PHYSICAL REVIEW X 12, 040501 (2022)

Emerging Research Landscape of Altermagnetism

Libor Šmejkal,1,2 Jairo Sinova,1,2 and Tomas Jungwirth 2,3
1
Institut für Physik, Johannes Gutenberg Universität Mainz, D-55099 Mainz, Germany
2
Institute of Physics, Czech Academy of Sciences, Cukrovarnická 10, 162 00 Praha 6 Czech Republic
3
School of Physics and Astronomy, University of Nottingham, Nottingham NG7 2RD, United Kingdom

(Received 15 March 2022; revised 15 August 2022; published 8 December 2022)

Magnetism is one of the largest, most fundamental, and technologically most relevant fields of
condensed-matter physics. Traditionally, two basic magnetic phases have been distinguished ferromag-
netism and antiferromagnetism. The spin polarization in the electronic band structure reflecting the
magnetization in ferromagnetic crystals underpins the broad range of time-reversal symmetry-breaking
responses in this extensively explored and exploited type of magnets. By comparison, antiferromagnets
have vanishing net magnetization. Recently, there have been observations of materials in which strong
time-reversal symmetry-breaking responses and spin-polarization phenomena, typical of ferromagnets, are
accompanied by antiparallel magnetic crystal order with vanishing net magnetization, typical of
antiferromagnets. A classification and description based on spin-symmetry principles offers a resolution
of this apparent contradiction by establishing a third distinct magnetic phase, dubbed altermagnetism. Our
perspective starts with an overview of the still emerging unique phenomenology of this unconventional
d-wave (or higher even-parity wave) magnetic phase, and of the wide array of altermagnetic material
candidates. We illustrate how altermagnetism can enrich our understanding of overarching condensed-
matter physics concepts and how it can have impact on prominent condensed-matter research areas.
DOI: 10.1103/PhysRevX.12.040501 Subject Areas: Condensed Matter Physics
Magnetism, Spintronics

I. INTRODUCTION space. This makes the conventional antiferromagnets, in

many aspects, akin to nonmagnetic materials and invisible
Magnetic solids are traditionally divided into two
to the macroscopic electrical or optical probes commonly
elementary phases—ferromagnets and antiferromagnets
used in ferromagnets.
[1]. Ferromagnets, known for several millennia, are char-
Recently, diverse condensed-matter research commun-
acterized by a strong macroscopic magnetization. Because
ities have been intrigued by theoretical predictions of T -
of the spin polarization in the reciprocal momentum space,
symmetry breaking macroscopic phenomena [2–16] and
reflecting the magnetization order parameter in direct
spin-split band structures [2–9,13–25], which are typical
space, they generate a range of phenomena originating
of ferromagnets, in crystals with compensated antiparallel
from electronic band structures with broken time-reversal
magnetic ordering, which is characteristic of antiferro-
(T ) symmetry and spin splitting. Antiferromagnets, on the
magnets. The apparent ferromagnetic-antiferromagnetic
other hand, were discovered only a century ago, because of
dichotomy in these materials challenges the traditional
their vanishing net magnetization, which makes them
division of materials by the two basic magnetic phases.
behave, in many aspects, as nonmagnetic materials. In
A recent symmetry classification and description [16],
conventional antiferromagnetism, a compensating antipar-
employing a generalized symmetry formalism for describ-
allel ordering of atomic magnetic moments in the direct
ing spin arrangements on crystals [26–28], resolves this
physical space, i.e., the effective cancellation of atomic
contradiction. The formalism allows for describing a
moments leading to the vanishingly small macroscopic net
broad landscape of magnetic phases by considering sym-
magnetization, implies the absence of any counterpart spin-
metry transformations that combine generally different
polarization order parameter in the reciprocal momentum
operations in real and spin space. The classification based
on this symmetry formalism establishes, apart from the
traditional ferromagnetism and antiferromagnetism, a third
Published by the American Physical Society under the terms of distinct and comparably abundant phase [16]. This third
the Creative Commons Attribution 4.0 International license.
Further distribution of this work must maintain attribution to phase is characterized by a compensated magnetic order in
the author(s) and the published article’s title, journal citation, direct space with opposite-spin sublattices connected by
and DOI. crystal-rotation symmetries, and by a corresponding

2160-3308=22=12(4)=040501(27) 040501-1 Published by the American Physical Society

ŠMEJKAL, SINOVA, and JUNGWIRTH PHYS. REV. X 12, 040501 (2022)

FIG. 1. Illustrative models of collinear ferromagnetism, antiferromagnetism, and altermagnetism in crystal-structure real space
and nonrelativistic electronic-structure momentum space. (a) Ferromagnetic model with one spin sublattice, and corresponding distinct
spin-group form, nonzero magnetization, T -symmetry breaking momentum-independent spin splitting of bands, and isotropic s-wave
spin-split Fermi surfaces. (b) Antiferromagnetic model with opposite-spin sublattices (red, blue) connected by inversion or translation,
and corresponding distinct spin-group form, zero net magnetization, and T -invariant spin-degenerate bands reminiscent of nonmagnetic
systems. (c) Altermagnetic model with opposite-spin sublattices connected by rotation and not by translation or inversion, and
corresponding distinct spin-group form, zero net magnetization, T -symmetry breaking spin splitting with alternating sign, anisotropic
sublattice spin densities, and anisotropic d (g or i)-wave spin-split Fermi surfaces. See Sec. II B for the definitions of the distinct
spin-group forms Ris .

unconventional spin-polarization order in the reciprocal how an unconventional d-wave counterpart could be
momentum space that reflects the same rotation sym- realized in magnetism (Fig. 2) [29].
metries. The direct-to-reciprocal-space correspondence Earlier considerations focused on possible realizations
results in electronic band structures with broken T sym- of the unconventional d-wave magnetism due to strong
metry and alternating momentum-dependent sign of the electronic correlations [31–33]. As we will discuss, alter-
spin splitting. The alternating spin polarizations in both magnetism is a realization of the unconventional d-wave
direct physical space and reciprocal momentum space, (or high even-parity wave) magnetism that emerges already
characteristic of the third phase, suggest the term alter- on the basic level of an effective single-particle non-
magnetism [16]. The distinction of the three phases is relativistic description of collinear magnets. It is, therefore,
highlighted in Fig. 1. a robust, elementary, magnetically ordered phase. The
Turning to a broader condensed-matter physics context, theoretical prediction of altermagnetism thus complements,
we point out that altermagnetism allows for a realization of in a fundamentally unique way, modern studies of spin
unconventional d-wave magnetism, a long-sought mag- quantum phases associated with more complex, and often
netic counterpart of unconventional d-wave superconduc- more subtle, many-body correlations, relativistic physics,
tivity [29]. Conventional superconductivity and magnetism noncollinear magnetic ordering, topological phenomena, or
are connected by a striking analogy [29], highlighted in frustrated magnetic interactions [34–43]. Simultaneously,
Fig. 2: The electron-electron Cooper pairs forming around altermagnetism can coexist and constructively interplay
the Fermi surface and driving the conventional s-wave with these additional physical complexities.
superconductivity have the counterpart in the majority spin Altermagnetism is expected to be abundant in nature and
electron—minority spin-hole pairs distributed isotropically to occur in both three-dimensional and two-dimensional
around the Fermi surface in the conventional model of crystals, in diverse structural or chemistry types, and in
(s-wave) ferromagnetism [29]. The discovery of the uncon- conduction types covering the whole spectrum from
ventional d-wave superconductivity not only opened an insulators to superconductors. In Sec. II, we give an
entirely new research landscape of this many-body phase overview of the predicted characteristic features, sym-
[30] but also raised a fundamental question of whether and metries, and material landscape of the altermagnetic phase.

040501-2
EMERGING RESEARCH LANDSCAPE OF ALTERMAGNETISM PHYS. REV. X 12, 040501 (2022)

Superconductivity Magnetism

s-wave
Cooper-pair Spin
phase orientation
d-wave

ky
nodes nodes

kx

FIG. 2. Analogy between superconductivity and magnetism. Top: conventional s-wave superconductivity, which has the same
Cooper-pair phase around the Fermi surface and is analogous to conventional magnetism (ferromagnetism) with an excess of one spin
orientation around the Fermi surface. Bottom: unconventional d-wave superconductivity where the Cooper-pair phase changes sign
around the Fermi surface, and the anticipated analogy of unconventional d-wave magnetism where the excess spin orientation changes
sign around the Fermi surface. The unconventional phases have the characteristic nodes highlighted by gray circles. This figure was
inspired by Ref. [29].

The properties connected to the spin-polarized magnetic order and the T -symmetry breaking electronic
T -symmetry broken band structures of altermagnets responses, supporting evidence has been discovered by
open up a potential for previously unanticipated develop- initial experiments [5,10–12], as highlighted in Table I.
ments in a broad condensed-matter physics field. In Apart from the fundamental physics interest, we expect that
Sec. III, we highlight the distinct properties of altermagnets intense experimental research will also be driven by
in the context of overarching physical concepts of lifted the potential impact of altermagnetism on technology.
Kramers spin degeneracy, Fermi-liquid instabilities, elec- Altermagnetism can occur in crystals with common light
tron and magnon quasiparticles, and Berry phase and elements, high-magnetic-ordering temperatures, and strong
nondissipative transport. In Sec. IV, we outline the potential spin coherence, which are among the key prerequisites for
we foresee of altermagnets in selected active research areas, practical device applications [16].
including spintronics, ultrafast photomagnetism, neuro-
morphics, thermoelectrics, field-effect electronics, multi- II. ALTERMAGNETIC PHASE
ferroics, and superconductivity.
While our focus in the following sections is on the A. Ab initio band structures
emerging field of altermagnetism from the theory perspec- In Fig. 3, we show representative nonrelativistic band
tive, we point out here that first measurements have already structures of metallic RuO2 [3,8,13,14,16,18] and insulat-
indicated that altermagnetism can soon become an active ing FeF2 [47] and MnF2 [20,48], on which we illustrate key
experimental field. Shortly after the theoretical predictions characteristics of the spin-polarization order in the alter-
of the possible coexistence of the compensated antiparallel magnetic band structures that break T symmetry. These

TABLE I. Theoretical predictions, supported by experiments, of T -symmetry breaking macroscopic phenomena in RuO2 , and a list of
other altermagnetic materials in which these macroscopic responses were also theoretically predicted. The anomalous Hall effect is
a T -odd off-diagonal component of the electrical conductivity tensor [2]. The T -odd spin current can be generated along or transverse to
an applied electrical bias; an out-of-plane spin current generated by an in-plane electrical bias in an altermagnetic layer can exert a torque
on magnetic moments in an adjacent layer in a multilayer stack [8].

Theory in Experiment in
Macroscopic response RuO2 RuO2 Theory in other materials
Anomalous Hall effect 2019 [3] 2020 [5] SrRuO3 [44], Mn5 Si3 [6], κ-Cl [45], (Cr,Fe)Sb2 [7], perovskites [46]
Spin current and torque 2020 [8,16] 2021 [10–12] κ-Cl [4], CaCrO3 [9]

040501-3
ŠMEJKAL, SINOVA, and JUNGWIRTH PHYS. REV. X 12, 040501 (2022)

(a) (c) (d)

(b) (e) (f)

FIG. 3. (a) Schematics of the rutile XY 2 crystal structure with antiparallel magnetic moments on X A and X B magnetic sublattices.
(b) Brillouin zone of the rutile crystal and ab initio nonrelativistic calculation of a wave-vector kz ¼ 0 cut of the anisotropic d-wave spin-
polarized Fermi surface of metallic RuO2 . (c) Ab initio altermagnetic spin splitting of bands in RuO2 , calculated without (red and blue)
and with (black) relativistic spin-orbit coupling. (d) Ab initio altermagnetic spin-split Fermi surface for selected kz values in RuO2 with
correlations accounted for within the dynamical mean-field theory. (d,e) Ab initio altermagnetic spin-split bands of insulating FeF2 and
MnF2 , respectively. This figure is adapted from Refs. [3,18,20,47].

altermagnetic material examples belong to the family of exceed, by an order of magnitude, the record relativistic
crystals with rutile structure [Fig. 3(a)]. For several insu- spin splittings in bulk crystals with heavy elements [55].
lating members of the rutile family, the compensated Moreover, unlike the momentum-dependent spin textures
antiparallel arrangement of magnetic moments [Fig. 3(a)] in the relativistic bands, spin is a good quantum number,
was well known already to Néel and his contemporaries and the electronic states share a common momentum-
who, ironically, introduced them into the literature as a independent spin quantization axis in the nonrelativistic
classic representation of antiferromagnetism [49,50]. The bands of altermagnets.
notion was based on focusing on the lattice of magnetic The spin-split parts of the altermagnetic band structure
atoms alone while omitting the essential role of non- are accompanied by spin degeneracies along certain sur-
magnetic atoms on magnetism in the rutile crystals. This faces in the Brillouin zone. In Sec. II B, we show that
may be one of the reasons why the unconventional spin- the altermagnetic spin-group symmetries of the given
polarization order in the reciprocal momentum space, crystal characterize the prominent symmetries of the
resulting in the T -symmetry breaking and alternating spin-polarization order in the electronic band structure [16].
spin splitting of their nonrelativistic band structures Fermi-surface cuts shown in Figs. 3(b) and 3(d) highlight
[Figs. 3(b)–3(f)], remained unnoticed for nearly a century. the typical anisotropic nature of the spin-polarized Fermi
Remarkably, the room-temperature antiparallel magnetic surfaces, with an equal number of states in the opposite spin
ordering in metallic rutile RuO2 was discovered [51,52] and channels, and with spin-momentum locking that is even
investigated [53,54] only recently. The subsequent theo- under the inversion of the momentum and breaks T
retical and experimental exploration of the T -symmetry symmetry.
breaking electronic responses [3,5,8,10–14,16,18,19] has The altermagnetic phase is robust in a broad range of
made RuO2 one of the workhorse materials of the emerging materials, as it can be described within the effective single-
research of altermagnetism. particle Kohn-Sham theory, and the nonrelativistic crystal
Figures 3(b)–3(f) show that the altermagnetic spin split- potential can play a dominant role in both uncorrelated and
ting is strongly momentum dependent in all three rutiles. correlated, and in both clean and disordered altermagnets.
In RuO2 , it reaches, in parts of the Brillouin zone, This is illustrated, for example, by ab initio calculations in
close to 1-eV scale, which is comparable to the spin- RuO2 shown in Figs. 3(c) and 3(d). The calculations
splitting magnitudes in ferromagnets. Unlike ferromagnets, demonstrate that the altermagnetic spin splitting is only
however, the altermagnetic spin splitting in the nonrelativ- weakly affected by the relativistic spin-orbit coupling and
istic bands is accompanied by a symmetry-protected zero net that the prominent features of the altermagnetic spin-
magnetization. momentum locking are preserved when including correla-
Figure 3 also illustrates that spin splittings in altermag- tion effects beyond the local-spin-density approximation of
nets, which are of the strong nonrelativistic origin, can the density-functional theory [3,18,20]. A stable itinerant

040501-4
EMERGING RESEARCH LANDSCAPE OF ALTERMAGNETISM PHYS. REV. X 12, 040501 (2022)

altermagnetism is further confirmed in calculations without

Hubbard correlations in other material candidates, such as (a) (b)
Mn5 Si3 [6] or KRu4 O8 [16]. A sizable altermagnetic spin
splitting also survives in the presence of a strong alloying
disorder, as shown in altermagnetic Cr0.15 Fe0.85 Sb2 [7].
This robustness can be understood by the fact that the
altermagnetic spin-group symmetries, discussed in the
following section, can hold equally well for the effective
single-particle Kohn-Sham potential, as well as for the
Dyson-equation description of many-body systems.

B. Symmetry classification and description

We now move from the microscopic ab initio theory to a
symmetry-based framework for classification and descrip-
tion of altermagnetism as a distinct phase from ferromag-
netism and antiferromagnetism [16]. The framework allows
for describing a broad landscape of phases by considering
symmetry transformations in real space, or in spin space, or
transformations that combine generally different operations FIG. 4. (a) Schematic top view (upper panel) and 3D view
in real and spin space [26–28]. Classical examples of the (lower panel) of the RuO2 crystal with opposite spin directions on
first two cases are liquid and solid phases of matter, RuA and RuB sublattices depicted in red and blue, oxygen atoms
distinguished by the presence vs absence of a real-space shown in black, and with the depicted nonrelativistic spin-group
rotation symmetry, and nonmagnetic vs ferromagnetic symmetries corresponding to spin group 24=1m2m1m in the
phases, distinguished by the presence vs absence of a notation of Ref. [16]. The curved red arrow and its label highlight
spin-space rotation symmetry. Below, we show that the the generator of opposite-spin sublattice transformations, and the
classification and description of the distinct ferromagnetic, generators of the halving subgroup of same-spin sublattice
antiferromagnetic, and altermagnetic phases will consider transformations are also highlighted (in black). Here, C2 on
the combinations of generally different operations acting the left of the double vertical bar is a 180° spin-space rotation
transformation around an axis perpendicular to the spins, and E is
simultaneously in real and spin space [16].
the spin-space identity. On the right of the double vertical bar,
Microscopically, the above symmetry formalism corre- C4z t is a fourfold real-space rotation combined with translation,
sponds to nonrelativistic quantum mechanics in which real and M i are real-space mirror transformations. (b) Schematic spin
space and spin space are decoupled. In magnetism, non- arrangement on the RuO2 crystal with antiparallel (upper panel)
relativistic physics typically dominates over the relativistic and parallel (lower panel) spin directions and the crystallographic
spin-orbit coupling [26,56]. Nevertheless, the correspond- spin-axis orientation depicted by red and blue arrows, and with
ing nonrelativistic spin-group formalism [26–28] has been the depicted generators of the relativistic magnetic symmetry
employed only sporadically in the magnetic literature group m0 m0 m. Here, T is time reversal. The crossed arrow
[16,24,25,56–61], in contrast to the commonly employed highlights that the magnetic group contains no opposite spin-
relativistic magnetic groups [62–67]. sublattice transformation elements. The same magnetic group
describes a fully compensated antiparallel magnetic order, a
The nonrelativistic spin groups consider pairs of trans-
parallel magnetic order with a strong nonrelativistic ferromag-
formations ½Ri kRj , where the transformation on the left of netic moment, as well as an antiparallel magnetic order with a
the double vertical bar acts only in spin space, and the weak uncompensated relativistic magnetization [16]. This exam-
generally different transformation on the right of the double ple illustrates that the relativistic magnetic groups generally do
vertical bar simultaneously acts only in real space; an not separate between relativistic and nonrelativistic, compensated
example is the important spin-group symmetry of RuO2 , and noncompensated, or collinear and noncollinear magnetic
½C2 kC4z t, highlighted in Fig. 4(a), which combines a phases. This figure is adapted from Ref. [16].
twofold spin-space rotation with a fourfold real-space
rotation (nonsymmorphic with translation labeled by t).
In contrast, the relativistic magnetic groups contain only ½C2 kC4z t spin symmetry [Fig. 4(b)]. This symmetry,
the same transformations acting simultaneously in spin and however, signals the unconventional d-wave magnetic
real space, Ri ≡ ½Ri kRi  [26–28]. Therefore, the magnetic phase in RuO2 and governs prominent responses, such
groups cover a much narrower symmetry landscape than as the giant magnetoresistance [13,16].
the spin groups. Correspondingly, the relativistic magnetic A recent theoretical development has shown that the spin
groups can omit phases and phenomena that emerge when groups lead to the classification and description of three
considering the richer nonrelativistic spin-group formalism. distinct phases of collinear magnetism [16] (Fig. 1): The
For example, the magnetic groups principally omit the first phase has one spin lattice (or opposite-spin sublattices

040501-5
ŠMEJKAL, SINOVA, and JUNGWIRTH PHYS. REV. X 12, 040501 (2022)

TABLE II. Spin-group symmetries and corresponding magnetic crystal-structure and nonrelativistic band-structure characteristics.
The first two lines regard symmetries that apply to all three nonrelativistic collinear phases. The remaining lines apply only to
altermagnets. For the definitions of symbols in the first column, see Sec. II B. This table is adapted from Ref. [16].

Spin-group symmetries Magnetic crystal structure Nonrelativistic band structure

½C∞ kE Collinear Spin is a good quantum number and k independent
½C̄2 kT  Coplanar Invariance under inversion of k
½EkH Sublattice spin-density anisotropy Spin-Fermi-surface anisotropy
½C2 kAH Compensated Broken T and spin splitting at general k
Lk ∩ AH ≠ ∅ Spin degeneracy at high-symmetry k
Lk ∩ AH ¼ ∅ þ=− spin splitting at general k=AHk

not connected by any symmetry transformation). It corre- and the electronic structure is strictly separated into non-
sponds to conventional ferromagnetism (ferrimagnetism) mixing spin-up and spin-down channels [16], as high-
[62]. The second phase has opposite-spin sublattices lighted on the first line of Table II. Since all collinear
connected by translation or inversion and corresponds to magnetic orderings are coplanar, they also have the
conventional antiferromagnetism [1,26,68]. The third phase coplanarity symmetry (½C̄2 kT ) combining the spin-space
has opposite-spin sublattices connected by rotation (proper inversion, i.e., time reversal, with the 180° spin-space
or improper and symmorphic or nonsymmorphic) but not rotation around an axis perpendicular to the spins
connected by translation or inversion. [16,27]. This symmetry protects a general invariance of
The conventional ferromagnetic phase has a net mag- the nonrelativistic electronic energy bands under real-space
netization of a strong nonrelativistic origin in the direct (crystal-momentum) inversion, i.e., ϵðs; kÞ ¼ ϵðs; −kÞ [16]
physical space and a corresponding spin-polarization order (second line of Table II).
in the reciprocal momentum space that is principally The nontrivial spin groups that are specific exclusively to
isotropic (s wave) [Fig. 1(a)]. The electronic structure is the nonrelativistic electronic energy bands of altermagnets
split into majority-spin and minority-spin bands that break are given by Ref. [16], RIII
s ¼½EkHþ½C2 kG−H¼½EkHþ
T symmetry [62]. The conventional antiferromagnetic ½C2 kAH. Here, E on the left of the double vertical bar
phase has a staggered order of the magnetic moments in denotes the spin-space identity transformation, and H is a
the direct space of a strong nonrelativistic origin with a zero subgroup containing half of the real-space transformations
net magnetization and no corresponding spin-polarization of the nonmagnetic crystallographic group G (including the
order in the reciprocal momentum space [Fig. 1(b)]. The real-space identity). The remaining half of the transforma-
electronic energy bands are spin degenerate and T invariant tions, G − H, can be written as AH, where A is a real-space
in the limit of zero relativistic spin-orbit coupling, remi- rotation (proper or improper). In contrast, the nontrivial
niscent of nonmagnetic systems [1,69–73]. The third, spin groups describing the momentum-space electronic
altermagnetic phase has an alternating order of the mag- structure of the conventional ferromagnetic phase have a
netic moments in the direct space of a strong nonrelativistic distinct form, RIs ¼ ½EkG, and those of the conventional
origin with a zero net magnetization in the limit of zero antiferromagnetic phase have another distinct form, RIIs ¼
relativistic spin-orbit coupling, and a corresponding alter- ½EkG þ ½C2 kG [16].
nating spin-polarization order in the reciprocal momentum Because of the inversion symmetry of bands in collinear
space [Fig. 1(c)]. The spin-up and spin-down energy bands magnets, the above classification is limited to G’s that are
are split, break T symmetry, and are equally populated in crystallographic point groups (translations replaced with
the limit of zero relativistic spin-orbit coupling [16]. identity) containing the real-space inversion symmetry
The spin groups contain both symmetries that are (crystallographic Laue groups), independent of whether
common to all three collinear magnetic phases (so-called the magnet does or does not have real-space inversion
spin-only groups) and symmetries corresponding exclu- symmetry [16].
sively to one of the three phases (so-called nontrivial spin In altermagnets, the corresponding nontrivial spin sub-
groups) [16]. Among the common symmetries, arbitrary group ½EkH contains symmetry transformations that
rotations of the spin space around the axis of spins (½C∞ kE interchange atoms belonging only to one of the two spin
with E on the right of the double vertical bar denoting the sublattices. These symmetries determine the characteristic
real-space identity transformation) [16,27] protect spin as a anisotropy of the real-space sublattice spin densities and the
good quantum number with a momentum-independent anisotropy of the individual spin-channel Fermi surfaces
spin-quantization axis across the whole Brillouin zone. (third line of Table II). Note that G − H ¼ AH contains the
The energy bands ϵðs; kÞ can then be indexed by the spin s, other half of the real-space transformations. These are

040501-6
EMERGING RESEARCH LANDSCAPE OF ALTERMAGNETISM PHYS. REV. X 12, 040501 (2022)

generated by a real-space transformation A, which is, structure, composition, or type of order (e.g., magnetic).
exclusively, a rotation (proper or improper) [16]. The As a reference, we can recall how observations in the
corresponding spin-group symmetries ½C2 kAH are trans- early 20th century, initially prompting conflicting notions
formations that interchange atoms between opposite-spin of paramagnetic or ferromagnetic anomalies, were later
sublattices. The presence of symmetry transformations resolved by Néel’s symmetry-based delimitation of the
connecting the opposite-spin sublattices protects the zero antiferromagnetic phase [1]. We can compare this to the
net magnetization of the nonrelativistic magnetic structure recent works that have raised conflicting notions of
[16] (fourth line of Table II). Simultaneously, since the ferromagnetic or antiferromagnetic anomalies, and the
symmetry transformations connecting the opposite-spin resolution of the conflict by the symmetry-based classi-
sublattices include only real-space rotations (proper or fication and description of the third, altermagnetic
improper), they allow for broken T symmetry in the band phase [16].
structure, i.e., ϵðs; kÞ ≠ ϵð−s; −kÞ, and for spin splitting, While above we have only provided a brief summary, a
i.e., ϵðs; kÞ ≠ ϵð−s; kÞ [16] (fourth line of Table II). detailed spin-group classification and description of alter-
The sign of the spin splitting alternates across the magnetism, including a discussion within the general
Brillouin zone, in line with the zero net magnetization, context of phases of condensed-matter systems, is given
which implies the presence of spin-degenerate high- in Ref. [16]. Each phase in a material system exhibits a
symmetry momenta in the band structure. The spin degen- characteristic set of physical properties and responses,
eracy is protected by symmetry when the little group of which, for altermagnetism, are yet to be fully explored.
momentum k (Lk ) contains at least one of the real-space Our understanding of, and outlook on, the distinct phe-
symmetry transformations AH connecting opposite-spin nomenology of altermagnetism in a broad context of basic
sublattices [16] (Lk ∩ AH ≠ ∅; see fifth line of Table II). and applied condensed-matter physics fields will be the
Recall that a little group contains real-space symmetry focus of Secs. III and IV. First, we summarize the basic
transformations that map momentum k onto itself or onto a rules for identifying altermagnets and list representative
momentum that differs from k by a reciprocal lattice vector. material candidates.
For a general momentum k with spin splitting allowed
by symmetry (Lk ∩ AH ¼ ∅; see sixth line of Table II), C. Identification rules
the opposite-spin equal-energy states are, at rotated Elementary rules for identifying the altermagnetic phase
momenta, given by AHk. of a crystal can be summarized as follows:
Reference [16] gives a list of all ten spin Laue groups (1) There is an even number of magnetic atoms in the
RIII
s (or 37 corresponding spin point groups) classifying unit cell, and the number of atoms in the unit cell
and describing altermagnets. They are constructed from does not have to change between the nonmagnetic
only eight different crystallographic Laue groups. The three and magnetic phases of the crystal [cf. two Ru atoms
remaining crystallographic Laue groups, G ¼ 1̄, 3̄, or m3, in the RuO2 unit cell, shown in Fig. 4(a)].
do not allow for the altermagnetic phase. (2) There is no inversion center between the sites
Below, we summarize the basic elements of the algo- occupied by the magnetic atoms from the oppo-
rithm for determining the altermagnetic spin group, which site-spin sublattices [cf. the absence of the inversion
can be constructed by identifying center between the RuA and RuB sites in RuO2
(1) the crystallographic group of the material, because of the oxygen atoms, shown in Fig. 4(a)].
(2) the crystallographic group of the spin sublattice (in (3) The two opposite-spin sublattices are connected
the case of a bipartite lattice, the spin sublattice point by crystallographic rotation transformation, pos-
group corresponds directly to the Wyckoff position sibly combined with translation or inversion
point group), and transformation [cf. the opposite-spin sublattices in
(3) the crystallographic rotation transformation connect- RuO2 connected by C4z t transformation, shown in
ing the opposite-spin sublattices. Fig. 4(a)].
Taking RuO2 as an example, the crystallographic (4) The spin group is determined by the algorithm
point group G ¼ 4=mmm, the sublattice (Wyckoff) described in Sec. II B (cf. the RuO2 spin group with
point group H ¼ mmm, and the crystallographic point- A ¼ C4z and H ¼ mmm).
group rotation generating the symmetry transforma-
tion connecting the opposite-spin sublattices is C4z D. Material candidates
[cf. Fig. 4(a)]. The rules from Sec. II C can be used for high-throughput
Finally, let us note that symmetry is one of the scanning of altermagnetic material candidates. This section
fundamental principles in physics for identifying distinct gives an overview of the predicted range of material types,
phases of matter [62,74] and that a phase is commonly illustrated on specific examples.
associated with a uniform state of a physical system Symmetry prohibits the realization of altermagnetism
distinguished from other phases by, among others, crystal in one-dimensional (1D) chains because of the absence

040501-7
ŠMEJKAL, SINOVA, and JUNGWIRTH PHYS. REV. X 12, 040501 (2022)

(a) (c) of rotation transformations in 1D. On the other hand,

Figs. 6–8, and the list of material candidates given below,
illustrate that altermagnetism can occur in two-dimensional
(2D) and three-dimensional (3D) crystals; the conduction
types can cover the whole spectrum from insulators,
semiconductors, and semimetals, to metals and supercon-
ductors; and the structure and chemistry types can also be
diverse:
(i) quasi-2D oxide insulator V2 Se2 O [15] or semimetal
Cr2 O [75],
(ii) 3D rutile fluoride or oxide insulators FeF2 [47],
MnF2 [20,48], MnO2 [17], and metal RuO2 [3,18],
(b) (d) (iii) perovskite oxide insulators LaMnO3 [22,76],
CaCrO3 [9], and parent cuprate of high-T c super-
conductor La2 CuO4 [16],
(iv) ferrite insulator Fe2 O3 [16],
(v) pnictide with metal-insulator transition FeSb2 [7,16]
and metal CrSb [16],
(vi) chalcogenide semiconductor MnTe [16] and (semi)
metal VNb3 S6 [16] and CoNb3 S6 [3],
(vii) silicide metal Mn5 Si3 [6],
FIG. 5. (a) Spin-polarized relativistic Fermi surface highlight- (viii) organic insulator κ-Cl [4].
ing the presence of the approximate spin symmetry ½C2 jjC4z , Crystallographic and spin groups, and other characteristics
omitted by the magnetic group. (b) Relativistic momentum- of the selected altermagnetic material candidates are
resolved Berry curvature hotspots originating from the avoided summarized in Table III.
crossings along the kx;y ¼ 0 lines, whose position in the mo-
mentum space is determined by the spin symmetries in the
A list of crystallographic symmetry groups that, in
absence of the relativistic spin-orbit coupling. (c) Fermi-surface- principle, allow for hosting the altermagnetic phase is
resolved Berry curvature of the FeSb2 altermagnet illustrates given in Ref. [16]. We also point out that altermagnetism
pronounced contributions to Berry curvature from the quasinodal can occur in structures with inversion symmetry (e.g.,
surface kx ¼ 0, ky ¼ 0. (d) Brillouin zone notation. This figure is rutiles) or without inversion symmetry (e.g., VNb3 S6
adapted from Refs. [5,7,16]. or CoNb3 S6 ).

Insulator La2CuO4

Semiconductor MnTe

Semimetal VNb3S6

Metal FeSb2

3D Layered / Quasi-2D

FIG. 6. Crystal structures of selected altermagnetic candidates organized by dimensionality and conduction type. The crystal structure
are discussed in detail in Refs. [7,16].

040501-8
EMERGING RESEARCH LANDSCAPE OF ALTERMAGNETISM PHYS. REV. X 12, 040501 (2022)

(a) (b)

(c) (d)

FIG. 7. (a)–(d) Ab initio spin-split band structures of depicted altermagnetic candidate materials. This figure is adapted from
Refs. [6,7,15,116].

FIG. 8. Altermagnetic candidates identified from ab initio calculations, organized in a altermagnetic transition temperature vs
altermagnetic spin-splitting strength diagram. The size of the balls scales with the largest atomic number in the crystal. This figure is
adapted from Ref. [16] and references therein.

040501-9
ŠMEJKAL, SINOVA, and JUNGWIRTH PHYS. REV. X 12, 040501 (2022)

TABLE III. Altermagnetic candidates identified from ab initio calculations. We list the nonmagnetic space group, spin point group,
even-parity wave anisotropy, metallic (M) or insulating/semiconducting (I) conduction type, altermagnetic transition temperature, and
altermagnetic spin-splitting magnitude and anisotropy type.

Space group Spin point group Anisotropy Conduction T AM (K) Splitting (meV) References
2 1 1 1
RuO2 P42 =mnm 4= m m m d-wave M 400 1400 [3,18]
KRu4 O8 I4=m 24=1m d-wave M  300 [16]
Mn5 Si3 P63 =mcm 2m2m1m d-wave M ≈200 150 [6]
ðCr; FeÞSb2 Pnma 2 2 1
mmm d-wave M  200 [7]
CaCrO3 Pnma 2 2 1 d-wave M 90 200 [9]
mmm
2 2 2 1
CrSb P63 =mmc 6= m m m g-wave M 705 1200 [16]
2
MnF2 P42 =mnm 4=1m1m1m d-wave I 67 297 [20,48]
2
MnO2 P42 =mnm 4=1m1m1m d-wave I  900 [17]
2 2
CuF2 P21 =c 2= m d-wave I 69 350 [16]
La2 CuO4 Bmab 2m2m1m d-wave I 317 10 [16]
LaMnO3 Pnma 2m2m1m d-wave I 139,5 20 [22,76]
κ-Cl Pnma 2 2 1
mmm d-wave I 23 50 [4]
1 2
Fe2 O3 R3̄c 3̄ m g-wave I 966 200 [16]
2 2 2 1
MnTe P63 =mmc 6= m m m g-wave I 310 1100 [16]

III. PHYSICAL CONCEPTS spin-split energy bands with typically noncollinear spin
textures in the reciprocal momentum space. An example of
To illustrate the potential and stimulate future research of
a Rashba spin splitting in an inversion-asymmetric non-
altermagnetism in a broad condensed-matter physics field,
magnetic 2D system is illustrated in Fig. 9(a).
we now discuss our understanding of, and outlook on,
unique features of the altermagnetic phase in the context of
several overarching physical concepts. (a) Relativistic (c) Altermagnetic

A. Lifted Kramers spin degeneracy

Energy bands are Kramers spin degenerate [77,78]
across the whole Brillouin zone in all crystals that
are invariant under the symmetry transformation that
combines T and space inversion. Lifting Kramers spin
degeneracy by breaking the symmetry has brought forth a
plethora of physically intriguing and technologically
relevant phenomena, ranging from topological phases
of matter [41,42,79–84] and dissipationless Hall transport
[2,37,42], to charge-spin conversion effects in spintronic
memory devices [40,85–87].
For the many decades of spin-physics research, lifting of
the Kramers spin degeneracy in energy bands has been
considered to originate from two basic mechanisms— (b) (d)
relativistic and nonrelativistic—where the latter is due to
an internal magnetization in ferromagnets or an applied
magnetic field. We start by briefly recalling these two
mechanisms to highlight their distinction from the uncon-
ventional mechanism in altermagnets. Before moving on to
altermagnets, we also briefly review the physics of (lifted)
Kramers spin degeneracy in antiferromagnets to further
FIG. 9. (a) Model relativistic Rashba spin-split bands. (b) Model
emphasize the distinct physics of altermagnetism. of antiferromagnetic zero-magnetization crystal of BiCoO3
The first conventional mechanism of lifting the Kramers with magnetic symmetry T t, and with broken space-inversion
spin degeneracy that does not require magnetic order links symmetry. (c) Model nonrelativistic altermagnetic spin split-
the broken space-inversion symmetry in the direct crystal ting. (d) Model of altermagnetic crystal of RuO2 with non-
space to the spin space by the electron’s relativistic spin- relativistic spin symmetry ½C2 kC4 t. The crystal are discussed in
orbit coupling [88,89]. It results in inversion-asymmetric Refs. [16,101].

040501-10
EMERGING RESEARCH LANDSCAPE OF ALTERMAGNETISM PHYS. REV. X 12, 040501 (2022)

The second mechanism is associated with T -symmetry Breaking of inversion symmetry and T symmetry in the
breaking by internal magnetization of ferromagnets band-structure reciprocal space of this type of collinear
(ferrimagnets) or by an external magnetic field [62]. antiferromagnets is, therefore, purely of relativistic origin.
Microscopically, the former tends to be dominated by a To complete the above discussion, we add the following
nonrelativistic magnetic-exchange interaction and is com- remarks on studies of lifted Kramers spin degeneracy in
monly modeled by a momentum-independent effective antiferromagnets. While magnets with compensated mag-
Zeeman term in the band structure, as illustrated in netic order in the direct space were commonly associated
Fig. 1(a). with spin-degenerate bands in the reciprocal momentum
Next, we discuss Kramers spin degeneracy in antiferro- space, Zeeman or relativistic spin-splitting mechanisms
magnets. All magnetically ordered crystals have broken T were also discussed in antiferromagnets. An effective
symmetry in the direct space. While this leads to the ferromagneticlike Zeeman splitting, and the correspond-
effective Zeeman spin splitting of the band structure in the ing T -symmetry breaking in the band-structure reciprocal
reciprocal momentum space in ferromagnets, spin splitting space, was associated, in antiferromagnets, with a net
has been commonly considered to be excluded in crystals moment induced by an external magnetic field [97–100].
with a compensating antiparallel arrangement of magnetic A T -symmetric relativistic Rashba splitting was predicted
moments [1,69–73]. Indeed, there are two types of Kramers in antiferromagnets such as BiCoO3 with broken space-
spin-degenerate antiferromagnets. inversion symmetry in the direct space, and with the
The first type has a symmetry combining T with opposite Co sublattices connected by the real-space T t
translation t in the direct space of the antiferromagnetic symmetry, as shown in Figs. 9(a) and 9(b) [101]. Another
crystal. The T t symmetry defines type-IV magnetic space example of a magnetic and relativistic splitting was
groups. Among those, only the antiferromagnetic crystals experimentally demonstrated in surface states of antifer-
with space-inversion symmetry have the Kramers spin- romagnetic NdBi [102]. Both of these types of spin
degenerate bands. As a result, the bands have T symmetry splitting offer intriguing interplay with antiferromagnet-
in the reciprocal momentum space, ϵðs; kÞ ¼ ϵð−s; −kÞ, ism. However, they also inherit a net magnetization
and inversion symmetry, ϵðs; kÞ ¼ ϵðs; −kÞ, apart from [Fig. 1(a)] or a noncollinear spin texture [Fig. 9(a)] in
being spin degenerate, ϵðs; kÞ ¼ ϵð−s; kÞ. Examples are the band-structure momentum space, characteristic of
FeRh or MnBi2 Te4 [90,91]. In the nonrelativistic limit and conventional ferromagnets or relativistic spin-orbit-
for collinear antiferromagnetic order (cf. Sec. II B), the coupled systems, respectively.
Kramers spin degeneracy is protected by the spin-group Having provided the necessary background for contrast,
symmetry ½C2 kt alone, i.e., independent of whether the we now discuss the lifting of the Kramers spin degeneracy
antiferromagnetic crystal is or is not inversion symmetric in the altermagnetic phase. As illustrated in Figs. 9(c) and
in the direct space. We note that the nonrelativistic 9(d), it is principally distinct from the conventional
collinear symmetry ½C2 kt does not imply that all materials mechanisms. Unlike the relativistic spin-orbit coupling in
described by type-IV magnetic space groups necessarily nonmagnetic (or magnetic) systems, lifting of the Kramers
have vanishing spin splitting in the nonrelativistic limit. spin degeneracy by altermagnetism does not require break-
This is because type-IV magnetic space groups encompass ing of the space-inversion symmetry. In fact, the non-
also noncollinear magnets. However, all materials from relativistic band structure of altermagnets has inversion
type-IV magnetic space groups have T -symmetric bands in symmetry protected by spin-group symmetry correspond-
the reciprocal space, whether or not relativistic effects are ing to the coplanarity of magnetic order, as shown on the
included. second line in Table II. This applies independently of
The second type of antiferromagnetic crystals with the presence or absence of inversion symmetry in the
Kramers spin-degenerate bands break space-inversion magnetic crystal structure [16]. Also, unlike the relativistic
and T (or T t) symmetries in the direct space on their spin-orbit coupling mechanism, the nonrelativistic elec-
own but have a symmetry in the direct space combining tronic states in the altermagnetic bands have a common
the two transformations. In this case, the Kramers spin- spin-quantization axis, and spin is a good quantum number.
degenerate bands in the reciprocal momentum space have These characteristics are protected by the spin-group
broken inversion symmetry and broken T symmetry. Here, symmetry corresponding to the collinearity of magnetic
CuMnAs or Mn2 Au are among the prominent material order, as highlighted on the first line in Table II.
examples [92–96]. The Kramers spin degeneracy of non- Comparing to ferromagnets, altermagnets share the
relativistic bands in the reciprocal space in these collinear strong nonrelativistic T -symmetry breaking and spin split-
antiferromagnets is protected by spin symmetry ½C2 kĒ, in ting in the band structure. In altermagnets, these character-
combination with the symmetry ½C̄2 kT  (cf. second line in istics are allowed by the spin-group symmetry shown on
Table II). In addition, the ½C2 kĒ symmetry protects T the fourth line in Table II. The distinction from ferromag-
symmetry of the nonrelativistic bands, and ½C̄2 kT  protects nets is that the same spin-group symmetry also protects the
inversion symmetry of bands in the nonrelativistic limit. zero nonrelativistic net magnetization in altermagnets.

040501-11
ŠMEJKAL, SINOVA, and JUNGWIRTH PHYS. REV. X 12, 040501 (2022)

B. Fermi-liquid instabilities
(a) (b)
As discussed in previous sections, altermagnetism can
occur in both insulating and metallic systems as a
consequence of certain (broken) symmetries of the spin
arrangement on the crystal. In this section, we highlight
analogies and differences between altermagnetism and
the theory of Fermi-liquid instabilities in metallic systems.
In Fermi-liquid theory, interactions among electron (c) (d)
quasiparticles are described by Landau parameters in
spin-singlet and spin-triplet channels, using an orbital
angular-momentum partial-wave expansion. Large (neg-
ative) values of Landau parameters lead to Pomeranchuk
Fermi-liquid instabilities [31,103]. A prominent example
of an isotropic s-wave instability in the spin-triplet
channel is Stoner ferromagnetism, which corresponds
to the momentum-independent effective Zeeman spin
splitting in the electronic band structure.
Theoretically, a rich landscape of quantum ordered phases FIG. 10. (a) Schematic diagram of the anisotropic (d-wave)
exchange Fermi-liquid instability in the altermagnet. The black
is linked to anisotropic (nonzero angular-momentum)
line corresponds to the spin-degenerate band in the nonmagnetic
Landau parameters [31]. However, experimental indica- phase, while red and blue lines are spin-split bands in the
tions of anisotropic Fermi-liquid instabilities are rare: for altermagnetic phase. (b) Spin-projected (and orbital-projected)
example, nematic-phase instabilities in the spin-singlet ab initio bands of RuO2 in the energy window corresponding to
channel with nonzero angular momenta. Their typical panel (a). (c) Schematic diagram of the isotropic (s-wave)
characteristics are anisotropic distortions of Fermi exchange Fermi-liquid instability combined with anisotropic
surfaces. Nematic instabilities have been considered in crystal potential in the altermagnet. Solid and dashed black lines
fractional quantum Hall systems, Mott insulators, or high- correspond to the spin-degenerate bands dominated by one or the
T c superconductors—all belonging to the family of com- other sublattice in the nonmagnetic phase, respectively. Red and
plex, strongly correlated systems [31]. blue lines are spin-split bands in the altermagnetic phase.
(d) Spin-projected (and orbital-projected) ab initio bands of
In analogy to Stoner ferromagnetism, nonzero angular-
RuO2 in the energy window corresponding to panel (c). Pink
momentum instabilities in the spin-triplet channel typically boxes with labels 1 and 2 correspond to the full ab initio bands of
break the SUð2Þ symmetry of the nonrelativistic non- RuO2 shown in Fig. 3(c). This figure is adapted from Ref. [16].
magnetic Fermi liquid [31]. Altermagnetic symmetries
are reminiscent of even-parity wave Fermi-liquid instabil-
ities in the spin-triplet channel [16,18]. Despite the rem- hopping due to the anisotropic exchange interaction in the
iniscence, we now illustrate, on the band structure of RuO2 , altermagnetic state [6,7,13]. The model band structure is
that the predicted characteristics of the altermagnetic plotted in Figs. 11(a) and 11(b). The energy spectrum
symmetries are extraordinary. exhibits spin-degenerate nodal surfaces at kx;y ¼ 0; π,
A spin-splitting mechanism due to an anisotropic marked in gray in Fig. 11(a), and protected by mirror
exchange interaction [13,18,20,21] can be identified in plane symmetries that transform one spin sublattice on the
parts of the RuO2 band structure with a single twofold spin- opposite-spin sublattice and are contained in the little
degenerate band in the nonmagnetic phase, which, in the groups of the nodal-surface momenta (cf. fifth line in
altermagnetic phase, undergoes an anisotropic, momentum- Table II). The resulting nodal structure and spin-splitting
dependent spin splitting with alternating sign [16]. This is modulation pattern correspond to a dxy -wave symmetry.
illustrated in Figs. 10(a) and 10(b) on a schematic diagram The characteristic dxy -wave spin-up and spin-down Fermi
and ab initio bands of RuO2 . Remarkably, the mechanism surfaces are anisotropic and mutually rotated by 90°,
is dominated by anisotropic exchange interactions; i.e., it following the ½C2 kC4  spin-group symmetry. Ab initio
persists without including many-body correlation effects bands of RuO2 and the corresponding model thus illustrate
beyond the effective single-particle, local-spin-density altermagnetic symmetries reminiscent of a d-wave, spin-
approximation [16,18]. triplet, Fermi-liquid instability. However, the d-wave alter-
An effective single-particle, two-band Hamiltonian, magnetism is not ascribed to strong correlations but
originates from the (broken) symmetries of the spin
H ¼ 2t cos kx cos ky þ 2tJ sin kx sin ky σ z ; ð1Þ arrangement on the crystal.
Remarkably, the ab initio band structure of RuO2
that models this mechanism contains, apart from the demonstrates an additional, distinct spin-splitting mecha-
common kinetic-energy hopping term, a spin-dependent nism. In this case, the size and momentum dependence of a

040501-12
EMERGING RESEARCH LANDSCAPE OF ALTERMAGNETISM PHYS. REV. X 12, 040501 (2022)

(a) (c)

(b) (d)

(e) (f)

FIG. 11. (a,b) Model of the altermagnetic quasiparticle with quadratic dispersion around the spin-degenerate Γ point, and spin-
winding number 2 around the Γ point. The model corresponds to Eq. (1). (c,d) Model of the altermagnetic spin-polarized valley
quasiparticle with no spin winding in the valley. The model corresponds to Eq. (2). The center of a valley is at TRIM, and the spin
polarization is opposite at TRIM X and Y. (e) Schematic illustration of distinct symmetries of band structures with nonrelativistic
altermagnetic and relativistic nonmagnetic valleys. (f) Real-space spin-dependent hoppings used to construct model band structures in
panels (a)–(c). This figure is adapted from Refs. [6,7,13].

strong nonrelativistic altermagnetic spin splitting is deter- size and momentum dependence of the spin splitting is a
mined by the band splitting due to an anisotropic crystal copy of the size and momentum dependence of the orbital
potential of the nonmagnetic phase [16]. This unconven- splitting in the nonmagnetic state. The presence of this
tional electric spin-splitting mechanism is illustrated on a microscopic spin-splitting mechanism in RuO2 is again
schematic diagram in Fig. 10(c). In the nonmagnetic state, confirmed by ab initio calculations shown in Fig. 10(d). In
there are two spin-degenerate bands that cross at the this case, the altermagnet can be viewed as two interpen-
fourfold spin and orbital-degenerate Γ point, while the etrating s-wave Stoner ferromagnets with opposite mag-
orbital degeneracy is lifted away from the Γ point. One of netizations that, because of the interplay with anisotropies
the two spin-degenerate bands has a dominant projection of the crystal potential, generate spin-split d-wave-like
on one sublattice, while the other band has a dominant Fermi surfaces.
projection on the other sublattice. The bands are anisotropic The potential richness of the landscape of altermagnet-
due to the anisotropy of the crystal potential (cf., third line ism can be further inferred by inspecting the symmetries of
in Table II). The anisotropies of the spin-degenerate bands all altermagnetic spin groups. Each altermagnetic spin
corresponding to the two sublattices are mutually rotated by point group can be associated with a given even-parity
90°, reflecting the real-space C4 rotation symmetry that wave anisotropy of the spin-split Fermi surfaces around the
transforms one crystal sublattice on the other. As a result, Γ point [16]. Apart from the d-wave form, this anisotropy
there is a mutual momentum-dependent splitting between can have a g-wave or i-wave form [16].
the two spin-degenerate bands away from the Γ point in the Finally, we recall that the discovery of the unconven-
nonmagnetic phase. Upon the transition to the magnetically tional d-wave superconductivity not only opened an
ordered altermagnetic phase, an additional momentum- entirely new research landscape of this many-body phase
independent (isotropic) exchange interaction occurs, with [30] but also raised a fundamental question of whether and
opposite sign in the bands corresponding to opposite-spin how an unconventional d-wave counterpart could be
sublattices. As a result, two pairs of spin-split bands form realized in magnetism [29]. Earlier considerations focused
with opposite sign of the spin splitting. For a given pair, the on possible realizations of the unconventional d-wave

040501-13
ŠMEJKAL, SINOVA, and JUNGWIRTH PHYS. REV. X 12, 040501 (2022)

magnetism due to strong electronic correlations [31–33]. spin-winding number 1 in the relativistic Rashba model.
In contrast, the d-wave (or higher even-parity wave) We also point out that the spin-winding number in the
altermagnetism discussed above is directly linked to relativistic systems is associated with continuously varying
symmetries of the effective single-particle crystal potential spin direction in the momentum space. In contrast, alter-
and does not require strongly correlated systems. This magnets show that nonzero integer invariants, describing
makes the altermagnetic materials discussed in this work how many times the quasiparticle spin reverses when
realistic candidates for a robust unconventional d-wave completing a closed path around the Γ point, can also
(or higher even-parity wave) magnetism, which can exist in systems where all spins share a common spin
host time-reversal symmetry-breaking responses of com- quantization axis, and spin is a good quantum number.
parable strength to the conventional s-wave ferromagnets. A different type of predicted altermagnetic electron
Moreover, we have already emphasized that the altermag- quasiparticle can be illustrated on a two-band model
netic symmetries can equally hold for the Dyson-equation Hamiltonian [6,7,13]:
description of correlated systems, as for the effective
single-particle Kohn-Sham potentials. Therefore, altermag- H ¼ 2tJ ðcos kx − cos ky Þσ z ; ð2Þ
nets may also facilitate the originally perceived realization
of the unconventional d-wave magnetism via strong whose energy spectrum is shown in Fig. 11(d). [Around the
correlations. Γ point, Eq. (2) is related to the model in Eq. (1) by a 45°
rotation of the momentum space, and by setting t ¼ 0.]
C. Electron quasiparticles In the k · p approximation around time-reversal invariant
The predicted extraordinary Fermi-liquid instabilities in momenta (TRIMs) X and Y, highlighted by dashed
altermagnets can generate a variety of unconventional rectangles in Fig. 11(d), the spectrum takes the form of
electron quasiparticles. An example can be illustrated on spin-split valleys given by [see Figs. 11(c) and 11(d)],
the energy bands of the model two-band Hamiltonian (1) in
the k · p approximation around the Γ point, highlighted by E ðX; kÞ ¼ tJ ð4 − k2 Þ;
a dashed rectangle in Fig. 11(b). The spin-dependent part of E ðY; kÞ ¼ ∓ tJ ð4 − k2 Þ: ð3Þ
the band structure is given by 2tJ kx ky σ z. The spin degen-
eracy of the Γ point is generally protected in altermagnets (Recall that a momentum k is time-reversal invariant when
by the spin-group symmetry ½C2 kA because the Γ point is it differs from −k by a reciprocal lattice vector.)
invariant under any real-space symmetry transformation The possibility to observe spin-split valleys around
(including the rotations A). The Γ-point spin degeneracy is TRIMs in real materials is predicted by the spin-group
analogous to the T -symmetric relativistic bands. Here, an symmetry analysis and ab initio band-structure calculations
example is the Rashba model whose spin-dependent part is of the altermagnetic phase in, e.g., Mn5 Si3 [6] [Fig. 7(c)].
given by λðkx σ y − ky σ x Þ [Fig. 9(a)]. However, unlike the Besides 3D crystals, the altermagnetically spin-split valleys
linearly dispersing quasiparticles around the Γ point of the can also form in 2D materials [15,23,104], as predicted,
inversion-asymmetric relativistic bands, the altermagnetic e.g., in ab initio calculations of the band structure of a
quasiparticles in the above model have a quadratic monolayer insulator V2 Se2 O [15] [Fig. 7(d)].
dispersion around the Γ point, in line with the general Locally, the individual valleys around the X and Y
inversion symmetry of bands in altermagnets (cf. second TRIMs are isotropic in the above model. This illustrates
line in Table II). that altermagnets can host spin-polarized quasiparticles
The altermagnetic quasiparticles are spin polarized away analogous to the model nonrelativistic s-wave Stoner
from the Γ point and can be assigned a characteristic even ferromagnet, with no spin winding around the TRIM.
integer, defined as follows [16]: When making a closed However, unlike ferromagnets, the altermagnetic spin-
loop in the momentum space around the Γ point in a plane group symmetries impose that each spin-split TRIM has
orthogonal to a spin-degenerate nodal surface crossing the a counterpart TRIM elsewhere in the Brillouin zone with
Γ point, the spin rotates by 360° following two discrete opposite spin splitting. The presence of these TRIM pairs is
reversals. Each spin-degenerate nodal surface crossing the protected by the ½C2 kAH spin-group symmetries (cf. fourth
Γ point, which is present in the crystal momentum space, line in Table II).
generates such a spin rotation. The characteristic spin- The altermagnetic spin-polarized quasiparticles in sep-
group integer is defined as the number of these spin- arate local valleys in the momentum space are reminiscent
degenerate nodal surfaces crossing the Γ point. It can be 2 of relativistic spin-polarized valley quasiparticles in non-
(d-wave), 4 (g-wave), or 6 (i-wave) in altermagnets [16]. magnetic hexagonal 2D materials, such as transition metal
This illustrates the potential richness of the spin-polarized dichalcogenides [105]. The common features shared by the
quasiparticles in altermagnets around the Γ point. altermagnetic and relativistic quasiparticles are the opposite
The characteristic spin-group integer of the above alter- spin polarization in valleys occupying different parts of the
magnetic quasiparticle model is 2. This is in contrast to the Brillouin zone, and the zero net spin polarization when

040501-14
EMERGING RESEARCH LANDSCAPE OF ALTERMAGNETISM PHYS. REV. X 12, 040501 (2022)

integrated over the whole Brillouin zone. However, only alternating across the magnon Brillouin zone [106]. On one
altermagnets allow these valleys to be centered at TRIMs, hand, altermagnetic magnons can be chiral and carry spin
as highlighted in Fig. 11(e). In the nonmagnetic relativistic currents, similar to ferromagnets but with highly aniso-
systems, spin splitting is excluded by T symmetry not only tropic characteristics. On the other hand, the dispersion of
at the Γ point but at all TRIMs. altermagnetic magnons can be linear around the degenerate
So far, we have discussed the electron quasiparticles Γ point, similar to antiferromagnets.
from the symmetry perspective limited to the spin-group For metallic systems, the Landau damping in altermag-
transformations acting on the spin- and momentum-depen- nets can be suppressed due to the spin splitting of the
dent band structure. Additional, rich, quasiparticle physics, electron quasiparticles. This is more reminiscent of the low
including higher-order degeneracy quasiparticles, can Landau damping in metallic ferromagnets and contrasts
emerge from the analysis of spin-group transformations with the large Landau damping associated with the spin
acting on the electron wave functions (spin-group repre- degeneracy of the electronic bands in antiferromagnets
sentations) [16,24,59,61]. [106,108]. All these unconventional characteristics make
altermagnets a promising material platform for magnonics,
D. Magnons including the exploration and exploitation of the emission,
Besides the electron energy spectra and quasiparticles, propagation, and detection of ultrashort (ps-scale) THz
we foresee that the symmetries of the direct-space spin- magnon pulses of wavelengths orders of magnitude smaller
polarization order parameter in altermagnets will also be than the wavelengths of the THz photons.
reflected in unconventional characteristics in the reciprocal
momentum space of the spin-wave spectra and magnon E. Berry phase, nondissipative transport,
quasiparticles [106]. and band topology
The typically leading contribution to the magnon
The Berry phase is a general concept in quantum
spectra can be obtained by mapping the spin-dependent
mechanics [111]. A prototypical example is the Aharonov-
electronic
P structure on the Heisenberg Hamiltonian, H ¼
Bohm phase given by a real-space path integral of the
− i≠j Jij êi êj [59,107,108]. Here, êi is the direction
electrodynamic vector potential along a closed loop or,
of the magnetic moment around an atom at position Ri , equivalently, by an integral of the magnetic field over an
and Jij are Heisenberg exchange coupling parameters. In area enclosed by the loop. The phase can be macroscop-
the Heisenberg model, the real- and spin-space trans- ically observable by resistance oscillations in an applied
formations are decoupled, and the symmetries of the magnetic field.
corresponding magnon bands can be described by the In the crystal momentum space, a Berry connection
nonrelativistic spin-group formalism [26,59]. analogue of the electrodynamic vector potential, and a
In antiferromagnets, translation or inversion symmetry Berry curvature analogue of the magnetic field,
transformations connecting opposite-spin sublattices pro-
tect the degeneracy of opposite-chirality magnon bands
[26,59]. This has been commonly illustrated on the A n ðkÞ ¼ ihunk j∇k unk i;
opposite-spin sublattices of rutile crystals while omitting Bn ðkÞ ¼ ∇k × A n ðkÞ; ð4Þ
the presence of nonmagnetic atoms in these crystals
[26,59,109]. We have seen in Sec. II, however, that the
nonmagnetic O atoms in metallic rutile RuO2 , or F atoms in can also generate macroscopic observables. A prominent
insulating rutiles FeF2 or MnF2, break the translation and example is the nondissipative Hall current given by the
inversion symmetries connecting the opposite-spin sublat- transverse conductivity [37],
tices. Instead of classic antiferromagnets [26,49,59,109],
Z
rutiles are the prototypical representatives of d-wave e2 X d3 k
altermagnetism [16], with interlinked unconventional prop- σ Hall
xy ¼− 3
f½εn ðkÞBzn ðkÞ: ð5Þ
ℏ n BZ ð2πÞ
erties of the direct-space crystal structure and reciprocal-
space electronic, as well as magnonic, band structure [106].
Hematite Fe2 O3 is another prominent insulating compen- Here, f½εn ðkÞ is the Fermi-Dirac distribution function,
sated magnet in magnonic research [110]. In analogy to the εn ðkÞ is the energy of the Bloch state in band n with crystal
rutiles, the nonmagnetic O atoms in Fe2 O3 break the momentum k, and unk ðrÞ is the periodic part of the Bloch
translation and inversion symmetries while preserving a wave function.
rotation symmetry, connecting the opposite-spin sublatti- Altermagnets are predicted to bring unique elements into
ces. Fe2 O3 is an example of a g-wave altermagnet [16]. the physics of Berry phase phenomena [2,3,5–7,44,45].
In analogy to the electronic band structure, the degen- The Berry curvature near the Γ point of a k · p altermagnet-
eracy of magnon bands with opposite chirality is predicted Rashba model [2], tk2 þ 2tJ kx ky σ z þ λðkx σ y − ky σ x Þ, is
to be lifted in altermagnets, with the sign of the splitting given by

040501-15
ŠMEJKAL, SINOVA, and JUNGWIRTH PHYS. REV. X 12, 040501 (2022)

2tJ λ2 kx ky becomes quantized when the integration covers the full

BðkÞ ¼∓ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi : ð6Þ Brillouin zone in 2D insulators [118]. The corresponding
4t2J ðkx ky Þ2 þ λ2 k2 quantum Hall effect [119] was demonstrated in graphene at
room temperature [120], but it requires a strong magnetic
Equation (6) illustrates that the characteristic even-parity field. The ferromagnetic quantum anomalous Hall counter-
wave (e.g., d-wave) anisotropy in the nonrelativistic band part [121] can be observed at zero magnetic field but, so far,
structure of altermagnets can also be reflected in their has been limited to Kelvin temperatures [42,122]. Since
relativistic Berry curvature. In contrast, a counterpart altermagnetism can host the Berry phase phenomena and
ferromagnet-Rashba model, tk2 þ Δσ z þ λðkx σ y − ky σ x Þ, can occur in 2D crystals and in insulators, it opens new
gives an isotropic Berry curvature near the Γ point possibilities in the search for high-temperature zero-field
[2,37,112–115], reflecting the principally isotropic s-wave quantum Hall phenomena. For a further in-depth discussion
nature of ferromagnetism. of Berry phase physics and nondissipative Hall transport in
The Berry curvature tends to reach the highest values altermagnets, we refer to the recent topical review [2].
near band (anti)crossings [2,37], which implies another While above we have discussed topological Berry phase
outstanding feature of altermagnets. In contrast to the physics, which is a consequence of perturbative relativistic
typically accidental (anti)crossings in ferromagnets, the spin-orbit coupling effects, altermagnets can also lead to
spin-group symmetries of altermagnets impose the pres- identification of unconventional magnetic topologies of a
ence of the spin-degenerate nodal surfaces in the non- nonrelativistic origin [7,16]. The spin degeneracies in
relativistic band structure (cf. fifth line in Table II). When nonrelativistic altermagnetic bands are protected by sym-
the relativistic spin-orbit coupling is included, these nodal metries of the spin point group (spin Laue group) because
surfaces (which may be weakly gapped by the spin-orbit the electronic structure is strictly separated into nonmixing
coupling) may become symmetry-defined Berry-curvature spin-up and spin-down channels [16]. However, additional
hotspots. This is illustrated in Fig. 5 on relativistic ab initio band degeneracies can exist within one spin channel, i.e.,
band structures of RuO2 and FeSb2 [2,7,13]. degeneracies in band indices other than spin, which are
Since T is antiunitary in quantum mechanics, the Berry protected by space-group symmetries (symmetry trans-
curvature (4) is odd under T , T Bn ðkÞ ¼ −Bn ð−kÞ, imply- formations also containing translations). These features
ing that the integral in Eq. (5) vanishes in T -symmetric can be included in the symmetry analysis based on the
band structures. Breaking of T symmetry in the band nonrelativistic spin-group formalism and can be important
structure of altermagnets is, therefore, the key property that when exploring exotic topological quasiparticles near such
allows for the observation of macroscopic responses, such degeneracies [16,24,25,57,58]. The nonrelativistic spin-
as the anomalous Hall effect [2,3]. Recent experiments [5] space groups are also well suited for exploring magnon
have detected the anomalous Hall effect in RuO2 of a band topology [59] of altermagnets.
comparable strength to typical Hall signals in ferromagnets.
This is consistent with the predicted strong altermagnetic IV. RESEARCH AREAS
T -symmetry breaking in the band structure of this com-
We now move to the discussion of the potential of
pensated collinear magnet [3,5] (cf. Table I).
altermagnetism in specific areas of condensed-matter
RuO2 is an example in which the lattice of the magnetic
research. We start from spintronics in which, besides the
Ru atoms alone would have the opposite-spin sublattices
anomalous Hall effect, initial theory predictions of non-
connected by a translation. As mentioned above, this
relativistic spin-dependent responses have recently been
symmetry would imply T symmetry of the band structure
tested by experiments.
(and in combination with inversion symmetry of the crystal,
also spin degeneracy). The presence of the nonmagnetic
oxygen atoms is, therefore, essential for the T -symmetry A. Spintronics
breaking (and spin splitting) in the altermagnetic band To highlight the novelty of the emerging concept of
structure of RuO2 and, consequently, for the anomalous spintronics without magnetization and relativity, enabled
Hall effect [3]. The term “crystal Hall effect” [3,7,44,117] by altermagnetism, we first briefly recall the principles of
was introduced to highlight this feature. One of the the established fields of spintronics in ferromagnets and
implications, unparalleled in the conventional anomalous antiferromagnets.
Hall effect in ferromagnets, is that in altermagnets the Hall The nonrelativistic electronic structure of ferromagnets is
signal is predicted to flip sign not only when reversing the split into majority and minority spin bands that break the T
magnetic moments but also when the symmetry-breaking symmetry and where spin is a good quantum number
arrangement of nonmagnetic atoms reverses between the independent of momentum. This results in different con-
two magnetic sublattices [3]. ductances of the two conserved spin channels, which makes
Next, we recall that in 2D systems, Eq. (5) turns into a electrical currents in ferromagnets strongly spin polarized.
surface integral proportional to the Berry phase, which Passing the spin-polarized current in a multilayer structure

040501-16
EMERGING RESEARCH LANDSCAPE OF ALTERMAGNETISM PHYS. REV. X 12, 040501 (2022)

with reference and sensing (or recording), ferromagnetic spin-dependent conductivities in d-wave altermagnets can
electrodes can generate spin-transfer torque (STT) and be estimated from the conventional current-in-plane GMR
giant/tunneling magnetoresistance (GMR/TMR) effects. expression derived in ferromagnets [85],
These strong nonrelativistic responses facilitate reorienta-  
tion between parallel and antiparallel magnetizations of the 1 1
GMR ¼ Rσ þ −2 ; ð7Þ
two ferromagnetic electrodes and corresponding resistive 4 Rσ
changes on a scale of around 10%–100% utilized in
commercial spintronic memory devices [85–87,123]. where Rσ ¼ σ þ;xx =σ −;xx ¼ σ þ;xx =σ þ;yy ¼ σ −;yy =σ −;xx . The
Direct antiferromagnetic counterparts of STT and GMR/ ab initio calculations in RuO2 give GMR reaching a 100%
TMR effects are principally excluded by the T invariance scale [13] [Fig. 12(e)], highlighting the expected large
and spin degeneracy of the nonrelativistic electronic GMR ratios in altermagnets.
structure. As a result, the research of antiferromagnetic As noted above, the polarization of the longitudinal spin-
spintronics turned to the typically weaker relativistic polarized current in altermagnets is predicted to reverse not
phenomena, such as the spin-orbit torque reorientation of only with the reversal of the altermagnetic-order vector but
the Néel vector detected by a scale of about 0.1%–1% also with the reorientation (e.g., by 90°) of the applied
anisotropic magnetoresistance [39,40,92,93,124]. The con- electrical bias in the d-wave altermagnet. A directly related
cept inherits the general weakness of relativistic spintronic effect, also unparalleled in ferromagnets, is illustrated in
phenomena that are simultaneously generated and sup- Fig. 12(c). For a bias applied in the diagonal direction
pressed by the spin nonconserving nature of the relativistic between the two anisotropy axes of the spin-split Fermi
spin-orbit coupling [40]. surfaces, the longitudinal current is unpolarized. However,
The strong nonrelativistic T -symmetry breaking and a nonrelativistic spin current is generated in the transverse
spin splitting in altermagnetic bands directly opens a direction. The effect has been predicted in a range of
possibility to not only replicate the concepts known from inorganic and organic materials [4,8,9,15]. The d-wave
ferromagnets but also to enrich nonrelativistic spintronics altermagnet acts here as an electrical spin splitter, with a
by new effects and functionalities linked to the zero propagation angle between spin-up and spin-down currents
nonrelativistic net magnetization [4,8–15]. reaching 34° in RuO2 [8]. The corresponding charge-spin
The anisotropy of the split and equally populated conversion ratio reaches a remarkable 28% [Fig. 12(e)], and
spin-up and spin-down Fermi surfaces in altermagnets the spin conductivity is a factor of 3 larger than the record
[cf. Figs. 11(a) and 11(b)] results in spin-dependent value from a survey of 20 000 nonmagnetic relativistic
anisotropic group velocities, ∂Eþ ðkÞ=∂ki ≠ ∂E− ðkÞ=∂ki , spin-Hall materials [125].
where þ=− refers to the spin index. The corresponding The above outstanding charge-spin conversion efficiency
conductivities in d-wave altermagnets are then also spin of altermagnetic RuO2 prompted a theoretical proposal of a
dependent and anisotropic. Considering the x and y spin-splitter torque (SST) [8], in part already supported by
directions as the anisotropic axes of the spin-split initial experiments [10–12] (cf. Table I). In the geometry
Fermi surfaces, we get σ þ;xx ≠ σ −;xx , σ þ;yy ≠ σ −;yy , and schematically illustrated in Fig. 12(d), an in-plane bias
σ ;xx ¼ σ ∓;yy . The electrical current then becomes spin generates the nonrelativistic spin current in the d-wave
polarized when the bias is applied along the x or y altermagnetic film along the out-of-plane direction, with
direction, as schematically illustrated in Fig. 12(a). the polarization of the spin current controlled by the
Moreover, as a consequence of the T -symmetry breaking orientation of the altermagnetic-order vector. The spin
of the spin-split bands, the sign of the spin polarization current then exerts a torque on the adjacent altermagnetic
reverses when reversing the altermagnetic-order vector. (In (or ferromagnetic) layer. SST does not inherit the problems
analogy to the antiferromagnetic Néel vector, we define the of STT associated with the out-of-plane direction of the
altermagnetic-order vector as the difference between the applied electrical bias [86,126]. Instead, it shares the in-
magnetization vectors of the opposite-spin sublattices.) plane electrical-bias geometry of the spin-orbit torque
The above nonrelativistic spin-current characteristics are generated by the relativistic spin-Hall polarizer while
analogous to ferromagnets. However, in contrast to ferro- circumventing the limitations of the more subtle relativistic
magnets, the altermagnetic spin splitting is also predicted to spintronic effects [40].
cause the reversal of the spin polarization of the current Another nonrelativistic spintronic effect that can be
when the applied electrical bias is flipped between the x and generally expected to exist in altermagnets is a variant
y directions. of the TMR in a tunnel junction with an insulating spacer
The spin-polarized current directly implies a GMR effect separating the two altermagnetic electrodes [13,14]. The
in a stack comprising two altermagnets, separated by a altermagnetic TMR can be illustrated on the model band
conductive nonmagnetic spacer, with the altermagnetic- structure with spin-split valleys [cf. Figs. 11(c) and 11(d)].
order vectors oriented either parallel or antiparallel, as The pairs of valleys with opposite spin polarization result in
illustrated in Fig. 12(b). The contribution to GMR from the the equal net population of spin-up and spin-down states,

040501-17
ŠMEJKAL, SINOVA, and JUNGWIRTH PHYS. REV. X 12, 040501 (2022)

(a) Longitudinal spin-current (b) Giant magnetoresistance

(e)

(c) Transverse spin-current (d) Spin-splitter torque

e
n(m)
n( )
jS

AM(FM)

E AM

FIG. 12. (a) Schematics of the longitudinal spin current in altermagnets. For an electric bias E applied along one of the main
anisotropy axes of the spin-split Fermi surfaces, the spin-up and spin-down charge currents are parallel but of different magnitudes due
to the Fermi-surface anisotropies. As a result, the longitudinal charge current is spin polarized. (b) Schematics of a GMR stack in a
current-in-plane geometry. As an example, we show the antiparallel configuration of the altermagnetic-order vectors in the two
electrodes AM1 and AM2 . Interfaces are oriented along one of the main anisotropy axes of the spin-split bands. Energy band cuts
highlight the anisotropy around the Γ point, resulting in anisotropic spin-dependent conductivities. (c) Schematics of the transverse spin
current. For E applied in the diagonal direction between the two anisotropy axes of the spin-split Fermi surfaces, the spin-up and spin-
down charge currents combine in an unpolarized longitudinal charge current and in a pure transverse spin current. (d) Spin-splitter-
torque concept on a schematic of altermagnetic-altermagnetic (ferromagnetic) bilayer. A spin-current from the bottom altermagnet
propagates in the out-of-plane direction and generates a spin-splitter torque on the altermagnetic (or ferromagnetic) order vector in the
top layer. (e) Ab initio longitudinal spin-up and spin-down conductivities (red and blue), GMR, and the ratio of the transverse spin
current relative to the longitudinal charge current (SCR) in RuO2 . This figure is adapted from Refs. [8,13].

while the densities of states within a given valley become Finally, we note that symmetry-wise, TMR is, in
spin dependent, nþ ðM1 Þ ≠ n− ðM1 Þ, nþ ðM2 Þ ≠ n− ðM2 Þ, principle, expected in all altermagnetic spin groups [13]
and n ðM1 Þ ¼ n∓ ðM2 Þ. For tunneling that conserves the and can reach large magnitudes as long as the spin-
valley index, parallel and antiparallel configurations of polarized quasiparticles are well separated in the momen-
altermagnetic-order vectors in the two layers, illustrated tum space to provide for the sufficiently decoupled spin
in Fig. 13(a), are predicted to give different conductances, transport channels [Fig. 13(c)]. On the other hand, the
in analogy to ferromagnetic TMR. This can be seen by GMR derived from the anisotropy of the macroscopic
applying the Jullière formula [85] per valley [13], (averaged over momentum) spin-dependent conductivities
  is predicted to be allowed only in d-wave altermagnets
1 1 [13,16].
TMR ¼ Rn þ −2 ; ð8Þ
2 Rn We now proceed to the discussion of how altermagnet-
ism can contribute to the research of space, time, and
where the ratio of the spin-up and spin-down densities of energy downscaling in spintronic devices. The nonrelativ-
states in the valley is given by Rn ¼ nþ ðM1 Þ=n− ðM1 Þ ¼ istic nature and corresponding large strength of the GMR
nþ ðM1 Þ=nþ ðM2 Þ ¼ n− ðM2 Þ=n− ðM1 Þ. (TMR) and SST (STT) effects are important prerequisites
The ab initio calculations of about 100% TMR ratios in for temporal and spatial scalability of memory devices
RuO2 [Fig. 13(b)] or Mn5 Si3 [13,14] illustrate the potential utilizing these phenomena for reading and writing infor-
for achieving large TMR responses in tunnel junctions with mation. This highlights the importance of the presence
altermagnetic electrodes. of these phenomena in altermagnets. In contrast, the

040501-18
EMERGING RESEARCH LANDSCAPE OF ALTERMAGNETISM PHYS. REV. X 12, 040501 (2022)

(a) Parallel Antiparallel

AM1 Tunneling AM2 AM1 Tunneling AM2

barrier barrier

(b) (c)

FIG. 13. (a) Schematics of a TMR stack with an insulating barrier and altermagnetic electrodes with parallel and antiparallel order
vectors. Energy band cuts highlight the oppositely split valleys, resulting in valley and spin-dependent densities of states. (b) Ab initio
quantum-transmission calculations of the TMR in a RuO2 j TiO2 j RuO2 tunnel junction. (c) Model spin and transport-momentum
projected band structure, and relative difference between the conductances in the parallel and antiparallel configurations of the
altermagnetic-order vectors in the two electrodes. The TMR is maximized for transport energies corresponding to the spin-split valleys
well separated in momentum. This figure is adapted from Refs. [13,14].

scalability is limited in antiferromagnetic devices that rely close-to THz range of the resonance frequency of collective
on the typically weaker relativistic spintronic phenomena. spin excitations (magnons), corresponding to the close-to
Realizing the strong nonrelativistic effects and function- ps internal timescale of spin dynamics. This opens the
alities in altermagnets will bring advantages compared to possibility for achieving the least-dissipative and fastest
antiferromagnets and to ferromagnets. The large nonrela- control of a bistable memory bit [128]. According to the
tivistic magnetization, accompanying the spin-polarized Landauer thermodynamic principle [129,130], confirmed
band structure of ferromagnets, represents a significant in magnetic systems at above around μs timescales
limitation for spintronics. The low tolerance to perturbing [131,132], the minimal energy dissipation when restoring
magnetic fields implies that stray magnetic fields from the one memory bit is given by kB T ln 2, where T is the
ferromagnetic films in the GMR(TMR)/STT stacks have to temperature and kB denotes the Boltzmann constant. At
be suppressed by elaborate device engineering to eliminate room temperature, this Landauer limit is in the range of
magnetic cross-talk within and between individual memory about meV; i.e., it corresponds to the energy of a photon of
bits. This is achieved at the expense of synthesizing about THz. Remarkably, switching by an energy corre-
complex multilayer stacks, with typically over ten different sponding to a single THz photon per magnetic atom, i.e.,
materials, in which the reference or recording electrodes around meV per atom, and with the limiting ps duration of
comprise two or more ferromagnetic layers with mutually the delivered THz pulse has been demonstrated recently in
compensating antiparallel magnetizations (the so-called a compensated collinear magnet TmFeO3 [128]. This
synthetic antiferromagnets) [85,127]. Even in these elabo- shows that reaching the Landauer energy limit of about
rate multilayers, the net magnetic stray fields can only be meV per memory bit at the limiting ps timescale is
partially eliminated. In contrast, altermagnetic crystals with becoming a legitimate, albeit still very challenging,
vanishing magnetization and stray fields will offer a research goal in ultrafast photomagnetism [133]. Since
magnetic robustness combined with the absence of mag- the orthoferrite TmFeO3 is an insulating d-wave altermag-
netic cross-talk, which will allow for a significant sim- net, it also opens the possibility for detecting the ultrafast
plification of the structure of spintronic devices. In switching dynamics of the altermagnetic-order vector by
combination with the strong GMR (TMR) and SST the T -symmetry breaking magneto-optical responses com-
(STT) effects, this can enable spatial downscaling beyond monly employed in ferromagnets. In metallic altermagnets,
the limits of ferromagnetic devices. the research towards the limiting energy and timescales can
Another principal advantage compared to ferromagnets, further benefit from the presence of the strong nonrelati-
which altermagnets share with antiferromagnets, is the vistic GMR (TMR) and SST (STT) responses.

040501-19
ŠMEJKAL, SINOVA, and JUNGWIRTH PHYS. REV. X 12, 040501 (2022)

So far, we have discussed concepts of spintronic devices Ferromagnets have been considered for a direct heat
in which the magnetic-order vector is uniformly switched conversion to electricity [138]. Here, the anomalous
across a manmade memory element. Alternatively, infor- Nernst effect, a thermo-electric counterpart of the anoma-
mation can be coded in self-assembled magnetic nano- lous Hall effect, is regarded as an attractive candidate
textures, such as domain walls. This is another research phenomenon [139]. The anomalous Nernst effect gener-
area for which the stray-field-free compensated magnets, ates an electric field in a transverse direction to the thermal
and altermagnets in particular, can be highly favorable. gradient. Particularly in thin-film or nanostructured
Self-assembled magnetic nano-textures in Kramers spin- heat-charge conversion devices, the transverse geometry
degenerate antiferromagnets have been suggested as the can significantly enhance the conversion efficiency com-
origin of the recently observed, highly reproducible, analog pared to the conventional longitudinal Seebeck effect
time-dependent switching over the full range from ms [139]. A complementary research area to the anomalous
electrical writing pulses to sub-ps optical pulses, and with a Nernst effect are thermal counterparts of the GMR/TMR
resistivity increase in the switched state on a 10%–100% and STT phenomena. Here, the energy harvesting concept
scale [134,135]. The devices mimic a logic-in-memory is based on employing heat gradients, instead of electrical
functionality of spiking neuromorphic elements, thus bias, to directly read or write information in a memory
offering additional space, time, and energy downscaling device [138].
prospects unparalleled in the conventional digital devices Altermagnets significantly enlarge the material land-
with logic and memory separated by the von Neumann scape for realizing and optimizing these thermo-electric
bottleneck. The resistivity increase in the switched state responses that originate, in analogy to their electronic
was ascribed to the formation of atomically sharp domain counterparts, from the T -symmetry-broken spin-polarized
walls [134–136]. They correspond to the ultimate down- band structure. Unlike the typically metallic ferromagnets,
scaling of a magnetic domain-wall width since the Néel altermagnets are predicted to span a broad range of
vector flips by 180° from one to the neighboring atomic conduction types (cf. Sec. II D). This prediction is favor-
plane [136]. If generated in an altermagnet, each atomically able because, from a general thermo-electric perspective,
sharp domain wall separating domains with opposite sign semimetals or semiconductors are more suitable material
of the altermagnetic-order vector would represent a local types than metals due to the strong dependence of their
GMR junction that is self-assembled and corresponds to the electronic structure on energy near the Fermi level. A
ultimate downscaling of the width of the junction spacer. particularly intriguing class of materials are those exhibit-
This atomic-scale GMR would generate a strong additional ing a metal-insulator transition. Among the altermagnetic
contribution to the resistances of the atomically sharp candidate materials, FeSb2 [7,16] is an example in which
domain walls in altermagnets. The altermagnetic phase earlier studies reported an extraordinary (spin-independent)
thus offers a possibility to combine, in one physical system, thermo-electric response, linked to the metal-insulator
the phenomena and functionalities of the commercial transition [140].
digital ferromagnetic memory devices and of the exper- So far, we have discussed electrical-current (voltage)
imental antiferromagnetic analog neuromorphic devices. responses to the applied thermal gradients. Altermagnets
may also open intriguing new directions in research and
B. Spin caloritronics, field-effect electronics, application of spin-current responses to thermal gradients
and multiferroics in the field commonly referred to as spin caloritronics
The key merit of ferromagnets from an energy-saving [138]. In analogy to the anomalous Nernst and Hall effects,
perspective is nonvolatility; i.e., they can store information there is also a spin Nernst effect [141], driven by a thermal
even when the power is switched off. On the other hand, bias, that is a counterpart of the spin Hall effect [38], driven
electrical reading and especially writing information into by an electrical bias. Conventionally, these transverse spin-
ferromagnetic memory devices can generate significant current responses to the applied thermal or electrical bias
Joule heating [85]. This can be directly harvested during the did not require a magnetically ordered system and were
writing process in which the elevated temperature effec- ascribed to the relativistic spin-orbit coupling. As a result of
tively reduces the equilibrium energy barrier separating the the relativistic origin, their magnitudes were typically
states with opposite magnetization orientation. In the latest weak. In contrast, the transverse spin-current response in
generation of hard disks, elevating the temperature of a bit altermagnets can be expected to have a strong nonrelativ-
while recording is provided through an external laser heat istic contribution, exceeding, by orders of magnitude, the
source. Similarly, all-optical switching by laser pulses is relativistic spin Nernst contribution. The expectation is
typically accompanied by significant heating effects [137]. based on the analogy with the electrically generated
This brings us to a discussion of how altermagnetism, transverse spin current discussed earlier in Sec. IVA. In
rather than generating heat, can contribute to energy RuO2 , ab initio calculations have shown that the electri-
harvesting in devices combining heat, charge, and spin cally induced transverse spin current can have a non-
phenomena. relativistic contribution, the so-called spin-splitter effect,

040501-20
EMERGING RESEARCH LANDSCAPE OF ALTERMAGNETISM PHYS. REV. X 12, 040501 (2022)

which is orders of magnitude stronger than the relativistic As shown in Sec. II D, altermagnetism is compatible with
spin Hall effect in the same altermagnetic material [8]. this material family. Here, CaMnO3 is an example multi-
As a final remark on spin caloritronics, we recall the spin ferroic [149] that is a candidate for hosting the d-wave
magnon Nernst effect [142], where, instead of electrons, the altermagnetic phase [16].
spin current is carried by magnons. In Sec. III D, we have
highlighted the expected unconventional phenomenology C. Superconductivity
of magnons in altermagnets. Their unparalleled combina-
tion of linear dispersion and alternating chirality splitting in The family of insulating perovskite oxides brings us to
the magnon Brillouin zone is potentially highly favorable its prominent cuprate member La2 CuO4 that, upon doping,
for realizing spin-caloritronic devices. turns into a high-temperature d-wave superconductor [30].
Similar to spin caloritronics, nonmetallic materials are The recognition that this cuprate crystal belongs to a
also favorable for combining spin physics with field-effect d-wave altermagnetic spin group [16] leads to the anticipation
electronics. Earlier research demonstrated the intriguing of a new research direction exploring an interplay between
interplay of spintronic and transistor functionalities in altermagnetism and superconductivity [16,150]. Research in
ferromagnetic semiconductors [143,144]. However, the this context may include areas such as the coexistence of
field suffered from the notorious incompatibility of robust altermagnetism and unconventional superconductivity with
high-temperature ferromagnetism with semiconducting anisotropic Cooper pairing [151], altermagnetic fluctuations
band structures. Altermagnets thus open a new prospect as a pairing mechanism [151], or phenomena at altermagnet
of materials with high magnetic transition temperatures, or superconductor interfaces [152,153].
allowing us to combine spintronic and semiconducting Since altermagnets have spin-degenerate nodal surfaces
device functionalities. The initial experimental demonstra- protected by spin-group symmetries, a spin-singlet Cooper
tion of a spontaneous anomalous Hall effect in the absence pairing may occur for the corresponding momenta. For the
of an external magnetic field has been reported in a room- spin-singlet case, in analogy to conventional antiferromag-
temperature g-wave altermagnetic semiconductor MnTe nets, the 2 × 2 Cooper-pair potential matrix (gap function
[145]. Moreover, theory predicts a high sensitivity of this or order parameter) Δ̂ðkÞ satisfies Δ↑↑ ðkÞ ¼ Δ↓↓ ðkÞ ¼ 0,
and other spintronic responses to small changes of the Δ↓↑ ðkÞ ¼ −Δ↑↓ ðkÞ, and Δ↑↓ ðkÞ ¼ Δ↑↓ ð−kÞ [151]. The
Fermi level position near the valence or conduction band matrix is unitary, with corresponding zero net spin average
edge, which can be controlled by doping or electrostatic of the pairing state, and it describes even-parity wave
gating [145]. Copper pairing, including the anisotropic, e.g., d-wave,
Finally, we discuss the prospect of an interplay of pairing.
altermagnetism with dielectric and ferroelectric (ferroe- However, the altermagnetic spin-group symmetries also
lastic) materials or phases. The initial measurements of the allow for spin-split and broken T -symmetry parts of the
anomalous Hall effect and spin currents in the d-wave Brillouin zone, where ϵðs; kÞ ≠ ϵð−s; −kÞ and AHk ≠ k
altermagnet RuO2 exploited the high structural compat- (cf. fourth and sixth lines in Table II). These momenta
ibility with TiO2 [5,10,12], which is one of the most can support spin-triplet Cooper pairing. In analogy to
commonly used dielectrics in consumer electronics, as ferromagnets, the spin-triplet Cooper-pair potential matrix
well as a commonly employed memristive element in corresponding to a spin-split spin-up Fermi surface of the
ð↑Þ ð↑Þ
experimental neuromorphic devices [146]. Another altermagnet, Δ̂ð↑Þ ðkÞ, takes the form Δ↑↓ ðkÞ ¼ Δ↓↑ ðkÞ ¼
closely related dielectric, HfO2 , plays an important role ð↑Þ ð↑Þ ð↑Þ
in modern CMOS devices, and the recent discovery of Δ↓↓ ðkÞ ¼ 0 and Δ↑↑ ðkÞ ¼ −Δ↑↑ ð−kÞ [151]. The matrix
ferroelectricity in HfO2 thin films has triggered a renewed in this case is nonunitary and describes odd-parity wave
interest in the development of scalable, nonvolatile, Copper pairing [151]. Unlike ferromagnets, however, the
ferroelectric memories [147]. Interfaces of structurally altermagnetic spin-group symmetries impose the presence
compatible altermagnetic and ferroelectric (ferroelastic) of a counterpart spin-down Fermi surface with a corre-
ð↓Þ ð↓Þ
films pave the way for nonvolatile electric-field control of sponding Δ̂ð↓Þ ðk0 Þ that satisfies Δ↑↓ ðk0 Þ ¼ Δ↓↑ ðk0 Þ ¼
magnetism via a coupling between the ferroic orders. ð↓Þ ð↓Þ ð↓Þ ð↓Þ
Δ↑↑ ðk0 Þ ¼ 0, Δ↓↓ ðk0 Þ ¼ −Δ↓↓ ð−k0 Þ, and Δ↓↓ ðk0 Þ¼
Such a coupling can also be realized in multiferroic
ð↑Þ
materials hosting the magnetic and ferroelectric orders Δ↑↑ ðkÞ, where k0 ¼ AHk. On one hand, altermagnets
within the same crystal [148]. Since only insulating can thus share the spin-triplet symmetry of Cooper pairing
(semiconducting) materials can be ferroelectric, a coex- with ferromagnets. On the other hand, unlike ferromagnets,
isting ferromagnetic order is rare because ferromagnetism the altermagnetic spin-group symmetries protect a zero net
favors metallic structures, which again highlights the new spin average of the spin-triplet superconducting state. In the
opportunities brought up by altermagnetism. The promi- context of unconventional superconductivity, we again see
nent multiferroic materials are noncentrosymmetric per- that altermagnets can share features typical of ferromagnets
ovskite oxides with a compensated magnetic order [148]. or typical of antiferromagnets, and they can also show

040501-21
ŠMEJKAL, SINOVA, and JUNGWIRTH PHYS. REV. X 12, 040501 (2022)

FIG. 14. Summary of the emerging research landscape of altermagnetism.

features unparalleled in either of the two conventional V. CONCLUSION

magnetic phases.
In this work, we have described a third basic magnetic
Apart from the compatibility of altermagnetism with the
phase that emerges already on the fundamental level of
different types of Cooper pairing, altermagnetic fluctua-
tions can provide mechanisms for generating electron nonrelativistic, uncorrelated band theory of nonfrustrated
pairing that have not been explored before. Since the collinear magnets. The altermagnetic phase has unconven-
electron-phonon coupling mechanism tends to be limited tional spin-polarization orders in the direct physical space
to the conventional spin-singlet s-wave pairing [151], and reciprocal momentum space; this phase is systemati-
altermagnets can be particularly relevant for the research cally classified and described by symmetry-group theory,
of unconventional superconductivity, including both the and it is predicted to be abundant among diverse material
spin-singlet and spin-triplet anisotropic types of Cooper types. The significance of this unconventional d-wave (or
pairing. higher even-parity wave) magnetic phase is further under-
Finally, we foresee intriguing new physics at altermagnet lined by the unique ways in which altermagnetism can
or superconductor interfaces in areas including Andreev contribute to the development of fundamental physical
reflection [152] or Majorana fermion quasiparticles [153]. concepts and to the research in modern condensed-matter
On one hand, the behavior of altermagnets at these physics fields. Given the still relatively early stage of our
interfaces can be reminiscent of conventional antiferro- understanding of altermagnetism, and the limited space,
magnets when dominated by the spin-symmetry-protected our choice of discussion topics in this work should be
nodal surface. On the other hand, interface orientations regarded as broadly illustrative and provisional. We can
exposing the strong altermagnetic spin splitting can gen- anticipate that in the near future, altermagnetism will have
erate a phenomenology similar to the ferromagnet or an impact on other fields including magnetic topological
superconductor interfaces. As in the case of bulk crystals, matter or axion electrodynamics. Our current view of the
the research of interface effects can exploit the predicted emerging research landscape of altermagnetism is summa-
broad range of altermagnetic material types. rized in Fig. 14.

040501-22
EMERGING RESEARCH LANDSCAPE OF ALTERMAGNETISM PHYS. REV. X 12, 040501 (2022)

ACKNOWLEDGMENTS [11] Hua Bai, Lei Han, X. Y. Feng, Y. J. Zhou, R. X. Su, Qian
Wang, L. Y. Liao, W. X. Zhu, X. Z. Chen, Feng Pan, X. L.
We acknowledge fruitful interactions with Igor Mazin. Fan, and Cheng Song, Observation of Spin Splitting
This work was supported by Ministry of Education Torque in a Collinear Antiferromagnet RuO2 , Phys.
of the Czech Republic Grants No. LNSM-LNSpin and Rev. Lett. 128, 197202 (2022).
No. LM2018140, Czech Science Foundation Grant [12] Shutaro Karube, Takahiro Tanaka, Daichi Sugawara,
No. 19-28375X, EU FET Open RIA Grant No. 766566, Naohiro Kadoguchi, Makoto Kohda, and Junsaku Nitta,
SPIN þ X (DFG SFB TRR 173 268565370), and Elasto-Q- Observation of Spin-Splitter Torque in Collinear Antifer-
Mat (DFG SFB TRR 288 422213477). We acknowledge the romagnetic RuO2 , Phys. Rev. Lett. 129, 137201 (2022).
computing time granted on the supercomputer Mogon at [13] Libor Šmejkal, Anna Birk Hellenes, Rafael González-
Johannes Gutenberg University Mainz. Hernández, Jairo Sinova, and Tomas Jungwirth, Giant and
Tunneling Magnetoresistance in Unconventional Collin-
ear Antiferromagnets with Nonrelativistic Spin-Momen-
tum Coupling, Phys. Rev. X 12, 011028 (2022).
[14] Ding-Fu Shao, Shu-Hui Zhang, Ming Li, Chang-Beom
[1] Louis Néel, Magnetism and Local Molecular Field, Eom, and Evgeny Y. Tsymbal, Spin-Neutral Currents for
Science 174, 985 (1971). Spintronics, Nat. Commun. 12, 7061 (2021).
[2] Libor Šmejkal, Allan H. MacDonald, Jairo Sinova, Satoru [15] Hai-Yang Ma, Mengli Hu, Nana Li, Jianpeng Liu, Wang
Nakatsuji, and Tomas Jungwirth, Anomalous Hall Anti-
Yao, Jin-Feng Jia, and Junwei Liu, Multifunctional Anti-
ferromagnets, Nat. Rev. Mater. 7, 482 (2022).
ferromagnetic Materials with Giant Piezomagnetism and
[3] Libor Šmejkal, Rafael González-Hernández, Tomáš
Noncollinear Spin Current, Nat. Commun. 12, 2846 (2021).
Jungwirth, and Jairo Sinova, Crystal Time-Reversal Sym-
[16] Libor Šmejkal, Jairo Sinova, and Tomas Jungwirth,
metry Breaking and Spontaneous Hall Effect in Collinear
Beyond Conventional Ferromagnetism and Antiferro-
Antiferromagnets, Sci. Adv. 6, eaaz8809 (2020).
magnetism: A Phase with Nonrelativistic Spin and Crystal
[4] Makoto Naka, Satoru Hayami, Hiroaki Kusunose, Yuki
Rotation Symmetry, Phys. Rev. X 12, 031042 (2022).
Yanagi, Yukitoshi Motome, and Hitoshi Seo, Spin Current
[17] Yusuke Noda, Kaoru Ohno, and Shinichiro Nakamura,
Generation in Organic Antiferromagnets, Nat. Commun.
Momentum-Dependent Band Spin Splitting in Semicon-
10, 4305 (2019).
ducting MnO2 : A Density Functional Calculation, Phys.
[5] Zexin Feng, Xiaorong Zhou, Libor Šmejkal, Lei Wu,
Zengwei Zhu, Huixin Guo, Rafael González-Hernández, Chem. Chem. Phys. 18, 13294 (2016).
Xiaoning Wang, Han Yan, Peixin Qin, Xin Zhang, [18] Kyo-Hoon Ahn, Atsushi Hariki, Kwan-Woo Lee, and Jan
Haojiang Wu, Hongyu Chen, Ziang Meng, Li Liu, Kuneš, Antiferromagnetism in RuO2 as d-wave Pomer-
Zhengcai Xia, Jairo Sinova, Tomáš Jungwirth, and Zhiqi anchuk Instability, Phys. Rev. B 99, 184432 (2019).
Liu, An Anomalous Hall Effect in Altermagnetic [19] Satoru Hayami, Yuki Yanagi, and Hiroaki Kusunose,
Ruthenium Dioxide, Nat. Electron. 5, 735 (2022). Momentum-Dependent Spin Splitting by Collinear Anti-
[6] Helena Reichlova, Rafael Lopes Seeger, Rafael ferromagnetic Ordering, J. Phys. Soc. Jpn. 88, 123702
González-Hernández, Ismaila Kounta, Richard Schlitz, (2019).
Dominik Kriegner, Philipp Ritzinger, Michaela Lammel, [20] Lin-Ding Yuan, Zhi Wang, Jun-Wei Luo, Emmanuel I.
Miina Leiviskä, Vaclav Petiček, Petr Doležal, Eva Rashba, and Alex Zunger, Giant Momentum-Dependent
Schmoranzerova, Antonín Badura, Andy Thomas, Spin Splitting in Centrosymmetric Low-Z Antiferro-
Vincent Baltz, Lisa Michez, Jairo Sinova, Sebastian T. B. magnets, Phys. Rev. B 102, 014422 (2020).
Goennenwein, Tomáš Jungwirth, and Libo Smejkal, Macro- [21] Satoru Hayami, Yuki Yanagi, and Hiroaki Kusunose,
scopic Time Reversal Symmetry Breaking Arising from Bottom-up Design of Spin-Split and Reshaped Electronic
Antiferromagnetic Zeeman Effect, arXiv:2012.15651v1. Band Structures in Antiferromagnets without Spin-Orbit
[7] I. I. Mazin, K. Koepernik, M. D. Johannes, Rafael Coupling: Procedure on the Basis of Augmented Multi-
González-Hernández, and Libor Šmejkal, Prediction of poles, Phys. Rev. B 102, 144441 (2020).
Unconventional Magnetism in Doped FeSb2 , Proc. Natl. [22] Lin-Ding Yuan, Zhi Wang, Jun-Wei Luo, and Alex Zunger,
Acad. Sci. U.S.A. 118, e2108924118 (2021). Prediction of Low-Z Collinear and Noncollinear Anti-
[8] Rafael González-Hernández, Libor Šmejkal, Karel ferromagnetic Compounds Having Momentum-Dependent
Výborný, Yuta Yahagi, Jairo Sinova, Tomáš Jungwirth, Spin Splitting Even without Spin-Orbit Coupling, Phys.
and Jakub Železný, Efficient Electrical Spin Splitter Based Rev. Mater. 5, 014409 (2021).
on Nonrelativistic Collinear Antiferromagnetism, Phys. [23] Sergei A. Egorov and Robert A. Evarestov, Colossal Spin
Rev. Lett. 126, 127701 (2021). Splitting in the Monolayer of the Collinear Antiferro-
[9] Makoto Naka, Yukitoshi Motome, and Hitoshi Seo, magnet MnF2 , J. Phys. Chem. Lett. 12, 2363 (2021).
Perovskite as a Spin Current Generator, Phys. Rev. B [24] Pengfei Liu, Jiayu Li, Jingzhi Han, Xiangang Wan, and
103, 125114 (2021). Qihang Liu, Spin-Group Symmetry in Magnetic Materials
[10] Arnab Bose, Nathaniel J. Schreiber, Rakshit Jain, Ding-fu with Negligible Spin-Orbit Coupling, Phys. Rev. X 12,
Shao, Hari P. Nair, Jiaxin Sun, Xiyue S. Zhang, David A. 021016 (2022).
Muller, Evgeny Y. Tsymbal, Darrell G. Schlom, and [25] Jian Yang, Zheng-Xin Liu, and Chen Fang, Symmetry
Daniel C. Ralph, Tilted Spin Current Generated by an Invariants of Spin Space Groups in Magnetic Materials,
Antiferromagnet, Nat. Electron. 5, 263 (2022). arXiv:2105.12738.

040501-23
ŠMEJKAL, SINOVA, and JUNGWIRTH PHYS. REV. X 12, 040501 (2022)

[26] W. F. Brinkman and R. J. Elliott, Theory of Spin-Space Hall Effect in κ-Type Organic Antiferromagnets, Phys.
Groups, Proc. R. Soc. A 294, 343 (1966). Rev. B 102, 075112 (2020).
[27] D. B. Litvin and W. Opechowski, Spin Groups, Physica 76, [46] Makoto Naka, Yukitoshi Motome, and Hitoshi Seo,
538 (1974). Anomalous Hall Effect in Antiferromagnetic Perovskites,
[28] D. B. Litvin, Spin Point Groups, Acta Crystallogr. Sect. A arXiv:2208.11823.
33, 279 (1977). [47] S. López-Moreno, A. H. Romero, J. Mejía-López, A.
[29] A. Schofield, There and Back Again: From Magnets to Muñoz, and Igor V. Roshchin, First-Principles Study of
Superconductors, Physics 2, 93 (2009). Electronic, Vibrational, Elastic, and Magnetic Properties
[30] Qimiao Si, Rong Yu, and Elihu Abrahams, High- of FeF2 as a Function of Pressure, Phys. Rev. B 85,
Temperature Superconductivity in Iron Pnictides and 134110 (2012).
Chalcogenides, Nat. Rev. Mater. 1, 16017 (2016). [48] S. López-Moreno, A. H. Romero, J. Mejía-López, and A.
[31] Congjun Wu, Kai Sun, Eduardo Fradkin, and Shou-Cheng Muñoz, First-Principles Study of Pressure-Induced Struc-
Zhang, Fermi Liquid Instabilities in the Spin Channel, tural Phase Transitions in MnF2 , Phys. Chem. Chem.
Phys. Rev. B 75, 115103 (2007). Phys. 18, 33250 (2016).
[32] Laura Classen, Andrey V. Chubukov, Carsten [49] Louis Néel, Some New Results on Antiferromagnetism and
Honerkamp, and Michael M. Scherer, Competing Orders Ferromagnetism, Rev. Mod. Phys. 25, 58 (1953).
at Higher-Order Van Hove Points, Phys. Rev. B 102, [50] N. Kurti, Selected Works of Louis Neel, 1st ed. (CRC Press,
125141 (2020). London, 1988).
[33] R. A. Borzi, S. A. Grigera, J. Farrell, R. S. Perry, S. J. S. [51] T. Berlijn, P. C. Snijders, O. Delaire, H. D. Zhou, T. A.
Lister, S. L. Lee, D. A. Tennant, Y. Maeno, and A. P. Maier, H. B. Cao, S. X. Chi, M. Matsuda, Y. Wang, M. R.
Mackenzie, Formation of a Nematic Fluid at High Fields Koehler, P. R. C. Kent, and H. H. Weitering, Itinerant
in Sr3 Ru2 O7 , Science 315, 214 (2007). Antiferromagnetism in RuO2 , Phys. Rev. Lett. 118,
[34] F. Duncan M. Haldane, Nobel Lecture: Topological 077201 (2017).
Quantum Matter, Rev. Mod. Phys. 89, 040502 [52] Z. H. Zhu, J. Strempfer, R. R. Rao, C. A. Occhialini, J.
(2017). Pelliciari, Y. Choi, T. Kawaguchi, H. You, J. F. Mitchell, Y.
[35] J. Georg Bednorz and K. Alex Müller, Perovskite-Type Shao-Horn, and R. Comin, Anomalous Antiferro-
Oxides—The New Approach to High-T c Superconducti- magnetism in Metallic RuO2 Determined by Resonant
vity. Nobel Lecture, Angew. Chem., Int. Ed. Engl. 27, 735 X-ray Scattering, Phys. Rev. Lett. 122, 017202
(1988). (2019).
[36] Horst L. Stormer, Nobel Lecture: The Fractional Quantum [53] S. W. Lovesey, D. D. Khalyavin, and G. van der Laan,
Hall Effect, Rev. Mod. Phys. 71, 875 (1999). Magnetic Properties of Ruthenium Dioxide (RuO2 ) and
[37] Naoto Nagaosa, Jairo Sinova, Shigeki Onoda, A. H. Charge-Magnetic Interference in Bragg Diffraction of
MacDonald, and N. P. Ong, Anomalous Hall Effect, Circularly Polarized X-rays, Phys. Rev. B 105, 014403
Rev. Mod. Phys. 82, 1539 (2010). (2022).
[38] Jairo Sinova, Sergio O. Valenzuela, J. Wunderlich, C. H. [54] Connor A. Occhialini, Valentina Bisogni, Hoydoo
Back, and T. Jungwirth, Spin Hall Effects, Rev. Mod. Phys. You, Andi Barbour, Ignace Jarrige, J. F. Mitchell,
87, 1213 (2015). Riccardo Comin, and Jonathan Pelliciari, Local Electronic
[39] T. Jungwirth, X. Marti, P. Wadley, and J. Wunderlich, Structure of Rutile RuO2 , Phys. Rev. Res. 3, 033214 (2021).
Antiferromagnetic Spintronics, Nat. Nanotechnol. 11, 231 [55] K. Ishizaka et al., Giant Rashba-Type Spin Splitting in
(2016). Bulk BiTeI, Nat. Mater. 10, 521 (2011).
[40] A. Manchon, J. Železný, I. M. Miron, T. Jungwirth, J. [56] Aleksandr F. Andreev and V. I. Marchenko, Symmetry and
Sinova, A. Thiaville, K. Garello, and P. Gambardella, the Macroscopic Dynamics of Magnetic Materials, Usp.
Current-Induced Spin-Orbit Torques in Ferromagnetic Fiz. Nauk 130, 39 (1980).
and Antiferromagnetic Systems, Rev. Mod. Phys. 91, [57] Peng-Jie Guo, Yi-Wen Wei, Kai Liu, Zheng-Xin Liu, and
035004 (2019). Zhong-Yi Lu, Eightfold Degenerate Fermions in Two
[41] Contemporary Concepts of Condensed Matter Science, Dimensions, Phys. Rev. Lett. 127, 176401 (2021).
edited by M. Franz and L. Molenkamp, Topological [58] Pengfei Liu, Jingzhi Han, and Qihang Liu, Flavor Weyl
Insulators Vol. 6 (Elsevier, New York, 2013). Fermions Protected by SU(2) Isospin Symmetry in Spin-
[42] Yoshinori Tokura, Kenji Yasuda, and Atsushi Tsukazaki, Orbit-Free Antiferromagnetic Semimetals, arXiv:2107
Magnetic Topological Insulators, Nat. Rev. Phys. 1, 126 .09984.
(2019). [59] A. Corticelli, R. Moessner, and P. A. McClarty, Spin-Space
[43] Satoru Nakatsuji and Ryotaro Arita, Topological Magnets: Groups and Magnon Band Topology, Phys. Rev. B 105,
Functions Based on Berry Phase and Multipoles, Annu. 064430 (2022).
Rev. Condens. Matter Phys. 13, 119 (2022). [60] Ilja Turek, Altermagnetism and Magnetic Groups with
[44] Kartik Samanta, Marjana Ležaić, Maximilian Merte, Frank Pseudoscalar Electron Spin, Phys. Rev. B 106, 094432
Freimuth, Stefan Blügel, and Yuriy Mokrousov, Crystal (2022).
Hall and Crystal Magneto-Optical Effect in Thin Films of [61] Patrick M. Lenggenhager, Xiaoxiong Liu, Titus Neupert,
SrRuO3 , J. Appl. Phys. 127, 213904 (2020). and Tomáš Bzdušek, Triple Nodal Points Characterized by
[45] Makoto Naka, Satoru Hayami, Hiroaki Kusunose, Yuki Their Nodal-Line Structure in All Magnetic Space Groups,
Yanagi, Yukitoshi Motome, and Hitoshi Seo, Anomalous Phys. Rev. B 106, 085128 (2022).

040501-24
EMERGING RESEARCH LANDSCAPE OF ALTERMAGNETISM PHYS. REV. X 12, 040501 (2022)

[62] L. D. Landau and E. M. Lifshitz, in Electrodynamics of [80] Libor Šmejkal, Yuriy Mokrousov, Binghai Yan, and
Continuous Media, Course of Theoretical Physics Vol. 8, Allan H. MacDonald, Topological Antiferromagnetic
2nd ed. (Pergamon Press, Oxford, 1965). Spintronics, Nat. Phys. 14, 242 (2018).
[63] N. V. Shubnikov and A. V. Belov, Colored Symmetry [81] Topology in Magnetism, edited by Jiadong Zang, Vincent
(MacMillan Publishers, New York, 1964). Cros, and Axel Hoffmann, Springer Series in Solid-State
[64] B. A. Tavger and V. M. Zaitsev, Magnetic Symmetry of Sciences Vol. 192 (Springer International Publishing,
Crystals, Sov. Phys. JETP 3, 430 (1956). Cham, 2018).
[65] C. J. Bradley and A. P. Cracknell, The Mathematical [82] M. G. Vergniory, L. Elcoro, Claudia Felser, Nicolas
Theory of Symmetry in Solids (Oxford University Press, Regnault, B. Andrei Bernevig, and Zhijun Wang, A
New York, 1972). Complete Catalogue of High-Quality Topological
[66] D. B. Litvin, Magnetic Group Tables (International Union Materials, Nature (London) 566, 480 (2019).
of Crystallography, Chester, England, 2013). [83] Yuanfeng Xu, Luis Elcoro, Zhi Da Song, Benjamin J.
[67] Samuel V. Gallego, J. Manuel Perez-Mato, Luis Elcoro, Wieder, M. G. Vergniory, Nicolas Regnault, Yulin Chen,
Emre S. Tasci, Robert M. Hanson, Koichi Momma, Mois I. Claudia Felser, and B. Andrei Bernevig, High-Throughput
Aroyo, and Gotzon Madariaga, MAGNDATA: Towards a Calculations of Magnetic Topological Materials, Nature
Database of Magnetic Structures. I. The Commensurate (London) 586, 702 (2020).
Case, J. Appl. Crystallogr. 49, 1750 (2016). [84] Luis Elcoro, Benjamin J. Wieder, Zhida Song, Yuanfeng
[68] A. Corticelli, R Moessner, and P. A. McClarty, Spin-Space Xu, Barry Bradlyn, and B. Andrei Bernevig, Magnetic
Groups and Magnon Band Topology, Phys. Rev. B 105, Topological Quantum Chemistry, Nat. Commun. 12, 5965
064430 (2022). (2021).
[69] E. A. Turov, Physical Properties of Magnetically Ordered [85] Claude Chappert, Albert Fert, and Frédéric Nguyen Van
Crystals (Academic Press, New York, 1965). Dau, The Emergence of Spin Electronics in Data Storage,
[70] A. S. Núñez, R. A. Duine, Paul Haney, and A. H. Nat. Mater. 6, 813 (2007).
MacDonald, Theory of Spin Torques and Giant Magneto- [86] D. C. Ralph and M. D. Stiles, Spin Transfer Torques,
resistance in Antiferromagnetic Metals, Phys. Rev. B 73, J. Magn. Magn. Mater. 320, 1190 (2008).
214426 (2006). [87] Sabpreet Bhatti, Rachid Sbiaa, Atsufumi Hirohata, Hideo
[71] Christoph Sürgers, Gerda Fischer, Patrick Winkel, and Ohno, Shunsuke Fukami, and S. N. Piramanayagam,
Hilbert V. Löhneysen, Large Topological Hall Effect Spintronics Based Random Access Memory: A Review,
in the Non-collinear Phase of an Antiferromagnet, Nat. Mater. Today 20, 530 (2017).
Commun. 5, 3400 (2014). [88] Roland Winkler, Spin–Orbit Coupling Effects in Two-
[72] Christoph Sürgers, Wolfram Kittler, Thomas Wolf, and Dimensional Electron and Hole Systems, Springer Tracts
Hilbert V. Löhneysen, Anomalous Hall Effect in the Non- in Modern Physics (Springer, Berlin, Heidelberg, 2003),
collinear Antiferromagnet Mn5 Si3 , AIP Adv. 6, 055604 Vol. 191.
(2016). [89] N. P. Armitage, E. J. Mele, and Ashvin Vishwanath, Weyl
[73] Nirmal J. Ghimire, A. S. Botana, J. S. Jiang, Junjie Zhang, and Dirac Semimetals in Three-Dimensional Solids, Rev.
Y.-S. Chen, and J. F. Mitchell, Large Anomalous Hall Mod. Phys. 90, 015001 (2018).
Effect in the Chiral-Lattice Antiferromagnet CoNb3 S6 , [90] X. Marti, I. Fina, C. Frontera, Jian Liu, P. Wadley, Q. He,
Nat. Commun. 9, 3280 (2018). R. J. Paull, J. D. Clarkson, J. Kudrnovský, I. Turek, J.
[74] Joel E. Moore, The Birth of Topological Insulators, Nature Kuneš, D. Yi, J.-H. Chu, C. T. Nelson, L. You, E.
(London) 464, 194 (2010). Arenholz, S. Salahuddin, J. Fontcuberta, T. Jungwirth,
[75] Xin Chen, Duo Wang, Linyang Li, and Biplab and R. Ramesh, Room-Temperature Antiferromagnetic
Sanyal, Room Temperature Two-Dimensional Antiferro- Memory Resistor, Nat. Mater. 13, 367 (2014).
magnetic Weyl Semimetal CrO with Giant Spin-Splitting [91] M. M. Otrokov et al., Prediction and Observation of an
and Spin-Momentum Locked Transport, arXiv:2104 Antiferromagnetic Topological Insulator, Nature (London)
.07390. 576, 416 (2019).
[76] Takuya Okugawa, Kaoru Ohno, Yusuke Noda, and [92] J. Železný, H. Gao, K. Výborný, J. Zemen, J. Mašek,
Shinichiro Nakamura, Weakly Spin-Dependent Band Struc- Aurélien Manchon, J. Wunderlich, Jairo Sinova, and T.
tures of Antiferromagnetic Perovskite LaMO3 (M ¼ Cr, Jungwirth, Relativistic Néel-Order Fields Induced by
Mn, Fe), J. Phys. Condens. Matter 30, 075502 (2018). Electrical Current in Antiferromagnets, Phys. Rev. Lett.
[77] H. A. Kramers, Théorie Générale de la Rotation Para- 113, 157201 (2014).
magnétique Dans les Cristaux., Proc. Amsterdam Acad. 33 [93] P. Wadley et al., Electrical Switching of an Antiferromag-
(1930). net, Science 351, 587 (2016).
[78] E. Wigner, Ueber die Operation der Zeitumkehr in der [94] Libor Šmejkal, J. Železný, J. Sinova, and T. Jungwirth,
Quantenmechanik, Nachrichten von der Gesellschaft der Electric Control of Dirac Quasiparticles by Spin-Orbit
Wissenschaften zu Göttingen, Mathematisch-Physikalische Torque in an Antiferromagnet, Phys. Rev. Lett. 118,
Klasse 1932, 546 (1932). 106402 (2017).
[79] Barry Bradlyn, L. Elcoro, Jennifer Cano, M. G. Vergniory, [95] H. J. Elmers, S. V. Chernov, S. W. D’Souza, S. P.
Zhijun Wang, C. Felser, M. I. Aroyo, and B. Andrei Bommanaboyena, S. Yu Bodnar, K. Medjanik, S.
Bernevig, Topological Quantum Chemistry, Nature Babenkov, O. Fedchenko, D. Vasilyev, S. Y. Agustsson,
(London) 547, 298 (2017). C. Schlueter, A. Gloskovskii, Yu Matveyev, V. N. Strocov,

040501-25
ŠMEJKAL, SINOVA, and JUNGWIRTH PHYS. REV. X 12, 040501 (2022)

Y. Skourski, L. Šmejkal, J. Sinova, J. Minár, M. Kläui, G. [110] R. Lebrun, A. Ross, S. A. Bender, A. Qaiumzadeh, L.
Schönhense, and M. Jourdan, Néel Vector Induced Baldrati, J. Cramer, A. Brataas, R. A. Duine, and M. Kläui,
Manipulation of Valence States in the Collinear Anti- Tunable Long-Distance Spin Transport in a Crystalline
ferromagnet Mn2 Au, ACS Nano 14, 17554 (2020). Antiferromagnetic Iron Oxide, Nature (London) 561, 222
[96] Olena Fedchenko, Libor Smejkal, Michael Kallmayer, (2018).
Yarina Lytvynenko, Katerina Medjanik, Sergey Babenkov, [111] M. V. Berry, Quantal Phase Factors Accompanying Adia-
Dmitry Vasilyev, Matthias Klaeui, Jure Demsar, Gerd batic Changes, Proc. R. Soc. A 392, 45 (1984).
Schoenhense, Martin Jourdan, Jairo Sinova, and Hans- [112] Dimitrie Culcer, Allan MacDonald, and Qian Niu, Anoma-
Joachim Elmers, Direct Observation of Antiferromagnetic lous Hall Effect in Paramagnetic Two-Dimensional Sys-
Parity Violation in the Electronic Structure of Mn2 Au, tems, Phys. Rev. B 68, 045327 (2003).
arXiv:2110.12186. [113] Tamara S. Nunner, N. A. Sinitsyn, Mario F. Borunda, V. K.
[97] Revaz Ramazashvili, Kramers Degeneracy in a Magnetic Dugaev, A. A. Kovalev, Ar. Abanov, Carsten Timm, T.
Field and Zeeman Spin-Orbit Coupling in Antiferromag- Jungwirth, Jun-ichiro Inoue, A. H. MacDonald, and Jairo
netic Conductors, Phys. Rev. Lett. 101, 137202 (2008). Sinova, Anomalous Hall Effect in a Two-Dimensional
[98] Revaz Ramazashvili, Kramers Degeneracy in a Magnetic Electron Gas, Phys. Rev. B 76, 235312 (2007).
Field and Zeeman Spin-Orbit Coupling in Antiferro- [114] V. K. Dugaev, J. Barnaś, M. Taillefumier, B. Canals, C.
magnetic Conductors, Phys. Rev. B 79, 184432 (2009). Lacroix, and P. Bruno, Anomalous Hall Effect and Berry
[99] Emil J. Rozbicki, James F. Annett, Jean-René Souquet, Phase in Two-Dimensional Magnetic Structures, J. Phys.
and Andrew P. Mackenzie, Spin-Orbit Coupling and Conf. Ser. 104, 012018 (2008).
k-Dependent Zeeman Splitting in Strontium Ruthenate, [115] Di Xiao, Ming Che Chang, and Qian Niu, Berry Phase
J. Phys. Condens. Matter 23, 094201 (2011). Effects on Electronic Properties, Rev. Mod. Phys. 82, 1959
[100] T. Suzuki, R. Chisnell, A. Devarakonda, Y. T. Liu, W. (2010).
Feng, D. Xiao, J. W. Lynn, and J. G. Checkelsky, Large [116] Hitoshi Seo and Makoto Naka, Antiferromagnetic State in
Anomalous Hall Effect in a Half-Heusler Antiferromagnet, κ-type Molecular Conductors: Spin Splitting and Mott
Nat. Phys. 12, 1119 (2016). Gap, J. Phys. Soc. Jpn. 90, 064713 (2021).
[101] Kunihiko Yamauchi, Paolo Barone, and Silvia Picozzi, [117] Ding Fu Shao, Jun Ding, Gautam Gurung, Shu Hui Zhang,
Bulk Rashba Effect in Multiferroics: A Theoretical Pre- and Evgeny Y. Tsymbal, Interfacial Crystal Hall Effect
diction for BiCoO3, Phys. Rev. B 100, 245115 (2019). Reversible by Ferroelectric Polarization, Phys. Rev. Appl.
[102] Benjamin Schrunk, Yevhen Kushnirenko, Brinda 15, 024057 (2021).
Kuthanazhi, Junyeong Ahn, Lin-Lin Wang, Evan [118] M. Z. Hasan and C. L. Kane, Colloquium: Topological
O’Leary, Kyungchan Lee, Andrew Eaton, Alexander Insulators, Rev. Mod. Phys. 82, 3045 (2010).
Fedorov, Rui Lou, Vladimir Voroshnin, Oliver J. Clark, [119] K. V. Klitzing, G. Dorda, and M. Pepper, New Method for
Jaime Sánchez-Barriga, Sergey L. Bud’ko, Robert-Jan High-Accuracy Determination of the Fine-Structure
Slager, Paul C. Canfield, and Adam Kaminski, Emergence Constant Based on Quantized Hall Resistance, Phys.
of Fermi Arcs Due to Magnetic Splitting in an Antiferro- Rev. Lett. 45, 494 (1980).
magnet, Nature (London) 603, 610 (2022). [120] K. S. Novoselov, Z. Jiang, Y. Zhang, S. V. Morozov, H. L.
[103] I. I. Pomeranchuk, On the Stability of a Fermi Liquid, Stormer, U. Zeitler, J. C. Maan, G. S. Boebinger, P. Kim,
Phys. JETP 8, 361 (1959). and A. K. Geim, Room-Temperature Quantum Hall Effect
[104] Sergei A. Egorov, Daniel B. Litvin, and Robert A. in Graphene, Science 315, 1379 (2007).
Evarestov, Antiferromagnetism-Induced Spin Splitting in [121] C.-Z. Chang et al., Experimental Observation of the
Systems Described by Magnetic Layer Groups, J. Phys. Quantum Anomalous Hall Effect in a Magnetic Topo-
Chem. C 125, 16147 (2021). logical Insulator, Science 340, 167 (2013).
[105] John R. Schaibley, Hongyi Yu, Genevieve Clark, Pasqual [122] Yujun Deng, Yijun Yu, Meng Zhu Shi, Zhongxun Guo,
Rivera, Jason S. Ross, Kyle L. Seyler, Wang Yao, and Zihan Xu, Jing Wang, Xian Hui Chen, and Yuanbo Zhang,
Xiaodong Xu, Valleytronics in 2D Materials, Nat. Rev. Quantum Anomalous Hall Effect in Intrinsic Magnetic
Mater. 1, 16055 (2016). Topological Insulator MnBi2 Te4, Science 367, 895
[106] A. Marmodoro, L. Šmejkal, J. Sinova, and T. Jungwirth (2020).
(to be published). [123] Samuel D. Bader and S. S. P. Parkin, Spintronics, Annu.
[107] S. V. Halilov, H. Eschrig, A. Y. Perlov, and P. M. Oppeneer, Rev. Condens. Matter Phys. 1, 71 (2010).
Adiabatic Spin Dynamics from Spin-Density-Functional [124] Peter Wadley, Sonka Reimers, Michal J. Grzybowski, Carl
Theory: Application to Fe, Co, and Ni, Phys. Rev. B 58, Andrews, Mu Wang, Jasbinder S. Chauhan, Bryan L.
293 (1998). Gallagher, Richard P. Campion, Kevin W. Edmonds,
[108] Alberto Marmodoro, Sergiy Mankovsky, Hubert Ebert, Jan Sarnjeet S. Dhesi, Francesco Maccherozzi, Vit Novak,
Minár, and Ondej Ondej Ondejšipr, Electric Field Control Joerg Wunderlich, and Tomas Jungwirth, Current Polarity-
of Magnons in Magnetic Thin Films: Ab Initio Predictions Dependent Manipulation of Antiferromagnetic Domains,
for 2D Metallic Heterostructures, Phys. Rev. B 105, Nat. Nanotechnol. 13, 362 (2018).
174411 (2022). [125] Yang Zhang, Qiunan Xu, Klaus Koepernik, Roman
[109] Sergio M. Rezende, Antonio Azevedo, and Roberto L. Rezaev, Oleg Janson, Jakub Železný, Tomáš Jungwirth,
Rodríguez-Suárez, Introduction to Antiferromagnetic Claudia Felser, Jeroen van den Brink, and Yan Sun,
Magnons, J. Appl. Phys. 126, 151101 (2019). Different Types of Spin Currents in the Comprehensive

040501-26
EMERGING RESEARCH LANDSCAPE OF ALTERMAGNETISM PHYS. REV. X 12, 040501 (2022)

Materials Database of Nonmagnetic Spin Hall Effect, npj [140] Qing Jie, Rongwei Hu, Emil Bozin, A. Llobet, I.
Comput. Mater. 7, 167 (2021). Zaliznyak, C. Petrovic, and Q. Li, Electronic Thermo-
[126] Arne Brataas, Andrew D. Kent, and Hideo Ohno, Current- electric Power Factor and Metal-Insulator Transition in
Induced Torques in Magnetic Materials, Nat. Mater. 11, FeSb2 , Phys. Rev. B 86, 115121 (2012).
372 (2012). [141] S. Meyer, Y. T. Chen, S. Wimmer, M. Althammer, T.
[127] R. A. Duine, Kyung-Jin Lee, Stuart S. P. Parkin, and M. D. Wimmer, R. Schlitz, S. Geprags, H. Huebl, D.
Stiles, Synthetic Antiferromagnetic Spintronics, Nat. Phys. Kodderitzsch, H. Ebert, G. E. W. Bauer, R. Gross, and
14, 217 (2018). S. T. B. Goennenwein, Observation of the Spin Nernst
[128] S. Schlauderer, C. Lange, S. Baierl, T. Ebnet, C. P. Schmid, Effect, Nat. Mater. 16, 977 (2017).
D. C. Valovcin, A. K. Zvezdin, A. V. Kimel, R. V. [142] Hantao Zhang and Ran Cheng, A Perspective on Magnon
Mikhaylovskiy, and R. Huber, Temporal and Spectral Spin Nernst Effect in Antiferromagnets, Appl. Phys. Lett.
Fingerprints of Ultrafast All-Coherent Spin Switching, 120, 090502 (2022).
Nature (London) 569, 383 (2019). [143] Tomasz Dietl and Hideo Ohno, Dilute Ferromagnetic
[129] R. Landauer, Irreversibility and Heat Generation in the Semiconductors: Physics and Spintronic Structures, Rev.
Computing Process, IBM J. Res. Dev. 5, 183 (1961). Mod. Phys. 86, 187 (2014).
[130] Charles H. Bennett, The Thermodynamics of Computation [144] T. Jungwirth, J. Wunderlich, V. Novák, K. Olejník, B. L.
—A Review, Int. J. Theor. Phys. 21, 905 (1982). Gallagher, R. P. Campion, K. W. Edmonds, A. W.
[131] Jeongmin Hong, Brian Lambson, Scott Dhuey, and Jeffrey Rushforth, A. J. Ferguson, and P. Némec, Spin-Dependent
Bokor, Experimental Test of Landauer’s Principle in Phenomena and Device Concepts Explored in (Ga,Mn)As,
Single-Bit Operations on Nanomagnetic Memory Bits, Rev. Mod. Phys. 86, 855 (2014).
Sci. Adv. 2 (2016). [145] R. D. Gonzalez Betancourt, J. Zubáč, R. J. Gonzalez-
[132] R. Gaudenzi, E. Burzurí, S. Maegawa, H. S. J. van der Hernandez, K. Geishendorf, Z. Šobá, G. Springholz, K.
Zant, and F. Luis, Quantum Landauer Erasure with a Výborný, K. Olejník, L. Šmejkal, J. Sinova, T. Jungwirth,
Molecular Nanomagnet, Nat. Phys. 14, 565 (2018). S. T. B. Goennenwein, A. Thomas, H. Reichlová, J.
[133] A. V. Kimel, A. M. Kalashnikova, A. Pogrebna, and A. K. Železný, and D. Kriegner, Spontaneous Anomalous Hall
Zvezdin, Fundamentals and Perspectives of Ultrafast Effect Arising from Antiparallel Magnetic Order in a
Photoferroic Recording, Phys. Rep. 852, 1 (2020). Semiconductor, arXiv:2112.06805.
[134] Z. Kašpar, M. Surýnek, J. Zubáč, F. Krizek, V. Novák, R. P. [146] Georgii A. Illarionov, Sofia M. Morozova, Vladimir V.
Campion, M. S. Wörnle, P. Gambardella, X. Marti, P. Chrishtop, Mari-Ann Einarsrud, and Maxim I. Morozov,
Němec, K. W. Edmonds, S. Reimers, O. J. Amin, F. Memristive TiO2 : Synthesis, Technologies, and Applica-
Maccherozzi, S. S. Dhesi, P. Wadley, J. Wunderlich, K. tions, Front. Chem. 8, 724 (2020).
Olejník, and T. Jungwirth, Quenching of an Antiferromagnet [147] Writam Banerjee, Alireza Kashir, and Stanislav Kamba,
into High Resistivity States Using Electrical or Ultrashort Hafnium Oxide (HfO2 )—A Multifunctional Oxide: A
Optical Pulses, National electronics review 4, 30 (2021). Review on the Prospect and Challenges of Hafnium Oxide
[135] Jan Zubáč, Zdeněk Kašpar, Filip Krizek, Tobias Förster, in Resistive Switching and Ferroelectric Memories, Small
Richard P. Campion, Vít Novák, Tomáš Jungwirth, and 18, 2107575 (2022).
Kamil Olejník, Hysteretic Effects and Magnetotransport of [148] R. Ramesh and Nicola A. Spaldin, Multiferroics:
Electrically Switched CuMnAs, Phys. Rev. B 104, 184424 Progress and Prospects in Thin Films, Nat. Mater. 6,
(2021). 21 (2007).
[136] Filip Krizek, Sonka Reimers, Zdeněk Kašpar, Alberto [149] Satadeep Bhattacharjee, Eric Bousquet, and Philippe
Marmodoro, Jan Michalička, Ondej Man, Alexander Ghosez, Engineering Multiferroism in CaMnO3 , Phys.
Edstrom, Oliver J. Amin, Kevin W. Edmonds, Richard Rev. Lett. 102, 117602 (2009).
P. Campion, Francesco Maccherozzi, Sarnjeet S. Dnes, Jan [150] Igor I. Mazin, Notes on Altermagnetism and Super-
Zubáč, Jakub Železný, Karel Výborný, Kamil Olejník, Vít conductivity, arXiv:2203.05000.
Novák, Jan Rusz, Juan C. Idrobo, Peter Wadley, and Tomas [151] Manfred Sigrist and Kazuo Ueda, Phenomenological
Jungwirth, Atomically Sharp Domain Walls in an Anti- Theory of Unconventional Superconductivity, Rev. Mod.
ferromagnet, Sci. Adv. 8, eabn3535 (2022). Phys. 63, 239 (1991).
[137] Alexey V. Kimel and Mo Li, Writing Magnetic Memory with [152] I. I. Mazin, A. A. Golubov, and B. Nadgorny, Probing Spin
Ultrashort Light Pulses, Nat. Rev. Mater. 4, 189 (2019). Polarization with Andreev Reflection: A Theoretical Basis,
[138] Gerrit E. W. Bauer, Eiji Saitoh, and Bart J. Van Wees, Spin J. Appl. Phys. 89, 7576 (2001).
Caloritronics, Nat. Mater. 11, 391 (2012). [153] Karsten Flensberg, Felix von Oppen, and Ady Stern,
[139] Masaki Mizuguchi and Satoru Nakatsuji, Energy- Engineered Platforms for Topological Superconductivity
Harvesting Materials Based on the Anomalous Nernst and Majorana Zero Modes, Nat. Rev. Mater. 6, 944
Effect, Sci. Technol. Adv. Mater. 20, 262 (2019). (2021).

040501-27