Bond–electromagnetic origin of superconducting pairing across materials Authors: Min Tae Kim Published: 2026 ABSTRACT AbstractWe develop a bond–electromagnetic framework in which superconducting pairing originates from dynamic covalent bonds whose fluctuations generate a real–space pairing kernel. Bond–centered singlets emerge from these structural fluctuations, while long–range superconductivity appears when they become electromagnetically frozen into a phase–coherent network, as quantified by the London kernel and the superfluid stiffness. Within this framework, the microscopic pairing scale is determined directly from experimentally accessible structural quantities—including bond geometry, hybridization symmetry, dielectric screening, and the logarithmic bond–frequency moment—through a structural reformulation of the Allen–Dynes expression. The conversion of into the observed superconducting transition temperature is then governed by the electromagnetic rigidity of the material. Using only structural inputs, the framework reproduces pairing scales across representative materials and coupling regimes. For elemental Al, is obtained in a structurally parameter–free manner, while for Pb it is constrained solely by independently established ranges of the coupling strength. To probe structure–driven pairing in systems with reduced phase stiffness, we analyze FeSe under hydrostatic pressure. A structurally constrained form naturally captures the pressure–induced rise, maximum, and curvature of the superconducting dome, with the resulting trajectory spanning the experimentally observed transition region from the resistive onset to the zero–resistance state without invoking pressure–dependent spectroscopic input. These results establish a structure–anchored origin of superconducting pairing across distinct materials classes and demonstrate that experimentally accessible bond dynamics provide a practical materials–level criterion for linking lattice structure, electromagnetic response, and superconducting coherence.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1038/s41598-026-55635-1.Subject terms: Materials science, Physics Introduction Superconductivity—the emergence of zero resistance and perfect diamagnetism below a critical temperature —is a paradigmatic form of macroscopic quantum order. For many elemental metals, the Bardeen–Cooper–Schrieffer framework and its Eliashberg extension provide a quantitatively successful description of pairing and electrodynamic response in terms of electron–phonon interactions1–5. The discovery of unconventional materials—including cuprates6,7, iron-based compounds8,9, superconducting hydrides at megabar pressures10,11, and nickelates12—has broadened this landscape and highlighted that superconductivity can arise across a wide range of coupling strengths, dimensionalities, and electronic environments. Across these varied systems, numerous empirical observations point to a central role of bonding geometry and orbital hybridization in determining pairing. In cuprates, epitaxial strain or lattice distortions shift by unusually large amounts13, while in iron pnictides, hydrostatic or chemical pressure strongly modifies Fe–As bond angles and the resulting superconducting properties14. Stoichiometric FeSe provides a particularly clean setting: hydrostatic pressure tunes its superconductivity almost entirely through changes in Fe–Se bond geometry and layer spacing, without introducing chemical disorder15,16. Further indicators of bond–governed behavior appear across both conventional and unconventional materials. Isotope effects frequently deviate from the canonical behavior, revealing sensitivity beyond simple phonon frequency scaling. Superfluid-density systematics—including Uemura scaling and Homes’ law17–19—suggest that structural geometry, coherence, and electrodynamics are tightly linked. Classical ultrasonic anomalies below 20,21 demonstrate that bond stiffness and electromagnetic response evolve together, indicating that superconductivity may involve not only quasiparticle pairing but also a real-space reorganization of valence bonds. These observations motivate a formulation in which the microscopic degrees of freedom include the dynamic rearrangement of covalent bonds—specifically their geometry, hybridization symmetry, dielectric environment, and stiffness. Consistent with this viewpoint, related approaches have shown that electron–phonon interactions can be described in terms of bond-length modulations and local structural distortions within hopping- and bond-resolved frameworks22–24. These studies indicate that the coupling between lattice dynamics and electronic states can be naturally understood at the level of real-space bonds. Building on this perspective, the present work develops a structural description of pairing in which the relevant microscopic degrees of freedom are associated with bond-resolved lattice dynamics. In this approach, a Cooper pair is interpreted as a dynamic covalent bond: a time-dependent overlap of valence orbitals whose oscillatory interaction acts as a bond–energy mode. Superconductivity corresponds to the regime in which these dynamic bonds acquire long-range phase coherence, forming an electromagnetically frozen network consistent with Meissner electrodynamics25–27. Formally, the Eliashberg kernel is expressed in terms of directly measurable structural quantities—orbital overlap, hybridization symmetry, dielectric screening, and bond spectral stiffness—providing a transparent real-space interpretation of the effective coupling parameters. To access this quantity, we construct the bond spectral density B(Ω ), where Ω denotes the bond-resolved vibrational frequency, from symmetry-resolved lattice dynamics—obtainable through Raman or inelastic x-ray scattering—together with measured structural and elastic parameters. These experimentally anchored inputs define a bond–electromagnetic kernel that characterizes the pairing interaction without phenomenological fitting and is applicable to weak-coupling metals such as Al, strong-coupling systems such as Pb, and stiffness-limited compounds such as FeSe. The goal is to identify how measurable bond geometry and mode symmetry encode the microscopic scale at which pairing initiates. In the remainder of this work we establish the formal structure of this kernel and analyze its microscopic components from a materials-oriented perspective. Applications to representative systems spanning distinct coupling regimes—Al (weak coupling), Pb (strong coupling), and FeSe (pressure-tunable, stiffness-limited)—demonstrate how experimentally accessible structural information provides a materials-level route to determining the pairing scale across diverse families of superconductors. A full treatment of superfluid stiffness and vortex dynamics is deferred to a companion study. Bond–electromagnetic framework for superconductivity Superconductivity is the emergence of an electromagnetically rigid state, in which the internal fields satisfy and in the bulk and screening currents flow without dissipation within a London layer28–30. In the present framework this rigidity does not arise from quasiparticle kinematics but from the collective behaviour of dynamic covalent bonds: time–dependent electromagnetic bonds between neighbouring valence orbitals. Before presenting the formal construction, Fig. 1 summarizes the structural–electromagnetic pathway from crystal geometry to the microscopic pairing scale. Conceptual structure of the bond–electromagnetic kernel. Local bond geometry (bond length, hybridization symmetry, elastic stiffness) determines the dynamics of dynamic covalent bonds. Their collective fluctuations define the bond spectral density B(Ω ) as a structurally projected vibrational spectrum, which is constructed by projecting phonon modes with frequencies and polarization vectors onto bond directions via weights . The resulting spectral weight and characteristic logarithmic frequency together determine the microscopic pairing scale . Electromagnetic freezing, interpreted as the onset of phase coherence of bond-centered singlets, gives rise to the superconducting state. Structural origin of dynamic covalent bonds Local bond geometry (bond length, orbital overlap, dielectric screening, and hybridization symmetry) sets the stiffness of dynamic covalent bonds— the elementary units of electromagnetic energy exchange in a metal. Unlike conventional electron–phonon pictures, the fluctuating bond is itself an electromagnetic object whose dynamics depends only on structural inputs. The collective motion of these dynamic bonds defines a structural spectral density B(Ω ), a real-space generalisation of the Eliashberg function 2–4. To make this definition explicit, the bond spectral density can be constructed from phonon normal modes. For a mode m with frequency and polarization vector , we define the projection onto a bond direction as The bond spectral density is then given by In practice, this construction can be carried out using experimentally measured phonon spectra or first-principles phonon calculations, where both and are available. In the isotropic monatomic limit, the projection factor averages to unity, so that B(Ω ) reduces to the phonon density of states in this limit, thereby continuously connecting to the conventional Eliashberg spectral description. Within the conventional electron–phonon framework, the spectral function combines lattice vibrational phase space with electronic matrix elements. In the present formulation, we explicitly construct the structural component of this interaction by defining a bond-resolved spectral density B(Ω ) from phonon eigenmodes. This construction is consistent with bond-resolved descriptions of electron–phonon coupling, in which the interaction originates from bond-length modulations that affect electronic hopping amplitudes or local potentials22–24. Accordingly, B(Ω ) captures the vibrational phase space associated with such bond distortions and establishes a direct correspondence at the level of spectral moments, while the full electronic weighting is not introduced at the same level of microscopic detail (see Supplementary Information). In the harmonic and isotropic limit, where bond geometry becomes uniform and mode-dependent screening is absent, B(Ω ) reduces to a phonon-density-of-states-like distribution, thereby recovering the same leading spectral moments that enter the Eliashberg formulation, consistent with the standard strong-coupling framework31,32. From B(Ω ) to the microscopic pairing scale As derived in Supplementary Information (SI) Sec. 1, the structural bond spectrum B(Ω ) yields two symmetry–resolved spectral moments that play the role of microscopic coupling parameters: Here measures the integrated strength of bond–fluctuation modes, while captures the effective geometric frequency of the structural spectrum. Both quantities are computed directly from experimentally determined phonon dispersions, elastic constants, or Raman/IXS mode symmetries. The present formulation is best understood as a structural continuation of the Eliashberg framework, in which the structural contribution to pairing is isolated while remaining consistent with established electron–phonon descriptions. The use of an Allen–Dynes–type functional form is justified by the fact that, in Eliashberg theory, the superconducting scale is governed primarily by low-order spectral moments of the bosonic kernel, which determine its asymptotic exponential structure. In this sense, the functional form is controlled by spectral moments, not by the detailed microscopic representation of the interaction. In the present framework, B(Ω ) provides the corresponding structural spectrum, which reduces to a phonon-density-of-states-like distribution in the harmonic limit and reproduces the same leading moments. Accordingly, and enter in direct analogy, supporting the retention of the Allen–Dynes expression as a moment-controlled approximation. These spectral moments enter a structurally determined Allen–Dynes–type expression for the microscopic pairing scale, derived in SI Sec. 1. Here denotes the explicit Allen–Dynes functional form, with all inputs replaced by their bond–resolved structural counterparts. Because this mapping introduces no fitted electron–phonon matrix elements or material-specific tuning parameters, represents a structurally grounded prediction of the microscopic singlet–formation scale. In elemental close–packed metals such as aluminum (Al) and lead (Pb), the large superfluid stiffness and nearly isotropic coherence length imply that macroscopic phase ordering occurs essentially simultaneously with pair formation. As a result, the experimentally observed transition temperature satisfies to within a few percent, consistent with London rigidity and the absence of strong phase–fluctuation effects in three–dimensional metals5,29,30,33. Cooper pairs as dynamic covalent bonds In this structural viewpoint, a Cooper pair corresponds to a dynamic covalent bond of singlet character, whose phase rigidity governs electromagnetic response. In this work, we use the term “electromagnetic freezing” to denote the onset of macroscopic phase coherence of bond-centered singlet correlations, governed by the emergence of finite superfluid stiffness (London rigidity). This distinguishes it from the microscopic pair-formation scale , which may remain higher than the observed superconducting transition temperature in systems with reduced phase stiffness, and may manifest as a true phase transition or as a crossover depending on dimensionality and phase stiffness. As temperature is lowered, dynamic bonds lose their temporal fluctuations and lock into a macroscopically ordered state—a process corresponding to electromagnetic freezing as defined above. Directional bond networks, in which singlet correlations are tied to specific real–space bond orientations, naturally generate alternating–sign gap structures—including d–wave–like patterns—consistent with the experimentally observed order–parameter symmetries in cuprates7 and with theoretical frameworks emphasizing real–space singlet formation34,35. Bond–electromagnetic kernel Figure 1 outlines the structural pathway leading from local bond geometry to the microscopic pairing scale. To make this connection explicit, we introduce the bond–electromagnetic kernel, which provides the momentum–frequency representation of the interaction mediated by dynamic covalent bonds. Whereas the structural spectrum B(Ω ) characterizes the real–space fluctuations of these bonds, the kernel specifies how such fluctuations generate an effective pairing interaction across momenta and energies. As derived in SI Sec. 2, fluctuations of dynamic covalent bonds generate a pairing interaction of the form where and Ω denote momentum and frequency, respectively. The first term describes the dynamically screened Coulomb repulsion, with the bare Coulomb interaction and the dielectric function. The second term represents an attractive contribution mediated by bond–energy fluctuations, encoded in the symmetry–resolved bond propagator . Here β is an effective bond–electromagnetic coupling constant that characterizes the coupling between bond distortions and electronic energy, with its cumulative effect captured by the spectral moment defined in Eq. (1). The short–range component of is particularly important, because pair formation in real materials is strongly influenced by local bond geometry. Projecting onto real space therefore defines an effective structural exchange scale , whose magnitude depends on the bond length a, the local dielectric environment \varepsilon, and the coordination number ν. Within the present framework, this exchange scale acts as the microscopic driver for bond–centered singlet formation. In the harmonic limit of a monatomic crystal, reduces smoothly to the usual phonon spectral function, and recovers the conventional Eliashberg interaction in this limit5,32. Strong–coupling theory thus emerges as the infrared (low–energy, coarse–grained) limit of a more general bond–resolved description, obtained when local geometric and symmetry information is averaged out. Extension to high- families Although the present work benchmarks the framework on elemental low- metals—where the large phase stiffness enforces —the bond–electromagnetic kernel provides a structural perspective that may be extended to more complex superconducting systems. The present work does not explicitly treat cuprates, pnictides, nickelates, or hydrides, where strong electronic correlations can play a dominant role. Accordingly, the current results should be understood as establishing a structural pairing scale within a lattice-based framework. Any extension to strongly correlated systems, including a detailed formulation of the phase-ordering scale and its material dependence, remains an open direction for future investigation. In these high- families, bond-centered singlet correlations— identified here with the microscopic pairing scale — may emerge at temperatures well above the scale at which long-range phase coherence is established17,33. The resulting reduction of superfluid stiffness, relative to three-dimensional elemental metals, then naturally leads to a separation reflected empirically in Uemura scaling17 and Homes’ law18. The structural pairing scale therefore remains a well-defined and meaningful quantity even when superconductivity is limited by phase ordering rather than pair formation. Results Pairing scale extracted from structural bond dynamics The microscopic condensation scale is evaluated using the bond–electromagnetic kernel developed in the previous Theory section, with the bond spectral density B(Ω ) reconstructed directly from structural inputs (phonon dispersions, elastic constants, or Raman/IXS modes). From B(Ω ) we obtain the bond moments and as defined in Eq. (1)–(2). To convert these structural quantities into a pairing scale, we reorganize the conventional Allen–Dynes expression so that all coupling parameters are replaced by their bond–resolved counterparts. This yields the structural pairing estimator where the logarithmic bond frequency is converted to temperature units through and and are the standard Allen–Dynes correction factors (defined in SI, Sec. 1). The screened Coulomb pseudopotential is taken in the standard Morel–Anderson range –0.13 for simple metals36. It is treated as a fixed physical input within its conventional range, not as a material-specific fitting parameter. The Allen–Dynes correction factors and are likewise used without adjustment. In this sense, “parameter-free” refers to the absence of adjustable parameters in the structural construction of the spectral moments, while external quantities such as are fixed within their established physical range. In the harmonic and isotropic limit, the bond spectral density B(Ω ) reduces continuously to the Eliashberg function , so that and recover their conventional definitions. Equation (6) therefore reduces to the standard Allen–Dynes expression, ensuring consistency with the established framework in the appropriate limit. For simple elemental metals with high carrier density and broadly dispersive electronic bands, the electromagnetic phase stiffness is typically large, so electromagnetic freezing occurs immediately upon pair formation. Accordingly , enabling a direct comparison between predicted pairing scales and measured transition temperatures, as demonstrated for Al in SI Sec. 3. In Pb the phonon density of states (DOS) still fixes the structural frequency scale, but the overall coupling strength carries an electronic prefactor, so the prediction becomes semi–structural: is determined by lattice dynamics alone, whereas must be constrained independently as shown in SI Sec. 4. FeSe provides a third and complementary test. Although its absolute ambient–pressure pairing scale is not fully structural, the pressure derivatives of the bond kernel are determined directly by the hardening of Fe–Se modes under compression. Consequently the structural form reproduces the observed superconducting dome15,37. Detailed procedures are given in SI, Sec. 5. This quadratic exponential form arises naturally from a second-order expansion of the structurally defined Allen–Dynes expression after linearizing the pressure dependence of the bond spectral moments and . At low pressure, increasing Fe–Se bond stiffness and hybridization enhance the effective coupling, producing a positive linear term, while the intrinsic concavity of the Allen–Dynes exponent generates a negative quadratic correction, yielding and . This structural pairing scale can then be directly compared with experimentally defined transition temperatures. The experimentally observed separation between the resistive onset and zero-resistance temperatures reflects phase fluctuations and vortex-mediated phase ordering, which are pronounced in reduced-stiffness systems such as FeSe17,33. In this context, the structural pairing scale naturally lies between these two temperatures, consistent with a separation between pair formation and macroscopic phase coherence. For clarity, we emphasize that the present material set serves as a proof-of-principle benchmark that establishes the framework across representative regimes, and that the pressure dependence in FeSe follows directly from the structure of the pairing expression. Fully structural prediction: elemental Al Among elemental superconductors, Al provides the most stringent benchmark for a structural pairing theory. As a monatomic, nearly harmonic fcc solid with an extensively measured phonon spectrum, Al allows the microscopic pairing scale to be computed entirely from experimental lattice dynamics, without electronic matrix elements, tunneling inversion, or any adjustable parameters. In contrast to multicomponent or anisotropic materials, the full phonon DOS g(Ω ) reported in classic neutron–scattering measurements by Stedman and coworkers 38,39 already encodes all structural quantities relevant for the bond spectrum: bond geometry, elastic stiffness, and mode symmetries. For a monatomic fcc lattice with isotropically distributed phonon polarizations, the branch–summed structural projection factor averages identically to unity, so that the bond spectrum satisfies with no normalization or fitting (see SI Sec. 3). This makes Al a parameter–free verification case within the structural construction of the spectral moments. Using the digitized longitudinal and transverse components of the experimental DOS, the structural Eliashberg moments are evaluated via the definitions in Eq. (1)–(2): with the latter corresponding , in excellent agreement with canonical Eliashberg analyses of Al. These values follow directly from the DOS and involve no model dependence. Substituting these structural moments into the pairing expression, Eq. (6), with a standard screened Coulomb range –0.14 gives Experimentally, Al exhibits and because phase stiffness is large in a three–dimensional metal, holds to within a small quantitative deviation expected from the conventional Coulomb pseudopotential range. Thus the structural calculation reproduces the superconducting scale within the expected accuracy of the Allen–Dynes framework, matching the accuracy of conventional Eliashberg theory despite using only the phonon DOS. Elemental Al thus establishes the internal consistency of the bond–electromagnetic framework in the monatomic, harmonic limit: when bond geometry and lattice dynamics are fully specified, the resulting structural kernel yields the microscopic pairing scale without recourse to electronic matrix elements or phenomenological input. This case serves as a baseline against which more complex, electronically renormalized systems can be meaningfully assessed. Semi–structural analysis: Pb Pb provides a complementary strong–coupling test of the structural pairing framework. In contrast to Al, however, the absolute electron–phonon coupling strength in Pb cannot be fixed in a fully parameter–free manner from lattice dynamics alone. This is because the relevant electronic prefactor—commonly expressed through the Hopfield parameter or an equivalent deformation potential—depends on electronic matrix elements that are not directly accessible from neutron–scattering measurements of the phonon spectrum5,31,32. To isolate what is accessible strictly from structure, we deliberately avoid any use of electronic matrix elements or tunneling inversion data and work only with the phonon information reported by Stedman et al. 39; full details are given in SI Sec. 4. From Figure 2 of Ref. 39, we digitize the resolved longitudinal and transverse components of the phonon density of states, , , and , and construct the total DOS For monatomic fcc Pb at low temperature, phonon polarization directions are approximately isotropic with respect to any nearest–neighbour bond. As a result, the bond spectral density inherits the shape of the measured phonon DOS, with the overall proportionality constant cancelling identically from all normalized spectral moments, so that only the frequency distribution (i.e. the shape of the DOS) enters and . Using the digitized dataset and the same trapezoidal integration scheme employed for Al, the logarithmic bond–frequency moment defined in Eq. (2) yields This value lies within the range expected from strong–coupling analyses based on tunneling and thermodynamic measurements and demonstrates that the characteristic phonon energy scale entering the pairing kernel of Pb is a genuinely structural quantity fixed by the measured phonon density of states. By contrast, the overall magnitude of the bond spectrum contains an additional electronic prefactor reflecting strong screening effects in Pb. As a consequence, the dimensionless coupling constant cannot be fixed from phonon DOS data alone. We therefore treat the coupling strength λ as an independently constrained electronic parameter, while its characteristic frequency scale is supplied structurally by the bond kernel. Adopting the experimentally established strong–coupling range consistent with tunneling and heat–capacity analyses, and inserting the structurally determined value into Eq. (6) yields in excellent agreement with the experimental transition temperature , as established by tunneling spectroscopy and thermodynamic measurements3,5,40. Pb therefore serves as a semi–structural testbed for the present framework: the logarithmic phonon scale is fixed entirely by structural bond dynamics, whereas the absolute coupling strength is supplied by independently established electronic information. Together with the structually parameter–free case of Al, this demonstrates that bond spectra extracted from lattice dynamics provide a consistent structural backbone for strong–coupling superconductivity in elemental metals, without recourse to tunneling inversion or additional material–specific fitting. Pressure–tunable pairing in FeSe FeSe offers a clean test of the structural pairing mechanism because hydrostatic pressure modifies mainly the Fe–Se bond geometry and screening. As shown in SI Sec. 5, two robust structural responses dominate under compression: (i) hardening of the Fe–Se optical modes and (ii) enhanced Fe–Se hybridization due to the large compressibility. Both effects increase the bond moments and therefore raise the pairing scale at low pressure. Expanding the structurally defined Allen–Dynes pairing scale to second order in pressure yields the generic form where the coefficients and are not obtained by fitting to the experimental curve, but are constrained by explicit functions of the linear pressure derivatives of and , as derived in SI Sec. 5. The signs and follow directly from experimentally established structural trends in FeSe, including the hardening of Fe–Se optical modes under pressure, the large lattice compressibility, and the intrinsic concavity of the Allen–Dynes exponent. Using representative structural derivatives consistent with available phonon and elastic data, the resulting pairing trajectories cluster into a narrow structural envelope, with of order 10 K and modest positive curvature. Figure 2 shows a representative structural trajectory ( K, GPa, GPa) together with the experimentally determined onset and zero–resistivity transition temperatures reported by Mizuguchi et al. and Okabe et al.41,42. The red dashed curve illustrates a structurally constrained pairing scale, rather than a point–by–point fit to either experimental criterion. Pressure-dependent pairing scale in FeSe. Experimental data from Mizuguchi et al. (open symbols) and Okabe et al. (filled symbols). The red dashed curve shows a structural pairing trajectory using with K, GPa, GPa. While does not coincide with a single experimental definition of , its magnitude and pressure evolution systematically span the transition region between the initial resistive onset and the fully phase–coherent zero–resistance state. This behavior is consistent with the interpretation of FeSe as a stiffness–limited superconductor, in which pair formation and macroscopic phase coherence occur at distinct temperature scales. Table 1 summarizes the status of the bond–electromagnetic pairing analysis across the three representative superconductors studied in this work. Al serves as a fully structural benchmark in which the measured phonon DOS alone determines , , and without any electronic input. In Pb the phonon DOS fixes the structural frequency scale, while the overall coupling strength remains an electronic quantity that must be taken from independent strong–coupling analyses. Table 1Structural pairing scales in representative superconductors. The bond–electromagnetic framework yields a fully parameter–free prediction of the microscopic pairing scale in monatomic Al from the measured phonon density of states. In Pb, lattice dynamics fix the structural frequency scale, while the absolute coupling strength is constrained by independent electronic information. In FeSe, the pressure evolution of the structurally defined pairing scale reproduces the overall location, width, and curvature of the superconducting dome without fitting to the experimental curve, systematically spanning the experimentally observed transition region bounded above by the resistive onset temperature and below by the zero–resistance state under hydrostatic pressure.MaterialStructural input (rad s)Structural levelAlfull phonon DOS g(Ω )0.511.8–3.5 K1.18 Kfully structuralPbphonon DOS —reproduces at –1.557.2 Ksemi–structuralFeSeP–dependent bond kernel——–36 K–40 Kstructural P response Structural pairing scales in representative superconductors. The bond–electromagnetic framework yields a fully parameter–free prediction of the microscopic pairing scale in monatomic Al from the measured phonon density of states. In Pb, lattice dynamics fix the structural frequency scale, while the absolute coupling strength is constrained by independent electronic information. In FeSe, the pressure evolution of the structurally defined pairing scale reproduces the overall location, width, and curvature of the superconducting dome without fitting to the experimental curve, systematically spanning the experimentally observed transition region bounded above by the resistive onset temperature and below by the zero–resistance state under hydrostatic pressure. FeSe provides a complementary test in which the pressure evolution of the structural bond kernel yields a pairing dome that reproduces the experimental profile without fitting to the experimental curve. Discussion The present work demonstrates that superconducting pairing can be formulated within a structurally anchored bond–electromagnetic framework. In this picture, pairing correlations emerge from fluctuations of dynamic covalent bonds, while macroscopic superconductivity appears when these bond-centered singlets become electromagnetically frozen into a phase-coherent state. This freezing manifests as a long-wavelength electromagnetic response consistent with London–Meissner electrodynamics28–30. In the harmonic and isotropic limit, the framework connects smoothly to conventional Eliashberg theory5,32, while making explicit the structural origin of the pairing interaction. The real-space spectral density B(Ω ) provides a direct structural link between bond-level dynamics and the microscopic pairing scale, enabling a unified description of pairing and electromagnetic response5,32. A central outcome of the present analysis is that the microscopic pairing scale can be anchored to experimentally accessible structural quantities, most notably the bond spectral density B(Ω ). Across the three representative systems studied here—Al, Pb, and FeSe—the framework reveals a clear hierarchy of structural control: (i) in Al, the pairing scale is fully determined by structure alone; (ii) in Pb, the characteristic frequency scale is fixed structurally, while the overall coupling strength remains constrained by independent electronic input; and (iii) in FeSe, hydrostatic pressure modifies superconductivity primarily through changes in bond geometry and lattice stiffness, allowing the pressure evolution of to be reproduced without fitting to the experimental curve from structurally constrained parameters. Together, these cases clarify which aspects of superconductivity are genuinely structural and establish B(Ω ) as a unifying object controlling the microscopic pairing scale. A broader implication concerns systems with reduced superfluid stiffness, including many high–temperature and low–dimensional superconductors. In such materials, the superconducting transition temperature is no longer set by pair formation alone but is limited by the onset of global phase coherence17,18,33. The structurally defined pairing scale identified here therefore provides a well-defined microscopic input for a complementary theory of the phase-ordering scale , in which electromagnetic rigidity and bond stiffness govern macroscopic coherence. Because the bond kernel depends only on real-space geometry and vibrational spectra, the same structural–electromagnetic framework is expected, within the limits discussed above, to extend to layered, anisotropic, and strongly correlated materials. Overall, the bond–electromagnetic framework offers a falsifiable and quantitatively predictive structural foundation for superconducting pairing. Its consistency across weak–coupling (Al), strong–coupling (Pb), and stiffness–limited (FeSe) systems suggests that pair formation is governed by a common structural basis, while the observed diversity of superconducting transition temperatures reflects differences in electromagnetic rigidity and phase-coherence constraints. By providing a materials-level route to linking lattice structure, bond dynamics, and superconducting coherence, the present approach establishes a practical framework for comparative analysis and structure-guided exploration of superconductivity across materials classes. While broader validation across additional material classes remains an important direction for future work, the present results establish a consistent structural benchmark across representative regimes. Methods Bond–electromagnetic kernel Pairing–scale predictions are obtained from the bond–electromagnetic framework introduced in the main text. The bond spectral density B(Ω ) is constructed by projecting lattice vibrations onto nearest–neighbour bond directions, providing a structural generalization of the Eliashberg function. The corresponding bond moments and the pairing scale follow directly from B(Ω ) as defined in Eqs. (1)–(2). Full details are provided in SI Sec. 1. Structural inputs All structural inputs are taken from experimental phonon and elastic data reported in the literature. When phonon eigenvectors are available (e.g., Al), they are projected directly onto bond directions; when only phonon density–of–states data are available (e.g., Pb), published DOS curves are digitized to construct B(Ω ). No ab initio calculations or tunneling inversions are employed, and no adjustable parameters enter the structural construction of B(Ω ). Details are given in SI Secs. 3–4. Pressure–dependent structural response in FeSe The pressure evolution of the bond kernel in FeSe is inferred from experimentally established trends in Fe–X bond hardening and lattice compressibility. These structural inputs determine the pressure dependence of the bond moments and lead to the generic form , with parameter ranges constrained by structural considerations rather than fitted directly to experimental data. Details are provided in SI Sec. 5. Supplementary Information Below is the link to the electronic supplementary material. Supplementary file 1 Author contributions M.T.K. conceived the study, developed the theoretical framework, performed the analysis, and wrote the manuscript. Data availability All data supporting the findings of this study are available from the published literature cited in the manuscript. Derived data and analysis codes are available from the author upon reasonable request. Declarations Acknowledgements The author gratefully acknowledges Hyung Pyo Kwon, an SI Auditor, for valuable advice and support related to this work. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Change history 9/11/2026 The original online version of this Article was revised: In the original version of this Article the Supplementary Information file was omitted. The Supplementary Information file now accompanies the original Article. References Associated Data Supplementary Materials Supplementary file 1 Data Availability Statement All data supporting the findings of this study are available from the published literature cited in the manuscript. Derived data and analysis codes are available from the author upon reasonable request.