betaFeedback

WorksPhysics and Astronomy

Theory of Chiral Transport in Carbon Nanotubes

Abstract

Based on the similarity between the chiral nanotube and the classical solenoid, we study chiral transport along the circumferential direction in a carbon nanotube. We calculate the chiral conductivity, representing a circumferential current induced by an electric field along the nanotube axis for various chiralities of carbon nanotubes. We find that the chiral conductivity in a chiral nanotube is in general non zero, and oscillates as a function of the Fermi energy. This oscillating behavior is attributed to the subband structure of the nanotubes and the warping of the Fermi surfaces.

Introduction

Carbon nanotubes[1, 2] have been attracting much interest for their unique properties and for their possibilities of applications for electronic devices, such as a field-emission display[3] and nanocomposite materials,[4] nanosensors.[5] Its novel transport properties has been revealed experimentally. For example, quantized conductance is measured,[6] and the ballistic conduction for both single[7] and multi-wall[8] carbon nanotubes are reported. Theoretically, a carbon nanotube can be either a metal or a semiconductor, depending on its chirality. This property is understood from the band structure of a graphene and periodic boundary conditions along the circumference direction. Because of the singular band structure and unique structure of a carbon nanotube, its transport properties are expected to be intriguing.[9, 10, 11, 12]

In a chiral carbon nanotube, lower crystallographic symmetry allows chiral transport, i.e. a current in the circumferential direction induced by an electric field along the tube. Namely, when carriers are doped, an electric current along the tube axis induces a current around the tube. In a previous study, the chiral transport along the circumferential direction has been suggested in a BC2N\mathrm{BC}_{2}\mathrm{N} nanotube theoretically,[9] although the BC2N\mathrm{BC}_{2}\mathrm{N} nanotube has not been fabricated yet.[13] In Ref.,[9] both first-principles and tight-binding calculations are performed for the BC2N\mathrm{BC}_{2}\mathrm{N} nanotube, and they predicted that an electric field in BC2N\mathrm{BC}_{2}\mathrm{N} nanotube induces chiral transport, analogous to a solenoid. In Ref.,[14] the chiral conductivity is calculated for a BC2N\mathrm{BC}_{2}\mathrm{N} nanotube and a carbon nanotube and is shown to be non-zero. Here dependence on the chirality and on the doping are not explicitly shown. For the BN nanotube, the dynamical current directing the chiral shift has been calculated. However, the dependence on the chiral angle and the electronic concentration has not been studied.[15] Furthermore, chiral current is investigated for metallic carbon nanotubes[16] and stretched carbon nanotubes.[17] Experimental detection of the chiral current is also proposed.[18] The oscillation of the chiral currents in a carbon nanotube as a function of the Fermi energy is reported, in which the giant magnetization is expected.[19] Moreover, chiral transport in three-dimensional chiral crystals has been also proposed recently,[20, 21] as well as in three-dimensional topological insulators[22] and in two-dimensional surface and interfaces without inversion symmetry.[23]

In the present paper, we study dependence of the chiral current on the chirality of the carbon nanotubes. We calculate the chiral conductivity in carbon nanotubes, which is given by the ratio between the applied electric field along the nanotube axis and the induced current around the tube. As a function of the Fermi energy, the chiral conductivity is shown to oscillate with kinks. ThMoreover, its sign depends on m−nm-n modulo 3, where (m,n)(m,n) is the chirality of the nanotube. This singular behavior is interpreted in terms of the subband structure of the nanotube, and warping of the Fermi surfaces is crucial here. Thus, it shows that the carbon nanotube is a unique stage for chiral transport, where chiral nature can be controlled in a unique way out of the same structure.

We first consider symmetry requirements to have chiral transport. When a current JJ flows along the tube axis, a chiral current is induced and a magnetization MM along the tube will be induced. Thus the chiral response within the linear response is described as M=CJM=CJ with constant CC, and JJ being the current along the tube axis. This response is called kinetic magnetoelectric effect and orbital Edelstein effect,[20, 21, 24, 25, 26, 27, 28, 29, 30, 31, 32] and has been studied in the three-dimensional bulk metals[20, 21] and in topological insulators.[22] This effect is an orbital analog of the spin Edelstein effect.[33, 34, 35] We will see in the following that the carbon nanotubes are special in that it contains various structures with various chiralities, based on the same two-dimensional graphene sheet. Then we can see that the system should simultaneously break three types of symmetries: (i) inversion symmetry, (ii) mirror symmetry with respect to the plane perpendicular to the tube axis, and (iii) mirror symmetry with respect to the plane including the tube axis, in order to realize chiral transport induced by a current along the tube.

The response coefficient CC vanishes when the system is invariant under either of these operations. In the three types of carbon nanotubes, i.e. armchair, zigzag and chiral ones, the symmetries are different as summarized in Table 1. We see that only chiral nanotubes break all the three types of symmetries. in the same way as in classical solenoids. Thus, chiral carbon nanotubes are candidates for a chiral transport, similar to a classical solenoid. We note that in some three-dimensional chiral materials, similar response has been discussed. In carbon nanotubes, we will see that based on a same bulk band structure of graphene, one can have varieties of chiral structures, and we can systematically study various chiral effects.

(i)(ii)(iii)
armchair✓✓✓
zigzag✓✓✓
chiral

Table 1. Symmetries of three types of the carbon nanotubes. (i) inversion symmetry, (ii) mirror symmetry w.r.t. the plane including the tube axis, and (iii) mirror symmetry w.r.t. the plane perpendicular to the tube axis.

We here explain the chirality of the the carbon nanotube. Figure 1(a)(Figure 1b) show the chiral vector Ch\mathbf{C}_{h} and the translational vector T\mathbf{T} for the chirality (4,2)(4,2). The chiral vector Ch\mathbf{C}_{h} is a translation vector in graphene, which constitutes the perimeter of the nanotube. The chiral vector Ch\mathbf{C}_{h} is written as Ch=OA→=na1+ma2\mathbf{C}_{h}=\overrightarrow{OA}=n\mathbf{a}_{1}+m\mathbf{a}_{2}, with integers mm and nn, where a1=a(1,0)\mathbf{a}_{1}=a(1,0) and a2=a(1/2,3/2)\mathbf{a}_{2}=a(1/2,\sqrt{3}/2) are translation vectors in graphene with lattice constant a=0.246 nma=0.246\,\mathrm{nm}.[36, 37, 38] Chirality of the carbon nanotube is then labeled by index (n,m)(n,m). The vector T\mathbf{T}, which is a primitive translation vector of the nanotube, is expressed as

Structure of carbon nanotubes and their wavevectors

Figure 1. Structure of carbon nanotubes and their wavevectors. (a) Chiral vector ChC_h and the translational vector T\mathbf{T} for a carbon nanotube. (a) illustrates the case with the chirality (4,2). (b) The vectors ChC_h and T\mathbf{T} in the nanotube. (c) First Brillouin-zone for the (4,2) carbon nanotube is represented in blue. The gray hexagon represents the first Brillouin zone in graphene. The wavevector along the K2\mathbf{K}_2 is quantized, which is shown as a set of short segments shown in red. For convenience, we re-express the allowed wavevectors in (d) as a set of parallel segments inside the Brillouin zone of the graphene.

T=OB→=t1a1+t2a2  with  t1=−(2m+n)/dR, t2=(2n+m)/dR\mathbf{T}=\overrightarrow{OB}=t_{1}\mathbf{a}_{1}+t_{2}\mathbf{a}_{2}\ \text{ with }\ t_{1}=-(2m+n)/d_{R},\ t_{2}=(2n+m)/d_{R}

and dRd_{R} is the least common multiple of 2m+n2m+n and 2n+m2n+m. The number of hexagon NN, contained in unit cell described by the rectangle spanned by Ch\mathbf{C}_{h} and T\mathbf{T} in Figure 1(a) is given by N=2(n2+m2+nm)/dRN=2(n^{2}+m^{2}+nm)/d_{R}. The chiral angle θc\theta_{c}, which is an angle between ChC_h and a1\mathbf{a}_1, is given as θc=tan⁡−1[3m/(2n+m)]\theta_c=\tan^{-1}\left[\sqrt{3}m/(2n+m)\right]. The zigzag, armchair and chiral carbon nanotubes can be distinguished by the chiral angle. The chiral angle is given by θc=0\theta_c=0, π/3\pi/3, θc=π/6\theta_c=\pi/6 and θc≠0,π/3,π/6\theta_c\ne0,\pi/3,\pi/6, for zigzag, armchair and chiral carbon nanotubes, respectively.

In the carbon nanotube, the wave number kk along the circumferential direction is quantized. [36, 37, 38] Namely, the wave vector in the carbon nanotube is expanded in terms of K1\mathbf{K}_1 and K2\mathbf{K}_2 as

k=k1K1+k2K2∣K2∣,(1)\mathbf{k}=k_1\mathbf{K}_1+k_2\frac{\mathbf{K}_2}{\lvert\mathbf{K}_2\rvert}, \tag*{(1)}

with k1k_1 quantized as

k1=0,1,2,⋯ ,N−1,−π∣T∣≤k2<π∣T∣.(2)k_1=0,1,2,\cdots,N-1,\quad-\frac{\pi}{\lvert\mathbf{T}\rvert}\le k_2<\frac{\pi}{\lvert\mathbf{T}\rvert}. \tag*{(2)}

K1\mathbf{K}_1 and K2\mathbf{K}_2 are the reciprocal vectors of the carbon nanotube corresponding to T\mathbf{T} and ChC_h, respectively, satisfying

Ch⋅K1=2π,Ch⋅K2=0,T⋅K1=0,T⋅K2=2π.(3)C_h\cdot\mathbf{K}_1=2\pi,\quad C_h\cdot\mathbf{K}_2=0,\quad\mathbf{T}\cdot\mathbf{K}_1=0,\quad\mathbf{T}\cdot\mathbf{K}_2=2\pi. \tag*{(3)}

Namely, K1\mathbf{K}_1 and K2\mathbf{K}_2 are given by K1=(−t2b1+t1b2)/N\mathbf{K}_1=(-t_2\mathbf{b}_1+t_1\mathbf{b}_2)/N and K2=(mb1−nb2)/N\mathbf{K}_2=(m\mathbf{b}_1-n\mathbf{b}_2)/N, with b1=2π(1,1/3)\mathbf{b}_1=2\pi(1,1/\sqrt{3}) and b2=(0,4π/3)\mathbf{b}_2=(0,4\pi/\sqrt{3}). As an example, the first Brillouin zone of the carbon nanotube for the chirality (4, 2) is shown in blue in Figure 1(c).

The energy dispersion in the carbon nanotube can be described from that in graphene. Namely, the energy band in the carbon nanotube is given by that in graphene with the wave vector quantized along the K1\mathbf{K}_1 direction. In this paper, we use the tight-binding model with a nearest-neighbor hopping, [9, 39, 40, 41]

H=−∑⟨i,j⟩tai†aj,(4)H=-\sum_{\langle i,j\rangle}t a_i^\dagger a_j, \tag*{(4)}

where tt is the hopping between the sites ii and jj, and ⟨⋯ ⟩\langle\cdots\rangle means a summation over nearest-neighbor sites, with hopping integral t∼3.0eVt\sim3.0\mathrm{eV} for the tight-binding model. The energy dispersion in this system is given by

Ek±=±t1+4cos⁡2akx2+4cos⁡akx2cos⁡aky2.(5)E_{\mathbf{k}}^\pm=\pm t\sqrt{1+4\cos^2\frac{ak_x}{2}+4\cos\frac{ak_x}{2}\cos\frac{ak_y}{2}}. \tag*{(5)}

Here, we note that the energy dispersion is exactly the same as that in graphene, except the fact that the wave number is discretized in the carbon nanotube. The Fermi energy is set at EF=0E_F=0. It is known that the carbon nanotube is metallic when the wave vector crosses the Dirac points at KK and K′K' points in graphene which corresponds to (n−m)≡0(mod3)(n-m)\equiv0\pmod{3}, and semiconducting for the other chiralities. [11]

Results and discussion

We calculate the chiral conductivity when an electronic field is applied along the axis. With the Boltzmann transport equation, the conductivity is given by

σij=−e2∑kBZvivj∂f(Ek)∂Ekτ,i,j=x,y,(6)\sigma_{ij}=-e^2\sum_{\mathbf{k}}^{\mathrm{BZ}}v_i v_j\frac{\partial f(E_{\mathbf{k}})}{\partial E_{\mathbf{k}}}\tau,\quad i,j=x,y, \tag*{(6)}

where f(Ek)f(E_{\mathbf{k}}) is the Fermi distribution function, and the summation is taken over the allowed wavevectors within the Brillouin zone (BZ), v=(vx,vy)\mathbf{v}=(v_x,v_y) is the velocity of an electron, and τ\tau is the relaxation time assumed to be constant. By using vi=ℏ−1dEk/dkiv_i=\hbar^{-1}dE_{\mathbf{k}}/dk_i and partial integration, at zero temperature, we obtain

σij=e2τℏ2∑k1∫dk2∂2Ek∂ki∂kj,(7)\sigma_{ij}=\frac{e^2\tau}{\hbar^2}\sum_{k_1}\int dk_2\frac{\partial^2 E_{\mathbf{k}}}{\partial k_i\partial k_j}, \tag*{(7)}

which is a convenient form for numerical calculation. The sum over k1k_1 in eq. (7) is taken over the discretized wave vector k1k_1 in eq. (1).

We can calculate the chiral conductivity σ12\sigma_{12} from σxx\sigma_{xx}, σxy\sigma_{xy} and σyy\sigma_{yy} in eq. (7), by rotating the coordinate axes from x,yx,y to 1,21,2 by the angle θc\theta_c. As a result, the chiral conductivity σ12\sigma_{12} is obtained as

\begin{pmatrix} \sigma_{11} & \sigma_{12}\\ \sigma_{21} & \sigma_{22} \end{pmatrix} = \begin{pmatrix} \cos\theta_c & \sin\theta_c\\ -\sin\theta_c & \cos\theta_c \end{pmatrix} \begin{pmatrix} \sigma_{xx} & \sigma_{xy}\\ \sigma_{yx} & \sigma_{yy} \end{pmatrix} \begin{pmatrix} \cos\theta_c & -\sin\theta_c\\ \sin\theta_c & \cos\theta_c \end{pmatrix}, \tag*{(8)}\end{pmatrix}

where 1 and 2 represent the circumferential direction and the direction along the tube, respectively. We calculate the chiral conductivity σ12\sigma_{12} by varying two parameters, the chirality (n,m)(n,m) and the electronic concentration nin_i. Because the band dispersion depends on chirality, the Fermi level is numerically determined for an electronic concentration nin_i for each carbon nanotube.

Figures 2 (a) and (b) shows the chiral conductivity σ12\sigma_{12} as a function of the chiral angle for the electronic concentration ni=0.001n_i=0.001, measured from the undoped value. We set the chirality as (8,m)(8,m) in (a) and (15,m)(15,m) in (b), where integer mm is changed from 0 to 8. Generally, the sign of the chiral conductivity σ12\sigma_{12} depends on mm through the value (n−m)≡0,1,2(mod3)(n-m)\equiv0,1,2\pmod{3}. Namely, for (n−m)≡1(n-m)\equiv1 (2) (mod3)\pmod{3}, the sign of σ12\sigma_{12} is positive (negative). For (n−m)≡0(mod3)(n-m)\equiv0\pmod{3} when the carbon nanotube becomes metallic, σ12\sigma_{12} is almost zero. We note that σ12\sigma_{12} becomes zero for m=0m=0 because the carbon nanotube is of the armchair type and chiral transport is prohibited by symmetry. Furthermore, ∣σ12∣|\sigma_{12}| tends to become larger as the chiral angle increases, with an exception of m=5,10m=5,10 for n=15n=15.

We found that σ12\sigma_{12} tend to become larger as the integers mm and nn in the chirality (m,n)(m,n) become larger in general, by calculating also for the chiralities (4,m)(4,m), (12,m)(12,m), (20,m)(20,m) and (30,m)(30,m), although not shown here. This is because the contribution from the Fermi surfaces to σ12\sigma_{12} becomes larger for large chirality. The density of states on the Fermi level increases as chirality increases, with some exceptions with accidental local minima of σ12\sigma_{12} such as those for m=5,10m=5,10 in Figure 2(b). We interpret these remarkable dependences of σ12\sigma_{12} on chirality in terms of the warping of the Fermi surfaces. For large doping, the warping effect is prominent and the contributions to σ12\sigma_{12} do not perfectly cancel each other in metallic carbon nanotube, as we discuss later.

Chiral conductivity $\sigma_{12}$ as a function of chiral angle and electron concentration

Figure 2. Chiral conductivity σ12\sigma_{12}. (a)(b) Chiral conductivity as a function of the chiral angle for ni=0.001n_i = 0.001. The chirality is set as (a) (8,m)(8,m) with m=0,1,…,8m=0,1,\ldots,8 and (b) (15,m)(15,m) with m=0,1,…,15m=0,1,\ldots,15. (c)(d) Chiral conductivity as a function of the electron concentration for (c) (8,6)(8,6) and (d) (20,16)(20,16) nanotubes.

Figures 2(c) and (d) shows σ12\sigma_{12} as a function of electronic concentration nin_i for 0≤ni≤0.20\leq n_i\leq0.2 for the chiralities (a) (8,6) and (b) (20,16). Generally, σ12\sigma_{12} oscillates as a function of nin_i, with sharp kinks where σ12\sigma_{12} changes between an increasing and a decreasing function of nin_i. As nin_i becomes even larger and approaches 1, which is not realistic but is studied only for a theoretical interest, the chiral conductivity converges to zero, because electrons occupy all the bands in the model, giving no carriers in the hole picture.

To explain the oscillation of σ12\sigma_{12}, we analyze the dependence of σ12\sigma_{12} on nin_i in the (8,6) carbon nanotube for 0≤ni≤0.20\leq n_i\leq0.2 in Figure 2(c). As an example, in Figure 3(a) we show the Fermi surface at ni=0.03n_i=0.03, where σ12\sigma_{12} monotonically increases. The rhombus in the figure is the unit cell for the reciprocal space of graphene, which is equivalent to the first Brillouin zone. The wavevector k\mathbf{k} is discretized along the K1\mathrm{K}_1 direction (see eq. (1)), shown as the parallel lines, giving rise to subband structures. We label each subband by the value of the integer k1k_1, and they are shown as the indices from 5 to 9 for the subbands near the Fermi level. There are two Fermi surfaces, near the K and the K′' points for the graphene. At ni=0.03n_i=0.03, the Fermi level crosses only the 7th and 8th lines. The chiral conductivity σ12\sigma_{12} in eq. (6) is attributed to the states at the Fermi energy. Here we focus on the Fermi surface near the K point, because the contribution from the states near the K point and that near the K′' point are the same by the time-reversal symmetry. The states on the Fermi level near the K point are represented by four intersection points between the Fermi surface and the allowed wavevectors specified by the parallel lines. On the other hand, for ni=0.055n_i=0.055, where the chiral conductivity σ12\sigma_{12} shows a peak in Figure 2 (c), the Fermi surface is illustrated in Figure 3 (a). In addition to the four intersection points on the 7th and the 8th lines in Figure 3 (a), the Fermi surface and the 6th line come in contact with each other in this case. To see how each band contributes to the chiral conductivity σ12\sigma_{12}, we show in Figure 3 (c) the contributions to σ12\sigma_{12} from each subband as a function of nin_i. We can see the negative contribution from the 6th band in Figure 3 (c) appears for ni>0.055n_i>0.055. Therefore, the sharp kink of σ12\sigma_{12} at ni=0.055n_i=0.055 is attributed to the negative contribution of the 6th band to σ12\sigma_{12} for ni>0.055n_i>0.055. As the electron concentration increases, the Fermi surface crosses the 7th, 8th, 6th, 9th and 5th bands consecutively, and they contribute to the chiral conductivity with various signs. This corresponds to the points of the sharp kinks in Figure 2 (c). Besides, the signs of the contributions changes alternately in the order, 7th, 8th, 6th, 9th and 5th, which explains the oscillation of σ12\sigma_{12} as a function of nin_i.

Singular behaviors of chiral conductivities from subband contributions

Figure 3. Singular behaviors of chiral conductivities from subband contributions. The Fermi surface of the (8,6)(8,6) carbon nanotube is shown for (a) ni=0.03n_i=0.03 and (b) ni=0.055n_i=0.055 in the first Brillouin zone in graphene. The lines are the discretized wave vectors of carbon nanotubes. The indexes 5-9 are subband indices showing the values of k1k_1. (c) The contribution from the ll-th band (l=5,6,7,8,9l=5,6,7,8,9) to σ12\sigma_{12} as a function of the electronic concentration nin_i for the chirality (8,6)(8,6). Each color and number correspond to those in (a) and (b).

Figure 2 (d) shows nin_i dependence of σ12\sigma_{12} for the chirality (20,16)(20,16). The range of the oscillation is larger, and the period of the oscillation is smaller than those for the chirality (8,6)(8,6).

To understand the reason for the singular behavior of σ12\sigma_{12}, we first assume the band structure around the KK point to be a perfect Dirac cone with perfect linear dispersion Ek=γ∣k−k0∣E_k = \gamma|\mathbf{k} - \mathbf{k}_0| where γ\gamma is a constant, and k0\mathbf{k}_0 is either the K or K’ point. Then σ12\sigma_{12} is completely zero for any chiralities and the electronic concentrations. Since the linear dispersion Ek=γ∣k−k0∣E_k = \gamma|\mathbf{k} - \mathbf{k}_0| makes the Fermi surface to be a circle without warping, there are always a pair of Fermi points in the same subband, where the values of v1v_1 are the same while the signs of v2v_2 are opposite (the points P and Q in Figure 4 (a)). Their contributions to σ12\sigma_{12} in eq. (6) completely cancels each other because of symmetry. In reality, the Fermi surfaces in carbon nanotubes are distorted from a circle and have a warping (Figure 4 (b)). For the Fermi surface with the warping, the chiral conductivity has a nonzero value in general, because the contributions from the Fermi points do not necessarily cancel each other.

Fermi surfaces without and with warping near the K point in graphene

Figure 4. Effect of the warping of the the Fermi surface onto the chiral conductivities. The Fermi surface (a) without and (b) with warping near the KK point in graphene. The lines describe the subbands of the carbon nanotube. The arrows normal to the Fermi surface denote the Fermi velocity, which can be decomposed into the circumferential velocity v1v_1 and the velocity along the tube v2v_2.

For n−m≡0(mod3)n - m \equiv0 \pmod{3}, one of the subbands crosses the Dirac point in graphene. In this case, the contributions cancel each other almost completely for the low electronic concentration such as ni=0.001n_i = 0.001, because the Fermi surface is almost symmetric with respect to the subband line going across the Dirac point. However, the contribution does not vanish completely each other for a higher electronic concentration, because of the large warping of the Fermi surface.

We now estimate the value of the current flowing along the circumferential direction. When the electric field is applied along the axis, the current per unit length along the circumferential direction j1j_1 is given by

j1=σ12σ22I22πR,(9)j_1 = \frac{\sigma_{12}}{\sigma_{22}}\frac{I_2}{2\pi R}, \tag*{(9)}

where I2I_2 is the current along the nanotube, and RR is the radius of the carbon nanotube. We note that σ22\sigma_{22} is the longitudinal conductivity along the tube axis, calculated from eq. (8). In Figure 5(a), we show the comparison of σ22\sigma_{22} and σ12\sigma_{12} in the (8, 6) carbon nanotube, where the diameter is R=0.47nmR = 0.47\mathrm{nm}. σ22\sigma_{22} increases with some oscillations as nin_i increases. σ22\sigma_{22} shows a moderate kink at the electronic concentrations where σ12\sigma_{12} shows sharp peaks and dips. This is because new subbands begin to contribute to σ12\sigma_{12} and σ22\sigma_{22}, as we increase the concentration, as we have seen in this paper. In Figure 5(b), we show the ratio between σ22\sigma_{22} and σ12\sigma_{12}, which represents a dimensionless figure of merit for the chiral transport. We note that the ratio becomes largest at a very low concentration, and that it oscillates, converging to zero for larger nin_i because σ22\sigma_{22} increases rapidly as the Fermi level increases. The positions of the peaks of the oscillation correspond to that of σ12\sigma_{12} and σ22\sigma_{22}. For a very low concentration region, the ratio takes a value ∼−0.15\sim-0.15 in this case, and becomes zero at ni=0n_i = 0. For the (8, 6) nanotube, the maximum current which can be applied to the carbon nanotube is I1max⁡×S=6.6×10−8AI^{\max}_1 \times S = 6.6 \times10^{-8}\mathrm{A},1 with the cross section of the (8, 6) nanotube S=6.6×10−17cm2S = 6.6 \times10^{-17}\mathrm{cm}^2. For σ12/σ22∼−1.5\sigma_{12}/\sigma_{22} \sim-1.5, j1j_1 is j1=−3.5A/cmj_1 = -3.5\mathrm{A/cm} from eq. (9), corresponding to 4.4 gauss. The (orbital) magnetization per unit length of the nanotube is estimated as M=j2S=−0.23×10−15A⋅cmM = j_2S = -0.23 \times10^{-15}\mathrm{A\cdot cm}.

The longitudinal conductivity $\sigma_{22}$ and the chiral conductivity $\sigma_{12}$ as functions of electron concentration

Figure 5. The longitudinal conductivity σ22\sigma_{22} and the chiral conductivity σ12\sigma_{12}. (a) σ12\sigma_{12} and σ22\sigma_{22} as functions of nin_i in the (8, 6) nanotube. (b) σ12/σ22\sigma_{12}/\sigma_{22} as a function of nin_i in the (8, 6) nanotube.

Concluding remarks

In conclusion, we studied chiral transport in carbon nanotubes. We calculated the chiral conductivity by using the tight-binding model and the Boltzmann transport equation for the electric field applied along the axis of the carbon nanotube. We found that the chiral conductivity σ12\sigma_{12} is nonzero for semiconducting chiral carbon nanotubes with doping. For a constant electronic concentration, its sign alternately changes as the chiral angle increases, and it becomes positive, negative and zero when for the cases with n−m≡1,2,0(mod3)n-m \equiv1, 2, 0(\mathrm{mod}3), respectively. As a function of the electron concentration, σ12\sigma_{12} oscillates. This is attributed to the different signs of the contributions from the individual subbands to σ12\sigma_{12}. As the electronic concentration increases, the Fermi surface crosses the subbands with positive and negative contributions to σ12\sigma_{12}, giving rise to a peculiar oscillatory dependence on the concentration. Finally, we have estimated the value of chiral current and magnetization induced by the chiral current.

This kind of chiral transport has been also studied in bulk metals and topological insulators in three dimensions. For example, in tellurium, this chiral transport has been evaluated by ab initio calculation[28] and a related experiment on current-induced magnetization has been performed.[26] Tellurium hss left-handed and right-handed crystals, which consist of one-dimensional chains in shapes of helices, and one can see its analogy with a classical solenoid easily. On the other hand, in chiral nanotubes, its chirality is not fixed but can switch between both directions by changing an electron concentration.

This effect induces an orbital magnetization in carbon nanotubes, and through the spin-orbit coupling it might also lead to spin magnetization, which might be one route to induce the chirality-induced spin selectivity (CISS) effect.[43, 44, 45, 46, 47]

Notes

The authors declare no competing financial interest.

Acknowledgement This work was supported by JSPS KAKENHI Grants No. JP30578216, No. 23740236, No.26287062, and No. 18H03678, by the JSPS-EPSRC Core-to-Core program ”Oxide Superspin”, by MEXT KAKENHI Grants No. 25103709, and No.26103006, and by Elements strategy Initiative to Form Core Research Center (TIES), from MEXT Grant Number JP-MXP0112101001.

References

  1. [1]Iijima, S. Helical microtubules of graphitic carbon. Nature 1991, 354, 56–58.DOI
  2. [2]Iijima, S.; Ichihashi, T. Single-shell carbon nanotubes of 1-nm diameter. Nature 1993, 363, 603–605.DOI
  3. [3]Liu, Z.; Jiao, L.; Yao, Y.; Xian, X.; Zhang, J. Aligned, Ultralong Single-Walled Carbon Nanotubes: From Synthesis, Sorting, to Electronic Devices. Advanced Materials 2010, 22, 2285–2310.DOI
  4. [4]Byrne, M. T.; Gun’ko, Y. K. Recent Advances in Research on Carbon Nanotube-Polymer Composites. Advanced Materials 2010, 22, 1672–1688.DOI
  5. [5]Allen, B.-L.; Kichambare, P.-D.; Star, A. Carbon Nanotube Field-Effect-Transistor-Based Biosensors. Advanced Materials 2007, 19, 1439–1451.
  6. [6]Kong, J.; Yenilmez, E.; Tombler, T. W.; Kim, W.; Dai, H.; Laughlin, R. B.; Liu, L.; Jayanthi, C. S.; Wu, S. Y. Quantum Interference and Ballistic Transmission in Nanotube Electron Waveguides. Phys. Rev. Lett. 2001, 87, 106801.
  7. [7]Javey, A.; Guo, J.; Wang, Q.; Lundstrom, M.; Dai, H. Ballistic carbon nanotube field-effect transistors. Nature 2003, 424, 654–657.pubmed.ncbi.nlm.nih.gov/12904787
  8. [8]Berger, Y., C.and Yi; Wang, Z. L.; de Heer, W. A. Multiwalled carbon nanotubes are ballistic conductors at room temperature. Appl Phys A 2002, 74, 363–365.DOI
  9. [9]Miyamoto, Y.; Rubio, A.; Cohen, M. L.; Louie, S. G. Chiral tubules of hexagonal BC₂N. Phys. Rev. B 1994, 50, 4976–4979.pubmed.ncbi.nlm.nih.gov/9976827
  10. [10]Ando, T. Theory of transport in carbon nanotubes. Semiconductor Science and Technology 2000, 15, R13–R27.DOI
  11. [11]Ando, T.; Nakanishi, T. Impurity Scattering in Carbon Nanotubes - Absence of Back Scattering -. Journal of the Physical Society of Japan 1998, 67, 1704–1713.DOI
  12. [12]Nakanishi, T.; Ando, T. Numerical Study of Impurity Scattering in Carbon Nanotubes. Journal of the Physical Society of Japan 1999, 68, 561–566.DOI
  13. [13]Ayala, P.; Arenal, R.; Loiseau, A.; Rubio, A.; Pichler, T. The physical and chemical properties of heteronanotubes. Rev. Mod. Phys. 2010, 82, 1843–1885.DOI
  14. [14]Miyamoto, Y.; Louie, S. G.; Cohen, M. L. Chiral Conductivities of Nanotubes. Phys. Rev. Lett. 1996, 76, 2121–2124.pubmed.ncbi.nlm.nih.gov/10060611
  15. [15]Král, P.; Mele, E. J.; Tománek, D. Photogalvanic Effects in Heteropolar Nanotubes. Phys. Rev. Lett. 2000, 85, 1512–1515.DOI
  16. [16]Tsuji, N.; Takajo, S.; Aoki, H. Large orbital magnetic moments in carbon nanotubes generated by resonant transport. Phys. Rev. B 2007, 75, 153406.DOI
  17. [17]Miyamoto, Y. Mechanically stretched carbon nanotubes: Induction of chiral current. Phys. Rev. B 1996, 54, R11149–R11152.pubmed.ncbi.nlm.nih.gov/9985008
  18. [18]Miyamoto, Y.; Rubio, A.; Louie, S. G.; Cohen, M. L. Self-inductance of chiral conducting nanotubes. Phys. Rev. B 1999, 60, 13885–13889.DOI
  19. [19]Lambert, C. J.; Bailey, S. W. D.; Cserti, J. Oscillating chiral currents in nanotubes: A route to nanoscale magnetic test tubes. Phys. Rev. B 2008, 78, 233405.DOI
  20. [20]Yoda, T.; Yokoyama, T.; Murakami, S. Current-induced Orbital and Spin Magnetizations in Crystals with Helical Structure. Sci. Rep. 2015, 5, 12024.
  21. [21]Yoda, T.; Yokoyama, T.; Murakami, S. Orbital Edelstein Effect as a Condensed-Matter Analog of Solenoids. Nano Letters 2018, 18, 916–920, PMID: 29373028.DOI
  22. [22]Osumi, K.; Zhang, T.; Murakami, S. Orbital Edelstein effect in topological insulators. 2021; arXiv:2101.01869.
  23. [23]Hara, D.; Bahramy, M. S.; Murakami, S. Current-induced orbital magnetization in systems without inversion symmetry. Phys. Rev. B 2020, 102, 184404.DOI
  24. [24]Shalygin, V. A.; Sofronov, A. N.; Vorob’ev, L. E.; Farbshtein, I. I. Current-induced spin polarization of holes in tellurium. Physics of the Solid State 2012, 54, 2362–2373.DOI
  25. [25]Koretsune, T.; Arita, R.; Aoki, H. Magneto-orbital effect without spin-orbit interactions in a noncentrosymmetric zeolite-templated carbon structure. Phys. Rev. B 2012, 86, 125207.DOI
  26. [26]Furukawa, T.; Shimokawa, Y.; Kobayashi, K.; Itou, T. Observation of current-induced bulk magnetization in elemental tellurium. Nat. Commun. 2017, 8, 954.DOI
  27. [27]Zhong, S.; Moore, J. E.; Souza, I. Gyrotropic Magnetic Effect and the Magnetic Moment on the Fermi Surface. Phys. Rev. Lett. 2016, 116, 077201.DOI
  28. [28]Tsirkin, S. S.; Puente, P. A.; Souza, I. Gyrotropic effects in trigonal tellurium studied from first principles. Phys. Rev. B 2018, 97, 035158.DOI
  29. [29]Furukawa, T.; Watanabe, Y.; Ogasawara, N.; Kobayashi, K.; Itou, T. Chirality-Induced Electrical Generation of Magnetism in Nonmagnetic Elemental Tellurium. 2020; arXiv:2010.09210.DOI
  30. [30]Rou, J.; Şahin, C.; Ma, J.; Pesin, D. A. Kinetic orbital moments and nonlocal transport in disordered metals with nontrivial band geometry. Phys. Rev. B 2017, 96, 035120.DOI
  31. [31]Şahin, C.; Rou, J.; Ma, J.; Pesin, D. A. Pancharatnam-Berry phase and kinetic magnetoelectric effect in trigonal tellurium. Phys. Rev. B 2018, 97, 205206.DOI
  32. [32]Hara, D.; Bahramy, M. S.; Murakami, S. Current-induced orbital magnetization in systems without inversion symmetry. Phys. Rev. B 2020, 102, 184404.DOI
  33. [33]Edelstein, V. Spin polarization of conduction electrons induced by electric current in two-dimensional asymmetric electron systems. Solid State Commun. 1990, 73, 233–235.DOI
  34. [34]Ivchenko, E. L.; Pikus, G. E. New photogalvanic effect in gyrotropic crystals. ZhETF Pisma Redaktsiiu 1978, 27, 640.
  35. [35]Levitov, L. Nazarov Yu. V. and Eliashberg G. M. Sov. Phys. JETP 1985, 61, 133.
  36. [36]Saito, R.; Dresselhaus, G.; Dresselhaus, M. S. Physical Properties of Carbon Nanotubes; Published by Imperial College Press and distributed by World Scientific Publishing Co., 1998.
  37. [37]Dresselhaus, M.; Dresselhaus, G.; Eklund, P. In Science of Fullerenes and Carbon Nanotubes; Dresselhaus, M., Dresselhaus, G., Eklund, P., Eds.; Academic Press: San Diego, 1996.
  38. [38]Ajiki, H.; Ando, T. Electronic States of Carbon Nanotubes. Journal of the Physical Society of Japan 1993, 62, 1255–1266.DOI
  39. [39]Hamada, N.; Sawada, S.-I.; Oshiyama, A. New one-dimensional conductors: Graphitic microtubules. Phys. Rev. Lett. 1992, 68, 1579–1581.pubmed.ncbi.nlm.nih.gov/10045167
  40. [40]Saito, R.; Fujita, M.; Dresselhaus, G.; Dresselhaus, M. S. Electronic structure of graphene tubules based on C₆₀. Phys. Rev. B 1992, 46, 1804–1811.pubmed.ncbi.nlm.nih.gov/10003828
  41. [41]Saito, R.; Fujita, M.; Dresselhaus, G.; Dresselhaus, M. S. Electronic structure of chiral graphene tubules. Applied Physics Letters 1992, 60, 2204–2206.DOI
  42. [43]Naaman, R.; Waldeck, D. H. Chiral-Induced Spin Selectivity Effect. The Journal of Physical Chemistry Letters 2012, 3, 2178–2187.DOI
  43. [44]Gohler, B.; Hamelbeck, V.; Markus, T. Z.; Kettner, M.; Hanne, G. F.; Vager, Z.; Naaman, R.; Zacharias, H. Spin Selectivity in Electron Transmission Through Self-Assembled Monolayers of Double-Stranded DNA. Science 2011, 331, 894–897.DOI
  44. [45]Kettner, M.; Maslyuk, V. V.; Nürenberg, D.; Seibel, J.; Gutierrez, R.; Cuniberti, G.; Ernst, K.-H.; Zacharias, H. Chirality-Dependent Electron Spin Filtering by Molecular Monolayers of Helicenes. The Journal of Physical Chemistry Letters 2018, 9, 2025–2030.
  45. [46]Bloom, B. P.; Kiran, V.; Varade, V.; Naaman, R.; Waldeck, D. H. Spin Selective Charge Transport through Cysteine Capped CdSe Quantum Dots. Nano Letters 2016, 16, 4583–4589.DOI
  46. [47]Dor, O. B.; Morali, N.; Yochelis, S.; Baczewski, L. T.; Paltiel, Y. Local Light-Induced Magnetization Using Nanodots and Chiral Molecules. Nano Letters 2014, 14, 6042–6049.DOI

Paper details

vlnc:W_d678q2kr
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents