Infrared photodetector based on photo-thermionic effect of graphene

Subscriber access provided by The University of Melbourne Libraries

Functional Nanostructured Materials (including low-D carbon)

Infrared photodetector based on photo-thermionic effect of graphene-nanowalls/silicon heterojunction Xiangzhi Liu, Quan Zhou, Shi Luo, Haiwei Du, Zhensong Cao, Xiaoyu Peng, Wenlin Feng, Jun Shen, and Dapeng Wei ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b03329 • Publication Date (Web): 22 Apr 2019 Downloaded from http://pubs.acs.org on April 22, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Infrared Photodetector Based on PhotoThermionic Effect of GrapheneNanowalls/Silicon Heterojunction Xiangzhi Liu,† ¶ ‡ Quan Zhou,† § ‡ Shi Luo,† Haiwei Du,† Zhensong Cao,ǁ Xiaoyu Peng,† Wenlin Feng,¶ Jun Shen,† * Dapeng Wei,† *

†Chongqing Key Laboratory of Multi-scale Manufacturing Technology, Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing, 400714, P. R. China. ¶Department of Applied Physics, Chongqing University of Technology, Chongqing, 400054, P. R. China. §China Aerodynamics Research and Development Center, Mianyang, 621000, P. R. China. ǁKey Laboratory of Atmospheric Optics, Chinese Academy of Sciences, Hefei, 230031, P. R. China. KEYWORDS: Graphene, Nanoparticle, Photodetector, Hot electron, Schottky barrier, Thermionic emission

ABSTRACT: Due to the slow relaxation process according to weak acoustic phonon interaction, photo-thermionic effect in graphene could be much more obvious than

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

metal film, so graphene heterojunction photodetector based on photo-thermionic effect is promising for infrared imaging applications. However, the 2.3% absorption rate of graphene film presents a severe limitation. Here, in-situ grown graphene nanowalls (GNWs) were integrated on silicon substrate interfaced with Au nanoparticles. Due to the strong infrared absorption and hot-carrier relaxation process in GNWs, as-prepared GNWs/Au/silicon heterojunction has a photo to dark ratio of 2×104, responsivity of 138 mA/W, and linear dynamic range of 89.7 dB, with specific detectivity of 1.4×1010 cm Hz1/2/W and 1.6×109 cm Hz1/2/W based on calculated and measured noise respectively in 1550 nm at room temperature, which has best performance among silicon-compatible infrared photodetectors without any complicated waveguide structures. Obvious photoresponse are also detected in the mid-infrared and terahertz band. The interface Au particle are found to reduce the barrier height and enhance absorption. The device structure in this report could be compatible with semiconductor process, so infrared photodetectors with high integration density and low cost could be potentially realized.

1. INTRODUCTION Infrared detectors have been widely applied in military fields,1 but they are hard to use widely in the civilian market due to their high price. The reason for this high price is that the material preparation cost is high and materials are not compatible with standard silicon (Si) -based fabrication processes because of cross-contamination and the high lattice mismatch.2 Although uncooled detectors based on bolometer which mostly work in the far-infrared are compatible with Si-based processes, it is generally


Page 2 of 25

Page 3 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


difficult to cover short-wave and mid-wave infrared.3 In the other hand, metal-silicon detectors based on photo-thermionic effect (PTE) are well compatible with silicon processes, such as platinum-silicon infrared focal plane array (FPA) detectors, which can partially integrate information processing circuits and detector elements on silicon. Compared with other compound detection, it is the only short-wave infrared detector that can be completed by the all-silicon manufacturing process. In 1993, the 1040×1040 charge scanning device (CSD) infrared FPA was developed by Mitsubishi Corporation of Japan. It is the first silicide array that can reach megapixels, with the highest spatial resolution and minimum image.4 However, due to the low hot carrier emission efficiency of the metal material and the momentum mismatch between the electron states in the metal and Si of silicide Schottky barrier photodetector, the responsivity (R) of these silicon-compatible devices is limited to less than 0.1 A/W.5,6 Compared with metal materials, graphene (Gr) has longer hot carrier relaxation time and higher hot carrier temperature, which can achieve higher hot carrier emission efficiency.7-9 At present, research on hot carrier emission of graphene materials is one of the focuses about graphene photodetectors.10-14 The high thermal carrier emission efficiency of graphene materials, combined with its infrared broad spectral absorption characteristics,15,16 and good compatibility with silicon-based processes,17 made graphene-based Schottky heterojunction promising for new infrared detectors.18 However, the low absorption of the two-dimensional graphene film is always limited to 2.3% caused by its monoatomic layer structure.19 As a result, the quantum efficiency of graphene photodetectors is low, and the advantages of high thermal carrier emission



efficiency of graphene materials cannot be utilized. One way to increase the light absorption of graphene optoelectronic devices is to use quantum dot recombination,20 optical waveguide integration,21 microresonator enhancement,22 surface plasmon,23 etc. To a certain extent, the quantum efficiency of the device is improved, but the absorption band is determined by materials such as quantum dots and waveguides, which loses the intrinsic advantage of the broad spectral photoelectric response of graphene materials.19 Using multiple layers of graphene instead of a single layer as the electrode of Schottky barrier photodetector could provide a possible solution. According to Y. Chen et al., the absorption coefficient of graphene is as high as 6.8×107 m-1,24 which is much higher than that of metallic materials, such as Pt is 2.2×104 m-1.25 Increasing the number of graphene layers can allow for more efficient absorption. Common way to achieve multilayer graphene is to use multiple transfer methods26or tape stripping methods.13 There are some intrinsic drawbacks of these methods. since the metal particle from the growth procedure and polymer residue from the transfer process, the interface between silicon and graphene is usually not ideal, and the quality of the Schottky barrier as well as the depletion area will be influenced in the bad way.27 By in-situ growth of highquality graphene nanowalls (GNWs) on a silicon substrate,28 the contamination problems caused by metal catalysts and photoresist residues can be effectively solved, which is beneficial to the formation of high-quality Schottky junction interfaces. GNWs are three-dimensional materials formed by the overlap of longitudinally grown graphene, with graphene Raman peaks and atomic structures,29 which can take advantage of the high efficiency of hot carrier emission of graphene materials. In


Page 4 of 25

Page 5 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


addition, heterojunction photodetectors based on PTE effect have been widely used in broadband detection and have good performance.30,31 In this report, we use directly grown GNWs as light absorbing material to form GNWs/Si Schottky junctions with irregular gold (Au) nanoparticles as interface (GNWs/Au/Si). The barrier height was reduced by using Au nanoparticles. lower barrier height is favorable for infrared thermionic emission. The Schottky junction formed by GNWs in-situ growth in this way reduces polymer residue during the transfer of graphene, improves the quality of the Schottky junction and passivates dangling bond on silicon surface. The special three-dimensional structure of GNWs solves the problem of weak absorption of graphene, and the irregular Au nanoparticles further enhance the absorption. Therefore, the fabricated device has a high photo to dark ratio of 2×104, responsivity is up to 138 mA/w and 0.44 μA/W at 1550 nm and 3.5 μm, respectively. Rise-time of 370 μs and fall-time of 510 μs were detected. Furthermore, an obvious signal was detected in the terahertz (THz) band. 2. EXPERIMENTAL SECTION 2.1. Growth of GNWs. The growth of GNWs was in 750 °C with RF power of 50 W at chamber pressure of 50 Pa. The gas flow rate for CH4 and H2 is 6:4. The thickness of GNWs depends on growth time. Unless otherwise mentioned, the growth time of the sample is 20 min and the effective area is 4.48 mm2 in this paper. The mentioned Au nanoparticles is annealed 2 nm Au film. 2.2. Device fabrication. The substrate was selected as lightly N-type doped silicon with



300 nm thickness SiO2 (The resistivity is 2-4 Ω • cm, and the thickness is 500 μm). These wafers were divided into 15*15 mm2. After etching a square window by HF solution, the wafer ultrasonically clean in acetone, alcohol and deionized (DI) water. To ensure that the contact between the Au and Si surface is clean, the silicon wafer soak with diluted HF (HF: H2O = 1:50) for about 3 minutes to remove the SiO2 formed by natural oxidation. After cleaning with isopropanol and drying by N2, the sample put into the magnetron sputtering chamber for sputtering 2 nm Au. The sample was then immediately placed into PECVD to grow GNWs. Excess GNWs and Au at the edge of the sample are removed by tape to prevent their connection with the silicon substrate from causing excessive leakage current. Silver paste was brushed around the window as upper electrode to contact GNWs, and Ga-In alloy was used as lower electrode on the back of the sample to keep ohm contact with silicon. 2.3. Photoresponse Measurement. The photoresponse characteristics were tested under the condition of light and dark by Keithley semiconductor analyzer 4200 and four probes. Power-tunable optical fiber coupling 1550 nm laser (TSL-1550) was illuminated to the sample to measure the photocurrent. To obtain 3.5 μm and THz test data, bias voltage output and data acquisition was applied by a Keithley 2450 Source Meter DC voltage source. LDC-3724 laser diode controller was used to control the output of 3.5 μm light. The device was fixed on the edge of a table to guarantee light and function area at the same horizontal height, then cover it with an opening box outside. The test environment of THz is under pulse energy of about 0.1 μJ, peak power of 0.5 mW, peak electric field of 100 kV/cm. The T-Ray 5000 system (Advanced


Page 6 of 25

Page 7 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Photonix, Inc. USA) was used for the THz system. The THz pulse is generated by a femtosecond pulsed laser beam pumping bias photoconductive switch with a center wavelength of 1064 nm. This femtosecond laser pulse has 80 fs duration and a 100 MHz repetition rate. The spectral bandwidth of the THz system is 0.1 to 10 THz, with a spectral resolution of 12.5 GHz, and its dynamic range can reach 70 dB. All the tests were conducted in darkness and at room temperature. In order to obtain the time response and the 3dB cut-off frequency of device, the device was illuminated by pulse light which is obtained by a chopper. 3. RESULTS AND DISCUSSION 3.1. Choosing of Device Structure Figure 1(a) shows the device fabrication process GNWs/Au/Si photodetector. GNWs was grown by plasma-enhanced chemical vapor deposition (PECVD), as previously reported.28,29 Irregular Au nanoparticles were obtained by sputtering a 2 nm

Figure 1. (a) The fabrication process of GNWs/Au/Si photodetector. (b) SEM image



of GNWs with irregular Au nanoparticles. gold film and annealing at the same time as the growth of GNWs. The surface feature of the device is displayed in Figure 1(b). The GNWs was distributed in some interstices and surfaces of Au nanoparticles. To ensure the advantage of GNWs/Au/Si, various kinds of device with GNWs were compared. Firstly, the Au film was deposited before and after GNWs growth for GNWs/Au/Si and Au/GNWs/Si device respectively. The I-V curves of three structures, GNWs/Si, GNWs/Au/Si, and Au/GNWs/Si were tested, respectively, as shown in Figure 2(a). It was observed that the rectification characteristics of this GNWs/Au/Si structure is the best. Moreover, the dark current under zero bias is the lowest. When the device is illuminated with the wavelength of 1550 nm, the photocurrent of the device is the largest. However, the dark current curve and the photocurrent curve of the other two structures are basically the same, which cannot be distinguished. This indicates that the net photocurrent generated is small. Secondly, heterojunctions based on graphene and GNWs were compared. The rectification and net photocurrent of Gr/Au/Si, GNWs/Au/Si and Au/Si were shown in Figure 2(b). In order to ensure the consistency of conditions as much as possible, the Gr/Au/Si and Au/Si device were prepared by sputtering 2 nm Au before the silicon wafer was placed in a PECVD chamber. Then, graphene was transferred after annealing at 750 °C for 20 min. The rectification characteristics of Gr/Au/Si are worse than GNWs/Au/Si, and the dark current is one order of magnitude larger than GNWs/Au/Si at zero bias. When the three devices are irradiated with the same power of 1550 nm, the GNWs/Au/Si photo-dark ratio is at least


Page 8 of 25

Page 9 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


two order of magnitudes larger than that of Gr/Au/Si, due to the three-dimensional special structure of GNWs which has a certain light trapping effect to enhance light absorption. However, no obvious photoresponse was observed in the Au/Si device. Thirdly, Figure 2(c) shows the I-V curve of three different metals as the interface layer. The thickness of Au, Ag, Ti is 2nm (Figure S1 for Au with different thickness). Under the illumination of 1550 nm, the photocurrent and the dark current of the device in the middle layer of Ag, Ti are coincident and the net photocurrent are very small. Conversely, the net photocurrent of GNWs/Au/Si can reach to 3.3 μA. Finally, the I-V curves of GNWs devices with different growth times were also tested. From Figure 2(d), GNWs has the best performance for 20 min growth. The device has the lowest dark current and the open circuit voltage is 0.08 V which is twice as large as than 10 min. It is well known that the longer the GNWs grows, the thicker the GNWs is. Although the thicker the GNWs attributed to the stronger the light absorption capacity, the dark current increases with growth time. Hence, the effective signals of 40 min devices is submerged in the noise.



Figure 2. (a) Current-voltage curve of three different structures (GNWs/Si, GNWs/Au/Si, Au/GNWs/Si). (b) The comparison between graphene and GNWs with the same structure. (c) Current-voltage curve of three metal (Au, Ag, Ti) with the same structure of GNWs/Metal/Si. (d) Current-voltage curve of the photodetector based on GNWs growing at different time (10 min - 40 min). All the current-voltage curves were measured under the illumination of 1550 nm laser beam. The power of laser is 4.5 mW. 3.2. Photoresponse of GNWs/Au/Si Photodetector For photoresponse characteristics, typical current–voltage curves under dark and light at the wavelength of 1550 nm with the power of 2.3 mw was described in Figure 3(a). Noted that the dark current of GNWs/Au/Si heterojunction with the area of 4.48 mm2 is 0.13 nA at the zero bias. As shown in Figure 3(b) the photo-to-dark current ratio is larger than four magnitudes at the zero bias. Figure 3(c) demonstrated the time response. The device can work at zero bias and reverse bias, but there is a lower dark


Page 10 of 25

Page 11 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. Photoresponse characteristics for the light at wavelength of 1550 nm and the power of 2.3 mW. (a) current–voltage curve. the dark current of GNWs/Au/Si is 0.13 nA at the zero bias, (b) Time-resolved photoresponse under light on/off switching. (c) Time-resolved analysis of the rise and fall time. rise-time and fall-time are 370 μs and 510 μs, respectively. (d) 3 dB cut-off frequency. current at zero bias. Here, we mainly discuss the time response of devices at zero bias. we measured the time response with oscilloscope at zero bias by pulsating light of 1550 nm. It exhibits that photovoltage time response is the sum of the two parts which included rise-time lower than 370 μs and fall-time lower than 510 μs, respectively. At the same time, we measured the 3 dB cut-off frequency, as shown in the Figure 3(d). The 3 dB cut-off frequency of our device is about 800 Hz.



Figure 4. (a) current–voltage curves of GNWs/Au/Si heterojunction photodetector measured at different incident power (1.7 µW - 1.5 mW). (b) Linear dynamic range of GNWs/Au/Si heterojunction photodetector. LDR is 89.7dB at the -0.2 V. (c) The responsivity behavior of GNWs/Au/n-Si heterojunction with two different effective area (triangle: 4.48 mm2 and square: 0.27 mm2) under different powers. The responsivity of the 4.48 mm2 and 0.27 mm2 device is 21 mA/W and 138 mA/W, respectively. (d) The specific detectivity of GNWs/Au/Si heterojunction photodetector measured and calculated. All tests above were measured under the illumination of 1550 nm laser beam. To test the linearity of the device, we also measured the current–voltage curves at different incident power using the illumination of 1550 nm laser beam in Figure 4(a). We can see that the photocurrent has significantly increased as the power increased at


Page 12 of 25

Page 13 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


reverse bias. The device still maintains supreme rectification characteristics. With a linear fit to the photocurrent, a linear dynamic range (LDR)32 of 89.7dB is obtained at the -0.2 V, as shown in Figure 4(b). Since no obvious photoresponse at lower power have been detected, the LDR at zero bias is smaller than reversal bias.33 Responsivity and specific detectivity (D*) are the key figure-of-merit parameters of a photodetector. Responsivity is the physical quantity that reflects the photoelectric 𝐼𝑝ℎ

conversion capability of the device and is given by R = 𝑃𝑖𝑛 =

𝐼𝑙𝑖𝑔ℎ𝑡 ― 𝐼𝑑𝑎𝑟𝑘 𝑃𝑖𝑛

.34 Where Iph

represents the photocurrent generated under light illumination, which is calculated by subtracting the current measured in the dark from the current measured under the light illumination (Ilight – Idark). Pin is the incident light power on the effective area of the device. Figure 4(c) is the responsivity curve for two devices of different effective area. When the effective incident power is 0.19 µW, the responsivity of the 4.48 mm2 device is 21 mA/W. Due to laser power limitations, we increase the effective power by reducing the effective area of the device. When the effective incident power is 18 nW, the responsivity of the 0.27 mm2 device is increased to 138 mA/W. The D*of the 4.48 mm2 device by measured and calculated was shown in Figure 4(d). D* represents the ability of a device to detect weak light signals and can be obtained from 𝐷 ∗ = (based on calculated noise)34 and 𝐷 ∗ =

𝐴∗𝑅 𝐼𝑛

𝐴∗𝑅 2𝑞𝐼𝑑𝑎𝑟𝑘

(based on measured noise). Where A is

the effective area of the device, In is the measured noise current. The noise properties in this device are determined mainly by the dark current. Then we can calculate D* according to above equation. The specific detectivity of 5.8×109 cm Hz1/2/W and 1.4×1010 cm Hz1/2/W were obtained at 0 V and -1 V respectively. For practical utility,



Page 14 of 25

the current noise measured directly using the Fast Fourier Transform (FFT) method35 is much more credible. Moreover, the frequency-dependent noise spectral density of the photodetector was measured under reverse bias (Figure S2). The dark current noise of 1.6 pA/Hz1/2 was obtained at zero bias. the noise equivalent power (NEP) is about 2×10-9 W/Hz1/2 and 1.3×10-10 W/Hz1/2 at 1 Hz under 0 V and -1 V respectively. The specific detectivity based on measured noise is 1.6×109 cm Hz1/2/W at -1V. When the bias voltage is -1 V, our photodetector achieved maximum responsivity in these structures of graphene-si36,37 and metal-si38. Since these photodetectors were combined with the waveguide structure, the effective area of photodetectors will be small (µm2 or even nm2). It can be estimated that the specific detectivity is much smaller than ours from the equation. The high specific detectivity of our device shows promise for practical applications. 3.3. Disscussion of high responsivity To discuss the photocurrent generation, the band diagram of GNWs/Au/Si photodetector is illustrated in Figure 5(a). To quantitatively analyze the device characteristic, we used the diode equation ф𝐵

𝑞𝑉

I = 𝐴 ∗ 𝑇2exp ( ― 𝐾𝐵𝑇)[exp(𝑛𝑘𝐵𝑇 ―1)]

(1)

where A* is the effective Richardson constant, T is the temperature, ϕB is the Schottky barrier height, kB is Boltzmann’s constant. In the reverse bias saturation 𝑞𝑉

regime, 𝑒𝑥𝑝𝑛𝑘𝐵𝑇 ≪ 1. Therefore, we can extract ϕB by fitting the plot (ln (Isat/T2) versus e/kBT) in the reverse bias saturation regime as depicted in Figure 5(b), where Isat is reverse bias saturation current. I-V curves of our device had been


Page 15 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


measured between 308 K and 343 K (Figure S3). The barrier height of GNWs/Au/Si is 0.4 eV at -1V. A barrier height фB of 0.46 eV at 0V is obtained by extrapolating to get the barrier height of the detector. The lowering of the barrier with the increasing reverse voltage is typical of Gr/Si Schottky diodes due to both the limited density of states of graphene and the image force barrier lowering.39 The previous work has got the GNWs/Si barrier height of 0.69 eV.29 The introduction of Au nanoparticles adjusts the work function of GNWs and reduces the barrier height of the device. Compared with Ag or Ti nanoparticles, it is guaranteed that dark current should not be too large. The absorption spectra of the heterojunction were first measured by UV–vis–NIR spectrometry, as shown in Figure 5(c). The absorption of Single layer graphene is about 3% at 1550 nm. Due to the three-dimensional graphene structure of GNWs, it can achieve 16% light absorption. The absorption of GNWs/Au can reach to 18%, the part of which comes from Au nanoparticles. We used Finite Difference Time Domain (FDTD) software to calculate the electric field distribution of the Au nanoparticles at 1550 nm. The dielectric constant parameters of Au nanoparticles and silicon substrates are derived from the Luxpop database. The gold nanoparticle size was measured by Atomic Force Microscope (AFM) (Figure S4) to build the model, as shown in Figure 5(d). For Au nanoparticles, the area where the electric field is enhanced is mainly concentrated at the edge position. Au nanoparticles are used to confine incident light to these regions by the coupling of electric fields and photons. As Au nanoparticles size increase, the enhancement area around the particle increase. The relative electric field strength gradually decreases, and the resonance peak moves toward the long wave



(Figure S5). We can make two conclusions. First, the particles after annealing of the Au film have an effect of enhancing absorption. Second, the barrier height of GNWs/Au/Si was reduced by Au nanoparticles, which can enhance GNWs thermal electron emission efficiency.

Figure 5. (a) Hot electrons can be injected into a semiconductor via surmounting Schottky barrier. (b) Schottky barrier height of different voltage fitted by temperaturedependent measurements. The Schottky barrier of 0.4 eV at -1 V. inset: The extracted Schottky barrier height versus bias voltage. The Schottky barrier of 0.46 eV at 0 V. (c) NIR absorption spectra of GNWs/Au nanoparticles, GNWs, Gr, Au nanoparticles respectively. (d) Electric field distribution of Au nanoparticles at 1550 nm. Red and blue represent the strongest and weakest positions of the electric field, respectively.


Page 16 of 25

Page 17 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3.4. Response in Mid-infrared and THz Due to the ultra-wide absorption band of GNWs, the photoresponse of the device at a mid-infrared wavelength of 3.5 μm with an optical power of 4.5 mW was tested. As shown in Figure 6(a), the photoresponse of our device in the mid-infrared is obvious. It can be calculated that the responsivity of the device in the mid-infrared is 0.44 μA/W at zero bias. Response at THz is also measured with a broadband THz source of 0.110THz. Since the absorption of GNWs in the mid-infrared is less than that in the farinfrared or even THz, we can get a significant photoresponse of the device in THz from Figure 6(b). The Keithley 2450 Source Meter was applied for the THz test. By calculating the time interval between the collected data points, it can be judged that there is a mains cycle (50 HZ) under the illumination state and the dark condition, but the signal acquisition is not affected.

Figure 6. (a) Photoresponse of the GNWs/Au/Si device at the illumination of 3.5μm and (b) terahertz. 4. CONCLUSIONS In summary, through directly-grown GNWs on silicon surface modified interlayer of

irregular

Au

nanoparticles,

high-performance


Schottky-junction

infrared


photodetectors were obtained. The irregular Au nanoparticles further enhance the broad-spectrum absorption and thermal electron emission efficiency, so as-fabricated device has a photo to dark ratio of 2×104, specific detectivity of 5.8×109 cm Hz1/2/W and 1.4×1010 cm Hz1/2/W at 0 V and -1 V with calculated noise respectively, The specific detectivity based on measured noise is 1.6×109 cm Hz1/2/W at -1 V. Furthermore, responsivity is up to 138 mA/w and 0.44 μA/W at 1550 nm and 3.5 μm, respectively. Rise-time of 370 μs and fall-time of 510 μs were obtained, which present a LDR of 89.7 dB. There was also an obvious signal detected in the THz band. The energy of photon in the mid-infrared and THZ bands is lower, resulting in less hot electrons surmounting the Schottky barrier. Therefore, photoresponse of mid-infrared and THz seems much lower than that in 1550nm. In the other hand, since the photoresponse in mid-infrared and THz is limited by the junction barrier height not the GNWs material, further research is being carried out in our lab to adjust the barrier height suitable for these wavebands. ASSOCIATED CONTENT Supporting Information Current-voltage of different thicknesses of Au nanoparticles; noise spectral density of device; Current-voltage at different temperatures; AFM characterization of 2nm Au film after annealing; the variation of the electric field maximum value of Au nanoparticles with different thickness with wavelength. AUTHOR INFORMATION Corresponding Author


Page 18 of 25

Page 19 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


*Email: [email protected]. *Email: [email protected]. Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. ‡Xiangzhi Liu, ‡Quan Zhou contributed equally. Notes The authors declare no competing financial interests. ACKNOWLEDGMENT This work was supported by NSFC (No. 61705229), Youth Innovation Promotion Association of CAS (2015316, 2018416), Project of Chongqing brain science Collaborative Innovation Center, Project of CAS Western Young Scholar, Project of CQ CSTC (cstc2017zdcy-zdyfX0001, cstc2017zdcy-zdyfX0078). Xiangzhi Liu and Quan Zhou contributed equally to this work. REFERENCES 1.

Rogalski, A., Infrared Detectors for the Future. Acta Physica Polonica A 2009, 116 (3), 389-406.

2.

Eng, P. C.; Song, S.; Ping, B., State-of-the-Art Photodetectors for Optoelectronic Integration at Telecommunication Wavelength. Nanophotonics 2015, 4 (1), 277302.



3.

Yan, J.; Kim, M. H.; Elle, J. A.; Sushkov, A. B.; Jenkins, G. S.; Milchberg, H. M.; Fuhrer, M. S.; Drew, H. D., Dual-Gated Bilayer Graphene Hot-Electron Bolometer. Nature nanotechnology 2012, 7, 472-478.

4.

Akiyama, A.; Kimata, M., 1040×1040 Infrared Charge Sweep Device Imager with PtSi Schottky-Barrier Detectors. Optical Engineering 1994, 33 (1), 64-71.

5.

Brongersma, M. L.; Halas, N. J.; Nordlander, P., Plasmon-Induced Hot Carrier Science and Technology. Nature nanotechnology 2015, 10 (1), 25-34.

6.

Casalino, M.; Coppola, G.; La Rue, R. M.; Logan, D. F., State-of-the-Art AllSilicon Sub-Bandgap Photodetectors at Telecom and Datacom Wavelengths. Laser & Photonics Reviews 2016, 10 (6), 895-921.

7.

Graham, M. W.; Shi, S. F.; Ralph, D. C.; Park, J.; Mceuen, P. L., Photocurrent Measurements of Supercollision Cooling in Graphene. Nature Physics 2013, 9 (2), 103-108.

8.

Tielrooij, K. J.; Piatkowski, L.; Massicotte, M.; Woessner, A.; Ma, Q.; Lee, Y.; Myhro, K. S.; Lau, C. N.; Jarilloherrero, P.; Hulst, N. F. V., Generation of Photovoltage in Graphene on a Femtosecond Timescale through Efficient Carrier Heating. Nature nanotechnology 2015, 10 (5), 437-443.

9.

Han, J.; Wang, J.; Yang, M.; Kong, X.; Chen, X.; Huang, Z.; Guo, H.; Gou, J.; Tao, S.; Liu, Z.; Wu, Z.; Jiang, Y.; Wang, X., Graphene/Organic Semiconductor Heterojunction Phototransistors with Broadband and Bi-directional Photoresponse. Advanced materials 2018, 30 (49), e1804020.

10. Ma, Q.; Andersen, T. I.; Nair, N. L.; Gabor, N. M.; Massicotte, M.; Lui, C. H.;


Page 20 of 25

Page 21 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Young, A. F.; Fang, W.; Watanabe, K.; Taniguchi, T., Tuning Ultrafast Electron Thermalization Pathways in a Van Der Waals Heterostructure. Nature Physics 2016, 12 (5), 455–459. 11. Massicotte, M.; Schmidt, P.; Vialla, F.; Watanabe, K.; Taniguchi, T.; Tielrooij, K.

J.; Koppens, F. H. L., Photo-Thermionic Effect in Vertical Graphene Heterostructures. Nature communications 2016, 7, 12174. 12. Rodrigueznieva, J. F.; Dresselhaus, M. S.; Song, J. C., Enhanced Thermionic-

Dominated Photoresponse in Graphene Schottky Junctions. Nano letters 2016, 16 (10), 6036-6041. 13. Urcuyo, R.; Duong, D. L.; Sailer, P.; Burghard, M.; Kern, K., Hot Carrier

Extraction from Multilayer Graphene. Nano letters 2016, 16 (11), 6761-6766. 14. Li, X.; Tao, L.; Chen, Z.; Fang, H.; Li, X.; Wang, X.; Xu, J.-B.; Zhu, H., Graphene

and Related Two-Dimensional Materials: Structure-Property Relationships for Electronics and Optoelectronics. Applied Physics Reviews 2017, 4 (2), 021306. 15. Zhang, B. Y.; Liu, T.; Meng, B.; Li, X.; Liang, G.; Hu, X.; Wang, Q. J., Broadband

High Photoresponse from Pure Monolayer Graphene Photodetector. Nature communications 2013, 4 (6), 1811. 16. Zhou, H.; Wang, J.; Ji, C.; Liu, X.; Han, J.; Yang, M.; Gou, J.; Xu, J.; Jiang, Y.,

Polarimetric Vis-NIR Photodetector Based on Self-Aligned Single-Walled Carbon Nanotubes. Carbon 2019, 143, 844-850. 17. Pospischil, A.; Humer, M.; Furchi, M. M.; Bachmann, D.; Guider, R.; Fromherz,

T.; Mueller, T., CMOS-Compatible Graphene Photodetector Covering All Optical



Communication Bands. Nature Photonics 2013, 7 (11), 892-896. 18. Bartolomeo, A. D., Graphene Schottky diodes: An Experimental Review of the

Rectifying Graphene/Semiconductor Heterojunction. Physics Reports 2016, 606, 1-58. 19. Koppens, F. H.; Mueller, T.; Avouris, P.; Ferrari, A. C.; Vitiello, M. S.; Polini, M.,

Photodetectors Based on Graphene, Other Two-Dimensional Materials and Hybrid Systems. Nature nanotechnology 2014, 9 (10), 780-793. 20. Konstantatos, G.; Badioli, M.; Gaudreau, L.; Osmond, J.; Bernechea, M.; Arquer,

F. P. G. D.; Gatti, F.; Koppens, F. H. L., Hybrid Graphene-Quantum Dot Phototransistors with Ultrahigh Gain. Nature nanotechnology 2011, 7 (6), 363-368. 21. Wang, X.; Cheng, Z.; Xu, K.; Tsang, H. K.; Xu, J. B., High-Responsivity

Graphene/Silicon-Heterostructure Waveguide Photodetectors. Nature Photonics 2013, 7 (11), 888-891. 22. Marco, F.; Alexander, U.; Andreas, P.; Govinda, L.; Karl, U.; Hermann, D.; Pavel,

K.; Maxwell, A. A.; Werner, S.; Gottfried, S., Microcavity-Integrated Graphene Photodetector. Nano letters 2012, 12 (6), 2773-2777. 23. Echtermeyer, T. J.; Milana, S.; Sassi, U.; Eiden, A.; Wu, M.; Lidorikis, E.; Ferrari,

A. C., Surface Plasmon Polariton Graphene Photodetectors. Nano letters 2015, 16 (1), 8-20. 24. Chen, Y. L.; Feng, X. B.; Hou, D. D., Optical Absorptions in Monolayer and

Bilayer Graphene. Acta Physica Sinica 2013, 62 (18), 187301-187301. 25. Haynes, W. M., Handbook of Chemistry and Physics 96th. CRC Press, Boca Raton,


Page 22 of 25

Page 23 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


FL, 2012; pp 2298-2299. 26. Jiao, T.; Wei, D.; Liu, J.; Sun, W.; Jia, S.; Zhang, W.; Feng, Y.; Shi, H.; Du, C.,

Flexible Solar Cells Based on Graphene-Ultrathin Silicon Schottky Junction. Rsc Advances 2015, 5 (89), 73202-73206. 27. Lupina, G.; Kitzmann, J.; Costina, I.; Lukosius, M.; Wenger, C.; Wolff, A.; Vaziri,

S.; M, Ã. s.; Pasternak, I.; Krajewska, A., Residual Metallic Contamination of Transferred Chemical Vapor Deposited Graphene. ACS nano 2015, 9 (5), 47764785. 28. Song, X.; Jian, L.; Yu, L.; Yang, J.; Liang, F.; Shi, H.; Du, C.; Wei, D., Direct

Versatile PECVD Growth of GrapheneNanowalls on Multiple Substrates. Materials Letters 2014, 137 (137), 25-28. 29. Shen, J.; Liu, X.; Song, X.; Li, X.; Wang, J.; Zhou, Q.; Luo, S.; Feng, W.; Wei, X.;

Lu, S., High-Performance Schottky Heterojunction Photodetector with Directly Grown Graphene Nanowalls as Electrodes. Nanoscale 2017, 9 (18), 6020-6025. 30. Yang, M.; Wang, J.; Zhao, Y.; He, L.; Ji, C.; Liu, X.; Zhou, H.; Wu, Z.; Wang, X.;

Jiang, Y., Three-Dimensional Topological Insulator Bi2Te3/Organic Thin Film Heterojunction Photodetector with Fast and Wideband Response from 450 to 3500 Nanometer. ACS nano 2019, 13 (1), 755-763. 31. Yang, M.; Wang, J.; Han, J.; Ling, J.; Ji, C.; Kong, X.; Liu, X.; Huang, Z.; Gou, J.;

Liu, Z.; Xiu, F.; Jiang, Y., Enhanced Performance of Wideband Room Temperature Photodetector Based on Cd3As2 Thin Film/Pentacene Heterojunction. ACS Photonics 2018, 5 (8), 3438-3445.



32. Nikitskiy, I.; Goossens, S.; Kufer, D.; Lasanta, T.; Navickaite, G.; Koppens, F. H.;

Konstantatos, G., Integrating an Electrically Active Colloidal Quantum Dot Photodiode with a Graphene Phototransistor. Nature communications 2016, 7, 11954. 33. Bartolomeo, A. D.; Giubileo, F.; Luongo, G.; Iemmo, L.; Martucciello, N.; Gang,

N.; Fraschke, M.; Skibitzki, O.; Schroeder, T.; Lupina, G., Tunable Schottky Barrier and High Responsivity in Graphene/Si-Nanotip Optoelectronic Device. 2D Materials 2016, 4 (1), 015024. 34. Xiao, P.; Mao, J.; Ding, K.; Luo, W.; Hu, W.; Zhang, X.; Zhang, X.; Jie, J.,

Solution-Processed 3D RGO-MoS2 /Pyramid Si Heterojunction for Ultrahigh Detectivity and Ultra-Broadband Photodetection. Advanced materials 2018, 30 (31), e1801729. 35. Liu, C. H.; Chang, Y. C.; Norris, T. B.; Zhong, Z., Graphene Photodetectors with

Ultra-Broadband and High Responsivity at Room Temperature. Nature nanotechnology 2014, 9 (4), 273-278. 36. Goykhman, I.; Sassi, U.; Desiatov, B.; Mazurski, N.; Milana, S.; de Fazio, D.;

Eiden, A.; Khurgin, J.; Shappir, J.; Levy, U.; Ferrari, A. C., On-Chip Integrated, Silicon-Graphene Plasmonic Schottky Photodetector with High Responsivity and Avalanche Photogain. Nano letters 2016, 16 (5), 3005-3013. 37. Bartolomeo, A. D.; Luongo, G.; Giubileo, F.; Funicello, N.; Gang, N.; Schroeder,

T.; Lisker, M.; Lupina, G., Hybrid Graphene/Silicon Schottky Photodiode with Intrinsic Gating Effect. 2D Materials 2017, 4 (2), 025075.


Page 24 of 25

Page 25 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


38. Muehlbrandt, S.; Melikyan, A.; Harter, T.; Köhnle, K.; Muslija, A.; Vincze, P.;

Wolf, S.; Jakobs, P.; Fedoryshyn, Y.; Freude, W., Silicon-Plasmonic InternalPhotoemission Detector for 40 Gbit/s Data Reception. Optica 2016, 3 (7), 741-747. 39. Bartolomeo, A. D.; Luongo, G.; Iemmo, L.; Urban, F.; Giubileo, F., Graphene-

Silicon

Schottky

Diodes

for

Photodetection.

Nanotechnology 2018, 17 (6), 1133-1137.

Table Of Contents (TOC) Graphic


IEEE

Transactions

on

Infrared photodetector based on photo-thermionic effect of graphene

Recommend Documents