Graphene films transfer using marker-frame method

Graphene on a copper substrate has attracted great attention of researchers and industrialists all over the world, mostly due to its high quality and the possibility of achieving a monolayer graphene film which can be efficiently transferred and easily implemented in mass production.^1,2 A standard approach to graphene's transferring from copper substrates onto arbitrary substrates involves employing poly(methyl methacrylate) as a supporting layer preventing the collapse of the graphene layer during the transferring process.^3,4 Other popular methods use polydimethylsiloxane as a stamp enabling the detachment of graphene from Cu and its placement on substrates.^5,6 Moreover, it has been shown that the roll-to-roll method with a thermal release tape as a support can be successfully used for mass-scale production.^2,7 Nevertheless, each of these techniques requires the removal of the supporting layers by supplying dissolvent agents or adjusting the temperature. In turn, this causes additional impurities on the graphene surface. What is more, the application of dissolvents limits the range of potential substrates only to those resistant to the used solvent. It is also known that PMMA and PDMS films are hard to remove from the graphene surface, hence leaving residues on top of it. Putting graphene on top of nanowires or nanoparticles by polymer-evolving methods to form nanocomposites is another serious challenge due to the fact that graphene's adhesion to polymers is stronger than to extended surfaces. To solve the aforementioned problem, one can search for methods which do not require any polymer-like supporting layers. One of them involves the usage of a drop of isopropanol and a TEM grid; however, this compound restricts the transfer of the graphene layer to TEM grids.⁸ Recently, a polymer-free process of graphene transferring has been demonstrated.⁹ It has been realized by applying a graphite holder keeping the graphene/copper system in place during the stages of copper etching and cleaning. This technique necessitates a specially prepared holder though. Such mechanical support can both destroy the graphene layer when taken out and limit the shape and size of the transferred graphene films.

In the present work, we demonstrate how to transfer graphene from copper foils onto arbitrary substrates without using polymer-like supporting films. Our technique guarantees cleaner graphene surface and no polymer residues on it. Moreover, the said procedure allows transferring graphene onto substrates not resistant to acetone, which are typically used for removing polymers. Consequently, it significantly increases the applicability of graphene. What is more, one does not need a special holder or any other sophisticated tools. When following the method, it is possible to transfer an unrestricted shape of graphene layers even onto rough and expended surfaces like nanowires or materials functionalized with nanoparticles.

The graphene films were synthesized by chemical vapor deposition on the surface of 25 μm thick copper foils as described in Ref. 10. In the first step we aimed at confirming the quality of the obtained graphene films by performing Raman spectroscopy measurements. Then we removed graphene from the backside of copper foil to avoid impurities between the top and lateral graphene films formed during copper etching. Afterwards, we selected the region of the samples which we wanted to transfer onto arbitrary substrates and then we marked it with a waterproof marker. In consequence, a stable plastic frame was formed. The marker-frame was on the graphene surface and, therefore, the binder present in the marker's ink associated with the graphene layer underneath. It is recommended that the marker-frame should be larger than the arbitrary substrate. As a result, we can prevent the removal of the marker, e.g. in contact with alcohol. Owing to the fact that the connection between the graphene layer and the frame exists, the polycrystalline graphene film does not rip. Moreover, the marker-frame was very thin and light, thus preventing graphene from cracking and drowning. Next, the graphene sample with the marker-frame was put on the surface of an aqueous solution of ammonium persulfate. When Cu was completely etched, graphene floating on the surface of the solution was cleaned with continuous and controlled flow of DI water. In the end, the water in the vessel was released through a tap and graphene surrounded by the marker-frame fell onto the target substrate. Then they were all gradually heated at above 100°C in an air atmosphere to improve both adhesion and contact between them. Figure 1 presents a scheme of the applied method. In this way we transferred graphene films onto GaN nanowires, the density of whose clusters equaled 30/μm². The graphene/SiO₂/Si samples were prepared as described in Ref. 10. The characterization of the properties of graphene transferred following the enhanced marker-frame method was performed by Raman spectroscopy using a Renishaw system with a 532 nm Nd:YAG laser as an excitation source, as well as by carrying out SEM imaging and conducting TEM investigations.

Each of the samples with the transferred graphene films on top was first characterized by performing spatial Raman mapping. The main goal of Raman measurements was to collate information on the features of graphene as a function of substrate roughness. There were areas of reduced contact on the surface, which means that the examined graphene bears stronger resemblance to suspended graphene. The examination of the samples was carried out at a micro-scale to analyze the homogeneity and continuity of the graphene structure. Spatial Raman mapping was carried out on a micro-scale (101 × 101 μm²) to show the average graphene parameters and to present the homogeneity of the graphene structure. The laser spot diameter on the sample surface was approximately 0.3 μm. Raman spectroscopy provided information on the formation of the graphene structure and made distinguishing between mono and bilayer graphene and its strains possible.^11–14 Consequently, in the presented work we showed the histograms of the positions of the G and 2D bands, the width of the 2D band and the relative intensities of the 2D and G bands. For monolayer graphene, which was the case in our investigation, the G and 2D peaks were fitted to a single Lorentzian. Taking into account the abovementioned considerations, one can conclude that graphene transferred onto GaN NWs and SiO₂/Si substrates is homogenous. In the case of graphene transferred onto GaN NWs the mean positions of the 2D and G bands are located at 2678 cm⁻¹ and 1585 cm⁻¹ (Figure 2(a) and 2(b)), respectively, which suggests no stress in the graphene films, and these values are very close to the values characteristic of free standing graphene.^14,15 FWHM (Full Width of Half Maximum) of the G band presented in Fig. 3(b) (average 16 cm⁻¹) indicates small carrier concentration (below 2*10¹² cm⁻²).^16,17 The histogram of the graphene film clearly shows a mean 2D/G intensity ratio of above 2 (maximum at 4.5), thus confirming low carrier density in this sample. Fig. 3(a) confirms the presence of graphene (FWHM 2D of about 34 cm⁻¹) in the case of the graphene/SiO₂/Si samples. In contrast to the sample on GaN NWs, this sample shows a slight compressive strain (blueshift of the G and the 2D band position (1591 cm⁻¹ and 2687 cm⁻¹ respectively) with respect to freestanding graphene). In addition, the FWHM of G band (average 11 cm⁻¹) suggests that the sample is slightly doped (carrier concentration of about 5*10¹² cm⁻²). Moreover, a lower 2D/G ratio confirms higher carrier concentration for graphene on SiO₂ than for GaN NWs.

The analysis of the Raman maps proves that identical graphene samples transferred on GaN NWs and SiO₂ have slightly different parameters. It suggests that differences in the structure of graphene on both materials are mainly linked with its interaction with the substrates. Based on the abovementioned detailed results one can conclude that the features of graphene transferred on top of GaN NWs correspond closer to those of freestanding graphene.

To demonstrate the morphology and cleanliness of our samples, a low-kV SEM imaging was applied. We investigated the surface of graphene covering the GaN nanowire films using the In-Lens secondary electrons detector (true SE1) and the Energy selective Backscattered electron (EsB, low-loss BSE) detector, both positioned on the optical axis of the Gemini™ column of the Auriga CrossBeam Workstation (Carl Zeiss). The energy of primary electrons in the scanning beam was selected for 500 eV so as to reveal the morphology of the ultra-thin layer of graphene (single layer) and simultaneously distinguish different phases present on the substrate basing on the compositional contrast (low-loss BSE). The images presented in Fig. 4 reveal the surface without external impurities.

For the purpose of TEM analysis carried out using the Titan Cubed 80–300 transmission electron microscope at 300 kV, the samples were cleaned in argon-oxygen plasma (containing 80% of argon and 20% of oxygen) for 2 sec. A longer cleaning time caused significant damage to graphene. Microscopic studies revealed the presence of a single layer of graphene over the entire area supporting the grid. Fig. 5(a) presents the Fourier Filtered HRTEM image that was obtained with a filter applied to the FFT of the raw images. The figure also shows the dimensioned distance that corresponds to the values shown in the FFT image in Fig. 5(b). No other reflections than those of a single layer of graphene were observed. The sample's tilting at an angle of −/+ 32 degrees did not result in additional reflections, which also proves the presence of a single layer of graphene.

In summary, we demonstrated a new method of graphene films transferring applying a marker-frame instead of polymer-like films as a support. This method is an alternative that is much faster, cheaper and freely available to all. Furthermore, it opens up all kinds of possibilities. For instance, one can think of using graphene as a component of nanocomposites, MEMS membranes or coupling it with other nanomaterials. As an example we showed graphene films suspended on nanowires. For the graphene films transferred by the marker-frame method, Raman spectra proved the formation of graphitic structures, whereas SEM images showed the morphology of the graphene-substrates interface and confirmed the uniformity of the graphene layer. TEM investigations revealed a hexagonal lattice of graphene.

This work was partially supported by the EU-FET grant GRAPHENICS 618086, the National Science Centre Nr DEC-2012/05/N/ST3/03163, UMO-2013/09/N/ST5/02481 and No. POIG.02.01-00-14-032/08. The research leading to these results has also received funding from the European Union Seventh Framework Programme under grant agreement n 604391 Graphene Flagship.