Graphene has a range of unique physical properties and could be of use in the development of a variety of electronic, photonic and photovoltaic devices. For most applications, large-area high-quality graphene films are required and chemical vapour deposition (CVD) synthesis of graphene on copper surfaces has been of particular interest due to its simplicity and cost effectiveness. However, the rates of growth for graphene by CVD on copper are less than 0.4 μm s, and therefore the synthesis of large, single-crystal graphene domains takes at least a few hours. Here, we show that single-crystal graphene can be grown on copper foils with a growth rate of 60 μm s. Our high growth rate is achieved by placing the copper foil above an oxide substrate with a gap of ∼15 μm between them. The oxide substrate provides a continuous supply of oxygen to the surface of the copper catalyst during the CVD growth, which significantly lowers the energy barrier to the decomposition of the carbon feedstock and increases the growth rate. With this approach, we are able to grow single-crystal graphene domains with a lateral size of 0.3 mm in just 5 s.

Vertical heterostructures based on two-dimensional layered materials, such as stacked graphene and hexagonal boron nitride (G/h-BN), have stimulated wide interest in fundamental physics, material sciences and nanoelectronics. To date, it still remains challenging to obtain high quality G/h-BN heterostructures concurrently with controlled nucleation density and thickness uniformity. In this work, with the aid of the well-defined poly(methyl methacrylate) seeds, effective control over the nucleation densities and locations of graphene domains on the predeposited h-BN monolayers was realized, leading to the formation of patterned G/h-BN arrays or continuous films. Detailed spectroscopic and morphological characterizations further confirmed that ∼85.7% of such monolayer graphene domains were of single-crystalline nature with their domain sizes predetermined throughout seed interspacing. Density functional theory calculations suggested that a self-terminated growth mechanism can be applied for the related graphene growth on h-BN/Cu. In turn, as-constructed field-effect transistor arrays based on such synthesized single-crystalline G/h-BN patterning were found to be compatible with fabricating devices with nice and steady performance, hence holding great promise for the development of next-generation graphene-based electronics.

To fully understand the kinetics of graphene growth, large-scale atomic simulations of graphene islands evolution up to macro sizes (i.e., graphene islands of a few micrometers or with billions of carbon atoms) during growth and etching is essential, but remains a great challenge. In this paper, we developed a low computational cost large-scale kinetic Monte Carlo (KMC) algorithm, which includes all possible events of carbon attachments and detachments on various edge sites of graphene islands. Such a method allows us to simulate the evolution of graphene islands with sizes up to tens of micrometers during either growth or etching with a single CPU core. With this approach and the carefully fitted parameters, we have reproduced the experimentally observed evolution of graphene islands during both growth or etching on Pt(111) surface, and revealed more atomic details of graphene growth and etching. Based on the atomic simulations, we discovered a complementary relationship of graphene growth and etching—the route of graphene island shape evolution during growth is exactly the same as that of the etching of a hole in graphene and that of graphene island etching is exactly same as that of hole growth. The complementary relation brings us a basic principle to understand the growth and etching of graphene, and other 2D materials from atomic scale to macro size and the KMC algorithm is expected to be further developed into a standard simulation package for investigating the growth mechanism of 2D materials on various substrates.

In chemical vapor depositiongrowth, depending on the growth condition, graphene islands present various shapes, such as circular, hexagonal, square, rectangular star-like, and fractal. We present a systematic phase-field model study to explore the role of three key factors, carbon flux, precursor concentration on the metal surface, and diffusion of carbon precursors, in the determination of the graphene domain shape. We find that the size of the depletion zone, that is, the area with a carbon precursor concentration gradient around the circumference of the growing graphene island, is of critical importance. In case of no depletion zone or if the size of the depletion zone is much larger than that of the graphene island, the graphene island will present the shape of a regular hexagon. However, star-like graphene islands will form if the size of the depletion zone is comparable to that of the graphene island. Further increasing the size of the graphene island will lead to a fractal-like graphene island with multiscaled branches. Although an extremely small depletion zone will lead to a graphene island with a regular shape, its edges are found to be rough. The three key parameters affect the shape of graphene by tuning the size of the depletion zone. Based on this study, a series of experimental puzzles are properly explained and the conditions for growing high-quality graphene emerge.

High-quality single-layered and bilayered graphene (SLG and BLG) was synthesized on copper foil surfaces by controllable chemical vapor deposition (CVD). Impurity nanoparticles formed on the copper foil surface by hightemperature annealing were found to play a crucial role in the growth of BLG. Analysis of energy-dispersive spectrometry (EDS) data indicated that these nanoparticles consisted of silicon and aluminum. According to the inverted wedding cake model, these nanoparticles served as nucleation centers for BLG growth and the free space between a nanoparticle and graphene served as the center of C injection for the continuous growth of the adlayer beneath the top layer. By combining phase-field theory simulations, we confirmed the mechanism of BLG growth and revealed more details about it in comparison with SLG growth. For the first time, this study led to a complete understanding of the BLG growth mechanism from nucleation to continuous growth in the CVD process, and it has opened a door to the thickness-controllable synthesis of graphene.