Quantitative theory of electrical conduction in transition metals: the critical-region magnetoresistance of nickel and the Cu-Ni (constantan) resistance maximum left open by Mott's 1936 two-band model

Mott's 1936 two-band theory attributed the high resistivity of the transition metals Ni, Pd and Pt and of their Cu-Ni alloys to s-d scattering, the transitions of s-like conduction electrons into unoccupied d states, and explained qualitatively the sublinear high-temperature resistance of Pd and Pt, the resistance anomaly of nickel at its Curie point, and the resistance maximum of constantan. Its simplifying assumptions (parabolic d band, equalized spin currents, Einstein phonons with elastic scattering, order-of-magnitude matrix elements, neglected five-fold d-band degeneracy) left quantitative failures, chiefly the magnetoresistance coefficient of nickel near the Curie point, predicted as $b = 1/9 \approx 0.11$ against the corrected experimental $b \approx 0.5$, and the absence of any computation of the constantan maximum. Later literature resolved much of the program by framework replacement: the two-current model with spin mixing, critical-dynamics treatments of the Curie region, and first-principles (density-functional) spin-disorder and electron-phonon transport on realistic band structures. What survives is stated explicitly in current work: a purely theoretical, quantitative description of the resistivity of magnetic materials at finite temperature remains elusive. The open core is a parameter-free, first-principles account of (a) the field- and temperature-dependent magnetoresistance of nickel through the ferromagnetic critical region and (b) the resistance-temperature curves of Cu-Ni alloys including the position and height of the constantan maximum.

Background

The transition metals Ni, Pd and Pt owe their transport anomalies to an electronic structure in which a broad s-like conduction band overlaps a narrow, five-fold orbitally degenerate d band, so that the Fermi surface (the constant-energy surface of the occupied electron states in crystal-momentum space) intersects both bands. In his 1936 two-band theory, Mott explained the relatively high resistivity of these metals by s-d scattering: a conduction electron, which is mainly s-like, can be scattered by a displaced atom (a lattice vibration) into an unoccupied d state, and because the density of d states $N_d(E)$ at the Fermi energy is large, this probability exceeds ordinary s-s scattering several-fold. Since the unoccupied d states also carry the ferromagnetism or high paramagnetism of these elements, magnetic and transport properties are directly linked.

Mott's quantitative apparatus rested on simplifications: a parabolic d band with $N_d(E) \propto \sqrt{E_0 - E}$ near the band edge $E_0$; scattering matrix elements $A_{ss}(\theta)$ and $A_{sd}(\theta)$ assumed isotropic, of the same order, and not evaluated from first principles; an Einstein model of lattice vibrations (each atom vibrating independently at a single frequency) with the exchange of energy between electrons and phonons neglected, permissible only at temperatures well above the Debye temperature $\Theta$ (the temperature scale below which quantized, dispersive phonons and inelastic electron-phonon scattering qualitatively change the resistivity, the Bloch-Gruneisen regime); and, for nickel, conduction currents in the two spin channels treated as equalized. With these, the theory gave a $T^3$-type downward deviation from the linear high-temperature resistance of Pd and Pt, a sharp change of nickel's temperature coefficient at the Curie point (the temperature at which ferromagnetic order disappears) with a resistance decrease in an applied field, and a qualitative explanation of the resistance maximum of Cu-Ni alloys such as constantan, the alloy used for its near-zero temperature coefficient at room temperature.

The quantitative failures were explicit. For the magnetoresistance of nickel near the Curie point, writing $\Delta R/R = -b\,\Delta(\sigma^2)/\sigma_\infty^2$ with $\sigma$ the magnetization and $\sigma_\infty$ its saturation value, the two-band model gives $b = 1/9 \approx 0.11$, while the experimental value, corrected for the true (spontaneous) magnetization via the magnetocaloric effect, is $b \approx 0.5$. The specific heats computed from the same $N_d(E)$ model came out higher than the observed atomic heats of Pd and Pt. The constantan discussion was explicitly qualitative, with no computation of the position or height of the maximum. The five-fold d-band degeneracy was mentioned but not included in the calculation.

Later work transformed the framework rather than repairing it. The two-current model, in which the two spin orientations carry largely independent currents coupled by spin-flip (spin-mixing) scattering, replaced the equalized-spin-current assumption (Fert 1969; Fert and Campbell 1976) and became the standard language of ferromagnetic transport and of spintronics. Critical-dynamics work showed that the resistivity anomaly near $T_C$ reflects short-range spin correlations, not the squared magnetization alone (Fisher and Langer 1968). Modern transport calculations start from density-functional band structures with first-principles matrix elements, treating thermal spin disorder in the disordered-local-moment (DLM) picture (thermal spin disorder modeled as a random alloy of locally fixed magnetic moments) within the Korringa-Kohn-Rostoker coherent-potential approximation (KKR-CPA, a mean-field treatment of substitutional disorder in alloys), with the conductivity evaluated by the Kubo-Greenwood linear-response formula; electron-phonon resistivity, including the inelastic low-temperature regime, is likewise now computed from first principles. These advances realize the 1936 requests for realistic electronic structure and computed matrix elements, and they reproduce the zero-field spin-disorder resistivity of Fe well (Ni less well, because of local-moment collapse). What they do not deliver is a parameter-free quantitative account of the two benchmarks that failed in 1936: the field-dependent magnetoresistance of nickel through the ferromagnetic critical region, and the resistance-temperature curves of Cu-Ni alloys including the constantan maximum.

Problem Statement

Construct a quantitative, predictive theory of electrical conduction in the transition metals Ni, Pd and Pt and their Cu-Ni-type alloys, starting from realistic electronic structure, hybridized s and d bands with allowance for the five-fold orbital degeneracy of the d shell and scattering matrix elements evaluated from the band wavefunctions rather than assumed of the same order, and removing, where they matter for the target observables, the further simplifications acknowledged in the 1936 source: the Einstein phonon model with electron-phonon energy exchange neglected, valid only above the Debye temperature $\Theta$, and the equalization of the two spin currents, in favour of spin-dependent s-d scattering with spin mixing; and demonstrate quantitative agreement, without fitted scattering parameters, with the two targets that the 1936 source itself leaves open, each compared separately against the measured data within its experimental uncertainty: (1) the field-dependent magnetoresistance of ferromagnetic nickel through the critical region near and above its Curie point, that is, the field- and temperature-resolved resistivity $\rho(H,T)$ over nonzero field (the coefficient $b$ being the critical-region anchor, not the whole requirement), where the two-band model predicts $\Delta R/R = -b\,\Delta(\sigma^2)/\sigma_\infty^2$ with $b = 1/9 \approx 0.11$ against the corrected experimental $b \approx 0.5$ obtained via the magnetocaloric effect, and where the source explicitly declines to give a quantitative theory of the zero-field resistance-temperature curve, that zero-field curve being the already-computed, context-only comparison identified below; and (2) the resistance-temperature curves of Cu-Ni alloys, including the position and height of the maximum in constantan (the Cu$_{55}$Ni$_{45}$ constantan composition), which the source treats only qualitatively, compared against the measured Cu-Ni resistivity-temperature data that include the constantan resistance maximum. The single common objective is predictive, parameter-free accuracy for the temperature- and field-dependent resistivity of transition metals and their concentrated alloys, the two targets being linked by the same microscopic scattering physics. The source's other quantitative comparisons, the sublinear high-temperature resistance of Pd and Pt with the $T^3$-type deviation and degeneracy temperatures of order $2100$ K (Pd) and $3100$ K (Pt), the low-temperature resistivity regime $T \lesssim \Theta$, the electronic specific heats of Pd and Pt, and the zero-field resistivity of nickel near $T_C$, were already computed in the 1936 paper or lay outside its model's claimed domain, and are substantially covered by modern first-principles transport; they serve as context and consistency checks for any proposed theory, not as part of the two targets. Thermal spin disorder with magnetic short-range order at and above $T_C$ is a physically motivated possibility for target (1), since mean-field spin-disorder treatments give a constant contribution in that regime.

References

1. N. F. Mott, 'The electrical conductivity of transition metals', Proceedings of the Royal Society of London A 153, 699 (1936), DOI 10.1098/rspa.1936.0031, https://doi.org/10.1098/rspa.1936.0031
2. N. F. Mott, 'The resistance and thermoelectric properties of the transition metals', Proceedings of the Royal Society of London A 156, 368 (1936), DOI 10.1098/rspa.1936.0154, https://doi.org/10.1098/rspa.1936.0154
3. H. H. Potter, 'On the change of resistance of nickel in a magnetic field', Proceedings of the Royal Society of London A 132, 560 (1931), DOI 10.1098/rspa.1931.0118, https://doi.org/10.1098/rspa.1931.0118
4. E. Englert, 'Widerstandsänderung und Magnetisierung am Curiepunkt', Zeitschrift für Physik 74, 748 (1932), DOI 10.1007/bf01340424, https://doi.org/10.1007/bf01340424
5. N. F. Mott and H. H. Potter, 'Sharpness of the magnetic Curie point', Nature 139, 411 (1937), DOI 10.1038/139411a0, https://doi.org/10.1038/139411a0
6. N. F. Mott, 'Electrons in transition metals', Advances in Physics 13, 325 (1964), DOI 10.1080/00018736400101041, https://doi.org/10.1080/00018736400101041
7. P. G. de Gennes and J. Friedel, 'Anomalies de résistivité dans certains métaux magnétiques', Journal of Physics and Chemistry of Solids 4, 71 (1958), DOI 10.1016/0022-3697(58)90196-3, https://doi.org/10.1016/0022-3697(58)90196-3
8. M. E. Fisher and J. S. Langer, 'Resistive Anomalies at Magnetic Critical Points', Physical Review Letters 20, 665 (1968), DOI 10.1103/physrevlett.20.665, https://doi.org/10.1103/physrevlett.20.665
9. A. Fert, 'Two-current conduction in ferromagnetic metals and spin wave-electron collisions', Journal of Physics C: Solid State Physics 2, 1784 (1969), DOI 10.1088/0022-3719/2/10/311, https://doi.org/10.1088/0022-3719/2/10/311
10. A. Fert and I. A. Campbell, 'Electrical resistivity of ferromagnetic nickel and iron based alloys', Journal of Physics F: Metal Physics 6, 849 (1976), DOI 10.1088/0305-4608/6/5/025, https://doi.org/10.1088/0305-4608/6/5/025
11. F. C. Zumsteg and R. D. Parks, 'Electrical resistance near ferromagnetic Curie points', Journal de Physique Colloques 32, C1-534 (1971), DOI 10.1051/jphyscol:19711179, https://doi.org/10.1051/jphyscol:19711179
12. L. Berger, 'Spin relaxation in metallic ferromagnets', Physical Review B 83, 054410 (2011), DOI 10.1103/PhysRevB.83.054410, https://doi.org/10.1103/PhysRevB.83.054410
13. A. L. Wysocki, K. D. Belashchenko, J. P. Velev and M. van Schilfgaarde, 'Calculations of spin-disorder resistivity from first principles', Journal of Applied Physics 101, 09G506 (2007), DOI 10.1063/1.2670472, https://doi.org/10.1063/1.2670472
14. A. L. Wysocki, R. F. Sabirianov, M. van Schilfgaarde and K. D. Belashchenko, 'First-principles analysis of spin-disorder resistivity of Fe and Ni', Physical Review B 80, 224423 (2009), DOI 10.1103/PhysRevB.80.224423, https://doi.org/10.1103/PhysRevB.80.224423
15. J. Kudrnovský, V. Drchal, I. Turek, S. Khmelevskyi, J. K. Glasbrenner and K. D. Belashchenko, 'Spin-disorder resistivity of ferromagnetic metals from first principles: The disordered-local-moment approach', Physical Review B 86, 144423 (2012), DOI 10.1103/PhysRevB.86.144423, https://doi.org/10.1103/PhysRevB.86.144423
16. J. Kudrnovský, V. Drchal, I. Turek, D. Wagenknecht and S. Khmelevskyi, 'The Spin-Disorder Resistivity: The Disordered Local Moment Approach', Solid State Phenomena 289, 185 (2019), DOI 10.4028/www.scientific.net/SSP.289.185, https://doi.org/10.4028/www.scientific.net/SSP.289.185
17. G. D. Samolyuk, S. Mu, A. F. May, B. C. Sales, S. Wimmer, S. Mankovsky, H. Ebert and G. M. Stocks, 'Temperature dependent electronic transport in concentrated solid solutions of the 3d-transition metals Ni, Fe, Co and Cr from first principles', Physical Review B 98, 165141 (2018), DOI 10.1103/PhysRevB.98.165141, https://doi.org/10.1103/PhysRevB.98.165141
18. A. Giri, M. V. Tokina, O. V. Prezhdo and P. E. Hopkins, 'Electron-phonon coupling and related transport properties of metals and intermetallic alloys from first principles', Materials Today Physics 12, 100175 (2020), DOI 10.1016/j.mtphys.2019.100175, https://doi.org/10.1016/j.mtphys.2019.100175
19. S. Zhao, M. Zhu and Y. Zheng, 'Linear temperature dependence of intrinsic resistivity of metals determined by the electronic structure', Journal of Materials Chemistry C 14, 9298 (2026), DOI 10.1039/d5tc04381f, https://doi.org/10.1039/d5tc04381f
20. F. Engelke and C. Heiliger, 'Realistic modelling of transport properties at finite temperature in magnetic materials by local quantization of a Heisenberg model', arXiv:2508.11405 (2025; v2 January 2026), arXiv ID 2508.11405, https://arxiv.org/abs/2508.11405