Linear Magnetoelectric Phase in Ultrathin MnPS3 Probed by Optical
Second Harmonic Generation
Hao Chu,1,2 Chang Jae Roh,3 Joshua O. Island,4 Chen Li,1,2 Sungmin Lee,5,6 Jingjing Chen,7 Je-Geun Park,5,6
Andrea F. Young,4 Jong Seok Lee,3 and David Hsieh 1,2,*
1Department of Physics, California Institute of Technology, Pasadena, California 91125, USA
2Institute for Quantum Information and Matter, California Institute of Technology, Pasadena, California 91125, USA
3Department of Physics and Photon Science, Gwangju Institute of Science and Technology, Gwangju 61005, Republic of Korea
4Department of Physics, University of California, Santa Barbara, California 93106, USA
5Center for Correlated Electron Systems, Institute for Basic Science (IBS), Seoul 08826, Republic of Korea
6Department of Physics and Astronomy, Seoul National University (SNU), Seoul 08826, Republic of Korea
7School of Physics, Nankai University, Tianjin 300071, China
(Received 22 May 2019; published 16 January 2020)
The transition metal thiophosphates MPS3 (M ¼ Mn, Fe, Ni) are a class of van der Waals stacked
insulating antiferromagnets that can be exfoliated down to the ultrathin limit. MnPS3 is particularly
interesting because its Ne´el ordered state breaks both spatial-inversion and time-reversal symmetries,
allowing for a linear magnetoelectric phase that is rare among van der Waals materials. However, it is
unknown whether this unique magnetic structure of bulk MnPS3 remains stable in the ultrathin limit. Using
optical second harmonic generation rotational anisotropy, we show that long-range linear magnetoelectric
type Ne´el order in MnPS3 persists down to at least 5.3 nm thickness. However an unusual mirror symmetry
breaking develops in ultrathin samples on SiO2 substrates that is absent in bulk materials, which is likely
related to substrate induced strain.
DOI: 10.1103/PhysRevLett.124.027601
Thin film materials that exhibit the magnetoelectric (ME)
effect—a coupling between magnetic (electric) polarization
and external electric (magnetic) fields—have potentially
broad applications in spintronics, sensing, and energy
harvesting technologies [1,2]. Although ME effects in
single-phase bulk crystals have been continuously pursued
since their discovery in Cr2O3 in 1960 [3,4], advances in
thin film deposition techniques over the past two decades
have opened new pathways to stabilize and to control high
quality materials with large ME coupling strengths via
epitaxial strain and heterostructure engineering, allowing
the possibility of integration into functional nanoscale
devices. At present, searching for both single-phase and
composite thin film materials with stronger ME coupling,
and developing methods to scale them down to the ultrathin
few-unit-cell limit, remain active areas of research.
The recent discoveries of long-range magnetic ordering
in exfoliated van der Waals (vdW) semiconductors [5–8]
potentially offer a new route to realizing ME materials in
the ultrathin limit. The simplest type of ME effect, which
involves a linear coupling between the external field and
induced polarization, is allowed in materials that lack both
spatial-inversion and time-reversal symmetries. As most
of the naturally occurring vdW crystals are structurally
centrosymmetric, a convenient strategy is to rely on the
magnetic ordering itself to break inversion symmetry. This
suggests that one should focus on antiferromagnetic (AF)
rather than ferromagnetic (FM) materials because the
latter generally do not break the inversion symmetry of
the underlying lattice. It was recently reported that upon
exfoliating CrI3 down to a single bilayer, its magnetic order
transforms from being FM to AF, breaking inversion
symmetry and turning on a linear ME coupling in the
process [5,9–11]. However, so far there are no reports of an
ultrathin material that directly inherits linear ME properties
from its bulk precursor.
The transition metal thiophosphates MPS3 (M ¼ Mn,
Fe, Ni) present an interesting family of AF vdW materials
for such a study [12–15]. While the AF orders in FePS3 and
NiPS3 preserve inversion symmetry [16,17], neutron dif-
fraction studies have shown that the AF order in bulk
MnPS3 breaks inversion symmetry and allows a linear ME
effect [18]. However, it is not clear if the linear ME-type AF
order persists down to the ultrathin limit. Because such
order does not exhibit any net magnetization, a magneto-
optical Kerr rotation experiment is not applicable. Although
Raman spectroscopy has detected phonon anomalies in
ultrathin MnPS3 that are potentially associated with AF
ordering [19], and spin transport measurements have shown
evidence of persistent magnons in few layer MnPS3 devices
[20], a technique that directly probes the AF structure in
nanoscopic exfoliated samples is still urgently anticipated.
Leveraging the sensitivity of optical second harmonic
generation (SHG) to AF order [21], we demonstrate here
PHYSICAL REVIEW LETTERS 124, 027601 (2020)
0031-9007=20=124(2)=027601(6) 027601-1 © 2020 American Physical Society
that SHG rotational anisotropy (RA) can directly couple to
the AF order parameter in MnPS3 nanoflakes, and use it to
show that the linear ME-type AF order found in bulk
MnPS3 persists down to the ultrathin limit.
Bulk MnPS3 crystallizes in a monoclinic structure with
centrosymmetric 2=m point group symmetry [18]. It has a
twofold rotation axis along the crystallographic b axis and
a mirror plane perpendicular to bˆ [Figs. 1(a) and 1(b)]. The
Mn atoms are octahedrally coordinated by S atoms and
form a honeycomb lattice in the ab plane, but the in-plane
sixfold rotational symmetry of the honeycomb lattice is
absent in the bulk crystal due to the displacement of
adjacent layers along aˆ. Despite the similar lattice struc-
tures of MnPS3, FePS3, and NiPS3 [Fig. 1(c)], MnPS3 hosts
an inversion broken Ne´el-type AF order [18] whereas
FePS3 and NiPS3 exhibit inversion symmetric zigzag-type
AF order [16,17]. The Ne´el-type AF order preserves the
size of the unit cell and exhibits no net moment, therefore it
is challenging to detect via Raman spectroscopy and
magneto-optical Kerr rotation, respectively. However,
because it breaks inversion symmetry, it should exhibit a
finite second-order electric-dipole (ED) susceptibility that
is responsible for SHG [22]. Therefore, we expect to see a
finite SHG yield below the AF ordering temperature TAF
from MnPS3 and but not FePS3 and NiPS3.
The SHG-RA experiments were performed with a Ti:
sapphire oscillator delivering laser pulses with a photon
energy of ℏω ¼ 1.5 eV, a pulse width of 80 fs, and a
repetition rate of 80 MHz. The SHG photons produced at
3 eV are resonant with the band gap of MnPS3 [23].
A 5× (50×) microscope objective was used to focus light
onto the bulk (exfoliated) samples at normal incidence
with a spot size of approximately 30 μm (2 μm), and the
intensity of the reflected SHG beam was measured using a
photomultiplier tube. The pulse energy of the incoming
beam was kept below 50 pJ. The SHG-RA patterns were
acquired by rotating the linear polarization of the incoming
and outgoing beams (parametrized by the angle ϕ), which
were maintained parallel to each other in the ab plane
[Fig. 1(c)]. BulkMPS3 single crystals were grown by a self-
flux method described elsewhere [24].
Despite having a centrosymmetric crystallographic point
group, we observe weak but finite SHG-RA signals from all
three bulk crystals even above TAF [Fig. 2(a)]. This may
arise from surface ED SHG or higher-rank bulk SHG
processes such as electric-quadrupole (EQ) SHG [22], both
of which are generally allowed in centrosymmetric materi-
als and were found to fit the data equally well [Fig. 2(a)]
[25]. For simplicity, we therefore only consider the bulk EQ
term in our later fitting. The loss of sixfold rotational
symmetry that arises from the stacking offset between
adjacent honeycomb layers is apparent in the data, although
the degree of departure from sixfold symmetry varies across
samples as well as across spots within a single sample. We
speculate that this may be due to spatial variations in the
strength of interlayer coupling and/or variations in the
concentration of 120° twins or stacking faults [29].
Below TAF we observe no changes in the SHG intensity
from both FePS3 and NiPS3, but an increase in the SHG
intensity fromMnPS3 as anticipated. As shown in Fig. 2(b),
the low temperature SHG-RA patterns from MnPS3 can be
well fit using the coherent sum of a nonmagnetic EQ
contribution and an AF order induced time-noninvariant
ED contribution described by the equation [25]
P2ωi ¼ χEQijklEωj ∇kEωl þ χEDijkðTÞEωj Eωk ; ð1Þ
where P2ω is the induced electric polarization at the SHG
frequency, Eω is the magnitude of the incident electric
field, χEQijkl is the temperature independent EQ susceptibil-
ity from a 2=m crystallographic point group, and χEDijk ðTÞ is
a temperature dependent ED susceptibility from the
20=m magnetic point group describing the Ne´el phase
[18]. As shown in Fig. 2(c), the SHG intensity from
MnPS3 shows an order parameterlike increase below TAF.
Since χEDijk ðTÞ is directly proportional to the inversion
broken Ne´el order parameter, we can extract the critical
exponent of the order parameter (β) by fitting the temper-
ature dependent SHG intensity to the phenomenological
function I2ω ∝ ½aþ bðTAF − TÞβ2, where a is fixed by the
intensity of the EQ contribution above TAF and both b and
β are free parameters. Best fits to the region 60 K ≤ T ≤
TAF yield β ¼ 0.37ð8Þ [Fig. 2(c)], which is close to the
(a)
(c)  
MnPS3 FePS3 NiPS3E( ) a
Inversion center
c
b
M P S
a
b
c
a
b
c
(b)
FIG. 1. Crystal and magnetic structure of MPS3. MPS3 lattice
viewed along the (a) c and (b) b axis. Adjacent ab planes are
displaced by a=3 along the aˆ direction. (c) AF structures of
MPS3. Arrows denote spin orientation. Star denotes an inversion
center of the AF structure. The inset shows the in-plane
orientation of the incident electric field.
PHYSICAL REVIEW LETTERS 124, 027601 (2020)
027601-2
numerical calculation of ∼0.369 for the 3D Heisenberg
model [30].
To investigate whether the long-range Ne´el order in
MnPS3 survives in the ultrathin limit, we exfoliated bulk
crystals onto an amorphous SiO2 substrate in a nitrogen
purged glove box. The choice of pure SiO2 over SiO2=Si as
a substrate was made to reduce laser induced heating
arising from optical absorption by Si at 800 and
400 nm. In contrast, SiO2 is transparent to both 800 and
400 nm light. Because of the poor thermal conductivity
of SiO2 and the relatively high laser power needed for
our SHG-RA measurements on MnPS3 compared to other
optical techniques for studying vdW magnets such as
magneto-optical Kerr microscopy or Raman spectroscopy,
we face more stringent sample cooling demands [25]. To
increase cooling efficiency, we deposited gold rings around
the MnPS3 flakes, which are thermally anchored to the
cryostat sample holder by gold electrodes. Figure 3(a)
shows an optical image of a typical device. Using atomic
force microscopy, we identified ultrathin MnPS3 nanoflakes
with 5.3 and 12.5 nm step sizes above the substrate on this
device [Fig. 3(b)]. Based on previously published atomic
force microscopy data on MnPS3 [24], these correspond to 7
and 16 single layers of MnPS3, respectively. Figure 3(c)
shows typical SHG-RA patterns obtained from these flakes
at a temperature of 10 K, compared with both thicker
(75 nm) flakes and the bare substrate. We find that the
overall SHG intensity approximately scales with the sample
thickness, consistent with a bulk dominated SHG signal. The
SiO2 substrate contributes an isotropic background and is
thus easily distinguished from the MnPS3 signal.
As shown in Fig. 2(c), the ED SHG signal from MnPS3
below TAF is of comparable magnitude to the high temper-
ature EQ signal and is thus relatively weak overall. This is
likely related to our incident 1.5 eV photon energy being
well below the band gap (∼3 eV) of MnPS3. Consequently,
when we attempted to protect the MnPS3 flakes by
T > T(b)
(c)
200 K
0.4 0.6 0.8 1.0 1.2
1
2
3 NiPS3 (TAF = 155 K)
MnPS3 (TAF = 78 K)
FePS3 (TAF = 123 K)
T/TAF
SH
G
 In
te
ns
ity
 (N
orm
)
10 K 30 K 50 K 100 K
50 K 90 K 130 K 150 K
50 K 100 K 140 K 180 K
0.2
0
1
0.5
2.5
1
1 1
1
MnPS3 FePS3 NiPS3
1
(a)
AF T < TAF
0°
FIG. 2. SHG-RA patterns and long-range Ne´el order in MnPS3.
(a) SHG-RA patterns of MPS3 above and (b) below their
respective AF ordering temperatures: TAF ¼ 78 (MnPS3), 123
(FePS3), and 155 K (NiPS3). Solid circles are experimental data
and the solid lines are best fits to the phenomenological model
described in the main text. For T > TAF, all data were fit using
only an EQ term. For T < TAF, the FePS3 and NiPS3 data were fit
using only an EQ term, whereas the MnPS3 data were fit using a
coherent sum of an EQ and ED term. (c) Temperature dependence
of the SHG intensity along the ϕ ¼ 60° direction. Solid line on
the MnPS3 data is a best fit to the power law function described in
the main text, which accounts for a constant EQ term and a
temperature dependent ED term.
2
x20
x20
SiO2
75 nm
12.5 nm
SiO
12.5 nm
5.3 nm
5 µm
(b)
(c)(a) 50 µm
MnPS3
5.3 nm
x1
x5
SiO2
FIG. 3. MnPS3 nanodevice for probing long-range Ne´el order.
(a) Optical image of exfoliated MnPS3 flakes on SiO2 with gold
ring and electrode on top to improve cooling efficiency. The 5.3
and 12.5 nm flakes are found within the white box. (b) Atomic
force microscopy scan of the area bounded by the white box in
(a). Green lines indicate the positions of line scans, with
corresponding magnified line profiles shown in white.
(c) SHG-RA patterns from various regions of the device at 10 K.
PHYSICAL REVIEW LETTERS 124, 027601 (2020)
027601-3
encapsulation with a hexagonal boron nitride (hBN) thin
flake, we found that the SHG signal was dominated by the
hBN. Therefore we had to work with exposed MnPS3
flakes, which are more prone to degradation. At cryogenic
temperatures, we found that the SHG intensity from the
few-layer regions starts to decrease over a timescale of
several hours. This is likely due to surface adsorption of gas
molecules and/or chemical reaction processes activated by
laser exposure, as is observed in CrI3 nanoflakes [31].
Therefore we were only able to acquire a limited number
of SHG-RA scans at low temperatures before the onset of
sample degradation. Nevertheless, our data clearly show an
order parameterlike increase in the SHG intensity from the
MnPS3 nanoflakes below a temperature close to the bulk
TAF value (Fig. 4), which again saturates at only several
times the high temperature value. This indicates that the
linear ME-type Ne´el ordering observed in bulk crystals
persists at least down to 7 layer thick samples.
Measurements collected from 3 layer samples also show a
markedly higher SHG intensity at 10 K compared to 100 K
[25], but their faster degradation prevented a full temperature
dependence measurement from being taken.
Given that the low temperature SHG signal from MnPS3
involves the interference between a time-noninvariant ED
response and a time-invariant EQ response, the existence
of two different 180° AF domains related by time reversal
should produce different SHG intensities, analogous to
what has previously been observed in Cr2O3 [32].
Assuming that the roughly twofold SHG intensity increase
in our 5.3 nm flakes at low temperature [Fig. 4(d)] arises
from a pure þ domain where the ED and EQ contributions
interfere constructively, the ratio of the ED to EQ SHG
electric fields should be roughly 1
2
. By extension, the SHG
intensity from a − domain is expected to be ∼25% of the
high temperature value, or ð1þ 1
2
Þ2=ð1 − 1
2
Þ2 ¼ 9 times
lower in intensity compared to the þ domain. By raster
scanning our beam of spot size ∼2 μm over the roughly
4 μm × 4 μm area of our 5.3 nm MnPS3 flake at 10 K, we
found the SHG intensity varies by only approximately
30% about the mean intensity. These small variations
may be due to sample inhomogeneity and/or slight changes
in alignment and are inconsistent with 180° domains.
Therefore we believe our flake to be a single AF domain,
which is comparable to the FM domain sizes observed in
ultrathin vdW materials like CrI3 [5] and Cr2Ge2Te6 [6].
We note that the low temperature SHG-RA patterns from
the ultrathin 5.3 nm flakes exhibit an unusual symmetry. In
particular, the ac mirror plane (reflection about the hori-
zontal line in the SHG-RA patterns) that is preserved by
both the 2=m and 20=m point groups is absent [Fig. 3(c)].
This mirror symmetry breaking only becomes apparent as
the material thickness is reduced and as the temperature is
lowered below TAF [Fig. 4(c)]. We believe that this is
related to a substrate induced strain because for ultrathin
MnPS3 flakes that are exfoliated onto SiO2=Si substrates,
which are much smoother than pure SiO2 substrates, there
is no clear evidence of ac mirror breaking in the low
temperature SHG-RA patterns [25]. Model Hamiltonian
calculations [33] show that a spiral spin texture that breaks
acmirror symmetry is favored over the collinear Ne´el order
only if the second nearest-neighbor Dzyaloshinskii-Moriya
interaction D2 is comparable to the nearest-neighbor
exchange J1 in MnPS3, which is around 1.5 meVaccording
to inelastic neutron diffraction experiments [34]. However,
spin Hall based measurements ofD2 in MnPS3 put its value
at merely 0.3 meV [35]. Since it is unlikely that D2 is
several times larger in ultrathin flakes compared to bulk
crystals, especially given that Raman spectroscopy studies
show no drastic changes in TAF or the phonon spectrum as a
function of thickness [19], we rule out a noncollinear spin
texture as the cause for ac mirror breaking. Instead, it is
possible that the substrate induced strain tilts the easy axis,
causing the Ne´el ordered moments to rigidly cant out of the
ac plane. Further structural and magnetic characterization
of thin MnPS3 flakes will be necessary to confirm this
hypothesis.
In conclusion, we have demonstrated SHG-RA to be a
direct and effective probe of inversion breaking AF order
4 2.5
(a)
(b)
(c)
(d)
Temperature (K)
75 nm 12.5 nm 5.3 nm
75
 n
m
 
5.
3 
nm
10 K 50 K 60 K 70 K 120 K
SH
G
 In
te
ns
ity
 (N
orm
)
 
12
.5
 n
m
0 40 80
1
7
0 40 80 0 40 80 120
FIG. 4. Long-range Ne´el order in few-layer MnPS3. (a)–(c) Temperature dependent SHG-RA patterns from MnPS3 flakes of different
thicknesses. (d) Normalized temperature dependence of SHG Intensity along the ϕ ¼ 0° direction.
PHYSICAL REVIEW LETTERS 124, 027601 (2020)
027601-4
parameters in exfoliated vdW materials. A linear
ME-type Ne´el order that features in bulk crystals of
MnPS3 was found to survive down to the few layer limit.
Future quantitative measurements of the ME coupling
strength in ultrathin MnPS3 samples will help to assess
its potential for applications in nanoscale spintronics and
optoelectronics devices.
Work at Caltech and UCSB was supported by ARO
MURI Grant No. W911NF-16-1-0361. Work at GISTwas
supported by National Research Foundation of Korea
(NRF) grants funded by the Korea government (MSIP)
(No. 2018R1A2B2005331). Work at IBS CCES was
supported by Institute for Basic Science (IBS) in Korea
(IBS-R009-G1). D. H. and J. S. L. also acknowledge
support from a GIST-Caltech collaborative grant.
J. O. I. acknowledges the support of the Netherlands
Organization for Scientific Research (NWO) through
the Rubicon grant, Project No. 680-50-1525/2474.
*Corresponding author.
dhsieh@caltech.edu
[1] M. Fiebig, Revival of the magnetoelectric effect, J. Phys. D
38, R123 (2005).
[2] N. A. Spaldin and R. Ramesh, Advances in magnetoelectric
multiferroics, Nat. Mater. 18, 203 (2019).
[3] I. E. Dzyaloshinskii, On the magneto-electrical effects in
antiferromagnets, Sov. Phys. JETP 10, 628 (1960).
[4] D. N. Astrov, The magnetoelectric effect in antiferromag-
netics, Sov. Phys. JETP 11, 708 (1960).
[5] B. Huang et al., Layer-dependent ferromagnetism in a van
der Waals crystal down to the monolayer limit, Nature
(London) 546, 270 (2017).
[6] C. Gong et al., Discovery of intrinsic ferromagnetism in
two-dimensional van der Waals crystals, Nature (London)
546, 265 (2017).
[7] K. S. Burch, D. Mandrus, and J.-G. Park, Magnetism in two-
dimensional van der Waals materials, Nature (London) 563,
47 (2018).
[8] M. Gibertini, M. Koperski, A. F. Morpurgo, and K. S.
Novoselov, Magnetic 2D materials and heterostructures,
Nat. Nanotechnol. 14, 408 (2019).
[9] B. Huang et al., Electrical control of 2D magnetism in
bilayer CrI3, Nat. Nanotechnol. 13, 544 (2018).
[10] S. Jiang, L. Li, Z. Wang, K. Mak, and J. Shan, Controlling
magnetism in 2D CrI3 by electrostatic doping. Nat. Nano-
technol. 13, 549 (2018).
[11] Z. Sun et al., Giant nonreciprocal second harmonic gen-
eration from layered antiferromagnetism in bilayer CrI3,
Nature (London) 572, 497 (2019).
[12] X. Li, T. Cao, Q. Niu, J. Shi, and J. Feng, Coupling the
valley degree of freedom to antiferromagnetic order, Proc.
Natl. Acad. Sci. U.S.A. 110, 3738 (2013).
[13] N. Sivadas, M.W. Daniels, R. H. Swendsen, S. Okamoto,
and D. Xiao, Magnetic ground state of semiconducting
transition-metal trichalcogenide monolayers, Phys. Rev. B
91, 235425 (2015).
[14] B. L. Chittari, Y. Park, D. Lee, M. Han, A. H. MacDonald,
E. Hwang, and J. Jung, Electronic and magnetic properties
of single-layer MPX3 metal phosphorous trichalcogenides,
Phys. Rev. B 94, 184428 (2016).
[15] J.-U. Lee, S. Lee, J. H. Ryoo, S. Kang, T. Y. Kim, P. Kim,
C.-H. Park, J.-G. Park, and H. Cheong, Ising-type magnetic
ordering in atomically thin FePS3, Nano Lett. 16, 7433
(2016).
[16] A. R. Wildes et al., Magnetic structure of the quasi-two-
dimensional antiferromagnet NiPS3, Phys. Rev. B 92,
224408 (2015).
[17] D. Lancon, H. C. Walker, E. Ressouche, B. Ouladdiaf, K. C.
Rule, G. J. McIntyre, T. J. Hicks, H. M. Rønnow, and A. R.
Wildes, Magnetic structure and magnon dynamics of the
quasi-two-dimensional antiferromagnet FePS3, Phys. Rev.
B 94, 214407 (2016).
[18] E. Ressouche, M. Loire, V. Simonet, R. Ballou, A. Stunault,
and A. Wildes, Magnetoelectric MnPS3 as a candidate for
ferrotoroidicity, Phys. Rev. B 82, 100408(R) (2010).
[19] K. Kim et al., Antiferromagnetic ordering in van der
Waals two-dimensional magnetic material MnPS3
probed by Raman spectroscopy, 2D Mater. 6, 041001
(2019).
[20] W. Xing, L. Qiu, X. Wang, Y. Yao, Y. Ma, R. Cai, S. Jia,
X. C. Xie, and W. Han, Magnon Transport in Quasi-Two-
Dimensional van der Waals Antiferromagnets, Phys. Rev. X
9, 011026 (2019).
[21] M. Fiebig, V. V. Pavlov, and R. V. Pisarev, Second-harmonic
generation as a tool for studying electronic and magnetic
structures of crystals: Review, J. Opt. Soc. Am. B 22, 96
(2005).
[22] Y. R. Shen, The Principles of Nonlinear Optics (Wiley,
Hoboken, New Jersey, 2003).
[23] V. Grasso, F. Neri, P. Perillo, L. Silipigni, and M. Piacentini,
Optical-absorption spectra of crystal-field transitions in
MnPS3 at low temperatures, Phys. Rev. B 44, 11060
(1991).
[24] S. Lee, K.-Y. Choi, S. Lee, B. H. Park, and J.-G. Park,
Tunneling transport of mono- and few-layers magnetic van
der Waals MnPS3, APL Mater. 4, 086108 (2016).
[25] See Supplemental Material at http://link.aps.org/
supplemental/10.1103/PhysRevLett.124.027601 for ex-
tended derivations and experimental details, which includes
Refs. [26–28].
[26] L. Zhao, C. A. Belvin, R. Liang, D. A. Bonn, W. N. Hardy,
N. P. Armitage, and D. Hsieh, Evidence of an odd-parity
hidden order in a spin-orbit coupled correlated iridate, Nat.
Phys. 12, 32 (2016).
[27] A.-Y. Lu et al., Janus monolayers of transition metal
dichalcogenides, Nat. Nanotechnol. 12, 744 (2017).
[28] G. De Luca, N. Strkalj, S. Manz, C. Bouillet, M. Fiebig,
and M. Trassin, Nanoscale design of polarization in
ultrathin ferroelectric heterostructures, Nat. Commun. 8,
1419 (2017).
[29] C. Murayama, M. Okabe, D. Urushihara, T. Asaka, K.
Fukuda, M. Isobe, K. Yamamoto, and Y. Matsushita,
Crystallographic features related to a van der Waals cou-
pling in the layered chalcogenide FePS3, J. Appl. Phys. 120,
142114 (2016).
PHYSICAL REVIEW LETTERS 124, 027601 (2020)
027601-5
[30] M. Campostrini, M. Hasenbusch, A. Pelissetto, P. Rossi, and
E. Vicari, Critical exponents and equation of state of the
three-dimensional Heisenberg universality class, Phys. Rev.
B 65, 144520 (2002).
[31] D. Shcherbakov et al., Raman spectroscopy, photocatalytic
degradation, and stabilization of atomically thin chromium
tri-iodide, Nano Lett. 18, 4214 (2018).
[32] M. Fiebig, D. Frohlich, G. Sluyterman v. L., and R. V.
Pisarev, Domain topography of antiferromagnetic Cr2O3
by second-harmonic generation, Appl. Phys. Lett. 66, 2906
(1995).
[33] R. Cheng, S. Okamoto, and D. Xiao, Spin Nernst Effect of
Magnons in Collinear Antiferromagnets, Phys. Rev. Lett.
117, 217202 (2016).
[34] A. R. Wildes, B. Roessli, B. Lebech, and K.W. Godfrey,
Spin waves and the critical behaviour of the magnetization
in MnPS3, J. Phys. Condens. Matter 10, 6417 (1998).
[35] Y. Shiomi, R. Takashima, and E. Saitoh, Experimental
evidence consistent with a magnon Nernst effect in the
antiferromagnetic insulator MnPS3, Phys. Rev. B 96,
134425 (2017).
PHYSICAL REVIEW LETTERS 124, 027601 (2020)
027601-6