- Email: [email protected]
A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition
G Model
ARTICLE IN PRESS
APSUSC-34892; No. of Pages 6
Applied Surface Science xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Full Length Article
A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition S. Chandramohan a,b,∗ , Tae Hoon Seo c , V. Janardhanam a , Chang-Hee Hong a , Eun-Kyung Suh a,∗ a Department of Semiconductor Science and Technology, Semiconductor Physics Research Center, Chonbuk National University, Jeonju 54896, Republic of Korea b Department of Physics & Nanotechnology, SRM University, Kattankulathur 603 203, Tamil Nadu, India c Soft Innovative Materials Research Center, Korea Institute of Science and Technology, Jeonbuk 55324, Republic of Korea
a r t i c l e
i n f o
Article history: Received 17 October 2016 Received in revised form 9 January 2017 Accepted 11 January 2017 Available online xxx Keywords: Graphene Charge transfer doping Molybdenum trioxide Rapid thermal annealing
a b s t r a c t Charge transfer doping is a renowned route to modify the electrical and electronic properties of graphene. Understanding the stability of potentially important charge-transfer materials for graphene doping is a crucial first step. Here we present a systematic comparison on the doping efficiency and stability of single layer graphene using molybdenum trioxide (MoO3 ), gold chloride (AuCl3 ), and bis(trifluoromethanesulfonyl)amide (TFSA). Chemical dopants proved to be very effective, but MoO3 offers better thermal stability and device fabrication compatibility. Single layer graphene films with sheet resistance values between 100 and 200 ohm/square were consistently produced by implementing a two-step growth followed by doping without compromising the optical transmittance. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Since its discovery in 2004, graphene has become the most studied material owing to its two-dimensionality and unique band structure, which is fundamentally responsible for the remarkable electro-optical properties [1]. Especially, the superior carrier transport properties and exceptionally high optical transparency unified in an atomically thin sheet make it promising as a transparent conductive electrode (TCE) in place of brittle TCE of common use such as indium tin oxide for applications in a range of optoelectronic devices and flexible touch screens [2–8]. However, the low intrinsic carrier concentration and defects generated during the growth and transfer processes are known limiting the high conductivity requirement for practical applications [9,10], a key problem averting graphene’s progress to budge from the laboratory to industry. While integration of graphene with already established flexible TCE materials like metal nanowires seems to be an appealing solution [11], this approach has limited relevance when it is necessary to
∗ Corresponding authors at: Department of Semiconductor Science and Technology, Semiconductor Physics Research Center, Chonbuk National University, Jeonju, 54896, Republic of Korea. Department of Physics and Nanotechnology, SRM University, Kattankulathur 603 203, Tamil Nadu, India. E-mail addresses: [email protected] (S. Chandramohan), [email protected] (E.-K. Suh).
modulate the work function for specific device applications. Alternatively, a renowned charge transfer doping [12] has become the universal approach to realize low sheet resistance with a simultaneous work function regulation by exploiting the unprecedented charge-transfer interaction property of graphene i.e., being atomically thin with a vanishing density of states at the Dirac point, the Fermi level is very sensitive to external perturbations such as adsorption of metals, molecules, etc. Indeed, experiments have demonstrated that charge doping of graphene has the potential in achieving low sheet resistance values down to 30 ohm/square [2,4]. Following the pioneering work by Bae et al.[2], interest has arisen in developing doped graphene TCE and in recent years numerous studies have been conducted using different chemicals [5,12–15], strong acids [4,6], and organic molecules [16–18]. Of various p-type chemical dopants explored to date, metal chlorides are known to give the maximum doping efficiency [2,4,13] whereas the bis(trifluoromethanesulfonyl)amide (TFSA) provides better stability under normal ambient conditions [16]. However, majority of the chemical dopants including metal chlorides and TFSA on graphene surface have been found resoluble in water or organic solvents [19], inhibiting their use in device manufacturing that involves lithography and etching processes. Recently, few studies showed that relatively stable doping of graphene with sheet resistances below 20 ohm/square can be achievable through FeCl3 intercalation [19,20], but it depreciates the optical transmittance drastically.
http://dx.doi.org/10.1016/j.apsusc.2017.01.097 0169-4332/© 2017 Elsevier B.V. All rights reserved.
Please cite this article in press as: S. Chandramohan, et al., A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.01.097
G Model APSUSC-34892; No. of Pages 6
ARTICLE IN PRESS S. Chandramohan et al. / Applied Surface Science xxx (2017) xxx–xxx
2
On the other hand, high work function transition metal oxides have been recognized as potential candidates for p-type doping of graphene and carbon nanotubes [21–27]. The advantage of using metal oxides is that they can be simply deposited onto graphene as thin film coating to not only dope graphene effectively without compromising the optical transparency [23], but also to provide graphene a conformal protective layer for integration in device technology [25]. Especially, molybdenum trioxide (MoO3 ), a wide band gap material with unusually large electron affinity of 6.8 eV, has been broadly studied for graphene doping in organic electronic devices [23–25]. Unlike chemical dopants, the MoO3 is demonstrated to be stable to environment [24], chemical [21], and high temperatures [22,26]. The doping mechanism is also studied to some extent through spectroscopic investigations for both stoichiometric MoO3 [23] and substoichiometric MoOx [28]. Majority of the previous studies considered growth and in situ evaluation of doping in high vacuum conditions and a major issue yet to be resolved is the instability of MoO3 surface potential to air exposure and its impact on the doping efficiency and stability for long time exposure to environment. Despite promising results, disagreement exists between studies reported by different groups on the doping stability against air exposure. Therefore, understanding of the stability of potentially important charge-transfer materials for graphene doping is a crucial first step. Herein we give a systematic comparison on the doping efficiency and stability of single layer graphene doped by MoO3 , AuCl3 , and TFSA with respect to various factors such as the surface chemistry of graphene (amount of polymer residues), ambient exposure, and dopant morphology.
2. Materials and methods Graphene used in this study was synthesized on electrochemically polished copper foil by low pressure CVD. Single layer graphene with reduced defect density aided by a two-step growth process was grown by using methane as a carbon precursor [29]. The transfer of graphene to arbitrary substrates (Al2 O3 and SiO2 /Si) was performed by etching the Cu in ammonium persulphate after spin coating a thin polymethyl methacrylate (PMMA) layer on the surface of graphene as a protective layer. The detached PMMAsupported graphene was then washed in deionized water several times and transferred to the target substrates. The PMMA was dissolved in acetone followed by annealing at 400 ◦ C for 600 s in a rapid thermal annealing system. In our experiments, the temperature ramp up and cool down times were 30 s and 300–600 s, respectively. The cool down was done under the flow of 10 slm N2 until the temperature reduces below 50 ◦ C. The MoO3 (99.98%, Aldrich) deposition was carried out in a thermal evaporator under a base pressure of 10−6 Torr. The deposition rate was below 0.5 Å/s in most cases. For comparison, both AuCl3 and TFSA are employed in our study, because these two materials in solution form have been shown very effective for the hole doping of graphene. Doping of graphene with AuCl3 and TFSA was performed by dissolving the respective materials in nitromethane and spin-coating the solution on the surface of graphene. The three different doping processes are illustrated in Fig. 1. The optical transmittance measurements were carried out by using a JASCO V-570 ultraviolet-visible-near-infrared (UV–vis-NIR) spectrophotometer. Raman spectra of graphene were obtained by using a 514 nm line of an argon ion laser. The resistivity measurements were carried out using a Hall Effect measurement system at room temperature with a magnetic field of 0.556 T. The surface topography was probed by atomic force microscopy in tapping mode. X-ray photoelectron spectroscopy (XPS) measurements were performed on a Thermo Scientific K-alpha instrument with a monochromized Al K␣ x-ray source (1486.6 eV). Ultraviolet
photoelectron spectroscopy (UPS) measurements were performed on an AXIS Ultra DLD spectrometer where He I discharge lamp (h = 21.22 eV) was used as an excitation source.
3. Results and discussion The opto-electrical characteristics of graphene films before and after doping with different materials are evaluated. Fig. 2(a) shows the optical transmittance spectra taken on 15 mm2 graphene films doped with 10 mM AuCl3 , 20 mM TFSA, and 5 nm MoO3 . It can be seen that the optical transmittance of graphene is not deteriorated at moderate solution concentrations (for TFSA and AuCl3 cases) or MoO3 film thickness. The dependence of optical transmittance with respect to MoO3 film thickness is shown in Fig. 2(b). It is observed that the MoO3 /graphene hybrid possess high transparency exceeding 90% at all the MoO3 film thickness studied from 1 to 20 nm. In addition, the maximum percentage of reduction in transmittance is estimated to be only 3.7% at 550 nm, corresponding to the MoO3 film thickness of 20 nm. On the other hand, the optical transmittance of the AuCl3 doped graphene films is known to have a strong dependence on the dopant concentration such that increasing the AuCl3 concentration leads to increasing density of Au nanoparticles on the graphene surface [13,30]. In our study, 10 mM AuCl3 and 20 mM TFSA solutions are considered optimum because the maximum decrease in sheet resistance can be achieved at those optimum concentrations [13,31], which will be shown later. Fig. 2(c) shows the sheet resistance values of graphene measured before and immediately after the MoO3 deposition for different film thicknesses. The measurements were performed on graphene transferred to SiO2 /Si substrates with van der pauw contact geometry under normal ambient conditions. Some of the graphene films employed in this study were subjected to rapid thermal annealing (RTA) at 400 ◦ C in an inert ambient for 10 min in order to eliminate the PMMA residues prior to the deposition of MoO3 (the respective data are marked with an asterisk symbol in the figure). As can be seen in Fig. 2(c), the deposition of MoO3 lowers the sheet resistance of graphene to values nearly half to that of its initial values for MoO3 thickness above 3 nm. The results bring into picture that the effect of film thickness is insignificant beyond this optimal thickness (3 nm) at which a maximum decrease in sheet resistance of 56% is observed. This result is consistent with recent findings that the change in sheet resistance of graphene is approximately 50% and saturation occurs for 3–4 nm MoO3 [23]. In addition, for majority of annealed samples (sample #4–7), the decrease in sheet resistance after MoO3 deposition is estimated to vary between 43% and 50%. It is known that air exposure of graphene could result in change in the electronic properties of graphene and therefore the actual doping efficacy can vary from sample to sample if the experimental conditions are not invariant in all the cases. To avoid the inconsistency, we carried out similar experiments in a clean room condition (samples 9 and 10) and prior to MoO3 deposition the samples were subjected to mild heat treatment at 150 ◦ C. In this case, the doping efficacy is estimated to be 67% and 56%, respectively, for the pristine and annealed graphene. It is apparent that the doping efficacy has increased for graphene subjected to mild heat treatment prior to the deposition of MoO3 and the doping efficacy of annealed graphene is relatively low compared to pristine graphene. Similar experiments were performed for AuCl3 and TFSA doping and the results are shown in Fig. 2(d). In this case also annealed films showed relatively low doping efficacy compared to pristine graphene. Residual polymer is known to impact the electronic properties of graphene and many studies have been reported in the past demonstrating that residual polymer leads to p-type doping and carrier mobility degradation in graphene transferred to foreign
Please cite this article in press as: S. Chandramohan, et al., A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.01.097
G Model APSUSC-34892; No. of Pages 6
ARTICLE IN PRESS S. Chandramohan et al. / Applied Surface Science xxx (2017) xxx–xxx
3
Fig. 1. Schematic showing the doping of graphene using MoO3 , TFSA and AuCl3 .
Fig. 2. (a, b) Optical transmittance spectra of pristine graphene and graphene coated with different dopants. (c) Sheet resistance of graphene films before and after the deposition of MoO3 film of various thicknesses. (d) Sheet resistance of graphene films before and after the deposition of AuCl3 and TFSA. * indicates the samples subjected to annealing prior to doping.
substrates [32–34]. Thermal annealing has been considered as a promising route to reduce the polymer residues. In our study, the rapid thermal annealing of graphene is found to reduce the amount of polymer residue, which can be clearly envisaged by comparing Fig. 3(a) and (b). Here complete elimination of the polymer residue did not happen which could be likely due to annealing being conducted for a short time. But, annealing caused strong p-doping in graphene (will be discussed later under Raman analysis) complicating the interpretation of the results. Since maximum doping efficiency is observed for the graphene subjected to no prior thermal annealing, one can assume that the residual polymer has little influence on the doping efficacy of graphene. On the other hand, the doping efficacy is found to be a strong function of surface coverage and initial graphene quality. Figs. 3(c)–(e) show the surface topography of representative MoO3 coated graphene films. Figs. 3(c) and (d) correspond to annealed and pristine graphene films, respectively, after being coated with a 3 nm MoO3 . It can be seen that the surface of annealed graphene with a 3 nm MoO3 coating is very smooth with a root mean square roughness (Rq ) of 0.264 nm. On the other hand, the surface of the pristine graphene is relatively rough (Rq = 0.992 nm)
due to the presence of large amount of residual PMMA. Note that these two samples were prepared under different conditions; so the surface morphology of the samples differed significantly. However, its influence on the extent of graphene doping by MoO3 deposition is negligible because the relative change in sheet resistance for this sample is estimated to be 53%, which is comparable to other samples. Conversely, the doping efficiency largely depends on the surface coverage of MoO3, which likely depends on the surface chemistry of graphene. As shown in Fig. 3(e), the deposition of 11 nm MoO3 on a defective graphene (ID /IG ∼ 0.6) forms nanodots and the relative change in sheet resistance for this sample is estimated to be only 32%. Similarly, in the case of TFSA and AuCl3 , the doping efficiency largely depends on dopant concentration and surface roughness usually plays no significant role in controlling the sheet resistance. Few recent studies [30,35] demonstrated that for dopants such as AuCl3 , the doping efficiency is limited by the dopant morphology especially at high concentrations due to the formation of large dopant clusters or agglomeration of Au particles. The maximum reduction in the sheet resistance for graphene doped with AuCl3 , TFSA, and MoO3 is found to be 79%, 69%, and 56%, respectively. Thus, it is clear that among the three doping
Please cite this article in press as: S. Chandramohan, et al., A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.01.097
G Model APSUSC-34892; No. of Pages 6
ARTICLE IN PRESS S. Chandramohan et al. / Applied Surface Science xxx (2017) xxx–xxx
4
Fig. 3. (a, b) SEM images of graphene before and after rapid thermal annealing. (c-e) AFM surface topography of MoO3 coated graphene. Table 1 Electrical properties of graphene before and after doping. Dopant
Carrier concentration /cm2
Mobility cm2 /V sec
Sheet resistance ohm/square
Before doping 10 nm MoO3 Before doping 10 mM AuCl3 Before doping 20 mM TFSA RTA
3.295E + 12 2.510E + 13 6.764E + 12 5.773E + 13 7.530E + 12 2.367E + 13 1.068E + 13
1.988E + 03 7.844E + 02 1.396E + 03 6.835E + 02 1.070E + 03 1.094E + 03 1.180E + 03
953 317 661 158 775 241 495
materials studied, the AuCl3 leads to a significant reduction in sheet resistance of 79%, a value comparable to previous results reported by Kim et al. [6]. For graphene doped with TFSA the respective value is 69%, which is similar to the values reported in the literature [16]. Despite the fact that chemical doping is more effective in improving the sheet resistance, its stability with time has been the subject of central issue for device applications. To have a more insight, the sheet resistance of the films is measured over a period of three months during which all the three samples were stored in air atmosphere. Fig. 4 shows variations in sheet resistance values with respect to time for representative graphene films doped with different dopants. It is observed that chemically doped graphene films have better ambient stability, especially the graphene doped with TFSA shows excellent stability over time. In the case of AuCl3 doping, the sheet resistance increased from 158 ohm/square on day 1 to 268 ohm/square on day 90. Contrary to chemical doping, MoO3 coated graphene showed dissimilar behavior. In this case, the sheet resistance first decreased to a minimum value, slightly increased during the first few days, and then remained constant thereafter. Since air exposure is known to affect the stoichiometry and degrade the work function of MoO3 [23,24], the observed increase in sheet resistance suggests that the doping is principally governed by the work function difference between the two materials. In other words, the doping efficiency depends on the electronic properties of MoO3 . It is also noticed that prolonged exposure of graphene to ambient air prior to doping reduces the doping efficacy to values as low as 12%. The electrical parameters such as sheet resistance, carrier mobility, and sheet carrier concentration
of graphene before and after doping with three different dopants are given in Table 1. It can be seen that it is possible to increase the carrier concentration by approximately one order and bring down the sheet resistance of graphene by 79% from its initial value. The change in sheet resistance is a direct consequence of charge transfer induced p-doping in graphene, which is verified by Raman measurements. Fig. 5(a) shows the Raman spectra of graphene before and after the doping process for the three dopant materials studied. The pristine graphene exhibits three characteristic peaks known as D, G, and 2D bands whose positions usually differ with respect to local variations in thickness uniformity and post-transfer annealing. Prior to annealing, the G and 2D bands are positioned at 1586 cm−1 and 2686 cm−1 , respectively; while the disorder-induced band is negligible implying that the graphene is high-quality. This is further supported by a low intensity ratio between D band and G band (ID /IG ∼ 0.06). We note that the graphene is predominantly single layer as the intensity ratio of 2D band to G band (I2D /IG ) is almost two or higher when measured at different spots and the 2D band is symmetric with a peak width of ∼32 cm−1 , a value widely quoted for single layer graphene. A blue shift in the G band together with an increase in the D band intensity is observed for rapid thermally annealed samples (Fig. 5(b)). In addition, an I2D /IG ratio between 1 and 1.5 is consistently observed when measured at different spots on various samples. These results show unintentional p-doping in graphene caused due to annealing, similar to the findings of a recent study that exploited RTA for eliminating the polymer residues on the surface of graphene [36]. We observed similar effects for graphene annealed in a quartz tube
Please cite this article in press as: S. Chandramohan, et al., A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.01.097
G Model APSUSC-34892; No. of Pages 6
ARTICLE IN PRESS S. Chandramohan et al. / Applied Surface Science xxx (2017) xxx–xxx
5
Fig. 6. Ultraviolet photoelectron spectra of pristine and doped graphene on SiO2 /Si substrate.
Fig. 4. Variations in sheet resistance of doped graphene as a function of storage time.
furnace under Ar + H2 mixture at 400 ◦ C. A close look at the spectrum in Fig. 5(b) reveals a significant increase in the background intensity around the D band position and broadening of both G and 2D bands. This observation is analogous to findings reported by Hong et al. [37] that the broadening is a consequence of additional broad D and G peaks originating from amorphous carbon formed on the surface of graphene by carbonization of polymeric residues. Our previous study on the thermal annealing of graphene on metal oxide also showed oxidation of graphene under similar annealing conditions [38]. Despite the fact that RTA induces some disorder within the graphene as evidenced from the increase in D band intensity (ID /IG ∼ 0.2), the sheet resistance of graphene is significantly reduced. For instance, sheet resistance values between 300 and 500 ohm/sq. are consistently achieved on annealed samples (Fig. 2). Doping of either graphene (pristine and annealed) up shifts both the G and 2D bands and brings the I2D /IG ratio <1, signifying charge transfer induced p-doping in graphene. Of the three dopant materi-
als studied, the AuCl3 causes a maximum shift in the G band of about 15 cm−1 whereas the TFSA and MoO3 doped graphene are identified to have a shift of 11 cm−1 and 8 cm−1 , respectively. This result is consistent with the sheet resistance measurements. Similar results are observed when doping was performed on annealed graphene. However, the doping efficiency is relatively low compared to pristine samples. Similarly, exposing the sample to ambient air reduces the sheet resistance and hence the doping efficacy. Work function of graphene is yet another important electronic properties need to be engineered for device specific applications. For instance, a high work function is required in order to use graphene as a transparent electrode in GaN-based light-emitting diodes and Si solar cells. Therefore, understanding the effect of different dopants on the work function of graphene is essential. Fig. 6 shows the UPS spectra of graphene coated with different dopants measured at secondary electron cut-off and Fermi level positions. The surface work function is estimated to be 4.3 eV, 5.1 eV, 4.9 eV, and 4.7 eV, respectively, for the pristine graphene and graphene doped with MoO3 , AuCl3 , and TFSA. The work function of pristine graphene is not a constant, can vary over a wide range depending on the amount of PMMA residue on the surface of graphene (not shown here). One can notice that the work function of graphene after doping has increased significantly and the maximum work function is observed for MoO3 coated graphene. It has been shown that the deposition of MoO3 could result an increase in work function up to 6.8 eV in high vacuum conditions [26]. In our experiment, as the sample has been exposed to ambient a degradation in the work function is expected, because air exposure of MoO3 has been shown to degrade its electronic properties [26]. The thermal stability of
Fig. 5. (a) Raman spectra of graphene before and after doping. (b) Raman spectra of pristine and annealed graphene before and after MoO3 deposition.
Please cite this article in press as: S. Chandramohan, et al., A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.01.097
G Model APSUSC-34892; No. of Pages 6
ARTICLE IN PRESS S. Chandramohan et al. / Applied Surface Science xxx (2017) xxx–xxx
6
the dopants is studied by annealing the samples at 150 ◦ C for 300 s and measuring the sheet resistance. Among the three dopants, the MoO3 proved to be highly stable without showing any change in the electrical properties. The device fabrication compatibility of charge transfer doping is also taken into consideration. To study this, the samples were subjected to a series of lithography steps (photoresist coating, exposure, development, and lift-off). We observed excellent stability for the MoO3 coated graphene while chemically doped graphene films were poor in stability when treated with organic solvents. 4. Conclusion In conclusion, we demonstrated the competence and limitations of different charge transfer materials for graphene doping. Chemically doped graphene proved to be effective in terms of doping efficiency and stability in ambient conditions. Despite relatively low doping efficacy, MoO3 still can be a potential material for graphene doping as it provides thermal stability in addition to its ability to serve as a protective coating for device integration. Rapid thermal annealing is identified to induce p-doping in graphene, which eventually reduces the doping efficacy, but overall it improves the electrical conductivity of graphene. By further optimizing the growth conditions, single layer graphene films with sheet resistance values below 100 ohm/square is possible via the charge transfer doping. Acknowledgments This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2015R1A6A1A04020421) and a grant from the Korea Institute of Science and Technology (KIST) institutional program. The authors thank Prof. Hyunsoo Kim (Semiconductor Physics Research Center, Chonbuk National University, Korea) for extending the facilities for electrical measurement. References [1] A.H. Castro, F. Neto, N.M.R. Guinea, K.S. Peres, A.K. Novoselov, Rev. Mod. Phys. 81 (2009) 109. [2] S. Bae, H. Kim, Y. Lee, X. Xu, J.-S. Park, Y. Zheng, J. Balakrishnan, T. Lei, H.R. Kim, Y.I. Song, Y.-J. Kim, K.S. Kim, B. Ozyilmaz, J.-H. Ahn, B.H. Hong, S. Lijima, Nat. Nanotechnol. 5 (2010) 574. [3] G. Jo, M. Choe, S. Lee, W. Park, Y.H. Kahng, T. Lee, Nanotechnology 23 (2012) 112001. [4] T.-H. Han, Y. Lee, M.-R. Choi, S.-H. Woo, S.-H. Bae, B.H. Hong, J.-H. Ahn, T.-W. Lee, Nat. Photon. 6 (2012) 105. [5] N. Li, S. Oida, G.S. Tulevski, S.-J. Han, J.B. Hannon, D.K. Sadana, T.-C. Chen, Nat. Commun. 4 (2013) 2294.
[6] H. Kim, S.-H. Bae, T.-H. Han, K.-G. Lim, J.-H. Ahn, T.-W. Lee, Nanotechnology 25 (2014) 014012. [7] K. Rana, J. Singh, J.-H. Ahn, J. Mater. Chem C 2 (2014) 2646. [8] H. Lee, I. Kim, M. Kim, H. Lee, Nanoscale 8 (2016) 1789. [9] S. De, J.N. Coleman, ACS Nano 5 (2010) 2713. [10] Y. Song, W. Fang, R. Brenes, J. Kong, Nano Today 10 (2015) 681. [11] I.N. Kholmanov, C.W. Magnuson, A.E. Aliev, H. Li, B. Zhang, J.W. Suk, L.L. Zhang, E. Peng, S.H. Mousavi, A.B. Khanikaev, R. Piner, G. Shvets, R.S. Ruoff, Nano Lett. 12 (2012) 5679. [12] (a) C.N.R. Rao, R. Voggu, Mater. Today 13 (2010) 34; (b) H. Liu, Y. Liu, D. Zhu, J. Mater. Chem 21 (2011) 3335. [13] K.K. Kim, A. Reina, Y. Shi, H. Park, L.-J. Li, Y.H. Lee, J. Kong, Nanotechnology 21 (2010) 285205. [14] S. Chandramohan, J.H. Kang, Y.S. Katharria, N. Han, Y.S. Beak, K.B. Ko, J.B. Park, B.D. Ryu, H.K. Kim, E.K. Suh, C.H. Hong, J. Phys. D Appl. Phys. 45 (2012) 145101. [15] D.H. Shin, K.W. Lee, J.S. Lee, J.H. Kim, S. Kim, S.-H. Choi, Nanotechnology 25 (2014) 125701. [16] S. Tongay, K. Berke, M. Lemaitre, Z. Nasrollahi, D.B. Tanner, A.F. Hebard, B.R. Appleton, Nanotechnology 22 (2011) 425701. [17] Y. Kim, J. Ryu, M. Park, E.S. Kim, J.M. Yoo, J. Park, J.H. Kang, B.H. Hong, ACS Nano 8 (2014) 868. [18] C.L. Hsu, C.-T. Lin, J.-H. Huang, C.-W. Chu, K.-H. Wei, L.-J. Li, ACS Nano 6 (2012) 5031. [19] Y. Song, W. Fang, A.L. Hsu, J. Kong, Nanotechnology 25 (2014) 395701. [20] T.H. Bointon, G.F. Jones, A.D. Sanctis, R. Hill-Pearce, M.F. Cracium, S. Russo, Sci. Rep. 5 (2015) 16464. [21] Z. Chen, I. Santoso, R. Wang, L.F. Xie, H.Y. Mao, H. Huang, Y.Z. Wang, X.Y. Gao, Z.K. Chen, D. Ma, A.T.S. Wee, W. Chen, Appl. Phys. Lett. 96 (2010) 213104. [22] S.L. Hellstrom, M. Vosqueritchian, R.M. Sroltengberg, I. Irfan, M. Hammock, Y.B. Wang, C. Jia, X. Guo, Y. Gao, Z. Bao, Nano Lett. 12 (2012) 3574. [23] J. Meyer, P.R. Kidambi, B.C. Bayer, C. Weijtens, A. Kuhn, A. Centeno, A. Pesquera, A. Zurutuza, J. Robertson, S. Hofmann, Sci. Rep. 4 (2014) 5380. [24] A. Kuruvila, P.R. Kidambi, J. Kling, J.B. Wagner, J. Robertson, J. Meyer, S. Hofmann, J. Mater. Chem. C 2 (2014) 6940. [25] P.R. Kidambi, C. Weijtens, J. Robertson, S. Hofmann, J. Meyer, Appl. Phys. Lett. 106 (2015) 063304. [26] L. D’Arsié, S. Esconjauregui, R. Weatherup, Y. Guo, S. Bhardwaj, A. Centeno, A. Zurutuza, C. Cepek an, J. Robertson, Appl. Phys. Lett. 105 (2014) 103103. [27] S. Esconjauregui, L. D’Arsié, Y. Guo, J. Yang, H. Sugime, S. Caneva, C. Cepek, J. Robertson, ACS Nano 9 (2015) 10422. [28] Q.-H. Wu, Y. Zhao, G. Hong, J.-G. Ren, C. Wang, W. Zhang, S.-T. Lee, Carbon 65 (2013) 46. [29] T.H. Seo, S. Lee, H. Cho, S. Chandramohan, E.-K. Suh, H.S. Lee, S. Bae, S.M. Kim, M. Park, J.K. Lee, M.J. Kim, Sci. Rep. 6 (2016) 24143. [30] Mir Abdullah-Al-Galib, B. Hou, T. Shahriad, S. Zivanovic, A.D. Radadia, App. Surf. Sci. 366 (2016) 78–84. [31] D. Kim, D. Lee, Y. Lee, D.Y. Jeon, Adv. Funct. Mater. 23 (2013) 5049. [32] A. Pirkle, J. Chan, A. Venugopal, D. Hinojos, C.W. Magnuson, S. McDonnell, L. Colombo, E.M. Vogel, R.S. Ruoff, R.M. Wallace, Appl. Phys. Lett. 99 (2011) 122108. [33] J.W. Suk, W.H. Lee, J. Lee, H. Chou, R.D. Piner, Y. Hao, D. Akinwande, R.S. Ruoff, Nano Lett. 13 (2013) 1462. [34] E. Koo, S.-Y. Ju, Carbon 86 (2015) 318. [35] M. Hofmann, Y.-P. Hsieh, K.-W. Chang, H.-G. Tsai, T.-T. Chen, Sci. Rep. 5 (2015) 17393. [36] C.W. Jang, J.H. Kim, J.M. Kim, D.H. Shin, S. Kim, S.-H. Choi, Nanotechnology 24 (2013) 405301. [37] J. Hong, M.K. Park, E.J. Lee, D. Lee, D.S. Hwang, S. Ryu, Sci. Rep. 3 (2013) 2700. [38] S. Chandramohan, B.D. Ryu, T.H. Seo, H. Kim, E.-K. Suh, C.-H. Hong, J. Phys. D Appl. Phys. 48 (2015) 095102.
Please cite this article in press as: S. Chandramohan, et al., A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.01.097
ARTICLE IN PRESS
APSUSC-34892; No. of Pages 6
Applied Surface Science xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Full Length Article
A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition S. Chandramohan a,b,∗ , Tae Hoon Seo c , V. Janardhanam a , Chang-Hee Hong a , Eun-Kyung Suh a,∗ a Department of Semiconductor Science and Technology, Semiconductor Physics Research Center, Chonbuk National University, Jeonju 54896, Republic of Korea b Department of Physics & Nanotechnology, SRM University, Kattankulathur 603 203, Tamil Nadu, India c Soft Innovative Materials Research Center, Korea Institute of Science and Technology, Jeonbuk 55324, Republic of Korea
a r t i c l e
i n f o
Article history: Received 17 October 2016 Received in revised form 9 January 2017 Accepted 11 January 2017 Available online xxx Keywords: Graphene Charge transfer doping Molybdenum trioxide Rapid thermal annealing
a b s t r a c t Charge transfer doping is a renowned route to modify the electrical and electronic properties of graphene. Understanding the stability of potentially important charge-transfer materials for graphene doping is a crucial first step. Here we present a systematic comparison on the doping efficiency and stability of single layer graphene using molybdenum trioxide (MoO3 ), gold chloride (AuCl3 ), and bis(trifluoromethanesulfonyl)amide (TFSA). Chemical dopants proved to be very effective, but MoO3 offers better thermal stability and device fabrication compatibility. Single layer graphene films with sheet resistance values between 100 and 200 ohm/square were consistently produced by implementing a two-step growth followed by doping without compromising the optical transmittance. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Since its discovery in 2004, graphene has become the most studied material owing to its two-dimensionality and unique band structure, which is fundamentally responsible for the remarkable electro-optical properties [1]. Especially, the superior carrier transport properties and exceptionally high optical transparency unified in an atomically thin sheet make it promising as a transparent conductive electrode (TCE) in place of brittle TCE of common use such as indium tin oxide for applications in a range of optoelectronic devices and flexible touch screens [2–8]. However, the low intrinsic carrier concentration and defects generated during the growth and transfer processes are known limiting the high conductivity requirement for practical applications [9,10], a key problem averting graphene’s progress to budge from the laboratory to industry. While integration of graphene with already established flexible TCE materials like metal nanowires seems to be an appealing solution [11], this approach has limited relevance when it is necessary to
∗ Corresponding authors at: Department of Semiconductor Science and Technology, Semiconductor Physics Research Center, Chonbuk National University, Jeonju, 54896, Republic of Korea. Department of Physics and Nanotechnology, SRM University, Kattankulathur 603 203, Tamil Nadu, India. E-mail addresses: [email protected] (S. Chandramohan), [email protected] (E.-K. Suh).
modulate the work function for specific device applications. Alternatively, a renowned charge transfer doping [12] has become the universal approach to realize low sheet resistance with a simultaneous work function regulation by exploiting the unprecedented charge-transfer interaction property of graphene i.e., being atomically thin with a vanishing density of states at the Dirac point, the Fermi level is very sensitive to external perturbations such as adsorption of metals, molecules, etc. Indeed, experiments have demonstrated that charge doping of graphene has the potential in achieving low sheet resistance values down to 30 ohm/square [2,4]. Following the pioneering work by Bae et al.[2], interest has arisen in developing doped graphene TCE and in recent years numerous studies have been conducted using different chemicals [5,12–15], strong acids [4,6], and organic molecules [16–18]. Of various p-type chemical dopants explored to date, metal chlorides are known to give the maximum doping efficiency [2,4,13] whereas the bis(trifluoromethanesulfonyl)amide (TFSA) provides better stability under normal ambient conditions [16]. However, majority of the chemical dopants including metal chlorides and TFSA on graphene surface have been found resoluble in water or organic solvents [19], inhibiting their use in device manufacturing that involves lithography and etching processes. Recently, few studies showed that relatively stable doping of graphene with sheet resistances below 20 ohm/square can be achievable through FeCl3 intercalation [19,20], but it depreciates the optical transmittance drastically.
http://dx.doi.org/10.1016/j.apsusc.2017.01.097 0169-4332/© 2017 Elsevier B.V. All rights reserved.
Please cite this article in press as: S. Chandramohan, et al., A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.01.097
G Model APSUSC-34892; No. of Pages 6
ARTICLE IN PRESS S. Chandramohan et al. / Applied Surface Science xxx (2017) xxx–xxx
2
On the other hand, high work function transition metal oxides have been recognized as potential candidates for p-type doping of graphene and carbon nanotubes [21–27]. The advantage of using metal oxides is that they can be simply deposited onto graphene as thin film coating to not only dope graphene effectively without compromising the optical transparency [23], but also to provide graphene a conformal protective layer for integration in device technology [25]. Especially, molybdenum trioxide (MoO3 ), a wide band gap material with unusually large electron affinity of 6.8 eV, has been broadly studied for graphene doping in organic electronic devices [23–25]. Unlike chemical dopants, the MoO3 is demonstrated to be stable to environment [24], chemical [21], and high temperatures [22,26]. The doping mechanism is also studied to some extent through spectroscopic investigations for both stoichiometric MoO3 [23] and substoichiometric MoOx [28]. Majority of the previous studies considered growth and in situ evaluation of doping in high vacuum conditions and a major issue yet to be resolved is the instability of MoO3 surface potential to air exposure and its impact on the doping efficiency and stability for long time exposure to environment. Despite promising results, disagreement exists between studies reported by different groups on the doping stability against air exposure. Therefore, understanding of the stability of potentially important charge-transfer materials for graphene doping is a crucial first step. Herein we give a systematic comparison on the doping efficiency and stability of single layer graphene doped by MoO3 , AuCl3 , and TFSA with respect to various factors such as the surface chemistry of graphene (amount of polymer residues), ambient exposure, and dopant morphology.
2. Materials and methods Graphene used in this study was synthesized on electrochemically polished copper foil by low pressure CVD. Single layer graphene with reduced defect density aided by a two-step growth process was grown by using methane as a carbon precursor [29]. The transfer of graphene to arbitrary substrates (Al2 O3 and SiO2 /Si) was performed by etching the Cu in ammonium persulphate after spin coating a thin polymethyl methacrylate (PMMA) layer on the surface of graphene as a protective layer. The detached PMMAsupported graphene was then washed in deionized water several times and transferred to the target substrates. The PMMA was dissolved in acetone followed by annealing at 400 ◦ C for 600 s in a rapid thermal annealing system. In our experiments, the temperature ramp up and cool down times were 30 s and 300–600 s, respectively. The cool down was done under the flow of 10 slm N2 until the temperature reduces below 50 ◦ C. The MoO3 (99.98%, Aldrich) deposition was carried out in a thermal evaporator under a base pressure of 10−6 Torr. The deposition rate was below 0.5 Å/s in most cases. For comparison, both AuCl3 and TFSA are employed in our study, because these two materials in solution form have been shown very effective for the hole doping of graphene. Doping of graphene with AuCl3 and TFSA was performed by dissolving the respective materials in nitromethane and spin-coating the solution on the surface of graphene. The three different doping processes are illustrated in Fig. 1. The optical transmittance measurements were carried out by using a JASCO V-570 ultraviolet-visible-near-infrared (UV–vis-NIR) spectrophotometer. Raman spectra of graphene were obtained by using a 514 nm line of an argon ion laser. The resistivity measurements were carried out using a Hall Effect measurement system at room temperature with a magnetic field of 0.556 T. The surface topography was probed by atomic force microscopy in tapping mode. X-ray photoelectron spectroscopy (XPS) measurements were performed on a Thermo Scientific K-alpha instrument with a monochromized Al K␣ x-ray source (1486.6 eV). Ultraviolet
photoelectron spectroscopy (UPS) measurements were performed on an AXIS Ultra DLD spectrometer where He I discharge lamp (h = 21.22 eV) was used as an excitation source.
3. Results and discussion The opto-electrical characteristics of graphene films before and after doping with different materials are evaluated. Fig. 2(a) shows the optical transmittance spectra taken on 15 mm2 graphene films doped with 10 mM AuCl3 , 20 mM TFSA, and 5 nm MoO3 . It can be seen that the optical transmittance of graphene is not deteriorated at moderate solution concentrations (for TFSA and AuCl3 cases) or MoO3 film thickness. The dependence of optical transmittance with respect to MoO3 film thickness is shown in Fig. 2(b). It is observed that the MoO3 /graphene hybrid possess high transparency exceeding 90% at all the MoO3 film thickness studied from 1 to 20 nm. In addition, the maximum percentage of reduction in transmittance is estimated to be only 3.7% at 550 nm, corresponding to the MoO3 film thickness of 20 nm. On the other hand, the optical transmittance of the AuCl3 doped graphene films is known to have a strong dependence on the dopant concentration such that increasing the AuCl3 concentration leads to increasing density of Au nanoparticles on the graphene surface [13,30]. In our study, 10 mM AuCl3 and 20 mM TFSA solutions are considered optimum because the maximum decrease in sheet resistance can be achieved at those optimum concentrations [13,31], which will be shown later. Fig. 2(c) shows the sheet resistance values of graphene measured before and immediately after the MoO3 deposition for different film thicknesses. The measurements were performed on graphene transferred to SiO2 /Si substrates with van der pauw contact geometry under normal ambient conditions. Some of the graphene films employed in this study were subjected to rapid thermal annealing (RTA) at 400 ◦ C in an inert ambient for 10 min in order to eliminate the PMMA residues prior to the deposition of MoO3 (the respective data are marked with an asterisk symbol in the figure). As can be seen in Fig. 2(c), the deposition of MoO3 lowers the sheet resistance of graphene to values nearly half to that of its initial values for MoO3 thickness above 3 nm. The results bring into picture that the effect of film thickness is insignificant beyond this optimal thickness (3 nm) at which a maximum decrease in sheet resistance of 56% is observed. This result is consistent with recent findings that the change in sheet resistance of graphene is approximately 50% and saturation occurs for 3–4 nm MoO3 [23]. In addition, for majority of annealed samples (sample #4–7), the decrease in sheet resistance after MoO3 deposition is estimated to vary between 43% and 50%. It is known that air exposure of graphene could result in change in the electronic properties of graphene and therefore the actual doping efficacy can vary from sample to sample if the experimental conditions are not invariant in all the cases. To avoid the inconsistency, we carried out similar experiments in a clean room condition (samples 9 and 10) and prior to MoO3 deposition the samples were subjected to mild heat treatment at 150 ◦ C. In this case, the doping efficacy is estimated to be 67% and 56%, respectively, for the pristine and annealed graphene. It is apparent that the doping efficacy has increased for graphene subjected to mild heat treatment prior to the deposition of MoO3 and the doping efficacy of annealed graphene is relatively low compared to pristine graphene. Similar experiments were performed for AuCl3 and TFSA doping and the results are shown in Fig. 2(d). In this case also annealed films showed relatively low doping efficacy compared to pristine graphene. Residual polymer is known to impact the electronic properties of graphene and many studies have been reported in the past demonstrating that residual polymer leads to p-type doping and carrier mobility degradation in graphene transferred to foreign
Please cite this article in press as: S. Chandramohan, et al., A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.01.097
G Model APSUSC-34892; No. of Pages 6
ARTICLE IN PRESS S. Chandramohan et al. / Applied Surface Science xxx (2017) xxx–xxx
3
Fig. 1. Schematic showing the doping of graphene using MoO3 , TFSA and AuCl3 .
Fig. 2. (a, b) Optical transmittance spectra of pristine graphene and graphene coated with different dopants. (c) Sheet resistance of graphene films before and after the deposition of MoO3 film of various thicknesses. (d) Sheet resistance of graphene films before and after the deposition of AuCl3 and TFSA. * indicates the samples subjected to annealing prior to doping.
substrates [32–34]. Thermal annealing has been considered as a promising route to reduce the polymer residues. In our study, the rapid thermal annealing of graphene is found to reduce the amount of polymer residue, which can be clearly envisaged by comparing Fig. 3(a) and (b). Here complete elimination of the polymer residue did not happen which could be likely due to annealing being conducted for a short time. But, annealing caused strong p-doping in graphene (will be discussed later under Raman analysis) complicating the interpretation of the results. Since maximum doping efficiency is observed for the graphene subjected to no prior thermal annealing, one can assume that the residual polymer has little influence on the doping efficacy of graphene. On the other hand, the doping efficacy is found to be a strong function of surface coverage and initial graphene quality. Figs. 3(c)–(e) show the surface topography of representative MoO3 coated graphene films. Figs. 3(c) and (d) correspond to annealed and pristine graphene films, respectively, after being coated with a 3 nm MoO3 . It can be seen that the surface of annealed graphene with a 3 nm MoO3 coating is very smooth with a root mean square roughness (Rq ) of 0.264 nm. On the other hand, the surface of the pristine graphene is relatively rough (Rq = 0.992 nm)
due to the presence of large amount of residual PMMA. Note that these two samples were prepared under different conditions; so the surface morphology of the samples differed significantly. However, its influence on the extent of graphene doping by MoO3 deposition is negligible because the relative change in sheet resistance for this sample is estimated to be 53%, which is comparable to other samples. Conversely, the doping efficiency largely depends on the surface coverage of MoO3, which likely depends on the surface chemistry of graphene. As shown in Fig. 3(e), the deposition of 11 nm MoO3 on a defective graphene (ID /IG ∼ 0.6) forms nanodots and the relative change in sheet resistance for this sample is estimated to be only 32%. Similarly, in the case of TFSA and AuCl3 , the doping efficiency largely depends on dopant concentration and surface roughness usually plays no significant role in controlling the sheet resistance. Few recent studies [30,35] demonstrated that for dopants such as AuCl3 , the doping efficiency is limited by the dopant morphology especially at high concentrations due to the formation of large dopant clusters or agglomeration of Au particles. The maximum reduction in the sheet resistance for graphene doped with AuCl3 , TFSA, and MoO3 is found to be 79%, 69%, and 56%, respectively. Thus, it is clear that among the three doping
Please cite this article in press as: S. Chandramohan, et al., A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.01.097
G Model APSUSC-34892; No. of Pages 6
ARTICLE IN PRESS S. Chandramohan et al. / Applied Surface Science xxx (2017) xxx–xxx
4
Fig. 3. (a, b) SEM images of graphene before and after rapid thermal annealing. (c-e) AFM surface topography of MoO3 coated graphene. Table 1 Electrical properties of graphene before and after doping. Dopant
Carrier concentration /cm2
Mobility cm2 /V sec
Sheet resistance ohm/square
Before doping 10 nm MoO3 Before doping 10 mM AuCl3 Before doping 20 mM TFSA RTA
3.295E + 12 2.510E + 13 6.764E + 12 5.773E + 13 7.530E + 12 2.367E + 13 1.068E + 13
1.988E + 03 7.844E + 02 1.396E + 03 6.835E + 02 1.070E + 03 1.094E + 03 1.180E + 03
953 317 661 158 775 241 495
materials studied, the AuCl3 leads to a significant reduction in sheet resistance of 79%, a value comparable to previous results reported by Kim et al. [6]. For graphene doped with TFSA the respective value is 69%, which is similar to the values reported in the literature [16]. Despite the fact that chemical doping is more effective in improving the sheet resistance, its stability with time has been the subject of central issue for device applications. To have a more insight, the sheet resistance of the films is measured over a period of three months during which all the three samples were stored in air atmosphere. Fig. 4 shows variations in sheet resistance values with respect to time for representative graphene films doped with different dopants. It is observed that chemically doped graphene films have better ambient stability, especially the graphene doped with TFSA shows excellent stability over time. In the case of AuCl3 doping, the sheet resistance increased from 158 ohm/square on day 1 to 268 ohm/square on day 90. Contrary to chemical doping, MoO3 coated graphene showed dissimilar behavior. In this case, the sheet resistance first decreased to a minimum value, slightly increased during the first few days, and then remained constant thereafter. Since air exposure is known to affect the stoichiometry and degrade the work function of MoO3 [23,24], the observed increase in sheet resistance suggests that the doping is principally governed by the work function difference between the two materials. In other words, the doping efficiency depends on the electronic properties of MoO3 . It is also noticed that prolonged exposure of graphene to ambient air prior to doping reduces the doping efficacy to values as low as 12%. The electrical parameters such as sheet resistance, carrier mobility, and sheet carrier concentration
of graphene before and after doping with three different dopants are given in Table 1. It can be seen that it is possible to increase the carrier concentration by approximately one order and bring down the sheet resistance of graphene by 79% from its initial value. The change in sheet resistance is a direct consequence of charge transfer induced p-doping in graphene, which is verified by Raman measurements. Fig. 5(a) shows the Raman spectra of graphene before and after the doping process for the three dopant materials studied. The pristine graphene exhibits three characteristic peaks known as D, G, and 2D bands whose positions usually differ with respect to local variations in thickness uniformity and post-transfer annealing. Prior to annealing, the G and 2D bands are positioned at 1586 cm−1 and 2686 cm−1 , respectively; while the disorder-induced band is negligible implying that the graphene is high-quality. This is further supported by a low intensity ratio between D band and G band (ID /IG ∼ 0.06). We note that the graphene is predominantly single layer as the intensity ratio of 2D band to G band (I2D /IG ) is almost two or higher when measured at different spots and the 2D band is symmetric with a peak width of ∼32 cm−1 , a value widely quoted for single layer graphene. A blue shift in the G band together with an increase in the D band intensity is observed for rapid thermally annealed samples (Fig. 5(b)). In addition, an I2D /IG ratio between 1 and 1.5 is consistently observed when measured at different spots on various samples. These results show unintentional p-doping in graphene caused due to annealing, similar to the findings of a recent study that exploited RTA for eliminating the polymer residues on the surface of graphene [36]. We observed similar effects for graphene annealed in a quartz tube
Please cite this article in press as: S. Chandramohan, et al., A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.01.097
G Model APSUSC-34892; No. of Pages 6
ARTICLE IN PRESS S. Chandramohan et al. / Applied Surface Science xxx (2017) xxx–xxx
5
Fig. 6. Ultraviolet photoelectron spectra of pristine and doped graphene on SiO2 /Si substrate.
Fig. 4. Variations in sheet resistance of doped graphene as a function of storage time.
furnace under Ar + H2 mixture at 400 ◦ C. A close look at the spectrum in Fig. 5(b) reveals a significant increase in the background intensity around the D band position and broadening of both G and 2D bands. This observation is analogous to findings reported by Hong et al. [37] that the broadening is a consequence of additional broad D and G peaks originating from amorphous carbon formed on the surface of graphene by carbonization of polymeric residues. Our previous study on the thermal annealing of graphene on metal oxide also showed oxidation of graphene under similar annealing conditions [38]. Despite the fact that RTA induces some disorder within the graphene as evidenced from the increase in D band intensity (ID /IG ∼ 0.2), the sheet resistance of graphene is significantly reduced. For instance, sheet resistance values between 300 and 500 ohm/sq. are consistently achieved on annealed samples (Fig. 2). Doping of either graphene (pristine and annealed) up shifts both the G and 2D bands and brings the I2D /IG ratio <1, signifying charge transfer induced p-doping in graphene. Of the three dopant materi-
als studied, the AuCl3 causes a maximum shift in the G band of about 15 cm−1 whereas the TFSA and MoO3 doped graphene are identified to have a shift of 11 cm−1 and 8 cm−1 , respectively. This result is consistent with the sheet resistance measurements. Similar results are observed when doping was performed on annealed graphene. However, the doping efficiency is relatively low compared to pristine samples. Similarly, exposing the sample to ambient air reduces the sheet resistance and hence the doping efficacy. Work function of graphene is yet another important electronic properties need to be engineered for device specific applications. For instance, a high work function is required in order to use graphene as a transparent electrode in GaN-based light-emitting diodes and Si solar cells. Therefore, understanding the effect of different dopants on the work function of graphene is essential. Fig. 6 shows the UPS spectra of graphene coated with different dopants measured at secondary electron cut-off and Fermi level positions. The surface work function is estimated to be 4.3 eV, 5.1 eV, 4.9 eV, and 4.7 eV, respectively, for the pristine graphene and graphene doped with MoO3 , AuCl3 , and TFSA. The work function of pristine graphene is not a constant, can vary over a wide range depending on the amount of PMMA residue on the surface of graphene (not shown here). One can notice that the work function of graphene after doping has increased significantly and the maximum work function is observed for MoO3 coated graphene. It has been shown that the deposition of MoO3 could result an increase in work function up to 6.8 eV in high vacuum conditions [26]. In our experiment, as the sample has been exposed to ambient a degradation in the work function is expected, because air exposure of MoO3 has been shown to degrade its electronic properties [26]. The thermal stability of
Fig. 5. (a) Raman spectra of graphene before and after doping. (b) Raman spectra of pristine and annealed graphene before and after MoO3 deposition.
Please cite this article in press as: S. Chandramohan, et al., A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.01.097
G Model APSUSC-34892; No. of Pages 6
ARTICLE IN PRESS S. Chandramohan et al. / Applied Surface Science xxx (2017) xxx–xxx
6
the dopants is studied by annealing the samples at 150 ◦ C for 300 s and measuring the sheet resistance. Among the three dopants, the MoO3 proved to be highly stable without showing any change in the electrical properties. The device fabrication compatibility of charge transfer doping is also taken into consideration. To study this, the samples were subjected to a series of lithography steps (photoresist coating, exposure, development, and lift-off). We observed excellent stability for the MoO3 coated graphene while chemically doped graphene films were poor in stability when treated with organic solvents. 4. Conclusion In conclusion, we demonstrated the competence and limitations of different charge transfer materials for graphene doping. Chemically doped graphene proved to be effective in terms of doping efficiency and stability in ambient conditions. Despite relatively low doping efficacy, MoO3 still can be a potential material for graphene doping as it provides thermal stability in addition to its ability to serve as a protective coating for device integration. Rapid thermal annealing is identified to induce p-doping in graphene, which eventually reduces the doping efficacy, but overall it improves the electrical conductivity of graphene. By further optimizing the growth conditions, single layer graphene films with sheet resistance values below 100 ohm/square is possible via the charge transfer doping. Acknowledgments This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2015R1A6A1A04020421) and a grant from the Korea Institute of Science and Technology (KIST) institutional program. The authors thank Prof. Hyunsoo Kim (Semiconductor Physics Research Center, Chonbuk National University, Korea) for extending the facilities for electrical measurement. References [1] A.H. Castro, F. Neto, N.M.R. Guinea, K.S. Peres, A.K. Novoselov, Rev. Mod. Phys. 81 (2009) 109. [2] S. Bae, H. Kim, Y. Lee, X. Xu, J.-S. Park, Y. Zheng, J. Balakrishnan, T. Lei, H.R. Kim, Y.I. Song, Y.-J. Kim, K.S. Kim, B. Ozyilmaz, J.-H. Ahn, B.H. Hong, S. Lijima, Nat. Nanotechnol. 5 (2010) 574. [3] G. Jo, M. Choe, S. Lee, W. Park, Y.H. Kahng, T. Lee, Nanotechnology 23 (2012) 112001. [4] T.-H. Han, Y. Lee, M.-R. Choi, S.-H. Woo, S.-H. Bae, B.H. Hong, J.-H. Ahn, T.-W. Lee, Nat. Photon. 6 (2012) 105. [5] N. Li, S. Oida, G.S. Tulevski, S.-J. Han, J.B. Hannon, D.K. Sadana, T.-C. Chen, Nat. Commun. 4 (2013) 2294.
[6] H. Kim, S.-H. Bae, T.-H. Han, K.-G. Lim, J.-H. Ahn, T.-W. Lee, Nanotechnology 25 (2014) 014012. [7] K. Rana, J. Singh, J.-H. Ahn, J. Mater. Chem C 2 (2014) 2646. [8] H. Lee, I. Kim, M. Kim, H. Lee, Nanoscale 8 (2016) 1789. [9] S. De, J.N. Coleman, ACS Nano 5 (2010) 2713. [10] Y. Song, W. Fang, R. Brenes, J. Kong, Nano Today 10 (2015) 681. [11] I.N. Kholmanov, C.W. Magnuson, A.E. Aliev, H. Li, B. Zhang, J.W. Suk, L.L. Zhang, E. Peng, S.H. Mousavi, A.B. Khanikaev, R. Piner, G. Shvets, R.S. Ruoff, Nano Lett. 12 (2012) 5679. [12] (a) C.N.R. Rao, R. Voggu, Mater. Today 13 (2010) 34; (b) H. Liu, Y. Liu, D. Zhu, J. Mater. Chem 21 (2011) 3335. [13] K.K. Kim, A. Reina, Y. Shi, H. Park, L.-J. Li, Y.H. Lee, J. Kong, Nanotechnology 21 (2010) 285205. [14] S. Chandramohan, J.H. Kang, Y.S. Katharria, N. Han, Y.S. Beak, K.B. Ko, J.B. Park, B.D. Ryu, H.K. Kim, E.K. Suh, C.H. Hong, J. Phys. D Appl. Phys. 45 (2012) 145101. [15] D.H. Shin, K.W. Lee, J.S. Lee, J.H. Kim, S. Kim, S.-H. Choi, Nanotechnology 25 (2014) 125701. [16] S. Tongay, K. Berke, M. Lemaitre, Z. Nasrollahi, D.B. Tanner, A.F. Hebard, B.R. Appleton, Nanotechnology 22 (2011) 425701. [17] Y. Kim, J. Ryu, M. Park, E.S. Kim, J.M. Yoo, J. Park, J.H. Kang, B.H. Hong, ACS Nano 8 (2014) 868. [18] C.L. Hsu, C.-T. Lin, J.-H. Huang, C.-W. Chu, K.-H. Wei, L.-J. Li, ACS Nano 6 (2012) 5031. [19] Y. Song, W. Fang, A.L. Hsu, J. Kong, Nanotechnology 25 (2014) 395701. [20] T.H. Bointon, G.F. Jones, A.D. Sanctis, R. Hill-Pearce, M.F. Cracium, S. Russo, Sci. Rep. 5 (2015) 16464. [21] Z. Chen, I. Santoso, R. Wang, L.F. Xie, H.Y. Mao, H. Huang, Y.Z. Wang, X.Y. Gao, Z.K. Chen, D. Ma, A.T.S. Wee, W. Chen, Appl. Phys. Lett. 96 (2010) 213104. [22] S.L. Hellstrom, M. Vosqueritchian, R.M. Sroltengberg, I. Irfan, M. Hammock, Y.B. Wang, C. Jia, X. Guo, Y. Gao, Z. Bao, Nano Lett. 12 (2012) 3574. [23] J. Meyer, P.R. Kidambi, B.C. Bayer, C. Weijtens, A. Kuhn, A. Centeno, A. Pesquera, A. Zurutuza, J. Robertson, S. Hofmann, Sci. Rep. 4 (2014) 5380. [24] A. Kuruvila, P.R. Kidambi, J. Kling, J.B. Wagner, J. Robertson, J. Meyer, S. Hofmann, J. Mater. Chem. C 2 (2014) 6940. [25] P.R. Kidambi, C. Weijtens, J. Robertson, S. Hofmann, J. Meyer, Appl. Phys. Lett. 106 (2015) 063304. [26] L. D’Arsié, S. Esconjauregui, R. Weatherup, Y. Guo, S. Bhardwaj, A. Centeno, A. Zurutuza, C. Cepek an, J. Robertson, Appl. Phys. Lett. 105 (2014) 103103. [27] S. Esconjauregui, L. D’Arsié, Y. Guo, J. Yang, H. Sugime, S. Caneva, C. Cepek, J. Robertson, ACS Nano 9 (2015) 10422. [28] Q.-H. Wu, Y. Zhao, G. Hong, J.-G. Ren, C. Wang, W. Zhang, S.-T. Lee, Carbon 65 (2013) 46. [29] T.H. Seo, S. Lee, H. Cho, S. Chandramohan, E.-K. Suh, H.S. Lee, S. Bae, S.M. Kim, M. Park, J.K. Lee, M.J. Kim, Sci. Rep. 6 (2016) 24143. [30] Mir Abdullah-Al-Galib, B. Hou, T. Shahriad, S. Zivanovic, A.D. Radadia, App. Surf. Sci. 366 (2016) 78–84. [31] D. Kim, D. Lee, Y. Lee, D.Y. Jeon, Adv. Funct. Mater. 23 (2013) 5049. [32] A. Pirkle, J. Chan, A. Venugopal, D. Hinojos, C.W. Magnuson, S. McDonnell, L. Colombo, E.M. Vogel, R.S. Ruoff, R.M. Wallace, Appl. Phys. Lett. 99 (2011) 122108. [33] J.W. Suk, W.H. Lee, J. Lee, H. Chou, R.D. Piner, Y. Hao, D. Akinwande, R.S. Ruoff, Nano Lett. 13 (2013) 1462. [34] E. Koo, S.-Y. Ju, Carbon 86 (2015) 318. [35] M. Hofmann, Y.-P. Hsieh, K.-W. Chang, H.-G. Tsai, T.-T. Chen, Sci. Rep. 5 (2015) 17393. [36] C.W. Jang, J.H. Kim, J.M. Kim, D.H. Shin, S. Kim, S.-H. Choi, Nanotechnology 24 (2013) 405301. [37] J. Hong, M.K. Park, E.J. Lee, D. Lee, D.S. Hwang, S. Ryu, Sci. Rep. 3 (2013) 2700. [38] S. Chandramohan, B.D. Ryu, T.H. Seo, H. Kim, E.-K. Suh, C.-H. Hong, J. Phys. D Appl. Phys. 48 (2015) 095102.
Please cite this article in press as: S. Chandramohan, et al., A comparison of various surface charge transfer hole doping of graphene grown by chemical vapour deposition, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.01.097