reSeArCH Letter
temperatures resembles measurement results of a 40-nm-thick infinite-
layer copper oxide film with Tc ≈ 10.8 K and extrapolated London
penetration depth λL(T = 0) = 2.2 μm (ref.^31 ). This indicates that λL
for Nd0.8Sr0.2NiO 2 is similarly large compared to the film thickness.
Given the numerical uncertainties arising from the finite sample size
(substantially wider films show indications of laterally inhomogeneous
reduction), the order parameter symmetry and the scale of disorder, we
did not attempt to extract λL (ref.^32 ). Nevertheless, these data suggest
that this is a type-II superconductor with second critical field Hc2,⊥
approximately given in the inset to Fig. 4a.
Clearly the analogy to copper oxides motivated this finding, and
much remains to be explored in this new superconducting compound.
However, several important dissimilarities between these two systems
are apparent. One key difference is the energy level alignments in their
orbital electronic structure. Holes in copper oxides are often discussed
in terms of Zhang–Rice singlets with strong oxygen character, owing
to the close spatial overlap and near-energetic degeneracy of the Cudxy (^22) − orbitals and the O 2p orbitals
(^33). This naturally leads to large
in-plane antiferromagnetic coupling, which many consider to be cen-
tral for superconducting pairing^24. Because Ni+ is one column to the
left of Cu^2 + on the periodic table and one oxidation state lower, the
chemical potential in the infinite-layer nickelates is several electronvolts
higher than that of comparable copper oxides; therefore, in hole-doped
nickelates, much less hybridization with the O 2p band is expected^6.
Furthermore, powder neutron diffraction studies of LaNiO 2 and
NdNiO 2 show no indication of magnetic order down to 5 K and 1.7 K,
respectively^15 ,^16 , and the resistivity of NdNiO 2 (Fig. 3b) is inconsistent
with a robust insulator (although interface effects may contribute to
conductivity). Consequently, two features that are central to copper
oxides—the Zhang–Rice singlet and large planar spin fluctuations—
may be absent (or considerably diminished) in these nickelate
superconductors.
On the materials side, one immediate question is the effect of
the various substrates on the topotactic structural transition of this
–100
–50
0
50
100
E (V cm
–1
)
–200 –100 0 100 200
J (kA cm–2)
2 K 10 K
3 K 11 K
4 K 12 K
5 K 13 K
6 K 14 K
7 K 15 K
8 K 16 K
9 K
1.0
0.8
0.6
0.4
0.2
0.0
Resistivity (m
Ω
cm)
0 5 10 15 20 25 30
Temperature (K)
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Resistivity (m
Ω
cm)
0 100 200 300
Temperature (K)
–8
–6
–4
–2
0
2
Hall coef cient (×10
–3
cm
3 C
–1
)
0 100 200 300
Temperature (K)
NdNiO 2
Nd0.8Sr0.2NiO 2
2.0
1.5
1.0
0.5
0.0
Resistivity (m
Ω
cm)
0 100 200 300
Temperature (K)
NdNiO 2
10 –1 Nd0.8Sr0.2NiO 2
100
101
102
Resistivity (m
Ω
cm)
0 100 200 300
Temperature (K)
NdNiO 3
Nd0.8Sr0.2NiO 3
ab
d e
c
f
Fig. 3 | Transport properties and superconductivity of the nickelate thin
films. a, Resistivity versus temperature ρ(T) plots of the as-grown NdNiO 3
and Nd0.8Sr0.2NiO 3 films. b, c, Resistivity (b) and normal-state Hall
coefficient (c) as a function of temperature for the corresponding reduced
films (NdNiO 2 and Nd0.8Sr0.2NiO 2 ). d, e, ρ(T) for multiple Nd0.8Sr0.2NiO 2
films, showing resistive superconducting transitions. Dotted lines indicate
samples without a capping layer, for which the XRD Scherrer thickness
was used to estimate the resistivity. f, Electric field (E) versus current
density (J) characteristics for varying temperature.
1.8
1.7
1.6
1.5
1.4
1.3
1.2
Re(
Vp
) (
μV)
0 2 4 6 8 10 12 14
Temperature (K)
140
120
100
80
60
Im(
Vp
) (nV)
500
400
300
200
100
0
Resistivity (
μΩ
cm)
0 10 20 30 40 50
Temperature (K)
0 T 5 T
1 T 7 T
2 T 9 T
3 T 11 T
4 T 13 T
20
15
10
5
0
0 4 8 12 16
Temperature (K)
P^0
Hc,
(T)⊥
ab
Fig. 4 | Magnetic-field response of superconducting Nd0.8Sr0.2NiO 2.
a, ρ(T) under a varying magnetic field perpendicular to the a–b plane.
The inset shows the variation of the upper critical field Hc,⊥ (as estimated
by the midpoint of the resistive transition) with a linear fit in the vicinity
of Tc. b, The real (Re(Vp)) and imaginary (Im(Vp)) parts of the voltage
as a function of temperature in the pickup coil on a Nd0.8Sr0.2NiO 2 film,
measured using a two-coil mutual-inductance measurement. μ 0 , magnetic
constant.
626 | NAtUre | VOL 572 | 29 AUGUSt 2019