2D Materials in Infrared Detector Family †
Total Page:16
File Type:pdf, Size:1020Kb
Proceedings 2D Materials in Infrared Detector Family † Antoni Rogalski Institute of Applied Physics, Military University of Technology, 2 Kaliskiego Str., 00-908 Warsaw, Poland; [email protected] † Presented at the 15th International Workshop on Advanced Infrared Technology and Applications (AITA 2019), Florence, Italy, 17–19 September 2019. Published: 26 September 2019 Abstract: The paper compares two-dimensional (2D) material detectors performance with traditionally and commercially available ones operating in high temperature conditions. The most effective single graphene detectors are THz detectors which utilize plasma rectification phenomena in FETs. Most of 2D layered semiconducting material photodetectors operate at the visible and near-infrared regions and generally their high sensitivity does not coincide with a fast response time, which limits real detector functions. Keywords: 2D materials; infrared and terahertz detectors; detector performance 1. Introduction This paper gathers performance data of different types of single 2D material-based infrared (IR) and terahertz (THz) detectors. Their performance is compared with existing mature material systems including HgCdTe, InGaAs, III-V compounds and microbolometers. The intent is to concentrate on device architectures that can have the most impact on future position in wide high operating temperature (HOT) IR and THz detector family. 2. Graphene-Based Detectors Graphene detectors can be divided into two categories: either photon and thermal (the bolometer effect) detectors. The majority of graphene pristine photodetectors exploit graphene metal junctions, graphene p-n junctions, and FET transistors. The most recent approach also employs the photo-thermoelectric (PTE) effect (Seebeck effect) to create an electric field due to electron diffusion into metal contacts. The photocurrent direction due to both PV and PTE effects are the same. The development of the graphene high responsivity photodetectors is determined by the two major challenges: the low optical absorption in the detector’s active region (~100–200 nm) and short photocarrier lifetime meaning that graphene photodetectors are mainly limited by trade-off between high responsivity, ultrafast time response and broadband operation. Responsivity improvement in grapheme detectors can be reached by increasing the photocarrier lifetime through both band structure and defect engineering. Table 1 shows schematically the differences between pure graphene photodetector and hybrid photodetector [1]. Additionally, the distinction between ultrafast and ultrasensitive graphene photodetectors is presented. The main feature of the hybrid photodetector is ultrahigh gain originating from the high carrier mobility of the graphene sheet and the recirculation of charge during the lifetime of the carriers remaining trapped in light absorbing region. Multiple circulation of holes in the graphene channel following a single e-h photogeneration leads to strong photoconductive gain. Proceedings 2019, 27, 33; doi: 10.3390/proceedings2019027033 www.mdpi.com/journal/proceedings Proceedings 2019, 27, 33 2 of 4 Table 1. Ultrafast and ultrasensitive graphene photodetectors. Structure Band Structure Recombination Mechanism Diffusion Ultra-fast (pristine) Drift Silicon Ec Drain Source (Ti) (Pd) EF + + + SiO 2 VGate Drift Ev Recombination (picosecond Diffusion Ultra-sensitive (hybrid) Electron trapping Hole Light extraction τ (2) lifetime V DS Source (1) Quantum dots Hole Drain injection + + τ (3)(2)(3) + Hole drift transit VG ττ te lifetime» transit Ga (1) Light absorption in heteromedium (2) Trapping electrons (3) Injection of holes in graphene Hybrid photodetectors offer improvements in responsivity, however the majority of these devices have a limited linear dynamic range due to the charge relaxation time, which quickly saturates the available states for photoexcitation, leading to a drop in responsivity with incident optical power. 3. Graphene-Based Detector Performance—The Present Status The spectral responsivity of graphene photodetectors operating in visible and NIR spectral ranges are compared with commercially available silicon and InGaAs photodiodes in Figure 1. As shown, graphene’s high mobility together with the trapped charge lifetimes in the quantum dots produced a photodetector responsivity of 107 A/W. However, due to the long lifetime of the traps, the demonstrated frequency response is very slow (<10 Hz). 109 108 Konstantatos et al., Nature Commun. 7, 363 (2012) Hybrid: smal QD phototransistors Chang et al., Sci. Rep. 7, 46281 (2017) 7 Hybrid: MAPbI perovskite phototransistors 10 3 106 Konstantatos et al., Nature Commun. 7, 363 (2012) Hybrid: large QD phototransistors 105 Red color - ≤ 1-ns response times ) Green color - ≤ 1-ns response times 104 Blue color - ≥ 1-sec response times A/W ( Adv. Mater points - Gr/FGr detector; Du et al., . 29, y t 1700463 (2017); ~200 ms response time 103 sivi n o 102 Chen et al., ACS Nano 11, 430 (2017) esp Hybrid + plasmonic R Ferreira et al., Phys. Rev. B 85, 115438 (2012) 1 two cavities 10 Si APD InGaAs PD (Hamamatsu) (Hamamatsu) η=100% 0 10 Zhang et al., Nature Commun. 4, 1811 (2013) With electron trapping centres 10-1 Si PD Gan et al., Nature Photonics 7, 883 (2013) (Hamamatsu) Waveguide integrated Furchi et al., Nano Letters 12, 2773 (2012) Microcavity-integrated 10-2 0.5 1.0 1.5 2.0 2.53.0 3.5 4.0 Wavelength (μ m) Figure 1. Spectral responsivity of graphene photodetectors compared with commercial ones. Literature data for longer wavelength infrared graphene-based photodetectors, with cut-off wavelength above 3 μm, are limited. Figure 2 compares their detectivities with commercially available HgCdTe photodiodes and type-II InAs/GaSb superlattice interband quantum cascade infrared photodetectors (IB QCIPs) operating at room temperature. The upper detectivity of Gr/FGr photodetector in SWIR range is close to HgCdTe photodiodes. This figure also shows experimental data for black phosphorus arsenic (bAsP) photodetectors entering the second atmospheric transmission window. However, instability of the surface of bAsP, due to chemical degradation in ambient conditions, remains a major impediment to its prospective applications. Proceedings 2019, 27, 33 3 of 4 1011 300 K bPAs: Amani et al., ACS Nano 11, 11724 (2017) 2π FOV 300 K background HgCdTe photovoltaic detectors (VIGO System) 1010 ) IB QCIP with T2SL InAs/GaSb absorbers -1 2 1/ 9 (cmHz10 W y t i v i t ec t De 108 Fluorographene - Du et al ., Adv. Mater. 29, 1700463 (2017) bAs0.83 P 0.17 - Long et al., Sci. Adv. 3, e1700589 (2017) bAs0.83 P 0.17 /MoS 2 - Long et al., Sci. Adv. 3, e1700589 (2017) 107 1 2 34567 8 91011 Wavelength (μ m) Figure 2. Comparison of spectral detectivity graphene-based detectors with commercial detectors. Particular attention in development of THz imaging systems is devoted to the fabrication of sensors with a large potential for real-time imaging while maintaining room temperature operation. Much of recent development is directed towards three types of focal plane arrays (FPAs): Schottky barrier diodes (SBDs) compatible with the CMOS process, FETs relying on plasmonic rectification effect, and adaptation of IR bolometers to the THz frequency range. At present stage of technology, the most effective graphene THz detectors employ plasma rectification effect in FETs, where plasma wave in the channel are excited by incoming THz wave modulating the potential difference between gate and source/drain and being rectified via non-linear coupling and transfer characteristics in FET. Next two figures compare NEP values of graphene-based FET room temperature detectors with existing THz photon detectors (Figure 3a) and thermal detectors (Figure 3b) dominating on the market. A lot of experimental data are gathered from world-wide literature. Frequency (THz) Frequency (THz) 1 0.1 100 10 100 10 1 0.1 10000 10000 bP FET, Vitti et al., APL Mater. 5, 035602 (2017) Graphene FETs Graphene FETs bP FET, Vitti et al., APL Mater. 5, 035602 (2017) Spirito et al., APL 104, 061111 (2014) Spirito et al., APL 104, 061111 (2014) Yang et al., APL 111, 021102 (2017) Yang et al., APL 111, 021102 (2017) Zak et al., Nano Lett. 14, 5834 (2014) Zak et al., Nano Lett. 14, 5834 (2014) Generalov et al., IEEE TST 7, 614 (2017) 1000 1000 Generalov et al., IEEE TST 7, 614 (2017) Tong et al., Nano Lett. 15, 5295 (2015) Tong et al., Nano Lett. 15, 5295 (2015) Bianko et al., APL 107, 131104 (2015) Calculation Liu et al., NPG Asia Mater. 10, 318 (2018) Bianko et al., APL 107, 131104 (2015) ) Liu et al., Adv. Opt. Mater. 1800836 (2018) Liu et al., NPG Asia Mater. 10, 318 (2018) ) 2 / 2 / 1 1 Liu et al., Adv. Opt. Mater.1800836 (2018) z 100 z 100 Schottky diodes CMOS-based and plasma detectors WR1.0 ZBD ErAs/InGaAlAs (pW/HP CMOS, 150nm, NMOS, single element GaAs (VDI) E NEP (pW/H NEP 10 InGaAs HEMT asymmetric dual-grating gate N 10 CMOS SiGe, 65 nm NMOS, 3x5 array, chopper 1 kHz WR1.5 ZBD CMOS SiGe, 65 nm NMOS, 32x32 array, chopper 5 kHz Microbolometers InGaAs HEMT, single element WR2.8 ZBD μ + TV/4 (320x240), 49.5- m pitch, VOx bolometer, NEC TV/4 (320x240), 23.5-μ m pitch, a-Si nolometer, Leti 1.0 1.0 (320x240), 50-μ m pitch, a-Si bolometer, Leti Vanadium oxide bolometer array Vanadium oxide bolometer array μ + (160x120), 23.5- m pitch, VOx bolometer, INO μ (384x288), 35- m pitch, VOx bolometer, INO 0.1 0.1 1 10 100 1000 10000 1 10 100 1000 10000 Wavelength (μ m) Wavelength (μ m) (a) (b) Figure 3. NEP spectral dependence for graphene FET detectors and different terahertz photon (CMOS-based, Schottky diodes) (a) and microbolometers (b) FPAs. Most experimental data gathered in literature is given for single graphene detectors operating above wavelength of 100 μm. Generally, the performance of graphene FET detectors is lower in comparison with CMOS based and plasma detectors fabricated using silicon, SiGe and InGaAs based materials. In comparison with VOx and amorphous silicon microbolometers, the performance of graphene detectors is close to the trend line estimated for microbolometers in THz spectral region (see Figure 3b).