Photoconductivity of reduced graphene oxide and graphene oxide composite films

Photoconductivity of reduced graphene oxide and graphene oxide composite films

Thin Solid Films 521 (2012) 163–167 Contents lists available at SciVerse ScienceDirect Thin Solid Films journal homepage: www.elsevier.com/locate/ts...

735KB Sizes 43 Downloads 295 Views

Thin Solid Films 521 (2012) 163–167

Contents lists available at SciVerse ScienceDirect

Thin Solid Films journal homepage: www.elsevier.com/locate/tsf

Photoconductivity of reduced graphene oxide and graphene oxide composite films Haifeng Liang ⁎, Wen Ren, Junhong Su, Changlong Cai Key Laboratory for Optical Measurement and Thin Films of Shannxi Province, Xi'an Technological University, Xi'an 710032, China

a r t i c l e

i n f o

Available online 8 January 2012 Keywords: Photo-conductivity RGO/GO Light irradiation Photovoltaic response

a b s t r a c t A photoconductive device was fabricated by patterning magnetron sputtered Pt/Ti electrode and Reduced Graphene Oxide (RGO)/Graphene Oxide (GO) composite films with a sensitive area of 10 × 20 mm2. The surface morphology of as-deposited GO films was observed by scanning electronic microscopy, optical microscopy and atomic force microscopy, respectively. The absorption properties and chemical structure of RGO/GO composite films were obtained using a spectrophotometer and an X-ray photoelectron spectroscopy. The photoconductive properties of the system were characterized under white light irradiation with varied output power and biased voltage. The results show that the resistance decreased from 210 kΩ to 11.5 kΩ as the irradiation power increased from 0.0008 mW to 625 mW. The calculated responsiveness of white light reached 0.53 × 10 − 3 A/W. Furthermore, the device presents a high photo-conductivity response and displays a photovoltaic response with an open circuit voltage from 0.017 V to 0.014 V with irradiation power. The sources of charge are attributed to efficient excitation dissociation at the interface of the RGO/GO composite film, coupled with cross-surface charge percolation. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Graphene, as the thinnest material known in the universe, is a single-atom-thick sheet of sp2-bonded carbon atoms in a hexagonal two-dimensional lattice. These thin graphene sheets have attracted considerable attention for next-generation electronics as a substitute for silicon [1]. Recent research indicates that graphene sheets offer extraordinary electronic, thermal, and mechanical properties that are expected to provide a variety of applications in various technological fields, such as field effect transistors [1], transparent conductors [2,3], photovoltaics [4,5], ultrafast photonics applications [6] and detectors [7]. Due to their high electron mobility (up to 170,000 cm2/Vs) [8], much effort has been made to develop graphene-based electronic devices as a substitute for traditional semiconductor silicon materials. Besides applying graphene to the development of electric devices, there is a strong interest in opening a band gap in graphene for optoelectronic applications. Single-layer graphene should transmit ~ 97% of the incident light and absorb 2.3%, independent of the wavelength of the light [9]. Indeed, a number of experimental studies have verified the 2.3% inter-band absorption in a graphene monolayer over a wide wavelength scale, spanning the visible and infrared ranges [9]. The wide range and constant absorption of graphene make it a promising material for photo detectors.

⁎ Corresponding author. Tel./fax: + 86 29 83208210. E-mail address: hfl[email protected] (H. Liang). 0040-6090/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2011.12.086

Graphene has metallic-like properties and good electrical conductivity, which limits its applications in conventional photoconductors. Hwang et al. verified that there is no clear photoconductive behavior in a monolayer of graphene under irradiation with visible light [10]. Reduced Graphene Oxide (RGO) is a known semiconductor. Progress has been made to modify Graphene Oxide (GO) or RGO samples by compositing other materials that are intended to act as sensitive layer photoconductors. Obvious photocurrents have been detected under bias voltages and light irradiation [11–14]. However, Lv Xin observed high photocurrent generation efficiency for graphene-based films [15]. There is some controversy regarding the charge generation mechanism of RGO and GO/RGO composite films. Venkatran et al. reported a conjugated polymer-graphene oxide with better photoconductivity than that of GO. They determined that the GO is the source of photo-electron generation, while the polymer merely collected and mobilized these electrons [11]. Yao et al. presented RGO and poly-diallyldimethylammonium\titania hybrid films that exhibit high photocurrent generation. The photo-electron generation was ascribed to the photo-electro conversion of TiO2 nano-sheets, where the RGO was only an electronic collector and transporter [12]. However, the efficient photocurrent conversion in RGO and titania multilayered films, reported by Lon et al., was attributed to the efficient excitation dissociation at the interface coupled with cross-surface charge percolation [13]. Moreover, the charge generation for RGO and copper phthalocyanine composite, reported by Zhai et al., was also attributed to the presence of donor/acceptor materials and large donor/acceptor interfaces [14]. So, a large number of experiments under different conditions, using different films (Graphene, GO, GO/RGO, others Graphene/GO/RGO

164

H. Liang et al. / Thin Solid Films 521 (2012) 163–167

based composite films), are needed in order to clarify the fundamental understanding of these systems, as well as to enhance the photoconductivity properties before the commercialization of Graphene/GO/ RGO-based photoconductor devices is possible. Semiconductor light detectors can be divided into two major categories: junction and bulk effect devices. Junction devices, including Schottky photodiodes, Metal-Semiconductor-Metal (MSM) photodiodes, p-i-n/p-n and avalanche photodiodes, and phototransistors, utilize the reverse characteristic of a junction when operated in the photoconductive mode. Under reverse bias, the PN junction acts as a light controlled current source. The output is proportional to the incident illumination and is relatively independent of applied voltage. In contrast, bulk effect photoconductors have no junction. The bulk resistivity decreases with increasing illumination, allowing more photocurrent to flow. This resistive characteristic gives bulk effect photoconductors a unique quality: the signal current from the detector can be varied over a wide range by adjusting the applied voltage. In our work, GO films were patterned on Pt/Ti electrodes with 20 mm gaps on the SiO2/Si substrate to form typical bulk effect light detector devices. After annealing at a temperature of 1173 K, with a pressure of 0.1 Pa, RGO and GO composite films were formed because parts of the GO were reduced to graphene. We observed high photocurrents under visible light. Our results show that only carbon nano-sheet (RGO and GO) composite films could be applied to the development of sensitive photoconductor or energy conversion materials. The sources of charge were attributed to efficient excitation dissociation at the interface of the RGO/GO composition, coupled with cross-surface charge percolation. 2. Experiments 2.1. Device preparation The GO was prepared by oxidation of graphite using the method of Hummer and Offeman [15]. A GO water solution with a density of 8 mg/10 ml was spin-coated on a quartz slide for absorption properties and on a SiO2/Si substrate for fabricating device, scanning electronic microscopy (SEM), optical microscopy, X-ray photoelectron spectroscopy (XPS) and atomic force microscopy (AFM) measurements. The device electrodes were prepared by argon sputtering at sequential Ti and Pt targets with a 150 W power supply under a working pressure of 4 Pa. The thickness of Pt and Ti were determined to be 40 nm and 10 nm, respectively. Fig. 1 presents a schematic illustration of the fabricated device, along with the electrical transport measurement setup. A GO

Fig. 1. Schematic illustration of the fabricated device, along with the electrical transport measurement setup.

thickness of about 10 nm was obtained by cycling the spinning of the GO water solution with a 600 rpm velocity, resulting in an effective sensitive area of 10× 20 mm2. A cross section, obtained by SEM, illustrates the thin GO layer under the SiO2/Si substrate, as shown in Fig. 2(b). The entire device was annealed at a temperature of less than 1173 K with a pressure of 0.1 Pa for 30 min. 2.2. Photoconductivity measurements Photoconductivity measurements were performed by illuminating the composite film under the visible light from a fiber with halogen light source. The output was focused onto the device with a spot size of 30 mm, which covered the entire device, including the electrodes. A variable voltage was applied between the electrodes. All the measurements were carried out in air at room temperature. 2.3. Properties of RGO/GO composite films The absorption properties for wavelengths ranging from 200 nm to 2000 nm were acquired using UV spectroscopy produced by Hitachi UV-3501. The morphology of the GO was obtained using AFM (produced by VECCO with type of Innova) with a silicon nitride tip under tapping mode and optical microscopy (produced by Nikon with type of L150) under dark field mode, respectively. The chemical structure was acquired by XPS produced by Thermo fisher scientific with model of K-Alpha. Aluminum Kα X-rays, whose beam energy, beam spot size and beam current density are 1486.68 eV 400 μm and 4.778A/cm− 2 respectively, were applied to evaluate the composite films' chemical structure. The original XPS data were collected under scanning mode and analyzed in Thermo Scientific Advantage 4.51 software, respectively. The peak fitting processes were performed as follows. Firstly, the linear background was deducted from the original XPS data. Secondly, subtracted a 0.4 eV to each of the experimental XPS data according to the C–C bond energy (284.6 eV), and then fitted the remaining XPS data using Gauss peak. 3. Results and discussion The formation of single-layered GO sheets and measurements of their thickness were confirmed by AFM measurements, as shown in Fig. 2(a). The thickness was found to be in the range of 1–1.3 nm, which is typical for the thickness of a single, functionalized GO sheet. Large numbers of flakes, dozens of micrometers in size, were observed, as shown in Fig. 2(c), depicting the surface structure and morphology of the GO/RGO composite film coated on a SiO2/Si substrate for photocurrent measurements using dark-field model microscopy with 100× objective. Fig. 3 displays the transmittance spectra of a RGO coated quartz substrate and an uncoated quartz substrate. Also shown in Fig. 3 is the difference between these two transmittance spectra. From the difference curve, it is evident that as-deposited film after annealing presents an absorption maximum at 266 nm and a tail up to full measurement range. Venkatram Nalla discovered the absorption maximum for GO at 235 nm [11]. Other authors observed the GO absorption maximum at 225 nm [16]. The red-shift of the absorption maximum of the as-deposited film after annealing, as well as the absorption in the whole spectrum in our device suggested an extended-stacking interlayer conjunction and furthermore indicated that the GO was reduced to graphene [12]. However, the fully reduced GO film did not show any absorption peak [16], indicating that the part organic function groups were removed in our work. Hence, we have confirmed that the as-deposited films after annealing reduction are RGO and GO composite films. Fig. 4(a) shows the XPS data taken on a thin film of GO before reduction on Si/SiO2. Here we can see the three components of carbon-based atom in different functional groups of GO: (i) the non-oxygenated C–C bond (284.6 eV), (ii) C–O bond (286.6 eV) and (iii) the C=O bond

H. Liang et al. / Thin Solid Films 521 (2012) 163–167

165

Fig. 2. (a) Morphology of the GO films in AFM, (b) Cross section of the device, (c) Morphology of the GO films in optical microscopy, (d) Schematic diagram of the energy levels.

(288.3 eV). Fig. 1(b) shows the XPS data after reduction (RGO/GO). The peaks for oxygen functional groups in RGO were reduced. This also implies that the oxygen in the GO is removed by the reduction process. However, the resident peaks for oxygen functional groups in RGO also indicated that the composite films contained part GO after reduction which is comparable to that of absorption curves. Fig. 5 shows the current-voltage properties under light illumination with light powers of 144 mW, 245 mW and 431 mW. We have determined that the photovoltage at the RGO/metal electrode interface has been modulated by the Schottky barrier [17], giving a localized potential that is equal and opposite when the anode and cathode electrode junction (positive and negative) are separately illuminated by the light source. In order to cancel out the interface effect, we have used a light source that is bigger than the sample size and illuminates both interfaces simultaneously, which is similar to the arrangement in reference 17. We conclude that the photo response reported here is due to the RGO/GO composite only and not due to interface effects. From this

picture, we can see that the dark current linearly increases with the applied voltage, which shows good Ohm contact. However, the net photo-current exhibits distinct diode behavior under light irradiation. For applied voltages, ranging from −1 V to +1 V, the photo-current dramatically increases and tends to saturate with applied voltages of more than 1 V and less than −1 V, under various light irradiation powers. This dramatic increase at low applied voltages shows that increasing the applied field causes the capture of more charge carriers, which matches the modified kinetic model for the Onsager model, which is used widely to analyze the features of photo generated charge carriers [18]. Moreover, evidence of photocurrent saturation at higher applied fields suggests that there exists a charge carrier limitation. This photocurrent tendency is in accordance with the results of bulk-film-based graphene sheets reported by Lv Xin [16]. In addition, the device displays a photovoltaic response when using a RGO/GO active layer with an

Fig. 3. Transmittance spectrum of RGO/GO composite films.

Fig. 4. XPS of the GO (a) before reduction and (b) after reduction.

166

H. Liang et al. / Thin Solid Films 521 (2012) 163–167

Fig. 5. Current-voltage properties of the device.

Fig. 7. Photo-current varying with the light power.

open circuit voltage from 0.017 V to 0.014 V with irradiation power, shown in the inset of Fig. 5, which means that RGO/GO composite films could be used as energy conversion materials. Some authors have also reported a type of carbon material that presents a photovoltaic response [19]. In this case, it was determined that the active layer was C60 where GO's main function is in materials processing, which is quite different from the main function of our device. As the irradiation power increases the RGO/GO composite films present stronger photovoltaic properties under 0 V applied voltage, specifically the short circuit current increases from 1.54 μA to 4.23 μA, as shown in Fig. 6. With an applied voltage of −3 V, the net photocurrent increases from 0 μA to 249 μA and the corresponding resistance decreases from 210 kΩ to 11.5 kΩ as the irradiation power increases from 0.0008 mW to 625 mW, as shown in Fig. 7. The calculated responsiveness of white light reached 0.53× 10− 3 A/W. We aim to answer the question: Where does photo-current come from? Does it come from the GO, the RGO, or the GO/RGO junction? GO is known to be a semiconductor material with excellent electrical properties. However, some authors have proven that GO not a good photo-conductor [11]. Since we found no evidence of current flow under applied voltage and light irradiation before annealing in our device, we deduced that the charge did not come from the GO. Pure RGO exhibits metallic-like properties and has good electrical conductivity, which limits its applications for photoconductors. Gilgueng Hwang verified that there is no clear photoconductive behavior in monolayer graphene under visible light irradiation [10]. Therefore, we conclude that the photo-current comes from the GO/RGO junction. In our device, we observed obvious photoconductive behavior in partially reduced GO. This charge generation is likely due to efficient excitation dissociation at the interface of RGO/GO. Previous work has shown that as-synthesized GO undergoes insulator-semiconductorsemi-metal transitions with reduction, yielding an apparent transport gap ranging from 10–50 meV that approaches zero with extensive reduction [20]. In addition, The RGO/GO work function varies from 5.4

to 4.2 eV according to various literatures [21–24] and is affected by metal contact [21,22], electric fields [23], and reduction levels [20]. According to the analysis mentioned above, the RGO/GO composite sheets in our devices should be a mixture of many large fused aromatic molecules, which have a similar structure with some minor differences in size, shape, and edge group, as shown in Fig. 2(c). These minor differences would affect the GO or the RGO band structure and the work function. Furthermore, the RGO/GO composite sheets are partially reduced GO, namely some of the RGO displayed semiconductor properties and other RGO demonstrated semi-metal properties. Hence, the RGO/GO composite work function profiler comprises a wide range, from 4.2 to 5.5 eV. The corresponding schematic diagram of the energy levels is shown in Fig. 2(d). The work functions for Pt are taken from the literature [25]. The photoconductivity is caused by exciton creation via absorbing photons, which dissociate excitons into free charge carriers (electrons and holes). In fact, in the RGO/GO composite, the presence of Schottky barriers at the RGO/GO junctions are a consequence of the band-gap in GO [26]. As a result, a built-in electric field is generated near the contact. Photoexcitation near a contact generates electron/ hole pairs that can be separated by the built-in field, and then transported to the corresponding electrodes through the semi-metal RGO network, generating a net photocurrent.

4. Conclusions A photoconductive device was fabricated by patterning the magnetron sputtered Pt/Ti electrode and multi-layer graphene/graphene oxide films with a sensitive area of 10× 20 mm2. Absorption curves, XPS spectrum and AFM images indicate that the as-deposited films are a GO/RGO composite. In this system, the resistance decreased from 210 kΩ to 11.5 kΩ as the irradiation power increased from 0.0008 mW to 625 mW. The calculated responsiveness of white light reached 0.53× 10− 3 A/W. Furthermore, high photocurrents were detected under visible light, which indicates that only carbon nanosheet (RGO/GO) composite films could be applied to the development of sensitive photoconductor materials. Moreover, the device displays a photovoltaic response with an open circuit voltage from 0.017 V to 0.014 V with irradiation power, implying that RGO/GO composite films could be used as energy conversion materials. We have determined that the sources of charge are due to efficient excitation dissociation at the interface of the RGO/GO composite, coupled with crosssurface charge percolation. The photoconductive behavior of graphene makes it a promising candidate for an optoelectronic detector.

Acknowledgements

Fig. 6. Short circuit current varying with the light power.

This work was supported by the National Natural Science Foundation of China (No. 61007015 and 60978040).

H. Liang et al. / Thin Solid Films 521 (2012) 163–167

References [1] A.K. Geim, K.S. Novoselov, Nat. Mater. 6 (2007) 183. [2] Xuan Wang, Linjie Zhi, Klaus M llen, Nano Lett. 8 (2008) 323. [3] Junbo Wu, Héctor A. Becerril, Bao Zhenan, Zunfeng Liu, Chen Yongsheng, Peumans Peter, Appl. Phys. Lett. 92 (2008) 263302. [4] Z.F. Liu, Q. Liu, Y. Huang, Y.F. Ma, S.G. Yin, X.Y. Zhang, W. Sun, Y.S. Chen, Adv. Mater. 20 (2008) 3924. [5] G. Hwang, S. Haliyo, S. R gnier, Micro Nano Lett. 5 (2010) 140. [6] Q.L. Bao, H. Zhang, J.X. Yang, S. Wang, D.Y. Tang, R. Jose, S. Ramakrishna, C.T. Lim, K.P. Loh, Adv. Funct. Mater. 20 (2010) 782. [7] Meryl D. Stoller, Sungjin Park, Yanwu Zhu, Jinho An, Rodney S. Ruoff, Nano Lett. 8 (2008) 3498. [8] K.I. Bolotin, K.J. Sikes, J. Hone, H.L. Stormer, P. Kim, Phys. Rev. Lett. 101 (2008) 096802. [9] Jahan M. Dawlaty, Shriram Shivaraman, Jared Strait, Paul George, Mvs Chandrashekhar, Farhan Rana, Michael G. Spencer, Dmitry Veksler, Yunqing Chen, Appl. Phys. Lett. 93 (2008) 131905. [10] Gilgueng Hwang, Juan Camilo Acosta, Emir Vela, Sinan Haliyo, Stéphane Régnie, ISOT2009 - International Symposium on Optomechatronic Technologies Istanbul, Turkey, Septemner 21 23, 2009, ISOT2009 - International Symposium, 2009, p. 169. [11] Venkatram Nalla, Lakshminarayana Polavarapu, Kiran Kumar Manga, BeeMin Goh, Kian Ping Loh, Qing-Hua Xu, Xei Ji, Nanotechnology 21 (2010) 415203. [12] Hong-Bin Yao, Li-Heng Wu, Chun-Hua Cui, Hai-Yu Fang, Shu-Hong Yu, J. Mater. Chem. 20 (2010) 5190. [13] Kiran Kumar Manga, Yong Zhou, Yongli Yan, Kian Ping Loh, Adv. Funct. Mater. 19 (2009) 3638.

167

[14] Anindarupa Chunder, Tanusri Pal, Saiful I. Khondaker, Lei Zhai, J. Phys. Chem. C 114 (2010) 15129. [15] W.S. Hummers, R.E. Offeman, J. Am. Chem. Soc. 80 (1958) 1339. [16] Xin Lv, Yi Huang, Zhibo Liu, Jianguo Tian, Yan Wang, Yanfeng Ma, Jiajie Liang, Fu. Shipeng, Xiangjian Wan, Yongsheng Chen, Small 5 (2009) 1682. [17] Ph. Avouris, in: 2010 10th IEEE Conference on Nanatehnology, NANO 2010, Seoul, Republic of Korea, August 17-20, 2010, 2010 10th IEEE Conference on Nanatehnology (2010)6. [18] Charles L. Braun, J. Chem. Phys. 80 (1984) 4157. [19] Vincent C. Tung, Jen-Hsien Huang, Ian Tevis, Franklin Kim, Jaemyung Kim, Chih-Wei Chu, Samuel I. Stupp, Jiaxing Huang, J. Am. Chem. Soc. 133 (2011) 4940. [20] Goki Eda, Cecilia Mattevi, Hisato Yamaguchi, HoKwon Kim, Manish Chhowalla, J. Phys. Chem. C 113 (2009) 15768. [21] G. Giovannetti, P.A. Khomyakov, G. Brocks, V.M. Karpan, J. van den Brink, P.J. Kelly, Phys. Rev. Lett. 101 (2008) 026803 1. [22] Y. Murata, E. Starodub, B.B. Kappes, C.V. Ciobanu, N.C. Bartelt, K.F. McCarty, S. Kodambaka, Appl. Phys. Lett. 97 (2010) 143114. [23] Yu. Young-Jun, Yue Zhao, Sunmin Ryu, Louis E. Brus, Kwang S. Kim, Philip Kim, Nano Lett. 9 (2009) 3430. [24] Anass Benayad, Hyeon-Jin Shin, Hyeon Ki Park, Seon-Mi Yoon, Ki Kang Kim, Mei Hua Jin, Hae-Kyung Jeong, Jae Cheol Lee, Jae-Young Choi, Young Hee Lee, Chem. Phys. Lett. 47 (5) (2009) 91. [25] S.V. Jagadeesh Chandra, Jin-Sung Kim, Kyung-Won Moon, Chel-Jong Choi, Microelectron. Eng. 89 (2011) 76. [26] Wu. Xiaosong, Mike Sprinkle, Xuebin Li, Fan Ming, Claire Berger, Walt A. de Heer, Phys. Rev. Lett. 101 (2008) 026801.