseem, and S.-Q. Yu, “Si based GeSn photoconductors with a 1.63 A/W peak responsivity and a 2.4 μm longwavelength cutoff,” Appl. Phys. Lett. 105(22), 221117 (2014). [CrossRef]
8. Y. Dong, W. Wang, X. Xu, X. Gong, D. Lei, Q. Zhou, Z. Xu, W. K. Loke, S.-F. Yoon, G. Liang, and Y.-C. Yeo, “Germanium-tin on Si avalanche photodiode: Device design and technology demonstration,” IEEE Trans. Electron Dev. 62(1), 128–135 (2015). [CrossRef]
9. Y. Dong, W. Wang, S. Y. Lee, D. Lei, X. Gong, W. K. Loke, S.-F. Yoon, G. Liang, and Y.-C. Yeo, “Germanium-tin multiple quantum well on silicon avalanche photodiode for photodetection at two micron wavelength,” Semicond. Sci. Technol. 31(9), 095001 (2016). [CrossRef]
10. J. Mathews, R. Roucka, J. Xie, S. Yu, J. Menéndez, and J. Kouvetakis, “Extended performance GeSn/Si(100) pin photodetectors for full spectral range telecommunication applications,” Appl. Phys. Lett. 95(13), 133506 (2009). [CrossRef]
11. J. Mathews, R. Roucka, C. Weng, R. Beeler, J. Tolle, J. Menéndéz, and J. Kouvetakis, “Near IR photodiodes with tunable absorption edge based on Ge1-ySny alloys integrated on silicon,” ECS Trans. 33(6), 765–773 (2010).
12. J. Werner, M. Oehme, M. Schmid, M. Kaschel, A. Schirmer, E. Kasper, and J. Schulze, “Germanium-tin p-i-n photodetectors integrated on silicon grown by molecular beam epitaxy,” Appl. Phys. Lett. 98(6), 061108 (2011). [CrossRef]
13. S. Su, B. Cheng, C. Xue, W. Wang, Q. Cao, H. Xue, W. Hu, G. Zhang, Y. Zuo, and Q. Wang, “GeSn p-i-n photodetector for all telecommunication bands detection,” Opt. Express 19(7), 6400–6405 (2011). [CrossRef] [PubMed]
14. M. Oehme, M. Schmid, M. Kaschel, M. Gollhofer, D. Widmann, E. Kasper, and J. Schulze, “GeSn p-i-n detectors integrated on Si with up to 4% Sn,” Appl. Phys. Lett. 101(14), 141110 (2012). [CrossRef]
15. H. H. Tseng, H. Li, V. Mashanov, Y. J. Yang, H. H. Cheng, G. E. Chang, R. A. Soref, and G. Sun, “GeSn based p-i-n photodiodes with strained active layer on a Si wafer,” Appl. Phys. Lett. 103(23), 231907 (2013). [CrossRef]
16. R. T. Beeler, J. Gallagher, C. Xu, L. Jiang, C. L. Senaratne, D. J. Smith, J. Menéndéz, A. V. G. Chizmeshya, and J. Kouvetakis, “Band gap-engineered group-IV optoelectronic semiconductors, photodiodes and prototype photovoltaic devices,” ECS J. Solid State Sci. Technol. 2(9), Q172–Q177 (2013). [CrossRef]
17. M. Oehme, K. Kostecki, K. Ye, S. Bechler, K. Ulbricht, M. Schmid, M. Kaschel, M. Gollhofer, R. Körner, W. Zhang, E. Kasper, and J. Schulze, “GeSn-on-Si normal incidence photodetectors with bandwidths more than 40 GHz,” Opt. Express 22(1), 839–846 (2014). [CrossRef] [PubMed]
18. M. Oehme, D. Widmann, K. Kostecki, P. Zaumseil, B. Schwartz, M. Gollhofer, R. Koerner, S. Bechler, M. Kittler, E. Kasper, and J. Schulze, “GeSn/Ge multiquantum well photodetectors on Si substrates,” Opt. Lett. 39(16), 4711–4714 (2014). [CrossRef] [PubMed]
19. Y.-H. Peng, H. H. Cheng, V. I. Mashanov, and G.-E. Chang, “GeSn p-i-n waveguide photodetectors on silicon substrates,” Appl. Phys. Lett. 105(23), 231109 (2014). [CrossRef]
20. Y. Dong, W. Wang, D. Lei, X. Gong, Q. Zhou, S.-Y. Lee, W.-K. Loke, S.-F. Yoon, E. S. Tok, G. Liang, and Y.-C. Yeo, “Suppression of dark current in germanium-tin on silicon p-i-n photodiode by a silicon surface passivation technique,” Opt. Express 23(14), 18611–18619 (2015). [CrossRef] [PubMed]
21. B. R. Conley, Y. Zhou, A. Mosleh, S. A. Ghetmiri, W. Du, R. A. Soref, G. Sun, J. Margetis, J. Tolle, H. A. Naseem, and S.-Q. Yu, “Infrared spectral response of a GeSn p-i-n photodiode on Si,” in Group Four Photonics (GFP) 2014, 17–18.
22. T. Pham, W. Du, H. Tran, J. Margetis, J. Tolle, G. Sun, R. A. Soref, H. A. Naseem, B. Li, and S.-Q. Yu, “Systematic study of Si-based GeSn photodiodes with 2.6 μm detector cutoff for short-wave infrared detection,” Opt. Express 24(5), 4519–4531 (2016). [CrossRef]
23. W. Wang, S. Vajandar, S. L. Lim, Y. Dong, V. R. D’Costa, T. Osipowicz, E. S. Tok, and Y.-C. Yeo, “In-situ gallium-doping for forming p+ germanium-tin and application in germanium-tin p-i-n photodetector,” J. Appl. Phys. 119(15), 155704 (2016). [CrossRef]
24. 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 Photonics J. 8(5), 6804706 (2016). [CrossRef]
25. M. Morea, C. E. Brendel, K. Zang, J. Suh, C. S. Fenrich, Y.-C. Huang, H. Chung, Y. Huo, T. I. Kamins, K. C. Saraswat, and J. S. Harris, “Passivation of multiple-quantum-well Ge0.97Sn0.03/Ge p-i-n photodetectors,” Appl. Phys. Lett. 110(9), 091109 (2017). [CrossRef]
26. Y.-H. Kuo, Y. K. Lee, Y. Ge, S. Ren, J. E. Roth, T. I. Kamins, D. A. B. Miller, and J. S. Harris, “Strong quantum-confined Stark effect in germanium quantum-well structures on silicon,” Nature 437(7063), 1334–1336 (2005). [CrossRef] [PubMed]
27. P. Chaisakul, D. Marris-Morini, J. Frigerio, D. Chrastina, M. Rouifed, S. Cecchi, P. Crozat, G. Isella, and L. Vivien, “Integrated germanium optical interconnects on silicon substrates,” Nat. Photonics 8(6), 482–488 (2014). [CrossRef]
28. L. M. Giovane, H.-C. Luan, A. M. Agarwal, and L. C. Kimerling, “Correlation between leakage current density and threading dislocation density in SiGe p-i-n diodes grown on relaxed graded buffer layers,” Appl. Phys. Lett. 78(4), 541–543 (2001). [CrossRef]
29. 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]
30. S. M. Sze and K. K. Ng, Physics of Semiconductor Devices (John Wiley & Sons, Inc., 2007).
31. W. Shockley and W. T. Read, “Statistics of the recombinations of holes and electrons,” Phys. Rev. 87(5), 835–842 (1952). [CrossRef]
32. R. N. Hall, “Electron-hole recombination in germanium,” Phys. Rev. 87(2), 387 (1952). [CrossRef]
33. P. N. Grillot, S. A. Ringel, E. A. Fitzgerald, G. P. Watson, and Y. H. Xie, “Electron trapping kinetics at dislocations in relaxed Ge0.3Si0.7/Si heterostructures,” J. Appl. Phys. 77(7), 3248–3256 (1995). [CrossRef]
34. P. N. Grillot, S. A. Ringel, E. A. Fitzgerald, G. P. Watson, and Y. H. Xie, “Minority- and majority-carrier trapping in strain-relaxed Ge0.3Si0.7/Si heterostructure diodes grown by rapid thermal chemical-vapor deposition,” J. Appl. Phys. 77(2), 676–685 (1995). [CrossRef]
35. J. Frenkel, “On pre-breakdown phenomena in insulators and electronic semi-conductors,” Phys. Rev. 54(8), 647–648 (1938). [CrossRef]
36. G. Vincent, A. Chantre, and D. Bois, “Electric field effect on the thermal emission of traps in semiconductor junctions,” J. Appl. Phys. 50(8), 5484–5487 (1979). [CrossRef]
37. L. Colace, P. Ferrara, G. Assanto, D. Fulgoni, and L. Nash, “Low dark-current germanium-on-silicon near-infrared detectors,” IEEE Photonics Technol. Lett. 19(22), 1813–1815 (2007). [CrossRef]
38. J. M. Hartmann, J.-F. Damlencourt, Y. Bogumilowicz, P. Holliger, G. Rolland, and T. Billon, “Reduced pressure-chemical vapor deposition of intrinsic and doped Ge layers on Si(001) for microelectronics and optoelectronics purposes,” J. Cryst. Growth 274(1−2), 90–99 (2005). [CrossRef]
39. G. Wang, R. Loo, E. Simoen, L. Souriau, M. Caymax, M. M. Heyns, and B. Blanpain, “A model of threading dislocation density in strain-relaxed Ge and GaAs epitaxial films on Si(100),” Appl. Phys. Lett. 94(10), 102115 (2009). [CrossRef]
40. S. B. Samavedam, M. T. Currie, T. A. Langdo, and E. A. Fitzgerald, “High-quality germanium photodiodes integrated on silicon substrates using optimized relaxed graded buffers,” Appl. Phys. Lett. 73(15), 2125–2127 (1998). [CrossRef]
41. H.-C. Luan, D. R. Lim, K. K. Lee, K. M. Chen, J. G. Sandland, K. Wada, and L. C. Kimerling, “High quality Ge epilayers on Si with low threading-dislocation densities,” Appl. Phys. Lett. 75(19), 2909–2911 (1999). [CrossRef]
42. M. Oehme, J. Werner, E. Kasper, M. Jutzi, and M. Berroth, “High bandwidth Ge p-i-n photodetector integrated on Si,” Appl. Phys. Lett. 89(7), 071117 (2006). [CrossRef]
43. A. Novack, M. Gould, Y. Yang, Z. Xuan, M. Streshinsky, Y. Liu, G. Capellini, A. E.-J. Lim, G.-Q. Lo, T. Baehr-Jones, and M. Hochberg, “Germanium photodetector with 60 GHz bandwidth using inductive gain peaking,” Opt. Express 21(23), 28387–28393 (2013). [CrossRef] [PubMed]
44. Y. Yang, K. L. Low, W. Wang, P. Guo, L. Wang, G. Han, and Y.-C. Yeo, “Germanium-tin n-channel tunneling field-effect transistor: Device physics and simulation study,” J. Appl. Phys. 113(19), 194507 (2013). [CrossRef]
45. S. R. Forrest, O. K. Kim, and R. G. Smith, “Optical response time of In0.53Ga0.47As/lnP avalanche photodiodes,” Appl. Phys. Lett. 41(1), 95–98 (1982). [CrossRef]
46. S. Klinger, “Germanium pin photodiodes on silicon and photonic integrated circuits: Components for high-speed optical data communications,” Ph.D. dissertation, University of Stuttgart, 2011.
