CVD GrapheneAug 1, 2017 | ACS MATERIAL LLC
Graphene, as manufactured by the CVD method, has been in the forefront of many research initiatives due to the beneficial extremes it exhibits for many different characteristics. Synthesis requires a catalyst (usually copper or nickel), carrier gases (usually from H2, Ar), feed material (methane) and a proper atmospheric environment. The list of potential use-cases for this remarkable substance grows daily, and techniques for growing, handling and transferring the graphene are also increasing. In this paper, we will explore these aspects of CVD graphene and look at likely future scenarios.
Introduction to CVD Graphene
With growing interests in graphene on the rise due to its attractive properties (i.e. high elasticity and mechanical strength), fabricating this material in large quantities is essential for research and application purposes. Chemical vapor deposition (CVD) has been the most popular and current method in producing graphene molecularly and is becoming more common as it can produce graphene with a huge surface area. With this method, graphene is deposited onto a transition metal substrate that is easily etched by an acid solution and can therefore be transferred onto another substrate such as silicon dioxide; this provides the opportunity to utilize graphene for various applications that can potentially replace silicon technology. However, low sheet resistance per square of graphene on a copper substrate along with a high optical transparency can potentially make great transparent conductive films.
The CVD Method
The CVD process involves a thin metal substrate that is placed into a heated furnace at high temperatures (900 to 1000°C) and in low vacuum. Decomposed methane gas then supplies the carbon and hydrogen gases needed to flow through a chamber to cause a reaction between methane and the surface of the metal film, typically copper, but nickel and cobalt has reportedly been used.1 The dilemma with using nickel, for instance, is that large amounts of carbon sources are absorbed on nickel foils and form thick pieces of graphite crystals rather than a more desired layer of graphene. Thin nickel films of less than 300 nm must then be evaporated onto a SiO2/Si surface before graphene growth. Copper, however, attracts carbon at its surface which can form a more preferred graphene structure. The chamber is then set with a fast cooling rate to suppress the formation of multiple layers and separate the graphene layers from the substrate.2
Figure 1. A single layer of graphene is deposited onto a copper substrate and can then be etched.
While this method has its perks, the outcome of the graphene can be affected by a number of ways. First, the varying cooling conditions (cooling rate and hydrocarbon concentration) can influence different nucleation and growth behavior that affects the quality of the graphene being deposited onto the substrate.3 Meanwhile, poor quality of the copper substrate devalues the outcome of the graphene by reducing nucleation density but this can be subsided by a pre-treatment, such as a wet-chemical treatment, in which the copper substrate is soaked in acetic acid to prevent oxidation.4 An increased supply of carbon atoms that is deposited onto the substrate along with an increased amount of hydrogen gases that promotes a reaction on the substrate surface due to a large quantity of reaction gases from methane can also affect the graphene’s quality.
Before analyzing the graphene obtained from the CVD method, the copper substrate used in the process can be etched and transferred onto another substrate of choice before inspecting the quality of the graphene obtained. While other substrates may be used, a SiO2/Si substrate with approximately 300 nm of SiO2 layer, is typically chosen to be adsorbed by graphene for its insulating characteristic. Silicon dioxide wafers tend to have a smoother surface and graphene is highly compatible with silicon technology and it could potentially take over future applications.5 After the graphene transfer, using a Raman spectroscopy can begin to determine the characterization of the graphene sample.
Raman spectroscopy is commonly used to observe molecular vibrations from the graphene sample and it utilizes monochromatic light from a laser ranging from visible, near infrared or near ultraviolet sources. Most of the scattered light has the same frequency as the excitation source which is considered to be Rayleigh (or elastic) scattering but it would have to be filtered since the Raman spectroscopy relies on Raman (or inelastic) scattering. With inelastic scattering, the incident photon either gains (anti-Stokes scattering) or loses (Stokes scattering) an amount of energy equal to the vibrational mode.8 In addition, the anti-Stokes lines in the Raman spectra can be found when molecules of the material are already in an excited state. This spectroscopy instrument is used to determine the quality and types of edge, number of layers in multilayer graphene (up to 4 layers), the effects of perturbations in order to gain insight into the carbon allotrope, structural defects, functional groups, unwanted by-products, chemical modifications during preparation, and differentiating between a single or bilayer graphene.6
There are different peaks, or bands, that reveal the characteristics of the graphene. The G band is a sharp peak that appears around 1587 cm-1 for a single layer in the spectrum and indicates the number of layers present in the sample. Meanwhile, the D band signifies any defects or deformity of the graphene; a prominent peak will indicate many defects, or even residue, in the material and the defects in the sample is directly proportional to the intensity of the D band. The 2D band is known as a second order of the D band but can also referred to as a G’ band.7 It does not determine the graphene defects but it can analyze the thickness of the graphene layer while differentiating graphene layers for up to four layers. The more layers of graphene that exists, the more the 2D band will split into overlapping modes. The 2D peak is commonly found around 2650 cm-1 on the spectra while the use of a silicon substrate will leave its own peak around 500 cm-1.
Figure 2. Raman spectrum for Graphene on a silicon dioxide substrate for a single layer graphene.
The G and 2D Raman peaks will change shapes, positions and relative intensity according to the number of graphene layers.9 The 2D band is usually more intense than the G band for a single-layer graphene and is observed due to resonant processes while exhibiting dispersive tendencies as it is dependent on the Raman band position from the laser excitation.11 The peak intensity in the G band gets higher as the number of layers increases while the 2D band decreases. The different band shape compared to the G band allows the 2D band to differentiate between a single or multilayer graphene and it is dependent on not only its band position but its band shape. For more than two layers, the 2D peak shifts to a greater wavenumber and the full-width at half-maximum (FWHM) broadens as it determines layer thickness of the graphene sample.
The G band intensity shows a linear trend as the graphene layer increases from a single layer to a multilayer. Due to the different positions and shapes of the G and 2D bands, the relative intensity will greatly differ and its intensity ratio will increase with each additional layers of graphene. The intensity can be found with the following ratio:
IG/I2D ~ 0.5 (for single-layer graphene).
A Raman spectrum shows an intensity-versus-wavelength shift measured in arbitrary units and cm-1, respectively.10
While the CVD technique has proven to be an effective method in obtaining quality graphene, a high sheet resistance is to blame for the poor performance of organic electronic devices that uses graphene-based transparent electrodes.11 An undoped graphene is a one-atom thick, two-dimensional crystalline allotrope that has a sheet resistance (Rsheet) of approximately 6 kΩ with 98% transparency. When utilizing the CVD method with a copper substrate, the sheet resistance was found to be 350 Ω/sq while holding a 90% transparency; this signifies the advance of CVD graphene to be used as transparent conductive films. Graphene that has been synthesized by this process shows a better transparency/Rsheet ratio.12 When there are more graphene layers, the sheet resistance decreases for each layer that has been added. Hypothetically, sheet resistance should be constant and related to that of a multilayer film if the layers behave independent of each other.13
Polymethyl-methacrylate (PMMA) is commonly used as a temporary support layer, typically around 1-2 nm in thickness, to be placed on top of graphene to prevent tearing or cracking during the transfer after the CVD process.14 This polymer is highly desirable based on its availability, admirable material properties and ease of processing. However, there is an issue that stems from residual polymer left behind on the graphene surface which can alter its intrinsic properties. In general, rinsing the PMMA film that carries the graphene layer in distilled water can reduce contamination before transferring to a different substrate. From there, an organic solvent such as acetone can proceed to remove the polymer from the graphene film.
While a Raman spectroscopy may show clear G and 2D bands with little to no defects of a high-quality graphene film, it will reveal that the polymer has been removed after using an acetone treatment when viewing the spectrum. However, recent studies have shown a new way of detecting polymer residue by means of an isotope labeling of PMMA and time-of-flight secondary ion mass spectroscopy (ToF-SIMS). This instrument bombards the sample with a primary ion beam that leads to the sputtering of the surface. A small portion of the sputtered particles are then ionized (secondary ions) and the sample composition can be determined.15,16 ToF-SIMS can be a complementary approach alongside the Raman analysis when it comes to characterizing the chemical composition and surface contamination of materials.17
CVD synthesis has been the most popular and cost-effective method for preparing and producing large quantities of graphene for further research and technical applications. Using this process to deposit graphene onto metal substrates (i.e. copper) has paved its way towards new discoveries in electronic properties, such as those that are found in transistors, over the past decade. The ability to create a single layer, high-quality structure and its ease in transferring graphene from one substrate to another makes CVD graphene ideal in the advancement of technology.
ACS Material Products:
1. Bhaviripudi, Sreekar, et al. “Role of Kinetic Factors in Chemical Vapor Deposition Synthesis of Uniform Large Area Graphene Using Copper Catalyst.” ACS Publications, pubs.acs.org/doi/abs/10.1021/n1102355e. Accessed 25 Sept. 2017.
2. Kim, Keun Soo, et al. “Large-Scale pattern growth of graphene films for stretchable transparent electrodes.” Nature, 14 Jan. 2009, doi: 10.1038/nature07719.
3. Choi, Dong Soo, et al. “Effect of Cooling Condition on Chemical Vapor Deposition Synthesis of Graphene on Copper Catalyst.” ACS Applied Materials & Interfaces, vol. 6, no. 22, 2014, pp. 19574-19578., doi: 10.1021/am503698h.
4. Braeuninger-Weimer, Philipp, et al. “Understanding and Controlling Cu-Catalyzed Graphene Nucleation: The Role of Impurities, Roughness, and Oxygen Scavenging.” Chemistry of Materials, American Chemical Society, 27 Dec. 2016, www.ncbi.nlm.nih.gov/pmc/articles/PMC5261424/. Accessed 25 Sept. 2017.
5. Dang, Xuejie, et al. “Semiconducting Graphene on Silicon from First-Principles Calculations.” ACS Publications, pubs.acs.org/doi/abs/10.1021/acsnano.5b03722. Accessed 25 Sept. 2017.
6. Ferrari, Andrea C., and Denis M. Basko. “Raman spectroscopy as a versatile tool for studying the properties of graphene.” Nature Nanotechnology, 4 Apr. 2013, doi: 10.1038/nnano.2013.46.
7. Park, J. S., et al. “G’ band Raman spectra of single, double and triple layer graphene.” Carbon, Pergamon, 14 Jan. 2009, www.sciencedirect.com/science/article/pii/S000862230900030X. Accessed 25 Sept. 2017.
8. Camp Jr., Charles H., and Marcus T. Cicerone. “Chemically sensitive bioimaging with coherent Raman scattering.” Nature Photonics, 29 Apr. 2015, doi: 10.1038/nphoton.2015.60.
9. Ferrari, Andrea C. “Raman Spectroscopy of graphene and graphite: Disorder, electron-phonon coupling, doping and nonadiabatic effects.” Solid State Communications, 27 Apr. 2007.
10.Bumbrah, Gurvinder Singh, and Rakesh Mohan Sharma. “Raman spectroscopy – Basic principle, instrumentation and selected applications for the characterization of drugs of abuse.” Egyptian Journal of Forensic Sciences, 23 June 2015. Accessed 3 Oct. 2017.
11.Choy, Wallace C. H. Organic Solar Cells: Materials and Device Physics. Springer, 2013.
12.Arco, Lewis Gomez De, et al. “Continuous, Highly Flexible, and Transparent Graphene Films by Chemical Vapor Deposition for Organic Photovoltaics.” ACS Nano, vol. 4, no. 5, 2010, pp. 2865-2873., doi: 10.1021/nn901587x.
13. Li, Xuesong, et al. “Transfer of Large-Area Graphene Films for High-Performance Transparent Conductive Electrodes.” Nano Letters, vol. 9, no.12, Sept. 2009, pp. 4359-4363., doi: 10.1021/nl902623y.
14.Wang, Xiaohan, et al. “Direct Observation of Poly(Methyl Methacrylate) Removal from a Graphene Surface.” ACS Publications, pubs.acs.org/doi/abs/10.1021/acs.chemmater.6b03875?journalCode=cmatex.
15.Chou, Harry, et al. “Revealing the planar chemistry of two-Dimensional heterostructures at the atomic level.” Nature News, Nature Publishing Group, 23 June 2015, www.nature.com/articles/ncomms8482.
16.Michalowski, Pawel Piotr, et al. “Graphene Enhanced Secondary Ion Mass Spectrometry (GESIMS).” Nature News, Nature Publishing Group, 7 Aug. 2017, www.nature.com/articles/s41598-017-07984-1.
17.Pollard, Andrew J. “Metrology for Graphene and 2-D Materials.” 17th International Congress of Metrology, 2015, doi: 10.1051/metrology/201514001.