Abrikosov (2001) computed the superconducting transition temperature for a mechanism proposed by Suhl, in which pairing of conduction electrons is mediated by their exchange coupling to ferromagnetically ordered localized spins, concluding that the mechanism yields only singlet s-wave pairing, that TcT_c is non-monotonic in the exchange field HH on the localized spins, and that survival against the pair-breaking exchange field hh on the electrons requires the hierarchy H>Tc>hH > T_c > h, equivalent to Jv≫1Jv \gg 1 and hence a heavy-fermion metal, a qualitative fit to pressurized UGe2. The derivation, however, rests on a ladder-diagram summation (the standard BCS technique of summing repeated particle-particle scattering diagrams) controlled only for Tc≪HT_c \ll H, treats HH and hh as phenomenological parameters of undetermined microscopic origin, and leaves open whether the instability survives vertex corrections and crossed diagrams (diagrammatic terms beyond that ladder, for which no small parameter is known), competing interactions, and realistic band structure, and whether any real material meets the hierarchy. The later literature reports that the uranium ferromagnetic superconductors are now widely judged, on indirect evidence, to show equal-spin triplet pairing, superseding the s-wave premise of the UGe2 application; the closest microscopic treatment of localized-spin-mediated pairing (magnon exchange) found triplet pairing in a finite temperature window, at odds with the observed persistence of superconductivity down to the lowest temperatures in the known compounds; and first-principles studies of UGe2 did not determine HH and hh. The model-intrinsic questions - a controlled gap equation valid for Tc∼HT_c \sim H, the microscopic origin and values of HH and hh, and the realizability of the hierarchy - remain open.

Background

Superconductivity and ferromagnetism were long thought incompatible: an internal magnetic (exchange) field tends to align the two spins of a Cooper pair and, above the Clogston (paramagnetic) limit, destroys conventional spin-singlet pairing. The discovery of superconductivity inside the ferromagnetic phase of the heavy-electron (heavy-fermion) compound UGe2 under pressure (Saxena et al. 2000), followed by reports in the itinerant ferromagnet ZrZn2 (Pfleiderer et al. 2001) and in URhGe (another uranium-based ferromagnetic superconductor), made a microscopic explanation urgent. Spin-triplet pairing, in which the two electrons of a pair carry parallel spins (equal-spin pairing), evades the paramagnetic limit because an exchange field does not depair parallel spins; Fay and Appel (1980) had already proposed p-state triplet pairing mediated by ferromagnetic spin fluctuations in itinerant ferromagnets. Against this background, Suhl (2001) proposed a different mechanism: superconductivity mediated by the exchange interaction of conduction electrons with ferromagnetically ordered localized spins. Abrikosov (2001) then treated this model in detail, motivated by UGe2, where he took the uranium 5f moments to be localized and argued from the persistence of superconductivity in polycrystalline samples - where non-magnetic impurity scattering suppresses anisotropic pairing states while leaving isotropic s-wave pairing intact (Anderson's theorem) - that the order parameter must be s-wave. In Abrikosov's model, conduction electrons couple with exchange constant JJ to localized spins of magnitude SS and number density NN; the ferromagnetic order is represented by an exchange field HH acting on the localized spins, while the conduction electrons additionally feel an exchange field hh that breaks singlet pairs. The electron density of states at the Fermi level is v=p0m/(2π2)v=p_0m/(2\pi^2), with p0p_0 the Fermi momentum and mm the effective mass; a large vv means a large mass, i.e. a heavy-fermion metal, in which conduction electrons acquire very large effective masses through hybridization with localized f-electrons. The pairing instability is computed by a ladder summation, i.e. by summing the infinite series of repeated particle-particle scattering diagrams (the standard BCS linearized-gap-equation technique), with the localized-spin response entering through the Brillouin function BSB_S, normalized to saturate at 11 in the strong-field, low-temperature limit, so that the equilibrium magnetization of a localized spin SS in a field is S BS(H/T)S\,B_S(H/T). In the regime Tc≪HT_c \ll H the calculation gives Tc≃2γHπexp⁡(−H2J2NvS),T_c \simeq \frac{2\gamma H}{\pi}\exp\left(-\frac{H}{2J^2NvS}\right), with γ≈1.78\gamma \approx 1.78, and more generally the implicit equation 2J2NvS BS(H/Tc)Hln⁡2γHπTc=12J^2NvS\,\frac{B_S(H/T_c)}{H}\ln\frac{2\gamma H}{\pi T_c}=1, modified by a subtracted integral when h≠0h \neq 0. Abrikosov concluded that (i) the mechanism allows only singlet s-wave pairing; (ii) TcT_c grows with HH, peaks at H=J2NvS/2H=J^2NvS/2 (a claim about the extrapolated regime Tc∼HT_c \sim H; the asymptotic form above, valid for Tc≪HT_c \ll H, itself peaks at H=2J2NvSH=2J^2NvS), and decreases for stronger ferromagnetism, so superconductivity persists only while ferromagnetism is weak; (iii) hh suppresses TcT_c as Tc≈Tc0−7ζ(3)4π2h2Tc0T_c \approx T_{c0}-\frac{7\zeta(3)}{4\pi^2}\frac{h^2}{T_{c0}} and destroys it for hh of order Tc0T_{c0}; and (iv) the hierarchy H>Tc>hH > T_c > h, with hh estimated as the molecular field h∼JNSh \sim JNS, reduces to Jv≫1Jv \gg 1, implying a heavy-fermion metal, consistent with the estimated m∼50mem \sim 50m_e in UGe2. Two structural weaknesses are visible in the paper itself: the Matsubara-frequency sums (sums over the discrete imaginary frequencies of finite-temperature perturbation theory) are cut off at HH, justified only for Tc≪HT_c \ll H, so the analysis of the regime Tc∼HT_c \sim H (including the claim that BS(H/Tc)/HB_S(H/T_c)/H is HH-independent there) is an uncontrolled extrapolation; and the ladder summation neglects vertex corrections and crossed diagrams, for which no Migdal-type theorem exists - unlike in electron-phonon superconductivity, where the small ratio of sound velocity to Fermi velocity suppresses such corrections, quasi-static localized spins provide no small parameter. The fields HH and hh are phenomenological inputs rather than derived quantities, competing interactions and band-structure details are not included, and the model presumes well-defined localized moments, so itinerant ferromagnets such as ZrZn2, where magnetism and superconductivity were attributed to the same itinerant bands, lie outside it as formulated. Contemporaneous alternatives (Fay and Appel 1980; Karchev et al. 2001; Mineev 2002; Watanabe and Miyake 2002) developed coexistence theories in the triplet and spin-fluctuation frameworks.

Problem Statement

The question asks for the microscopic completion, or controlled refutation, of the Suhl-Abrikosov mechanism of s-wave superconductivity mediated by ferromagnetically ordered localized spins. For the model of conduction electrons exchange-coupled (constant JJ) to localized spins (magnitude SS, density NN) through a contact (ss-dd-type) exchange interaction between the conduction-electron spins and the localized spins at lattice sites (the model of Suhl 2001), with the localized spins treated as equilibrium moments whose magnetization responds to HH through the Brillouin function BSB_S, with exchange field HH acting on the localized spins and pair-breaking exchange field hh acting on the electrons - hh being the mean exchange field felt by the electron spins, proportional to JJ and to the mean localized spin, with molecular-field estimate h∼JNSh \sim JNS - and with electron density of states at the Fermi level v=p0m/(2π2)v=p_0m/(2\pi^2) (p0p_0 the Fermi momentum, mm the effective mass), determine:

(1) Regime of validity. Derive the pairing instability and the equation for the superconducting transition temperature TcT_c without the assumption Tc≪HT_c \ll H, and settle the extrapolated claims made for Tc∼HT_c \sim H: that TcT_c grows approximately linearly with HH, that it is maximal at H=J2NvS/2H=J^2NvS/2, and that the factor BS(H/Tc)/HB_S(H/T_c)/H - where BSB_S is the Brillouin function, giving the equilibrium magnetization of a spin SS in a field - can be treated as HH-independent in this regime. Include an estimate of the non-ladder corrections (vertex corrections and crossed diagrams, i.e. diagrammatic terms beyond the repeated particle-particle scattering chain; unlike in electron-phonon superconductivity, no small parameter is known that suppresses them), so that the ladder-based instability is either controlled or shown to fail.

(2) Microscopic exchange fields. Determine the origin and magnitudes of the phenomenological fields HH and hh microscopically - whether they arise from direct exchange, superexchange, or the same electron-spin coupling that produces the pairing, together with band-structure contributions - replacing the molecular-field estimate h∼JNSh \sim JNS.

(3) Realizability. Decide whether any real ferromagnetic metal can satisfy the hierarchy H>Tc>hH > T_c > h together with Jv≫1Jv \gg 1 (large density of states, hence large effective mass) once competing interactions and realistic band structure are included, and confront the mechanism with the experimental record of the known ferromagnetic superconductors: the pressure dependence of TcT_c and of the ordered moment in UGe2, the evidence for equal-spin triplet pairing in the uranium compounds, and the elusive superconductivity of ZrZn2.

These three components jointly serve one scientific objective: a verdict on whether localized-spin-mediated s-wave superconductivity is a real and realizable mechanism. A resolution requires all three components (1)-(3).

References

  1. A. A. Abrikosov, "Superconductivity due to ferromagnetically ordered localized spins", J. Phys.: Condens. Matter 13 (2001) L943-L947. DOI: 10.1088/0953-8984/13/48/104, https://doi.org/10.1088/0953-8984/13/48/104
  2. H. Suhl, "Simultaneous Onset of Ferromagnetism and Superconductivity", Phys. Rev. Lett. 87 (2001) 167007. DOI: 10.1103/PhysRevLett.87.167007, https://doi.org/10.1103/PhysRevLett.87.167007
  3. S. S. Saxena, P. Agarwal, K. Ahilan, F. M. Grosche, R. K. W. Haselwimmer, M. J. Steiner, E. Pugh, I. R. Walker, S. R. Julian, P. Monthoux, G. G. Lonzarich, A. Huxley, I. Sheikin, D. Braithwaite, J. Flouquet, "Superconductivity on the border of itinerant-electron ferromagnetism in UGe2", Nature 406 (2000) 587-592. DOI: 10.1038/35020500, https://doi.org/10.1038/35020500
  4. D. Fay, J. Appel, "Coexistence of p-state superconductivity and itinerant ferromagnetism", Phys. Rev. B 22 (1980) 3173-3182. DOI: 10.1103/PhysRevB.22.3173, https://doi.org/10.1103/PhysRevB.22.3173
  5. N. I. Karchev, K. B. Blagoev, K. S. Bedell, P. B. Littlewood, "Coexistence of Superconductivity and Ferromagnetism in Ferromagnetic Metals", Phys. Rev. Lett. 86 (2001) 846-849. DOI: 10.1103/PhysRevLett.86.846, https://doi.org/10.1103/PhysRevLett.86.846
  6. N. Karchev, "Magnon-mediated superconductivity in itinerant ferromagnets", J. Phys.: Condens. Matter 15 (2003) L385-L391. DOI: 10.1088/0953-8984/15/24/105, https://doi.org/10.1088/0953-8984/15/24/105
  7. V. P. Mineev, "Superconducting states in ferromagnetic metals", Phys. Rev. B 66 (2002) 134504. DOI: 10.1103/PhysRevB.66.134504, https://doi.org/10.1103/PhysRevB.66.134504
  8. S. Watanabe, K. Miyake, "Coupled CDW and SDW Fluctuations as an Origin of Anomalous Properties of Ferromagnetic Superconductor UGe2", J. Phys. Soc. Jpn. 71 (2002) 2489-2499. DOI: 10.1143/JPSJ.71.2489, https://doi.org/10.1143/JPSJ.71.2489
  9. A. Harada, S. Kawasaki, H. Mukuda, Y. Kitaoka, Y. Haga, E. Yamamoto, Y. Ōnuki, K. M. Itoh, E. E. Haller, H. Harima, "Experimental evidence for ferromagnetic spin-pairing superconductivity emerging in UGe2: A 73Ge-nuclear-quadrupole-resonance study under pressure", Phys. Rev. B 75 (2007) 140502. DOI: 10.1103/PhysRevB.75.140502, https://doi.org/10.1103/PhysRevB.75.140502
  10. C. Pfleiderer, M. Uhlarz, S. M. Hayden, R. Vollmer, H. v. Löhneysen, N. R. Bernhoeft, G. G. Lonzarich, "Coexistence of superconductivity and ferromagnetism in the d-band metal ZrZn2", Nature 412 (2001) 58-61. DOI: 10.1038/35083531, https://doi.org/10.1038/35083531
  11. S. M. Stishov, V. A. Sidorov, A. V. Tsvyashchenko, E. D. Bauer, A. E. Petrova, T. Park, J. D. Thompson, "Phase diagram of ZrZn2 at high pressure: Low-temperature features and elusive superconductivity", Physica B 378-380 (2006) 411-412. DOI: 10.1016/j.physb.2006.01.144, https://doi.org/10.1016/j.physb.2006.01.144
  12. D. Aoki, F. Hardy, A. Miyake, V. Taufour, T. D. Matsuda, J. Flouquet, "Properties of ferromagnetic superconductors", C. R. Physique 12 (2011) 573-583. DOI: 10.1016/j.crhy.2011.04.007, https://doi.org/10.1016/j.crhy.2011.04.007
  13. D. Aoki, J. Flouquet, "Ferromagnetism and Superconductivity in Uranium Compounds", J. Phys. Soc. Jpn. 81 (2012) 011003. DOI: 10.1143/JPSJ.81.011003, https://doi.org/10.1143/JPSJ.81.011003
  14. Y. Ōnuki, R. Settai, Y. Haga, Y. Machida, K. Izawa, F. Honda, D. Aoki, "Fermi surface, magnetic, and superconducting properties in actinide compounds", C. R. Physique 15 (2014) 616-629. DOI: 10.1016/j.crhy.2014.06.005, https://doi.org/10.1016/j.crhy.2014.06.005
  15. L. Bulaevskii, R. Eneias, A. Ferraz, "Triplet superconductivity in ferromagnets due to magnon exchange", Phys. Rev. B 99 (2019) 064506. DOI: 10.1103/PhysRevB.99.064506, https://doi.org/10.1103/PhysRevB.99.064506
  16. E. Kądzielawa-Major, M. Fidrysiak, P. Kubiczek, J. Spałek, "Spin-triplet paired phases inside a ferromagnet induced by Hund's rule coupling and electronic correlations: Application to UGe2", Phys. Rev. B 97 (2018) 224519. DOI: 10.1103/PhysRevB.97.224519, https://doi.org/10.1103/PhysRevB.97.224519
  17. M. Samsel-Czekała, M. Werwiński, A. Szajek, G. Chełkowska, R. Troć, "Electronic structure of UGe2 at ambient pressure: Comparison with X-ray photoemission spectra", Intermetallics 19 (2011) 1411-1419. DOI: 10.1016/j.intermet.2011.05.008, https://doi.org/10.1016/j.intermet.2011.05.008
  18. V. G. Storchak, J. H. Brewer, D. G. Eshchenko, P. W. Mengyan, O. E. Parfenov, D. Sokolov, "Spin-polaron band in the ferromagnetic heavy-fermion superconductor UGe2", J. Phys.: Conf. Ser. 551 (2014) 012016. DOI: 10.1088/1742-6596/551/1/012016, https://doi.org/10.1088/1742-6596/551/1/012016