The search for appropriate materials for technological applications is challenging, as real materials are subject to uncontrolled doping and thermal effects. Tetragonal NaMnBi of the I-Mn-V class of antiferromagnetic semiconductors with a Néel transition (TN), above room temperature, can exhibit an extreme magnetoresistance (MR), greater than 10 000% at 2 K and 600% at room temperature and 9 T by quenching disorder into the system. Coupled with the large MR is a re-orientation of the magnetic moment, from a collinear spin arrangement along c to a canted one along the (011) crystallographic axis. The extreme MR is observed in samples with about 15% of Bi vacancies which in turn effectively introduces charge carriers into the lattice, leading to a drastic change in the electronic transport, from semiconducting to metallic, and to the very large MR under the magnetic field. In the absence of Bi defects, the MR is severely suppressed, suggesting that the hybridization of the Mn and Bi orbitals may be key to the field induced large MR. This is the only material of its class that exhibits the extreme MR and may potentially find use in microelectronic devices.

The emerging field of antiferromagnetic (AFM) spintronics aims at using room temperature AFM semiconductors to complement or replace ferromagnetic (FM) electrodes in hard drive read heads and magnetic random access memories.1,2 Their resilience to external magnetic fields and the absence of stray fields make AFMs suitable for ultrafast, ultrahigh-density magnetic storage devices and AFM spin valves.3 AFM semiconductors can potentially integrate spintronics and microelectronic functionalities in a single material.4,5 Traditional spin-based electronics are FM ordered systems exhibiting giant magnetoresistance (GMR) in which the ordered magnetic moment couples to the electrical resistance.6–8 Devices using semiconductors with FM ground states are however limited due to low Curie temperatures, chemical solubility, and metastability, although recent advances have reached a Curie temperature of up to 230 K.9–12 Several candidate AFM semiconductors have been identified, with zinc-blende structures, half-Heusler compounds, and spin-orbit coupled oxides.4,5,13–15 Early theoretical work proposed that sizeable GMR and current induced torque can materialize in trilayers made of AFM Cr/Au/Cr layers.16,17 More recently, tunneling devices with AFM IrMn/non-magnetic metal layers yielded large MR signals,4,5,13 demonstrating that electronic device characteristics can be controlled by an AFM. Single phase AFM semiconductors with large MR are rare however.

New materials engineering holds the promise for realizing AFM spintronics. Three-dimensional AFM semiconductors of the I-Mn-V (with I = Li, Na, and K, and V = As, Sb, and Bi) class with Néel temperatures above 300 K are promising candidates for the next generation of AFM spintronics.4,9,13–15,18–20 Of this class, LiMnAs has been predicted to exhibit large spin orbit coupling induced magnetic anisotropy.4 Tetragonal phases of CuMnSb and CuMnAs are compatible with existing semiconducting technologies but show little magnetoresistance in bulk form.1 We report that the narrow-gap NaMnBi semiconductor can exhibit a very large MR when defects are introduced in the crystal lattice. When quenched from high temperature, the MR increases thousand-fold, with the large increase observed upon cooling from room temperature to 2 K, and is not limited to the vicinity of TN. Quenching introduces vacancies on the Bi and Na sublattices, accompanied by the reorientation of the AFM coupled Mn moments which involves spin canting. Deviations of the local magnetization can propagate through magnons and may find use in information processing such as in magnonic devices.21 

NaMnBi has a tetragonal crystal structure, consisting of layers of Mn tetrahedrally coordinated with Bi and with Na intercalated between the layers as seen in Fig. 1(a).22,23 As-grown and quenched single crystals of NaMnBi can be synthesized in bulk form as seen in the inset of Fig. 1(b), a major advantage over existing (III, Mn)V semiconductors.24 The volatility of Na limits the amount of the alkali metal in the crystal structure. From chemical composition analysis by energy-dispersive x-ray spectroscopy (EDX) using a FEI Quanta LV200 Scanning Electron Microscope, we estimate the as-grown and quenched chemical ratios to be N a 0.92 MnBi and N a 0.82 M n B i 0.86 , respectively. This was also confirmed by the Rietveld analysis of diffraction data, with refined parameters in Table I. The backscattered electron micrographs of Figs. 1(c)–1(f) obtained using the Scanning Electron Microscope (SEM) show a homogeneous surface morphology that is similar between the two types of crystals at the two magnifications. No evidence of metallic inclusions or other types can be observed at these length scales.

FIG. 1.

(a) The bulk magnetic susceptibility of as-grown and quenched NaMnBi using a 0.1 T field parallel to the c-axis. The inset shows the tetragonal crystal structure of pristine NaMnBi with its collinear spin configuration. (b) Resistivity at zero magnetic field for a quenched crystal, Na0.82MnBi0.86, and an as-grown Na0.92MnBi crystal. The inset shows a photograph of the quenched crystal. A transition is observed in the transport of the quenched crystal, which is absent in the as-grown crystal. (c)–(f) Back-scattered electron SEM images on the cleaved surface of two typical crystals. Shown in (c) and (e) is the topography of the as-grown crystal. Shown in (d) and (f) is the topography of the quenched crystal. Measurements were done at a voltage of 20 kV.

FIG. 1.

(a) The bulk magnetic susceptibility of as-grown and quenched NaMnBi using a 0.1 T field parallel to the c-axis. The inset shows the tetragonal crystal structure of pristine NaMnBi with its collinear spin configuration. (b) Resistivity at zero magnetic field for a quenched crystal, Na0.82MnBi0.86, and an as-grown Na0.92MnBi crystal. The inset shows a photograph of the quenched crystal. A transition is observed in the transport of the quenched crystal, which is absent in the as-grown crystal. (c)–(f) Back-scattered electron SEM images on the cleaved surface of two typical crystals. Shown in (c) and (e) is the topography of the as-grown crystal. Shown in (d) and (f) is the topography of the quenched crystal. Measurements were done at a voltage of 20 kV.

Close modal
TABLE I.

A list of crystal structure parameters from the Rietveld refinement of as-grown and quenched NaMnBi. The space group is P4/nmm, and the Mn site is ( 3 4 , 1 4 , 0 ) .

χ 2 a (Å) c (Å) Na ( 1 4 , 1 4 , c ) Bi ( 3 4 , 3 4 , c ) Na Mn Bi
c c Uiso Uiso Uiso
As-grown  3.494  4.5444(1)  7.6997(5)  0.6526(8)  0.7693(4)  0.0068  0.0137  0.0060 
Quenched  1.312  4.5460(2)  7.7070(9)  0.6527(9)  0.7694(7)  0.0204  0.0176  0.0076 
χ 2 a (Å) c (Å) Na ( 1 4 , 1 4 , c ) Bi ( 3 4 , 3 4 , c ) Na Mn Bi
c c Uiso Uiso Uiso
As-grown  3.494  4.5444(1)  7.6997(5)  0.6526(8)  0.7693(4)  0.0068  0.0137  0.0060 
Quenched  1.312  4.5460(2)  7.7070(9)  0.6527(9)  0.7694(7)  0.0204  0.0176  0.0076 

The magnetotransport properties are tailored by quenching defects into the lattice. Distinctly different material characteristics are observed between quenched and as-grown crystals due to manifestation of disorder. The zero-field transport shown in Fig. 1(b) is typical of crystal grown by the two methods (quenched versus as-grown). The upper curve shows a semiconducting temperature dependence with the resistivity rising upon cooling for the as-grown crystal. The lower curve shows semi-metallic behavior with the resistivity dropping upon cooling and an anomaly appearing near 250 K which shows thermal hysteresis for the quenched crystal. Consistent with the resistivity data, the bulk magnetization measured at 0.1 T also shows a kink near 250 K [Fig. 1(a)]. On the other hand, the as-grown crystal shows no anomaly around 250 K, neither in the resistivity nor in the magnetization.

When the magnetic field is turned on, the electrical response of the as-grown and quenched crystals is starkly different. The magnetotransport results are summarized in Fig. 2. Between 0 and 9 T, the resistivity of the as-grown crystal shows a modest increase upon cooling (20%) [Fig. 2(a)]. However, the resistivity of the quenched crystal increases significantly between 0 and 9 T as seen in Fig. 2(b). This gives rise to an anomalously large positive MR that increases substantially upon cooling. Classical MR is a weak effect that commonly appears in non-magnetic systems under a magnetic field, H. Defined as the ratio of the change of resistivity with a field to the resistivity without a field, Δ ρ ( H ) ρ ( 0 ) = [ ρ ( H ) ρ ( 0 ) ] ρ ( 0 ) , it is usually positive.25 

FIG. 2.

(a) A resistivity measurement at 0 and 9 T with H//c on the as-grown crystal. The resistivity was measured along the a-direction. (b) A more detailed resistivity measurement as a function of the magnetic field for the quenched single crystal. (c) MR versus magnetic field at 2 K (left panel, red color) and 300 K (right panel, black color) for the as-grown crystal. (d) MR versus magnetic field at 2 K (left panel, red color) and 300 K (right panel, black color) for the quenched crystal.

FIG. 2.

(a) A resistivity measurement at 0 and 9 T with H//c on the as-grown crystal. The resistivity was measured along the a-direction. (b) A more detailed resistivity measurement as a function of the magnetic field for the quenched single crystal. (c) MR versus magnetic field at 2 K (left panel, red color) and 300 K (right panel, black color) for the as-grown crystal. (d) MR versus magnetic field at 2 K (left panel, red color) and 300 K (right panel, black color) for the quenched crystal.

Close modal

Associated with the orbital motion of the charge carriers, the MR shows a quadratic dependence in H at low fields but quickly saturates as the field increases.26 An anomalously large positive MR, as shown here, is particularly uncommon. Plotted at two temperatures as a function of the field, the MR increases by more than 10 000% by 2 K and 9 T, while at room temperature, it reaches 600%, an appreciable value [Fig. 2(d)]. On the other hand, the MR observed in the as-grown crystal is significantly smaller [Fig. 2(c)]. To date, the observed MR reported here is the largest MR observed in a I-Mn-V semiconductor.

The neutron powder diffraction measurements were performed on crushed single crystals as a function of the magnetic field up to 7 T. The results are summarized in Fig. 3. In both the quenched and as-grown compositions, the crystal symmetry is P4/nmm, consistent with the tetragonal space group previously reported.19 The a- and c-lattice constants are 4.5206 and 7.6713 Å in the quenched sample and 4.5443 and 7.6995 Å in the as-grown sample, respectively. The magnetic refinement of the diffraction data of the as-grown NaMnBi shown in Fig. 3(a) yields a collinear magnetic structure along c, in agreement with previous reports and with a Néel temperature TN 340 K.20 With quenching, spin canting due to disorder occurs where the spins align along the (011) direction as shown in the inset of Fig. 3(b). The refined magnetic moment in the quenched sample is around 4.5 μB/Mn just as in the as-grown. The order parameter determined at 7 T (black symbols) is shown in Fig. 3(c) which is a plot of the integrated intensity of the magnetic ( 200 ) M Bragg peak as a function of temperature for the quenched crystal. Also shown in this plot is the order parameter for the as-grown crystal, indicating that the number of vacancies does not affect TN but clearly affects the MR.

FIG. 3.

The neutron powder diffraction data in (a) are for the as-grown sample at 4 K and in (b) for the quenched sample at 13 K. The vertical marks represent the position of Bragg peaks. The insets show the refined magnetic configuration at H = 0 T. (c) The magnetic order parameter at 0 and 7 T for the quenched crystal and at 0 T for the as-grown crystal. The inset shows the magnetic field dependence of the unit cell volume at 5 and 300 K. (d) The temperature dependence of the magnetic (200)M peak at 7 T. The color indicates the neutron diffraction intensity. Error bars correspond to one standard deviation.

FIG. 3.

The neutron powder diffraction data in (a) are for the as-grown sample at 4 K and in (b) for the quenched sample at 13 K. The vertical marks represent the position of Bragg peaks. The insets show the refined magnetic configuration at H = 0 T. (c) The magnetic order parameter at 0 and 7 T for the quenched crystal and at 0 T for the as-grown crystal. The inset shows the magnetic field dependence of the unit cell volume at 5 and 300 K. (d) The temperature dependence of the magnetic (200)M peak at 7 T. The color indicates the neutron diffraction intensity. Error bars correspond to one standard deviation.

Close modal

When the magnetic field is turned on with H c , the spin structure of the quenched crystal remains the same, yielding the same refined moment. As a function of the field, the unit cell volume shows a noticeable expansion at 300 K, as seen in the inset of Fig. 3(c), but not at 5 K. Moreover, the intensity of the magnetic order parameter at 7 T reaches a peak 250 K as seen in Fig. 3(d), suggesting a gradual transition from the paramagnetic (PM) to the AFM phase due to persistent spin fluctuations that are not suppressed under the field. It is clear that defects lead to the reorientation of the AFM moments. The introduction of defects most likely affects the Mn orbital states, but it is not understood at present how the defect concentration affects the band structure.

At present, it is difficult to specify the microscopic origin of the large MR in NaMnBi. One possible mechanism may arise from spin fluctuations. It is known that enhanced spin fluctuation in antiferromagnets under a magnetic field can raise MR.27,28 This is related to the well-known GMR, originating from the suppression of spin fluctuation of ferromagnets by the stabilization of the ferromagnetic order under a magnetic field, which, on the other hand, does not stabilize the AFM order. NaMnBi may be useful as a single magnetic electrode in devices because it carries both magnetism and a very large magnetoresistance.4,14

High purity elements of Na (99.99%), Mn (99.99%), and Bi (99.9999%) with a ratio of 1:1:6 were placed in a quartz tube for the single crystal growth. The quartz tube was heated to 600° and held there for 48 h to allow soaking with a Bi metal which is a liquid at this temperature. Following this, the tube was slow-cooled down to 300 °C. As-grown crystals were obtained by decanting at 300°, followed by furnace cooling to room temperature. Quenched crystals were obtained by after-growth annealing at 350° and quenched into liquid nitrogen. The resulting crystals can easily cleave, exposing a clean surface.

The conventional four-probe technique was used for the resistivity measurements. Four gold wires were attached on the surface of the crystal using silver epoxy. The silver epoxy was cured at room temperature in a glove box for 48 h prior to the measurement. The resistivity and magnetic susceptibility measurements were performed using a Quantum Design Physical Properties Measurement System. Single crystals were ground to powder for the neutron powder diffraction measurements performed using the BT1 diffractometer at the NIST Center for Neutron Research (NCNR) located in Gaithersburg, Maryland, with a neutron wavelength of λ = 2.079 Å , as well as using the HB2A diffractometer at the HFIR of Oak Ridge National Laboratory with a neutron wavelength of λ = 2.4111 Å . Energy dispersive X-ray (EDX) analysis was performed using an FEI Quanta LV200 Scanning Electron Microscope (SEM).

The work at the University of Virginia was supported by the Department of Energy (Grant No. DE-FG02-01ER45927). The authors would like to acknowledge valuable discussions with R. Arita and M. Ochi.

1.
F.
Máca
,
J.
Masek
,
O.
Stelmakhovych
,
X.
Martí
,
H.
Reichlová
,
K.
Uhlírová
,
P.
Beran
,
P.
Wadley
,
V.
Novák
, and
T.
Jungwirth
,
J. Magn. Magn. Mater.
324
,
1606
1612
(
2012
).
2.
X.
Marti
,
I.
Fina
,
C.
Frontera
,
J.
Liu
,
P.
Wadley
,
Q.
He
,
R. J.
Paull
,
J. D.
Clarkson
,
J.
Kudrnovský
,
I.
Turek
,
J.
Kunes
,
D.
Yi
,
J.-H.
Chu
,
C. T.
Nelson
,
L.
You
,
E.
Arenholz
,
S.
Salahuddin
,
J.
Fontcuberta
,
T.
Jungwirth
, and
R.
Ramesh
,
Nat. Mater.
13
,
367
(
2014
).
3.
C.
Chappert
,
A.
Fert
, and
F. N.
Van Dau
,
Nat. Mater.
6
,
813
823
(
2007
).
4.
T.
Jungwirth
,
T. V.
Novák
,
X.
Martí
,
M.
Cukr
,
F.
Máca
,
A. B.
Shick
,
J.
Masek
,
P.
Horodyská
,
P.
Nĕmec
,
V.
Holý
,
J.
Zemek
,
P.
Kuzel
,
I.
Nĕmec
,
B. L.
Gallagher
,
R. P.
Campion
,
C. T.
Foxon
, and
J.
Wunderlich
,
Phys. Rev. B
83
,
035321
(
2011
).
6.
S. A.
Wolf
,
D. D.
Awschalom
,
R. A.
Buhrman
,
J. M.
Daughton
,
S.
von Molnár
,
M. L.
Roukes
,
A. Y.
Chtchelkanova
, and
D. M.
Treger
,
Science
294
,
1488
1495
(
2001
).
7.
8.
T.
Dietl
,
D. D.
Awschalom
,
M.
Kaminska
, and
H.
Ohno
,
Semiconductors and Semimetals Spintronics
(
Elsevier
,
Amsterdam
,
2008
), p.
82
.
9.
Z.
Deng
,
C. Q.
Jin
,
Q. Q.
Liu
,
X. C.
Wang
,
J. L.
Zhu
,
S. M.
Feng
,
L. C.
Chen
,
R. C.
Yu
,
C.
Arguello
,
T.
Goko
,
F.
Ning
,
J.
Zhang
,
Y.
Wang
,
A. A.
Aczel
,
T.
Munsie
,
T. J.
Williams
,
G. M.
Luke
,
T.
Kakeshita
,
S.
Uchida
,
W.
Higemoto
,
T. U.
Ito
,
B.
Gu
,
S.
Maekawa
,
G. D.
Morris
, and
Y. J.
Uemura
,
Nat. Commun.
2
,
422
(
2011
).
10.
K.
Zhao
,
B.
Chen
,
G.
Zhao
,
Z.
Yuan
,
Q.
Liu
,
Z.
Deng
,
J.
Zhu
, and
C.
Jin
,
Chin. Sci. Bull.
59
,
2524
(
2014
).
11.
K.
Zhao
,
Z.
Deng
,
X. C.
Wang
,
W.
Han
,
J. L.
Zhu
,
X.
Li
,
Q. Q.
Liu
,
R. C.
Yu
,
T.
Goko
,
B.
Frandsen
, and
L.
Liu
,
Nat. Commun.
4
,
1442
(
2013
).
12.
G. Q.
Zhao
,
C. J.
Lin
,
Z.
Deng
,
G. X.
Gu
,
S.
Yu
,
X. C.
Wang
,
Z. Z.
Gong
,
Y. J.
Uemera
,
Y. Q.
Li
, and
C. Q.
Jin
,
Sci. Rep.
7
,
14473
(
2017
).
13.
P.
Wadley
,
V.
Novák
,
R. P.
Campion
,
C.
Rinaldi
,
X.
Martí
,
H.
Reichlová
,
J.
Elezný
,
J.
Gazquez
,
M. A.
Roldan
,
M.
Varela
,
D.
Khalyavin
,
S.
Langridge
,
D.
Kriegner
,
F.
Máca
,
J.
Masek
,
R.
Bertacco
,
V.
Holý
,
A. W.
Rushforth
,
K. W.
Edmonds
,
B. L.
Gallagher
,
C. T.
Foxon
,
J.
Wunderlich
, and
T.
Jungwirth
,
Nat. Commun.
4
,
2322
(
2013
).
14.
A. B.
Shick
,
S.
Khmelevskyi
,
O. N.
Mryasov
,
J.
Wunderlich
, and
T.
Jungwirth
,
Phys. Rev. B
81
,
212409
(
2010
).
15.
B. J.
Kim
,
H.
Ohsumi
,
T.
Komesu
,
S.
Sakai
,
T.
Morita
,
H.
Takagi
, and
T.
Arima
,
Science
323
,
1329
1332
(
2009
).
16.
A. S.
Núñez
,
R. A.
Duine
,
P.
Haney
, and
A. H.
MacDonald
,
Phys. Rev. B
73
,
214426
(
2006
).
17.
R. A.
Duine
,
P. M.
Haney
,
A. S.
Núñez
, and
A. H.
MacDonald
,
Phys. Rev. B
75
,
14433
(
2007
).
18.
W.
Yao
,
D.
Xiao
, and
Q.
Niu
,
Phys. Rev. B
77
,
235406
(
2008
).
19.
Y. Y.
Wang
,
C.
Song
,
B.
Cui
,
G. Y.
Wang
,
F.
Zeng
, and
F.
Pan
,
Phys. Rev. Lett.
109
,
137201
(
2012
).
20.
B. G.
Park
,
J.
Wunderlich
,
X.
Martí
,
V.
Holý
,
Y.
Kurosaki
,
M.
Yamada
,
H.
Yamamoto
,
A.
Nishide
,
J.
Hayakawa
,
H.
Takahashi
,
A. B.
Shick
, and
T.
Jungwirth
,
Nat. Mater.
10
,
347
(
2011
).
21.
R. E.
Troncoso
,
C.
Ulloa
,
F.
Pesce
, and
A. S.
Nunez
,
Phys. Rev. B
92
,
224424
(
2015
).
22.
W.
Bronger
,
P.
Müller
,
R.
Huppner
, and
H.-U.
Schuster
,
Z. Anorg. Allg. Chem.
539
,
175
(
1986
).
23.
R.
Müller
,
M.
Kuckel
,
H.-U.
Schuster
,
P.
Müller
, and
W.
Bronger
,
J. Alloys Compd.
176
,
167
(
1991
).
25.
J. M.
Ziman
,
Principles of the Theory of Solids
(
Cambridge University Press
,
London
,
1972
).
26.
H.
Jones
,
Theory of Electrical and Thermal Conductivity in Metals
(
Springer
Berlin, Heidelberg
,
1956
).
27.
H.
Yamada
and
S.
Takada
,
J. Phys. Soc. Jpn.
34
,
51
(
1973
).
28.
I.
Balberg
and
J. S.
Helman
,
Phys. Rev. B
18
,
303
(
1978
).