Open Access
Add to BibSonomyAbstract
The intrinsic photo-response of chemical vapor deposited (CVD) graphene photodetectors were investigated after eliminating the influence of photodesorption using an atomic layer deposited (ALD) Al2O3 passivation layer. A general model describing the intrinsic photocurrent generation in a graphene is developed using the relationship between the device dimensions and the level of intrinsic photocurrent under UV illumination.
© 2013 Optical Society of America
1. Introduction
Graphene, a single layer of carbon atoms, has attracted much attention as a candidate material for photodetectors because graphene can absorb 2.3% of incident light in the infrared to visible wavelength range, even at a one atomic layer thickness [1]. This is remarkably high absorption efficiency because a 15 nm layer of Si or 20 nm layer of GaAs is required to absorb the same amount of incident light [2–4]. The photon-induced carrier multiplication in graphene is another advantage for high efficiency photonic device applications [5,6]. The combination of high carrier mobility and strong interaction with light at a wide range of wavelengths make graphene an excellent candidate material for novel photonic devices [1,7–10]. In particular, a two-dimensional structure of graphene is another merit for scaled optoelectronic device integration, especially in high speed monolithic optical interconnect systems [3,11].
Thus, the photo-electrical response characteristics of carbon-based materials including carbon nanotubes (CNTs) and graphene have been investigated by several research groups [12–16], which have uncovered several challenges associated with photonic applications of graphene. For example, Chen et al. [15] reported that an oxygen photodesorption lowered the conductance of p-type single-walled carbon nanotubes (SWNTs) and a gradual re-adsorption onto the SWNTs leads to the recovery of sample conductance when the light is turned off. Shi et al. [12] showed a negative shift of Id-Vg curves and gradual decrease in drain current due to photon-induced oxygen desorption which dedopes p-type graphene. Sun et al. [13] reported a change in a resistance due to a photo-induced molecular desorption in a graphene film under UV illumination. A rapid decrease in photocurrent due to the photodesorption in the visible to UV wavelength range has also been reported [12]. These are serious problems for photonic device applications because the photodesorption shifts the baseline of photocurrent, which results in significant scattering in the sensitivity of graphene photodetectors. These issues are related to the high surface-to-volume ratio of graphene, which makes graphene especially sensitive to environmental factors, thereby incurring unintentional doping, unstable device operation, and poor performance. In particular, the slow change in resistance under illumination caused by the photodesorption limits the speed of photo-response from a few seconds to hundreds of seconds. Furthermore, since this slow response often dominates the photo-response characteristics, it has been difficult to investigate the intrinsic photocurrent generation in graphene photodetectors [17].
In this work, the intrinsic photo-response of chemical vapor deposited (CVD) graphene photodetectors were investigated after eliminating the photodesorption using an atomic layer deposited (ALD) Al2O3 passivation layer. A general model describing the intrinsic photocurrent generation in a graphene was also developed using the relationship between the device dimensions and intrinsic photocurrent under UV illumination.
2. Experiments
A single layer CVD graphene grown on Cu foil [18] was transferred to a SiO2 (90 nm)/Si substrate [19]. The device was fabricated as follows. Au (100 nm) source/drain electrodes were patterned using i-line photolithography. At this stage, the entire layer of graphene was under the electrode contacts the source/drain metal, minimizing a contact resistance. After the metal patterning, the graphene channel (W = 6 μm ~15 μm, L = 4 μm ~7 μm) was patterned using photolithography and reactive ion etching (RIE) in an oxygen. The RIE was performed for 50 seconds at 50 W at room temperature in O2 ambient (200 mTorr). After the RIE step, photoresist was stripped by dipping the sample in a photoresist stripper solution (AZ400T) at 50 °C for 5 minutes. Then, the samples were annealed in high vacuum (~10−7 Torr) at 200 °C for 2 hour to minimize the residue of photoresist on the graphene surface. Afterwards, the graphene channel was passivated using 30 nm of Al2O3 deposited at 130 °C by a low temperature ALD process. For ALD, trimethylaluminium (TMA) and a H2O precursor were used with N2 carrier gas. Then, the Al2O3 was patterned using a lift-off process to open a part of the source/drain electrodes for an electrical contact. Finally, a PDA was performed at 200 þC for 30 minutes in a high vacuum of ~10−7 Torr to densify the film and drive out residual H2O molecules near the graphene. The final structure of the graphene channel photodetector consisted of a source/drain electrode, a highly doped Si substrate as a back gate, 90 nm of SiO2 gate dielectric, and 30 nm of an Al2O3 passivation layer.
Electrical and photonic properties of the graphene photodetector were characterized using a semiconductor parameter analyzer (Keithley 4200) and 365 nm wavelength LED lamp with 200 µW/cm2 intensity. The drain current was measured with various drain biases ranging from 1 to 100 mV while modulating the illumination intensity of the UV lamp. All the measurements reported here were carried out in air ambient.
3. Results and discussion
Figure 1(a) shows a top-down scanning electron microscope (SEM) image of a graphene photodetectors. After the source/drain patterning, the quality of the graphene channel was measured using Raman spectroscopy at 0.514 nm as shown in Fig. 1(b). Raman spectra showed a small D peak and a high 2D/G ratio of 2.67, the signature of high quality monolayer graphene. The inset of Fig. 1(b) is a SEM image of a graphene channel in which the circled area shown in Fig. 1(a) is magnified. The Al2O3 passivation layer is deposited at 130 °C on the graphene channel region using a lift-off photoresist patterning method. After the Al2O3 deposition, the devices are annealed to densify the Al2O3 layer and to drive H2O molecules out of the graphene/Al2O3 interface. Some devices did not undergo Al2O3 passivation so that the role of the dielectric passivation could be explored.
Fig. 1 (a) Top-down SEM image of a graphene photodetector. (b) Raman spectra of the graphene channel. The inset is a SEM image of the graphene channel that magnifies the white circle in (a). (c) The DC Id-Vg curve of graphene photodetector before the passivation. The graphene channel is exposed to air ambient. (d) The DC Id-Vg curve of the graphene photodetector after the passivation and 200 °C PDA. The graphene channel is passivated with Al2O3.
Before investigating the photo characteristics, the quality of the graphene channel in back gate graphene photodetectors was examined using basic electrical characterization methods. The drain current-gate voltage (Id-Vg) characteristics of these photodetectors were measured at Vd = 10 mV with a Vg sweep range of −40 V to 40 V before and after the Al2O3 passivation. The Id-Vg curves were asymmetric before the Al2O3 passivation, and the Dirac point was observed at Vg = 15 V (30 V for reverse voltage sweep). Hysteresis due to a tunneling-induced charge trapping and an electrochemical reaction between the graphene and ambient species was around 15 V, equivalent to 3.7 x 1012 charge traps/cm2 [20–23] (Fig. 1(c)). These are typical behaviors of air-exposed graphene photodetector or CNT based photodetectors on SiO2 substrates [22,24–27]. The apparent interaction of graphene with the ambient species, causing asymmetric carrier transport and a large hysteresis, is primarily attributed to the oxygen/water redox couple-induced charge transfer reaction [23,28].
On the other hand, after the Al2O3 passivation followed by a 200 þC post-deposition anneals (PDA), Id-Vg curves became almost symmetric with significantly less hysteresis (Fig. 1(d)). The Dirac point moved to Vg = −6 V ( + 4 V for reverse voltage sweep) due to the self-cleaning effect of the ALD Al2O3 process [29]. As expected, the Al2O3 passivation was effective in protecting the graphene channel from interacting with the ambient species [23].
Finally, the photonic characteristics of the graphene photodetectors were measured by detecting the current change under illumination from a light-emitting diode (LED) lamp (power = 200 μW/cm2 at 365 nm wavelength). Figure 2(a) shows the conductance modulation of air-exposed graphene under UV illumination. When the UV lamp was turned on, the drain current steeply increased by ~100 nA with a ~0.3 sec rise time due to the generation of a photocurrent, but gradually decreased even below the dark current level with a tens or hundreds of seconds time constant [12,13]. As the illumination cycle was repeated, similar responses were observed, but the baseline of the drain current continued to decrease. The steep current increase (ΔIP) at the beginning of the illumination cycle is due to the electron hole pair (EHP) generated in the graphene channel [17,30]. The gradual decrease of the drain current under illumination is attributed to the gradual shift of the charge neutrality level caused by photodesorption (ΔIPD) and re-adsorption (ΔIR) of molecular species on the surface of graphene channel [12]. When the graphene channel region was passivated with Al2O3, however, a photo-response became very abrupt ~200 ms of falling time and a baseline shift was negligible as shown in Fig. 2(b). Interestingly, the drain current decreased when the UV lamp was turned on. This behavior is opposite of the photo-response of the graphene exposed to air ambient. The mechanism changing the direction of the photocurrent before and after Al2O3 passivation is explained below.
Fig. 2 Photo-response of (a) an air-exposed graphene photodetector and (b) a Al2O3-passivated graphene photodetector with a UV on/off cycle of 20 seconds (200 μW/cm2 power and 365 nm wavelength UV lamp was used and the graphene photodetectors were biased with Vd = 10 mV and Vg = 0 V).
To explain the origin and direction of the photocurrent (ΔIP) in our devices, band diagrams of graphene photodetectors with and without an Al2O3 passivation layer are presented in Figs. 3(a)-3(c). The black dashed line represents the Fermi level, EF, while the solid black line denotes the charge neutrality level. ΔΦ represents the EF of graphene pinned by the contact metal [30]. ΔΦ is ~0.2 eV at the graphene in contact with Au electrode [31]. The band alignment and the direction of band bending are determined by the difference between EF of the metal and bulk graphene, ΔΦ [32,33].
Fig. 3 Band profile of the graphene photodetector (a) exposed to air ambient without illumination, (b) exposed to air ambient under illumination and (c) after Al2O3 passivation under illumination. ΔΦ (ΔE) denotes the Fermi level shift of graphene due to metal-induced doping (external doping). Shaded area adjacent to the electrode is the transition region (LTS, LTD), black (open) circle denotes electron (hole), arrow denotes the direction in which the carrier is moving, and () means the direction of the drain (source) side photocurrent. (d) The photocurrent of Al2O3 passivated graphene photodetector measured while modulating the back gate bias with 100 mV of drain bias. (e) Schematic illustration of carrier generation and recombination, photodesorption and re-adsorption of an air-exposed area and Al2O3 passivated area of graphene photodetector under illumination. Red (blue) circle denotes electron (hole). Only photo generated carriers within the drift distance from the electrode can contribute to the photocurrent. UV-induced photodesorption occurred at the air-exposed graphene photodetector resulting hole extraction from the graphene.
Before the Al2O3 passivation, the EF of the graphene channel is in the lower cone as shown in Fig. 3(a) due to the hole doping effect. Since the Dirac point was located at + 15 V (30 V for reverse voltage sweep) before the Al2O3 passivation, the density of hole in the channel region is ~3.7 x 1012 /cm2, calculated using the net charge Eq. (1) [34]:
where εε0 and tox are the permittivity and thickness of the gate dielectric, e is the charge of the electron, and V0 = 0 V. This hole carrier density can be translated to the Fermi level of the graphene using Eq. (2) [20]:where g(E) = 2|E-Eig|/πħ2vF2 is the graphene DOS, Eig is the intrinsic graphene Fermi level, E is the initial doping level, and vF is the Fermi velocity. A hole density of 3.7 x 1012 /cm2 corresponds to ΔEF ~0.23 eV as shown in Fig. 3(a). The positive ΔEF means that the graphene is doped as a p-type, i.e., populated with holes. At this stage, ΔEF was reduced during the illumination cycle due to a photo induced molecular desorption which reduces the concentration of holes (Fig. 3(b)). On the other hand, VDirac = −6 V after the Al2O3 passivation is equivalent to ΔEF ~-0.14 eV, implying that the graphene is weakly doped as an n-type and the band structure is changed as shown in Fig. 3(c).When graphene is illuminated, photo excited electron-hole pairs (EHPs) are continuously generated, but they recombine within picoseconds due to small or zero bandgap [35,36] and only a few EHPs generated near metal contacts can be transferred to the metal contacts. Depending on the direction of the band bending in the transition region at the boundary of the metal contact and the graphene channel, the directions of electron and hole drift are changed. When the graphene is doped as a p-type, holes generated by UV excitation gather towards the center of the channel as shown in Fig. 3(b). The amount of the net photocurrent is then determined by the difference between the . Theoretically, a graphene photodetector having exactly same source/drain metal contacts has a symmetric band alignment and cannot generate a photocurrent without a drain bias because the photocurrents generated at both edges of the source and drain cancel each other out [32]. In this study, a small drain bias was applied to increase the field gradient of the transition region at the drain side. When a drain bias was applied, the potential gradient at the drain side gets steeper and the transition region gets narrower, resulting in faster drift at the drain side than the source side, i.e., more photocurrent is generated at the drain side.
When the graphene channel in air ambient is illuminated with UV light, a photo-induced molecular desorption that decreases the hole density in the graphene occurs. As a result, the drain current increases at the beginning of UV illumination, but exponentially decreases due to the reduction of hole concentration induced by the photodesorption of oxygen/water combined adsorbates [12–14]. When the UV illumination is turned off, the EF of the graphene gradually recovers to the hole doping state due to the re-adsorption of p-type doping species and the drain current gradually increases as shown in Fig. 2(a). On the other hand, an intrinsic photocurrent can be measured after the passivation because the photodesorption (ΔIPD) and re-adsorption (ΔIR) were almost eliminated. As a result, the photo-response time dramatically improved from tens or hundreds of seconds to less than 200 ms. Under UV illumination, the net photocurrent () flows from the source to the drain because the is larger than. The direction of the intrinsic photocurrent is opposite to the direction of the drift current with a positive drain bias because the direction of the band bending changes after the passivation. Thus, the net current of Al2O3-passivated graphene photodetectors decreases under UV illumination.
Above explanation on the direction of photocurrent was experimentally confirmed by emulating the modulation of the charge neutrality level using a back gate bias as shown in Fig. 3(d). As the Fermi level of graphene moves upward by modulating the back gate bias, the direction of net photocurrent also changes from positive to negative side [17,37]. Here, the state I shown in Fig. 3(d) is similar to the air exposed graphene photodetector and the state IV is similar to the Al2O3 passivated graphene photodetector. No net photocurrent was observed at ~-5 V. Zero photocurrent bias is determined by the difference between metal induced doping level (ΔΦ) and external doping level (ΔE) of graphene. It might be possible to use this value to determine the band alignment at metal-graphene interface.
Figure 3(e) graphically summarizes the various mechanisms associated with the photocurrent generation. Only the photo carriers generated near the metal contact can contribute to the photocurrent. The carrier life time of exfoliated graphene is in picoseconds order [35] and that of CVD graphene is even in sub picosecond order [36]. Thus, the travel distance during the carrier lifetime is very limited [35].
Since the intrinsic photocurrent of graphene was successfully measured with minimal influence from the photodesorption, sensitivity and response speed were examined systematically to develop a guideline for future photodetector designs. Figure 4(a) shows the power dependence of the intrinsic photocurrent generation. As the incident power of UV light increases from 20 to 200 μW/cm2 using a neutral density filter, the photocurrent, ΔIP, increases with incident power, P0.2. The power law exponent of ~0.2 is much lower than 0.6 for a highly ordered pyrolytic graphite (HOPG) photodetector [33], 0.3~0.8 for a ZnO photodetector [38,39], and 1 for a CNT photodetector [40]. The low exponent of our device seems to be due to the high defect density of the CVD graphene and the symmetric band profile of our device [32,33].
Fig. 4 (a) Variation in photocurrent generation as a function of the illumination intensity. (b) 1Hz and 20-second on/off characteristics of graphene photodetectors. The inset is on/off characteristics at 1Hz that magnifies the red rectangle in (b). (c) Photocurrent generation in an Al2O3 passivated graphene measured at different channel width, (d) Photocurrent generation in an Al2O3 passivated graphene measured at different length. Power = 200 μW/cm2, wavelength = 365nm, and Vd = 100 mV were used at for (c) and (d) measurement.
Figure 4(b) shows on/off characteristics of graphene photodetectors at 1 Hz and 20 second cycles. The inset shows the modulation of the photocurrent at 1 Hz. Even though the operation speed was limited by the speed of the chopper system, the operation speed of this device seems to be at least at 15 Hz as the rise time of photocurrent was 70 ms. The operation speed of our device can be enhanced if the modulation amplitude of the photocurrent is reduced further. While there are several cases reporting very high speed photocurrent modulation up to 40 GHz, those are measured while modulating the intensity of light. The on-off speed of photocurrent has been orders of magnitude slower than the light modulation speed. Especially, the on-off speed of UV photodetector was severely limited by the slow photodesorption. Our device showed a few hundred times faster on-off operation speed compared to the previously reported UV photodetectors [12,13,16]. Also, there was no baseline shift even after a long period of illumination as the influence of the photodesorption is minimized.
Finally, the photocurrents of devices with various channel widths and lengths were measured at different drain biases to investigate the photocurrent generation mechanism further (Figs. 4(c) and 4(d)). Since the photocurrent is primarily generated at the transition region near the graphene–metal boundary, the intrinsic photocurrent increases as the channel gets wider (Fig. 4(c)). On the other hand, the photocurrent decreases as the channel gets longer even though the width of the region generating EHP is the same (Fig. 4(d)). The reduced photocurrent in long channel devices can be explained by a reduction in the field gradient at the transition region due to a lower effective drain bias. However, as the drain bias increases, the photocurrent likewise increases as the source/drain transition region becomes more asymmetrical.
The photocurrent of graphene photodetectors can be described using the concept of a metal-semiconductor-metal (MSM) Schottky-type photodetector. The photocurrent generation in a MSM Schottky photodetector is described using Eq. (3) [41]:
where q is the electron charge, G is the carrier generation rate, d1 (d2) is the source (drain) side depletion width, Lp is hole diffusion length, x2-x1 is the bulk region (channel length subtracted by the depletion region of both sides), and f is the effective diffusion length. The first term of Eq. (3) is related to the photocurrent generation at the source/drain depletion region, and the other terms are related to the diffusion term around the boundary of the depletion region and bulk. The depletion region (i.e. transition region) of metal-graphene junction is reported around 200 ~300 nm. The carrier transit time to metal contact will increase to 1.5 ~3.2 ps for 200 ~300 nm (transit time, τt = l2/(µΔV) [35], µ = 1,000 cm2/(Vs), ΔV = 340 mV). Since the carrier life time of CVD graphene is only ~0.5 ps [36], most of photo carriers generated beyond this limit cannot reach the electrode. Even some of the carriers generated in the transition region cannot reach the metal electrode with this short carrier lifetime. As a result, most of the photocurrent generated in channel region can be ignored and Eq. (3) can be simplified to Eq. (4).The depletion width, d1 and d2, should be replaced with v1τ and v2τ, representing the maximum drift distance of the generated carrier. Here, v1 (v2) is the source (drain) side drift velocity and τ is the carrier life time of the graphene. Since the graphene is 2D material, the photocurrent Iph is expressed with J multiplied by graphene channel width Weff equivalent to width × thickness of 3D material. Therefore, Eq. (4) can be modified for graphene photodetectors with channel width Weff as follows:where α is the fitting parameter representing the ratio of the photo carriers survived to the metal contact and the photo carriers generated in the photo active region, Weff is the width of the effective photocurrent generation region at the graphene and metal boundary, and G is the generation rate of the electron hole pair. G can be calculated using γMPopt/EphA where γ is the absorption coefficient, M is the carrier multiplication factor of graphene, Popt is the power of the incident light, Eph is the energy of the incoming photon, and A is the area of the device [37]. Equation (5) simply represents the net photocurrent of the graphene which is the difference between the photocurrent generated at the transition regions of source and drain sides, respectively. Experimental data shown in Fig. 4(c) and 4(d) were successfully modeled using M = 5 [5], τ = 0.5 ps [36], and G = 6 x 1020 /cm2s. α matching with the experimental data was around 16 ~37% [17,35]. For W = 9 µm, L = 5 µm, Vd = 10 mV, v1 = 2.3 x 107 cm/s, v2 = 3.4 x 107 cm/s, Iph was 141 nA. Calculated photocurrent value matches well with the experimental Iph of 150 nA, indicating our assumption on the physical parameters was reasonable. Here, v was calculated from µE where E = ∆V/d, E is electric field, ∆V is potential drop and d is distance. According to this model, the sensitivity of graphene photodetectors can be improved using a wide device and an asymmetric band profile, which can be achieved using an external bias or different metal contacts for the source and drain. Since the channel region does not contribute to the photocurrent generation, the channel length should be minimized to 2vτ in the order of several hundred nanometers.4. Summary
In summary, a strong photodesorption-induced sensitivity drift has been one of the major obstacles in using graphene for photodetector applications. This problem has been solved using an Al2O3 passivation layer. Intrinsic photocurrent generation in graphene with high speed on-off characteristics has been achieved by suppressing the photodesorption current. Also, carrier mobility was improved by removing traps in graphene [23] and dark current was suppressed after the Al2O3 passivation. We also proposed and examined the simple drift current model to describe the intrinsic photocurrent in graphene photodetectors. Since the photocurrent was primarily generated at the metal-graphene boundary, narrow interdigitated electrodes with a gap of 2 × vτ in the order of several hundred nanometers are suggested as an optimal structure for CVD graphene photodetectors.
Acknowledgments
This work was supported by the inter-ER cooperation projects funded by the MOTIE/KIAT and by the industrial strategic technology development program (10039174) of MOTIE, Korea.
References and links
1. R. R. Nair, P. Blake, A. N. Grigorenko, K. S. Novoselov, T. J. Booth, T. Stauber, N. M. Peres, and A. K. Geim, “Fine structure constant defines visual transparency of graphene,” Science 320(5881), 1308 (2008). [CrossRef] [PubMed]
2. Q. Bao, H. Zhang, Y. Wang, Z. Ni, Y. Yan, Z. X. Shen, K. P. Loh, and D. Y. Tang, “Atomic-layer graphene as a saturable absorber for ultrafast pulsed lasers,” Adv. Funct. Mater. 19(19), 3077–3083 (2009). [CrossRef]
3. Q. Bao and K. P. Loh, “Graphene photonics, plasmonics, and broadband optoelectronic devices,” ACS Nano 6(5), 3677–3694 (2012). [CrossRef] [PubMed]
4. F. Wang, Y. Zhang, C. Tian, C. Girit, A. Zettl, M. Crommie, and Y. R. Shen, “Gate-variable optical transitions in graphene,” Science 320(5873), 206–209 (2008). [CrossRef] [PubMed]
5. T. Winzer, A. Knorr, and E. Malic, “Carrier multiplication in graphene,” Nano Lett. 10(12), 4839–4843 (2010). [CrossRef] [PubMed]
6. K. J. Tielrooij, J. C. W. Song, S. A. Jensen, A. Centeno, A. Pesquera, A. Zurutuza Elorza, M. Bonn, L. S. Levitov, and F. H. L. Koppens, “Photoexcitation cascade and multiple hot-carrier generation in graphene,” Nat. Phys. 9(4), 248–252 (2013). [CrossRef]
7. K. I. Bolotin, K. J. Sikes, Z. Jiang, M. Klima, G. Fudenberg, J. Hone, P. Kim, and H. L. Stormer, “Ultrahigh electron mobility in suspended graphene,” Solid State Commun. 146(9–10), 351–355 (2008). [CrossRef]
8. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, M. I. Katsnelson, I. V. Grigorieva, S. V. Dubonos, and A. A. Firsov, “Two-dimensional gas of massless Dirac fermions in graphene,” Nature 438(7065), 197–200 (2005). [CrossRef] [PubMed]
9. F. Bonaccorso, Z. Sun, T. Hasan, and A. C. Ferrari, “Graphene photonics and optoelectronics,” Nat. Photonics 4(9), 611–622 (2010). [CrossRef]
10. P. Avouris, “Graphene: electronic and photonic properties and devices,” Nano Lett. 10(11), 4285–4294 (2010). [CrossRef] [PubMed]
11. S. Assefa, S. Shank, W. Green, M. Khater, E. Kiewra, C. Reinholm, S. Kamlapurkar, A. Rylyakov, C. Schow, F. Horst, H. Pan, T. Topuria, P. Rice, D. M. Gill, J. Rosenberg, T. Barwicz, M. Yang, J. Proesel, J. Hofrichter, B. Offrein, X. Gu, W. Haensch, J. Ellis-Monaghan, and Y. Vlasov, “A 90nm CMOS integrated nano-photonics technology for 25Gbps WDM optical communications applications,” in Electron Devices Meet. Iedm 2012 IEEE Int., 33.8.1–33.8.3 (2012) [CrossRef]
12. Y. Shi, W. Fang, K. Zhang, W. Zhang, and L.-J. Li, “Photoelectrical response in single-layer graphene transistors,” Small 5(17), 2005–2011 (2009). [CrossRef] [PubMed]
13. P. Sun, M. Zhu, K. Wang, M. Zhong, J. Wei, D. Wu, Y. Cheng, and H. Zhu, “Photoinduced molecular desorption from graphene films,” Appl. Phys. Lett. 101(5), 053107 (2012). [CrossRef]
14. Z. Luo, N. J. Pinto, Y. Davila, and A. T. Charlie Johnson, “Controlled doping of graphene using ultraviolet irradiation,” Appl. Phys. Lett. 100(25), 253108 (2012). [CrossRef]
15. R. J. Chen, N. R. Franklin, J. Kong, J. Cao, T. W. Tombler, Y. Zhang, and H. Dai, “Molecular photodesorption from single-walled carbon nanotubes,” Appl. Phys. Lett. 79(14), 2258–2260 (2001). [CrossRef]
16. M. Kim, N. S. Safron, C. Huang, M. S. Arnold, and P. Gopalan, “Light-driven reversible modulation of doping in graphene,” Nano Lett. 12(1), 182–187 (2012). [CrossRef] [PubMed]
17. J. Park, Y. H. Ahn, and C. Ruiz-Vargas, “Imaging of photocurrent generation and collection in single-layer graphene,” Nano Lett. 9(5), 1742–1746 (2009). [CrossRef] [PubMed]
18. X. Li, W. Cai, J. An, S. Kim, J. Nah, D. Yang, R. Piner, A. Velamakanni, I. Jung, E. Tutuc, S. K. Banerjee, L. Colombo, and R. S. Ruoff, “Large-area synthesis of high-quality and uniform graphene films on copper foils,” Science 324(5932), 1312–1314 (2009). [CrossRef] [PubMed]
19. A. Reina, X. Jia, J. Ho, D. Nezich, H. Son, V. Bulovic, M. S. Dresselhaus, and J. Kong, “Large area, few-layer graphene films on arbitrary substrates by chemical vapor deposition,” Nano Lett. 9(1), 30–35 (2009). [CrossRef] [PubMed]
20. P. L. Levesque, S. S. Sabri, C. M. Aguirre, J. Guillemette, M. Siaj, P. Desjardins, T. Szkopek, and R. Martel, “Probing charge transfer at surfaces using graphene transistors,” Nano Lett. 11(1), 132–137 (2011). [CrossRef] [PubMed]
21. S. Ryu, L. Liu, S. Berciaud, Y.-J. Yu, H. Liu, P. Kim, G. W. Flynn, and L. E. Brus, “Atmospheric oxygen binding and hole doping in deformed graphene on a SiO2 substrate,” Nano Lett. 10(12), 4944–4951 (2010). [CrossRef]
22. C. M. Aguirre, P. L. Levesque, M. Paillet, F. Lapointe, B. C. St-Antoine, P. Desjardins, and R. Martel, “The role of the oxygen/water redox couple in suppressing electron conduction in field-effect transistors,” Adv. Mater. 21(30), 3087–3091 (2009). [CrossRef]
23. C. G. Kang, Y. G. Lee, S. K. Lee, E. Park, C. Cho, S. K. Lim, H. J. Hwang, and B. H. Lee, “Mechanism of the effects of low temperature Al2O3 passivation on graphene field effect transistors,” Carbon 53, 182–187 (2013). [CrossRef]
24. K. Nagashio, T. Yamashita, T. Nishimura, K. Kita, and A. Toriumi, “Electrical transport properties of graphene on SiO2 with specific surface structures,” J. Appl. Phys. 110(2), 024513 (2011). [CrossRef]
25. T. Lohmann, K. von Klitzing, and J. H. Smet, “Four-terminal magneto-transport in graphene p-n junctions created by spatially selective doping,” Nano Lett. 9(5), 1973–1979 (2009). [CrossRef] [PubMed]
26. S. S. Sabri, P. L. Lévesque, C. M. Aguirre, J. Guillemette, R. Martel, and T. Szkopek, “Graphene field effect transistors with parylene gate dielectric,” Appl. Phys. Lett. 95(24), 242104 (2009). [CrossRef]
27. M. Lafkioti, B. Krauss, T. Lohmann, U. Zschieschang, H. Klauk, K. V. Klitzing, and J. H. Smet, “Graphene on a hydrophobic substrate: doping reduction and hysteresis suppression under ambient conditions,” Nano Lett. 10(4), 1149–1153 (2010). [CrossRef] [PubMed]
28. Y. G. Lee, C. G. Kang, C. Cho, Y. Kim, H. J. Hwang, and B. H. Lee, “Quantitative analysis of hysteretic reactions at the interface of graphene and SiO2 using the short pulse I–V method,” Carbon 60, 453–460 (2013). [CrossRef]
29. C. L. Hinkle, A. M. Sonnet, E. M. Vogel, S. McDonnell, G. J. Hughes, M. Milojevic, B. Lee, F. S. Aguirre-Tostado, K. J. Choi, H. C. Kim, J. Kim, and R. M. Wallace, “GaAs interfacial self-cleaning by atomic layer deposition,” Appl. Phys. Lett. 92(7), 071901 (2008). [CrossRef]
30. F. Xia, T. Mueller, R. Golizadeh-Mojarad, M. Freitag, Y. M. Lin, J. Tsang, V. Perebeinos, and P. Avouris, “Photocurrent imaging and efficient photon detection in a graphene transistor,” Nano Lett. 9(3), 1039–1044 (2009). [CrossRef] [PubMed]
31. G. Giovannetti, P. A. Khomyakov, G. Brocks, V. M. Karpan, J. van den Brink, and P. J. Kelly, “Doping graphene with metal contacts,” Phys. Rev. Lett. 101(2), 026803 (2008). [CrossRef] [PubMed]
32. F. Xia, T. Mueller, Y. M. Lin, A. Valdes-Garcia, and P. Avouris, “Ultrafast graphene photodetector,” Nat. Nanotechnol. 4(12), 839–843 (2009). [CrossRef] [PubMed]
33. T. Mueller, F. Xia, and P. Avouris, “Graphene photodetectors for high-speed optical communications,” Nat. Photonics 4(5), 297–301 (2010). [CrossRef]
34. H. E. Romero, N. Shen, P. Joshi, H. R. Gutierrez, S. A. Tadigadapa, J. O. Sofo, and P. C. Eklund, “n-Type behavior of graphene supported on Si/SiO2 substrates,” ACS Nano 2(10), 2037–2044 (2008). [CrossRef] [PubMed]
35. A. Urich, K. Unterrainer, and T. Mueller, “Intrinsic response time of graphene photodetectors,” Nano Lett. 11(7), 2804–2808 (2011). [CrossRef] [PubMed]
36. K.-J. Yee, J.-H. Kim, M. H. Jung, B. H. Hong, and K.-J. Kong, “Ultrafast modulation of optical transitions in monolayer and multilayer graphene,” Carbon 49(14), 4781–4785 (2011). [CrossRef]
37. M. Freitag, T. Low, F. Xia, and P. Avouris, “Photoconductivity of biased graphene,” Nat. Photonics 7(1), 53–59 (2012). [CrossRef]
38. H. Kind, H. Yan, B. Messer, M. Law, and P. Yang, “Nanowire ultraviolet photodetectors and optical switches,” Adv. Mater. 14(2), 158–160 (2002). [CrossRef]
39. W. Park, G. Jo, W.-K. Hong, J. Yoon, M. Choe, S. Lee, Y. Ji, G. Kim, Y. H. Kahng, K. Lee, D. Wang, and T. Lee, “Enhancement in the photodetection of ZnO nanowires by introducing surface-roughness-induced traps,” Nanotechnology 22(20), 205204 (2011). [CrossRef] [PubMed]
40. J. Zhang, N. Xi, H. Chen, K. W. C. Lai, G. Li, and U. C. Wejinya, “Design, manufacturing, and testing of single-carbon-nanotube-based infrared sensors,” IEEE Trans. NanoTechnol. 8(2), 245–251 (2009). [CrossRef]
41. L. C. Liou and B. Nabet, “Simple analytical model of bias dependence of the photocurrent of metal-semiconductor-metal photodetectors,” Appl. Opt. 35(1), 15–23 (1996). [CrossRef] [PubMed]
