Graphene from Nanomembranes
May 29, 2013 – Researchers from Bielefeld, Ulm and the PTB in Braunschweig turned monolayers of aromatic molecules, linked by electrons into carbon nanomembranes, into graphene. AC-HRTEM at 80 kV in Ulm shows atomically perfect single-layer graphene on copper and nanocrystalline graphene on gold whose grains are stitched together by pentagons and heptagons.
Graphene can be made from molecules: researchers from Bielefeld University, Ulm University and the Physikalisch-Technische Bundesanstalt (PTB) in Braunschweig let a monolayer of biphenylthiol (BPT) molecules assemble on copper, linked it with low-energy electrons into a carbon nanomembrane (CNM) one nanometre thick and heated it into graphene (Fig. 1). Aberration-corrected high-resolution transmission electron microscopy (AC-HRTEM) at 80 kV in Ulm showed that the result is atomically perfect single-layer graphene, with the same lattice as graphene from graphite (Fig. 2). Six weeks later, a second study with the Max Planck Institute for Polymer Research in Mainz and the University of Frankfurt applied the scheme to many aromatic molecules on gold; there the HRTEM images show nanocrystalline graphene whose grains are stitched together by pentagons and heptagons (Fig. 3). Two boxes compare the Ulm results with graphene from graphite imaged by a FEI Titan at 80 kV in Oxford and with CVD graphene imaged by a C3/C5-corrected Nion UltraSTEM at 60 kV at Cornell University.
From molecules to graphene
The route has three steps. Aromatic thiol molecules arrange themselves upright into a self-assembled monolayer (SAM) on a metal surface. Electrons of only 50 or 100 eV, about 3000 to 3500 per nm2, break C–H bonds and link neighbouring molecules into a mechanically stable network, the CNM, which the Bielefeld group had first lifted off as a free-standing film in 2005[3]. Heated in ultra-high vacuum, the membrane turns into graphene, and its thickness and size are set by the molecules and by the irradiated area. For the microscope, the graphene is covered with a polymer, released from the metal, placed on a TEM grid and freed from the polymer again.
A single layer on copper
In the first study[1], BPT monolayers on copper foil were linked and annealed at up to 830 °C. The AC-HRTEM image at 80 kV clearly resolves the honeycomb lattice of the suspended sheet, with the carbon atoms dark under the chosen imaging conditions (Fig. 2a in the box below). Already the image contrast shows a single layer, and the diffraction pattern from a much wider area confirms it: the intensity ratio of its first- and second-order reflections is that of single-layer graphene. “In TEM the atomic potential is probed showing both in imaging (…) and in diffraction (…) that atomically perfect single-layer graphene has been formed,” the authors write. A series of dark-field TEM images, coloured by lattice orientation, maps the crystallites: up to about 1.2 µm and about 300 nm on average.
The images can be compared with an earlier HRTEM study of graphene from graphite with the same type of microscope at the same voltage, in the box below (Fig. 2).
Many molecules on gold
The second study extended the scheme from BPT to three types of aromatic thiols on gold: molecules with a straight backbone of benzene rings, condensed molecules such as anthracene and pyrene thiols, and bulky, disc- or propeller-shaped molecules. All of them formed stable CNMs, and after heating to 900 °C, the Ulm team imaged the resulting graphene by AC-HRTEM at 80 kV (Fig. 3). Without the catalytic copper, the membranes did not become a closed single layer but nanocrystalline graphene.
Graphene seen atom by atom
Fig. 3a shows one of the thinnest membranes, 800 pm thick before annealing, made from hexaphenylbenzene (HPB). About half of the sheet is single-layer graphene with its clearly visible hexagonal arrangement of carbon atoms. Randomly oriented graphene nanocrystallites are joined by the typical grain boundaries of pentagons and heptagons (inset of Fig. 3a), the defects Ulm had followed atom by atom in graphene a year earlier[4]. Such defects are not mere flaws: grain boundaries and their non-hexagonal rings change the electronic, chemical and mechanical properties of graphene[5]. About 20 % of the sheet consists of double layers, recognizable by their moiré pattern[6]; the rest is disordered carbon and holes.
Why pentagons and heptagons come in pairs was explained in a Perspective in ACS Nano later that year by Jiong Lu, Yang Bao, Chen Liang Su and Kian Ping Loh from the National University of Singapore[7]. A single pentagon or heptagon would bend the flat sheet; paired, their curvatures cancel, and the pair forms the core of a dislocation. Such pairs are the building blocks of grain boundaries in polycrystalline graphene, and HRTEM has revealed them in carbon nanotubes and graphene; clusters such as the Stone–Wales defect[8], the 555-777 defect and the flower defect can even be created in situ by the electron beam of the microscope, as in the Ulm studies the authors cite[4]. Real grain boundaries are often aperiodic, like those in Fig. 3a, although the state of lowest energy should be a periodic array of dislocation cores whose density depends on the angle between the grains; at 32.2° the boundary consists of pentagon–heptagon pairs only. Clusters of such defects may even open a local band gap of about 200 meV.
Graphene from the other precursors shows the same nanocrystalline structure (Fig. 3b–g), but it is up to about three times thicker, depending on the molecule. Side by side, BPT gives the largest crystalline areas (Fig. 3c), HBC-Br the fewest, with much amorphous carbon (Fig. 3g); a measurement of the surely crystalline area confirms this order, at least 19 % for BPT and about 1 % for HBC-Br (Table 2). The team checked this visual impression by measuring the film thickness with quantitative electron energy-loss spectroscopy (EELS) at the carbon K edge[9], as the box below shows.
Validation: FEI Titan at 80 kV vs. Nion UltraSTEM at 60 kV
Grain boundaries made of pentagons and heptagons had been imaged atom by atom in 2011 by a team from Cornell University in graphene grown by chemical vapour deposition (CVD) on copper, with a C3/C5-corrected Nion UltraSTEM at 60 kV. The box below compares the two microscopes and shows both grain boundaries side by side (Fig. 4).
Challenges named by the authors
The authors name the limits of the scheme. On gold, the graphene stays nanocrystalline, with double layers, disordered carbon and holes beside the single-layer areas; a closed single layer with grains of hundreds of nanometres formed only on copper. Not every aromatic thiol works: short molecules such as pyridine-4-thiol and anthracene-9-thiol form no membranes, and molecules that pack poorly leave pores of about 1–10 nm. The monolayers must be linked before heating, because unlinked aromatic thiol monolayers leave the gold at about 120 °C, long before they could turn into graphene. And HRTEM sees only a few tens of nanometres at a time, so that the thickness over larger areas needs EELS.
In a Perspective in the same issue of ACS Nano, Richard Palmer, Alex Robinson and Quanmin Guo from the University of Birmingham called the scheme an instructive case of how nanoscience turns into technology[12]. Its key new point, they wrote, is that the arrangement of the molecules in the monolayer is passed on to the morphology of the membranes and sheets, so that, for example, regular arrays of pores can be made, and the membranes are only 0.5 to 3 nm thick. The authors consider the method scalable to square metres with a defocused electron source; meanwhile, the Bielefeld group had set up a small company offering novel grids for transmission electron microscopy.
What the molecules turn into is decided atom by atom in the microscope: “The single layer nature of this film can be determined already from the HRTEM image contrast,” the authors of the first study write.
Resource:
Angelova, P., Vieker, H., Weber, N.-E., Matei, D., Reimer, O., Meier, I., Kurasch, S., Biskupek, J., Lorbach, D., Wunderlich, K., Chen, L., Terfort, A., Klapper, M., Müllen, K., Kaiser, U., Gölzhäuser, A., Turchanin, A. (2013).
A Universal Scheme to Convert Aromatic Molecular Monolayers into Functional Carbon Nanomembranes.
ACS Nano 7(8), 6489–6497.
https://doi.org/10.1021/nn402652f
Update – January 9, 2017
The news article of May 29, 2013 ends above. Later work is listed below.
Grain boundaries at a glance
Grain boundaries decide much of the quality of CVD graphene, and TEM is the usual tool to probe their structure, as researchers from Seoul, Sydney and New York wrote in 2017[13]. The atoms that stitch the grains together, pentagons and heptagons as in the inset of Fig. 3a and in Fig. 4, can only be seen in the electron microscope.
Download this news article as PDF – for reading on paper, taking notes or filing.
-
Matei, D. G., Weber, N., Kurasch, S., et al. (2013). Functional Single-Layer Graphene Sheets From Aromatic Monolayers. Adv. Mater., 25, 4146–4151 (first published 29 May 2013). https://doi.org/10.1002/adma.201300651
-
Warner, J. H., Rümmeli, M. H., Ge, L., Gemming, T., Montanari, B., Harrison, N. M., Büchner, B., Briggs, G. A. D. (2009). Structural Transformations in Graphene Studied With High Spatial and Temporal Resolution. Nat. Nanotechnol., 4, 500–504. https://doi.org/10.1038/nnano.2009.194
-
Eck, W., Küller, A., Grunze, M., Völkel, B., Gölzhäuser, A. (2005). Freestanding Nanosheets From Crosslinked Biphenyl Self-Assembled Monolayers. Adv. Mater., 17, 2583–2587. https://doi.org/10.1002/adma.200500900
-
SALVE news (2012). Grain Boundary Migration in Graphene. salve-project.de, May 10, 2012. https://salve-project.de/newspress/grain-boundary-migration-in-graphene.html
-
Terrones, H., Lv, R., Terrones, M., Dresselhaus, M. S. (2012). The Role of Defects and Doping in 2D Graphene Sheets and 1D Nanoribbons. Rep. Prog. Phys., 75, 062501. https://doi.org/10.1088/0034-4885/75/6/062501
-
Warner, J. H., Rümmeli, M. H., Gemming, T., Büchner, B., Briggs, G. A. D. (2009). Direct Imaging of Rotational Stacking Faults in Few Layer Graphene. Nano Lett., 9, 102–106. https://doi.org/10.1021/nl8025949
-
Lu, J., Bao, Y., Su, C. L., Loh, K. P. (2013). Properties of Strained Structures and Topological Defects in Graphene. ACS Nano, 7, 8350–8357 (online 22 October 2013). https://doi.org/10.1021/nn4051248
-
SALVE news (2011). Transformations in Carbon Nanostructures. salve-project.de, June 23, 2011. https://salve-project.de/newspress/transformation-in-carbon-nanostructures.html
-
Egerton, R. F., Malac, M. (2002). Improved Background-Fitting Algorithms for Ionization Edges in Electron Energy-Loss Spectra. Ultramicroscopy, 92, 47–56. https://doi.org/10.1016/S0304-3991(01)00155-3
-
Huang, P. Y., Ruiz-Vargas, C. S., van der Zande, A. M., et al. (2011). Grains and Grain Boundaries in Single-Layer Graphene Atomic Patchwork Quilts. Nature, 469, 389–392. https://doi.org/10.1038/nature09718
-
Angelova, P., Vieker, H., Weber, N.-E., et al. (2013). A Universal Scheme to Convert Aromatic Molecular Monolayers Into Functional Carbon Nanomembranes. ACS Nano, 7, 6489–6497. https://doi.org/10.1021/nn402652f
-
Palmer, R. E., Robinson, A. P. G., Guo, Q. (2013). How Nanoscience Translates Into Technology: The Case of Self-Assembled Monolayers, Electron-Beam Writing, and Carbon Nanomembranes. ACS Nano, 7, 6416–6421 (online 14 August 2013). https://doi.org/10.1021/nn403955e
-
Lee, J.-Y., Lee, J.-H., Kim, M. J., Dash, J. K., Lee, C.-H., Joshi, R., Lee, S., Hone, J., Soon, A., Lee, G.-H. (2017). Direct Observation of Grain Boundaries in Chemical Vapor Deposited Graphene. Carbon, 115, 147–153. https://doi.org/10.1016/j.carbon.2017.01.009




