Current-Field Cooling in Superconductors
- Current-field cooling is a history-dependent protocol that uses magnetic fields generated by electrical currents to control vortex nucleation and screening in superconductors.
- In Nb thin films, inhomogeneous-field cooling creates a graded vortex landscape, leading to a measurable 20% enhancement in effective critical current as indicated by Bean and Kim models.
- This method encodes current history into persistent vortex or flux structures, offering practical insights for optimizing pinning strategies and superconducting device performance.
Current-field cooling denotes a family of history-dependent preparation protocols in which a system is cooled through a transition in the presence of a magnetic field generated or structured by electrical current, or under a transport current whose self-field participates in vortex nucleation. In the Nb thin-film realization that most directly formulates the concept, cooling in an inhomogeneous magnetic field generated by a concentric Nb ring enhances the film’s screening capacity relative to zero-field cooling and homogeneous-field cooling, and both Bean-type and Kim-type analyses indicate an increased effective critical current density (Chaves et al., 2021). Related work in cuprate strips, granular ceramics, bulk high- superconductors, and annular Josephson tunnel junctions shows that the cooling protocol can be encoded as vortices, trapped fluxoids, persistent currents, or other frozen current structures that subsequently control flux penetration, transport, and memory effects (Nishimura et al., 25 Nov 2025).
1. Protocol definition and experimental variants
The canonical Nb experiment distinguishes three cooling procedures. In zero-field cooling (ZFC), the film is cooled from above with no applied field. In homogeneous-field cooling (HFC), it is cooled under a uniform perpendicular magnetic field. In inhomogeneous-field cooling (IFC), it is cooled under a spatially varying field generated by current through a narrow concentric Nb ring surrounding the film; the field profile was calculated via the Biot–Savart law. The signs “+” and “–” denote initial states containing vortices or antivortices, respectively (Chaves et al., 2021).
The device used for this IFC protocol was a $200$ nm-thick Nb film of size $2.48$ mm $2.48$ mm, surrounded by a concentric narrow Nb ring $0.06$ mm wide with a $0.08$ mm gap. After cooling, a uniform out-of-plane field was applied and flux penetration was studied by quantitative magneto-optical imaging using a Bi:YIG indicator and calibrated image processing in MATLAB/ImageJ (Chaves et al., 2021).
A related but distinct protocol is current-biased cooling in a YBCO strip. There, the external drive is the transport current itself rather than an externally imposed cooling field. Imaging after cooldown shows that even in zero magnetic field, current-biased cooling nucleates vortices within the strip; with a small external magnetic field, the resulting distribution is polarized opposite to the Lorentz-force direction. The study interprets this as direct encoding of current history into a frozen vortex configuration (Nishimura et al., 25 Nov 2025).
This usage concerns state preparation rather than temperature reduction. A plausible implication is that “current-field cooling” is best understood as a protocol class centered on history-dependent freezing of current- and field-generated textures.
2. Nb thin-film realization: flux-front hierarchy and effective critical current
The central empirical observable in the Nb thin film is the flux front penetration depth, defined as the distance from the film edge to the farthest flux entry into the superconducting film. Magneto-optical images establish a strict ordering of penetration depth from deepest to shallowest: Thus, cooling in a positive inhomogeneous field, which prepares vortices in the film, yields the shallowest penetration, while cooling in the opposite polarity, which prepares antivortices, yields the deepest penetration (Chaves et al., 2021).
For , the reconstructed values reported for penetration depth and effective current metrics are:
| Procedure | 0 (mm) | Relative 1 |
|---|---|---|
| IFC– | 0.383 | Bean 94.2%, Kim 95.6% |
| HFC– | 0.375 | Bean 95.6%, Kim 97.5% |
| ZFC | 0.350 | Bean 100.0%, Kim 100.0% |
| HFC+ | 0.300 | Bean 110.3%, Kim 103.4% |
| IFC+ | 0.288 | Bean 113.2%, Kim 107.3% |
Within the Bean analysis, IFC+ corresponds to 2, which is 3 higher than ZFC, whereas IFC– gives 4, which is 5 lower than ZFC. The full IFC– to IFC+ span corresponds to an approximately 6 enhancement. The B-dependent Kim analysis preserves the same ordering and likewise identifies IFC+ as the maximal effective-7 state (Chaves et al., 2021).
The immediate physical interpretation used in the study is that shallower penetration requires more screening current to exclude flux, so a reduced penetration depth functions as an operational indicator of a higher effective critical current.
3. Microscopic and mesoscopic mechanisms
The Nb thin-film mechanism is formulated in terms of interactions between entering vortices and the vortex population prepared during cooling. If the initial state contains pinned vortices, as in IFC+ or HFC+, newly entering vortices repel the existing ones, and further penetration becomes more difficult. If the initial state contains antivortices, as in IFC– or HFC–, entering vortices attract them and can annihilate, which facilitates penetration and reduces shielding effectiveness (Chaves et al., 2021).
The inhomogeneous case is stronger than the homogeneous one because IFC creates a non-uniform, graded vortex landscape. The study explicitly relates this to optimal pinning landscapes, including conformal vortex arrangements. This suggests that the improvement is not merely due to the presence of trapped vortices, but to their spatial distribution. In this reading, the cooling field acts as a non-contact method for writing a collective pinning background into the film (Chaves et al., 2021).
Current-biased cooling in YBCO provides a complementary self-field mechanism. The strip current generates a non-uniform out-of-plane self-field that is strongest near the edges and changes sign at the strip center. Near the transition, positive and negative vortices nucleate at opposite edges and are driven toward the strip interior by the Lorentz force; oppositely signed vortices can then annihilate near the center, suppressing the vortex population there. When a small external field is added, the point where the total field changes sign shifts, and the vortex landscape becomes polarized (Nishimura et al., 25 Nov 2025).
Across both systems, the common principle is that the cooling history determines the sign, density, and spatial organization of frozen vortices. A plausible implication is that current-field cooling is most effective when the frozen state is not uniform, but deliberately biased to oppose later flux entry.
4. Modelling frameworks
The Nb study analyzes the enhanced screening with both B-independent and B-dependent critical-state descriptions. In the Bean model for a thin film of thickness 8, half-width 9, applied field $200$0, and flux-front penetration $200$1,
$200$2
The B-dependent alternative is the Kim form
$200$3
The numerical inversion of current distributions normalized to $200$4 yields the same qualitative ranking as the Bean analysis: the effective $200$5 is maximal for IFC+ and minimal for IFC– (Chaves et al., 2021).
In bulk high-temperature superconductors after field-cooled magnetization, the critical state has also been modeled by a backward computation method. The method initializes a large surface magnetization current after rapid FC magnetization, then relaxes that current inward step-by-step according to the critical state model. After approximately $200$6–$200$7 backward iterations, the distribution reaches a steady state. For a finite-element model with $200$8 million degrees of freedom, the method takes less than $200$9 hours and agrees closely with H-formulation calculations in both electromagnetic and electromagnetic-mechanical coupled analyses (Zhang et al., 2020).
Current-biased cooling in the YBCO strip is described by a self-consistent relation between current and local field. For a strip of half-width $2.48$0,
$2.48$1
and in steady flux flow,
$2.48$2
With flux creep, the generalized relation becomes
$2.48$3
with $2.48$4. The reconstructed current density exhibits a central peak, consistent with the self-consistent freezing of an out-of-equilibrium current profile during cooldown (Nishimura et al., 25 Nov 2025).
Taken together, these models emphasize that current-field cooling is not only a matter of trapped flux amplitude; it is also a boundary-value problem involving field geometry, self-consistency, pinning, and the path by which the system approaches its low-temperature state.
5. Frozen flux, persistent currents, and memory states
Field-cooled YBa$2.48$5Cu$2.48$6O$2.48$7 ceramics provide a classical frozen-flux analogue. In the FC regime, cooling to $2.48$8 K in the excitation field and then switching the field off produces a “double vortex” consisting of two oppositely directed, coaxial macroscopic vortex currents. In the alternative ZFC regime, where the sample is cooled in the Earth’s field and the excitation field is only applied afterward, the resulting state is a “multivortex” structure made of many smaller current loops. The local critical excitation field for the ZFC multivortex structure is approximately $2.48$9, about ten times the uniform perpendicular critical field of the slab, while the FC double-vortex structure can be formed by practically any weak excitation field (Bondarenko et al., 2013).
Because the FC double vortex behaves as a macroscopic current object, its motion under transport current can be observed directly. The passage of transport current displaces the structure via the Lorentz force, enabling extraction of a pinning force 0 and a viscosity 1 (Bondarenko et al., 2013).
Annular Josephson tunnel junctions exhibit a related frozen-current mechanism at the device level. When long Nb/Al-AlO2/Nb annular junctions are cooled through 3 in a transverse magnetic field, the doubly connected electrode traps an integer number of flux quanta and generates a persistent current. The associated radial magnetic field modifies both static and dynamic junction properties: it shifts the first lobe of the magnetic diffraction pattern and, combined with a d.c. bias current, induces viscous flow of dense trains of Josephson vortices that appears as displaced linear slopes, Fiske step staircases, and Eck steps in the current-voltage characteristic. The effect can be mitigated or canceled by an external magnetic field perpendicular to the junction plane, and it is enhanced in confocal annular geometries while vanishing in the circular limit where the relevant forcing term tends to zero (Monaco et al., 2020).
These cases illustrate the same general principle: cooling can freeze topological or quasi-topological current structures whose later dynamical role is equivalent to an internally stored bias field.
6. Applications, limitations, and adjacent directions
For thin superconducting devices, the immediate application proposed for IFC is enhancement of effective critical current without micro- or nano-patterning. The study explicitly presents the method as a potentially simple way to boost performance, describes the graded vortex arrangement as a strong pinning landscape, and suggests a “field-learning” approach for engineering critical currents. It also notes that graded vortex landscapes suppress flux channeling and thermomagnetic avalanches, which would further stabilize device operation (Chaves et al., 2021).
In bulk HTS undulators, field-cooled critical-state modelling is used to study the trade-off between electromagnetic performance and mechanical stress. Without mechanical degradation, the undulator field amplitude is about 4; with Lorentz force only it drops to about 5; with Lorentz force and pre-stress it drops further to about 6. The computational framework therefore serves design optimization as much as state reconstruction (Zhang et al., 2020).
A common misconception is that any field-cooled or current-field-cooled state with trapped flux improves performance. The Nb thin film shows the opposite for antivortex-prepared states: IFC– has deeper penetration and a lower effective critical current than ZFC, whereas IFC+ yields the enhancement (Chaves et al., 2021). Another misconception is that anomalous low-temperature thermodynamic signatures after field cooling necessarily reflect equilibrium phase transitions. In Sn-Pb solders field-cooled at 7, the first specific-heat run at a new temperature can show anomalously low 8 together with a temperature rise of 9–$2.48$0, rather than the nominal $2.48$1, because trapped-flux reduction during warming causes self-heating through flux flow and flux jumps. The reported “lower-$2.48$2 transition” is therefore interpreted as an artifact of dissipative flux dynamics rather than a true phase transition (Mizuguchi et al., 2024).
A plausible broader implication is that cooling-protocol-controlled memory is not restricted to superconducting vortices. In exchange-coupled $2.48$3 films, cooling below the blocking temperature under moderate magnetic fields yields up to $2.48$4 magnetic domain memory throughout the magnetization loop, whereas nearly saturating field cooling suppresses memory except at nucleation and saturation (Chesnel et al., 2020). Although that system is not a superconductor, it underscores the same general point: the state reached after cooling can retain a detailed record of the field or current environment present during the transition.