Chemical Vapor Deposition-Derived Graphene with Electrical ...
Chemical Vapor Deposition-Derived Graphene with Electrical Performance of Exfoliated Graphene Nicholas Petrone†, Cory R. Dean†,‡, Inanc Meric‡, Arend M. van der Zande†, Pinshane Y. Huang§, Lei Wang†, David Muller§, , Kenneth L. Shepard‡, James Hone*,† †
Department of Mechanical Engineering, ‡Department of Electrical Engineering, Columbia University, New York, New York 10027, United States
§
School of Applied and Engineering Physics, Kavli Institute at Cornell University, Cornell University, Ithaca, New York 14853, United States
1
Graphene Growth by Chemical Vapor Deposition Discrete, star-shaped patches of large-grain graphene were grown by low-pressure chemical vapor deposition (CVD) using methane as a carbon feedstock. The graphene was grown on the inside surface of 25 µm thick copper foil (Alfa Aesar #10950) folded into a pocket shape with the edges tightly crimped in order to reduce the carbon flux on the inside growth surface. The copper pocket was annealed at 1 mTorr and 1030 °C under a hydrogen background for 15 minutes. Graphene was grown at 10 mTorr and 1000 °C by flowing methane and hydrogen over the foil at flow rates of 1 sccm and 2 sccm, respectively. By varying the growth time between 30 – 120 minutes, the size of the graphene patches can be controlled between 20-250 µm in dimension. Following growth, the sample was cooled rapidly by removing the oven from the growth region of the process tube. The sample was cooled to ambient temperature under a flow of methane and hydrogen at growth flow rates. Continuous sheets of polycrystalline, small-grain graphene were grown on 25 µm thick copper foil (Alfa Aesar #13382). The copper foil was annealed under a hydrogen background at a pressure of 50 mTorr and a temperature of 800 °C for 12 hours. Graphene was subsequently grown by flowing methane and hydrogen over the foil at rates of 35 sccm and 2 sccm, respectively. The growth was conducted at 300 mTorr and 1000 °C for 30 minutes. The sample was cooled rapidly using a similar procedure to that described for large-grain graphene growth. Figure S1 (a) shows an optical micrograph of film of small-grain graphene transferred onto SiO2. The grain size of our small-grain graphene was determined using dark-field transmission electron microscopy (DF-TEM). Figure S1 (b) shows a false-colored DF-TEM map of the crystal domains of small-grain CVD graphene, where each color indicates a different crystallographic orientation. The DFTEM image presented in Figure S1 (b) demonstrates that the small-grain graphene produced in this work has grain dimensions ranging between 0.2-5 µm. The inset of Figure S1 (b) shows the corresponding electron diffraction pattern, with each orientation present in the DF-TEM image yielding
2
one set of six-fold symmetric spots; the multiple sets of spots present in the electron diffraction pattern indicate that the graphene film is polycrystalline, with randomly oriented grains.
Figure S1. (a) Optical micrograph of a sheet of small-grain graphene transferred onto SiO2. The underlying SiO2 is visible at the edges of the transferred graphene sheet. (b) DF-TEM image of smallgrain CVD graphene. The inset of (b) is the electron diffraction patterns corresponding to the area of DF-TEM image; colored circles outline spots corresponding to different grain orientations.
Dry-Transfer Method Graphene was transferred onto SiO2 substrates for analysis using a dry-transfer method. A diagram of the dry-transfer process is shown in Figure S2. Prior to transfer from the copper growth substrate, the graphene surface was spin-coated with a layer of poly(methyl methacrylate) (PMMA) which provides mechanical support to the graphene throughout the transfer process. The copper/graphene/PMMA stack was then cut into square sections, and the copper side was subsequently adhered over windows cut in pieces of polyimide tape (3M #5413). The exposed copper was chemically etched in an ammonium persulfate copper etchant (Transene APS-100). The resulting graphene/PMMA membrane, suspended across the window in the tape, was then rinsed in DI water, followed by isopropanol, and gently dried in
3
a stream of nitrogen gas. The exposed graphene surface of the suspended membrane was then directly applied to a dielectric surface and subsequently freed from the tape support.
Figure S2. Process flow of dry-transfer procedure for CVD graphene.
The dry-transfer process is also compatible with an aligned transfer procedure used to transfer singlecrystals of large-grain CVD graphene onto pre-selected flakes of hexagonal boron nitride (h-BN) exfoliated onto silicon substrates with 285 nm of thermally grown SiO2. Figure S3 shows a diagram of the aligned transfer process. The CVD graphene is coated in PMMA, and the graphene/PMMA membrane is suspended across a window cut in polyimide tape following the procedures outlined for the dry-transfer method. Next, the PMMA surface of the suspended graphene/PMMA membrane is applied over a hole cut in a glass slide, and the tape support is cut free and removed. The slide is then inserted into a micropositioner stage, aligned over an exfoliated h-BN flake under a microscope, and brought into contact with the h-BN flake. The sample is heated at 180 °C for 10 minutes prior to removing the
4
PMMA in acetone, leaving the CVD graphene crystal dry-transferred onto the surface of the selected hBN flake.
Figure S3. Process flow for aligned, dry-transfer method used to transfer CVD graphene crystals onto exfoliated h-BN flakes.
Hall Bar Fabrication Once transferred, standard electron-beam lithography, reactive ion etching, and physical vapor deposition processes were used to pattern the graphene samples into Hall bar geometry and fabricate Cr/Pd/Au electrodes to contact the graphene structures. After device processing, samples were annealed in a tube furnace under a forming gas background for 4.5 hours at 345 °C. Completed Hall bar devices were measured with atomic force microscopy (AFM) using silicon cantilevers operated in non-contact mode. Figure S4 shows an AFM image of a completed Hall bar device fabricated from large-grain CVD graphene transferred onto h-BN.
5
Figure S4. AFM image of completed Hall bar fabricated from large-grain CVD graphene transferred onto h-BN. The graphene Hall bar is apparent in the center of the image, and metal electrodes contacting the graphene are visible near the edges of the AFM image.
Electrical Transport Data Parameters extracted from electrical transport data are presented in Table S1 for all devices measured. Tabulated parameters are: location of charge neutrality point with respect to gate voltage, CNP; full width at half maximum of resistivity peaks as a function of carrier density, ∆WCNP; minimum conductivity, σmin; contribution to resistivity from short-range scattering, ρs; density-independent mobility attributed to charged-impurity scattering, µc; and density-dependent, field effect mobility, µFE. Tabulated parameters correspond to transport data measured at 1.6 K.
6
Table S1. Parameters extracted from electrical transport data. Values are presented for all devices measured at 1.6 K.
Figure S5 shows a histogram which plots density-independent mobility, µc, values measured for multiple exfoliated samples fabricated on h-BN measured at 1.6 K. Mobility values range between 16,100 – 60,000 cm2V-1s-1 in the samples measured. Large-grain CVD graphene samples fabricated on h-BN measured in this work demonstrated values of µc ranging between 27,200 – 44,900 cm2V-1s-1. For samples fabricated on h-BN, figure S5 shows that the mobility values measured in large-grain CVD graphene are equivalent to those obtained measured in exfoliated samples.
7
Figure S5. Histogram of all density-independent mobility, µc, values measured for exfoliated samples fabricated on h-BN. Data taken at 1.6 K.
Figure S6 plots (a) resistivity as a function of gate voltage and (b) field-effect mobility as a function of carrier density for the highest mobility sample measured fabricated from small-grain CVD graphene on h-BN. The value of field-effect mobility is seen to vary from ~15,000 cm2V-1s-1 at high-density to over 70,000 cm2V-1s-1 near the CNP.
Figure S6. (a) Resistivity plotted as a function of gate voltage and (b) field-effect mobility plotted as a function of carrier density for the highest mobility sample measured fabricated from small-grain CVD sample on h-BN. 8
Department of Mechanical Engineering, ‡Department of Electrical Engineering, Columbia University, New York, New York 10027, United States
§
School of Applied and Engineering Physics, Kavli Institute at Cornell University, Cornell University, Ithaca, New York 14853, United States
1
Graphene Growth by Chemical Vapor Deposition Discrete, star-shaped patches of large-grain graphene were grown by low-pressure chemical vapor deposition (CVD) using methane as a carbon feedstock. The graphene was grown on the inside surface of 25 µm thick copper foil (Alfa Aesar #10950) folded into a pocket shape with the edges tightly crimped in order to reduce the carbon flux on the inside growth surface. The copper pocket was annealed at 1 mTorr and 1030 °C under a hydrogen background for 15 minutes. Graphene was grown at 10 mTorr and 1000 °C by flowing methane and hydrogen over the foil at flow rates of 1 sccm and 2 sccm, respectively. By varying the growth time between 30 – 120 minutes, the size of the graphene patches can be controlled between 20-250 µm in dimension. Following growth, the sample was cooled rapidly by removing the oven from the growth region of the process tube. The sample was cooled to ambient temperature under a flow of methane and hydrogen at growth flow rates. Continuous sheets of polycrystalline, small-grain graphene were grown on 25 µm thick copper foil (Alfa Aesar #13382). The copper foil was annealed under a hydrogen background at a pressure of 50 mTorr and a temperature of 800 °C for 12 hours. Graphene was subsequently grown by flowing methane and hydrogen over the foil at rates of 35 sccm and 2 sccm, respectively. The growth was conducted at 300 mTorr and 1000 °C for 30 minutes. The sample was cooled rapidly using a similar procedure to that described for large-grain graphene growth. Figure S1 (a) shows an optical micrograph of film of small-grain graphene transferred onto SiO2. The grain size of our small-grain graphene was determined using dark-field transmission electron microscopy (DF-TEM). Figure S1 (b) shows a false-colored DF-TEM map of the crystal domains of small-grain CVD graphene, where each color indicates a different crystallographic orientation. The DFTEM image presented in Figure S1 (b) demonstrates that the small-grain graphene produced in this work has grain dimensions ranging between 0.2-5 µm. The inset of Figure S1 (b) shows the corresponding electron diffraction pattern, with each orientation present in the DF-TEM image yielding
2
one set of six-fold symmetric spots; the multiple sets of spots present in the electron diffraction pattern indicate that the graphene film is polycrystalline, with randomly oriented grains.
Figure S1. (a) Optical micrograph of a sheet of small-grain graphene transferred onto SiO2. The underlying SiO2 is visible at the edges of the transferred graphene sheet. (b) DF-TEM image of smallgrain CVD graphene. The inset of (b) is the electron diffraction patterns corresponding to the area of DF-TEM image; colored circles outline spots corresponding to different grain orientations.
Dry-Transfer Method Graphene was transferred onto SiO2 substrates for analysis using a dry-transfer method. A diagram of the dry-transfer process is shown in Figure S2. Prior to transfer from the copper growth substrate, the graphene surface was spin-coated with a layer of poly(methyl methacrylate) (PMMA) which provides mechanical support to the graphene throughout the transfer process. The copper/graphene/PMMA stack was then cut into square sections, and the copper side was subsequently adhered over windows cut in pieces of polyimide tape (3M #5413). The exposed copper was chemically etched in an ammonium persulfate copper etchant (Transene APS-100). The resulting graphene/PMMA membrane, suspended across the window in the tape, was then rinsed in DI water, followed by isopropanol, and gently dried in
3
a stream of nitrogen gas. The exposed graphene surface of the suspended membrane was then directly applied to a dielectric surface and subsequently freed from the tape support.
Figure S2. Process flow of dry-transfer procedure for CVD graphene.
The dry-transfer process is also compatible with an aligned transfer procedure used to transfer singlecrystals of large-grain CVD graphene onto pre-selected flakes of hexagonal boron nitride (h-BN) exfoliated onto silicon substrates with 285 nm of thermally grown SiO2. Figure S3 shows a diagram of the aligned transfer process. The CVD graphene is coated in PMMA, and the graphene/PMMA membrane is suspended across a window cut in polyimide tape following the procedures outlined for the dry-transfer method. Next, the PMMA surface of the suspended graphene/PMMA membrane is applied over a hole cut in a glass slide, and the tape support is cut free and removed. The slide is then inserted into a micropositioner stage, aligned over an exfoliated h-BN flake under a microscope, and brought into contact with the h-BN flake. The sample is heated at 180 °C for 10 minutes prior to removing the
4
PMMA in acetone, leaving the CVD graphene crystal dry-transferred onto the surface of the selected hBN flake.
Figure S3. Process flow for aligned, dry-transfer method used to transfer CVD graphene crystals onto exfoliated h-BN flakes.
Hall Bar Fabrication Once transferred, standard electron-beam lithography, reactive ion etching, and physical vapor deposition processes were used to pattern the graphene samples into Hall bar geometry and fabricate Cr/Pd/Au electrodes to contact the graphene structures. After device processing, samples were annealed in a tube furnace under a forming gas background for 4.5 hours at 345 °C. Completed Hall bar devices were measured with atomic force microscopy (AFM) using silicon cantilevers operated in non-contact mode. Figure S4 shows an AFM image of a completed Hall bar device fabricated from large-grain CVD graphene transferred onto h-BN.
5
Figure S4. AFM image of completed Hall bar fabricated from large-grain CVD graphene transferred onto h-BN. The graphene Hall bar is apparent in the center of the image, and metal electrodes contacting the graphene are visible near the edges of the AFM image.
Electrical Transport Data Parameters extracted from electrical transport data are presented in Table S1 for all devices measured. Tabulated parameters are: location of charge neutrality point with respect to gate voltage, CNP; full width at half maximum of resistivity peaks as a function of carrier density, ∆WCNP; minimum conductivity, σmin; contribution to resistivity from short-range scattering, ρs; density-independent mobility attributed to charged-impurity scattering, µc; and density-dependent, field effect mobility, µFE. Tabulated parameters correspond to transport data measured at 1.6 K.
6
Table S1. Parameters extracted from electrical transport data. Values are presented for all devices measured at 1.6 K.
Figure S5 shows a histogram which plots density-independent mobility, µc, values measured for multiple exfoliated samples fabricated on h-BN measured at 1.6 K. Mobility values range between 16,100 – 60,000 cm2V-1s-1 in the samples measured. Large-grain CVD graphene samples fabricated on h-BN measured in this work demonstrated values of µc ranging between 27,200 – 44,900 cm2V-1s-1. For samples fabricated on h-BN, figure S5 shows that the mobility values measured in large-grain CVD graphene are equivalent to those obtained measured in exfoliated samples.
7
Figure S5. Histogram of all density-independent mobility, µc, values measured for exfoliated samples fabricated on h-BN. Data taken at 1.6 K.
Figure S6 plots (a) resistivity as a function of gate voltage and (b) field-effect mobility as a function of carrier density for the highest mobility sample measured fabricated from small-grain CVD graphene on h-BN. The value of field-effect mobility is seen to vary from ~15,000 cm2V-1s-1 at high-density to over 70,000 cm2V-1s-1 near the CNP.
Figure S6. (a) Resistivity plotted as a function of gate voltage and (b) field-effect mobility plotted as a function of carrier density for the highest mobility sample measured fabricated from small-grain CVD sample on h-BN. 8
