Thank you for visiting Nature.com. The version of browser you are using has limited CSS support. For best results, we recommend that you use a newer version of your browser (or disable Compatibility Mode in Internet Explorer). In the meantime, to ensure ongoing support, we are displaying the site without styling or JavaScript.
Nanoscale graphite films (NGFs) are robust nanomaterials that can be produced by catalytic chemical vapor deposition, but questions remain about their ease of transfer and how surface morphology affects their use in next-generation devices. Here we report the growth of NGF on both sides of a polycrystalline nickel foil (area 55 cm2, thickness about 100 nm) and its polymer-free transfer (front and back, area up to 6 cm2). Due to the morphology of the catalyst foil, the two carbon films differ in their physical properties and other characteristics (such as surface roughness). We demonstrate that NGFs with a rougher backside are well suited for NO2 detection, while smoother and more conductive NGFs on the front side (2000 S/cm, sheet resistance – 50 ohms/m2) can be viable conductors. channel or electrode of the solar cell (since it transmits 62% of visible light). Overall, the described growth and transport processes may help realize NGF as an alternative carbon material for technological applications where graphene and micron-thick graphite films are not suitable.
Graphite is a widely used industrial material. Notably, graphite has the properties of relatively low mass density and high in-plane thermal and electrical conductivity, and is very stable in harsh thermal and chemical environments1,2. Flake graphite is a well-known starting material for graphene research3. When processed into thin films, it can be used in a wide range of applications, including heat sinks for electronic devices such as smartphones4,5,6,7, as an active material in sensors8,9,10 and for electromagnetic interference protection11. 12 and films for lithography in extreme ultraviolet13,14, conducting channels in solar cells15,16. For all of these applications, it would be a significant advantage if large areas of graphite films (NGFs) with thicknesses controlled in the nanoscale <100 nm could be easily produced and transported.
Graphite films are produced by various methods. In one case, embedding and expansion followed by exfoliation were used to produce graphene flakes10,11,17. The flakes must be further processed into films of the required thickness, and it often takes several days to produce dense graphite sheets. Another approach is to start with graphitable solid precursors. In industry, sheets of polymers are carbonized (at 1000–1500 °C) and then graphitized (at 2800–3200 °C) to form well-structured layered materials. Although the quality of these films is high, the energy consumption is significant1,18,19 and the minimum thickness is limited to a few microns1,18,19,20.
Catalytic chemical vapor deposition (CVD) is a well-known method for producing graphene and ultrathin graphite films (<10 nm) with high structural quality and reasonable cost21,22,23,24,25,26,27. However, compared with the growth of graphene and ultrathin graphite films28, large-area growth and/or application of NGF using CVD is even less explored11,13,29,30,31,32,33.
CVD-grown graphene and graphite films often need to be transferred onto functional substrates34. These thin film transfers involve two main methods35: (1) non-etch transfer36,37 and (2) etch-based wet chemical transfer (substrate supported)14,34,38. Each method has some advantages and disadvantages and must be selected depending on the intended application, as described elsewhere35,39. For graphene/graphite films grown on catalytic substrates, transfer via wet chemical processes (of which polymethyl methacrylate (PMMA) is the most commonly used support layer) remains the first choice13,30,34,38,40,41,42. You et al. It was mentioned that no polymer was used for NGF transfer (sample size approximately 4 cm2)25,43, but no details were provided regarding sample stability and/or handling during transfer; Wet chemistry processes using polymers consist of several steps, including the application and subsequent removal of a sacrificial polymer layer30,38,40,41,42. This process has disadvantages: for example, polymer residues can change the properties of the grown film38. Additional processing can remove residual polymer, but these additional steps increase the cost and time of film production38,40. During CVD growth, a layer of graphene is deposited not only on the front side of the catalyst foil (the side facing the steam flow), but also on its back side. However, the latter is considered a waste product and can be quickly removed by soft plasma38,41. Recycling this film can help maximize yield, even if it is of lower quality than face carbon film.
Here, we report the preparation of wafer-scale bifacial growth of NGF with high structural quality on polycrystalline nickel foil by CVD. It was assessed how the roughness of the front and back surface of the foil affects the morphology and structure of NGF. We also demonstrate cost-effective and environmentally friendly polymer-free transfer of NGF from both sides of nickel foil onto multifunctional substrates and show how the front and back films are suitable for various applications.
The following sections discuss different graphite film thicknesses depending on the number of stacked graphene layers: (i) single layer graphene (SLG, 1 layer), (ii) few layer graphene (FLG, < 10 layers), (iii) multilayer graphene ( MLG, 10-30 layers) and (iv) NGF (~300 layers). The latter is the most common thickness expressed as a percentage of area (approximately 97% area per 100 µm2)30. That’s why the whole film is simply called NGF.
Polycrystalline nickel foils used for the synthesis of graphene and graphite films have different textures as a result of their manufacture and subsequent processing. We recently reported a study to optimize the growth process of NGF30. We show that process parameters such as annealing time and chamber pressure during the growth stage play a critical role in obtaining NGFs of uniform thickness. Here, we further investigated the growth of NGF on polished front (FS) and unpolished back (BS) surfaces of nickel foil (Fig. 1a). Three types of samples FS and BS were examined, listed in Table 1. Upon visual inspection, uniform growth of NGF on both sides of the nickel foil (NiAG) can be seen by the color change of the bulk Ni substrate from a characteristic metallic silver gray to a matte gray color (Fig. 1a); microscopic measurements were confirmed (Fig. 1b, c). A typical Raman spectrum of FS-NGF observed in the bright region and indicated by red, blue and orange arrows in Figure 1b is shown in Figure 1c. The characteristic Raman peaks of graphite G (1683 cm−1) and 2D (2696 cm−1) confirm the growth of highly crystalline NGF (Fig. 1c, Table SI1). Throughout the film, a predominance of Raman spectra with intensity ratio (I2D/IG) ~0.3 was observed, while Raman spectra with I2D/IG = 0.8 were rarely observed. The absence of defective peaks (D = 1350 cm-1) in the entire film indicates the high quality of NGF growth. Similar Raman results were obtained on the BS-NGF sample (Figure SI1 a and b, Table SI1).
Comparison of NiAG FS- and BS-NGF: (a) Photograph of a typical NGF (NiAG) sample showing NGF growth at wafer scale (55 cm2) and the resulting BS- and FS-Ni foil samples, (b) FS-NGF Images/ Ni obtained by an optical microscope, (c) typical Raman spectra recorded at different positions in panel b, (d, f) SEM images at different magnifications on FS-NGF/Ni, (e, g) SEM images at different magnifications Sets BS -NGF/Ni. The blue arrow indicates the FLG region, the orange arrow indicates the MLG region (near the FLG region), the red arrow indicates the NGF region, and the magenta arrow indicates the fold.
Since growth depends on the thickness of the initial substrate, crystal size, orientation, and grain boundaries, achieving reasonable control of NGF thickness over large areas remains a challenge20,34,44. This study used content we previously published30. This process produces a bright region of 0.1 to 3% per 100 µm230. In the following sections, we present results for both types of regions. High magnification SEM images show the presence of several bright contrast areas on both sides (Fig. 1f,g), indicating the presence of FLG and MLG regions30,45. This was also confirmed by Raman scattering (Fig. 1c) and TEM results (discussed later in the section “FS-NGF: structure and properties”). The FLG and MLG regions observed on FS- and BS-NGF/Ni samples (front and back NGF grown on Ni) may have grown on large Ni(111) grains formed during pre-annealing22,30,45. Folding was observed on both sides (Fig. 1b, marked with purple arrows). These folds are often found in CVD-grown graphene and graphite films due to the large difference in the coefficient of thermal expansion between the graphite and the nickel substrate30,38.
The AFM image confirmed that the FS-NGF sample was flatter than the BS-NGF sample (Figure SI1) (Figure SI2). The root mean square (RMS) roughness values of FS-NGF/Ni (Fig. SI2c) and BS-NGF/Ni (Fig. SI2d) are 82 and 200 nm, respectively (measured over an area of 20 × 20 μm2). The higher roughness can be understood based on the surface analysis of the nickel (NiAR) foil in the as-received state (Figure SI3). SEM images of FS and BS-NiAR are shown in Figures SI3a–d, demonstrating different surface morphologies: polished FS-Ni foil has nano- and micron-sized spherical particles, while unpolished BS-Ni foil exhibits a production ladder. as particles with high strength. and decline. Low and high resolution images of annealed nickel foil (NiA) are shown in Figure SI3e–h. In these figures, we can observe the presence of several micron-sized nickel particles on both sides of the nickel foil (Fig. SI3e–h). Large grains may have a Ni(111) surface orientation, as previously reported30,46. There are significant differences in nickel foil morphology between FS-NiA and BS-NiA. The higher roughness of BS-NGF/Ni is due to the unpolished surface of BS-NiAR, the surface of which remains significantly rough even after annealing (Figure SI3). This type of surface characterization before the growth process allows the roughness of graphene and graphite films to be controlled. It should be noted that the original substrate underwent some grain reorganization during graphene growth, which slightly decreased the grain size and somewhat increased the surface roughness of the substrate compared to the annealed foil and catalyst film22.
Fine-tuning the substrate surface roughness, annealing time (grain size)30,47 and release control43 will help reduce regional NGF thickness uniformity to the µm2 and/or even nm2 scale (i.e., thickness variations of a few nanometers). To control the surface roughness of the substrate, methods such as electrolytic polishing of the resulting nickel foil can be considered48. The pretreated nickel foil can then be annealed at a lower temperature (< 900 °C) 46 and time (< 5 min) to avoid the formation of large Ni(111) grains (which is beneficial for FLG growth).
SLG and FLG graphene is unable to withstand the surface tension of acids and water, requiring mechanical support layers during wet chemical transfer processes22,34,38. In contrast to the wet chemical transfer of polymer-supported single-layer graphene38, we found that both sides of the as-grown NGF can be transferred without polymer support, as shown in Figure 2a (see Figure SI4a for more details). Transfer of NGF to a given substrate begins with wet etching of the underlying Ni30.49 film. The grown NGF/Ni/NGF samples were placed overnight in 15 mL of 70% HNO3 diluted with 600 mL of deionized (DI) water. After the Ni foil is completely dissolved, FS-NGF remains flat and floats on the surface of the liquid, just like the NGF/Ni/NGF sample, while BS-NGF is immersed in water (Fig. 2a,b). The isolated NGF was then transferred from one beaker containing fresh deionized water to another beaker and the isolated NGF was washed thoroughly, repeating four to six times through the concave glass dish. Finally, FS-NGF and BS-NGF were placed on the desired substrate (Fig. 2c).
Polymer-free wet chemical transfer process for NGF grown on nickel foil: (a) Process flow diagram (see Figure SI4 for more details), (b) Digital photograph of separated NGF after Ni etching (2 samples), (c) Example FS – and BS-NGF transfer to SiO2/Si substrate, (d) FS-NGF transfer to opaque polymer substrate, (e) BS-NGF from the same sample as panel d (divided into two parts), transferred to gold plated C paper and Nafion (flexible transparent substrate, edges marked with red corners).
Note that SLG transfer performed using wet chemical transfer methods requires a total processing time of 20–24 hours 38 . With the polymer-free transfer technique demonstrated here (Figure SI4a), the overall NGF transfer processing time is significantly reduced (approximately 15 hours). The process consists of: (Step 1) Prepare an etching solution and place the sample in it (~10 minutes), then wait overnight for Ni etching (~7200 minutes), (Step 2) Rinse with deionized water (Step – 3). store in deionized water or transfer to target substrate (20 min). Water trapped between the NGF and the bulk matrix is removed by capillary action (using blotting paper)38, then the remaining water droplets are removed by natural drying (approximately 30 min), and finally the sample is dried for 10 min. min in a vacuum oven (10–1 mbar) at 50–90 °C (60 min) 38.
Graphite is known to withstand the presence of water and air at fairly high temperatures (≥ 200 °C)50,51,52. We tested samples using Raman spectroscopy, SEM, and XRD after storage in deionized water at room temperature and in sealed bottles for anywhere from a few days to one year (Figure SI4). There is no noticeable degradation. Figure 2c shows free-standing FS-NGF and BS-NGF in deionized water. We captured them on a SiO2 (300 nm)/Si substrate, as shown at the beginning of Figure 2c. Additionally, as shown in Figure 2d,e, continuous NGF can be transferred to various substrates such as polymers (Thermabright polyamide from Nexolve and Nafion) and gold-coated carbon paper. The floating FS-NGF was easily placed on the target substrate (Fig. 2c, d). However, BS-NGF samples larger than 3 cm2 were difficult to handle when completely immersed in water. Usually, when they begin to roll in water, due to careless handling they sometimes break into two or three parts (Fig. 2e). Overall, we were able to achieve polymer-free transfer of PS- and BS-NGF (continuous seamless transfer without NGF/Ni/NGF growth at 6 cm2) for samples up to 6 and 3 cm2 in area, respectively. Any remaining large or small pieces can be (easily seen in the etching solution or deionized water) on the desired substrate (~1 mm2, Figure SI4b, see sample transferred to copper grid as in “FS-NGF: Structure and Properties (discussed) under “Structure and Properties”) or store for future use (Figure SI4). Based on this criterion, we estimate that NGF can be recovered in yields of up to 98-99% (after growth for transfer).
Transfer samples without polymer were analyzed in detail. Surface morphological characteristics obtained on FS- and BS-NGF/SiO2/Si (Fig. 2c) using optical microscopy (OM) and SEM images (Fig. SI5 and Fig. 3) showed that these samples were transferred without microscopy. Visible structural damage such as cracks, holes, or unrolled areas. The folds on the growing NGF (Fig. 3b, d, marked by purple arrows) remained intact after transfer. Both FS- and BS-NGFs are composed of FLG regions (bright regions indicated by blue arrows in Figure 3). Surprisingly, in contrast to the few damaged regions typically observed during polymer transfer of ultrathin graphite films, several micron-sized FLG and MLG regions connecting to the NGF (marked by blue arrows in Figure 3d) were transferred without cracks or breaks (Figure 3d). 3). . Mechanical integrity was further confirmed using TEM and SEM images of NGF transferred onto lace-carbon copper grids, as discussed later (“FS-NGF: Structure and Properties”). The transferred BS-NGF/SiO2/Si is rougher than FS-NGF/SiO2/Si with rms values of 140 nm and 17 nm, respectively, as shown in Figure SI6a and b (20 × 20 μm2). The RMS value of NGF transferred onto the SiO2/Si substrate (RMS < 2 nm) is significantly lower (about 3 times) than that of NGF grown on Ni (Figure SI2), indicating that the additional roughness may correspond to the Ni surface . In addition, AFM images performed on the edges of FS- and BS-NGF/SiO2/Si samples showed NGF thicknesses of 100 and 80 nm, respectively (Fig. SI7). The smaller thickness of BS-NGF may be a result of the surface not being directly exposed to the precursor gas.
Transferred NGF (NiAG) without polymer on SiO2/Si wafer (see Figure 2c): (a,b) SEM images of transferred FS-NGF: low and high magnification (corresponding to the orange square in the panel). Typical areas) – a). (c,d) SEM images of transferred BS-NGF: low and high magnification (corresponding to the typical area shown by the orange square in panel c). (e, f) AFM images of transferred FS- and BS-NGFs. Blue arrow represents the FLG region – bright contrast, cyan arrow – black MLG contrast, red arrow – black contrast represents the NGF region, magenta arrow represents the fold.
The chemical composition of the grown and transferred FS- and BS-NGFs was analyzed by X-ray photoelectron spectroscopy (XPS) (Fig. 4). A weak peak was observed in the measured spectra (Fig. 4a, b), corresponding to the Ni substrate (850 eV) of the grown FS- and BS-NGFs (NiAG). There are no peaks in the measured spectra of transferred FS- and BS-NGF/SiO2/Si (Fig. 4c; similar results for BS-NGF/SiO2/Si are not shown), indicating that there is no residual Ni contamination after transfer. Figures 4d–f show the high-resolution spectra of the C 1 s, O 1 s and Si 2p energy levels of FS-NGF/SiO2/Si. The binding energy of C 1 s of graphite is 284.4 eV53.54. The linear shape of graphite peaks is generally considered to be asymmetrical, as shown in Figure 4d54. The high-resolution core-level C 1 s spectrum (Fig. 4d) also confirmed pure transfer (i.e., no polymer residues), which is consistent with previous studies38. The linewidths of the C 1 s spectra of the freshly grown sample (NiAG) and after transfer are 0.55 and 0.62 eV, respectively. These values are higher than those of SLG (0.49 eV for SLG on a SiO2 substrate)38. However, these values are smaller than previously reported linewidths for highly oriented pyrolytic graphene samples (~0.75 eV)53,54,55, indicating the absence of defective carbon sites in the current material. The C 1 s and O 1 s ground level spectra also lack shoulders, eliminating the need for high-resolution peak deconvolution54. There is a π → π* satellite peak around 291.1 eV, which is often observed in graphite samples. The 103 eV and 532.5 eV signals in the Si 2p and O 1 s core level spectra (see Fig. 4e, f) are attributed to the SiO2 56 substrate, respectively. XPS is a surface-sensitive technique, so the signals corresponding to Ni and SiO2 detected before and after NGF transfer, respectively, are assumed to originate from the FLG region. Similar results were observed for transferred BS-NGF samples (not shown).
NiAG XPS results: (ac) Survey spectra of different elemental atomic compositions of grown FS-NGF/Ni, BS-NGF/Ni and transferred FS-NGF/SiO2/Si, respectively. (d–f) High-resolution spectra of the core levels C 1 s, O 1s and Si 2p of the FS-NGF/SiO2/Si sample.
The overall quality of the transferred NGF crystals was assessed using X-ray diffraction (XRD). Typical XRD patterns (Fig. SI8) of transferred FS- and BS-NGF/SiO2/Si show the presence of diffraction peaks (0 0 0 2) and (0 0 0 4) at 26.6° and 54.7°, similar to graphite. . This confirms the high crystalline quality of NGF and corresponds to an interlayer distance of d = 0.335 nm, which is maintained after the transfer step. The intensity of the diffraction peak (0 0 0 2) is approximately 30 times that of the diffraction peak (0 0 0 4), indicating that the NGF crystal plane is well aligned with the sample surface.
According to the results of SEM, Raman spectroscopy, XPS and XRD, the quality of BS-NGF/Ni was found to be the same as that of FS-NGF/Ni, although its rms roughness was slightly higher (Figures SI2, SI5) and SI7).
SLGs with polymer support layers up to 200 nm thick can float on water. This setup is commonly used in polymer-assisted wet chemical transfer processes22,38. Graphene and graphite are hydrophobic (wet angle 80–90°) 57 . The potential energy surfaces of both graphene and FLG have been reported to be quite flat, with low potential energy (~1 kJ/mol) for the lateral movement of water at the surface58. However, the calculated interaction energies of water with graphene and three layers of graphene are approximately − 13 and − 15 kJ/mol,58 respectively, indicating that the interaction of water with NGF (about 300 layers) is lower compared to graphene. This may be one of the reasons why freestanding NGF remains flat on the surface of water, while freestanding graphene (which floats in water) curls up and breaks down. When NGF is completely immersed in water (results are the same for rough and flat NGF), its edges bend (Figure SI4). In the case of complete immersion, it is expected that the NGF-water interaction energy is almost doubled (compared to floating NGF) and that the edges of the NGF fold to maintain a high contact angle (hydrophobicity). We believe that strategies can be developed to avoid curling of the edges of embedded NGFs. One approach is to use mixed solvents to modulate the wetting reaction of the graphite film59.
The transfer of SLG to various types of substrates via wet chemical transfer processes has been previously reported. It is generally accepted that weak van der Waals forces exist between graphene/graphite films and substrates (be it rigid substrates such as SiO2/Si38,41,46,60, SiC38, Au42, Si pillars22 and lacy carbon films30, 34 or flexible substrates such as polyimide 37). Here we assume that interactions of the same type predominate. We did not observe any damage or peeling of NGF for any of the substrates presented here during mechanical handling (during characterization under vacuum and/or atmospheric conditions or during storage) (e.g., Figure 2, SI7 and SI9). In addition, we did not observe a SiC peak in the XPS C 1 s spectrum of the core level of the NGF/SiO2/Si sample (Fig. 4). These results indicate that there is no chemical bond between NGF and the target substrate.
In the previous section, “Polymer-free transfer of FS- and BS-NGF,” we demonstrated that NGF can grow and transfer on both sides of nickel foil. These FS-NGFs and BS-NGFs are not identical in terms of surface roughness, which prompted us to explore the most suitable applications for each type.
Considering the transparency and smoother surface of FS-NGF, we studied its local structure, optical and electrical properties in more detail. The structure and structure of FS-NGF without polymer transfer were characterized by transmission electron microscopy (TEM) imaging and selected area electron diffraction (SAED) pattern analysis. The corresponding results are shown in Figure 5. Low magnification planar TEM imaging revealed the presence of NGF and FLG regions with different electron contrast characteristics, i.e. darker and brighter areas, respectively (Fig. 5a). The film overall exhibits good mechanical integrity and stability between the different regions of NGF and FLG, with good overlap and no damage or tearing, which was also confirmed by SEM (Figure 3) and high magnification TEM studies (Figure 5c-e). In particular, in Fig. Figure 5d shows the bridge structure at its largest part (the position marked by the black dotted arrow in Figure 5d), which is characterized by a triangular shape and consists of a graphene layer with a width of about 51 . The composition with an interplanar spacing of 0.33 ± 0.01 nm is further reduced to several layers of graphene in the narrowest region (end of the solid black arrow in Figure 5 d).
Planar TEM image of a polymer-free NiAG sample on a carbon lacy copper grid: (a, b) Low magnification TEM images including NGF and FLG regions, (ce) High magnification images of various regions in panel-a and panel-b are marked arrows of the same color. Green arrows in panels a and c indicate circular areas of damage during beam alignment. (f–i) In panels a to c, SAED patterns in different regions are indicated by blue, cyan, orange, and red circles, respectively.
The ribbon structure in Figure 5c shows (marked with red arrow) the vertical orientation of the graphite lattice planes, which may be due to the formation of nanofolds along the film (inset in Figure 5c) due to excess uncompensated shear stress30,61,62. Under high-resolution TEM, these nanofolds 30 exhibit a different crystallographic orientation than the rest of the NGF region; the basal planes of the graphite lattice are oriented almost vertically, rather than horizontally like the rest of the film (inset in Figure 5c). Similarly, the FLG region occasionally exhibits linear and narrow band-like folds (marked by blue arrows), which appear at low and medium magnification in Figures 5b, 5e, respectively. The inset in Figure 5e confirms the presence of two- and three-layer graphene layers in the FLG sector (interplanar distance 0.33 ± 0.01 nm), which is in good agreement with our previous results30. Additionally, recorded SEM images of polymer-free NGF transferred onto copper grids with lacy carbon films (after performing top-view TEM measurements) are shown in Figure SI9. The well suspended FLG region (marked with blue arrow) and the broken region in Figure SI9f. The blue arrow (at the edge of the transferred NGF) is intentionally presented to demonstrate that the FLG region can resist the transfer process without polymer. In summary, these images confirm that partially suspended NGF (including the FLG region) maintains mechanical integrity even after rigorous handling and exposure to high vacuum during TEM and SEM measurements (Figure SI9).
Due to the excellent flatness of NGF (see Figure 5a), it is not difficult to orient the flakes along the [0001] domain axis to analyze the SAED structure. Depending on the local thickness of the film and its location, several regions of interest (12 points) were identified for electron diffraction studies. In Figures 5a–c, four of these typical regions are shown and marked with colored circles (blue, cyan, orange, and red coded). Figures 2 and 3 for SAED mode. Figures 5f and g were obtained from the FLG region shown in Figures 5 and 5. As shown in Figures 5b and c, respectively. They have a hexagonal structure similar to twisted graphene63. In particular, Figure 5f shows three superimposed patterns with the same orientation of the [0001] zone axis, rotated by 10° and 20°, as evidenced by the angular mismatch of the three pairs of (10-10) reflections. Similarly, Figure 5g shows two superimposed hexagonal patterns rotated by 20°. Two or three groups of hexagonal patterns in the FLG region can arise from three in-plane or out-of-plane graphene layers 33 rotated relative to each other. In contrast, the electron diffraction patterns in Figure 5h,i (corresponding to the NGF region shown in Figure 5a) show a single [0001] pattern with an overall higher point diffraction intensity, corresponding to greater material thickness. These SAED models correspond to a thicker graphitic structure and intermediate orientation than FLG, as inferred from the index 64. Characterization of the crystalline properties of NGF revealed the coexistence of two or three superimposed graphite (or graphene) crystallites. What is particularly noteworthy in the FLG region is that the crystallites have a certain degree of in-plane or out-of-plane misorientation. Graphite particles/layers with in-plane rotation angles of 17°, 22° and 25° have previously been reported for NGF grown on Ni 64 films. The rotation angle values observed in this study are consistent with previously observed rotation angles (±1°) for twisted BLG63 graphene.
The electrical properties of NGF/SiO2/Si were measured at 300 K over an area of 10×3 mm2. The values of electron carrier concentration, mobility and conductivity are 1.6 × 1020 cm-3, 220 cm2 V-1 C-1 and 2000 S-cm-1, respectively. The mobility and conductivity values of our NGF are similar to natural graphite2 and higher than commercially available highly oriented pyrolytic graphite (produced at 3000 °C)29. The observed electron carrier concentration values are two orders of magnitude higher than those recently reported (7.25 × 10 cm-3) for micron-thick graphite films prepared using high-temperature (3200 °C) polyimide sheets 20 .
We also performed UV-visible transmittance measurements on FS-NGF transferred to quartz substrates (Figure 6). The resulting spectrum shows a nearly constant transmittance of 62% in the range 350–800 nm, indicating that NGF is translucent to visible light. In fact, the name “KAUST” can be seen in the digital photograph of the sample in Figure 6b. Although the nanocrystalline structure of NGF is different from that of SLG, the number of layers can be roughly estimated using the rule of 2.3% transmission loss per additional layer65. According to this relationship, the number of graphene layers with 38% transmission loss is 21. The grown NGF mainly consists of 300 graphene layers, i.e. about 100 nm thick (Fig. 1, SI5 and SI7). Therefore, we assume that the observed optical transparency corresponds to the FLG and MLG regions, since they are distributed throughout the film (Figs. 1, 3, 5 and 6c). In addition to the above structural data, conductivity and transparency also confirm the high crystalline quality of the transferred NGF.
(a) UV-visible transmittance measurement, (b) typical NGF transfer on quartz using a representative sample. (c) Schematic of NGF (dark box) with evenly distributed FLG and MLG regions marked as gray random shapes throughout the sample (see Figure 1) (approx. 0.1–3% area per 100 μm2). The random shapes and their sizes in the diagram are for illustrative purposes only and do not correspond to actual areas.
Translucent NGF grown by CVD has previously been transferred to bare silicon surfaces and used in solar cells15,16. The resulting power conversion efficiency (PCE) is 1.5%. These NGFs perform multiple functions such as active compound layers, charge transport pathways, and transparent electrodes15,16. However, the graphite film is not uniform. Further optimization is necessary by carefully controlling the sheet resistance and optical transmittance of the graphite electrode, since these two properties play an important role in determining the PCE value of the solar cell15,16. Typically, graphene films are 97.7% transparent to visible light, but have a sheet resistance of 200–3000 ohms/sq.16. The surface resistance of graphene films can be reduced by increasing the number of layers (multiple transfer of graphene layers) and doping with HNO3 (~30 Ohm/sq.)66. However, this process takes a long time and the different transfer layers do not always maintain good contact. Our front side NGF has properties such as conductivity 2000 S/cm, film sheet resistance 50 ohm/sq. and 62% transparency, making it a viable alternative for conductive channels or counter electrodes in solar cells15,16.
Although the structure and surface chemistry of BS-NGF are similar to FS-NGF, its roughness is different (“Growth of FS- and BS-NGF”). Previously, we used ultra-thin film graphite22 as a gas sensor. Therefore, we tested the feasibility of using BS-NGF for gas sensing tasks (Figure SI10). First, mm2-sized portions of BS-NGF were transferred onto the interdigitating electrode sensor chip (Figure SI10a-c). Manufacturing details of the chip were previously reported; its active sensitive area is 9 mm267. In the SEM images (Figure SI10b and c), the underlying gold electrode is clearly visible through the NGF. Again, it can be seen that uniform chip coverage was achieved for all samples. Gas sensor measurements of various gases were recorded (Fig. SI10d) (Fig. SI11) and the resulting response rates are shown in Figs. SI10g. Likely with other interfering gases including SO2 (200 ppm), H2 (2%), CH4 (200 ppm), CO2 (2%), H2S (200 ppm) and NH3 (200 ppm ). One possible cause is NO2. electrophilic nature of the gas22,68. When adsorbed on the surface of graphene, it reduces the current absorption of electrons by the system. A comparison of the response time data of the BS-NGF sensor with previously published sensors is presented in Table SI2. The mechanism for reactivating NGF sensors using UV plasma, O3 plasma or thermal (50–150°C) treatment of exposed samples is ongoing, ideally followed by the implementation of embedded systems69.
During the CVD process, graphene growth occurs on both sides of the catalyst substrate41. However, BS-graphene is usually ejected during the transfer process41. In this study, we demonstrate that high-quality NGF growth and polymer-free NGF transfer can be achieved on both sides of the catalyst support. BS-NGF is thinner (~80 nm) than FS-NGF (~100 nm), and this difference is explained by the fact that BS-Ni is not directly exposed to the precursor gas flow. We also found that the roughness of the NiAR substrate influences the roughness of the NGF. These results indicate that the grown planar FS-NGF can be used as a precursor material for graphene (by exfoliation method70) or as a conductive channel in solar cells15,16. In contrast, BS-NGF will be used for gas detection (Fig. SI9) and possibly for energy storage systems71,72 where its surface roughness will be useful.
Considering the above, it is useful to combine the current work with previously published graphite films grown by CVD and using nickel foil. As can be seen in Table 2, the higher pressures we used shortened the reaction time (growth stage) even at relatively low temperatures (in the range of 850–1300 °C). We also achieved greater growth than usual, indicating potential for expansion. There are other factors to consider, some of which we have included in the table.
Double-sided high-quality NGF was grown on nickel foil by catalytic CVD. By eliminating traditional polymer substrates (such as those used in CVD graphene), we achieve clean and defect-free wet transfer of NGF (grown on the back and front sides of nickel foil) to a variety of process-critical substrates. Notably, NGF includes FLG and MLG regions (typically 0.1% to 3% per 100 µm2) that are structurally well integrated into the thicker film. Planar TEM shows that these regions are composed of stacks of two to three graphite/graphene particles (crystals or layers, respectively), some of which have a rotational mismatch of 10–20°. The FLG and MLG regions are responsible for the transparency of FS-NGF to visible light. As for the rear sheets, they can be carried parallel to the front sheets and, as shown, can have a functional purpose (for example, for gas detection). These studies are very useful for reducing waste and costs in industrial scale CVD processes.
In general, the average thickness of CVD NGF lies between (low- and multi-layer) graphene and industrial (micrometer) graphite sheets. The range of their interesting properties, combined with the simple method we have developed for their production and transport, makes these films particularly suitable for applications requiring the functional response of graphite, without the expense of the energy-intensive industrial production processes currently used.
A 25-μm-thick nickel foil (99.5% purity, Goodfellow) was installed in a commercial CVD reactor (Aixtron 4-inch BMPro). The system was purged with argon and evacuated to a base pressure of 10-3 mbar. Then nickel foil was placed. in Ar/H2 (After pre-annealing the Ni foil for 5 min, the foil was exposed to a pressure of 500 mbar at 900 °C. NGF was deposited in a flow of CH4/H2 (100 cm3 each) for 5 min. The sample was then cooled to temperature below 700 °C using Ar flow (4000 cm3) at 40 °C/min. Details on optimization of the NGF growth process are described elsewhere30.
The surface morphology of the sample was visualized by SEM using a Zeiss Merlin microscope (1 kV, 50 pA). The sample surface roughness and NGF thickness were measured using AFM (Dimension Icon SPM, Bruker). TEM and SAED measurements were carried out using an FEI Titan 80–300 Cubed microscope equipped with a high brightness field emission gun (300 kV), an FEI Wien type monochromator and a CEOS lens spherical aberration corrector to obtain the final results. spatial resolution 0.09 nm. NGF samples were transferred to carbon lacy coated copper grids for flat TEM imaging and SAED structure analysis. Thus, most of the sample flocs are suspended in the pores of the supporting membrane. Transferred NGF samples were analyzed by XRD. X-ray diffraction patterns were obtained using a powder diffractometer (Brucker, D2 phase shifter with Cu Kα source, 1.5418 Å and LYNXEYE detector) using a Cu radiation source with a beam spot diameter of 3 mm.
Several Raman point measurements were recorded using an integrating confocal microscope (Alpha 300 RA, WITeC). A 532 nm laser with low excitation power (25%) was used to avoid thermally induced effects. X-ray photoelectron spectroscopy (XPS) was performed on a Kratos Axis Ultra spectrometer over a sample area of 300 × 700 μm2 using monochromatic Al Kα radiation (hν = 1486.6 eV) at a power of 150 W. Resolution spectra were obtained at transmission energies of 160 eV and 20 eV, respectively. NGF samples transferred onto SiO2 were cut into pieces (3 × 10 mm2 each) using a PLS6MW (1.06 μm) ytterbium fiber laser at 30 W. Copper wire contacts (50 μm thick) were fabricated using silver paste under an optical microscope. Electrical transport and Hall effect experiments were carried out on these samples at 300 K and a magnetic field variation of ± 9 Tesla in a physical properties measurement system (PPMS EverCool-II, Quantum Design, USA). Transmitted UV–vis spectra were recorded using a Lambda 950 UV–vis spectrophotometer in the 350–800 nm NGF range transferred to quartz substrates and quartz reference samples.
The chemical resistance sensor (interdigitated electrode chip) was wired to a custom printed circuit board 73 and the resistance was extracted transiently. The printed circuit board on which the device is located is connected to the contact terminals and placed inside the gas sensing chamber 74. Resistance measurements were taken at a voltage of 1 V with a continuous scan from purge to gas exposure and then purge again. The chamber was initially cleaned by purging with nitrogen at 200 cm3 for 1 hour to ensure removal of all other analytes present in the chamber, including moisture. The individual analytes were then slowly released into the chamber at the same flow rate of 200 cm3 by closing the N2 cylinder.
A revised version of this article has been published and can be accessed through the link at the top of the article.
Inagaki, M. and Kang, F. Carbon Materials Science and Engineering: Fundamentals. Second edition edited. 2014. 542.
Pearson, H.O. Handbook of Carbon, Graphite, Diamond and Fullerenes: Properties, Processing and Applications. The first edition has been edited. 1994, New Jersey.
Tsai, W. et al. Large area multilayer graphene/graphite films as transparent thin conductive electrodes. application. physics. Wright. 95(12), 123115(2009).
Balandin A.A. Thermal properties of graphene and nanostructured carbon materials. Nat. Matt. 10(8), 569–581 (2011).
Cheng K.Y., Brown P.W. and Cahill D.G. Thermal conductivity of graphite films grown on Ni (111) by low-temperature chemical vapor deposition. adverb. Matt. Interface 3, 16 (2016).
Hesjedal, T. Continuous growth of graphene films by chemical vapor deposition. application. physics. Wright. 98(13), 133106(2011).
Post time: Aug-23-2024