Structural Phase Transitions and Superconductivity Induced in Antiperovskite Phosphide CaPd3P

Article abstract of DOI:10.1021/acs.inorgchem.0c01482

In this study, we succeeded in synthesizing new antiperovskite phosphides MPd3P (M = Ca, Sr, Ba) and discovered the appearance of a superconducting phase (0.17 ≤ x ≤ 0.55) in a solid solution (Ca1-xSrx)Pd3P. Three perovskite-related crystal structures were identified in (Ca1-xSrx)Pd3P, and a phase diagram was built on the basis of experimental results. The first phase transition from centrosymmetric (Pnma) to noncentrosymmetric orthorhombic (Aba2) occurred in CaPd3P near room temperature. The phase transition temperature decreased as Ca2+ was replaced with a larger-sized isovalent Sr2+. Bulk superconductivity at a critical temperature (Tc) of approximately 3.5 K was observed in a range of x = 0.17-0.55; this was associated with the centrosymmetric orthorhombic phase. Thereafter, a noncentrosymmetric tetragonal phase (I41md) remained stable for 0.6 ≤ x ≤ 1.0, and superconductivity was significantly suppressed as samples with x = 0.75 and 1.0 showed Tc values as low as 0.32 K and 57 mK, respectively. For further substitution with a larger-sized isovalent Ba2+, namely, (Sr1-yBay)Pd3P, the tetragonal phase continued throughout the composition range. BaPd3P no longer showed superconductivity down to 20 mK. Since the inversion symmetry of structure and superconductivity can be precisely controlled in (Ca1-xSrx)Pd3P, this material may offer a unique opportunity to study the relationship between inversion symmetry and superconductivity.

Full text of DOI:10.1021/acs.inorgchem.0c01482

pubs.acs.org/IC
Article
Structural Phase Transitions and Superconductivity Induced in
Antiperovskite Phosphide CaPd3P
Akira Iyo,* Hiroshi Fujihisa, Yoshito Gotoh, Shigeyuki Ishida, Hiroki Ninomiya, Yoshiyuki Yoshida,
Hiroshi Eisaki, Hishiro T. Hirose, Taichi Terashima, and Kenji Kawashima
Cite This: https://dx.doi.org/10.1021/acs.inorgchem.0c01482
Read Online
Metrics & More
Article Recommendations
ACCESS
ABSTRACT: In this study, we succeeded in synthesizing new antiperovskite phosphides
MPd3P (M = Ca, Sr, Ba) and discovered the appearance of a superconducting phase (0.17 ≤ x
≤ 0.55) in a solid solution (Ca1−xSrx)Pd3P. Three perovskite-related crystal structures were
identified in (Ca1−xSrx)Pd3P, and a phase diagram was built on the basis of experimental
results. The first phase transition from centrosymmetric (Pnma) to noncentrosymmetric
orthorhombic (Aba2) occurred in CaPd3P near room temperature. The phase transition
temperature decreased as Ca2+ was replaced with a larger-sized isovalent Sr2+. Bulk
superconductivity at a critical temperature (Tc) of approximately 3.5 K was observed in a
range of x = 0.17−0.55; this was associated with the centrosymmetric orthorhombic phase.
Thereafter, a noncentrosymmetric tetragonal phase (I41md) remained stable for 0.6 ≤ x ≤ 1.0,
and superconductivity was significantly suppressed as samples with x = 0.75 and 1.0 showed Tc
values as low as 0.32 K and 57 mK, respectively. For further substitution with a larger-sized
isovalent Ba2+, namely, (Sr1−yBay)Pd3P, the tetragonal phase continued throughout the
composition range. BaPd3P no longer showed superconductivity down to 20 mK. Since the inversion symmetry of structure and
superconductivity can be precisely controlled in (Ca1−xSrx)Pd3P, this material may offer a unique opportunity to study the
relationship between inversion symmetry and superconductivity.
Synthesis of phosphides is relatively difficult due to the high
vapor pressure and high reactivity of P at high temperatures.
We speculated that this may have hindered the exploration of
new compounds for antiperovskite phosphides. Our intensive
search for new antiperovskite phosphides resulted in the
successful synthesis of MPd3P (M = Ca, Sr, Ba). SrPd3P and
BaPd3P exhibited no superconductivity above 2 K. On the
other hand, CaPd3P exhibited slight superconductivity at ∼4 K.
Furthermore, CaPd3P had a different crystal structure than that
of the other two compounds, which motivated us to synthesize
a solid solution between CaPd3P and SrPd3P. As a result, we
detected structural phase transitions and bulk superconductiv-
ity in the solid solution (Ca1−xSrx)Pd3P. Herein, we present the
phase diagram of (Ca1−xSrx)Pd3P, which was constructed
based on experimental results, such as those of X-ray
diffraction (XRD) analysis and resistivity and magnetization
measurements, as functions of composition and temperature.
INTRODUCTION
■
Superconductivity has significant research interest due to its
potential practical applications, in addition to the profound
physics behind superconductors. The discovery of new
superconductors can considerably affect material science as
well as the research on superconductivity. Most recently, we
synthesized a new compound Mg2Rh3P and found that
superconductivity can be induced at 3.9 K by introducing a
small amount of Mg deficiency.1,2 Mg2Rh3P can be regarded as
an antiperovskite structure due to its crystallographic
characteristic. That is, there is a three-dimensional network
of corner-sharing Rh6P octahedrons, and two Mg2+ ions
occupy the space surrounded by the octahedrons. The
antiperovskite-type superconductors have attracted much
attention due to their unique features, e.g., a heavy-Fermion
superconductor with broken space inversion symmetry CePt3Si
(Tc = 0.75 K),3 a possible topological superconductor
Sr3−δSnO (Tc = ∼5 K),4 spin-triplet superconductor Li2Pt3B
(Tc = ∼2.8 K),5,6 Ni-based compounds MgNi3C (Tc = 8 K),7
ZnNi3N (Tc = ∼3 K),8 and CdNi3C (Tc = ∼3 K).8,9 Regarding
antiperovskite-related phosphide, APt3P (Tc = 8.4, 6.6, and 1.5
K for A = Sr, Ca, and La, respectively),10 SrPt6P2 (Tc = 0.6
K),11 and SrPt10P4 (Tc = 1.4 K)12 are known superconductors
other than Mg2Rh3P.
Received: May 20, 2020
https://dx.doi.org/10.1021/acs.inorgchem.0c01482
© XXXX American Chemical Society
Inorg. Chem. XXXX, XXX, XXX−XXX
A

Inorganic Chemistry
pubs.acs.org/IC
Article
We first show X-ray diffraction (XRD) patterns and lattice
parameters of the samples as functions of composition and
temperature. Next, we indicate three types of crystal structures
that appeared in (Ca1−xSrx)Pd3P. We also demonstrate how
the structural phase transition temperature changed and how
superconductivity developed in (Ca1−xSrx)Pd3P, indicating the
composition dependence of resistivity and magnetization data.
Finally, we present the phase diagram of (Ca1−xSrx)Pd3P
constructed based on the experimental results.
EXPERIMENTAL SECTION
■
Material Synthesis. Polycrystalline samples were synthesized by
solid-state reaction. A Pd powder (Kojundo Chemical, 99.9%) and
precursors with nominal compositions of CaP, SrP, and BaP were
used as starting materials. The precursors were obtained by reacting
Ca (Furuuchi Chemical, 99.5%), Sr (Furuuchi Chemical, 99.9%), or
Ba (Rare Metallic, 99%) with P (Furuuchi Chemical, 99.999%) at 700
°C for 24 h in evacuated quarts tubes. The Ca, Sr, or Ba was
processed into flaky shapes prior to the synthesis of precursors to
accelerate the reaction. A sample with a nominal composition of
(Ca1−xSrx)Pd3P or (Sr1−yBay)Pd3P was ground using a mortar. The
ground powder was pressed into a pellet (∼0.15 g), which was then
enclosed in an evacuated quartz tube (inner diameter of 8 mm, length
of ∼70 mm). The sample was heated at 970 °C for 12 h followed by
furnace cooling. Raw materials and precursors were processed in an
N2-filled glovebox to prevent oxidation. Samples of (Ca1−xSrx)Pd3P
and (Sr1−yBay)Pd3P remained stable in air for at least several months.
Measurements. Powder XRD patterns were obtained at room
temperature (RT) (∼293 K) using a diffractometer (Rigaku, Ultima
IV) with Cu−Kα radiation. Powder XRD patterns of CaPd3P were
also measured at 150 K using a laboratory-made cooling system. The
obtained diffraction patterns were indexed and the space groups were
searched using BIOVIA Materials Studio (MS) X-Cell software.13
TOPAS ver. 5 software was used to search for the initial atomic
coordinates.14 Pawley and Rietveld analyses were performed using
BIOVIA MS Reflex ver. 2020 software.15 We removed the constraints
of the space groups from the candidate models and obtained stable
atomic coordinates using DFT calculations with MS Castep
software,16 while confirming that the original space groups had
been preserved.
Figure 1. Powder XRD patterns at RT for representative
compositions of (Ca1−xSrx)Pd3P and (Sr1−yBay)Pd3P. Major diffrac-
tion peaks from the NCS- and CS-orthorhombic and NCS-tetragonal
phases are indicated by closed circles, triangles, and squares,
respectively. The contribution of Cu−Kα2 to the diffraction data
was eliminated through software processing.
Figure 2 shows the composition dependence of reduced lattice
parameters and reduced cell volume (Vr) for the CS-
Magnetization (M) measurements were performed under a
magnetic field (H) of 10 Oe, using a magnetic-property measurement
system (Quantum Design, MPMS-XL7). Measurements were
performed with zero-field-cooling (ZFC) and field-cooling (FC)
modes. Electrical resistivity in the temperature range of 2−320 K were
measured by a four-probe method, using a physical property
measurement system (Quantum Design, PPMS). Electrical resistiv-
ities of (Sr0.75Ca0.25)Pd3P, SrPd3P, and BaPd3P were measured down
to 20 mK using a dilution refrigerator.
RESULTS AND DISCUSSION
■
Powder XRD Patterns at RT. Three types of crystal
structures appeared depending on the composition and
temperature in (Ca1−xSrx)Pd3P and (Sr1−yBay)Pd3P. Figure 1
shows the powder XRD patterns at RT for representative
compositions (x = 0, 0.05, 0.25, 0.55, 0.6, 0.75, and 1.0, y = 0.5
and 1.0). CaPd3P (x = 0) has a more complex diffraction
pattern than that of others due to mixing of centrosymmetric
(CS) and noncentrosymmetric (NCS) orthorhombic phases3.
Only a small amount of the Sr substitution (x = 0.05) made
the CS-orthorhombic phase dominant. The CS-orthorhombic
phase was stable up to x = 0.55. Thereafter, the XRD pattern
clearly changed for x = 0.6 due to a structural phase transition
from CS-orthorhombic to NCS-tetragonal. The NCS-tetrago-
nal phase was stable for a composition range 0.6 ≤ x ≤ 1 and
continued for a whole composition range of (Sr1−yBay)Pd3P.
Figure 2. Composition x dependence of reduced lattice parameters
and reduced cell volumes for CS-orthorhombic (0 ≤ x < 0.6) and
NCS-tetragonal (0.6 ≤ x ≤ 1.0) phases in (Ca1−xSrx)Pd3P.
orthorhombic (0 ≤ x < 0.6) and NCS-tetragonal (0.6 ≤ x ≤
1.0) phases in (Ca1−xSrx)Pd3P. The lattice parameters and Vr
increased almost linearly according to Vegard’s law in each
phase as the substitution of the larger size isovalent Sr2+ for
Ca2+ progressed, which ensured that systematic samples were
successfully prepared. Lattice parameters and the Vr of
(Sr1−yBay)Pd3P (not shown) also increased as a proportion
B
https://dx.doi.org/10.1021/acs.inorgchem.0c01482
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
of larger Ba2+ ions (y) increased. The Vr increased by nearly
10.4% from CaPd3P (Vr = 88.36 Å3) to BaPd3P (Vr = 97.51
Å3). The Pd6P octahedron network appears to be flexible with
respect to the size of the cations acceptable to form crystals.
Therefore, other cations such as alkali metals and lanthanides
may also be accepted into the Pd6P network.
Powder XRD Pattern of CaPd3P at Low Temperature.
Figure 3 shows the powder XRD patterns of CaPd3P at 150 K
Figure 4. Rietveld fittings for (a) NCS-orthorhombic CaPd3P (150
K), (b) CS-orthorhombic (Ca0.75Sr0.25)Pd3P, and (c) NCS-tetragonal
SrPd3P. Iobs and Ical indicate the observed and calculated diffraction
intensities, respectively. The peaks from Cu−Kα2 line were not
eliminated for the Rietveld analyses.
Table 1. Results of Rietveld Structure Refinement for
a
CaPd3P at 150 K
atoms Wyckoff position
x
y
z
Figure 3. Powder XRD patterns of CaPd3P at (a) RT and (b) 150 K.
The contribution of Cu−Kα2 to the diffraction data was eliminated
through software processing.
Ca1
Ca2
Pd1
Pd2
Pd3
P
4a
4a
8b
8b
8b
8b
0
0
0
0
0.2496
0.7800(20)
0.0299(16)
0.2688(18)
−0.0027(17)
0.0319(22)
0.2742(6)
0.2026(5)
0.0328(5)
0.2903(13)
0.0378(5)
0.2461(7)
0.2743(6)
0.2795(13)
with those at RT. The diffraction pattern at RT could be
analyzed assuming the existence of the two orthorhombic
phases. The diffraction peaks from the CS-orthorhombic phase
almost disappeared at 150 K. Note that 150 K is sufficiently
lower than the temperatures (269 and 283 K) at which CaPd3P
exhibited hysteretic behavior in resistivity due to the structural
phase transition. Note that a phase transition from a CS to an
NCS structure with decreasing temperature (∼200 K) is also
observed in the pyrochlore superconductor Cd2Re2O7 (Tc = 1
K).17 In Cd2Re2O7, an additional structural phase transition
occurs at ∼120 K. The structural changes are attributed to the
distortion of Re tetrahedra.
Crystal Structure Refinements. Crystal structures were
refined for the representative samples of the NCS-ortho-
rhombic (150 K) CaPd3P, CS-orthorhombic (Ca0.75Sr0.25)-
Pd3P, and NCS-tetragonal SrPd3P (BaPd3P). As shown in
Figure 3 (a), CaPd3P contained both NCS- and CS-
orthorhombic phases, which complicated the structure refine-
ment (only the lattice parameters could be determined).
Therefore, the structure of CaPd3P was refined at a low
temperature (150 K). Rietveld fittings are shown for the three
samples in Figure 4, and the refined structure parameters are
summarized in Tables 1−3. For CaPd3P at 150 K, we used the
Pawley method to find that its crystal system is orthorhombic
with a base-centered symmetry. The diffraction intensities
could be fitted to the CS-orthorhombic space group Cmce (a =
8.75 Å, b = 8.99 Å, c = 8.93 Å) and the NCS-orthorhombic
Aba2 (a = 8.93 Å, b = 8.99 Å, c = 8.75 Å). Rwp of the Rietveld
refinement for the former model was 2.2% higher than that for
the latter. An enthalpy comparison based on DFT calculations
showed that the value for the former was 0.31 eV higher than
that for the latter. For these reasons, we concluded that the
a
Aba2 (orthorhombic no. 41), a = 8.9224(7) Å, b = 8.9841(7) Å, c =
8.7470(16) Å, V = 701.2(1) Å3 (Z = 8), Rwp = 24.00%, Re = 11.74%, S
= 2.04, Uiso = 0.026 (1) Å2 for all atoms. The occupancy for each
atom was fixed at 1.
Table 2. Results of Rietveld Structure Refinement for
(Ca0.75Sr0.25)Pd3P at RT
a
atoms
Wyckoff position
x
y
z
CaSr
Pd1
Pd2
P
4c
4c
8d
4c
−0.0016(7)
0.4568(3)
0.2268(2)
0.1935(9)
1/4
1/4
0.0145(5)
1/4
0.7832(7)
0.2040(4)
0.4554(2)
0.1693(11)
a
Pnma (orthorhombic no. 62), a = 8.9944(3) Å, b = 6.1545(2) Å, c =
6.4989(2) Å, V = 359.7(1) Å3 (Z = 4), Rwp = 12.37%, Re = 7.76%, S =
1.59, Uiso = 0.033(1) Å2 for all atoms. The occupancy for each atom
was fixed at 1.
NCS-orthorhombic Aba2 was the correct space group for
CaPd3P at 150 K. Note that Rwp for CaPd3P at 150 K (24%)
was worse than those for the other samples because of the use
of a sample cover to protect the CaPd3P from frost. The cover
decreased the peak intensity below diffraction angles of 30°.
However, we believe that the atomic coordinates are reliable
because the fitting between 30° and 140° in the Rietveld
analysis is as satisfactory as in the other analyses. In the analysis
of (Ca0.75Sr0.25)Pd3P, Rwp decreased by only 0.4% when using
individual isotropic atomic displacement parameter Uiso per
atomic site. This resulted in Uiso values of 0.009, 0.012, and
C
https://dx.doi.org/10.1021/acs.inorgchem.0c01482
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
a
Table 3. Results of Rietveld Structure Refinements for SrPd3P and BaPd3P at RT
atoms
Wyckoff position
x
y
z
Sr1 [Ba1]
Sr2 [Ba2]
Sr3 [Ba3]
Sr4 [Ba4]
Pd1
Pd2
Pd3
Pd4
Pd5
4a
4a
4a
4a
16c
8b
8b
8b
8b
16c
0
0
1/2
1/2
0
1/2
0
0.0118 [0.0089]
−0.0102(10) [-0.0215(6)]
−0.0255(5) [-0.0272(4)]
−0.0044(10) [-0.0047(5)]
0.0254(6) [0.0225(4)]
0.1147(7) [0.1101(6)]
0.1191(7) [0.1177(5)]
0.3689(7) [0.3625(6)]
0.3727(6) [0.3657(6)]
0.1560(7) [0.1550(4)]
1/2
0.2474(5) [0.2462(4)]
1/2
0
0
0.2489(3) [0.2484(7)]
0.2722(5) [0.2374(7)]
0.2871(5) [0.3264(5)]
0.1772(5) [0.2218(6)]
0.2536(6) [0.2280(6)]
0.2511(10) [0.2585(12)]
1/2
P
0.2453(10) [0.2478(13)]
a
Data of BaPd3P are written in brackets. I41md (tetragonal no. 109): For SrPd3P, a = 8.8064(2) Å, c = 19.2228(6) Å, V = 1491(1) Å3 (Z = 16),
and Rwp = 8.51%, Re = 6.14%, S = 1.39. For BaPd3P, a = 8.9298(2) Å, c = 19.5646(7) Å, V = 1560(1) Å3 (Z = 16), and Rwp = 7.92%, Re = 5.32%, S
= 1.49. Uiso of all atoms were fixed to be 0.033(1) Å2 and 0.04(1) Å2 for SrPd3P and BaPd3P, respectively. The occupancy for each atom was fixed
at 1.
0.046 for cations, P, and Pd, respectively. The cation Uiso was
considered to be too small, while that of the Pd was too large,
for 150 K. Therefore, we fixed Uiso for all elements. Crystal
structures are schematically drawn for CaPd3P, (Ca0.75Sr0.25)-
Pd3P, and SrPd3P in Figure 5. They commonly have three-
dimensional networks of corner sharing Rh6P octahedrons.
However, as is often seen in perovskite structures, their space
groups changed due to the deformation and rotation of the
Rh6P octahedrons.
CaPd3P had the lowest symmetric structure with a space
group of Aba2 among them. Compared to a simple perovskite
structure, periodicity is doubled along each three-axis direction
(formula unit Z = 8). (Ca0.75Sr0.25)Pd3P (Pnma) can be
considered to be related to the BaAu3Ge(SrPt3P)-type
structure (P4/nmm). That is, P atoms adjacent in the b- and
c-axes directions are alternately displaced in the opposite
direction in the bc plane. Both Pnma and P4/nmm belong to a
CS space group. The distortion of the Pd6P octahedrons causes
√2 and 2 times periodicity (Z = 4) along the b(c)-axis and the
a-axis directions, respectively.
On the other hand, SrPd3P and BaPd3P (I41md) can be
understood based on the NCS CePt3B(CePt3Si)-type structure
(P4mm). That is, P atoms are displaced in the same direction
along the c-axis. For this reason, the crystal structure of SrPd3P
and BaPd3P lacks an inversion symmetry. Periodicity is
doubled in the a- and b-axis directions and quadrupled in
the c-axis direction (Z = 16).
Superconductivity in (Ca1−xSrx)Pd3P (0 ≤ x ≤ 0.20).
Resistivity and magnetization data are displayed separately in
the two composition regions of 0 ≤ x ≤ 0.2 and 0.3 ≤ x ≤ 1.0.
Figure 6 shows the temperature (T) dependence of resistivity
(ρ) and 4πM/H for representative x (= 0, 0.1, 0.15, 0.17, and
0.2) in 0 ≤ x ≤ 0.2. Resistivity anomalies with hysteretic
behaviors were observed in the resistivity data for x = 0, 0.1,
0.15, and 0.17. The temperatures at which the resistivity
changed abruptly in the cooling and warming processes were
defined as Tsc and Tsw, respectively, as indicated for x = 0. The
hysteresis loop closed above 320 K for CaPd3P, which is the
reason for the presence of the two orthorhombic phases at RT.
The resistivity anomaly with the characteristic of a first-order
transition can be attributed to the structural phase transition
detected by the XRD measurements. Tsc and Tsw were
determined to be 269 and 283 K, respectively. Then, Tsc and
Tsw decreased with increasing x, and the anomaly and
hysteresis disappeared for x = 0.2.
Below the structural transition temperature, namely, in the
NCS-orthorhombic phase, the resistivity decreased more
steeply with temperature. Consequently, the normal state
resistivity of CaPd3P decreased to 3.16 μΩ·cm (at 5 K), which
was relatively small as a polycrystalline sample. A residual
resistivity ratio (RRR) defined by ρ(300 K)/ρ(5 K) was as
large as ∼80.2 for CaPd3P. In contrast, the resistivity was less
sensitive to temperature in the CS-orthorhombic phase, e.g.,
RRR was only 1.6 for x = 0.2.
Figure 5. Schematic refined crystal structures of (a) NCS-
orthorhombic CaPd3P (150 K), (b) CS-orthorhombic (Ca0.75Sr0.25)-
Pd3P, and (c) NCS-tetragonal SrPd3P produced using the VESTA
software.7 Solid lines indicate the unit cells. Pd6P octahedrons are
shaded.
D
https://dx.doi.org/10.1021/acs.inorgchem.0c01482
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
Figure 6. Temperature dependence of resistivity (left panel) and
4πM/H (right panel) for (Ca1−xSrx)Pd3P in the composition range 0
≤ x ≤ 0.20. The resistivity measurements with decreasing and
increasing temperature are indicated by blue and red curves,
respectively. Insets show increments in resistivity near super-
conducting transitions.
Figure 7. Temperature dependence of resistivity (left panel) and
4πM/H (right panel) for (Ca1−xSrx)Pd3P in the composition range
0.3 ≤ x ≤ 1.0 and BaPd3P. Resistivities of (Sr0.75Ca0.25)Pd3P, SrPd3P,
and BaPd3P were measured down to 20 mK. Insets show resistivity in
low-temperature regions.
The magnetization data clearly exhibited the evolution of
superconductivity in (Ca1−xSrx)Pd3P. Superconducting tran-
sitions were more or less observed for all the samples in Figure
5. However, the transitions were broad, and the super-
conducting shielding (ZFC) volume fractions at 2 K (Vs) were
as small as ∼7% and 35% for x = 0 and 0.1, respectively.
Transitions were still broad for x = 0.15, though Vs increased
to ∼100%. We believe that only the CS-orthorhombic phase is
responsible for the superconductivity at ∼3.5 K. The broad
superconducting transitions at 0 ≤ x ≤ 0.15 is probably caused
by the CS-orthorhombic phase partially remaining in the
samples as a result of the imperfect structural phase transition.
Sharp transitions in both resistivity and magnetization began to
appear at x = 0.17. Tc determined by resistivity and
magnetization for x = 0.17 as indicated in Figure 5 were
3.51 and 3.32 K, respectively.
resistivity; therefore, structural phase transitions appeared to
be absent for the samples in Figure 7. Abrupt superconducting
transitions were observed at ∼3.5 K for samples (from x =
0.17) up to x = 0.55. Tc determined by magnetization data
were 3.32, 3.32, 3.33, 3.37, 3.44, and 3.47 K for x = 0.17, 0.2,
0.3, 0.4, 0.5, and 0.55, respectively. Thus, Tc slightly enhanced
as x increased.
In the NCS-tetragonal phase samples (0.6 ≤ x ≤ 1.0, 0 ≤ y
≤ 1.0), superconductivity was suppressed drastically. The
sample with x = 0.6 exhibited only a trace of superconductivity
at 3.5 K, most probably due to unavoidable compositional
fluctuations in solid solution polycrystalline samples. Tc
decreased significantly down to 0.32 and 57 mK for x = 0.75
and 1.0, respectively. Moreover, no superconductivity was
observed above 20 mK in BaPd3P. Thus, the NCS-tetragonal
as well as the NCS-orthorhombic structures were unfavorable
for superconductivity in this material. Note that RRRs of the
Superconductivity in (Ca1−xSrx)Pd3P (0.3 ≤ x ≤ 1. 0)
and BaPd3P. Figure 7 shows the temperature dependence of
resistivity and 4πM/H for representative x (= 0.3, 0.55, 0.6,
0.8, and 1.0) in 0.3 ≤ x ≤ 1.0 and BaPd3P. Resistivity data
were measured down to 20 mK for (Sr0.75Ca0.25)Pd3P, SrPd3P,
and BaPd3P. No anomalies were detected in the normal state
E
https://dx.doi.org/10.1021/acs.inorgchem.0c01482
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
endmembers SrPd3P and BaPd3P are large (10.6 and 11.9,
respectively). Therefore, the disorder of lattice caused by the
mixing of different cations may have resulted in the small RRR
values (∼2) of the solid solutions.
Phase Diagram. A schematic phase diagram of (Ca1−xSrx)-
Pd3P drawn based on the experimental results is shown in
Figure 8. There were two critical compositions regarding the
a structural phase transition, and most of the sample retains the
CS-orthorhombic phase. Tc monotonically and slightly
increased with the composition x and reached a maximum
value of 3.65 K (determined by resistivity data for x = 0.55)
just before the phase boundary. Then, superconductivity was
significantly suppressed in the NCS-tetragonal phase.
(Ca0.25Sr0.75)Pd3P and SrPd3P showed zero resistivity at as
low as 0.32 K and 57 mK, respectively. The lack of inversion
symmetry appears to be fatal to superconductivity in
(Ca1−xSrx)Pd3P. This material may offer a unique opportunity
to study the effect of inversion symmetry on superconductivity
which is currently attracting considerable attention in the field
of superconductivity.20,21
It is currently unclear why superconductivity varies
significantly depending on the crystal structure and how
inversion symmetry is related to superconductivity in this
material. Detailed physical property measurements and
theoretical work such as band structure calculations are
necessary to obtain a better understanding regarding this.
Finally, we stress that there is room to explore new
antiperovskite Pd phosphides in combination with other
cations such as alkali metals and rare earth elements. In
addition, the previously reported Zr(P,Se)2 (PbClF-type)
superconductor was also observed in the solid solution
between ZrP2 (PbCl2-type) and ZrSe2 (CdI2-type).22 There-
fore, searching for superconductivity in such a hidden phase
can also lead to the discovery of new materials/super-
conductors.
Figure 8. Schematic phase diagram of (Ca1−xSrx)Pd3P drawn based
on the powder XRD, resistivity, and magnetization measurements. Tc
in 0 ≤ x ≤ 0.6 were plotted for samples with −4πM/H > 0.5 (ZFC) at
2 K.
structure and superconductivity. The endmember CaPd3P had
the structural phase transition from CS- to NCS-orthorhombic
system near RT. As x increases, the transition temperature Ts
(= Tsc/2 + Tsw/2) decreases toward x = 0.17−0.2. The CS-
orthorhombic phase was stable from x = 0.17−0.2 to x = 0.55.
The structure then changed to the NCS-tetragonal system for
further substitution (0.6 ≤ x < 1.0 and 0 ≤ y < 1.0). Thus, the
crystal structure (inversion symmetry) was precisely controlled
by the substitution using electronically similar elements.
Subedi et al. calculated the difference in total energies
between the antipolar (CS) and polar (NCS) structures of
M′Pt3P and M′Pt3Si (M′ = Sr, Ca, La).18 They found that the
energy differences in phosphides, especially CaPt3P, are small,
and indicated that the structure can be tuned from CS to NCS
through appropriate synthesis conditions or partial replace-
ment of elements. They showed that such structural control
opens new perspectives toward the understanding of
unconventional superconductivity. We realized their proposal
in the CaPd3P system.
CONCLUSIONS
■
In our attempt to discover new antiperovskite phosphide
materials, we succeeded in synthesizing new MPd3P com-
pounds and discovered a superconducting phase in the solid
solutions between CaPd3P and SrPd3P. Three perovskite-
related structures were identified in (Ca1−xSrx)Pd3P and
(Sr1−yBay)Pd3P, and superconductivity was found to depend
strongly on the structures, possibly because of the presence of
inversion symmetry. After the suppression of the NCS-
orthorhombic phase by the Sr substitution, superconductivity
was induced in the CS-orthorhombic phase (0.17 ≤ x ≤ 0.55)
at ∼3.5 K. For further substitution (0.6 ≤ x ≤ 1.0 and 0 ≤ y ≤
1.0), the NCS-tetragonal phase was stable, and the super-
conductivity was significantly suppressed. In order to under-
stand the effect of the crystal structure variation on
superconductivity, further experiments with theoretical support
are currently in progress.
Hu et al. investigated the effect of Pd substitution for
SrPt3P.19 They synthesized Sr(Pt1−zPdz)3P for 0 ≤ z ≤ 0.4 and
found that Tc decreases monotonously as z increases. In
addition, they found that the electron correlation is enhanced
while the electron−phonon coupling and spin−orbit coupling
are suppressed by Pd substitution. However, samples with z >
0.4 were not synthesized due to the substitution limit of Pd. If
samples can be synthesized by optimizing synthesis conditions,
there must be a phase transition from CS to NSC structure in
0.4 < z < 1.0, which can be a suitable platform to study the
effect of structure on superconductivity.
Bulk superconductivity at ∼3.5 K began to occur in the CS-
orthorhombic phase at x = ∼0.17. Note that the sample of x =
0.17 exhibited the normal state resistivity anomaly. However,
the anomalies are smaller than those of other samples,
suggesting that only a small portion of the sample undergoes
ASSOCIATED CONTENT
■
Accession Codes
CCDC 2005001−2005004 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
■
Corresponding Author
Akira Iyo − National Institute of Advanced Industrial Science
and Technology (AIST), Tsukuba, Ibaraki 3058568, Japan;
orcid.org/0000-0002-9610-647X; Email: iyo-akira@
aist.go.jp
F
https://dx.doi.org/10.1021/acs.inorgchem.0c01482
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
Zandbergen, H. W.; Ong, N. P.; Cava, R. J. Superconductivity in the
non-oxide perovskite MgCNi3. Nature 2001, 411, 54.
Authors
Hiroshi Fujihisa − National Institute of Advanced Industrial
Science and Technology (AIST), Tsukuba, Ibaraki 3058568,
Japan
Yoshito Gotoh − National Institute of Advanced Industrial
Science and Technology (AIST), Tsukuba, Ibaraki 3058568,
Japan
(8) Uehara, M.; Uehara, A.; Kozawa, K.; Yamazaki, T.; Kimishima,
Y. New Antiperovskite Superconductor ZnNNi3, and Related
Compounds CdNNi3 and InNNi3. Phys. C 2010, 470, S688−S690.
̂
(9) Uehara, M.; Yamazaki, T.; Kori, T.; Kashida, T.; Kimishima, Y.;
Hase, I. Superconducting Properties of CdCNi3. J. Phys. Soc. Jpn.
2007, 76, 034714.
Shigeyuki Ishida − National Institute of Advanced Industrial
Science and Technology (AIST), Tsukuba, Ibaraki 3058568,
Japan; orcid.org/0000-0001-9445-8079
Hiroki Ninomiya − National Institute of Advanced Industrial
Science and Technology (AIST), Tsukuba, Ibaraki 3058568,
Japan; orcid.org/0000-0001-7236-1241
Yoshiyuki Yoshida − National Institute of Advanced Industrial
Science and Technology (AIST), Tsukuba, Ibaraki 3058568,
Japan
Hiroshi Eisaki − National Institute of Advanced Industrial
Science and Technology (AIST), Tsukuba, Ibaraki 3058568,
Japan
Hishiro T. Hirose − National Institute for Materials Science,
Tsukuba, Ibaraki 3050003, Japan
Taichi Terashima − National Institute for Materials Science,
Tsukuba, Ibaraki 3050003, Japan
Kenji Kawashima − National Institute of Advanced Industrial
Science and Technology (AIST), Tsukuba, Ibaraki 3058568,
Japan; IMRA Material R&D Co., Ltd., Kariya, Aichi 4480032,
Japan; orcid.org/0000-0003-4786-3498
(10) Takayama, T.; Kuwano, K.; Hirai, D.; Katsura, Y.; Yamamoto,
A.; Takagi, H. Strong Coupling Superconductivity at 8.4 K in an
Antiperovskite Phosphide SrPt3P. Phys. Rev. Lett. 2012, 108, 23700.
(11) Lv, B.; Jawdat, B. I.; Wu, Z.; Sorolla, M.; Gooch, M.; Zhao, K.;
Deng, L.; Xue, Y.; Lorenz, B.; Guloy, A. M.; Chu, C. Synthesis,
Structure, and Superconductivity in the New-Structure-Type
Compound: SrPt6P2. Inorg. Chem. 2015, 54, 1049−1054.
(12) Lv, B.; Jawdat, B. I.; Wu, Z.; Li, S.; Chu, C. Superconductivity
in the Ternary Compound SrPt10P4 with Complex New Structure.
Phys. Rev. Mater. 2017, 1, 064801.
(13) Neumann, M. X-Cell: a novel indexing algorithm for routine
tasks and difficult cases. J. Appl. Crystallogr. 2003, 36, 356.
(14) Bruker. DIFFRAC.TOPAS; https://www.bruker.com/
products/x-ray-diffraction-and-elemental-analysis/x-ray-diffraction/
xrd-software/topas.html (accessed July 20, 2020).
̀
(15) Dassault Systemes Americas Corp. BIOVIA Materials Studio
Reflex; https://www.3ds.com/products-services/biovia/products/
molecular-modeling-simulation/biovia-materials-studio/analytical-
and-crystallization/ (accessed July 17, 2020).
(16) Clark, S. J.; Segall, M. D.; Pickard, C. J.; Hasnip, P. J.; Probert,
M. I. J.; Refson, K.; Payne, M. C. First principles methods using
CASTEP. Z. Kristallogr. - Cryst. Mater. 2005, 220, 567.
(17) Hiroi, Z.; Yamaura, J.; Kobayashi, T. C.; Matsubayashi, Y.;
Hirai, D. Pyrochlore Oxide Superconductor Cd2Re2O7 Revisited. J.
Phys. Soc. Jpn. 2018, 87, 024702.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.inorgchem.0c01482
(18) Subedi, A.; Ortenzi, L.; Boeri, L. Electron-phonon super-
conductivity in APt3P (A = Sr, Ca, La) compounds: From weak to
strong coupling. Phys. Rev. B: Condens. Matter Mater. Phys. 2013, 87,
144504.
Notes
The authors declare no competing financial interest.
(19) Hu, K.; Gao, B.; Ji, Q.; Ma, Y.; Li, W.; Xu, X.; Zhang, H.; Mu,
G.; Huang, F.; Cai, C.; Xie, X.; Jiang, M. Effects of Electron
Correlation, Electron-Phonon Coupling, and Spin-Orbit Coupling on
the Isovalent Pd-Substituted Superconductor SrPt3P. Phys. Rev. B:
Condens. Matter Mater. Phys. 2016, 93, 214510.
(20) Smidman, M.; Salamon, M. B.; Yuan, H. Q.; Agterberg, D. F.
Superconductivity and spin−orbit coupling in non-centrosymmetric
materials: a review. Rep. Prog. Phys. 2017, 80, 036501.
ACKNOWLEDGMENTS
■
This work was supported by the JSPS KAKENHI
(JP19K04481 and JP19H05823) and TIA collaborative
research program KAKEHASHI “Tsukuba−Kashiwa−Hongo
Superconductivity Kakehashi Project”.
REFERENCES
■
(21) Kneidinger, F.; Bauer, E.; Zeiringer, I.; Rogl, P.; Blaas-Schenner,
C.; Reith, D.; Podloucky, R. Superconductivity in non-centrosym-
metric materials. Phys. C 2015, 514, 388.
(1) Iyo, A.; Hase, I.; Fujihisa, H.; Gotoh, Y.; Takeshita, N.; Ishida,
S.; Ninomiya, H.; Yoshida, Y.; Eisaki, H.; Kawashima, K. Super-
conductivity Induced by Mg Deficiency in Noncentrosymmetric
Phosphide Mg2Rh3P. Phys. Rev. Mater. 2019, 3, 124802.
(2) Hase, I.; Yanagisawa, T.; Iyo, A.; Fujihisa, H.; Goto, Y.; Eisaki,
H.; Kawashima, K. Electronic Structure of Novel Non-Centrosym-
metric Superconductor Mg2Rh3P. J. Phys.: Conf. Ser. 2019, 1293,
012028.
̂
(22) Kito, H.; Yanagi, Y.; Ishida, S.; Oka, K.; Gotoh, Y.; Fujihisa, H.;
Yoshida, Y.; Iyo, A.; Eisaki, H. New Intermetallic Ternary Phosphide
Chalcogenide AP2‑xXx (A = Zr, Hf; X = S, Se) Superconductors with
PbFCl-Type Crystal Structure. J. Phys. Soc. Jpn. 2014, 83, 074713.
(3) Bauer, E.; Hilscher, G.; Michor, H.; Paul, C.; Scheidt, E. W.;
̈
Gribanov, A.; Seropegin, Y.; Noel, H.; Sigrist, M.; Rogl, P. Heavy
Fermion Superconductivity and Magnetic Order in Noncentrosym-
metric CePt3Si. Phys. Rev. Lett. 2004, 92, 027003.
(4) Oudah, M.; Ikeda, A.; Hausmann, J. N.; Yonezawa, S.;
Fukumoto, T.; Kobayashi, S.; Sato, M.; Maeno, Y. Superconductivity
in the Antiperovskite Dirac-Metal Oxide Sr3‑xSnO. Nat. Commun.
2016, 7, 13617.
(5) Badica, P.; Kondo, T.; Togano, K. Superconductivity in a New
Pseudo-Binary Li2B(Pd1‑xPtx)3 (x = 0−1) Boride System. J. Phys. Soc.
Jpn. 2005, 74, 1014.
(6) Nishiyama, M.; Inada, Y.; Zheng, G. Spin Triplet Super-
conducting State due to Broken Inversion Symmetry in Li2Pt3B. Phys.
Rev. Lett. 2007, 98, 047002.
(7) He, T.; Huang, Q.; Ramirez, A. P.; Wang, Y.; Regan, K. A.;
Rogado, N.; Hayward, M. A.; Haas, M. K.; Slusky, J. S.; Inumara, K.;
G
https://dx.doi.org/10.1021/acs.inorgchem.0c01482
Inorg. Chem. XXXX, XXX, XXX−XXX