A broadband, high-performance infrared Ge photodetector decorated with Au nanoparticles (NPs) is proposed. Photoelectronic characterization demonstrated that the responsivity of devices decorated with Au NPs is as high as 3.95 A/W at a wavelength of 1550 nm. Compared with a Ge photodetector without Au NPs, the responsivity of a device decorated with Au NPs is significantly increased, i.e., by more than 10 times in the entire range of infrared communication wavelengths, including the O, E, S, C, L, and U bands. The increase is ascribed to type-II energy-band alignment between Ge covered with Au NPs and bare Ge, instead of the localized surface-plasmon-resonance effect. The type-II energy-band alignment enhances the spatial electron-hole separation and restrains the electron-hole recombination, thus a larger photocurrent is observed. These results reflect the potential of this approach for achieving broadband, high-performance Ge photodetectors operating in the near-infrared communication band.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Detection of infrared (IR) wavelengths, specifically near-infrared (NIR) wavelengths, is of significant importance in military applications, night vision, and medicine, among others [1, 2]. A Si-based NIR photodetector has been widely used due to its mature technology and competitive price [3, 4]. However, Si photodetectors are at a disadvantage because of the relatively large Si bandgap of 1.12 eV, corresponding to an absorption cutoff wavelength of approximately 1.1 μm. Fortunately, another group-IV material germanium (Ge) is one of the most promising materials for NIR detection owing to its small bandgap of 0.67 eV and full compatibility with Si complementary metal-oxide-semiconductor processes [5, 6]. It is highly desirable to develop Ge photodetectors with a high responsivity at the NIR communication band. In the past few decades, MSM (metal-semiconductor-metal) and PIN type Ge photodetector have been widely studied. In general, the dark current of PIN photodetectors is relatively smaller than the MSM photodetector, owing to the on/off characteristic of the PN junction. Improving the crystal quality or surface passivation can further reduce the dark current of photodetectors . And the responsivity of PIN photodetectors can be easily improved by widening the intrinsic region of the device. In addition, MSM photodetectors have been attracted lots of interests for its high speed performance and ease of fabrication. Moreover, the symmetrical characteristic of MSM photodetectors is benefit for some special applications. Whereas, a significant obstacle, the relatively low absorption coefficient of Ge at the C, L, and U bands, severely limits its application [8–11].
A hybridization photodetector with metal nanostructures is considered an effective method by which to improve the properties of photodetectors [12–19]. The advantages of using noble-metal nanostructures such as Au nanoparticles (NPs) to improve the performances of optical devices have been proved in previous reports [20–22]. Most importantly, Au NPs/semiconductor hybrid systems exhibit remarkable enhancement behavior under ultraviolet and visible-light irradiation. For instance, ZnO nanowires with Au NPs exhibit a large enhancement of photoresponse as compared with a pure ZnO nanowire device under illumination of 350 nm ; the ratio of photocurrent to dark current of a device with Au NPs was approximately 106. For MoTe2 phototransistors decorated with Au NPs, the photocurrent increased by more than 200 times under illumination of both 365- and 405-nm light .In a hybrid black phosphorus/Au NP photodetector , responsivities of 60 and 500 times higher than those of BP photodetectors under 655- and 980-nm light, respectively, were observed. All of these positive results in various structures decorated with Au NPs are mainly based on the localized surface-plasmon-resonance (LSPR) effect. The LSPR effect only appears at a particular wavelength, which is strongly sensitive to the size, shape, and crystalline structure of the nanocrystal, as well as the surrounding environment . Therefore, the LSPR generated by Au NPs is mainly applied to improve the performance of photodetectors at a narrow band in the ultraviolet and visible-light regions, but it can hardly achieve broadband or NIR communication band enhancement.
In this work, a broadband, high-performance NIR hybridization Ge photodetector is proposed. Au NPs of different sizes were fabricated on Ge substrates via sputtering followed by annealing techniques. The responsivity increases by more than 10 times after being covered with Au NPs in the entire range of NIR communication bands due to the type-II energy-band alignment between Ge covered with Au NPs and bare Ge. In addition, the effect of enhanced responsivity weakens with increasing Au NP size. These structures, as demonstrated, are incredibly valuable for the further development of Si-based Ge optoelectronic devices and their applications.
To fabricate Ge photodetectors decorated with Au NPs, initially an array of Au/Ni (100 nm/5 nm) electrodes were patterned on Ge substrates. The substrate in our work is an undoped intrinsic substrate with a thickness of 525 µm and a resistivity greater than 50 Ω·cm. The Ni layer is used as adhesive layer between the substrate and Au electrodes. The gap between the Au pads is 100 μm. Then, using the sputtering technique, Au thin films of different thicknesses, i.e., 8, 12, and 30 nm, were deposited on the surface. Finally, the samples were annealed at 350°C for 30 min to form Au NPs. The Au NPs formed in the gap of two-terminal photodetector are used to modulate the energy-band structure of Ge and further affect the carrier transport. The prepared structures are designated Au NPs (d nm), where d is the initial Au deposition thickness.
The photoelectronic characteristics of the photodetectors were measured using a Keithley 4200A-SCS semiconductor parameter analyzer with suitable light sources. Specifically, a 1550-nm wavelength laser was used during the current-versus-voltage (I-V) measurements. The transmission and reflection spectra were measured using a Fourier infrared spectrometer (Nicolet iS20, Thermo Scientific), which adopted the integrating sphere method. The absorption (A) was obtained from the measured reflection (R) and transmission (T) by the formula A = 1−R−T. Additionally, a tungsten halogen lamp was used as a broadband light source (0.3–2.5 μm, Gloria-T250A, Zolix Instruments) during the response spectra measurements. This broadband light was initially monochromatized by a monochromator (Omni-λ300i, Zolix Instruments), whereupon the monochromatic light was coupled into a multi-mode optical fiber and ultimately shined on the device. In the testing process, a 532-nm-wavelength green light was used for alignment operations, and the power of the monochromatic light from the fiber was calibrated by an optical power meter (PM100D, Thorlabs) with an InGaAs sensor (1.2–2.5 μm, S148C, Thorlabs).
3. Results and discussion
Figure 1(a) is a schematic of an Au-NP-decorated Ge photodetector. Au NPs were formed on substrate between two Au electrodes. The surface morphology of the device decorated with Au NPs (8 nm) was characterized by atomic force microscopy (AFM) as shown in Fig. 1(b). More morphology details for this sample can be found in Figs. S1 and S2 in the Supporting Information. The average height and diameter of Au NPs (8 nm) are approximately 17 and 520 nm, respectively.
The photodetection properties of both devices with or without Au NPs (8 nm) were investigated through the I-V curves, the current–versus–input-optical-power (I-Pin) curves, as well as the current-versus-time (I-t) curves. The dark and illuminated I-V curves of both devices at wavelength of 1550 nm are shown in Fig. 2(a). For the Ge device decorated with Au NPs, an obvious increase on the photocurrent (Iph, Iph=Ilight−Idark, Ilight, and Idark are currents under light and dark conditions, respectively) is observed, indicating that the introduction of Au NPs can improve the photodetection behavior. Moreover, the dark current is increased after decoration with Au NPs. There are two reasons for this increase of dark current: On one hand, the Au NPs on the substrate can serve as a conductive channel for the carriers in photodetectors, and thus increase the surface leakage current; on the other hand, the carrier distribution in Ge was changed due to the decoration of Au NPs on the Ge surface, thus modulating the dark current. Figure 2(b) shows the photocurrent of devices with or without Au NPs as a function of input optical power at 1550 nm at 0.5-V bias. The photocurrent of the device with Au NPs (8 nm) is much larger than that of the device without Au NPs, which confirms the enhancement effect of photoresponse. In the entire range of NIR communication bands, responsivity was calculated according to Iph/Pin and the results shown in Fig. 2(c). Interestingly, an obvious broadband enhancement on the responsivity of the device with Au NPs is observed over the entire range of IR communication wavelengths, including the O, E, S, C, L, and U bands. In particular, the highest responsivity of a device with Au NPs reaches 3.95 A/W at 1550 nm, which is approximately 10 times higher than that of devices without Au NPs. Figure 2(d) shows the light current Ilight of the photodetectors with or without Au NPs under the chopped light illumination at 0.5 V bias. The photocurrent of devices with Au NPs (8 nm) is approximately 10 µA, while that of devices without Au NPs is approximately 1 µA. This result is in good agreement with Fig. 2(c). The transient photocurrent Iph of the photodetectors with Au NPs (8 nm) under the chopped 1550 nm light illumination at 0.5 V bias was shown in Fig. S3(a). The rise time of the device is ∼40 µs and the fall time is ∼75 µs.
To further investigate the mechanism of responsivity enhancement, optical properties of the devices with or without Au NPs (8 nm) were characterized through transmission and reflection spectra. Figure 3(a) shows the transmission of the two samples, and no significant difference is observed at 1300–1700nm, whereas transmittance of the device without Au NPs is slightly higher than that of a device with Au NPs (8 nm) at 1700–1900nm. The reflection of a device with Au NPs (8 nm) is approximately 10% higher than that of samples without Au NPs in the range 1200–1800nm, as shown in Fig. 3(b). Because of the high reflectivity of Au, the reflectivity of the device increases after being covered with Au NPs. As shown in Fig. 3(c), the absorptivity of a device decorated with Au NPs slightly decreases at 1300–1800nm, while the absorptivity of the sample with Au NPs (8 nm) is higher than that of the sample without Au NPs at wavelengths above 1800nm. In other words, by introducing Au NPs, the absorptivity in the NIR communication band is not improved and no resonance absorption peak can be observed in the absorption spectra. Therefore, the LSPR effect is ruled out in the proposed devices owing to the fact that LSPR effect usually enhances absorption and only occurs at a specific wavelength.
In addition to the LSPR effect, Au NPs also may have other physical influences [24–27]. In the present work, the distribution of carriers in Ge could be modulated by the Au NPs. To investigate the effect of Au NPs on carrier distribution, the energy-band diagram and distribution of carrier concentrations in the Au NPs/Ge structures were simulated using Silvaco technology computer aided design tools and the results shown in Figs. 4(b)–4(d). The simulated structure is illustrated in Fig. 4(a). The energy band and distribution of carriers of devices with or without Au NPs are presented in Figs. 4(b) and 4(c), respectively. The obvious band bending in the gap of two electrodes was observed after decoration with Au NPs. This clearly demonstrates that type-II energy-band alignment is formed between Ge covered with Au NPs and bare Ge. At the same time, due to the change of energy band, the carrier distribution is also affected. The spatial electron-hole separation behavior is displayed as well. The holes mainly distribute in Ge area under Au NPs and electrons mainly in bare Ge. The recombination rates of the device with or without Au are shown in Fig. 4(d). The results show that the recombination rate significantly decreased via decoration with Au NPs. This demonstrates that the recombination of electrons and holes was restrained, and thus the photocurrent and response increased.
Based on the above simulated results, energy-band diagrams and distribution of photogenerated carriers of devices are discussed as shown in Fig. 5. Under equilibrium, due to the work-function difference between Au and Ge [Fig. 5(a)], charge transfer and band bending of metal NP/semiconductor Schottky junctions essentially occurs below the NPs , and hence a nanoscale depletion region and built-in electric field are formed below the Au NPs/Ge interface [Fig. 5(b)]. As a result, photogenerated electron-hole pairs are separated by the built-in electric field under illumination [Fig. 5(c)]. Meanwhile, the depletion region beneath the Au NPs/Ge interface can “trap” holes. Consequently, electrons have a greater likelihood of being collected into the electrode, which generates a current gain.
The potential energy and energy-band diagram in a Au-NP-decorated Ge photodetector are schematically shown in Figs. 5(d) and 5(e), respectively. After decoration with Au NPs, the surface potential of the substrate changes accordingly [Fig. 5(d)]. In Fig. 5 (e), because of the pronounced energy-band bending in Ge, a type-II energy-band alignment between Ge covered with Au NPs and bare Ge is formed, which is consistent with the simulation results in Fig. 4(b). The spatially indirect nature of the transitions, which can enhance the spatial electron-hole separation effect, is beneficial for a long carrier lifetime through the reduction of carrier recombination. As a result, the recombination rate of the device with Au NPs decreases, as shown in the simulation results [Fig. 4(d)]. The current gain is proportional to the carrier lifetime when the transit time is constant (gain=τ/ttr, where τ is the time for the carrier trapping and ttr is the time for the transfer of carriers through the device channel) . Therefore, the response gain increases with increasing photogenerated carrier lifetime. Compared to the LSPR effect, this novel mechanism exhibits significant enhancement behavior and no wavelength selectivity.
In addition, the effect of Au size on the photodetection properties was also investigated. Figure 6(a) depicts the responsivity spectra of the devices decorated with Au NPs of different sizes at a wavelength of 1200–1900nm. The responsivity of the Au NPs shows clear size dependence. Moreover, as shown in Fig. 6(b), it describes the Iph of these three devices under the chopped light illumination, the curves confirm that the responsivity decreases. The transient photocurrent of the photodetectors with Au NPs (8 nm), Au NPs (12 nm) and Au NPs (30 nm) under the chopped 1550 nm light illumination at 0.5 V bias is shown in Fig. S3 in the Supporting Information. The rise and fall time of three devices are similar, thus it demonstrates that the NPs size has no obvious influence on the response time. Figures 6(c)–6(e) show the AFM results of the samples with Au NPs (8, 12, and 30 nm), respectively. When the Au deposition thickness increases, the sizes of the Au NPs also increase. To investigate the influence of NPs size on absorption, the transmission, reflection and absorption spectra of devices with Au NPs (8 nm), Au NPs (12 nm) and Au NPs (30 nm) are shown in Fig. S4 in the Supporting Information. The lowest absorption is observed in the sample decorated with Au NPs (8 nm). Therefore, the decrease in responsivity is not originate from the increase in reflection/scattering of incident light. In addition, the width of the Schottky barrier increases at the Au NPs/Ge interface with increasing Au NP size . According to the simulation results shown in Fig. S5, larger Au NPs leads to a wider depletion. But it will enhance the spatial separation effect of photogenerated electron-hole pairs, rather than increase the recombination rate. Therefore, the reduction in responsivity could be ascribed to the larger spacing between Au NPs. As the size of the Au NPs increases, larger spacing between the Au NPs is observed from surface morphology images. And there are a high recombination rate in the spacing area between Au NPs (Fig. 4(d)). Thus, the responsivity reduces as the size of Au NPs increases.
In summary, the responsivity of Ge-based photodetectors increased remarkably over the entire range of NIR communication wavelengths, including the O, E, S, C, L, and U bands, by the introduction of Au NPs. The responsivity of a device with Au NPs increased to 3.95 A/W for 1550-nm wavelength light. This obvious enhancement of the photodetection property can be ascribed to the formation of type-II energy-band alignment between Ge covered with Au NPs and bare Ge. Compared to the LSPR effect, this novel mechanism exhibits a larger broadband enhancement behavior and no wavelength selectivity. These results provide an effective approach to improving the performance of the optoelectronic elements and competitiveness of Ge photodetectors over the entire range of NIR communication bands.
National Key Research and Development Program of China (No. 2019YFB2204400).
The authors declare no conflicts of interest.
No data were generated or analyzed in the presented research.
See Supplement 1 for supporting content.
1. I. Ka, L. F. Gerlein, I. M. Asuo, R. Nechache, and S. G. Cloutier, “An ultra-broadband perovskite-PbS quantum dot sensitized carbon nanotube photodetector,” Nanoscale 10(19), 9044–9052 (2018). [CrossRef]
2. L. Ye, P. Wang, W. Luo, F. Gong, L. Liao, T. Liu, L. Tong, J. Zang, J. Xu, and W. Hu, “Highly polarization sensitive infrared photodetector based on black phosphorus-on-WSe 2 photogate vertical heterostructure,” Nano Energy 37, 53–60 (2017). [CrossRef]
3. P. Chaisakul, V. Vakarin, J. Frigerio, D. Chrastina, G. Isella, L. Vivien, and D. Marris-Morini, “Recent Progress on Ge/SiGe Quantum Well Optical Modulators, Detectors, and Emitters for Optical Interconnects,” Photonics 6(1), 24 (2019). [CrossRef]
4. T. Hu, B. Dong, X. Luo, T.-Y. Liow, J. Song, C. Lee, and G.-Q. Lo, “Silicon photonic platforms for mid-infrared applications [Invited],” Photonics Res. 5(5), 417–429 (2017). [CrossRef]
5. J. Michel, J. Liu, and L. C. Kimerling, “High-performance Ge-on-Si photodetectors,” Nat. Photonics 4(8), 527–534 (2010). [CrossRef]
6. V. Sorianello, A. D. Iacovo, L. Colace, and G. Assanto, “Near-Infrared Photodetectors in Evaporated Ge: Characterization and TCAD Simulations,” IEEE Trans. Electron Devices 60(6), 1995–2000 (2013). [CrossRef]
7. X. Zhao, M. Moeen, M. S. Toprak, G. Wang, J. Luo, X. Ke, Z. Li, D. Liu, W. Wang, C. Zhao, and H. H. Radamson, “Design impact on the performance of Ge PIN photodetectors,” J. Mater. Sci.: Mater. Electron. 31(1), 18–25 (2020). [CrossRef]
8. G.-E. Chang, R. Basu, B. Mukhopadhyay, and P. K. Basu, “Design and Modeling of GeSn-Based Heterojunction Phototransistors for Communication Applications,” IEEE J. Sel. Top. Quantum Electron. 22(6), 425–433 (2016). [CrossRef]
9. 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]
10. 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]
11. 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]
12. C. C. Guo, Z. H. Zhu, X. D. Yuan, W. M. Ye, K. Liu, J. F. Zhang, W. Xu, and S. Q. Qin, “Experimental Demonstration of Total Absorption over 99% in the Near Infrared for Monolayer-Graphene-Based Subwavelength Structures,” Adv. Opt. Mater. 4(12), 1955–1960 (2016). [CrossRef]
13. N. K. Emani, T. F. Chung, X. Ni, A. V. Kildishev, Y. P. Chen, and A. Boltasseva, “Electrically tunable damping of plasmonic resonances with graphene,” Nano Lett. 12(10), 5202–5206 (2012). [CrossRef]
14. B. Feng, J. Zhu, C. Xu, J. Wan, Z. Gan, B. Lu, and Y. Chen, “All-Si Photodetectors with a Resonant Cavity for Near-Infrared Polarimetric Detection,” Nanoscale Res. Lett. 14(1), 39 (2019). [CrossRef]
15. B. Feng, J. Zhu, B. Lu, F. Liu, L. Zhou, and Y. Chen, “Achieving Infrared Detection by All-Si Plasmonic Hot-Electron Detectors with High Detectivity,” ACS Nano 13(7), 8433–8441 (2019). [CrossRef]
16. R. Liu, B. Liao, X. Guo, D. Hu, H. Hu, L. Du, H. Yu, G. Zhang, X. Yang, and Q. Dai, “Study of graphene plasmons in graphene-MoS2 heterostructures for optoelectronic integrated devices,” Nanoscale 9(1), 208–215 (2017). [CrossRef]
17. P. K. Venuthurumilli, P. D. Ye, and X. Xu, “Plasmonic Resonance Enhanced Polarization-Sensitive Photodetection by Black Phosphorus in Near Infrared,” ACS Nano 12(5), 4861–4867 (2018). [CrossRef]
18. B. Zhao, J. M. Zhao, and Z. M. Zhang, “Enhancement of near-infrared absorption in graphene with metal gratings,” Appl. Phys. Lett. 105(3), 031905 (2014). [CrossRef]
19. Y. Fan, C. Guo, Z. Zhu, W. Xu, F. Wu, X. Yuan, and S. Qin, “Monolayer-graphene-based broadband and wide-angle perfect absorption structures in the near infrared,” Sci. Rep. 8(1), 13709 (2018). [CrossRef]
20. K. Liu, M. Sakurai, M. Liao, and M. Aono, “Giant Improvement of the Performance of ZnO Nanowire Photodetectors by Au Nanoparticles,” J. Phys. Chem. C 114(46), 19835–19839 (2010). [CrossRef]
21. W. Chen, R. Liang, Y. Liu, S. Zhang, W. Cheng, L. Zhao, and J. Xu, “Surface plasmon-enhanced photodetection in MoTe2 phototransistors with Au nanoparticles,” Appl. Phys. Lett. 115(14), 142102 (2019). [CrossRef]
22. S. Jeon, J. Jia, J. H. Ju, and S. Lee, “Black phosphorus photodetector integrated with Au nanoparticles,” Appl. Phys. Lett. 115(18), 183102 (2019). [CrossRef]
23. R. Jiang, B. Li, C. Fang, and J. Wang, “Metal/Semiconductor hybrid nanostructures for plasmon-enhanced applications,” Adv. Mater. 26(31), 5274–5309 (2014). [CrossRef]
24. A. Llopis, J. Lin, S. M. S. Pereira, T. Trindade, M. A. Martins, I. M. Watson, A. A. Krokhin, and A. Neogi, “Electrostatic mechanism of strong enhancement of light emitted by semiconductor quantum wells,” Phys. Rev. B 87(20), 201304 (2013). [CrossRef]
25. R. T. Tung, “The physics and chemistry of the Schottky barrier height,” Appl. Phys. Rev. 1(1), 011304 (2014). [CrossRef]
26. J. Lin, A. Llopis, A. Krokhin, S. Pereira, I. M. Watson, and A. Neogi, “Comparison of electrostatic and localized plasmon induced light enhancement in hybrid InGaN/GaN quantum wells,” Appl. Phys. Lett. 104(24), 242106 (2014). [CrossRef]
27. M. Kimura, N. Tarutani, M. Takahashi, S. Karna, A. Neogi, and R. Shimada, “Enhanced photoluminescence emission from anthracene-doped polyphenylsiloxane glass,” Opt. Lett. 38(24), 5224–5227 (2013). [CrossRef]
28. Y. Yin, Z. Wang, S. Wang, Y. Bai, Z. Jiang, and Z. Zhong, “Electrostatic effect of Au nanoparticles on near-infrared photoluminescence from Si/SiGe due to nanoscale metal/semiconductor contact,” Nanotechnology 28(15), 155203 (2017). [CrossRef]
29. Z. Ni, L. Ma, S. Du, Y. Xu, M. Yuan, H. Fang, Z. Wang, M. Xu, D. Li, J. Yang, W. Hu, X. Pi, and D. Yang, “Plasmonic Silicon Quantum Dots Enabled High-Sensitivity Ultrabroadband Photodetection of Graphene-Based Hybrid Phototransistors,” ACS Nano 11(10), 9854–9862 (2017). [CrossRef]