betaFeedback

WorksPhysics and Astronomy

Strain-Tunable High-TcT_c Superconductivity in Layered R3m B2_2CN

Abstract

Using DFPT and Wannier-interpolated matrix elements in EPW, we find that isotropic tensile strain of 0–6% narrows the B:2pp/C:2pp bands at the Fermi level while softening the phonon branches most strongly coupled to them, driving the coupling constant λλ from 0.967 to 2.620. All four strained configurations remain mechanically and dynamically stable, satisfying the Born–Huang criteria with no imaginary phonon modes at any strain level. The fully anisotropic Migdal–Eliashberg equations give a TcT_c that rises monotonically from 53.5~K to 84.5~K across the strain series, crossing the liquid-nitrogen boiling point at 6% strain. The SISSO-based machine-learning model in EPW agrees with the anisotropic Migdal–Eliashberg TcT_c values to within 10~K across the strain series.

Introduction

Conventional phonon-mediated superconductors have a key advantage over unconventional ones. Their properties can be described quantitatively from first principles [1, 2]. Cuprates and iron-based compounds reach high transition temperatures, but their pairing mechanisms remain questioned [3, 4]. Bardeen-Cooper-Schrieffer (BCS) theory [5, 6] and its strong-coupling Migdal-Eliashberg extension [7, 8] describe phonon-mediated materials with no such ambiguity. Advances in density-functional perturbation theory (DFPT) [9] and Wannier-based electron-phonon interpolation [10] have made this theoretical framework more predictive. Computation-driven discovery of new high critical temperatures (TcT_c) phonon-mediated superconductors is now a realistic goal [11]. High TcT_c in conventional superconductors tends to arise from a combination of covalent bonding and light constituent elements [12, 13]. Light atoms push phonon frequencies up, while covalent states near the Fermi level support strong electron-phonon matrix elements, a principle articulated early by Ashcroft [14, 15]. MgB2_2 is the closest ambient-pressure case, with Tc=39T_c = 39 K driven by coupling between boron σ\sigma-bands and bond-stretching phonon modes [16]. More recently, hydrogen-rich clathrates under pressure extend the logic further: H3_3S reaches Tc≈203T_c \approx203 K [17] and LaH10_{10} reaches Tc≈250T_c \approx250 K [18], alongside related hydride phases reaching comparable or higher TcT_c [19]. CaH6_6 independently reaches Tc≈210T_c \approx210–215 K [20, 21]. However megabar pressures are required, and that constraint has pushed attention back toward ambient-pressure architectures built from light elements. Recent high-throughput searches have begun mapping how far conventional superconductivity can reach at ambient pressure [22, 23].

Ternary boron-carbon-nitrogen compounds have both ingredients that drive high TcT_c in conventional superconductors: light elements and strong covalent bonding [24].

B–C–N materials are exceptionally stiff, carry large Debye temperatures, and cover electronic behavior from insulating cubic BN to semimetallic graphite networks [25, 26, 27, 28, 29]. Diamond-like P3m1 B2_2CX (X=N,P) phases have been predicted to reach TcT_c near 46 K through coupling of metallic σ\sigma electrons to B-C stretching modes [30]. BC5_5, the diamond-like boron-carbide analog, was predicted computationally [31] and subsequently synthesized [32]. Hard BN clathrates have been predicted to superconduct [33] and H or Li-doped cubic BN was recently predicted to reach TcT_c above 120 K at ambient pressure [34]. More recently, anisotropic Migdal-Eliashberg calculations on the related superhard ternaries B2_2C3_3N and B4_4C5_5N3_3 found Tc≈20T_c \approx20–40 K using the same EPW-based methodology adopted here [35]. Layered metal borocarbides have separately been proposed to reach TcT_c up to 73 K via anisotropic Migdal-Eliashberg analysis [36]. Despite the growing interest in the B–C–N family, R3m B2_2CN has received comparatively little attention as a superconducting candidate.

R3m B2_2CN is metallic, with Fermi-level states dominated by B and C p-orbitals a layered quasi-2D covalent framework [37]. That combination of light atoms, stiff bonds, and metallic p-states is what the Ashcroft-MgB2_2 model [15, 16] points toward for strong electron-phonon coupling. Stiff covalent bonding raises phonon frequencies but lowers N(EF)N(E_F) through highly dispersive bands [38]. Lattice expansion offers a route to enhance electron-phonon coupling without changing chemical composition [39, 40, 41, 42]. In light-element covalent compounds, the mechanism is especially direct, since phonon softening and band narrowing reinforce each other. R3m B2_2CN satisfies both conditions, yet the strain dependence of its superconductivity has not been characterized. Reaching TcT_c above 77 K at ambient pressure is also practically relevant, since liquid-nitrogen cooling is far cheaper and simpler than the liquid-helium or megabar-pressure requirements of other high TcT_c routes [17, 18]. In this work, we report a first-principles study of R3m B2_2CN under isotropic tensile strains of 0%, 2%, 4%, and 6%. We examine structural stability, electronic structure, lattice dynamics, and superconducting properties. The motivation of this work is to use isotropic tensile strain as a viable tuning parameter to enhance the superconducting properties of R3m B2CN.

Computational methods

Our density functional theory (DFT) [43, 44] calculations were performed using the Quantum ESPRESSO (QE) [45, 46, 47] package. Electron-ion interactions were described using ultrasoft pseudopotentials (USPP) [48]. Exchange-correlation effects were treated with the Perdew-Burke-Ernzerhof (PBE) functional under the generalized gradient approximation (GGA) [49, 50]. Kohn-Sham wavefunctions were expanded in a planewave basis at a kinetic energy cutoff of 60 Ry, with the charge density cutoff set to 480 Ry. A Gaussian broadening scheme [51] with a smearing width of σ=0.02\sigma= 0.02 Ry was applied to handle the metallic occupancies of B2_2CN. NSCF calculations used a Γ\Gamma-centered 24×24×224 \times24 \times2 Monkhorst-Pack mesh [52]. Total and atom-projected densities of states used the tetrahedron [53] method on this kk-mesh. Lattice dynamics were computed using DFPT as implemented in ph.x. Dynamical matrices were calculated on a 6×6×26 \times6 \times2 qq-mesh. These meshes reflect the quasi-two-dimensional geometry of the layered structure. Band structures were computed along Γ\Gamma–T–H2_2|H0_0–L–Γ\Gamma–S0_0|–S2_2–F–Γ\Gamma of the hexagonal Brillouin zone from a non-self-consistent calculation on the converged charge density. Fermi surfaces were visualized using FermiSurfer [54].

To examine the effect of strain on the superconducting properties, isotropic tensile strain was applied to the equilibrium R3m structure by uniformly scaling the lattice parameters while preserving the crystal symmetry. Fixing the internal coordinates isolates the volumetric contribution to the electron-phonon response from any competing ionic relaxation. The mechanical and dynamical stability of each strained configuration was confirmed independently via Murnaghan equation of state [55, 56] and phonon calculations. The equation of state (EOS) and elastic tensor for each strained configuration were computed using the thermo_pw package [57, 58]. For the EOS, total energies at a set of uniformly scaled volumes were fitted to the Murnaghan equation of state to extract the equilibrium volume, bulk modulus, and its pressure derivative.

The elastic tensors were computed for each strained configuration using the stress-strain approach as implemented in thermo_pw. Mechanical stability was checked against the Born-Huang criteria [59]. Various elastic parameters were calculated using ELATE: an elastic tensor analysis tool [60]. The Debye temperature ΘD\Theta_{\mathrm{D}} was calculated from the VRH-averaged [61] sound velocities via Anderson’s procedure [62].

Electron-phonon matrix elements and superconducting properties were computed with the EPW code [63, 64, 65]. EPW combines maximally localized Wannier functions (MLWFs) [66, 67] with Wannier interpolation to access dense Brillouin zone grids at modest computational cost. MLWFs were constructed via Wannier90 [68] using initial projections onto B-2p, C-2p, and N-2p orbitals. These projections capture the dominant orbital character at the Fermi level (EFE_{\mathrm{F}}) in R3m B2_2CN and give well-localized functions with small spread. NSCF calculations for the Wannierization step used a Γ\Gamma-centered 24×24×424 \times24 \times4 kk-mesh. Coupling matrix elements were calculated on a coarse 12×12×212 \times12 \times2 kk-mesh and 6×6×26 \times6 \times2 qq-mesh, then interpolated onto fine 24×24×824 \times24 \times8 and 12×12×412 \times12 \times4 qq grids. The Coulomb pseudopotential [69] was set to μ∗=0.10\mu^{*} = 0.10, consistent with values used for covalent light-element systems such as MgB2_2 [70, 71]. The Eliashberg spectral function [72] α2F(ω)\alpha^{2}F(\omega) and cumulative coupling [73] λ(ω)\lambda(\omega) were also extracted at this fine grid. The Allen-Dynes temperature (TcADT_{\mathrm{c}}^{\mathrm{AD}}) was also evaluated from the interpolated α2F(ω)\alpha^{2}F(\omega) and logarithmic phonon frequency [74] ωlog⁡\omega_{\log} to provide an isotropic reference alongside the anisotropic result, following the formula of Allen and Dynes.

The anisotropic Migdal-Eliashberg equations were solved on the Matsubara frequency axis as implemented in EPW [75]. Unlike the Allen-Dynes formula, this approach retains the full kk-point dependence of the gap function Δnk(iωm)\Delta_{nk}(i\omega_{m}) and mass renormalization Znk(iωm)Z_{nk}(i\omega_{m}) and has been shown to give transition temperatures closer to experiment in systems with pronounced Fermi surface anisotropy. R3m B2_2CN falls in this category, with multiple bands crossing the Fermi level. The superconducting transition temperature (TcT_{\mathrm{c}}) was estimated from the temperature dependence of the anisotropic gap, taking the gap-closing point as TcT_{\mathrm{c}}. Gap distributions at each temperature were obtained from the kk-resolved values Δnk\Delta_{nk} on the fine mesh. The quasiparticle DOS and band-resolved coupling λnk\lambda_{nk} were computed from the same solution to characterize the pairing state across the strain series. As an independent check, TcT_{\mathrm{c}} was also estimated from the SISSO-based machine-learning model implemented in EPW, and the resulting TcMLT_{\mathrm{c}}^{\mathrm{ML}} values were compared against the Migdal-Eliashberg predictions across the strain series.

Crystal structure & stability

Crystal Structure and Strain Series

R3m B2_2CN is a rhombohedral layered structure (space group No. 160) with 12 atoms per unit cell – three formula units. B, C and N layers alternate along cc-axis, connected by strong covalent B–C and B–N bonds, with each atom in a threefold-coordinate planar environment (Fig. 1). The equilibrium lattice parameters are aa=2.586 Å and cc=13.177 Å, with c/a≈5.09c/a \approx5.09.

The Murnaghan EOS has the form,

P(V)=B0B0′[(VV0)−B0′−1](1)P(V) = \frac{B_{0}}{B'_{0}}\left[\left(\frac{V}{V_{0}}\right)^{-B'_{0}} - 1\right] \tag*{(1)}
Crystal structure of B$_2$CN

Figure 1. Crystal structure of B2_2CN

where V0V_0, B0B_0, and B0′B'_0 are the equilibrium volume, bulk modulus, and its pressure derivative. The fit gives an equilibrium volume of V0=76.40V_0 = 76.40 Å3^3 at 0% strain. The bulk modulus increases monotonically with strain - 320.0, 322.1, 330.5, and 347.5 GPa at 0%, 2%, 4%, and 6%. As bond lengths in the B–C–N network extend, the structure resists further volumetric compression more strongly. This is consistent with the behavior expected of a stiffening covalent framework.

Mechanical Stability

For the elastic constants, small symmetry-inequivalent strains were applied to the relaxed cell and the resulting stress tensors used to construct the full CijC_{ij} tensor. The R3m group yields six independent constants C11C_{11}, C12C_{12}, C13C_{13}, C33C_{33}, C44C_{44} and C66C_{66} – with C66=(C11−C12)/2C_{66} = (C_{11} - C_{12})/2 fixed by symmetry. The results for all four strain levels are collected in Table I.

Mechanical stability was checked against the Born-Huang criteria for trigonal crystals [59, 76], which require: C11>∣C12∣C_{11} > |C_{12}|, C44>0C_{44} > 0, (C11+C12)C33>2C132(C_{11} + C_{12})C_{33} > 2C_{13}^{2} and (C11−C12)C44>0(C_{11} - C_{12})C_{44} > 0. All four conditions are satisfied at every strain level. The constants are largely insensitive to strain below 6%: C11C_{11}, C33C_{33} and C44C_{44} each shifted by less than 0.2% across 0–4% range. At 6% strain all components soften slightly – C11C_{11} by 1.1% and C33C_{33} by 0.9% – but the Born-Huang conditions remain satisfied. Brillouin light scattering (BLS) can measure bulk and surface acoustic modes in thin-film and polycrystalline specimens [78] [77]. The computed CijC_{ij} values reported here can therefore be compared with future BLS measurements to experimentally validate the predicted mechanical response of B2_2CN under strain.

CijC_{ij} (GPa)ε=0%\varepsilon= 0\%ε=2%\varepsilon= 2\%ε=4%\varepsilon= 4\%ε=6%\varepsilon= 6\%
C11C_{11}567.97568.32567.12561.85
C12C_{12}225.39225.61224.86221.51
C13C_{13}111.56111.75111.09108.15
C33C_{33}851.44851.85850.44844.19
C44C_{44}292.97293.06292.72291.20
C66C_{66}171.29171.35171.13170.17
B0B_0320.0322.1330.5347.5
ΘD\Theta_D1555.01555.21554.41550.9
KVK_V320.49320.74319.86315.95
EVE_V602.83603.09602.19598.19
GVG_V254.04254.12252.83253.52
ν\nu0.18650.18660.18620.1844
KV/GVK_V/G_V1.261.261.261.26

Table 1. Elastic constants CijC_{ij} (GPa), Murnaghan bulk modulus B0B_0 (GPa), Debye temperature ΘD\Theta_D (K), Voigt bulk modulus KVK_V (GPa), Young’s modulus EVE_V (GPa), Shear modulus GVG_V (GPa), Poisson ratio ν\nu, and Pugh ratio KV/GVK_V/G_V for R3m B2_2CN at each strain level

The elastic moduli place B2_2CN firmly in the hard covalent ceramic class [79]. KV≈320K_V \approx320 GPa and EV≈600E_V \approx600 GPa shift by less than 1.5% across the full 0–6% strain. This indicates that the B–C–N network is not meaningfully disrupted at these strain levels. At ν≈0.186\nu\approx0.186 the Poisson ratio is lower than typical transition-metal nitrides and carbides (ν≈0.20\nu\approx0.20-0.30) [80] but sits above cubic BN (ν≈0.12\nu\approx0.12) [81] and diamond (ν≈0.148\nu\approx0.148) [82], consistent with the mixed ionic covalent character of the B–C–N framework. The Pugh ratio KV/GV=1.26K_V/G_V = 1.26 falls well below ductile-brittle threshold of 1.75 [83, 84] at all strain levels, indicating that the material is intrinsically brittle across the series. The Debye temperature ΘD\Theta_D is ≈1555\approx1555 K at 0% strain and falls by just 0.3% to 1551 K at 6%. This places B2_2CN well within empirical bound TC<ΘDT_C < \Theta_D proposed for phonon-mediated superconductors [85], later validated across the Materials Project database using machine-learned ΘD\Theta_D predictions [86]. For comparison, cubic BN has ΘD≈1700\Theta_D \approx1700 K [87] and MgB2≈750_2 \approx750 K [88]. The high value in B2_2CN follows from the light atomic masses and stiff B–C–N bonds.

Electronic structure

The electronic band structure and atom-projected density of states (PDOS) R3m B2_2CN at 0–6% isotropic tensile strain are shown in Figure 2. The k-path follows the standard rhombohedral Brillouin zone: Γ\Gamma–T–H2_2|H0_0–L–Γ\Gamma–S0_0|–S2_2–F–Γ\Gamma. The compound remains metallic across the entire strain range, with multiple bands crossing the Fermi level at all four strain levels. Near EFE_F the spectral weight is dominated by B:2pp and C:2pp states. N:2pp states lie largely below EFE_F, with a broad peak centered roughly 1–1.5 eV beneath it. These N:pp states form the occupied bonding network rather than contributing to metal-

Electronic band structure and density of states for R3m B2CN under isotropic tensile strain

Figure 2. Electronic band structure and DOS for R3mR3m B2CNB_2CN under isotropic tensile strain of (a) 0%, (b) 2%, (c) 4%, and (d) 6%.

lic conduction. This follows from the relatively higher electronegativity of nitrogen compared to boron and carbon [89]. The Fermi surface is therefore defined by B and C character at all strain levels considered.

The orbital character of the Fermi-level crossing bands remains unchanged across the strain series. The same B:p and C:p bands cross EFE_F at all strains from 0% to 6%. Fermi-level crossings fall in the same regions of the Brillouin zone, and the orbital character of the relevant bands is preserved. N:p bands remain predominantly occupied. No metal-to-insulator transition [90], Lifshitz transition [91, 92], or symmetry-lowering distortion is observed. R3m B₂CN is neither magnetically ordered nor symmetry-lowering. Isotropic tensile strain therefore acts as a clean perturbation, uniformly scaling the lattice without introducing additional complexity.

Bandwidth compression is the dominant electronic trend across the strain series [93]. As the lattice expands, orbital overlaps between neighboring B, C, and N sites weaken and effective hopping integrals decrease. The B:p and C:p bands near EFE_F become progressively less dispersive. In Figure 2, the energy range spanned by the Fermi-level bands at 6% is visibly narrower than at 0%, while the Fermi crossings themselves remain intact. The bands narrow gradually, with no discontinuity. This smooth trend matches the way strain steadily reduces orbital overlap.

Band narrowing at EFE_F concentrates states into a narrower energy window, so N(EF)N(E_F) rises monotonically with strain. The total DOS at EFE_F increases from 2.33 states eV−1^{-1} at 0% to 2.41, 2.52, and 2.63 states eV−1^{-1} at 2%, 4%, and 6% strain, respectively. The DOS panels in Figure 2 confirm this directly. The total DOS at EFE_F increases at each successive strain level. B:p and C:p contributions account for most of the increase, as both sit at the Fermi level. The N:p contribution shifts modestly with lattice strain but remains anchored below EFE_F, because these nitrogen states are strongly hybridized into deep bonding orbitals. As a result, it has little effect on N(EF)N(E_F).

Figure 3 shows the three-dimensional Fermi surfaces of R3m B₂CN under isotropic tensile strains of (a) 0%, (b) 2%, (c) 4%, and (d) 6% within the first Brillouin zone of the rhombohedral lattice. The color scale represents the local magnitude of the Fermi velocity vFv_F, ranging from minimum values in blue to maximum values in red. At all strain levels, the Fermi surface contains two different types of sheets. The large central pocket around the Γ point is electron-like and has a rounded and convex shape. It originates from bands that cross the Fermi level with upward dispersion near the zone center. This is also consistent with the band structure in Figure 2, where C-pp and B-pp dominated bands cross EFE_F near Γ. At 0% strain in panel (a), the central electron pocket is irregular and strongly anisotropic, with large regions of high Fermi velocity shown by the orange-red colors. This indicates strongly dispersive bands near the Fermi level. As the tensile strain increases to 2% and 4% in panels (b) and (c), the central electron pocket becomes smoother and more isotropic with a more rounded shape. The mainly green color indicates intermediate Fermi velocities and suggests that the anisotropy of the band dispersion around Γ is reduced by strain. At 6% strain in panel (d), the central pocket becomes more compact and nearly hexagonally symmetric with mostly low-to-intermediate vFv_F values in the blue-green range. This suggests that the related bands near Γ become less dispersive under larger tensile strain.

Fermi surface of R3m B₂CN under isotropic tensile strain

Figure 3. Fermi surface of R3m B₂CN under isotropic tensile strain of (a) 0%, (b) 2%, (c) 4%, and (d) 6% with the color mapping representing the magnitude of the Fermi velocity vFv_F.

The outer star-shaped sheets extend toward the boundaries of the hexagonal Brillouin zone and form six arm-like features. These sheets are hole-like and originate from bands that cross the Fermi level with downward dispersion near the zone-boundary regions, including the T, L, and F points. This is consistent with the downward-dispersing bands observed near T and along the Γ-T and L-Γ directions in Figure 2. At 0% strain, the hole-like arms are broad and interconnected and show a wide range of Fermi velocities. The neck regions mainly show blue-to-green colors, while the outer parts change from green to yellow. This indicates that the hole-like surface contains states with different band curvatures and dispersions. With increasing tensile strain, the hole-like arms become smaller and narrower. The connecting regions between neighboring arms become thinner at 2% and 4% strain, while the overall volume of the hole-like sheets decreases. This suggests that the contribution of hole carriers at the Fermi level is reduced with strain. At 6% strain, the hole-like arms are much smaller and appear less connected, with mainly low Fermi velocities shown by the blue-green colors. This suggests that tensile strain shifts the related hole bands away from the Fermi level and limits their contribution to a smaller region of momentum space.

Overall, the Fermi surface changes continuously with tensile strain. The electron-like pocket around Γ\Gamma becomes more regular and less anisotropic, while the hole-like sheets near the zone boundaries become smaller. At the same time, the variation in Fermi velocity across the surface is reduced, and the total density of states at the Fermi level is modified. These changes are directly relevant to the superconducting properties of B2_2CN, since the density of states at EFE_F and the Fermi velocity enter directly into the electron-phonon coupling and the superconducting transition temperature TcT_c through the McMillan–Allen–Dynes framework.

The superconducting implications of this trend are discussed in Section VI. According to the McMillan-Hopfield expression [94, 95], λ∝N(EF)⟨I2⟩M⟨ω2⟩\lambda\propto\frac{N(E_F)\langle I^2\rangle}{M\langle\omega^2\rangle}, where ⟨I2⟩\langle I^2\rangle is the Fermi-surface-averaged electron-phonon matrix element, MM the atomic mass, and ⟨ω2⟩\langle\omega^2\rangle the mean-square phonon frequency. Because λ\lambda scales directly with N(EF)N(E_F), the band narrowing established above is favorable for electron-phonon coupling. Tensile strain drives this rise in N(EF)N(E_F) purely by narrowing the B:p and C:p bands at the Fermi level. The metallic character and orbital composition of the Fermi surface, and the sheet count established in Figure 3, remain unchanged. As shown in Section VI, however, the rise in N(EF)N(E_F) is modest compared to the accompanying phonon softening, which is the dominant contributor to the strain-induced enhancement of λ\lambda and TcT_c. Angle-resolved photoemission spectroscopy (ARPES) on strained samples can detect both the bandwidth narrowing shown in Figure 2 and the associated rise in N(EF)N(E_F) [96, 97]. ARPES’s momentum resolution allows direct examination of the underlying electronic mechanism, providing an experimentally accessible way to test these predictions before superconductivity measurements.

Phonon Properties

Phonon dispersions were computed for all four strained structures along Γ\Gamma–T–H2_2|H0_0–L–Γ\Gamma–S0_0|–S2_2–F–Γ\Gamma of the hexagonal Brillouin zone. No imaginary frequencies appeared at any q-point or strain level. All four configurations are dynamically stable [9].

The phonon spectrum of B2_2CN is shown in Figure 4(i). At zero strain the spectrum extends to approximately 1100 cm−1^{-1}. Atom-projected phonon density of states (PHDOS) analysis (Figure 4(ii)) shows that the vibrational character is governed by interatomic force constants rather than atomic mass alone. N-derived modes dominate the high-frequency region (roughly 600–1100 cm−1^{-1}), where N sits in a stiff bonding environment in the lattice. C modes fall mainly in the mid-frequency range. B modes are concentrated at lower frequencies. This happens not because boron is the lightest atom, but because the restoring forces at B sites are comparatively softer.

Phonon dispersions, projected phonon density of states, and Eliashberg spectral function for R3m B2CN under tensile strain

Figure 4. i) Phonon dispersions along the high-symmetry path, ii) projected phonon density of states (PHDOS), and iii) Eliashberg spectral function α2F(ω)\alpha^{2}F(\omega) with cumulative coupling λ(ω)\lambda(\omega) for R3m B2_{2}CN under (a) 0%, (b) 2%, (c) 4%, and (d) 6% isotropic tensile strain. Branch colors in (i) indicate mode-resolved coupling strength λqν\lambda_{q\nu} from low (cyan) to high (magenta).

Phonon frequencies scale as ω∝keff/m\omega\propto\sqrt{k_{\mathrm{eff}}/m}, so a soft local force constant can outweigh a light atomic mass and pin the associated modes toward lower frequency [38]. This is consistent with the bond-projected force constants in Table ??. This force-constant hierarchy remains qualitatively unchanged across the strain series, and the identity of the vibrational eigenstates is preserved even as their frequencies shift.

Isotropic tensile strain progressively softens the entire phonon spectrum [98]. The maximum phonon frequency falls from approximately 1100 cm−1^{-1} at 0% to roughly 900 cm−1^{-1} at 6% strain. All three PHDOS envelopes shift to lower frequencies with each lattice increment (Figure 4(ii)). This global softening follows from the weakening of interatomic force constants as B–C and B–N bonds elongate under lattice expansion. At 6% strain, the lowest acoustic and low-optical branches near Γ\Gamma, T & F show a pronounced downward shift without developing imaginary character. The system is approaching but has not yet reached a lattice instability. The strain-induced softening of the B-dominated acoustic branches can be tested using inelastic X-ray or neutron scattering [99, 100]. Measurements of the phonon dispersion at different strain levels would directly test the calculated softening and provide experimental insight into its connection with the enhanced electron-phonon coupling predicted here.

The mode-resolved coupling strength [72] λq\lambda_{\mathbf{q}} is shown by the color scale from cyan (low) to magenta (high) in Figure 4(iii). It is strongly concentrated in the low-to-mid frequency branches at all strain levels. At 0% strain, the most strongly coupled modes cluster around 200–400 cm−1^{-1}, in predominantly B-character and mixed B–C branches. As strain increases, these high-λq\lambda_{\mathbf{q}} modes shift downward — to roughly 150–250 cm−1^{-1} at 6% strain — driven by phonon softening in the low-frequency sector. The high-frequency N-dominated optical branches remain weakly coupled throughout. N contributes little weight at EFE_F, so these modes couple poorly to the Fermi surface.

Although no imaginary frequencies appear within the studied strain range, if the strain is increased to 8% or 10%, the lowest acoustic branches develop imaginary character, indicating a dynamical instability, as shown in Section ??. The 6% strain level is therefore near the upper limit of mechanical and dynamical stability for R3m B2_2CN.

Electron–Phonon Coupling & Superconducting Critical Temperature

Formalism

Sections IV and V established the two inputs: band narrowing raises N(EF)N(E_F) and tensile strain softens the phonon spectrum. This section quantifies how they com-

bine to determine λ\lambda and TcT_{\mathrm{c}}. The isotropic reference point is set by the Eliashberg spectral function,

α2F(ω)=12πN(EF)∑qνγqνωqνδ(ω−ωqν),(2)\alpha^{2}F(\omega) = \frac{1}{2\pi N(E_{\mathrm{F}})} \sum_{\mathbf{q}\nu} \frac{\gamma_{\mathbf{q}\nu}}{\omega_{\mathbf{q}\nu}} \delta(\omega-\omega_{\mathbf{q}\nu}), \tag*{(2)}

and its energy-weighted integral,

λ=2∫0∞α2F(ω)ω dω(3)\lambda= 2 \int_{0}^{\infty} \frac{\alpha^{2}F(\omega)}{\omega}\,d\omega \tag*{(3)}

Both were extracted from the EPW calculations following Section II. The logarithmic average phonon frequency,

ωlog⁡=exp⁡[2λ∫0∞α2F(ω)ln⁡ωω dω](4)\omega_{\log} = \exp\left[\frac{2}{\lambda}\int_{0}^{\infty}\frac{\alpha^{2}F(\omega)\ln\omega}{\omega}\,d\omega\right] \tag*{(4)}

enters the Allen-Dynes formula,

TcAD=ωlog⁡1.2exp⁡[−1.04(1+λ)λ−μ∗(1+0.62λ)].(5)T_{\mathrm{c}}^{\mathrm{AD}} = \frac{\omega_{\log}}{1.2}\exp\left[-\frac{1.04(1+\lambda)}{\lambda-\mu^{*}(1+0.62\lambda)}\right]. \tag*{(5)}

Equation (5) is an isotropic reference point. The TcT_{\mathrm{c}} values reported in § 1.0.5 C come from fully anisotropic Migdal-Eliashberg equations solved on the Matsubara frequency axis in EPW.

Electron-Phonon Coupling Strength

The Eliashberg spectral function [72] α2F(ω)\alpha^{2}F(\omega) and its cumulative integral [73] λ(ω)\lambda(\omega) (Figure 4, iii) track how electron-phonon coupling evolves with isotropic tensile strain. At 0% strain, spectral weight in α2F(ω)\alpha^{2}F(\omega) is distributed broadly across low and mid frequencies, giving a coupling constant λ=0.967\lambda= 0.967. As strain increases, a pronounced low-frequency peak develops in α2F(ω)\alpha^{2}F(\omega) and grows. By 6% strain, this peak dominates the spectral function, and λ\lambda reaches 2.620, which is 2.7 times its unstrained value. The key electron-phonon and superconducting parameters for all four strain levels are collected in Table II.

The cumulative λ(ω)\lambda(\omega) curves make the regime change concrete. At 0% strain, λ(ω)\lambda(\omega) rises gradually across the full frequency range, saturating near 1000 cm−1^{-1}. By 6% strain it rises steeply below ≈300\approx300 cm−1^{-1} and then flattens — more than 80% of the total λ=2.620\lambda= 2.620 accumulates in this low-frequency window. This is a strong-coupling limit, and it has direct consequences for the Allen–Dynes approximation discussed below.

The superlinear growth of λ\lambda with strain compounds two mechanisms acting together. Band narrowing raises N(EF)N(E_{\mathrm{F}}) (Section IV). At the same time, softening of the most strongly coupled low-frequency phonon branches (§ 1.0.4) amplifies the Fermi-surface-averaged electron–phonon matrix elements. Their combined impact is sharpest between 4% and 6% strain, where λ\lambda increases by 1.04 — more than the total gain over the entire 0–4% range.

Superconducting Tc: Allen-Dynes vs. Anisotropic Eliashberg

The strain dependence of the superconducting properties is summarized in Figure 6. The phonon softening that drives this enhancement simultaneously reduces ωlog⁡\omega_{\log} as

R3m B2CN superconducting properties versus isotropic tensile strain

Figure 6. R3m B2_2CN superconducting properties versus isotropic tensile strain. (a) EPC strength λ\lambda and logarithmic average phonon frequency ωlog⁡\omega_{\log}. (b) Superconducting TcT_{\mathrm{c}} from the anisotropic Eliashberg equations, Allen-Dynes formula, and machine-learning predictor. Dashed line: liquid nitrogen boiling point (77 K).

Strainλ\lambdaωlog⁡\omega_{\log}BCS Δ0\Delta_0Allen–Dynes TcT_cML TcT_cAniso TcT_c
0%0.96755.046.4042.2 K44.9 K53.5 K
2%1.24250.328.2754.5 K61.0 K68.5 K
4%1.57943.649.1860.5 K72.4 K78.5 K
6%2.62030.749.1360.2 K89.5 K84.5 K

caption: Superconducting parameters and critical temperatures of R3m B2_2CN under different isotropic tensile strain, showing the electron–phonon coupling constant λ\lambda, the logarithmic average phonon frequency ωlog⁡\omega_{\log} (in meV), the superconducting gap at zero temperature Δ0\Delta_0 (in meV), and the superconducting transition temperatures TcT_c calculated using the Allen–Dynes equation, a machine-learning (ML) model, and an anisotropic Eliashberg calculation (Aniso), respectively.

:::

spectral weight shifts to lower frequencies. This competition produces the non-monotonic Allen–Dynes TcADT_c^{\mathrm{AD}} [74]. TcADT_c^{\mathrm{AD}} rises from 42.2 K at 0% to a peak of 60.5 K at 4% strain, then stalls at 60.2 K at 6%. This stall happens because the sharp drop in ωlog⁡\omega_{\log}, from 43.64 to 30.74 meV, partially offsets the gain in λ\lambda. Above λ≈1.5\lambda\approx1.5 the Allen–Dynes formula loses accuracy [101, 102, 103], a threshold crossed between 2% and 4% strain, exactly where the TcADT_c^{\mathrm{AD}} curve begins to flatten.

The anisotropic Eliashberg equations [104] yield a monotonically increasing TcT_c across the strain series (Figure 6b). The gap between TcADT_c^{\mathrm{AD}} and TcanisoT_c^{\mathrm{aniso}} widens from 11.3 K at 0% to 24.3 K at 6%. This growing gap is a direct measure of how badly the Allen–Dynes approximation breaks down as the system moves deeper into the strong-coupling regime. The predicted transition temperatures can be experimentally tested using four-probe resistivity and AC susceptibility measurements under controlled strain or in intercalated bulk specimens [105, 106].

a different Fermi sheet [107, 108]. This produces spectral complexity absent in a single-gap ss-wave superconductor. The multi-peak quasiparticle DOS in Figure 5(i) can be tested by low-temperature scanning tunneling spectroscopy (STS) [109]. In STS, the differential conductance provides a measure of the superconducting density of states. Point-contact Andreev reflection (PCAR) gives a complementary measurement of the superconducting gap and can also separate multiple gap scales in multiband superconductors [110].

Anisotropic Eliashberg results for R3m B2CN under isotropic tensile strain

Figure 5. Anisotropic Eliashberg results for R3m B2CN at (a) 0%, (b) 2%, (c) 4%, and (d) 6% isotropic tensile strain. Panel (i): anisotropic gap distribution Δnk\Delta_{nk} vs. temperature (cyan: low TT, magenta: near TcT_c). Panel (ii): quasiparticle DOS Ns(ω)/NFN_s(\omega)/N_F showing coherence peaks collapsing toward TcT_c. Panel (iii): kk-resolved coupling distribution ρ(λnk)\rho(\lambda_{nk}); dashed line marks ⟨λ⟩\langle\lambda\rangle

The kk-resolved coupling distribution ρ(λnk)\rho(\lambda_{nk}) (Figure 5(ii)) maps how coupling strength varies across the Fermi surface. At 0% strain, the distribution is compact, contained almost entirely within λnk<1.5\lambda_{nk}<1.5, with Fermi-surface average ⟨λ⟩=0.740\langle\lambda\rangle=0.740. As the strain increases it broadens and a high-coupling tail emerges. ⟨λ⟩\langle\lambda\rangle rises to 0.969, 1.392, and 1.941 at 2%, 4%, and 6% strain, and the tail expands beyond λnk>4\lambda_{nk}>4 at 6% strain. A subset of kk-states on the Fermi surface becomes extremely strongly coupled under high strain. These states drive the super-linear growth of λ\lambda and the large anisotropic TcT_c.

Anisotropic Gap Structure and Quasiparticle DOS

The gap distribution Δnk\Delta_{nk} as a function of temperature (Figure 5(i)) is strongly anisotropic at all strain levels. It spans continuously across a finite range rather than taking a single isotropic value. At 0% strain, gaps span roughly 0–12 meV at low temperature. With increasing strain, both the width and the upper edge grow, reaching ≈15\approx15 meV at 4% and ≈20\approx20 meV at 6% strain. At high strain the distribution stratifies into distinct bands. Each band maps onto one of the seven Fermi sheets identified in Section IV, and sheets with stronger Fermi-surface coupling develop proportionally larger gaps. The distribution collapses uniformly to zero at TcT_c at every strain level, confirming a single superconducting transition with no partial gap above TcT_c.

The quasiparticle density of states Ns(ω)/NFN_s(\omega)/N_F (Figure 5(ii)) encodes the gap structure directly. At each strain level sharp coherence peaks appear symmetrically about ℏω=0\hbar\omega=0 and collapse toward NsN_s as T→TcT\rightarrow T_c, as expected for a superconducting DOS [5]. Their energy increases with strain, consistent with the growing maximum gap. At 6% strain the spectrum develops a multi-peak structure extending to ∣ℏω∣≈20|\hbar\omega|\approx20 meV – each peak arising from a different Fermi sheet [107, 108]. This produces spectral complexity absent in a single-gap ss-wave superconductor. The multi-peak quasiparticle DOS in Figure 5(i) can be tested by low-temperature scanning tunneling spectroscopy (STS) [109]. In STS, the differential conductance provides a measure of the superconducting density of states. Point-contact Andreev reflection (PCAR) gives a complementary measurement of the superconducting gap and can also separate multiple gap scales in multiband superconductors [110].

The kk-resolved coupling distribution ρ(λnk)\rho(\lambda_{nk}) (Figure 5(ii)) maps how coupling strength varies across the Fermi surface. At 0% strain, the distribution is compact, contained almost entirely within λnk<1.5\lambda_{nk}<1.5, with Fermi-surface average ⟨λ⟩=0.740\langle\lambda\rangle=0.740. As the strain increases it broadens and a high-coupling tail emerges. ⟨λ⟩\langle\lambda\rangle rises to 0.969, 1.392, and 1.941 at 2%, 4%, and 6% strain, and the tail expands beyond λnk>4\lambda_{nk}>4 at 6% strain. A subset of kk-states on the Fermi surface becomes extremely strongly coupled under high strain. These states drive the super-linear growth of λ\lambda and the large anisotropic TcT_c.

Independent Cross-Validation

The SISSO-based machine-learning (ML) model implemented in EPW provides a complementary estimate of TcT_c [65, 111, 112]. The ML TcT_c rises monotonically from 44.9 K at 0% to 89.5 K at 6% strain, staying within 10 K of TcanisoT_c^{\mathrm{aniso}} across the full range. Agreement between two independent methods – one from first-principles Eliashberg theory and one from a data-driven model – validates the predicted TcT_c values. It also confirms the monotonic trend that the Allen–Dynes formula fails to capture.

The BCS superconducting gap [5] Δ0\Delta_0 increases from 6.40 meV at 0% to 9.18 meV at 4% strain, then holds at 9.13 meV at 6% as λ\lambda continues to rise sharply. In this strong-coupling regime the gap saturates while λ\lambda keeps growing – the two are no longer simply proportional. The picture across the strain levels is clear: λ\lambda grows superlinearly, TcanisoT_c^{\mathrm{aniso}} increases by 31 K over the 0–6% range.

Conclusion

Isotropic strain is a direct route to stronger superconductivity in R3m B2_2CN. Across 0–6% strain, λ\lambda rises from 0.967 to 2.620 and ωlog⁡\omega_{\log} falls from 55.04 to 30.74 meV. The anisotropic Eliashberg TcT_c rises monotonically from 53.5 K to 84.5 K. All expanded structures are dynamically stable, with no imaginary modes at any volume. Phonon calculations at 8% and 10% strain reveal imaginary modes at both volumes, placing the stability boundary just above 6% and makes it the natural upper limit of the series. The enhancement is driven by genuine softening and redistribution of spectral weight, not by approach to a structural instability.

The Allen-Dynes and anisotropic Eliashberg predictions split above λ≈1.5\lambda\approx1.5. TcADT_c^{\mathrm{AD}} peaks at 4% strain (60.5 K, λ=1.579\lambda= 1.579) and stalls at 6% (60.2 K). The anisotropic Eliashberg TcT_c keeps rising and reaches 84.5 K at 6%, and the ML estimate of 89.5 K follows it rather than Allen-Dynes. Agreement between these two independent methods indicates that gap anisotropy becomes significant under expansion and needs to be treated explicitly, not approximated away.

These calculations confirm a dynamically stable, first-principles route to strain-tunable, phonon-mediated high-TcT_c superconductivity in a light-element covalent compound [113]. This result holds across the full 0–6% strain series, with no sign of structural instability at any point.

Reaching this experimentally means finding a way to expand the lattice isotropically. B2_2CN’s stiffness (bulk modulus ≈320\approx320 GPa) rules out mechanical loading, and epitaxial strain is directional [114]. Intercalation of spacer species between the layers - negative chemical pressure - is the natural candidate. In practice, however, intercalation typically expands the c-axis preferentially rather than isotropically. Whether c-axis-dominated expansion reproduces the coupling enhancement seen here is the immediate open question.

The isotropic protocol isolates the volumetric contribution to electron-phonon coupling from directional and ionic-relaxation effects. It is a deliberate choice that cleanly separates lattice expansion from everything else. Modeling specific intercalants and accounting for anharmonicity [115] at large expansions are beyond the scope of this work. The stability and scale of the predicted enhancement give B2_2CN and related ternary boron-carbon-nitrogen phases a clear case for experimental follow-up. Brillouin light scattering can give the strain-dependent elastic tensor. ARPES on strained specimens can resolve the bandwidth compression and N(EF)N(E_F) rise. Inelastic X-ray or neutron scattering can map the full phonon softening. Four-probe resistivity and AC susceptibility can test the predicted TcT_c values, and scanning tunneling spectroscopy or PCAR can resolve the anisotropic multi-gap structure in Figure 5(i). Each measurement targets a different link in the coupling chain, but taken together, they form a tractable experimental test of the strain-tuning mechanism proposed here.

Data availability

The datasets generated and analysed during this study are not publicly available due to their large file sizes, but are available from the corresponding author upon reasonable request.

References

Strain-Tunable High-TcT_c Superconductivity in Layered R3ˉm\mathrm{R}\bar{3}\mathrm{m} B2_2CN

Durjay Dhali; Arafat Rahman; Tareq Mahmud; Alamgir Kabir

Supplementary.pdf

Supplemental Material for: Strain-Tunable High-TcT_c Superconductivity in Layered R3ˉm\mathrm{R}\bar{3}\mathrm{m} B2_2CN

Computational Convergence Tests

The kinetic-energy cutoff of 60 Ry and charge-density cutoff of 480 Ry are the values recommended for the ultrasoft pseudopotentials (USPP) in the Standard Solid-State Pseudopotentials (SSSP) precision library [1]. These are benchmarked against all-electron references across a wide range of crystal structures and were used directly without additional system-specific testing.

k-Mesh Convergence

Table S1 lists N(EF)N(E_F) computed with Gaussian smearing (σ=0.02\sigma= 0.02 Ry) at three k-mesh densities across all four strain levels. The monotonic decrease with mesh density reflects shrinking smearing-induced bias rather than a lack of convergence. N(EF)N(E_F) shifts by less than 3% between the 24×24×224 \times24 \times2 and 32×32×232 \times32 \times2 meshes at all strain levels, confirming the production mesh is converged. Main-text values were obtained with the tetrahedron method on the 24×24×224 \times24 \times2 mesh, giving slightly different absolute numbers but the same convergence behavior.

k-meshε=0%\varepsilon= 0\%ε=2%\varepsilon= 2\%ε=4%\varepsilon= 4\%ε=6%\varepsilon= 6\%
16×16×216 \times16 \times22.402.532.672.80
24×24×224 \times24 \times22.332.422.522.63
32×32×232 \times32 \times22.262.382.502.63

Table S1. N(EF)N(E_F) (states eV−1^{-1}) evaluated with Gaussian smearing (σ=0.02\sigma= 0.02 Ry) at three k-mesh densities for all four strain levels. The selected production mesh is 24×24×224 \times24 \times2.

Wannier Function Construction and Interpolation Accuracy

Wannier Spread and Localization

Wannierization used initial projections onto B:2p, C:2p, and N:2p orbitals, giving 36 maximally localized Wannier functions per unit cell (12 atoms ×\times 3 pp-orbitals). Table S2 reports the total spread decomposed into ΩI\Omega_I, ΩD\Omega_D, and ΩOD\Omega_{OD} at each strain level.

Quantityε=0%\varepsilon= 0\%ε=2%\varepsilon= 2\%ε=4%\varepsilon= 4\%ε=6%\varepsilon= 6\%
ΩI\Omega_I38.229533.731442.677749.2011
ΩD\Omega_D3.823637.04920.25260.4162
ΩOD\Omega_{OD}6.75535.68456.05518.0404
ΩTotal\Omega_{Total}48.808476.465148.985557.6577

Table S2. Wannier spread components (Å2^2) for B:2p, C:2p, and N:2p projections at each strain level. ΩI\Omega_I is gauge-invariant; ΩD\Omega_D and ΩOD\Omega_{OD} are gauge-dependent parts minimized during Wannierization.

Interpolation Accuracy

Wannier interpolation quality was assessed by comparing interpolated band velocities and Fermi surface smoothness to directly computed NSCF eigenvalues. The interpolated bands reproduce the correct number of Fermi crossings and orbital character at EFE_{\mathrm{F}} across all four strain levels. A direct band overlay is omitted because the DFT and Wannier structures were computed along different k-paths; the Wannier path follows the hexagonal zone convention used in EPW, while the DFT path follows the rhombohedral convention used in the main text. Interpolation accuracy is instead reflected in the physical consistency of α2F(ω)\alpha^{2}F(\omega) and the gap distributions, neither of which shows spurious features attributable to interpolation error.

Structural Details

Fractional Coordinates and Bond Lengths

Atomic fractional coordinates were held fixed at all strain levels, preserving R3ˉmR\bar{3}m symmetry by construction (see Sec. II of the main text). Table S3 lists the resulting interatomic distances. Since fractional coordinates are fixed, all bond lengths scale uniformly with aa and cc. The B–C and B–N distances elongate by 6% across the full strain range, weakening the interatomic force constants that drive phonon softening and the increase in electron-phonon coupling reported in the main text.

Bondε=0%\varepsilon= 0\%ε=2%\varepsilon= 2\%ε=4%\varepsilon= 4\%ε=6%\varepsilon= 6\%
B–C1.11451.13681.15911.1814
B–N1.02631.04681.06741.0879
C–N2.32562.37212.41862.4651

Table S3. Key interatomic distances (A˚\mathring{\mathrm{A}}) in R3ˉmR\bar{3}m B2_2CN at each strain level. Bond lengths scale uniformly with the lattice parameters since fractional coordinates are fixed throughout.

Bond-Projected Force Constants

To quantify the bond softening implied by the lattice elongation in Sec. S3.1, we projected the real-space interatomic force-constant tensor Φαβ(i,j)\Phi_{\alpha\beta}(i,j) onto each bond’s unit vector, kbond=u^TΦu^k_{\mathrm{bond}} = \hat{u}^{T}\Phi\hat{u}, using the force constants from q2r.x. Table S4 lists the resulting nearest-neighbor B–C and B–N force constants at each strain level. Both bonds soften monotonically with strain: the B–C force constant decreases by 47.7% and the B–N force constant by 70.2% across the full 0–6% range, with B–N softening consistently faster than B–C at every strain step. The preferential softening of B–N drives the phonon softening and enhanced electron-phonon coupling reported in the main text.

Bondε=0%\varepsilon= 0\%ε=2%\varepsilon= 2\%ε=4%\varepsilon= 4\%ε=6%\varepsilon= 6\%
B–C13.1610.748.676.88
B–N10.187.375.023.04

Table S4. Nearest-neighbor bond-projected force constants (eV/A˚2\mathring{\mathrm{A}}^{2}) in R3ˉmR\bar{3}m B2_2CN at each strain level, obtained by projecting the real-space force-constant tensor onto the bond direction. Values are reported as magnitudes; the underlying tensor elements carry the restoring-force sign convention of the interatomic force-constant matrix.

Elastic Properties

Born–Huang Stability Criteria

The Born–Huang stability conditions for trigonal crystals are given in Sec. III of the main text. Table S5 lists their numerical values at each strain level, computed from the elastic constants in Table I. All four conditions are satisfied with a large margin across the 0–6% strain range, and no strain level approaches a mechanical instability.

Conditionε=0%\varepsilon= 0\%ε=2%\varepsilon= 2\%ε=4%\varepsilon= 4\%ε=6%\varepsilon= 6\%
C11−∣C12∣C_{11} - \lvert C_{12}\rvert (GPa)342.58342.71342.26340.34
C44C_{44} (GPa)292.97293.06292.72291.20
(C11+C12)C33−2C132(C_{11} + C_{12})C_{33} - 2C_{13}^{2} (10510^{5} GPa2^{2})6.516.516.496.38
(C11−C12)C44(C_{11} - C_{12})C_{44} (10510^{5} GPa2^{2})1.001.001.000.99

Table S5. Numerical evaluation of the Born–Huang stability conditions for trigonal crystals at each strain level. Conditions 1 and 2 are in GPa; conditions 3 and 4 are in 10510^{5} GPa2^{2}. All values are positive, confirming mechanical stability throughout the strain series.

Directional Elastic Properties from ELATE

The full angular dependence of the elastic response was computed with ELATE [2]. Figures S1 and S2 show the directional Young’s modulus E(n^)E(\hat{n}), linear compressibility β(n^)\beta(\hat{n}), shear modulus G(n^)G(\hat{n}), and Poisson’s ratio ν(n^)\nu(\hat{n}) at 0% and 6% strain. E(n^)E(\hat{n}) remains nearly spherical at both strain levels, consistent with the three-fold rotational symmetry of the rhombohedral lattice and the absence of any pronounced soft direction. The slight flattening along the cc-axis in G(n^)G(\hat{n}) reflects the smaller C44C_{44} relative to C11C_{11} reported in Table I.

Phonon Dispersions at 8% and 10% Strain

Phonon calculations at 8% and 10% isotropic tensile strain reveal imaginary branches over broad portions of the Brillouin zone, placing the dynamical stability boundary just above 6% strain (§ 1.0.5 of the main text). Figure S3 shows the full dispersion curves at both strain levels. Imaginary modes, plotted as negative frequencies, appear across the entire zone, consistent with a soft-mode instability from excessive weakening of the interatomic force constants.

Phonon dispersions of $\mathrm{R}\bar{3}\mathrm{m}$ B$_2$CN at 8% and 10% isotropic tensile strain

Figure S3. Phonon dispersions of R3ˉm\mathrm{R}\bar{3}\mathrm{m} B2_2CN at 8% and 10% isotropic tensile strain. Imaginary frequencies (plotted as negative values) confirm that 6% is near the upper limit of dynamical stability.

Fermi Surface Sheet Decomposition

R3ˉmR\bar{3}m B2_2CN has seven Fermi sheets (bands 8–14) at all strain levels. Figure S4 shows each sheet at 0% and 6% strain, colored by the local Fermi velocity vFv_{\mathrm{F}}. The total N(EF)N(E_{\mathrm{F}}) at each strain level is given in Section IV of the main text.

Individual Fermi surface sheets from bands 8–14 for $\bar{R}3$m B$_2$CN at (a) 0% and (b) 6% strain, colored by the magnitude of the Fermi velocity $v_F$.

Figure S4. Individual Fermi surface sheets from bands 8–14 for Rˉ3\bar{R}3m B2_2CN at (a) 0% and (b) 6% strain, colored by the magnitude of the Fermi velocity vFv_F.

Eliashberg Spectral Function: Supplementary Analysis

Sensitivity to the Coulomb Pseudopotential *

The Coulomb pseudopotential μ∗=0.10\mu^{*} = 0.10 used throughout is the standard value for covalent light-element systems. Table S6 reports TcADT_{\mathrm{c}}^{\mathrm{AD}} and TcisoT_{\mathrm{c}}^{\mathrm{iso}} for μ∗∈{0.08,0.10,0.12,0.15}\mu^{*} \in\{0.08, 0.10, 0.12, 0.15\} at 0% and 6% strain, using the isotropic Eliashberg solver as a computationally cheaper proxy for testing μ∗\mu^{*} sensitivity. This isotropic solver systematically underestimates TcT_{\mathrm{c}} relative to the fully anisotropic calculation reported in the main text (e.g., Tciso=48.5T_{\mathrm{c}}^{\mathrm{iso}} = 48.5 K vs. Tcaniso=53.5T_{\mathrm{c}}^{\mathrm{aniso}} = 53.5 K at 0% strain and μ∗=0.10\mu^{*} = 0.10), consistent with the anisotropy enhancement discussed in § 1.0.5.C. The μ∗\mu^{*}-dependence trend, rather than the absolute values, is the relevant result here.

μ∗\mu^{*}TcADT_{\mathrm{c}}^{\mathrm{AD}} (0%)TcisoT_{\mathrm{c}}^{\mathrm{iso}} (0%)TcADT_{\mathrm{c}}^{\mathrm{AD}} (6%)TcisoT_{\mathrm{c}}^{\mathrm{iso}} (6%)
0.0846.552.562.385.5
0.10 (used)42.248.560.282.5
0.1238.045.558.080.5
0.1531.941.554.877.5

Table S6. Dependence of TcADT_{\mathrm{c}}^{\mathrm{AD}} and TcisoT_{\mathrm{c}}^{\mathrm{iso}} (K) on the Coulomb pseudopotential μ∗\mu^{*} at ε=0%\varepsilon= 0\% and 6%. The adopted value μ∗=0.10\mu^{*} = 0.10 is marked. The liquid-nitrogen boiling point (77 K) is exceeded by TcisoT_{\mathrm{c}}^{\mathrm{iso}} at 6% strain for all values of μ∗\mu^{*} in the table.

Four directional elastic property plots: linear compressibility, Poisson's ratio, shear modulus, and Young's modulus

Figure S1. Directional elastic properties of R3ˉmR\bar{3}m B2_2CN at ε=0%\varepsilon= 0\% from ELATE [2]. (a) Young’s modulus E(n^)E(\hat{n}), (b) linear compressibility β(n^)\beta(\hat{n}), (c) shear modulus G(n^)G(\hat{n}), and (d) Poisson’s ratio ν(n^)\nu(\hat{n}).

Four surface plots showing linear compressibility, Poisson's ratio, shear modulus, and Young's modulus

Figure S2. Same as Fig. S1 for ε=6%\varepsilon= 6\% isotropic tensile strain. The overall topology of each surface is preserved relative to the unstrained case, consistent with the negligible shift in CijC_{ij} values across the strain series.

Anisotropic Eliashberg Equations: Additional Results

Sheet-Resolved Gap Mapping

The gap distribution Δnk\Delta_{nk} stratifies into distinct bands at high strain, as discussed in § 1.0.5.D of the main text, with each band mapping onto one of the seven Fermi sheets (bands 8–14). Table S7 reports the mean gap on each sheet normalized by kBTck_{\mathrm{B}}T_{\mathrm{c}}, extracted from the anisotropic Eliashberg solution at the lowest computationally accessible temperature at each strain level. Normalization by kBTck_{\mathrm{B}}T_{\mathrm{c}} permits comparison across the strain series despite differing minimum temperatures. The BCS weak-coupling limit gives Δ0/kBTc=1.764\Delta_{0}/k_{\mathrm{B}}T_{\mathrm{c}} = 1.764 from 2Δ0=3.528kBTc2\Delta_{0} = 3.528 k_{\mathrm{B}}T_{\mathrm{c}}. The systematic enhancement above this value at 6% strain follows from the strong-coupling regime (λ=2.620\lambda= 2.620) established in Sec. V. Band indices at ε=4%\varepsilon= 4\% are re-indexed from 1 in the EPW output and are mapped to the DFT band numbering (8-14) used throughout.

Band (sheet)ε=0%\varepsilon= 0\% (25 K)ε=2%\varepsilon= 2\% (40 K)ε=4%\varepsilon= 4\% (55 K)ε=6%\varepsilon= 6\% (55 K)
81.691.791.552.14
91.711.771.802.09
101.791.791.782.12
111.761.931.762.22
121.862.031.862.32
131.861.991.712.30
141.861.941.742.28
BCS ref.1.764

Table S7. Sheet-resolved mean superconducting gap normalized by kBTck_{\mathrm{B}}T_{\mathrm{c}} for each Fermi sheet of Rˉ3m\bar{R}3m B2CN\mathrm{B}_{2}\mathrm{CN}. Values are extracted from the anisotropic Eliashberg solution at the lowest computationally accessible temperature at each strain level: 25 K (0%), 40 K (2%), and 55 K (4% and 6%). The BCS weak-coupling limit Δ0/kBTc=1.764\Delta_{0}/k_{\mathrm{B}}T_{\mathrm{c}} = 1.764 serves as a reference.

References

References

  1. [1]F. Giustino, “Electron-phonon interactions from first principles,” Reviews of Modern Physics, vol. 89, no. 1, p. 015003, 2017.DOI
  2. [2]J. A. Flores-Livas, L. Boeri, A. Sanna, G. Profeta, R. Arita, and M. Eremets, “A perspective on conventional high-temperature superconductors at high pressure: Methods and materials,” Physics Reports, vol. 856, pp. 1–78, 2020.
  3. [3]B. Keimer, S. A. Kivelson, M. R. Norman, S. Uchida, and J. Zaanen, “From quantum matter to high-temperature superconductivity in copper oxides,” Nature, vol. 518, no. 7538, pp. 179–186, 2015.DOI
  4. [4]J. Paglione and R. L. Greene, “High-temperature superconductivity in iron-based materials,” Nature physics, vol. 6, no. 9, pp. 645–658, 2010.DOI
  5. [5]J. Bardeen, L. N. Cooper, and J. R. Schrieffer, “Theory of superconductivity,” Physical review, vol. 108, no. 5, p. 1175, 1957.
  6. [6]J. Bardeen, L. N. Cooper, and J. R. Schrieffer, “Microscopic theory of superconductivity,” Physical Review, vol. 106, no. 1, p. 162, 1957.
  7. [7]A. Migdal, “Interaction between electrons and lattice vibrations in a normal metal,” Sov. Phys. JETP, vol. 7, no. 6, pp. 996–1001, 1958.
  8. [8]G. Eliashberg, “Interactions between electrons and lattice vibrations in a superconductor,” Sov. Phys. JETP, vol. 11, no. 3, pp. 696–702, 1960.
  9. [9]S. Baroni, S. De Gironcoli, A. Dal Corso, and P. Giannozzi, “Phonons and related crystal properties from density-functional perturbation theory,” Reviews of modern Physics, vol. 73, no. 2, p. 515, 2001.DOI
  10. [10]F. Giustino, M. L. Cohen, and S. G. Louie, “Electron-phonon interaction using wannier functions,” Physical Review B—Condensed Matter and Materials Physics, vol. 76, no. 16, p. 165108, 2007.
  11. [11]C. J. Pickard, I. Errea, and M. I. Eremets, “Superconducting hydrides under pressure,” Annual Review of Condensed Matter Physics, vol. 11, no. 1, pp. 57–76, 2020.DOI
  12. [12]T. F. Cerqueira, A. Sanna, and M. A. Marques, “Sampling the materials space for conventional superconducting compounds,” Advanced Materials, vol. 36, no. 1, p. 2307085, 2024.DOI
  13. [13]J. Kortus, I. Mazin, K. D. Belashchenko, V. P. Antropov, and L. Boyer, “Superconductivity of metallic boron in mg b2,” Physical Review Letters, vol. 86, no. 20, p. 4656, 2001.
  14. [14]N. W. Ashcroft, “Metallic hydrogen: a high-temperature superconductor?,” Physical Review Let- ters, vol. 21, no. 26, p. 1748, 1968. ity in the atomically thin limit: Strain-enhanced three-gap superconductivity in monolayer mgb _2," arXiv preprint arXiv:1708.04327, 2017.
  15. [15]N. Ashcroft, “Hydrogen dominant metallic alloys: high temperature superconductors?,” Physical Review Letters, vol. 92, no. 18, p. 187002, 2004.pubmed.ncbi.nlm.nih.gov/15169525
  16. [16]J. Nagamatsu, N. Nakagawa, T. Muranaka, Y. Zenitani, and J. Akimitsu, “Superconductivity at 39 k in magnesium diboride,” nature, vol. 410, no. 6824, pp. 63–64, 2001.pubmed.ncbi.nlm.nih.gov/11242039
  17. [17]A. Drozdov, M. Eremets, I. Troyan, V. Ksenofontov, and S. I. Shylin, “Conventional superconductivity at 203 kelvin at high pressures in the sulfur hydride system,” Nature, vol. 525, no. 7567, pp. 73–76, 2015.DOI
  18. [18]A. Drozdov, P. Kong, V. Minkov, S. Besedin, M. Kuznovikov, S. Mozaffari, L. Balicas, F. F. Balakirev, D. Graf, V. Prakapenka, et al., “Superconductivity at 250 k in lanthanum hydride under high pressures,” Nature, vol. 569, no. 7757, pp. 528–531, 2019.DOI
  19. [19]D. Duan, Y. Liu, F. Tian, D. Li, X. Huang, Z. Zhao, H. Yu, B. Liu, W. Tian, and T. Cui, “Pressure-induced metallization of dense (h2s) 2h2 with high-t c superconductivity,” Scientific reports, vol. 4, no. 1, p. 6968, 2014.DOI
  20. [20]Z. Li, X. He, C. Zhang, X. Wang, S. Zhang, Y. Jia, S. Feng, K. Lu, J. Zhao, J. Zhang, et al., “Superconductivity above 200 k discovered in superhydrides of calcium,” Nature communications, vol. 13, no. 1, p. 2863, 2022.DOI
  21. [21]L. Ma, K. Wang, Y. Xie, X. Yang, Y. Wang, M. Zhou, H. Liu, X. Yu, Y. Zhao, H. Wang, et al., “High-temperature superconducting phase in clathrate calcium hydride cah 6 up to 215 k at a pressure of 172 gpa,” Physical Review Letters, vol. 128, no. 16, p. 167001, 2022.DOI
  22. [22]A. Sanna, T. F. Cerqueira, Y.-W. Fang, I. Errea, A. Ludwig, and M. A. Marques, “Prediction of ambient pressure conventional superconductivity above 80 k in hydride compounds,” npj Computational Materials, vol. 10, no. 1, p. 44, 2024.DOI
  23. [23]K. Gao, T. F. Cerqueira, A. Sanna, Y.-W. Fang, D. o. e. Dangić, I. Errea, H.-C. Wang, S. Botti, and M. A. Marques, “The maximum t c of conventional superconductors at ambient pressure,” Nature Communications, vol. 16, no. 1, p. 8253, 2025.DOI
  24. [24]A. Y. Liu, R. M. Wentzovitch, and M. L. Cohen, “Atomic arrangement and electronic structure of bc 2 n,” Physical Review B, vol. 39, no. 3, p. 1760, 1989.DOI
  25. [25]Y. Tateyama, T. Ogitsu, K. Kusakabe, S. Tsuneyuki, and S. Itoh, “Proposed synthesis path for heterodiamond bc 2 n,” Physical Review B, vol. 55, no. 16, p. R10161, 1997.DOI
  26. [26]V. L. Solozhenko, D. Andrault, G. Fiquet, M. Mezouar, and D. C. Rubie, “Synthesis of superhard cubic bc 2 n,” Applied Physics Letters, vol. 78, no. 10, pp. 1385–1387, 2001.
  27. [27]E. Knittle, R. M. Wentzovitch, R. Jeanloz, and M. L. Cohen, “Experimental and theoretical equation of state of cubic boron nitride,” Nature, vol. 337, no. 6205, pp. 349–352, 1989.DOI
  28. [28]K. Watanabe, T. Taniguchi, and H. Kanda, “Direct-bandgap properties and evidence for ultraviolet lasing of hexagonal boron nitride single crystal,” Nature materials, vol. 3, no. 6, pp. 404–409, 2004.pubmed.ncbi.nlm.nih.gov/15156198
  29. [29]J. Bekaert, A. Aperis, B. Partoens, P. M. Oppeneer, and M. Milosevic, “Evolution of multi-gap superconductivity in the atomically thin limit: Strain-enhanced three-gap superconductivity in monolayer mgb 2,” arXiv preprint arXiv:1708.04327, 2017.DOI
  30. [30]Y. Wan, Y.-J. Chen, S.-X. Qiao, K.-Y. Jiang, G.-H. Liu, N. Jiao, P. Zhang, and H.-Y. Lu, “Prediction of high-temperature superconductors at ambient pressure with diamond-like structures: B 2 cx (x= n, p),” Physical Chemistry Chemical Physics, vol. 26, no. 45, pp. 28643–28649, 2024.DOI
  31. [31]M. Calandra and F. Mauri, “High-t c superconductivity in superhard diamondlike bc 5,” Physical review letters, vol. 101, no. 1, p. 016401, 2008.DOI
  32. [32]V. L. Solozhenko, O. O. Kurakevych, D. Andrault, Y. Le Godec, and M. Mezouar, “Ultimate metastable solubility of boron in diamond: synthesis of superhard diamondlike bc 5,” Physical Review Letters, vol. 102, no. 1, p. 015506, 2009.pubmed.ncbi.nlm.nih.gov/19257210
  33. [33]X. Li, X. Yong, M. Wu, S. Lu, H. Liu, S. Meng, J. S. Tse, and Y. Li, “Hard bn clathrate superconductors,” The Journal of Physical Chemistry Letters, vol. 10, no. 10, pp. 2554–2560, 2019.DOI
  34. [34]H.-B. Ding, R. Niu, S.-A. Li, Y.-M. Liu, X.-J. Chen, H.-Q. Lin, and G.-H. Zhong, “Hydrogen-doped c-bn as a promising path to high-temperature superconductivity above 120 k at ambient pressure,” Advanced Science, vol. 11, no. 44, p. 2408275, 2024.DOI
  35. [35]A. D. Smith, Y. K. Vohra, and C.-C. Chen, “First-principles calculation of the superconducting t c in superhard bcn metals,” Physical Review B, vol. 113, no. 21, p. 214509, 2026.DOI
  36. [36]C. R. Tomassetti, G. P. Kafle, E. T. Marcial, E. R. Margine, and A. N. Kolmogorov, “Prospect of high-temperature superconductivity in layered metal borocarbides,” Journal of Materials Chemistry C, vol. 12, no. 13, pp. 4870–4884, 2024.DOI
  37. [37]S. Saha, S. Di Cataldo, M. Amsler, W. Von Der Linden, and L. Boeri, “High-temperature conventional superconductivity in the boron-carbon system: Material trends,” Physical Review B, vol. 102, no. 2, p. 024519, 2020.DOI
  38. [38]N. W. Ashcroft and N. D. Mermin, Solid State Physics. New York: Holt, Rinehart and Winston, 1976.
  39. [39]J.-P. Locquet, J. Perret, J. Fompeyrine, E. Mächler, J. W. Seo, and G. Van Tendeloo, “Doubling the critical temperature of la1. 9sr0. 1cuo4 using epitaxial strain,” Nature, vol. 394, no. 6692, pp. 453–456, 1998.DOI
  40. [40]A. V. Pogrebnyakov, J. M. Redwing, S. Raghavan, V. Vaithyanathan, D. G. Schlom, S. Y. Xu, Q. Li, D. A. Tenne, A. Soukassian, X. X. Xi, M. D. Johannes, J. D. Kasinathan, W. E. Pickett, J. S. Wu, and J. C. H. Spence, “Enhancement of the superconducting transition temperature of mgb 2 by a strain-induced bond-stretching mode softening,” Phys. Rev. Lett., vol. 93, p. 147006, Sep 2004.pubmed.ncbi.nlm.nih.gov/15524834
  41. [41]E. Henríquez-Guerra, L. Almonte, H. Li, D. Elvira, M. R. Calvo, and A. Castellanos-Gomez, “Modulation of the superconducting phase transition in multilayer 2h-nbse2 induced by uniform biaxial compressive strain,” Nano letters, vol. 24, no. 34, pp. 10504–10509, 2024.
  42. [42]E. Johansson, F. Tasnádi, A. Ektarawong, J. Rosen, and B. Alling, “The effect of strain and pressure on the electron-phonon coupling and superconductivity in mgb2—benchmark of theoretical methodologies and outlook for nanostructure design,” Journal of Applied Physics, vol. 131, no. 6, 2022.
  43. [43]W. Kohn and L. J. Sham, “Self-consistent equations including exchange and correlation effects,” Physical review, vol. 140, no. 4A, p. A1133, 1965.DOI
  44. [44]P. Hohenberg and W. Kohn, “Inhomogeneous electron gas,” Physical review, vol. 136, no. 3B, p. B864, 1964.
  45. [45]P. Giannozzi, S. Baroni, N. Bonini, M. Calandra, R. Car, C. Cavazzoni, D. Ceresoli, G. L. Chiarotti, M. Cococcioni, I. Dabo, et al., “Quantum espresso: a modular and open-source software project for quantum simulations of materials,” Journal of physics: Condensed matter, vol. 21, no. 39, p. 395502, 2009.DOI
  46. [46]P. Giannozzi, O. Andreussi, T. Brumme, O. Bunau, M. Buongiorno Nardelli, M. Calandra, R. Car, C. Cavazzoni, D. Ceresoli, M. Cococcioni, et al., “Advanced capabilities for materials modelling with quantum espresso,” Journal of physics: Condensed matter, vol. 29, no. 46, p. 465901, 2017.DOI
  47. [47]P. Giannozzi, O. Baseggio, P. Bonfà, D. Brunato, R. Car, I. Carnimeo, C. Cavazzoni, S. De Gironcoli, P. Delugas, F. Ferrari Ruffino, et al., “Quantum espresso toward the exascale,” The Journal of chemical physics, vol. 152, no. 15, 2020.DOI
  48. [48]D. Vanderbilt, “Soft self-consistent pseudopotentials in a generalized eigenvalue formalism,” Physical review B, vol. 41, no. 11, p. 7892, 1990.pubmed.ncbi.nlm.nih.gov/9993096
  49. [49]J. P. Perdew, K. Burke, and M. Ernzerhof, “Generalized gradient approximation made simple,” Physical review letters, vol. 77, no. 18, p. 3865, 1996.pubmed.ncbi.nlm.nih.gov/10062328
  50. [50]G. Prandini, A. Marrazzo, I. E. Castelli, N. Mounet, and N. Marzari, “Precision and efficiency in solid-state pseudopotential calculations,” npj Computational Materials, vol. 4, no. 1, p. 72, 2018.DOI
  51. [51]C. Elsässer, M. Fähnle, C. Chan, and K. Ho, “Density-functional energies and forces with gaussian-broadened fractional occupations,” Physical Review B, vol. 49, no. 19, p. 13975, 1994.pubmed.ncbi.nlm.nih.gov/10010346
  52. [52]H. J. Monkhorst and J. D. Pack, “Special points for brillouin-zone integrations,” Physical review B, vol. 13, no. 12, p. 5188, 1976.
  53. [53]P. E. Blöchl, O. Jepsen, and O. K. Andersen, “Improved tetrahedron method for brillouin-zone integrations,” Physical Review B, vol. 49, no. 23, p. 16223, 1994.pubmed.ncbi.nlm.nih.gov/10010769
  54. [54]M. Kawamura, “Fermisurfer: Fermi-surface viewer providing multiple representation schemes,” Computer Physics Communications, vol. 239, pp. 197–203, 2019.DOI
  55. [55]F. D. Murnaghan, “The compressibility of media under extreme pressures,” Proceedings of the National Academy of Sciences, vol. 30, no. 9, pp. 244–247, 1944.pubmed.ncbi.nlm.nih.gov/16588651
  56. [56]F. Birch, “Finite elastic strain of cubic crystals,” Physical review, vol. 71, no. 11, p. 809, 1947.DOI
  57. [57]X. Wu, D. Vanderbilt, and D. Hamann, “Systematic treatment of displacements, strains, and electric fields in density-functional perturbation theory,” Physical Review B—Condensed Matter and Materials Physics, vol. 72, no. 3, p. 035105, 2005.DOI
  58. [58]A. Dal Corso, “Thermo-pw.” https://dalcorso.github.io/thermo_pw/, 2024. Accessed: 2026-07-24.
  59. [59]M. Born and K. Huang, Dynamical theory of crystal lattices. Oxford university press, 1996.
  60. [60]R. Gaillac, P. Pullumbi, and F.-X. Coudert, “Elate: an open-source online application for analysis and vi-DOI
  61. [61]R. Hill, “The elastic behaviour of a crystalline aggregate,” Proceedings of the Physical Society. Section A, vol. 65, no. 5, pp. 349–354, 1952.DOI
  62. [62]O. L. Anderson, “A simplified method for calculating the debye temperature from elastic constants,” Journal of Physics and Chemistry of Solids, vol. 24, no. 7, pp. 909–917, 1963.DOI
  63. [63]J. Noffsinger, F. Giustino, B. D. Malone, C.-H. Park, S. G. Louie, and M. L. Cohen, “Epw: A program for calculating the electron–phonon coupling using maximally localized wannier functions,” Computer Physics Communications, vol. 181, no. 12, pp. 2140–2148, 2010.
  64. [64]S. Poncé, E. R. Margine, C. Verdi, and F. Giustino, “Epw: Electron–phonon coupling, transport and superconducting properties using maximally localized wannier functions,” Computer Physics Communications, vol. 209, pp. 116–133, 2016.DOI
  65. [65]H. Lee, S. Poncé, K. Bushick, S. Hajinazar, J. Lafuente-Bartolome, J. Leveillee, C. Lian, J.-M. Lihm, F. Macheda, H. Mori, et al., “Electron–phonon physics from first principles using the epw code,” npj Computational Materials, vol. 9, no. 1, p. 156, 2023.arxiv.org/abs/2302.08085
  66. [66]N. Marzari and D. Vanderbilt, “Maximally localized generalized wannier functions for composite energy bands,” Physical review B, vol. 56, no. 20, p. 12847, 1997.DOI
  67. [67]I. Souza, N. Marzari, and D. Vanderbilt, “Maximally localized wannier functions for entangled energy bands,” Physical Review B, vol. 65, no. 3, p. 035109, 2001.DOI
  68. [68]G. Pizzi, V. Vitale, R. Arita, S. Blügel, F. Freimuth, G. Géranton, M. Gibertini, D. Gresch, C. Johnson, T. Koretsune, et al., “Wannier90 as a community code: new features and applications,” Journal of Physics: Condensed Matter, vol. 32, no. 16, p. 165902, 2020.DOI
  69. [69]P. Morel and P. Anderson, “Calculation of the superconducting state parameters with retarded electron-phonon interaction,” Physical Review, vol. 125, no. 4, p. 1263, 1962.
  70. [70]H. J. Choi, D. Roundy, H. Sun, M. L. Cohen, and S. G. Louie, “The origin of the anomalous superconducting properties of mgb2,” Nature, vol. 418, no. 6899, pp. 758–760, 2002.pubmed.ncbi.nlm.nih.gov/12181561
  71. [71]Y. Kong, O. Dolgov, O. Jepsen, and O. Andersen, “Electron–phonon interaction in the normal and superconducting states of mgb 2,” Physical Review B, vol. 64, no. 2, p. 020501, 2001.DOI
  72. [72]P. B. Allen, “Neutron spectroscopy of superconductors,” Physical Review B, vol. 6, no. 7, p. 2577, 1972.DOI
  73. [73]R. Dynes, “MacMillan’s equation and the tc of superconductors,” Solid State Communications, vol. 10, no. 7, pp. 615–618, 1972.DOI
  74. [74]P. B. Allen and R. Dynes, “Transition temperature of strong-coupled superconductors reanalyzed,” Physical Review B, vol. 12, no. 3, p. 905, 1975.DOI
  75. [75]E. R. Margine and F. Giustino, “Anisotropic migdal-eliashberg theory using wannier functions,” Physical Review B—Condensed Matter and Materials Physics, vol. 87, no. 2, p. 024505, 2013.DOI
  76. [76]F. Mouhat and F.-X. Coudert, “Necessary and sufficient elastic stability conditions in various crystal systems,” Physical review B, vol. 90, no. 22, p. 224104, 2014. G. Carlotti, G. Gubbiotti, F. Hicker­nell, H. Liaw, and G. Socino, “Comparative study of the elastic properties of polycrystalline aluminum nitride films on silicon by brillouin light scattering,” Thin Solid Films, vol. 310, no. 1-2, pp. 34–38, 1997.
  77. [78]G. Carlotti, N. Cherault, N. Casanova, C. Goldberg, and G. Socino, “Elastic constants of low-k and barrier dielectric films measured by brillouin light scattering,” Thin Solid Films, vol. 493, no. 1-2, pp. 175–178, 2005.DOI
  78. [79]D. Clerc and H. Ledbetter, “Mechanical hardness: A semimpirical theory based on screened electrostatics and elastic shear,” Journal of Physics and Chemistry of Solids, vol. 59, no. 6-7, pp. 1071–1095, 1998.DOI
  79. [80]C. Kral, W. Lengauer, D. Rafaja, and P. Ettmayer, “Critical review on the elastic properties of transition metal carbides, nitrides and carbonitrides,” Journal of Alloys and Compounds, vol. 265, no. 1-2, pp. 215–233, 1998.DOI
  80. [81]S. M. Kastuar, Z.-L. Liu, S. Najmaei, and C. E. Ekuma, “Mechanical properties of cubic boron nitride and diamond at dynamical pressure and temperature,” Applied Physics Letters, vol. 123, no. 23, 2023.DOI
  81. [82]C. A. Klein and G. F. Cardinale, “Young’s modulus and poisson’s ratio of cvd diamond,” Diamond and Related Materials, vol. 2, no. 5-7, pp. 918–923, 1993.
  82. [83]S. Pugh, “Xcii. relations between the elastic moduli and the plastic properties of polycrystalline pure metals,” The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, vol. 45, no. 367, pp. 823–843, 1954.DOI
  83. [84]X.-Q. Chen, H. Niu, D. Li, and Y. Li, “Modeling hardness of polycrystalline materials and bulk metallic glasses,” Intermetallics, vol. 19, no. 9, pp. 1275–1281, 2011.DOI
  84. [85]I. Esterlis, S. Kivelson, and D. Scalapino, “A bound on the superconducting transition temperature,” npj Quantum Materials, vol. 3, no. 1, p. 59, 2018.DOI
  85. [86]A. D. Smith, S. B. Harris, R. P. Camata, D. Yan, and C.-C. Chen, “Machine learning the relationship between Debye temperature and superconducting transition temperature,” Physical Review B, vol. 108, no. 17, p. 174514, 2023.DOI
  86. [87]M. Grim­discht, E. Zouboulis, and A. Polian, “Elastic constants of boron nitride,” Journal of applied physics, vol. 76, no. 2, pp. 832–834, 1994.DOI
  87. [88]S. I. B. Ud’ko, G. Papper­t, C. Petrovic, C. Cunningham, N. Anderson, and P. Canfield, “Boron isotope effect in superconducting mgb 2,” Physical Review Letters, vol. 86, no. 9, p. 1877, 2001.
  88. [89]L. Pauling, “The nature of the chemical bond. iv. the energy of single bonds and the relative electronegativity of atoms,” Journal of the American Chemical Society, vol. 54, no. 9, pp. 3570–3582, 1932.DOI
  89. [90]N. F. Mott, “Metal-insulator transition,” Reviews of Modern Physics, vol. 40, no. 4, p. 677, 1968.
  90. [91]I. Lifshitz et al., “Anomalies of electron characteristics of a metal in the high pressure region,” Sov. Phys. JETP, vol. 11, no. 5, pp. 1130–1135, 1960.
  91. [92]Y. M. Blanter, M. Kaganov, A. Pantsulaya, and A. Varlamov, “The theory of electronic topological transitions,” Physics Reports, vol. 245, no. 4, pp. 159–257, 1994.DOI
  92. [93]V. M. Pereira and A. Castro Neto, “Strain engineering of graphene’s electronic structure,” Physical review letters, vol. 103, no. 4, p. 046801, 2009.pubmed.ncbi.nlm.nih.gov/19659379
  93. [94]W. McMillan, “Transition temperature of strong-coupled superconductors,” Physical Review, vol. 167, no. 2, p. 331, 1968.
  94. [95]J. Hopfield, “Angular momentum and transition-metal superconductivity,” Physical Review, vol. 186, no. 2, p. 443, 1969.DOI
  95. [96]A. Damascelli, Z. Hussain, and Z.-X. Shen, “Angle-resolved photoemission studies of the cuprate superconductors,” Reviews of modern physics, vol. 75, no. 2, p. 473, 2003.DOI
  96. [97]H. Yang, A. Liang, C. Chen, C. Zhang, N. B. Schroeter, and Y. Chen, “Visualizing electronic structures of quantum materials by angle-resolved photoemission spectroscopy,” Nature Reviews Materials, vol. 3, no. 9, pp. 341–353, 2018.DOI
  97. [98]W. Kohn, “Image of the fermi surface in the vibration spectrum of a metal,” Physical Review Letters, vol. 2, no. 9, p. 393, 1959.DOI
  98. [99]A. Q. Baron, “Phonons in crystals using inelastic x-ray scattering,” arXiv preprint arXiv:0910.5764, 2009.arxiv.org/abs/0910.5764
  99. [100]L. Pintschovius, “Electron–phonon coupling effects explored by inelastic neutron scattering,” physica status solidi (b), vol. 242, no. 1, pp. 30–50, 2005.
  100. [101]J. Carbotte, “Properties of boson-exchange superconductors,” Reviews of Modern Physics, vol. 62, no. 4, p. 127, 1990.DOI
  101. [102]F. Marsiglio and J. Carbotte, “Strong-coupling corrections to bardeen-cooper-schrieffer ratios,” Physical Review B, vol. 33, no. 9, p. 6141, 1986.pubmed.ncbi.nlm.nih.gov/9939162
  102. [103]F. Marsiglio, “Eliashberg theory: A short review,” Annals of Physics, vol. 417, p. 168102, 2020.DOI
  103. [104]P. B. Allen and B. Mitrovic, “Theory of superconducting tc,” Solid state physics, vol. 37, pp. 1–92, 1983.DOI
  104. [105]M. Somayazulu, M. Ahart, A. K. Mishra, Z. M. Geballe, M. Baldini, Y. Meng, V. V. Struzhkin, and R. J. Hemley, “Evidence for superconductivity above 260 k in lanthanum superhydride at megabar pressures,” Physical review letters, vol. 122, no. 2, p. 027001, 2019.DOI
  105. [106]F. Gömöry, “Characterization of high-temperature superconductors by ac susceptibility measurements,” Superconductor Science and Technology, vol. 10, no. 8, pp. 523–542, 1997.DOI
  106. [107]S. Souma, Y. Machida, T. Sato, T. Takahashi, H. Matsui, S.-C. Wang, H. Ding, A. Kaminski, J. Campuzano, S. Sasaki, et al., “The origin of multiple superconducting gaps in mgb2,” Nature, vol. 423, no. 6935, pp. 65–67, 2003.pubmed.ncbi.nlm.nih.gov/12721624
  107. [108]P. Szabó, P. Samuel, J. Kačmarčík, T. Klein, J. Marcus, D. Fruchart, S. Miraglia, C. Marcenat, and A. Jansen, “Evidence for two superconducting energy gaps in mgb 2 by point-contact spectroscopy,” Physical review letters, vol. 87, no. 13, p. 137005, 2001.
  108. [109]Ø. Fischer, M. Kugler, I. Maggio-Aprile, C. Berthod, and C. Renner, “Scanning tunneling spectroscopy in high-temperature superconductors,” Reviews of Modern Physics, vol. 79, no. 1, pp. 353–419, 2007.DOI
  109. [110]D. Daghero and R. Gonnelli, “Probing multiband superconductivity by point-contact spectroscopy,” Superconductor Science and Technology, vol. 23, no. 4, p. 043001, 2010.DOI
  110. [111]S. Xie, Y. Quan, A. Hire, B. Deng, J. DeStefano, I. Salinas, U. Shah, L. Fanfarillo, J. Lim, J. Kim, et al., “Machine learning of superconducting critical temperature from eliashberg theory,” npj Computational Materials, vol. 8, no. 1, p. 14, 2022.arxiv.org/abs/2106.05235
  111. [112]V. Stanev, C. Oses, A. G. Kusne, E. Rodriguez, J. Paglione, S. Curtarolo, and I. Takeuchi, “Machine learning modeling of superconducting critical temperature,” npj Computational Materials, vol. 4, no. 1, p. 29, 2018.DOI
  112. [113]J. E. Moussa and M. L. Cohen, “Constraints on t c for superconductivity in heavily boron-doped diamond,” Physical Review B—Condensed Matter and Materials Physics, vol. 77, no. 6, p. 064518, 2008.
  113. [114]D. G. Schlom, L.-Q. Chen, C.-B. Eom, K. M. Rabe, S. K. Streiffer, and J.-M. Triscone, “Strain tuning of ferroelectric thin films,” Annu. Rev. Mater. Res., vol. 37, no. 1, pp. 589–626, 2007.DOI
  114. [115]I. Errea, M. Calandra, and F. Mauri, “Anharmonic free energies and phonon dispersions from the stochastic self-consistent harmonic approximation: Application to platinum and palladium hydrides,” Physical Review B, vol. 89, no. 6, p. 064302, 2014.DOI

Paper details

vlnc:W_27bw3f2s
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