a comparison of conventional and iron- … int. j. struct. & civil engg. res. 2014 mihir mishra...

96

Int. J. Struct. & Civil Engg. Res. 2014 Mihir Mishra and Amin Anish Ravindra, 2014

A COMPARISON OF CONVENTIONAL AND IRON-BASED SHAPE MEMORY ALLOYS AND THEIRPOTENTIAL IN STRUCTURAL APPLICATIONS

Mihir Mishra1* and Amin Anish Ravindra2

1 Civil Engineering Department, Vellore Institute of Technology, Vellore - 632014, Tamil Nadu, India.2 Mechanical Engineering Department, Vellore Institute of Technology, Vellore - 632014, Tamil Nadu, India.

*Corresponding author:Mihir Mishra [email protected]

ISSN 2319 – 6009 www.ijscer.comVol. 3, No. 4, November 2014

© 2014 IJSCER. All Rights Reserved

Int. J. Struct. & Civil Engg. Res. 2014

Research Paper

INTRODUCTIONA Shape Memory Alloy (SMA) is an alloy that“remembers” its original shape, and whendeformed, returns to its pre-deformed shapewhen heated, or upon removal of stress(Otsuka and Wayman, 1998). The prominentcharacteristics of SMAs, based on the type ofalloy, are Superelasticity (SE), Shape MemoryEffect (SME), increased damping capacity ortwo-way shape memory effects (Janke et al.,2005). Superelasticity refers to the ability of

Shape Memory Alloys (SMAs) and their applications in various fields have been known fordecades. The application of SMAs in civil engineering is limited and the research is still in thepioneer phase. This paper presents the applications of SMAs in structural engineering, focusedon the rehabilitation and seismic retrofit of monuments as well as their Shape Memory Effect(SME) in structures. A comparison between conventional Nitinol (NiTi) SMAs and their lowercost iron-based counterparts has also been highlighted, to determine the extent of their practicalapplications based on key properties that have undergone prior research. Also stated is thecontrast between reinforcement grade steel and SMAs, and consequently the advantages ofboth, including the discussion of possible solutions to overcome their limitations. Finally, basedon existing paradigms, potential regions where SMAs can be utilized to conserve cultural heritagestructures have also been mentioned.

Keywords: Shape Memory Alloys, Damping, Structural Rehabilitation, Cultural Heritage,Structural Concrete, Civil Engineering

SMAs to undergo significant inelasticdeformation under stress, and revert to theiroriginal shapes upon unloading (Dureig et al.,1990). The shape memory effect, on the otherhand, occurs due to the reversible phasetransformation between the two crystallinestructures – Martensite and Austenite (Claredaet al., 2014). In the martensitic transformation,a diffusionless solid state transformation isbought about by the relative movement ofatoms in an organized manner, creating a new

97


crystal structure that does not have anycompositional change, but can bring aboutsignificant macroscopic deformations. Thistransformation is generally achieved by eitherquenching or mechanical deformation(Clareda et al., 2014).

Austenite is stronger and stable at highertemperatures, while the weaker Martensitephase, at low temperatures, primarily due tothe difference in their crystal structures, withAustenite having a body-centered cubic crystalstructure whereas Martensi te has anasymmetric parallelogram structure (which has24 variations) (Song et al., 2006). For thisreason, the Martensite phase is weaker, beingeasily deformed. In contrast, Austenite, havingonly a single orientation, has increasedresistance to external stress (Dureig et al.,1990).

When applied stresses are absent, thetemperature at which the martensitic (forward)transformation begins is known as Ms(martensite start) and at Mf (martensite finish),the transformation finishes. If the material is inMartensite phase (T < Mf), then upon heatingthe material, the transformation can bereversed. Austenite phase formation beingsat temperature As (austenite start) andconcludes at Af (austenite finish). Thereactions with the range of temperatures havebeen represented in Figure 1. Due to thermalhysteresis during the transformation, theforward and reverse transformations do notoccur at the same temperature (Kumar, 2008).

In 1962, Buechler and co-researchersdiscovered the shape memory effect in a Ni–Ti alloy (named Nitinol) (Buehler et al., 1963).It was the most commonly used SMA until

1982, when the SME was observed in Fe–Mn–Si alloy and since then new iron-based SMAshave been looked into, which offer greaterpotential (Sato et al., 1982). Two groups of ironbased SMAs exist. The first group containsalloys such as Fe–Pt, Fe–Pd and Fe–Ni–Co,displaying characteristics similar to Ni–Tithermoelastic martensitic transformations witha narrow thermal hysteresis. The latter consistsof alloys like Fe–Ni–C and Fe–Mn–Si, which,despite having larger thermal hysteresis intransformation, also display SME (Maruyama,2011).

Possible applications of SMAs in civilengineering structures were presented byauthors Janke et al. (2005) which includedpassive vibration damping and energydissipation, active vibration control, actuatorapplications and the utilization of the SME fortensioning applications or sensors. Thedamping capacity arises due to Martensitevariations reorientation which exhibits SME,and also stress-induced martensi tictransformation of the austenite phase, whichexhibits superelasticity. The vibrationsuppression of civil structures can be pursuedusing passive control which relies on dampingcapacity of SMA (Song et al., 2006).

The main limitation of SMAs is due to thesize of the civil engineering structures andrelatively high stresses, hence to beeconomically viable, the SMAs utilized neededto be lower in cost. Research regarding thecase of damping in civil structures using ironbased shape memory alloys is still in itspioneer stage. Rehabilitation of structures byusing SMAs holds great promise, both in termsof damage reduction and repair cost. Thispaper explores the damping capacity and

98


other properties of Fe based SMAs and theiradvantages over conventional Ni-Ti alloys, aswell as their potential for restoration ofstructures, especially cultural monuments.

TYPES AND PROPERTIES OFSHAPE MEMORY ALLOYSThe first incidence of SME was observed in1932 by Chang and Read in gold-cadmium(AuCd). It was only in 1962, when the sameeffect was seen in nickel-titanium by Buechlerand his co-researchers, that the properties ofSMAs were explored further (Buehler et al.,1963). As mentioned earlier in the paper, thisSMA, named Nitinol, was the most popular ofits type. It possessed superior thermochemicaland thermoelectrical properties and wasreadily available (Duerig et al., 1963). Severalproperties and applications of Nitinol havebeen well documented. Czaderski et al. usedNiTi wires as flexural reinforcement in aconcrete beam; The stiffness and strength ofthe beam changed by varying the temperaturein the SMA wires with electrical resistiveheating (Czaderski, 2006). NiTi wires havealso been used in short-fiber concrete, wherethe fibers were prestressed upon heating thetest specimen in an oven (Moser et al., 2005).In addition, recovery stresses of NiTi wires(with a specific composition and thermaltreatment) were observed to be more than 200MPa at room temperature, by Tran et al.(2011).

Although still holding the primary positionin the industrial market, the commercialapplications of NiTi are offset by the high costsof both raw materials and processing,reducing their viability. Also, their effectivenessfor prestressing is restricted by their relatively

narrow thermal hysteresis, inadequate for civilengineering applications where large andstable recovery stresses are required(Lagoudas, 2008).

Alternatives in the form of iron-based SMAsare deemed more feasible for civil engineeringapplications due to their properties(Maruyama and Kubo, 2008). The two groupsof i ron-based SMAs have di fferentapplications. The first group - Fe–Pt, Fe–Pdand Fe–Ni–Co al loys, exhibi t thecharacteristics of thermoelastic martensitictransformations similar to Ni–Ti and shows anarrow thermal hysteresis. Despite extensivestudies however, pseudoelasticity at roomtemperature has not been reported with Fe–Pt or Fe–Pd alloys. In 2010, Tanaka et al.(2010) presented an Fe–29Ni–18Co–5Al–8Ta–0.01B (mass %) SMA with a recoverystrain of over 13% at room temperature andhigh tensile strength of 1200 MPa, placing ithigh in the list of new materials (Ma, 2010; Maet al., 2013). It could prove to be useful forapplications related to pseudoelasticity anddamping capacity. Similarly, good superelasticproperties have been observed at roomtemperature in Fe–36Mn–8Al–8.6Ni (mass %)alloy (Janke et al., 2005), showing a recoverystrain of over 5% and a fracture tensile straingreater than 8%. However, these alloys stillneed to be developed further, in order toproduce large quantities for real-scaleelements in the construction industry. The costof the material would probably be high forconstruction standards since they need to becast in special conditions due to theircomposition (Cladera et al., 2014).

The second group, with Fe–Ni–C and Fe–Mn–Si, shows a larger thermal hysteresis in

99


transformation but still exhibits properties ofSME. The primary focus, for about twodecades, has been on Fe–Mn–Si SMAs, dueto their low cost, good workability, goodmachinability and good weldability (Kajiwara,1999), although the real applications are stilllimited, barring some remarkable exceptions,i.e. large size joining pipes for tunnelconstruction and crane rail joint bars(Maruyama et al., 2011).

Seen from Figure 2, Fe–Mn–Si displaysSME due to the stress induced martensitetransformation from a parent -austenite (fcc–face-centered cubic) phase to an -martensitephase (hcp – hexagonal closed-packed) atlow and intermediate temperature and thereverse transformation ( - to -phase) at hightemperature. Fe–Mn alloys have been knownto undergo this transformation, but such anSME had not been obtained, issue being forhigh amounts of Mn, -austenite stabilized, thestress induced martensitic transformation wasdifficult to induce. On the other hand, for loweramounts of Mn, when the alloy was subjectedto stress, not only -martensite but also ’-martensite (bct – body-centered tetragonal)was generated, with this phase beingirreversible (Figure 2c). The occurrence of ’-martensite creates marked dislocations,

preventing the SME from developing (Otsuka,1992). The addition of Si allowed the SME tooccur, and this was recorded by Sato and hisco-workers (Sato et al., 1982). Among otherelements, Chromium (Cr) was observed to beeffective to some extent, suggesting that it wasa suitable fourth element for the Fe–Mn–Sialloy to improve its corrosion resistance(Otsuka, 1992). The SME is reduced if the Cramount exceeds 7%, as the brittle phaseintrudes in the alloy. Ni was effective inrestraining the formation of phase and thus,had to be added for increasing amounts of Cr(Otsuka, 1992).

Although in Fe–Mn–Si alloys, superelasticitycannot be observed, new research shows thatsome iron-based alloys such as Fe–29 Ni–18Co–5Al–8Ta–0.01B (mass %) or Fe–36Mn–8Al–8.6Ni (mass %) show super-elasticity at room temperature, but these alloysstill require further development to be producedat the scale needed for civil engineeringapplications. As a result, they would not be costeffective (Cladera et al., 2014). However, analternative production process for the Fe–Mn–Si SMAs by mechanical alloying andsubsequent sintering has been documented(Saito et al., 2013). This technique, based on

Figure 1: Schematic definition of the forward and reversemartensitic transformation temperatures [4]

100


powder metallurgy, involved solid state reactionamong powders by high energy collisions. Thismay offer several advantages formanufacturing industrial products, since itallows the production of alloys in near netshape, thus minimizing the additionalmachining required for the final product. Itproduced comparable results between themechanical properties and the shape memoryeffect of alloys cast using this technique, toconventional casting (Saito et al., 2013).

Recovery stress is paramount for the useof iron-based SMAs as prestressing tendonsin civil engineering structures. Shape recovery

properties for Fe–28Mn–6Si–5Cr SMAincrease upon thermomechanical training(Maruyama et al., 2011), but require additionalprocessing, increasing the production costand only suiting components that have simplegeometries (Leinbach et al., 2012; Li et al.,2013). The modulus of elasticity of the Fe–28Mn–6Si–5Cr is 170 GPa (Maruyama et al.,2011), higher than that reported for Ni–Ti alloys,which varies from 30 to 98 GPa in austenite to21–52 GPa in martensite (Wen et al., 2011).The recovery stresses and transformationtemperatures for various Fe-Mn-Si alloys aresummarized in Table 1.

Table 1: Recovery Stresses and transformation temperaturesfor different Fe-Mn-Si alloys. Adapted from [4]

S.No.

1.

2.

3.

4.

5.

6.

7.

8.

Composition in Mass %

Fe-28Mn-6Si-5Cr

Fe-28Mn-6Si-5Cr-0.5(Nb,C)

Fe-19Mn-5Si-8Cr-5Ni

Fe-16Mn-5Si-10Cr-4Ni-1(V,N)

Fe-15Mn-4Si-8Cr-4Ni-0.012C



Fe-17Mn-5Si-10Cr-4Ni-1(V,C)

Comments

Without training-5-8% pre strainWith training -5-8% pre strain

Non pre rolled6% pre rolled14% pre rolled70% pre rolled

Equal channel angularpressing and 4.5% pre strain atRT

4% pre-strain at 45LC4% pre-strain at RT

Cold drawn alloy-6% pre strainat RT-Annealing temperature650C



4% pre-strain at RT

RecoveryStress MPa

130180

145255295200

460

500440

520

535

565

580

Temperature(0C)

350350

400400400400

500

225160

–

–

–

130

Reference

Maruyamaet al., 2011

Baruj et al.,2002

Zhang et al.,2007

Li et al.,2013

Wang et al.,2011

Wang et al.,2011

Wang et al.,2011

Leinenbachet al., 2012

101


COMPARISON BETWEENNITINOL AND IRON BASEDSMASTo clarify, advantages of Nitinol include thecapability of recentering and recovering morethan 80% of the induced strains, high modulesof elasticity, similar behavior in tension andcompression, high quali ty for makingmachined components such as bolts and highresistance to corrosion as well as fatigue.However, certain points must also beconsidered. Nitinol has a small area understress-strain curve and little cyclic energydissipation capacity.

It exhibits semibrittle and alarmless fracturewith little ultimate fracture strain (less than15%). Other problems include difficulty inwelding and machining, as well as its highprice (Jalaeefer, 2013).

Based on the above-mentioned properties,Fe–Mn–Si alloys have shown a widertemperature transformation hysteresis and ahigher elastic stiffness than Ni–Ti alloys. Also,their workability, corrosion resistance andweldability is better, in comparison. Over theyears, new Fe–Mn–Si alloys with fineprecipitates have been developed, allowinghigher recovery stresses without,thermomechanical training (Cladera et al.,2014).

Most of the research conducted for Ni–Tialloys based on its properties and applicationscan be applied to iron-based shape memoryalloys as well such as concrete that isprestressed by Fe–Mn–Si alloy short fibers,an idea already originally given to Ni–Ti alloys(Moser et al., 2005; Shajil et al., 2013), or theconfinement of concrete columns, already

applied to Ni–Ti or Ni–Ti–Nb alloys (Shin,2010; Park et al., 2011). However, severalsuch applications would demand moreinformation about bond strength reduction ofthe SMA due to temperature rise. Thus,provided that in-depth future research on ironbased SMAs is carried out, they seem toprovide greater application opportunities.

COMPARISON BETWEENSMAS AND STEELThe main advantages of structural steel asreinforcement are the capability to bear up to35% strain before rupture, a large area understress-strain curve or high cyclic energydissipation capacity, gradual and alarmingfracture (showing obvious deformations) andsoftening phase in the stress-strain curve. It isalso known to exhibit similar behavior in tensionand compression. Other Major factors in itsfavor are the simplicity of machining andwelding, and especially the cost, making itviable for large scale construction.

On the offside, it has several disadvantagesas well. Steel undergoes large permanentplastic strain and has a weak recenteringcapability. Its quality for making machinedcomponents, such as bolts, due to porositycoupled with its low resistance to fatigue,mandate the need for alternatives (Jalaeefer,2013).

Several benefits of SMAs such as Nitinolhave already been discussed above in thispaper. Apart from those, iron-based SMAtendons hold great promise for eitherreinforcing or repairing existing structures.Some of their advantages are no frictionlosses, no requirement for anchor heads andducts and no space necessity for hydraulic

102


Figure 2: Crystal lattices: (a) -Austenite (fcc – face-centered cubic);(b) -Martensite (hcp – hexagonal close-packed structure);

(c) -Martensite (bct – body-centered tetragonal)

Figure 3: Schematic of SMA damper for a stay cable bridge

Figure 4: Schematic of intelligent reinforced concrete specimen

103


devices to apply force. New iron-based SMAshave lower costs due to the use of cheap iron(Fe) and also by melting and producing themunder normal atmospheric conditions (Claderaet al., 2014)

However, being highly sensi tive tocompositional changes, any such variations inthe composition of an SMA can cause majorundesirable changes in its mechanicalproperties. For large scale use SMAs,enforcing quality control measures would beof prime importance for proper and consistentcomposition which yield appropriateproperties (Jalaeefer, 2013). Also, Nitinol,though the most commonly used SMA, is highlyexpensive, and its iron based counterparts,despite being economic, sti l l requiresignificant research to be viable as effectivereplacements.

After considering all options, shapememory alloys can only be considered for useparallel to structural grade steel, by finding amiddle ground where optimum benefits can beobtained, while negating maximum possible

deficiencies.

DAMPING PROPERTIES OFSHAPE MEMORY ALLOYSThe damping capacity of SMAs come frommartensite variations reorientation, whichexhibit SME and stress-induced martensitictransformation of the austenite phase, whichexhibit superelasticity. Three methods tosupress vibration in civil structures due toexternal dynamic loading exist, namely activecontrol, semi-active control, and passivecontrol. Passive structure control is dependenton the SMA’s ability to dissipate vibrationenergy of structures subject to dynamicloading. Dolce and Cardone (2001) studiedsuperelastic Nitinol wires under tensionloading. The dependence of the dampingcapacity on temperature, loading frequencyand the number of loading cycles was noted.Their results indicated that the mechanicalbehavior of the wires was stable within a usefulrange, considering seismic applications, andsuggested that in order to increase efficiencyof energy dissipation, the austenite wires

Figure 5: Specimens tested by Sawaguchi et al. (a) bending strength of mortarspecimens with SMA, steel and without reinforcement (b) [42]

104


should be pre-tensioned (Dolce, 2001; Ip,2000; Piedboeuf, 1998).

Liu et al found that the damping capacity ofa martensite Nitinol bar under tension–compression cycles increased upon increasingthe strain amplitude and decreased withloading cycles, before reaching a stableminimum value (Liu et al., 1999). Themartensite and austenite damping of Nitinolbars, subjected to torsion was observed Dolceand Cardone. They found that the dampingcapacity of the martensite Nitinol bar washigher than that of the austenite Nitinol bar.However with the accumulation of residualstrain, the martensite Nitinol bar could notremain at its highest value. Also observed wasthe martensite bar’s mechanical behaviorbeing independent of loading frequency andthat of the austenite bar slightly depending onit (Dolce, 2001). It can be inferred that bothmartensite and austenite Nitinol bars arecapable of working in a wide frequency rangeand are suitable for seismic protection.

APPLICATION OF SMAS INCIVIL STRUCTURESDamping in Bridge Systems

Li et al. used a stay-cable bridge totheoretically study the vibration mitigation of acombined cable–SMA damper system, shownin Figure 3. The SMA damped cable wasmade to vibrate at its first mode or at its firstfew modes respectively, to simulate itsdynamic responses. In both cases, the authorsused the results to propose the ability ofsuperelastic SMA dampers to suppress thecable’s vibration (Li et al., 2004).

The utility of large martensite Nitinol bars inseismic protection devices for bridges was

studied by Casciati et al. (1998). To analyzethe static and dynamic response of the devicesto strong earthquakes a Finite Element Model(FEM) was used. The results were inagreement with the FEM analysis and verifiedthe application of martensite Nitinol bars inenergy dissipation. Hence, superelastic andmartensite SMAs can be used as dampingelements for bridges.

Application of SMAs for StructuralSelf-Rehabilitation

Structural self-rehabilitation using SMAscomes under active structural control. Song andMo proposed the concept of IntelligentReinforced Concrete (IRC), using the actuationproperty of SMA wires (Song, 2003). Itsintelligence is based on the ability to senseand self-rehabilitate. Figure 4 shows IRC withstranded Martensite SMA wires for post-tensioning. The electric resistance change ofthe shape memory alloy wires can bemonitored to exhibit the strain distributioninside the concrete. Occurrence of macro-sized cracks from earthquakes can bereduced by electrically heating the SMA wirestrands, causing them to contract and hence,minimize the cracks.

Small concrete blocks of dimension 13.5in. × 6 in. × 2 in., post-tensioned by Nitinol SMAwires were fabricated and tested at Universityof Houston, providing convincing preliminaryresults (Mo et al., 2004). Seven 0.015 in.-diameter SMA wires with a transformationtemperature of 90º C were present in eachrope. A threepoint bending test was conducted,during which upon loading, a crack measuringup to .32 in. (8 mm) opened up. Removal ofthe load and heating of the SMA wires reducedthe crack significantly, rendering it barely

105


visible to the naked eye. While loading, theelectrical resistance value of the SMA wirechanged up to 15%. This observation allowedthe monitoring of the crack opening without anyadditional sensors.

.In Michigan, a bridge was strengthenedthrough external post-tensioning, performedwith Fe–Mn–Si–Cr SMA rebars in 2001(Soroushian, 2001). A Fe–28Mn– 6Si–5Cral loy (manufactured by Nippon SteelCorporation) was used, capable of developingrecovery stresses up to 255 MPa after heatingup to 300º C. The iron-based alloy wassubjected to extreme environmentaltemperatures, and effects on the restrainedrecovery stresses were examined. The resultsshowed that the variations in stress wererelatively small, and the shear crack closed toa large extent when post-tensioned with SMAs,with the load capacity of the rehabilitated beamrecovered.

Shape Memory Effect in StructuralApplications

At the industrial level, crane rail fishplates havebeen manufactured to connect finite rail lengthsfor heavy duty cranes (Maruyam et al., 2008),as well as steel pipe joints as a form ofconstruction method for tunneling work(Maruyam et al., 2011). In both cases, theproducts were manufactured by Awaji MateriaCo.

A 1 mm diameter prestressed wire of Fe–27Mn–6Si–5Cr–0.05C alloy was used byWatanabe et al. as the reinforcement of an 80mm long plaster prism specimen (Watanabeet al., 2002). The wires were subjected topretensile strain at room temperature (1%, 2%and 3%), and embedded into a plaster matrix.

In order to generate a compressive stress inthe matrix, the specimens were heated up to250º C. Three-point bending tests, formechanical property characterization as wellas pull-out tests were carried out, revealingthat the fracture toughness of plaster could beimproved by the wires. Also, as the level of pre-strain in the SMA wires was increased, thebending strength of the composite specimenincreased as well.

Sawaguchi et al. cast small mortar prisms,reinforced with square Fe–28Mn–6Si–5Cr–1(NbC) bars (Sawaguchi et al., 2006). Theywere embedded with SMA square bars, steelbars with the same dimensions (shown inFigure 5), or without any reinforcement, andcured for two days. Upon extraction from themould, high-temperature curing was performedin an autoclave. The first stage of the autoclavecuring at 87º C (360 K) for 24 h was carriedout to increase the compressive strength ofthe mortar matrix. The second stage wasrequired to pre-stress in the SMA, for whichthe conditions 177º C for 6 h and 247º C for30 min, were applied. The mechanicalproperties of the mortar increased with the177º C curing, but reduced for the 247º Ctreatment, with the exception being the prismsstrengthened with steel subjected to the hightemperature curing ( from the results in Figure5). It was concluded that the iron-based SMAswere usable for producing pre-stress in themortar because the bending strengthincreased significantly in the specimens withthe SMA compared to those either reinforcedwith steel or not reinforced at all. The authorsfurther suggested that increased strength waspossible if the reverse transformationtemperature of the SMAs was reduced, and

106


as a result significant thermal damage to themortar matrix could be avoided.

For pre-stressing concrete structures, Leeet al. studied the recovery stress behavior ofan Fe–17Mn–5Si–10Cr–4Ni–1(V, C) alloy(Lee et al., 2013) The pre-stressing effect dueto SME was simulated by tests that involvingpre-straining the material succeeded byheating and cooling at a constant strain. Thetested SMA specimens were part of a 15 kgalloy ingot, induction melted under normalatmospheric conditions. The iron-based SMAused showed a wide hysteresis from -60 º Cto 103º C obtained by a differential scanningcalorimetry analysis. The highest recoverystress occurred at 400 MPa, for a pre-strainof 4% and a heating temperature of 160º C.Also, under additional loading following pre-stressing, the behavior was observed for cyclicmechanical loading and cyclic thermal loading.It was seen that the pre-stressed SMAundergoes inelastic deformation during the firstcycle, reducing the recovery stress. After thefirst cycle though, the SMA was elastic, stressremained stable for all subsequent cycles. Thestudy experimentally demonstrated that sucha loss could be recovered completely uponreheating the SMA element.

Shape Memory Alloys as SeismicRetrofitOwing to their potential, SMAs have been usedto retrofit existing or damaged structures, andprotect them from further damage brought onby disasters such as earthquakes. Theintensity of damage to structures is directlyrelated to the magnitude of the earthquake onthe Richter scale. Traditionally, in order toprotect structures such as cultural heritagemonuments from seismic behavior, localized

reinforcements such as steel bars or cableshave been used, to provide an increasedstability and ductility, but several cases existwhere despite such provisions, structuralcollapse has still occurred. In fact, during a pastearthquake and its resultant damage to achurch, the tympana collapsed despitereinforcement through steel ties, although theremaining facades at the lower portion of thechurch remained standing (Doglioni, 1994).

One significant example is the Basilica ofSt Francis in Assisi, severely damaged duringthe 1999 earthquake in central Italy (Croci etal., 2000; Mazzaloni, 2002). In its restoration,obtaining an adequate safety level whilesimultaneously maintaining its originalstructure was the major challenge. Aconnection between the tympanum and theroof was created using superelastic SMA rodsso as to reduce the effect of seismic forces onit. The SMA devices exhibited differentstructural properties based on horizontal forcelevels. Under low horizontal forces, they werestiff, allowing no significant displacements,under high horizontal actions the stiffnessreduced to make an allowance for controlleddisplacements of the masonry walls to occur,while extremely intense horizontal loadsincreased their stiffness to avoid structuralcollapse.

Another example of utilizing SMAs asseismic retrofit is in the bell tower of the SanGiorgio church in Trignano, Italy, severelydamaged during the 1996 earthquake(Mazzaloni, 2002; Indirli, 2000). For itsrehabilitation, four vertical pre-stressing steeltie bars were inserted in the internal cornersof the structure aiming to increase its flexuralstrength with the main aim being post-

107


tensioning the SMA devices so that a state ofconstant compression was induced. Six tight-screwing segments placed in series with fourSMA devices composed of severalsuperelastic wires formed the tie bars.

Shaking table tests on brick masonry wallmock-ups were used to simulate a part of amonument by Indirli et al. (2000), in order toevaluate the how effective SMAs as ties were.Test results indicated that SMAs for suchpurposes could be highly effective inpreventing out-of-plane collapse of peripheralwalls like church facades often having weakconnections at floor level. The SMA ties, incomparison to traditional ties, increasedresistance against out-of-plane seismicvibrations of such masonry walls by 50% (asmaximum bearable peak ground accelerationwithout damage), due to reduced topacceleration by a minimum of 50%. Also, thetympanum structures were protected by theseSMA ties, from further seismic-induceddamage.

In India, several earthquakes have wreakedhavoc in the past especially in the high riskzones of the Himalayan belt, HimachalPradesh, North-East states, Maharashtra andDelhi where two tectonic plates are in constantcollision. The 1905 earthquake damaged mostof the monuments in Himachal Pradesh, theworst hit being Baijnath and Kangra Valley. TheKangra Fort was partially damaged while theBrajeshwari Devi temple and the belfry of SaintJohn’s Church were totally destroyed. The restof the church building however escapedserious damage. The most disastrousearthquake in India’s recent history took placeon 26 January, 2001 in Bhuj, Gujarat,

measuring 7.7 on the Richter scale. It destroyedseveral monuments in the region, dating backto 9th century AD. Among them are the AinaMahal Palace (~ 200 years old), Lakhpatji’sChhatardi (AD 1752–1761) in the city of Bhujand some of the old temples in its vicinity. TheSun Temple at Kotai (11th century AD) and thePunvareshwar Temple at Manjal (9th centuryAD) are among the oldest temples that werehit. While the Chhatardi collapsed completely,Aina Mahal Palace is still standing withextensive damage. These structures seem tohave survived the 1819 and 1956earthquakes, but the tremors were apparentlytoo severe during 2001 (Sinha, 2001).

Shape Memory Alloys, however, have notbeen used in India in ancient monumentsbecause they are needed at the point offoundation. Monuments are generally repairedwith the top portions like the domes and roofsincluding walls. The foundation is usually leftuntouched unless the structure is completelyrazed. AS mentioned in this paper, significantexamples such as the Basilica of St Francis inAssisi and the bell tower of the San Giorgiochurch in Trignano, which were severelydamaged during earthquakes in Italy, havebeen restored using SMAs. Given theseresults, the potential for SMAs in this field isvast. Preservation of monuments in disaster-prone regions is possible by analyzing suitableproperties of SMAs, depending on the extentof damage to the structure or the degree ofbracing needed.

RESULTS AND CONCLUSIONSome key properties of iron-based shapememory alloys as well as Nitinol, such asdamping, and their applications in the

108


rehabilitation and seismic retrofit of civilstructures have been reviewed in this paper.With the emergence of several iron-basedSMAs in the past decade, the research of theirproperties has increased manifold. Bycomparison, the potential for iron-based SMAsseems to be vast, and are bound to overtaketheir Nitinol counterparts, since they showsimilar, if not better properties under the sameconditions and are cost effective as well.Parallels between SMAs and reinforcementsteel have also been drawn, which show thatwhile SMAs may not be suitable as the solereinforcement component, i f used inconjunction with steel and a certain balancebe achieved, it can serve to not only lowercosts, but also optimize performance of saidstructures.

A review of the damping properties of SMAson the basis of experiments that have beencarried out, show that there is ample scopefor utility in civil engineering structures, suchas bridge systems. Self-rehabilitation ofstructures is a promising field for the nearfuture, although further studies would beneeded to assess its possibility on a largerand more commercial scale. Similarly, theshape memory effect appears to have positiveresul ts regarding the properties andstrengthening of plaster and mortar systems,reinforced with SMAs, but again, research isneeded to validate its feasibility with respectto actual structures.

CONCLUSIONPast earthquakes have inflicted severe losseson India’s national heritage. Given theexamples on successful implementation of

SMAs to retrofit structures in Italy, which havebeen damaged in the past, several monumentslying in prone areas have been listed out,where such SMAs would have the samepurpose. Not only has it been shown that SMAscan aid the reparation process, but structuresreinforced with them have been capable ofwithstanding further damage as well.

Granted, for SMAs to replace conventionalmaterials is a distant reality. However, the keytopics summarized in this paper could serveas reference points to allow significantanalysis of their behavior and furtherunderstanding of properties relevant to the fieldwhere they can be ideally utilized.

REFERENCES1. Croci G, Bonci A and Viskovic A (2000).

“Use of shape memory alloy devices in the

basil ica of St Francis of Assisi,”

Proceedings of the Final Workshop of

ISTECH Project - Shape Memory Alloy

Devices for Seismic Protection of Cultural

Heritage Structures, pp. 110–13.

2. Baruj A, Kikuchi T, Kajiwara S and Shinya

N (2002), “Effect of pre-deformation of

austenite on shape memory properties in

Fe–Mn–Si-based alloys containing Nb

and C”, Mater Trans; Vol. 43, pp. 585–8.

3. Buehler W J, Gilfrich J V and Wiley R C

(1963), “Effect of low-temperature phase

changes on the mechanical properties of

alloys near composition TiNi”, J Appl

Phys, Vol. 34, pp. 1475–7.

4. Casciati F, Faravelli L and Petrini L

(1998), “Energy dissipation in shape

memory alloy devices”, Computed-Aided

109


Civil and Infrastructure Engineering; Vol.

13, pp. 433–42.

5. Cladera A, Weber B, Leinenbach C,Czaderski C, Shahverdi M and MotavalliM (2014), “Iron based shape memoryalloys for civil engineering structures: Anoverview”, Construction and buildingmaterials, Vol. 63, pp. 281-293.

6. Czaderski C, Hahnebach B and MotavalliM (2006), “RC beam with variable stiffnessand strength”, Constr Build Mater; Vol. 20,No. 9, pp. 824–33.

7. Doglioni F, Moretti A and Petrini V (Eds),(1994), “Le Chiese e il Terremoto (TheChurches and the Earthquake)”, EdizioniLint, Trieste (in Italian).

8. Dolce M and Cardone D (2001),“Mechanical behaviour of shape memoryalloys for seismic application 1. Martensiteand Austenite NiTi bars subjected totorsion”, International Journal ofMechanical Sciences; Vol. 43, pp. 2631–56.

9. Duerig T W et al. (1990), Engineeringaspects of shape memory alloys ,London: Butterworth-Heinemann.

10. Indirli M (2000), “The demo-interventionof the ISTECH Project: the bell tower of SGiorgio in Trignano (Italy),” Proceedingsof the Final Workshop of ISTECH Project- Shape Memory Alloy Devices forSeismic Protection of Cultural HeritageStructures, pp. 134–146.

11. Indirli M, Carpani B, Martelli A, CastellanoM G, Infanti S, Croci G, Biritognolo M,Bonci A, Viskovic A and Viani S (2000),“Experimental tests on masonry structures

provided with shape memory alloyantiseismic devices,” 12th WorldConference in Earthquake Engineering,(pp. 1–8).

12. Ip K H (2000), “Energy dissipation inshape memory alloy wire under cyclicbending”, Smart Materials andStructures; Vol. 9, pp. 653–9.

13. JalaeefarA and Asgarian B (2013),“Experimental Investigation of MechanicalProperties of Nitinol, Structural Steel, andTheir Hybrid Component.” J. Mater. Civ.Eng., Vol. 25, No. 10, pp. 1498–1505.

14. Janke L, Czaderski C and Motavalli Mand Ruth J (2005), “Applications of shapememory alloys in civil engineeringstructures – overview, limits and newideas”, Mater Struct; Vol. 38, No. 279, pp.578–92.

15. Janke L, Czaderski C, Motavalli M andRuth J. (2005), “Applications of shapememory alloys in civil engineeringstructures – overview, limits and newideas”, Mater Struct; Vol. 38, pp. 578–92.

16. Kajiwara S (1999), “Characteristicfeatures of shape memory effect andrelated transformation behaviour in Fe-based alloys”, Mater Sci Eng A; pp. 273–275:67–88.

17. Krstulovic-Opara N and Thiedeman P D(2000), “Active confinement of concretemembers with sel f-stressingcomposites”, ACI Struct J; Vol. 97, pp.297–308.

18. Kumar P K and Lagoudas D C (2008),In: Lagoudas D C, Shape memory alloys.

110


modeling and engineering applications,Springer, pp. 1–51 [Chapter 1].

19. Lagoudas D C (2008), Shape memoryalloys, Springer, pp. 435.

20. Lee W J, Weber B, Feltrin G, CzaderskiC, Motavalli M and Leinenbach C. (2013),“Stress recovery behaviour of an Fe–Mn–Si–Cr–Ni–VC shape memory alloy usedfor prestressing”, Smart Mater Struct; Vol.22, pp. 1–9.

21. Leinenbach C, Kramer H, Bernhard C,Eifler D (2012), “Thermo-mechanicalproperties of an Fe–Mn–Si–Cr–Ni–VCshape memory alloy with lowtransformation temperature”, Adv EngMater; Vol. 14, pp. 62–7.

22. Li H, Liu M and Ou J P (2004), “Vibrationmitigation of a stay cable with one shapememory alloy damper”, Structural Controland Health Monitoring; Vol. 11, pp. 1–36.

23. Li K, Dong Z, Liu Y, and Zhang L (2013),“A newly developed Fe-based shapememory alloy suitable for smart civilengineering”, Smart Mater Struct: p. 22.

24. Liu Y, Xie Z and Humbeeck J V (1999),“Cyclic deformation of NiTi shape memoryal loys”, Materials Science andEngineering, pp. A273–275:673–8.

25. Ma J and Karaman I (2010), “Expandingthe repertoire of shape memory alloys”,Science; Vol. 327, pp. 1468–9.

26. Ma J, Hornbuckle B C, Karaman I,Thompson G B, Luo Z P and ChumlyakovY I (2013), “The effect of nanoprecipitateson the superelastic properties ofFeNiCoAlTa shape memory alloy single

crystals”, Acta Mater; Vol. 61, pp. 3445–55.

27. Maruyama T and Kubo H (2011) In:Yamauchi K, Ohkata I, Tsuchiya K,Miyazaki S, editors, Shape memory andsuperelastic alloys: technologies andapplications , Cambridge (UK):

Woodhead Publishing Limited; (pp. 141–59) [Chapter 12].

28. Maruyama T and Kubo H (2011), “Ferrous(Fe-based) shape memory alloys (SMAs):properties, processing and applications”,Woodhead Publishing Limited., pp. 141–59.

29. Maruyama T, Kurita T, Kozaki S, AndouK, Farjami S and Kubo H. (2008),“Innovation in producing crane rail

fishplate using Fe–Mn–Si–Cr basedshape memory alloy”, Mater Sci Technol;Vol. 24: pp. 908–12.

30. Mazzolani F M and Mandara A (2002),“Modern trends in the use of specialmetals for the improvement of historicaland monumental structures,” EngineeringStructures, Vol. 24, pp. 843–856.

31. Mo Y L, Song G and Otero K (2004),“Development and testing of a proof-of-

concept smart concrete structure”, In:Proceeding of smart structurestechnologies and earthquakeengineering.

32. Moser K, Bergamini A, Christen R andCzaderski C (2005), “Feasibility ofconcrete prestressed by shape memoryalloy short fibers”, Mater Struct; Vol. 38,

No. 279, pp. 593–600.

111


33. Moser K, Bergamini A, Christen R andCzaderski C (2005), “Feasibility ofconcrete prestressed by shape memoryalloy short fibers”, Mater Struct/MaterConstr; Vol. 38, pp. 593–600.

34. Otsuka H (1992), “Fe–Mn–Si basedshape memory alloys”, Mat Res SocSymp Proc; Vol. 246, pp. 309–20.

35. Otsuka K and Wayman C M (1998),Shape memory materials , UnitedKingdom: Cambridge University Press.

36. Park J, Choi E, Park K and Kim H T.(2011), “Comparing the cyclic behaviourof concrete cylinders confined by shapememory alloy wire or steel jackets”, SmartMater Struct; Vol. 20, pp. 1–11.

37. Piedboeuf M C and Gauvin R (1998),“Damping behaviour of shape memoryalloys: strain amplitude, frequency andtemperature effects”, Journal of Soundand Vibration; Vol. 214, No. 5, pp. 885–901.

38. Saito T, Kapusta C and Takasaki A(2013), “Synthesis and characterization ofFe–Mn–Si shape memory alloy bymechanical alloying and subsequentsintering”, Mater Sci Eng A; Vol. 592, pp.88–94.

39. Sato A, Chishima E, Soma K and Mori T(1982), “Shape memory effect in y transformation in Fe–30Mn–1Si alloysingle crystals”, Acta Metall, Vol. 30, pp.1177–83.

40. Sawaguchi T, Kikuchi T, Ogawa K,Kajiwara S, Ikeo Y, Kojima M, et al. (2006),“Development of prestressed concreteusing Fe–Mn–Si-based shape memory

alloys containing NbC”, Mater Trans; Vol.47, pp. 580–3.

41. Shajil N, Srinivasan S M and SanthanamM (2013), “Self-centering of shapememory alloy fiber reinforced cementmortar members subjected to strongcyclic loading”, Mater Struct; Vol. 46, pp.651–61.

42. Shin M and Andrawes B (2010),“Experimental investigation of activelyconfined concrete using shape memoryalloys”, Eng Struct; Vol. 32, pp. 656–64.

43. Sinha A K (2001), The GujaratEarthquake, Asian Disaster ReductionCenter.

44. Song G and Mo Y L (2003), “Increasingconcrete structural survivability using smartmaterials”, A proposal submitted to Grantsto Enhance and Advance Research(GEAR), University of Houston; January.

45. Song G, Ma N and Li N M (2006),“Applications of shape memory alloys incivil structures”, Engineering Structures,Vol. 28, pp. 1266-1274.

46. Soroushian P, Ostowari K, Nossoni A andChowdhury H (2001), “Repair andstrengthening of concrete structuresthrough appl ication of correctiveposttensioning forces with shape memoryalloys”, Transp Res Rec: pp. 20–6.

47. Tanaka Y, Himuro Y, Kainuma R, Sutou Y,Omori T and Ishida K (2010), “Ferrouspolycrystall ine shape-memory alloyshowing huge superelasticity”, Science;Vol. 327, pp. 1488–90.

48. Tran H, Balandraud X and Destrebecq JF (2011), “Recovery stresses in SMA

112


wires for civil engineering applications:experimental analysis and thermo-

mechanical modelling”, MaterialwissWerkstofftech; Vol. 42, No. 5, pp. 435–43.

49. Wang C P, Wen Y H, Peng H B, Xu D Qand Li N (2011), “Factors affecting

recovery stress in Fe–Mn–Si–Cr–Ni–Cshape memory alloys”, Mater Sci Eng A;

Vol. 528, pp. 1125–30.

50. Watanabe Y, Miyazaki E and Okada H

(2002). Enhanced mechanical propertiesof Fe–Mn–Si–Cr shape memory fiber/

plaster smart composite”, Mater Trans;Vol. 43, pp. 974–83.

51. Wen Y H, Xie W L, Li N and Li D (2007),“Remarkable difference between effects ofcarbon contents on recovery strain andrecovery stress in Fe–Mn–Si–Cr–Ni–C alloys”,Mater Sci Eng A; Vol. 457, pp. 334-7.

52. Zhang W, Wen Y H, Li N and Huang S K(2007), “Remarkable improvement ofrecovery stress of Fe–Mn–Si shapememory alloy fabricated by equal channelangular pressing”, Mater Sci Eng A; pp.454–455:19–23.

a comparison of conventional and iron- … int. j. struct. & civil engg. res. 2014 mihir mishra...

Documents