PHYSICAL REVIEW MATERIALS 8, 074407 (2024)

Tuning magnetotransport properties through chemical substitution

in Pr- and Sm-doped NdRuO3 thin films

Takahiro C. Fujita ,* Shunsuke Senzaki, and Ling-Fei Zhang

Department of Applied Physics and Quantum-Phase Electronics Center (QPEC), University of Tokyo, Tokyo 113-8656, Japan

Masashi Kawasaki
Department of Applied Physics and Quantum-Phase Electronics Center (QPEC), University of Tokyo, Tokyo 113-8656, Japan
and Center for Emergent Matter Science (CEMS), RIKEN, Saitama 351-0198, Japan

(Received 10 April 2024; accepted 25 June 2024; published 16 July 2024)

We have investigated the magnetotransport properties for solid-solution thin films of Nd1−x Pr x RuO3 and
Nd1−y Smy RuO3 prepared through a solid phase epitaxy. It is found that the peculiar magnetotransport properties
that emerge in NdRuO3 are modified by the systematic control of the electron correlation strength through the
tuning of the bandwidth as well as in the total angular momentum J of rare earth magnetic ions. The transition
from metal to insulator is observed with increasing the concentration y of smaller Sm3+ in Nd1−y Smy RuO3 ,
while Nd1−x Pr x RuO3 is kept metallic with increasing x for larger Pr 3+ . An anomalous Hall effect is observed
for all the conducting samples with the maximum Hall angle appearing for NdRuO3 . The topological Hall effect
observed for NdRuO3 gradually diminishes with increasing the doping concentrations x and y.

DOI: 10.1103/PhysRevMaterials.8.074407

I. INTRODUCTION Recently, we have demonstrated an epitaxial stabiliza-

tion of LaRuO3 and NdRuO3 thin films with a solid phase
Transition metal perovskite oxide (AMO3 , M: transition
epitaxy technique and performed comprehensive magneto-
metal) is an intriguing platform for studying the physical
transport measurements at ultra-low temperatures down to
properties of strongly correlated electron systems. The intri-
50 mK and under high magnetic fields up to 54 T [15]. Despite
cate interplay between the charge, orbital, and spin degrees
similarly metallic behaviors in both compounds, magneto-
of freedom manifests a myriad of emergent phenomena. The
transport properties are quite different, which can be attributed
interplay between electron correlation (U) and spin-orbit in-
to the distinction between the nonmagnetic La3+ and mag-
teraction (SOI) has recently attracted considerable attention
netic Nd3+ ions. LaRuO3 is a paramagnetic metal exhibiting
as an important origin for realizing novel electronic and mag-
a positive magnetoresistance and linear Hall effect down to
netic phases [1]. In perovskite iridates, the presence of the
50 mK. NdRuO3 , on the other hand, exhibits a nonlinear
Jeff = 1/2 state for 5d 5 electrons on Ir 4+ has led to the dis-
Hall effect and negative magnetoresistance, as in the case of
covery of various unconventional properties, such as Dirac
typical magnetic metals. NdRuO3 shows an unconventional
semimetal for AIrO3 [2–4] and spin-orbit Mott-insulator state
anomalous Hall effect (AHE), namely, the topological Hall
for Sr 2 IrO4 [5]. Similarly, 4d 5 electron configuration in Ru3+
effect (THE), below 1.5 K plausibly originating from the
is of considerable interest, while the SOI is not as strong as
noncoplanar spin texture of Nd3+ ions and resultant scalar
in Ir 4+ . Indeed, owing to the Jeff = 1/2 state, α-RuCl3 is the-
spin chirality. These findings indicate a crucial role in the
oretically proposed to be Kitaev quantum spin liquid [6] and
transport properties of the f -d exchange interaction between
experimentally substantiated through the half-integer thermal
Ln-4 f localized moments and Ru-4d itinerant electrons. Ad-
quantum Hall effect [7]. In oxides, the stable oxidation state
ditionally, the cooperation between U and SOI may potentially
of Ru is generally +4 or higher. LnRuO3 (Ln: lanthanide),
give rise to abundant emergent quantum states as in the case
which possesses an orthorhombic perovskite structure, rep-
of Ln2 Ir 2 O7 [16–18]. To gain further insights, it is a promising
resents rare exceptions wherein Ru adopts a +3 oxidation
way to control magnetic properties and U through the doping
state. Due to the inherent instability of Ru3+ , however, studies
on Ln site in LnRuO3 .
on LnRuO3 have been rather limited to bulk polycrystals
In this work, we have investigated the magnetotransport
[8–11] and single crystals [12]. Consequently, the electronic
properties for solid-solution thin films of Nd1−x Pr x RuO3 and
and magnetic properties of these compounds have not been
Nd1−y Smy RuO3 . Starting from NdRuO3 , we have substituted
revealed well, in stark contrast to an immense amount of
Nd3+ with Pr 3+ or Sm3+ , where ionic radius and total angular
previous literature dedicated to AeRu4+ O3 (Ae: alkaline earth)
momentum J are Sm3+ (1.132 Å, J = 2.5), Nd3+ (1.163 Å,
[13,14].
J = 4.5), and Pr 3+ (1.179 Å, J = 4). Chemical substitution is
a well-established way to control effective U by modulating
the one-electron bandwidth, which is the so-called “chem-
*
Contact author: fujita@ap.t.u-tokyo.ac.jp ical pressure effect” [19]. Indeed, the relationship between

2475-9953/2024/8(7)/074407(6) 074407-1 ©2024 American Physical Society

FUJITA, SENZAKI, ZHANG, AND KAWASAKI PHYSICAL REVIEW MATERIALS 8, 074407 (2024)

FIG. 1. (a) Schematics of the sample preparation method. Amorphous precursor film is deposited by pulsed laser deposition while
switching two targets, and then annealed ex situ to obtain a solid-solution epitaxial film with a thickness of ∼10 nm. (b) X-ray diffraction θ -2θ
scans for Nd1−x Pr x RuO3 and Nd1−y Smy RuO3 films taken around STO (002) peak (vertically shifted for clarity). Peaks from STO substrate and
films are marked with an asterisk and triangles, respectively. (c) Out-of-plane lattice spacing d001 deduced from the (002) peak position of the
films as a function of the nominal doping concentration x and y. Error bars are indicated for Sm-rich films (y  0.5) due to the ambiguity of
the peak positions.

the bandwidth and the average size of A-site ions has been deposited at a substrate temperature of 600 ◦ C and an oxygen
well studied for perovskite oxides. Therefore, the present pressure of 0.1 mTorr as shown in Fig. 1(a). The doping
solid-solution films based on NdRuO3 provide an ideal plat- concentration was controlled by the thickness ratio of two
form to study the bandwidth control effect on its electronic constituent layers. Here it is worth noting that the doping con-
state, and to gain a hint for the origin of the unconventional centrations x and y for Nd1−x Pr x RuO3 and Nd1−y Smy RuO3
AHE. We have revealed that the electronic state changes from in this work are nominal values. Because all three Ln-site
metal to insulator upon doping smaller Sm3+ reminiscent elements (Pr, Nd, Sm) are located close to each other on the
of the Mott transition. Negative magnetoresistance at lower periodic table, it is unable to determine the actual dopant
temperatures in NdRuO3 is enhanced (suppressed) with dop- levels of the films through conventional energy-dispersive
ing Sm (Pr) and eventually becomes positive for PrRuO3 . In x-ray spectroscopy technique. The total thicknesses of the
addition, we have observed the reduction of anomalous Hall films were designed to be ∼10 nm, which was confirmed
angle and the disappearance of the THE on both sides of with x-ray reflectivity measurements at a low-angle range.
heavy doping. The amorphous films were then annealed ex situ with RuO2
powder at 1000 ◦ C for 2 h under 160 ml/min N2 flow. The
annealing process promotes the epitaxial crystallization as
II. METHODS
well as the formation of a solid solution of the two constituent
Nd1−x Pr x RuO3 and Nd1−y Smy RuO3 thin films with x, y = compounds [Fig. 1(a)].
0, 0.25, 0.5, 0.75, and 1 were fabricated on SrTiO3 (STO)(001) Figure 1(b) summarizes the x-ray diffraction (XRD) pat-
substrates with a combination of pulsed laser deposition terns of the Nd1−x Pr x RuO3 and Nd1−y Smy RuO3 films. The
and ex situ annealing, following the procedure described in films are confirmed to have a (001) oriented single phase in a
Ref. [15]. Targets for LnRuO3 layers were prepared by solid- pseudocubic setting. The clear peak shift indicates that the
state reaction using Ln2 O3 (Ln = Nd or Sm) or Pr 6 O11 lattice constant systematically expands (shrinks) upon Pr (Sm)
and RuO2 powders as chemical agents. To avoid the defi- doping. Figure 1(c) shows the out-of-plane lattice spacing d001
ciency of Ru, these powders were mixed with molar ratio of deduced from the (002) peak position of the films. Note that
Ln:Ru = 1:1.25. The mixed powders were milled and calcined Sm-rich films (y  0.5) have large error bars, because their
for 24 h at 1150◦ C, forming pyrochlore Ln2 Ru2 O7 phases peak positions are adjacent to the peaks from STO substrate.
while excess Ru being remained as RuO2 . Before deposition, The deduced d001 is almost linear to the nominal doping
chemically etched STO substrates were annealed in situ at concentrations x and y, indicating uniformity of the dopants.
950 ◦ C under 10−5 Torr oxygen to obtain a clear step-terrace Coherently strained lattice structure of the films is con-
structure with single-unit-cell height [20]. A KrF excimer firmed by XRD reciprocal space mappings around STO (103)
laser (λ = 248 nm) with a pulse frequency of 5 Hz and a peak (see Supplemental Material, Fig. S1 [21], including
laser fluence of ∼2 J/cm2 was employed to ablate the targets. Refs. [15,22,23]).
By alternately ablating NdRuO3 and dopant (i.e., PrRuO3 Magnetotransport measurements were performed in a liq-
or SmRuO3 ) targets, amorphous multilayer films were first uid He cryostat with a 9 T superconducting magnet (Physical

074407-2
TUNING MAGNETOTRANSPORT PROPERTIES THROUGH … PHYSICAL REVIEW MATERIALS 8, 074407 (2024)

Property Measurement System, Quantum Design Co.) down

to 0.5 K. Ni (10 nm) and Au (50 nm) bilayer films were
deposited by electron beam evaporation as electrodes of a
Hall-bar structure fabricated by a scriber. Aluminum wires
were attached to the six electrodes to obtain longitudinal re-
sistivity (ρxx ) and Hall resistivity (ρyx ) by the four-terminal
measurement method. The magnetic field (B) was applied
perpendicularly to the film surface to deduce ρxx and ρyx by
conventional symmetrization and antisymmetrization proce-
dures, respectively.

III. RESULTS AND DISCUSSION

The temperature (T) dependence of ρxx for Nd1−x Pr x RuO3
and Nd1−y Smy RuO3 is presented in Figs. 2(a)–2(c).
Nd1−x Pr x RuO3 exhibits an overall metallic behavior accom-
panied by a slight upturn at lower temperatures as in the
case of NdRuO3 , while the upturn is totally suppressed in
PrRuO3 as can be seen in the magnified Fig. 2(a). In contrast,
Nd1−y Smy RuO3 shows a systematic transition from metal to
insulator upon increasing y. This difference in ρxx -T curves
with chemical substitution shows an implication that the elec-
tronic state is systematically shifted from metal in PrRuO3
to insulator in SmRuO3 . This is, indeed, consistent with the
chemical pressure effects; the larger Ln ion increases the Ru-
O-Ru bond angle to suppress effective U and enhance metallic
states [19]. We fit the ρxx -T curves for Nd1−y Smy RuO3 , as-
suming a three-dimensional Mott variable-range hopping to
confirm the enhanced localization (Supplemental Material,
Fig. S2 [21]). Our thin films are strained on the substrate as
discussed above (Fig. S1 [21]), thus the films are imposed by
not only chemical pressure but also a physical one. Neverthe-
less, the observed trend in ρxx -T curves can be explained by
the chemical pressure effects, thus U may be responsible for
the transport properties in the range of Ln discussed in the
present work.
We measured the T dependence of the magnetoresistance
ratio under B = 9 T (MRR ≡ ρxx (9 T)/ρxx (0) − 1) for
the NdRuO3 , SmRuO3 , and PrRuO3 films, as displayed in
Fig. 2(d). MRR is almost zero at higher temperatures than
∼100 K for all compounds, which is followed by positive
MRR upon further cooling. As temperature decreases, posi-
tive MRR saturates at ∼10 (∼20) K for NdRuO3 (SmRuO3 )
and turns negative at lower temperatures. In contrast, positive
MRR keeps increasing as one lowers the temperature for
PrRuO3 . We also measured B dependence of MRR at various
temperatures (see Supplemental Material, Fig. S3 [21]). This
contrasting trend of MRR depending on Ln is of interest as FIG. 2. Temperature dependence of longitudinal resistivity (ρxx )
will be discussed later. for (a), (b) Nd1−x Pr x RuO3 and (c) Nd1−y Smy RuO3 films. The vertical
We now turn to the Hall effect of the films. Magnetic axis is in linear scale in panel (a) and the logarithmic scales for panels
field dependences of ρyx are shown in Fig. S4 [21] at various (b) and (c). (d) Temperature dependence of magnetoresistance ratio
temperatures. As we have reported for NdRuO3 in Ref. [15], under B = 9 T [MRR ≡ ρxx (9 T)/ρxx (0) − 1] for PrRuO3 , NdRuO3 ,
ρyx starts to contain the anomalous Hall contribution below and SmRuO3 films.
∼100 K, and the ordinary Hall coefficient (RH ) is almost
constant below 150 K, which is confirmed by the high to discuss the low-temperature behavior, where the f -d inter-
field measurements up to 54 T (see Supplemental Material, action is more prominent. Because of the increasing ρxx as
Fig. S5 [21]). Therefore, by assuming that RH is constant shown in Fig. 2(c), the data become noisy for Sm-rich films (y
below 150 K, ρAHE is calculated by ρAHE (B) ≡ ρyx (B) − = 0.5, 0.75, 1), especially at 0.5 K. The measurement at 0.5 K
RH (150 K)B for the present solid-solution films as displayed was even not possible for films with y = 0.75, 1. All the films
in Fig. 3. Here the data at 0.5 (black) and 3 K (blue) are shown exhibit nonlinear ρyx as expected for AHE. NdRuO3 has been

074407-3
FUJITA, SENZAKI, ZHANG, AND KAWASAKI PHYSICAL REVIEW MATERIALS 8, 074407 (2024)

FIG. 3. (a) Magnetic field dependence of anomalous Hall resistivity (ρAHE ) at low temperatures for (a)–(e) Nd1−y Smy RuO3 and (f)–(j)
Nd1−x Pr x RuO3 thin films. The hump structure characteristic of the topological Hall effect (see text) is denoted by the triangles. Note that the
data in panels (e) and (f) are identical, and that data at 0.5 and 3 K are overlapped in panel (j).

revealed to exhibit THE below ∼1 K due to a noncoplanar regardless of the species of A and M [25,26]. The smaller f
spin texture realized during the magnetic transition, which is indicates the larger bond angle distortion, leading to the reduc-
indicated by the triangles in Figs. 3(e) and 3(f). The unique tion of one-electron bandwidth or electron conductivity [19].
hump structures at around B = 1 T originating from THE In the present system, with using the ionic size of Ln3+ and
seemingly remain for x, y  0.5 [Figs. 3(c)–3(h)] although the Ru3+ , f is calculated as 0.859 (SmRuO3 ), 0.870 (NdRuO3 ),
data are noisy in the Sm-doped side. In particular for Pr-rich and 0.875 (PrRuO3 ); thus the observed trend is reasonably un-
films (x = 0.75, 1) in Figs. 3(i)–3(j), THE is indiscernible. derstood in the framework of a bandwidth-control driven Mott
At the same time, the kink structure in MRR observed for transition.
NdRuO3 also disappears with increasing x and y (see Sup- As shown in Fig. 2(d), the T dependence of MRR for
plemental Material, Fig. S6 [21]), indicating that these two SmRuO3 and NdRuO3 contrasts with that of PrRuO3 es-
features are attributed to the same origin: plausibly a sort of pecially at lower temperatures. This tendency is further
magnetic transition. confirmed for the solid-solution films as shown in Fig. 4(b).
To quantitatively discuss the magnetotransport properties, MRR of NdRuO3 is negative at both 3 and 0.5 K while
the data are summarized in Fig. 4 for (a) ρxx , (b) MRR, the magnitude is larger at 0.5 K. With increasing Pr-doping
and (c) anomalous Hall angle (θAHE ≡ ρxx ρAHE /(ρxx 2 + concentration x, negative MRR is suppressed and turns into
ρAHE 2 )) at B = 9 T deduced from the transport results positive at both 3 and 0.5 K for PrRuO3 , the magnitude
(Figs. 2 and 3). Pr doping does not cause appreciable effects of which is ∼4%. In contrast, with increasing Sm-doping
on ρxx . Upon doping Sm, ρxx systematically increases both concentration y, negative MRR is enhanced, the magnitude
at 3 and 300 K; especially for SmRuO3 , ρxx (3 K) diverges of which reaches as much as −25% for y = 0.5 at 0.5 K.
and is one order of magnitude higher than Nd1−x Pr x RuO3 Considering that all the Ln ions employed in this work are
[Fig. 4(a)]. The lattice distortion of perovskite AMO3 is gov- magnetic with finite J, the observed positive MRR in PrRuO3
erned√by the tolerance factor f , which is defined as f = (rA + is unusual since negative MRR is expected due to the suppres-
rO )/ 2(rM + rO ). Here ri (i = A, M, or O) represents the ionic sion of magnetic scattering by the applied field. Indeed, in the
size of each element [24]. When f is close to 1, the cubic per- case of LnNiO3 thin films, negative MRR has been commonly
ovskite structure is realized. As rA or equivalently f decreases, observed for Ln = Sm [27], Nd [28], and Pr [29] at lower
the lattice structure distorts to an orthorhombic structure, to temperatures, where they are in antiferromagnetic insulating
which LnRuO3 belongs [10]. In the case of the orthorhombic phases. This positive MRR in PrRuO3 may be related to the
lattice, the M-O-M bond angle varies continuously with f , emergence of a topological semimetallic state, similar to that

074407-4
TUNING MAGNETOTRANSPORT PROPERTIES THROUGH … PHYSICAL REVIEW MATERIALS 8, 074407 (2024)

experimental and theoretical investigations are required to

validate this analogy.
The anomalous Hall angle is ∼0.3% for NdRuO3 and
decreases for both Nd1−x Pr x RuO3 and Nd1−y Smy RuO3 with
similar doping concentration dependences [Fig. 4(c)]. At the
same time, the feature of THE diminishes as shown in Fig. 3,
as well as the kink structure in MRR (Fig. S6 [21]). Due to the
small volume and low ordering temperature, it is currently not
feasible to perform magnetization measurements for the films
to gain further insights and compare them with the observed
transition in AHE and THE. One possible reason may be the
difference in ordering temperatures of Ln3+ spins as THE in
this system appears to be related to the spin texture of the
Ln site. Indeed, the ordering temperature of Ln3+ spins has
been reported to remarkably vary depending on Ln in other
LnMO3 systems such as M = Cr, Mn, Fe [31]. However,
additional studies are desired to unambiguously clarify its
origin, which is indeed challenging and we wish to see as
future work.

IV. CONCLUSION
In summary, we have clarified the magnetotransport prop-
erties of Nd1−x Pr x RuO3 and Nd1−y Smy RuO3 thin films. At
both ends of the solid solutions, PrRuO3 and SmRuO3 respec-
tively show metallic and insulating temperature dependences
of ρxx , implying the transition of the electronic ground state.
AHE is consistently observed across the entire range of dop-
ing concentrations, while the Hall angle decays and THE
disappears on both sides of the doping from NdRuO3 . Our
work on systematic chemical substitution on the novel mate-
rials platform of perovskite oxide with Ru3+ not only enables
the systematic tuning of transport properties but also paves the
way for investigating electronic phase transition and poten-
FIG. 4. (a) Longitudinal resistivity (ρxx ) at 3 K and 300 K under tially uncovering more exotic transport phenomena associated
B = 0 T, (b) MRR at 3 K and 0.5 K under B = 9 T, and (c) anomalous with magnetic topological materials.
Hall angle at 3 K and 0.5 K under B = 9 T, as a function of the
nominal doping concentrations x and y. On the top, total angular ACKNOWLEDGMENTS
momentum (J) and ionic radii of each Ln3+ are also shown. Note
that data at 0.5 and 3 K are overlapped for x = 1. This work was supported by JSPS Grants-in-Aid for Sci-
entific Research (S) No. JP22H04958, JSPS Grant-in-Aid
for Early-Career Scientists No. JP20K15168, the Mitsubishi
observed in SrIrO3 with large positive MRR [30], which orig- Foundation, the Izumi Science and Technology Foundation,
inates from the d 5 electronic configuration of Ir 4+ . Although the Tokuyama Science Foundation, and the Toyota Physical
PrRuO3 also possesses d 5 electronic configuration, further and Chemical Research Institute.

[1] W. Witczak-Krempa, G. Chen, Y. B. Kim, and L. Balents, Annu. [6] H.-S. Kim, V. Vijay Shankar, A. Catuneanu, and H.-Y. Kee,
Rev. Condens. Matter Phys. 5, 57 (2014). Phys. Rev. B 91, 241110(R) (2015).
[2] M. A. Zeb and H.-Y. Kee, Phys. Rev. B 86, 085149 (2012). [7] Y. Kasahara, T. Ohnishi, Y. Mizukami, O. Tanaka, S. Ma, K.
[3] Y. F. Nie, P. D. C. King, C. H. Kim, M. Uchida, H. I. Wei, B. D. Sugii, N. Kurita, H. Tanaka, J. Nasu, Y. Motome, T. Shibauchi,
Faeth, J. P. Ruf, J. P. C. Ruff, L. Xie, X. Pan, C. J. Fennie, D. G. and Y. Matsuda, Nature (London) 559, 227 (2018).
Schlom, and K. M. Shen, Phys. Rev. Lett. 114, 016401 (2015). [8] R. J. Bouchard and J. F. Weiher, J. Solid State Chem. 4, 80
[4] J. Fujioka, R. Yamada, M. Kawamura, S. Sakai, M. Hirayama, (1972).
R. Arita, T. Okawa, D. Hashizume, M. Hoshino, and Y. Tokura, [9] T. Sugiyama and N. Tsuda, J. Phys. Soc. Jpn. 68, 3980
Nat. Commun. 10, 362 (2019). (1999).
[5] B. J. Kim, H. Jin, S. J. Moon, J.-Y. Kim, B.-G. Park, C. S. Leem, [10] A. Sinclair, J. A. Rodgers, C. V. Topping, M. Míšek, R. D.
J. Yu, T. W. Noh, C. Kim, S.-J. Oh, J.-H. Park, V. Durairaj, G. Stewart, W. Kockelmann, J.-W. G. Bos, and J. P. Attfield,
Cao, and E. Rotenberg, Phys. Rev. Lett. 101, 076402 (2008). Angew. Chem., Int. Ed. 53, 8343 (2014).

074407-5
FUJITA, SENZAKI, ZHANG, AND KAWASAKI PHYSICAL REVIEW MATERIALS 8, 074407 (2024)

[11] K. Ji, A. Paul, E. Solana-Madruga, A. M. Arevalo-Lopez, U. V. for the Sm-doped samples; the magnetic field dependence of
Waghmare, and J. P. Attfield, Phys. Rev. Mater. 4, 091402(R) magnetoresistance and Hall resistivity; high field measurements
(2020). data for NdRuO3 ; the analysis for the correlation between mag-
[12] B. K. Patel, M. T. K. Kolambage, C. D. McMillen, and J. W. netoresistance and the topological Hall effect.
Kolis, J. Cryst. Growth 602, 126979 (2023). [22] N. F. Mott, J. Non-Cryst. Solids 1, 1 (1968).
[13] G. Koster, L. Klein, W. Siemons, G. Rijnders, J. S. Dodge, C.-B. [23] B. I. Shklovskii and A. L. Efros, in Variable-Range Hopping
Eom, D. H. A. Blank, and M. R. Beasley, Rev. Mod. Phys. 84, Conduction, edited by B. I. Shklovskii and A. L. Efros, Springer
253 (2012). Series in Solid-State Sciences (Springer, Berlin, 1984), pp.
[14] J.-G. Cheng, J.-S. Zhou, and J. B. Goodenough, Proc. Natl. 202–227.
Acad. Sci. USA 110, 13312 (2013). [24] J. B. Torrance, P. Lacorre, A. I. Nazzal, E. J. Ansaldo, and C.
[15] L. Zhang, T. C. Fujita, Y. Masutake, M. Kawamura, T.-h. Arima, Niedermayer, Phys. Rev. B 45, 8209 (1992).
H. Kumigashira, M. Tokunaga, and M. Kawasaki, Commun. [25] M. Marezio, J. P. Remeika, and P. D. Dernier, Acta Crystallogr.,
Mater. 5, 35 (2024). Sect. B 26, 2008 (1970).
[16] K. Ueda, J. Fujioka, C. Terakura, and Y. Tokura, Phys. Rev. B [26] D. A. MacLean, H.-N. Ng, and J. E. Greedan, J. Solid State
92, 121110(R) (2015). Chem. 30, 35 (1979).
[17] Z. Tian, Y. Kohama, T. Tomita, H. Ishizuka, T. H. Hsieh, J. J. [27] K. Ramadoss, N. Mandal, X. Dai, Z. Wan, Y. Zhou, L.
Ishikawa, K. Kindo, L. Balents, and S. Nakatsuji, Nat. Phys. 12, Rokhinson, Y. P. Chen, J. Hu, and S. Ramanathan, Phys. Rev. B
134 (2016). 94, 235124 (2016).
[18] B. Cheng, T. Ohtsuki, D. Chaudhuri, S. Nakatsuji, M. Lippmaa, [28] A. Stupakov, O. Pacherova, T. Kocourek, M. Jelinek, A.
and N. P. Armitage, Nat. Commun. 8, 2097 (2017). Dejneka, and M. Tyunina, Phys. Rev. B 99, 085111 (2019).
[19] M. Imada, A. Fujimori, and Y. Tokura, Rev. Mod. Phys. 70, [29] D. D. K., A. Singh, S. Sathapathy, K. K. Maurya, P. K. Siwach,
1039 (1998). V. K. Malik, S. S. Kushvaha, and H. K. Singh, J. Supercond.
[20] M. Kawasaki, K. Takahashi, T. Maeda, R. Tsuchiya, M. Novel Magn. 36, 623 (2023).
Shinohara, O. Ishiyama, T. Yonezawa, M. Yoshimoto, and H. [30] J. Fujioka, T. Okawa, A. Yamamoto, and Y. Tokura, Phys. Rev.
Koinuma, Science 266, 1540 (1994). B 95, 121102(R) (2017).
[21] See Supplemental Material at http://link.aps.org/supplemental/ [31] E. Bousquet and A. Cano, J. Phys.: Condens. Matter 28, 123001
10.1103/PhysRevMaterials.8.074407 for XRD reciplocal space (2016).
mappings of the samples; the variable range hopping analysis

074407-6