betaFeedback

WorksPhysics and Astronomy

Sensitive dependence of pairing symmetry on Ni-ege_g crystal field splitting in the nickelate superconductor La3_3Ni2_2O7_7

Abstract

The discovery of high-temperature superconductivity in La3_3Ni2_2O7_7 under pressure has drawn great attention. However, consensus has not been reached on its pairing symmetry in theory. By combining density-functional-theory (DFT), maximally-localized-Wannier-function, and linearized gap equation with random-phase-approximation, we find that the pairing symmetry of La3_3Ni2_2O7_7 is dxyd_{xy}, if its DFT band structure is accurately reproduced by a downfolded bilayer two-orbital model. More importantly, we reveal that the pairing symmetry of La3_3Ni2_2O7_7 sensitively depends on the crystal field splitting between two Ni-ege_g orbitals. A slight increase in Ni-ege_g crystal field splitting alters the pairing symmetry from dxyd_{xy} to s±s_{\pm}. Such a transition is associated with the change in inverse Fermi velocity and susceptibility, while the shape of Fermi surface remains almost unchanged. Our work highlights the sensitive dependence of pairing symmetry on low-energy electronic structure in multi-orbital superconductors, which calls for care in the downfolding procedure when one calculates their pairing symmetry.

Introduction

The recent discovery of superconductivity in La3Ni2O7\mathrm{La}_3\mathrm{Ni}_2\mathrm{O}_7 under pressure has drawn great attention [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54]. It not only adds a new member to the family of nickelate superconductors [55, 56, 57, 58, 59, 60, 61, 62, 63], but also substantially increases the transition temperature to near 80 K [1].

Under a pressure above 19 GPa, La3Ni2O7\mathrm{La}_3\mathrm{Ni}_2\mathrm{O}_7 transforms from an orthorhombic structure (space group AmamAmam) to a tetragonal structure (space group I4/mmmI4/mmm) at low temperatures [45, 54].

Extensive density-functional-theory (DFT) calculations find that the Fermi surface of tetragonal La3Ni2O7\mathrm{La}_3\mathrm{Ni}_2\mathrm{O}_7 has three Fermi sheets that are derived from two Ni-ege_g orbitals [4, 7, 8, 9]. This is different from the well-known single-orbital Fermi surface of cuprate superconductors [64, 65, 66]. Based on this low-energy electronic structure, a number of studies use many-body methods and propose different mechanisms for the superconductivity found in La3Ni2O7\mathrm{La}_3\mathrm{Ni}_2\mathrm{O}_7. Yet no consensus has been reached so far on the most favorable pairing symmetry of La3Ni2O7\mathrm{La}_3\mathrm{Ni}_2\mathrm{O}_7—a central property of unconventional superconductivity. Refs. [5, 6, 7, 8, 9, 10, 11, 12, 29] find s±s_\pm symmetry (group theory notation A1gA_{1g}), reminiscent of iron-based superconductors [67, 68, 69], while Refs. [13, 16, 17, 18, 19, 20] obtain dxyd_{xy} (B2gB_{2g}) or dx2−y2d_{x^2-y^2} (B1gB_{1g}) symmetry.

In this work, we combine density-functional-theory (DFT) [70, 71], maximally-localized-Wannier-function (MLWF) [72, 73], and linearized gap equation calculations with random-phase-approximation (RPA) [74, 75, 76, 77, 78, 79] to study the superconducting pairing symmetry of tetragonal La3Ni2O7\mathrm{La}_3\mathrm{Ni}_2\mathrm{O}_7 by downfolding the DFT band structure to a bilayer two-orbital model. We find that the leading superconducting instability of La3Ni2O7\mathrm{La}_3\mathrm{Ni}_2\mathrm{O}_7 has dxyd_{xy} symmetry, if its DFT band structure is accurately reproduced by the downfolded model and the interaction strengths are in a physically reasonable range. More importantly, we show that a slight increase in the crystal field splitting between two Ni-ege_g orbitals (<0.2< 0.2 eV) changes the pairing symmetry from dxyd_{xy} to s±s_\pm. We elucidate that such a transition is associated with the change in inverse Fermi velocity and susceptibility, while the shape of Fermi surface remains almost unchanged. Our work highlights the sensitive dependence of La3Ni2O7\mathrm{La}_3\mathrm{Ni}_2\mathrm{O}_7 pairing symmetry on Ni-ege_g crystal field splitting, which provides one source of discrepancy in the pairing symmetry reported in the literature. Such sensitivity calls for care in the procedure of downfolding DFT band structure to low-energy effective models when one calculates the pairing symmetry of multi-orbital superconductors.

We perform DFT calculations, as implemented in the Vienna ab initio simulation package (VASP) [80, 81]. We use MLWF to do the downfolding from DFT band structure [82] and obtain a bilayer two-orbital model for subsequent linearized gap equation calculations. We refer to our bilayer two-orbital model as Wannier’s model. For comparison, we also study another bilayer two-orbital model [4], referred to as Luo’s model in this study, which is widely used to study the superconducting pairing symmetry of La3_3Ni2_2O7_7 in literature. Wannier’s model and Luo’s model have the same bases: (Ni1-d3z2−r2, Ni1-dx2−y2, Ni2-d3z2−r2, Ni2-dx2−y2)(\mathrm{Ni1}\text{-}d_{3z^{2}-r^{2}},\ \mathrm{Ni1}\text{-}d_{x^{2}-y^{2}},\ \mathrm{Ni2}\text{-}d_{3z^{2}-r^{2}},\ \mathrm{Ni2}\text{-}d_{x^{2}-y^{2}}) where 1 and 2 label different NiO2_2 layers. The difference between the two models is that Wannier’s model takes into account all the hoppings between different orbitals, while Luo’s model only includes nearest and second-nearest hoppings. Thus the two models have slightly different Ni-ege_g crystal field splitting and hopping parameters (see Supplementary Note 3). We use RPA to treat the local Hubbard interaction on Ni-ege_g orbitals in the Slater-Kanamori form [83] and obtain the effective pairing potential. The onsite and hopping parameters of Wannier’s model and Luo’s model, the local Hubbard interaction, as well as the computational details of DFT, RPA and linearized gap equation calculations are found in Methods.

Results

Electronic structure and susceptibility

Figure 1(a) shows the crystal structure of La3_3Ni2_2O7_7 under high pressure. In our DFT calculations, we find that after atomic relaxation, oxygen octahedral rotations in La3_3Ni2_2O7_7 are completely suppressed under 30 GPa, consistent with the experiment [1, 45, 54] and other theoretical calculations [8, 23]. We obtain a tetragonal structure whose theoretical lattice constants are in good agreement with the experiment [1, 45, 54] (see the comparison in Supplementary Note 1). Following the previous studies [4, 7, 8, 9, 10, 13, 14, 15, 16], because the inter-block interaction is weak, we focus on one-block of La3_3Ni2_2O7_7, whose DFT band structure is used to do the downfolding to a bilayer two-orbital model whose first Brillouin zone is a simple two-dimensional (2D) square.

Image not extracted

Figure 1. Crystal structure and electronic structure. (a) Crystal structure of La3_3Ni2_2O7_7 under high pressure. (b) Comparison of DFT band structure of La3_3Ni2_2O7_7 under 30 GPa (blue solid curves) to the band structure of Wannier’s model (red dashed curves). (c) The Fermi surface of Wannier’s model. The green color indicates the magnitude of inverse Fermi velocity ∣vF(k)∣−1\lvert v_{\mathrm{F}}(\mathbf{k})\rvert^{-1}. (d) Comparison of DFT band structure of La3_3Ni2_2O7_7 under 30 GPa (blue solid curves) to the band structure of Luo’s model (red dashed curves). (e) The Fermi surface of Luo’s model. The green color indicates the magnitude of inverse Fermi velocity ∣vF(k)∣−1\lvert v_{\mathrm{F}}(\mathbf{k})\rvert^{-1}. Source data are provided as a Source Data file.

Figure 1(b) shows the comparison of DFT band structure of La3_3Ni2_2O7_7 and the band structure from Wannier’s model. Wannier’s model exactly reproduces the four Ni-ege_g-derived bands close to the Fermi level. The Ni-ege_g atomic projections onto those bands are shown in Supplementary Note 4. Figure 1(c) shows the Fermi surface of Wannier’s model, which has three sheets in the 2D Brillouin zone, labeled as α\alpha, β\beta, and γ\gamma. The α\alpha and β\beta sheets are mainly derived from Ni-dx2−y2d_{x^{2}-y^{2}} orbital, while the γ\gamma sheet dominantly consists of Ni-d3z2−r2d_{3z^{2}-r^{2}} (see Supplementary Note 4). We use green color to indicate the magnitude of inverse Fermi velocity ∣vF(k)∣−1\lvert v_{\mathrm{F}}(\mathbf{k})\rvert^{-1}. The γ\gamma sheet has a large inverse Fermi velocity, while the inverse Fermi velocity of the α\alpha and β\beta sheets is substantially smaller. Next we switch to Luo’s model for comparison. Figure 1(d) shows the band structure of Luo’s model, which overlays the DFT band structure of La3_3Ni2_2O7_7. Luo’s model also reasonably reproduces the four Ni-ege_g-derived bands around the Fermi level. The minor difference in the band structure between DFT and Luo’ model is probably due to the fact that Luo’s model only takes into account the nearest and second nearest hoppings [4], while DFT (as well as Wannier’s model) includes

all the long-range hoppings. However, the Fermi surface of Luo’s model is very similar to that of Wannier’s model, as shown in Figure 1(e). We find that in Luo’s model, the inverse Fermi velocity of the γ\gamma sheet is slightly larger than that of Wannier’s model. We will return to this important point later when we discuss the superconducting pairing symmetry.

Next we compare the total and orbital-projected bare susceptibility χ0(q)\chi_0(\mathbf{q}) of the two models (the definition of different kinds of susceptibility is found in Methods). Figure 2(a) and (b) show χ0(q)\chi_0(\mathbf{q}) of Wannier’s model on a high-symmetry q\mathbf{q}-path and in the 2D Brillouin zone. In panel (a), both total and orbital-projected χ0(q)\chi_0(\mathbf{q}) are shown, while in panel (b), we only show total χ0(q)\chi_0(\mathbf{q}). In Wannier’s model, the total χ0(q)\chi_0(\mathbf{q}) has a peak at MM. The peak value would be substantially enhanced in the spin susceptibility χs(q)\chi_s(\mathbf{q}) when local Hubbard interactions are added on Ni orbitals in RPA calculations (see Supplementary Note 5). By comparison, we perform the same calculations on Luo’s model and show the results in Figure 2(c) and (d). We find that while χ0(q)\chi_0(\mathbf{q}) of Luo’s model is overall similar to that of Wannier’s model, there are some notable differences. In Luo’s model, total χ0(q)\chi_0(\mathbf{q}) at MM gets suppressed and the highest peak of total χ0(q)\chi_0(\mathbf{q}) is along Γ\Gamma to MM, close to (π2,π2)\left(\frac{\pi}{2},\frac{\pi}{2}\right). Compared to Wannier’s model, the total χ0(q)\chi_0(\mathbf{q}) is also slightly increased around the zone center (Γ\Gamma point) in Luo’s model. The implications of those changes in χ0(q)\chi_0(\mathbf{q}) for the superconducting pairing symmetry will be discussed at the end.

Image not extracted

Figure 2. Susceptibility. Bare static susceptibility χ0(q)\chi_0(\mathbf{q}) at T=116T = 116 K. (a) Total and orbital-projected χ0(q)\chi_0(\mathbf{q}) of Wannier’s model along the high-symmetry q\mathbf{q}-path. (b) Total χ0(q)\chi_0(\mathbf{q}) of Wannier’s model in the 2D Brillouin zone. (c) Total and orbital-projected χ0(q)\chi_0(\mathbf{q}) of Luo’s model along the high-symmetry q\mathbf{q}-path. (d) Total χ0(q)\chi_0(\mathbf{q}) of Luo’s model in the 2D Brillouin zone. Source data are provided as a Source Data file.

Superconducting pairing symmetry

Now we study the superconducting pairing symmetry of La₃Ni₂O₇ by solving the linearized gap equation with RPA method to treat the local Hubbard interaction for the effective pairing potential. RPA method has been widely used to calculate the pairing symmetry of unconventional superconductors such as cuprates and iron-pnictides [84, 85], which proves to be reliable and consistent with other methods including fluctuation exchange approximation (FLEX) and functional renormalization group (fRG) [68, 77]. We study a range of UU with JH=0.15UJ_{\mathrm{H}} = 0.15U. The key results do not change if other physically reasonable ratios of JH/UJ_{\mathrm{H}}/U are used (see Supplementary Note 9). In Figure 3(a), we show the eigenvalues Λ\Lambda of the linearized gap equation using Wannier’s model. Since the superconducting transition temperature Tc∝e−1/λT_{\mathrm{c}} \propto e^{-1/\lambda}, the eigenvector of the largest λ\lambda corresponds to the most favorable superconducting pairing symmetry. We find that the leading superconducting instability has dxyd_{xy} symmetry with a sign difference between α\alpha sheet and β/γ\beta/\gamma sheets. In addition, we also show two sub-leading superconducting instabilities that have dx2−y2d_{x^2-y^2} and s±s_{\pm} pairing symmetries. The leading superconducting pairing symmetry does not change with the interaction strength UU (we find Uc=1.4U_{\mathrm{c}} = 1.4 eV that triggers the formation of spin density wave). Figure 3(b) shows the gap function for the largest eigenvalue of linearized gap equation, using Wannier’s model. This gap function clearly exhibits dxyd_{xy} symmetry (the gap function of the two sub-leading pairing symmetries is shown in Supplementary Note 12). In Figure 3(c), we show the superconducting pairing symmetry of Luo’s model. By contrast, we find that the leading superconducting instability has s±s_{\pm} symmetry, with two sub-leading superconducting instabilities that have dxyd_{xy} and dx2−y2d_{x^2-y^2} pairing symmetries, consistent with the previous studies [8, 9, 10, 12]. The leading superconducting pairing symmetry in Luo’s model does not change with the interaction strength either (Uc=1.3U_c = 1.3 eV for spin density wave). Figure 3(d) shows the gap function for the largest eigenvalue of linearized gap equation, using Luo’s model. This gap function exhibits s±s_{\pm} symmetry, in which α\alpha and γ\gamma sheets are of the same

Image not extracted

Figure 3. Superconducting pairing symmetry. (a) The eigenvalues λ\lambda of linearized gap equation with RPA method, using Wannier’s model. The leading superconducting instability has dxyd_{xy} (red squares), while two sub-leading superconducting instabilities have dx2−y2d_{x^2-y^2} (green triangles) and s±s_{\pm} (blue circles) symmetries, respectively. (b) Gap function for the leading superconducting instability of Wannier’s model calculated at U=1.2U = 1.2 eV and JH=0.15U=0.18J_H = 0.15U = 0.18 eV, which exhibits dxyd_{xy} symmetry. (c) The eigenvalues λ\lambda of linearized gap equation with RPA method, using Luo’s model. The leading superconducting instability has s±s_{\pm} (blue circles), while two sub-leading superconducting instabilities have dxyd_{xy} (red squares) and dx2−y2d_{x^2-y^2} (green triangles) symmetries, respectively. (d) Gap function for the leading superconducting instability of Luo’s model calculated at U=1.2U = 1.2 eV and JH=0.15U=0.18J_H = 0.15U = 0.18 eV, which exhibits s±s_{\pm} symmetry. Source data are provided as a Source Data file.

sign, while β\beta sheet is of the opposite sign. We note that a superconducting order parameter that changes sign between different Fermi sheets is widely observed in iron-pnictides [67, 68, 69, 86, 87, 88].

Sensitivity to Ni-ege_g crystal field splitting

Next we trace the main source for the different pairing symmetries predicted by Wannier’s model and Luo’s model, despite the fact that they exhibit very similar Fermi surfaces. As we mentioned earlier, the two models have the same bases but slightly different parameters. By comparing the onsite energies and hopping matrix elements, we find that the “biggest” difference between the two models is Ni-ege_g crystal field splitting Δ\Delta. The difference in Δ\Delta is about 0.15 eV, while the differences in the hopping matrix elements are a few times or even one order of magnitude smaller (see details in Supplementary Note 3). Next we perform a “thought-experiment”, in which we treat Δ\Delta as a free parameter and manually tune its value. For a given value of Δ\Delta, we re-calculate the band structure and superconducting pairing symmetry using the linearized gap equation with RPA method. We use U=1.2U = 1.2 eV as a representative interaction strength. Our main results do not depend on this specific value of UU (see Supplementary Note 14). In Figure 4(a), we compare the band structure of modified Wannier’s model to DFT band structure of La3_3Ni2_2O7_7 under 30 GPa, as Δ\Delta is fine-tuned in a small range from 0.20 eV to 0.40 eV. The optimal value of Δ\Delta in Wannier’s model is 0.22 eV, based on the fitting. We find that tuning Δ\Delta in this small range yields marginal effects on the overall band structure, since the total bandwidth is about 4 eV. Correspondingly, the shape of the resulting Fermi surfaces remains almost unchanged with Δ\Delta (see Supplementary Note 10). In Figure 4(b), we show the eigenvalues for the leading superconducting instability, using modified Wannier’s model, as Δ\Delta varies. For comparison, the eigenvalues for two sub-leading superconducting instabilities are also shown. We find that the leading superconducting instability has dxyd_{xy} symmetry (s±s_{\pm} symmetry) if Δ\Delta is smaller (larger) than 0.37 eV. The brown dashed line highlights the original value of Δ\Delta in Wannier’s model, at which the leading superconducting pairing symmetry is dxyd_{xy}. Next we switch to Luo’s model. In Figure 4(c), we compare the band structure of modified Luo’s model to DFT band structure of La3_3Ni2_2O7_7 under 30 GPa, as Δ\Delta is fine-tuned in the same range. The optimal value of Δ\Delta in Luo’s model is 0.37 eV, larger than Wannier’s model. Similar to

Image not extracted

Figure 4. Tuning Ni-ege_g crystal field splitting. (a) Comparison of DFT band structure of La3_3Ni2_2O7_7 under 30 GPa (gray) to the band structure of modified Wannier’s model at a few different Ni-ege_g crystal field splittings Δ\Delta. (b) The eigenvalues λ\lambda of linearized gap equation with RPA method for the leading superconducting instability as well as two sub-leading superconducting instabilities in modified Wannier’s model, calculated at U=1.2U = 1.2 eV and JH=0.18J_H = 0.18 eV. The leading superconducting pairing symmetry is dxyd_{xy} (s±s_{\pm}) when Δ\Delta is smaller (larger) than 0.37 eV. The dashed brown line highlights the original value of Δ=0.22\Delta= 0.22 eV in Wannier’s model. (c) Comparison of DFT band structure of La3_3Ni2_2O7_7 under 30 GPa (gray) to the band structure of modified Luo’s model at a few different Ni-ege_g crystal field splittings Δ\Delta. (d) The eigenvalues λ\lambda of linearized gap equation with RPA method for the leading superconducting instability as well as two sub-leading superconducting instabilities in modified Luo’s model, calculated at U=1.2U = 1.2 eV and JH=0.18J_H = 0.18 eV. The leading superconducting pairing symmetry is dxyd_{xy} (s±s_{\pm}) when Δ\Delta is smaller (larger) than 0.30 eV. The dashed brown line highlights the original value of Δ=0.37\Delta= 0.37 eV in Luo’s model. Source data are provided as a Source Data file.

Wannier’s model, we find insignificant changes in the overall band structure of Luo’s model by such fine-tuning of Δ\Delta. Then we re-calculate the superconducting pairing symmetry at the same interaction strengths, using modified Luo’s model, as Δ\Delta varies. Figure 4(d) shows the eigenvalues for the leading superconducting instabilities, as well as two sub-leading superconducting instabilities. We find the same trend that a smaller (larger) Δ\Delta favors dxyd_{xy} symmetry (s±s_{\pm} symmetry) with the critical point at 0.30 eV. The dashed brown line highlights the original value of Δ\Delta in Luo’s model, which is larger than the critical value. Thus the leading superconducting pairing symmetry of Luo’s model is s±s_{\pm}.

To get a deeper understanding of the effects from a small change in Ni-ege_g crystal field splitting on the pairing symmetry, we study the inverse Fermi velocity ∣vF(k)∣−1\lvert v_F(\mathbf{k})\rvert^{-1} and bare total susceptibility χ0(q)\chi_0(\mathbf{q}), both of which play a crucial role in the linearized gap equation calculations [76]. As we find in Figure 1, ∣vF(k)∣−1\lvert v_F(\mathbf{k})\rvert^{-1} has large values on γ\gamma sheet, while the α\alpha and β\beta sheets have substantially smaller ∣vF(k)∣−1\lvert v_F(\mathbf{k})\rvert^{-1}. Thus we average ∣vF(k)∣−1\lvert v_F(\mathbf{k})\rvert^{-1} over the γ\gamma sheet of modified Wannier’s model and show (∣vF(k)∣−1)avg(\lvert v_F(\mathbf{k})\rvert^{-1})_{\mathrm{avg}} as a function of Ni-ege_g crystal field splitting Δ\Delta. The raw data of ∣vF(k)∣−1\lvert v_F(\mathbf{k})\rvert^{-1} as a function of Δ\Delta are found in Supplementary Note 10. Figure 5(a) shows that (∣vF(k)∣−1)avg(\lvert v_F(\mathbf{k})\rvert^{-1})_{\mathrm{avg}} monotonically increases with Δ\Delta. This indicates that with an increased Δ\Delta, the γ\gamma sheet that is mainly composed of d3z2−r2d_{3z^2-r^2} orbital, carries more weight in the linearized gap equation calculations. Figure 5(b) shows the bare total susceptibility χ0(q)\chi_0(\mathbf{q}) of modified Wannier’s model at a few different Δ\Delta (the orbital-projected χ0(q)\chi_0(\mathbf{q}) are shown in Supplementary Note 11). Figure 5(b) shows that when Δ=0.2\Delta= 0.2 eV, total χ0(q)\chi_0(\mathbf{q}) has a clear peak at MM point, which suggests that intralayer correlations are significant. This usually favors dd-wave pairing [64, 65, 66]. As Δ\Delta increases, total χ0(q)\chi_0(\mathbf{q}) gets reduced along the zone boundary (X→MX \to M), but increases around the zone center (Γ\Gamma point). The enhancement of χ0(q)\chi_0(\mathbf{q}) around Γ\Gamma point is associated with the strengthening of interlayer correlations. This, together with d3z2−r2d_{3z^2-r^2} orbital carrying more weight, favors ss-wave pairing, as a number of recent studies have shown [5, 22, 38]. Figure 5(c) and (d) show (∣vF(k)∣−1)avg(\lvert v_F(\mathbf{k})\rvert^{-1})_{\mathrm{avg}} and total χ0(q)\chi_0(\mathbf{q}) of modified Luo’s model as Δ\Delta varies. We find exactly the same trends as Wannier’s model: 1) (∣vF(k)∣−1)avg(\lvert v_F(\mathbf{k})\rvert^{-1})_{\mathrm{avg}} increases with Δ\Delta and 2) an increasing Δ\Delta reduces χ0(q)\chi_0(\mathbf{q}) along the zone boundary but enhances χ0(q)\chi_0(\mathbf{q}) around the zone center, both of which suggests that the favorable superconducting pairing symmetry would change from dd-wave to ss-wave. In particular, at the original value of Δ\Delta, (∣vF(k)∣−1)avg(\lvert v_F(\mathbf{k})\rvert^{-1})_{\mathrm{avg}} of Luo’s model is 4.2 eV−1^{-1}, larger than that of Wannier’s model (3.9 eV−1^{-1}), which shows the tendency of favoring s±s_{\pm} symmetry.

Image not extracted

Figure 5. Inverse Fermi velocity and bare static susceptibility. (a) Average inverse Fermi velocity (∣vF(k)∣−1)avg(|v_F(\mathbf{k})|^{-1})_{\mathrm{avg}} over the γ\gamma sheet of modified Wannier’s model as a function of Ni-ege_g crystal field splitting Δ\Delta. The dashed brown line highlights the original value of Δ=0.22\Delta= 0.22 eV in Wannier’s model. (b) Total χ0(q)\chi_0(\mathbf{q}) along the high-symmetry q\mathbf{q}-path of modified Wannier’s model at a few different Ni-ege_g crystal field splittings Δ\Delta. (c) Average inverse Fermi velocity (∣vF(k)∣−1)avg(|v_F(\mathbf{k})|^{-1})_{\mathrm{avg}} over the γ\gamma sheet of modified Luo’s model as a function of Ni-ege_g crystal field splitting Δ\Delta. The dashed brown line highlights the original value of Δ=0.37\Delta= 0.37 eV in Luo’s model. (d) Total χ0(q)\chi_0(\mathbf{q}) along the high-symmetry q\mathbf{q}-path of modified Luo’s model at a few different Ni-ege_g crystal field splittings Δ\Delta. We set the lattice constant a=1a = 1 and the reduced Planck constant ℏ=1\hbar= 1, and thus Fermi velocity has the unit of energy. Source data are provided as a Source Data file.

Discussions

Before we conclude, we make a few comments.

  1. The above analysis does not mean that the pairing symmetry of La3_3Ni2_2O7_7 only depends on Ni-ege_g crystal field splitting. Rather, what we want to highlight is the sensitive dependence

of La3Ni2O7La_3Ni_2O_7 pairing symmetry on its low-energy electronic structure, in particular Ni-ege_g crystal field splitting. This implies that a prediction of pairing symmetry based on a given tight-binding model could be biased and does not provide the complete picture, considering inevitable approximations and uncertainties introduced in the modelling. Instead, the trend that a smaller (larger) Ni-ege_g crystal field splitting favors dxyd_{xy} (s±s_{\pm}) pairing symmetry in La3Ni2O7La_3Ni_2O_7 is robust and conclusive.

  1. Compared to Ref. [41], we employ smaller UU values in the linearized gap equation calculations with RPA, because RPA is a weak-coupling theory and UU in RPA calculations is understood as a renormalized (i.e. reduced) interaction Uˉ\bar{U} [89, 90]. Spin density wave (SDW) instability is triggered at a small Uˉ\bar{U} value in the RPA method. This is an intrinsic property of RPA, as pointed out by Refs. [76, 90]. Our critical UU value for SDW in La3Ni2O7La_3Ni_2O_7 is consistent with the previous RPA studies [9, 17].

  1. Our calculations show that when the ratio JH/UJ_H/U is within a physically reasonable range between 0.1 and 0.2 [9, 10, 12, 17, 78], Wannier’s model favors dxyd_{xy} symmetry. Using Wannier’s model, we also calculate the phase diagram of pairing symmetry for a larger range of JH/UJ_H/U ratios (see Supplementary Note 9). We find that when JH/UJ_H/U were sufficiently large, Wannier’s model would favor dx2−y2d_{x^2-y^2} pairing symmetry, consistent with Ref. [17]. In addition, a number of other works use variants of tt-JJ model to study the superconducting pairing symmetry of La3Ni2O7La_3Ni_2O_7 [21, 22, 38]. They find that the pairing symmetry depends on exchange interactions JJ. This does not conflict with our results, because in multiorbital systems, when one derives the tt-JJ model from a Hubbard-like model, the exchange interaction JJ depends on hopping parameters and crystal field splitting, as well as UU and JHJ_H. In our work, we find that when Ni-ege_g crystal field splitting increases, the favorable pairing symmetry changes from dd-wave to ss-wave, because the d3z2−r2d_{3z^2-r^2} orbital carries more weight in the linearized gap equation calculations and the interlayer correlation is enhanced, as manifested by the susceptibility χ(q)\chi(\mathbf{q}) at Γ\Gamma point. This is consistent with Refs. [22, 38], which find that an increased interlayer exchange interaction J⊥J_{\perp} would change the pairing symmetry from dd-wave to ss-wave in La3Ni2O7La_3Ni_2O_7.

  1. We comment that tuning Ni-ege_g crystal field splitting of complex oxides is experimentally feasible because Ni-ege_g crystal field splitting is directly coupled to the NiO6_6 oxygen octahedra environment [91, 92]. Static or dynamical perturbations may change the Ni-ege_g crystal field splitting by up to 0.5 eV, which is sufficiently large to alter the pairing symmetry

of La3_3Ni2_2O7_7.

  1. Related to the preceding point, since the physical Ni-ege_g crystal field splitting is very close to the critical value, La3_3Ni2_2O7_7 may exhibit distinct pairing symmetries under experimental perturbations that can either alter the ratio of in-plane and out-of-plane Ni-O bond lengths, or move the position of Ni atoms in a NiO6_6 oxygen octahedron.

In conclusion, by performing first-principles and effective model calculations, we find that the superconducting pairing symmetry of La3_3Ni2_2O7_7 very sensitively depends on Ni-ege_g crystal field splitting. A slight increase in Ni-ege_g crystal field splitting (<0.2< 0.2 eV) changes the leading superconducting pairing symmetry from dxyd_{xy} to s±s_{\pm}. When the DFT band structure is exactly reproduced by the downfolded model, we find that the pairing symmetry of La3_3Ni2_2O7_7 is dxyd_{xy}. We also reveal that the transition from dxyd_{xy} to s±s_{\pm} pairing symmetry is associated with an increase in inverse Fermi velocity and zone-center susceptibility. Finally we conjecture that such sensitive dependence of pairing symmetry on low-energy electronic structure may exist in other multi-orbital unconventional superconductors (e.g. La4_4Ni3_3O10_{10} [43, 47, 93, 94, 95, 96, 97, 98]) and thus care is required in the downfolding procedure when ones studies unconventional superconductivity in realistic materials.

Methods

First-Principles Calculations

We perform DFT [70, 71] calculations, as implemented in VASP [80, 81] with the projector augmented wave (PAW) method. We use generalized gradient approximation (GGA) with Perdew-Burke-Ernzerhof parametrization [99] for the exchange-correlation functional. We use an energy cutoff of 600 eV. We employ the conventional orthogonal cell of La3_3Ni2_2O7_7 (48-atom) to perform the atomic relaxation. The k-mesh is 16×16×416 \times16 \times4. The total energy convergence criterion is 10−710^{-7} eV. The force convergence criterion is 1 meV/Å. The pressure convergence criterion is 0.1 kbar. We find that under a pressure of 30 GPa, oxygen octahedra rotations in La3_3Ni2_2O7_7 are completely suppressed and La3_3Ni2_2O7_7 is stabilized in a tetragonal structure (space group I4/mmmI4/mmm, see Supplementary Note 13 for details). The DFT optimized lattice constants and internal atomic positions, as well as the experimental lattice constants are shown in Supplementary Note 1.

We note that the bilayer two-orbital model is intended to study a monoblock of La3Ni2O7\mathrm{La}_3\mathrm{Ni}_2\mathrm{O}_7 because the inter-block interaction is weak [4, 6, 7, 8, 10, 12, 15]. Thus we focus on a monoblock of La3Ni2O7\mathrm{La}_3\mathrm{Ni}_2\mathrm{O}_7 with its optimized atomic positions fixed and a vacuum of 30 Å thick inserted in the cell. The corresponding Brillouin zone is a two-dimensional simple square. We perform a self-consistent calculation with a k-mesh of 20×20×120 \times20 \times1 on this bulk-like “freestanding” La3Ni2O7\mathrm{La}_3\mathrm{Ni}_2\mathrm{O}_7 to make direct connections to the bilayer two-orbital model in literature [4, 6, 7, 8, 10, 12, 15] (see Figure 1(a) in the main text for the crystal structure of La3Ni2O7\mathrm{La}_3\mathrm{Ni}_2\mathrm{O}_7 under high pressure).

The Downfolded Models

The details of Wannier’s model and Luo’s model are shown below. We set the in-plane lattice constants a=ba=b to be 1 throughout. Both models can be compactly written as:

H^=H^0+H^int(1)\hat{H}=\hat{H}_0+\hat{H}_{\mathrm{int}} \tag*{(1)}

where

H^0=∑kσΨ^kσ†H0(k)Ψ^kσ(2)\hat{H}_0=\sum_{\mathbf{k}\sigma}\hat{\Psi}^{\dagger}_{\mathbf{k}\sigma}\mathcal{H}_0(\mathbf{k})\hat{\Psi}_{\mathbf{k}\sigma} \tag*{(2)}

and Ψ^kσ†=(c^kσ1z†,c^kσ1x†,c^kσ2z†,c^kσ2x†)\hat{\Psi}^{\dagger}_{\mathbf{k}\sigma}=(\hat{c}^{\dagger}_{\mathbf{k}\sigma1z},\hat{c}^{\dagger}_{\mathbf{k}\sigma1x},\hat{c}^{\dagger}_{\mathbf{k}\sigma2z},\hat{c}^{\dagger}_{\mathbf{k}\sigma2x}) in which c^kσν†\hat{c}^{\dagger}_{\mathbf{k}\sigma\nu} is the creation operator for orbital ν\nu with momentum k\mathbf{k} and spin σ\sigma. (1z,1x,2z,2x)(1z,1x,2z,2x) is a shorthand for the orbital basis:

∣Ni1-d3z2−r2⟩,∣Ni1-dx2−y2⟩,∣Ni2-d3z2−r2⟩,∣Ni2-dx2−y2⟩(3)\lvert\mathrm{Ni1}\text{-}d_{3z^2-r^2}\rangle,\lvert\mathrm{Ni1}\text{-}d_{x^2-y^2}\rangle,\lvert\mathrm{Ni2}\text{-}d_{3z^2-r^2}\rangle,\lvert\mathrm{Ni2}\text{-}d_{x^2-y^2}\rangle \tag*{(3)}

where 1 and 2 label different NiO2\mathrm{NiO}_2 layers.

For H^int\hat{H}_{\mathrm{int}}, we use the full Slater-Kanamori-type interaction for multi-orbital systems [83]:

H^int=U∑α,νn^αν↑n^αν↓+∑α,σσ′[(U′−JHδσσ′)(n^α1zσn^α1xσ′+n^α2zσn^α2xσ′)]−JH∑α(c^α1z↓†c^α1x↑†c^α1x↓c^α1z↑+c^α2z↓†c^α2x↑†c^α2x↓c^α2z↑+h.c.)−J′∑α(c^α1z↓†c^α1z↑†c^α1x↓c^α1x↑+c^α2z↓†c^α2z↑†c^α2x↓c^α2x↑+h.c.)(4)\begin{aligned} \hat{H}_{\mathrm{int}}={}&U\sum_{\alpha,\nu}\hat{n}_{\alpha\nu\uparrow}\hat{n}_{\alpha\nu\downarrow} +\sum_{\alpha,\sigma\sigma'}\left[(U'-J_H\delta_{\sigma\sigma'})(\hat{n}_{\alpha1z\sigma}\hat{n}_{\alpha1x\sigma'}+\hat{n}_{\alpha2z\sigma}\hat{n}_{\alpha2x\sigma'})\right]\\ &-J_H\sum_{\alpha}\left(\hat{c}^{\dagger}_{\alpha1z\downarrow}\hat{c}^{\dagger}_{\alpha1x\uparrow}\hat{c}_{\alpha1x\downarrow}\hat{c}_{\alpha1z\uparrow}+\hat{c}^{\dagger}_{\alpha2z\downarrow}\hat{c}^{\dagger}_{\alpha2x\uparrow}\hat{c}_{\alpha2x\downarrow}\hat{c}_{\alpha2z\uparrow}+\mathrm{h.c.}\right)\\ &-J'\sum_{\alpha}\left(\hat{c}^{\dagger}_{\alpha1z\downarrow}\hat{c}^{\dagger}_{\alpha1z\uparrow}\hat{c}_{\alpha1x\downarrow}\hat{c}_{\alpha1x\uparrow}+\hat{c}^{\dagger}_{\alpha2z\downarrow}\hat{c}^{\dagger}_{\alpha2z\uparrow}\hat{c}_{\alpha2x\downarrow}\hat{c}_{\alpha2x\uparrow}+\mathrm{h.c.}\right) \tag*{(4)} \end{aligned}

where α\alpha labels a site, ν\nu labels an atomic orbital (ν=1x,1z,2x,2z\nu=1x,1z,2x,2z) and σ\sigma is spin. ‘h.c.’ means hermitian conjugate. UU and U′U' are the intraorbital and interorbital interaction strengths. JHJ_H is the Hund’s coupling strength and J′J' is the pair-hopping strength. The first line of Eq. (4) is the density-density interactions. The second line is the spin-flip interaction. The third line is the pair-hopping interaction. Throughout our calculations, we set U′=U−2JHU^{\prime}=U-2J_{\mathrm{H}} and J′=JHJ^{\prime}=J_{\mathrm{H}}.

Wannier’s model is obtained by using MLWF [72, 73] to fit the DFT band structure of La3_3Ni2_2O7_7 under 30 GPa. H0W(k)\mathcal{H}^{\mathrm{W}}_{0}(\mathbf{k}) is written in the form of:

H0W(k)=∑RH0W(R)eik⋅R(5)\mathcal{H}^{\mathrm{W}}_{0}(\mathbf{k})=\sum_{\mathbf{R}}\mathcal{H}^{\mathrm{W}}_{0}(\mathbf{R})e^{i\mathbf{k}\cdot\mathbf{R}} \tag*{(5)}

We use a Γ\Gamma-centered k-mesh of 20×2020 \times20 to sample the two-dimensional Brillouin zone. Hence there are altogether 441 R\mathbf{R} and ∣Rx∣≤10\lvert R_x\rvert\le10 and ∣Ry∣≤10\lvert R_y\rvert\le10. The onsite and hopping matrix elements up to the third-nearest-neighbor H0W(R)\mathcal{H}^{\mathrm{W}}_{0}(\mathbf{R}) are given below (in the unit of eV):

H0W(0,0)=(+1.153+0.000−0.602+0.000+0.000+1.375+0.000+0.010−0.602+0.000+1.153+0.000+0.000+0.010+0.000+1.375)(6)\mathcal{H}^{\mathrm{W}}_{0}(0,0)= \begin{pmatrix} +1.153 & +0.000 & -0.602 & +0.000 \\ +0.000 & +1.375 & +0.000 & +0.010 \\ -0.602 & +0.000 & +1.153 & +0.000 \\ +0.000 & +0.010 & +0.000 & +1.375 \end{pmatrix} \tag*{(6)}
H0W(1,0)=(−0.155−0.248+0.030+0.031−0.248−0.487+0.031+0.000+0.030+0.031−0.155−0.248+0.031+0.000−0.248−0.487)(7)\mathcal{H}^{\mathrm{W}}_{0}(1,0)= \begin{pmatrix} -0.155 & -0.248 & +0.030 & +0.031 \\ -0.248 & -0.487 & +0.031 & +0.000 \\ +0.030 & +0.031 & -0.155 & -0.248 \\ +0.031 & +0.000 & -0.248 & -0.487 \end{pmatrix} \tag*{(7)}
H0W(1,1)=(−0.023+0.000+0.004+0.000+0.000+0.064+0.000+0.003+0.004+0.000−0.023+0.000+0.000+0.003+0.000+0.064)(8)\mathcal{H}^{\mathrm{W}}_{0}(1,1)= \begin{pmatrix} -0.023 & +0.000 & +0.004 & +0.000 \\ +0.000 & +0.064 & +0.000 & +0.003 \\ +0.004 & +0.000 & -0.023 & +0.000 \\ +0.000 & +0.003 & +0.000 & +0.064 \end{pmatrix} \tag*{(8)}
H0W(2,0)=(−0.011+0.022−0.003+0.003+0.022−0.064+0.003−0.004−0.003+0.003−0.011+0.022+0.003−0.004+0.022−0.064)(9)\mathcal{H}^{\mathrm{W}}_{0}(2,0)= \begin{pmatrix} -0.011 & +0.022 & -0.003 & +0.003 \\ +0.022 & -0.064 & +0.003 & -0.004 \\ -0.003 & +0.003 & -0.011 & +0.022 \\ +0.003 & -0.004 & +0.022 & -0.064 \end{pmatrix} \tag*{(9)}

The C4C_4 rotation symmetry in the tetragonal cell leads to :

H0W(1,0)=H0W(−1,0)=H0W(0,1)=H0W(0,−1)\mathcal{H}^{\mathrm{W}}_{0}(1,0)=\mathcal{H}^{\mathrm{W}}_{0}(-1,0)=\mathcal{H}^{\mathrm{W}}_{0}(0,1)=\mathcal{H}^{\mathrm{W}}_{0}(0,-1)
H0W(1,1)=H0W(1,−1)=H0W(−1,1)=H0W(−1,−1)\mathcal{H}^{\mathrm{W}}_{0}(1,1)=\mathcal{H}^{\mathrm{W}}_{0}(1,-1)=\mathcal{H}^{\mathrm{W}}_{0}(-1,1)=\mathcal{H}^{\mathrm{W}}_{0}(-1,-1)
H0W(2,0)=H0W(−2,0)=H0W(0,2)=H0W(0,−2)(10)\mathcal{H}^{\mathrm{W}}_{0}(2,0)=\mathcal{H}^{\mathrm{W}}_{0}(-2,0)=\mathcal{H}^{\mathrm{W}}_{0}(0,2)=\mathcal{H}^{\mathrm{W}}_{0}(0,-2) \tag*{(10)}

Other H0(R)\mathcal{H}_0(\mathbf{R}) can be found in our Supplementary File “wannier90_hr.dat”. Using the occupancy of 3 electrons per formula, we obtain that the Fermi level of Wannier’s model is 0.782 eV. We note that for the furthest distance R=(10,10)\mathbf{R}=(10,10), the magnitude of each matrix element of H0W(10,10)\mathcal{H}_0^{\mathrm{W}}(10,10) does not exceed 0.03 meV. This shows that Wannier’s model practically includes all nonzero hoppings.

Luo’s model is constructed in Ref. [4], which takes the form of:

H0L(k)=[H11(k)H12(k)H12(k)H22(k)](11)\mathcal{H}_0^{\mathrm{L}}(\mathbf{k}) = \begin{bmatrix} \mathcal{H}_{11}(\mathbf{k}) & \mathcal{H}_{12}(\mathbf{k}) \\ \mathcal{H}_{12}(\mathbf{k}) & \mathcal{H}_{22}(\mathbf{k}) \end{bmatrix} \tag*{(11)}

where

H11(k)=H22(k)=[Tkx/zVkVkTkx],H12(k)=[t⊥x/zVk′Vk′t⊥x](12)\mathcal{H}_{11}(\mathbf{k})=\mathcal{H}_{22}(\mathbf{k})= \begin{bmatrix} T_{\mathbf{k}}^{x/z} & V_{\mathbf{k}} \\ V_{\mathbf{k}} & T_{\mathbf{k}}^{x} \end{bmatrix}, \qquad \mathcal{H}_{12}(\mathbf{k})= \begin{bmatrix} t_{\perp}^{x/z} & V_{\mathbf{k}}' \\ V_{\mathbf{k}}' & t_{\perp}^{x} \end{bmatrix} \tag*{(12)}

with

Tkx/z=2t1x/z(cos⁡kx+cos⁡ky)+4t2x/zcos⁡kxcos⁡ky+ϵx/zT_{\mathbf{k}}^{x/z}=2t_1^{x/z}(\cos k_x+\cos k_y)+4t_2^{x/z}\cos k_x\cos k_y+\epsilon^{x/z}
{#eq:13}

V_{\mathbf{k}}=2t_3^{xz}(\cos k_x-\cos k_y),\qquad V_{\mathbf{k}}'=2t_4^{xz}(\cos k_x-\cos k_y).

AlltheparametersinLuo’smodelarefoundinxref‘TableI‘.All the parameters in Luo’s model are found in {xref}`Table I`.
t1xt_1^xt1zt_1^zt2xt_2^xt2zt_2^zt3xzt_3^{xz}
-0.483-0.1100.069-0.0170.239
t⊥xt_{\perp}^xt⊥zt_{\perp}^zt4xzt_4^{xz}ϵx\epsilon^xϵz\epsilon^z
0.005-0.635-0.0340.7760.409

Table I. The parameters of Luo’s model H0L(k)\mathcal{H}_0^{\mathrm{L}}(\mathbf{k}) [4]. The unit is eV.

Susceptibility

We define different types of susceptibility as follow. We set the Boltzmann constant kBk_B and the reduced Planck constant ℏ\hbar to be 1 throughout. For a multi-orbital non-interacting model H0(k)\mathcal{H}_0(\mathbf{k}), the bare susceptibility is defined as [78]:

χ0(q)l1l2,l3l4=−∑k[G0(k+q)]l1l3[G0(k)]l4l2(13)\chi_0(q)_{l_1l_2,l_3l_4} = -\sum_k [G_0(k+q)]_{l_1l_3}[G_0(k)]_{l_4l_2} \tag*{(13)}

where q=(q,Ωm)q=(\mathbf{q},\Omega_m) and k=(k,ωn)k=(\mathbf{k},\omega_n) are four-component vectors. Ωm=2mπT\Omega_m=2m\pi T is a bosonic frequency and ωn=(2n+1)πT\omega_n=(2n+1)\pi T is a fermionic frequency. TT is the temperature and the summation ∑k=TNk∑k,ωn\sum_k = \frac{T}{N_k}\sum_{\mathbf{k},\omega_n}. For non-interacting Green function G0G_0, the summation over ωn\omega_n can be analytically performed and we obtain:

\chi_0(\mathbf{q})_{l_1l_2,l_3l_4}=-\frac{1}{N_k}\sum_{\mathbf{k}}\sum_{n,n'}(a_{\mathbf{k}+\mathbf{q}})_{l_1n}(a_{\mathbf{k}+\mathbf{q}})^{*}_{l_3n}(a_{\mathbf{k}})_{l_4n'}(a_{\mathbf{k}})^{*}_{l_2n'}\frac{n_F[(\epsilon_{\mathbf{k}+\mathbf{q}})_n]-n_F[(\epsilon_{\mathbf{k})_{n'}]}{(\epsilon_{\mathbf{k}+\mathbf{q}})_n-i\Omega_m-(\epsilon_{\mathbf{k}})_{n'}}} \tag*{(14)}

where aka_{\mathbf{k}} is the transition matrix that diagonalizes H0(k)\mathcal{H}_0(\mathbf{k}). nFn_F is the Fermi-Dirac occupancy function and ϵ(k)n\epsilon(\mathbf{k})_n is the nn-th eigenvalue of H0(k)\mathcal{H}_0(\mathbf{k}). If Ωm=0\Omega_m=0, we get the static susceptibility χ0(q)=χ0(q,Ωm=0)\chi_0(\mathbf{q})=\chi_0(\mathbf{q},\Omega_m=0):

χ0(q)l1l2,l3l4=−1Nk∑k∑n,n′(ak+q)l1n(ak+q)l3n∗(ak)l4n′(ak)l2n′∗nF[(ϵk+q)n]−nF[(ϵk)n′](ϵk+q)n−(ϵk)n′(15)\chi_0(\mathbf{q})_{l_1l_2,l_3l_4}=-\frac{1}{N_k}\sum_{\mathbf{k}}\sum_{n,n'}(a_{\mathbf{k}+\mathbf{q}})_{l_1n}(a_{\mathbf{k}+\mathbf{q}})^{*}_{l_3n}(a_{\mathbf{k}})_{l_4n'}(a_{\mathbf{k}})^{*}_{l_2n'}\frac{n_F[(\epsilon_{\mathbf{k}+\mathbf{q}})_n]-n_F[(\epsilon_{\mathbf{k}})_{n'}]}{(\epsilon_{\mathbf{k}+\mathbf{q}})_n-(\epsilon_{\mathbf{k}})_{n'}} \tag*{(15)}

The above bare static susceptibility is a four-index tensor. The total bare static susceptibility is defined as:

χ0tot(q)=12∑l,l′[χ0(q)]ll,l′l′(16)\chi_0^{\mathrm{tot}}(\mathbf{q})=\frac{1}{2}\sum_{l,l'}[\chi_0(\mathbf{q})]_{ll,l'l'} \tag*{(16)}

The orbital projected bare static susceptibility is defined as:

χ0ν(q)=12∑l,l′=ν[χ0(q)]ll,l′l′(17)\chi_0^\nu(\mathbf{q})=\frac{1}{2}\sum_{l,l'=\nu}[\chi_0(\mathbf{q})]_{ll,l'l'} \tag*{(17)}

where ν\nu is a given orbital. In the bilayer two-orbital model, ν\nu is either d3z2−r2d_{3z^2-r^2} or dx2−y2d_{x^2-y^2}.

The static spin and charge susceptibilities within random-phase-approximation are defined in the next subsection.

Random Phase Approximation Calculations

The details of random-phase-approximation (RPA) [74, 75, 76, 77, 78, 79] method for the calculations of spin and charge susceptibilities are shown below. Based on charge and spin susceptibilities, we construct an effective pairing potential for subsequent linearized gap equation calculations.

Throughout this section, the tensor multiplication of two four-index tensors C=ABC=AB is defined as follows:

Cij,kl=(AB)ij,kl=∑m,nAij,mnBmn,kl.(18)C_{ij,kl}=(AB)_{ij,kl}=\sum_{m,n}A_{ij,mn}B_{mn,kl}. \tag*{(18)}

where both AA and BB are four-index tensors. The spin and charge susceptibilities within RPA method are defined as [75, 76, 79]:

χs(q)=[I−χ0(q)Us]−1χ0(q)(19)\chi_s(\mathbf{q})=[I-\chi_0(\mathbf{q})U_s]^{-1}\chi_0(\mathbf{q}) \tag*{(19)}
χc(q)=[I+χ0(q)Uc]−1χ0(q)(20)\chi_{c}(\mathbf{q}) = [I+\chi_{0}(\mathbf{q})U_{c}]^{-1}\chi_{0}(\mathbf{q}) \tag*{(20)}

where the UU-matrices take the form of [78]:

(Us)αl1,αl2,αl3,αl4={Uαl1=l2=l3=l4Uα′l1=l3≠l2=l4JHαl1=l2≠l3=l4Jα′l1=l4≠l2=l3(U_{s})_{\alpha l_{1},\alpha l_{2},\alpha l_{3},\alpha l_{4}} = \begin{cases} U_{\alpha} & l_{1}=l_{2}=l_{3}=l_{4}\\ U_{\alpha}^{\prime} & l_{1}=l_{3}\ne l_{2}=l_{4}\\ J_{H\alpha} & l_{1}=l_{2}\ne l_{3}=l_{4}\\ J_{\alpha}^{\prime} & l_{1}=l_{4}\ne l_{2}=l_{3} \end{cases}
(Uc)αl1,αl2,αl3,αl4={Uαl1=l2=l3=l4−Uα′+2JHαl1=l3≠l2=l42Uα′−JHαl1=l2≠l3=l4Jα′l1=l4≠l2=l3(21)(U_{c})_{\alpha l_{1},\alpha l_{2},\alpha l_{3},\alpha l_{4}} = \begin{cases} U_{\alpha} & l_{1}=l_{2}=l_{3}=l_{4}\\ -U_{\alpha}^{\prime}+2J_{H\alpha} & l_{1}=l_{3}\ne l_{2}=l_{4}\\ 2U_{\alpha}^{\prime}-J_{H\alpha} & l_{1}=l_{2}\ne l_{3}=l_{4}\\ J_{\alpha}^{\prime} & l_{1}=l_{4}\ne l_{2}=l_{3} \end{cases} \tag*{(21)}

where α\alpha labels a site and lil_i (i=1,2,3,4i=1,2,3,4) labels an orbital. All the other entries of UcU_c, UsU_s are zero. The derivation of Eq. (21) based on the Hamiltonian Eq. (4) is found Supplementary Note 4.

In this work, we choose Uα′=Uα−2JHαU_{\alpha}^{\prime}=U_{\alpha}-2J_{H\alpha} and Jα′=JHαJ_{\alpha}^{\prime}=J_{H\alpha}. For the results in the main text, we use JH=0.15UJ_H=0.15U for each Ni site [78]. We also test other JH/UJ_H/U ratios and find that the key results do not qualitatively change (see Supplementary Note 9).

The effective pairing potential in singlet (s) or triplet (t) channel is given by [76]

Vs/t(k,k′)=12[V(k,k′)±V(k,−k′)](22)V^{\mathrm{s/t}}(\mathbf{k},\mathbf{k}^{\prime})=\frac{1}{2}[V(\mathbf{k},\mathbf{k}^{\prime})\pm V(\mathbf{k},-\mathbf{k}^{\prime})] \tag*{(22)}

where

Vij,mn(k,k′)=∑l1l2l3l4(ak)l2m∗(a−k)l3n∗Re⁡[Vl1l2,l3l4(k,k′)](ak′)l1i(a−k′)l4j(23)V_{ij,mn}(\mathbf{k},\mathbf{k}^{\prime})=\sum_{l_{1}l_{2}l_{3}l_{4}}(a_{\mathbf{k}})_{l_{2}m}^{*}(a_{-\mathbf{k}})_{l_{3}n}^{*}\operatorname{Re}[\mathcal{V}_{l_{1}l_{2},l_{3}l_{4}}(\mathbf{k},\mathbf{k}^{\prime})](a_{\mathbf{k}^{\prime}})_{l_{1}i}(a_{-\mathbf{k}^{\prime}})_{l_{4}j} \tag*{(23)}

with

Vl1l2,l3l4(k,k′)={12[Us+Uc+3Usχs(k−k′)Us−Ucχc(k−k′)Uc]l1l2,l3l4for singlet−12[Usχs(k−k′)Us+Ucχc(k−k′)Uc]l1l2,l3l4for triplet(24)\mathcal{V}_{l_{1}l_{2},l_{3}l_{4}}(\mathbf{k},\mathbf{k}^{\prime})= \begin{cases} \frac{1}{2}[U_s+U_c+3U_s\chi_s(\mathbf{k}-\mathbf{k}^{\prime})U_s-U_c\chi_c(\mathbf{k}-\mathbf{k}^{\prime})U_c]_{l_{1}l_{2},l_{3}l_{4}} & \text{for singlet}\\ -\frac{1}{2}[U_s\chi_s(\mathbf{k}-\mathbf{k}^{\prime})U_s+U_c\chi_c(\mathbf{k}-\mathbf{k}^{\prime})U_c]_{l_{1}l_{2},l_{3}l_{4}} & \text{for triplet} \end{cases} \tag*{(24)}

We note that in Eq. (23) (l1l2l_{1}l_{2}, l3l4l_{3}l_{4}) are labels of atomic orbitals and (ijij, mnmn) are labels of band indices.

Linearized Gap Equation

We show the details of linearized gap equation calculations below. We focus on a two-dimensional Brillouin zone (which is relevant to the bilayer two-orbital model in this work).

The linearized gap equation for multi-orbital system is:

Δi(k)=−∑j∫FSjdk′∣vj(k′)∣Γij(k,k′)Δj(k′)(25)\Delta_i(\mathbf{k})=-\sum_j\int_{FS_j}\frac{d\mathbf{k}'}{\lvert\mathbf{v}_j(\mathbf{k}')\rvert}\Gamma_{ij}(\mathbf{k},\mathbf{k}')\Delta_j(\mathbf{k}') \tag*{(25)}

where FSjFS_j means that the integration is performed on the jj-th Fermi surface sheet. vj(k′)\mathbf{v}_j(\mathbf{k}') is the Fermi velocity at the momentum k\mathbf{k} of jj-th Fermi surface sheet. The effective pairing potential Γij(k,k′)=Vjj,iis/t(k,k′)\Gamma_{ij}(\mathbf{k},\mathbf{k}')=V^{s/t}_{jj,ii}(\mathbf{k},\mathbf{k}') (see Eqs. (22)–(24)). Δi(k)\Delta_i(\mathbf{k}) is the gap function of ii-th Fermi surface sheet. With a discrete k\mathbf{k}-mesh, Eq. (25) is reduced to an eigenvalue problem:

λΔi(k)=Mij(k,k′)Δj(k′)(26)\lambda\Delta_i(\mathbf{k})=M_{ij}(\mathbf{k},\mathbf{k}')\Delta_j(\mathbf{k}') \tag*{(26)}

where the matrix Mij(k,k′)M_{ij}(\mathbf{k},\mathbf{k}') is:

Mij(k,k′)=−1(2π)2lj(k′)∣vj(k′)∣Γij(k,k′)(27)M_{ij}(\mathbf{k},\mathbf{k}')=-\frac{1}{(2\pi)^2}\frac{l_j(\mathbf{k}')}{\lvert\mathbf{v}_j(\mathbf{k}')\rvert}\Gamma_{ij}(\mathbf{k},\mathbf{k}') \tag*{(27)}

and lj(k′)l_j(\mathbf{k}') is the length element at k′\mathbf{k}' grid of jj-th Fermi surface sheet and vj(k′)\mathbf{v}_j(\mathbf{k}') is the Fermi velocity of the jj-th energy band at k′\mathbf{k}'. The calculation of lj(k′)l_j(\mathbf{k}') is shown in Supplementary Note 7. All momenta k\mathbf{k} are constrained close to the Fermi surface such that ∣ϵk−μ∣≤ξc\lvert\epsilon_{\mathbf{k}}-\mu\rvert\leq\xi_c where μ\mu is the chemical potential and ξc\xi_c is an energy cutoff. λ\lambda is the eigenvalue, which is real. The largest eigenvalue corresponds to the leading superconducting instability.

In this work, for the linearized gap equation calculations, we use a k\mathbf{k}-mesh of 200×200×1200\times200\times1 to sample the 2D square Brillouin zone and ξc=1\xi_c=1 meV. The temperature TT is set at 116 K. In Wannier’s model, we find 1576 k\mathbf{k} points on the Fermi surface. In Luo’s model, we find 1436 k\mathbf{k} points on the Fermi surface.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request. Source data are provided with this paper.

Code availability

The electronic structure calculations were performed using the proprietary code VASP [81] and the open-source code Wannier90 [73]. Wannier90 is freely distributed on academic use under the Massachusetts Institute of Technology (MIT) License.

Acknowledgments

We are grateful to Steffen Bötzel, Frank Lechermann, Xianxin Wu, Fan Yang, Yi-feng Yang and Dao-Xin Yao for useful discussions. This project was financially supported by the National Natural Science Foundation of China under project number 12374064, the National Key R&D Program of China under project number 2021YFE0107900, and Science and Technology Commission of Shanghai Municipality under grant number 23ZR1445400. NYU High-Performance-Computing (HPC) provides computational resources.

Author contributions

H.C. conceived and supervised the project. C.X., H.L., S.Z. and H.C. performed the calculations. H.C., C.X. and H.L. wrote the manuscript. All the authors participated in the discussion.

Competing interests

The authors declare no competing financial and non-financial interests.

References

  1. [1]H. Sun, M. Huo, X. Hu, J. Li, Z. Liu, Y. Han, L. Tang, Z. Mao, P. Yang, B. Wang, J. Cheng, D.-X. Yao, G.-M. Zhang, and M. Wang, Signatures of superconductivity near 80 K in a nickelate under high pressure, Nature 621, 493 (2023).DOI
  2. [2]J. Hou, P.-T. Yang, Z.-Y. Liu, J.-Y. Li, P.-F. Shan, L. Ma, G. Wang, N.-N. Wang, H.-Z. Guo, J.-P. Sun, Y. Uwatoko, M. Wang, G.-M. Zhang, B.-S. Wang, and J.-G. Cheng, Emergence of High-Temperature Superconducting Phase in Pressurized La₃Ni₂O₇ Crystals, Chin. Phys. Lett. 40, 117302 (2023).
  3. [3]D. A. Shilenko and I. V. Leonov, Correlated electronic structure, orbital-selective behavior, and magnetic correlations in double-layer La₃Ni₂O₇ under pressure, Phys. Rev. B 108, 125105 (2023).DOI
  4. [4]Z. Luo, X. Hu, M. Wang, W. Wú, and D.-X. Yao, Bilayer Two-Orbital Model of La₃Ni₂O₇ under Pressure, Phys. Rev. Lett. 131, 126001 (2023).
  5. [5]Q. Qin and Y.-f. Yang, High-Tc superconductivity by mobilizing local spin singlets and possible route to higher Tc in pressurized La₃Ni₂O₇, Phys. Rev. B 108, L140504 (2023).
  6. [6]J. Huang, Z. D. Wang, and T. Zhou, Impurity and vortex states in the bilayer high-temperature superconductor La₃Ni₂O₇, Phys. Rev. B 108, 174501 (2023).
  7. [7]Y. Zhang, L.-F. Lin, A. Moreo, T. A. Maier, and E. Dagotto, Trends in electronic structures and s±-wave pairing for the rare-earth series in bilayer nickelate superconductor R₃Ni₂O₇, Phys. Rev. B 108, 165141 (2023).DOI
  8. [8]Y. Zhang, L.-F. Lin, A. Moreo, T. A. Maier, and E. Dagotto, Structural phase transition, s±-wave pairing, and magnetic stripe order in bilayered superconductor La₃Ni₂O₇ under pressure, Nat. Commun. 15, 2470 (2024).DOI
  9. [9]Y.-B. Liu, J.-W. Mei, F. Ye, W.-Q. Chen, and F. Yang, s±-Wave Pairing and the Destructive Role of Apical-Oxygen Deficiencies in La₃Ni₂O₇ under Pressure, Phys. Rev. Lett. 131, 236002 (2023).DOI
  10. [10]Q.-G. Yang, D. Wang, and Q.-H. Wang, Possible s±-wave superconductivity in La₃Ni₂O₇, Phys. Rev. B 108, L140505 (2023).
  11. [11]X.-Z. Qu, D.-W. Qu, J. Chen, C. Wu, F. Yang, W. Li, and G. Su, Bilayer t−J−J⊥ Model and Magnetically Mediated Pairing in the Pressurized Nickelate La₃Ni₂O₇, Phys. Rev. Lett. 132, 036502 (2024).DOI
  12. [12]Y. Gu, C. Le, Z. Yang, X. Wu, and J. Hu, Effective model and pairing tendency in bilayer Ni-based superconductor La₃Ni₂O₇, arXiv:2306.07275 (2023).DOI
  13. [13]Z. Fan, J.-F. Zhang, B. Zhan, D. Lv, X.-Y. Jiang, B. Normand, and T. Xiang, Superconductivity in nickelate and cuprate superconductors with strong bilayer coupling, Phys. Rev. B 110, 024514 (2024).DOI
  14. [14]Y.-f. Yang, G.-M. Zhang, and F.-C. Zhang, Interlayer valence bonds and two-component theory for high-Tc superconductivity of La₃Ni₂O₇ under pressure, Phys. Rev. B 108, L201108 (2023).
  15. [15]Z. Luo, B. Lv, M. Wang, W. Wú, and D.-X. Yao, High-Tc superconductivity in La₃Ni₂O₇ based on the bilayer two-orbital t–J model, npj Quantum Matter. 9, 61 (2024).DOI
  16. [16]G. Heier, K. Park, and S. Y. Savrasov, Competing dxy and s± pairing symmetries in superconducting La₃Ni₂O₇: LDA + FLEX calculations, Phys. Rev. B 109, 104508 (2024).
  17. [17]F. Lechermann, J. Gondolf, S. Bötzel, and I. M. Eremin, Electronic correlations and superconducting instability in La₃Ni₂O₇ under high pressure, Phys. Rev. B 108, L201121 (2023).
  18. [18]S. Bötzel, F. Lechermann, J. Gondolf, and I. M. Eremin, Theory of magnetic excitations in the multilayer nickelate superconductor La₃Ni₂O₇, Phys. Rev. B 109, L180502 (2024).
  19. [19]S. Jiang, Z. Wang, and F.-C. Zhang, High-Temperature Superconductivity in La₃Ni₂O₇, Chin. Phys. Lett. 41, 017402 (2024).
  20. [20]Y. Wang, K. Jiang, Z. Wang, F.-C. Zhang, and J. Hu, The electronic and magnetic structures of bilayer La₃Ni₂O₇ at ambient pressure, arXiv:2401.15097 (2024).
  21. [21]Z. Liao, L. Chen, G. Duan, Y. Wang, C. Liu, R. Yu, and Q. Si, Electron correlations and superconductivity in La₃Ni₂O₇ under pressure tuning, Phys. Rev. B 108, 214522 (2023).
  22. [22]H. Oh and Y.-H. Zhang, Type-II t–J model and shared superexchange coupling from Hund’s rule in superconducting La₃Ni₂O₇, Phys. Rev. B 108, 174511 (2023).
  23. [23]B. Geisler, J. J. Hamlin, G. R. Stewart, R. G. Hennig, and P. Hirschfeld, Structural transitions, octahedral rotations, and electronic properties of A₃Ni₂O₇ rare-earth nickelates under high pressure, npj Quantum Mater. 9, 38 (2024).DOI
  24. [24]H. Lange, L. Homeier, E. Demler, U. Schollwöck, F. Grusdt, and A. Bohrdt, Feshbach resonance in a strongly repulsive ladder of mixed dimensionality: A possible scenario for bilayer nickelate superconductors, Phys. Rev. B 109, 045127 (2024).
  25. [25]H. LaBollita, V. Pardo, M. R. Norman, and A. S. Botana, Electronic structure and magnetic properties of La₃Ni₂O₇ under pressure: active role of the Ni-dx²−y² orbitals, arXiv:2309.17279 (2023).
  26. [26]Y. Zhang, D. Su, Y. Huang, H. Sun, M. Huo, Z. Shan, K. Ye, Z. Yang, R. Li, M. Smidman, M. Wang, L. Jiao, and H. Yuan, High-temperature superconductivity with zero resistance and strange-metal behaviour in La₃Ni₂O₇−δ, Nat. Phys. 20, 1269 (2024).DOI
  27. [27]J. Yang, H. Sun, X. Hu, Y. Xie, T. Miao, H. Luo, H. Chen, B. Liang, W. Zhu, G. Qu, C.-Q. Chen, M. Huo, Y. Huang, S. Zhang, F. Zhang, F. Yang, Z. Wang, Q. Peng, H. Mao, G. Liu, Z. Xu, T. Qian, D.-X. Yao, M. Wang, L. Zhao, and X. J. Zhou, Orbital-dependent electron correlation in double-layer nickelate La₃Ni₂O₇, Nat. Commun. 15, 4373 (2024).DOI
  28. [28]Z. Pan, C. Lu, F. Yang, and C. Wu, Effect of Rare-earth Element Substitution in Superconducting R₃Ni₂O₇ Under Pressure, arXiv:2309.06173 (2023).
  29. [29]Y.-H. Tian, Y. Chen, J.-M. Wang, R.-Q. He, and Z.-Y. Lu, Correlation effects and concomitant two-orbital s±-wave superconductivity in La₃Ni₂O₇ under high pressure, Phys. Rev. B 109, 165154 (2024).DOI
  30. [30]Y. Zhang, L.-F. Lin, A. Moreo, T. A. Maier, and E. Dagotto, Electronic structure, magnetic correlations, and superconducting pairing in the reduced Ruddlesden-Popper bilayer La₃Ni₂O₆ under pressure: Different role of d₃z²−r² orbital compared with La₃Ni₂O₇, Phys. Rev. B 109, 045151 (2024).
  31. [31]S. Ryee, N. Witt, and T. O. Wehling, Quenched Pair Breaking by Interlayer Correlations as a Key to Superconductivity in La₃Ni₂O₇, Phys. Rev. Lett. 133, 096002 (2024).DOI
  32. [32]C. Lu, Z. Pan, F. Yang, and C. Wu, Interplay of two E_g orbitals in superconducting La₃Ni₂O₇ under pressure, Phys. Rev. B 110, 094509 (2024).
  33. [33]G. Wang, N. N. Wang, X. L. Shen, J. Hou, L. Ma, L. F. Shi, Z. A. Ren, Y. D. Gu, H. M. Ma, P. T. Yang, Z. Y. Liu, H. Z. Guo, J. P. Sun, G. M. Zhang, S. Calder, J.-Q. Yan, B. S. Wang, Y. Uwatoko, and J.-G. Cheng, Pressure-Induced Superconductivity In Polycrystalline La₃Ni₂O₇−δ, Phys. Rev. X 14, 011040 (2024).
  34. [34]H. Yang, H. Oh, and Y.-H. Zhang, Strong pairing from small Fermi surface beyond weak coupling: Application to La₃Ni₂O₇, arXiv:2309.15095 (2023).arxiv.org/abs/2309.15095
  35. [35]T. Kaneko, H. Sakakibara, M. Ochi, and K. Kuroki, Pair correlations in the two-orbital Hubbard ladder: Implications for superconductivity in the bilayer nickelate La₃Ni₂O₇, Phys. Rev. B 109, 045154 (2024).DOI
  36. [36]L. C. Rhodes and P. Wahl, Structural routes to stabilize superconducting La₃Ni₂O₇ at ambient pressure, Phys. Rev. Mater. 8, 044801 (2024).DOI
  37. [37]J.-X. Zhang, H.-K. Zhang, Y.-Z. You, and Z.-Y. Weng, Strong Pairing Originated from an Emergent Z₂ Berry Phase in La₃Ni₂O₇, Phys. Rev. Lett. 133, 126501 (2024).
  38. [38]C. Lu, Z. Pan, F. Yang, and C. Wu, Interlayer-Coupling-Driven High-Temperature Superconductivity in La₃Ni₂O₇ under Pressure, Phys. Rev. Lett. 132, 146002 (2024).
  39. [39]X. Chen, P. Jiang, J. Li, Z. Zhong, and Y. Lu, Critical charge and spin instabilities in superconducting La₃Ni₂O₇, arXiv:2307.07154 (2023).
  40. [40]Y. Cao and Y.-f. Yang, Flat bands promoted by Hund’s rule coupling in the candidate double-layer high-temperature superconductor La₃Ni₂O₇ under high pressure, Phys. Rev. B 109, L081105 (2024).DOI
  41. [41]V. Christiansson, F. Petocchi, and P. Werner, Correlated Electronic Structure of La₃Ni₂O₇ under Pressure, Phys. Rev. Lett. 131, 206501 (2023).
  42. [42]Y. Shen, M. Qin, and G.-M. Zhang, Effective Bi-layer Model Hamiltonian and Density-Matrix Renormalization Group Study for the High-Tc Superconductivity in La₃Ni₂O₇ under High Pressure, Chin. Phys. Lett. 40, 127401 (2023).
  43. [43]Q. Li, Y.-J. Zhang, Z.-N. Xiang, Y. Zhang, X. Zhu, and H.-H. Wen, Signature of Superconductivity in Pressurized La₄Ni₃O₁₀, Chin. Phys. Lett. 41, 017401 (2024).
  44. [44]J. Chen, F. Yang, and W. Li, Orbital-selective superconductivity in the pressurized bilayer nickelate La₃Ni₂O₇: An infinite projected entangled-pair state study, Phys. Rev. B 110, L041111 (2024).DOI
  45. [45]L. Wang, Y. Li, S.-Y. Xie, F. Liu, H. Sun, C. Huang, Y. Gao, T. Nakagawa, B. Fu, B. Dong, Z. Cao, R. Yu, S. I. Kawaguchi, H. Kadobayashi, M. Wang, C. Jin, H.-k. Mao, and H. Liu, Structure Responsible for the Superconducting State in La₃Ni₂O₇ at High-Pressure and Low-Temperature Conditions, J. Am. Chem. Soc. 146, 7506 (2024).
  46. [46]G. Wang, N. Wang, Y. Wang, L. Shi, X. Shen, J. Hou, H. Ma, P. Yang, Z. Liu, H. Zhang, X. Dong, J. Sun, B. Wang, K. Jiang, J. Hu, Y. Uwatoko, and J. Cheng, Observation of high-temperature superconductivity in the high-pressure tetragonal phase of La₂PrNi₂O₇−δ, arXiv:2311.08212 (2023).arxiv.org/abs/2311.08212
  47. [47]Y. Zhu, D. Peng, E. Zhang, B. Pan, X. Chen, L. Chen, H. Ren, F. Liu, Y. Hao, N. Li, Z. Xing, F. Lan, J. Han, J. Wang, D. Jia, H. Wo, Y. Gu, Y. Gu, L. Ji, W. Wang, H. Gou, Y. Shen, T. Ying, X. Chen, W. Yang, H. Cao, C. Zheng, Q. Zeng, J. Guo, and J. Zhao, Superconductivity in pressurized trilayer La₄Ni₃O₁₀−δ single crystals, Nature 631, 531 (2024).DOI
  48. [48]H. Schlömer, U. Schollwöck, F. Grusdt, and A. Bohrdt, Superconductivity in the pressurized nickelate La₃Ni₂O₇ in the vicinity of a BEC-BCS crossover, arXiv:2311.03349 (2023).arxiv.org/abs/2311.03349
  49. [49]M. Zhang, C. Pei, Q. Wang, Y. Zhao, C. Li, W. Cao, S. Zhu, J. Wu, and Y. Qi, Effects of pressure and doping on Ruddlesden-Popper phases Laₙ₊₁NiₙO₃ₙ₊₁, J. Mater. Sci. Technol. 185, 147 (2024).
  50. [50]D.-C. Lu, M. Li, Z.-Y. Zeng, W. Hou, J. Wang, F. Yang, and Y.-Z. You, Superconductivity from Doping Symmetric Mass Generation Insulators: Application to La₃Ni₂O₇ under Pressure, arXiv:2308.11195 (2023).DOI
  51. [51]W. Wú, Z. Luo, D.-X. Yao, and M. Wang, Superexchange and charge transfer in the nickelate superconductor La₃Ni₂O₇ under pressure, Sci. China Phys. Mech. Astron. 67, 117402 (2024).DOI
  52. [52]Z. Liu, M. Huo, J. Li, Q. Li, Y. Liu, Y. Dai, X. Zhou, J. Hao, Y. Lu, M. Wang, and H.-H. Wen, Electronic correlations and partial gap in the bilayer nickelate La₃Ni₂O₇, Nat. Commun. 15, 7570 (2024).DOI
  53. [53]Y. Zhang, L.-F. Lin, A. Moreo, and E. Dagotto, Electronic structure, dimer physics, orbital-selective behavior, and magnetic tendencies in the bilayer nickelate superconductor La₃Ni₂O₇ under pressure, Phys. Rev. B 108, L180510 (2023).DOI
  54. [54]J. Li, P. Ma, H. Zhang, X. Huang, C. Huang, M. Huo, D. Hu, Z. Dong, C. He, J. Liao, X. Chen, T. Xie, H. Sun, and M. Wang, Pressure-driven right-triangle shape superconductivity in bilayer nickelate La₃Ni₂O₇, arxiv:2404.11369 (2024).
  55. [55]D. Li, K. Lee, B. Y. Wang, M. Osada, S. Crossley, H. R. Lee, Y. Cui, Y. Hikita, and H. Y. Hwang, Superconductivity in an infinite-layer nickelate, Nature 572, 624 (2019).DOI
  56. [56]D. Li, B. Y. Wang, K. Lee, S. P. Harvey, M. Osada, B. H. Goodge, L. F. Kourkoutis, and H. Y. Hwang, Superconducting Dome in Nd₁₋ₓSrₓNiO₂ Infinite Layer Films, Phys. Rev. Lett. 125, 027001 (2020).
  57. [57]S. Zeng, C. S. Tang, X. Yin, C. Li, M. Li, Z. Huang, J. Hu, W. Liu, G. J. Omar, H. Jani, Z. S. Lim, K. Han, D. Wan, P. Yang, S. J. Pennycook, A. T. S. Wee, and A. Ariando, Phase Diagram and Superconducting Dome of Infinite-Layer Nd₁₋ₓSrₓNiO₂ Thin Films, Phys. Rev. Lett. 125, 147003 (2020).
  58. [58]Q. Gu, Y. Li, S. Wan, H. Li, W. Guo, H. Yang, Q. Li, X. Zhu, X. Pan, Y. Nie, and H.-H. Wen, Single particle tunneling spectrum of superconducting Nd₁₋ₓSrₓNiO₂ thin films, Nat. Commun. 11, 6027 (2020).DOI
  59. [59]Y. Gu, S. Zhu, X. Wang, J. Hu, and H. Chen, A substantial hybridization between correlated Ni-d orbital and itinerant electrons in infinite-layer nickelates, Commun. Phys. 3, 84 (2020).DOI
  60. [60]S. Zeng, C. Li, L. E. Chow, Y. Cao, Z. Zhang, C. S. Tang, X. Yin, Z. S. Lim, J. Hu, P. Yang, and A. Ariando, Superconductivity in infinite-layer nickelate La₁₋ₓCaₓNiO₂ thin films, Sci. Adv. 8, eabI9927 (2022).DOI
  61. [61]M. Osada, B. Y. Wang, B. H. Goodge, S. P. Harvey, K. Lee, D. Li, L. F. Kourkoutis, and H. Y. Hwang, Nickelate Superconductivity without Rare-Earth Magnetism: (La, Sr)NiO₂, Adv. Mater. 33, 2104083 (2021).DOI
  62. [62]C. Xia, J. Wu, Y. Chen, and H. Chen, Dynamical structural instability and its implications for the physical properties of infinite-layer nickelates, Phys. Rev. B 105, 115134 (2022).
  63. [63]H. Chen, Y.-f. Yang, G.-M. Zhang, and H. Liu, An electronic origin of charge order in infinite-layer nickelates, Nat. Commun. 14, 5477 (2023).
  64. [64]C. C. Tsuei and J. R. Kirtley, Pairing symmetry in cuprate superconductors, Rev. Mod. Phys. 72, 969 (2000).
  65. [65]B. Keimer, S. A. Kivelson, M. R. Norman, S. Uchida, and J. Zaanen, From quantum matter to high-temperature superconductivity in copper oxides, Nature 518, 179 (2015).
  66. [66]E. Gull, O. Parcollet, and A. J. Millis, Superconductivity and the Pseudogap in the Two-Dimensional Hubbard Model, Phys. Rev. Lett. 110, 216405 (2013).
  67. [67]A. V. Chubukov, D. V. Efremov, and I. Eremin, Magnetism, superconductivity, and pairing symmetry in iron-based superconductors, Phys. Rev. B 78, 134512 (2008).
  68. [68]P. J. Hirschfeld, M. M. Korshunov, and I. I. Mazin, Gap symmetry and structure of Fe-based superconductors, Rep. Prog. Phys. 74, 124508 (2011).
  69. [69]G. R. Stewart, Superconductivity in iron compounds, Rev. Mod. Phys. 83, 1589 (2011).
  70. [70]P. Hohenberg and W. Kohn, Inhomogeneous Electron Gas, Phys. Rev. 136, B864 (1964).
  71. [71]W. Kohn and L. J. Sham, Self-Consistent Equations Including Exchange and Correlation Effects, Phys. Rev. 140, A1133 (1965).
  72. [72]N. Marzari, A. A. Mostofi, J. R. Yates, I. Souza, and D. Vanderbilt, Maximally localized Wannier functions: Theory and applications, Rev. Mod. Phys. 84, 1419 (2012).
  73. [73]A. A. Mostofi, J. R. Yates, Y.-S. Lee, I. Souza, D. Vanderbilt, and N. Marzari, Wannier90: A tool for obtaining maximally-localised Wannier functions, Comput. Phys. Commun. 178, 685 (2008).
  74. [74]J. E. Hirsch, Two-dimensional Hubbard model: Numerical simulation study, Phys. Rev. B 31, 4403 (1985).
  75. [75]Y. Yanase, T. Jujo, T. Nomura, H. Ikeda, T. Hotta, and K. Yamada, Theory of superconductivity in strongly correlated electron systems, Phys. Rep. 387, 1 (2003).
  76. [76]S. Graser, T. A. Maier, P. J. Hirschfeld, and D. J. Scalapino, Near-degeneracy of several pairing channels in multiorbital models for the fe pnictides, New J. Phys. 11, 025016 (2009).DOI
  77. [77]D. J. Scalapino, A common thread: The pairing interaction for unconventional superconductors, Rev. Mod. Phys. 84, 1383 (2012).DOI
  78. [78]X. Wu, D. Di Sante, T. Schwemmer, W. Hanke, H. Y. Hwang, S. Raghu, and R. Thomale, Robust dₓ²₋ᵧ²-wave superconductivity of infinite-layer nickelates, Phys. Rev. B 101, 060504 (2020).DOI
  79. [79]A. T. Rømer, D. D. Scherer, I. M. Eremin, P. J. Hirschfeld, and B. M. Andersen, Knight Shift and Leading Superconducting Instability from Spin Fluctuations in Sr₂RuO₄, Phys. Rev. Lett. 123, 247001 (2019).DOI
  80. [80]M. C. Payne, M. P. Teter, D. C. Allan, T. A. Arias, and J. D. Joannopoulos, Iterative minimization techniques for ab initio total-energy calculations: molecular dynamics and conjugate gradients, Rev. Mod. Phys. 64, 1045 (1992).DOI
  81. [81]G. Kresse and J. Furthmüller, Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set, Phys. Rev. B 54, 11169 (1996).
  82. [82]H. Chen, A. Hampel, J. Karp, F. Lechermann, and A. J. Millis, Dynamical Mean Field Studies of Infinite Layer Nickelates: Physics Results and Methodological Implications, Front. Phys. 10, 835942 (2022).DOI
  83. [83]J. Kanamori, Electron Correlation and Ferromagnetism of Transition Metals, Prog. Theor. Phys. 30, 275 (1963).
  84. [84]D. J. Scalapino, The case for dₓ²₋ᵧ² pairing in the cuprate superconductors, Phys. Rep. 250, 329 (1995).DOI
  85. [85]A. Chubukov, Pairing Mechanism in Fe-Based Superconductors, Annu. Rev. Condens. Matter Phys. 3, 57 (2012).
  86. [86]Z. Du, X. Yang, D. Altenfeld, Q. Gu, H. Yang, I. Eremin, P. J. Hirschfeld, I. I. Mazin, H. Lin, X. Zhu, and H.-H. Wen, Sign reversal of the order parameter in (Li₁₋ₓFeₓ)OHFe₁₋ᵧZnᵧSe, Nat. Phys. 14, 134 (2018).DOI
  87. [87]T. Hanaguri, S. Niitaka, K. Kuroki, and H. Takagi, Unconventional s-Wave Superconductivity in Fe(Se, Te), Science 328, 474 (2010).DOI
  88. [88]I. I. Mazin, D. J. Singh, M. D. Johannes, and M. H. Du, Unconventional Superconductivity with a Sign Reversal in the Order Parameter of LaFeAsO₁₋ₓFₓ, Phys. Rev. Lett. 101, 057003 (2008).
  89. [89]N. Bulut, D. J. Scalapino, and S. R. White, Comparison of Monte Carlo and diagrammatic calculations for the two-dimensional Hubbard model, Phys. Rev. B 47, 2742 (1993).
  90. [90]L. Chen, C. Bourbonnais, T. Li, and A.-M. S. Tremblay, Magnetic properties of the two-dimensional Hubbard model, Phys. Rev. Lett. 66, 369 (1991).
  91. [91]I. Ardizzone, M. Zingl, J. Teyssier, H. U. R. Strand, O. Peil, J. Fowlie, A. B. Georgeascu, S. Catalano, N. Bachar, A. B. Kuzmenko, M. Gibert, J.-M. Triscone, A. Georges, and D. van der Marel, Optical properties of LaNiO₃ films tuned from compressive to tensile strain, Phys. Rev. B 102, 155148 (2020).DOI
  92. [92]J. Hwang, Z. Feng, N. Charles, X. R. Wang, D. Lee, K. A. Stoerzinger, S. Muy, R. R. Rao, D. Lee, R. Jacobs, D. Morgan, and Y. Shao-Horn, Tuning perovskite oxides by strain: Electronic structure, properties, and functions in (electro)catalysis and ferroelectricity, Mater. Today 31, 100 (2019).DOI
  93. [93]H. Sakakibara, M. Ochi, H. Nagata, Y. Ueki, H. Sakurai, R. Matsumoto, K. Terashima, K. Hirose, H. Ohta, M. Kato, Y. Takano, and K. Kuroki, Theoretical analysis on the possibility of superconductivity in the trilayer Ruddlesden-Popper nickelate La₄Ni₃O₁₀ under pressure and its experimental examination: Comparison with La₃Ni₂O₇, Phys. Rev. B 109, 144511 (2024).DOI
  94. [94]J. Li, C.-Q. Chen, C. Huang, Y. Han, M. Huo, X. Huang, P. Ma, Z. Qiu, J. Chen, X. Hu, L. Chen, T. Xie, B. Shen, H. Sun, D.-X. Yao, and M. Wang, Structural transition, electric transport, and electronic structures in the compressed trilayer nickelate La₄Ni₃O₁₀, Sci. China Phys. Mech. Astron. 67, 117403 (2024).DOI
  95. [95]M. Zhang, H. Sun, Y.-B. Liu, Q. Liu, W.-Q. Chen, and F. Yang, s±-wave superconductivity in pressurized La₄Ni₃O₁₀, Phys. Rev. B 110, L180501 (2024).
  96. [96]C.-Q. Chen, Z. Luo, M. Wang, W. Wú, and D.-X. Yao, Trilayer multiorbital models of La₄Ni₃O₁₀, Phys. Rev. B 110, 014503 (2024).
  97. [97]Y. Zhang, L.-F. Lin, A. Moreo, T. A. Maier, and E. Dagotto, Prediction of s±-Wave Superconductivity Enhanced by Electronic Doping in Trilayer Nickelates La₄Ni₃O₁₀ under Pressure, Phys. Rev. Lett. 133, 136001 (2024).
  98. [98]M. Zhang, C. Pei, X. Du, W. Hu, Y. Cao, Q. Wang, J. Wu, Y. Li, H. Liu, C. Wen, Y. Zhao, C. Li, W. Cao, S. Zhu, Q. Zhang, N. Yu, P. Cheng, L. Zhang, Z. Li, J. Zhao, Y. Chen, H. Guo, C. Wu, F. Yang, S. Yan, L. Yang, and Y. Qi, Superconductivity in trilayer nickelate La₄Ni₃O₁₀ under pressure, arXiv:2311.07423 (2024).DOI
  99. [99]J. P. Perdew, K. Burke, and M. Ernzerhof, Generalized Gradient Approximation Made Simple, Phys. Rev. Lett. 77, 3865 (1996).

Paper details

vlnc:W_rzg7mktj
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents