Activating “Invisible” Glue: Using Electron Beam for Enhancement of

Activating “Invisible” Glue: Using Electron Beam for Enhancement of Interfacial Properties of Graphene−Metal Contact Songkil Kim,† Michael Russell,‡ Dhaval D. Kulkarni,‡ Mathias Henry,† Steve Kim,∥ Rajesh R. Naik,∥ Andrey A. Voevodin,∥ Seung Soon Jang,‡ Vladimir V. Tsukruk,‡ and Andrei G. Fedorov*,†,§ †

George W. Woodruff School of Mechanical Engineering, ‡School of Materials Science and Engineering, and §Parker H. Petit Institute for Bioengineering and Bioscience, Georgia Institute of Technology, Atlanta, Georgia 30332, United States ∥ Materials and Manufacturing Directorate, Air Force Research Laboratory, Wright Patterson AFB, Ohio 45433-7707, United States S Supporting Information *

ABSTRACT: Interfacial contact of two-dimensional graphene with threedimensional metal electrodes is crucial to engineering high-performance graphene-based nanodevices with superior performance. Here, we report on the development of a rapid “nanowelding” method for enhancing properties of interface to graphene buried under metal electrodes using a focused electron beam induced deposition (FEBID). High energy electron irradiation activates two-dimensional graphene structure by generation of structural defects at the interface to metal contacts with subsequent strong bonding via FEBID of an atomically thin graphitic interlayer formed by low energy secondary electron-assisted dissociation of entrapped hydrocarbon contaminants. Comprehensive investigation is conducted to demonstrate formation of the FEBID graphitic interlayer and its impact on contact properties of graphene devices achieved via strong electromechanical coupling at graphene−metal interfaces. Reduction of the device electrical resistance by ∼50% at a Dirac point and by ∼30% at the gate voltage far from the Dirac point is obtained with concurrent improvement in thermomechanical reliability of the contact interface. Importantly, the process is rapid and has an excellent insertion potential into a conventional fabrication workflow of graphene-based nanodevices through single-step postprocessing modification of interfacial properties at the buried heterogeneous contact. KEYWORDS: focused electron beam induced deposition (FEBID), graphene, graphitic interlayer, heterogeneous contact, nanowelding

G

A carrier transport at defect-free, contamination-free graphene−metal interfaces is affected by physicochemical nature of the contact and strength of interactions between graphene and metal.11 Strong chemical binding (chemisorption) of graphene with metal atoms is required to achieve superior mechanical (binding),10,11 electron,11,12 and phonon couplings.10 A shorter coupling length than a carrier mean free path results in a ballistic transport, yielding high carrier transmissivity at interfaces.13 Thus, selection of a metal is important to achieve stronger coupling and intrinsically better interfacial properties. In addition to the intrinsic nature of the graphene−metal interface, other factors such as metal deposition methods (evaporation vs sputtering),6 microstructure of metal contacts on graphene,14 and lithography resist (polymer) residues (contaminants)15,16 also influence

raphene, a two-dimensional hexagonal carbon structure, is a promising alternative to existing electronic materials in many applications.1 Although intrinsic electronic and thermal properties of graphene are excellent, several fundamental limitations still provide significant roadblocks to its application to real device platforms, including graphene-based composites for electronics2 and energy conversion devices.3,4 Specifically, when developing graphenebased nanodevices, formation of heterogeneous contacts with metal electrodes is inevitable, and the contact properties often govern overall device performance and reliability.2,5−10 That is, poor interfacial properties dramatically reduce electrical and thermal conductances across the contact which in turn degrade device performance, as well as increase power consumptions, resulting in severe thermal issues (e.g., local temperature rise, “hot spot”, due to Joule heating)7 and wasted energy. Thus, improving interfacial properties of graphene−metal contacts is critical to developing high-performance, reliable graphenebased nanodevices. © XXXX American Chemical Society

Received: October 8, 2015 Accepted: January 7, 2016

A

DOI: 10.1021/acsnano.5b06342 ACS Nano XXXX, XXX, XXX−XXX

Article

www.acsnano.org

Article

ACS Nano

Figure 1. Schematic of the FEBID nanowelding process for contact interface modification between graphene and metal, using electronassisted formation of graphitic interlayer, acting as an atomic-scale interfacial glue.

electrodes contacting mechanically exfoliated multilayer graphene can reduce electrical contact resistance after thermal treatments.33 In addition, it was suggested that graphitic medium between two different nanomaterials (e.g., SWCNTmetal34 and SiC-metal35) helps improve thermo-electromechanical properties of heterogeneous contacts. Guided by these advances and fundamental principles of electron−matter interactions, here we develop a facile method for interface modification of a buried graphene under metal electrodes using the FEBID technique, which can be broadly applied to any device structure fabricated using any state-of-the-art conventional device fabrication workflows. This new concept of an electron-beam-assisted nanowelding is based on electroninduced transformation of entrapped hydrocarbon contaminants (e.g., polymer residues resulting from lithography process or graphene transfer process) to atomically thin electrically/ thermally conductive graphitic “interlayer”, resulting in strong chemical bindings at graphene−metal interfaces.

graphene−metal interface properties. The latter frequently dominates interfacial properties in practical applications, even when theoretically the most optimal contact metal is chosen. For instances, titanium (Ti) metal contacts are theoretically known to be the best candidate with the strongest chemical coupling at the graphene−metal interfaces, but experimentally, Ti contacts on graphene are found to result in inconsistent electrical contact properties with poor stability.6,14,17 Also, significant disparities in electrical contact resistivities varying from ∼1 to ∼100 kΩ μm can be found in chromium (Cr) metal contacts, which is another possible candidate material.6,18,19 Among the factors affecting quality of graphene−metal interfaces, entrapped lithography resist (polymer) residues impede the direct, efficient carrier transport from graphene to metal or vice versa, regardless of specific metal used for making contact. What makes the contamination issue even more pressing is that it is difficult to perfectly clean a graphene surface, and especially in the course of a metal contact fabrication process.19−21 This work addresses this fundamental challenge of any workflow involved in making graphene based electronic devices by developing a rapid, highly controllable method based on electron beam processing to convert surface contamination polymer residues entrapped at a graphene− metal interface to a conductive “atomic glue” capable of forming a high performance interface. The approach is analogous to a conventional electromechanical “welding” technique with conductive solders, but performed with a high resolution on the atomic length scale. Welding is an additive manufacturing process for joining two or more dissimilar materials which is of great significance to fabrication of multicomponent engineering systems. For nanoscale contact formation, the welding process requires high precision and capabilities for selective consolidation of heterogeneous materials using an appropriate solder. Focused electron beam induced deposition (FEBID) has fundamental properties suitable for nanoscale welding, and has been demonstrated for site-specific, high-resolution patterning of nanomaterials for a broad spectrum of applications, that is, electromechanical welding of nanotube junctions,22−24 nanoplasmonics,25 nanosensors,26 and photomask repairs.27 FEBID enables controllable manipulations of nanomaterials and devices with an intrinsic resolution feasible down to a single atom for molecularly thin suspended substrates.27−29 Specifically, in its applications to carbon-based nanomaterial systems, the FEBID provides unique additive “direct-write” capabilities for local modification of structure and electronic properties of graphene-based materials30−32 and fabrication of electrically conducting heterojunctions of carbon nanotube-metal electrodes by connecting multiple inner shells of multiwalled carbon nanotubes (MWCNTs).24 Also, in our previous work, it was found that electron beam irradiation over source-drain metal

RESULTS AND DISCUSSION Figure 1 schematically describes an FEBID nanowelding process aimed to improve interfacial properties of graphene− metal heterogeneous contacts. Fabrication of graphene-based devices requires patterning of a graphene layer via a lithography process using a polymer resist (e.g., in this study, polymethyl(methacrylate) (PMMA) was used as a positive resist for ebeam lithography). During a lithography process, it is difficult to completely remove PMMA resists after development of designed patterns, and thus usually a molecularly thin residual PMMA layer remains between graphene and metal contacts.19,21 In the electron-beam-assisted nanowelding process, the entrapped PMMA (a form of hydrocarbon) residues are utilized as a precursor converted into a carbon “glue” to chemically bind graphene and metal. Primary electron beam with its energy of 25 keV can penetrate through metal contacts, e.g., down to ∼1.2 μm for Au and ∼2.5 μm for Cr.36 Interaction of the penetrating primary electrons with graphene and the graphene-supporting substrate results in defects on the graphene surface and simultaneous generation of low-energy secondary electrons,30 which dissociate the structurally stable PMMA residues transforming them into a structurally unstable hydrogenated carbon structure hereafter referred to as “FEBID carbon interlayer”. “Unstable” nature of as-deposited FEBID carbon interlayer is easily transformed to graphitic structure having higher thermal/electronic conductivities by postdeposition, low-temperature thermal annealing process37 or by localized ohmic heating simply by passing a small current through the device.24 This leads to stabilization of chemical binding of FEBID graphitic carbon with graphene and metal, for significant improvements of interfacial properties of the graphene−metal contact. B

DOI: 10.1021/acsnano.5b06342 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 2. (a) Change of carrier transfer characteristics of a graphene electronic device before and after FEBID graphitic carbon interlayer formation, showing reduction of device electrical resistance after postdeposition thermal annealing at 350 °C for 35 and 85 min, while (b) a graphene device without FEBID contact treatment shows degradation of electric properties after thermal annealing for 85 min. (c) Transfer characteristic of a graphene device with initially high electrical resistance. The inset shows the transfer curve of the device after thermal annealing for 35 min, showing severe degradation of graphene−metal interface. This highlights a significant benefit of nanowelding via FEBID graphitic carbon interlayer for improving the graphene−metal interface and its electrical and thermomechanical interfacial properties.

imaging that FEBID carbon is fully graphitized after thermal annealing at 350 °C. Due to the increased carrier scattering by carbon on the channel, total resistance initially increased immediately after the FEBID treatment over the entire range of gate voltages applied, as shown in Figure 2a. However, after postdeposition thermal annealing for 35 min, the total resistance decreased by ∼47% at Vbg = VDirac and by ∼27% at Vbg − VDirac = −35 V, respectively, as compared to the respective resistances of an as-fabricated device before FEBID. After the prolonged annealing in air at 350 °C up to 85 min, only a slight change in the resistance was found with the final reduction of total resistance by ∼50% at Vbg = VDirac and ∼30% at Vbg − VDirac= −35 V. This suggests that the annealing time of 35 min is sufficient for complete graphitization of FEBID carbon interlayer and forming stable interfacial glue. For comparison, as-fabricated devices without FEBID interlayer-assisted welding of interfaces were also tested upon the same thermal annealing process, with transfer characteristics of two different graphene devices shown in Figure 2b and 2c. It is worth noting that, in Figure 2, a scaled electrical performance metric, that is the total resistance multiplied by a ratio of geometric dimensions (W/Lch) of the conduction channel, is used to allow for side-by-side comparison of contact resistances between different devices. This metric is appropriate since the contact geometries (contact length ∼ 2.6 μm and width ∼ 1.5

The impact and unique benefits of nanowelding of a buried graphene−metal interface through FEBID graphitic interlayer formation was evaluated by assessing changes of electrical resistance of graphene devices. Figure 2a shows the change of the transfer characteristics upon a FEBID nanowelding process, measured at ambient conditions by sweeping a back gate voltage with a fixed drain-source bias as described in the Supporting Information Figure S1a. The FEBID carbon interlayer formation process was performed at two graphene− metal contacts by electron irradiations over a fixed area (1.5 μm × 1.5 μm) right on top of metal contacts (30 nm Au/10 nm Cr) with the beam energy of 25 keV and the electron dose of 5e18 e−/cm2 (total exposure time ∼ 45 s). It is worth noting that primary electron beam irradiation is accompanied by physisorbed carbon deposition on a graphene channel, which is readily eliminated by using the exposure to low power Raman laser30 or low-temperature thermal annealing in air (see Supporting Information, Figures S2 and S3) with no detrimental impact on the underlying graphene surface. In fact, postdeposition thermal annealing in air at 350 °C resulted in complete removal of any parasitic carbon deposits from the graphene channel with concurrent phase transformation of the FEBID carbon interlayer. (see Supporting Information, Figures S2 and S3) More importantly, in our previous work,37 we have shown using Raman spectroscopy and AFM topography C

DOI: 10.1021/acsnano.5b06342 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 3. Optical microscope images of as-fabricated Cu patterns on graphene before and after Cu wet etching for 30 min: (a, c) with exposure to electron beam irradiation and (b, d) without exposure to electron beam irradiation. Scale bar: 20 μm.

contacts. Two sets of Cu square (6 μm × 6 μm) arrays were fabricated on a monolayer graphene film, supported on a 90 nm SiO2/Si substrate using electron beam lithography followed by Cu evaporation and subsequent lift-off of Cu, as shown in Figure 3a and b. The Cu array in Figure 3a was irradiated by electron beam with energy of 25 keV and dose of 1e19 e−/cm2 over an area of 4 μm × 4 μm (total exposure time ∼ 10 min), using a FEI Quanta 200 Environmental scanning electron microscope. Prior to Cu deposition, a surface topology of the developed Cu square pattern was measured using an atomic force microscope to confirm presence of PMMA residues on the graphene surface as shown in Figure S4. Uniformly distributed, continuous PMMA residue film with height of 1−3 nm was measured over the entire graphene surface, including some relatively large domains with height up to ∼40 nm. To investigate formation of FEBID carbon interlayer and its contribution to enhancement of chemical (mechanical) binding between graphene and the Cu contacts, the Cu contacts were selectively removed by a chemical wet-etching method using 0.05 g/mL of ammonium persulfate in DI water. The sample substrate was placed in the etching solution heated on a hot plate at 40 °C and taken out from the solution after a variable etching time, followed by rinsing in DI water and dry air blow. After sufficient etching time (30 min) when no further change in all metal patterns can be observed in optical microscope, the arrays with and without electron-beam exposure have been compared. A clear difference can be found, as shown in Figure 3c and d, with the Cu squares exposed to electron beam irradiation still remaining on graphene, whereas Cu patterns not exposed to electrons were completely etched away. Measuring a surface topology of the square with FEBID after an attempted Cu etching (see Supporting Information Figure S5), it was found that ∼3.6 nm Cu still remained bound to graphene. This indicates that thin Cu layers strongly adhere to the graphene surface via formation of the FEBID carbon interlayer, thus enhancing chemical binding between Cu and graphene, while Cu as-deposited on graphene without FEBID

μm) are identical for all three devices. The electrical measurements of the as-fabricated graphene devices indicate that the device of Figure 2a before FEBID graphitic interlayer formation has lower quality graphene−metal interface as compared to the device of Figure 2b; and the device of Figure 2c has the worst interfacial properties of all. After thermal annealing for 35 min, the resistance in Figure 2b slightly decreased by ∼29% at Vbg = VDirac and ∼14% at Vbg − VDirac= −35 V, which is due to partial removal of residual photoresist contaminants from the graphene channel by thermal oxidation. However, the device resistance began to increase upon longer exposure to annealing conditions due to thermomechanical damage of the graphene−metal interfaces, whereas the device resistance with FEBID graphitic interlayer in Figure 2a further decreased after the same prolonged annealing. Further prolonged annealing of the device in Figure 2b will increase the resistance by continuously damaging the graphene−metal interfaces. Damage by thermal annealing was also found in other devices (10 devices were tested), with severe results even after the initial exposure to an elevated temperature, for example as shown in Figure 2c. After the first stage of thermal annealing for 35 min, the device lost the graphene’s intrinsic electrical characteristic (i.e., ambipolar behavior). The change of Ids−Vds curve at Vbg= −20 V is shown in Figure S1b. The device clearly showed an ohmic junction behavior before annealing, but it changed to rectifying junction behavior after thermal annealing and the prolonged annealing continued to significantly increase the device resistance. It suggests that the thermal annealing breaks down the interface between graphene and metal in the absence of interfacial “glue” such as the FEBID graphitic interlayer. Collectively, these observations highlight that the FEBID graphitic interlayer not only reduces the electrical contact resistance, but also improves the interfacial thermo-mechanical properties of graphene−metal contacts. For an in-depth assessment of the physicochemical nature and interfacial properties, we performed a comprehensive characterization of graphene−FEBID carbon interlayer−metal D

DOI: 10.1021/acsnano.5b06342 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 4. (a) Change of Raman spectrum of pristine graphene with a Cu contact and after 25 keV electron beam irradiation with various electron doses. Quantification of Raman spectra in terms of changes of (b) intensity and area ratios of the D to the G peaks and (c) full width half-maximum (fwhm) of the G and D peaks of graphene upon increasing the electron dose used for FEBID nanowelding of the buried interface.

presence of sp2 sites, including both carbon chains and rings.40,41 Thus, the decrease of the I(2D)/I(G) ratio indicates either the decrease of sp2 hexagonal sites on graphene by the defect generation or an increase of sp2 sites with additional carbon deposition. To quantitatively assess the effect of the electron dose on the FEBID carbon interlayer formation on graphene, evolution of signature characteristics of the Raman spectra are plotted in Figure 4b and c. With the electron dose of 2e17 e−/cm2, the I(D)/I(G) ratio increases from ∼0.25 to ∼1.0, which represents a higher density of defects on graphene generated by exposure to high energy electrons. The electron energy used for FEBID is sufficient for the generation of sp3-type defects, but less than the energy required for other kinds of defects, such as Stone−Wales (SW) defects or carbon vacancies.42,43 The sp3-type defect sites are energetically unstable, thus FEBID carbon atoms generated upon decomposition of residual PMMA can strongly bind to these defect sites.30,42,43 Figure 4c shows evolution of a full width half-maximum (fwhm) of the G and D peaks as a function of the electron dose. Formation of FEBID carbon interlayer can be confirmed by the observed broadening of G and D peaks, with the latter indicative of formation of graphitic domains (sp2 carbon bonds in nanosized hexagonal rings) within the FEBID carbon interlayer deposit.37,39,44 Interestingly, when being irradiated with the electron dose of 2e17 e−/cm2, broadening of the G peak is much more significant than that of the D peak. Thus, an increase of the A(D)/A(G) ratio must be mainly due to an increase of the D peak intensity, and it can be concluded that the defect formation in graphene is more dominant than formation of graphitic domains in the FEBID carbon interlayer. Increasing the electron dose to 1e18 e−/cm2 resulted in a decrease of both I(D)/I(G) and A(D)/A(G) ratios with broadening of both the G and D peaks. For the case of metal contact irradiation with this electron dose, no or negligible

has a weak physical (i.e., via van der Waals interactions) binding to graphene10,11 and is readily etched away upon exposure to the etching solution. Finally, to provide a direct evidence for the postulated mechanism of enhanced graphene−metal binding at the electron-beam-irradiated interface, we investigated the nature and spatially resolved properties of FEBID carbon interlayer between graphene and Cu via Raman spectroscopy. Details of experimental procedures are described in the Supporting Information. Figure 4a shows Raman spectra demonstrating different stages of FEBID carbon interlayer formation along with modification of graphene structural properties by changing the electron dose. The Raman spectra were obtained after metal electrode etching for 11 min to remove most of interfering bulk Cu and to expose the graphene surface, using confocal Raman spectroscopy with a 514 nm Ar+ laser. Pristine graphene as transferred to the substrate shows the Raman characteristics of a high quality, monolayer sp2 carbon bonded sheet with the characteristic peak intensity ratios, I(D)/I(G) ∼ 0.25 and I(2D)/I(G) ∼ 2.0, respectively. After Cu deposition, the D peak intensity slightly increases, which resulted from either an exposure to a small dose of electrons during electron beam lithography or generation of atomic-scale strain in graphene lattice structures by interaction with Cu on top of graphene.38 The Raman spectra of graphene are more significantly influenced by exposure to high energy electrons in the course of FEBID carbon interlayer formation. The I(D)/I(G) strongly increased with the electron dose of 2e17 e−/cm2, which indicates generation of defects in graphene. Increasing the electron dose broadened both the D and G peaks, which is a clear evidence of formation of the FEBID carbon interlayer between graphene and Cu.30,37,39 A decrease of the I(2D)/I(G) ratio occurs with the defect generation and also with the FEBID carbon interlayer formation.30,40,41 The 2D peak is a signature of graphene’s sp2 hexagonal sites, while G peak indicates E

DOI: 10.1021/acsnano.5b06342 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

∼300 nm. For electron beam lithography, a Quanta 200 ESEM (FEI, Inc.) instrument operated under ∼10−6 Torr was employed with a “NPGS” (Nanometer Pattern Generation System) software. Electron beam conditions were set to beam current of ∼400 pA and energy of 25 keV for a beam dwell time of 25 μs, corresponding to electron dose of ∼400 μC/cm2. After exposing PMMA layer according to the designed patterns, the PMMA layer was developed by soaking the substrate in methyl isobutyl ketone (MIBK) for 130 s. Pattern development was terminated by soaking the substrate in isopropyl alcohol for 30 s and washing it in DI water for 30 s followed by dry air blow. Finally, Cu was deposited on the substrate using electron beam evaporator, followed by the lift-off process to remove Cu everywhere except the pattern areas. For fabrication of graphene devices used for electrical measurements, the same procedures were done, but with the deposition of 30 nm Au/10 nm Cr. Characterization. Topography images were obtained using a Dimension-3000 atomic force microscope with a silicon tip in a tapping mode. Raman measurements of graphene before/after FEBID graphitic interlayer formation were carried out with a WITec (Alpha 300R) confocal Raman microscope using 514.5 nm Ar+ ion laser with a maximum power ∼ 1 mW. Ten spot (laser spot size ∼ 1 μm) measurements were performed and the measured spectra were averaged for analysis. The Raman data were analyzed for the spectral range between 1100 and 1800 cm−1 to observe the G and D peaks. Lorentzian peak fitting was applied to the Raman data in order to obtain the D to G peak intensity and area ratios.

additional defect generation occurs on graphene with FEBID carbon deposition, which only increases and broadens the G peak, and still no significant amount of graphitic domains is formed within the FEBID carbon deposits. Beyond the electron dose of 5e18 e−/cm2, a significant increase of the A(D)/A(G) is observed, while the I(D)/I(G) ratio decreases. An increase of the A(D)/A(G) ratio is due to significant broadening of the D peak, and it indicates formation of graphitic domains within the FEBID carbon interlayer deposit, featuring the amorphous matrix containing the sp2 carbon bonds in the form of rings.30,37,39,44 A further increase of the electron dose to 5e18 e−/cm2 does not lead to any changes in either of the two ratios. It implies that the electron dose of 5e18 e−/cm2 is sufficient for complete graphitization of the FEBID carbon interlayer.

CONCLUSION In summary, we have developed a novel approach to nanoscale welding of buried graphene−metal heterogeneous contacts and demonstrated its positive impact on contact properties of graphene devices. FEBID interlayer formation at the contact interface is induced and controlled by changing an electron dose, leading to tunable structural composition of the carbon “glue” interlayer and the underlying graphene. Significant reduction in device electrical resistance, as much as ∼50% at a minimum carrier density and ∼30% at a maximum carrier density, has been achieved via postdeposition thermal annealing for further graphitization of the FEBID graphitic interlayer and removal of the carbon contamination from the graphene conducting channel. In contrast, the graphene−metal interfaces with no interfacial modification have severely degraded upon the same thermal treatment, which emphasizes the superior thermomechanical properties of the graphene−metal heterogeneous interfaces enhanced with the FEBID graphitic interlayer formation. Owing to the direct-write nature of the FEBID process, this technique has a high insertion potential into existing graphene electronic device fabrication workflows and can be applicable to many other types of graphene device structures, which require improvements of interfacial properties of heterogeneous contacts. Among most intriguing applications of this technique is perhaps the contact fabrication in stacked, multilayer devices based on emerging 2D membrane materials, including graphene, hexagonal boron nitride, molybdenum disulfide, and others, which could take advantage of electron beam modification of buried interfaces with no direct line-ofsight access required for conventional fabrication methods of electronic devices.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.5b06342. Postdeposition graphene regeneration by removing FEBID amorphous carbon weakly interacting with graphene, electrical measurement setup in Figure 2 and Ids−Vds curves at Vbg = −20 V in Figure 2c, AFM image and cross-sectional profile to confirm the presence of PMMA residues after developing contact patterns, AFM image and cross-sectional profile of the Cu pattern remained after wet-etching in Figure 3c, and change of Cu patterns upon consecutive etch steps: results of Figure 4 (PDF)

AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENTS Research primarily supported by the U.S. Department of Energy (DOE), Office of Science, Basic Energy Sciences (BES), under Award #DE-SC0010729 (FEBID experiments, test structure fabrication and electrical measurements, and data analysis). AFOSR BIONIC Center Award No. FA9550-09-10162 provided support for synthesis of graphene samples, and Semiconductor Research Corporation GRC Contract 2011-OJ2221 supported work on graphene transfer optimization and Raman/AFM characterization.

METHODS Graphene Growth. A tube furnace (OTF-1200x-STM, MTI Corp., Richmond, CA) equipped with a scroll vacuum pump was used for chemical vapor deposition of monolayer graphene. A 4 × 4 in2 Cu foil was placed in the furnace and heated up to 1000 °C while a hydrogen flow was injected at a pressure of 125 mTorr. The hydrogenonly reduction step continued for 30 min at 1000 °C. Then, a methane gas was flowed at a pressure of 1.25 Torr for 30 min at 1000 °C. The furnace was powered off and allowed to cool down to room temperature while keeping the flow of methane and hydrogen. Fabrication of Metal Contacts onto Monolayer Graphene. Using the optimized PMMA-mediated wet transfer method,32 a monolayer graphene film was transferred onto a SiO2/Si substrate. Electron beam lithography was done to make square patterns for fabrication of metal contacts on graphene. PMMA was used as a positive e-beam resist spin-coated on graphene supported by the SiO2/ Si substrate at 3000 rpm for 30 s, yielding a thickness of a PMMA layer

REFERENCES (1) Novoselov, K. S.; Fal’ko, V. I.; Colombo, L.; Gellert, P. R.; Schwab, M. G.; Kim, K. A Roadmap for Graphene. Nature 2012, 490, 192−200. (2) 2012 International Technology Roadmap for Semiconductors (2012 ed. Interconnect). Available: http://www.itrs.net. F

DOI: 10.1021/acsnano.5b06342 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano (3) Zhang, N.; Yang, M.-Q.; Liu, S.; Sun, Y.; Xu, Y.-J. Waltzing with the Versatile Platform of Graphene to Synthesize Composite Photocatalysts. Chem. Rev. 2015, 115, 10307. (4) Yang, M.-Q.; Zhang, N.; Pagliaro, M.; Xu, Y.-J. Artificial Photosynthesis over Graphene-Semiconductor Composites. Are We Getting Better? Chem. Soc. Rev. 2014, 43, 8240. (5) Nagashio, K.; Nishimura, T.; Kita, K.; Toriumi, A. Contact Resistivity and Current Flow Path at Metal/Graphene Contact. Appl. Phys. Lett. 2010, 97, 143514. (6) Nagashio, K.; Nishimura, T.; Kita, K.; Toriumi, A. Metal/ Graphene Contact as a Performance Killer of Ultra-High Mobility Graphene Analysis of Intrinsic Mobility and Contact Resistance. Technol. Dig.− Int. Electron Devices Meet. 2009, 565−568. (7) Franklin, A. D.; Han, S. J.; Bol, A. A.; Perebeinos, V. Double Contacts for Improved Performance of Graphene Transistors. IEEE Electron Device Lett. 2012, 33, 17−19. (8) Smith, J. T.; Franklin, A. D.; Farmer, D. B.; Dimitrakopoulos, C. D. Reducing Contact Resistance in Graphene Devices through Contact Area Patterning. ACS Nano 2013, 7, 3661−3667. (9) Grosse, K. L.; Bae, M. H.; Lian, F.; Pop, E.; King, W. P. Nanoscale Joule Heating, Peltier Cooling and Current Crowding at Graphene-Metal Contacts. Nat. Nanotechnol. 2011, 6, 287−290. (10) Chen, L.; Huang, Z.; Kumar, S. Impact of Bonding at MultiLayer Graphene/Metal Interfaces on Thermal Boundary Conductance. RSC Adv. 2014, 4, 35852−35861. (11) Matsuda, Y.; Deng, W. Q.; Goddard, W. A., III Contact Resistance for “End-Contacted” Metal-Graphene and Metal-Nanotube Interfaces from Quantum Mechanics. J. Phys. Chem. C 2010, 114, 17845−17850. (12) Wang, L.; Meric, I.; Huang, P. Y.; Gao, Q.; Gao, Y.; Tran, H.; Taniguchi, T.; Watanabe, K.; Campos, L. M.; Muller, D. A.; Guo, J.; Kim, P.; Hone, J.; Shepard, K. L.; Dean, C. R. One-Dimensional Electrical Contact to a Two-Dimensional Material. Science 2013, 342, 614−617. (13) Xia, F.; Perebeinos, V.; Lin, Y.; Wu, Y.; Avouris, P. The Origins and Limits of Metal-Graphene Junction Resistance. Nat. Nanotechnol. 2011, 6, 179−184. (14) Watanabe, E.; Conwill, A.; Tsuya, D.; Koide, Y. Low Contact Resistance Metals for Graphene Based Devices. Diamond Relat. Mater. 2012, 24, 171−174. (15) Wei Chen, C.; Ren, F.; Chi, G. C.; Hung, S. C.; Huang, Y. P.; Kim, J.; Kravchenko, I. I.; Pearton, S. J. UV Ozone Treatment for Improving Contact Resistance on Graphene. J. Vac. Sci. Technol. B 2012, 30, 060604. (16) Robinson, J. A.; LaBella, M.; Zhu, M.; Hollander, M.; Kasarda, R.; Hughes, Z.; Trumbull, K.; Cavalero, R.; Snyder, D. Contacting Graphene. Appl. Phys. Lett. 2011, 98, 053103. (17) Moon, J. S.; Antcliffe, M.; Seo, H. C.; Curtis, D.; Lin, S.; Schmitz, A.; Milosavljevic, I.; Kiselev, A. A.; Ross, R. S.; Gaskill, D. K.; Campbell, P. M.; Fitch, R. C.; Lee, K. M.; Asbeck, P. Ultra-Low Resistance Ohmic Contacts in Graphene Field Effect Transistors. Appl. Phys. Lett. 2012, 100, 203512. (18) Song, S. M.; Park, J. K.; Sul, O. J.; Cho, B. J. Determination of Work Function of Graphene under a Metal Electrode and Its Role in Contact Resistance. Nano Lett. 2012, 12, 3887−3892. (19) Gahng, S.; Ra, C. H.; Cho, Y. J.; Kim, J. A.; Kim, T. S.; Yoo, W. J. Reduction of Metal Contact Resistance of Graphene Devices via CO2 Cluster Cleaning. Appl. Phys. Lett. 2014, 104, 223110. (20) Pirkle, A.; Chan, J.; Venugopal, A.; Hinojos, D.; Magnuson, C. W.; McDonnell, S.; Colombo, L.; Vogel, E. M.; Ruoff, R. S.; Wallace, R. M. The Effect of Chemical Residues on the Physical and Electrical Properties of Chemical Vapor Deposited Graphene Transferred to SiO2. Appl. Phys. Lett. 2011, 99, 122108. (21) Suk, J. W.; Lee, W. H.; Lee, J.; Chou, H.; Piner, R. D.; Hao, Y.; Akinwande, D.; Ruoff, R. S. Enhancement of the Electrical Properties of Graphene Grown by Chemical Vapor Deposition via Controlling the Effects of Polymer Residue. Nano Lett. 2013, 13, 1462−1467. (22) Banhart, F. The Formation of a Connection between Carbon Nanotubes in an Electron Beam. Nano Lett. 2001, 1, 329−332.

(23) Rykaczewski, K.; Henry, M. R.; Kim, S. K.; Kulkarni, D.; Singamaneni, S.; Tsukruk, V. V.; Fedorov, A. G. Effect of Electron Beam Induced Deposited (EBID) Carbon Joint Geometry and Material Properties on Electrical Resistance of Multiwalled Carbon Nanotubes (MWNT)-To-Metal Contact Interface. Nanotechnology 2010, 21, 035202. (24) Kim, S.; Kulkarni, D. D.; Rykaczewski, K.; Henry, M.; Tsukruk, V. V.; Fedorov, A. G. Fabrication of an Ultralow-Resistance Ohmic Contact to MWCNT-Metal Interconnect Using Graphitic Carbon by Electron Beam Induced Deposition (EBID). IEEE Trans. Nanotechnol. 2012, 11, 1223−1230. (25) Hoeflich, K.; Yang, R. B.; Berger, A.; Leuchs, G.; Christiansen, S. The Direct Writing of Plasmonic Gold Nanostructures by ElectronBeam-Induced Deposition. Adv. Mater. 2011, 23, 2657−2661. (26) Boero, G.; Utke, I.; Bret, T.; Quack, N.; Todorova, M.; Mouaziz, S.; Kejik, P.; Brugger, J.; Popovic, R. S.; Hoffmann, P. Submicrometer Hall Devices Fabricated by Focused Electron-Beam-Induced Deposition. Appl. Phys. Lett. 2005, 86, 042503. (27) van Dorp, W. F.; Hagen, C. W. A Critical Literature Review of Focused Electron Beam Induced Deposition. J. Appl. Phys. 2008, 104, 081301. (28) van Dorp, W. F.; Zhang, X.; Feringa, B. L.; Hansen, T. W.; Wagner, J. B.; De Hosson, J. Th. M. Molecule-By-Molecule Writing Using a Focused Electron Beam. ACS Nano 2012, 6, 10076−10081. (29) Meyer, J. C.; Girit, C. O.; Crommie, M. F.; Zettl, A. Hydrocarbon Lithography on Graphene Membranes. Appl. Phys. Lett. 2008, 92, 123110. (30) Kim, S.; Kulkarni, D. D.; Davis, R.; Kim, S. S.; Naik, R. R.; Voevodin, A. A.; Russell, M.; Jang, S. S.; Tsukruk, V. V.; Fedorov, A. G. Controlling the Physicochemical State of Carbon on Graphene Using Focused Electron-Beam-Induced Deposition. ACS Nano 2014, 8, 6805−6813. (31) Kim, S.; Kulkarni, D. D.; Henry, M.; Zackowski, P.; Jang, S. S.; Tsukruk, V. V.; Fedorov, A. G. Localized Conductive Patterning via Focused Electron Beam Reduction of Graphene Oxide. Appl. Phys. Lett. 2015, 106, 133109. (32) Kim, S.; Russell, M.; Henry, M.; Kim, S. S.; Naik, R. R.; Voevodin, A. A.; Jang, S. S.; Tsukruk, V. V.; Fedorov, A. G. Dynamic Modulation of Electronic Properties of Graphene by Localized Carbon Doping Using Focused Electron Beam Induced Deposition. Nanoscale 2015, 7, 14946−14952. (33) Fedorov, A. G.; Kim, S.; Henry, M.; Kulkarni, D.; Tsukruk, V. V. Focused-Electron-Beam-Induced Processing (FEBIP) for Emerging Applications in Carbon Nanoelectronics. Appl. Phys. A: Mater. Sci. Process. 2014, 117, 1659−1674. (34) Chai, Y.; Hazeghi, A.; Takei, K.; Chen, H.-Y.; Chan, P. C. H.; Javey, A.; Wong, H.-S. P. Low-Resistance Electrical Contact to Carbon Nanotubes with Graphitic Interfacial Layer. IEEE Trans. Electron Devices 2012, 59, 12−19. (35) Liu, F.; Hsia, B.; Carraro, C.; Pisano, A. P.; Maboudian, R. Enhanced Ohmic Contact via Graphitization of Polycrystalline Silicon Carbide. Appl. Phys. Lett. 2010, 97, 262107. (36) Kanaya, K. Okayama, S. Penetration and Energy-Loss Theory of Electrons in Solid Targets. J. Phys. D: Appl. Phys. 1972, 5, 43−58. (37) Kulkarni, D. D.; Rykaczewski, K.; Singamaneni, S.; Kim, S.; Fedorov, A. G.; Tsukruk, V. V. Thermally Induced Transformation of Amorphous Carbon Nanostructures Fabricated by Electron Beam Induced Deposition. ACS Appl. Mater. Interfaces 2011, 3, 710−720. (38) Shen, X.; Wang, H.; Yu, T. How Do the Electron Beam Writing and Metal Deposition Affect the Properties of Graphene during Device Fabrication? Nanoscale 2013, 5, 3352−3358. (39) Kulkarni, D. D.; Kim, S.; Fedorov, A. G.; Tsukruk, V. V. LightInduced Plasmon-Assisted Phase Transformation of Carbon on Metal Nanoparticles. Adv. Funct. Mater. 2012, 22, 2129−2139. (40) Cancado, L. G.; Jorio, A.; Martins Ferreira, E. H.; Stavale, F.; Achete, C. A.; Capaz, R. B.; Moutinho, M. V. O.; Lombardo, A.; Kulmala, T. S.; Ferrari, A. C. Quantifying Defects in Graphene via Raman Spectroscopy at Different Excitation Energies. Nano Lett. 2011, 11, 3190−3196. G

DOI: 10.1021/acsnano.5b06342 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano (41) Eckmann, A.; Felten, A.; Mishchenko, A.; Britnell, L.; Krupke, R.; Novoselov, K. S.; Casiraghi, C. Probing the Nature of Defects in Graphene by Raman Spectroscopy. Nano Lett. 2012, 12, 3925−3930. (42) Banhart, F.; Kotakoski, J.; Krasheninnikov, A. V. Structural Defects in Graphene. ACS Nano 2011, 5, 26−41. (43) Li, L.; Reich, S.; Robertson, J. Defect Energies of Graphite: Density-Functional Calculations. Phys. Rev. B: Condens. Matter Mater. Phys. 2005, 72, 184109. (44) Ferrari, A. C.; Robertson, J. Interpretation of Raman Spectra of Disordered and Amorphous Carbon. Phys. Rev. B: Condens. Matter Mater. Phys. 2000, 61, 14095.

H

DOI: 10.1021/acsnano.5b06342 ACS Nano XXXX, XXX, XXX−XXX