 Nano Review
 Open Access
 Published:
HOT Graphene and HOT Graphene Nanotubes: New Low Dimensional Semimetals and Semiconductors
Nanoscale Research Letters volume 15, Article number: 56 (2020)
Abstract
We report a new graphene allotrope named HOT graphene containing carbon hexagons, octagons, and tetragons. A corresponding series of nanotubes are also constructed by rolling up the HOT graphene sheet. Ab initio calculations are performed on geometric and electronic structures of the HOT graphene and the HOT graphene nanotubes. Dirac cone and high Fermi velocity are achieved in a nonhexagonal structure of HOT graphene, implying that the honeycomb structure is not an indispensable condition for Dirac fermions to exist. HOT graphene nanotubes show distinctive electronic structures depending on their topology. The (0,1) n (n ≥ 3) HOT graphene nanotubes reveal the characteristics of semimetals, while the other set of nanotubes (1,0) n shows continuously adjustable band gaps (0~ 0.51 eV) with tube size. A competition between the curvature effect and the zonefolding approximation determines the band gaps of the (1,0) n nanotubes. Novel conversion between semimetallicity and semiconductivity arises in ultrasmall tubes (radius < 4 Å, i.e., n < 3).
Introduction
Because of its bonding flexibility, carbonbased systems show an unlimited number of different structures with an equally large variety of physical properties. These physical properties are, in great part, the result of the dimensionality of these structures [1]. Graphene is a single twodimensional layer of carbon atoms bound in a hexagonal lattice structure [2] revealing a number of unique properties, such as massless carriers, high Fermi velocity [3], and Dirac cones [4, 5], which are characteristic of twodimensional Dirac fermions. The honeycomb lattice consisting of two equivalent carbon sublattices plays a crucial role in forming such intriguing properties [2]. Enyashin and Ivanovskii [6] constructed 12 artificial 2D carbon networks but found no structures other than the graphyne allotrope exhibit the graphenelike electronic behavior. It seems to imply that the Diraclike fermions in sp^{2}bonded carbon systems are dependent on the honeycomb structure. In the lower dimension, the carbon nanotube is a honeycomb structure rolled into a hollow cylinder with nanometric diameter and μm length [7,8,9,10]. As there is an infinite number of ways of rolling a sheet into a cylinder, the large variety of possible helical geometries, defining the tube chirality, provides a family of nanotubes with different diameters and microscopic structures [11,12,13]. The electronic and transport properties are certainly among the most significant physical properties of carbon nanotubes, and crucially depend on the diameter and chirality [14,15,16,17,18]. Graphene nanotubes can be either semimetallic [14] or semiconducting [19,20,21], with a band gap varying from zero to a few tenths of an eV, depending on their diameter and chirality [10, 14, 16]. Furthermore, the band gap of semiconducting tubes can be shown to be simply related to the tube diameter. The semimetallic nanotubes also maintain the unique properties from graphene, such as massless carriers, high Fermi velocity [22], and Dirac cones [23]. Such remarkable results can be obtained from a variety of considerations, starting from the socalled bandfolding approach, based on knowledge of the electronic properties of the graphene sheet, to the direct study of nanotubes using semiempirical tightbinding approaches [14, 16, 18, 23]. Comparing with more sophisticated ab initio calculations and available experimental results, finer considerations, such as curvature effects, k_{F} shifting [24, 25], σπ hybridization [26] are introduced. Graphene and graphenelike materials [6] are considered a revolutionary material for future generation of highspeed electronic, radio frequency logic devices [27, 28], thermally and electrically conductive reinforced composites [29, 30], catalyst [31], sensors [32,33,34,35], transparent electrodes [27, 36], etc. basing on the unusual properties all above. Over the past few decades, carbon nanotubes have also shown great potential in logic circuits, gas storage, catalysis, and energy storage because of their extraordinary electronic, mechanical, and structural properties [37,38,39]. Hence, the creation of new carbon allotropes (including 2D and 1D) has been the focus of numerous theoretical and experimental explorations because of their fundamental scientific and technological importance [40]. However, completely clarifying the structures of these exciting carbon phases through current experimental technologies is usually unrealistic due to their limited quantity, as well as the mixture of other phases. Theoretical prediction is necessary and has yielded great success [31,32,33,34,35, 40,41,42].
In this study, we designed a new allotrope of graphene that has twodimensional Dirac fermions without an exclusively hexagonal structure. The new allotrope was constructed with interlaced carbon hexagons, octagons, and tetragons, and was named HOT graphene. HOT graphene nanotubes were also constructed by rolling up HOT graphene sheet along with different directions. The electronic property, curvature effect, k_{F} shifting effect, etc. of HOT graphene and nanotubes were calculated using ab initio calculations based on density function theory (DFT).
Method of Calculation
The present calculations on HOT graphene and HOT graphene nanotubes were performed by using a firstprinciples method based on the densityfunctional theory (DFT) with the generalized gradient approximation (GGA) in the form of PerdewBurkeErnzerh (PBE) exchangecorrelation functional [43], as implemented in the Vienna Ab initio Simulation Package (VASP) [44, 45]. The wave functions were expanded in plane waves up to a cutoff of plane wave kinetic energy of 520 eV. The Brillouin zone (BZ) integrals were performed by using a MonkhorstPack [46] sampling scheme with a kpoint mesh resolution of 2π × 0.03 Å^{− 1}. The unit cell basis vectors (representing unit cell shape and size) and atomic coordinates were fully relaxed in each system until the forces on all the atoms were smaller than 0.01 eV/Å.
Results and Discussion
Geometric and Electronic Structures of HOT Graphene
The geometric structure of HOT graphene (Fig. 1a) shows a more complicated bonding situation than graphene. The variety of carbon polygons in HOT graphene results in various carbon bonding characters. These polygons in HOT graphene share common edges with each other, and the bonds can be distinguished by the two polygons they belong to. Therefore, in our research, they are named as 6–8 bonds, 4–8 bonds, 4–6 bonds, 6–6 bonds, and 8–8 bonds. The 4–8 bonds and 6–8 bonds have two different bond lengths: 1.44 Å and 1.47 Å for 4–8 bonds; 1.41 Å and 1.48 Å for 6–8 bonds. The 4–6 bonds, 6–6 bonds, and 8–8 bonds have unique bond lengths of 1.44 Å, 1.46 Å, and 1.34 Å, relatively. Figure 1b shows the band structure and density of states (DOS) of HOT graphene with the corresponding BZ depicted in Fig. 1c. The crossing point of energy bands at the Fermi level indicates semimetallicity of HOT graphene, which is confirmed by the vanishing DOS at the Fermi level. The Dirac point is located at (0, 0.0585, 0) adjacent to Γ. The 3D band structure (Fig. 2) presents the band surfaces near the Fermi level, where one can see the Dirac cones formed by upper and lower conical surfaces meeting at two Dirac points exactly at the Fermi surface. The corresponding Fermi velocity (v_{F}) of the Dirac fermions, evaluated from the gradient of the linear dispersions of the band structures, is 6.27 × 10^{5} m/s, which is a little lower than 8.1 × 10^{5} m/s [22] for graphene nanotube and 8.6 × 10^{5} m/s [47, 48] for graphene. The high v_{F} implies high mobility of carriers in the HOT graphene.
Geometric and Electronic Structures of HOT Graphene Nanotubes
The HOT graphene tubes are rolled up from the HOT graphene sheet in various directions symbolized by an index on the 2D HOT graphene lattice (Fig. 3a). This index is denoted as (l,m) n basing on two unit vectors a and b. Symbol (l,m) [18, 20, 21] stands for different directions on the HOT graphene sheet, n stands for the number of perimeter units (Fig. 3b and c) used in rolling up the tubes. Under the space reversal symmetry of the HOT graphene unit cell, the nanotube symbol is confined to 0 ≤ l and 0 ≤ m to avoid a manytoone correspondence. The number of possible geometric structures of the HOT graphene nanotubes is infinite since there are infinite choices for both diameters and rolling directions. Among such a large number of nanotubes, only two directions, (0,1, 1,0), are selected in our study. This is because the helical arrangement of atoms in other directions is barely periodic and possesses a nearly infinite length of unit cell along the tube axis. Such a large unit cell is unrealistic in our calculation. Different rolling directions between (1,0) n and (0,1) n tubes result in the differences in geometry and bonding situation. Two tubes, (1,0)6 and (0,1)4, are depicted in Fig. 3d and e to describe the geometric differences between the two rolling directions. The arrangement of polygons along (1,0) direction can be divided into two patterns: C_{4}–C_{6}–C_{8} (orange) and C_{8}–C_{6}–C_{4} (blue) which is exactly opposite to each other. These two opposite patterns alternate along the circumferential direction of the tube. In direction (0,1), polygons arranging along the tube axis also have two patterns: C_{4}–C_{8} (blue) and C_{6}–C_{6} (orange). Two C_{4}–C_{8} patterns alternate with one C_{6}–C_{6} pattern along the circumferential direction.
To reveal the energy cost in rolling up a sheet into tubes, we define the curvature energy (Fig. 4) E_{cur} as:
where \( {E}_{\mathrm{tube}}^{\mathrm{at}} \) is the average energy of atoms in nanotube, and \( {E}_{\mathrm{sheet}}^{\mathrm{at}} \) is the average energy of atoms in the 2D sheet. The (1,0) n tubes exhibit a lower energy cost than the graphene nanotubes while the (0,1) n tubes are nearly the same as the armchair graphene nanotubes except for several ultrasmall tubes. Such results also suggest that it is possible to synthesize the HOT graphene nanotubes in experiments. It should be noted that tube (1,0)1 is too small because its diameter is even shorter than the bond length of carbon.
The calculated electronic band structure and DOS of nanotube (0,1)6 (Fig. 5b) indicate a semimetallic character. When the HOT graphene sheet is rolled up into a nanotube, its 2D BZ reduces to 1D BZ as a result of the new periodic boundary conditions in nanotubes. The periodic boundary conditions along the circumferential direction of the tube only allow wave vectors “around” the nanotube circumference and these vectors are quantized [49]. The periodic boundary conditions along the nanotube axis remain the same as the 2D sheet, then the wave vectors remain continuous along the nanotube axis. According to the zonefolding scheme, the electronic band structure of a specific nanotube is given by the superposition of the electronic energy bands of the corresponding 2D sheet along the specifically allowed k lines [50]. As the quantized wave vectors in the middle of the BZ of the HOT graphene always cross the Dirac point (point F in Fig. 5a), a nondegenerate Dirac point (point F in Fig. 5b) and Dirac cone appears in the band structures of the (0,1) n HOT graphene nanotubes, resulting in the semimetallicity of all the (0,1) n nanotubes. Tube (0,1)6 is calculated to evaluate the semimetallicity of this set of (0,1) n nanotubes in Fig. 5. Band structure of the (0,1)6 HOT graphene nanotubes shows a crossing point of energy bands at the Fermi level and the corresponding DOS shows no states at the Fermi level which verifies the semimetallicity of the system. As the Dirac cone in (0,1) n tubes is originated from the HOT graphene sheet, the calculated coordinate of the Dirac point in the (0,1)6 nanotube is the same as that in HOT graphene sheet which is (0, 0.0585, 0). The calculated fermi velocity at the Dirac point in (0,1)6 nanotube is 6.76 × 10^{5} m/s, close to 6.27 × 10^{5} m/s in the HOT graphene sheet.
The band structure evolution of (0,1) n nanotubes with different tube radii (Fig. 6a) reveals that the HOT graphene nanotubes (0,1) n are semimetallic (n ≥ 3) and transform to metal (n = 2) and then turn back to semimetal (n = 1). Such a change under a small radius resulted from the socalled curvature effect [26]. In the situation of ultrasmall nanotubes (e.g., (0,1)2 and (0,1)1), the curvature takes a nonnegligible effect on the zonefolding scheme. The bond length and bond angle undergo a nonnegligible change within the big curvature, which has an influence on the electronic band structure. The band structures of such nanotubes are no longer a simple superposition of energy bands on the allowed wave vectors. This change of bonds modifies the conditions that define the k point at which occupied and unoccupied bands do cross (at a point we label k_{F}) and shift the k_{F} away from the original Dirac points, which is called the k_{F} shifting effect [24, 25]. As a result, in the (0,1) n HOT graphene nanotubes, the k_{F} shifts away from its original position (k_{HOT}) in the HOT graphene sheet (point F in Fig. 5a). And the shifting direction of k_{F} is calculated to be along the allowed wave vector (dash lines in Fig. 5a), resulting in no change of the semimetallicity (Fig. 6a (n ≥ 3)). According to the calculated coordination of Dirac points, the k_{F} shifting effect becomes nonnegligible at (0,1)5 with a tube radius of 5.988 Å, whose Dirac point (k_{F}) shifts to (0,0.0626,0) from original point (k_{HOT}) at (0,0.0594,0) in HOT graphene sheet. As the tube radius gets smaller, the k_{F} keeps shifting and reaches point (0,0.0712,0) in tube (0,1)3. In tube (0,1)2, the k_{F} shifts to (0,0.0835,0) where the Dirac point moves down below the Fermi level, resulting in a metallic system. The vanishing of the semimetallicity in (0,1)2 indicates a deviation from the k_{F} shifting effect in the (0,1) n nanotubes (n ≥ 3). Moreover, the (0,1)1 tube becomes semimetallic again in its band structure and the DOS (Fig. 6b). Our electron state analysis of the (0,1) n HOT nanotubes shows π states overlapping at n ≥ 2, which is usually considered the origin of the semimetallicity of graphene nanotubes [18, 24]. However, the corresponding electron state analysis of the HOT graphene nanotube (0,1)1 shows a σπ hybridization that a lowlying σ* band intersects the Fermi level and joins the Dirac cone (blue lines in Fig. 6a). The coordination of Dirac point, which is (0,0.18345,0), also exhibits a distinction from other (0,1) n HOT graphene nanotubes. The calculated Fermi velocity is 4.47 × 10^{5} m/s, lower than 6.27 × 10^{5} m/s in HOT graphene sheet and values for other (0,1) n nanotubes (~6.76 × 10^{5} m/s). An obviously different shape of its band structure is also shown in Fig. 6a. All these characters verify that the semimetallicity of the HOT graphene nanotube (0,1)1 is originated from the σπ hybridization. In summary, with the increasing curvature, the k_{F} shifting effect emerges at n = 5, becomes more effective at 4 ≥ n ≥ 2, and is finally replaced by the σπ hybridization effect at n = 1.
In another rolling direction, the calculated band structure of (1,0)6 (Fig. 7b) shows semiconductivity. The 6 wave vectors (dash lines) are parallel to ΓM, crossing the Г point in the middle of the BZ of HOT graphene (Fig. 7a), and open a 0.46 eV band gap as shown in the DOS (Fig. 7b). In this rolling direction, the allowed wave vectors in the middle of the BZ never include the Dirac points, which results in the nonzero band gaps in this set of nanotubes. The band evolution reveals a change of band gaps with different radii in Fig. 8a. The (1,0) n HOT graphene nanotubes are semiconductive (n ≥ 4). The valence band maximum (VBM) and conduction band minimum (CBM) get closer from tube (1,0)6 to tube (1,0)4 and then crosses with each other in tube (1,0)3 with a radius of 2.17 Å. This crossing point is exactly on the Fermi level located at (0.0791,0,0). The calculated DOS (Fig. 8b) presents 0 states at the Fermi level of (1,0)3, verifying the semimetallicity. When the tube radius decreases to (1,0)2, a 0.848 eV gap opens up again. The CBM and VBM of (1,0)2 are located at Γ point and M point, respectively, indicating an indirect band gap. This change in VBM implies a different origin for the semiconductivity in (1,0)2. A further study of the bandgap change in the (1,0) n HOT graphene nanotubes is shown in Fig. 9. The band gap evolution as a function of n (3 ≤ n ≤ 23) indicates that the band gap is adjustable with tube sizes. Also, instead of being monotonous, the dependence of the band gap as the tube size has a zigzag shape (Fig. 9 black curve). The global minimum at (1,0)15 presents a zero band gap. The semimetallicity of (1,0)15 is further confirmed by the band structure and DOS (Fig. 9g). From the zonefolding scheme, we know that the band structure of nanotubes is the superposition of the band structure of the 2D sheet along the corresponding quantized k lines [50]. Thus, the semimetallicity indicates that at least one of the allowed k lines (dash lines in Fig. 7a) intersects the Dirac points (red points in Fig. 7a) at n = 15. Otherwise, if the allowed k lines have a distance from the Dirac points (k_{HOT}), a band gap will appear in nanotube. Furthermore, this distance between k_{F} and the k lines is proportional to the band gaps, as the band dispersion near the Dirac cone is linear [25]. The Δk_{m} measures the shortest distance between k_{F} and k_{m} lines of quantum number m. We calculated this distance Δk_{m} between the Dirac points (k_{HOT}) and the allowed k lines in the HOT graphene and plot it (red line) together with the band gaps (black line) in Fig. 9. Firstly, all the allowed k lines have a quantum number m. When n ≤ 7 (e.g., Fig. 9a), the shortest distance (Δk_{m}) is between the k_{HOT} and the first k line, k_{1}, which sits at the Γ point constantly (insert graph (a) in Fig. 9). In this situation, Δk_{m} is constant as both k_{1} and k_{HOT} are constant independents of the tube size. However, as the allowed k lines become denser in larger nanotubes (7 ≤ n ≤ 17), the k_{2} becomes the nearest one to k_{HOT} (e.g. Fig. 9b). In this situation, the k_{2} approaches the Dirac point from the outer BZ with increasing tube radii; therefore, it shows a decline of Δk_{m} in Fig. 9 (7 ≤ n ≤ 17). n = 17 is a turning point where k_{2} almost intersects the Dirac point resulting in a local minimum of the distance Δk_{m} (Fig. 9c). As the radius keeps going up, the k_{2} traverse the k_{HOT} point and continues to move away from it to the Г point resulting in an increase of the distance again at 17 ≤ n ≤ 24 (e.g., Fig. 9d). At the same time, k_{3} is approaching the Dirac point. k_{3} gets closer to k_{HOT} than k_{2} and begins a new decrease in distance Δk_{m} at n ≥ 24. As the band gaps are proportional to this distance Δk_{m} [25], the band gap curve shows the same shape as the Δk_{m} plot (n ≥ 7). And it is revealed that the band gaps change in cycles: the k_{m} gets closer to the Dirac point (k_{HOT}) causing a decline of band gap, then traverses the Dirac point resulting in a local minimum, then gets farther from the Dirac point causing a rising of band gap, and is finally replaced by the next line k_{m + 1} entering the next cycle. In summary, the reason why the band gaps are changing with tube size (n ≥ 7) is that the k lines are moving with different tube sizes, thereby the change the distance Δk_{m} between k_{HOT} and the allowed k lines which is proportional to the band gaps.
Though the Δk_{m} curve and the band gap curve have similarities in shape, the differences between them are also obvious, which is that the Δk_{m} plot shows a “delay” in change at n ≥ 7 and becomes completely different from the band gap plot at 3 ≤ n ≤ 7. The reason is that the Fermi point in HOT graphene nanotubes (k_{F}) is assumed to have the same coordinate as the original Fermi point in HOT graphene (k_{HOT}) in the preceding section where the band gap changing was explained. However, the Fermi point (k_{F}) in nanotubes shifts away from the origin Fermi point (k_{HOT}) in HOT graphene sheet under the curvature effect. Therefore, the k_{F} shifting (Δk_{F} = k_{F}–k_{HOT}) effect contributes to the mismatches between the distance Δk_{m} and the band gaps. As calculated in the (0,1) n tubes (Fig. 6), the k_{F} in (1,0) n tubes also shifts to the outer BZ towards the symmetric point X under the curvature. Therefore, when the nearest k_{m} sits between the k_{HOT} and the Γ point, the distance Δk_{m} underestimates the band gap (e.g., 17 ≤ n ≤ 24 in Fig. 9d). Otherwise, the nearest k_{m} sitting at outer the k_{HOT} point, it results in an overestimate of the band gap (e.g., 7 ≤ n ≤ 17 in Fig. 9b). In small nanotubes (3 ≤ n ≤ 7), the k_{F} shifting effect is enhanced under the large growth rate of the curvature; consequently, it causes a drastic k_{F} shifting and changes the band gap. When the radius is getting smaller than n = 8, the k_{1} becomes the closest to the k_{HOT} indicating a constant Δk_{m}. However, the k_{F} shifting effect is so strong to move the k_{F} farther from k_{1} but closer to the k_{2} (Fig. 9e). The k_{F} shifting effect wins the competition with the k lines moving and begins to determine the band gaps since then (n ≤ 7). The k_{F} keeps shifting to the k_{2} at a high velocity so that the distance between k_{F} and k_{2} gets smaller and smaller. Therefore, from (1,0)7 to (1,0)3, the band gap decreases (Fig. 9). At last, the k_{F} catches up with k_{2} and crosses it at n = 3(Fig. 9f). This crossing of a k line with the k_{F} results in the Dirac point in tube (1,0)3 that gives rise to the semimetallicity (Fig. 8) as discussed in the preceding sections. Further decrease of radius to n = 2 opens a 0.848 eV gap in the tube (1,0)2 (Fig. 8). This gap is so large and is considered out of the k_{F} shifting scheme, and hence is not plotted in Fig. 9. In summary, there is a competition mechanism between the k line moving and the k_{F} shifting in determining the band gaps. The k_{F} shifting effect leads in small tubes (7 ≥ n ≥ 3), while the k line moving leads in big tubes (n ≥ 8) where the k_{F} shifting effect is faded. The amount of the k_{F} shifting is estimated to be 0.0015 2π/Å at n = 15 and 0.0238 2π/Å at n = 3 as the band gaps are 0 eV where the shifted k_{F} point is on the allowed k lines. It can be seen that the k_{F} shifting is 15.86 times bigger in a small tube (tube (1,0)3) than in a big tube (tube (1,0)15).
In HOT graphene, carbon atoms are all threefold coordinated, thus the fourth valence electron plays a key role in its conductivity. The calculated band decomposed charge density ± 0.15 eV around the Fermi level (Fig. 10) shows the distribution of the electrons in the Dirac cone. Only the electrons on the 8–8 bonds (Fig. 10a) have an overlapping and the side view (Fig. 10b) shows that the electrons distribute perpendicularly to the HOT graphene sheet, which indicates that the Dirac cone consists of π states. Therefore, the electron overlapping on the 8–8 bonds (enlarged side view in Fig. 10a) is considered to be localized π states. In big nanotubes such as (0,1)6 (Fig. 10c), the charge density is similar to the HOT graphene sheet showing localized π bonds on 8–8 bonds (enlarged side view in Fig. 10c). As the radius decreases to (0,1)2, whose conductivity transforms to metal (Fig. 6), the 8–8 bonds show several deformations (Fig. 10d). Firstly, these states are no longer symmetric with respect to the tube wall. The overlapping of the π state outside the tube wall breaks apart while the π states inside keep overlapping with each other. Besides the 8–8 π bonds, new π bonds form on 4–8 in (0,1)2. These bonds are similar to the deformed 8–8 bonds: separated π states outside the tube wall and overlapping π states inside the tube wall. Every 4–8 bond connects two 8–8 bonds adjacent to it, forming a delocalized π overlapping inside the tube along the tube axis direction. The enlarged side view in Fig. 10d shows the connection between 4 and 8 and 8–8 bonds as a segment of the whole delocalized bond. Therefore, the metallicity in tube (0,1)2 can be attributed to the delocalized π overlapping in the 4–8 and 8–8 bonds along the tube axis direction which provides a pathway for the electrons to travel along the tube. When the radius keeps decreasing, the conductivity disappears and the (0,1)1 tube becomes a semimetal again (Fig. 6). Different from all the other (0,1) n tubes, the 8–8 overlapping (Fig. 10e) in nanotube (0,1)1 is totally broken up; instead, 4–8 overlapping and 4–6 overlapping plays the major role in the Dirac cones. These two bonds belong to two opposite edges in the same carbon tetragons and are arranged parallel to the tube axis. Furthermore, they are no longer π states. Based on the electronic state analysis, a σπ hybridization takes place at the Fermi level under such a strong curvature in nanotube (0,1)1. It is verified by the charge density which shows the electron states distribute closely to the bond axis (enlarged side view in Fig. 10e). Strongly modified lowlying σ states are introduced at the Fermi level as discussed in the preceding sections (blue line in Fig. 6). Therefore, the σπ hybrid states in the 4–8 and 4–6 bonds are considered the reason for the semimetallicity in (0,1)1, which is essentially different from the other semimetallic tubes (0,1) n (n ≥ 3). In another set of HOT graphene nanotubes, the band gaps show adjustability with different tube radius (Fig. 9). The charge densities also present an evolution with the tube radii in Fig. 11. The band decomposed charge density of tube (1,0)9 in Fig. 11a and b shows the localized π states overlapping in both the VBM and CBM. The VBM is contributed by π states on 6–6 bonds and 8–8 bonds (Fig. 11a). The CBM is contributed by π states on part of 4–8 and 6–8 bonds (Fig. 11b). The 4–6 bonds have no states on both of the VBM and CBM. When the (1,0) n tubes become semimetallic at some specific radii, such as n = 15, the VBM and CBM meet with each other. Band decomposed charge density ± 0.15 eV around the Fermi level of (1,0)15 nanotube show the different distribution of electrons from the semiconductive tubes. More importantly, it exhibits a similar distribution to the semimetallic HOT graphene sheet and (0,1) n tubes. The localized π bond of (1,0)15 only locates on the 8–8 bonds. This redistribution of electron in (1,0) n tubes causes the conductivity change.
Conclusion
A new graphene allotrope named HOT graphene is constructed by carbon hexagons, octagons, and tetragons showing Dirac cone and high Fermi velocity, which implies that the honeycomb structure is not an indispensable condition for Dirac fermions to exist. The semiconductivity of HOT graphene is dependent on the localized π bonding. A corresponding series of nanotubes is rolled up from the HOT graphene sheet and shows distinctive electronic structures depending on the topology. The set of (0,1) n (n ≥ 3) HOT graphene nanotubes reveals a character of semimetallicity and Dirac cones that are composed by π states. A nonnegligible k_{F} shifting along the allowed k line arises under the curvature effect when the tube radius gets smaller (3 ≤ n ≤ 5). However, the ultrasmall nanotube (0,1)2 begins to deviate from the π statebased k_{F} shifting effect showing a transformation to metallicity. Finally, an σπ hybridization takes the place of the π states at Fermi level in nanotube (0,1)1, where a lowlying σ* band intersection appears at the Fermi level and forms a semimetallicity again. Another set of tubes (1,0) n shows various band gaps (0~ 0.51 eV), which is continuously adjustable with the tube size. The band gaps of (1,0) n (n ≥ 3) nanotubes turn out to be determined by a competition mechanism between the k line moving and the kF shifting effect. The zonefolding approximation indicates a k line moving and results in the zigzag and periodical band gap changing curve in big tubes (n ≥ 8), while the kF shifting effect gets stronger and causes a dramatic decrease of band gaps in small tubes (7 ≥ n ≥ 3). Zerogap semimetallic tubes appear periodically under the competition.
Availability of Data and Materials
Authors declare that the datasets used and/or analyzed during the current study are available to the readers and included in this article.
Abbreviations
 1D:

Onedimensional
 2D:

Twodimensional
 3D:

Threedimensional
 BZ:

Brillouin zone
 CBM:

Conduction band minimum
 DFT:

Density functional theory
 DOS:

Density of states
 GGA:

GeneralizedGradient Approximation
 HOT :

Hexagons, octagons, and tetragons
 PBE:

Perdew–Burke–Ernzerh
 VASP:

Vienna Ab initio Simulation Package
 VBM:

Valence band maximum
References
 1.
Castro Neto AH, Guinea F, Peres NMR, Novoselov KS, Geim AK (2009) The electronic properties of graphene. Rev Mod Phys 81:109
 2.
Cooper DR, D’Anjou B, Ghattamaneni N, Harack B, Hilke M, Horth A, Majlis N, Massicotte M, Vandsburger L, Whiteway E, Yu V (2012) Experimental review of graphene. ISRN Condensed Matter Physics 2012:1
 3.
Novoselov KS, Geim AK, Morozov SV, Jiang D, Katsnelson MI, Grigorieva IV, Dubonos SV, Firsov AA (2005) Twodimensional gas of massless Dirac fermions in graphene. Nature 438:197
 4.
Hass J, de Heer WA, Conra EH (2008) The growth and morphology of epitaxial multilayer graphene. J Phys Condens Matter 20:323202
 5.
Zhang Y, Tan YW, Stormer HL, Kim P (2005) Experimental observation of the quantum Hall effect and Berry’s phase in graphene. Nature 438:201
 6.
Enyashin AN, Ivanovskii AL (2011) Graphene allotropes. Phys Status Solidi B 248:1879
 7.
Iijima S (1991) Helical microtubules of graphitic carbon. Nature 354:56
 8.
Bethune DS, Kiang CH, de Vries MS, Gorman G, Savoy R, Vazquez J, Beyers R (1993) Cobaltcatalysed growth of carbon nanotubes with singleatomiclayer walls. Nature 363:605
 9.
Iijima S, Ichihashi T (1993) Singleshell carbon nanotubes of 1nm diameter. Nature 363:603
 10.
Guo T, Nikolaev P, Thess A, Colbert DT, Smalley RE (1995) Catalytic growth of singlewalled nanotubes by laser vaporization. Chem Phys Lett 243:49
 11.
Saito, R., G. Dresselhaus, and M. S. Dresselhaus, (1998) Physical Properties of Carbon Nanotubes Imperial College Press, London SW7 2BT. ISBN 1860940935
 12.
Damnjanovic M, Miloševic I, Vukovic T, Sredanovic R (1999) Full symmetry, optical activity, and potentials of singlewall and multiwall nanotubes. Phys Rev B 60:2728
 13.
Reich, S., C. Thomsen, and J. Maultzsch, (2004) Carbon nanotubes, WileyVCH Verlag GmbH & Co. KGaA, Weinheim ISBN 3527403868
 14.
Mintmire JW, Dunlap BI, White CT (1992) Are fullerene tubules metallic? Phys Rev Lett 68:631
 15.
Loiseau A, Launois P, Petit P, Roche S, Salvetat JP (2006) Understanding Carbon Nanotubes Lecture Notes in Physics Vol. 677. SpringerVerlag, Heidelberg
 16.
Saito R, Fujita M, Dresselhaus G, Dresselhaus MS (1992) Electronic structure of chiral graphene tubules. Appl Phys Lett 60:2204
 17.
Vukovic T, Milosevic I, Damnjanovic M (2002) Carbon nanotubes band assignation, topology, Bloch states, and selection rules. Phys Rev B 65:045418
 18.
Hamada N, Sawada S, Oshiyama A (1992) New OneDimensional Conductors: Graphitic Microtubules. Phys. Rev. Lett 68:1579
 19.
White CT, Mintmire JW (1998) Density of states reflects diameter in nanotubes. Nature 394:29
 20.
Saito R, Dresselhaus G, Dresselhaus MS (2000) Trigonal warping effect of carbon nanotubes. Phys Rev B 61:2981
 21.
Zhou X, Park JY, Huang S, Liu J, McEuen PL (2005) Band Structure, Phonon Scattering, and the Performance Limit of SingleWalled Carbon Nanotube Transistors. Phys Rev Lett 95:146805
 22.
Liang W, Bockrath M, Bozovic D, Hafner JH, Tinkham M, Park H (2001) FabryPerot interference in a nanotube electron waveguide. Nature 411(665)
 23.
Reich S, Maultzsch J, Thomsen C, Ordejon P (2002) Tightbinding description of graphene. Phys Rev B 66:035412
 24.
Kane CL, Mele EJ (1997) Size, shape, and low energy electronic structure of carbon nanotubes. Phys Rev Lett 78:1932
 25.
Yang L, Han J (2000) Electronic structure of deformed carbon nanotubes. Phys Rev Lett 85:154
 26.
Blase X, Benedict LX, Shirley EL, Louie SG (1994) Hybridization effects and metallicity in small radius carbon nanotubes. Phys Rev Lett 72:1878
 27.
Kumar TN, Mohan L, Karakkad S (2019). Carbon nanofiber based epoxy nano compositesan innovative material for electromagnetic shielding. AIP Conf Proc 2105:020029
 28.
Takahashi H, Suzuki Y, Yoshida N, Nakagawa K, Maki H (2019) Trionbased Highspeed Electroluminescence from Semiconducting Carbon Nanotube Films. arXiv preprint arXiv 1903:01640
 29.
Pokharela P, Xiaoa D, Erogbogbob F, Kelesb O, Lee DS (2019) A hierarchical approach for creating electrically conductive network structure in polyurethane nanocomposites using a hybrid of graphene nanoplatelets, carbon black and multiwalled carbon nanotubes. Composites Part B 161(169)
 30.
SánchezRomatea XF, Artigas J, JiménezSuárez A, Sánchez M, Güemes A, Ureña A (2019) Critical parameters of carbon nanotube reinforced composites for structural health monitoring applications: Empirical results versus theoretical predictions. Compos Sci Technol 171:44
 31.
Meng J, Niu C, Xu L, Li J, Liu X, Wang X, Wu Y, Xu X, Chen W, Li Q, Zhu Z, Zhao D, Mai L (2017) General Oriented Formation of Carbon Nanotubes from Metal−Organic Frameworks. J Am Chem Soc 139:8212
 32.
Gui Y, Tang C, Zhou Q, Xu L, Zhao Z, Zhang X (2018) The sensing mechanism of Ndoped SWCNTs toward SF6 decomposition products: A firstprinciple study. Appl Surf Sci 440:846
 33.
Wang Y, Gui Y, Ji C, Tang C, Zhou Q, Lia J, Zhang X (2018) Adsorption of SF6 decomposition components on Pt3TiO2(1 0 1) surface: A DFT study. Appl Surf Sci 459:242
 34.
Liu D, Gui Y, Ji C, Tang C, Zhou Q, Li J, Zhang X (2019) Adsorption of SF6 decomposition components over Pd (1 1 1): A density functional theory study. Appl Surf Sci 465:172
 35.
Wei H, Gui Y, Kang J, Wang W, Tang C (2018) A DFT Study on the Adsorption of H_{2}S and SO_{2} on Ni Doped MoS_{2} Monolayer. Nanomaterials 8:646
 36.
Subramanyam BVRS, Mahakul PC, Sa K, Raiguru J, Alam I, Das S, Mondal M, Subudhi S, Mahanandi P (2019) Improved stability and performance of organic photovoltaic cells by application of carbon nanostructures and PEDOT:PSS composites as additional transparent electrodes. Sol Energy 186:146
 37.
Wang H, Song Y, Cao Y, Yu H, Liang H, Peng F (2019) Facile Synthesis of Cobalt and Nitrogen Coordinated Carbon Nanotube as a HighPerformance Electrocatalyst for Oxygen Reduction Reaction in Both Acidic and Alkaline Media. ACS Sustain Chem Eng 7:10951
 38.
De Volder MFL, Tawfick SH, Baughman RH, Hart AJ (2019) Carbon Nanotubes: Present and Future Commercial Applications. Science 339:535
 39.
CarpenaNunez J, Boscoboinik JA, Saber S, Rao R, Zhong JQ, Maschmann MR, Kidambi PR, Dee NT, Zakharov DN, Hart AJ, Stach EA, Maruyama B (2019) Isolating the Roles of Hydrogen Exposure and Trace Carbon Contamination on the Formation of Active Catalyst Populations for Carbon Nanotube Growth. ACS Nano 13:8736
 40.
Hu M, Zhao Z, Tian F, Oganov AR, Wang Q, Xiong M, Fan C, Wen B, He J, Yu D, Wang HT, Xu B, Tian Y (2013) Compressed carbon nanotubes: A family of new multifunctional carbon allotropes. Sci Rep 3:1331
 41.
Hart M, Chen J, Michaelides A, Sella A, Shaffer MSP, Salzmann CG (2019) OneDimensional Pnictogen Allotropes inside SingleWall Carbon Nanotubes. Inorg Chem 58:15216
 42.
Jeon, Yoon J, Kim U, Lee C, Xiang R, Shawky A, Xi J, Byeon J, Lee HM, Choi M, Maruyama S, Matsuo Y (2019) HighPerformance SolutionProcessed DoubleWalled Carbon Nanotube Transparent Electrode for Perovskite Solar. Cells Adv Energy Mater 9:1901204
 43.
Perdew JP, Burke K, Ernzerhof M (1997) Generalized Gradient Approximation Made Simple. Phys Rev Lett 77:3865
 44.
Kresse G, Furthmüller J (1996) Efficient iterative schemes for ab initio totalenergy calculations using a planewave basis set. Phys Rev B 54:11169
 45.
Kresse G, Furthmüller J (1996) Efficiency of abinitio total energy calculations for metals and semiconductors using a planewave basis set. Comput Mater Sci 6:15
 46.
Monkhorst HJ, Pack JD (1976) Special points for Brillouinzone integration. Phys Rev B 13:5188
 47.
Yang B, Zhou H, Zhang X, Liu X, Zhao M (2017) Dirac cones and highly anisotropic electronic structure of supergraphyne. Carbon 113:40
 48.
Jo JY, Kim BG (2012) Carbon allotropes with triple bond predicted by firstprinciple calculation: Triple bond modified diamond and Tcarbon. Phys. Rev B 86:075151
 49.
Xu LH, Hu JQ, Zhang JH, Wu SQ, Chuang FC, Zhu ZZ, Ho KM (2019) Structural and electronic properties of T graphene nanotubes: a firstprinciples study. New J Phys 21:053015
 50.
Charlier JC, Blase X, Roche S (2007) Electronic and transport properties of nanotubes Rev Mod Phys 79:677
Acknowledgements
The authors are grateful to all of the sponsors.
Funding
This work was supported by the National Key R&D Program of China under grant No. 2016YFA0202601 and 2016YFB0901502. Financial support from the program of China Scholarships Council (No.201806310018).
Author information
Affiliations
Contributions
LHX proposed the work and carried out the computation. LHX and ZZZ wrote the manuscript. LHX, ZZZ, SQW, FCC, and ZQH analyzed the results. ZZZ and FZ revised the paper. KMH participated in the design of the study. All authors read and approved the final manuscript.
Corresponding authors
Correspondence to LinHan Xu or ZiZhong Zhu.
Ethics declarations
Competing Interests
The authors declare that they have no competing interests.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
About this article
Cite this article
Xu, L., Wu, S., Huang, Z. et al. HOT Graphene and HOT Graphene Nanotubes: New Low Dimensional Semimetals and Semiconductors. Nanoscale Res Lett 15, 56 (2020). https://doi.org/10.1186/s1167102032791
Received:
Accepted:
Published:
Keywords
 HOT graphene
 Nanotubes
 Electronic structures
 Firstprinciples calculations