flow behaviour of thermal cycled titanium (ti-6al-4v) alloy · components. titanium (ti-6al-4v)...

20

Int. J. Mech. Eng. & Rob. Res. 2013 S Manikandan and S Ramanathan, 2013

FLOW BEHAVIOUR OF THERMAL CYCLEDTITANIUM (Ti-6Al-4V) ALLOY

S Manikandan1* and S Ramanathan2

*Corresponding Author: S Manikandan,[email protected]

Titanium is recognized for its strategic importance as a unique lightweight, high strength alloyedstructurally efficient metal for critical, high-performance aircraft, such as jet engine and airframecomponents. Titanium (Ti-6Al-4V) alloy is an alpha beta alloy which is solution treated at atemperature of 950 °C to attain beta phase. This beta phase is maintained by quenching andsubsequent aging to increase strength. Thermal cycling process was carried out for Ti-6Al-4Vspecimens using forced air cooling. The compression test was conducted for varioustemperatures at three different strain rates. Flow behaviour of different heat treated and thermalcycled Titanium (Ti-6Al-4V) alloy were investigated.

Keywords: Thermal cycling, Heat treatment, Solutionizing, Aging, Compressive strength

INTRODUCTIONTitanium is called as the “space age metal”and is recognized for its high strength-to-weight ratio. Today, titanium alloys arecommon, readily available engineered metalsthat compete directly with stainless andSpecialty steels, copper alloys, nickel basedalloys and composites. Titanium (Ti-6Al-4V)alloy is an alpha +beta alloy, which generallycontains alpha and beta stabilizers and is heattreatable to various degrees. Alloying elementsin titanium are classified as stabilizers and stabilizers. Alpha stabilizers such as oxygen

ISSN 2278 – 0149 www.ijmerr.comVol. 2, No. 2, April 2013

© 2013 IJMERR. All Rights Reserved

Int. J. Mech. Eng. & Rob. Res. 2013

1 Department of Mechanical Engineering, Annamalai University, Annamalai Nagar 608002, Tamil Nadu, India.2 Department of Manufacturing Engineering, Annamalai University, Annamalai Nagar 608002, Tamil Nadu, India.

and aluminium rise to transformationtemperature. Beta stabilizers such asmanganese, chromium, iron, molybdenum,vanadium, lower to transformationtemperature and depending on the amountadded may result in the retention of some âphase at room temperature. Alpha stabilizers(such as Al. only) are not heat treatable. Betaalloys are metastable and contain betastabilizers (such as Vanadium) (Wood andAvor, 1972) are used to completely retain betaphase upon quenching. Ti-6Al-4V can besolution treated and aged to achieve significant

Research Paper

21


increase in strength. Titanium alloys areneeded to be heat treated to reduce residualstress developed during fabrication and tooptimize special properties such as fracturetoughness, fatigue strength and hightemperature creep strength. The response oftitanium and titanium alloys to heat treatmentdepends on the compositions of the metal andthe effects of alloying elements on - Crystaltransformation of titanium (Froes andHajliberger, 1985). In addition, not all heattreating cycles are applicable to all titaniumalloys because the various alloys are designedfor different purpose. Ti-6Al-4v is designed forhigh strength at low to moderate temperatures.This alloy can be fully heat treated and can beused up to approximately 400 °C. Over 70%of all Titanium alloy grades melted on a subgrade of Ti-6Al-4V which is used in aerospace,air frame and engine components. Ti-6Al-4Valloys are heat treated for two differentmethods (Material Properties Handbook,1994) in order to get optimum combination ofductility, machinability and structural stabilityby annealing and to increase the strength bysolution treating. The heat treated specimenswere subjected to thermal cycling up to 1500cycles. Variations of True stress at differentstrain rates and temperatures for differentthermal cycled specimens (500, 1000 and1500 cycles) were analysed.

EXPERIMENTAL STUDYThe material used in this study is Ti-6Al-4valloy. It has the following chemical composition(wt %). [Al-6.36, V-4.12, O-0.1562, C-0.015,Fe-0.04, N-0.0037, H-0.0045, Balance-Titanium]. Ti-6Al-4V specimens weremachined (wire cut) to a dimension of 10 mmin length and 10 mm in diameter. Ti-6Al-4V

specimens were subjected to two subsequentheat treatments. In the first heat treatment, thespecimens are heated to a temperature of750 °C for 4 hours and it is cooled in thefurnace to reach the atmospherictemperature. In the second heat treatment thespecimens are heated to a temperature of975 °C for 1 hour and it is quenched in asolution containing water with 5% of causticsoda. After quenching (delay of 6 second),aging was done at 450 °C for 4 hours andthen it is furnace cooled to reach atmospherictemperature.

The heat treated specimens are thermalcycled in a specially designed thermal cyclingapparatus as shown in Figure 1. The basicthermal cycling apparatus comprises a mufflefurnace, forced air cooling unit, electronicexperimental controls unit and timer unit. Thetimer unit controls the heating and cooling timeperiod of the cyclic process and regulates theflow of compressed air to the pneumatic cylinderthrough the electrically operated pneumaticvalve. Further the experimental set-up as shownin Figure 2 consist of pneumatic componentslike double acting reciprocating air compressor,air control main valve, air filter, air regulator, airlubricator, electrically operated pneumatic valveand pneumatic cylinder. The pneumatic cylinderpiston carries the cage, in which the test sampleis placed. The cage is made up of stainlesssteel with three compartments to carry similartest samples in it. The timer unit counts thenumber of cycles and stops the process aftercompleting the required number of cycles.Manual handling and monitoring of the sampleduring the thermal cycling process is minimizeddue to this automation. Pneumatic pistonactuator was built to cycle the specimen in and

22


out of the furnace. Where the requiredtemperature was maintained in the furnace. Thethermal cycle imposed had dwell time of2 minutes in and out of the furnace. Sampleswere cooled by forced air cooling at a pressureof 2 bars. Heat treated specimens weresubjected to thermal cycling in the range of 500,1000 and 1500 cycles.

The compression test was conducted forvarious temperatures in the range of 200, 300and 400 °C at three different strain rates in therange of 0.01, 0.1 and 1.00. Flow behaviourof different heat treated and thermal cycledTitanium (Ti-6Al-4V) alloy were investigated.

RESULTS AND DISCUSSIONTitanium (Ti-6Al-4V) alloys are heat treated(Solutionized and aged). The heat treatedspecimens were subjected to thermal cycling.The number of thermal cycle range is selectedfrom 500 cycles to 1500 cycles in steps of 500cycles. The compression test was conductedfor various temperatures at three different strainrates. Flow behaviour of different heat treatedand thermal cycled Titanium (Ti-6Al-4V) alloywere investigated.

Heat TreatmentsTi-6Al-4V specimens were subjected to twosubsequent heat treatments. Primarily, thespecimens are heated to a temperature of750 °C for 4 hours and it is cooled in thefurnace to reach the atmospherictemperature. Then the specimens are heatedto a temperature of 975 °C for 1 hour and it isquenched in a solution containing water with5% of caustic soda. After quenching (delayof 6 second), aging was done at 450 °C for 4hours and then it is furnace cooled to reachatmospheric temperature.

Heating + alloy to solution treatingtemperature, produces a higher ratio of phase. This partitioning of phases ismaintained by quenching and subsequentaging, decomposition of the unstable phaseoccurs, providing high strength. To obtain highstrength with adequate ductility, it is necessaryto solution treat at a temperature high in – field. Normally 25 to 85 °C (50 to 1500F) belowthe transus of the alloy. If high fracturetoughness or improved resistance to stresscorrosion is required, â annealing or solutiontreating may be desirable. However, heattreating – alloys in the range causes asignificant loss in ductility. These alloys areusually solution heat treated below the transus to obtain an optimum balance ofductility, fracture toughness, creep and stress-rupture properties. If the transus isexceeded, compressive properties of – alloys (especially ductility) are reduced andcannot be fully restored by subsequent thermaltreatment (Srinivasan and Venugopal, 2008).

Thermal Cycling

The heat treated specimens are thermal cycledin a specially designed thermal cyclingapparatus. The samples under theinvestigation are subjected to programmedcyclic heating and cooling temperatures in amuffle furnace for designed time period.Pneumatic piston actuator was built to cyclethe specimen in and out of the furnace. Wherethe required temperature was maintained inthe furnace. One heating and cooling operationof the sample is considered as one cycle. Thesamples are subjected to different number ofcycles during the experiment. The thermal cycleimposed had dwell time of 2 minutes in andout of the furnace. Samples were cooled by

23


forced air cooling at a pressure of 2 bars. Heattreated specimens were subjected to thermalcycling in the range of 500, 1000 and 1500cycles. Manual handling of the sample in thethermal cycling process is very difficult, tediousprocess and not more accurate. Henceautomation is necessary for material handling.

Thermal cycling is a temperaturemodulation process developed to improve the

performance, strength and longevity of a varietyof materials. During the thermal cyclingprocess, materials are alternately cooled andheated until they experience molecularreorganization. This reorganization “tightens”or optimizes the particulate structure of thematerial throughout, relieving stresses, andmaking the metal denser and more uniform(thereby minimizing flaws or imperfections).

Figure 1: Experimental Set Up

Figure 2: Schematic Representation of Experimental Set Up

24


The tighter structure also enhances the energyconductivity and heat distributioncharacteristics of the material. The behaviourof the materials under thermal cyclingconditions is affected by a number of factorssuch as the microstructure, thermalexpansion coefficients of micro constituents,loading level and thermal conditions (Hongbinet al., 1997; and Ismail and Kazim, 2004).Thermal Cycling minimizes “hotspots”,enhances cooling, and impedes the abilityand tendency of metals to vibrate.Significantly reducing vibration as a factor inmetal fatigue slows down the metal’s eventualfailure or breakage (Xu et al., 1994).

Hot Compression TestIsothermal hot compression tests wereconducted using FIE servo-controlled universaltesting machine. The hot deformation of thematerial was investigated by conducting hotcompression tests in the temperature rangeof 200-400 °C (in steps of 100 °C) and strainrate range of 0.01-1.0 s–1.

Hot compression testing has becomeincreasingly popular for several reasons, inparticular: (a) uniform deformation can bemaintained for large strains with properlubrication, (b) the compressive state closelyrepresents the conditions present in forging,extrusion and rolling process. Compressiontest on a cylindrical specimen is easy to obtaina constant true strain rate using anexperimental decay of the actuator speed. Itis convenient to measure the adiabatictemperature rise directly on the specimen andconduct the tests under isothermal conditions(Lin et al., 2008). The size of the specimenwas 10 mm in height and 10 mm in diameter.The load-stroke data were converted into true

stress-true strain curves using standardequations. The flow stress values werecorrected for the adiabatic temperature rise.The hot compression testing arrangement isshown in Figure 3.

Figure 3: Compression Testing Machine

Interpretations from Flow Curves

The flow curves are used to interpret the typeof mechanism involved in deformation underparticular condition. The flow curves for Heattreated and Thermal cycled Ti-6Al-4V alloyspecimen with various temperatures (200,300and 400 °C) and strain rate range of 0.01-1.0S–1 are shown in Figures 4-12.

Flow curves of different heat treated and500 thermal cycled Titanium alloy at differentstrain rates and at a constant temperature of200 °C is shown in Figure 4. It is observedthat, the flow stress decreases with theincrease of temperature, but its variation withstrain rate is low. At room temperature, the flowstress is found about 1240 MPa for specimensdeformed at a strain rate of 1 S–1. While at

25


400 °C, under the same strain rate level, itdecreases to 740 MPa. The general featuresof the curves are similar to those attemperatures of 200 and 300 °C. It is alsoevident that the rate of work hardening at lowertemperature in the initial stages of deformationis greater than that at 400 °C, and is slightlydependent on the strain rate (Lin et al., 2008).When comparing these curves, the mostobvious difference is the pronounced influenceof the test temperature. As before, the flowstress obtained at 200 °C is greater whencompared with those obtained for the other twohigher temperatures. At a strain rate of 0.01and at 200 °C temperature (Figure 4) thematerial exhibits a slight drop in flow stressindicating thermal softening behavior, However,for the rest of the temperatures, the stressstrain curves show steady state flow withoutany peak stress. Figures 5 to 12 the stressstrain behavior is different. For this strain rate,after‘ reaching the yield strength, the flow stressfor all temperatures decreases with strainfollowing different work hardening rates, except

for 200 °C where steady-state low wasobserved. With regard to the specimens testedat a strain rate of 0.01 for a given temperature,the flow stresses decreased with the increaseof strain, but a small rate of work hardeningappeared when the temperature was greaterthan 300 °C. From the flow curves, it can beconcluded that the work-hardening effect ispronounced at temperatures below 400 °C.Under the three tested strain rates and that therate of work hardening decreases markedlywith increase of temperature and strain rate.

From Figure 4, it is inferred that, true stressof the Solution Treated specimen (ST)increases upto 16% when compared toWithout Heat Treated specimens (WHT).solution treated specimen has higher stressvalue than Annealed specimen (AN) at StrainRate (SR) 0.1 and 1. True stress of theAnnealed specimen increases upto 4% whencompared to without heat treated specimens.

From Figure 5, it is inferred that, true stressof the solution treated specimen increases upto

Figure 4: Flow Curves of Different Heat Treated and 500 Thermal Cycled Titanium Alloyat Different Strain Rates and at a Constant Temperature of 200 °C

26


13% when compared to without heat treatedspecimens. solution treated specimen hashigher stress value than Annealed specimenat strain rate 0.1 and 1. There is no substantialdifference of true stress of the Annealedspecimen when compared to without heattreated specimens.

From Figure 6, it is inferred that, true stressof the solution treated specimen increasesupto11% when compared to without heattreated specimens. solution treated specimen

has higher stress value than Annealedspecimen at strain rate 0.1 and 1. Annealedspecimen. exhibits higher stress whencompared to without heat treated specimens.

From Figure 7, it is inferred that, true stressof the solution treated specimen increasesupto 6.7% when compared to without heattreated specimens. solution treated specimenhas higher stress value than Annealedspecimen at strain rate 0.1 and 1. Annealedspecimen.



27


From Figure 8, it is inferred that,true stressof the solution treated specimen increasesupto 5.8% when compared to without heattreated specimens. solution treated specimenhas higher stress value than Annealedspecimen at a strain rate of 0.1 and 1.Annealed specimens exhibits higher stressvalue when compared to Without heat treatedspecimens.

From Figure 9, it is inferred that, true stressof the solution treated specimen increases upto5.5% when compared to without heat treatedspecimens. When compared to Annealed andwithout heat treated specimens, solution treatedspecimen has higher stress value at strain rate0.1 and 1. Annealed specimen exhibits higherstress when compared to without heat treatedspecimens at strain rate of 0.1 and 1.



28


From Figure 11, it is inferred that, true stressof the solution treated specimen increasesupto 6% when compared to without heattreated specimens. Annealed specimenexhibits higher stress value when comparedto without heat treated specimens at strain rateof 0.1 and 1.

From the Figure 12, it is inferred that truestress of the solution treated specimen

From Figure 10, it is inferred that, true stressof the solution treated specimen increases 6%when compared to without heat treatedspecimens. When compared to Annealedspecimens, solution treated specimen hashigher stress value than at strain rate 0.1 and1. Annealed specimen exhibits higher stresswhen compared to without heat treatedspecimens at strain rate of 0.1 and 1.


Figure 10: Flow Curves of Different Heat Treated and 1500 Thermal Cycled TitaniumAlloy at Different Strain Rates and at a Constant Temperature of 200 °C

29


increases upto 6.3% when compared towithout heat treated specimens. Whencompared to Annealed specimens, solutiontreated specimen has higher stress value.Annealed specimen exhibits higher stressvalue when compared to without heat treatedspecimens at strain rate of 0.1 and 1.

At lower temperatures, the strain hardeningis more and steady state is obtained at higher

temperature. At lower temperatures, thematerial exhibits severe work hardeningfollowed by continuous flow softening. At highertemperatures, the flow softening behavior isdistinctly different. The initial work hardeningcomponent is reduced and the materialexhibits only steady-state flow behavior. Thisis good agreement with Wang et al. (2007). Itis observed that peak stress increases with



30


increasing of strain rates. But after peakstress, the rate of decrease in flow stress ishigh when the strain rate increases. The dropin flow stress after a peak at higher strain raterepresents flow softening (Manish et al., 2000).It can be found that flow stress decreases withdecreasing strain rate. Also flow softeningoccurs at higher strain rates and steady-stateflow occurs at lower strain rate. This is wellagreed with Sivakesavam and Prasad (2003).

It is observed that compressive specimenafter thermal cycling have good interfacebonding and fine sub grains in the size of 0.2-0.5 µm. The refinement of grains greatlyimproves the strength of the mattering .Solutiontreated specimen increases the resistingforces of dislocation movement and also

increases the difficulty of grain boundarysliding. Solution treated specimen attainsgreater strength and the hardness alsoincreases for higher number of cycles. Thereis no effect of thermal cycling after 1000cycles. True stress of the solution treatedspecimens shows higher value than Annealedand without heat treated specimens at400 °C. True stress of the is high at 500 cycleswhen compared to without heat treatedspecimen and than Annealed specimens.Truestress decreases upto 12% when the strainrate increases.

Vickers Hardness

Vickers hardness for Without Heat treated,annealed and solution treated specimens areshown in Table 1.

250 350 360 398

500 356 368 400

750 360 390 408

1000 360 398 413

1250 358 393 396

1500 355 383 388

Number of Cycles

Table 1: Vickers Hardness for Without Heat Treated, Annealed and Solution TreatedSpecimens

Vickers Hardness ofWithout Heat Treated

Specimens (WHT)

Vickers Hardness ofAnnealed Specimen (AN)

Vickers Hardness ofSolution TreatedSpecimens (ST)

It is observed that the Vickers hardness forWithout Heat Treated specimens (WHT),Annealed Specimens (AN) and SolutionTreated specimens (ST) are initially Increaseswith increase of number of thermal cycles upto 1000 thermal cycles and then it startsdecreasing.

Vickers hardness for Solution Treatedspecimens (ST) found more than Without HeatTreated and annealed specimens. A recent

study by Sharmilee et al. (2008) confirmedthat an increase in micro hardness with anincreasing number of thermal cycles, indicatingthat strain hardening is caused by dislocationgeneration and movement. In addition,repeated quenching from a high temperatureof range 200 to 400 °C is expected to add tothe concentration of vacancies, which formprismatic dislocation loops and increase thehardness as a result. The decrease in micro

31


hardness with increasing number of thermalcycles is suggestive of softening by damageaccumulation. The observations of decreasein micro hardness with increasing number ofthermal cycles and damage accumulation aresimilar to those reported by Nikhilesh andKrishan (2006).

CONCLUSIONHeat treated (annealed and solution treated)and thermal cycled titanium alloy specimens(500-1500 cycle at 450 °C and 2 minutes dwelltime) were subjected to Hot compression testsfor different strain rate (0.01, 0.1, 1) and atdifferent temperature (200 °C, 300 °C, 400 °C)by using a Universal testing machine. The flowbehaviour of different heat treated and thermalcycled specimens were studied. Comparativestudy of flow behaviour of titanium alloy wasmade. The optimum Thermal Cycling wasidentified.

The following conclusions were made fromthe present study.

• The flow stress decreases with decreasingstrain rate. Also flow softening occurs athigher strain rate and steady-state flowoccurs at lower strain rate. At lowertemperatures, the strain hardening is foundto be high and it decreases with increasein temperature.

• The temperature and strain rate significantlyaffect the flow stress in the isothermaldeformation.

• The flow stress decreases with increasingof deformation temperature and decreasingof strain rate.

• The flow stress of solution treated specimenshowed higher stress value than without

heat treated and annealed alloy over theranges of temperatures and strain rates dueto grain refinement and strengthening.

• The flow stress increases upto 500 cyclesand then decreases in all heat treatedspecimens.

• The strength of the annealed and solutiontreated specimens are increases withnumber of cycles increases.

• The hardness increases at greater level upto 500 cycles and marginally rises from1000 to 1500 cycles. When compared with,without heat treated specimen thepercentage of hardness increases up to9.5% for annealed and 12.8% for solutiontreated specimen.

• Solution treated and 1000 cycled Titanium(Ti-Al-4V) alloy was found to have morestrength.

ACKNOWLEDGEMENTThe authors wish to acknowledge the supportand encouragement for the fabrication ofThermal cycling equipment and extendedtesting facilities by the Department ofManufacturing Engineering, Faculty ofEngineering and Technology, AnnamalaiUniversity.

REFERENCES1. Froes F H and Hajliberger W T (1985),

“Synthesis of Coronets in TitaniumTechnology: Present Status and FutureResults”, pp. 95-102, TitaniumDevelopment Association.

2. Hongbin Geng, Shiyu He and Tingqan Lei(1997), “Thermal Cycling Behavior of As-Quenched and Aged Alloy”, Metallurgical

32


and Material Transactions A, Vol. 28A,pp. 1809-1814.

3. Ismail Ozdemir and Kazim Onel (2004),“Thermal Cycling Behaviour of anExtruded Aluminium Alloy/SiCpComposite”, Composites: Part B, Vol. 35,pp. 379-384.

4. Lin Y C, Ming-Song Chen and Jue Zhong(2008), “Prediction of 42CrMo SteelFlowstress at High Temperature andStrain Rate”, Mechanics ResearchCommunications, Vol. 35, No. 3,pp. 142-150.

5. Manish Dighe D, Arun Gokhale M, andMark Horstemeyer Fandmosher D A(2000), “Effect of Strain Rate on DamageEvolution in a Cast Al-Si-Mg Base Alloy”,Metallurgical and MaterialsTransactions A, Vol. 31, pp. 1725-1731.

6. Material Properties Handbook (1994),Titanium Alloys, ASM International.

7. Nikhilesh Chawla and Krishan K Chawla(2006), “Metal Matrix Composites”,Springer Science and Business MediaInc., New York, USA.

8. Prasad Y V R K and Rao K P (2005),“Processing Maps and Rate ControllingMechanisms of Hot Deformation ofElectrolytic Tough Pitch Copper in theTemperature Range 300-950 °C”,Materials Science and Engineering A,Vol. 391, pp. 141-150.

9. Sharmilee Pal, Bhanuprasad V V andMitra R (2008), “Damage Evaluation inDiscontinuously Reinforced AluminumMatrix Composites Subjected to Thermal

Cycling”, Materials Science andEngineering A, Vol. 489, pp. 11-20.

10. Sivakesavam O and Prasad Y V R K(2003), “Hot Deformation Behavior of as-Cast Mg-2Zn-1 Mn Alloy in Compression:A Study with Processing Map”, MaterialsScience and Engineering A, Vol. 362,pp. 118-124.

11. Srinivasan K K and Venugopal P (2008),“Compression Testing of Ti-6A1-4V in theTemperature Range of 303-873K”,Materials and Manufacturing Process,Vol. 23A, pp. 342-346.

12. Wang C Y, Wang X J, Chang H, Wu K andZheng M Y (2007), “Processing Maps forHot Working of ZK60 Magnesium Alloy”,Materials Science and Engineering A,Vol. 464, pp. 52-58.

13. Woei-Shyan Lee and Ming-Tong Lin(1997), “The Effect of Strain Rate andTemperature on the CompressiveDeformation Behavior of Ti-6Al-4V Alloy”,Journal of Materials ProcessingTechnology, Vol. 71, pp. 235-246.

14. Wood R A and Avor R J F (1972),“Titanium Alloy Handbook”, Report MCIC-HB-02, Battelle Memorial InstituteColumbia.

15. Xu Z R, Neuman A, Chawla K K, WolfendenA and Ligett G M (1994), “Thermal CyclingInduce Damage in Alumina Fiber/Magnesium Alloycomposite, in HighPerformance Composites, Commonality ofPhenomena”, October, TMS/ASM Materialsand Non Metallic Committee, TMSPublications.

flow behaviour of thermal cycled titanium (ti-6al-4v) alloy · components. titanium (ti-6al-4v)...

Documents