Structural and mechanical properties of partially unzipped carbon nanotubes

We report molecular dynamics simulations of structural and mechanical properties of partially unzipped carbon nanotubes. Our results show that in the absence of edge passivation, partially unzipped carbon nanotubes are unstable with rising temperature depending on the geometry of cutting. When the length-to-width ratio of the graphene segment is not sufficiently large, the dangling bonds at the cutting front tend to reconnect to each other and form back to carbon nanotube structure; otherwise the structures roll up at the graphene end due to the competition of bending stiffness between longitudinal direction and transverse direction. When the graphene edges are hydrogen saturated, the self-healing behavior is suppressed. Tensile tests show that partially unzipped carbon nanotubes exhibit brittle fracture behavior, with a Young’s modulus around 700 GPa, which is comparable to that of carbon nanotubes and graphene.

Figure 1

Partially unzipped (12,12) CNT relaxed at different temperatures. Temperatures in panels (a), (b), (c), and (d) are 5 K, 100 K, 300 K, and 400 K, respectively. Cutting length $L=7.4$ nm. Also shown in the right panel of (d) is the total energy vs simulation time at 400 K. Carbon atoms at the cutting edges of GNR are highlighted.

Figure 2

Structural evolution of partially unzipped (12,12) CNT relaxed at different temperatures. (a) Total energy vs time in the simulation; some snapshots at 300 K and 400 K are given in (b) and (c), respectively. The cutting length in this model is 10.4 nm.

Figure 3

Cutting-width-dependent healing rate of a (12,12) PUCNT. Inset presents the healing rate of a (20,0) PUCNT vs temperature.

Figure 4

(a) Partially unzipped (6,6) CNTs relaxed at different temperatures; (b) freestanding (16,16) PUCNT relaxed at 50 and 100 K; (c) edge fixed (16,16) PUCNT relaxed at 400 K. The fixed atoms in panels (a) and (c) are highlighted in black.

Figure 5

Edge state effect on the stability of partially unzipped CNTs. The edge atoms in PUCNT in (a) and (b) are saturated by hydrogen atoms. The PUCNT structure in (c) is obtained by etching one row of carbon atoms along the axis direction, as sketched in the top CNT structure. (d) shows a (20,0) PUCNT cut in the middle relaxed at 100 K.

Figure 6

Tensile properties of a partially unzipped (12,12) CNT at 300 K. Left panels show the strain-dependent strain energy and stress; typical structural patterns under tension are presented in the right panel. The cutting length in this simulation is 7.4 nm.

Figure 7

Tensile properties of a partially unzipped (20,0) CNT with different unzipping geometry; insets show corresponding stress vs strain curve. The edge atoms of the GNR segment in the right panel are saturated with hydrogen atoms.

Figure 8

Tensile properties of graphene nanoribbons obtained by unzipping the (12,12) and (20,0) SWCNTs. Also shown in (a) and (b) are the side view of the unzipped GNR from (12,12) CNT with intrinsic rippling vanished at large strain.