Viewpoint on fast creation of dense MgB₂ phase in wires made by internal Mg diffusion process

This is a viewpoint on the letter by P Kováč et al (2016 Supercond. Sci. Technol. 29 10LT01).

Since the discovery of MgB_2, significant progress has been made in recent years on improving conductor performance, and a number of fabrication techniques have been successfully developed. A series of important advances showed that substantial enhancement could be achieved in MgB₂ wires in terms of their critical current density, J_c(H), irreversibility field, H_irr, and upper critical field, H_c2, by nano-SiC carbon nanotube (CNT), or carbohydrate doping [1–5]. A record high H_c2 of 43 T has been achieved for nano-SiC doped MgB₂ [6]. Various attempts have been made to improve the flux pinning for MgB₂, using techniques such as strain engineering [7], irradiation [8], and reducing grain size [9]. Unlike its main competitor Nb₃Sn, MgB₂ is clearly able to incorporate a variety of nanoscale pins, leading to significant improvement in bulk pinning. Nevertheless, the practical difficulty for incorporating nanoparticles into MgB₂ is that the effective doping agents increase the normal state resistivity, segregate at the grain boundaries of MgB₂, and reduce the active cross-section for carrying the supercurrent [10, 11]. Their influence in blocking the current is particularly negative for the low-field J_c and can easily exceed the benefits of having effective pinning centres in high fields.

In situ and ex situ powder metallurgical processes are currently used for the industrial fabrication of mono- and multi-filamentary MgB₂ wires. The mass density in the MgB₂ filaments of the wires after the final heat treatment is quite low with respect to the theoretical density (2.36 g cm⁻³). Values of ∼50% and ∼70% have been reported for in situ and ex situ wires, respectively [10, 12]. To date, the low density within the filament cores has been a serious obstacle to reaching high J_c values in MgB₂ wires fabricated by in situ processing techniques. High porosity is well known to be difficult to avoid. As a result, grain connectivity is hampered by the presence of voids in the finished wires. The typical effective superconductor area in the wire is only about 10% [12]. Cold high pressure densification (CHPD) has been suggested as an alternative to enhance filament mass density [13]. The first successful CHPD of a dense MgB₂ conductor in a short length (<1 m) was reported by the University of Geneva in 2008 [14]. The density increased from 50% for normal wire to 73% for the CHPD wire after heat treatment. The J_c enhancement was particularly impressive at 20 K, since J_c increased by more than 300%. No degradation of J_c was detected in the overlapping pressure zones between two neighbouring regions when compared to short wire lengths, so the Geneva group extended the CHPD method to lengths exceeding 10 m, a first step towards industrial lengths [15].

Another very effective densification technique is called the infiltration method, first reported by Giunchi et al, in which the MgB₂ wire is fabricated using a composite billet composed of a steel pipe internally lined with a Nb tube filled with a coaxial internal pure Mg rod and B powder [16]. The J_c performance from the reported work was not especially high, however, there was no chemical doping associated with the technique. Following this work, an internal Mg diffusion (IMD) process using a pure Mg rod as the core, surrounded by B powder with a SiC addition, was developed for the fabrication of SiC doped MgB₂ wire. The composite was cold-worked into a wire and heat treated at temperatures above the melting point of Mg (650 °C). During the heat treatment, liquid Mg infiltrated the B layer and reacted to form MgB₂. The best J_c for this IMD SiC doped MgB₂ wire was 100 000 A·cm⁻² at 8 T [17], well exceeding the performance of commercial Nb-Ti wires and satisfactory for all technical applications.

Considerably higher values of J_c are expected for further enhancement of mass density. This is the case for MgB₂ wires produced by IMD [18] and advanced internal Mg infiltration (AIMI) [19] processes for wires, where the mass density approaches the theoretical maximum.

However, there is a serious limitation on scaling-up the IMD method. Although the AIMI wires exhibit the highest J_c values so far among all MgB₂ wires, some development is still required for these second generation wires. Indeed, it takes a long time to diffuse Mg into the B layer to complete the reaction during AIMI, in which a hollow wire and a thin layer of fully dense MgB₂ are formed inside the sheath tube, thus lowering the filling factor and leading to poor mechanical properties. It is obvious that, for this technique to be useful for large-scale production, a detailed investigation and clear understanding of the diffusion mechanism is required to develop an innovative process.

This is a timely publication on the IMD process by Kovac et al [20], in which the authors present a detailed analysis of a fast formation mechanism for MgB₂ during IMD that does not compromise the critical temperature or J_c in single-core MgB₂ wire. The highest J_c (10⁵ A cm⁻²) at 6 T was observed in the MgB₂ wire sintered at 635 °C for only 4 to 8 min. This result is very promising and well comparable with the previous highest J_c value (10⁵ A cm⁻² at 6 T) for an undoped wire sintered at 640 °C–670 °C for 3–5 h that was reported by the National Institute for Materials Science (NIMS, Japan) group [21]. The authors showed that 8 min sintering time is sufficient for the formation of a uniform MgB₂ phase 90 μm in thickness and without cracks, which is nearly identical to samples produced under the same sintering temperature conditions (range of 640 °C–670 °C) but longer sintering time of 3–5 h that were previously reported by the NIMS group. This finding is important for economical industrial production, but a question still remains as to how the round wires with a hollow filament will perform in magnet design and application, where enhancement of the filling factor, reduction of filament size, and mechanical and electrical stability and reliability are key factors.