Magnesium Diboride (MgB₂) [1] (See Fig. 1) was discovered as a phonon-mediated [2] superconductor in 2001 by Nagamatsu [3] and promptly sparked interest due to its transition temperature $\mathrm{T}_{\mathrm{c}}=39 \mathrm{~K}$. Superconducting components based on MgB₂ have been primarily produced as bulks, tapes, wires, and films to suit different applications in energy, electronics, medical, and particle physics applications [4], [5], [6], [7]. Among those modalities, some of the predominant techniques for MgB₂ wire and tape production are Powder-in-Tube (PIT) [8] and Mg Infiltration [9]. Unlike REBCO superconductors, the polycrystalline nature of MgB₂ per se does not limit its superconducting or mechanical performance [10]. Rather, phenomena related to the growth of MgO, MgB₄, amongst other non-superconducting phases generally culminates in the reduction of connectivity between MgB₂ grains, causing rapid decay of the magnetic field in response to increasing critical current densities ($J_C$) [11].

In this manuscript, the state-of-the-art manufacturing and challenges associated with superconducting MgB₂ tapes, wires, and bulks for a multitude of applications are explained. By pinpointing milestones in MgB₂ superconductor performance achieved so far, it serves as baseline to define next steps concerning manufacturing process needs and performance reproducibility. The proposed roadmap outlines the Process-Structure-Property-Performance (PSPP) needs, where potential links between process dynamics and intrinsic material properties are leveraged towards elaborating on MgB₂ performance for each superconductor modality.

MgB₂-based thin films exhibit promise for future superconducting electronic device applications. Compared to the 2–3 nm coherence length in YBCO [14], MgB₂ presents a 4–6 nm and rises as a candidate to compose Josephson Junctions [15]. Deposition of MgB₂ thin films via Molecular Beam Epitaxy (MBE) [16], [17], as well as Pulsed Laser Deposition (PLD) [18] has been reported, where the latter process has mostly yielded low-quality, amorphous phases of MgB₂ rather than epitaxial, crystalline MgB₂ films. Among those, Hybrid Physical-Chemical Vapor Deposition (HPCVD) is regarded as a more stable process for growth of MgB₂ thin films compared to other techniques, according to reports in [19], [20]. However, observed circular defect sites in MgB₂ films grown via HPCVD point to the Mg phase transformation as a key factor for originating such sites and thinning MgB₂ films in their vicinities [21]. While the thermodynamic behavior of Mg poses a challenge to the MgB₂ film epitaxial growth, the substrate choice may also impact the process thin film growth; hence leading to the deposition of low crystallinity layers. To unveil challenges in deposition of MgB₂ films, authors in [22] exhibited a systematic analysis of MgB₂ thin film morphologies, where the relationship between MgB₂ grain size, surface roughness, and deposition rates was built. Growth parameters have significantly impacted the film quality grown on SiC, MgO, and stainless-steel substrates. As expected, the growth of MgB₂ thin films grown via HPCVD obeys the Volmer-Weber modes [23], where small grain colonies are formed, then coalesce through a thermally activated mass diffusion process. Tight deposition control of HPCVD has enabled epitaxial, highly crystalline thin films as reported in [24] and [25]. $\mathrm{T}_{\mathrm{c}} 42 \mathrm{~K}$ and $\mathrm{J}_{\mathrm{c}} 1 \times 10^{8} \mathrm{~A} / \mathrm{cm}^{2}$ in self-field at $\mathrm{T}=1.8 \mathrm{~K}$ have been achieved as results of the property reproducibility of such thin films [23]. A process-agnostic CALPHAD model proposed by [26] studies the dependencies of the MgB₂ superconducting phase films with pressure, temperature, composition of precursors. Those ultimately led to MgB₂ phase diagrams for a multitude of process conditions, shedding light to the expected behavior of Mg and B precursors in forming MgB₂ and posterior growth in the thin film form.

Along with high $J_C$, a $T_C$ above the theoretical 39K first demonstrated in 2001 [5], [27], [28] demonstrated MgB₂ film smooth surfaces are suitable for applications in Josephson Junctions. Authors in [29], [30] presented a Hot Electron Bolometer (HEB) produced via HPCVD with MgB₂ thin films for precision Infrared radiation (IR) measurements. In [31], authors characterized submicron MgB₂ films grown on SiC, where the 10 nm and 40 nm MgB₂ films achieved $\mathrm{T}_{\mathrm{C}} 35 \mathrm{~K}$ and $\mathrm{T}_{\mathrm{C}} 41 \mathrm{~K}$, respectively. The 20nm films exhibited the highest $\mathrm{J}_{\mathrm{c}} 1.2 \times 10^{8} \mathrm{~A} / \mathrm{cm}^{2}$ under self-field conditions at 4.2K. This evidenced the dominant flux pinning mechanism (i.e., surface or volume pinning) according to the film thickness, culminating in potential increases of both $T_C$ and $J_C$. Investigations in [32] employed optical spectroscopy to analyze optical properties of an HPCVD-grown 1000-nm MgB₂ film on Al₂O₃ substrate, for which $T_C$40K. Authors reported the film electron–phonon density function, optical scattering rate and conductivity at T=8K and T=50K. Optical information of such thin films was utilized in generating data to support the development of multigap devices based on superconductors such as MgB₂. In [33], 50 mm-diameter MgB₂ films were fabricated on an Nb substrate via HPCVD and tested as superconducting radiofrequency cavities. The produced MgB₂ film achieved good performance as demonstrated by TC39.6K and surface resistance of 120 μΩ at 4K. Characterization of multiple thicknesses of MgB₂ thin films (i.e., 20 nm, 200 nm, 600 nm, and 1μ m) produced via HPCVD was done in [34] for analysis of surface morphology of films grown on SiC. Film growth according to the Volmer-Weber Mode was reported and accounted for the high RMS film roughness. In this process, the measured film critical thickness remained between 60–100 nm, after which the deposited MgB₂ islands would coalesce to form the film. Dominance of the surface pinning mechanism was observed for 60 nm and 200 nm films and, for thicker films with a-b growth, stronger flux pinning mechanism via volume pinning was reported.

Authors in [35] investigated the possibility to produce MgB₂ in three-dimensional structures via HPCVD on Al₂O₃ cylindrical substrates. The 10 μm film exhibited $T_C$39K, and $J_C$10⁵A/cm² in self-field. Reduction of $J_C$ is linked to inhomogeneity of films caused by non-epitaxial growth, Volmer-Weber modes, and high deposition rates. Besides, film properties were majorly affected by the inability to control the Mg vapor pressure in the HPCVD reactor. Such a factor culminated in decreasing the in-situ reaction rate of MgB₂ from Mg vapor and B₂H₆. As a result, deposits of both unreacted Boron and Magnesium were found on the Al₂O₃ substrate. The work presented in [36] built upon [35] concerning HPCVD of MgB₂ films on stainless steel cylinders, where authors presented a model to predict the feasibility of MgB₂ films in shielding magnetic fields with effectiveness of 65 dB.In such a case, the reported $T_C$37.5K with $J_C$7.6×10⁶A/cm² at 10 K, and 4.6×10⁶A/cm² at 20 K both in self-field. MgB₂ thin films were grown on Al₂O₃ with ZnO layers via HPCVD [37]. Different buffer layer thicknesses (i.e., 0, 9 nm, 15 nm, 23 nm, and 40 nm), were shown to improve $J_C$. ZnO layers diffused with MgB₂ throughout HPCVD, originating flux pinning centers in grain boundaries and point defects; hence improving $J_C$. Anisotropy analysis in off-axis MgB₂ [0 0 1] c-axis thin films on MgO [2 1 1] substrates via HPCVD perpendicular planes was reported in [38]. Through a Van der Pauw configuration, resistivity ratio values at both 40 K and 200 K were obtained with linear four-probe, refined, and simplified circuit models for a 40 nm thick MgB₂ layer. In [39], authors demonstrated thin films of MgB₂ on C-terminated 6H-SiC present better grain connectivity compared to Si-terminated in HPCVD. The study utilized low B₂H₆ gas flow rate (1-2sccm), high H₂ flow (400 sccm) and maintained pressure of 40 Torr. The oxidation of surface Si atoms and posterior polishing removed Si, then Si-C bilayer on top. The thicker film (10.7 nm) grown on Si presented $T_C$ of 37.8 K as opposed to the 36.4 K on C-terminated substrate. The reported 1 nm RMS roughness of the film on the C-terminated substrate was significantly lower than the 2.4 nm measured for the MgB₂ film on Si. Electrical resistivity at 42 K is 7 μΩ.cm for the MgB₂ thicker film on C-terminated substrate, and 17 μΩ.cm for the film on Si-terminated SiC. Authors in [40] built upon [39], and demonstrated the growth of a 2 nm MgB₂ film on C-terminated SiC substrate via HPCVD. A Molybdenum susceptor held Mg pellets heated to 730 °C, and the thickness calibration of MgB₂ films exhibited a deposition rate of 0.16 nm/s. For the 2 nm MgB₂ superconducting film, authors reported $T_C$27.2K, and $J_C$=2×10⁷A/cm² at 3 K in self-field, and RMS roughness of 0.62 nm. Also, the growth of a MgO layer prior to the MgB₂ film deposition caused the substrate surface to become smoother, allowing for a smaller RMS roughness of the film.

Mono and multifilamentary wires have been under development since the discovery of MgB₂ as a superconductor. Research groups and private companies based out of Europe [41], United States [42], [43], and Asia [44], [45] have engaged in the pursuit of optimizing MgB₂ wires to make it economically viable for applications ranging from 4.2-20 K. Most roadblocks to long-length MgB₂ wire manufacturing concern the mechanical and superconducting property reproducibility. Techniques such as Internal Magnesium Diffusion (IMD) [46] (See Fig. 3), Reactive Liquid Diffusion (RLI) [47], Powder-in-Tube (PIT) [48], and Continuous Tube Fill Forming (CTFF) [49] have been applied to producing kilometer-scale MgB₂ wires for low-field applications and power transmission. However, the synthesis, %wt. of the MgB₂ phase, contamination, low grain consolidation and orientation control have become ubiquitous challenges in MgB₂ mono or multifilamentary wires.

The literature has presented a plethora of studies concerning the property stability of MgB₂ wires for applications in poloidal coils of fusion reactors [51], low-field coils for Magnetic Resonance Imaging (MRI) [52], [53], [54], development of test coils for low-field magnets up to 3 T [55], [56], power transmission cables [57] and power generation [58], as well as launching systems for civil aircraft [59]. For low-field applications (i.e., <5T), MRI and poloidal coils for fusion reactors are candidate applications for operation at 4.2-20 K. To illustrate the roadmap to wire practical applications, authors in [60] defined the six phases that a material must go through from the discovery of the superconductor to the magnet-grade production (See Fig. 4).

Due to the brittleness of MgB₂, wire manufacturing is usually performed based on composite structures (i.e., metal matrices – MC) to provide the strain resistance that the sintered MgB₂ by itself does not have. Early in the manufacturing research of MgB₂ wires, Seven-filament MgB₂ wires were proposed by [61] in a process of mechanical deformation and post-annealing heat treatment. Compared to the mono-core wires produced via Powder-in-Tube, the Fe-sheathed wires caused the $J_C$ to be around four times larger than observed in mono-core wires. Although promising, outcomes of a four-month degradation test results showed the annealed wire would no longer superconduct without prior sintering. While the sintered cable exhibited $I_C$ =50.7A in self-field, the PIT (See Fig. 5) mono-core cable would still present a post-degradation $I_C$ 62.8A at 4.2 K in self-field. The presumed reason for the post-degradation performance decrease in the seven-filament, annealed wires is the Mg oxidation that limits the MgB₂ core density. Reduction of the %wt. superconducting phase in wires is the compromising agent of their $J_C$.

Through the Rietveld method, [58] performed full spectra neutron scattering studies with ex-situ, Ni-sheathed MgB₂Ni_2.5 wires. The MgB₂ lattice strain due to the wire deformation pointed to a decrease in the wire $T_C$. Mg deficiency in the starting powder was pointed out. Using neutron diffraction and SEM imaging, MgB₂ larger crystallite sizes showed to be inversely correlated with the wire critical temperature. Tapes analyzed under 1 T and 4 T magnetic fields exhibited their highest $J_C$ when reacted at 900° and 920 °C, respectively. Ni migration into the tape during the sintering process was determined to limit the current density of the tapes. In a study presented by [63], texture characterization of Ti-MgB₂ wires produced via PIT reported that the MgB₂ $[10 \overline{1} 0]$ phase, found in mono-core wires, disappeared after the rolling process for production of multifilamentary wires. Such findings evidenced the MgB₂ content may also be impacted by the wire drawing process. In [64], authors reported Reactive Liquid Infiltration (RLI) to produce mono and seven-filament, MgB₂ wires on Nb sheath enclosed by soft steel (SS). The hollow wire geometry is caused by the melting of an Mg rod and posterior infiltration, followed by the in-situ reaction with the Boron powder to form an MgB₂ phase on top of the internal surface of the Nb sheath to yield the MgB₂ mono-core wire (See Fig. 6). The reported superconducting area of the monofilamentary wire is ten times smaller than the seven-filamentary topology, determined to estimate the wire $I_C$. RLI monofilamentary wires tested at up to 10 T and 4.2–30 K exhibited $J_C$10×103A/mm² at 4.2 K-1 T.

The applicability of MgB₂ superconducting coils for space applications in [65] points to the weight factor, where MgB₂ filaments enclosed by either thin or thick walls represents a lighter option than BSCCO. $J_C$ exhibited inverse correlation with wire strain at 4.2 K. The Fe-sheathed annealed-bent wire exhibited a $J_C$ that is four times lower than the bent-annealed wire at a radius of 1 cm. However, constructive factors of MgB₂-based coils such as structural support and thermal expansion coefficients may detract from the MgB₂ wire superconducting performance. In [66], authors presented MgB₂ wires produced via RLI that yielded a 0.28 wire fill factor. Volume shrinkage of up to 25 % was reported, hence ascribed as a cause of low property homogeneity in tubular MgB₂ wires. Carbon-doped MgB₂ wires presented lower electrical resistance than the undoped form at B=0-4.5T. Nonetheless, tubular wires opened the precedent for cryogenic cooling with liquid Ne at temperatures around 27 K. The carbon-doped modality of MgB₂ wires produced via RLI exhibited an $B_{C_{2}}$ of 5 T, congruent with the findings for the undoped modality, but at 26 K [47]. In that case, the C-dopant at a 6 % of B weight was enveloped by an Nb barrier. Addition of SiC and carbon nanoparticles culminated in higher $B_{C_{2}}$ induced by the C-substitution in MgB₂ sites.

To reduce chances of Cu migration into the MgB₂ cores, various metal matrices such as Fe and Nb barriers for in-situ wires were reported in [67], where a technique named Continuous Tube Forming & Filling (CTFF) combined with PIT yielded monofilamentary MgB₂/Fe and MgB₂/Nb/Cu/SS. For both wire topologies, in-situ reaction of MgB₂ with an added 8 % of SiC yielded improvements in the $B_{C_{2}}$ and $J_C$ of the wire at 4.2 K. The estimated $J_C$10⁶A/cm² in self-field and high n-factor hinted about potential the use of CTFF + PIT to produce MgB₂-based coils for operation in persistent mode. Authors in [49] performed CTFF of MgB₂/Nb/Cu/Fe wires with seven filaments utilizing 8 % SiC. The SiC-doped, Fe mono-core wires exhibited higher $J_C$ and $B_{C_{2}}$ compared to undoped Fe-monocore and multifilamentary MgB₂. SiC- and ethyltoluene-doped MgB₂ tapes produced via in-situ PIT exhibited significant $J_C$ improvements from 33-225A/mm² at 4.2 K under 10 T [68]. With the heat treatment between 600–800 °C, Arc-plasma of MgB₂ with ethyl-toluene (C₆H₉) and 3.7 % SiC additive yielded the highest $J_C$ at both 4.2 K-10 T and 20 K-5 T. In [69], authors analyzed the sintering temperature effects in the ex-situ MgB₂ grain morphology. As expected, grain sizes increased with the sintering temperature. Although wires sintered at 1050 °C yielded slightly higher fill factor and grain size than the obtained at 950 °C, MgB₂-950 °C presented overall best superconducting performance with $J_C$3×10⁵A/cm² at 2 T.

Motivated by the challenges in IMD, CTFF and PIT for MgB₂ wire production, [70] presented a novel multi-stage manufacturing process for MgB₂ multifilamentary cables. The approach addresses the MgB₂ production from the in-process purification of precursor powders. Nanosized Boron powder is mixed with Mg and Cu then deposited on a U-shaped composite SS-Cu sheath. Unlike PIT and IMD, this method encompasses laser welding of the sheath containing the in-situ powder. While the process comprised a sintering step, no details on the technology or process temperature were reported. In a subsequent work, authors in [71] addressed the in-situ MgB₂ mono-core wire production with 10 % SiC on Monel/Titanium sheath through the same process demonstrated by [70]. The 0.75 mm-diameter samples sintered at 700 °C for 10 and 30 minutes, respectively, at 1 T, exhibited $I_C$ below 200 A.

Building upon [70], [71], authors in [72] compared the $J_C$ of SiC-added, in-situ MgB₂ wires sintered in (i) batches and (ii) in line. S2 (in-line continuous sintering) yielded wires with comparable $I_C$ (i.e., between 140A-180A) in self-field with S1 (direct insertion in pre-heated furnace) wires at comparable sintering times below 40 minutes. S3 (batch sintered) wires yielded $I_C$ 170A in self-field, but sintering the whole batch took 20 h to yield comparable properties with S1 and S2-sintered wires. Besides, the in-situ reaction dynamics of Mg + B + SiC requires further study for this process. The SiC as a dopant may trigger the growth of non-superconducting phases such as MgSi, B-C amongst other byproducts which may culminate in the reduction of %wt. MgB₂/wire length; hence deteriorating the wire superconducting performance. Mechanical milling of Mg with coronene and nano-sized Boron powders yielded a 0.5 mm-diameter, mono-core MgB₂ wire with $J_C$10³A/mm² under 10–20 K and 6.1–3.3 T conditions [73]. The Boron nano-powder reacted with Mg and C₂₄H₁₂ in-situ to form the superconducting phase doped with carbon. Heat treatment under 600 °C-3 h yielded a high packing ratio, caused by the reduced porosity in the in-situ process due to the addition of coronene. The addition of a carbon source triggered the Boron substitution in MgB₂ sites, which contributed to the creation of pinning centers; therefore, improving the wire $J_C$.

In [74], authors presented an electromechanical assessment of in-situ PIT of a 10-filament MgB₂ wire produced with Coronene dopant. Wire necking was reported at 4.2 K, B = 0 T and ∊ 0.4-0.45%, but no apparent signs of wear were seen at 10 T. Tests at 300 K, 77 K, and 4.2 K exhibited the wire critical tensile strain increases with the temperature reduction. Between 7–10 T, $I_C$ decreased exponentially from around 350 A to 50 A. Such values remain some of the best found in the literature for in-situ PIT MgB₂ wires. However, the application of PIT MgB₂ wires in transmission lines requires further testing on $I_C$ stability under strain and temperature variation. Indentation tests performed in Dai et al., 2017 [75] using a 36-filament, in-situ PIT-produced MgB₂ wires yielded an inverse correlation between indentation depth and $I_C$ in self-field. Under fields of 2–7 T at 4.2 K, an indentation depth of 0.12 mm yielded the maximum $I_C$ values. Comparison of $I_C$ in self-field measured in Bi-2212, Nb₃Sn, and NbTi cables of similar diameter evidenced the MgB₂ wire as the most sensitive to increases in the indentation depth. The impact of heat treatment conditions in MgB₂ multifilamentary wires was exhibited in [76]. The wires manufactured by Hypertech Inc and WST were cut to 135 cm pieces for $I_C$ measurement under magnetic fields between 1–6 T. Heat treatment for the WST wire, composed by 53μ m-thick filaments, took 4 h compared to the 2 h needed by the Hypertech wire with 38μ m-thick filaments. The heat treatment yielded a decrease of less than 5 % in $I_C$ for both wire topologies. Nevertheless, further study is required to understand how both wires perform under strain and what role degradation plays in the decrease of $I_C$ over time.

Supporting computational work for $I_C$ prediction in multi-strand MgB₂ superconducting wires was reported in [77]. Wires of 0.36 mm diameter from nine batches were utilized, for which IC100A at 25 K. The model encompassed the E-J power law to predict $J_C$ as a function of temperature targeting current improvement for use of MgB₂ in Superconducting Fault Current Limiters (SFCLs). Studies on the thermal conductivity of MMC-MgB₂ wires in MRIs were carried out in [78] where a finite element approach was taken to elucidate the thermodynamic compatibility between the MgB₂ core, Nb and copper barriers, as well as the Monel matrix. Thermal conductivities, Poisson Ratio, and Elasticity Modulus of each component measured between 10–300 K showed the need for understanding composite structure properties prior to magnet design such that failure mechanisms (i.e., quenching) are avoided.

Modeling of MgB₂ coils for application in SMES was presented in [79], where React and Wind (i.e., R&W) and Wind and React (i.e., W&R) coils produced by Hypertech Inc and Columbus superconductors served as testbeds. Strain analysis and distribution exhibited a direct relationship of strand deterioration with the $I_C$ decrease and quenching for 30 kJ of energy storage in the coils under liquid hydrogen cooling. A relevant study was then proposed in [80] where a 2D electromagnetic model was developed to find optimum filament positioning for reduction of AC losses in MgB₂ multifilamentary wires. The optimum ordering was exhibited for a 13+0f wire, where filaments are positioned near the outer surface of the metal matrix. In that case, the magnetic flux density mapping indicates low intensity of $B$ in the wire core, culminating in a linear increase of AC losses rather than a piecewise linear-exponential curve shown in the 6+6+1f and 7+6f wires.

Further experimental validation on such a modality of electromagnetic studies may support development of novel MgB₂ wire topologies with improved performance and enhanced quenching protection. A combined approach of $I_C$ prediction and magnet finite element analysis would help understand the role of material compatibility (i.e., thermal conductivities, heat expansion, stress and strains) in the operation of MgB₂ magnets.

Processes for MgB₂ powder consolidation have been largely explored over the years. Research initiatives explored grain growth, microstructure properties, and influence of dopants in bulks’ superconducting and mechanical properties for application in superconducting joints and magnetic levitation devices. Across the techniques employed, Hot Pressing (HP) [81], Hot Isostatic Pressing (HIP) [82], and Spark Plasma Sintering (SPS) [83], [84] remain as effective methods to achieving high-quality bulks for superconducting applications of MgB₂. Despite such efforts, a plethora of process-related challenges constrains aspects that affect the control of MgB₂ bulk properties (i.e., mechanical, and superconducting), consequently slowing down any progress toward increasing their Technology Readiness Level (TRL).

Initially, [85] reported MgB₂ bulks produced at 720 °C using ball-milled Magnesium Hydride (MgH₂) and powder B as precursors. For this case, the SPS processing temperature was kept below 897 °C to prevent the MgB₂ decomposition, and consequent creation of higher borides (i.e., MgB₄), and amorphous Mg. Yielded MgB₂ bulks presented a temperature onset of 37.3 K. Subsequently, [86] proved the relationship between pressure and temperature with the pinning mechanisms and levitation force in MgB₂ superconducting bulks. Additions of SiC, C, and Ti have been shown to improve the flux trapping performance, yielding improvements in $J_C$. Although MgB₂ exhibited only 2.5 %wt. compared with higher borides such as MgB₇ and MgB₁₂ for bulks processed under 2 GPa, a $T_C$37.6K remained relatively close to the theoretical value of 39 K. Bulks processed via SPS caused the boron and oxygen colonies to increase the flux pinning forces if compared to point or grain boundary pinning centers. In-situ hot-pressed MgB₂ bulks with added SiC at 1050 °C-2GPa showed the highest pinning force density (i.e.,$ \mathrm{F}_{\mathrm{P}}=10.9 \times 10^{9} \mathrm{~N} / \mathrm{m}^{3}$) amongst all samples. In a similar approach to [86], authors in [87] investigated the SPS of in-situ, SiC-added MgB₂ bulks with high Mg deficit on ratios B/Mg between 3.75–1.87. The sample with highest deficit of Mg (S1) presented the $\mathrm{T}_{\mathrm{C}} 36.6 \mathrm{~K}, \mathrm{~B}_{\mathrm{C}_{2}}=36.37 \mathrm{~T}$, and the highest $J_C$10⁶A/cm² in self-field amongst all S1-S3 samples. Besides, the absence of Mg₂Si hinted about the unreacted amount of SiC in the sample. In [88], authors presented nearly 100 % dense MgB₂ bulks processed via HP-SPS under 1.7-5GPa. This approach exhibited reduced %wt. of non-superconducting phases and controlled MgB₂ grain orientation that allows for the control of supercurrents. To illustrate both intra and inter-grain supercurrents, Fig. 7 shows the dynamics of both micro and macro $J_C$ based upon the work reported in [89].

In another attempt to improve superconducting properties of MgB₂ bulks, [90] presented $J_C$ improvement in MgB₂ bulks by adding Boron Nitride (BN) at 1, 3, and 5 %wt. In multiple cases, BN as a dopant caused the shrinkage of the MgB₂ lattice parameter a, while slightly stretching c. The samples ‘b’ with pure MgB₂ and ‘c’ with MgB₂(h-BN)_0.005 exhibited the highest $F_P$ values at both 5–20 K. Overall, the addition of h-BN in ex-situ MgB₂ sample did not yield significant changes in either the $F_P$ or $J_C$. Conversely, c-BN in sample ‘f’ (i.e., MgB₂(c-BN)_0.005) and ‘g’ (i.e., MgB₂(c-BN)_0.01) exhibited good improvement of FP to values between 4.77-5.37×10⁹N/m³, and yielding similar $J_C$ behavior for both samples at 5–20 K. Authors in [91], presented SiC-doped MgB₂ samples processed in-situ, via one- and two-step SPS at 850 °C. Pure MgB₂ bulks exhibited $T_C$37.8K, being highest than the others reported. The molten Mg in the in-situ reaction Mg + 2B bulks with 10 %wt. SiC was shown to weaken the Si-C bonds, incorporating C into the MgB₂ lattice, and causing Mg to eventually react with Si to form Mg₂Si. The impact of such in-situ reactions in the superconductivity of MgB₂ bulks was also studied by [86], [87].

Further, [92] reported that spark plasma sintered GeO₂-added MgB₂ bulks germanium-based dopants in MgB₂ bulks were reported by exhibited improved $B_{C2}$ as a result of the creation of pinning centers similar to the carbon effect in SiC added MgB₂ samples. The addition of GeO₂ increased flux pinning in grain boundaries, as shown in samples MgB₂(GeO₂)_0.005 and MgB₂(GeO₂)_0.01 samples that yielded comparable $J_C$ to the pure bulk, but stood out with the highest $B_{C2}$ between 8.2–7.5 T at $J_C$=103A/cm2. However, for such a work, pure MgB₂ samples yielded the $J_C$10⁶A/cm² in self-field at 5 K, the highest across all bulks. The addition of both germanium and carbon-based sources for flux pinning center creation was done by [93], where Ge₂C₆H₁₀O₇-added, ex-situ SPS MgB₂ bulks exhibited $J_C$10²A/cm² at 5.8 T for the MgB₂(Ge₂C₆H₁₀O₇)_0.0014 form. Such a precursor for Ge and C caused the $T_C$ onset of the bulk to decrease from 38 K to values as low as 33.4 K, which can be ascribed to the sharp reduction of MgB₂ %wt. in the bulk. Interestingly, compared to the doped samples, the pure MgB₂ bulk yielded the poorest $B_{C2}$ performance at temperatures between 5–20 K.

Ex-situ SPS of MgB₂ bulks doped with c-BN and C₆₀ was reported in [94]. Fullerene served as a primary C source in the MgB₂ bulks, where a 92.7 % relative density, 10.3 %wt. MgB₄, and the highest residual strain amongst all samples equaling σ=0.88% were reported. Tests for the pure MgB₂ bulk exhibited a $J_C$0.9×10⁶A/cm² in self-field, the highest amongst all samples. Pristine MgB₂ and carbon-added bulks showed similar performances for $J_C$ at 0–3 T and 5 K, but the pure sample exhibited a sharp $J_C$ decay past 4 T. At 20 K, MgB₂ and (MgB₂)_0.95(C)_0.05 bulks have similar $J_C$ in up to 2 T. The rate of decay for the MgB₂ past this region is sharper than in the (MgB₂)_0.95(C)_0.05. Such an effect can be ascribed to the C replacement of B in MgB₂ sites, creating flux pinning centers that improve the in-field $J_C$. Overall, the concomitant addition of C₆₀ and c-BN in ex-situ MgB₂ did not yield significant results concerning flux pinning, similar to the Ge and C added dopants in a previous attempt. In [95], authors obtained 99 % dense, pure MgB₂ bulks processed by ex-situ SPS, for which a $T_C$38.5K was reported. MgB₂ samples produced under 50 MPa-10^-3 bar vacuum pressure, and heated at 1050 °C-1200 °C for 20 min yielded $J_C$2×10⁵A/cm² in self-field. The measured trapped field equivalent to 1.4 T at 10 K was also reported for the 20 mm diameter sample. Authors in [96] built upon [95] and reported the flux trapping at 10–25 K by the surface and interior of a stacked MgB₂ bulk structure.

In [82], authors took an approach to solely characterize the mechanical performance of MgB₂ bulks produced via HIP. The 70-mm diameter bulks produced under 98 MPa-1173 K for 3 h yielded bulks with a 92 % packing ratio. On average, the fracture strength of such bulks at 77 K equated to 257 MPa, whereas testing at room temperature indicated an average of 220 MPa. Comparatively, MgB₂ bulks yielded higher fracture strength then high-density REBCO bulks produced under similar conditions. Furthermore, with 100 % packing ratio, the predicted fracture strength of the MgB₂ bulk is 419 MPa. Furthermore, test results in [97] also indicated that the processing temperature in SPS improves the MgB₂ bulk packing ratio, contributing to the increase of both Young Modulus and fracture strength. The three MgB₂ bulks were produced via SPS under a 10^-3 bar vacuum atmosphere and temperatures 950 °C, 1000 °C, and 1100 °C. [98] analyzed the influence of dwelling time in superconducting properties of MgB₂ bulk samples produced via SPS. In it, a variety of ex-situ samples were submitted to periods of 1, 4, 7, 10, and 20 minutes. The bulk ‘20 m’, whose description was ascribed according to its SPS dwelling time, presented 97.1 % relative density: the highest among all bulks. Nevertheless, $J_C$ measured between 10–25 K indicate that samples ‘1 m’ and ‘4 m’ exhibit better properties for 0–8 T altogether. At zero-field and 100 A/cm², the $B_{C2}$ of ‘1 m’ and ‘4 m’ presented smoother decays over temperature increases. Overall, results highlighted that the dwelling time did not impact either the microstructure or superconducting properties of the bulks in a significant manner.

In [99], authors measured the magnetic levitation force of in-situ MgB₂ bulks produced with a surplus range of 5–30 % Mg added to the Mg + 2B mixture. XRD patterns exhibited increases in the Mg content as the %wt. Mg added beyond the ratio Mg:2B. The optimal $T_C$ onset at 38.6 K for the MgB₂ + 10 %Mg bulk, whereas the highest repulsive force $F_z$ measured was slightly above 22 N for the MgB₂ + 15 %Mg sample. The lateral force $F_x$ also provided insights on how MgB₂ bulk properties can be tailored toward utilization in magnetic bearings for electromechanical energy conversion devices such as induction motors. Subsequently, [100] reported studies on the pinning force in MgB₂ bulks sintered via SPS and other sintering techniques. The SPS-processed MgB₂ bulk presented a $\mathrm{F}_{\mathrm{P}}=300 \times 10^{6} \mathrm{~N} / \mathrm{m}^{3}$, which is more than twice as high as the maximum achieved by the non-SPS sintered samples. A $J_CSPS $ of 300×10³A/cm², flux trapped of 1.45 T at 20 K, and high pinning forces are ascribed to the microstructural control provided by SPS compared to other sintering techniques. [101] reported improvements in $J_C$ and $T_C$ of ex-situ MgB₂ bulks with added Y₂O₃ nanosized powder. Both powders were ball-milled prior to the Field Assisted Sintering Technology (i.e., FAST, namely SPS) step. With all samples sintered at 1150 °C-50 MPa with dwell time of 5 minutes, $J_C$10⁹A/m² at 4.2 K-3 T for 0.5–2.0 %wt. of Y₂O₃ has been reported. Such samples were ball-milled for 12 h. At 20 K, the pure MgB₂ sample exhibited higher $J_C$ for fields up to 1.5 T, followed by the MgB_{2 +} 0.5 % wt. Y₂O₃ bulk. The bulk MgB₂ with 0.5 %wt. Y₂O₃ sintered for 12 h exhibited a $J_C$ that is twice as large as for pure MgB₂ sample. The increase in the $B_{C2}$ at both 4.2–20 K provided substantial evidence of superconducting property improvement via dispersion of Y₂O₃ as a mechanism to create artificial pinning centers in MgB₂ bulks.

In [102] sintered four ex-situ MgB₂ samples between 900–1200 °C to analyze the processing temperature effects over the sample properties. While no significant changes occur in $T_C$ according to the provided estimate, bulks processed at 1000–1200 °C exhibited higher relative densities and higher $J_C$ if compared to the 900 °C-sample at 4.2 K-5 T, 20 K-0 T, and 20 K-3 T. Nevertheless, the 900 °C-processed bulk exhibits a $B_{C2}$=6.5T at 4.2 K, which is slightly above the 6.1 T exhibited by the samples produced at 1000 °C and 1100 °C.

Technology Readiness Levels (TRL) were defined by NASA [114] in the 1980s as a benchmark for technologies applied to space missions, in which the maturity level of was assessed in a 1-9 scale (See Fig. 10). Since then, such a tool has been utilized in benchmarking technology according to their level of maturity. The TRL of fusion devices involves very specific challenges concerning the plasma stability, fuel fabrication and performance, and waste management as proposed in [115] but leaves superconducting technology aside. As discussed by [45], [51], [116], further development of MgB₂ poloidal coils may enable a potential replacement of Nb-based coils with a superconductor at higher temperature. However, the literature on MgB₂-based poloidal coils is scarce, with just a few publications since the discovery of its superconductivity. To leverage such challenges related to MgB₂ applications in large-scale devices, Fig. 10 provides a TRL map of milestones for superconducting device manufacturing, qualification, and operation.

MgB₂ wire topologies were explored in [117] for implementation in the IGNITOR fusion prototype, a collaboration between Italy and Russia. However, this project has not progressed further than the design stage, having had its last update in 2022 with a field-coil design revision [118]. Studies on the development of MgB₂ coils reported in [116] highlight the short-lived radioactivity of MgB₂ compared to the induced radioactivity in Nb-based superconductors. In such an effort, a 100-meter wire section was fabricated based on MgB₂ with added Cu, which yielded a 2.62 T field. Other studies concerning the effects of cable indentation in the superconducting properties were reported in [75], followed by experimental work on both DC and AC performance in MgB₂ Cable-In-Conduit Conductor (CICC) superconducting wires [45], and heat transfer analysis for quench mitigation in multifilamentary MgB₂ wires [44].

Additional barriers are yet to be overcome for the MgB₂ application in poloidal magnets of fusion systems, whose main requirements for stability and reliability related to withstanding extremely high mechanical stresses and electrical currents [119], [120], [121]. Fusion reactors require superconductors to carry high current densities in order to induce multi-Tesla, steady magnetic fields for plasma confinement for extended periods, while withstanding radiation effects along with a multitude of thermal, magnetic, and mechanical stresses. Currently, Nb₃Sn has been utilized in facilities such as ITER [122] and FNSF [123], [124], but the required cryogenic cooling power needed for Nb₃Sn superconductors is one of the energy expenditure-based roadblocks faced by current magnetic confinement NFRs. Compared to Nb₃Sn-based superconductors currently utilized in experimental reactor facilities, MgB₂ is a promising alternative for magnets in nuclear fusion in terms of affordability and manufacturability, $T_C$, $J_C$, and $B_C$.

Methods such as isotopic doping of MgB₂ have been assessed, where Mg¹¹ was employed to MgB₂ bulks [125]. Such experiments provide an understanding of how MgB₂-based joints would operate within a nuclear fusion-relevant environment. While the so-called “Isotope Effect” was proven to interfere in the phonon frequency of the MgB₂ lattice and cause improvement in superconducting properties of such a material, further work is required to benchmark such a behavior for MgB₂ poloidal coil structures. Notably, superconductivity testing for MgB₂ components in fusion-relevant environments still requires work and could hardly be found in the literature. This indicates the infancy of MgB₂ applications in fusion-related applications, consequently placing such a technology between TRLs 1-3.

This manuscript presented and described the state-of-the-art manufacturing techniques to produce MgB₂-based superconducting components. Four different modalities of MgB₂ components were described: thin films, tapes, wires, and bulks. To date, the HPCVD remains a step above other deposition processes of MgB₂ thin films due to yielding low RMS surface roughness. Nanofilms of MgB₂ produced via such a technique exhibit better properties at comparable scales. Nevertheless, applications of MgB₂-based Josephson Junctions still require addressing issues such as superconductivity of in the interfacial layers of electrode-barrier. Improvement in the coherence length, film roughness, and engineering of flux pinning centers remain open questions. Materials compatibility for tape production based on deposition techniques still require further control of grain growth modes.

Similarly, manufacturing of MgB₂ wires and bulks share several of the challenges shown in the growth of thin film. The control of in-process phase for property homogeneity remains a multi-scale issue that relates from the compound physical-chemical properties to the constructive aspects of deposition reactors or sintering devices. Production scales are connected to the process physics, in which the material modality (i.e., powder, precursor gas, and liquid) determines the thermodynamic behavior of the MgB₂ device. Growth modes are affected by process parameters such as pressure, temperature, deposition rate, and process atmosphere inertness. Yielded properties are driven by process conditions and controls, material purity and production scale. Such factors corroborate the performance of MgB₂ superconductor devices, stability, and lifecycle. Successful strides in the directions pointed out in this manuscript may support closing the gap for application of MgB₂ in a multitude of superconducting devices.

In summary, MgB₂ remains as an alternative to a future replacement of Nb₃Sn and NbTi for low-field applications and permanent magnets. To achieve this, hurdles concerning property reproducibility, in-operation stability, and improved process controllability are still open questions. Based on the applications evaluated in the literature, Fig. 11 exhibits a systematic diagram containing the modalities of MgB₂ components driven by their respective applications. While the dimensions of devices described range from nanometer to the kilometer scales, the challenges concerning performance of MgB₂ superconducting devices require control of the process thermodynamics. The precision of the manufacturing process dictates the superconducting phase stability, orientation, and stoichiometry; therefore, determining the operational stability and lifecycle of the devices.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Supplementary data to this article can be found online at https://doi.org/10.1016/j.supcon.2023.100083 https://doi.org/10.1016/j.supcon.2023.100083.