*3.1. Novel Strengthening Solution Based on SME of Fe-SMA*

High-cycle fatigue causes crack propagation in structural components, a case which in turn leads to the deterioration of stiffness/strength and shortens the service life of the structure. Practice has proven that through strengthening the cracked or damaged components, the service life can be effectively prolonged. Existing strengthening strategies include using external bonding reinforcing materials or applying prestressing. Overcoming some

possible shortcomings such as the difficulty in construction for the traditional reinforcing strategies, a new method utilizing Fe-SMA has received great attention. The main procedure of the Fe-SMA-based strengthening solution is similar to that of the aforementioned SME-triggered tightening method and can be summarized as follows (see Figure 13):


**Figure 13.** Schematic illustration of procedure for Fe-SMA-based strengthening/connecting solution.

3.1.1. Strengthening for Reinforced Concrete (RC) Structures

The earliest practical application of Fe-SMA in the field of prestressing can be traced back to 2001 where a bridge in Michigan, United States, experienced fatigue-induced cracking [120]. In this case, Fe-SMA tendons were installed perpendicularly to the shear cracks. After electric heating, a recovery stress of approximately 225 MPa was induced in the Fe-SMA tendons. Field measurements indicated that the generated recovery stress closed the width of the shear cracks to a large extent and the load-carrying capacity of this bridge was effectively recovered.

Since the initial success, laboratory research works have been conducted on concrete structures strengthened by Fe-SMA reinforcement, and suitable anchorage systems have been developed. For example, EMPA (Swiss Federal Laboratories for Materials Science and Technology) proposed an anchorage system for Fe-SMA tendons employed as near-surface mounted reinforcement (NSMR) in RC structures [121–123]. As shown in Figure 14a, Fe-SMA prestressing elements are embedded in pre-made grooves and covered with adhesive material such as cement-based mortar herein. Lap-shear experiments have been carried out to clarify the bonding behavior between the Fe-SMA strips and cement-based mortar [122]. Deeper embedment depth and ribbed surface for Fe-SMA strips are recommended for practical application [124]. It is also found that the current design guidelines would underestimate the necessary anchorage length for Fe-SMA bars [125]. The corresponding calculation methods are yet to be available.

With the aim of further simplifying the anchoring process and satisfying the objective of rapid recovery on site, a new anchoring method employing shotcrete is proposed (see Figure 14b) [126]. The previously tensioned Fe-SMA tendons are installed beneath the beam with an additional cementitious layer (shotcrete) sprayed on, covering the Fe-SMA tendons. After sufficient curing, current resistance heating is applied for activation and prestress is induced. Feasibility studies on flexural strengthening [126] and shear strengthening [127] of RC beams with this anchoring method have been conducted. Both the test results revealed that this strengthening system can efficiently increase the flexural/shear-resistant performance of RC beams. At the same time, beam deflections, number of cracks, and the widths of cracks were all reduced.

The flexural behavior of RC beams strengthened by the Fe-SMA NSMR system was investigated in [128], where ribbed Fe-SMA strips were longitudinally embedded at the bottom of the beam. Copper clamps were used to transmit the electric current. After current resistance heating to a target temperature of 160 ◦C, a permanent prestress of about 200 MPa was created in the Fe-SMA strips. Rojob et al. [129] conducted a comparative experimental study on the effect of strengthening through CFRP strips and Fe-SMA strips, and confirmed that the Fe-SMA strips lead to better ductility of the beam. Rojob et al. [130] further added expansion anchor to this system (see Figure 14c) and found that the ductility of the RC beam was further improved. This is because the additional expansion anchor

provides an extra force transmission path, which maintains beam function after the Fe-SMA tendons are stripped from the adhesive material.

Nail-based mechanical anchorage system is an alternative method for the Fe-SMA NSMR system. As shown in Figure 14d, Fe-SMA strips can be easily fixed to the surface of the base concrete layer by the aid of direct-fasteners (e.g., nails) and nail-setting devices [131]. The total duration for installing and activating a 5-m Fe-SMA strip is within 20 min [132]. This method has been applied to some retrofitting cases in Switzerland [132]. However, nail-based anchorage systems may not be the best solution for bridges since the nails tend to loosen under HCF-loading conditions.

It should be noted that for RC structures, the target temperature during electric heating should be carefully controlled, since high temperatures may cause concrete cracking/damage and could be detrimental to the bond strength between Fe-SMA and concrete [73]. As reported in [125], a longitudinal splitting crack with a width of about 0.05 mm appeared in the mortar surface when the Fe-SMA was heated up to 190 ◦C. Most existing studies adopt a maximum activation temperature of around 160 ◦C, which can be regarded as a feasible target temperature.

#### 3.1.2. Strengthening for Steel Structures

The friction-based mechanical anchorage system, which is feasible for strengthening steel structures with Fe-SMA prestressing strips, was first developed by EMPA, as shown in Figure 15a [133]. Glass-fiber-reinforced plastic (GFRP) laminates and friction foils are also involved in this anchorage system, along with clamping plates and bolts which are necessary for anchoring. The GFRP laminates electrically insulate the Fe-SMA strips from the steel plate during the activation procedure, thus avoiding energy waste and reduction of heating efficiency. Extra friction foils are often used to increase the static friction coefficient for this joint. The experiment was first conducted on simple steel plates, and it was shown that a 2% pre-strain of the Fe-SMA strips can produce a recovery stress of about 330–410 MPa after heating to 260 ◦C, resulting in a compressive stress of about 35–74 MPa in the base steel plates. A fatigue test was further conducted and the results proved that the fatigue life of these strengthened steel plates was evidently increased and the propagation of initial cracks was postponed and even arrested in some cases [134]. Appropriate modifications were subsequently made and the applications were extended to fatigue strengthening of metallic girders (see Figure 15b) [135,136] and connections (see Figure 15c) [137]. Similar conclusions were drawn from these works.

Recently, a novel fatigue strengthening solution for steel structures using adhesively bonded Fe-SMA strips was investigated by EMPA. The adhesive Sika1277 was used to bond the Fe-SMA strips to the steel plates [138]. It is reported that the bonding force is approximately twice the prestress achieved in the Fe-SMA strips. No softening behavior was observed during the activation process, which means that the adhesive can securely anchor the Fe-SMA strip throughout the whole strengthening process [139]. Due to the bridging mechanism of the adhesive anchorage, crack opening in the base structure was suppressed and stress singularity at the crack tip was also significantly reduced [139]. However, future studies are still needed to investigate the time-varying behavior of this bonding-based anchorage system during the entire service life.

**Figure 15.** Schematic illustration of friction-based mechanical anchorage system for Fe-SMA strips in steel structures: (**a**) Configuration details; (**b**) Strengthening of metallic girders; and (**c**) strengthening of metallic connections.

## *3.2. Seismic Dampers*

Conventional metal dampers are usually made of steel with a reasonably low yield strength, which encourages early participation in energy dissipation. Ductility and durability are also important characteristics, since many strong ground motions followed by a series of aftershocks have been recorded in the past decades [140]. A Japanese industry– academic–government joint research group had developed Fe-SMA-based (Fe-15Mn-4Si-10Cr-8Ni) buckling restrained shear dampers (BRS, see Figure 16a) and buckling-restrained braces (BRB, see Figure 16b), and used them in the JP Tower Nagoya and The Aichi International Convention & Exhibition Center, Tokoname, respectively [28]. Related experiments have been conducted [23,28,75] and the results confirmed that the Fe-SMA seismic dampers exhibit considerably longer fatigue life (around ten times) than conventional steel dampers. Loading tests with random seismic wave inputs were also performed and the results showed that the seismic dampers exhibit stable energy absorption behavior under a wide range of deformation angles, reflecting a reliable performance during earthquake sequences [28].

A more comprehensive experimental study on Fe-SMA (Fe-17Mn-5Si-10Cr-5Ni) BRS was conducted by the authors and co-workers recently [23] (Figure 17a). Loading protocols with constant and incremental symmetrical shear displacement amplitudes, marked as 'protocol I' and 'protocol II' in Figure 17b, respectively, were employed to investigate the hysteresis response of the Fe-SMA-based BRSs. Such loading protocols were also conducted on steel (Q235) BRSs with the same geometry, and the test results are compared in Figure 18. The hysteretic loops of Fe-SMA-based BRSs are slightly narrower than those of steel BRSs (half-life cycle EVD = 0.42 vs. 0.52 under constant displacement amplitude), which is consistent with the material-level observation described previously. Importantly, significantly enhanced fatigue resistance was achieved in the Fe-SMA-based BRSs, leading to a considerable increase in the total accumulated energy dissipation (*E*T) (Figure 19). Figure 20 shows the final crack patterns of Fe-SMA- and Q235-based BRSs in this experiment. The cracks of the Fe-SMA-based BRSs tended to be initiated in the center region of the core plates, with a subsequent crack propagation to the arc-shaped edge,

whereas the cracks of the Q235-based ones were initiated from the arc-shaped edge region. Research opportunities exist in further investigating the reasons behind the difference in the fracture mechanism between Fe-SMA and steel shear dampers.

**Figure 16.** Configuration of Fe-SMA-based (**a**) BRS; and (**b**) BRB developed by a Japanese industry– academic–government joint research group.

**Figure 17.** Laboratory experiment of Fe-SMA-based BRS: (**a**) illustration of test setup; and (**b**) loading protocols.

**Figure 18.** Hysteretic behavior for Fe-SMA- and Q235-based BRS under: (**a**) Protocol 'I'; and (**b**) Protocol 'II'.

**Figure 19.** The total amount of accumulated energy dissipation (*E*T) for Fe-SMA- and Q235-based BRS: (**a**) *E*<sup>T</sup> vs. Cycles under Protocol 'I'; and (**b**) a comparison of *E*<sup>T</sup> between these two metal BRS under different loading protocols.

**Figure 20.** Macroscopic fracture behavior of (**a**) Fe-SMA-based BRS; and (**b**) Q235-based BRS under loading protocol 'I'.

More recently, the authors and co-workers have completed a series of tests on Fe-SMA U-shaped dampers, as shown in Figure 21a. It was found that the LCF life of the Fe-SMA dampers is 5–7 times that of their steel counterparts. A representative Fe-SMA damper hysteretic curve is shown in Figure 21b, where full and stable hysteresis curves are observed. The details of this experimental program will be published in a separate paper.

**Figure 21.** Fe-SMA U-shaped damper tests: (**a**) test setup; and (**b**) hysteretic response under incremental loading protocol.

## *3.3. Advantages Compared with Alternative Solutions*

Summarizing the above studies and applications, the main advantages of the Fe-SMA solutions are further elaborated here. In the field of retrofitting, the prestressing process of Fe-SMA strips is easier than that of the CFRP tendon-based reinforcement which currently prevails [133,136,139,141]. This is mainly because the former can be activated through electrical heating without any heavy hydraulic jacks or dedicated mechanical clamps. Moreover, the required fire protection for Fe-SMA strips can be less demanding than that required for CFRP strips. Studies have found that Fe-SMA strips have a positive effect in countering the relaxation behavior of structures at elevated temperatures [142]. Furthermore, corrosion, creep, and relaxation behavior under extreme environments (e.g., high-temperature and chlorine environments) have been investigated which confirmed the suitability and reliability of Fe-SMA [120,143–150]. Another attractive advantage of Fe-SMA-based strengthening strategy is its re-prestressing property, i.e., HCF-induced relaxation in recovery stress can be restored through repeated rounds of thermal activation [18,134,151]. This process is simple and can be implemented without the necessity for the time and labor-intensive disassembling procedure.

As for the Fe-SMA-based seismic dampers which have attracted great attention in the community of seismic designers, the fatigue-free and maintenance-free seismic design ambition may become possible. These significantly benefit society since the maintenance costs including those related to hazards have become an important part of governments' expenditure over the years (for example, about 400 billion Euros are paid for maintenance of buildings in Europe [152]).

Despite the fact that the price of Fe-SMA is currently higher than conventional structural steel, the whole-life costing of Fe-SMA-based applications may be comparable with the conventional technologies. It is noted that Fe-SMA is much cheaper than NiTi SMA which is often criticized for its high cost [153–157]. Existing studies also found that the total costs of Fe-SMA- and CFRP-based strengthening solutions are comparable when the cost of dedicated mechanical clamps for prestressing CFRP strips as well as the cost of labor were considered [141]. Moreover, long term costs can be saved to some degree due to the low maintenance requirements and the corrosion-resistance property of Fe-SMA. Importantly, since Fe-SMA shares similar production process to stainless steel [9], it is optimistic to predict that the price of Fe-SMA has the potential to approach that of stainless steel as long as the demand matches with production quantities.

#### **4. Further Research Needs**

There is still a gap to be filled for a comprehensive understanding of Fe-SMA's mechanical properties towards more confident application in practice. Metal plastic forming and heat treatment have a strong influence on the mechanical properties of Fe-SMA. Coldworked Fe-SMA displays higher yield strength and higher shape recovery stress than that of hot-rolled ones [76], and the stress-strain curves may also show differences. The rationale behind the influence of these factors has not been comprehensively studied, and there are still limited constitutive models developed for Fe-SMA.

Welding is another important aspect for engineering application of Fe-SMA. Although the applicability of different welding technologies such as tungsten-inert gas, laser beam, and electron beam welding have been studied [158–162], some crucial welding properties and technologies are still under investigation. For example, in most of the existing experiments, Fe-SMA was welded to Fe-SMA or austenitic stainless steel in which the base metal shares the same metallography (γ-austenite). However, there has been little experimental research focusing on welding methods for connections between Fe-SMA and conventional structural steel. In addition, most of the existing studies focused on the effect of welding on the SME of Fe-SMA, whereas the mechanical properties of the weld itself and the heat-affected zone (HAZ) have been inadequately studied. Special attention should be paid to the fracture initiation mechanism of the weld, because the fatigue resistance as well as the plastic deformation capability of the weld and HAZ are usually inferior to the

parent material of Fe-SMA, a case which leads to early fracture of the weld zone before the LCF failure of Fe-SMA. Further research opportunities exist in seeking for reliable weld techniques for Fe-SMA-based components, especially in the field of seismic engineering.

From a fracture mechanics' point of view, previous research works have found that the crack distribution pattern of Fe-SMA-based dampers is very different from steel dampers [23], which may be related to the unique reversible martensitic transformation of Fe-SMA. However, an accurate explanation for the difference in the crack initiation and propagation behavior is still unavailable. Research opportunities exist in revealing how the microscopic feature of Fe-SMA would affect the macroscopic fatigue failure mode.

In future, Fe-SMAs may be used as next-generation structural steel due to its high strength, high ductility, excellent energy dissipation capability and LCF resistance during earthquakes. To this end, a more in-depth understanding of the mechanical behavior of Fe-SMA elements (such as springs, tendons, cables, etc.) and structural components (such as beams, columns, shear plate walls, etc.) is required. Since differences in mechanical properties exist between Fe-SMA and conventional structural steel, the applicability of the existing construction and design methods for normal steel to Fe-SMA-based structural components is questionable. In particular, a more unique system-level behavior is anticipated when the Fe-SMA elements are employed because of the special cyclic hardening and unloading spring-back behavior. Fe-SMA may also be used together with super-elastic Nitinol to reach a hybrid design which enables a good balance among self-centering capability, energy dissipation, and cost, compared with a pure Nitinol solution [44,163–167]. Preliminary seismic collapse safety assessment has revealed that the collapse capacity of beam-column joints equipped with Fe-SMA is significantly improved [168]. More studies are needed towards establishing a systematic and standardized design and construction process for Fe-SMA-based components and structural systems, and the entire design philosophy has to be revisited in future. Similarly, for the application of Fe-SMA in strengthening, further work is required to develop an integral design approach including the selection of materials, installation process, activation process, and quality check standard.

Last but not least, when one utilizes both the excellent LCF resistance and SME of Fe-SMA, a completely new structural design philosophy, namely, fatigue free and in situ recoverable structural design, may be developed. In other words, even if small residual deformation exists after an earthquake, a further deformation recovery may be promoted by heating the material via either electrical resistance or infrared heating, where the entire process is safe, quick, and economical. In this regard, the recoverable strain/recovery stress of cyclically 'trained' Fe-SMA (rather than that of the material under monotonic loading) is worth further studies.

## **5. Concluding Remarks**

Fe-SMA is an emerging high-performance metal with unique properties that make it well suited to many applications in the construction sector. Its shape recovery properties result from reversible martensitic transformation and have been utilized for prestressing/retrofitting structural components. It is foreseeable that the convenient prestressing process and sound re-activating properties would promote the material for wider use in the field of structural retrofitting. A series of anchoring systems for Fe-SMA strengthening solution have been developed for different structural types.

The use of Fe-SMA as seismic-resistant material began in the 2000s. Satisfactory ductility and energy dissipation capability are identified. The repeated tension and compressioninduced martensite, generated through cyclic loading, suppresses the micro-crack initiation and/or propagation which directly enhances the LCF life of Fe-SMA. These features make fatigue-free and maintenance-free seismic design philosophy possible in the near future.

Challenges and opportunities do remain for a more confident application of Fe-SMA in construction. For example, the influences of forming process and heating effects on the mechanical properties of Fe-SMA have not been fully investigated. Technologies for welding of Fe-SMA are needed. Constitutive models applicable to Fe-SMA should be proposed and the fundamental fracture mechanism of the material under different stress states needs to be clarified. Existing design approaches may be revisited when designing Fe-SMA-based structural components since many unique mechanical behaviors exist in Fe-SMA. The highly nonlinear stress-strain response, especially the substantial strain hardening characteristic may be a key design challenge for Fe-SMA-based seismic dampers. It is advised that more experiments on the component level and even system level should be conducted, seeking for a more comprehensive understanding of this material and its behavior in a structural system.

**Author Contributions:** Conceptualization, Z.-X.Z. and J.Z.; methodology, Z.-X.Z.; software, Z.-X.Z. and J.Z.; validation, Z.-X.Z., J.Z. and H.W.; formal analysis, Z.-X.Z.; investigation, Z.-X.Z. and J.Z.; resources, Z.-X.Z. and J.Z.; data curation, Z.-X.Z., Y.J. and J.Z.; writing—original draft preparation, Z.-X.Z.; writing—review and editing, Z.-X.Z., J.Z., D.D.K. and H.W.; visualization, Z.-X.Z., D.D.K. and J.Z.; supervision, Z.-X.Z. and H.W.; project administration, H.W.; funding acquisition, H.W. All authors have read and agreed to the published version of the manuscript.

**Funding:** The financial support from the National Natural Science Foundation of China (NSFC) with Grant Nos. 52078359 and 51778456 are gratefully acknowledged. Support for this study was also provided by the Shanghai Rising-Star Program (20QA1409400).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data presented in this study are available in this article.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Materials* Editorial Office E-mail: materials@mdpi.com www.mdpi.com/journal/materials

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com ISBN 978-3-0365-6606-1