Nickelate superconductors: structure before comparison

Scope

Nickelate superconductors include infinite-layer reduced films, bilayer Ruddlesden–Popper crystals and engineered layer sequences. The number of NiO2 layers, oxygen coordination and sample preparation distinguish these systems. An ambient-pressure epitaxial film and a compressed bulk crystal should retain separate records even when both contain lanthanum and nickel. Structural resemblance to cuprates motivates comparisons without establishing a shared pairing mechanism. R1 R2 R4

The 2026 La2PrNi2O7 high-field transport study concerns a bilayer Ruddlesden–Popper film. The inherited draft incorrectly placed it in the infinite-layer family. Its field-induced normal state is also distinct from the Eu-substituted infinite-layer material exhibiting re-entrant resistance suppression. Family, composition and probe must accompany these interpretations. R6 R7

Mathematical assumptions

Formal charge neutrality gives nickel valence $v_{\mathrm{Ni}}=1+x+2\delta$ for an idealized $R_{1-x}\mathrm{Sr}_x\mathrm{NiO}_{2+\delta}$ with trivalent $R$, divalent Sr and divalent oxygen. For $R_3\mathrm{Ni}_2\mathrm{O}_{7-\delta}$, it gives $v_{\mathrm{Ni}}=2.5-\delta$. Sixty-six exact bookkeeping checks preserve neutrality. These averages do not measure orbital occupancy, vacancy location, ligand holes or mobile carriers. In particular, an annealing-gas dose does not specify oxygen stoichiometry. R1 R5

Even the elementary two-carrier low-field Drude expression $R_H=(p\mu_h^2-n\mu_e^2)/[e(p\mu_h+n\mu_e)^2]$, with $e>0$, prevents a unique carrier count from Hall sign alone. The ambiguity can persist when zero-field conductivity is also known. In synthetic normalized units with $e=1$, the tuples $(n,p,\mu_e,\mu_h)=(2,1,2,1)$ and $(1,2,3,1)$ both give $\sigma_0=n\mu_e+p\mu_h=5$ and $R_H=-7/25$, despite opposite majority populations. Exact inversion of the finite-field conductivity tensor checks this common low-field limit; the two finite-field curves differ. These examples assume an ordinary isotropic two-carrier normal state with $\mu_e B,\mu_h B\ll1$, excluding anomalous Hall and vortex contributions. They are local counterexamples, not fits to nickelate data. Li and Hsu discuss multiband limitations, so a negative Hall coefficient must not be silently converted to a single-band electron density. R1 R6

Experimental criteria

For the representative Nd0.8Sr0.2NiO2 film in Li's 2019 discovery, the published text identifies a 14.9 K onset by maximum curvature, a 13.6 K midpoint and 9.1 K resistance indistinguishable from the noise floor. The final methods describe six-point electrical measurements, 9–11 nm perovskite precursor films and topotactic reduction; precursor thickness is not an independently measured reduced-film thickness. Two-coil mutual inductance supports a diamagnetic response, but the authors explicitly decline to extract a penetration depth because of finite-size, order-parameter and disorder uncertainties. The 2.2 μm value discussed nearby belongs to a cited 40 nm infinite-layer copper oxide comparison film, not a measured nickelate penetration depth. The three transition temperatures describe different criteria on the selected nickelate sample. R1

Sun's 2023 bilayer-crystal report gives an 80 K resistance onset at 18.5 GPa in run 1 without pressure medium. Run 2 uses KBr and reaches a 78 K onset at 18.9 GPa. Figure 3 distinguishes the 10 μA run-1 excitation from 300 μA in run 2. Its magnetic response near 77 K is measured separately at 25.2 GPa. Residual resistance and specimen differences prevent combining these observations into one zero-resistance or bulk-fraction record. R2

Oxygen, strain and transport

Bhatt's final strain study distinguishes nominal substrate mismatch from measured strain: the strongly compressed film is labelled approximately −1.6% measured strain, rather than the inherited draft's −1.8%. Growth method, oxygen treatment and substrate vary across comparisons. In-plane compression accompanied by out-of-plane expansion is not equivalent to hydrostatic compression. The strain article retains the source-specific structural comparison and its confounding variables. R3

Xiang's revised author manuscript demonstrates an oxygen-removal and ozone-annealing sequence that can recover degraded bilayer films. Repeating ozone treatment alone can instead damage the superconducting response. Its proposed phase diagram uses cumulative annealing-gas volume; oxygen concentration is not precisely measured, and unknown phases complicate interpretation. Figure 2 identifies a zero-resistance point near 5.5 K for sample S1049-4, far below its onset region. The recovery experiment constrains preparation sensitivity without establishing a calibrated carrier-density axis or a universal recycling recipe. R5

Hsu's final La2PrNi2O7 study defines 41 K using 90% of a fitted normal-state resistivity. Pulsed-field measurements access a low-temperature normal state with approximately quadratic resistivity; an extrapolated zero-field normal-state curve is not directly observed below the transition. The film is about 5 nm thick, grown on SrLaAlO4 with a SrTiO3 cap and kept cold between measurements. The reported effective mass uses an empirical Kadowaki–Woods relation, not direct calorimetry. These results describe a strained bilayer specimen, not every nickelate. R6

Recent structures and membranes

Nie's publisher abstract, published in April 2026, reports ambient-pressure onset temperatures spanning 46–50 K in engineered 1212 and 2323 layer sequences, alongside a nonsuperconducting 1313 comparison. The exact operational criterion and lower nondetection bound have not been extracted here, so this range is not assigned to a precise searchable Tc record. Associated orbital and Fermi-surface observations constrain structure–electronic-state relationships; they do not independently prove a unique pairing interaction. R4

Lee's April 2026 membrane preprint reports 74.2 K at 91.5 GPa using 99.5% of an extrapolated normal-state resistance. Extended Data Figure 1 gives 69.1 K with an intersection construction and 75.4 K with a second-derivative construction for the illustrated transition. These differences are criterion dependence, not separate superconducting phases. Zero resistance is defined by the noise floor and occurs lower. The preprint remains explicitly labelled and is not an ambient-pressure result. R8

The same preprint measures six membranes and fits an onset increase near 0.65 K/GPa over its sampled range. It uses room-temperature ruby calibration below 50 GPa and diamond Raman calibration above that, with the latter measured adjacent to the specimen. Silicone-oil solidification and nonhydrostatic stress complicate broadening. Its 13 T field measurements and fitted coherence lengths should not be mistaken for direct measurements of a much larger zero-temperature critical field. Extrapolating the pressure trend beyond the measured range is not a demonstrated materials prediction. R8

Limits and disputes

Yang's author manuscript identifies an approximately 54-unit-cell Eu-substituted infinite-layer SCE0.34 film whose resistance returns to zero above 6.23 T at 2 K with field along the c axis. Its separate temperature sweeps show a mutual-inductance screening response at 9 T and 12 T, absent at 3 T. This magnetic evidence supports re-entrant superconductivity beyond a resistance drop alone; 2 K is the field-sweep measurement temperature, not a new transition temperature. Magnetic compensation and triplet pairing remain proposed interpretations, and long-range magnetic order is not established. The re-entrant superconductivity and magnetic substitution articles retain the measurement windows and alternative mechanisms. None of these observations establishes a universal nickelate gap symmetry. R7

Complete scientific acceptance remains open. This is an editorial review-status statement: the cited studies supply examples of differing evidence and criteria, not certification of this article. Selected primary passages, figures and local model checks have been reviewed, while full supplements, complete final-author version comparisons, independent reproduction and raw-data analysis remain incomplete. Onset, midpoint, noise-floor zero, spectroscopic gaps and normal-state measurement temperatures retain distinct meanings. Pressure and strain comparisons require matching phase and preparation rather than ordering headline temperatures. R1 R2 R3 R8

Discovery implications

A useful comparison records layer sequence, oxygen history, nominal and measured composition, substrate, thickness, pressure calibration, probe and temperature criterion. Controlled oxygen cycles and structural contrasts are informative only with their phase and interface changes documented. A promising design hypothesis should specify which variable changes, which confounders remain and what observation would distinguish competing mechanisms. R3 R4 R5

Related topics

Continue with cuprates, strain, two-dimensional interfaces, multiband pairing, and magnetic substitution.

Material cases

Values describe individual reports, not a ranking. Unknown means the checked source did not supply the field at this review scope.