betaFeedback

WorksPhysics and Astronomy

Enhancement of spin-orbit coupling in Dirac semimetal Cd3_{3}As2_{2} films by Sb-doping

Abstract

We present a study on magnetotransport in films of the topological Dirac semimetal Cd3_{3}As2_{2} doped with Sb grown by molecular beam epitaxy. In our weak antilocalization analysis, we find a significant enhancement of the spin-orbit scattering rate, indicating that Sb doping leads to a strong increase of the pristine band-inversion energy. We discuss possible origins of this large enhancement by comparing Sb-doped Cd3_{3}As2_{2} with other compound semiconductors. Sb-doped Cd3_{3}As2_{2} will be a suitable system for further investigations and functionalization of topological Dirac semimetals.

Topologically non-trivial phases are commonly characterized by band inversion induced by large spin-orbit coupling[1, 2, 3]. As seen in the field of topological insulators, tuning the band inversion through modulation of the spin-orbit coupling has advanced the understanding such as of topological phase transitions[4, 5]. A topological Dirac semimetal, which is characterized by stable pairs of bulk Dirac dispersions in the presence of band inversion, is regarded as an ideal platform to study a variety of topological phases[6, 7, 8, 9]. In particular, tuning the band inversion in topological Dirac semimetals does not only serve as a source of various topological phase transitions but also allows to control the separation of the Dirac points in the momentum space, leading to exotic quantum transport phenomena. Moreover, the quantum spin Hall insulator phase in quantum confinement[8] can be stabilized by increasing the band-inversion energy. These few examples already suggest that gaining control over the tunability of the spin-orbit coupling is strongly desired for investigations of topological Dirac semimetals.

Cd3As2\mathrm{Cd}_3\mathrm{As}_2 is prototypical topological Dirac semimetal[10, 11, 12, 13, 14, 15], exhibiting a simple band structure, where only Cd 5s5s and As 4p4p bands, which form the Dirac dispersions, exist over a wide energy range around the Fermi level, as shown in Figure 1(a). As an example of exotic quantum transport, quantum Hall states mediated by both the surface and bulk states[16, 17] have been reported in this compound. Although chemical doping has been considered effective for tuning the band inversion as in the cases of topological insulators, it has been limited to Zn- and P-doping because of the high volatility of Cd3As2\mathrm{Cd}_3\mathrm{As}_2 itself[18, 19]. Also in high-crystallinity Cd3As2\mathrm{Cd}_3\mathrm{As}_2 films obtained by solid-phase epitaxy[13, 17, 20, 21], Zn-doping has been demonstrated to successfully modulate the quantum transport and to induce a topological phase transition to a trivial insulator simply by reducing the spin-orbit coupling[17, 22, 23]. Unfortunately, however, solid-phase epitaxy does not work on other dopant elements, and so far, chemical doping efficiently enhancing the spin-orbit coupling has not been reported for Cd3As2\mathrm{Cd}_3\mathrm{As}_2.

Schematic band-structure change induced by Sb doping in Cd3As2 and characterization of a Cd3(As0.85Sb0.15)2 film

Figure 1. (a) Schematic band-structure change induced by Sb doping in Cd3As2\mathrm{Cd}_{3}\mathrm{As}_{2}. The band inversion is expected to be enhanced by Sb doping which exhibits a larger spin-orbit coupling than As. Electron carriers caused by As and Sb deficiency are also expected to be reduced. (b) Cross-section HAADF-STEM image and (c) XRD θ−2θ\theta- 2\theta scan of a Cd3(As0.85Sb0.15)2\mathrm{Cd}_{3}(\mathrm{As}_{0.85}\mathrm{Sb}_{0.15})_{2} film grown on a CdTe substrate. The asterisks denote peaks from the CdTe substrate. (d) XRD θ−2θ\theta- 2\theta scans magnified around the (4 4 8) Cd3(As1−ySby)2\mathrm{Cd}_{3}(\mathrm{As}_{1-y}\mathrm{Sb}_{y})_{2} film peak with different Sb concentrations yy. The Sb concentrations are estimated using Vegard’s law, which linearly interpolates the lattice constants between y=0y = 0 and 0.15 films determined by EDX analyses.

Doping Sb which exhibiting a large spin-orbit coupling is promising to overcome this issue. From the band-gap change measured in P-doped Cd3As2\mathrm{Cd}_3\mathrm{As}_2, 10 % Sb-doping is expected to increase the valence-band-maximum energy by 80 meV, judged from the band alignment of conventional compound semiconductors[19, 24]. In addition to enhancement of the band-inversion energy, lowering of the Fermi level is also expected, as shown in Figure 1(a). This is because the main origin of the nn-type carriers in Cd3As2\mathrm{Cd}_3\mathrm{As}_2 is the deficiency at the anion site (As and Sb), and since Sb is less volatile than As, a decrease of the carrier density can be expected. Considering that it is difficult to dope especially volatile atoms into Cd3As2\mathrm{Cd}_3\mathrm{As}_2 under thermal equilibrium, molecular beam epitaxy, which is thermally a non-equilibrium process as compared to solid-phase epitaxy, is expected more suitable to dope Sb into Cd3As2\mathrm{Cd}_3\mathrm{As}_2 effectively. For undoped Cd3As2\mathrm{Cd}_3\mathrm{As}_2, high-quality films have been grown also by molecular-beam epitaxy, and a quantum Hall state has been observed for the films with two-dimensional Fermi surface[25]. Quantum Hall states have also been observed for the films with thickness of three-dimensional Fermi surface, where involved magnetic orbits can be described neither by a simple two-dimensional nor by a three-dimensional electronic state[26]. In this letter, we report transport signatures of an enhancement of spin-orbit coupling in single-crystalline Cd3(As1−ySby)2\mathrm{Cd}_3(\mathrm{As}_{1-y}\mathrm{Sb}_y)_2 films grown by molecular-beam epitaxy. A systematic analysis of weak antilocalization in films with various Sb concentrations reveals that the spin-orbit scattering rate is dramatically enhanced by Sb doping.

Cd3(As1−ySby)2\mathrm{Cd}_3(\mathrm{As}_{1-y}\mathrm{Sb}_y)_2 films with yy ranging from 0.005 to 0.24 were grown by molecular-beam epitaxy, following the same procedure described in Ref.[26]. The single-crystalline (1 1 1)A CdTe substrate was heated at 160 °C during the growth. The molecular beams were simultaneously provided from conventional Knudsen cells containing 6N Cd and 6N Sb and a valved cracker source containing 7N5 As. The beam equivalent pressures were set to 1×10−41 \times10^{-4} Pa for Cd and As4_4 and 0.05 - 5×10−55 \times10^{-5} Pa for Sb. Transport measurements were performed by using the conventional four-terminal method on film samples of 55 mm ×1\times1 mm size.

Figure 1(b) shows a cross-sectional scanning transmission electron microscopy (STEM) image taken on a Cd3(As1−ySby)2\mathrm{Cd}_3(\mathrm{As}_{1-y}\mathrm{Sb}_y)_2 film with y=0.15y = 0.15. Cd3(As1−ySby)2\mathrm{Cd}_3(\mathrm{As}_{1-y}\mathrm{Sb}_y)_2 layers are clearly formed from the heterointerface to the CdTe substrate. An x-ray diffraction (XRD) θ\theta-2θ2\theta scan for the y=0.15y = 0.15 film in Figure 1(c) shows the reflections from the {1 1 2}\{1\ 1\ 2\} lattice planes without any impurity phases. As shown in Figure 1(d), XRD scans show a systematic peak shift with increasing yy, indicating that the lattice constant systematically increases upon Sb doping. The film thicknesses determined by atomic force microscopy are 186 nm, 156 nm, 82 nm, 52 nm, and 83 nm for y=0.005y = 0.005, 0.07, 0.12, 0.15, and 0.24 samples, respectively. These thickness variation is caused by difficulty of estimating accurate growth rate for each sample due to Sb doping effects. However, transport properties of the Cd3(As1−ySby)2\mathrm{Cd}_3(\mathrm{As}_{1-y}\mathrm{Sb}_y)_2 samples show systematic dependence on the Sb concentration as presented below, indicating that the thickness variation does not have a considerable effect on discussions.

Figure 2(a) shows longitudinal resistance RxxR_{xx} and Hall resistance RyxR_{yx}, taken on the y=y = 0.15 film with sweeping the magnetic field at 2 K. RyxR_{yx} has a positive slope and is slightly bent at low fields. This indicates that the conduction is dominated by holes (p∼1018 cm−3p \sim10^{18}\ \mathrm{cm}^{-3}) and that there is a small number of remnant electrons (n∼1015 cm−3n \sim10^{15}\ \mathrm{cm}^{-3}). The low-temperature carrier mobility and carrier density obtained for all the Cd3(As1−ySby)2\mathrm{Cd}_{3}(\mathrm{As}_{1-y}\mathrm{Sb}_{y})_{2} films are summarized in Figure 2(b) and Figure 2(c). As also confirmed from the Hall resistance in Figure 2(d) - Figure 2(i), the electron carrier density is systematically reduced with Sb doping. For y=0.12y = 0.12, the Fermi energy measured from the Dirac points is estimated from its carrier density to be about 40 meV, and the saddle point energy is estimated to be about 100 meV[19, 24], suggesting that the Fermi energy is located inside the band-inversion region. At y=0.15y = 0.15, the hole carriers become dominant. This indicates that the deficiency at As sites is greatly suppressed by Sb doping as expected, and eventually the system becomes pp type. As commonly observed in III-V and II-VI compound semiconductors, the valence band level is higher in heavier anions[24, 27]. Based on these observations in the conventional compound semiconductors, the valence band level of Cd3As2\mathrm{Cd}_{3}\mathrm{As}_{2} is also expected to be shifted higher by Sb doping, and hole carriers are considered to become easier to be induced as a result of the shallower valence band level.

Plots of longitudinal and Hall resistance, mobility versus carrier density, Hall resistance times thickness, and magnetoresistance ratio

Figure 2. (a) Longitudinal resistance RxxR_{xx} and Hall resistance RyxR_{yx} of Cd3(As0.85Sb0.15)2\mathrm{Cd}_{3}(\mathrm{As}_{0.85}\mathrm{Sb}_{0.15})_{2} film at 2 K. The magnetic field BB is applied out of plane. Mobility vs carrier density summarized for our Cd3(As1−ySby)2\mathrm{Cd}_{3}(\mathrm{As}_{1-y}\mathrm{Sb}_{y})_{2} films in (b) nn-type and (c) pp-type regions. (d)-(i) Hall resistance times thickness RyxtR_{yx}t and (j)-(n) magnetoresistance ratio for our Cd3(As1−ySby)2\mathrm{Cd}_{3}(\mathrm{As}_{1-y}\mathrm{Sb}_{y})_{2} films at 2 K. In (g) and (i), the curves are multiplied by ten for clarity.

Magnetoresistivity ratio MRR ≡[ρ(B)−ρ(0)]/ρ(0)=Δρ(B)/ρ(0)\equiv[\rho(B)-\rho(0)]/\rho(0) = \Delta\rho(B)/\rho(0) of the Cd3(As1−ySby)2\mathrm{Cd}_{3}(\mathrm{As}_{1-y}\mathrm{Sb}_{y})_{2} films at 2 K are shown in Figure 2(j) - Figure 2(n). Due to Sb doping, the shape of the MRR curves changes from parabolic-like in the lightly doped region to cusp-like in the heavily-doped region. This change indicates that the Sb doping suppresses the classical magnetoresistivity and enhances weak antilocalization effects as discussed below. The magnitude of the classical magnetoresistivity originating from the Lorentz force is expressed by

Δρorb(B)ρ(0)≈(μB)2,(1)\frac{\Delta\rho_{\mathrm{orb}}(B)}{\rho(0)} \approx(\mu B)^2, \tag*{(1)}

with the carrier mobility μ\mu[28], and, therefore, its contribution decreases when the carrier mobility is reduced.

The observed magnetoresistivity ρtotal\rho_{\mathrm{total}} can be expressed as ρtotal=ρorb+ρWAL\rho_{\mathrm{total}} = \rho_{\mathrm{orb}} + \rho_{\mathrm{WAL}}, where ρorb\rho_{\mathrm{orb}} is the classical magnetoresistivity and ρWAL\rho_{\mathrm{WAL}} is a quantum correction to the magnetoresistivity due to weak antilocalization. According to Kohler’s rule[29], Eq. (1) can be rewritten as

Δρorb(B)ρ(0)=k(Bρ(0))2(2)\frac{\Delta\rho_{\mathrm{orb}}(B)}{\rho(0)} = k\left(\frac{B}{\rho(0)}\right)^2 \tag*{(2)}

with a temperature-independent constant kk, as long as the carrier density is independent of temperature. In the present case, kk can be obtained by fitting the total magnetoresistivity at high temperatures (T∼50T \sim50 K), above which only Δρorb(B)\Delta\rho_{\mathrm{orb}}(B) is observed. By using this kk value, ΔρWAL(B)\Delta\rho_{\mathrm{WAL}}(B) at low temperatures can be derived by subtracting Δρorb(B)\Delta\rho_{\mathrm{orb}}(B) from Δρtotal(B)\Delta\rho_{\mathrm{total}}(B). Details about the subtraction of the classical magnetoresistivity can be found in the Supplementary Materials[30].

Figure 3(a) shows ΔρWAL(B)≡Δρtotal(B)−Δρorb(B)\Delta\rho_{\mathrm{WAL}}(B) \equiv\Delta\rho_{\mathrm{total}}(B) - \Delta\rho_{\mathrm{orb}}(B) at 2 K for all examined Cd3(As1−ySby)2\mathrm{Cd}_{3}(\mathrm{As}_{1-y}\mathrm{Sb}_{y})_{2} films. Before moving on to a detailed discussion of the weak antilocalization, we first check the dimensionality of the Cd3(As1−ySby)2\mathrm{Cd}_{3}(\mathrm{As}_{1-y}\mathrm{Sb}_{y})_{2} films. The dimensionality of weak localization and weak antilocalization is determined by comparing the film thickness tt to the dephasing length Lϕ=DτϕL_{\phi} = \sqrt{D\tau_{\phi}}, with the diffusion coefficient DD and the dephasing scattering time τϕ\tau_{\phi}. The dephasing process is dominated by inelastic scatterings including electron-phonon and electron-electron scattering[31]. When LϕL_{\phi} is smaller than tt, the quantum interference process causing the weak antilocalization is treated as three-dimensional (Figure 3(b)), and otherwise, as two-dimensional (Figure 3(c)). In the case of our Cd3(As1−ySby)2\mathrm{Cd}_{3}(\mathrm{As}_{1-y}\mathrm{Sb}_{y})_{2} films, we employed both two-dimensional and three-dimensional models to obtain LϕL_{\phi} and compare it with tt to determine the dimensionality of the weak antilocalization. For y=0.07−0.24y = 0.07 - 0.24, the experimental data can be well fitted by the three-dimensional weak antilocalization model[31, 32, 33, 34] as presented in Figure 3(d) - Figure 3(g), and LϕL_{\phi} obtained from this model is smaller than tt. This suggests that the weak antilocalization observed in these films can be treated as three-dimensional. On the other hand, for y=0.005y = 0.005, a narrow dip observed at low fields (B<40B < 40 mT) cannot be described by the three-dimensional model. Instead, it is well fitted by the two-dimensional weak antilocalization model[35, 36], and LϕL_{\phi} for the y=0.005y = 0.005 film obtained from the two-dimensional model satisfies Lϕ>tL_{\phi} > t, suggesting that the weak antilocalization observed in this film is of two-dimensional nature. Two-dimensional weak antilocalization has also been reported for undoped Cd3As2\mathrm{Cd}_{3}\mathrm{As}_{2} films previously[37]. In the following discussion, we focus on the three-dimensional weak antilocalization observed in y=0.07−0.24y = 0.07 - 0.24 films. See also the Supplementary Materials for further discussion on the dimensionality and the fitting of the y=0.005y = 0.005 film[30].

Figure 3. (a) Normalized magnetoresistivity ΔρWAL/ρ2(0)=[Δρtotal−Δρorb]/ρtotal2(0)\Delta\rho_{\mathrm{WAL}}/\rho^2(0) = [\Delta\rho_{\mathrm{total}} - \Delta\rho_{\mathrm{orb}}]/\rho_{\mathrm{total}}^2(0) of the Cd3(As1−ySby)2\mathrm{Cd}_3(\mathrm{As}_{1-y}\mathrm{Sb}_y)_2 films at 2 K. Trajectories of the scattered electrons for (b) three-dimensional and (c) two-dimensional weak antilocalization. (d)-(g) Temperature dependence of the normalized magnetoresistivity for y=0.07y = 0.07, 0.120.12, 0.150.15, and 0.240.24. The red curves are fits to the three-dimensional weak antilocalization model (Eq. (3)). The theory assumes that the magnetic field is within the range satisfying γ<1\gamma< 1 (see Eq. (3) in main text). The magnetoresistivity below 7 K is not shown for y=0.07y = 0.07 because Lϕ<tL_{\phi} < t is not satisfied at lower temperature (see also the Supplementary Material).

Three-dimensional weak localization and weak antilocalization including the effect of the spin-orbit coupling are expressed as[31, 32, 33, 34, 38]

ΔρWAL(B)ρ2(0)=e22π2ℏeBℏ(121−γ[f3(BB−)−f3(BB+)]−f3(BB2)−4BSO3B[11−γ(t+−t−)+t−t+1]),(3)\frac{\Delta\rho_{\mathrm{WAL}}(B)}{\rho^2(0)} = \frac{e^2}{2\pi^2\hbar}\sqrt{\frac{eB}{\hbar}}\left(\frac{1}{2\sqrt{1-\gamma}}\left[f_3\left(\frac{B}{B_-}\right)-f_3\left(\frac{B}{B_+}\right)\right]-f_3\left(\frac{B}{B_2}\right)-\sqrt{\frac{4B_{\mathrm{SO}}}{3B}}\left[\frac{1}{\sqrt{1-\gamma}}\left(\sqrt{t_+}-\sqrt{t_-}\right)+\sqrt{t}-\sqrt{t+1}\right]\right), \tag*{(3)}

where

γ=[3g∗μBB4eD(2BSO−B0)]2,\gamma= \left[\frac{3g^*\mu_B B}{4eD(2B_{\mathrm{SO}}-B_0)}\right]^2,
B±=Bϕ+13(2BSO−B0)(1±1−γ),B_{\pm}=B_{\phi}+\frac{1}{3}(2B_{\mathrm{SO}}-B_0)\left(1\pm\sqrt{1-\gamma}\right),
Bϕ=Bi+B0,B2=Bi+13B0+43BSO,B_{\phi}=B_i+B_0,\qquad B_2=B_i+\frac{1}{3}B_0+\frac{4}{3}B_{\mathrm{SO}},
t=3Bϕ2(BSO−B0),t±=t+12(1±1−γ),t=\frac{3B_{\phi}}{2(B_{\mathrm{SO}}-B_0)},\qquad t_{\pm}=t+\frac{1}{2}\left(1\pm\sqrt{1-\gamma}\right),
f3(z)=∑N=0∞[2(N+1+z−N+z)−1N+12+z].f_3(z)=\sum_{N=0}^{\infty}\left[2\left(\sqrt{N+1+z}-\sqrt{N+z}\right)-\frac{1}{\sqrt{N+\frac{1}{2}+z}}\right].

Here, g∗g^* is the Landé factor, μB\mu_B is the Bohr magneton, and D=μkBT/eD=\mu_{\mathrm{kB}}T/e is the diffusion constant. The characteristic fields BjB_j (j=ij=i, SO, and 0) are related to the characteristic scattering times through Bj=ℏ/4eDτjB_j=\hbar/4eD\tau_j, where the indices denote inelastic, spin-orbit, and residual scattering, respectively. The dephasing scattering time τϕ\tau_{\phi} is defined by

1τϕ(T)=1τ0+1τi(T),(4)\frac{1}{\tau_{\phi}(T)}=\frac{1}{\tau_0}+\frac{1}{\tau_i(T)}, \tag*{(4)}

where its temperature dependence is governed by τi\tau_i, and its low-temperature limit is determined by temperature-independent τ0\tau_0 [31]. In the spin-orbit scattering process, on the other hand, not only the momentum but also the spin direction of the scattered electrons is changed due to the spin-orbit interaction [39]. In the limit of 1/τi≫1/τSO1/\tau_i\gg1/\tau_{\mathrm{SO}}, Eq. (3) describes weak localization, and in the opposite limit, 1/τSO≫1/τi1/\tau_{\mathrm{SO}}\gg1/\tau_i, it describes weak antilocalization.

Temperature dependences of 1/τϕ1/\tau_{\phi} and 1/τSO1/\tau_{\mathrm{SO}} are shown in Figure 4(a) and Figure 4(b). An exponential temperature dependence of 1/τϕ1/\tau_{\phi} is confirmed for all films. The extracted exponents p=1.9−3.1p=1.9-3.1 are quite typical for three-dimensional systems in the diffusive regime. Similar exponents p∼2−4p\sim2-4 have been previously reported for conventional compound semiconductors [31, 40, 41].

Figure 4. (a) Dephasing scattering rate 1/τϕ1/\tau_{\phi} and (b) spin-orbit scattering rate 1/τSO1/\tau_{\mathrm{SO}} as a function of temperature for y=0.07y = 0.07, 0.120.12, 0.150.15, and 0.240.24. The horizontal bars on the right axis in (a) represent the temperature-independent residual scattering rate extracted from the fit at 7 K for y=0.07y = 0.07 and at 2 K for y=0.12y = 0.12, 0.150.15, and 0.240.24. (c) Spin-orbit scattering rate 1/τSO1/\tau_{\mathrm{SO}} vs dephasing scattering rate 1/τϕ1/\tau_{\phi} summarized for the Cd3(As1−ySby)2\mathrm{Cd}_{3}(\mathrm{As}_{1-y}\mathrm{Sb}_{y})_{2} films. Data taken at different temperatures as shown in (a) and (b) are plotted. For comparison, 1/τSO1/\tau_{\mathrm{SO}} vs 1/τϕ1/\tau_{\phi} previously reported for three-dimensional systems (● In1−xPbxSb\mathrm{In}_{1-x}\mathrm{Pb}_{x}\mathrm{Sb}[40], ▲ Pb1−xEuxTe\mathrm{Pb}_{1-x}\mathrm{Eu}_{x}\mathrm{Te}[41]) and two-dimensional systems (▽ Ga1−xInxAs\mathrm{Ga}_{1-x}\mathrm{In}_{x}\mathrm{As}[46], ◇ GaAs/Al1−xGaxAs\mathrm{GaAs}/\mathrm{Al}_{1-x}\mathrm{Ga}_{x}\mathrm{As}[47], ◁▷ InSb\mathrm{InSb}[48], □ Al1−xGaxN/GaN\mathrm{Al}_{1-x}\mathrm{Ga}_{x}\mathrm{N}/\mathrm{GaN}[49, 50]) are also plotted. 1/τi1/\tau_i (1/τϕ1/\tau_{\phi} at τ0→∞\tau_0 \to\infty) is shown for Pb1−xEuxTe\mathrm{Pb}_{1-x}\mathrm{Eu}_{x}\mathrm{Te} instead of 1/τϕ1/\tau_{\phi}.

1/τSO1/\tau_{\mathrm{SO}} is significantly enhanced by Sb doping, as shown in Figure 4(b). The spin-orbit scattering cross section of an impurity σSO\sigma_{\mathrm{SO}} has been previously studied theoretically and experimentally[42, 43, 44, 45], and it is expressed as

σSO∼1τSO∼∑l≥0l(l+1)2l+1ξl2nl2(EF),(5)\sigma_{\mathrm{SO}} \sim\frac{1}{\tau_{\mathrm{SO}}} \sim\sum_{l \ge0} \frac{l(l+1)}{2l+1}\xi_l^2 n_l^2(E_{\mathrm{F}}), \tag*{(5)}

where ll is the azimuthal quantum number for impurity atoms, ξl\xi_l is half of the spin-orbit constant[42], and nl(EF)n_l(E_{\mathrm{F}}) is the angular momentum resolved local density of states of the impurity atoms at the Fermi level. Eq. (5) suggests that 1/τSO1/\tau_{\mathrm{SO}} largely depends on the local density of pp states (l=1l = 1) of impurity at the Fermi level (nl=1(EF)n_{l=1}(E_{\mathrm{F}}))[42], and, therefore, the observed enhancement of 1/τSO1/\tau_{\mathrm{SO}} in Cd3_3(As1−y_{1-y}Sby_y)2_2 is considered a result of a larger nl=1(EF)n_{l=1}(E_{\mathrm{F}}) with an increased impurity density by Sb doping.

The relation between 1/τSO1/\tau_{\mathrm{SO}} and 1/τϕ1/\tau_{\phi} obtained for the Cd3_3(As1−y_{1-y}Sby_y)2_2 films is summarized in Figure 4(c), and compared with previously reported data on both three-dimensional and two-dimensional systems[40, 41, 46, 47, 48, 49, 50]. In particular, a large enhancement of 1/τSO1/\tau_{\mathrm{SO}} has been observed for the pp-type Cd3_3(As1−y_{1-y}Sby_y)2_2 films, and this may suggest that a similar enhancement can be also expected for conventional compound semiconductors where the following two conditions are satisfied: (i) the doping takes place at the anion sites and (ii) the Fermi level is tuned into the valence band. For the compound semiconductors shown in Figure 4(c), at least one of these conditions is not satisfied. It is further suggested that a large enhancement of 1/τSO1/\tau_{\mathrm{SO}} can be expected, for example, for pp-type GaAs1−x_{1-x}Sbx_x and pp-type InP1−x_{1-x}Asx_x, which satisfy the aforementioned two conditions but these samples have not been examined yet.

In summary, we have systematically investigated magnetotransport of Cd3_3(As1−y_{1-y}Sby_y)2_2 films with varying Sb concentrations y=0.005−0.24y = 0.005 - 0.24. Weak antilocalization is commonly observed in the Cd3_3(As1−y_{1-y}Sby_y)2_2 films and has been analyzed using a three-dimensional model. In the analysis, 1/τSO1/\tau_{\mathrm{SO}} is greatly enhanced upon increasing the Sb concentration. In particular, the Fermi energy of the y=0.12y = 0.12 film is estimated to be located inside the band-inversion region, based on the band alignment of conventional compound semiconductors. Our demonstration of a large enhancement of the spin-orbit coupling in Cd3_3(As1−y_{1-y}Sby_y)2_2 films paves the way for further investigations and functionalization of topological Dirac semimetals, including the realization of a robust quantum spin Hall insulator phase under confinement conditions as in HgTe/CdTe quantum wells[8, 51, 52]. Tuning a confinement-induced sub-band position and the Fermi energy inside the band-inversion region is considered more feasible with the increased band inversion in Cd3_3(As1−y_{1-y}Sby_y)2_2.

We thank N. Nagaosa and M. Tokunaga for fruitful discussions. We also thank M. Kriener for proofreading of the manuscript. This work was supported by JST PRESTO Grant No. JPMJPR18L2 and CREST Grant No. JPMJCR16F1, Japan, and by Grant-in-Aids for Scientific Research (B) No. JP18H01866 from MEXT, Japan.

References

References

  1. [1]M. Z. Hasan and C. L. Kane, Colloquium : Topological insulators, Rev. Mod. Phys. 82, 3045 (2010).
  2. [2]X.-L. Qi and S.-C. Zhang, Topological insulators and superconductors, Rev. Mod. Phys. 83, 1057 (2011).
  3. [3]N. P. Armitage, E. J. Mele, and A. Vishwanath, Weyl and Dirac semimetals in three-dimensional solids, Rev. Mod. Phys. 90, 015001 (2018).
  4. [4]H. Guo, K. Sugawara, A. Takayama, S. Souma, T. Sato, T. Satoh, A. Ohnishi, M. Kitaura, M. Sasaki, Q.-K. Xue, and T. Takahashi, Evolution of surface states in Bi₁₋ₓSbₓ alloys across the topological phase transition, Phys. Rev. B 83, 201104(R) (2011).DOI
  5. [5]S.-Y. Xu, Y. Xia., L. A. Wray, S. Jia, F. Meier, J. H. Dil, J. Osterwalder, B. Slomski, A. Bansil, H. Lin, R. J. Cava, and M. Z. Hasan, Topological Phase Transition and Texture Inversion in a Tunable Topological Insulator, Science 332, 560 (2011).DOI
  6. [6]S. M. Young, S. Zaheer, J. C. Teo, C. L. Kane, E. J. Mele, and A. M. Rappe, Dirac Semimetal in Three Dimensions, Phys. Rev. Lett. 108, 140405 (2012).
  7. [7]Z. Wang, Y. Sun, X.-Q. Chen, C. Franchini, G. Xu, H. Weng, X. Dai, and Z. Fang, Dirac semimetal and topological phase transitions in A₃Bi (A = Na, K, Rb), Phys. Rev. B 85, 195320 (2012).DOI
  8. [8]Z. Wang, H. Weng, Q. Wu, X. Dai, and Z. Fang, Three-dimensional Dirac semimetal and quantum transport in Cd₃As₂, Phys. Rev. B 88, 125427 (2013).
  9. [9]B.-J. Yang and N. Nagaosa, Classification of stable three-dimensional Dirac semimetals with nontrivial topology, Nat. Commun. 5, 4898 (2014).
  10. [10]Z. K. Liu, J. Jiang, B. Zhou, Z. J. Wang, Y. Zhang, H. M. Weng, D. Prabhakaran, S.-K. Mo, H. Peng, P. Dudin, T. Kim, M. Hoesch, Z. Fang, X. Dai, Z. X. Shen, D. L. Feng, Z. Hussain, and Y. L. Chen, A stable three-dimensional topological Dirac semimetal Cd₃As₂, Nat. Mater. 13, 677 (2014).
  11. [11]S. Jeon, B. B. Zhou, A. Gyenis, B. E. Feldman, I. Kimchi, A. C. Potter, Q. D. Gibson, R. J. Cava, A. Vishwanath, and A. Yazdani, Landau quantization and quasiparticle interference in the three-dimensional Dirac semimetal Cd₃As₂, Nat. Mater. 13, 851 (2014).DOI
  12. [12]P. J. W. Moll, N. L. Nair, T. Helm, A. C. Potter, I. Kimchi, A. Vishwanath, and J. G. Analytis, Transport evidence for Fermi-arc-mediated chirality transfer in the Dirac semimetal Cd₃As₂, Nature 535, 266 (2016).DOI
  13. [13]M. Uchida, Y. Nakazawa, S. Nishihaya, K. Akiba, M. Kriener, Y. Kozuka, A. Miyake, Y. Taguchi, M. Tokunaga, N. Nagaosa, Y. Tokura, and M. Kawasaki, Quantum Hall states observed in thin films of Dirac semimetal Cd₃As₂, Nat. Commun. 8, 2274 (2017).
  14. [14]C. Zhang, A. Narayan, S. Lu, J. Zhang, H. Zhang, Z. Ni, X. Yuan, Y. Liu, J.-H. Park, E. Zhang, W. Wang, S. Liu, L. Cheng, L. Pi, Z. Sheng, S. Sanvito, and F. Xiu, Evolution of Weyl orbit and quantum Hall effect in Dirac semimetal Cd₃As₂, Nat. Commun. 8, 1272 (2017).DOI
  15. [15]B.-C. Lin, S. Wang, S. Wiedmann, J.-M. Lu, W.-Z. Zheng, D. Yu, and Z.-M. Liao, Observation of an Odd-Integer Quantum Hall Effect from Topological Surface States in Cd₃As₂, Phys. Rev. Lett. 122, 036602 (2019).DOI
  16. [16]C. Zhang, Y. Zhang, X. Yuan, S. Lu, J. Zhang, A. Narayan, Y. Liu, H. Zhang, Z. Ni, R. Liu, E. S. Choi, A. Suslov, S. Sanvito, L. Pi, H.-Z. Lu, and A. C. , Quantum Hall effect based on Weyl orbits in Cd₃As₂, Nature 565, 331 (2019).
  17. [17]S. Nishihaya, M. Uchida, Y. Nakazawa, R. Kurihara, K. Akiba, M. Kriener, A. Miyake, Y. Taguchi, M. Tokunaga, and M. Kawasaki, Quantized surface transport in topological Dirac semimetal films, Nat. Commun. 10, 2564 (2019).DOI
  18. [18]H. Lu, X. Zhang, Y. Bian, and S. Jia, Topological Phase Transition in Single Crystals of (Cd₁₋ₓZnₓ)₃As₂, Sci. Rep. 7, 3148 (2017).DOI
  19. [19]S. Thirupathaiah, I. Morozov, Y. Kushnirenko, A. V. Fedorov, E. Haubold, T. K. Kim, G. Shipunov, A. Maksutova, O. Kataeva, S. Aswartham, B. Büchner, and S. V. Borisenko, Spectroscopic evidence of topological phase transition in the three-dimensional Dirac semimetal Cd₃(As₁₋ₓPₓ)₂, Phys. Rev. B 98, 085145 (2018).DOI
  20. [20]Y. Nakazawa, M. Uchida, S. Nishihaya, M. Kriener, Y. Kozuka, Y. Taguchi, and M. Kawasaki, Structural characterisation of high-mobility Cd₃As₂ films crystallised on SrTiO₃, Sci. Rep. 8, 2244 (2018).DOI
  21. [21]M. Uchida, T. Koretsune, S. Sato, M. Kriener, Y. Nakazawa, S. Nishihaya, Y. Taguchi, R. Arita, and M. Kawasaki, Ferromagnetic state above room temperature in a proximitized topological Dirac semimetal, Phys. Rev. B 100, 245148 (2019).DOI
  22. [22]S. Nishihaya, M. Uchida, Y. Nakazawa, K. Akiba, M. Kriener, Y. Kozuka, A. Miyake, Y. Taguchi, M. Tokunaga, and M. Kawasaki, Negative magnetoresistance suppressed through a topological phase transition in (Cd₁₋ₓZnₓ)₃As₂ thin films, Phys. Rev. B 97, 245103 (2018).DOI
  23. [23]S. Nishihaya, M. Uchida, Y. Nakazawa, M. Kriener, Y. Kozuka, Y. Taguchi, and M. Kawasaki, Gate-tuned quantum Hall states in Dirac semimetal (Cd₁₋ₓZnₓ)₃As₂, Sci. Adv. 4, ea r5668 (2018).DOI
  24. [24]Y. Hinuma, A. Grü neis, G. Kresse, and F. Oba, Band alignment of semiconductors from density- functional theory and many-body perturbation theory, Phys. Rev. B 90, 155405 (2014).
  25. [25]T. Schumann, L. Galletti, D. A. Kealhofer, H. Kim, M. Goyal, and S. Stemmer, Observation of the Quantum Hall Effect in Confined Films of the Three-Dimensional Dirac Semimetal Cd₃As₂, Phys. Rev. Lett. 120, 016801 (2018).
  26. [26]Y. Nakazawa, M. Uchida, S. Nishihaya, S. Sato, A. Nakao, J. Matsuno, and M. Kawasaki, Molecular beam epitaxy of three-dimensionally thick Dirac semimetal Cd₃As₂ films, APL Mater. 7, 071109 (2019).DOI
  27. [27]Chris G. Van de Walle and J. Neugebauer, Universal alignment of hydrogen levels in semicon- ductors, insulators and solutions, Nature 423, 626 (2003).
  28. [28]A. B. Pippard, Magnetoresistance in Metals (Cambridge University Press, 1989).
  29. [29]N. Luo and G. H. Miley, Kohler’s rule and relaxation rates in high-Tc superconductors, Physica C 371, 259 (2002).
  30. [30]See Supplementary Materials for more information about the analysis of magnetoresistance.
  31. [31]J. J. Lin and J. P. Bird, Recent experimental studies of electron dephasing in metal and semi- conductor mesoscopic structures, J. Phys.: Condens. Matter. 14, R501 (2002).DOI
  32. [32]H. Fukuyama and K. Hoshino, Effect of Spin-Orbit Interaction on Magnetoresistance in the Weakly Localized Regime of Three-Dimensional Disordered Systems, J. Phys. Soc. Jpn. 50, 2131 (1981).
  33. [33]S. Maekawa and H. Fukuyama, Magnetoresistance in Two-Dimensional Disordered Systems : Effects of Zeeman Splitting and Spin-Orbit Scattering, J. Phys. Soc. Jpn. 50, 2516 (1981).DOI
  34. [34]D. V. Baxter, R. Richter, M. L. Trudeau, R. W. Cochrane, and J. O. Strom-Olsen, Fitting to magnetoresistance under weak localization in three dimensions, J. Physique 50, 1673 (1989).DOI
  35. [35]G. Bergmann, Weak localization in thin films : a time-of-flight experiment with conduction electrons, Physics Reports 107, 1 (1984).DOI
  36. [36]S. Hikami, A. I. Larkin, and Y. Nagaoka, Spin-Orbit Interaction and Magnetoresistance in the Two Dimensional Random System, Prog. Theor. Phys. 63, 707 (1980).
  37. [37]B. Zhao, P. Cheng, H. Pan, S. Zhang, B. Wang, G. Wang, F. Xiu, and F. Song, Weak antilocalization in Cd₃As₂ thin films, Sci. Rep. 6, 22377 (2016).
  38. [38]The summation function f₃ can be approximated by the formula f₃(z) ≈ 2(√(2 + 1/z − √(1/z)) − [(1/2 + 1/z)⁻¹ᐟ² + (3/2 + 1/z)⁻¹ᐟ²] + 1/48 (2.03 + 1/z)⁻³ᐟ²) with sufficient accuracy. We used this approximation in the present analysis.
  39. [39]A. A. Abrikosov and L. P. Gor’kov, Spin-orbit interaction and the Knight shift in superconductors, JETP 15, 752 (1962).
  40. [40]M. Oszwaldowski and T. Berus, Weak localization in InSb thin films heavily doped with lead, Phys. Rev. B 65, 235418 (2002).DOI
  41. [41]M. L. Peres, V. A. Chitta, N. F. Oliveira, Jr., D. K. Maude, P. H. O. Rappl, A. Y. Ueta, and E. Abramof, Antilocalization of hole carriers in Pb₁₋ₓEuₓTe alloys in the metallic regime, Phys. Rev. B 79, 085309 (2009).DOI
  42. [42]N. Papanikolaou, N. Stefanou, P. H. Dederichs, S. Geier, and G. Bergmann, First-principles calculations of the spin-orbit scattering cross section of sp impurities in Mg, Phys. Rev. Lett. 69, 2110 (1992).
  43. [43]S. Geier and G. Bergmann, Residual atomic behavior of (sp)-impurities in a Mg host, Phys. Rev. Lett. 68, 2520 (1992).
  44. [44]P. Monod and S. Schultz, Conduction electron spin-flip scattering by impurities in copper, Journal de Physique 43, 393 (1982).
  45. [45]D. V. Fedorov, P. Zahn, M. Gräfhand, and I. Mertig, First-principles calculations of spin relaxation times of conduction electrons in Cu with nonmagnetic impurities, Phys. Rev. B 79, 059901(E) (2009).
  46. [46]S. A. Studenikin, P. T. Coleridge, N. Ahmed, P. J. Poole, and A. Sachrajda, Experimental study of weak antilocalization effects in a high-mobility InₓGa₁₋ₓAs/InP quantum well, Phys. Rev. B 68, 035317 (2003).
  47. [47]B. Grbić, R. Leturcq, T. Ihn, K. Ensslin, D. Reuter, and A. D. Wieck, Strong spin-orbit interactions and weak antilocalization in carbon-doped p-type GaAs/AlₓGa₁₋ₓAs, Phys. Rev. B 77, 125312 (2008).
  48. [48]R. L. Kallaher and J. J. Heremans, Spin and phase coherence measured by antilocalization in n-InSb thin films, Phys. Rev. B 79, 075322 (2009).
  49. [49]N. Thillosen, S. Cabanas, N. Kaluza, V. A. Guzenko, H. Hardtdegen, and Th. Schüpers, Weak antilocalization in gate-controlled AlₓGa₁₋ₓN/GaN two-dimensional electron gases, Phys. Rev. B 73, 241311(R) (2006).
  50. [50]N. Thillosen, Th. Schäpers, N. Kaluza, H. Hardtdegen, and V. A. Guzenko, Weak antilocalization in a polarization-doped AlₓGa₁₋ₓN/GaN heterostructure with single subband occupation, Appl. Phys. Lett. 88, 0022111 (2006).
  51. [51]B. A. Bernevig, T. L. Hughes, and S.-C. Zhang, Quantum Spin Hall Effect and Topological Phase Transition in HgTe Quantum Wells, Science 314, 1757 (2006).
  52. [52]M. König, S. Wiedmann, C. Brüne, A. Roth, H. Buhmann, L. W. Molenkamp, X.-L. Qi, and S.-C. Zhang, Quantum Spin Hall Insulator State in HgTe Quantum Wells, Science 318, 766 (2007).

Paper details

vlnc:W_pyjkxgy8
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