Large linear magnetoresistance in a new Dirac material BaMnBi2
Wang Yi-Yan1, 2, †, , Yu Qiao-He1, 2, †, , Xia Tian-Long1, 2, ‡,
Department of Physics, Renmin University of China, Beijing 100872, China
Beijing Key Laboratory of Opto-electronic Functional Materials & Micro-nano Devices, Renmin University of China, Beijing 100872, China

 

† These authors contribute equally to this work.

‡ Corresponding author. E-mail: tlxia@ruc.edu.cn

Project supported by the National Natural Science Foundation of China (Grant No. 11574391), the Fundamental Research Funds for the Central Universities, and the Research Funds of Renmin University of China (Grant No. 14XNLQ07).

Abstract
Abstract

Dirac semimetal is a class of materials that host Dirac fermions as emergent quasi-particles. Dirac cone-type band structure can bring interesting properties such as quantum linear magnetoresistance and large mobility in the materials. In this paper, we report the synthesis of high quality single crystals of BaMnBi2 and investigate the transport properties of the samples. BaMnBi2 is a metal with an antiferromagnetic transition at TN = 288 K. The temperature dependence of magnetization displays different behavior from CaMnBi2 and SrMnBi2, which suggests the possible different magnetic structure of BaMnBi2. The Hall data reveals electron-type carriers and a mobility μ(5 K) = 1500 cm2/V·s. Angle-dependent magnetoresistance reveals the quasi-two-dimensional (2D) Fermi surface in BaMnBi2. A crossover from semiclassical MR ∼ H2 dependence in low field to MR ∼ H dependence in high field, which is attributed to the quantum limit of Dirac fermions, has been observed in magnetoresistance. Our results indicate the existence of Dirac fermions in BaMnBi2.

1. Introduction

Dirac material is a group of compounds whose low-energy excitation behaves as massless Dirac particles.[1] In recent years, a variety of Dirac materials have been discovered, such as graphene,[2] three-dimensional (3D) topological insulators Bi1−xSbx, Bi2Se3, and Bi2Te3,[3,4] and Dirac semimetals Na3Bi,[5] Cd3As2.[69] The energy spectrum of Dirac materials exhibits linear behavior and can be described by relativistic Dirac equation. More interestingly, when the time reversal symmetry or space inversion symmetry is broken, Dirac semimetals (DSM) evolve into Weyl semimetals (WSM), whose extraordinary properties such as Fermi arc and chiral anomaly have attracted great attention.[1024] One interesting consequence of the linear energy dispersion is quantum transport phenomena. For Dirac materials, a moderate magnetic field can compel all carriers occupy the lowest Landau level and lead to the quantum linear magnetoresistance (MR).[25,26]

In Dirac materials, there is a class of ternary 112 type compounds such as CaMnBi2,[2730] SrMnBi2,[3134] EuMnBi2,[35,36] and LaAgBi2,[37] which have been researched deeply. More recently, time-reversal symmetry breaking Weyl state has been suggested to exist in YbMnBi2,[38] and Sr1−yMn1−zSb2[39] is also identified as a promising candidate of WSM. Among these materials, the Bi/Sb square net is a common feature and has been considered as the platform that hosts Dirac/Weyl fermions. Therefore, it is of considerable interest to explore new materials which have similar structure and further study their physical properties.

In this work, we have synthesized the single crystals of BaMnBi2 and investigated the transport properties. Hall resistivity shows carriers in BaMnBi2 are electron type. Magnetic property measurement indicates the magnetic structure of BaMnBi2 may be slightly different from that of SrMnBi2 or CaMnBi2. Angle-dependent MR implies the anisotropic quasi-two-dimensional Fermi surface in BaMnBi2. The in-plane MR displays a crossover from semiclassical quadratic field dependence to linear field dependence with the increase of magnetic field. The linear MR indicates the possible existence of Dirac fermions in this material. It is worth doing ARPES experiments to check whether BaMnBi2 is a Weyl semimetal.

2. Experiment and crystal structure

Single crystals of BaMnBi2 were grown from Bi flux. The mixtures of Ba, Mn and Bi were placed in a crucible and sealed in a quartz tube with a ratio of Ba:Mn:Bi=1:1:6. The quartz tube was heated to 1180 °C in 60 h, held there for 30 h, and cooled to 370 °C at a rate of 3 °C/h, and then the excess Bi-flux was removed by centrifuging. Elemental analysis was performed using energy dispersive x-ray spectroscopy (EDS, Oxford X-Max 50). The determined atomic proportion was consistent with the composition of BaMnBi2 within instrumental error. Single crystal x-ray diffraction (XRD) pattern was collected from a Bruker D8 Advance x-ray diffractometer using Cu Kα radiation. Resistivity measurements were performed on a Quantum Design physical property measurement system (QD PPMS-14T) and the magnetic properties were measured with vibrating sample magnetometer (VSM) option.

BaMnBi2 is isostructural with SrMnBi2. As shown in Fig. 1(a), the crystal structure of BaMnBi2 is comprised of alternating MnBi and BaBi layers.[40] In the MnBi layer, each Mn atom is surrounded by four Bi atoms, which form the MnBi4 tetrahedra. In the BaBi layer, Bi atoms are separated and form a square net (highlighted by red atoms). Figure 1(c) shows the x-ray diffraction pattern of a BaMnBi2 single crystal on the (00l) plane. An additional small peak at about 27.2° also has been observed. It may come from the residual Bi flux on the surface. The inset is an optical image of a representative crystal.

Fig. 1. (a) The crystal structure of BaMnBi2. (b) The definition of polar angle θ and azimuthal angle φ. (c) Single crystal x-ray diffraction pattern of a BaMnBi2 crystal, showing only the (00l) reflections. The inset is an image of a typical single crystal with a scale of 4 mm.
3. Results and discussion

Both the in-plane resistivity ρxx(T) and out-plane resistivity ρz(T) exhibit a simple metallic behavior as shown in Fig. 2(a). The out-plane resistivity is nearly 50 times the in-plane resistivity in a range of 2.5 K–300 K. Such a significant anisotropy suggests quasi-2D electronic band structure in BaMnBi2.

Fig. 2. (a) Temperature dependence of in-plane resistivity ρxx(T) and out-plane resistivity ρz(T). (b) Hall resistivity ρxy versus magnetic field H with the temperature T from 5 K to 150 K. (c) and (d) Temperature dependence of carrier concentration and mobility, respectively. (e) Magnetization versus temperature with the magnetic field H = 1 T applied parallel to ab plane (up) and c axis (down) under zero field cooling (ZFC) (red) and field cooling (FC) (violet) conditions. The dashed lines denote two anomalous temperatures.

Figure 2(b) plots the magnetic field dependence of Hall resistivity ρxy(H) measured at various temperatures. The negative slope of ρxy suggests that the dominant charge carriers in BaMnBi2 are electrons. Single band model has been employed to analyze the Hall effect data. The carrier concentration and carrier mobility are shown in Figs. 2(c) and 2(d) respectively. The carrier concentration is given by n = 1/eRH, and the Hall coefficient RH is obtained by RH = ρxy(8 T)/H. At 5 K, the carrier concentration is 1.1×1019 cm−3. We calculate the carrier mobility by μ = 1/enρxx(0 T). With temperature reduced, the mobility becomes larger and reaches 1500 cm2/V·s at 5 K.

Figure 2(e) presents the temperature dependence of magnetization of BaMnBi2 measured under zero field cooling (ZFC) and field cooling (FC) conditions with an applied field H = 1 T parallel to ab plane and c axis respectively. Similar to CaMnBi2 and SrMnBi2,[33,34] there are two anomalous temperatures in BaMnBi2. T1 = TN = 288 K is the temperature of antiferromagnetic transition, below T1 the magnetization exhibits strong anisotropy, the linear temperature dependence of magnetization with positive slope above T1 indicates strong antiferromagnetic correlations.[41] There is a ZFC–FC splitting for magnetic field parallel to ab plane when T < T2 = 270 K. This is quite different from the behavior in CaMnBi2 or SrMnBi2, in which the ZFC–FC splitting occurs when the field is parallel to c axis. The difference implies that the magnetic structure of BaMnBi2 may be slightly different from that of CaMnBi2 or SrMnBi2.

Fig. 3. (a) Polar angle θ dependence of in-plane resistivity ρxx at H = 10, 6, 2 T. The temperature and azimuthal angle are fixed at T = 2.5 K and φ = 90° respectively. The red curve is the fitting of cosine function. (b) Magnetic field H dependence of in-plane resistivity ρxx with different polar angle θ at 2.5 K. (c) In-plane magnetoresistance MR versus magnetic field H with the temperature from 5 K to 150 K. (d) The field derivative of the MR as a function of magnetic field H. The straight lines in low field regions are linear fitting results using the relationship MR = A2H2. Lines in high field regions are fitting results using MR = A1H + O(H2). The intersection of fitting lines in low field and high field is defined as critical point H*. (e) Temperature dependence of critical point H*. The solid curve is the fit to the equation

The carriers of a metal in magnetic field are subject to the Lorentz force. The Lorentz force affects the carriers’ momentum components in the plane perpendicular to the field and MR is partially determined by the mobility in this plane. The carriers in a quasi-2D material will only be affected by the magnetic field component H|cosθ|. Figure 3(a) shows the angular-dependent in-plane MR measured in different magnetic fields. When θ = 0°, 180° (field parallel to c axis), MR has the maximum value. With the increase of the polar angle θ, MR decreases gradually and reaches the minimum value when θ = 90°. The |cosθ| function behavior of the whole curve indicates the existence of a quasi-2D Fermi surface in BaMnBi2. The small deviation implies the existence of 3D electronic transport in BaMnBi2.

Figure 3(c) shows the in-plane MR as a function of magnetic field at different temperatures. The MR exhibits quantum linear behavior in high field and semiclassical quadratic dependence in low field.[42] The transition from semiclassical to quantum can be seen more clearly from the field derivative of the MR [Fig. 3(d)]. Initially, dMR/dH is proportional to H, which indicates a semiclassical H2-dependent MR. With the increase of magnetic field, the field crosses a critical point H* and dMR/dH nearly becomes a constant. This implies that MR follows linear field dependence plus a small quadratic term in high field, namely MR = A1H+O(H2). The linear MR suggests the existence of Dirac fermions in BaMnBi2.[4346]

A sufficiently strong perpendicular magnetic field can cause the complete quantization of orbit of Dirac fermions and the quantized Landau level can be described as where vF is the Fermi velocity and n is the Landau index. So the energy splitting between the first and lowest Landau level is At a specific temperature T, with the increase of field H, the energy splitting becomes larger than the Fermi energy EF and the thermal fluctuations kBT. As a result, all carriers will occupy the lowest Landau level and reach the quantum limit. With these two relationships combined, we can get the temperature dependence of the critical magnetic field [47] However, the condition for absolute quantum limit is n < (eH/ħ)3/2, where n is the carrier concentration. Substituting the electron concentration n(5 K) = 1.1 × 1019 cm−3 obtained from the Hall measurement into the expression, we get the value H > 32.5 T. Although the system does not enter the absolute quantum limit in our situation, the linear quantum MR also could appear at lower field with more than one Landau level filled, which also has been observed in previous research on InSb.[45] In addition, considering the relationship between T and H* can be well fitted by the equation as shown in Fig. 3(e), it is suggested that the linear MR in BaMnBi2 originates from Dirac fermions.

4. Summary

In summary, single crystals of BaMnBi2 have been grown. Resistivity, Hall resistivity, magnetic property and magnetoresistance have been measured and analyzed. Compared with CaMnBi2 and SrMnBi2, the different behavior of magnetization brings the possibly different magnetic structure of BaMnBi2. Quasi-2D electronic transport is also observed in angle-dependent MR. The crossover from semiclassical parabolic field-dependent MR in low field to linear field-dependent MR in high field can be explained by combining semiclassical and quantum magnetoresistance. Our results clearly indicate the existence of Dirac fermions in BaMnBi2.

Note added When the paper was being finalized, we noticed one similar related work on BaMnBi2,[48] where the Shubnikov–de Hass oscillation is also observed under high magnetic field. However, the angle-dependent resistivity in their work is different from ours and deviates greatly from the |cosθ| relationship, which may be caused by misaligned leads on the samples besides the contribution from 3D Fermi surface.

Reference
1Wehling T OBlack-Schaffer A MBalatsky A V 2014 Adv. Phys. 63 1
2Castro Neto A HGuinea FPeres N M RNovoselov K SGeim A K 2009 Rev. Mod. Phys. 81 109
3Hasan M ZKane C L 2010 Rev. Mod. Phys. 82 3045
4Qi X LZhang S C 2011 Rev. Mod. Phys. 83 1057
5Liu Z KZhou BZhang YWang Z JWeng H MPrabhakaran DMo S KShen Z XFang ZDai XHussain ZChen Y L 2014 Science 343 864
6Borisenko SGibson QEvtushinsky DZabolotnyy VBüchner BCava R J 2014 Phys. Rev. Lett. 113 027603
7Neupane MXu S YSankar RAlidoust NBian GLiu CBelopolski IChang T RJeng H TLin HBansil AChou FHasan M Z 2014 Nat. Commun. 5 3786
8Liang TGibson QAli M NLiu MCava R JOng N P 2015 Nat. Mater. 14 280
9He L PHong X CDong J KPan JZhang ZZhang JLi S Y 2014 Phys. Rev. Lett. 113 246402
10Wan XTurner A MVishwanath ASavrasov S Y 2011 Phys. Rev. 83 205101
11Lv B QWeng H MFu B BWang X PMiao HMa JRichard PHuang X CZhao L XChen G FFang ZDai XQian TDing H 2015 Phys. Rev. 5 031013
12Xu S YBelopolski IAlidoust Net al. 2015 Science 349 613
13Lv B QXu NWeng H MMa J ZRichard PHuang X CZhao L XChen G FMatt C EBisti FStrocov V NMesot JFang ZDai XQian TShi MDing H 2015 Nat. Phys. 11 724
14Huang XZhao LLong YWang PChen DYang ZLiang HXue MWeng HFang ZDai XChen G 2015 Phys. Rev. 5 031023
15Yang LLiu Z KSun YPeng HYang H FZhang TZhou BZhang YGuo Y FRahn MPrabhakaran DHussain ZMo S KFelser CYan BChen Y L 2015 Nat. Phys. 11 728
16Huang S MXu S YBelopolski ILee C CChang GWang B KAlidoust NBian GNeupane MZhang CJia SBansil ALin HHasan M Z 2015 Nat. Commun. 6 7373
17Xu S YBelopolski ISanchez Daniel Set al. 2015 Sci. Adv. 1 e1501092
18Weng HFang CFang ZBernevig B ADai X 2015 Phys. Rev. 5 011029
19Liu Z KYang L XSun YZhang TPeng HYang H FChen CZhang YGuo Y FPrabhakaran DSchmidt MHussain ZMo S KFelser CYan BChen Y L 2016 Nat. Mater. 15 27
20Shekhar CNayak A KSun YSchmidt MNicklas MLeermakers IZeitler USkourski YWosnitza JLiu ZChen YSchnelle WBorrmann HGrin YFelser CYan B 2015 Nat. Phys. 11 645
21Arnold FShekhar CWu S CSun YRicardo D d RKumar NNaumann MAjeesh M OSchmidt MGrushin A GBardarson J HBaenitz MSokolov DBorrmann HNicklas MFelser CHassinger EYan B 2016 Nat. Commun. 7 11615
22Yang XLiu YWang ZZheng YXu Z2015arXiv:1506.03190v1 [cond-mat.mtrl-sci]
23Sun YWu S CYan B 2015 Phys. Rev. 92 115428
24Zhang JLiu F LDong J KXu YLi N NYang W GLi S Y 2015 Chin. Phys. Lett. 32 097102
25Zhang YTan Y WStormer H LKim P 2005 Nature 438 201
26Miller D LKubista K DRutter G MRuan Mde Heer W AFirst P NStroscio J A 2009 Science 324 924
27Wang KGraf DWang LLei HTozer S WPetrovic C 2012 Phys. Rev. 85 041101
28He J BWang D MChen G F 2012 Appl. Phys. Lett. 100 112405
29Ya FengWang ZChen Cet al. 2014 Sci. Rep. 4 5385 2014
30Lee GFarhan M AKim J SShim J H 2013 Phys. Rev. 87 245104
31Park JLee GWolff-Fabris Fet al. 2011 Phys. Rev. Lett. 107 126402
32Wang KGraf DLei HTozer S WPetrovic C 2011 Phys. Rev. 84 220401
33Wang J KZhao L LYin QKotliar GKim M SAronson M CMorosan E 2011 Phys. Rev. 84 064428
34Guo Y FPrincep A JZhang Xet al. 2014 Phys. Rev. 90 075120
35Masuda HSakai HTokunaga Met al. 2016 Sci. Adv. 2 e1501117
36May A FMcGuire M ASales B C 2014 Phys. Rev. 90 075109
37Wang KGraf DPetrovic C 2013 Phys. Rev. 87 235101
38Borisenko SEvtushinsky DGibson QYaresko AKim TAli MBuechner BHoesch MCava R J2015arXiv: 1507.04847v2 [cond-mat.mes-hall]
39Liu J YHu JZhang QGraf DCao H BRadmanesh S M AAdams D JZhu Y LChen G FLiu XWei JTennant D ADiTusa J FChiorescu ISpinu LMao Z Q2015arXiv: 1507.07978v2 [cond-mat.mtrl-sci]
40Cordier GSchäfer H1977Z. Naturforsch.32383
41Zhang G MSu Y HLu Z YWeng Z YLee D HXiang T 2009 Europhys. Lett. 86 37006
42Abrikosov A ABeknazarov A1988Fundamentals of the Theory of MetalsAmsterdamNorth Holland
43Abrikosov A A 1998 Phys. Rev. 58 2788
44Abrikosov A A 2000 Europhys. Lett. 49 789
45Hu JRosenbaum T 2008 Nat. Mater. 7 697
46Lee MRosenbaum T FSaboungi M LSchnyders H S 2002 Phys. Rev. Lett. 88 066602
47Huynh Khuong K.Tanabe Yoichi Tanigaki Katsumi 2011 Phys. Rev. Lett. 106 217004
48Li LWang KGraf DWang LWang APetrovic C 2016 Phys. Rev. 93 115141