Metal-catalyst-free growth of patterned graphene on SiO2 substrate by

Mar 25, 2019 - Illuminated by 792 nm laser, the responsivity and the specific detectivity of the detector measured at room temperature are 275.9 mA/W ...
0 downloads 0 Views 810KB Size
Subscriber access provided by UNIVERSITY OF ADELAIDE LIBRARIES

Surfaces, Interfaces, and Applications

Metal-catalyst-free growth of patterned graphene on SiO substrate by annealing plasma-induced crosslinked Parylene for optoelectronic device applications 2

Yibo Dong, Chuantong Cheng, Chen Xu, Xurui Mao, Yiyang Xie, Hongda Chen, Beiju Huang, Yongdong Zhao, Jun Deng, Weiling Guo, Guanzhong Pan, and Jie Sun ACS Appl. Mater. Interfaces, Just Accepted Manuscript • Publication Date (Web): 25 Mar 2019 Downloaded from http://pubs.acs.org on March 29, 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

Metal-Catalyst-Free Growth of Patterned Graphene on SiO2 Substrate by Annealing Plasma-Induced Cross-Linked Parylene for Optoelectronic Device Applications Yibo Dong,† Chuantong Cheng,‡ Chen Xu,*,† Xurui Mao,*,‡ Yiyang Xie,† Hongda Chen,‡ Beiju Huang,‡ Yongdong Zhao,† Jun Deng,† Weiling Guo,† Guanzhong Pan,† and Jie Sun*,§ †Key

Laboratory of Optoelectronics Technology, College of Microelectronics, Beijing

University of Technology, Beijing 100124, China ‡State

Key Laboratory of Integrated Optoelectronics, Institute of Semiconductor,

Chinese Academy of Sciences, Beijing 100083, China §National

and Local United Engineering Laboratory of Flat Panel Display

Technology, Fuzhou University, Fuzhou 350116, China KEYWORDS: graphene, metal-catalyst-free, direct growth, Parylene, cross-linking, Schottky junction, photodetector

1

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

Page 2 of 25

ABSTRACT A metal-catalyst-free method for the direct growth of patterned graphene on an insulating substrate is reported in this paper. Parylene N is used as the carbon source. The surface molecule layer of Parylene N is cross-linked by argon plasma bombardment. Under high-temperature annealing, the cross-linking layer of Parylene N is graphitized into nanocrystalline graphene, which is a process that transforms organic to inorganic and insulation to conduction, while the Parylene N molecules below the cross-linking layer decompose and vaporize at high temperature. Using this technique, the direct growth of a graphene film in a large area and with good uniformity is achieved. The thickness of the graphene is determined by the thickness of the cross-linking layer. Patterned graphene films can be obtained directly by controlling the patterns of the cross-linking

region

(lithography-free

patterning).

Graphene-silicon

Schottky

junction

photodetectors are fabricated using the as-grown graphene. The Schottky junction shows good performance. The application of direct-grown graphene in optoelectronics is achieved with a great improvement of the device fabrication efficiency compared with transferred graphene. When illuminated with a 792 nm laser, the responsivity and the specific detectivity of the detector measured at room temperature are 275.9 mA/W and 4.93×109 cm Hz1/2/W, respectively.

2

ACS Paragon Plus Environment

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

ACS Applied Materials & Interfaces

INTRODUCTION Graphene has received worldwide attention due to its unique combination of special properties such as high carrier mobility, high transmittance and high mechanical strength.1 Over the past decade, a variety of graphene synthesis techniques have been developed: micromechanical exfoliation,2 reduction of graphene oxide,3-5 epitaxial growth6,7 and chemical vapor deposition (CVD).8-10 Among them, CVD is an efficient method for the synthesis of large-area, high-quality graphene and is one of the most commonly used methods. Typically, CVD grows graphene on the surface of a metal catalyst (Cu, Ni, etc.). In the fabrication of graphene devices, graphene needs to be transferred from the metal substrate to the target substrate. The transfer of one atomic layer of carbon is complicated and inevitably leaves extra defects in the graphene, such as wrinkles, holes, and contaminants. Additionally, there is no mature transfer technology or equipment to achieve large-scale graphene transfer, resulting in low transfer efficiency, which constitutes a large obstacle in the path to industrialization. Under this background, the direct growth of graphene has become a topic of interest in the field of graphene synthesis. In the direct growth of graphene, graphene films are grown directly on target substrates without transfer. Currently, the direct growth techniques of graphene can be roughly divided into three categories:18 metal-catalyst-free growth,11,12 plasma enhanced CVD,13,14 and sacrificial metal-assisted growth.15–17 All these techniques are basically types of CVD, and each has its own advantages and disadvantages.18 Metal-catalyst-free growth has almost no requirement regarding substrates. However, this method often requires a long growth time (several hours) and/or very high temperature (sometimes >1000 °C). The second method can effectively reduce the growth temperature (650 °C) without decomposition and vaporization. After that, rapid annealing is performed in a hydrogen/argon (40 sccm/960 sccm) atmosphere for 30 s at 1050 °C. At high temperature, the cross-linking layer graphitizes and transforms from supramolecular sheets into nanocrystalline graphene,31 which is a process that transforms organic to inorganic and insulation to conductivity. Parylene molecules below the cross-linking layer decompose and vaporize at high temperature (>650 °C). Argon plasma bombardment is a very important step in the growth process. In our experiments, it was found that Parylene would vaporize completely without argon plasma bombardment during the annealing process, and there would be no graphene remaining on the substrates after annealing. Parylene thin film is prepared by a unique vapor deposition process that is more uniform than the films obtained by spinning. When the thickness of Parylene is greater than 100 nm, no pinholes can be guaranteed. All these conditions ensure the uniformity of the graphene. Using a 5

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

photoresist as a mask, the Parylene film can be cross-linked in a local area. As shown in Figure 1c, after argon plasma bombardment on the local area of the Parylene, the color of the Parylene in the bombardment area obviously changes. After annealing, a patterned graphene film is obtained directly (Figure 1d). In the local area, the pattern of the graphene is very fine and the edge is neat as viewed via optical microscope observation. Previous reports on the lithography-free growth of patterned graphene have been mostly based on metal-assisted growth. The graphene patterns have been controlled by defining the metal patterns.17,23 Our method does not involve metals, so there is no risk of metal contamination during high-temperature growth.23 A photoresist is a type of organic compound that is a heavy p-type dopant of graphene.24 The lithography-free growth of graphene aims at avoiding direct contact between the graphene and the photoresist to ensure the stability of graphene’s electrical properties. The X-ray photoelectron spectroscopy (XPS) measurement of the as-prepared graphene is shown in Figure 2a, confirming its composition and chemical bond information. The thickness of the graphene is related to the thickness of the cross-linking layer, while the thickness of the cross-linking layer is related to the argon plasma bombardment power. As shown in Figure 2b, we experimented with three bombardment powers: 50 W, 100 W and 220 W. The thickness of the graphene was measured by atomic force microscopy (AFM). The thickness was found to increase with an increase in the bombardment power. When the argon plasma bombardment power was 50 W, 100 W and 220 W, the obtained graphene thickness was about 2.1 nm, 3.2 nm and 4.9 nm, respectively. This phenomenon is understandable. Higher power means higher energy of the argon plasma and therefore a thicker cross-linking layer. This unique thickness control method makes the whole growth process more controllable. Although it is difficult to achieve the accurate control of a graphene layer by controlling the bombardment power, the control of graphene thickness can be accurate to the nanometer scale. The thickness of the Parylene film used in our experiment was 200 nm, and the graphene obtained was only a few nanometers. In previously reported metal-catalyst-free growth,11,12 with an increase in the growth time, the graphene gradually becomes thicker. Although the thickness of graphene can be controlled by the growth time, for large area graphene growth, it is difficult to achieve a uniform graphene film due to the influence of the gas flow in the CVD system. We measured the square resistance of the graphene via the 6

ACS Paragon Plus Environment

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

ACS Applied Materials & Interfaces

Van der Pauw method. An increase in the thickness causes the square resistance of the graphene to decrease gradually, but it is still large on the whole. Figure 2c shows the Raman spectra (=532 nm) of the graphene. The positions of D, G and 2D peaks are 1337 cm-1, 1597 cm-1 and 2661 cm-1, respectively. The in-plane vibration of the Sp2 carbon atoms produces G-peaks, so the existence of a G-peak represents the Sp2 hybrid structure of the carbon atoms.30 The D peaks are high, which indicates a high graphene defect density. The 2D peaks almost disappeared. An increase in the number of layers34 or excessive defect density35,36 may result in a decrease the 2D peaks. According to the results of our experiments, we believe that the main reason for the “diminished” 2D peak is the high graphene defect density. As shown in Figure 2b, the thickness of the graphene obtained using different argon plasma bombardments is different, but their corresponding Raman spectra have no obvious changes, which indicates that the increase in the number of graphene layers, i.e., being closer to the graphite state, is not the main reason for the “diminished” 2D peaks. We believe that there are two reasons for the high defect density. First, this growth method is a non-CVD method. In CVD growth, in the case of methane as the carbon source, methane is first cleaved into a single carbon atom, which is then combined with other carbon atoms to form graphene. However, the Parylene used in our method is a solid carbon source and grows graphene directly from its molecular structure to nanocrystalline graphene.31 Therefore, the graphene quality is worse than that of graphene grown by methane. Another reason, which is also a widespread characteristic in all metal-catalyst-free growth, is the lack of metal catalysts. Metal catalysts are very important in the growth of graphene. In the absence of metal catalysts, the most common way to improve the quality of graphene is to increase the growth temperature. Figure 2d shows the transmittance of the graphene (220 W argon plasma bombardment). With an increase in the wavelength, the transmittance of the graphene increases gradually. In the visible band, the transmittance is 86.6%~91.7%. Figure 3a and b show scanning electron microscopy (SEM) images of the graphene. In the right part of Figure 3a, there is an intentional tweezer scratch to generate some contrast. The obtained graphene is a continuous film without significant cracks, as seen in Figure 3a and b. Figure 3c-h show AFM images of the surface morphology of the Parylene before and after argon plasma bombardment and the as-grown graphene. The surface morphology of the Parylene has changed 7

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

significantly before and after argon plasma bombardment. The Parylene surface is smooth before bombardment (Figure 3c and d) but rough after bombardment (Figure 3e and f). The breaking of some C–C bonds or functional groups and the formation of crosslinks may cause this kind of roughness.25 At the same time, we can see that a series of supramolecular carbon nanosheets are formed on the surface of the cross-linked Parylene (Figure 3e and f). During annealing, these supramolecular carbon nanosheets will graphitize and transform into nanocrystalline graphene (Figure 3g and h). This nanocrystalline graphene will polymerize and grow larger under the action of high temperature. The initial transformation temperature of graphitization ranges from 700 K to 1300 K.32 The annealing temperature in our experiment is approximately 1300 K, which essentially meets this requirement. Therefore, during annealing, graphitization occurred in our experiments. However, the annealing temperature is much lower than that of highly oriented pyrolytic graphite (HOPG, 3200 °C~3600 °C), resulting in a small grain size (~100 nm) and low-quality nanocrystalline graphene. HOPG is the raw material for micromechanical exfoliated high-quality graphene.2 A very high temperature is needed to obtain high-quality graphene because a solid carbon source is used, and the growth process is a graphitization process rather than a CVD process. We have compared our method with other metal-catalyst-free growth methods (Table 1) 11, 40-43 in terms of five primary aspects: carbon source, substrates, growth time and temperature, graphene quality and patterned graphene growth. The quality comparison of the graphene mainly refers to the Raman spectra. We selected five representative studies for comparison. 11, 40-43 Many methods have been used to achieve metal-catalyst-free growth in these studies, including increasing growth temperature,40 using large-flow carbon sources11 or long growth time,41 using other organic carbon sources,42 and two-stage growth.43 Generally, these methods all use gaseous carbon sources and a CVD growth method. Our method is a non-CVD method. The quality of the graphene is lower than that used in some studies40,41,43 and similar to that from other studies.11,42 The graphene quality still needs to be improved. However, the advantages of our method are also obvious: very fast growth time and patterned growth. Specifically, patterned growth is difficult to achieve in metal-catalyst-free CVD methods. By increasing the annealing temperature, the nanocrystalline size and the conductivity of the graphene film can be increased,31 but a substrate material with a 8

ACS Paragon Plus Environment

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

ACS Applied Materials & Interfaces

higher heat resistance is needed. It is worth mentioning that not only argon plasma but oxygen plasma can also make the surface molecules of Parylene cross-link (Figure S1a, Supporting Information). However, the XPS measurement shows that oxygen atoms remain in the Parylene after bombardment as well as in the obtained graphene. Carbon atoms form a number of covalent bonds with oxygen atoms and oxygen-containing functional groups. The resulting graphene is graphene oxide. By observing the C1s peak (Figure S1c, Supporting Information), we can see that the intensity of the peak at approximately 285 eV of the oxygen-plasma-bombardment graphene is lower than that of the argon-plasma-bombardment graphene, indicating that -C-C- bonds to oxygen form.26 Compared with the argon-plasma-bombardment graphene, the conductivity of this type of graphene oxide is much worse (Figure S1d, Supporting Information). This experiment shows that it is possible to dope graphene with some specific elements by changing the gas used for plasma bombardment. There are many existing reports showing the direct growth of graphene.11-18 However, there are few reports on the application of direct-grown graphene on optoelectronic devices. Almost all graphene devices still use transferred graphene, which has many unstable factors. The polymer support layer (PMMA) used in the graphene transfer process can produce a p-type doping of graphene. Wet etching of copper foils may also cause metal residues on graphene. The contact between the transferred graphene and substrates in different regions may vary due to impurities on the substrate or on the surface of the graphene. All of these factors will cause performance differences between different graphene devices, even in the same batch. Using the direct growth technique, the fabrication efficiency and consistency of graphene devices can be greatly improved,. Most common silicon Schottky junction photodetectors on the market use metals to form the Schottky junction with silicon. Because of the opacity of metal, only a small part of the incident light can reach the depletion layer of the Schottky junction. Therefore, the responsivity and the specific detectivity of the detectors will be affected. Graphene has high transmittance, good conductivity and is very suitable as a transparent electrode for Si Schottky junction detectors. We fabricated graphene-silicon Schottky junction photodetectors using our direct-grown patterned graphene. It should be noted that although patterned graphene growth can be performed at the 9

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

graphene growth stage to protect the graphene from contact with the photoresist, the graphene still inevitably contacts the photoresist during the second-step lithography in the metal electrodes. One solution is to prepare the metal electrodes first and then deposit the graphene. In this case, we need electrode materials that can withstand the growth temperature, such as Pt, Ti and Ni. This is a subject of future studies. The device structure is shown in Figure 4a and b, and more details about the fabrication process can be found in the experimental section. The selected substrate is light-doped n-type silicon (100) with 300 nm SiO2 (resistivity of 1-10 Ω-cm). The bombardment power of the argon plasma is 220 W. The built-in electric field of the Schottky junction formed by the graphene and silicon is shown in Figure 4c. The area of the graphene-silicon Schottky junction window is 1 mm × 1 mm. Figure 4d shows the optical image of the device, and Figure 4e shows a digital photograph of the sample. We use two types of light source: a white light light-emitting diode (LED) and a near infrared laser (792 nm). When illuminated by different incident light power, we obtain the I-V characteristics of the device (Figure 5a and b). The I-V measurements show a very good reproducibility with the maximum standard error less than 5 %. When the incident light power of the white light LED is 0.19 mW, the responsivity of the device is 206 mA/W at 0 V bias and 237.8 mA/W at 4 V reverse bias. When the incident power of a 792 nm laser is 0.2 mW, the responsivity of the device is 205.7 mA/W at 0 V bias, 267.3 mA/W at 2 V reverse bias and 275.9 mA/W at 4 V reverse bias. Figure 5c shows the time-dependent photocurrent measurement over 5 on–off periods using the white light illumination. The response speed of the device is fast, and the rise time exceeds the measurement limit of our equipment (