Date Published: May 30, 2017
Publisher: John Wiley and Sons Inc.
Author(s): Zhihong Zhang, Xiaozhi Xu, Lu Qiu, Shaoxin Wang, Tianwei Wu, Feng Ding, Hailin Peng, Kaihui Liu.
The exceptional properties of graphene make it a promising candidate in the development of next‐generation electronic, optoelectronic, photonic and photovoltaic devices. A holy grail in graphene research is the synthesis of large‐sized single‐crystal graphene, in which the absence of grain boundaries guarantees its excellent intrinsic properties and high performance in the devices. Nowadays, most attention has been drawn to the suppression of nucleation density by using low feeding gas during the growth process to allow only one nucleus to grow with enough space. However, because the nucleation is a random event and new nuclei are likely to form in the very long growth process, it is difficult to achieve industrial‐level wafer‐scale or beyond (e.g. 30 cm in diameter) single‐crystal graphene. Another possible way to obtain large single‐crystal graphene is to realize ultrafast growth, where once a nucleus forms, it grows up so quickly before new nuclei form. Therefore ultrafast growth provides a new direction for the synthesis of large single‐crystal graphene, and is also of great significance to realize large‐scale production of graphene films (fast growth is more time‐efficient and cost‐effective), which is likely to accelerate various graphene applications in industry.
Graphene, a two‐dimensional material with carbon atoms bonded in a honeycomb lattice, has attracted immerse interests in the past decade due to its exceptional properties and various applications.1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 The growth of large‐area high‐quality graphene films is fundamental for the upcoming graphene applications. Chemical vapour deposition (CVD) method offers good prospects to produce large‐size graphene films due to its simplicity, controllability and cost‐efficiency.13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75 Many researches have verified that graphene can be catalytically grown on metallic substrates, like ruthenium (Ru),13, 14 iridium (Ir),15, 16 platinum (Pt),17, 18, 19, 20 nickel (Ni)21, 22, 23 and copper (Cu)30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73 et al., and also can be directly grown on insulating substrates, such as SiC,24, 25 SiO2,26, 27 sapphire,28 and h‐BN.29, 30 Considering that direct graphene growth on desired insulating substrate without transfer process is promising for integration into silicon‐based technologies, it would be a favourable way for industrial application of graphene. However, despite a lot of efforts, so far the synthesized graphene on insulating substrates is typically of poor quality, small domain size and low growth rate, as a result of that the direct growth relies on the thermal decomposition of the carbon resource. Up to now, many challenges remain in this direction and far‐away from the large‐area high‐quality graphene production. On the other hand, graphene grown on metallic substrates possesses higher quality and can reach large size easily, thus is a way close at hand for reliable graphene production to realize various applications. The carbon solubility in metals determine the graphene growth behaviour (see Figure1).76 Ru, Ir, Pt and Ni have high carbon solubility and their usage leads to a segregation growth process, in which carbon first dissolves into the bulk metals at a high temperature and segregates to form graphene after reaching the carbon saturation, and the graphene grows up with carpet growth mode. Ru and Ir are noble metals and economically unfavourable. Previous studies on graphene grown on Ru and Ir mainly focused on the surface science rather than graphene synthesis. Although, Pt is also expensive, the development of the bubbling transfer method makes the repeated growth on Pt available and greatly reduces cost.18 The high catalytic activity and high melting temperature of Pt enables the fast growth of graphene at high temperatures. Besides, the graphene growth on Pt can also be realized by a surface‐mediated process and millimetre‐scale single‐crystal graphene domains can be obtained.19 A smart surface engineering of polycrystalline Pt by coating with silicon‐containing film makes the graphene growth free from the effects of Pt substrate which improves the crystallinity, uniformity and the domain size of graphene.20 With this eutectic substrate the growth rate was increased to 120 µm min−1 at 1150 °C, more than an order of magnitude higher than typical reported ones. Pt is one of the promising substrates, while at the moment it still cannot meet the demands for the large‐scale synthesis of graphene. Ni and Cu are much cheaper and widely available substrates that make the industrialized production of graphene feasible. High solubility of carbon in Ni leads to the formation of multilayer graphene.21, 22 Achieving a uniform monolayer graphene on Ni is still a great challenge. Especially, Cu with low carbon solubility, serving as catalytic substrate, makes the continuous uniform high‐quality monolayer graphene films accessible.31, 32, 33, 34 Nowadays, the large‐scale production and transfer of graphene film from Cu foils has been realized by the roll‐to‐roll technique.69, 70 Thus, the CVD growth on Cu foils is believed to be the most promising method to realize the industrial production of single‐crystal graphene films.
After about a decade’s efforts, the domain size of single‐crystal graphene has increased more than four orders of magnitude, from micrometres to inches (Figure2).31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54 On the other hand, before 2016, the growth rate has not improved as much, varying in the range of several µm min−1. That means to grow 12‐inch single‐crystal graphene films typically needs almost one day. This time‐ and energy‐consuming process is not suitable for the large‐scale production. Only from 2016, several breakthroughs made the growth rate exceed 100 µm min−1,45, 46 even astonishingly up to 3600 µm min−1.36 Ultrafast growth can greatly shorten the process and its significance is far beyond this time or energy efficiency. It is known that the crystal nucleation is a random event, and thus new nuclei would inevitably form during the long growth process, as schematically illustrated in Figure3a. That’s also one reason why the single‐crystal domain size is still limited to inch size so far, even with the most advanced growth methods. If the growth rate can be extremely high, the nucleus would grow up so rapidly that new nuclei do not form (Figure 3b). Therefore, the ultrafast growth, if possible, is ideal to synthesize large single‐crystal graphene domains. If the domains are unaligned, the fast growth enables the polycrystalline graphene films consisting of large domains with less grain boundaries and therefore higher quality. If the domains are aligned and seamless stitched together, such as in the epitaxial growth on Ge(100) or Cu(111) substrate, large single‐crystal graphene film can be produced in a short time.58, 59 All in all, improving the growth rate is a new and effective route, although its importance has not been paid too much attention earlier, to grow high‐quality or super‐large single‐crystal graphene films.
Cu has very low carbon solubility and as a result, the graphene growth on Cu surface follows the surface‐mediated growth mode.79, 80, 81 The good thermal stability of CH4 makes it a preferred carbon source to prevent pyrolysis at high temperatures, which is favourable to surface‐mediate growth of high‐quality graphene.33 Generally speaking, the graphene CVD growth on Cu involves four processes:82 (i) absorption of CH4 onto Cu surfaces; (ii) catalytic dissociation of CH4 by Cu, resulting in C species CHx (x = 0–3); (iii) surface diffusion of C species; (iv) C attachment onto graphene domain edges. A large number of experiments have shown that these four processes in graphene growth are dominated by the catalytic substrate, feeding gas, reaction barriers, temperature and pressure (Figure4). So to make proper control over these conditions is essential to realize ultrafast growth of graphene.
To improve the growth rate of CVD graphene is essential for both fundamental research and industrial applications. In this review, the available ways towards the ultrafast graphene growth are discussed in details. Proper catalytic substrate can achieve the higher growth rate, while it is more powerful in the control of layer number. The optimization of feeding gas ratio has limited capacity to improve the grow rate. The deduction of reaction barrier, for example by the introduction of oxygen, can improve the growth rate exponentially and turns out to be a very powerful way to realize ultrafast growth.
The authors declare no conflict of interest.