Open Access
Add to BibSonomy
Abstract
An efficient and effective method to achieve high responsivity and specific detectivity, particularly for normal-incident quantum well infrared photodetectors (QWIPs), is proposed in this study. By combining superlattice (SL) structure, grating structures, and graphene monolayer onto traditional QWIP designs, a graphene-covered multicolor quantum grid infrared photodetector (QGIP) with improved optoelectrical properties is developed. The enhancements of the device’s responsivity and specific detectivity are about 7-fold and 20-fold, respectively, which resulted from an increase in the charge depletion region and the generation of extra photoelectrons due to graphene-semiconductor heterojunction. This method provides a potential candidate for future high-performance photodetectors.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Far infrared photodetection is an important technique applied in the military, medical, and meteorological fields among others. Its relevance in our daily lives is widely recognized; in fact, several researchers have already spent decades of studies to explore its capabilities and to describe the associated mechanisms [1–5]. In thermal imaging applications, quantum well infrared photodetectors (QWIPs) can be used to operate at around 6-12 µm wavelengths, which is about the range of radiation of an object at room temperature. This temperature region is closely accustomed to us as it covers, for instance, human body temperature and engine temperature, which we get along with every day. QWIPs are easily fabricated using III-V semiconductors since their material growth and processing techniques are already well developed [6,7]. The QWIP structure consists of very thin epitaxial heterostructures that provide a tunable feature and possess extremely high resolution at different temperatures, owing to its inherent quantum confinement and discrete energy band [8]. Nevertheless, QWIPs need a critical optical coupling scheme to reorient a normal incident light into the required direction for quantum well absorption – a major drawback that may limit their applications. Many device designs are built to overcome this obstacle, some of which include resonant QWIPs (R-QWIPs) [9,10], corrugated QWIPs [11,12] and photonic crystal resonators [13]. The main trend to increase quantum efficiency is by optical coupling with surface plasmons [14–16]. However, the energy concentrated region of surface plasmons is too shallow to reach a depth where the QWIP’s active layers are located; thus, the photoenergy cannot be utilized efficiently. In this paper, we propose a combination of three design improvements to enhance the quantum efficiency of QWIP. (1) Firstly, we combine traditional QWIP with a layer of superlattice (SL) structure. The SL structure does not just lower down the dark current in the device; it also provides a large photocurrent amplification and a bias-tunable spectrum [17–19]. Further, by combining with QW structure, the infrared photodetector can be selectively operated at two different wavelengths. This device functionality is called multicolor operation. (2) We also couple the QWIP with grating structures, which consequently forms what is generally known as the quantum grid infrared photodetector (QGIP) [20,21], to generate different modes of diffraction lights and to improve the confinement of energy within the active region. (3) Lastly, we inserted a graphene-semiconductor Schottky junction onto the photodetector [22–25]. This Schottky junction is expected to provide a very wide absorption band, covering the infrared range, due to generation of extra photoelectrons [26–30]. By combining traditional QWIP with SL structure, grating coupling, and graphene-GaAs Schottky junction, we can demonstrate an obviously enhanced multicolor-QGIP performance.
2. Experiments and methods
The schematic plot of our complete multicolor-QGIP with graphene overlay, as well as its stacking structure, is shown in Fig. 1(a). The device’s SL active layer consists of 15 periods of GaAs (40 Å, with doped concentration of 4×1017 cm−3)/Al0.27Ga0.73As (60 Å, without doping), whereas its multi-quantum well (MQW) layer is composed of 5 periods of GaAs (60 Å, with doped concentration of 4×1017 cm−3)/Al0.27Ga0.73As (500 Å, without doping). These two layers are separated by an intrinsic AlGaAs barrier with a thickness of around 120 nm. Top and bottom contacts are made up of 300 nm and 900 nm-thick highly doped GaAs, respectively; the entire grown layered structure was deposited on intrinsic GaAs substrates via molecular beam epitaxy. At this point, what has been built is a standard QWIP device (QWIP with SL structure).
Fig. 1. (a) Schematic structure of an as-prepared QGIP integrated with graphene on top. (b) A top-viewed SEM image of one of the samples. The inset shows the QGIP covered with a monolayer graphene sheet.
To overlay grating configurations on the top GaAs contact, e-beam lithography was performed to generate grating patterns. This is followed by inductive-coupled plasma etching to finally create the grating structure. Photolithography and wet etching were then adopted to define the isolation between each device. The size of a single device is equal to 300 µm × 300 µm, and the etching depth reaching the bottom contact is around 1 µm. Both top and bottom leads were constructed via another set of photolithography and e-gun deposition processes and are made up of a 50 nm Au/Ge/Ni alloy topped with a 50 nm Au layer. Rapid thermal annealing (RTA) process was finally conducted to form ohmic contacts between GaAs and metal layer and to complete the QGIP device structure.
To evaluate and compare the performance of QGIP devices with and without monolayer graphene film on top, a subset of the as-prepared QGIP devices underwent further fabrication processes. A monolayer graphene film grown by chemical vapor deposition (CVD) was transferred via polymethyl-methacrylate (PMMA) transfer method [31] onto the as-prepared QGIP. O2 plasma etching was executed on the graphene layer to define an active area and to prevent short circuit to occur among devices. A scanning electron microscopy (SEM) image of a multicolor-QGIP with graphene is shown in Fig. 1(b) with an inset displaying a graphene layer covering the top of a periodic grating structure.
The temperature of the device under measurement was initially lowered down to 60 K inside an enclosed cryo-system within which the photocurrent spectra of the device was measured using Fourier transform infrared (FTIR) spectroscopy while it was illuminated by a 1000 °C blackbody source. During measurement, to eliminate dark current in the samples, the photon flux of the blackbody source was modulated by an optical chopper; then the measured photocurrent was referred to a modulator to extract the sample’s optical signal via a lock-in amplifier (Stanford Research SR830). In this experimental setup, an optical chopper is used with a frequency of 1000 Hz between the blackbody source and the QGIP device in order to directly measure photocurrent levels using a lock-in amplifier. This setting frequency is used because the device response time is commonly within the range of microsecond to picosecond scale as pointed out in previous reports featuring a graphene/GaAs photodetector [28] and traditional QWIP devices [32,33]. In our case, we can make sure that the devices can operate at least at a speed of 1 ms or even shorter.
3. Results and discussion
Device responsivity and detectivity are figure of merit of photodetectors, including QWIPs. The peak responsivity (Rpeak) of QWIP is defined as the measured photocurrent (Ip) across the device divided by its absorbed photoenergy (EP),
For a calibrated 1000 °C blackbody source, the device’s EP is expressed as [34]To observe and compare solely the bandwidth of our standard QWIP device (QWIP with SL layer) at different bias voltages, the $\widehat {R}(\lambda)$ corresponding to their maximum peak responsivity is shown in Fig. 2. The $\widehat {R}(\lambda)$ of our QWIP at different bias voltages at 60 K could exhibit a tunable bandwidth under both narrowband and broadband detection mode.
Fig. 2. Normalized responsivity spectra of the standard QWIP at (a) positive bias voltages (0.3 V) and (b) negative bias voltages (−0.6 V). The insets show the corresponding band diagrams. (c) Responsivity spectra of the standard QWIP at various negative bias voltages. (d) The energy fraction in the absorption spectrum of QW (black square) and SL (red circle).
An absorption peak, originating from absorptions in the QW active layer, was found at a wavelength of 9.5 µm with an applied positive bias voltage [Fig. 2(a)]; while an additional peak, coming from absorptions in the SL active layer, appeared at around 7.8 µm as the device is subjected to a negative bias voltage [Fig. 2(b) to 2(c)] resulting to a broadband absorption [17]. At a positive bias condition, according to the layered structure of the utilized epi-wafer [Fig. 1(a)], photoelectrons generated from the QW region should travel through both barrier and SL regions, which will then be collected at the bottom contact. Since resistances of the QW layer and barrier regions are much higher than that of the SL layer, most of the bias will drop in these two regions. Therefore, only the QWs can be activated under a positive bias condition; photoelectrons generated within the SL region barely contribute to the device photocurrent. In a negative bias condition, on the other hand, the potential level of the SL layer changes into a high electric potential region. Photoelectrons within the wells of this region possess enough energy to escape toward the QW region due to a strong potential drop at the barrier. Consequently, abundant photoelectrons carrying signals from the SL are now capable of traversing through the barrier and QWs, which will then be collected at the bottom contact with the assistance of an external electric field [17]. In this case, both signals from the SL and QWs will appear in the spectrum [Fig. 2(b) and 2(c)]. The corresponding band diagrams for positive and negative bias conditions are shown in the inset of Fig. 2(a) and Fig. 2(b), respectively. Additionally, the peak value of the QW signal grew up more rapidly in contrast to the SL peak as negative bias was gradually increased; this can be seen in Fig. 2(c), where a tunable behavior is shown using three different bias voltages, that is, 0.3 V, −0.6 V, and −1 V. To verify the contribution weight (or energy fraction) of the QW and SL in the absorption spectra, individual intensities of the QW region (from 8.8 µm to 12 µm) and SL region (from 5 µm to 8.8 µm) were each integrated and then divided by the total spectrum intensity. In Fig. 2(d), it is obvious that the contribution weight gradually shifts from SL-dominated to QW-dominated with the increase of applied negative bias. This result strongly demonstrates the capability of our standard QWIP to switch between broadband mode (via SL) and narrowband mode (via QW) by simply adjusting the operating bias voltages.
Surface relief structures, which are grating patterns, were then adopted onto our standard QWIP device to generate different diffraction modes of light to enhance the absorption of normal incident light on the sample [37,38]. QWIP devices with grating structure, specifically called QGIP, are expected to exhibit optimal and improved performance, as can be seen in our previous work [39]. The schematic of the device’s grating structure with pertinent notations is shown in the inset of Fig. 3(a). Based on our tests, we had concluded that the optimal structure has a period of P = 3 µm, a duty cycle (ridge L/period P) of 50% and a depth of H = 3.3 µm. The responsivity of the device having such a structure is illustrated by a red-circle line in Fig. 3(a), which also features an additional sample possessing a grating structure of same period and duty cycle, but with a depth of H = 2 µm (green-triangle line). In contrast to our standard QWIP device (denoted by Ref), an apparent increase (1.6 times) in responsivity under a negative bias, or broadband condition, was observed in the former (with H = 3.3 µm) while the latter (with H = 2 µm) showed no enhancement due to the absence of a concentrating light onto the active regions. Another benefit of the QGIP design is that its responsivity is resonantly enhanced without generating extra levels of dark current. Figure 3(b) shows the dark current magnitude of our standard QWIP along with the one containing the optimal grating structure, which will simply be called QGIP from this point forward. The trend of dark current under positive bias (increasing) shows a contrasting result from that under negative bias (decreasing), which may be attributed to the asymmetrical junction structure of the SL and QW regions [17]. Moreover, due to partial removal of the active region and the increase of the charge depletion region (grating side walls) [40], the dark current level of QGIP devices decreased. The increase of responsivity (about 1.6 times) and decrease of dark current (about 6 times) show that the grating structure concentrates light energy in the active region effectively. To sum up, the grating structure not only endows the device with the capability of enabling normal incidence absorption, but it also improves the concentration condition of the absorbed photoenergy given suitable structure parameters.
Fig. 3. (a) Responsivity of the devices having various grating structures with and without overlaying graphene. (b) The measured dark current in each of the three relevant devices, namely the graphene-covered QGIP (blue-square line), the QGIP (red-circle line), and a standard QWIP (Reference, black-triangle line). All measurements are done at 60 K temperature. (c) The detectivities of the three relevant devices in (b). (d) Comparison of highest operating temperatures among graphene-covered QGIP, 25-QW graphene-covered QGIP (red-circle line), and the standard QWIP.
In general, photoelectrons are generated at the graphene/semiconductor interface while the junction is irradiated at a specific wavelength [23]. Therefore, we considered introducing graphene into our device to further improve its performance. A single layer of graphene was transferred onto the top surface of the completed QGIP via PMMA transfer method [31]. Going back to Fig. 3(b), it can be observed that the measured dark current in graphene-covered QGIP decreases to a lower level compared to that of the bare QGIP and standard QWIP devices. Meanwhile, in Fig. 3(a), the responsivity (R) of the graphene-covered QGIP shows up to 7-fold enhancement at negative bias condition relative to the standard QWIP. Additionally, the specific detectivity D* of the devices of interest, which is defined below, has also been derived and calculated [8].
In this formula, Ad denotes the area of the device, Id represents the dark current, and g signifies the conductive gain of the quantum well active layers, which is equal to 1/pcNw (Nw is the number of active layers significant for the peak responsivity). In this study, we use Nw = 15 for bias < −0.9 V due to the domination of the SL; and Nw = 5, equivalent to the number of quantum well layers, for bias > −0.9 V. The parameter pc represents the electron capture probability, which is assumed to be equal to 0.1 in this case [41]. Figure 3(c) shows the specific detectivity of the graphene-covered QGIP, bare QGIP, and standard QWIP at 60 K. Graphene-covered QGIP demonstrated an improved D* in the negative bias region, reaching up to 20 times amplification relative to standard QWIP for an applied bias greater than −1.1 V. Since the spectrum signals of the quantum wells and SLs are too weak to be detected under bias voltages ranging from −0.4 V to + 0.2 V, D* within this range cannot be presented in the plot. The highest operating temperature of the device is increased as well. Figure 3(d) displays the responsivities of graphene-covered QGIP, 25-QW graphene-covered QGIP, and standard QWIP as a function of the operating temperature at −1.2 V. It shows that the highest operating temperature for the responsivity of the standard QWIP device to be detected is about 80 K, while that of the graphene-covered QGIP can reach up to 110 K. When the temperature gets higher, the dark current becomes dominated by thermally excited carriers, which greatly interfere with the detection of optical signals. In graphene-covered QWIP, photocurrents generated by the vertical Schottky barrier heterostructure between graphene and GaAs rises with higher temperatures. This satisfies the classical Richardson-Dushman law. [42,43] Thus, signal to noise ratio enhances, increasing the maximum operating temperature of QWIP. Further improvements of the operating temperature can be achieved by using 25 layers of QW instead of just 5. This change has successfully extended the device’s operating temperature to 140 K, although its responsivity and specific detectivity both dropped in this case.Compared to other reports, the responsivity, specific detectivity and operating temperature enhancements achieved in this work are remarkable (Table 1). Hence, the mechanism is worthy of being discussed in detail.
Table 1. Comparison of QWIP’s improvement methods between various works
The mechanism behind the enhancements in the negative bias region is due to improved induced injection of photo-excited holes from the GaAs layer into graphene [26]. Detail of the proposed mechanism is described by the energy band diagram of the GaAs-graphene heterojunction under IR illumination as shown in Fig. 4. This concept is partially referred to from [24].
Fig. 4. Energy band diagram of the graphene-GaAs heterojunction under illumination (a) at zero bias, (b) negative bias, and (c) positive bias. (d) Energy band diagram of the whole QWIP structure and its charge transitions.
Figure 4(a) illustrates the original energy band structure of the heterojunction with no external bias applied. In general, p-doped graphene is commonly formed in graphene/GaAs heterostructures [28,47,48]. The interface between GaAs and graphene forms a Schottky barrier (eΦSBH), which induces energy band bending. Illuminating the junction with IR light forms a quasi-fermi level E’f (Gr) and induces photo-excited holes to be injected into the valence band (red area) of graphene. With an external bias applied, the position of the quasi-fermi level varies, affecting the population of the injected holes. Based on a Ref. [49], the relation between chemical potential shifting in a graphene layer and bias voltage can be well explained by studying the Schottky barrier height between graphene and GaAs. With an applied reverse (negative) bias, Schottky barrier height decreases causing an increase in injected carriers. In our case, E’f (Gr) lowers down with applied negative bias. This results in an increase in the accessible states for the injected photo-excited holes, as shown in Fig. 4(b). In a positive bias condition, on the other hand, E’f (Gr) raises from its unbiased position reducing the accessible states for hole-injection, as shown in Fig. 4(c). These additionally generated photoelectrons, especially in a negative bias condition, could be involved in the processes within the SL and QW regions, which may enhance the possibility of allowing excited photoelectrons to escape from the wells. Figure 4(d) shows the band diagram of a QWIP structure with graphene. Herein, photoelectrons flow through the device via an applied negative bias with respect to the bottom contact. Extra-generated photoelectrons from the GaAs-graphene interface are trapped in the well as it flows through the SL region. These photoelectrons fill the ground state of the well, increasing the excitation possibility. Thus, enhancing the photocurrent. The electrons excited from the wells of the SL traverse across the barriers and are recaptured immediately by the first well of the MQW due to high electron-capture probability of the quantum well at low bias voltage. This generates a capture current in the process. To maintain charge neutrality, an equal number of electrons are emitted from the well to compensate the number of electrons that are trapped in the well, forming a corresponding emission current [50]. This trap-and-excite mechanism keeps happening as the electrons travel through an entire series of wells in the MQW region, successfully preserving the extra-generated photocurrent contributed by the GaAs-graphene interface. Therefore, the output photocurrent and responsivity, which directly correlates with the photocurrent, significantly improved. Further improvement of the device’s performance is possible by adopting few-layer graphene, which provides a higher optical absorption rate and better electrical contact [51].
4. Summary
In conclusion, we integrated double active layers (SL/QW), surface-relief grating structure, and GaAs-graphene interface to fabricate a performance-improved graphene-covered multi-color QGIP. The detection is band-broadened and voltage-tunable with an absorption spectrum ranging from 5 µm to 11 µm. Overall responsivity is improved by about 7-fold, and the specific detectivity enhances up to 20 times relative to a reference standard QWIP sample. Additionally, the operating temperature greatly increased from 80 K to 110 K by increasing the number of QW active layers in the device. This method provides an effective way to amplify signal strength and to augment the applicability of the device. We believe that our method could be a potential candidate for high response thermal imaging sensor improvements in the future.
Funding
Ministry of Science and Technology, Taiwan (MOST-105-2221-E-002-185-MY3).
Acknowledgments
The authors thank Prof. Chien-Ping Lee from National Chiao Tung University and his laboratory for the technique support during the measurement process.
Disclosures
The authors declare no conflicts of interest.
References
1. J. Homola, “Surface Plasmon Resonance Sensors for Detection of Chemical and Biological Species,” Chem. Rev. 108(2), 462–493 (2008). [CrossRef]
2. J. N. Anker, W. P. Hall, O. Lyandres, N. C. Shah, J. Zhao, and R. P. Van Duyne, “Biosensing with plasmonic nanosensors,” in Nanoscience and Technology: A Collection of Reviews from Nature Journals, W. Jason, ed. (World Science, 2009).
3. K. A. Willets and R. P. V. Duyne, “Localized Surface Plasmon Resonance Spectroscopy and Sensing,” Annu. Rev. Phys. Chem. 58(1), 267–297 (2007). [CrossRef]
4. Y. S. Lee, Principles of terahertz science and technology (Springer Science & Business Media, 2009), Vol. 170.
5. B. Ferguson and X.-C. Zhang, “Materials for terahertz science and technology,” Nat. Mater. 1(1), 26–33 (2002). [CrossRef]
6. S. D. Gunapala, S. V. Bandara, J. K. Liu, J. M. Mumolo, S. B. Rafol, D. Z. Ting, A. Soibel, and C. Hill, “Quantum Well Infrared Photodetector Technology and Applications,” IEEE J. Sel. Top. Quantum Electron. 20(6), 154–165 (2014). [CrossRef]
7. A. Reisinger, R. Dennis, K. Patnaude, D. Burrows, J. Bundas, K. Beech, R. Faska, and M. Sundaram, “Broadband QWIP FPAs for hyperspectral applications,” Infrared Phys. Technol. 59, 112–117 (2013). [CrossRef]
8. H. Schneider and H. C. Liu, Quantum Well Infrared Photodetectors (Springer, 2007).
9. K. K. Choi, S. Allen, J. G. Sun, and E. A. DeCuir, “Resonant detectors and focal plane arrays for infrared detection,” Infrared Phys. Technol. 84, 94–101 (2017). [CrossRef]
10. K. K. Choi, J. G. Sun, K. A. Olver, and R. Fu, “Parameter study of resonator-quantum well infrared photodetectors,” IEEE J. Quantum Electron. 53(5), 1–10 (2017). [CrossRef]
11. C. J. Chen, K. K. Choi, M. Z. Tidrow, and D. C. Tsui, “Corrugated quantum well infrared photodetectors for normal incident light coupling,” Appl. Phys. Lett. 68(11), 1446–1448 (1996). [CrossRef]
12. K. K. Choi, D. P. Forrai, D. W. Endres, and J. G. Sun, “Corrugated quantum-well infrared photodetector focal plane arrays,” IEEE J. Quantum Electron. 45(10), 1255–1264 (2009). [CrossRef]
13. S. Kalchmair, R. Gansch, S. I. Ahn, A. M. Andrews, H. Detz, T. Zederbauer, E. Mujagić, P. Reininger, G. Lasser, W. Schrenk, and G. Strasser, “Detectivity enhancement in quantum well infrared photodetectors utilizing a photonic crystal slab resonator,” Opt. Express 20(5), 5622–5628 (2012). [CrossRef]
14. W. Wu, A. Bonakdar, R. Gelfand, and H. Mohseni, “A normal-incident quantum well infrared photodetector enhanced by surface plasmon resonance,” in Detectors and Imaging Devices: Infrared, Focal Plane, Single Photon (International Society for Optics and Photonics, 2010), pp. 77801A.
15. W. C. Hsu, H. S. Ling, S. Y. Wang, and C. P. Lee, “Dependence of Performance of Surface Plasmon Coupled Quantum Well Infrared Photodetectors on Doping,” J. Nanosci. Nanotechnol. 18(11), 7838–7842 (2018). [CrossRef]
16. W. C. Hsu, H. S. Ling, S. Y. Wang, and C. P. Lee, “Characteristics of surface plasmon coupled quantum well infrared photodetectors,” J. Appl. Phys. 121(24), 244503 (2017). [CrossRef]
17. S. H. Lin, Y. H. Wang, C. W. Chang, J. H. Lu, C. C. Chen, and C. H. Kuan, “Development of superlattice infrared photodetectors,” in Cutting Edge Nanotechnology (IntechOpen, 2010), pp. 113–136.
18. C. C. Chen, H. C. Chen, C. H. Kuan, S. D. Lin, and C. P. Lee, “Multicolor infrared detection realized with two distinct superlattices separated by a blocking barrier,” Appl. Phys. Lett. 80(13), 2251–2253 (2002). [CrossRef]
19. M. C. Hsu, C. H. Kuan, and S. Y. Wang, “Multicolor infrared detection using two stacks of superlattice structures in a back-to-back configuration,” Appl. Phys. Lett. 77(14), 2240–2242 (2000). [CrossRef]
20. J. Mao, A. Majumdar, K. K. Choi, D. C. Tsui, K. M. Leung, C. H. Lin, T. Tamir, and G. A. Vawter, “Light coupling mechanism of quantum grid infrared photodetectors,” Appl. Phys. Lett. 80(5), 868–870 (2002). [CrossRef]
21. K. K. Choi, G. Dang, J. W. Little, K. M. Leung, and T. Tamir, “Quantum grid infrared spectrometer,” Appl. Phys. Lett. 84(22), 4439–4441 (2004). [CrossRef]
22. X. An, F. Liu, Y. J. Jung, and S. Kar, “Tunable Graphene–Silicon Heterojunctions for Ultrasensitive Photodetection,” Nano Lett. 13(3), 909–916 (2013). [CrossRef]
23. X. Li, W. Chen, S. Zhang, Z. Wu, P. Wang, Z. Xu, H. Chen, W. Yin, H. Zhong, and S. Lin, “18.5% efficient graphene/GaAs van der Waals heterostructure solar cell,” Nano Energy 16, 310–319 (2015). [CrossRef]
24. W. Jie, F. Zheng, and J. Hao, “Graphene/gallium arsenide-based Schottky junction solar cells,” Appl. Phys. Lett. 103(23), 233111 (2013). [CrossRef]
25. S. Tongay, T. Schumann, and A. F. Hebard, “Graphite based Schottky diodes formed on Si, GaAs, and 4H-SiC substrates,” Appl. Phys. Lett. 95(22), 222103 (2009). [CrossRef]
26. J. Xu, J. Hu, R. Wang, Q. Li, W. Li, Y. Guo, F. Liu, Z. Ullah, L. Wen, and L. Liu, “Ultra-broadband graphene-InSb heterojunction photodetector,” Appl. Phys. Lett. 111(5), 051106 (2017). [CrossRef]
27. K. F. Mak, L. Ju, F. Wang, and T. F. Heinz, “Optical spectroscopy of graphene: From the far infrared to the ultraviolet,” Solid State Commun. 152(15), 1341–1349 (2012). [CrossRef]
28. L. B. Luo, H. Hu, X. H. Wang, R. Lu, Y. F. Zou, Y. Q. Yu, and F. X. Liang, “A graphene/GaAs near-infrared photodetector enabled by interfacial passivation with fast response and high sensitivity,” J. Mater. Chem. C 3(18), 4723–4728 (2015). [CrossRef]
29. J. Yuan, W. Jiang, W. Xu, J. Yang, S. Gong, S. Imran, L. Xie, Y. Ying, and Y. Ma, “Modulation of far-infrared light transmission by graphene-silicon Schottky junction,” Opt. Mater. Express 6(12), 3908–3915 (2016). [CrossRef]
30. S. Cakmakyapan, P. K. Lu, A. Navabi, and M. Jarrahi, “Gold-patched graphene nano-stripes for high-responsivity and ultrafast photodetection from the visible to infrared regime,” Light: Sci. Appl. 7(1), 20 (2018). [CrossRef]
31. X. Li, Y. Zhu, W. Cai, M. Borysiak, B. Han, D. Chen, R. D. Piner, L. Colombo, and R. S. Ruoff, “Transfer of large-area graphene films for high-performance transparent conductive electrodes,” Nano Lett. 9(12), 4359–4363 (2009). [CrossRef]
32. R. Paiella, F. Capasso, C. Gmachl, C. G. Bethea, D. L. Sivco, J. N. Baillargeon, A. L. Hutchinson, A. Y. Cho, and H. Liu, “Generation and detection of high-speed pulses of mid-infrared radiation with intersubband semiconductor lasers and detectors,” IEEE Photonics Technol. Lett. 12(7), 780–782 (2000). [CrossRef]
33. H. Liu, R. Dudek, A. Shen, E. Dupont, C.-Y. Song, Z. Wasilewski, and M. Buchanan, “QWIPs designed for high absorption and high operating temperature,” Int. J. High Speed Electron. Syst. 12(03), 803–819 (2002). [CrossRef]
34. B. F. Levine, “Quantum-well infrared photodetectors,” J. Appl. Phys. 74(8), R1–R81 (1993). [CrossRef]
35. R. D. Hudson, Infrared system engineering (Wiley-Interscience New York, 1969).
36. E. L. Dereniak and D. G. Crowe, Optical radiation detectors (John Wiley & Sons Inc., 1984).
37. J. Y. Andersson and L. Lundqvist, “Grating-coupled quantum-well infrared detectors: Theory and performance,” J. Appl. Phys. 71(7), 3600–3610 (1992). [CrossRef]
38. M. Z. Tidrow, K. K. Choi, A. J. DeAnni, W. H. Chang, and S. P. Svensson, “Grating coupled multicolor quantum well infrared photodetectors,” Appl. Phys. Lett. 67(13), 1800–1802 (1995). [CrossRef]
39. M. L. Lee, C. J. Hsieh, Y. H. You, V. C. Su, P. H. Chen, H. C. Lin, H. B. Yang, H. M. Chen, and C. H. Kuan, “Performance enhancement in quantum well infrared photodetector utilizing the grating structure,” in CLEO: Science and Innovations, OSA Technical Digest Series (Optical Society of America, 2013), paper CM3F.7.
40. D. C. Look, C. E. Stutz, and K. R. Evans, “Surface and interface free-carrier depletion in GaAs molecular beam epitaxial layers: Demonstration of high interface charge,” Appl. Phys. Lett. 56(7), 668–670 (1990). [CrossRef]
41. C. Schönbein, H. Schneider, R. Rehm, and M. Walther, “Noise gain and detectivity of n-type GaAs/AlAs/AlGaAs quantum well infrared photodetectors,” Appl. Phys. Lett. 73(9), 1251–1253 (1998). [CrossRef]
42. Y. S. Ang, H. Y. Yang, and L. K. Ang, “Universal Scaling Laws in Schottky Heterostructures Based on Two-Dimensional Materials,” Phys. Rev. Lett. 121(5), 056802 (2018). [CrossRef]
43. M. Trushin, “Theory of thermionic emission from a two-dimensional conductor and its application to a graphene-semiconductor Schottky junction,” Appl. Phys. Lett. 112(17), 171109 (2018). [CrossRef]
44. W. C. Hsu, H. S. Ling, S. Y. Wang, and C. P. Lee, “Wavelength tuning of surface plasmon coupled quantum well infrared photodetectors,” Opt. Express 26(1), 552–558 (2018). [CrossRef]
45. W. Wu, A. Bonakdar, and H. Mohseni, “Plasmonic enhanced quantum well infrared photodetector with high detectivity,” Appl. Phys. Lett. 96(16), 161107 (2010). [CrossRef]
46. D. Shao, X. Guo, Y. Zhu, S. Zhuang, Z. Fu, and J. Cao, “Surface-phonon-polariton-mediated photon response of terahertz quantum-well infrared photodetectors,” J. Phys. D: Appl. Phys. 52(3), 035105 (2019). [CrossRef]
47. A. Gamucci, D. Spirito, M. Carrega, B. Karmakar, A. Lombardo, M. Bruna, L. N. Pfeiffer, K. W. West, A. C. Ferrari, M. Polini, and V. Pellegrini, “Anomalous low-temperature Coulomb drag in graphene-GaAs heterostructures,” Nat. Commun. 5(1), 5824 (2014). [CrossRef]
48. W. Chen, X. Li, W. Yin, S. Lin, Z. Zhao, E. Li, and H. Zhou, “Modeling and simulation of graphene-gated graphene-GaAs Schottky junction field-effect solar cell for its performance enhancement,” IEEE Trans. Electron Devices 62(11), 3760–3766 (2015). [CrossRef]
49. S. Tongay, M. Lemaitre, X. Miao, B. Gila, B. R. Appleton, and A. F. Hebard, “Rectification at graphene-semiconductor interfaces: zero-gap semiconductor-based diodes,” Phys. Rev. X 2(1), 011002 (2012). [CrossRef]
50. M. Ershov, C. Hamaguchi, and V. Ryzhii, “Device physics and modeling of multiple quantum well infrared photodetectors,” Jpn. J. Appl. Phys. 35(Part 1, No. 2B), 1395–1400 (1996). [CrossRef]
51. A. S. Aji, P. Solís-Fernández, H. G. Ji, K. Fukuda, and H. Ago, “High mobility WS2 transistors realized by multilayer graphene electrodes and application to high responsivity flexible photodetectors,” Adv. Funct. Mater. 27(47), 1703448 (2017). [CrossRef]
