Boron Nitride Chains
October 17, 2014 – The electron beam of an aberration-corrected TEM cuts atomic chains out of hexagonal boron nitride and images them in real time. Suspended between two parts of the lattice at 60 kV on the JEOL JEM-2100F at AIST in Tsukuba, the chains live about one second; lying on a second BN layer at 80 kV on the FEI Titan in Ulm, they live much longer. Multislice simulations and DFT show that the chains consist of alternating boron and nitrogen atoms. Researchers from AIST, Aalto University, NUST MISiS in Moscow and Ulm University thus complete the analogy between carbon and boron nitride.
A research team from AIST in Tsukuba, Aalto University in Finland, Moscow and Ulm, led by Kazu Suenaga, Arkady V. Krasheninnikov and Ute Kaiser, has for the first time produced and observed atomic chains of boron nitride (BN)[1]. The electron beam of an aberration-corrected transmission electron microscope cut the chains out of hexagonal boron nitride (h-BN) and imaged them in real time: suspended between two parts of the lattice at 60 kV on the JEOL JEM-2100F with dodecapole correctors at AIST, and lying on a second BN layer at 80 kV on the image-side corrected FEI Titan in Ulm. On the second layer they live much longer. Multislice image simulations and density functional theory (DFT) serve as validation that the chains consist of alternating boron and nitrogen atoms.
The carbon family gets a BN counterpart
Carbon and boron nitride form closely related structures: diamond and cubic BN, graphite and h-BN, benzene and borazine, and fullerenes and their BN counterparts, sometimes called fulborenes[2],[3]. Nanotubes and single atomic layers exist for both. Monatomic carbon chains had already been made and watched in the electron microscope, so the question was whether BN chains exist too. Calculations had predicted in 2002 that a linear BN chain is favoured over zigzag forms and should be even more stable than a carbon chain[4].
The team irradiated few-layer h-BN with the electron beam at low voltage, which keeps the damage small enough to follow the transformation with sub-nanometre resolution. At AIST the sample was held at 650 °C. The beam etched large holes, and where two holes grew towards each other, a narrow bridge remained: in single layers a narrow ribbon, in bilayers a tube-like structure. Most bridges simply broke, but several times they first turned into a linear chain (Fig. 1). Pairs of chains formed under the same conditions as well. Suspended chains survived for about one second under the beam.
Which atoms make the chain?
A single, otherwise unprocessed frame shows dark spots along the chain, about 0.25 nm apart on average (Fig. 2). Because the signal is weak, the order of boron and nitrogen cannot be read directly from the image. The team therefore calculated BN, pure boron and pure nitrogen chains and the corresponding ribbons one hexagon wide. Nitrogen chains are unstable under the conditions of the experiment, as expected[5]. A BN ribbon one hexagon wide is not stable either and splits into two BN chains, as calculations had predicted for carbon ribbons of the same width[6]. Boron, in contrast, is known to form triangular structures[7], and the triangular boron ribbon turned out to be lower in energy than the boron chain, not the other way round.
By calculating the temperature-corrected displacement cross sections for the electron beam (Fig. 3), the authors find that BN chains withstand irradiation better than boron chains and boron ribbons. Spacing, energy and cross sections all point to BN, as the box below shows.
Damage below the threshold
The calculations also show how easily the beam removes atoms. In perfect h-BN, knock-on damage needs at least 80 kV. At the edges of large holes the threshold drops to 40 kV for boron atoms, while nitrogen atoms stay above 60 kV[8]. In the experiment, however, holes also formed and grew below these limits. Single-layer h-BN was damaged step by step even at 15 and 30 kV (Fig. 4), and the holes grew faster at 15 kV than at 30 kV.
Ute Kaiser, head of the group Electron Microscopy of Materials Science at Ulm University, said: “We have observed this effect already in studies of graphene with 20 keV electrons[9]. The damage below the knock-on threshold is probably formed by chemical reactions between the sample and the residuals in the microscope vacuum (high vacuum). To further avoid radiation damage, an ideal solution would be on the one hand to use acceleration voltages below the knock-on threshold (to prevent direct removal of atoms, but still allow sufficient resolution), and on the other hand, to achieve better vacuum (ultra-high vacuum) in the electron microscope to limit the effects of chemical etching.”
The AIST group measured this damage in detail a few months later; the box below holds its images (Figs. 5 and 6).
Chains on a second layer
In Ulm, supported chains formed in the same way at 80 kV and room temperature: two layers of a trilayer area were etched away, leaving narrow bridges that turned into chains on the remaining layer (Fig. 7). These chains resisted the beam better than the layer beneath them. In one case several chains formed a closed loop, which lived until the supporting layer itself was etched away.
From the displacement cross section and the beam current, the displacement rate follows from the formula given by Florian Banhart[11]. For the supported chains, the calculation, consistent with the experiments, gives lifetimes of several tens of minutes, limited by the degradation of the supporting layer, whose threshold is higher than that of the chains[8]. Suspended chains live much shorter, although a lower voltage was used for them. Strain is one possible cause of breaking: the calculations allow at most 7 %, less than calculated for carbon chains[6],[12]. The strain arises because the BN layer on both sides of the chain degrades. Vibration is another possible cause. Both matter much less for supported chains, because they can change their anchor points to defects on the surface or to other points along the edge of the shrinking layer (Fig. 7). Atoms diffusing on the surface of the support could also accommodate changes in the chain structure and bonding. The box below compares the two cases.
Carbon’s cousins
The authors point out further analogies with atomic carbon chains, the structural relatives of BN chains. Carbon chains have also been made from few-layer graphene and observed in the electron microscope[13], including in Ulm[14]. A common problem is that the images show little structure of the chain itself, even where resolution and signal-to-noise ratio should allow it. This has been explained by the vibration of the chains. Calculations show that the vibration of carbon chains increases with tension[15]. Compared with graphene, these chains are very poor conductors, with a band gap that also increases with strain[13],[15]. This agrees with the calculations of the authors, which predict that BN chains are insulating, with a band gap close to that of h-BN. The box below sets the two chain types side by side.
Challenges named by the authors
The short lifetime and low beam resistance of the chains force short exposures and low doses, so the signal-to-noise ratio is low. The alternating B–N order is supported by spacing, energies and cross sections, but seeing it directly would need chains that live longer or are more stable[1]. Suspended chains survive only about one second; strain and vibration limit them. Imaging at 650 °C helps the structures to reconstruct and drives off hydrocarbons, but thermal drift slightly blurs the images, and the experiments had to be done far from the areas where contamination gathered. Damage below the knock-on threshold hardly changed with temperature; avoiding it would need both a low voltage and an ultrahigh vacuum. Triple chains were never seen, although wider ribbons should split into them: overcoming the energy barrier would need several electrons to hit at the same moment, while at any time only one electron passes through the sample.
A new member of the BN family
The study completes the analogy between carbon and boron nitride nanostructures with the atomic chain, confirming earlier theoretical predictions[1]. In addition to demonstrating that the electron beam can produce and image this previously unobserved one-dimensional material, the authors highlight that the beam is a well-suited tool for making atomic chains of ionic materials. In a Perspective in the same issue of ACS Nano, Youmin Rong and Jamie Warner of the University of Oxford called the work the first direct experimental confirmation of linear atomic chains in h-BN, eleven years after carbon chains were first imaged between nanotubes and five years after they were seen in graphene[16]. They see such chains as interconnects between two-dimensional materials in electronic devices and expect mixed chains of carbon, boron and nitrogen in hybrid sheets, which could be semiconducting rather than metallic like carbon chains or insulating like BN chains[16]. The work was supported by the German Research Foundation (DFG) and the Ministry of Science, Research and the Arts of Baden-Württemberg within the SALVE (Sub-Ångström Low-Voltage Electron Microscopy) project.
Resource:
Cretu, O., Komsa, H.-P., Lehtinen, O., Algara-Siller, G., Kaiser, U., Suenaga, K., Krasheninnikov, A. V. (2014).
Experimental Observation of Boron Nitride Chains.
ACS Nano 8(12), 11950–11957.
https://doi.org/10.1021/nn5046147
Download this news article as PDF – for reading on paper, taking notes or filing.
-
Cretu, O., Komsa, H.-P., Lehtinen, O., Algara-Siller, G., Kaiser, U., Suenaga, K., & Krasheninnikov, A. V. (2014). Experimental Observation of Boron Nitride Chains. ACS Nano, 8(12), 11950–11957. https://doi.org/10.1021/nn5046147
-
Kroto, H. W., Heath, J. R., O’Brien, S. C., Curl, R. F., & Smalley, R. E. (1985). C60: Buckminsterfullerene. Nature, 318(6042), 162–163. https://doi.org/10.1038/318162a0
-
Golberg, D., Bando, Y., Stéphan, O., & Kurashima, K. (1998). Octahedral Boron Nitride Fullerenes Formed by Electron Beam Irradiation. Applied Physics Letters, 73(17), 2441–2443. https://doi.org/10.1063/1.122475
-
Abdurahman, A., Shukla, A., & Dolg, M. (2002). Ab Initio Many-Body Calculations on Infinite Carbon and Boron-Nitrogen Chains. Physical Review B, 65(11), 115106. https://doi.org/10.1103/PhysRevB.65.115106
-
Buchsbaum, S., Mills, R. L., & Schiferl, D. (1984). Phase Diagram of Nitrogen Determined by Raman Spectroscopy From 15 to 300 K at Pressures to 52 GPa. The Journal of Physical Chemistry, 88(12), 2522–2525. https://doi.org/10.1021/j150656a018
-
Liu, M., Artyukhov, V. I., Lee, H., Xu, F., & Yakobson, B. I. (2013). Carbyne From First Principles: Chain of C Atoms, a Nanorod or a Nanorope. ACS Nano, 7(11), 10075–10082. https://doi.org/10.1021/nn404177r
-
Boustani, I. (1997). Systematic Ab Initio Investigation of Bare Boron Clusters: Determination of the Geometry and Electronic Structures of Bn (n = 2–14). Physical Review B, 55(24), 16426–16438. https://doi.org/10.1103/PhysRevB.55.16426
-
Kotakoski, J., Jin, C. H., Lehtinen, O., Suenaga, K., & Krasheninnikov, A. V. (2010). Electron Knock-on Damage in Hexagonal Boron Nitride Monolayers. Physical Review B, 82(11), 113404. https://doi.org/10.1103/PhysRevB.82.113404
-
Meyer, J. C., Eder, F., Kurasch, S., Skakalova, V., Kotakoski, J., Park, H. J., Roth, S., Chuvilin, A., Eyhusen, S., Benner, G., Krasheninnikov, A. V., & Kaiser, U. (2012). Accurate Measurement of Electron Beam Induced Displacement Cross Sections for Single-Layer Graphene. Physical Review Letters, 108(19), 196102. https://doi.org/10.1103/PhysRevLett.108.196102
-
Cretu, O., Lin, Y.-C., & Suenaga, K. (2015). Inelastic Electron Irradiation Damage in Hexagonal Boron Nitride. Micron, 72, 21–27. https://doi.org/10.1016/j.micron.2015.02.002
-
Banhart, F. (1999). Irradiation Effects in Carbon Nanostructures. Reports on Progress in Physics, 62(8), 1181–1221. https://doi.org/10.1088/0034-4885/62/8/201
-
Nair, A. K., Cranford, S. W., & Buehler, M. J. (2011). The Minimal Nanowire: Mechanical Properties of Carbyne. EPL, 95(1), 16002. https://doi.org/10.1209/0295-5075/95/16002
-
Cretu, O., Botello-Mendez, A. R., Janowska, I., Pham-Huu, C., Charlier, J.-C., & Banhart, F. (2013). Electrical Transport Measured in Atomic Carbon Chains. Nano Letters, 13(8), 3487–3493. https://doi.org/10.1021/nl4018918
-
SALVE news (2009). Graphene: Chains and Edges. salve-project.de, August 19, 2009. https://salve-project.de/newspress/formation-of-linear-carbon-chains.html
-
Artyukhov, V. I., Liu, M., & Yakobson, B. I. (2014). Mechanically Induced Metal–Insulator Transition in Carbyne. Nano Letters, 14(8), 4224–4229. https://doi.org/10.1021/nl5017317
-
Rong, Y., & Warner, J. H. (2014). Wired Up: Interconnecting Two-Dimensional Materials With One-Dimensional Atomic Chains. ACS Nano, 8(12), 11907–11912. https://doi.org/10.1021/nn5065524







