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

High-efficiency GeSn/Ge multiple-quantum-well photodetectors with photon-trapping microstructures operating at 2 µm

Open Access Open Access

Abstract

We introduced photon-trapping microstructures into GeSn-based photodetectors for the first time, and achieved high-efficiency photo detection at 2 µm with a responsivity of 0.11 A/W. The demonstration was realized by a GeSn/Ge multiple-quantum-well (MQW) p-i-n photodiode on a GeOI architecture. Compared with the non-photon-trapping counterparts, the patterning and etching of photon-trapping microstructure can be processed in the same step with mesa structure at no additional cost. A four-fold enhancement of photo response was achieved at 2 µm. Although the incorporation of photo-trapping microstructure degrades the dark current density which increases from 31.5 to 45.2 mA/cm2 at −1 V, it benefits an improved 3-dB bandwidth of 2.7 GHz at bias voltage at −5 V. The optical performance of GeSn/Ge MQW photon-trapping photodetector manifests its great potential as a candidate for efficient 2 µm communication. Additionally, the underlying GeOI platform enables its feasibility of monolithic integration with other photonic components such as waveguide, modulator and (de)multiplexer for optoelectronic integrated circuits (OEICs) operating at 2 µm.

© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement

1. Introduction

The 2 µm wavelength communication band has been proposed as a promising solution for the projected ‘capacity crunch’ of conventional single mode fibers (SMFs) [1]. The hollow-core photonic bandgap fibers (HC-PBGFs) which have a predicted lowest-loss at 2 µm and high bandwidth thulium doped fiber amplifiers (TDFAs) operating at 1,910-2,050 nm collectively endorse the 2 µm communication window [26]. To continuously develop this communication band, myriads of research work has been focused on CMOS compatible Si-based optical components operating at 2 µm such as low bending loss waveguide [7], four-mode division (de)multiplexer [8] and high-speed Mach–Zehnder interferometer modulator and micro-ring modulator [9]. As a pivotal component in optical receivers, photodetectors operating at this wavelength are also indispensable [10]. The current solution for 2 µm photo detection mainly depends on III-V materials which have tunable bandgap to fulfill the requirements of varying communication bands [1113]. Group-IV materials GeSn alloy offers an alternative approach to extend the detection range of Ge towards 2 µm. Compared with III-V material, a CMOS compatible GeSn photodetector is more promising for integration with other on-chip devices to support the data communication at 2 µm [1422].

With regard to surface-illuminated p-i-n photodetectors, there is a trade-off between the responsivity and bandwidth since a thicker intrinsic layer which benefits the photon absorption hampers the response speed. Several solutions have been proposed to alleviate the trade-off for accomplishing high-efficiency and high-speed simultaneously. Resonant cavity enhanced (RCE) photodetectors are capable of achieving high external quantum efficiency (EQE) by elongating the effective optical path through the multi-pass of the incident light [23,24]. Waveguide photodetectors offer an alternative solution by separating different paths for photon absorption and carrier transmission [2528]. In addition, surface plasmon resonance (SPR) phenomena can be utilized to confine the optical field near the surface to enhance the optical absorption in photodetectors [29,30]. Another solution to address the trade-off is photon-trapping microstructure which can guide the normal-incident light to propagate laterally in the photodetectors [31,32]. It has been demonstrated that Ge photodetectors with optimized photon-trapping microstructures are able to achieve a high EQE of 73% at 1,550 nm [32]. Nevertheless, the optimization and application of photon-trapping microstructure for photodetectors operating at 2 µm have not been investigated.

In this work, GeSn/Ge MQW p-i-n photodetectors with photon-trapping microstructures are proposed and realized for high-efficiency photo detection at 2 µm. The pseudo-morphic GeSn/Ge MQW structure, serving as the active absorption layer, was transferred on an insulator platform by direct wafer bonding (DWB) process. By preventing strain relaxation, the dislocation density is significantly suppressed and a low dark current density of 31.5 mA/cm2 was achieved at −1 V at room temperature. The photon-trapping microstructure consisting of a square-array of circular holes was designed and optimized to enhance the optical absorption at 2 µm. It is noteworthy that the patterning and etching of photon-trapping structure were fabricated simultaneously in the same step for the mesa structure formation without introducing process complexity. A remarkably high responsivity of 0.11 A/W was achieved at 2 µm, which has a four-fold enhancement compared with the non-photon-trapping counterparts. Although the total thickness of GeSn layers is 150 nm, relatively high specific detectivity D* of 2.137 × 108 cm·Hz1/2/W was achieved at 2 µm. Furthermore, an improved 3-dB bandwidth was obtained with the incorporation of photon-trapping microstructure.

2. Material characterization and device fabrication

The GeSn/Ge MQW p-i-n structure was grown on a 300-mm Ge-buffered Si substrate by reduced pressure chemical vapor deposition (RPCVD). Heavily in-situ doping with a concentration higher than 1019 cm−3 was utilized in the formation of p+ and n+ contact layers. The intrinsic MQW region consists of 6 periods of 35 nm Ge barrier layer and 25 nm Ge0.92Sn0.08 well layer. The thickness of GeSn layer was designed to be below the critical thickness for the purpose of reducing threading dislocation density (TDD) by preventing strain relaxation. The as-grown wafer was subsequently transferred on an insulator platform by 200-mm DWB process as described in [33]. The insulator platform is beneficial for high-speed operation as it enables minimization of parasitic capacitance through proper design. Meanwhile, it plays a role as a bottom reflector and enhances the absorption of normal incident light. As compared with SOI substrate, the underlying GeOI substrate could be a more promising platform for the OEICs operating at 2 µm due to the higher refractive index, larger thermal-optic effect, stronger plasma dispersion effect, and higher carrier mobility of Ge.

The material quality after DWB and layer transfer process was evaluated by cross-sectional transmission electron microscopy (TEM) in Fig. 1(a). With the aid of SiN layer deposited on donor wafer, the voids were significantly suppressed at the bonding interface [34]. The seamless bonding interface and distinct MQW structure indicate high material quality was maintained after DWB. Secondary ion mass spectrometry (SIMS) depth profiles of Ge, Sn and Si elements are shown in Fig. 1(b). The similar Sn intensity peaks reveal uniform Sn composition in the MQW structure. The optical property of GeSn/Ge MQW was theoretically investigated by the calculation of band structure. In the calculation, the slightly tensile strain in the Ge layer was ignored and GeSn layer was assumed to be fully-strained. The band offsets between Ge and GeSn layers were estimated according to Vegard’s law and deformation potential parameters. Taking quantum confined effect (QCE) into consideration, the ground states of Γ−valley, L−valley, heavy hole and light hole bands were calculated individually by solving one-dimensional Schrödinger equation. The single band calculation results are plotted in Fig. 2. The direct bandgap of GeSn layer was estimated to be 612.1 meV which corresponds to a wavelength of 2,025 nm, manifesting the feasibility of the GeSn/Ge MQW wafer as a candidate for photo detection at 2 µm.

 figure: Fig. 1.

Fig. 1. (a) Cross-sectional TEM image of the GeSn/Ge MQW structure on an insulator platform after DWB process. (b) SIMS depth profiles of Ge, Sn and Si elements in GeSn/Ge MQW structure from surface to buried oxide layer.

Download Full Size | PDF

 figure: Fig. 2.

Fig. 2. Band structure of the GeSn/Ge MQW region calculated by single band model. Band offsets between Ge and GeSn layer are labelled using double-sided arrows. The dashed lines represent ground states of each bands in GeSn layer.

Download Full Size | PDF

The three-dimensional schematic of the GeSn/Ge MQW p-i-n photodetector is illustrated in Fig. 3(a). Electron beam lithography (EBL) and chlorine-based reactive ion etching (RIE) were used for the patterning and etching of mesa and photon-trapping structures. The key fabrication steps are listed in Fig. 3(b). It is noteworthy that the patterning and etching of hole-array were realized simultaneously with mesa structure without introducing an additional mask or process step. The hole-array is assumed to penetrate through both p+-Ge and GeSn/Ge MQW regions and has similar etching depth as the mesa structure. After the formation of mesa and photon-trapping structures, a 400 nm thick SiO2 passivation layer was deposited on the surface by plasma enhanced chemical vapor deposition (PECVD). The contact windows were opened using dry etching followed by wet etching to ensure a clean etch surface. The thickness of SiO2 passivation layer after wet etching process is ∼350 nm which is designed to reduce the reflection at 2 µm. Metal stack of Ti(20 nm)/TiN (50 nm)/Al (300 nm) was deposited by sputtering and the electrode was formed by lift-off process. Finally, rapid thermal annealing (RTA) at 450°C for 1 minute was performed to improve the Ohmic contact.

 figure: Fig. 3.

Fig. 3. (a) 3D schematic of GeSn/Ge MQW photodetector with photon-trapping microstructure. Passivation layer on the top surface is not shown for clarity. (b) Key process steps in the fabrication of the GeSn/Ge MQW photodetectors.

Download Full Size | PDF

Figure 4 shows the plane view scanning electron microscope (SEM) image of the GeSn/Ge MQW photodetector with a mesa diameter of 80 µm. Zoomed-in view of the photon-trapping microstructure at the detection window is shown in the inset. Decent uniformity in hole dimension and sharp edges of the circular holes can be clearly identified. The hole radii in this work vary from 500 to 800 nm and the periodicity is kept at 2 µm.

 figure: Fig. 4.

Fig. 4. Plane view SEM image of GeSn/Ge MQW photodetector with photon-trapping microstructure. The inset is zoomed-in view of the photon-trapping microstructure.

Download Full Size | PDF

3. GeSn/Ge MQW photodetector characterization

The photocurrent was measured at 2 µm using a commercial laser source (Thorlabs-FPL2000) with a central wavelength of 1,993 nm and spectrum bandwidth of 13 nm. Figure 5(a) plots the current-voltage (I-V) curves of photodetectors with and without photon-trapping microstructure under illumination. The incident optical power was fixed at 5.21 mW and the measured photodetectors have an identical mesa diameter of 250 µm. For the photon-trapping photodetector with a hole radius of 700 nm, a remarkably high optical responsivity of 0.11 A/W was achieved, which has a four-fold enhancement compared with the non-photon-trapping counterpart. The flat photo response curve under reverse bias voltage indicates efficient collection of photon-generated carriers. It demonstrates that the GeSn/Ge MQW photodetectors in this work are capable of low-energy consumption operation.

 figure: Fig. 5.

Fig. 5. Current-voltage (I-V) characteristics of the GeSn/Ge MQW photodetectors with and without photon-trapping microstructure at 2 µm.

Download Full Size | PDF

To investigate the photon-trapping microstructure dependent responsivity at 2 µm, photodetectors with various hole radii from 500 to 800 nm were fabricated while the periodicity of the hole array was kept at 2 µm. The bias voltage was fixed at −1 V and the photo currents were obtained at increasing optical power of 1.19, 5.21, 15.64, and 23.21 mW, respectively. The responsivities were extrapolated from the linear interpolation in Fig. 6(a) and were plotted in Fig. 6(b). Improvement in responsivity is observed for all photodetectors with photon-trapping microstructures. By increasing the hole radius, the responsivity firstly increases and then decreases with an optimized hole radius of 700 nm. For the targeted wavelength of 2 µm, the photon-trapping microstructure should be large enough to effectively couple the incident light into photodetectors. It is noteworthy that the absorptive volume of GeSn/Ge MQW region decreases with the increasing of hole radius. The decrease of responsivity for photodetectors with hole radius larger than 700 nm could be attributed to insufficient optical absorption resulted from the reduced volume of absorptive GeSn layers.

 figure: Fig. 6.

Fig. 6. (a) Photocurrents of photon-trapping photodetectors with various hole radii at increasing optical power of 1.19, 5.21, 15.64 and 23.21 mW. The reverse bias voltage was fixed at 1 V. (b) Extrapolated responsivities of photon-trapping photodetectors with various hole radii.

Download Full Size | PDF

The electric field density distribution in photodetectors with and without photon-trapping microstructures were simulated using finite-difference time-domain (FDTD) method. Given the square configuration of the photon-trapping microstructure, periodic boundary condition and plane-wave light source condition were adopted in the simulation. Figure 7 illustrates the simulated electric field density distribution within a unit cell of the photon-trapping structure. For photodetector without photon-trapping microstructure, electric filed is distributed uniformly in each layer as expected in Fig. 7(a) and Fig. 7(c). The relatively high electric field density is observed on the top surface, indicating a considerable proportion of optical loss, which degrades the performance of the photodetector. With regard to photodetector with photon-trapping microstructure in Fig. 7(b) and Fig. 7(d), the optical field is well-confined in the intrinsic layer. The guided lateral mode changes the propagation path of normal incident light. Propagation of photons in the larger lateral dimension elongates the effective absorption length, benefiting the enhanced optical absorption. Furthermore, the surface passivation layer and bottom insulator platform also collectively contribute to effective optical confinement in the vertical direction.

 figure: Fig. 7.

Fig. 7. Cross-sectional view and top view of electric filed density distribution in photodetectors with and without photon-trapping microstructure. The simulated region is a unit cell of hole-array structure and periodical boundary condition was adopted in the simulation.

Download Full Size | PDF

Micro Fourier-transform infrared spectroscopy (micro-FTIR) was used to measure the reflection spectral of photodetectors with and without photon-trapping microstructure. The spot size of micro-FTIR was scaled down to 50 µm to focus on the surface structure and the working wavelength ranges from 2 to 8 µm. The calibration of the reflection spectrum was performed on a standard gold sample. As illustrated in Fig. 8, the envelop curves of reflection spectral for both photodetectors have similar peak positions. With the incorporation of photon-trapping microstructure, multiple sub-peaks emerge around the main peaks. For photon-trapping photodetector with a hole radius of 700 nm, the normalized reflection decreases significantly from 0.72 to 0.29 at 2 µm. The photon-trapping microstructure also serves as an anti-reflection layer by reducing the refractive index difference between air and the device layers.

 figure: Fig. 8.

Fig. 8. Normalized reflection spectral of GeSn/Ge MQW photodetectors with and without photon-trapping microstructure from 2 to 3 µm measured by micro-FTIR.

Download Full Size | PDF

The responsivity of photon-trapping photodetectors was also investigated from 1,500 to 1,630 nm which covers the C-band and L-band. As shown in Fig. 9(a), the photon-trapping microstructure which was designed for optical absorption enhancement at 2 µm also benefits the optical response across the entire L-band. For the photodetector with a hole radius of 600 nm, the responsivity remains almost constant across the L-band with a value of 0.27 A/W. Compared with the non-photon-trapping counterpart, it has a 22% improvement at 1,565 nm and has a 153% improvement at 1,625 nm, respectively.

 figure: Fig. 9.

Fig. 9. (a) Responsivity spectral of GeSn/Ge MQW photodetectors with and without photon-trapping microstructure from 1500 to 1630 nm. (b) Photocurrent of GeSn/Ge MQW photon-trapping photodetector from 1700 nm to 2300 nm at −1 V.

Download Full Size | PDF

The cut-off wavelength of photon-trapping photodetector was measured based on an optical source setup which consists of a broad band laser (1.3−2.5 µm) and a laser line photonic filter (LLFT) (1.7−2.5 µm). As shown in Fig. 9(b), the photo current was obtained from 1,700 to 2,300 nm at −1 V. The minimum wavelength step of the LLFT was fixed at 5 nm. It can be observed that the photo current drops rapidly after 1,950 nm and becomes negligible after 2,260 nm. The peak shoulder of 1,950 nm is related with the direct bandgap of 2,025 nm. Meanwhile, the cut-off wavelength of 2,260 nm agrees well with the calculated indirect bandgap of 2,300 nm. The marginal derivation is probably due to the omission of the influence from other bands since single band model was used in the calculation. In the future, a 30 band k·p model will be adopted to calculate the band structure more precisely.

The effect of photon-trapping microstructure on dark current was systematically investigated. Figure 10(a) illustrates dark current of photodetectors with and without photon-trapping microstructure at room temperature. The mesa diameter varies from 60 to 250 µm. High on-off ratio of ∼2×104 was achieved for photodetector with mesa diameter of 60 µm, manifesting the prominent rectifying behavior. Owing to low dislocation density in the pseudo-morphic GeSn layers, a low dark current density of 31.5 mA/cm2 was achieved at a bias voltage of −1 V. Although the dark current density is higher as compared to reported III-V MQW photodetectors [3537], the achieved value in this work is among the lowest of group-IV photodetectors [16,18]. In the future, lattice matched GeSn/SiGeSn MQW structures could be designed and grown to further reduce the dark current density. Nevertheless, the dark current density increases to 45.2 mA/cm2 at −1 V when the photon-trapping microstructure was incorporated. In order to investigate the mechanism behind the increased dark current density, bulk and surface leakage current densities are extrapolated. The dark current density for the surface-illuminated p-i-n photodetector fabricated in this work can be expressed as:

$${J_{dark}} = {J_{bulk}} + \frac{{4D + 8Nr}}{{{D^2} - 4N{r^2}}}{J_{surf}}$$
where N is number of holes on top surface, D is diameter of the mesa and r is radius of each holes. The coefficient in front of surface leakage current density is known as perimeter-to-area ratio. For photodetector without photon-trapping microstructure, the perimeter-to-area ratio can be simplified as 4/D. Based on a simplified Eq. (1), the bulk and surface leakage current densities are extracted using liner interpolation in Fig. 10(b). The low surface leakage current density of 15.6 µA/cm indicates effective surface passivation, and the low bulk leakage current density of 28.7 mA/cm2 demonstrates high material quality with low threading dislocation density.

 figure: Fig. 10.

Fig. 10. (a) Dark current versus bias voltage (Idark-Vbias) characteristics of GeSn/Ge MQW photodetectors with varying mesa diameters at room temperature. The measured photon-trapping photodetectors have a hole radius of 700 nm. (b) Linear interpolation of dark current density versus 4/D for photodetectors without photon-trapping microstructure.

Download Full Size | PDF

By assuming the bulk leakage current density remains invariant for photodetectors with photon-trapping microstructure, the percentages of surface leakage for both photodetectors were calculated and plotted in Fig. 11(a). With regard to photodetectors without photon-trapping microstructure, smaller mesa diameter photodetector features higher surface leakage percentage due to larger perimeter-to-area ratio 4/D. In the case of photon-trapping photodetectors, the perimeter-to-area ratio becomes slightly complicated since the perimeters and areas of holes should be taken into consideration. The percentage reveals that the increased dark current density mainly results from the increased surface leakage current introduced by photon-trapping microstructure.

 figure: Fig. 11.

Fig. 11. Surface leakage percentages of photodetectors with and without photon-trapping microstructures.

Download Full Size | PDF

The rectifying dark current characteristic for photon-trapping photodetectors can be expressed using the diode current formula:

$$I = {I_0}\exp [\frac{{q(V - {R_S}{I_{net}})}}{{nkT}}] + \frac{V}{{{R_{Sh}}}}$$
where I0 is reverse saturation current, RS is series resistance, RSh is shunt resistance, and Inet is net current flow in the photodiode which is expressed as Inet= IV/RSh. The shunt resistance can be extracted by the first derivative of bias voltage V to dark current I near 0 V. The extracted shunt resistances at room temperature for photon-trapping photodetectors with mesa diameters of 60 and 250 µm are 396 and 27.5 kΩ, respectively. The high shunt resistances indicate effective surface passivation. It is expected that the photodetector with larger mesa diameter has lower shunt resistance owing to the increased surface recombination process on the larger surface area. Substituting I with Inet derived from shunt resistance in Eq. (2), the series resistance can be extrapolated according to the transformed expression:
$$\frac{{dV}}{{d{I_{net}}}} = \frac{{nkT}}{q}I_{net}^{ - 1} + {R_S}$$
As shown in Fig. 12, the y-axis intercept of linear interpolation curve is series resistance. The series resistance are 22.7 and 10.8 Ω for photon-trapping photodetectors with mesa diameters of 60 and 250 µm, respectively. Photodetector with larger surface area features lower series resistance. The high shunt resistance and low series resistance achieved in this work are desirable since high shunt resistance can effectively reduce the thermal noise and low series resistance benefits the resistance-capacitance (RC) delay limited bandwidth.

 figure: Fig. 12.

Fig. 12. Linear interpolation of Inet−1 versus dV/dInetr at room temperature for photon-trapping photodetector with mesa diameter of 60 and 250 µm.

Download Full Size | PDF

Specific detectivity D* is frequently used to evaluate the performance of miscellaneous photodetectors despite of materials, structures, and size. It is described as:

$${D^\ast} = \frac{{\sqrt {A\Delta f} }}{{NEP}} = \frac{{\sqrt {A\Delta f} }}{{{I_{rms}}}}R$$
where A is the active area of photodetector, Δf is noise bandwidth, Irms is root-mean-square noise current and R is the responsivity. The specific detectivity D* has a unit of cm·Hz1/2/W since it normalized the active area to 1 cm2 and noise bandwidth to 1 Hz. The root-mean-square noise current mainly consists of thermal noise and shot noise: ${I_{rms}} = \sqrt {I_{thermal}^2 + I_{shot}^2} $ when the measurement frequency is above 1 Hz. For thermal noise, it describes thermal motion of carriers which can be expressed as ${I_{thermal}} = \sqrt {4kT\Delta f/{R_{shunt}}} $. For shot noise, it evaluates carrier fluctuations and is described as ${I_{shot}} = \sqrt {2q({I_{photo}} + {I_{dark}})\Delta f} $. It should be noted that all the characterization in this work was performed at room temperature, therefore T is set to 300 K. With regard of photon-trapping photodetector with a mesa diameter of 250 µm and a hole radius of 700 nm, the calculated thermal noise is 7.7619×10−13 A·Hz−1/2 and shot noise is 1.1378×10−11 A·Hz−1/2 at −0.01 V when the surface is illuminated. Due to effective passivation, shot noise dominates minimum detectable signal of the photodetector over thermal noise even at room temperature. The specific detectivity D* achieved at 2 µm is 2.137×108 cm·Hz1/2/W at room temperature. It is slightly lower than specific detectivity D* of 3.668×108 cm·Hz1/2/W in bulk Ge0.9Sn0.1 photodiode in [17]. It is reasonable since the absorption layer in our GeSn/Ge MQW region is much thinner. At 2 µm, only GeSn well regions contribute to optical absorption and the total thickness is only 150 nm. The specific detectivity spectrum for photon-trapping photodetector with a hole radius of 600 nm and mesa diameter of 250 µm was also plotted in Fig. 13 from 1,500 to 1,630 nm. The peak point of specific detectivity D* spectrum is located at 1,500 nm with a value of 6.5699×108 cm·Hz1/2/W at room temperature.

 figure: Fig. 13.

Fig. 13. The specific detectivity D* spectrum of photon-trapping photodetector with hole radius of 600 nm and mesa diameter of 250 µm from 1500 to 1630 nm at room temperature. The red star point represents the specific detectivity D* of photon-trapping photodetector with hole radius of 700 nm and mesa diameter of 250 µm at 2 µm at room temperature. The specific detectivity D* of bulk Ge0.9Sn0.1 photodiode in [17] was also plotted for comparison.

Download Full Size | PDF

The influence of photon-trapping microstructure on frequency response was further investigated at wavelengths of 1.55 and 2 µm. As illustrated in Fig. 14(a), the 3-dB bandwidth increases with the reverse bias. For the GeSn/Ge MQW photodetectors, the 3-dB bandwidth is mainly limited by three factors: carrier transit time, RC delay, and carrier trapping effect at the heterojunction. The increase of reverse bias not only reduces the carrier transit time, but also makes the barrier at the heterojunction narrower which effectively alleviates the carrier trapping effect. It can be observed that the 3-dB bandwidth for photon-trapping photodetectors improves by ∼1.4 and ∼12.3% at reverse bias voltages of 3 and 5 V, respectively. The enhancement may be attributed to reduced RC delay since the reduced surface area leads to a smaller junction capacitance. Relatively high 3-dB bandwidth of 2.7 GHz is obtained on photon-trapping photodetector with a mesa diameter of 60 µm at 2 µm. Higher bandwidth larger than 10 GHz is expected when the mesa diameter scales down to 20 µm [16].

 figure: Fig. 14.

Fig. 14. (a) Small signal frequency response of GeSn/Ge MQW photodetectors with mesa diameters of 60 µm at 1.55 µm. (b) Small signal frequency response of GeSn/Ge MQW photodetectors with mesa diameters of 60 µm at 2 µm.

Download Full Size | PDF

4. Conclusion

GeSn/Ge MQW p-i-n photodetectors with photon-trapping microstructure were fabricated and characterized. The 300-mm RPCVD wafer growth technology and 200-mm layer transfer technology adopted in this work are promising for large-scale production. The MQW structure with pseudo-morphic GeSn layer ensures low threading dislocation by preventing strain relaxation, benefiting low dark current density of 31.5 mA/cm2 at −1 V. The introduction of photon-trapping microstructure enables high-efficiency photo detection with four-fold increased responsivity of 0.11 A/W at 2 µm. Furthermore, it benefits high-response photo detection due to reduced RC delay, with a ∼12% improvement in the 3-dB bandwidth at −5 V. This work enables GeSn photodetector to be a competent candidate for efficient data communication at 2 µm. It is desirable to monolithically integrate this work with GeSn lasers, transistors, waveguides and modulators on the GeOI platform in the future.

Funding

National Research Foundation Singapore (NRF– CRP19–2017–01); Ministry of Education - Singapore (R-263-000-D45-112); Ministry of Education - Singapore (R-263-000-C58-133).

Acknowledgements

The authors acknowledge Ms. Jin Zhou in Nanyang Technological University for the assistance in electron beam lithography.

Disclosures

The authors declare that there are no conflicts of interest related to this article.

References

1. D. J. Richardson, “Filling the light pipe,” Science 330(6002), 327–328 (2010). [CrossRef]  

2. P. Roberts, F. Couny, H. Sabert, B. Mangan, D. Williams, L. Farr, M. Mason, A. Tomlinson, T. Birks, and J. Knight, “Ultimate low loss of hollow-core photonic crystal fibres,” Opt. Express 13(1), 236–244 (2005). [CrossRef]  

3. H. Zhang, Z. Li, N. Kavanagh, J. Zhao, N. Ye, Y. Chen, N. Wheeler, J. Wooler, J. Hayes, and S. Sandoghchi, “81 Gb/s WDM transmission at 2µm over 1.15 km of low-loss hollow core photonic bandgap fiber,” in 2014 The European Conference on Optical Communication (ECOC), (IEEE, 2014), pp. 1–3.

4. H. Zhang, N. Kavanagh, Z. Li, J. Zhao, N. Ye, Y. Chen, N. Wheeler, J. Wooler, J. Hayes, and S. Sandoghchi, “100 Gbit/s WDM transmission at 2 µm: transmission studies in both low-loss hollow core photonic bandgap fiber and solid core fiber,” Opt. Express 23(4), 4946–4951 (2015). [CrossRef]  

5. Z. Li, A. Heidt, J. Daniel, Y. Jung, S. Alam, and D. J. Richardson, “Thulium-doped fiber amplifier for optical communications at 2 µm,” Opt. Express 21(8), 9289–9297 (2013). [CrossRef]  

6. Z. Li, A. Heidt, N. Simakov, Y. Jung, J. Daniel, S. Alam, and D. Richardson, “Diode-pumped wideband thulium-doped fiber amplifiers for optical communications in the 1800–2050nm window,” Opt. Express 21(22), 26450–26455 (2013). [CrossRef]  

7. J. Li, Y. Liu, Y. Meng, K. Xu, J. Du, F. Wang, Z. He, and Q. Song, “2 µm Wavelength Grating Coupler, Bent Waveguide, and Tunable Microring on Silicon Photonic MPW,” IEEE Photonics Technol. Lett. 30(5), 471–474 (2018). [CrossRef]  

8. S. Zheng, M. Huang, X. Cao, L. Wang, Z. Ruan, L. Shen, and J. Wang, “Silicon-based four-mode division multiplexing for chip-scale optical data transmission in the 2 µm waveband,” Photonics Res. 7(9), 1030–1035 (2019). [CrossRef]  

9. W. Cao, D. Hagan, D. J. Thomson, M. Nedeljkovic, C. G. Littlejohns, A. Knights, S.-U. Alam, J. Wang, F. Gardes, and W. Zhang, “High-speed silicon modulators for the 2 µm wavelength band,” Optica 5(9), 1055–1062 (2018). [CrossRef]  

10. R. Soref, “Group IV photonics: Enabling 2 µm communications,” Nat. Photonics 9(6), 358–359 (2015). [CrossRef]  

11. Y. Chen, Z. Xie, J. Huang, Z. Deng, and B. Chen, “High-speed uni-traveling carrier photodiode for 2 µm wavelength application,” Optica 6(7), 884–889 (2019). [CrossRef]  

12. J. M. Wun, Y. W. Wang, Y. H. Chen, J. E. Bowers, and J. W. Shi, “GaSb-Based pin Photodiodes With Partially Depleted Absorbers for High-Speed and High-Power Performance at 2.5 µm Wavelength,” IEEE Trans. Electron Devices 63(7), 2796–2801 (2016). [CrossRef]  

13. H. Yang, N. Ye, R. Phelan, J. O’Carroll, B. Kelly, W. Han, X. Wang, N. Nudds, N. MacSuibhne, and F. Gunning, “Butterfly packaged high-speed and low leakage InGaAs quantum well photodiode for 2000nm wavelength systems,” Electron. Lett. 49(4), 281–282 (2013). [CrossRef]  

14. S. Xu, Y.-C. Huang, K. H. Lee, W. Wang, Y. Dong, D. Lei, S. Masudy-Panah, C. S. Tan, X. Gong, and Y.-C. Yeo, “GeSn lateral pin photodetector on insulating substrate,” Opt. Express 26(13), 17312–17321 (2018). [CrossRef]  

15. Y. Lin, K. H. Lee, S. Bao, X. Guo, H. Wang, J. Michel, and C. S. Tan, “High-efficiency normal-incidence vertical pin photodetectors on a germanium-on-insulator platform,” Photonics Res. 5(6), 702–709 (2017). [CrossRef]  

16. 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(4), 5798–5813 (2019). [CrossRef]  

17. 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]  

18. H. Tran, T. Pham, W. Du, Y. Zhang, P. C. Grant, J. M. Grant, G. Sun, R. A. Soref, J. Margetis, and J. Tolle, “High performance Ge0.89Sn0.11 photodiodes for low-cost shortwave infrared imaging,” J. Appl. Phys. 124(1), 013101 (2018). [CrossRef]  

19. H. Tran, T. Pham, J. Margetis, Y. Zhou, W. Dou, P. C. Grant, J. M. Grant, S. Alkabi, G. Sun, and R. A. Soref, “Si-based GeSn photodetectors towards mid-infrared imaging applications,” ACS Photonics 6(11), 2807–2815 (2019). [CrossRef]  

20. M. Oehme, K. Kostecki, K. Ye, S. Bechler, K. Ulbricht, M. Schmid, M. Kaschel, M. Gollhofer, R. Körner, and W. Zhang, “GeSn-on-Si normal incidence photodetectors with bandwidths more than 40 GHz,” Opt. Express 22(1), 839–846 (2014). [CrossRef]  

21. Y. Dong, W. Wang, S. Xu, D. Lei, X. Gong, X. Guo, H. Wang, S.-Y. Lee, W.-K. Loke, and S.-F. Yoon, “Two-micron-wavelength germanium-tin photodiodes with low dark current and gigahertz bandwidth,” Opt. Express 25(14), 15818–15827 (2017). [CrossRef]  

22. M. Oehme, D. Widmann, K. Kostecki, P. Zaumseil, B. Schwartz, M. Gollhofer, R. Koerner, S. Bechler, M. Kittler, and E. Kasper, “GeSn/Ge multiquantum well photodetectors on Si substrates,” Opt. Lett. 39(16), 4711–4714 (2014). [CrossRef]  

23. B.-J. Huang, J.-H. Lin, H. Cheng, and G.-E. Chang, “GeSn resonant-cavity-enhanced photodetectors on silicon-on-insulator platforms,” Opt. Lett. 43(6), 1215–1218 (2018). [CrossRef]  

24. J. Song, S. Yuan, and J. Xia, “High efficiency resonant-metasurface germanium photodetector with ultra-thin intrinsic layer,” arXiv preprint arXiv:1904.05744 (2019).

25. T. Yin, R. Cohen, M. M. Morse, G. Sarid, Y. Chetrit, D. Rubin, and M. J. Paniccia, “31 GHz Ge nip waveguide photodetectors on Silicon-on-Insulator substrate,” Opt. Express 15(21), 13965–13971 (2007). [CrossRef]  

26. L. Vivien, J. Osmond, J.-M. Fédéli, D. Marris-Morini, P. Crozat, J.-F. Damlencourt, E. Cassan, Y. Lecunff, and S. Laval, “42 GHz pin Germanium photodetector integrated in a silicon-on-insulator waveguide,” Opt. Express 17(8), 6252–6257 (2009). [CrossRef]  

27. Y.-H. Peng, H. Cheng, V. I. Mashanov, and G.-E. Chang, “GeSn pin waveguide photodetectors on silicon substrates,” Appl. Phys. Lett. 105(23), 231109 (2014). [CrossRef]  

28. Y.-H. Huang, G.-E. Chang, H. Li, and H. Cheng, “Sn-based waveguide pin photodetector with strained GeSn/Ge multiple-quantum-well active layer,” Opt. Lett. 42(9), 1652–1655 (2017). [CrossRef]  

29. J. Tong, W. Zhou, Y. Qu, Z. Xu, Z. Huang, and D. H. Zhang, “Surface plasmon induced direct detection of long wavelength photons,” Nat. Commun. 8(1), 1660 (2017). [CrossRef]  

30. J. Tong, F. Suo, J. Ma, L. Y. Tobing, L. Qian, and D. H. Zhang, “[Opto-Electron Adv, 2019, 2 (1)] Surface plasmon enhanced infrared photodetection,” Opto-Electron. Rev. 3(1), 18002601–18002610 (2019). [CrossRef]  

31. Y. Gao, H. Cansizoglu, K. G. Polat, S. Ghandiparsi, A. Kaya, H. H. Mamtaz, A. S. Mayet, Y. Wang, X. Zhang, and T. Yamada, “Photon-trapping microstructures enable high-speed high-efficiency silicon photodiodes,” Nat. Photonics 11(5), 301–308 (2017). [CrossRef]  

32. H. Cansizoglu, C. Bartolo-Perez, Y. Gao, E. P. Devine, S. Ghandiparsi, K. G. Polat, H. H. Mamtaz, T. Yamada, A. F. Elrefaie, and S.-Y. Wang, “Surface-illuminated photon-trapping high-speed Ge-on-Si photodiodes with improved efficiency up to 1700nm,” Photonics Res. 6(7), 734–742 (2018). [CrossRef]  

33. S. Xu, K. Han, Y.-C. Huang, K. H. Lee, Y. Kang, S. Masudy-Panah, Y. Wu, D. Lei, Y. Zhao, and H. Wang, “Integrating GeSn photodiode on a 200 mm Ge-on-insulator photonics platform with Ge CMOS devices for advanced OEIC operating at 2 µm band,” Opt. Express 27(19), 26924–26939 (2019). [CrossRef]  

34. K. H. Lee, S. Bao, Y. Wang, E. A. Fitzgerald, and C. Seng Tan, “Suppression of interfacial voids formation during silane (SiH4)-based silicon oxide bonding with a thin silicon nitride capping layer,” J. Appl. Phys. 123(1), 015302 (2018). [CrossRef]  

35. B. Tossoun, J. Zang, S. J. Addamane, G. Balakrishnan, A. L. 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(20), 4981–4987 (2018). [CrossRef]  

36. B. Chen, W. Jiang, J. Yuan, A. L. Holmes, and B. M. Onat, “SWIR/MWIR InP-based pin photodiodes with InGaAs/GaAsSb type-II quantum wells,” IEEE J. Quantum Electron. 47(9), 1244–1250 (2011). [CrossRef]  

37. 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(26), 35034–35045 (2018). [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 (14)

Fig. 1.
Fig. 1. (a) Cross-sectional TEM image of the GeSn/Ge MQW structure on an insulator platform after DWB process. (b) SIMS depth profiles of Ge, Sn and Si elements in GeSn/Ge MQW structure from surface to buried oxide layer.
Fig. 2.
Fig. 2. Band structure of the GeSn/Ge MQW region calculated by single band model. Band offsets between Ge and GeSn layer are labelled using double-sided arrows. The dashed lines represent ground states of each bands in GeSn layer.
Fig. 3.
Fig. 3. (a) 3D schematic of GeSn/Ge MQW photodetector with photon-trapping microstructure. Passivation layer on the top surface is not shown for clarity. (b) Key process steps in the fabrication of the GeSn/Ge MQW photodetectors.
Fig. 4.
Fig. 4. Plane view SEM image of GeSn/Ge MQW photodetector with photon-trapping microstructure. The inset is zoomed-in view of the photon-trapping microstructure.
Fig. 5.
Fig. 5. Current-voltage (I-V) characteristics of the GeSn/Ge MQW photodetectors with and without photon-trapping microstructure at 2 µm.
Fig. 6.
Fig. 6. (a) Photocurrents of photon-trapping photodetectors with various hole radii at increasing optical power of 1.19, 5.21, 15.64 and 23.21 mW. The reverse bias voltage was fixed at 1 V. (b) Extrapolated responsivities of photon-trapping photodetectors with various hole radii.
Fig. 7.
Fig. 7. Cross-sectional view and top view of electric filed density distribution in photodetectors with and without photon-trapping microstructure. The simulated region is a unit cell of hole-array structure and periodical boundary condition was adopted in the simulation.
Fig. 8.
Fig. 8. Normalized reflection spectral of GeSn/Ge MQW photodetectors with and without photon-trapping microstructure from 2 to 3 µm measured by micro-FTIR.
Fig. 9.
Fig. 9. (a) Responsivity spectral of GeSn/Ge MQW photodetectors with and without photon-trapping microstructure from 1500 to 1630 nm. (b) Photocurrent of GeSn/Ge MQW photon-trapping photodetector from 1700 nm to 2300 nm at −1 V.
Fig. 10.
Fig. 10. (a) Dark current versus bias voltage (Idark-Vbias) characteristics of GeSn/Ge MQW photodetectors with varying mesa diameters at room temperature. The measured photon-trapping photodetectors have a hole radius of 700 nm. (b) Linear interpolation of dark current density versus 4/D for photodetectors without photon-trapping microstructure.
Fig. 11.
Fig. 11. Surface leakage percentages of photodetectors with and without photon-trapping microstructures.
Fig. 12.
Fig. 12. Linear interpolation of Inet−1 versus dV/dInetr at room temperature for photon-trapping photodetector with mesa diameter of 60 and 250 µm.
Fig. 13.
Fig. 13. The specific detectivity D* spectrum of photon-trapping photodetector with hole radius of 600 nm and mesa diameter of 250 µm from 1500 to 1630 nm at room temperature. The red star point represents the specific detectivity D* of photon-trapping photodetector with hole radius of 700 nm and mesa diameter of 250 µm at 2 µm at room temperature. The specific detectivity D* of bulk Ge0.9Sn0.1 photodiode in [17] was also plotted for comparison.
Fig. 14.
Fig. 14. (a) Small signal frequency response of GeSn/Ge MQW photodetectors with mesa diameters of 60 µm at 1.55 µm. (b) Small signal frequency response of GeSn/Ge MQW photodetectors with mesa diameters of 60 µm at 2 µm.

Equations (4)

Equations on this page are rendered with MathJax. Learn more.

J d a r k = J b u l k + 4 D + 8 N r D 2 4 N r 2 J s u r f
I = I 0 exp [ q ( V R S I n e t ) n k T ] + V R S h
d V d I n e t = n k T q I n e t 1 + R S
D = A Δ f N E P = A Δ f I r m s R
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.