Direct Observation of High Photoresponsivity in Pure Graphene Photodetectors
© The Author(s). 2017
Received: 5 December 2016
Accepted: 30 December 2016
Published: 7 February 2017
Ultrafast and broad spectral bandwidth photodetectors are desirable attributable to their unique bandstructures. Photodetectors based on graphene have great potential due to graphene’s outstanding optical and electrical properties. However, the highest reported values of the photoresponsivity of pure graphene are less than 10 mA/W at room temperature, which significantly limits its potential applications. Here, we report a photoresponsivity of 32 A/W in pure monolayer graphene photodetectors, an improvement of over one order of magnitude for functional graphene nanostructures (<3 A/W). The high photocurrent generation in our devices can be attributed to the high sensitivity of graphene’s resistivity to a local change of the electric field induced by photo-excited carriers generated in the light-doping substrate. This dramatically increases the feasibility of using graphene for the next generation of photodetectors.
As a novel two-dimensional (2D) material, graphene possesses a unique conic band structure [1–3] and at low energy shows a linear energy dispersion of massless Dirac fermions [4–6] and thus is popularly considered for many optoelectronic and electronic applications . For example, when the Fermi level of monolayer or bilayer graphene is tuned such that it is larger than half of the incident photon energy, graphene becomes completely transparent as a result of Pauli blocking [4–8]. This tunable property of graphene makes it highly suitable for many photonic devices [8–20]. A straightforward application of graphene, graphene photodetectors, has attracted numerous researchers, as graphene exhibits many advantageous behaviors compared to semiconductor counterparts [21, 22]. In addition to the aforementioned wide spectral absorption, with its high carrier mobility, graphene photodetectors are intrinsically capable of ultrafast operation not achievable by semiconductor-based detectors. The maximum bandwidth value of a reported graphene photodetector was 640 GHz , limited by the RC time constant of the measurement circuit. Additionally, the carrier extraction mechanism in graphene photodetectors is different from that of semiconductor photodetectors due to the built-in voltage at the metal–graphene junction . Currently, photo-generated carriers extract from graphene photodetectors because the metal–graphene interfaces produce local potential variations . However, the reported maximum responsivity in pure graphene photodetectors is very low (a few mA/W) , originating from the limited absorption (~2.3%) of the thin material, the small effective detection area of graphene sheet , and the short photo-generated carrier lifetime due to the gapless energy bands [4–8].
To date, work on creating graphene photodetectors with a high response speed and a broad spectral range at room temperature is still ongoing. More recent efforts concentrate on graphene functionalization techniques by creating a bandgap to enhance the photoresponsivity. However, not much work has been done on understanding and controlling the substrate effect on graphene photodetectors. Several known mechanisms contribute to photocurrent generation in biased graphene with a heavily doped silicon substrate [7, 24, 26]. Theoretical understanding and experimental investigations of generation mechanisms of photocurrents in graphene on lightly p-doped silicon substrates still have room for improvement, in how the thermoelectric and photovoltaic effects of graphene can affect the photocurrent. In this experiment, we investigate the mechanisms of photocurrent generation in biased graphene with a graphene field-effect transistor (FET) configuration fabricated on a lightly p-doped silicon/silicon oxide (Si/SiO2) substrate. We demonstrate a higher photoresponsivity at room temperature at visible range. The photoresponsivity value reaches a maximum of 32 AW−1 at room temperature measured from biased source-drain and back-gate voltage modulation, one order of magnitude higher than the values reported in recently published works. Our work potentially reveals the importance of the effects of the doping substrate for graphene photodetectors and offers insight on how to optimize pure graphene-based photodetectors and transistors for optoelectronic and electronic practical applications.
Graphene photodetectors with an FET configuration were fabricated by mechanical exfoliation from bulk highly oriented pyrolytic graphite (grade ZYA, SPI Supplies). Photolithography and e-beam evaporation of 5 and 80 nm of chromium and gold, respectively, were used to create the source and drain electrodes. The sample of monolayer graphene was identified by optical contrast on the 300 nm SiO2 dielectric and confirmed with Raman spectroscopy. In order to detect the excessive photo field-effect, we used a lightly doped (p-type 10–20 ohm-cm) silicon dioxide on silicon substrate. In situ sample cleaning by the electric field were performed to clean the surface, enhance carrier mobility, and reduce intrinsic defects. After this treatment, the Dirac point of the devices were near 25 V. The photocurrent measurements were performed with a helium–neon laser with a wavelength of λ ≈ 632 nm at room temperature. Incident laser power values were fixed at 5 mW, and the frequency of the laser pulse could be adjusted from 5–5000 Hz. The laser spot diameter on the sample could be adjusted from d ≈ 1 μm to 1 mm using a microscope objective lens. The photoresponsivity data was collected with a semiconductor device analyzer (Agilent, B1500A). Considering the electric field enhancement attributable to the gate stack, we calculated that the absorption of the incident light in the graphene device was 2.5%. The photocurrent amplitudes shown are peak-to-background values throughout. The photoresponsivity measurements were carried out on multiple samples.
Results and discussion
To clarify the origin of the photocurrent, we investigated the photocurrent dependence on the laser spot size. This allowed us to clearly identify the photocurrent contributions from both the silicon substrate and the graphene–electrode interfaces, where internal electric fields are produced and separate the photo-generated carriers . In real metal–graphene electrodes, the bending of the energy bands depends on the types and density of the surface states [29–32]. Figure 2b shows the photoresponsivity dependence on the laser spot size under the back-gate voltage bias of V G = − 1V. As shown in Fig. 2b, the magnitude of the photocurrent I ph and photoresponsivity are strongly dependent on the size of the laser spot. The photocurrent I ph increases linearly with the laser spot size and saturates at spot size d = 1.0 mm. Such behavior indicates that the photocurrent originates from several different mechanisms [4, 25, 26]. We propose that the high photocurrent generation in our device arises from the high sensitivity of graphene’s resistivity to the local change of the electric field. In our experiments, lightly doped silicon substrates were used, resulting in the creation of a large vertical gate-like voltage on the substrate, increasing the total photocurrent. The local change of the electric field can be attibuted to the photo-excited carrier generation in the underlying lightly doped substrate. The inset in Fig. 2b represents the schematic illustration of the positive charge accumulation at the Si/SiO2 interface under illumination without a source-drain bias, leading to a photogating effect in the graphene FET [33, 34]. The graphene channel shows a high sensitivity to external electrostatic perturbation, as interfacial charge traps switch the gate voltage of the graphene FET, leading to efficient photocurrent. As the spot size decreases to a size matching the sample device, the photocurrent I ph reduces to 10 nA(S = 0.8 mAW − 1, G = 6.8 × 10− 4). This value agrees with the previously reported results of a low photoresponsivity of pure graphene of about 10 mAW−1 .
Owing to the work function mismatch between silica and silicon, the valence and conduction bands in silicon bend at the interface . In our case of the p-type doped silicon substrate, the energy bands in silicon bend downward, leading to a triangular potential well for the electrons at the interface [35–40]. This downward bending of the energy bands near the metal electrodes enables the photon-generated electrons and holes to easily enter the metal without threshold energy barriers. This substrate effect would yield an enlarged vertical gate voltage, leading to high photoresponsivity . Photo-generated electrons diffuse toward the interface, while holes are repelled away from the interface, allows for an additional negative voltage across the interface, creating a similarly biased negative gate voltage in the graphene field-effect transistor and changing the source-drain current. Interestingly, the spatial extension and the magnitude of the photo field-effect is determined by the substrate’s chemical doping . For the heavily doped silicon, the carrier lifetime is relatively short. For intrinsically or lightly doped silicon (as in our case), the carrier lifetime is much longer, and the spatial extension of the effect can be as large as 1 mm. The intrinsic photocurrent decays fast when the laser spot is away from the graphene channel, but the photo field-effect can still be observed several millimeters away. Such properties enable us to estimate the magnitude of the photo field-effect via adjusting the size of the laser spot.
Figure 4a presents the bias dependence of the external gain and photoresponsivity. As V s-d increases, the gain and photoresponsivity increases linearly and the value is not being saturated. This observation potentially indicates that a higher photoresponsivity can be achieved with pure graphene devices. A maximum photoresponsivity of S = 32AW − 1 was achieved at room temperature with a source-drain bias of V s-d = 200 mV, representing an order of magnitude improvement over previously reported functional graphene nanostructure based photodetectors. Figure 4b, c presents the bias dependence of the external gain and photoresponsivity when the laser spot size is exactly the sample size, where the effect of the substrate can be screened. As V s-d increases, photoresponsivity (S) and gain (G) rise linearly and saturate at V s − d = 2V. A maximum photoresponsivity of S = 0.17 AW − 1 is achieved at room temperature at a source-drain bias of V s-d = 2000 mV, representing a 17-fold improvement over our previously reported value for our graphene photodetector at room temperature. The enhanced photoresponsivity with the addition of the substrate indicates that the lightly doped substrate significantly improves the functionality of the graphene photodetector. This result holds promise for prospective applications in a new era of high-performance optoelectronic devices.
In conclusion, we have demonstrated a highly effective method of increasing the sensitivity of a pure monolayer graphene photodetector by using lightly p-doped silicon dioxide on silicon as the substrate. Our pure graphene photodetector exhibits an increased photoresponsivity, one order of magnitude higher than that of similar graphene photodetectors in previous reports at room temperature. The observed phenomena in our experiments point to a clear pathway toward practical applications of pure graphene in the design of photodetectors that can be manipulated by back-gate (V G) and source-drain (V s-d) voltages. Moreover, the proposed configuration is superior to the previously reported structure, and a larger scale of the proposed device can be easily fabricated. Further device performance improvement can also be obtained through plasmonic nanostructures and waveguide-integrated configurations, providing that a method of growing high-quality graphene with high carrier mobility that is compatible with modern semiconductor technologies can be developed.
YPL would like to thank Dr. Wang Bin, Prof. Su, Prof. Lew, Prof. Wang, and Prof. Yao for their useful discussions. ZWL acknowledges the support from the Australian Research Council (ARC DP130104231). This work was also partially supported by the National Natural Science Foundation of China (Grant No.11174371, 61222406, 51272291, and 11404410).
YPL fabricated the device and performed the experiments. ZWL and JH coordinated the project. YPL, ZWL, QLX, and JH provided key interpretation of the data. YPL, JH, and WSL drafted the paper. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
- Nair RR et al (2008) Fine structure constant defines visual transparency of graphene. Science 320:1308–1308View ArticleGoogle Scholar
- Geim AK, Novoselov KS (2007) The rise of graphene. Nat Mater 6:183–191View ArticleGoogle Scholar
- Geim AK (2009) Graphene: status and prospects. Science 324:1530–1534View ArticleGoogle Scholar
- Koppens FHL et al (2014) Photodetectors based on graphene, other two-dimensional materials and hybrid systems. Nat Nanotechnol 9:780–793View ArticleGoogle Scholar
- Li ZQ et al (2008) Dirac charge dynamics in graphene by infrared spectroscopy. Nat Phys 4:532–535View ArticleGoogle Scholar
- He XY, Liu ZB, Wang DN (2012) Wavelength-tunable, passively mode-locked fiber laser based on graphene and chirped fiber Bragg grating. Opt Lett 37:2394–2396View ArticleGoogle Scholar
- Xia FN, Mueller T, Lin YM, Valdes-Garcia A, Avouris P (2009) Ultrafast graphene photodetector. Nat Nanotechnol 4:839–843View ArticleGoogle Scholar
- Fang ZY et al (2012) Graphene-antenna sandwich photodetector. Nano Lett 12:3808–3813View ArticleGoogle Scholar
- Pospischil A et al (2013) CMOS-compatible graphene photodetector covering all optical communication bands. Nat Photonics 7:892–896View ArticleGoogle Scholar
- Yoon D et al (2008) Strong polarization dependence of double-resonant Raman intensities in graphene. Nano Lett 8:4270–4274View ArticleGoogle Scholar
- Liu M et al (2011) A graphene-based broadband optical modulator. Nature 474:64–67View ArticleGoogle Scholar
- Zhou YX et al (2013) Tunable magnetoplasmons for efficient terahertz modulator and isolator by gated monolayer graphene. Phys Chem Chem Phys 15:5084–5090View ArticleGoogle Scholar
- Bao QL et al (2009) Atomic-layer graphene as a saturable absorber for ultrafast pulsed lasers. Adv Funct Mater 19:3077–3083View ArticleGoogle Scholar
- Sun ZP et al (2010) Graphene mode-locked ultrafast laser. ACS Nano 4:803–810View ArticleGoogle Scholar
- Xu JL, et al (2011) Performance of large-area few-layer graphene saturable absorber in femtosecond bulk laser. Appl Phys Lett 99:3672213Google Scholar
- Tan WD, et al (2010) Mode locking of ceramic Nd: yttrium aluminum garnet with graphene as a saturable absorber. Appl Phys Lett 96:3292018Google Scholar
- Rana F (2008) Graphene terahertz plasmon oscillators. Ieee T Nanotechnol 7:91–99View ArticleGoogle Scholar
- Bi L et al (2011) On-chip optical isolation in monolithically integrated non-reciprocal optical resonators. Nat Photonics 5:758–762View ArticleGoogle Scholar
- Xia FN et al (2001) An asymmetric twin-waveguide high-bandwidth photodiode using a lateral taper coupler. Ieee Photonic Tech L 13:845–847View ArticleGoogle Scholar
- Capasso F et al (2000) New frontiers in quantum cascade lasers and applications. Ieee J Sel Top Quant 6:931–947View ArticleGoogle Scholar
- Ishibashi T et al (2000) InP/InGaAs uni-traveling-carrier photodiodes. Ieice T Electron E83c:938–949Google Scholar
- Ishikawa Y, Wada K (2010) Near-infrared Ge photodiodes for Si photonics: operation frequency and an approach for the future. Ieee Photonics J 2:306–320View ArticleGoogle Scholar
- Meric I et al (2008) Current saturation in zero-bandgap, topgated graphene field-effect transistors. Nat Nanotechnol 3:654–659View ArticleGoogle Scholar
- Xia FN et al (2009) Photocurrent imaging and efficient photon detection in a graphene transistor. Nano Lett 9:1039–1044View ArticleGoogle Scholar
- Mueller T, Xia FNA, Avouris P (2010) Graphene photodetectors for high-speed optical communications. Nat Photonics 4:297–301View ArticleGoogle Scholar
- Freitag M, Low T, Xia FN, Avouris P (2013) Photoconductivity of biased graphene. Nat Photonics 7:53–59View ArticleGoogle Scholar
- Ni ZH et al (2010) On resonant scatterers as a factor limiting carrier mobility in graphene. Nano Lett 10:3868–3872View ArticleGoogle Scholar
- Ferrari AC, et al (2006) Raman spectrum of graphene and graphene layers. Phys Rev Lett 97:187401Google Scholar
- Loomis J, Panchapakesan B (2012) Large photocurrents in single layer graphene thin films: effects of diffusion and drift. Nanotechnology 23:265203Google Scholar
- Lu SX, Panchapakesan B (2006) Photoconductivity in single wall carbon nanotube sheets. Nanotechnology 17:1843–1850View ArticleGoogle Scholar
- Sze SM (1982) Citation classic—physics of semiconductor-devices. Cc/Eng Tech Appl Sci:28-28.Google Scholar
- Hatano T, Ishihara T, Tikhodeev SG, Gippius NA (2009) Transverse photovoltage induced by circularly polarized light. Phys Rev Lett 103:103906Google Scholar
- Ju L, et al (2014) Photoinduced doping in heterostructures of graphene and boron nitride. Nat Nano 60:348-352Google Scholar
- Liu CH, Chang YC, Norris TB, Zhong Z (2014) Graphene photodetectors with ultra-broadband and high responsivity at room temperature. Nat Nano 9:273-278Google Scholar
- Nicollian, Eh and Brews, Jr. MOS(Metal Oxide Semiconductor) Physics and Technology (Wiley, New York, 1982)Google Scholar
- Gabor NM et al (2011) Hot carrier-assisted intrinsic photoresponse in graphene. Science 334:648–652View ArticleGoogle Scholar
- Song JCW, Rudner MS, Marcus CM, Levitov LS (2011) Hot carrier transport and photocurrent response in graphene. Nano Lett 11:4688–4692View ArticleGoogle Scholar
- Ifuku R, Nagashio K, Nishimura T, Toriumi A (2013) The density of states of graphene underneath a metal electrode and its correlation with the contact resistivity. Appl Phys Lett 103:4815990Google Scholar
- Fang T, Konar A, Xing HL, Jena D (2007) Carrier statistics and quantum capacitance of graphene sheets and ribbons. Appl Phys Lett 91:2776887Google Scholar
- Nagashio K, Nishimura T, Toriumi A (2013). Estimation of residual carrier density near the Dirac point in graphene through quantum capacitance measurement. Appl Phys Lett 102:4804430Google Scholar